Pre-comment: As I have remarked in past editions of the YEAR BOOK, changes in life-styles frequently result in opportunities for new types of infections to appear or, as in this case, old infections to occur in new circumstances. We usually encounter Pseudomonas pneumonia as an infection of immunocompromised hosts. However, in this instance Pseudomonas pneumonia occurred in a healthy man who indulged himself in a very contemporary fashion.--D.E.R.
Abstract: The occurrence of Pseudomonas pneumonia in a previously healthy person living in the community is rare. The authors report a case of acute Pseudomonas pneumonia in a previously healthy man who apparently acquired the infection during use of a home whirlpool spa.
Man, 47, suddenly developed fever, pleuritic pain, and a cough productive of bloody sputum and was hospitalized 3 days later. He had smoked heavily for 25 years but had been well apart from nephrotic syndrome 8 years previously. Temperature was 39.4 C, and there were dullness to percussion and bronchial breathing over the right upper posterior part of the chest. A left shift of the white blood cells was found. Chest films showed a right upper lobe infiltrate with cavitation (Fig 1-1). The P$sub{O}{sub-sub-2} during oxygen inhalation at 2 L/minute was 48 mm Hg. Cefamandole and tobramycin were given intravenously, and ticarcillin was substituted for cefamandole when Pseudomonas aeruginosa was cultured from the sputum. Treatment continued for 3 weeks. The patient was well 2 months after discharge and showed complete clearing of the infiltrate and a decrease in size of the cavity. The sputum did not contain P. aeruginosa. Pulmonary function was normal. Studies of polymorphonuclear leukocyte function yielded normal results.
The patient was found to have installed a whirlpool spa a month before becoming ill and had used it four times. Two other persons had used the spa on one occasion without ill effects. Pseudomonas aeruginosa was isolated from spa water. Both this isolate and that from the patient's sputum were of serogroup 1, and they had identical antimicrobial susceptibility patterns.
Direct inhalation of aerosolized organisms during a prolonged session in the contaminated whirlpool spa was probably responsible for infection in this patient. No underlying disease or defect in defenses against bacterial infection was detected. Apparently the repeated inhalation of a large inoculum of P. aeruginosa can cause pneumonia in a normal adult. In contrast to the superficial, self-limited Pseudomonas folliculitis, Pseudomonas pneumonia carries a mortality approaching 70%. Early, appropriate antimicrobial therapy contributed to the favorable outcome in this case.
Post-Comment: In recent years, whirlpools, hot tubs, and even swimming pools have been associated with the development of Pseudomonas folliculitis. This relatively mild but troublesome skin infection occurs when the body is immersed in contaminated warm water. The lesions are small, itchy pustules scattered over the body, often following the outline of the bathing suit. Patients usually are afebrile but may complain of swollen lymph glands, particularly in the axilla. Both men and women may have mastitis--usually mild but not invariably so. External otitis can occur among those who dunked their heads under the water.
These infections first attracted attention when they occurred in groups of people, usually associated with use of public facilities such as whirlpools in hotels or health spas. Intensive use would exhaust the water's chlorine content, permitting Pseudomonas organisms to thrive. It was predicted that similar events could occur among the users of home whirlpools. Sure enough, such infections among family members and friends using home facilities are now being reported. The folliculitis usually requires no treatment. Be warned: Local use of steroid creams seems to prolong the course.
This episode of Pseudomonas pneumonia is a distinctly unusual consequence of using a whirlpool. The paper was chosen because we wanted to remind you of these new life-style-associated Pseudomonas infections--they seem to be increasing in frequency. I was impressed with the severity of this man's illness. That right upper lobe cavity was certainly a big price to pay for watching a televised baseball game for 90 minutes in sybaritic immersion!--D.E.R.
Is the Clean-Catch Midstream Void Procedure Necessary for Obtaining Urine Culture Specimens From Men?
Lipsky, Benjamin A.; Inui, Thomas S.; Plorde, James J.; Berger, Richard E.
Am. J. Med., February 1984, Vol. 76, Pg. 257-262, Ch. 4-3
Affiliation: Seattle
Pre-comment: Few concepts in the day-to-day management of patients have seemed so secure as the ``clean-catch'' urine and the ``10$sup{5}'' definition of a UTI. The following article and article 4-4 (N. Engl. J. Med. 311:560-564, Aug. 30, 1984) amply demonstrates that nothing taught during our formal medical training is sacrosanct. Thoughtful physician-scientists have reevaluated these two ``givens'' of clinical medicine and come up with important and practical new insight.--D.E.R.
Abstract: To determine whether the results of voided urine cultures in men are affected by meatal cleansing, midstream sampling, or circumcision status, 308 paired (initial and midstream) specimens were collected from 254 patients at a urology clinic. One half of the patients cleansed their urethral meatus with povidone-iodine prior to voiding. The rates of true bacteriuria (growth of 10$sup{4} or more colony-forming units/ml of urine with a single predominant species) and contamination (growth of 10$sup{3} or more colony-forming units/ml of urine with two or more colonial types) were compared in the various subgroups of collection techniques.
Of the 616 cultures (initial and midstream), 46 (7.5%) reflected bacteriuria, 10 (1.6%) were indeterminate, 66 (10.7%) were contaminated, and 494 (80.2%) were negative. Of the bacteriuric specimens, 38 (83%) grew organisms recognized as uropathogens (coliforms, Pseudomonas aeruginosa, and enterococci), and 8 (17%) grew other gram-positive cocci (staphylococci and nonenterococcal streptococci). Of the 46 specimens with true bacteriuria, the initial and midstream collections were identical in 22 of the 23 pairs of specimens.
There were no significant differences in the rate of positive cultures between circumcised and uncircumcised men (8.1% vs. 6.9%), between those who cleansed their urethral meatus and those who did not (8.9% vs 6.0%), or between midstream and initial urine specimens (7.8% vs. 7.1%). When only midstream specimens were considered, no significant differences in rates of bacteriuria were noted between cleansed and noncleansed patients (8.9% vs. 6.6%) or between circumcised and uncircumcised (8.5% vs. 7.2%) patients. When evaluation was limited to specimens from subjects who cleansed, there was no difference in rates of bacteriuria between initial and midstream specimens; there was, however, a difference between circumcised and uncircumcised men (12.3% vs. 6.0%). Contamination rates were not significantly different between circumcised and uncircumcised men or between those who cleansed their urethral meatus and those who did not. Midstream specimens were, however, less frequently contaminated than initial specimens (8.4% vs. 13.0%) by McNemar's test.
These data strongly suggest that circumcision status, meatal cleansing, and midstream sampling do not affect the results of voided urine cultures in men. Although only ambulatory veterans attending a urology clinic were studied, it is suspected that these findings are generalizable to other male populations. The present observations and those of others imply that the clean-catch midstream void procedure is often improperly performed. In view of its marginal impact on culture results, the difficulty elderly men frequently experience with this procedure, and the time required for instruction, it is recommended that routine voided urine specimens be collected by male patients without specific instructions. If the culture grows a single or predominant organism in a concentration of 10$sup{3} colony-forming units/ml, a second specimen should be collected either by the clean-catch midstream void technique with personnel supervision or by an invasive technique (i.e., catheterization or suprapubic aspiration). All other results are likely to be as reliable as those found with the clean-catch midstream void technique.
Post-Comment: Meatal cleansing did not lower contamination rates in this study, even in uncircumcised patients. With this information, urine specimens should clearly be considered adequate for processing even if breaks in the standard technique of collection have occurred. However, until confirmatory data are available, it is still advisable, whenever possible, to offer instruction for foreskin retraction, meatal cleansing, and midstream collection of urine for culture. One additional comment: The authors of this study were careful to refrigerate immediately the urine specimens--a crucial and frequently ignored step in the process of urinalysis and culture.--D.E.R.
Single-Dose Antibiotic Therapy of Urinary Tract Infection: Is It Appropriate in the Emergency Department?
McCabe, John B.; Hamilton, Glenn C.
Ann. Emerg. Med., June 1984, Vol. 13, Pg. 432-439, Ch. 4-6
Affiliation: Wright State Univ.
Pre-comment: Although it is now clear that single-dose antimicrobial therapy may be curative in infections of the lower urinary tract (1979 YEAR BOOK OF MEDICINE, pp. 33), it is not at all clear how best to use this information in an outpatient setting. The following article reviews the promise and problems of single-dose therapy.--D.E.R.
Abstract: Single-dose treatment of urinary tract infections would be advantageous for patients seen in the emergency department, compared with antibiotic treatment for 1-2 weeks. Treatment could be completed before the patient leaves, minimizing costs and compliance problems. Definitive diagnosis depends on demonstrating bacteriuria by urine culture. The accuracy of predicting bacteriuria from the presence of pyuria is 50% or less; the finding of bacteria in the urine is a better predictor of significant bacteriuria. The differentiation of upper from lower tract infection by clinical examination alone is imprecise. The only direct method of localization is bilateral ureteral catheterization, which is too invasive for routine use. Emergency patients tend to be less compliant than others, making supervised treatment especially appropriate. Treatment of urinary tract infection can help prevent long-term renal dysfunction.
Most studies of single-dose antibiotic therapy for urinary tract infection have been in nonpregnant women having acute, symptomatic infection. Cure rates with single-dose therapy have ranged from 33% to 100%, and have averaged 70% (table). The cure rate for upper tract infections is 50%, compared with 67% after prolonged treatment. In patients with lower tract infection, single-dose therapy yields cure rates comparable to those obtained with more prolonged treatment, except when cephalosporins are given. Side effects are less frequent with single-dose therapy. Treatment failures can be detected by appropriate follow-up of patients with upper tract infection.
Single-dose antibiotic therapy is appropriate for selected patients with urinary tract infection seen on an emergency basis and is the preferred method in nonpregnant women who have acute symptomatic infection. All patients should be followed up within a week of treatment.
Post-Comment: This article nicely reviews single-dose antimicrobial therapy from the perspective of outpatient practice. The authors summarize the difficulties involved in determining, at the time of the initial visit, whether infection is present and whether the upper tract is involved. In a related study, however, Schultz et al. (Mayo Clin. Proc. 59:391-397, 1984) have discouraged the use of single-dose therapy in an outpatient setting. Noting that a slightly higher relapse rate was observed with single-dose vs. multiple-dose trimethoprim-sulfamethoxazole therapy, they calculated that the additional physician and laboratory expenses related to the relapses overshadowed any cost savings of single-dose therapy. They further suggested that young women with symptoms of lower urinary tract infection and without vaginitis, diabetes, urologic disorders, or a history of recent infection may safely receive multiple-dose therapy (10 days) without pretreatment or posttreatment laboratory studies. According to their data, follow-up urinalyses and cultures are rarely useful once patients become asymptomatic, regardless of the choice and duration of therapy. It is clear from both of these studies that if persistent asymptomatic infection is routinely to be detected and treated, follow-up cultures will be required of all patients.
Those involved in primary care will need to carefully evaluate their laboratory resources and develop an algorithm of management appropriate to the type of patient seen in their environment.--D.E.R.
Bacteriuria in Elderly Institutionalized Men
Nicolle, Lindsay E.; Bjornson, Janet; Harding, Godfrey K. M.; MacDonell, J. A.
N. Engl. J. Med., Dec. 8, 1983, Vol. 309, Pg. 1420-1425, Ch. 4-7
Pre-comment: For elderly institutionalized men, the problems in managing urinary tract infection are not any easier. High prevalence of infection and an abysmal response to therapy, single-dose or multiple, seem to be the rule.--D.E.R.
Abstract: The authors obtained monthly urine samples from all noncatheterized male residents on two geriatric wards over a 2-year period to determine the occurrence and optimal management of bacteriuria in this population. Patients included 59 residents on the two wards on January 1, 1980, and 29 admitted during the 2-year study period after 29 residents died and 1 transferred. Mean number of chronic diseases diagnosed per patient was 2.6. The most frequent diagnoses were neurologic diseases. Urine specimens were collected at the first morning voiding using a clean-catch technique. When patients' urine specimens grew at least 10$sup{5} colony-forming units (CFU) of one or more organisms per milliliter (at least 10$sup{8} CFU/L) cultures were repeated 1 week later. Asymptomatic residents with two consecutive urine cultures containing the same organism(s) with at least 10$sup{5} CFU/ml received a single dose of trimethoprim-sulfamethoxazole orally (TMP-SMX; 320 mg of TMP and 1,600 mg of SMX) or tobramycin intramuscularly (100 mg). If single-dose therapy failed or a relapse occurred, the patient was randomly assigned to a group to be given antimicrobial therapy to eradicate bacteriuria or assigned to a group to receive no therapy. The treated group received a 2-week course of TMP-SMX or tobramycin.
Among the 88 men, the prevalence of bacteriuria was 33%, and the incidence was 45 infections per 100 patients per year. Outcomes after single-dose therapy for asymptomatic bacteriuria with 43 courses of trimethoprim-sulfamethoxazole and 23 of tobramycin included 15 cultures, 40 relapses, and 11 treatment failures. Thirty-six residents who had a relapse or experienced failure of single-dose therapy were randomly assigned to receive therapy to eradicate bacteriuria or to receive no therapy. All 20 residents who received no therapy remained bacteriuric. The 16 residents who received therapy had fewer months of bacteriuria after randomization, but at the end of the study, only 1 remained free of bacteriuria. Mortality (Fig 4-5) and infectious morbidity after randomization were similar in the two groups.
The authors conclude that asymptomatic bacteriuria is a common and persistent finding in elderly institutionalized men. Therapy is ineffective in curing bacteriuria. Thus, such therapy may result in superinfection, emergence of resistant organisms, or drug-related adverse effects. The authors recommend that asymptomatic bacteriuria should not be treated in elderly institutionalized men.
Post-Comment: It occurs to me that this high prevalence of asymptomatic bacteriuria creates another problem for acute-care hospitals. Elderly institutionalized patients are often transferred to acute-care facilities when infectious complications arise. Because the prevalence of asymptomatic bacteriuria approaches 30% in such populations, serious management errors will be made if the bacteriuria and pyuria are assumed to be the source of an acute febrile illness. Influenza, meningitis, pneumonia, diverticulitis, and a supervening symptomatic pyelonephritis are all diagnostic possibilities to be considered. The pyuria and bacteriuria may only represent a chronic asymptomatic infection and lead the physician away from more definitive diagnostic and therapeutic considerations.
For further discussion of prophylaxis of urinary tract infection in high-risk populations, see the excellent review by Stamm (Am. J. Med. 76:148-143, 1984).--D.E.R.
A Double-Blind Study of Oral Acyclovir for Suppression of Recurrences of Genital Herpes Simplex Virus Infection
Douglas, John M.; Critchlow, Cathy; Benedetti, Jacqueline; Mertz, Gregory J.; Connor, James D.; Hintz, Marie A.; Fahnlander, Anita; Remington, Michael; Winter, Carol; Corey, Lawrence
N. Engl. J. Med., June 14, 1984, Vol. 310, Pg. 1551-1556, Ch. 5-3
Pre-comment: In contrast to the failure of topical acyclovir in recurrent herpetic infections, orally administered acyclovir has now joined parenteral therapy as effective prophylaxis of recurrent disease.--D.E.R.
Abstract: Most patients with symptomatic recurrent genital herpes have five or more recurrences a year. A double-blind, placebo-controlled study of daily oral acyclovir therapy was undertaken in patients with frequent recurrences to determine its efficacy in preventing recurrences. Patients with culture-proved genital herpes simplex virus type 2 infection who had had six or more recurrences in the past year, but were in good general health, were studied. Patients received either acyclovir in capsule form in a dose of 200 mg five times daily or twice daily for 4 months or placebo capsules. Fifty-one patients received acyclovir five times daily, 52 were treated twice daily, and 50 received placebo. The 4-month medication period was completed by 143 patients.
Genital herpes recurred during the 4-month study period in 94% of placebo patients, 35% of patients treated twice daily with acyclovir, and 29% of those treated five times a day (Fig 5-2). The median times to the first clinical recurrence were 18 days in the placebo group, but more than 4 months in both active treatment groups. The duration of clinical recurrences was shorter in acyclovir-treated patients. Only 1 patient withdrew because of a possible adverse drug effect. The recurrence rate returned to baseline after acyclovir was discontinued.
Acyclovir taken orally is the first treatment that clearly suppresses the rate of clinical recurrence of genital herpes. It is a maintenance treatment, since its therapeutic effect persists only for as long as the drug is continuously used. Long-term oral acyclovir therapy has not eliminated recurrences. The drug was well tolerated over 4 months in this study, but longer-term studies are needed. A choice between episodic and continuous oral acyclovir therapy cannot yet be made.
Post-Comment: The conclusions of this paper are similar to those reported by Mindel et al. (Lancet 2:57-59, 1984) and Straus et al. (N. Engl. J. Med. 310:1545-1550, 1984): oral acyclovir is highly successful in preventing recurrent herpes genitalis. The prophylactic effectiveness of this drug is further dramatized by its ability to suppress recurrent genital or oral disease in immunocompromised populations as reported by Straus et al. (Ann. Intern. Med. 100:522-524, 1984) and Saral et al. (Ann. Intern. Med. 99:773-776, 1983).
Therapy (as opposed to prophylaxis) of genital herpes simplex infection with oral acyclovir has been less effective. Mertz et al. (JAMA 252:1147-1151, 1984) noted more rapid resolution of signs, symptoms, and viral shedding in patients with primary genital herpes simplex infection, but neither these authors nor Reichman et al. (JAMA 251:2103-2107, 1984) could demonstrate that acyclovir lessened any of the symptoms of recurrent infection after symptoms developed.
All studies to date indicate that once the prophylactic acyclovir is discontinued, relapses continue at the same frequency and intensity as before. Apparently, acyclovir does not eliminate the latent virus from the nerve ganglia.
For the present, the use of the topical preparation of acyclovir is limited to herpes keratoconjunctivitis and severe primary herpes genitalis. Intravenous therapy should be reserved for symptomatic relief of severe zoster infection and in the therapy of disseminated varicella-zoster and mucocutaneous herpes simplex virus in immunosuppressed children and adults. The oral preparation has now proved useful as therapy of primary genital herpes and for the suppression of recurrent infection in both normal and immunocompromised patients. Therapy of recurrent herpes genitalis with any form of acyclovir is not effective.
There are two very important new indications for acyclovir. First, Skoldenberg et al. (Lancet 2:707-711, 1984) have reported increased survival and decreased morbidity with acyclovir when compared with vidarabine in patients with herpes simplex encephalitis. This is important news. The results of a similar American trial have not yet been reported. Second, a decreased incidence of viridans streptococcal septicemia associated with use of prophylactic acyclovir in an immunosuppressed population has been noted by Ringden et al. (Lancet 1:744, 1984). They reasoned that a decreased incidence of mucocutaneous herpes infection secondary to the use of acyclovir led to the decreased incidence of secondary infection and septicemia.
Before we rush off to add acyclovir to the water supply of our hospitals, the inevitable side effects should be mentioned. Intravenous acyclovir can cause renal toxicity if given as a bolus or if given to dehydrated patients. Encephalopathy in severely ill patients continues to be reported by Vartian and Shlaes (Ann. Intern. Med. 99:568, 1983). Finally, herpes simplex isolates resistant to acyclovir have been recovered in patients receiving prophylaxic therapy. This finding deserves close scrutiny as oral acyclovir is expected to be used widely now that it has been released for general use.--D.E.R.
Apparent Failures of Endocarditis Prophylaxis: Analysis of 52 Cases Submitted to a National Registry
Durack, David T.; Kaplan, Edward L.; Bisno, Alan L.
Abstract: The authors reviewed 52 cases of apparent failure of endocarditis prophylaxis that were reported to a national American Heart Association (AHA) registry since 1979. Twenty-two other cases were excluded because of inadequate information. Seven of the 52 evaluable patients were children. One third of 42 patients had reported hypersensitivity to penicillin. Mitral valve prolapse accounted for one third of the cases, various congenital abnormalities for 29%, and rheumatic heart disease for 21%. Ten patients had prosthetic valves. Dental procedures provided 92% of events implicated as the cause of endocarditis. Symptoms began within 2 weeks of the responsible procedure in half of the cases and within 5 weeks in almost 80% of the cases.
Three fourths of the cases of endocarditis were due to viridans streptococci, and 7 were due to Staphylococcus aureus. Most patients had received penicillin orally as prophylaxis, but only 12% received regimens currently recommended by the AHA (table). The infecting organism was sensitive to the medications used for prophylaxis in 63% of 43 evaluable patients. Ninety percent of the patients were cured, and 5 died. Only 1 of the 39 patients with viridans streptococcal infection died. Seven patients required valve replacement.
Failures of endocarditis prophylaxis may be more prevalent than has been thought. They can occur even when the infecting microorganism is sensitive in vitro to the antibiotic being used. Most of the present patients were receiving regimens not conforming to current recommendations. A prolapsing mitral valve is a common site for development of endocarditis in this setting. Prophylaxis for most dental procedures may be warranted in these cases, but the cost-benefit ratio remains uncertain. The outcome of treatment for infective endocarditis developing despite attempted prophylaxis is usually favorable.
Post-Comment: This interesting paper raises almost as many questions as it answers. Be that as it may, I'm glad the authors did this retrospective analysis of the apparent failures of antimicrobial prophylaxis in endocarditis. First, but not really discussed by the authors, ``failure'' in some of the patients might simply have meant inadequate antimicrobial absorption and, in essence, ``no prophylaxis.'' Most of the patients received oral penicillin V. Studies dating back in the 1950s show that some people don't absorb it predictably (N. Eng. J. Med. 257:249, 1957). The predominance of penicillin-sensitive Streptococcus viridans as the infecting microorganism would also suggest this possibility. Further, as the authors note, in only 12% of cases were AHA-recommended regimens followed. The absence of any denominator--or a control group--leaves us, as the authors note, still uncertain as to whether antimicrobial prophylaxis does or does not protect against endocardial infection. Thus, we are in the same boat we started out in.
But the article does add several new pieces of information to think about. First, how remarkable the change in underlying valve lesion from times past! Mitral prolapse was noted first, then congenital lesions, and bringing up the rear, rheumatic valvular disease. Very, very different from the 1950s.
Second, dental procedures precede endocarditis in a remarkably high number of these cases.
Third, these cases were, in general, just as easy to treat as infections arising in individuals who have not received prophylaxis.
My advice: antimicrobial prophylaxis makes good intellectual sense. Continue to use it when the patient (1) is undergoing a procedure with a high risk of producing bacteremia, particularly dental procedures, and (2) has a valve lesion (though some will debate whether this includes asymptomatic mitral valve prolapse) known to predispose to bacterial endocarditis. When you do it, give enough antimicrobial.--D.E.R.
Treatment of Streptomycin-Susceptible and Streptomycin-Resistant Enterococcal Endocarditis
Wilson, Walter R.; Wilkowske, Conrad J.; Wright, Alan J.; Sande, Merle A.; Geraci, Joseph E.
Ann. Intern. Med., June 1984, Vol. 100, Pg. 816-823, Ch. 6-2
Abstract: Fifty-six patients seen in a 12-year period at the Mayo Clinic and Foundation with enterococcal endocarditis were given antimicrobial therapy for 4 weeks. Thirty-six had streptomycin-susceptible infection. All patients had at least two positive blood culture results within 48 hours and compatible clinical findings. Aqueous penicillin G was given by continuous infusion for 24 hours in a dose of 20 X 10$sup{6} units. Patients with streptomycin-susceptible infections also received that drug, 7.5 mg/kg intramuscularly every 12 hours. The 20 with streptomycin-resistant infections received gentamicin in a nominal dose of 3 mg/kg daily. The dose was adjusted to produce 1-hour peak serum levels of about 3 or 5 mu g/ml. Urinary and biliary tract infections were the most common suspected portals of entry in these patients.
The overall relapse rate was 12.5%; the rate was 8% in patients with streptomycin-susceptible infections and 20% in those with resistant infections. No patient with symptoms for less than 3 months relapsed. The dose of gentamicin could not be related to the relapse rate. All patients who relapsed had mitral valve infection; more of these patients had symptoms for longer than 3 months. Mortality was 15% in patients with streptomycin-resistant infections and 5.5% in those with streptomycin-susceptible infections. There were 5 deaths in all, for a mortality of 9%. Seven streptomycin-treated patients had abnormal vestibular dysfunction, and 1 patient had vestibular toxicity from gentamicin. Nephrotoxicity was most frequent in patients given the higher dose of gentamicin.
Patients with mitral valve enterococcal endocarditis and symptoms for 3 months or longer should receive penicillin and aminoglycoside therapy for at least 6 weeks. Patients with symptoms for a shorter time and those with aortic valve infection can be treated effectively in 4 weeks. Gentamicin should be used in place of streptomycin if the organism is resistant to the latter drug. Patients who relapse should be retreated for at least 6 weeks.
Post-Comment: This is a nice article from a Mayo Clinic group that has had a vast experience with infective endocarditis.
Enterococci cause 10%-15% of cases of bacterial endocarditis now seen in the United States. Enterococci are outranked only by viridans streptococci and Staphylococcus aureus. Treatment of these patients has been more difficult because enterococci are relatively resistant to most antibiotics. Thus, combination therapy, generally penicillin plus streptomycin, has been commonly used--and with good success. Now, and very much in evidence here, strains highly resistant to streptomycin are appearing with increasing frequency. They were rare before 1970. Now they account for as much as 40% of the isolates in some series. Gentamicin looks like the drug to couple with penicillin or ampicillin in these cases.
Several observations in the above report bear comment. First, the increased frequency of relapse in patients whose disease was long-standing (more than 3 months) was striking (0 vs. 44%). The same was true for mortality (2.5% vs. 25%). Thus, the authors suggest that all patients with disease lasting longer than 3 months receive 6 rather than 4 weeks of treatment. Patients with mitral valve involvement also had higher relapse rates than did those with aortic disease (25% vs. 0). Thus, the authors recommend prolonging treatment for these patients.
In an editorial accompanying this article, Scheld and Mandell make the point that duration of disease before you make the diagnosis is often tough to decide. Thus, they continue to recommend 6 weeks of treatment for this disease. They also recommend a vancomycin-gentamicin combination for penicillin-allergic patients (Arch. Intern. Med. 100:904-905, 1984).--D.E.R.
The Effect of Penicillin Therapy on the Symptoms and Signs of Streptococcal Pharyngitis
Abstract: No placebo-controlled study of the effect of antibiotic therapy on the course of streptococcal pharyngitis in children has ever been reported. The author compared the clinical responses of children with streptococcal pharyngitis during 48 hours when penicillin was given, or when specific treatment was withheld for 2 days. Thirty-five children, aged 5 and older, who had acute pharyngitis without rhinorrhea or cough, were included in the study. Each of 17 children received 300,000 units of penicillin G intramuscularly, while 18 of them received an oral placebo syrup. All children in the study had group A streptococci isolated from throat cultures. The two groups were demographically and clinically comparable.
Objective responses were compared. The total duration of fever, tender cervical nodes, injection of the pharynx, and tonsillar exudate was significantly less in treated patients. The duration of sore throats, and symptomatic responses to penicillin and placebo are shown in Figure 8-1. The parents felt that their children were completely well significantly sooner when given penicillin than when given placebo. The injections consis tently caused local soreness, but no serious adverse reactions occurred.
These findings confirm the clinical impression that specific antibiotic therapy promptly relieves symptoms in most children with streptococcal pharyngitis. Prompt treatment of classic cases could provide rapid symptomatic relief, and might reduce the likelihood of suppurative complications.
Post-Comment: I wonder how many of you feel this simply belabors the obvious? Interestingly, there have been diametrically opposed opinions at large about whether antimicrobial therapy does or does not hasten symptomatic improvement ever since the classic Warren Air Force Base studies of streptococcal pharyngitis reported 33 years ago.
This simple study seems to answer the question. Yes, kids feel better faster when their streptococcal pharyngitis is treated early. That this is not inconsequential--for both child plus probable working mother--is stressed in an editorial by Hall and Breese which accompanys this article (Pediatr. Infect. Dis. 3:7-8, 1984). Further, Hall and Breese went back and reanalyzed the lovely earlier studies of Rammelkamp, Wannamaker, Denny, and their associates. In retrospect (although not recognized then) these studies on young adult male Air Force recruits seemed to show the same thing.
For those of you who are wondering about the ethics of withholding definitive therapy from children with proved streptococcal disease, relax. They didn't. When seen at revisit 48 hours after the initial examination, all children with streptococcal pharyngitis confirmed by culture were given an intramuscular injection of 600,000 units of benzathine penicillin G.
The author closes his article with a nice little limerick by Dr. Wallace Clyde, Jr., of Chapel Hill.
Pharyngitis from strep of group A
Should be treated without much delay
For glomerulitis,
carditis, arthritis
May be just days to three weeks away.
Although this may not be as true today as in 1948, it still has meaning.--D.E.R.
Clinical Efficacy of Ceftazidime: Treatment of Serious Infection Due to Multiresistant Pseudomonas and Other Gram-Negative Bacteria
Abstract: The authors evaluated the beta-lactamase-stable cephalosporin ceftazidime in 57 patients with suspected or proved infections against which the agent was expected to be effective. Ceftazidime was given intravenously or intramuscularly in doses of 1-6 gm daily. Treatment was continued for 4-42 days. Forty-nine patients were evaluated for efficacy and 56 for toxic effects. Most patients had significant underlying disease; half were in poor or critical condition. Mean age was 40 years. The most common disorders were urinary tract infection and acute bacterial exacerbation of pulmonary infection.
Ceftazidime inhibited all initial isolates of Enterobacteriaceae at a concentration of 8 mg/L or below and a majority of Pseudomonas aeruginosa isolates at 12 mg/L or less. The mean serum drug concentration after infusion of 1 gm over 30 minutes was 62 mg/L. The overall clinical response rate was 84%, and the bacteriologic response rate was 72%, excluding cases of cystic fibrosis. Nine patients in the series had bacteremia and 5 had pneumonia. Seven patients had abscesses or soft tissue infection and 4 had osteomyelitis or septic arthritis. No major adverse effects were observed. Resistance to ceftazidime occurred for Pseudomonas in 5 patients with cystic fibrosis and for Enterobacteriaceae in 2 other patients.
Ceftazidime is an effective and safe antimicrobial for use in patients with serious infections due to multiply antibiotic-resistant P. aeruginosa and other aerobic gram-negative bacilli. More than 80% of the patients in this study responded clinically, and major adverse effects did not occur. Suggested doses of ceftazidime are 2 gm every 8-12 hours for systemic Pseudomonas infections and 1 gm every 8-12 hours for infections due to most other gram-negative bacilli. Dose adjustment is necessary in cases of renal insufficiency.
Post-Comment: Newer third-generation cephalosporins continue to appear, and an annual update in this market is probably warranted. The pharmacy and therapeutics committees of many hospitals have considered the differences between the third-generation cephalosporins to be so slight that price alone becomes the determining factor for use. Others have included the half-life of these agents as a determining factor, choosing those with improved pharmacokinetics that allow for dosing two or three times per day. Ceftazidime may ultimately prove to have no unique advantages and be required to stand in the ``price line'' with the rest. However, as demonstrated in this article, the drug appears to offer improved anti-Pseudomonas activity. The drug should be carefully considered in hospitals where multiply resistant Pseudomonas strains are prevalent.
Two cautions should be noted. First, the drug had relatively poor activity against gram-positive cocci and anaerobic bacteria. Ceftazidime alone would be inadequate empiric therapy if meningitis or abdominal infections are possibilities. Second, hospitals with extensive use of other third-generation cephalosporins have noted a rapid emergence of resistant hospital flora. I suspect ceftazidime will suffer a similar fate against hospital-acquired Pseudomonas infections. Careful monitoring of the antibiograms of nosocomial infections by the hospital infections control committees is essential.--D.E.R.
Intravenous Sulfamethoxazole and Trimethoprim for Serious Gram Negative Bacillary Infection
Abstract: A parenteral form of sulfamethoxazole (SMX) and trimethoprim (TMP) is now available. The drugs are primarily bacteriostatic, but the combination rapidly kills most susceptible gram-negative bacteria. The authors evaluated parenteral SMX-TMP treatment in 7 patients with serious infections caused by facultative gram-negative bacilli. Three patients were bacteremic, 3 had pneumonia, and 1 each had meningitis, peritonitis, pyogenic liver abscesses, and urinary tract infection. Three patients with renal failure were given SMX-TMP to prevent aminoglycoside-induced nephrotoxicity. Three were allergic to penicillin, and bacterial resistance to other agents was present in 2 patients. The usual doses were 8-14 mg of TMP per kg and 40-70 mg of SMX per kg daily.
The patients were cured of their serious infections by treatment with SMX-TMP alone. Acinetobacter meningitis due to an organism resistant to several effective agents responded to SMX-TMP therapy. Both components penetrate well into the cerebrospinal fluid in the presence of meningeal inflammation. One of the patients with renal failure became worse during treatment, but SMX-TMP can be safely used in such patients if serum concentrations are closely monitored and appropriate dose modifications are made. Adverse drug reactions in 3 patients included oral moniliasis, transient leukopenia, and fluid overload.
Parenteral SMX-TMP therapy is a promising approach to serious or life-threatening infections in patients allergic to beta-lactam antibiotics, with mild renal failure, or with infection by organisms resistant to other available drugs. The cost is much less than that of the new broad-spectrum cephalosporin antibiotics. Bacterial resistance has not emerged during SMX-TMP therapy. Prospective clinical trials are needed to compare the efficacy of SMX-TMP with that of other broad-spectrum antimicrobial agents.
Post-Comment: Although commonly used in the United Kingdom for therapy of life-threatening gram-negative infection, parenteral sulfamethoxazole (SMX) and trimethoprim (TMP) have received little attention in the American literature, partly because of the ``bacteriostatic'' label of the individual agents. However, the combination is often bactericidal against susceptible gram-negative pathogens. And, as noted in this article, SMX-TMP has proved efficacy against gram-negative bacillary infection when the pathogen is known to be sensitive. However, I would be reluctant to use SMX-TMP for empirical therapy of life-threatening illness when the broad-spectrum, albeit more expensive, cephalosporins are available.
The oral preparation of SMX-TMP has already gained prominence in the therapy of acute otitis media and sinusitis, primarily because of its effectiveness against most ampicillin-resistant H. influenzae. The parenteral form of SMX-TMP also holds promise in the empirical therapy of acute bacterial meningitis. Penetration into the CSF is excellent. One caution: despite good in vitro activity against the enterococcus, clinical failures have been reported with SMX-TMP for therapy for enterococcal urinary tract infection and sepsis (Goodhart G.L.: JAMA 252:2748-2749, 1984).--D.E.R.
Complications Associated With Staphylococcus aureus Bacteremia
Abstract: The authors reviewed the complications of 39 episodes of Staphylococcus aureus bacteremia occurring in 37 patients in a 1-year period. All patients were males, with a mean age of 62 years. Thirty-four episodes of bacteremia were nosocomial, associated with intravascular catheters in 20 instances and with dialysis access sites in 6. No source of infection was identified in 2 nosocomial and 2 community-acquired episodes. All patients but 1 had underlying medical illness, which was often chronic. Cardiovascular, pulmonary, renal, and gastrointestinal disorders were most frequent.
Complications developed in 36% of bacteremias and in 30% of catheter-associated bacteremias (table). Only 1 suppurative complication occurred in 13 catheter-associated episodes in which antistaphylococcal therapy was begun empirically at the time of the initial positive blood culture. No infective endocarditis was observed. Complications of bacteremia were responsible for 4 of the 11 deaths. These deaths were related to shock or adult respiratory distress syndrome. Medical treatment of bacteremia was highly variable. Surgical treatment was necessary in 11 patients, 3 of whom had two procedures. Seven operations were related to sources of infection and 7 to complications. Most suppurative complications required operation.
Staphylococcus aureus bacteremia causes considerable morbidity and mortality in hospitalized patients, even apart from endocarditis. Chronic underlying medical illness is frequently present. Teichoic acid antibody titers are a sensitive but nonspecific marker of infectious complications associated with staphylococcal bacteremia. Shorter courses of antibiotic therapy might be feasible in patients without heart valve disease if there is a rapid response to treatment with clearance of bacteremia and if no complications occur in the first 2 weeks.
Post-Comment: This review of one hospital's recent experience with staphylococcal bacteremia permits several comments about this all-too-common infection, so frequently iatrogenic in origin.
First, staphylococcal bacteremias are primarily an intrahospital problem. In most series today, 60%-85% originate within the hospital walls. Second, as the British would put it, ``vascular access sites'' are far and away the most common point of origin. In this study, 20 of 39 episodes were associated with indwelling catheters. The same was true in two other nice studies of the problem reported this year (South. Med. J. 76:1131-1135, 1983; Br. Med. J. 288:300-303, 1984). Third, and again noted here and in the other series, staphylococcal bacteremia is generally an infection of the sick, compromised host. Mortality continues to range upwards from 25%, irrespective of therapy.
Fourth, assays of teichoic acid antibodies really don't help very much. Titers tend to be high in patients with the most serious and extensive staphylococcal infections and lower or detectably absent in those with less severe infections, but I would not recommend relying very heavily on them--and the investigators who have done most of the work on them come out about the same (Arch. Intern. Med. 144:261-264, 1984).
Lastly, I've had to modify my thinking about the frequency of endocarditis as a complication of persistent staphylococcal bacteremia. I used to state that endocardial involvement occurred in about 60% of patients with staphylococcal bacteremia, but today I am clearly wrong. Endocarditis is primarily a disease of patients who acquire their bacteremia outside the hospital. In hospital-acquired staphylococcal bacteremia the incidence of endocardial infection is low: none in this series, only 2 of 47 patients in a prospective study in Buffalo (South. Med. J. 76:1131-1135, 1983), and only 79 in 400 episodes of staphylococcal bacteremia reported from St. Thomas's Hospital in London (Br. Med. J. 288:300-303, 1984).
So, this is a common intrahospital problem. It is easy to identify. The primary site is usually obvious--and usually one of those damned intravenous catheters. It is easy to treat. But the treatment takes a lot of expensive hospital days, and the frequency of serious underlying disease means it continues to kill at least 1 of 4 who acquire this infection. We need to bend all efforts to prevent it.--D.E.R.
Pulmonary Cryptococcosis: Radiologic-Pathologic Correlates of Its Three Forms
Feigin, David S.
AJR, December 1983, Vol. 141, Pg. 1263-1272, Ch. 9-7
Affiliation: Univ. of California, La Jolla
Abstract: The author reviewed the findings in 23 human cases of pulmonary cryptococcosis and 3 cases in zoo animals in which both radiographic and pathologic data were available for comparison. There were 21 cases of inflammatory infection, chiefly granuloma, and 5 of fungal mass formation without significant inflammatory reaction. The animals involved were proboscis monkeys. In 44 other clinical cases, only radiographic data were available.
The fungal masses were well-delineated formations 2-5 cm in diameter, none abutting the pleura (Fig 9-4). Three masses were slightly lobulated. No associated abnormalities were apparent. The inflammatory cases were associated with widely varying clinical presentation and symptoms. Segmental consolidation was the main radiographic finding in 9 cases (Fig 9-5). All these infiltrates were sharply circumscribed, and 4 had prominent air bronchograms. Cavitation was noted in only 1 case. Five patients had poorly delineated mass lesions with irregular, hazy borders. One of these masses was definitely, and another questionably, cavitary. ``Infiltrative masses'' were present in 4 cases. Frontal and lateral views showed a poorly defined or lobulated mass with irregular, hazy borders. The 90-degree view showed an infiltrate having the shape of a specific lung segment. Air bronchograms were present in 3 cases. Two other patients had a cavitary nodular infiltrate (Fig 9-6) and multiple tiny nodules, respectively, with marked lymph node enlargement. Consolidation of an entire lung was noted in one animal. Consolidative infiltration was present in 13 of the 44 cases without pathologic data, and poorly defined nodules in 16 cases. Six other patients had ``infiltrative masses.'' Cavitation was probably or definitely present in 5 masses and 2 infiltrates. Node enlargement was present in 11 cases and pleural effusion in 5.
The ``infiltrative mass'' or ``mass-like infiltrate'' was the most distinctive radiographic feature of pulmonary cryptococcosis in this series. It represents a granulomatous inflammatory reaction, often accompanied by lymph node enlargement. The well-circumscribed lung mass represents a gelatinous collection of fungal organisms with thick capsules. Airway colonization without parenchymal infiltration leads to no significant radiographic findings.
Post-Comment: Many physicians are unaware of the fact that cryptococcosis can present as a pulmonary problem. They tend to think about it only as an infection of the CNS. But this is an infection acquired principally through inhalation. While pulmonary infections are inapparent in most patients, the infection can produce significant symptomatic disease. The microbe is a budding yeast form of fungus that is commonly found in soil throughout the world. Recently pigeons have been implicated as the villains responsible for its spread, and the organism is much more common where pigeon excrement is found.
The radiographic findings, as noted here, can be fairly characteristic. Infiltrative masses are very commonly misdiagnosed as tumor. The radiographic features of pulmonary cryptococcosis worth remembering include the presence of such masses. Cavitation, fibrosis, lobar collapse, and calcification are uncommon. Enlarged hilar nodes occur in some instances.
Now for some irrelevant information that may improve either your roundsmanship or your batting average in the game Trivial Pursuit: Most microbes that infect man are quite fastidious--they attack humans but not other creatures. Not so with cryptococci. Like salmonellae, they'll infect almost any member of the animal kingdom. The authors tell us that cases of pulmonary or CNS cryptococcosis have been documented in cheetahs, sea lions, Galapagos tortoises, geese, diamondback rattlers, goats, langurs, lemurs, proboscis monkeys, and of course, pigeons--all at the San Diego Zoo. Wow!--D.E.R.
Controlled Evaluation of the Protective Efficacy of Pneumococcal Vaccine for Patients at High Risk of Serious Pneumococcal Infections
Shapiro, Eugene D.; Clemens, John D.
Ann. Intern. Med., September 1984, Vol. 101, Pg. 325-330, Ch. 9-9
Affiliation: Yale Univ.
Pre-comment: Some of you may be getting a bit weary of my annual harping about the suboptimal use of pneumococcal vaccine. In the several years since it was licensed, pneumococcal vaccine has been spectacularly underutilized. There are probably several reasons for this. Among them is the continuing uncertainty about how effectively the vaccine can prevent pneumococcal disease among those patients for whom it is recommended. The following paper demonstrates quite nicely that it works rather well.--D.E.R.
Abstract: The use of pneumococcal vaccine in patients at risk of serious infection is controversial, partly because efficacy trials have involved chiefly healthy young adults in epidemic circumstances. A case-control study has been done to evaluate pneumococcal vaccine in adults with current indications for vaccination. Cases with a first episode of documented systemic pneumococcal infection acquired in the community were compared with controls matched for age, date of hospitalization, and duration of indications for vaccination before hospitalization. Nearly 90% of the 90 cases had bacteremic pneumonia.
Six (7%) of the 90 cases and 16 (18%) controls had received pneumococcal vaccine, for an odds ratio of 0.33. The protective efficacy of the vaccine was 67%, and this was virtually unchanged by adjustment for potential confounding variables. The vaccine was 77% effective for patients at a moderately increased risk of pneumococcal infection, but it was ineffective for severely immunocompromised patients. Its overall protective efficacy for all patients aged 55 and older, after control for other indications for vaccination, was 70%.
Pneumococcal vaccine appears to confer substantial protection against systemic pneumococcal infection in the elderly and in patients who have disorders associated with a moderately increased risk of pneumococcal infection. More and larger studies of the vaccine are needed. It would be helpful to assess the efficacy of the vaccine against infections caused only by serotypes contained in the vaccine.
Post-Comment: I told you so. This carefully conducted study indicates that the vaccine was 77% effective among patients at risk of serious pneumococcal infection. Now, that's not bad, especially when you consider that all pneumococcal types (there are over 80 of them) were included in the study. The vaccine contains antigens of the 23 most common types, and thus cannot be expected to protect against all pneumococcal infections.
I believe pneumococcal vaccine should be used routinely by internists to protect their patients who are at special risk of serious pneumococcal infection. This includes adults with certain chronic diseases: cardiovascular and pulmonary disease, cirrhosis, renal failure, alcoholism, and CSF leaks. Quite predictably, the vaccine does not work as well in immunocompromised patients. However, many prevention-minded physicians also immunize their patients with multiple myeloma, renal failure, and Hodgkin's disease. Most important, anyone with splenic dysfunction or asplenia should be immunized. The vaccine is also indicated for all healthy adults over age 65. (Medicare will provide the vaccine free.)
I believe sufficient data are now available to satisfy the doubting Thomases. The time has come to stop ruminating and start immunizing.--D.E.R.
Role of the Parasympathetic System in Airway Obstruction Due To Emphysema
Gross, Nicholas J.; Skorodin, Morton S.
N. Engl. J. Med., Aug. 16, 1984, Vol. 311, Pg. 421-425, Ch. 10-3
Affiliation: Loyola Univ., Maywood, Ill.
Abstract: There is some evidence that anticholinergic agents are more potent bronchodilators than adrenergic agents are in patients having chronic bronchitis and emphysema. The roles of parasympathetic and sympathetic mechanisms were examined in a single-blind, placebo-controlled crossover study in 10 men with emphysema and moderate to severe obstruction of large and small airways. All had a reduced CO diffusing capacity. None had asthmatic features. Patients inhaled albuterol, 180 mu g, for 4 hours and then inhaled atropine methonitrate, 1.5 mg. The next day they inhaled the drugs in reverse order. Lung function was monitored throughout.
The FEV$sub{1} rose to a plateau after the second dose of either agent (Fig 10-3). All but 1 of the patients who received the adrenergic agent first had a further rise in FEV$sub{1} after atropine treatment, but the reverse was not the case. The response to atropine was greater than that to albuterol at most intervals, but the responses to the two agents correlated closely. Lung volumes (table) were consistent with the data on airflow. Both drugs were well tolerated.
Atropine alone provided as much bronchodilation as was possible in these emphysematous patients, suggesting that noncholinergic mechanisms of bronchoconstriction are not very important in this setting. Adrenergic and anticholinergic agents may act through a common cholinergic pathway, because considerable bronchodilation was obtained with maximal doses of both. Anticholinergic therapy provides a logical approach to treatment of patients with emphysema.
Post-Comment: The bronchodilator effect of atropine has been known for a long time, but its usefulness has always been compromised by its small therapeutic index, regardless of whether the drug is administered by aerosol or by mouth. Ipratropium, oxytropium, and atropine methonitrate are synthetic quaternary ammonium compounds with atropine-like activity which have been in use in Europe for a long time. Their advantage is that they are poorly absorbed when administered via aerosol but have just as much bronchodilator activity as atropine. Ipratropium (SCH 1000) will almost certainly be released in the United States in 1985 under the name of Atrovent, and it is likely that oxytropium, which differs only in potency, may be released soon after. The following is a brief summary of an excellent review of the field by the authors of the article abstracted above.
Atropine and related anticholinergic alkaloids have been used for breathing problems for at least 200 years but always with the problem of toxicity. (Cubeb cigarettes which were popular 50 years ago contained stramonium.) The efferent nerve supply to airways structures are largely parasympathetic postganglionic fibers, and adrenergic terminals are absent or sparse on bronchial muscle. Bronchomotor tone is largely constrictor in nature and is the result of parasympathetic neural impulses, there being little if any opposing sympathetic neural drive. The purinergic-peptidergic system, discussed in the introduction to Chapter 11 as the ``third nervous system in the lung,'' is inhibitory to bronchomotor tone and may be the major factor opposing cholinergic bronchospastic activity. Nociceptive stimuli to the lower or upper respiratory tract cause vagal efferent activity and bronchoconstriction. Even bronchoconstriction following allergen inhalation is mediated in part by vagal activity. Stimulation of cut adrenergic nerve endings produces bronchodilation only if there is substantial baseline parasympathetic bronchoconstrictor tone, and this sympathetic effect is thought to be indirect and due to inhibitory activity of adrenergic nerves terminating on the parasympathetic ganglia. Importantly, the bronchoconstrictor effect of beta-adrenergic blockade depends on prior parasympathetic tone; it is absent or mild in normal subjects but is severe in asthmatic subjects, and it is antagonized by atropine. Although there is little evidence for neural sympathetic activity in the lungs except for inhibition of parasympathetic ganglia, humoral adrenergic agents do act directly on adrenergic receptors in the airways.
The quaternary atropine analogues are poorly absorbed but very active locally when inhaled. They are excluded from the CNS. Ipratropium is twice as effective a bronchodilator as atropine, possibly because it is poorly removed from the site of deposition. In contrast to atropine with its very small therapeutic margin, ipratropium and the other two drugs are nearly free of toxicity. Ipratropium is partially effective against the bronchoconstriction produced by histamine but much less so than adrenergic agents. It is, as mentioned, very effective against bronchoconstriction produced by beta-adrenergic blockade and may be the treatment of choice for bronchospasm due to inadvertent administration of beta-blockers to asthmatics or bronchospastic COPD patients.
The usefulness of these agents in bronchospasm due to inhalation of antigen in extrinsic asthma is variable. They reduce somewhat the severity of exercise-induced bronchospasm but are not as effective as adrenergic agents. They may be very effective against psychogenic bronchoconstriction, as demonstrated by prevention of classic Pavlovian conditioned bronchospasm in animals. Peak activity is not reached for at least 60 minutes, much slower than adrenergic inhalants, but activity persists for 4 hours or longer, making it practical to administer these atropine analogues in two or three doses per day. In clinical asthma, anticholinergics are usually less effective than adrenergic agents, but the use of the two classes of drugs together is often additive. When atopic and nonatopic subjects were compared, salbutamol and fenoterol were found to be marginally more potent than oxytropium in intrinsic asthma and significantly more potent in extrinsic asthma. The use of the nonabsorbable atropine agents together with adrenergic aerosols may produce bronchodilatation comparable to larger doses of either but with fewer side effects, better control of patients who have poor response to adrenergic agents, and possibly more prolonged bronchodilatation. (Combined aerosols are available in Britain and probably will be in the United States.)
In contrast to their variable activity in asthma, these agents are as effective as or more effective than adrenergic agents in patients with chronic bronchitis and emphysema, as indicated in the abstracted article. Dryness of the mouth is the only problem. Careful studies have excluded effects on pupil size or urinary retention. Ipratropium has only a minimal effect on mucus viscosity or ciliary beat and does not impair mucociliary clearance of particulate materials. It is much less liable than adrenergic agents to cause hypoxemia by alteration of ventilation-perfusion relationships. These quaternary ammonium analogues of atropine may be the agents of choice in bronchospastic COPD, and the data in the abstracted article suggest that the effects of adrenergic agents will usually not be additive in these patients.--R.M.D.
The Bronchoconstrictor Effect of Inhaled Prostaglandin D{sub-2} in Normal and Asthmatic Men
Hardy, Christopher Charles; Robinson, Clive; Elizabeth Tattersfield, Anne; Holgate, Stephen Townley
N. Engl. J. Med., July 26, 1984, Vol. 311, Pg. 209-213, Ch. 11-1
Affiliation: Southampton Gen. Hosp., Southampton, England
Pre-comment: As discussed in Chapter 10, prognosis in chronic obstructive pulmonary disease (COPD) appears to be related to bronchial hyperreactivity as defined by increased responsiveness to aerosol bronchoconstrictors and bronchodilators; and bronchial hyperreactivity is a characteristic of the asthmatic state. Natural airways responsiveness unaltered by exogenous factors is probably a unimodal variable; those with the most responsiveness (farthest to the right on the bell-shaped curve) are, or tend to become, asthmatics. Whatever a person's natural responsiveness is, it can be greatly increased by a number of factors, including viral infection or exposure to ozone or other airways pollutants. In allergic asthma it is increased by IgE-based immune reactions. In the 1984 YEAR BOOK an article was quoted that demonstrated marked increase in sensitivity to inhaled methacholine during the pollen season in patients with established ragweed sensitivity and some combination of rhinitis and asthma. This past year a similar study was published (Am. Rev. Respir. Dis. 130:56, 1984) in which 10 patients, 3 with hay fever, 10 with asthma and hay fever, and all with established pollen allergy, had carbachol challenge inhalations in March (before pollen) and then repeatedly during the pollen season while alternately receiving corticosteroids or placebo. Bronchial hyperreactivity was not present in March but was present in the pollen season, and could then be reversed with steroid therapy.
Asthmatics with specific allergy are said to react to inhalation of specific allergen with either a single immediate response or with a dual (immediate plus late--6 to 8 hours) response. Persons with a dual response are characteristically hyperreactive to mecholyl inhalation and get bronchospasm from a variety of challenges, both immunologic and nonimmunologic. The state underlying the late response is thought to be responsible for the fact that the more severe cases of asthma are subacute or chronic, lasting for days or weeks rather than hours. Steroids ablate the late but not the early response. It has been proposed that development of the late response is what determines the evolution of seasonal and trivial asthma into perennial and severe asthma. Lichtenstein and associates state that IgG-blocking antibody, developed in response to specific immunotherapy, blocks the late but not the early response to antigen challenge, indicating that the late response and chronic bronchial hyperreactivity in asthmatics may have immunologic specificity.
A great deal of attention is now being paid to the increasing numbers of putative mediators of the allergic reaction, especially their possible role in the induction of chronic bronchial hyperreactivity. There are now many such mediators, including histamine, products of arachidonic acid (prostaglandins, thromboxanes, and leukotrienes), bradykinin, serotonin, substance P, and platelet-activating factor. The folloowing study concerns prostaglandins.--R.M.D.
Abstract: Prostaglandin D$sub{2} is produced by human mast cells in vivo and is released after IgE-dependent challenge of human lung fragments. It is a contractile agonist of guinea pig airways, five times more potent than prostaglandin F$sub{2}$sub{alpha}, and has a preferential effect on peripheral rather than central airways. The present studies evaluated the effects of inhaled prostaglandin D$sub{2}, F$sub{2}$sub{alpha}, and histamine on airways caliber in normal and mildly asthmatic subjects. In normals the two highest concentrations of prostaglandin D$sub{2} produced a small but significant fall in specific airway conductance, whereas there was no significant change with the same concentrations of prostaglandin F$sub{2}$sub{alpha}. In the asthmatic subjects, prostaglandin D$sub{2}, prostaglandin F$sub{2}$sub{alpha}, and histamine all caused dose-related falls in specific airway conductance. Prostaglandin D$sub{2} was 3.5 times more potent than F$sub{2}$sub{alpha} and 10.2 times more potent than histamine. Maximum effects were seen at 3 minutes and were still reduced by 40% 45 minutes after prostaglandin D$sub{2}. The good correlation between the quantitative responses to D$sub{2} and F$sub{2}$sub{alpha} and those of histamine indicates that the increased response of asthmatic airways to the inhaled prostanoids is related to an index of enhanced nonspecific airway reactivity. The results of this study demonstrate that prostaglandin D$sub{2} is the most potent prostaglandin bronchoconstrictor so far studied in humans. Since it is the predominant prostanoid released from activated human lung mast cells, this finding is of considerable interest in the pathogenesis of asthma.
Post-Comment: In an accompanying editorial it is stated that these findings ``unequivocally establish that the hyperreactivity of pulmonary airways in asthma extends to prostaglandin D$sub{2}, the principal endogenous product of the cyclo-oxygenation of arachidonic acid in human lung mast cells." Prostaglandin D$sub{2} is thought to be released in the immediate but not the late response after allergen challenge. Its role in asthma overall is a little questionable in view of the fact that cyclo-oxygenase inhibitors (aspirin, indomethacin, etc.) ordinarily do not make asthma better and in a substantial minority (10%-15%) make it worse. For a long time the most potent bronchoconstrictor known was a substance referred to as slow-reacting substance of anaphylaxis (SRS-A). This is now known to be a mixture of the products of the lipogenase pathway of arachidonate metabolism (leukotrienes B$sub{4}, C$sub{4}, and D$sub{4}). The actions of these substances are very complex and include bronchoconstriction of small airways (20,000 times more potent than histamine), permeability changes, chemotaxis, and stimulation of mucus production. It has been suggested that the mechanism underlying aspirin sensitivity in asthmatics is that cyclo-oxygenase inhibition by aspirin increases the substrate available for leukotriene production via lypoxygenation. Given the complexity of these many mediators and their potential interaction, solution of the asthma problem will undoubtedly occupy investigators for a number of years; however, the field is certainly interesting, and new treatment principles will almost certainly evolve.--R.M.D.
Vasoactive Intestinal Peptide Causes Bronchodilatation and Protects Against Histamine-Induced Bronchoconstriction in Asthmatic Subjects
Morice, A.; Unwin, R. J.; Sever, P. S.
Lancet, Nov. 26, 1983, Vol. 2, Pg. 1225, Ch. 11-2
Affiliation: St. Mary's Hosp. Med. School, London
Abstract: Vasoactive intestinal peptide (VIP) is widely distributed in autonomic nerve fibers and relaxes smooth muscle both in vitro and in vivo. In order to demonstrate a possible relationship to bronchospasm in asthma, 7 mildly asthmatic but otherwise normal volunteers had VIP administered by vein in constant infusion (6 pmole/kg/minute for 15 minutes). During infusion there was significant improvement in mean FEV$sub{1} values determined 5, 10, and 15 minutes after the infusion was begun. Further, the bronchoconstrictor effect of nebulized histamine was considerably inhibited during infusion of VIP. This was highly significant at all times.
A nonadrenergic inhibitory nervous system has been demonstrated in vitro in both animal and human airways. During electrical field stimulation of isolated respiratory smooth muscle, VIP, a smooth muscle relaxant neuropeptide with neurotransmitter properties, is released. The amount released correlates with the degree of relaxation, and relaxation is inhibited by preincubation with antibodies to VIP. Similar, more complex experiments involving human lung tissue give similar results. The relaxing effect of VIP extends from midtrachea to the distal bronchi. It is not modified by adrenergic-blocking agents. Vasoactive intestinal peptide inhibits not only smooth muscle contraction but also the secretion of macromolecules and lysozyme into the bronchial lumen. It has been shown to inhibit not only histamine-induced but also prostaglandin F$sub{2}$sub{alpha}-induced bronchoconstriction. It has been suggested that a deficiency of VIP or impaired binding to its receptors could be the basis of bronchial hyperreactivity in asthma.
Post-Comment: The nonadrenergic noncholinergic (NANC) nervous system, its possible role in asthma, the probability that VIP is the mediator thereof, and the general area of autonomic regulation of the airways were discussed in two good reviews this past year (Thorax 39:561, 1984; and Ann. Rev. Med. 35:451, 1984). Very briefly, it appears that there are virtually no adrenergic nerve endings on bronchial smooth muscle. Postganglionic adrenergic nerves end on cholinergic ganglia, and the results of their stimulation are probably mediated by inhibition of cholinergic neurotransmission at the ganglionic level. The vagus contains at least three functionally different components: afferents from sensory endings in the respiratory mucosa which may release a polypeptide, substance P, which is locally bronchoconstrictive; parasympathetic efferents which result in release of acetylcholine and stimulation of muscarinic receptors on the airways muscles, leading to bronchoconstriction; and, it is now believed, efferents of this third autonomic neural system, the nerve endings of which release the neurotransmitter VIP, which causes relaxation of airways musculature. In addition to being a bronchodilator, VIP inhibits antigen-induced release of histamine from lung fragments. Like beta-adrenoceptors, specific stimulation of VIP receptors activates adenylate cyclase, resulting in increased intracellular cyclic AMP content. It has been postulated that the state of bronchial hyperreactivity may be due to decreased activity or tone of this bronchodilator nervous system with resulting lack of opposition to bronchoconstrictor cholinergic tone.
Although there are no adrenergic nerve endings on the respiratory muscles, there are many adrenergic receptors, principally beta-adrenoceptors, which respond to circulating catecholamines. Hormonal or injected catecholamines also decrease histamine release from mast cells and decrease microvascular leakage produced by histamine.
Figure 11-1, which illustrates some of these relationships, was adapted from the editorial by Barnes mentioned above.--R.M.D.
Effect of Outpatient Treatment of Asthma With {beta} Agonists on the Response to Sympathomimetics in an Emergency Room
Rossing, Thomas H.; Fanta, Christopher H.; McFadden, E. R., Jr.
Am. J. Med., November 1983, Vol. 75, Pg. 781-784, Ch. 11-3
Affiliation: Brigham and Women's Hosp., Boston
Abstract: It has been suggested that tolerance to the bronchodilating effects of sympathomimetic drugs may develop in asthmatic patients after long-term use of these agents. In an emergency room setting, the effects of inhaled and injected sympathomimetic drugs in 58 patients who had pretreated themselves with beta agonists were compared with results observed in 38 patients who had not used such drugs. The two groups had similar degrees of obstruction on presentation and were well matched with respect to clinical features of their illness.
In the emergency room, both groups experienced significant improvement in lung function after 1 hour of treatment. No significant differences were found between those who had used adrenergic agonists as outpatients and those who had not (Fig 11-2). The lack of tolerance to sympathomimetic drugs was independent of type of catecholamine administered in the emergency room. The immediate responses to both injected epinephrine and inhaled isoproterenol were not influenced by prior use of sympathomimetic agents. Treatment with repetitive doses of either drug was well tolerated. In none of the patients was it necessary to discontinue or modify therapy because of palpitations, tachycardia, hypertension, or other signs or symptoms of drug toxicity.
These results demonstrate that failure of self-administered sympathomimetics to terminate acute exacerbations of asthma does not predict emergency room response to this class of drugs. In the present series, prior use of beta agonists by any route of administration did not adversely influence the bronchodilator activity of, or enhance adverse effects from, epinephrine or isoproterenol. These findings do not support the concept of acutely acquired refractoriness to sympathomimetic drug therapy.
Post-Comment: This is one of a series of excellent articles by the Respiratory Division and the Medical House Staff at the Brigham and Women's Hospital about treatment of asthma in the emergency room. The first said that isoproterenol, 2.5 mg administered in 2 ml of saline by wet inhalation, or subcutaneous epinephrine, 0.3 ml (0.3 mg) of a 1:1,000 solution, each administered every 20 minutes for three doses, were equally good; aminophylline was reported to add very little. The second article said that aminophylline did add a little to the rapidity of recovery. The third article indicated that it did not appear to make any difference whether the aminophylline levels were therapeutic or subtherapeutic, and the data indicated that it contributed nothing. The editorial discussion that followed referred to some other articles which indicated that the contribution of aminophylline in this setting is minimal. All these taken together indicate that aminophylline may add a little but certainly not much to emergency room therapy and that inhaled isoproterenol (by wet nebulization) is only very slightly superior if at all to subcutaneous epinephrine in the doses mentioned above.
This most recent article from the Brigham group provides another important guide to emergency room asthma therapy. The authors attempted to determine whether decreased responsiveness to adrenergic aerosols could be demonstrated in emergency room asthma patients who had recently used these drugs as compared to those who had not, and the answer was no.
The Brigham group believe that decisions on admission are best based on response to therapy and that they cannot be reliably predicted on the basis of findings present before institution of therapy, the contention of a widely quoted article by Fischl and associates. The latter group attempted to develop a numerical index to predict whether or not admission would eventually be required in the emergency room asthmatic. They gave one point each to a pulse rate higher than 120, respiratory rate higher than 30, pulsus paradoxus greater than 18 mm Hg, peak expiratory flow rate (PEFR) less than 120 L/min, the presence of dyspnea, use of accessory muscles, and prominent wheezing on physical examination. In their experience, patients with a numerical index of over 4 required admission. (Some other material concerning prognosis was discussed in the 1983 YEAR BOOK, including the observation that patients who are diaphoretic or who don't want to lie flat do poorly and that blood gases per se do not reliably predict the need for eventual hospitalization.) An article this year reported that the Fischl index has no real predictive value beyond the fact that sicker people are sicker. All of the factors mentioned by Fischl and others are helpful in the assessment of patients, but the attempt to compress these clinical data into a quantitative ``admission index'' probably adds little.
The related question of how to tell which emergency room asthmatic patient should get steroids and how much has also been addressed by the Brigham group. They showed that intravenous steroids (2-mg bolus of hydrocortisone per kilogram followed by 0.5 mg/kg per hour intravenously) greatly improved outcome. Improvement due to steroid therapy was apparent at 6 hours and continued to increase for 24 hours. An article quoted in the discussion that followed indicated that 40 mg and 125 mg of methylprednisolone given intravenously over 30 minutes every 6 hours were equally good, but a 15-mg dose was clearly inferior. The answer to ``how much'' is ``quite a bit,'' somewhere between 1 and 2 gm of hydrocortisone or its equivalent over 24 hours. So asthmatics with severe obstruction (PEFR, 120 L/min or less; FEV$sub{1}, 1.2 L or less) who do not get better promptly (within a couple of hours) with simple catecholamine therapy (by injection or inhalation) should be given steroids; those who are not clearly well enough to go home in 18-24 hours should be admitted. Where facilities for observation are limited, failure to improve rapidly on catecholamine therapy probably constitutes sufficient reason for admission. Another point also mentioned in the 1984 YEAR BOOK is that most patients who come to the emergency room with asthma will get better quicker and come back less frequently if they are sent out with steroids to be taken orally in decreasing doses for a week.
English authors have emphasized the desirability of determining the pattern of asthma by frequent use of the PEFR using a patient-owned inexpensive PEFR meter, an approach that has been little utilized in the United States. Hetzel and associates have shown that diurnal variation in PEFR is a normal phenomenon, that this diurnal cycle is greatly accentuated in many asthmatics, and that most asthma deaths occur in the early morning hours at the PEFR nadir. When long-acting theophylline preparations became available, there was hope that they might provide early morning protection. According to Hetzel they do not, and others have found that ``morning dipping'' asthma is better treated by regular and intensive adrenergic aerosol therapy together with inhaled steroids. As mentioned above, the value of fairly intensive adrenergic aerosol therapy is not only dilation of the bronchial musculature, but also inhibition of histamine release from mast cells and decrease in bronchial hyperreactivity as assessed by aerosol challenge with histamine or mecholyl. Significant reduction in ``morning dipping'' has now been observed with ipratropium bromide (four puffs at 40 mu g per puff) on retiring. The role of calcium channel blockers in asthma is not yet well defined and to date seems to be marginal, but there is some indication that these might benefit very refractory cases.
Some data indicating that immunotherapy was probably not helpful in asthma were mentioned in last year's YEAR BOOK, and special attention was given to a review by Lichtenstein which seemed to support that point of view. Accordingly, it is fair to mention another review by Lichtenstein on this subject which seems to say that there is an emerging base of scientific data to support immunotherapy now, although the number of good antigens is, to date, quite limited.--R.M.D.
Airway Response to Inhaled Salbutamol in Hyperthyroid and Hypothyroid Patients Before and After Treatment
Harrison, R. N.; Tattersfield, A. E.
Thorax, January 1984, Vol. 39, Pg. 34-39, Ch. 11-4
Affiliation: Univ. of Southampton, England
Abstract: Thyroid disease and asthma are related, but the mechanism by which changes in thyroid function influence asthma is unclear. The effects of thyroid dysfunction on airway responses to salbutamol, an inhaled beta-agonist, were examined in 18 patients with thyrotoxicosis and in 6 with hypothyroidism. Eight of the hyperthyroid patients participated in a time course study, inhaling a single dose of 400 mu g of salbutamol; the other 10 were reassessed after treatment. The hypothyroid patients were reevaluated after 4-11 months of treatment with thyroxine when they were clinically and biochemically euthyroid. Dose-response studies were done with 10-410 mu g of salbutamol.
Specific airway conductance (sGaw) increased after salbutamol inhalation, the maximum elevation occurring at 1 hour. There were no significant changes in FEV$sub{1} or baseline specific airway conductance (sGaw) after treatment in either patient group, but the mean forced vital capacity increased in the treated thyrotoxic patients. Responses of sGaw to salbutamol treatment were more marked after treatment of thyrotoxicosis. In the hypothyroid group, administration of salbutamol lowered the absolute sGaw after treatment of thyrotoxicosis.
These findings suggest an inverse relationship between adrenergic airway responsiveness and the level of thyroid function. Changes distal to the beta-adrenoceptor may be involved. The mechanism is not clear, but the findings are supported by in vitro observations and probably account for deterioration of the asthmatic condition that occurs in patients who become thyrotoxic.
Post-Comment: The literature about the association of hyperthyroidism and asthma goes back at least to 1944 and possibly to 1929 but the article that is said to mark the beginning of serious clinical interest was by Settipane et al. in 1972. They reported 5 patients admitted to the hospital for status asthmaticus who were found to have hyperthyroidism and whose asthma substantially improved or disappeared entirely when PTU was given, often within a matter of days. They also stated that hyperthyroid patients with asthma often demonstrated an untoward reaction to aminophylline and epinephrine, with hyperexcitability and sometimes worsening of the asthma. There was little documentation other than the quite convincing clinical case reports.
This year there have been several articles discussing this association. Omar and associates report an asthmatic lady who got worse with the onset of thyrotoxicosis and better with treatment. Again there were no data beyond the case report itself. Fitzpatrick and associates reported a patient with severe thyrotoxicosis treated with beta-blockers who developed severe asthma. If this relationship is real, and it probably is, then those who like to use propranolol in the treatment of thyrotoxicosis need to know that it might act together with the thyrotoxicosis to bring out a serious asthmatic state. Also, if epinephrine and aminophylline may act funny in thyrotoxic patients, as Settipane says, then the treatment for propranolol-enhanced bronchospasm in asthma may well be ipratroprium (see section on ipratroprium in Chapter 10). Many mechanisms have been suggested, including increased catabolism of cortisol or of prostaglandins, decrease in cyclic AMP concentrations, and alteration in the number or function of adrenergic receptors, but none has been established.
The article abstracted above was chosen because it is more than a case report. The association of bronchospasm with hyperthyroidism and bronchodilation with hypothyroidism was confirmed by response to treatment in each of the cases.--R.M.D.
Cyclical Variation of the Heart Rate in Sleep Apnea Syndrome: Mechanisms and Usefulness of 24-Hour Electrocardiography as a Screening Technique
Guilleminault, Christian; Connolly, Stuart; Winkle, Roger; Melvin, Kenneth; Tilkian, Ara
Pre-comment: According to a graceful review by Neil Cherniack, sleeping and breathing are in many ways antagonistic. Recumbency increases airways resistance, favors hypoxemia by basilar airways closure as a result of increased functional residual capacity, tends to produce posterior displacement of the tongue, changes the alignment of the respiratory muscles in the upper airway so as to compromise patency, and accentuates the respiratory handicaps of obesity. Sleep itself decreases the sensitivity and output of neural respiratory control mechanisms and decreases general sensory stimulation. On the other hand, the swings in blood gas tensions that occur with apnea and the increased efforts to breathe that occur in obstructive apneas produce arousal and terminate sleep. ``While these opposing tendencies are normally reconciled, sleep apnea patients seem unable to arrive at an adequate compromise.'' Sleep apnea is best regarded as one end of a continuum rather than a separate abnormal state. Abnormal sleep apneic events result from the interaction of two sets of factors: those that favor collapse or obstruction of the upper airways and those that compromise or decrease the neural output of the respiratory center so as to quantitatively exaggerate the effects of sleep on respiratory control. Included in the first is increased upstream resistance to air flow in the nose, which increases the negative pressure to which the pharynx is subjected in inspiration. Compromise of respiratory center output is probably a necessary condition for both obstructive and nonobstructive apneas. If the decrease in respiratory drive silences the upper airways (patency-maintaining) muscles before silencing the diaphragm, obstructive (peripheral) apneas occur. If the diaphragm is silenced first or simultaneously, nonobstructive (central) apneas will be observed, often followed, however, by upper airways obstruction due to collapse of upper airways muscles when breathing is resumed. Both the anatomic and the neural predispositions are mostly observed in obese older males. Alcohol and sedative drugs further depress the respiratory center. Anatomic factors such as tonsillar hypertrophy, micrognathia, macroglossia, and, as mentioned, nasal obstruction contribute to the tendency to airways closure on inspiration.
It has been estimated that 1% of Israeli workers demonstrate sleep apnea to a degree that would be called pathologic. Obviously it does not make sense to hold that such a large number of men should be studied with polysomnographic studies. The article abstracted below says that significant sleep apnea can be detected by the much easier and less expensive procedure of overnight Holter monitoring. The article comes from the premier sleep laboratory in the world.--R.M.D.
Abstract: Various arrhythmias can occur in sleep apnea syndrome, and cyclical variation of heart rate (CVHR), a pattern of sinus arrhythmia-bradycardia followed by abrupt tachycardia, has been observed in a small number of patients. The authors performed 24-hour ECG recording with simultaneous polygraphic recording at night in 400 patients with sleep apnea syndrome and an intact autonomic nervous system. Computer analysis was used to obtain plots of the R-R interval between QRS complexes during a 24-hour period. There were 384 men and 16 women in the study. The men had a median age of 47.5 years. Daytime somnolence was the most common presenting feature. Forty-six nonobese patients had a diagnosis of secondary sleep apnea syndrome. The control group included 50 patients with psychophysiologic insomnia, narcolepsy, or restless-leg syndrome, and 100 consecutive patients referred for ECG recording because of non-sleep-related indications.
The CVHR pattern was observed during nocturnal sleep in all 400 study patients. It disappeared during waking periods. The pattern was not observed in the 50 control patients seen for sleep-related problems, but it was present in 8 of the 100 patients assessed for cardiac reasons. Five of the 8 patients had polygraphic studies, and all 5 exhibited intermittent sleep apnea. None of the 50 study patients who were reexamined after surgery, with the tracheostomy open during sleep, had the CVHR pattern. The pattern was not observed in 3 study patients with Shy-Drager syndrome, 3 with heart transplants, or 7 with autonomic neuropathy due to diabetes or uremia. Atropine blocked the CVHR pattern in study patients by eliminating the bradycardia. Administration of pure oxygen led only to moderate blunting of the variation in heart rate.
It is concluded that the ECG abnormalities that characterize the CVHR pattern in sleep apnea patients are mediated by the autonomic nervous system, and that hypoxia is not the only factor involved. With 24-hour ECG recording, one can expect to identify patients with sleep apnea other than those with mild, intermittent abnormality or serious abnormalities of the autonomic nervous system.
Post-Comment: Abnormal sleep requires treatment only when it results in cor pulmonale, unacceptable daytime hypersomnolence (the sleep apnea hypersomnolence syndrome [SAHS]), or dangerous arrhythmias occurring at the time of serious hypoxic swings. (It has been suggested that systemic hypertension may also be a result of sleep apneic events, but whether or not it makes sense to study all fat snoring hypertensives is not clear.) Of these pathologic consequences, only the arrhythmias might not be detected by simple clinical evaluation. The use of nighttime Holter monitoring as a screening device, as demonstrated in the article abstrac abstracted above, is not only far easier and cheaper than formal polysomnography but also has the added attraction of detecting the life-threatening arrhythmias if they are present during sleep. It should be emphasized that although these arrhythmias have been well documented, whether or not they should be sought and treated remains a clinical opinion without guidance from controlled studies.
A recent paper from the Baltimore Regional Sleep Disorders Center says that clinical evaluation is not enough, and polysomnography is necessary. (Never ask your barber if you need a haircut.) Their data indicated that simple clinical observation for several minutes during sleep detected 20 of 31 cases, and 7 more should have been detected according to the events that were recorded while the doctor was watching. The other 4 had no diagnostic findings while the doctor was watching, although these did occur at other times and would have been detected by a longer period of observation. As the authors point out, whether or not formal sleep studies are needed before undertaking therapy depends a lot on how invasive or risky or expensive the contemplated therapy is. Simple weight loss in obese SAHS patients will be curative in many, and even a little bit goes a long way since the relationship between weight and sleep apnea index is logarithmic. There are a lot of drugs that can be tried, including progesterone, acetazolamide, protriptyline, aminophylline, and almitrine, although the results of these are either equivocal (progesterone) or not yet documented. As discussed in the article by Anthonisen on oxygen therapy abstracted in Chapter 10, there are some SAHS patients who improve greatly with oxygen therapy, apparently rehabilitating their respiratory centers by eliminating serious hypoxic insults and causing a reintegration of sleep architecture. According to Bradley and Phillipson, uvulopalatopharyngoplasty, a surgical procedure designed to remove redundant soft tissue in the oropharynx, has produced variable results and cannot be recommended for routine use. The idea that reduction of nasal resistance to inspiratory flow might be curative is very attractive, especially if some cases are found in which simple decongestants at night do the trick. Certainly if correction of a deviated nasal septum might be curative, it would be an attractive alternative to tracheostomy. It turns out that most hypothyroid patients will have sleep apnea when studied. Rajagopal et al. recorded sleep patterns diagnostic of SAHS in 9 of 11 consecutive hypothyroid patients. Six of this group were obese, but all demonstrated blunted respiratory center responses, and these became normal after thyroid replacement therapy. Certainly, formal sleep studies are indicated before tracheostomy is undertaken. Whether or not they are indicated before simpler procedures in a clinical (nonresearch) situation is at present a matter of opinion. Bradley and Phillipson believe that the most exciting new approach to treatment is the use of positive airways pressure, although, as they point out, the mechanism by which this works is at present a matter of dispute.
A lot of information about SAHS was covered in last year's YEAR BOOK: increased upper airways resistance in SAHS; demonstration of narrowing of the upper airways by computed tomography scan in SAHS patients; the use of flow-volume curves in diagnosis; protriptyline and diamox treatment; micrognathia, hypothyroidism, and sickle cell disease as causal factors. More factors were discussed in the 1983 YEAR BOOK; improvement of sleep architecture over time after tracheostomy; demonstration of the central component in obstructive sleep apnea; the role of Cheyne-Stokes breathing; more on hypothyroidism; more on flow-volume curves in diagnosis; stroke and SAHS; protriptyline therapy; oxygen therapy of SAHS; failure of hypoxemia as an arousal stimulus; interaction of sleep fragmentation and arousal threshhold.--R.M.D.
Intermittent Positive Pressure Breathing Trial Group
Ann. Intern. Med., November 1983, Vol. 99, Pg. 612-620, Ch. 14-3
Abstract: Previous studies of long-term intermittent positive pressure breathing (IPPB) in chronic obstructive lung disease have failed to show significant benefit, but all of these studies have serious shortcomings. The present multicenter trial of IPPB therapy compared it with compressor nebulizer therapy using the same aerosols in ambulatory patients whose condition was stable with standard management exclusive of IPPB. The 985 patients who entered the trial during a 34-month period were given metaproterenol, salbutamol, or isoetharine. Nebulization was completed in about 10 minutes at a tidal volume of 15 ml/kg, or 75% of inspiratory capacity if this was less than 15 ml/kg. The same type of machine was used for both forms of treatment.
The two treatment groups were clinically comparable at baseline. Dropout rates were about 13% or 14% in both groups. The average follow-up was 32 months. Compliance with treatment was variable in both groups. Mortality was comparable in the two groups, as were the numbers of hospitalizations, changes in pulmonary function, and changes in quality of life. All physical measures indicated deterioration during follow-up. Type of treatment did not influence mortality, hospitalizations, or the rate of change of FEV$sub{1} in patients with relatively mild or severe airway obstruction, great or little reversibility, or more or less evidence of emphysema.
Thus, IPPB treatment was not of greater benefit than compressor nebulizer therapy in this multicenter study of patients with chronic obstructive lung disease. The patients' conditions covered most of the spectrum of disease, excluding those who were most hypoxemic.
Post-Comment: One other area in which IPPB has been widely used is in postoperative therapy of patients with upper abdominal operations, in which group there is a significant incidence of respiratory complications due to incisional pain, with consequent depression of vital capacity and functional residual capacity associated with hypoxemia. Dr. Anthonisen's group from Winnipeg studied 30 such patients, 15 with IPPB (every 4 hours for 4 days) and 15 who had no IPPB but were otherwise treated the same. Both groups had postoperative physiotherapy. Whatever little difference there was favored the non-IPPB group. An older article says that IPPB is of no use in patients with pneumonia. It is likely that there is no place at all for this expensive form of therapy. It will probably disappear when it is no longer a profit center, not before.
There have been a lot of interesting articles on Swan-Ganz catheterization this past year. One major prospective study of 528 consecutive pulmonary artery catheterizations recorded complications in 126, which were judged to be serious in only 23 (4.4%), and no deaths. There were no catheter-associated deaths during the study period, but it had been the occurrence of four deaths due to pulmonary artery rupture that had stimulated the study, and two more were observed after it was over. In their overall experience the authors observed six episodes of massive hemoptysis and five deaths in about 5,000 procedures, an incidence of one pulmonary artery injury in each 800 catheterizations (0.125%). They stated that ``pulmonary artery injury in a heparinized patient is so likely to result in fatality . . . that we presently do not measure wedge pressures in patients on cardiopulmonary bypass, but rather use the pulmonary artery diastolic (PAD) pressure.'' The great majority of these procedures (390 catheterizations in the study group of 528) were in open heart surgical cases, a generally more favorable group than medical ICU patients, and 73% had their catheters in place less than 24 hours.
In an editorial commenting on this study, Santos points out that although most agree that diagnosis (between cardiac and noncardiac pulmonary edema) is improved by Swan-Ganz catheterization, the question as to whether the benefit is worth the risk remains unanswered. ``To date there are no prospective controlled series which demonstrate improvement in morbidity or mortality as a direct result of data obtained from thermodilution pulmonary artery catheters.'' The author quotes a figure of 15.8% complications and 3.7% serious complications in 6,000 procedures sufficiently analyzed to provide reliable data. The specific problem of pulmonary artery rupture was discussed by Hannan and associates. They were able to find 28 reported cases, of which 13 were fatal. They estimated the incidence of rupture to be about 1:500 or 0.2%. This complication occurs more in elderly people, more in women than men (2.7:1), and is frequently associated with pulmonary hypertension. Incidence and mor tality are both higher in patients who receive anticoagulation therapy. Hemoptysis was the major finding in 23, hemothorax was seen in 7, and 4 had both hemoptysis and hemothorax. The authors recommended that the catheter should be kept in place and immediate embolization performed by autologous clot when hemorrhage occurs.
Fine and associates asked again whether or not pulmonary artery catheterization was necessary for diagnosis and said that it was. They observed complications in 25% of their 70 patients and three catheter-related deaths (4%), two with bradyarrhythmia and electrical-mechanical dissociation and one with pulmonary artery rupture. Sepsis occurred in 3 patients (4%), all of whom had the catheter in place over 48 hours. The above were all pretty much clinical studies. An autopsy study from Yale has reported that the incidence of right-sided bacterial endocarditis in 55 autopsied patients who died with a Swan-Ganz catheter in place was 7%. In addition, 29 (53%) had one or more right-sided endocardial lesions, 12 (22%) subendocardial hemorrhage, 11 (20%) sterile thrombus, and 2 (4%) hemorrhage and thrombus. The authors concluded that right-sided endocarditis should be suspected in septic patients with catheter in place. Occurrence of the superior vena cava syndrome complicating Swan-Ganz catheterization in 6 cases, with fatal outcome in 2, was reported by Gore and associates. They also emphasized the frequency of pulmonary embolism when catheter-related central venous thrombosis had occurred, a frequency greater than 50% in the series that they studied.
The technical aspects of Swan-Ganz catheterization have been reviewed in an excellent article from Yale (Chest 85:537, 1984).--R.M.D.
Sustained Improvement in Primary Pulmonary Hypertension During Six Years of Treatment With Sublingual Isoproterenol
Pietro, Daniel A.; LaBresh, Kenneth A.; Shulman, Robert M.; D.Folland, Edward; Parisi, Alfred F.; Sasahara, Arthur A.
N. Engl. J. Med., Apr. 19, 1984, Vol. 310, Pg. 1032-1034, Ch. 15-5
Abstract: Primary pulmonary hypertension is a disorder of unknown cause that results in severe disability and death 3-7 years after the onset of symptoms. Symptomatic improvement has been gained in patients having the pulmonary artery pressure lowered, but the disease usually progresses. The present patient had sustained improvement in pulmonary hemodynamics and exercise capacity during 6 years of sublingual isoproterenol therapy.
Man, 36, a long-time smoker, presented with progressive exercise-induced dyspnea and episodic syncope, followed by peripheral edema, abdominal distention, and weight gain of 6 kg. Cardiac catheterization showed pulmonary hypertension with a mean pulmonary artery pressure of 45 mm Hg and normal coronary arteries. Moderate peripheral cyanosis was present 10 months after presentation, and an ECG showed right ventricular hypertrophy with strain; the mean pulmonary artery pressure was 47 mm Hg with a wedge pressure of 12 mm Hg. The cardiac output was 4.8 L/minute. Pulmonary angiography showed no evidence of pulmonary embolism.
Intravenous isoproterenol was found to produce a marked reduction in pulmonary artery pressure (Fig 15-2), and the patient was therefore placed on 10 mg sublingually every 2 hours. Dyspnea and fatigue were markedly relieved, permitting the patient to return to work. The long-term effects of treatment on rest and exercise hemodynamics are given in the table. Treatment at 90-minute intervals was effective when symptoms recurred after several months. The patient was able to paint the exterior of his house and work full-time as a factory machinist.
If the early stage of primary pulmonary hypertension is secondary to pulmonary arterial constriction, reversal of the vasoconstriction might interrupt the course of the disorder and provide lasting benefit. The present patient has continued to respond to isoproterenol with markedly improved exercise capacity. Testing of individual patients with different drugs is necessary. All patients with primary pulmonary hypertension should be screened for potential response to vasodilator therapy because of the poor prognosis without treatment.
Post-Comment: Primary pulmonary hypertension and its therapy have been discussed in the last several YEAR BOOKS OF MEDICINE without much clarification. Its usually ineluctable course is illustrated by the fact that reports of single favorable responses to therapy continue to be published in good journals, the abstracted article being an example. An impressive experience with nitroglycerin was reported in an article abstracted in the 1984 YEAR BOOK; bad results with the initially promising hydralazine were discussed the year before; and prostacyclin, discussed in 1981, has been again reported as effective this past year. Nifedipine seems to be a promising agent, and some experts start with it now. There is clearly a phase in which one or another vasoactive agent will be beneficial, as described in the abstracted article, followed by a phase in which the pulmonary vascular resistance is fixed. In the latter circumstance a low cardiac output state develops in which vasodilator therapy may lower systemic resistance while not altering pulmonary vessels, with resulting hypotension and sometimes death. One investigation concerning prognostic factors in pulmonary hypertension indicates that a low stroke-volume index and an elevated right atrial pressure were the best independent prognostic indicators. In any event, treatment is risky, and the disease is clearly best investigated by those with substantial experience in it. One expert's approach is to study the effects of prostacyclin (epoprostanol) at formal right heart catheterization, since its effect is transient, and an untoward result can be promptly reversed simply by stopping the infusion. He subsequently assesses the effects of other agents which can be given by other routes in the intensive care unit with a pulmonary artery catheter in place.
Primary pulmonary hypertension is usually considered a disease of young women with symptoms of fatigue, dyspnea on exertion, syncope (particularly exercise-induced), and chest pains. One article this year describes its occurrence in 2 patients older than 60. Whether or not this is the same disorder that occurs in the younger patients is not clear.
Pulmonary hypertension complicating collagen vascular diseases is an important and sometimes difficult differential diagnosis, as these conditions can present with very few clinical manifestations other than the pulmonary hypertension. These collagen vascular cases can be divided into two subsets: those in which the pulmonary vascular resistance is due in part to interstitial inflammation or fibrosis, and those in which the process is primarily vascular. Most physicians would probably treat the former with corticosteroids with or without immunosuppressive agents. In the latter situation, in which the pathologic changes are largely confined to the vessels, the outlook for steroid-immunosuppressive treatment is less good. Most believe that pulmonary hypertension complicating scleroderma is not treatable, and whether or not lupus or rheumatoid arthritis cases respond is not really well known. However, a recent, very good article on pulmonary involvement in mixed connective tissue disease (a syndrome with overlapping clinical features of systemic sclerosis, lupus, rheumatoid arthritis, and polymyositis, together with a high titer of antibody to ribonucleoprotein) reported improvement in a majority of their cases treated with steroids or steroids plus cytoxan. Case 2 in that article, a case reported in some detail, had definite proliferative vascular involvement and little interstitial involvement on lung biopsy, and the patient clearly had a very good response to steroid cyclophosphamide therapy.
The interesting association of pulmonary hypertension and chronic liver disease has been mentioned several times in these pages, and it has been implied that this might be due to circulating immune complexes, as is thought to be the case in collagen vascular diseases. However, an interesting article this past year reports a case of pulmonary hypertension associated with extrahepatic portal hypertension and a normal liver. The implication is that portal-systemic collateral circulation may be the major pathogenic factor in this association, presumably by allowing vasoactive substances ordinarily removed or altered by the liver to reach the pulmonary vascular bed.--R.M.D.
Results of Resection in Non-Oat-Cell Carcinoma of the Lung With Mediastinal Lymph Node Metastases
Martini, Nael; Flehinger, Betty J.; Zaman, Muhammad B.; J. Beattie, Edward, Jr.
Ann. Surg., September 1983, Vol. 198, Pg. 386-397, Ch. 16-4
Affiliation: New York
Pre-comment: Discussion of the surgical therapy of lung cancer requires using the now conventional TNM surgical staging system. This was briefly presented in each of the last two YEAR BOOKS OF MEDICINE and will be printed here again (Table 1). This system was based on a retrospective analysis of the clinical features of tumors demonstrating fairly similar prognosis. The 5-year postsurgical survival of patients with stage I tumors was 59%-70%; with stage II, 20%-35%; and with stage III, 1%-19%. Over the past several years, however, it has become clear that the three stages are too broad, including, in stage I and stage III, anatomic subsets with very different prognoses. In last year's YEAR BOOK an article was abstracted which expressed the opinion that the stage I subset with hilar nodes (T1 N1 M0) had a surgical prognosis much closer to stage II cancer (T2 N1 M0) than to stage I disease without hilar nodes. It is of interest to note that regarding hilar nodes as one thing and mediastinal nodes as another reflected the surgical opinion that the former could be encompassed in a pneumonectomy and the latter could not. It is also of interest to note that, as mentioned in the published discussion of the article abstracted below, the first pneumonectomy for bronchogenic carcinoma done by Evarts Graham, which produced a 38-year survival, had pathologic evidence of mediastinal node involvement when the specimen was obtained and reexamined years later.
An article this past year by Lillimoe et al. confirms the notion that N1 tumors are better grouped in stage II and also demonstrates the effect of tumor size (T1 vs. T2) on survival (Table 2). The importance of tumor size as an independent variable was also emphasized by Clee et al. from Scotland. In their study of independent predictors of survival in resected cases, a tumor size of more than 5 cm was associated with an 18% 5-year survival, whereas tumors less than 5 cm demonstrated a 44% 5-year survival. The other independent variables influencing survival in that study were pain (21% survival with, 37% without), weight loss (17% survival with, 39% without), and cell type (37% survival with squamous cell, 25% with all others).
Stage III also contains anatomic subsets with highly different prognoses. Patients whose cancers are Stage III by virtue of distant metastases (M1) are probably all dead by 5 years. The much better prognosis of those that are Stage III by virtue of having chest wall involvement, mentioned in last year's YEAR BOOK, was again emphasized this past year. Trastek et al. did not believe that involvement of the phrenic nerve, the esophagus, the diaphragm, the left atrium, or the superior vena cava necessarily contraindicated the attempt at operative cure, an attitude that is generally shared by the English thoracic surgical fraternity.
Most controversy in this area, however, concerns tumors that are stage III by virtue of having mediastinal node involvement (N2). This has obvious implications not only for treatment principles but also for the vigor with which staging procedures such as mediastinoscopy and perhaps CT scanning are to be carried out. The early justification for the policy of mediastinoscopy in all cases, advocated by many, was that a substantial percentage of cases with normal mediastinal contours on conventional chest radiograph would be shown to have mediastinal nodal deposits on mediastinoscopy, and would thereby be spared ``unnecessary thoractomy.'' As discussed extensively in the last two YEAR BOOKS OF MEDICINE, there is now widespread agreement that at least some cancers with mediastinal node involvement have a reasonable surgical prognosis. However, substantial lack of consensus remains in very important details, such as whether or not mediastinoscopy should be used. (Pearson of Canada says always; Kirsh of Michigan says only when there is an abnormal mediastinum or in order to document a cell type other than squamous cell in which latter circumstances he does not operate; Martini of Memorial says rarely; and Abbey-Smith of England says only in persons over 65 when the findings might make pneumonectomy necessary.) There is also a lack of consensus on whether or not postoperative irradiation should be used (Pearson, yes; Kirsh, yes; Martini, yes; Abbey-Smith, no) and on whether or not the prognosis of adenocarcinoma with mediastinal nodal implants is good enough to warrant operation (Kirsh, no; Pearson, probably not; Martini, yes indeed). Another incompletely explored area is the importance of the location of the mediastinal node. Some authorities, Kirsh and Abbey-Smith for instance, believe that high mediastinal nodes contraindicate resection, but other surgeons do not.
In an effort to make it possible to gather more precise information concerning the prognostic implication of node location, a much more detailed node mapping plan has been proposed. The following article from the Memorial Sloan-Kettering group (Martini) represents the most extensive recent experience utilizing this system.--R.M.D.
Abstract: From 40% to 45% of patients with non-oat-cell lung cancer have N2 disease, with mediastinal node involvement but no known spread outside the chest. They generally have been considered incurable. The authors reviewed the data on 706 patients with N2 disease, among a total of 1,598 patients with non-oat-cell lung cancer seen in 1974-1981. The distribution of nodes of interest in these cases is shown in Figure 16-4. Complete resection of the tumor and ipsilateral mediastinal nodes was possible in 151 (21%) N2 cases. Ninety-four patients had adenocarcinoma, 46 had epidermoid cancer, and 11 had large cell cancer. Lobectomy was performed in 119 cases, pneumonectomy in 26, and wedge resection or segmentectomy in 6. The involved nodes are shown in Figure 16-5, and the location of nodes in the 80 patients who had a single level of N2 involvement is shown in Figure 16-6. Thirty-five patients had direct extension of disease outside the lung, while 41 had lesions 3 cm or less in diameter that were confined to the primary site.
There were 2 postoperative deaths. External radiotherapy was given postoperatively to 13 patients, and 1 patient had preoperative irradiation. Survival is related to clinical stage in Figure 16-7. Survival rates at 3 and 5 years were 43% and 29%, respectively. Survival was related to tumor size, and was better in patients with node involvement at one level only. Patients with N2 nodes in the inferior mediastinum did less well than those with normal nodes at this site. Forty-nine patients had no recurrence after initial treatment. The brain was the most common site of distant metastasis. Thirty-seven patients are alive and well, and 14 are alive with distant disease.
Patients with clinical stage I or II non-oat-cell lung cancer may do well after curative resection despite the presence of N2 nodes. Very few patients with gross mediastinal node involvement have benefited from resection, and they should not be considered for thoracotomy unless innovative treatment measures can be offered. Cisplatin and vindesine have proved useful in patients with inoperable non-oat-cell lung cancer, and a trial of their adjuvant value in patients with resectable disease is under way.
Post-Comment: This article comes from one of the major groups advocating resection of stage III N2 disease and the only one that maintains that there is no difference in the prognosis of N2 adenocarcinomas and N2 squamous cell carcinomas. In order to understand the data, it is necessary to understand the makeup of their patient population and their mode of procedure, as both are really quite different from other centers. They do not use mediastinoscopy and did not mention using computed tomographic (CT) scan in a systematic way in this group of patients. They eliminated from further surgical consideration patients with pleural effusion, superior vena cava syndrome, widened mediastinum on chest radiograph suggesting bilateral mediastinal nodal involvement, and those with palpable supraclavicular nodes. Those with normal mediastinal shadows on conventional chest x-ray and no tracheal fixation or carinal abnormality on bronchoscopy were considered to be clinically free of mediastinal involvement (N0). It is not clear from their description just what mediastinal abnormalities on chest x-ray were allowed in those who went on to surgery. Clearly some were abnormal, as there was a group of clinical (as opposed to postsurgical) stage III N2 cases, and these actually did quite poorly (see below). Of 1,598 cancer patients, 706 had N2 disease. Complete resection of primary tumor and nodes was thought to have been possible in 151 of these (9.4% of the entire group or 21% of those found to have N2 disease at surgery). One important piece of data that is not presented here is how many of the 555 cases with N2 disease who were not resected had an exploratory thoracotomy; but judging from the authors' previous report, almost all did. As mentioned before in these pages, one practice almost unique to Memorial Hospital is the extensive use of radon seed implantation in tumors that are found to be unresectable at operation. In other words, since they believe that this is a good thing to do in inoperable cases, they have no reason to avoid thoracotomy. This also provides the unique opportunity of deciding at thoracotomy which patients might benefit from resection and which from radon seed implantation. It might be a much less reasonable policy to avoid doing mediastinoscopy or CT scans in institutions in which little is offered patients who are opened and found to be inoperable. (However, as mentioned above, it is the position of many that a normal mediastinal shadow on chest x-ray is a good indicator of a resectable case, even if mediastinal nodes are found at operation.)
In any event, this difference in practice at Memorial may be one reason why Martini and associates contend that patients with stage III N2 adenocarcinomas do as well as those with stage III N2 squamous cell carcinomas, whereas most authorities believe that the prognosis of adenocarcinoma, which grows more slowly than squamous cell cancer but metastasizes earlier and more widely, is worse overall than squamous cell carcinoma. Another factor that may possibly contribute to this difference of opinion is that Memorial is one of the three big screening enterprises. And, as has been discussed above, adenocarcinomas picked up on annual screening x-rays are almost all slow-growing tumors with a long doubling time and an intrinsically favorable prognosis. So to the extent that Memorial's population is made up of patients from the screening study, the population may contain patients with biologically more favorable tumors. A recent compilation from the Little Rock VA Hospital supports the generally held opinion that adenocarcinomas are worse. Of their patients who underwent lobectomy or pneumonectomy for lung cancer, 49.3% of those with squamous cell tumors and no lymph node metastases survived 5 years. If lymph node metastases were present, the survival rate fell to 22.7%. However, only 19% of patients with adenocarcinomas and no lymph node metastases survived 5 years, and this rate fell to 8.3% with lymph node metastases.
In any event, in the Memorial experience reported in the abstracted article, clinical stage (before thoracotomy and without mediastinoscopy) made a big difference. The best combination was a small tumor (T1) which was clinically N0 and at operation had positive node or nodes (surgical N2) at only one anatomic level. In these patients, the 1-, 3-, and 5-year survival rates were 83%, 78%, and 39%. Importantly, patients with nodes in the inferior mediastinum (subcarinal, paraesophageal, or pulmonary ligament nodes) did substantially worse, only about half as many surviving as those with upper mediastinal nodes. It has generally been believed, by Kirsh and by Abbey-Smith for instance, that nodes above the midtracheal level portend a bad prognosis. This was not found to be the case in the Memorial experience. Finally, one striking and important feature of this series is that it demonstrates clearly the importance of postoperative irradiation, which was used in almost all of these patients. Note that of their 102 cancer failures, 77% failed with distant metastases and only 13 (13%) had local-regional failure. This is to be contrasted with large autopsy series indicating that approximately 75% of squamous cell cancers fail locally, 75% of oat cell cancers fail due to distant metastases, and with adenocarcinomas the figures are about 50-50. The same point is made in a recent radiologic publication.--R.M.D.
Reduction in Local Recurrence and Improved Survival in Surgically Treated Patients With Small Cell Lung Cancer
Shepherd, Frances A.; Ginsberg, Robert J.; Evans, William K.; Feld, Ronald; Cooper, Joel D.; Ilves, Riivo; Todd, Thomas R. J.; Pearson, F. Griffith; Waters, Paul F.; Baker, Michael A.; DeBoer, Gerrit
J. Thorac. Cardiovasc. Surg., October 1983, Vol. 86, Pg. 498-506, Ch. 16-5
Affiliation: Toronto
Pre-comment: We participated in a review of small cell lung cancer this past year. Some points that interested us will be mentioned here.
First, there is at least some reason to doubt the presently widely taught concept that small cell lung cancers (SCLCs) are different in kind from other lung cancers, a point of view that has been used to provide rationale for treating them differently from other lung cancers. Reference was made to the opinion of Raymond Yesner, also discussed in the 1983 YEAR BOOK, that SCLCs are endodermal (like other lung cancers) rather than neuroectodermal in origin, are unrelated to carcinoids, and, rather than being different in kind, are just the least differentiated and therefore the most aggressive of lung cancers. While many disagree with this opinion, we believe that Dr. Yesner's argument is sound and explains far more than the alternative, and at present more popular, idea that SCLCs are APUD-omas that have a unique histogenesis and are malignant variants of pulmonary carcinoid adenomas. Second, the present benefits and limits of chemotherapy were discussed. (One third of patients present with limited disease; complete response in 50% of these provides a median survival of 14 months and a 3-year survival of 15%-20% [of the limited disease group], serious toxicity in 25%, and chemotherapy-related death in 2%-5%). Third, the opinion was discussed that further modifications in amounts or kinds of chemotherapy have not and are not going to improve things further. It was also mentioned that retreatment of failed cases had produced minimal benefit at best and, in view of the added toxicity, has not been shown to date to be worthwhile. The still open question as to whether chest irradiation contributes anything to long-term results in limited disease was discussed. The tentative conclusion from the literature is that although irradiation can not be shown to increase survival overall, it might provide increased 2-year survival in those who have limited disease and demonstrate complete response. It will be mentioned here that a small but good study addressing this point this past year has concluded that there is no demonstrable benefit to be gained by adding irradiation even in that specific subset (limited disease, complete response). Data concerning the value of prophylactic cranial irradiation as compared with symptomatic cranial irradiation was also reviewed, with the conclusion, similar to the case made for chest irradiation, that no overall increased survivorship was produced thereby, but there seemed to be some increase in prolonged (2-year) survival in the subset with limited disease and complete response to chemotherapy. Finally, the increasing evidence that surgery is beneficial in some instances of SCLC was reviewed.
The following is an abstract of yet another article on surgery in SCLC, differing from the others in that the patients were treated not only with chemotherapy and surgery in some order, but in many cases with irradiation as well. This triple therapy is an attempt to control local-regional recurrence, a pattern observed in 25%-50% of limited disease oat cell cancers treated with either chemotherapy alone or chemotherapy and irradiation.--R.M.D.
Abstract: Small cell lung cancer accounts for 15%-20% of all primary lung carcinomas, but fewer than 5% of cases in most surgical series. Failure of local control remains a major problem. The authors reviewed the results of surgical resection in 35 patients seen in 1976-1982 with small cell lung cancer, 27 men and 5 women with a median age of 62 years. In 28 cases, surgery was done because the disease was technically resectable and small cell cancer had not been definitively diagnosed. Most of these patients received chemotherapy postoperatively. Seven others received combination chemotherapy before resection as part of a planned combined treatment approach. Twenty-one patients received prophylactic cranial irradiation, and 22 received irradiation to the tumor bed and mediastinum.
Eighteen patients had a lobectomy, 8 a bilobectomy, and 9 a pneumonectomy. Resection was considered ``curative'' in 33 cases. Fifteen patients relapsed and died of recurrent tumor. Relapse was most frequent in the brain and skin; only 2 lesions recurred locally. Brain metastases developed in nearly one fourth of patients given prophylactic cranial irradiation. Three of the 7 patients given preoperative chemotherapy had residual disease at surgery. A definite survival advantage was apparent for patients who underwent surgical resection, even in comparison with nonsurgical patients who had complete remission. Survival could not be related to the stage of disease.
A policy of ``adjuvant resection'' of small cell lung cancer has led to prolonged survival for some patients with limited disease. Resection probably contributes to survival by reducing recurrence at the primary site, and it may even have a role in local control in cases of large primary tumor or N2 disease. Surgery seems indicated in selected patients with limited disease who seem to have operable disease and who respond significantly to induction chemotherapy. Prophylactic cranial irradiation also should be used, and irradiation of the primary site after surgery is also indicated. Further chemotherapy with non-cross-resistant drug combinations may be considered for patients with residual disease postoperatively.
Post-Comment: One important point to be emphasized is that only 2 of the 15 patients who relapsed had local-regional (primary or nodes) failure. This is an incidence of 6% (2 out of 35). The authors emphasized that in other series 50% of treated patients have local-regional recurrence, and in half of these this is the only site of relapse. As mentioned, 28 of 35 patients were operated on because the tumor appeared to be resectable and the cell type was not known. The other 7 were known to have SCLC and were treated with chemotherapy before operation. Of these, all but 2 were thought to have had complete resection of all tumor. Only 9 of the total group had no involved nodes at surgery (surgical stage I); 20 had N1 disease, and 6 had N2 disease (stage III). The median survival in the total group was 92.4 weeks, and the actuarial 5-year survival was 25%.
This seems to be the most promising new approach to SCLC. But it raises difficulties because those who survive the vigorous chemotherapy are at risk of developing other serious problems. Volk et al. report follow-up of 8 patients (out of 76) who were alive and disease free for at least 12 months following completion of treatment for SCLC. Five of these developed second malignancies, and in 2 they were acute leukemia. Another patient developed a serious dementia. (The long-term results of cranial irradiation in terms of dementia are by no means clear at present.) If the combined use of irradiation, surgery, and chemotherapy (as presently construed for oat cell cancer) turns out to be as good as indicated in the abstracted article, it will almost certainly be associated with lots of the sort of long-term complications reported by Volk et al., particularly hematologic malignancy. One possible solution is suggested by English authors who believe that maximum combined chemotherapy may not be needed. Another article advocating much less chemotherapy appeared this past year. Dosage was cyclophosphamide, 50 mg orally daily, methotrexate, 1.25 mg orally daily, procarbazine hydrochloride, 50 mg orally daily, and vincristine sulfate, 10 mu g/kg intravenously weekly, continued until relapse or for 2 years. Median survival of the patients with extensive disease was 20 weeks. Fifty-seven percent of the 21 patients with limited disease had a complete response; median survival was 55 weeks; 18-month survival was 36%; and 3 patients were disease free after discontinuation of therapy for 48, 22, and 7 months. Most importantly, only 4% of the patients had significant neutropenia (compared with almost 100% in what is now regarded as conventional multiple-drug chemotherapy for SCLC), and there were no treatment-related deaths. This is a small experience, but the results are impressive. Particularly hopeful for the future is the possibility that this much more gentle chemotherapeutic regimen might, when combined with surgery and irradiation, be as good as the more rigorous multiple-drug chemotherapy usually given to patients with SCLC and at the same time might avoid not only its short-term morbidity and mortality but also the long-term complications discussed in the article by Volk et al. discussed above.
Two useful SCLC articles will be briefly mentioned here. Kelly et al. report that bilateral bone marrow biopsy was required to demonstrate tumor, when present, in only 2 of 76 instances, and both of these were in cases being ``restaged'' after relapse. In 78% both aspirate and biopsy were positive and there were no patients with a positive aspirate and negative biopsy. There was no difference in survival between those with positive and those with negative marrows. Brennan and Craddock report a case of limbic encephalopathy (dementia, a seizure disorder, and inappropriate affect) as a paraneoplastic syndrome complicating SCLC, which cleared after chemotherapy with non-CNS-active drugs and without cranial irradiation.--R.M.D.
Reexpansion Pulmonary Edema After Pneumothorax
Kernodle, Douglas S.; DiRaimondo, Carol R.; Fulkerson, William J.
Abstract: Reexpansion pulmonary edema after treatment of pneumothorax is an unusual occurrence that can result in severe morbidity or death. Two cases, one fatal and one life-threatening, were encountered within 2 weeks. At least 25 cases now have been reported in the world literature. The pneumothorax is generally present for several days before reexpansion. Most cases are of spontaneous rather than traumatic origin. Most patients have had chronic obstructive lung disease. The symptoms include chest pain, dyspnea, and coughing. Coughing paroxysms may occur at the time of reexpansion or shortly afterward. Pulmonary edema is evident within minutes to several hours after reexpansion. Patients commonly expectorate frothy secretions. Hypotension is a frequent finding and occurred in both the present patients. Deaths have been attributed to hypoxia and bradycardia-asystole. The present death was due to hypotension with low cardiac output.
The pathophysiology of reexpansion pulmonary edema is uncertain. Increased pulmonary vascular permeability may result from anoxic injury or mechanical vascular damage. Decreased surfactant in the atelectatic lung may also be a factor. It has been suggested that gradual reexpansion of collapsed lung using underwater seal alone or small repeated needle aspirations may help prevent pulmonary edema, but the syndrome can develop despite this precaution. Patients with prolonged pneumothorax should be closely observed for 24 hours after evacuation of air. Supplemental oxygen should be routinely used during and after lung reexpansion.
Post-Comment: According to the literature, this is an uncommon event, although the occurrence of two cases on one service over a brief period makes it likely that it is actually not all that rare. The edema fluid is high in protein content, and the hemodynamic data reported in the abstracted article, said to be the only such yet recorded, demonstrated a normal pulmonary capillary wedge pressure. Accordingly this seems to be a ``permeability'' type of pulmonary edema. The usual clinical setting (fairly prolonged duration of pneumothorax, coincident COPD, and perhaps too rapid reexpansion) and the frequency of a fatal outcome are worth knowing.
Mathru et al. report on another unusual situation in which pulmonary edema may develop, i.e., tracheal intubation in a patient with burns. This also is a permeability type of pulmonary edema (high edema fluid protein). The authors believe that both burn and intubation are necessary for development of pulmonary edema, but we have seen one case in which an otherwise well young woman with obstructing epiglottitis rapidly developed pulmonary edema after tracheostomy. Mathru and associates suggest that abrupt changes in alveolar pressure as a result of bypassing the glottic expiratory resistance may be important, and they cite an article advising continuous positive pressure to avoid pulmonary edema when burn patients are intubated.
It has become conventional to divide pulmonary edema into two sets, one due to increased pulmonary microvascular permeability (high edema fluid protein) and the other due to increased microvascular pressure (low edema fluid protein). The latter is usually thought to be due to left heart failure (or to other causes of increased left atrial pressure) and accordingly always associated with an increased pulmonary capillary wedge pressure. That is to say, when the pulmonary capillary wedge pressure is normal, it is assumed that the cause of the edema is not hydrostatic but rather increased permeability of the pulmonary microvasculature. It now seems likely, however, that there is a third general set of conditions associated with pulmonary edema in which the common factor is increased pulmonary arteriolar pressure or perfusion or both. This can distend the capillaries and increase the area of the capillary bed where fluid exchange occurs without affecting the pulmonary capillary wedge pressure.
This ``hydrostatic pulmonary edema with a normal wedge pressure'' seems to be the most likely explanation for postpneumonectomy pulmonary edema, as reported by Zeldin et al. They note that ``although . . . sporadic, its consequence is usually death.'' Nine of their 10 cases (collected from several hospitals and surgeons) involved right pneumonectomy. Chart review suggested excess crystalloid fluid. Wedge pressures were measured in two cases and were normal, but the mean pulmonary artery pressures were elevated. Zeldin et al. believed that several factors were involved: the sudden decrease of the pulmonary capillary bed by more than 50% (in the case of a right pneumonectomy) and consequent sudden overperfusion of the remaining lung; vigorous fluid therapy; and, interestingly, pain-induced catecholamine excess, which would further increase the cardiac output. They did not think that decrease of plasma oncotic pressure due to excessive use of crystalloid fluids played much of a role Staub and others believe that sudden massive overperfusion and high pressures in a limited pulmonary vascular bed produce endothelial injury, with resulting permeability changes as well, so it is possible that both hydrostatic and microvascular permeability changes are involved in this sort of pulmonary edema. The authors warned against overzealous crystalloid therapy when pneumonectomy is to be done (indicating that their anesthesiologists are inclined to ``load the patient up'') and advised the use of sufficient narcotics to control pain in order to decrease catecholamine discharge.
Production of pulmonary edema by pulmonary arterial hypertension is thought to be a part of the explanation of neurogenic pulmonary edema, although some studies suggest catecholamine-induced increase in pulmonary vascular permeability and vasoconstriction at the postcapillary venule level are also contributory. An experimental study in the chronically instrumented awake sheep model by Minnear et al. this past year indicated that the increased pulmonary lymph flow and protein clearance following bolus epinephrine was not due to increased pulmonary vascular permeability but rather to an increase in the surface area of the fluid-exchanging capillary bed as a result of an abrupt increase in pulmonary arteriolar pressure. They also suggested that norepinephrine constricts the lymphatic outflow tract and concluded that a ``combination of high capillary pressure and impaired lymphatic flow may be the bases for the development of neurogenic and catecholamine-induced pulmonary edema.''
Pulmonary edema attributed to catecholamine excess has been reported in some other uncommon circumstances, including pheochromocytoma and hypoglycemia complicating insulin shock therapy. Mabie et al. report 10 instances of noncardiogenic pulmonary edema complicating the use of beta-adrenergic drugs to arrest premature labor, which is called tocolytic therapy.
Noncardiogenic pulmonary edema is a rare but serious complication of radiocontrast material injection. The mechanisms of this, as well as those underlying the better-known anaphylactoid reaction, are unknown. Neither is an immunologic reaction in the usual sense, since both can occur without prior exposure. Radiocontrast material is known to release histamine and to activate the alternate complement pathway. Histamine does cause transient increase in pulmonary vascular permeability in the sheep model, so perhaps that is the critical mechanism.{--}R.M.D.
Systemic Vasculitis With Asthma and Eosinophilia: A Clinical Approach to the Churg-Strauss Syndrome
Lanham, John G.; Elkon, Keith B.; Pusey, Charles D.; Hughes, Graham R.
Medicine (Baltimore), March 1984, Vol. 63, Pg. 65-81, Ch. 17-4
Abstract: Churg-Strauss syndrome (CSS) is a systemic vasculitis associated with eosinophilia occurring in persons with asthma and allergic rhinitis. Findings in 16 patients seen at Hammersmith Hospital in London between 1976 and 1982 were compared with those in 138 previously reported CSS patients. Twelve of the 16 patients had allergic upper respiratory tract disease, and 9 of the 12 required surgery, usually nasal polypectomy. Thirteen patients were anemic at some stage. Circulating immune complexes were found in only 2 of 14 patients. All patients but 1 responded rapidly to prednisolone or prednisone therapy. The disease was fatal in only 1 patient. Six patients relapsed during a mean follow-up of 38 months. Necrotizing vasculitis was found in the skin in 5 patients, the kidneys in 3, and muscle tissue in 2. A shorter duration of asthma before the onset of vasculitis was associated with a poorer prognosis.
Asthma may precede the development of systemic vasculitis by as much as 30 years in patients with CSS. Pulmonary infiltration is a central feature of the disorder. Eosinophilia and asthma are closely associated in patients with vasculitis. About half of the deaths from CSS are caused by cardiac involvement. Most patients have neuropathy, typically the mononeuritis multiplex form. The typical renal lesion is focal segmental glomerulonephritis with necrotizing features. The disease responds well to steroid therapy, but some patients may require adjunctive immunosuppressive treatment. Hypertension may be a major management problem both in the vasculitis phase and afterward. Variant forms of CSS occur, including a ``limited'' form in which none of the typical clinical features is present.
Post-Comment: This scholarly review is far too long to be abstracted. It brings order to a confused area. The authors contend that CSS has been greatly underdiagnosed because of too rigid adherance to the three defining histologic criteria: tissue infiltration with eosinophils, necrotizing vasculitis, and granuloma formation. They also believe that the clinical picture is often sufficiently typical to allow diagnosis without histologic confirmation. They describe a three-phase illness, often years in duration. The first phase is late (adult) onset allergic disease, always including asthma but often beginning with allergic rhinitis. This may be followed by a period characterized by peripheral blood and tissue eosinophilia, which may include a number of conditions including L{o"}ffler's syndrome, chronic eosinophilic pneumonia, eosinophilic gastroenteritis, eosinophilic pericarditis, and eosinophilic myocarditis. These may wax and wane over a period of years, but ultimately may be interrupted by a third life-threatening vasculitic phase characterized by systemic symptoms (fever, weight loss), skin rash, and vasculitic and granulomatous lesions in a number of organs.
Churg-Strauss syndrome was fatal in only 1 of 16 cases in the present study but was fatal in many gathered from the literature. Autopsy findings in 45 cases were compared with histologic findings in various biopsies in nonfatal cases. Granulomas were demonstrated in 40% and 38% respectively, eosinophilic tissue infiltration in 42% and 57%, and vasculitis in 90% and 70%. All three histopathologic features were present together in only 24% of the patients examined at autopsy and 13% of tissue biopsy specimens in nonfatal cases. However, criteria for inclusion in this review were clinical rather than pathological and consisted of (1) asthma, (2) peak eosinophilia greater than 1,500 cells/cu mm, and (3) systemic vasculitis in 2 or more extrapulmonary organs.
Asthma, usually mild and adult-onset, was present in all of the patients in this study and had been present in some for many years before onset of vasculitis. The authors note that asthma is quite unusual in Wegener's granulomatosis (WG) and polyarteritis nodosa (PAN). Pulmonary infiltration was also an important and frequent clinical feature, present in 72%. This could be anything really, including fleeting patchy infiltrates, the peripheral shadows characteristic of chronic eosinophilic pneumonia, a miliary pattern, or multiple nodules or masses. These last were not observed to cavitate, however, in contrast to WG. Eosinophilic pleural effusions were common. Allergic rhinitis was a frequent component of the initial phase of allergic illness and often led to nasal obstruction, recurrent sinusitis, and nasal polyposis requiring surgery. The authors contrasted this to the clinical picture of nasal pain and bloody discharge that characterizes nasal involvement in WG. Eosinophilia was present in most, but a few cases demonstrated tissue eosinophilic infiltrate without blood eosinophilia. The association of asthma, eosinophilia, and vasculitis, or the occurrence of these in sequence was sufficient to include the patient as a case of CSS and to exclude WG and PAN in the authors' opinion. They stated that both asthma and eosinophilia often faded with the onset of vasculitis.
They regarded eosinophilic gastrointestinal disease (characterized pathologically by tissue eosinophilia and clinically by a number of syndromes including colitis, bowel obstruction, peritonitis, unexplained pain, and cholecystitis) and eosinophilic cardiac disease (pericarditis, myocarditis, and myocardial infarction due to vasculitis) as counterparts of the pulmonary eosinophilic syndromes. Myocardial replacement by granulomas and scars were observed in the cases examined at autopsy. Skin lesions were frequent, including urticaria, maculopapular rashes, palpable purpura, and a distinctive subcutaneous vascular lesion consisting of granulomas and tending to ulceration. The most typical neurologic finding was mononeuritis multiplex, but symmetrical sensorimotor peripheral neuropathy was also seen, and a variety of central nervous system abnormalities, ranging from confusion or psychosis to coma, were present in a substantial minority. Cerebral hemorrhage and infarction were major causes of morbidity and death. Renal involvement was pathologically the same as in WG, i.e., focal segmental necrotizing glomerulonephritis with a tendency to crescent formation, but clinically it was much more benign than in either WG or PAN, and renal failure was very uncommon. However, hypertension was a common clinical feature. Migratory polyarthritis of a nondestructive sort was often seen.
The immunopathogenesis is unclear, but IgE concentrations were elevated in the vasculitic phase in all of the Hammersmith cases. Circulating immune complexes were sought and for the most part not found. Atopy was present in most patients and seemed to be a necessary precondition for the disease to develop, but atopy in family members was not common, and the allergic diathesis was unusual in that it was adult in onset. Visceral angiography showed vasculitis or aneurysm formation in 4 of 9 cases so studied.
Treatment with steroids (40-60 mg of prednisone per day) was generally very successful. It should be mentioned that cases called CSS at one time and WG at another are very uncommon. These syndromes tend to be distinct. Also, as will be discussed below, most people now regard lymphomatoid granulomatosis as a lymphatic neoplasm with a tissue response to the neoplastic cells rather than an allergic granulomatous process.
Another good review of these illnesses has been published this past year by Andrew Churg (not the Churg of Churg-Strauss). His point of departure was the landmark 1973 essay by Liebow in which it was suggested that WG, lymphomatoid granulomatosis, bronchocentric granulomatosis, and necrotizing sarcoid granulomatosis were all closely associated diseases, sharing the histopathologic characteristics of angiitis and granululomatosis in the lung. In this essay, using WG as the condition of reference, Churg differentiates the CSS (allergic angiitis and granulomatosis) much in the same way as was done in the abstracted article by Lanham, agreeing that it is in point of fact not at all like WG. Necrotizing sarcoid granulomatosis is discussed at some length with excellent references and a lot of collected cases. Clinically this is a condition that occurs mostly in women (4:1) with a wide age range, characterized by nodular pulmonary infiltrates, often associated with hilar adenopathy, and no evidence of extrapulmonary involvement of a vasculitic sort. The extrapulmonary involvement that was seen in 4 of the 32 cases with which Churg had personal experience was essentially a sarcoidosis type of picture, with uveitis in 3 and hypothalamic insufficiency in 1. Histologically the distinctive feature is the presence of numerous sarcoid-like granulomas along with a little vasculitis in the lung and some tendency to necrosis. The prognosis is always good and treatment appears to be unnecessary. The vasculitis is entirely confined to the lungs and is little different from vasculitis which may be seen in ordinary sarcoidosis, except for a tendency to be somewhat more destructive. The author feels that what is called necrotizing sarcoid granulomatosis pathologically corresponds exactly to the clinical entity of nodular sarcoidosis. Nodular sarcoidosis was discussed in the 1978 YEAR BOOK. Our own experience with it indicates that it is probably an early manifestation of sarcoidosis with an excellent prognosis for complete spontaneous resolution. Dr. Churg's review mentions all of the literature concerning nodular sarcoidosis as well as his own clinical analysis. It is his opinion that nodular sarcoidosis should no longer be considered as a relation, even distant, of WG. Lymphomatoid granulomatosis is discussed as almost certainly a pulmonary lymphoma rather than a vasculitis. He notes that the contention of Fauci that this represents a condition related to WG is based on the somewhat strained argument that some cases appear to respond to the same sort of therapy as WG. In point of fact the therapeutic response of lymphomatoid granulomatosis is much less clear-cut. Dr. Churg states that necrosis and extensive severe vasculitis are common occurrences in many types of primary and secondary lymphomas involving the lung. The evolution of lymphomatoid granulomatosis into frank lymphoma in a fairly large proportion of cases is of course the best argument for its neoplastic nature. He notes that ``there are several lymphoproliferative malignancies (angioimmunoblastic lymphadenopathy, Mediterranean lymphoma, mycosis fungoides), in which apparently `benign' lymphoid proliferation of a more or less long duration develops into a cytologic lymphoma, and some cases of lymphomatoid granulomatosis that are not overtly lymphomas or angioimmunoblastic lymphadenopathies may fit this general category.'' In any event he feels that lymphomatoid granulomatosis should be classified as a malignant or premalignant process, as will be further discussed below. Bronchocentric granulomatosis, it is now widely agreed, is a nonaggressive process with no extrapulmonary manifestations and is probably the same process as mucoid impaction of the bronchi or allergic bronchopulmonary aspergillosis, different only in that diffusion of antigen from its bronchial location induces a local peribronchial granulomatous and vasculitic response. Churg divides lethal midline granuloma into two conditions: one quite similar to lymphomatoid granulomatosis both histologically and in terms of its evolution into frank lymphoma, and a second condition histologically similar to WG which he regards as a local form of it. One other interesting point made in this essay is that tuberculosis and fungal infections can cause identical pathological processes of necrotizing granulomas associated with vasculitis. He raises the question whether some instances of very benign ``localized WG'' confined to the lungs might not in point of fact represent pulmonary infection with an organism (mycobacterial or fungal) to which the tissue response is granuloma formation. So of the original five conditions discussed in Dr. Liebow's 1973 essay, only WG and localized WG remain. Dr. Churg's review is excellent and very much worth reading in the original.
Another excellent review article in this area by Dr. Fauci will be mentioned briefly. He divides vasculitides into a group of systemic necrotizing vasculitis including classic PAN the CSS, and what he terms the polyangiitis overlap syndrome with features of both; a hypersensitivity vasculitis group including Henoch-Sch{o"}nlein purpura, serum sickness, drug-related vasculitis, and those associated with infections, cancers, collagen vascular diseases, and congenital deficiencies of the complement system; WG; the giant cell arteritides including temporal arteritis and Takayasu's arteritis; and others with considerable individual distinctiveness including Kawasaki disease, Beh{c-ced}et's, and Buerger's disease. He differs from the opinion about CSS expressed in the article summarized in abstract 17-4 in that he believes the prognosis is much worse and resembles that of PAN, and he identifies an overlap syndrome with features of both PAN and CSS. Classic PAN is defined as a segmental disease of small- and medium-size muscular arteries, with a predilection for bifurcations and branchings, associated with nonspecific but serious systemic symptoms together with some symptoms with a little more specificity. These include mononeuritis multiplex and abdominal pain (due to visceral ischemia), in which visceral angiography (kidneys, liver, and intestine) demonstrating small (1-cm) aneurysmal dilatations may be very important diagnostically, blind tissue biopsy being often nonproductive. This is the illness that is often associated with hepatitis B and in which circulating immune complexes (67%), depression of C3 titers (70%), and cryoglobulins (25%) are common. He says the 5-year survival untreated is 13%, with death occurring as a result of renal disease or, less commonly, of the effects of gastrointestinal infarction. The hypersensitivity group, he says, are usually much less serious illnesses with pathologic involvement of smaller vessels, arteries, and veins and with clinically predominant dermatologic manifestations including palpable purpura. He says that there may be arthritis in 40%, renal disease in 37%, and gastrointestinal disease in 15%, but that these are not usually progressive. The review contains considerable information on details of therapy which will not be repeated here. (For instance, the hypersensitivity group is said not to do well with steroids usually, and Dr. Fauci feels it is best treated with nonsteroidal anti-inflammatory agents.)
Wegener's granulomatosis, the condition to which all of these vasculitides are compared for purposes of definition, has occupied a lot of space in these pages over the past several years. The most extensive analysis of clinical findings and response to therapy is the article by Fauci and associates abstracted last year, in which a success rate of 93% is claimed for cytoxan and alternate-day steroids. In the editorial discussion following that abstract, another much smaller experience was mentioned that reported much worse results, with only 5 of 13 patients surviving 1 year. (It is of note that the disorders of 2 of their patients clinically resembled Goodpasture's syndrome{--}see below.) Another article from the Hammersmith Hospital this past year also reported much poorer treatment results than the NIH group. Seven of 18 patients, all of whom had severe renal disease, died during the induction phase of treatment, and another 5 died later of the disease. (This report emphasized the frequency of previous suppurative lung disease or tuberculosis, as did the early report of ``pulmonary polyarteritis'' by Rose and Spencer.) This less good prognosis seems closer to our own experience. On the other end of the spectrum, MacFarlane and associates report 3 cases of WG that ran a prolonged, indolent course without treatment, in spite of major manifestations outside of the respiratory tract distinguishing them from so-called ``localized Wegener's granulomatosis.'' It seems at least possible that the good results reported by the NIH group represent the sort of patient selection of a somewhat less ill population which is inherent in referral to a national center. The patients reported by Brandwein et al. and Pinching et al., who died quickly, would probably not be sent on any long journeys.
There is a tendency in our city to diagnose WG without the characteristic pulmonary or upper airways findings, often on the basis of segmental necrotizing glomerulonephritis, which is not specific for WG, the same lesion being seen, for instance, in CSS. Part of the impulse to do this has to do with the good results reported for cytoxan and alternate-day steroids in the NIH study. To us it seems that there remains good reason to restrict the term WG to disease characterized not only by vasculitis with or without glomerulonephritis but also by necrotizing granulomas in the lungs or upper airways or both. This seems advisable not only for clarity but also because it is not yet clear which drug does what in regimens containing cytoxan and steroids. It is possible, even likely, that cytoxan is good for immune-complex-mediated vasculitis; but it is very certain that cytoxan is effective, and steroids are not, in causing involution of the necrotizing granulomatous part of WG. Steroids seem to be the agent of choice when the process is uncomplicated vasculitis, such as in PAN, as well as in diseases characterized by vasculitis plus tissue eosinophilia, typically CSS.{--}R.M.D.
Ventilatory Failure Due to Asbestos Pleurisy
Miller, Albert; Teirstein, Alvin S.; Selikoff, Irving J.
Am. J. Med., December 1983, Vol. 75, Pg. 911-919, Ch. 18-1
Affiliation: Mt. Sinai School of Medicine
Abstract: The authors report the cases of 5 patients with ventilatory failure associated with asbestos-induced pleural fibrosis. All had been exposed to asbestos and had no other pleuropulmonary disease. They presented with severe dyspnea; a vital capacity less than 50% of the predicted value; a maximal voluntary ventilation less than 80% of the predicted value; CO{sub-2} retention, with a PaCO{sub-2} of 52 torr or above; and no airway obstruction. Two other patients had similar but less severe functional impairment. Three of the 7 patients were suspected of having a pulmonary or pleural neoplasm because of extension of pleural fibrosis into the lung.
Pleural effusions were observed in 4 patients and were recurrent or bilateral in 2 of them. All 7 patients had progressive shortness of breath. Pleural fibrosis was bilateral in all cases but 1. Five patients had little or no radiographic evidence of interstitial fibrosis, including 4 of the 5 with ventilatory failure. Four patients had thoracotomy with decortication, which did not produce clinical improvement in the 3 patients explored for neoplasia. Four patients with respiratory failure died, and the other is severely disabled. The average vital capacity was 33% of the predicted value. The radiographic findings in a patient without CO{sub-2} retention are shown in Figures 18-1 and 18-2. Pulmonary compliance was measured in 3 patients and was markedly reduced in all 3, 2 of whom did not have CO{sub-2} retention.
Some degree of interstitial fibrosis is likely in all patients with asbestos-induced pleural changes. Further cases of ventilatory failure are expected because of the increasing duration of exposure in the 9 million workers exposed to asbestos. Gallium scanning cannot be used to distinguish between fibrosis and neoplasia in asbestos-exposed subjects. Pleurectomy may improve ventilatory function by relieving entrapment of lung tissue by fibrocalcific pleura.
Post-Comment: There are six different pleural syndromes that have been attributed to asbestos exposure: benign asbestos pleural effusion, noncalcified or hyaline pleural plaques, calcified pleural plaques, benign pleural fibrosis (hyalinosis simplex), massive pleural fibrosis (hyalinosis complicata) described in the article abstracted above, and mesothelioma. It is difficult to know exactly how these relate to each other.
In the 1983 YEAR BOOK a landmark prospective survey of asbestos pleural effusion in a large number of Boston shipyard workers by Dr. Edward Gaensler was abstracted. The astounding fact to emerge was that this developed in 3.1% of all shipyard personnel studied and in 14.3% of those who actually worked with the materials, excluding office workers and supervisory personnel. Dr. Gaensler says that this is much the most frequent cause of pleural effusion in his Boston consulting practice. Benign pleural effusion, when it occurs, is the earliest clinical manifestation of asbestos disease, appearing in the first and second, and a little in the third decade after exposure has begun. The effusions are described as often asymptomatic and transient, but some are blood stained, some are bilateral, and a minority persist or recur to the extent that decortication is eventually required. Their relationship to pleural plaques, pleural calcification, and mesothelioma is not clear, but it is believed that they are often the initiating event leading to both simple and massive pleural fibrosis. Pleural plaques and pleural calcification usually don't occur until the second or third decade after exposure, and their relationship, if any, to either earlier pleural effusion or to subsequent mesothelioma is not known. Most likely they represent independent manifestations of asbestos exposure. Mesothelioma characteristically demonstrates a latent period of 20 years or more. About 2% of the Boston shipyard patients with asbestos pleural effusion went on to develop mesothelioma, but again it was not clear whether this was a causal sequence or just independent consequences of a single cause.
An article several years ago from Prague (Environ. Res. 6:455, 1973) used the terms hyalinosis simplex and hyalinosis complicata for simple and massive pleural fibrosis. This seems to be a good term, as it sets the process apart from hyaline plaques and pleural calcifications. They described 50 cases, half of whom also had parenchymal asbestosis. The size of the total population from which these cases were derived is not clear from the article, other than the statement that ``several hundred persons'' from a factory with asbestos risk were followed. Pleural fibrosis (hyalinosis) occurred about 30 years after initiation of exposure. There were 1 probable and 1 certain case of mesothelioma observed, an incidence of 4%. ``The regular follow-up . . . made it possible for us to gain an important insight: some persons live for many years with pleural hyalinosis without complaints, without any striking functional disorder, and without any significant changes in their clinical findings aside from progression of the calcified hyalinosis. In other persons the chronological course is characterized by acute inflammation of the pleura with effusion. The effusion was sterile in all the patients and the findings towards the diagnosis of tuberculosis were negative . . . . No cells suspected of malignancy were [seen]. The most interesting fact, however, is that after recessing of the acute phase a progression marked radiologically by diffuse opacity of the lung fields proceeded upward from the bottom. A functional disturbance of restrictive type proceeded simultaneously. It was evident at the autopsy that a considerable part of pleural surface had infiltrated the thorax wall and typical were coarse hyalinosis of the diaphragmatic pleura penetrating into the abdominal cavity . . . . In one case . . . the acute inflammatory phase occurred on the left and the later diffuse hyalinosis developed on both sides . . . . After five years, the patient died of the acute heart failure due to the cardiosclerosis and the autopsy proved, besides extensive compact hyalinosis of the right side, coarse hyalinosis of the diaphragmatic pleura penetrating as far as inside the liver tissue.'' Twenty-five percent of the total hyalinosis group (35% of those with complicating pulmonary parenchymal asbestosis and 9% of those without) had hyalinosis complicata (massive involvement with restrictive ventilatory defect). Three died of pleural disease; these patients constituted 11% of the total group but 57% of the group with hyalinosis complicata. In the authors' experience this complication was more frequent than parenchymal asbestosis, was almost as frequent a cause of death as bronchogenic carcinoma (which accounted for 4 deaths in this population), and was a more frequent cause of death than mesothelioma (2 cases, as mentioned above). It would be very interesting to know how many cases of benign asbestos pleural effusion progressed to ``hyalinosis,'' but these data are not in the article. It seems probable that many cases of benign asbestos pleural effusion go undetected, and it seems at least possible that all of the more chronic pleural manifestations{--}fibrosis, plaques, and mesothelioma{--}will have been preceded by subtle and transient pleural effusions. That just is not known.
The functional disability caused by asbestos-related diffuse pleural fibrosis was analyzed by McGavin and Sheers. Forty percent of the group they studied were disabled to the extent that they were unable to walk along level ground at a normal pace. Vital capacity was inversely correlated both with increasing breathlessness and with increasing radiographic pleural abnormality. There was no obstruction as such, and carbon monoxide diffusion was increased when adjusted for alveolar volume. These patients did not have parenchymal asbestosis, and the authors concluded that ``diffuse pleural fibrosis is a cause of impaired lung function and of disability, irrespective of the presence of other asbestos-related disease.'' The type of pulmonary functional defect that develops with hyalinosis complicata without associated parenchymal asbestosis is pure restrictive disease, analogous to that which complicates kyphoscoliosis, and respiratory failure with or without cor pulmonale is the way such patients die. In the article abstracted above it is implied that some patients will benefit from pleurectomy. However, the data indicate a substantial mortality, and the operation must be terribly difficult. Accordingly, any decision to operate would have to be made very carefully, and surgery is probably indicated only for those with far advanced and potentially fatal respiratory compromise.
As mentioned, the article abstracted above suggests that 4% of patients with asbestos pleural fibrosis will develop mesothelioma, and it is Dr. Gaensler's published opinion that 1%-3% of asbestos workers will eventually develop this tumor. It seems almost certain that mesothelioma will be observed more and more frequently in the next decade. Dr. Turner-Warwick and associates have reported their experience with 140 patients this year. In 56% there was a large pleural effusion on presentation, and dyspnea was the principal symptom. The others presented with pain, weight loss, and fever. In some cases recurrent and poorly understood pleurisy had been present for several years before the tumor became manifest, and 3 patients (2%) presented with a spontaneous pneumothorax. Intermittent fever and sweating occurred in 45 of 140 patients. Cytologic studies of pleural fluid yielded positive results in only 25% of those examined, and these were not diagnostic of mesothelioma. Closed pleural biopsy was diagnostic in only 21% on a single attempt, and in another 8% on a second attempt. Even open pleural biopsy missed the diagnosis in 12% of those in whom it was done. Diagnosis was sometimes established by biopsy of cutaneous deposits that developed after thoracentesis, but in the authors' opinion these deposits seldom caused problems, and the recognized possibility that open biopsy could lead to cutaneous implants was not thought to contraindicate the procedure. Treatment was symptomatic. Recurrent pleural effusions were controlled in some by intrapleural administration of bleomycin, preceded by aspiration to dryness and followed by suction. Pain was the most difficult clinical problem and often required neurosurgical procedures.
A recent article by Hillerdal reviews 4,710 published cases of mesothelioma. He says that the malignant variety is entirely unrelated to benign mesothelioma (which is itself unrelated to asbestos exposure). In persons unexposed to asbestos the incidence of malignant mesothelioma is 1 per million per year. The confusion between mesothelioma and pleural metastases from other tumors is discussed. He states that lymphatic and blood-borne metastases of mesothelioma are found in 50%-70% of cases; however, these rarely cause clinical trouble, and ``of patients with asbestosis, 10%-15% will die from mesothelioma.'' Smoking does not increase the risk. ``Signs of past exposure to asbestos, i.e., pleural plaques and/or basal parenchymal fibrosis on the other side, are of additional diagnostic value and can be seen in 20% or more of the cases.'' Finger clubbing and osteoarthropathy are common with localized benign mesothelioma but rare in diffuse malignant mesothelioma. There is a good deal about prognosis and treatment in this huge review. Hillerdal states, ``Operative measures are of doubtful value, but might give palliation in selected cases. Radiotherapy may prolong life, though there is a lack of controlled prospective studies, and has some palliative effect. Chemotherapy, especially . . . `sarcoma combinations,' have [sic] been shown to cause tumour reduction in individual cases, but the necessity of confirming this by large series with randomization according to cell type is indisputable.''
Pleural plaques and calcifications are not of themselves premalignant lesions, although they may or may not be preceded by pleural effusions and later can also be followed by mesothelioma. Even this latter association may be simply two different results of the same inciting event rather than a direct causal relationship. Pleural calcifications are, according to some, just late dystrophic changes in pleural plaques. This is not at all clear to the editor.
A ``Landmark Article'' on asbestosis and cancer dating from 1964 was recently republished in the JAMA and is worth noting here and having for one's files. Of deaths recorded from asbestos workers, 17.8% were due to neoplasia. Forty-five (of 632) asbestos workers died of cancer of the lung or pleura, whereas only 6.6 cases would have been expected from the general population of comparable age. Forty-two were bronchogenic, and 3 were mesothelioma. Twenty-nine cases of cancer of the stomach, colon, and rectum were observed, while 9.4 would have been expected. The relative risk of respiratory cancer in this population was therefore 6.8, and that of gastrointestinal cancer was 3.1.{--}R.M.D.
Lupus Pleuritis: Clinical Features and Pleural Fluid Characteristics With Special Reference to Pleural Fluid Antinuclear Antibodies
Good, James T., Jr.; King, Talmage E.; Antony, Veena B.; Sahn, Stevan A.
Chest, December 1983, Vol. 84, Pg. 714-718, Ch. 18-2
Affiliation: University of Colorado Health Sciences Center
Abstract: Pleuritis due to systemic lupus erythematosus (SLE) itself can be confused with pleuritis due to pulmonary embolism, viral infections, tuberculosis, and congestive heart failure occurring in patients with SLE. The authors studied 14 patients with lupus pleural effusions, comparing them with a group of lupus patients whose pleuritis was due to another process and to another group with pleuritis but without lupus. The most common presenting symptoms in the lupus pleuritis patients were pleuritic pain and dyspnea. Although other manifestations of lupus were present in all, the pleuropulmonary findings were the most striking in these patients, and in some the other findings of SLE were initially overlooked. The pleural effusion was bilateral in half; most were small to moderate in size, and other x-ray abnormalities (alveolar infiltrates, atelectasis, or cardiac enlargement) were present in 11 of the 14.
All 14 patients had exudative effusions by protein criteria (pleural fluid to serum ratio [PF/S] greater than 0.5) or lactate dehydrogenase (LDH) criteria (PF/S greater than 0.5 or LDH level greater than 200 IU/L). The LDH values never exceeded 550 IU/L. Fluid cell count ranged from 230 to 15,000 white blood cells with 10%-100% polymorphonuclear leukocytes; in half the patients, polymorphonuclear leukocytes predominated. Eleven had normal pleural fluid glucose levels (more than 60 mg/dl; PF/S greater than 0.5) and normal pH (greater than 7.3). In 3 the fluid had low glucose and pH measurements but was not otherwise different from the rest. Neither the total leukocyte count, differential, effusion size, nor duration of the effusion correlated with either pH or glucose content. There was marked variation in the appearance of the fluid, which ranged from serous to frankly sanguineous. Of the 8 patients who had pleural fluid examined for SLE cells by Wright stain, 7 had at least one typical SLE cell. In 11 of 13 patients with lupus pleuritis the pleural fluid antinuclear antibody titer was greater than 1:160, and in 9 the PF/S of antinuclear antibody (ANA) was greater than 1. Nine patients had pleural fluid hemolytic complement values determined, and 5 were low (less than 20 units/ml).
In patients with SLE and acute pleural effusion of other etiologies the physical findings and symptoms were indistinguishable from those of lupus pleuritis. All had positive serum ANAs; however, only 2 had pleural fluid ANAs, and in these the PF/S ratio of ANA was less than 1. In 67 patients with pleural effusion but without lupus the pleural fluid ANA was negative in all. Thus all patients with lupus pleuritis had either SLE cells present, high pleural fluid ANA titers, or a PF/S ratio of ANA greater than or equal to 1. It is of note that pH, glucose content, and leukocyte count with differential could not differentiate between pleuritis due to lupus and that due to other causes.
Post-Comment: Good pictures of lupus erythematosus cells in the pleural fluid, along with a description of how this examination is done (let fluid stand at room temperature for a couple of hours and Wright stain either the unspun or centrifuged sample, depending on the cell count), appear in articles by Carel et al. and by Reda and Baigelman. These articles also emphasize that the pleural effusion can be the initial manifestation of lupus, either drug-induced or naturally occurring. The article abstracted above provides easy guidelines as to how to make this diagnosis and will probably be definitive for some time. Previous articles have discussed pleural fluid complement in lupus and in rheumatoid arthritis. Since complement studies are hard for many people to obtain, it is helpful to know that they are not really needed.
Rheumatoid arthritis is the other collagen vascular disease most typically complicated by pleural effusion and in which pleural effusion may be the initial clinical manifestation. A rheumatoid effusion will demonstrate a low pleural fluid glucose level more typically than any other exudative effusion except empyema. This is thought to be due to an entrance block to glucose. Such effusions are also typically low in pH, presumably because acidic glucose metabolism end-products accumulate and can't get out. These changes may develop very rapidly. The exudate may be opalescent but is frequently turbid, and may be milky. Rheumatoid factor may be present in the pleural fluid in higher titer than in blood or may be present in pleural fluid and completely absent in the blood; thus this would seem to be another useful and easy thing to measure. The effusion can accumulate very rapidly and, as mentioned, may precede other manifestations of the rheumatoid state. Granulocytes containing cytoplasmic inclusions, thought by some to be immune complexes, have been seen in joint fluids and in pleural fluids of rheumatoid arthritis patients, but while these inclusions are suggestive, they are in no way specific for rheumatoid arthritis. However, an article from Oxford contends that reliable cytologic diagnosis of a rheumatoid pleural effusion can be made by demonstrating degenerating polymorphonuclear leukocytes with amorphous extracellular material and epithelioid cells, many of which are multinucleate. As mentioned, complement levels are usually low in rheumatoid pleural effusions.
The other pleural manifestation of rheumatoid arthritis is empyema, thought to be due either to rupture of a subpleural rheumatoid nodule or to buildup of progressively thick pleural fluid as lymphatic paths for removal of cells and protein become clogged. These may be secondarily infected or may remain sterile.{--}R.M.D.
Pneumothorax in the Marfan Syndrome: Prevalence and Therapy
Hall, John R.; Pyeritz, Reed E.; Dudgeon, David L.; Haller, J. Alex, Jr.
Ann. Thorac. Surg., June 1984, Vol. 37, Pg. 500-504, Ch. 18-3
Affiliation: Johns Hopkins University School of Medicine
Abstract: The cardinal manifestations of Marfan's syndrome are in the ocular, skeletal, and cardiovascular systems, but disorders of the pulmonary system, including congenital lung malformations, emphysema, and spontaneous pneumothorax, have been reported in some patients. At least 24 instances of spontaneous pneumothorax have been reported. In an attempt to determine the incidence and course of this complication, a group of 249 patients with Marfan's syndrome was reviewed. Eleven were found to have had a spontaneous pneumothorax, a prevalence of 4.4%; in 7 it was bilateral or recurrent. Nine patients had underlying lung disease in the form of diffuse or apical bullae visible on chest x-ray films. Seven patients, 5 of whom had recurrent pneumothorax, required thoracotomy with resection of the portion of lung causing the air leak.
No other survey of patients with Marfan's syndrome and lung problems has been reported. However, in an unpublished thesis in which the literature between 1943 and 1969 was reviewed, 15 instances of pneumothorax occurred in about 300 patients, a prevalence of 5%. Thus spontaneous pneumothorax is about 30 times more frequent in patients with Marfan's syndrome than in the general population.
Seven patients, 5 of whom had recurrent pneumothorax, required thoracotomy and resection of the portion of the lung causing the air leak. On the other hand, some of the patients were successfully treated by chest tube on the first occasion and did not have recurrence. Both the authors' experience and the literature review suggest that if apical bullae are visible on the chest film or if there is a persistent air leak, resection and apical pleurodesis should be performed. Finally, all persons with Marfan's syndrome should be advised against contact sports (to protect both the lung and the aortic root) and scuba diving (to protect against rupture of lung bullae). Pneumothorax should be considered immediately, along with aortic dissection, in any patient with Marfan's syndrome in whom acute chest pain, with or without dyspnea, develops.
Post-Comment: Some other good articles on pleural diseases published this past year will be mentioned here. Broghamer et al. studied the association of malignancy and serosanguineous effusions and say that the bleeding is not due to the cancer, but rather to very fragile blood vessels on the pleural surface, mostly related to underlying inflammation rather than malignancy. They defined serosanguineous as more than 10,000 cells/cu mm. This was the case in 35% of the 786 effusions they studied. Half of the cancer-associated effusions were bloody, but so were one quarter of those from patients with nonneoplastic disease. So the finding of a bloody effusion tilts a little towards cancer but not much, really. Nearly half of the malignancy-associated serosanguineous effusions were cytologically negative.
Three cases of chronic pancreatic pleural effusion were reported by Dewan et al. Two of the 3 had no demonstrable pancreatic disease, and the condition responded to conservative therapy. The third patient had a pancreatic pseudocyst and an internal pancreatic fistula which was corrected only after multiple surgical procedures. The fluid in such cases usually looks thick and purulent, often brownish, but the main clue is demonstration of a sky-high amylase level, often with minimal or no elevation of serum amylase.
Pleural ascites due to liver disease was discussed by Hartz et al. The 2 cases reported had no associated abdominal ascites, presumably because the fluid got sucked up into the chest by respiratory pumping through a diaphragmatic hole or defect which was functioning as a one-way valve closing during expiration.
A very impressive radiological article from Europe said that all patients with calcified pleural thickening exceeding 2 cm in depth will have fluid hidden in the shadow, that this can be detected by ultrasound and by computed tomographic scan, and that it can cause a lot of trouble. Long-term collections of fluid or pus such as these can result in late perforation either to the outside or to the lung.
Wied et al. say that tetracycline is better than silver nitrate for pleurodesis in spontaneous pneumothorax. Davis and associates also say that tetracycline is very satisfactory for the treatment of malignant pericardial effusion.{--}R.M.D.
Acid-Fast Bacilli in Sputum Smears of Patients With Pulmonary Tuberculosis: Prevalence and Significance of Negative Smears Pretreatment and Positive Smears Posttreatment
Kim, Taik Chae; Blackman, Raymond S.; Heatwole, Kenneth M.; Kim, Taeho; Rochester, Dudley F.
Abstract: The authors undertook a retrospective study of 977 patients with culture-proved pulmonary tuberculosis to determine the frequency and significance of smear negativity before treatment and of smear positivity despite culture conversion during treatment. All patients presented with active disease. Triple drug therapy was followed by dual drug therapy after the susceptibility results were known or the sputum smear converted to negative.
One fourth of the patients were repeatedly smear-negative before treatment. The finding was related inversely to both the extent of disease and the presence of cavities. Smear-negative patients whose organisms were drug-sensitive converted most rapidly to a sputum culture-negative status during treatment. Patients with far-advanced cavitary disease had the slowest conversion rate. Those with drug-resistant organisms were slower to convert than the other patients, but all except 4 eventually had negative cultures and sputum smears. Positive smears persisted in 20% of patients during an average of 6.1 {+-} 4.0 weeks (range, 4-20 weeks). This phenomenon was related to the extent of disease (Fig 19-1) and to the use of rifampin therapy. Sputum smears eventually converted to negative during continuance of the same regimen in all instances.
Sputum smears were consistently negative in about one fourth of patients in this series. Treatment should be initiated despite negative smears if appropriate clinical indications are present. About one fifth of patients continued to have positive sputum smears after sputum culture conversion on treatment. A change in treatment is not necessary in these cases, as long as clinical and radiographic evidence of improvement persists. Only if positive smears continue for more than 14 weeks after culture conversion should a change in therapy be considered for this reason alone.
Post-Comment: This article was chosen because it documented, with large numbers, something that some of our associates don't seem to know: the sputum smear often remains positive for periods of time after the culture has converted to negative, and this per se does not require modification of therapy. The article said that positive smears had to be taken seriously only if they occurred or persisted 14 weeks after sputum culture conversion, but we know of no data to suggest that that is necessarily so. With extensive caseous disease, often there will be sporadic sputum smear (and even culture) positivity after many months of treatment. This in all probability results from the fact that caseous material is unstable and may intermittantly break off and release dormant bacilli and that these may even grow in culture without having been metabolically active in the caseum in vivo. Another variation on this theme is the appearance of a positive acid-fast bacilli smear (or culture) during the course of some other destructive pulmonary process such as cancer, lung abscess, or even lung resection. If this occurs when a patient is not recei receiving antituberculosis therapy, it probably constitutes good reason for ``chemoprophylaxis,'' since bacilli disturbed and spread in this fashion can take hold and cause progressive disease if not treated. It does not mean that the particular destructive lung process is assoc associated with active tuberculosis; merely that it may reactivate it. It is to prevent this sort of activation due to anatomical disturbance of an inactive focus that chemoprophylaxis is often advised to cover pulmonary resec resection in someone thought to have quiescent old apical tuberculosis or with evidence of it on chest radiograph (old fibrotic disease or apical small calcific Simon foci). In following the course of tuberculosis under treatment, much more importance should be attached to a biphasic course, with initial disappearance of smear posit positivity and then, after 3 or 4 months of treatment, reappearance of increasing smear positivity which is consistent rather than sporadic. This used to almost always mean the development of a high degree of microbial drug resistance. It is now almost never seen, since regimens containing isoniazid and rifampin are so effective.
While talking about the meaning of positive smears, it is perhaps worth trying to rehabilitate the procedure of aspiration of gastric contents for diagnosis of pulmonary tuberculosis in persons without sputum production. It is a conservative alternative to fiberoptic bronchoscopy, which is what seems to be done more and more in such cases. Smears of gastric contents (which must be aspirated first thing in the morning) were long thought to be unreliable because of the potential of contamination with other mycobacteria contained in water or foodstuffs. Actually, when studied carefully, most positive gastric smears containing more than a single organism will be found to be due to active tuberculosis.{--}R.M.D.
Six-Months' Isoniazid-Rifampin Therapy for Pulmonary Tuberculosis: Report of a United States Public Health Service Cooperative Trial
Snider, Dixie E., Jr.; Long, Mary W.; Cross, Floy S.; Farer, Laurence S.
Abstract: Patients with pulmonary tuberculosis from 15 centers throughout the United States were entered into a trial of 2 regimens. All were aged 18 or over and had positive sputum cultures. All received isoniazid, 300 mg, and rifampin, 600 mg, daily for 6 months. In the following 9 months, patients received either isoniazid, 300 mg, and ethambutol, 15 mg/kg, daily or matching placebos. Patients with impaired hepatic or renal function or significant visual defects were excluded from the trial.
Of 672 patients who entered the trial, 46% completed both the initial and maintenance phases.
About one fifth of the patients failed to keep appointments. Delinquency was associated with alcohol abuse, younger age, and unmarried status; these factors, however, accounted for only 6% of the observed variation in delinquency. Adverse drug effects, particularly hepatotoxicity, led to the withdrawal of 37 patients (6%) from the trial. Another 6% were withdrawn because of persistently positive cultures between 12 and 24 weeks of treatment. One percent of patients (7 cases) were considered treatment failures because of clinical progression of their disease. Three hundred seventy-nine patients completed the initial 6-month trial, and 309 completed the 9-month maintenance phase. During the maintenance phase, 3 patients receiving active treatment and 16 receiving placebo relapsed, as evidenced by two or more positive sputum cultures; the difference was significant (table). No patient had visual toxicity from ethambutol.
This study showed the inadequacy of treatment with isoniazid and rifampin alone for 6 months in the management of pulmonary tuberculosis. Failure to complete treatment was a major problem in this trial; this finding sensitized clinicians to the magnitude of the problem of noncompliance. Because the potential advantages of short-course chemotherapy remain numerous, a new trial was initiated comparing a 9-month regimen of isoniazid and rifampin with a 6-month regimen containing pyrazinamide as well for the first 2 months. Addition of this agent and closer attention to compliance might make a 6-month regimen suitable for routine use in the United States.
Post-Comment: The conclusion of this large U.S. Public Health Service study is that 6 months of a regimen containing isoniazid and rifampin but not pyrazinamide is suboptimal, as illustrated in the table. There were several other important points. (1) Noncompliance was a very large problem, so large that the authors felt that this was the major risk facing the patient. (2) There was no increase in drug-related hepatotoxicity in alcoholic patients. (3) There was no evidence of ocular toxicity at an ethambutol dosage of 15 mg/kg per day, and the authors stated that screening for this was unnecessary (see below). (4) They provided considerable data to support the position that a single positive culture or smear occurring after initial sputum conversion during the course of chemotherapy did not represent a relapse. (5) As observed in virtually every study of short-course chemotherapy, bacteriologic relapse after a suboptimal regimen, such as the 6-month one in this experience, yielded organisms with the same drug sensitivity as was present at the outset of treatment.
The notion that 15 mg of ethambutol per kg is devoid of ocular toxicity, while true in this experience and true in one very large U.S. Public Health Service study, is not absolutely true. Retrobulbar neuritis is related to dosage, the incidence, in one good study, being 18% at doses greater than 35 mg/kg, 5% at about 30 mg/kg, and 3% at 25 mg/kg. Since almost all biologic phenomena including susceptibility to drug toxicity demonstrate a bell-shaped curve of incidence, it is really not sensible to assume that there will be no cases at 15 mg/kg; and there are definite cases of irreversible blindness occurring on an ethambutol dose of 15 mg/kg. We have had personal experience with one case in which the courts decided, rightfully in our opinion, that this dose of ethambutol had caused retrobulbar neuritis and blindness in an elderly diabetic woman.
While it is generally agreed that regimens containing only isoniazid and rifampin but not pyrazinamide need to be given for 9 months in order to achieve an acceptable chemotherapeutic maximum, there is emerging consensus that regimens containing isoniazid, rifampin, and pyrazinamide, usually with ethambutol or streptomycin as a fourth drug, can be given for a total of only 6 months, and that even much shorter periods provide substantial chemotherapeutic effectiveness. Rifampin, isoniazid, and pyrazinamide all appear to be essential to these abbreviated regimens for different reasons. It has become fashionable to speak of an infecting population of tubercle bacilli as consisting of three different subpopulations of organisms: those that are extracellular and rapidly multiplying, such as in a pulmonary cavity; those that are intracellular at an acid pH and usually multiplying much more slowly; and those that are contained within the acid and highly inhibitory milieu of caseous necrosis. Isoniazid remains the keystone of treatment, active against both rapidly and slowly metabolizing organisms in and out of cells and unaffected by pH. Rifampin is probably as effective as isoniazid except in certain circumstances such as CNS infection, and it has the added advantage of being particularly effective against organisms that are metabolizing very slowly or only intermittently. In other words, what rifampin does better than isoniazid is to mop up on the so-called ``persisters,'' organisms that for reasons of location and environmental milieu are barely metabolizing and are therefore resistant to isoniazid. This mopping-up action is the reason that rifampin treatment should not be terminated before the full course of chemotherapy has been given. Pyrazinamide is particularly effective against intracellular organisms and organisms at an acid pH and is, as a matter of fact, ineffective at an extracellular pH. For whatever reason, it is clear that addition of pyrazinamide to a regimen containing isoniazid and rifampin provides more rapid sterilization and reduces relapses, especially in short-course therapy. ``These differences provide strong evidence that pyrazinamide has a capacity to kill a part of the bacillary population relatively u unaffected by the other drugs, a plausible explanation for its important sterilizing action in short-course chemotherapy.'' Unlike isoniazid and rifampin, pyrazinamide appears to exhaust its effectiveness in 2 or 3 months and can be discontinued thereafter.
The essence of short-course chemotherapy is a 2-month (or 3-month) period of daily treatment with isoniazid, rifampin, and pyrazinamide. Usually ethambutol at 25 mg/kg, or streptomycin is given during this intensive phase, but there is some doubt as to whether either really adds to isoniazid-rifampin-pyrazinamide. After 2 (or 3) months of treatment with this three-drug combination plus ethambutol or streptomycin, there is a variety of daily and less-than-daily schedules that can provide equally good results when continued for a total of 6 months of therapy; the only requirement is that both isoniazid and rifampin be included in the continuation regimen. Toxicity is no greater in these regimens than in more prolonged regimens containing isoniazid and rifampin, and, very importantly, drug resistance at the outset affects outcome very little if at all. It still remains true that, given a drug-sensitive infection, this intensive chemotherapy provides results no better than isoniazid and ethambutol for 18 months in moderate disease, or with initial reinforcement with streptomycin in advanced disease, and the cost of drug toxicity is greater with the more intensive, shorter regimen. However, the number of people who can really take drugs reliably for 18 months is a lot smaller than most physicians would guess, and there is accordingly much to recommend the wider use of ultra-short-course, multiple-drug chemotherapy. One good thing is that isoniazid-rifampin-pyrazinamide with streptomycin or ethambutol given for only a couple of months, while certainly not an acceptable regimen, will still permanently cure greater than 50% of cases, no matter how far advanced, and will cure the great majority of patients with less advanced disease, particularly those whose sputum smear and culture are negative. Short-course chemotherapy was also discussed at some length in the 1983 YEAR BOOK.
Some other tuberculosis articles will be mentioned here. Carpenter et al. record a local incidence of antituberculous drug resistance of 16.4% for isoniazid, 3.9% for ethambutol, 10.6% for rifampin, and 7.8% for streptomycin in south Texas. This was not influenced by age or ethnic group. Drug resistance appears to be a problem for everybody in that part of the country.
The fact that rifampin is a potent inducer of the hepatic P{sub-4}{sub-5}{sub-0} mixed oxidase drug detoxifying system is well known, and increased coumadin requirement, increased metabolism of opiates, degradation of birth control pills with upsetting consequences, and enhanced cortisol catabolism have all been observed. Kyriazopoulou and associates have extensively studied and reported a case of adrenal crisis in a patient with Addison's disease who was given rifampin.
Dautzenberg et al. studied 30 immunocompromised patients with tuberculosis and concluded that the tuberculosis infection was improved with conventional therapy and that modification of immunosuppressive therapy was a bad thing to do. All of their bad results were due to having changed immunosuppressive treatment in the face of tuberculosis, only to have the underlying disease get worse.
Also see an article by Banks and associates which says that results of treatment of Mycobacterium xenopi infections are not as good as previously reported; an article by Ahn et al. describing good results with 12 months' treatment of Mycobacterium kansasii infections; and an article with more about the computed tomographic scan in tuberculosis of the CNS, as was discussed in last year's YEAR BOOK.{--}R.M.D.
Clinical and Laboratory Features of Disseminated Histoplasmosis During Two Large Urban Outbreaks
Sathapatayavongs, Boonmee; Batteiger, Byron E.; Wheat, Joseph; Slama, Thomas G.; Wass, Justin L.
Medicine (Baltimore), 1983, Vol. 62, Pg. 263-270, Ch. 19-4
Affiliation: Indianapolis
Pre-comment: Histoplasmosis is a common but elusive disease. In endemic areas literally millions of cases of asymptomatic acute pulmonary histoplasmosis occur almost entirely unnoticed. Sporadic, relatively heavy inhalation exposures to histoplasma spores cause nonspecific, self-limiting, febrile respiratory illnesses, which are rarely diagnosed in the individual case and are most frequently not recognized for what they are even in ``epidemics.'' However, two huge epidemic waves of clinically significant histoplasmosis, now totaling 741 cases, have been recognized in Indianapolis since 1978. Clinical and laboratory studies of the many hospitalized, symptomatic patients with acute pulmonary histoplasmosis, and some of the complications thereof (e.g., pericarditis), were discussed in the 1982 YEAR BOOK OF MEDICINE and in the 1984 YEAR BOOK. Wheat and his associates are to be commended for recognizing, evaluating, and describing this largest recorded epidemic of histoplasmosis. It could well have gone unnoticed, as did one in a nearby city when 15 cases of chronic pulmonary histoplasmosis were recognized and traced to the bulldozing of a blackbird roost. It is virtually certain that this latter experience also produced a large number of symptomatic acute cases that went unnoticed.
In order to define the magnitude of an epidemic of histoplasmosis and to provide a base for calculating the incidence of various clinical manifestations, it is necessary to estimate accurately the total number of new infections (acute pulmonary histoplasmosis, both asymptomatic and symptomatic). Wheat and associates estimate that there were 150,000 new infections in the Indianapolis epidemic, an estimate that must be considered very rough, since it was based on serologic studies which do not detect light infections. The profile of the epidemic, however, is impressive: new cases of acute histoplasmosis (estimated), 150,000 cases; hospitalized symptomatic cases, 741; pericarditis, 45; disseminated histoplasmosis, 61; chronic pulmonary histoplasmosis, 24. It is certain that there were many more people with symptomatic illness in the community who were not hospitalized.
This past year the Indianapolis experiences with disseminated histoplasmosis and chronic pulmonary histoplasmosis were recorded, the first of which is abstracted below.{--}Robert Goodwin
Abstract: Sixty-one patients experienced 66 episodes of disseminated histoplasmosis in the course of two outbreaks occurring in Indianapolis since 1978. The mean age was 39 years. The chief predisposing factors were hematologic malignancy and immunosuppressive drug therapy. About half of the patients were immunosuppressed. All but 13 patients had underlying illness. Fever was a universal symptom, and weight loss and respiratory symptoms were features in more than half of the patients (Table 1). The chest roentgenographic findings are given in Table 2. Infiltrates progressed in all patients who died before receiving adequate treatment. Positive culture results were obtained in most instances (Table 3), but 3-4 weeks were required before the growth could be identified. The culture and histologic findings are given in Table 4.
Mortality was 23% (Table 5). Five of the 29 nonimmunosuppressed patients died, all with serious underlying conditions. All patients given at least 500 mg of amphotericin B had a good clinical response; none of 23 patients who received at least 30 mg/kg, including 12 immunosuppressed patients, relapsed. The average total dose of amphotericin B given to 28 patients who were without evidence of persistent infection after 12-52 months of follow-up was 35 mg/kg. Five episodes of meningeal involvement in 4 patients responded only to intravenously administered amphotericin B. Six patients recovered without treatment after an average of 8 weeks of illness and remained well on follow-up for more than 3 years.
Untreated disseminated histoplasmosis is nearly always fatal. Amphotericin B treatment is regularly effective when a total dose of at least 500 mg is given; relapses occur if the total dose is less than 30 mg/kg. Ketoconazole appears to be effective in the treatment of nonimmunosuppressed patients, but a controlled trial comparing its effect with that of amphotericin B is needed.
Post-Comment: Two questions raised by this report merit comment. The first has to do with the definition of disseminated histoplasmosis, and the second, which is a related problem, concerns the distribution of observed cases among the susceptible elements of the population.
Human macrophages and those of other mammals are entirely susceptible (under normal circumstances of native immunity) to parasitization by the yeast phase of Histoplasma capsulatum derived from inhaled spores, and they remain entirely susceptible until specific cell-mediated immunity develops. At that time the infection is almost always entirely and permanently arrested. However, there is a period of probably 2 or 3 weeks before specific cell-mediated immunity develops in which the infected macrophages have unimpaired access to the lymphatic and blood channels, through which they spread throughout the body. This is entirely analogous to the situation in a primary tuberculous infection which, prior to the establishment of tuberculin hypersensitivity (cellular immunity), spreads throughout the body and establishes foci in both pulmonary and extrapulmonary sites. This preallergic bacillemic phase of tuberculous infection is believed to take place in every infection, and of course can in no way be regarded as miliary (disseminated) tuberculosis, although it may, especially in infants, evolve into such.
During this period when infected macrophages traffic in uninhibited fashion throughout the body, they lodge in various tissues and, with the onset of cellular immunity, tuberculoid granulomas are formed. Healed granulomas of this sort are almost the rule at autopsy in the livers and spleens of persons who have lived long periods in areas endemic for histoplasmosis. Moreover, when heavy aerosol infections occur in the previously exposed and immune or partially immune host, there will also be a period of usually nonprogressive blood stream dissemination. This is also entirely analogous to the situation in tuberculosis. Active advanced pulmonary tuberculosis will commonly be associated with the development of hepatic granulomas and eye-ground tubercles without the development of progressive blood stream spread, the key word here being progressive. In like fashion, persons who have some degree of immunity but inhale a large inoculum of histoplasmosis spores will have a period of blood stream dissemination which in the vast majority of instances will not be progressive. In some persons, for reasons of intercurrent immunosuppression or age or some unknown circumstance compromising cellular immune mechanisms, this event may in fact lead to progressive disseminated infection of the macrophage system.
In our opinion, the pathologic hallmark of established disseminated histoplasmosis is the finding of persistently infected macrophages without localizing tissue reaction well after a normal period of illness (at least 3 and perhaps as many as 6 to 8 weeks). The criteria for diagnosis in the Indianapolis study included a febrile illness of more than 3 weeks' duration or the finding of organisms or granulomas in extrapulmonary tissues. We feel that this sort of definition will certainly include many nonprogressive sorts of infection. One wonders how many of the 61 reported cases of disseminated histoplasmosis were in fact cases of fairly severe but self-limiting acute pulmonary histoplasmosis with positive blood or bone marrow cultures early in the illness, or cases with intrinsically nonprogressive organized granulomas in an extrapulmonary biopsy. Our interpretation that many cases of intrinsically nonprogressive disease were included in their definition of disseminated histoplasmosis would account for the discrepancy in the article between the statement that there were 6 untreated cases that showed spontaneous recovery, and the statement (certainly true) that untreated disseminated histoplasmosis is nearly always fatal.
Previous studies of mostly endemic cases extending well back into the past have always demonstrated that there are three groups in the population particularly susceptible to disseminated histoplasmosis: (1) infants (immature immunity) comprising roughly 50% of all cases; (2) immunosuppressed individuals (perhaps 25%); and (3) the remainder of otherwise normal individuals in whom there is a transient and at present not understood defect in immunity permitting progressive disseminated disease (although we believe the argument for immunosuppression by intercurrent viral illness is strong and is strengthened further by the analogous situation in the history of the acquired immunodeficiency syndrome). Studies on which these proportions are based antedate the recent sharp increase in the number of immunosuppressed individuals in the past few years. None the less, one wonders why more infants did not appear in the Indianapolis study. There were only 3 patients younger than 5 years of age, and 2 of these were immunosuppressed. Perhaps the study did not fully encompass the pediatric wards or hospitals in the community, or perhaps the very young were in some way shielded from exposure. However another explanation is that the actual number of disseminated cases as we would define them was far smaller than reported. In any event, it is important to remember that infancy does represent an important risk factor for dissemination of histoplasmosis.
The other large Indianapolis study to appear this year described the experience with chronic pulmonary histoplasmosis. The difference we would hold with this study is the implication that not all patients had emphysema. Just as the immune defect of the very young and the immunosuppressed renders their macrophages susceptible to chronic and progressive infection, so do the abnormal pulmonary air spaces of emphysema serve as the focus for chronic pulmonary infection. In spite of normal systemic immunity, abnormal pulmonary air spaces allow for colonization and persisting infection sequestered from normal tissues. (Again there are analogies to tuberculosis. It is well known that when cavitation occurs in pulmonary tuberculosis, the situation greatly favors massive multiplication of the infecting organisms, the local milieu of the cavity being ideally suited to the propagation of the infection. But whereas the intrinsically much more pathogenic tubercle bacillus is able to cause sufficient necrosis to make its own cavity, infection of preexisting bullous emphysema by the actually quite effete yeast forms accounts for ``cavitary'' histoplasmosis.)
In the Indianapolis experience the risk factors for cavitary histoplasmosis were listed as three: clinical or radiographic diagnosis of chronic bronchopulmonary disease, age, and white male. We would contend that these are all actually the same, as emphysema is mostly a disease of older white men. Also clinical evidence and radiographic evidence of emphysema are notoriously late manifestations indicating already advanced disease. Whereas the presence of emphysema is by no means always apparent at the onset of chronic pulmonary histoplasmosis, in our own experience we were never able to identify a single case in which clear indications of emphysema did not develop if we were able to follow the patient long enough. In those in whom it was not apparent at age 40, it was clearly so by age 50 to 60.
The incidence of chronic pulmonary histoplasmosis in the Indianapolis epidemic can be roughly calculated by dividing the 24 cases recognized as related to the epidemic into the estimated 150,000 people infected. The round number of one case in 6,000 infected individuals is fairly close to the figure of one case of chronic pulmonary histoplasmosis in 2,000 infected white males estimated in our study of 238 gradually accumulated endemic-type cases.{--}Robert Goodwin
Correlation of Two In Vitro Tests With Clinical Response to Immunosuppressive Therapy in 54 Patients With Severe Aplastic Anemia
Torok-Storb, Beverly; Doney, Kristine; Brown, Susan L.; Prentice, Ross L.
Blood, February 1984, Vol. 63, Pg. 349-355, Ch. 21-3
Affiliation: Univ. of Washington
Abstract: The techniques of detection of committed hematopoietic precursors in vitro as colony-forming units (CFU) have been applied to the study of aplastic anemia (AA). Colony-forming units of any kind (CFU-C, BFU-E, CFU-E) are either not detectable or are found in significantly reduced numbers in the bone marrow and blood of AA patients. Evidence for the concept of ``immune-mediated'' aplasia was first provided by AA patients who recovered autologous bone marrow function after immunosuppression.
Over a 2-year period, 54 patients who met the criteria for severe AA entered the study. All patients were treated with horse antithymocyte globulin (ATG), 16 mg/kg four times daily for 10 days, and oxymetholone, 3 mg/kg for 3 months; 30 of the patients also received an HLA-haploidentical marrow infusion. In the first test, peripheral blood mononuclear cells were collected from each patient before ATG therapy and then treated with a panel of monoclonal antibodies and complement. The treated monocytes were assayed for erythroid burst-forming units (EBF-E). This test was designed to determine whether removing various subpopulations of peripheral blood monocytes would increase the number of detectable BFU-E. In the second test, peripheral blood was collected within 48 hours after completion of ATG therapy, and cells were immediately assayed for BFU-E. Data from both tests were analyzed to determine whether the in vitro results correlated with patient response to therapy. The minimum follow-up time was 18 months.
Binary logistic regression analyses indicate that a modest correlation exists between test 1 in vitro results and clinical response. However, the strength of this association appears to decrease as the interval between diagnosis and treatment increases. In contrast, test 2 had a very significant correlation with clinical outcome among patients diagnosed more than 1 month prior to treatment, but this association was not apparent among those treated within 1 month of diagnosis.
There is increasing evidence that lymphocytes and monocytes play a role in regulating normal hematopoiesis. This suggests that the potential role of lymphocytes in mediating AA extends beyond the classical one of autoreactivity to include the possibility of defective regulation. It appears from the present data that test 1 and test 2 are detecting different phenomena. Growth of BFU-E after monoclonal antibody and normal rabbit serum complement treatment (test 1) may reflect an immune disease mechanism, and the correlation with clinical response decreases as disease duration increases because the ability of the patient to respond also decreases with time after diagnosis. This could explain the large numbers of false-positive in vitro scores obtained with test 1. Test 2 may be detecting the actual response to ATG, which would explain the significant correlation with clinical response. This, however, is an important observation, as in most patients, test 2 scores were obtained 4-6 weeks before any indication of peripheral blood recovery. The failure of test 2 to predict response accurately in patients less than 1 month after diagnosis might be due to the fact that patients in the early postdiagnosis period, regardless of etiology of their disease, may have more circulating BFU-E than patients with longer disease history. In such cases, BFU-E growth would be detected after ATG therapy, not necessarily because of a response but because the therapy decreases the absolute number of lymphocytes and thereby increases the concentration of BFU-E in peripheral blood monocytes to detectable levels.
Post-Comment: This is the best of a series of articles describing attempts at predicting the response to treatment of patients with aplastic anemia. Unfortunately, as shown here, none of these tests has sufficient predictive value to be effectively utilized in making clinical judgments.{--}M.J.C.
Prenatal Diagnosis of Sickle Cell Anemia in the First Trimester of Pregnancy
N. Engl. J. Med., Oct. 6, 1983, Vol. 309, Pg. 831-833, Ch. 21-6
Affiliation: Paris
Abstract: The development of restriction-enzyme methods of diagnosing sickle cell anemia prenatally has led to attempts to make the diagnosis earlier in pregnancy. The authors evaluated chorionic biopsy for the prenatal diagnosis of sickle cell anemia by restriction-endonuclease analysis of fetal DNA. Chorionic biopsy specimens were obtained from 30 women at 7-10 weeks of gestation, before elective abortion. Five subsequent women requested testing at 6-8 weeks of gestation. Two had had a child with sickle cell anemia, and 1 woman with the disease had already terminated a pregnancy because of an affected fetus, diagnosed in utero by fetal blood analysis. The restriction enzyme CvnI was used to analyze trophoblast DNA.
Enough chorionic villi were obtained to yield 5-40 {mu}g of DNA. Testing led to a decision to terminate 2 of the test pregnancies. In the 3 other cases a normal genotype was demonstrated. One normal infant has been born, and another pregnancy is proceeding normally. In a case in which fetal death occurred 3 weeks after sampling, placental abnormalities found on histologic study were compatible with a chromosomal aberration.
Chorionic biopsy is a suitable means of diagnosing sickle cell disease before 10 weeks of gestation by restriction-endonuclease analysis of fetal DNA. If further experience shows chorionic biopsy to be sufficiently safe for mother and fetus, the procedure may make prenatal diagnosis acceptable to more couples at risk of serious genetic disorders.
Post-Comment: The use of DNA analysis of chorion samples to make prenatal diagnosis during the first trimester of pregnancy was first proposed in an article in The Lancet in 1981. This technique has now been used in several countries around the world, although it has not been applied extensively in the United States. The use of this procedure appears to be logical; however, the incidence of obstetrical complications associated with chorion biopsy is still uncertain.{--}M.J.C.
Recurrence of Ph'-Positive Leukemia in Donor Cells After Marrow Transplantation for Chronic Granulocytic Leukemia
Marmont, Alberto; Frassoni, Francesco; Bacigalupo, Andrea; Podesta, Marina; Piaggio, Giovanna; Van Lint, Maria T.; Caimo, Attilio; de Filippi, Sandor
N. Engl. J. Med., Apr. 5, 1984, Vol. 310, Pg. 903-906, Ch. 23-4
Affiliation: Genoa
Abstract: Recurrence of leukemia in donor cells after bone marrow transplantation for acute leukemia is a rare form of relapse. The authors report a case of Ph'-positive leukemic relapse in donor cells after allogeneic marrow transplantation in a patient with Ph'-positive chronic granulocytic leukemia in a second chronic phase.
Woman, 21, was given busulfan for Ph'-positive chronic granulocytic leukemia. A lymphoid blast crisis developed, and treatment with vincristine and prednisone induced a second chronic phase. A bone marrow transplant from an HLA-identical brother was then carried out after preparation with cyclophosphamide and total body irradiation. Prompt engraftment occurred. Cyclosporine was used to prevent graft-vs.-host disease. Mild chronic skin involvement was treated with low-dose corticosteroids. More than 80% lymphoid blasts were present in a bone marrow aspirate when the patient was seen with facial palsy on day 170. Vincristine and prednisone led to a partial remission, but further chemotherapy was necessary a month later, and the patient died with infection. The Ph' chromosome was demonstrated in all metaphases in the patient's marrow cells obtained during the posttransplantation relapse. The Y body was clearly seen in the lymphoid blasts. The donor is well.
Seven cases of recurrent leukemia in donor cells after marrow transplantation have been reported. All the patients had advanced leukemia, and they usually had had numerous relapses before transplantation. That recurrent leukemia in the present case was Ph' positive suggests that a new leukemic cell clone may arise at relapse, with the same chromosomal abnormality that was detected in the original leukemia. The case also confirms that the 9;22 translocation is involved in the development of chronic granulocytic leukemia but is not necessarily causative. The agent of the original leukemia may have initiated leukemogenesis in the donor cells, and the appearance of the 9;22 translocation may be an expression of this event.
Post-Comment: This is an extremely interesting and surprising observation. The explanation of why leukemia should recur in donor cells is not immediately apparent. The usual hypotheses put forth suggest either that a virus is involved or that cancer genes (oncogenes) from the malignant cells of the patient somehow get into the proliferating bone marrow cells of the donor. Strong evidence does not exist for either of these hypotheses.{--}M.J.C.
Clinical Staging for Cutaneous T-Cell Lymphoma
Lamberg, Stanford I.; Green, Sylvan B.; Byar, David P.; Block, Jerome B.; Clendenning, William E.; Douglass, Margaret C.; Epstein, Ervin H., Jr.; Fuks, Zvi Y.; Golitz, Loren E.; Lorincz, Allan L.; McBurney, Elizabeth I.; Michel, Beno; Roenigk, Henry H., Jr.; Van Scott, Eugene J.; Vonderheid, Eric C.
Ann. Intern. Med., February 1984, Vol. 100, Pg. 187-192, Ch. 23-6
Abstract: The authors developed a staging system for cutaneous T-cell lymphomas on the basis of the findings in 376 patients seen at 21 centers between 1974 and 1977. Preliminary analyses identified the extent of skin involvement and the number of sites of clinically enlarged lymph nodes as significant prognostic variables. These variables were used to classify 340 cases into four clinical stages. The diagram used to distinguish between ``limited'' and ``widespread'' skin disease is shown in Figure 23-1.
A total of 124 patients had died at the time of analysis. The rest of the patients were followed for a median of 57 months. The clinical staging system proved to be useful in identifying patients with different survival experiences. The extent of skin involvement was clearly a discriminating prognostic variable. The number of enlarged node sites was also a significant prognostic variable, although somewhat less than the extent of skin involvement. An alternative grouping suggested by fitting a survival model to the data did not significantly improve predictions of survival.
Staging schemes based only on skin involvement and lymph node status are recommended for general use because the necessary information is readily at hand from physical examination alone. The present system for staging cutaneous T-cell lymphomas should help in comparing different studies, planning clinical trials, and increasing the numbers of patients available for treatment protocols. Treatment effects constitute a bias that is difficult to overcome, but no treatment presently in use has been shown to influence long-term survival.
Post-Comment: Beginning as long ago as 1862 attempts have been made to discriminate among various groups of patients with cutaneous T-cell lymphomas for the purposes of clinical classification and establishing prognosis. In 1972 Epstein and his colleagues reported a retrospective analysis of 144 patients with mycosis fungoides. Cutaneous tumors, skin ulcers, and clinically enlarged lymph nodes are the most important variables.{--}M.J.C.
Acute Myelogenous Leukemia: Maintenance Chemotherapy After Early Consolidation Treatment Does Not Prolong Survival
Sauter, C.; Berchtold, W.; Fopp, M.; Gratwohl, A.; Imbach, P.; Maurice, P.; Tschopp, L.; von Fliedner, V.; Cavalli, F.
Abstract: High complete remission rates have been obtained in adults with acute myelogenous leukemia (AML), but most patients die within 3 years of diagnosis. Prolongation of survival using conventional maintenance regimens has not been proved in a randomized study. The value of maintenance chemotherapy was assessed in 162 untreated patients with AML, seen between 1977 and 1982, in whom an attempt was made to induce remission. The 96 females and 66 males had a median age of 43 years; the median observation time was 44 months. Induction therapy included administration of cytarabine, daunorubicin, and vincristine. The 74 patients who remained in remission after early consolidation therapy, given for 3-5 months, were assigned to receive maintenance chemotherapy every 8 weeks for 2 years or observation only. Alternate cycles of cytarabine and 6-thioguanine, and cytarabine, prednisone, and vincristine were used. Doses were lowered if indicated by blood cell depression.
Patients who failed to respond to induction therapy had a median survival of 3 months, and those who experienced early relapse had a survival of 7.5 months. The median survival for patients completing the consolidation phase was 30 months. The prognosis was significantly related only to age. Patients older than 40 years of age had a median survival of 4 years, compared with 1$frac{1/2} years for younger patients. There were no significant differences in duration of remission or survival in patients given maintenance therapy and those who were merely observed (Fig 23-3).
Conventional maintenance chemotherapy, given after early consolidation therapy, does not appear to prolong remission of AML or patient survival. A second cycle of therapy usually is indicated for those who do not respond to the first induction cycle. Patients who are younger than age 65 years should be managed as aggressively as the youngest ones are. Patients less than 40 years old should receive a bone marrow transplant in the first year of illness if a suitable donor is available.
Post-Comment: I agree with the conclusions reached by these authors.{--}M.J.C.
Prognostic Importance of Cytogenetic Abnormalities in Patients With Chronic Lymphocytic Leukemia
Han, Tin; Ozer, H.; Sadamori, N.; Emrich, L.; Gomez, G. A.; Henderson, E. S.; Bloom, M. L.; Sandberg, A. A.
N. Engl. J. Med., Feb. 2, 1984, Vol. 310, Pg. 288-292, Ch. 23-13
Abstract: Chronic lymphocytic leukemia (CLL) is recognized as having a variable prognosis, but its staging has depended exclusively on anatomical sites of involvement and the presence or absence of anemia and thrombocytopenia. The recent availability of cytogenetic analysis of malignant B lymphocytes led the authors to examine the karyotypic abnormalities in CLL and to correlate them with clinical stage, progression of disease, and survival.
Karyotypic analyses of leukemic B cells in 114 patients with CLL of B-cell origin were performed. Cytogenetic evaluation data were available for 53 patients (37 men and 16 women), aged 48-88 years, in whom adequate metaphases were obtained for analyses. Duration of disease ranged from 1 month to 12 years and was not less than 5 years in most of the patients. Of these patients, 24 had not been treated, and 29 had previously received chlorambucil, prednisone (or both), cyclophosphamide, or irradiation. According to the clinical staging method of Rai et al., 10 patients were in stage 0, 2 in stage I, 18 in stage II, and 23 in stage III or IV.
Of the 53 patients, 21 (40%) had abnormal karyotypes and 32 (60%) had normal karyotypes. Abnormal karyotypes were found in 33% of all metaphases analyzed, with ranges from 6% to 100% in individual patients. Trisomy 12 was found as the sole abnormality in 8 patients, trisomy 12 plus other changes was found in 5, and changes without trisomy 12 were found in 8. There was no correlation between clonal chromosomal changes and duration of disease, treatment status, or leukemic phenotype, or with sex or age. Abnormal karyotypes were found in 2 (17%) of 12 patients with stage 0 or I disease, 3 (17%) of 18 with stage II disease, and 16 (17%) of 23 with stage III or IV disease. These observations clearly indicate a significantly higher prevalence of abnormal karyotypes in patients with advanced disease than in those with early disease. The follow-up time ranged from 12 to 36 months. Each of 10 patients who died of CLL had disease staged as either III or IV at the time of cytogenetic study. In these 10 cases there were normal karyotypes in 3, trisomy 12 in 1, trisomy 12 plus other changes in 3, and other karyotypic changes without trisomy 12 in 3. The median survival time for patients with abnormal karyotypes was 24 months, whereas for patients with normal karyotypes it exceeded 37 months.
In the present study, although the prevalence of clonal chromosomal abnormalities failed to correlate with sex, age, duration of disease, treatment status, or leukemic cell type, the presence of these clonal abnormalities strongly correlated with clinical stage and survival time. Among patients with trisomy 12 as the sole abnormality, 84% survived at least 10 years after diagnosis; this corresponds closely with the survival times of patients with normal karyotypes. Only 27% of the patients with trisomy 12 in combination with other chromosomal changes survived 8 years from diagnosis. Among patients with chromosomal changes other than trisomy 12, 58% survived at least 10 years from diagnosis. Also, it appears that trisomy 12 is the most frequent chromosomal change in CLL. The results additionally indicate that any other chromosomal abnormality could be a poor prognosis in patients with CLL.
Post-Comment: Chronic lymphocytic leukemia is associated with the proliferation of monoclonal small lymphocytes, usually of B-cell origin. The disease has a variable course with survival varying from a few years to more than a decade. Prognosis in CLL correlates with several clinical and laboratory abnormalities.
Until recently, cytogenetic abnormalities in CLL were not well defined, primarily because of the very low spontaneous mitotic index of the leukemic cells. The availability of B-cell mitogens such as the Epstein-Barr virus permit this very interesting study. It is now well recognized that an extra chromosome 12 is the most frequent karyotype abnormality in patients with CLL.{--}M.J.C.
A Clonal Complete Remission in a Patient With Acute Nonlymphocytic Leukemia Originating in a Multipotent Stem Cell
Jacobson, Robert J.; Temple, Matthew J.; Singer, Jack W.; Raskind, Wendy; Powell, Jerry; Fialkow, Philip J.
N. Engl. J. Med., June 1984, Vol. 310, Pg. 1513-1517, Ch. 25-1
Abstract: In some cases of acute nonlymphocytic leukemia, remission may be partly or totally derived from a clonal preleukemic precursor, and chemotherapy may destroy only a blastic subclone and restore a clonal preleukemic state. In the present case, assays of glucose-6-phosphate dehydrogenase (G6PD) and cytogenetic studies showed the disease to be clonal and karyotypically abnormal at presentation. After a chemotherapy-induced complete clinical remission, the bone marrow progenitors and blood cells were normal morphologically and had a normal karyotype, but still were derived from a single clone.
Woman, 67, presented after 6 weeks of fatigue, exertional dyspnea, and dizziness, with no history of a blood disorder or exposure to leukemogens. The bone marrow findings were consistent with acute nonlymphocytic leukemia. A complete remission was produced with cytarabine and daunomycin therapy, and maintained with cytarabine and thioguanine. Only type B G6PD activity was observed before treatment. Type A enzyme was present during remission, but subsequently type B enzyme predominated. The chromosomal abnormalities observed at presentation were not present during remission. The Philadelphia chromosome was not detected.
This case demonstrates the heterogeneous nature of the differentiative expression in the stem cells giving rise to acute nonlymphocytic leukemia and suggests that remissions also can be heterogeneous. A clonal preleukemic state can be restored by chemotherapy in some cases. The present patient remained in remission for 9 months, but the clinical implications of clonal remission are unknown.
Post-Comment: Remission of acute myeloid leukemia is thought to be the result of destruction of leukemic cells and repopulation of the bone marrow by normal hematopoietic stem cells. The evidence to support this notion is as follows: (1) the bone marrow karyotype generally returns to normal during remission induced by chemotherapy; (2) in 5 patients with acute nonlymphocytic leukemia who were heterozygous for glucose-6-phosphate dehydrogenase only a single enzyme type was found in the leukemic blast cells during relapse, but a normal enzyme pattern was found in mature leukocytes and their precursors during remission.
These findings indicate that mature blood cells are generally derived from stem cells that are not part of the leukemic clone. This report, however, suggests that in occasional patients a preleukemic clone may generate the cells of the remission state.{--}M.J.C.
Suppression of High-Grade Ventricular Ectopic Activity by Antiarrhythmic Drug Treatment as a Marker for Survival in Patients With Chronic Coronary Artery Disease
Hoffman, A.; Schutz, E.; White, R.; Follath, F.; Burckhardt, D.
Abstract: Fifty consecutive patients with chronic coronary artery disease (CAD) were studied retrospectively during a mean observation period of 16 months to investigate the relationship between the suppression of complex ventricular premature beats (VPBs) and subsequent survival. All patients had persistent frequent (more than 30/hour) multiform (Lown grade III) or complex (Lown grades IVa, IVb, or V) VPBs. One to three single drugs and one or two combined regimens were tested with a total of 124 drug trials done. Treatment was effective (``responders'') when all repetitive ventricular ectopic activity (VEA) (Lown grade IVa) was eliminated and when frequent multiform VPBs were reduced to fewer than 30 unifocal VPBs/hour during Holter monitoring.
Thirty-nine patients were considered ``responders,'' and 11 patients were not adequately suppressed (``nonresponders''). There were no significant differences in age and in presence of congestive heart failure in the two groups. There were 3 deaths (1 sudden) among the 39 ``responders.'' Five deaths (3 sudden) occurred among the 11 ``nonresponders.'' Significant side effects occurred in four patients. The cumulative probability of survival at 12 months was 0.93 for ``responders'' and 0.64 for ``nonresponders.''
This study shows that patients with chronic CAD in whom frequent multiform and complex VPBs can be suppressed by antiarrhythmic drugs have better prognosis than ``nonresponders'' as to total and sudden cardiac mortality. The association of improved survival with suppression of high-grade VEA may be explained by the treatment with antiarrhythmic drugs, which prevented sudden death, and by differences in severity of underlying disease. The authors recommend that antiarrhythmic drug treatment should be tried in patients with chronic CAD presenting with frequent and multiform or complex ventricular ectopy.
Post-Comment: Believe it or not, despite the enormous amount of work in this field, it is still not clear whether or not life is prolonged by antiarrhythmic therapy in patients with coronary artery disease and frequent ventricular premature contractions and/or high-grade ventricular ectopic activity! The results of this study are consistent with the notion that treatment is desirable, but one could argue that the responders might have had a better survival even without treatment, i.e., that the response to treatment somehow distinguished patients with an intrinsic favorable prognosis from those with an intrinsic poor prognosis. The outcome of trials comparing treatment with active drug vs. placebo in patients with arrhythmias are anxiously awaited. However, until the results of such studies are available, I believe that the authors' conclusion that antiarrhythmic drug therapy should be tried in patients with chronic coronary artery disease and frequent, multiform, or complex ventricular ectopy followed by long-term treatment of those patients who respond is a very reasonable one.
In a related study, Benditt et al. evaluated the usefulness of serial provocative electropharmacologic testing in predicting the efficacy of prophylactic antiarrhythmic treatment in patients without acute myocardial infarction resuscitated from sudden cardiac arrest. In those patients in whom a regimen could be found in the laboratory that prevented inducible life-threatening tachyarrhythmia, the incidence of sudden death was extremely low, suggesting that provocative electropharmacologic testing is useful in predicting the response to therapy in survivors of sudden cardiac arrest.{--}E.B.
Western Washington Randomized Trial of Intracoronary Streptokinase in Acute Myocardial Infarction
Kennedy, J. Ward; Ritchie, James L.; Davis, Kathryn B.; Fritz, James K.
N. Engl. J. Med., Dec. 15, 1983, Vol. 309, Pg. 1477-1482, Ch. 29-2
Affiliation: Seattle
Abstract: The authors studied data on 213 male and 37 female patients in a multicenter, community-based trial of the efficacy of intracoronary streptokinase thrombolysis in acute myocardial infarction; 134 were randomly assigned to streptokinase therapy and 116 served as controls. All patients (mean age, 56.3 years) underwent left ventricular angiography and coronary arteriography before the random assignment. Thirty-day mortality was assessed. The average dose of streptokinase was 286,000 {+-} 77,800 units given during a period of 72 {+-} 24 minutes.
Of 108 patients with complete occlusions, 68% had reperfusion, whereas 81% of the 16 patients with subtotal occlusions had reperfusion with streptokinase. Mortality during the first 30 days after acute myocardial infarction was 7.2% for all patients. Five of the 134 streptokinase-treated patients died (3.7%), and 13 of the 116 control patients died (11.2%), a significant difference (Fig 29-1). Among hypotensive patients, 1 of 16 streptokinase-treated patients died, whereas 4 of 23 controls died. Mortality was related to the location of infarction. After anterior infarction, mortality was 18.9% in controls compared with 7.9% in the treatment group. Among patients with inferior infarction, mortality was 4.8% and 0% in control and treated patients, respectively. Length of hospital stay was similar for both groups. No difference in radionuclide ejection fraction was found between the two groups. Complications of the angiographic procedure and the infusion of streptokinase occurred in 14.2% of the treated patients and 7.8% of the controls. Bleeding occurred more often in the former group, 5.2% vs. 0.9%, respectively.
The authors conclude that the data showed that intracoronary streptokinase can reduce early mortality after myocardial infarction. Investigations are under way to determine whether this early reduction in mortality is sustained.
Post-Comment: It is now clear that the intracoronary administration of thrombolytic agents in the early hours of acute myocardial infarction can achieve revascularization of an obstructed artery in approximately 80% of patients. This important subject was previously discussed in the 1984 YEAR BOOK. The paper abstracted above is important in that it represents the first randomized, controlled trial that has demonstrated that survival can be improved by this mode of therapy. However, there are several caveats: (1) previous randomized studies did not demonstrate an improvement in survival {--}although these previous trials were smaller than the Western Washington trial; (2) although mortality was improved in the short-term in the Western Washington trial, the difference became of borderline significance with the passage of time; (3) the improvement in survival did not appear to be associated with any reduction of infarct size or improvement of left ventricular function, although other laboratories have shown that left ventricular function can be improved by successful recanalization using intracoronary streptokinase; (4) both laboratory and clinical investigations have shown that recovery of myocardial function following reperfusion may take days or even weeks; (5) coronary thrombolysis has also been demonstrated to improve regional myocardial metabolism as reflected in an improvement of {sup-1}{sup-1}C palmitate uptake by previously ischemic myocardium.
It has been shown experimentally that reperfused infarcts are usually hemorrhagic, while clinical studies suggest that the distribution of the hemorrhagic areas following streptokinase treatment are probably confined to muscle that was already necrotic at the time of reperfusion. Postmortem observations on patients have demonstrated that reperfused infarcts show predominantly contraction-band necrosis, whereas myocardial infarcts without reperfusion show coagulation necrosis of the muscle fibers.
There is now considerable interest in the use of fibrin-specific thrombolytic agents which, unlike streptokinase and urokinase, do not produce a systemic lytic state and therefore are less likely to produce serious bleeding complications. Major attention has focused on two fibrin-specific thrombolytic agents: tissue plasminogen activator (tPA) (which can be manufactured by recombinant techniques) and acylated streptokinase-plasminogen complex. These two substances are apparently capable of producing effective intracoronary thrombolysis with only minor effects on peripheral coagulation variables in the majority of patients. Therefore, they should prove to be safer than the standard fibrinolytic agents{--}streptokinase and urokinase{--}but this point remains to be established.{--}E.B.
Rethrombosis After Reperfusion With Streptokinase: Importance of Geometry of Residual Lesions
Harrison, David G.; Ferguson, David W.; Collins, Steve M.; Skorton, David J.; Ericksen, Elizabeth E.; Kioschos, J. Michael; Marcus, Melvin L.; White, Carl W.
Circulation, May 1984, Vol. 69, Pg. 991-999, Ch. 29-4
Affiliation: Univ. of Iowa
Abstract: Emergency coronary bypass grafting or percutaneous angioplasty may be considered in patients with residual high-grade coronary stenosis after reperfusion with thrombolysis, because such stenoses are thought to predispose to rethrombosis and, possibly, to infarct extension. The present study of 24 patients with successful reperfusion by intracoronary streptokinase therapy was done to determine whether rethrombosis is related to the luminal extent of the residual stenosis. Angiography was repeated a mean of 10 days after initial reperfusion. Angiography results were analyzed by both quantitative coronary angiography and computer-based videodensitometry to obtain independent estimates of the cross-sectional area of a lesion with minimal observer variability.
One of 17 patients with patent reperfused vessels at recatheterization had unstable angina. Five of the other 7 patients had rethrombosis documented by coronary angiography; 4 of them had recurrent chest pain at least 24 hours after reperfusion in association with ST-segment elevation. Two patients died, and autopsy documented rethrombosis in 1 of them. The residual minimal cross-sectional area of rethrombosed vessels was significantly less than that of lesions remaining patent (Fig 29-3). A cross-sectional area of 0.4 sq mm provided the best distinction. The minimal cross-sectional area increased by 116% on average within 8-14 days after reperfusion in patients whose vessels remained patent. Integrated optical density also was helpful in identifying those at a high risk of rethrombosis.
Rethrombosis of a coronary artery after successful thrombolysis is related in part to the size of the residual lesion. Vessels with a residual stenotic cross-sectional area of less than 0.4 sq mm are at high risk of rethrombosis. These patients may be candidates for emergency transluminal angioplasty or coronary bypass grafting. In other instances, management may be based on the appearances at recatheterization 8-14 days after thrombolysis.
Post-Comment: There has been considerable debate concerning the management of patients with acute myocardial infarction following successful reperfusion with thrombolytic agents. One view is that the condition of these patients is inherently unstable, that they have a lesion that has already shown itself to be likely to thrombose, and that they therefore should be treated aggressively, i.e., by early angioplasty or coronary artery bypass surgery. Other investigators have found that the majority of their patients have done well and have not rethrombosed when treated conservatively. This interesting study provides a possible explanation. After thrombolysis there appeared to be two groups of patients: those in whom the residual stenotic cross-sectional area was markedly reduced (less than 0.4 sq mm) who were at high risk of rethrombosis and those with vessels having larger cross-sectional areas who exhibited a stable course and actually demonstrated an increase in cross-sectional area during the subsequent 8-14 days. If these results can be confirmed they would suggest that if intravenous thrombolytic therapy is administered in a community hospital the patient should be transferred to an institution in which arteriography can be performed promptly and the residual obstruction can be assessed. If it is very severe, then aggressive therapy is indicated; if not, the patient can be followed by treatment with antiplatelet agents, aspirin, and Persantine.{--}E.B.
Is Transluminal Coronary Angioplasty Mandatory After Successful Thrombolysis?
Serruys, P. W.; Wijns, W.; Van Den Brand, M.; Ribeiro, V.; Fioretti, P.; Simoons, M. L.; Kooijman, C. J.; Reiber, J. H. C.; Hugenholtz, P. G.
Affiliation: Erasmus Center, Rotterdam, The Netherlands
Abstract: It has been argued that percutaneous transluminal coronary angioplasty is mandatory to prevent reocclusion after thrombolysis in patients with acute myocardial infarction. Using a computer-based coronary angiography analysis system, the authors quantitatively assessed selected coronary artery segments in 105 patients catheterized in the acute phase of myocardial infarction. Sixty-four of the 105 patients were recanalized with 250,000 units of streptokinase; recanalization could not be achieved in 25 patients. The infarct-related vessel was patent at the time of the procedure in 16 patients. Transluminal angioplasty was performed immediately afterwards in 18 of the 78 patients who had a patent infarct-related vessel after the recanalization procedure. The contours of the arterial segments were detected automatically, and the reference diameter, minimal obstruction diameter, length of the lesions, and percentage diameter stenosis were averaged from multiple views (Fig 29-4).
Follow-up angiograms of 57 patients who had not undergone percutaneous transluminal coronary angioplasty 2 weeks after thrombolysis suggested a reocclusion rate of 17%. Eight additional patients had a reinfarction in the same myocardial region at a mean of 8.3 months; 4 lesions that had remained occluded at the acute stage were still occluded at the chronic stage. The percentage diameter stenosis for all patients was less than 50% in 31% of the patients and greater than or equal to 70% in only 19%. Reocclusion and reinfarction occurred more frequently in patients with a stenosis greater than 58% in diameter after recanalization. The 18 patients who underwent transluminal angioplasty showed a decrease in the average percentage diameter stenosis from 59% to 30% (P < .0001) and an increase in the minimal obstruction diameter from 1.3 to 2.2 mm. The reduction in diameter stenosis was associated with a significant decrease in mean pressure gradient from 41 to 8 mm Hg. At a mean follow-up of 4 months (range, 10 days to 11 months), angiography demonstrated a patent dilated vessel in all but 1 of the 15 patients who had undergone angioplasty.
Although percutaneous transluminal coronary angioplasty does not appear to be mandatory at the acute stage of myocardial infarction in the majority of patients, the procedure can be performed in one sitting and seems to prevent reocclusion in patients selected on the basis of quantitative angiographic criteria.
Post-Comment: The conclusions of this study from The Netherlands are consistent with the observations in the paper by Harrison et al. from Iowa (article 29-4). Serruys et al. have shown clearly that intracoronary streptokinase infusion can be followed immediately by percutaneous transluminal angioplasty (PCTA). While this therapeutic strategy is possible for those patients with acute myocardial infarction who reach a center that can carry out coronary arteriography within a few (preferably less than 4) hours after the onset of symptoms, clearly it is not applicable to the majority of patients with this condition who are treated in community hospitals.
Yasuno et al. studied the effects of PTCA following intracoronary thrombolysis with urokinase. They made several interesting observations: (1) Intracoronary urokinase is an effective coronary thrombolytic agent and can be used as an alternative to streptokinase; it is particularly valuable in patients who have recently had a streptococcal infection or who have previously received streptokinase and have developed antibodies to streptokinase. (2) The efficacy of urokinase is dose dependent. (3) Chronic patency of the coronary artery occurred in 60% of patients with successful PTCA following thrombolysis. (4) In patients with chronic arterial patency there was an improvement in left ventricular ejection fraction and regional wall motion, improvement that did not occur in patients with unsuccessful PTCA or late reocclusion. (5) Continuous infusion with glucose-insulin-potassium after PTCA may enhance the beneficial effect of reperfusion.{--}E.B.
Early Bypass Grafting Following Intracoronary Thrombolysis With Streptokinase
Sterling, Rosalyn P.; Walker, William E.; Weiland, Anne P.; Freund, Gregory C.; Fuentes, Francisco; Smalling, Richard W.; Lance Gould, K.
J. Thorac. Cardiovasc. Surg., April 1984, Vol. 87, Pg. 487-492, Ch. 29-6
Affiliation: Univ. of Texas, Houston
Abstract: Fibrinolytic therapy may only partially restore blood flow to an infarct-related coronary artery, and the fixed stenosis can become a site for new thrombus formation and recurrent tissue damage. Forty-one patients (32 men) with acute myocardial infarction, aged 29-80 years, underwent intracoronary streptokinase infusion and myocardial revascularization 3-10 days later. All were seen within 18 hours of the onset of chest pain. Hemodynamic instability did not contraindicate the procedure. Revascularization was performed in patients with unstable angina, left main or triple-vessel disease, or severe proximal stenosis of a large coronary artery.
Thirty-four patients had total occlusion of the infarct-related vessel at initial catheterization, and 4 failed to gain reperfusion of the distal portion of the coronary artery. Nearly 80% of all patients had multivessel disease. There was 1 operative death. None of the 40 patients followed for 6-24 months has had recurrent myocardial infarction. One late death was due to sepsis. Bypass grafting did not improve ventricular performance beyond that achieved after fibrinolytic therapy in the group as a whole, but significant improvement was evident in many patients who had reperfusion of a totally occluded coronary artery. Ejection fraction values are given in the table.
Patients with severe coronary heart disease and acute evolving myocardial injury can be effectively managed by intracoronary thrombolysis with streptokinase, followed by coronary bypass grafting where significant anatomical stenosis is present. Sustained improvement in left ventricular function can be effected by this approach. The risk of bypass grafting 3-10 days after thrombolysis is no greater than the risk of elective operation performed at a time remote from acute infarction.
Post-Comment: There are many alternative therapies following thrombolysis. The most conservative approach is simply to observe the patient and carry out mechanical revascularization (percutaneous transluminal angioplasty [PTCA] or coronary artery bypass grafting [CABG]) if there is evidence of recurrent ischemia. A second approach is to carry out PTCA if the patient has a severe residual stenotic lesion even in the absence of recurrent ischemia. Perhaps the most radical approach was used by these authors, who performed early CABG routinely in a series of 41 consecutive patients who had evidence of persistent unstable angina, left main or triple vessel disease, or severe proximal stenosis of a large coronary artery.
The Houston group has reported in this and in other papers that intracoronary streptokinase may be effective when begun as late as 18 hours after the onset of symptoms of myocardial infarction. These findings differ from those of other investigators who believe that effectiveness of reperfusion is limited to a much shorter time window{--}no more than 4 hours and perhaps even less. Regardless of the time interval between the onset of pain and reperfusion, many patients are left with life-threatening lesions either in the vessel that was recanalized or in other vessels, and early surgical treatment can apparently be carried out at a low risk. This approach, i.e., immediate thrombolysis (whether intravenous or intracoronary) followed by coronary arteriography and either CABG or PTCA, depending on the anatomy, may turn out to be the most effective ``one-two'' punch in the treatment of evolving myocardial infarction.
A number of groups have now also reported on emergency CABG in patients who had unsuccessful coronary thrombolysis. For example, Phillips et al. believe that in patients with acute myocardial infarction and single-vessel disease, thrombolytic therapy should be attempted, followed by PTCA if a significant residual lesion persists or by CABG if PTCA is unsuccessful. They recommend that patients with significant multivessel disease in the presence of an acute myocardial infarction should have emergency CABG. However, the resources{--}both human and institutional{--}required for the latter approach are enormous, and although surgical therapy early in the course of acute myocardial infarction may prove to be useful in patients with severe residual anatomic disease, for logistic and economic reasons, it may not be feasible except in a very small fraction of patients.{--}E.B.
Enhancement of Salvage of Reperfused Myocardium by Early {beta}-Adrenergic Blockade (Timolol)
Hammerman, Haim; Kloner, Robert A.; Briggs, Lance L.; Braunwald, Eugene
J. Am. Coll. Cardiol., June 1984, Vol. 3, Pg. 1438-1443, Ch. 29-7
Affiliation: Harvard Med. School
Abstract: The time after onset of infarction during which thrombolysis can salvage ischemic myocardium has been limited to a few hours. Treatment with {beta}-blockers can reduce the severity of ischemic injury after coronary occlusion and might extend the time of effective reperfusion. The left anterior descending coronary artery was occluded in dogs, after which the area at risk and the regional myocardial blood flow were assessed using labeled microspheres. The animals received either saline or timolol maleate 15 minutes after occlusion, the latter drug being given until the heart rate or systolic arterial pressure decreased by about 20%. The mean infused dose was 0.85 mg/kg. Coronary occlusion was maintained for 3 hours and was followed by 3 hours of reperfusion.
The amount of left ventricular mass at risk was similar in both groups of dogs, but the mass of necrosis was significantly less in the timolol-treated animals. Blood flow in the ischemic zone was comparable in the two groups 5 minutes after coronary occlusion. Left ventricular end-diastolic pressure tended to be higher in treated animals, but not significantly so. No significant impairment of regional function was evident on echocardiographic study in the timolol-treated animals.
In this study, myocardial salvage was enhanced by {beta}-blockade when treatment was started shortly after coronary occlusion and followed by reperfusion. Although clinical situations may differ, {beta}-blockade could increase the amount of tissue salvaged by reperfusion shortly after coronary occlusion, or prolong the time after occlusion during which reperfusion can salvage severely ischemic tissue. A prophylactic approach might be evaluated in patients at high risk of acute infarction, especially those with previous infarction.
Post-Comment: While the role of {beta}-blockers in reducing infarct size may be controversial (see comments following article 29-8), {beta}-blockers probably can at the very least delay the death of ischemic myocardial cells. Therefore, the prophylactic administration of {beta}-blockers to patients at high risk of coronary occlusion, such as those with unstable angina pectoris or previous myocardial infarction, or the administration of {beta}-blockers very early in the course of myocardial infarction prior to or concurrent with thrombolytic therapy, may enhance the effectiveness of reperfusion. This approach deserves evaluation in clinical trials with {beta}-blockers as well as other cardioprotective drugs, such as verapamil.{--}E.B.
Reduction of Infarct Size With the Early Use of Timolol in Acute Myocardial Infarction
International Collaborative Study Group
N. Engl. J. Med., Jan. 5, 1984, Vol. 310, Pg. 9-15, Ch. 29-8
Abstract: {beta}-Blockers minimize ischemic damage in animals, but patient studies have failed conclusively to show their potential benefit. A double-blind, placebo-controlled, multicenter evaluation of timolol in patients seen within 4 hours of the onset of symptoms of myocardial infarction is reported. Patients aged 21-70 years without bradycardia, hypotension, severe left ventricular failure, or heart block were entered into the trial. Patients with previous infarction were excluded. Patients with chest pain but no ECG abnormality and those with diagnostic ECG changes were stratified to receive timolol in a bolus of 1 mg or normal saline. Timolol was then infused at a rate of 0.6 mg/hour for 24 hours, or an equal volume of saline was given. Subsequent oral treatment was with 10 mg of timolol twice daily or a matching placebo.
The 73 patients assigned to receive timolol and the 71 given placebo were well matched clinically and demographically. Heart rate, blood pressure, and pain scores were reduced in timolol-treated patients compared with the placebo group. The need for analgesia was significantly reduced in the timolol group. Adverse effects were comparably frequent in the timolol and placebo groups. Release of creatine kinase was reduced in the timolol-treated patients (Fig 29-5). The maximal QRS-vector change for a given initial ST-vector magnitude was significantly reduced in the timolol group.
The results support the use of timolol early in the course of myocardial infarction. It seems to be helpful in both the early and the long-term management of patients with infarction. Timolol presumably reduces infarct size through a combination of blood pressure reduction and heart rate reduction, with oxygen-sparing benefit to the myocardium, and a direct metabolic effect.
Post-Comment: Two major trials of the effects of {beta}-blockers on infarct size appeared last year. The one that provided the positive results is abstracted above. The other was MILIS, multicenter investigation sponsored by the National Institutes of Health (NIH) that involved 269 patients randomized to placebo or propranolol, 0.1 mg/kg, intravenously, followed by oral therapy (N. Engl. J. Med. 311:218, l984). The primary end point, infarct size estimated from plasma MB creatine kinase activity, was virtually identical in the two groups. How do we account for the difference in results between these two studies? I believe it was related to the timing of the administration of the drugs. The patients studied by the International Collaborative Study Group had all been admitted to the hospital within 4 hours of the onset of symptoms, whereas the NIH multicenter study allowed treatment as long as 18 hours after the onset of symptoms, and only 2% of the patients were treated within 4 hours. It is likely that myocardial necrosis proceeds relatively rapidly following coronary occlusion and that there is relatively little tissue left to salvage after 4 hours of severe ischemia. Hence, any favorable intervention, whether it is the use of a {beta}-blocker or reperfusion induced by thrombolysis is likely to be effective only when it is applied very soon after the onset of the symptoms.
Geary et al. observed in the baboon that even pretreatment with propranolol failed to reduce the ultimate size of a myocardial infarct after 2 hours of coronary occlusion followed by reperfusion. This contrasts with several observations in the dog that have shown an infarct-sparing effect of {beta}-blockers, and it teaches us an important lesson about salvage of the ischemic myocardium. The baboon heart, in contrast to the canine heart, has very sparse collaterals, and therefore it would appear that following coronary occlusion a {beta}-blocker is capable of myocardial salvage in the presence but not the absence of collateral circulation. Perhaps a small quantity of residual blood flow to the ischemic zone either through a subtotally occluded vessel or through collaterals will enhance the ability of a variety of interventions to salvage myocardium.{--}E.B.
The Influence of Changes in Lipid Values Induced by Cholestyramine and Diet on Progression of Coronary Artery Disease: Results of the NHLBI Type II Coronary Intervention Study
Levy, R. I.; Brensike, J. F.; Epstein, S. E.; Kelsey, S. F.; Passamani, E. R.; Richardson, J. M.; Loh, I. K.; Stone, N. J.; Aldrich, R. F.; Battaglini, J. W.; Moriarty, D. J.; Fisher, M. L.; Friedman, L.; Friedewald, W.; Detre, K. M.
Circulation, February 1984, Vol. 69, Pg. 325-337, Ch. 29-11
Affiliation: Natl. Inst. of Health
Abstract: It has been hypothesized that a lowering of low-density lipoprotein cholesterol (LDL{sub-c}) levels can arrest or retard the progression of coronary artery disease (CAD). The National Heart, Lung, and Blood Institute Type II Coronary Intervention Study, a randomized, double-blind, placebo-controlled study, evaluated the efficacy of cholestyramine-induced reduction in cholesterol levels on the progression of CAD in 143 patients with LDL{sub-c} concentrations above the 95th percentile. Seventy-one received cholestyramine plus diet and 72 received placebo plus diet (controls). The patients were followed at monthly intervals over a 5-year period. The progression of CAD was evaluated by angiography.
Patients treated with cholestyramine showed a significant decrease in total cholesterol (TC) and LDL{sub-c} and increases in high-density lipoprotein cholesterol (HDL{sub-c}), as well as increases in HDL{sub-c}/TC and HDL{sub-c}/LDL{sub-c} ratios. Changes in HDL{sub-c} were attributable to increases in HDL{sub-2}{sub-A} and HDL{sub-2}{sub-B}. Analysis of the relationship between the progression of CAD and lipid changes independently of a specific treatment group revealed a significant inverse relationship between the progression of CAD at 5 years and both the increase in HDL{sub-c} and the decrease in LDL{sub-c}. The best predictors of CAD change were changes in the HDL{sub-c}/TC and HDL{sub-c}/LDL{sub-c} ratios. Although evaluation of these relationships independently of a specific treatment group was not intended in the initial study design, these trends were seen in both placebo-treated patients and those treated with cholestyramine. In addition, multivariate analysis showed that the effect of cholestyramine treatment on the progression of CAD could be eliminated by adding changes in the HDL{sub-c}/TC ratio to the regression model.
The results support the hypothesis that the progression of CAD can be prevented or delayed by increases in HDL{sub-c} and decreases in TC or LDL{sub-c}.
Post-Comment: In a similar study by Nikkila et al., the LDL{sub-c} was reduced by 19% while HDL{sub-c} rose by an average of 10% in the treatment group. Coronary arteriography showed less progression in the treated patients and cardiac survival was improved over a 7-year period.
The importance of HDL{sub-c} suggested by these trials is also supported by the work of Herbert et al., who found that 5 trained runners with a greatly expanded aerobic exercise capacity had HDL{sub-c} levels of 65 mg/dl compared with 41 mg/dl in controls. In the runners, the biologic half-life of HDL was much longer than it was in sedentary men, and this may result from augmented lipid transfer to HDL by lipoprotein lipase or diminished HDL clearance by hepatic lipase. The effects of physical exercise on lipoproteins were also studied by Swedish investigators who found that healthy middle-aged men who carried out mild regular physical exercise had a 26% increase in levels of HDL{sub-2}, the subfraction that presumably protects against ischemic heart disease; this was paralleled by increases in levels of 6-keto-prostaglandin F{sub-1}{sub-alpha}, suggesting that HDL{sub-2} contributes to the metabolism of prostacyclin, whereas HDL{sub-3}, which presumably does not exert a protective effect, was not affected by exercise.{--}E.B.
Treatment of Hyperlipidemia Retards Progression of Symptomatic Femoral Atherosclerosis: A Randomized Controlled Trial
Duffield, R. G. M.; Lewis, B.; Miller, N. E.; Jamieson, C. W.; Brunt, J. N. H.; Colchester, A. C. F.
Abstract: The authors began a controlled trial of treatment for hyperlipidemia in patients with symptomatic femoral atherosclerosis in 1977. Twenty-four patients with stable intermittent claudication due to femoropopliteal disease for 6 months or longer were assigned to either usual care or management with dietary advice and cholestyramine, nicotinic acid, or clofibrate, depending on the lipoprotein phenotype. All patients initially had a total serum cholesterol level greater than 6.5 mmole/L, a triglyceride level exceeding 1.8 mmole/L while fasting, or both. Biplanar transfemoral arteriography was performed at entry to the study and after 18-20 months of observation.
The plasma cholesterol and triglyceride levels in treated patients were 19% and 37% lower, respectively, than in the group receiving the usual care. The low-density lipoprotein (LDL) cholesterol was 24% lower in the intervention group, and the high-density lipoprotein (HDL) cholesterol was 41% higher. No significant group differences in blood pressure or body weight were noted during the study. The proportion of arterial segments showing progression of atherosclerotic disease was reduced by 60% in the study group. The rate of increase in plaque area, assessed visually, was one third of that in the group with usual care (table). The LDL cholesterol correlated with the rate of disease progression in both groups, but only a weak negative correlation was noted between HDL cholesterol and disease progression.
Effective hypolipidemic treatment appears to favorably influence the course of symptomatic peripheral atherosclerosis in hyperlipoproteinemic patients. Further studies are needed to determine the effect of such treatment on early, asymptomatic lesions and on atherosclerosis at other sites. No adverse effects of treatment were evident in the present study.
Post-Comment: While the experimental design was somewhat different, the results of this study on femoral atherosclerosis closely parallel those of the NIH study on coronary atherosclerosis (article 29-11). The message is loud and clear: A reduction of circulating lipids is associated with a decreased rate of progression of atherosclerosis. The lipid hypothesis got a tremendous boost in 1984 from the results of the Lipid Research Clinic's Coronary Primary Prevention Trial, in which the effects of cholesterol lowering, using the bile-acid sequestrant cholestyramine, were studied. In the treated patients a reduction of LDL cholesterol, which was 12.6% greater than that in the control group was associated with a 24% reduction in coronary heart disease deaths and a 19% reduction in nonfatal myocardial infarction, as well as a reduction in the development of angina. Further, it was observed that the incidence of coronary heart disease in patients sustaining a fall of 35% in LDL cholesterol was half of that of those who remained at pretreatment levels. This landmark study makes a strong case for efforts to reduce LDL cholesterol in subjects without apparent disease in whom it is elevated.
While cigarette smoking has long been known to be an important risk factor for the development of coronary heart disease, its deleterious mechanism is not clear. Brischetto et al. have demonstrated that cigarette smokers have significantly lower HDL and higher VLDL and plasma triglyceride levels than exsmokers and nonsmokers, suggesting that cigarette smoking might operate at least in part through this mechanism. One additional recently identified potentially important risk factor is the fibrinogen concentration, which appears to be statistically significant for the development of myocardial infarction and particularly of stroke.{--}E.B.
Risk Factors for Sudden Death After Acute Myocardial Infarction: Two-Year Follow-Up
Mukharji, Jhulan; Rude, Robert E.; Poole, W. Kenneth; Gustafson, Nancy; Thomas, Lewis J., Jr.; Strauss, H. William; Jaffe, Allan S.; Muller, James E.; Roberts, Robert; Raabe, Daniel S., Jr.; Croft, Charles H.; Passamani, Eugene; Braunwald, Eugene; Willerson, James T.; MILIS Study Group
Am. J. Cardiol., July 1, 1984, Vol. 54, Pg. 31-36, Ch. 29-13
Affiliation: Dallas
Abstract: This study identifies the various risk factors for sudden coronary death using laboratory and clinical data obtained from 533 patients who survived 10 days after myocardial infarction (MI) and followed up for as long as 24 months in the Multicenter Investigation of the Limitation of Infarct Size (MILIS). In particular, the relation between sudden death and the presence of left ventricular (LV) dysfunction and complex ventricular premature beats (VPBs) was examined.
Frequent VPBs (more than 10 beats per hour) on ambulatory ECG monitoring and LV dysfunction (radionuclide LV ejection fraction, greater than 0.40) were the most significant independent markers of risk for subsequent sudden death. Seventy-nine percent of sudden deaths occurred within the first 7 months of follow-up. The incidence of sudden death was 18% in patients with both frequent VPBs and LV dysfunction (P < .001). With the use of the four-way life-table analysis, it was determined that patients with both LV dysfunction and repetitive VPBs had a 17% incidence of sudden death, accounting for 34% of all sudden deaths in a group that represented 11% of the total patient population. When, 6 months after MI, 280 survivors were reclassified with regard to the presence or absence of frequent VPBs and LV dysfunction, these risk factors could not be associated with sudden death during a further follow-up period as long as 18 months; the overall incidence of sudden cardiac death was low (1.4%).
The presence of frequent VPBs in association with LV dysfunction early after MI identifies patients at high risk for sudden death over the next 7 months. However, since 79% of all sudden deaths occurred within 7 months of the index MI, this raises the question of how long empirical treatment of postinfarction arrhythmia is beneficial. The ability to define a small group of patients at high risk for sudden death may be a key for future studies that test the efficacy of antiarrhythmic drugs in preventing sudden death after MI.
Post-Comment: There are two major lessons from this study: (1) a very simple measure of electrical instability, i.e., frequent ventricular premature contractions on ambulatory ECG monitoring was as useful, perhaps even more useful, than more complex methods of arrhythmia analysis in identifying patients at risk of sudden death after myocardial infarction; (2) both left ventricular dysfunction, i.e., an ejection fraction of less than 40%, as determined from a radionuclide angiogram at rest 10 days following infarction and more than 10 VPBs per hour are independent risk factors for sudden death. Almost 80% of all sudden deaths occurred within 7 months after the index myocardial infarction and assessment of patients 6 months after the infarct for these risk factors was no longer useful in establishing their later prognosis.{--}E.B.
Prognostic Importance of a Clinical Profile and Exercise Test in Medically Treated Patients With Coronary Artery Disease
Weiner, Donald A.; Ryan, Thomas J.; McCabe, Carolyn H.; Chaitman, Benard R.; Sheffield, L. Thomas; Ferguson, James C.; Fisher, Lloyd D.; Tristani, Felix
J. Am. Coll. Cardiol., March 1984, Vol. 3, Pg. 772-779, Ch. 29-15
Affiliation: University Hosp., Boston
Abstract: To determine factors predictive of mortality in medically treated patients with symptomatic coronary artery disease, 30 variables were analyzed in 4,083 patients from the National Collaborative Artery Surgery Study (CASS).
Two hundred twelve patients died during the follow-up period; 82% had cardiac-related causes of death or died suddenly. Multivariate regression analysis showed that the most important clinical variables predictive of survival were cardiac enlargement and congestive heart failure scores. Stratification of the latter variable gave 4-year survival rates ranging from 90% for patients with no or mild congestive heart failure to 18% for patients with severe heart failure. Analysis of the clinical and exercise variables together showed that the final exercise stage achieved and the ST-segment response to exercise were the most important exercise determinants of survival. The probability of survival at 5 years ranged from 72% for those patients showing 1 mm or greater ST-segment depression who were able to exercise to only stage I, to 95% for patients with less than 1 mm of ST depression who were able to exercise to stage 3 or beyond.
When all clinical, exercise, and catheterization variables were analyzed together, left ventricular (LV) score was found to be the most important varible related to survival (P < .0001), followed by the number of stenosed coronary vessels (P < .003). In addition, gender, use of diuretics, digitalis therapy, peak heart rate, and final exercise stage each yielded prognostic information when combined with the catheterization data. In a subgroup of 572 patients with three-vessel disease and relatively preserved LV function (LV scores, 5-9), the overall 4-year survival rate was 82%, but ranged from 53% for patients able to exercise only to stage 1/2 to 100% for those able to attain stage 5 or beyond (P < .004) (Fig 29-6).
The results indicate that clinical and exercise variables primarily related to the functional status of the left ventricle are helpful in assessing the prognosis of patients with defined coronary pathoanatomy.
Post-Comment: The importance of the determination of prognosis of patients surviving acute myocardial infarction has already been discussed (articles 29-10 and 29-13). There is also considerable interest in determining the prognosis of patients with chronic coronary artery disease so that proper therapy, particularly surgical therapy, can be selected. Although in this study observations made from the angiogram (left ventricular contraction pattern and the number of diseased coronary vessels) were of principal importance, substantial information can be obtained noninvasively from the exercise stress test; the ability to exercise into stage 3 of the Bruce protocol without the development of ST-segment depression is a useful sign of good prognosis.{--}E.B.
Imaging by Nuclear Magnetic Resonance in Patients With Chronic Ischemic Heart Disease
Higgins, Charles B.; Lanzer, Peter; Stark, David; Botvinick, Elias; Schiller, Nelson B.; Crooks, Lawrence; Kaufman, Leon; Lipton, Martin J.
Circulation, March 1984, Vol. 69, Pg. 523-531, Ch. 29-17
Affiliation: Univ. of California, San Francisco
Abstract: Nuclear magnetic resonance (NMR) imaging provides good soft tissue contrast and requires no contrast medium, making it of potential use in assessing patients with ischemic heart disease. The authors performed gated NMR imaging in 8 normal subjects and 10 patients thought to have previously had acute myocardial infarction. Spin-echo sequencing was employed. An ECG gating technique was used in all patients and in 5 of the normal subjects.
In normal subjects, the cardiac chambers and great vessels were sharply defined on the gated NMR images. The endocardial interface and the interface between the myocardium and the pericardial fat or lung were clearly defined. Axial images demonstrated papillary muscles, parts of the atrioventricular valves, and parts of the coronary arteries. Regions of previous infarction appeared as areas of wall thinning on NMR images. The abnormality correlated with the left ventricular angiographic or sector-scan echocardiographic findings. Patients with aneurysms exhibited areas of extreme wall thinning and bulging of segments of the left ventricle. In 2 patients, left ventricular thrombi were demonstrated as structures of medium signal intensity projecting into the left ventricular chamber.
Gated NMR imaging clearly demonstrates anatomical abnormalities of the left ventricle in patients with remote myocardial infarction. Contrast injection is not necessary to delineate the blood-tissue interfaces of the cardiovascular system. It is not clear whether myocardial ischemia without infarction is detectable by this means, and the role of NMR imaging in evaluating ischemic heart disease remains to be established. The method holds potential for assessing the response of patients with ischemic heart disease to various therapeutic interventions.
Post-Comment: The group at the University of California, San Francisco, led by Dr. Higgins has also demonstrated the use of magnetic resonance imaging (MRI) in evaluating patients with acute myocardial infarction (article 29-9). The present paper extends these observations to patients with chronic coronary artery disease. With the use of a relatively brief imaging time, 5-12 minutes, the myocardial wall-blood interface could be defined sharply, and a variety of changes in patients with chronic ischemic heart disease, including postinfarction wall thinning, aneurysm formation, and mural thrombi were demonstrated. This sophisticated noninvasive technique exhibits great promise for the evaluation of patients with chronic coronary disease and for establishing prognosis and thereby helping to develop a strategy for the long-term management of such patients.
Barter et al. have demonstrated the value of computed tomography (CT) in patients with ischemic heart disease. Which of these two imaging methods, CT or MRI, will prove to be superior has yet to be determined. Computed tomographic imaging has been found to be useful in detecting graft patency following coronary bypass surgery in the early postoperative period. Gated CT is superior to fluoroscopy or cineradiography in detecting coronary artery calcification and in allowing precise noninvasive measurement of the diameter of the pulmonary arteries and thereby estimating pulmonary arterial pressure noninvasively.{--}E.B.
Myocardial Ischemia During Daily Life in Patients With Stable Angina: Its Relation to Symptoms and Heart Rate Changes
Deanfield, J. E.; Selwyn, A. P.; Chierchia, S.; Maseri, A.; Ribeiro, P.; Krikler, S.; Morgan, M.
Abstract: Objective measurement of transient myocardial ischemia during everyday activities could provide a more complete understanding of the disease and its underlying mechanisms. The authors used ambulatory ST-segment monitoring to document episodes of transient ST-segment changes during unrestricted daily life in 30 patients with chronic stable angina pectoris and positive results in exercise tests. All patients were monitored for 4 consecutive days, and long-term variability was assessed by repeated 48-hour monitoring and exercise testing in 20 patients over an 18-month period.
During 446 days of recording, there were 1,934 episodes of rectilinear or down-sloping ST depression (911, 1 mm; 638, 2 mm; 385, more than 3 mm), of which only 470 (24%) were accompanied by angina. Heart rates at the onset of ST depression and of angina were 126 {+-} 15 and 132 {+-} 18 beats per minute during the initial exercise test. Chest pain occurred without significant ST-segment change on 310 occasions, 120 times in 1 patient with a peptic ulcer as well as angina. The proportion of episodes of ST depression of 3 mm or more accompanied by pain was significantly higher than that of episodes of less than 2 mm (P < .01). However, 63% of the episodes of 3 mm or greater ST depression were asymptomatic, and 70% of all episodes accompanied by angina were associated with an ST depression of 1 or 2 mm. The duration of episodes of ST depression, both with and without angina, ranged from 1 to 90 minutes (mean duration of painful episodes, 20.2 {+-} 18 minutes vs. 15.3 {+-} 16 minutes for painless episodes). Heart rate at the onset of episodes of ST depression did not differ significantly between episodes with and those without pain. However, on average, heart rate at the onset of episodes of ST depression was significantly lower than at the onset of ST depression during exercise testing (P < .01). Of 19 patients exercised until the onset of angina, all developed ST depression and serial tomograms showed evidence of reversible regional myocardial ischemia. Tomographic abnormalities that developed during painless ST depression or in the absence of ST change occurred in the same segments of myocardium in which defects subsequently developed during exercise (Fig 29-7). In the 20 patients who underwent serial exercise testing and ambulatory monitoring over 18 months, there was considerable long-term variability in the number of episodes of ST depression during ambulatory monitoring, with periods of both improvement and exacerbation.
In this series, the patients showed frequent, variable, and often asymptomatic electrocardiographic evidence of ischemia. This variability of ST depression was consistently underestimated by the occurrence of symptoms and was not reflected by exercise testing. As an increase in heart rate before myocardial ischemia was not common, a transient impairment in coronary blood supply may be at least as important as an excessive increase in demand in the genesis of ischemia during daily life.
Post-Comment: This important paper teaches us a great deal about coronary artery disease. The first major lesson is that angina pectoris, which has commonly been considered to be the hallmark of transient myocardial ischemia, seems to be just the tip of the iceberg, since more than three quarters of the episodes of ST-segment depression in the patients studied by Deanfield et al. were painless, i.e., they were not accompanied by angina. It is important to note that this study was not carried out in patients with an elevated anginal threshold but rather in ``run of the mill'' patients with chronic stable angina. Therefore, when a patient tells us that he averages two episodes of angina per day, it is likely that he has many more (perhaps an average of eight) episodes of significant transient myocardial ischemia daily. The second major lesson is that we should reconsider our understanding of the mechanism of the ischemia in patients with chronic stable angina with positive exercise tests. We always thought that ischemia was was brought about by increased myocardial oxygen demand, resulting in part from an increase in heart rate, in the face of a fixed supply; but when these patients were studied in daily life it was observed that both symptomatic and asymptomatic episodes of ischemia, as reflected in ST-segment depressions, were in fact, not preceded or accompanied by tachycardia. This finding suggests that ischemia was not caused primarily by an increase in myocardial oxygen demand but rather by a reduction in myocardial oxygen supply due to coronary vasoconstriction. This interpretation was confirmed by perfusion scans using a short-acting positron emitting radioisotope.
While it has been clear that coronary spasm is responsible for attacks of angina in patients with variant (``Prinzmetal's'') angina, these observations by Deanfield et al. imply a much broader role for coronary vasoconstriction.
Perhaps in our evaluation of patients with chronic stable angina we should determine not only the frequency of episodes of painful ischemia, i.e., angina, but also the number of episodes of painless ischemia. Perhaps we should also determine the cause of the ischemia, regardless of whether it is associated with pain. Is it due to an increase in demand, as reflected in an increase in heart rate, or does it appear to be due to a reduction in supply, i.e., without an increase in heart rate? In the former case, therapy might focus on {beta}-blockers (to reduce oxygen demand), whereas in the latter, coronary vasodilators (to increase oxygen supply) should be used.{--}E.B.
Cigarette Smoking and the Treatment of Angina With Propranolol, Atenolol, and Nifedipine
Deanfield, John; Wright, Christine; Krikler, Shirley; Ribeiro, Paulo; Fox, Kim
N. Engl. J. Med., Apr. 12, 1984, Vol. 310, Pg. 951-954, Ch. 29-20
Affiliation: Hammersmith Hosp., London
Abstract: As cigarette smoking stimulates an increase in heart rate, myocardial oxygen requirements are increased. To determine the clinical importance of cigarette smoking in the management of angina pectoris, the authors studied 6 male and 4 female cigarette smokers with angina who received nifedipine (60 mg/day), propranolol (240 mg/day), atenolol (100 mg/day), and placebo, each for 1 week. The 4-week double-blind protocol was repeated with the same randomly determined order of drugs during a nonsmoking period.
Averaged over the smoking and nonsmoking periods, all three drugs showed a significant effect compared with placebo. The patients showed a reduction in the frequency of angina, workload before ST-segment depression and chest pain, area and severity of ST-segment depression, and both the number of episodes of ST-segment depression and their duration on ambulatory monitoring. Atenolol produced a significantly greater reduction than nifedipine in the frequency of angina and the severity of ST-segment depression during exercise.
Stopping smoking significantly increased the effects of the drugs on heart rate at rest and during exercise (P < .01), the frequency of angina (P < .05), workload to ST-segment depression (P < .01), the area and severity of ST depression (P < .01). These effects were always most pronounced with nifedipine. Blood levels of propranolol were significantly higher when the patients stopped smoking (P < .05), whereas blood levels of nifedipine and atenolol were unchanged.
The findings indicate that smoking has direct and adverse effects on the heart and interferes with the efficacy of all three antianginal drugs, especially nifedipine.
Post-Comment: Cigarette smoking is deleterious to patients with heart disease in that it is a serious independent risk factor, but it also lowers high-density lipoprotein cholesterol, as well as the threshold for angina, and as shown very clearly in the paper abstracted above, it interferes quite strikingly with the effects of common antianginal drugs.{--}E.B.
Combined Carotid and Coronary Operations: When Are They Necessary?
Jones, Ellis L.; Craver, Joe M.; Michalik, Richard A.; Murphy, Douglas A.; Guyton, Robert A.; Bone, David K.; Hatcher, Charles R.; Reichwald, Norman A.
J. Thorac. Cardiovasc. Surg., January 1984, Vol. 87, Pg. 7-16, Ch. 29-24
Affiliation: Emory Univ.
Abstract: The study population comprised three groups of patients operated on since 1973. Group 1 (132 patients; 73% men; mean age, 62 years) had combined carotid and coronary operations; they were compared with 167 patients without evidence of extracranial carotid vascular disease who had elective coronary artery bypass (CAB). Group 2 comprised 51 patients who had a stroke after isolated CAB; they were compared with 52 nonstroke patients who were operated on, so as to identify factors that increase the likelihood of developing stroke as a complication of elective CAB. Group 3 consisted of 169 patients having asymptomatic cervical bruit, high-risk neurologic symptoms, or prior carotid endarterectomy, and undergoing CAB alone. This group was retrospectively studied to determine the extent of carotid vascular obstruction and frequency of perioperative stroke.
Hypertension was significantly more prevalent in group 1 than in the elective CAB control group (67% vs. 47%), whereas there was no difference in incidence of unstable angina, or history of congestive heart failure or diabetes. Asymptomatic bruit was the reason for combining carotid and coronary operations in most patients. Significant unilateral or bilateral obstructive disease (75% or greater reduction in cross-sectional area) was present in 89% of patients. Incidence of left main coronary artery obstruction (equal to or greater than 50% reduction in cross-sectional area) was 15% in the combined carotid group and 10% in patients having CAB alone.
Hospital mortality and perioperative stroke rate in group 1 were 4 (3.0%) of 132 and 2 (1.6%) of 126, respectively. These rates were not significantly different from those of the control group having CAB alone. The rehospitalization rate for group 1 patients was 2.3%, not significantly greater than for patients without extracranial cerebrovascular obstruction undergoing CAB alone. The overall incidence of postoperative stroke in 5,676 patients having CAB alone was 0.9% (51 patients).
Patients in group 3, those with asymptomatic cervical bruit, history of stroke or transient ischemic attacks, or carotid endarterectomy prior to elective CAB, had perioperative stroke at a rate of 3.3% (2 of 60), 8.6% (6 of 70), and 5.1% (2 of 39), respectively.
The overwhelming majority of strokes after CAB are embolic in origin and not related to flow states through the extracranial carotid vascular circulation.
It is suggested that simultaneous carotid and coronary operations should be performed in patients with bilateral carotid disease and symptomatic carotid vascular disease associated with unstable angina, left main obstruction, or diffuse multivessel disease. Staged procedures are suggested for patients with stable angina and symptomatic carotid lesions and for difficult carotid revascularization procedures. Coronary artery bypass alone may be performed for most patients with asymptomatic cervical bruit, moderate or mild carotid artery obstruction, and unstable angina associated with prior stroke, although in the latter case postoperative risk of neurologic injury may be increased. An algorithm for preoperative evaluation of patients with evidence of carotid or coronary disease is shown in Figure 29-10.
Post-Comment: Coronary artery disease is usually a manifestation of a generalized disease involving vascular beds in addition to the coronary. The coincidence of extracranial carotid artery disease and coronary disease is high, and the management of these patients with involvement of both vascular beds poses a difficult problem. Fortunately there have been substantial developments in the noninvasive assessment of carotid artery disease. Digital subtraction angiography and duplex ultrasonography (a combination of high resolution ultrasonography with pulsed Doppler and spectral analysis) are extremely effective in assessing atherosclerosis of the carotid bifurcation. Less widely used, but also useful clinically is ocular pneumoplethysmography, which simultaneously measures pressures in both ophthalmic arteries and is capable of detecting pressure-reducing or hemodynamically significant carotid lesions. The application of these noninvasive techniques allows selection of patients for cerebral angiography and carotid endarterectomy.
In patients who are coronary artery bypass candidates, the presence of a carotid bruit is insufficient to warrant carotid angiography, but a duplex ultrasonogram will detect those patients in whom a carotid arteriogram is indicated prior to bypass surgery. The approach suggested by Jones and co-workers in the paper abstracted above seems rational to me. Among patients who require coronary artery bypass surgery, an asymptomatic bruit or a previous stroke occurring in those patients with less than severe extracranial carotid vascular disease do not necessitate carotid arterial surgery. Patients with an asymptomatic bruit, a previous stroke, transient ischemic attacks, and severe extracranial carotid vascular disease should have combination carotid and coronary surgery. Patients with transient ischemic attacks or those who have significant bilateral carotid disease and/or who require complicated carotid revascularization should have staged procedures with carotid artery surgery preceding coronary bypass grafting.
Neurologic complications occur frequently in patients following CABG; indeed, in a series of 3,206 consecutive coronary bypass procedures, 89 patients died and 32 patients suffered major neurologic syndromes.{--}E.B.
Effects of Blood Pressure Measurement by the Doctor on Patient's Blood Pressure and Heart Rate
Abstract: An alarm reaction by patients to clinic visits could compromise the reliability of diagnosing hypertension. The authors monitored intraarterial pressure continuously as a physician repeatedly measured cuff blood pressure in 48 hospitalized patients with a wide range of arterial pressure. Only patients with uncomplicated essential hypertension were included in the study. The 25 men and 23 women were aged 17-67 years. Intraarterial pressure was monitored over 24 hours, and three or four repeated cuff measurements were made during 10- and 15-minute visits from the doctor.
Visits were accompanied by increases in systolic and diastolic blood pressure. The pressure began rising when the doctor arrived and peaked at about 2 minutes, followed by a slow decline. Systolic blood pressure increased in 47 of the subjects, and diastolic pressure in 45. There were large interindividual differences in the degree of pressure elevation. Average pressures at the end of the visit remained significantly above previsit levels. Similar responses accompanied two visits by the same doctor. A heart rate increase also was observed. Average blood pressure and heart rate responses were similar in men and women, and did not correlate with age, the 24-hour average mean arterial pressure, or variability in mean arterial pressure over the 24-hour monitoring period.
The usual clinical method of measuring blood pressure often is associated with pressure increases that do not result from cuff application or inflation. The effect probably is due to the doctor's visit and the expectation of having the blood pressure measured. It can be minimized by using the last of 3 or 4 measurements made over about 10 minutes. It is not clear whether more precise blood pressure measurements, made by avoiding the alarm reaction, are more closely related to cardiovascular morbidity and mortality.
Post-Comment: It is fascinating, although perhaps not altogether surprising, that the average peak increases in systolic pressure were as high as 26.7 mm Hg; in diastolic pressure, 14.9 mm Hg; and in mean arterial pressure, 18.6 mm Hg. This peak pressure response occurred during the first 4 minutes of the physicians' visits to the patients' bedside. Perhaps these changes represent the exaggerated lability of arterial pressure that occurs in hypertensives. (The authors did not study normal subjects.)
Computerized sphygmomanometers allow the patient or an untrained family member to record blood pressure, and this may well have some advantages over the recording of arterial pressure by the physician. The decision to treat a patient with borderline hypertension might be more rational if the pressure is determined by an automatic or semiautomatic method during the patient's ordinary life rather than under the more artificial, anxiety-provoking conditions of the physician's office. Such an approach will identify patients who are hypertensive most of the time as opposed to those who are hypertensive during only a small fraction of the day or week.
It is probably useful to specify in which arm the blood pressure is recorded, since in one study more than 5% of healthy persons were found to have a systolic pressure difference exceeding 7.5 mm Hg between the two arms. In recording blood pressure, the lightest possible pressure should be placed on the stethoscope; excessive pressure, while not affecting systolic pressure, erroneously lowers diastolic pressure.{--}E.B.
Calcium-Channel Blockade With Nifedipine and Angiotensin Converting-Enzyme Inhibition With Captopril in the Therapy of Patients With Severe Primary Hypertension
Guazzi, Maurizio D.; De Cesare, Nicoletta; Galli, Claudia; Salvioni, Alessandro; Tramontana, Cristina; Tamborini, Gloria; Bartorelli, Antonio
Circulation, August 1984, Vol. 70, Pg. 279-284, Ch. 30-3
Affiliation: Univ. of Milan
Abstract: The beneficial effect of captopril in 2 severely hypertensive patients who responded poorly to nifedipine treatment suggested that the combination of these drugs might be worth trying in such individuals. Fourteen hospitalized patients, 8 men and 6 women with an average age of 48 years, were treated for essential hypertension that was poorly responsive to previous therapy or had not been treated. Diastolic pressures at hospital admission exceeded 120 mm Hg. After a placebo period, patients received nifedipine in 10-mg doses; then, captopril was substituted for placebo in doses of 25 mg. Captopril treatment then was continued as nifedipine was replaced by placebo. In some cases this order of medication was reversed. Each study period lasted for 1 week.
A peak decline in blood pressure occurred within an hour of nifedipine administration. Pressures remained significantly reduced throughout the day with a 6-hour dosing regimen. Captopril had similar but less pronounced effects on blood pressure. The antihypertensive response to combined treatment was definitely greater, with normal blood pressure maintained for several hours during the day. The increased pulse rate and ankle edema associated with nifedipine treatment did not occur during combined treatment. Plasma renin activity increased with captopril therapy alone, returning toward baseline when nifedipine was added. Captopril treatment caused no significant side effects.
The combination of a calcium blocker and a converting-enzyme inhibitor appears to be more effective than the use of either agent alone in patients with high diastolic blood pressure and vasoconstriction. A long-term follow-up trial is under way.
Post-Comment: Almost every combination of antihypertensive agents has been employed in clinical practice, but there is little information about the combination of a calcium-channel blocker and an angiotensin-converting-enzyme inhibitor, which would appear to be a logical one in the treatment of severe hypertension. Nifedipine is a potent vasodilator (as the study abstracted above shows), but when used alone it may cause tachycardia. Although the latter can be overcome by the simultaneous use of a {beta}-adrenergic blocker, the combination of a calcium antagonist and a {beta}-blocker may be undesirable in patients with impaired left ventricular reserve. Hence, as demonstrated by Guazzi et al., a combination of a calcium antagonist with a converting-enzyme inhibitor is attractive. Evans et al. have reported that nifedipine is an effective adjunct in patients with a poor response to standard antihypertensive therapy.{sup-6} Nifedipine (in a retard formulation so that it need be given only twice daily) has also been used successfully and was well tolerated as monotherapy in patients with mild to moderate hypertension.
A few comments about other new antihypertensive drugs: Labetalol was recently approved by the FDA. This is a combined {alpha}- and {beta}-adrenergic-receptor-blocking agent which is useful either as monotherapy or adjunctive therapy in patients with moderate hypertension. The intravenous administration of labetalol has been shown to be particularly effective in patients with severe hypertension. It lowers blood pressure without causing reflex tachycardia or other serious side effects. It is a suitable alternative to the direct vasodilating agents in the management of patients with severe hypertension and hypertensive emergencies.
Guanfacine, a clonidine-like drug that induces an antihypertensive effect by decreasing sympathetic output through activation of central {alpha}{sub-2}-adrenergic receptors in the central nervous system, has also been released recently. Studies by Hedner et al. have shown that guanfacine lowers circulating norepinephrine and epinephrine levels and reduces sympathoadrenal function and arterial pressure equally during rest and exercise.
This is also the year of transdermal patches. Their use for the continuous administration of nitroglycerin has already been described (article 29-19). Recently clonidine has become available as a transdermal preparation with a duration of action of at least 1 week. This preparation is said to have the following advantages: Patients are protected continuously throughout an entire week; daily fluctuations in plasma clonidine concentration are minimized; side effects are reduced; and compliance is improved. In a multicenter study of 85 patients with mild essential hypertension, transdermal patches of clonidine were well tolerated and effective as monotherapy for mild essential hypertension.{--}E.B.
Defect in the Sodium-Modulated Tissue Responsiveness to Angiotensin II in Essential Hypertension
Shoback, Dolores M.; Williams, Gordon H.; Moore, Thomas J.; Dluhy, Robert G.; Podolsky, Stephen; Hollenberg, Norman K.
J. Clin. Invest., December 1983, Vol. 72, Pg. 2115-2124, Ch. 30-4
Affiliation: Harvard Med. School
Abstract: Some hypertensive patients have both a blunted aldosterone response to a volume deficit and a renal perfusion rate inappropriate for the level of sodium intake. Angiotensin II (AII) has a dominant role in the normal responses of both aldosterone and renal blood flow to volume challenge. Eighteen hypertensive and 9 normal persons were evaluated during dietary sodium restriction to 10 mEq and during loading with 200 mEq sodium. Renal plasma flow was estimated by the para-aminohippurate clearance method. Angiotensin II was infused in graded doses of 0.3-3.0 ng/kg per minute for 45-minute periods as the plasma aldosterone, renin activity, and electrolyte levels were monitored.
Eight hypertensive patients had normal renovascular responses to AII infusion, with renal plasma flow values exceeding 125 ml/minute/1.73 sq m. Renal blood flow responses were greater with the high-sodium diet in normal responders, but not in abnormal responders. Basal renal blood flow rose significantly with sodium repletion in the normal responders only. Adrenal responses to AII were significantly greater after sodium restriction in normal responders, but not in abnormal responders (table).
A substantial number of essential hypertensive persons have abnormal responsiveness to angiotensin II in the two systems central to volume homeostasis; renal blood flow and aldosterone responses to AII are not modulated appropriately by changes in dietary sodium intake. The abnormalities may result from increased local production of AII or from a defect in the AII receptors in target tissues.
Post-Comment: These investigators have defined a subgroup of patients who failed to augment aldosterone secretion normally in response to angiotensin when placed on a low-sodium diet, who failed to demonstrate a normal increment in renal blood flow when placed on a high sodium diet, and in whom diastolic arterial pressure failed to rise more with a given dose of angiotensin when on a high-sodium compared with a low-sodium diet. Thus, in these ``nonresponder'' hypertensives, renovascular and adrenal responses to angiotensin II and basal renal blood flow were not appropriately modified by the level of sodium intake. If confirmed, these observations would identify a relatively homogeneous group of hypertensive patients out of the large wastebasket now termed essential hypertension.
The renin-angiotensin system remains important for the clinical management of hypertension. The converting-enzyme inhibitor captopril is an extremely effective antihypertensive drug in patients with severe hypertension, although it often needs to be supplemented{--}a thiazide diuretic being a better supplement than a {beta}-blocker. Enalapril, also a converting-enzyme inhibitor, is an inactive ``pro-drug'' that is absorbed from the gastrointestinal tract and then de-esterified in the liver to its active form. Because of this step, enalapril has a delayed onset and prolonged duration of action and may be useful when given on a once-a-day basis. It has been found to be effective when given in a single daily dose in patients with renal artery stenosis.{--}E.B.
Abstract: Although alcohol intake and blood pressure appear to be related, conflicting evidence on the acute effects of alcohol on blood pressure has been obtained in nonalcoholics. The authors examined the effects of alcohol consumption and abstention on blood pressure in 16 men with essential hypertension who regularly drank moderate amounts of alcohol. None had taken more than 80 gm of alcohol daily in the 4 weeks prior to the study. Half the patients continued to drink for 3 days in hospital and then stopped for 4 days. The others took no alcohol for the first 3 days and than drank beer providing 1 gm of alcohol per kilogram of body weight daily to a maximum of 80 gm daily.
Five of 8 subjects had a rise in systolic blood pressure, and 3 had an increase in diastolic blood pressure after admission while consuming alcohol. Seven of the 8 had a fall in systolic blood pressure, and 6 had a fall in diastolic blood pressure, when alcohol was stopped. In the other group, 7 of 8 men had a rise in systolic blood pressure, and all 8 had a rise in diastolic blood pressure when alcohol consumption was resumed. Liver function remained normal except for serum {gamma}-glutamyl transferase levels, which were elevated at admission but showed no consistent trend during the study. Plasma cortisol changes were not marked except, in the second group, for a significant rise by day 7 on alcohol. Changes in plasma renin activity were not significant. No substantial changes in urine volume or urinary electrolytes were noted. Changes in urinary metanephrine levels were not significant.
Alcohol has a pressor effect on moderate drinkers with essential hypertension. The effect is reversible when drinking ceases, but it could ultimately lead to irreversible changes. The role of alcohol in the pathogenesis of essential hypertension and the need to counsel patients to reduce their alcohol intake remain insufficiently recognized.
Post-Comment: The effects of alcohol are mixed; on the one hand it has been shown that moderate alcohol intake reduces the risk of atherosclerosis by increasing circulating high-density lipoprotein. At the same time alcohol has a pressor effect, particularly in patients with moderate hypertension. The pleasures of moderate alcohol intake and the small but significant increase in high-density lipoprotein on the one hand must be weighed against the deleterious effects of a 15-mm Hg rise in systolic pressure on the other.{--}E.B.
Noninvasive Assessment of Valvular Heart Disease: Surgery Without Catheterization
Borow, Kenneth M.; Wynne, Joshua; Sloss, Laurence J.; Cohn, Lawrence H.; Collins, John J.
Abstract: Accurate, reproducible, noninvasive methods of assessing the anatomy of the heart valves now are available. The authors reviewed data on 41 patients who had valve surgery solely on the basis of the clinical, M-mode echocardiographic, phonocardiographic, and external pulse record findings without preoperative cardiac catheterization. The patients had aortic valve surgery, mitral valve surgery, or both and were followed up for a mean of 4$frac{1/2} years. None of them had a history of chest pain or ECG evidence of coronary artery disease before surgery. Twenty-three patients were hemodynamically unstable at the time surgery was being considered. In the other 18 cases the chance of obtaining further pertinent data from catheterization was thought to be low.
Preoperative noninvasive studies indicated significant mitral valve disease in 24 patients, aortic valve disease in 15, and combined aortic and mitral disease in 2. The findings proved accurate in all cases, allowing the appropriate surgical procedure to be carried out. In no case were there significant unexpected findings at operation. No deaths were related to an incorrect or incomplete preoperative diagnosis.
Successful heart valve surgery can be carried out in selected cases solely on the basis of clinical and noninvasive evaluation. Cardiac catheterization need no longer be considered the definitive study before valve surgery in all instances. It should be done for clarification when the patient's condition is unclear after a planned, comprehensive, noninvasive assessment of the cardiac anatomy and pathophysiology.
Post-Comment: In view of the current emphasis on cost containment, this report is quite timely, and the results are consistent with those of St. John Sutton et al. This study utilized M-mode echocardiography as the principal noninvasive method. Since then, two-dimensional echocardiography and Doppler echocardiography (article 31-4) have been greatly improved and provide substantially more information, suggesting that preoperative catheterization can be avoided in many patients. For too long we have accepted as axiomatic that patients should not undergo cardiac surgery unless they had previously been studied in the cardiac catheterization-angiography laboratory. Clearly this is not always necessary. There is one caveat: Cardiac catheterization is necessary in patients with valvular heart disease in whom coexisting coronary artery disease is suspected. Therefore, coronary arteriography should be performed in patients with valvular heart disease who have evidence of myocardial ischemia{--}either clinically or as a result of noninvasive testing. Valve replacement in the presence of severe coronary disease which is not corrected could lead to a catastrophic outcome and/or a missed opportunity to bypass the coronary obstruction.{--}E.B.
The Natural History of Asymptomatic Patients With Aortic Regurgitation and Normal Left Ventricular Function
Bonow, Robert O.; Rosing, Douglas R.; McIntosh, Charles L.; Jones, Michael; Maron, Barry J.; Lan, K. K. Gordon; Lakatos, Edward; Bacharach, Stephen L.; Green, Michael V.; Epstein, Stephen E.
Circulation, September 1983, Vol. 68, Pg. 509-517, Ch. 31-2
Affiliation: Natl. Inst. of Health
Abstract: Aortic valve replacement has been recommended for asymptomatic patients with severe aortic regurgitation and normal left ventricular (LV) function, but their natural history is unknown. The authors performed serial studies in 77 patients who had asymptomatic aortic regurgitation and normal LV systolic function and were seen consecutively from January 1973 to February 1982. All fulfilled both M-mode echographic and radionuclide angiographic criteria for normal resting systolic LV function. Cardiac catheterization was done in 63 patients, and it confirmed isolated, severe aortic regurgitation in 59 of them. Twelve patients underwent aortic valve replacement and were reevaluated 6-8 months postoperatively.
No patient died during follow-up, and 65 patients have remained asymptomatic with normal LV function for a mean of 49 months. Aortic valve replacement was done in 12 cases after a mean of 40 months of follow-up. Five of the 11 patients operated on for cardiac symptoms developed LV dysfunction before or coincident with the onset of symptoms. The proportion of patients remaining asymptomatic with normal LV function is illustrated in Figure 31-1. The LV ejection fraction increased from 45% to 58% after aortic valve replacement, and the LV diastolic dimension decreased. The LV ejection fraction on maximum exercise also increased after aortic valve replacement.
When aortic valve replacement is delayed in asymptomatic patients with normal LV function until symptoms or LV dysfunction develop, postoperative survival is excellent, and LV dimensions and function improve after operation. Prophylactic aortic valve replacement does not seem warranted to preserve ventricular function in asymptomatic patients with aortic regurgitation who have normal resting LV function.
Post-Comment: The management of asymptomatic patients with severe aortic regurgitation remains controversial. The problem results from the fact that patients with free aortic regurgitation who were symptomatic with left ventricular dysfunction preoperatively are at risk of developing irreversible myocardial dysfunction and death from heart failure after aortic valve replacement. Yet patients with severe aortic regurgitation may remain asymptomatic for years, and in view of the continuous improvement in valvular protheses, it seems inappropriate to operate upon them years before such irreversible changes develop. This paper by the NIH group, which has contributed so much to this important subject, provides a useful approach to these patients. Bonow et al. have demonstrated that asymptomatic patients with severe aortic regurgitation and normal left ventricular function do not require immediate operation. Only about 4% per year develop symptoms or left ventricular dysfunction or both. These patients should be followed carefully by means of clinical examination as well as noninvasive testing, including echocardiography and radionuclide ventriculography every 6-12 months. Aortic valve replacement may be delayed until symptoms or left ventricular dysfunction develop without fear of development of irreversible myocardial changes.
Stone et al.(J. Am. Coll. Cardiol. 3:1118, 1984) have also concluded that the chances of normalization of left ventricular performance in aortic regurgitation correlate best with mild impairment of left ventricular function preoperatively. However, these authors pointed out that even patients with serious preoperative clinical disability and grossly impaired left ventricular function may experience a significant improvement in the quality of life. While this does not mean that one should wait until the patient experiences serious preoperative clinical disability, it does mean that surgery may not be too late even in such patients.
Turina et al. described a 10-year experience with the surgical management of chronic aortic regurgitation and concluded that operation should be performed even in asymptomatic patients who have distinct cardiomegaly. In analyzing their patients who died postoperatively, they found that those who were asymptomatic or mildly symptomatic had distinct cardiac enlargement preoperatively. They also pointed out that patients with cardiomegaly who undergo successful aortic value replacement may die suddenly because of cardiac arrhythmias during the first 2 postoperative years.{--}E.B.
Sensitivity and Specificity of Pulsed Doppler Echocardiography in Detection of Aortic and Mitral Regurgitation
Wautrecht, J. C.; Vandenbossche, J. L.; Englert, M.
Abstract: The accuracy of pulsed Doppler echocardiography in detecting aortic and mitral regurgitation, using both audiosignal and graphic display methods, was examined in 63 patients clinically suspected of having valvular disease who underwent echocardiography and cineangiography within a 48-hour period. Most patients had rheumatic heart disease. A 3-MHz M-mode echocardiograph was used in conjunction with the ultrasonic directional pulsed Doppler flowmeter. The patient was examined in the supine position, sometimes with left lateral rotation, and with the transducer on the left precordium.
The results of Doppler studies were negative in 21 of 24 patients without valve regurgitation. Both laminar, smooth, double-peaked mitral diastolic flow and laminar, smooth, continuous aortic systolic flow were observed in these patients. The 3 others had a modest systolic flow signal in the left atrium just posterior to the mitral valve; 2 had moderate anterior mitral prolapse, and 1 had a stenosed prosthetic mitral valve ring. Significant angiographic regurgitation was consistently detected by the Doppler study (Fig 31-3). Grade 1 mitral regurgitation was missed in 3 patients. The Doppler study was 94% sensitive in detecting both aortic and mitral regurgitation. Its specificity was 87.5%.
Pulsed Doppler echocardiography is a highly sensitive means of detecting aortic and mitral regurgitation, especially the former. It also has high specificity for detecting these lesions, particularly aortic insufficiency. Use of the Doppler method, perhaps simplified by the Fourier transform technique, along with real-time echocardiography, is expected to improve the noninvasive detection of valvular regurgitation.
Post-Comment: Doppler echocardiography has added a new dimension to noninvasive methods for assessing valvular heart disease. The aortic valve gradient may be estimated simply by measuring the velocity of blood flow across the aortic orifice using a simple hydrodynamic principle; the pressure gradient in millimeters of mercury is calculated as four times the square of the maximal velocity across the aortic valve expressed in meters per second. The combination of Doppler and two-dimensional echocardiography{--}which outlines the dimensions and shape of the left ventricular cavity and provides information concerning the thickness of the left ventricular wall and the degree of calcification and anatomy of the aortic orifice{--}provide most of the information required for making a decision concerning the advisability of cardiac surgery in a patient suspected of having aortic stenosis. Doppler echocardiography is also very specific in detecting aortic regurgitation and only slightly less so in detecting mild mitral regurgitation.
Doppler ultrasound uses the change in frequency, i.e., the so-called Doppler shift, produced by moving red blood cells, which reflects transmitted ultrasound to the transducer. This shift in ultrasound frequency is proportional to the velocity of red cell movement. Continuous wave Doppler techniques utilize two transducers, one to transmit and the other to receive ultrasound, while the so-called pulsed Doppler technique uses the same transducer both to transmit and receive ultrasound. Continuous wave Doppler is useful for estimating the transvalvular pressure gradient, while the pulsed Doppler is useful for detecting abnormal direction of flow, as in valvular regurgitation. Doppler echocardiography is also very helpful in following patients with bioprostheses and in diagnosing malfunction of a bioprosthetic valve.{--}E.B.
High Incidence of Myocarditis by Endomyocardial Biopsy in Patients With Idiopathic Congestive Cardiomyopathy
Zee-Cheng, Chi-Sung; Tsai, Cheng Chang; Palmer, Diane C.; Codd, John E.; Pennington, Glenn; Williams, George A.
J. Am. Coll. Cardiol., January 1984, Vol. 3, Pg. 63-70, Ch. 32-1
Affiliation: St. Louis Univ.
Abstract: The incidence of active lymphocytic myocarditis in patients with unexplained congestive heart failure (CHF) may be higher than previously believed. The authors report the results of percutaneous endomyocardial biopsy in 35 patients with unexplained CHF.
With the use of microscopic, ultrastructural, and immunofluorescent studies, the biopsy specimens were classified as showing either no inflammation (cardiomyopathy) or active lymphocytic myocarditis, grades I to IV (Fig 32-1). Twenty-two (63%) of the 35 patients were shown to have active lymphocytic myocarditis; 18 (81%) had grade I or II myocarditis. Electrocardiograms showed a sinus rhythm with nonspecific ST-T wave changes in 15 (68%) of the 22 patients. All but 2 of 14 patients studied by echocardiography exhibited moderate to marked left ventricular dilation with a mean diastolic left ventricular internal dimension of 6.6 {+-} 0.7 cm. Most patients had moderate to severe diffuse left ventricular dysfunction. Treatment was begun with prednisone and azathioprine. Four of 10 patients with chronic active myocarditis responded to immunosuppression with stabilization or improvement in clinical function, hemodynamic status, and biopsy grade. One patient developed an acute exacerbation of CHF and died before therapy could be evaluated. One of 8 patients with subacute myocarditis responded dramatically to immunosuppression and is continuing therapy, and another patient showed symptomatic improvement during 6 months of treatment. Two of the 4 patients with grade III or IV myocarditis died within days of presentation, despite maximal medical therapy. A third developed sepsis and died, even though she had shown transient improvement with the use of antithymocyte globulin.
The findings suggest that there is a high incidence of active myocarditis in patients with idiopathic congestive cardiomyopathy and that a majority of these patients have low grade myocarditis. The presence of T lymphocytes in the myocardium supports the hypothesis that immune-mediated myocardial damage plays an important role in the pathogenesis of active lymphocytic myocarditis presenting clinically as idiopathic congestive cardiomyopathy. Humoral immunity may also be involved in myocardial damage.
Post-Comment: In the past, most patients with unexplained congestive heart failure were considered to have idiopathic congestive cardiomyopathy. The results of this study differ from others in that patients with mild degrees (grade 1 and 2 changes) of lymphocyte infiltration were considered to have myocarditis. The finding that this disease process either halted or improved with immunosuppression is tantalizing. If confirmed, it would substantiate that the disease indeed is a low-grade myocarditis which causes immune-mediated myocardial damage. Since the evaluation of the response to treatment was uncontrolled, i.e., there was no control group, these patients might have improved spontaneously. However, in 3 patients, withdrawal of the immunosuppressive agent resulted in clinical deterioration.
Unverferth et al. determined the histologic characteristics of myocardium obtained by endocardial biopsy in 109 patients with various causes of congestive heart failure. No morphologic differences could be detected between valvular, idiopathic, and alcohol-induced heart failure. The doxorubicin-treated patients with heart failure had the least hypertrophy for a given loss of systolic myocardial function, whereas the alcohol-induced group had the greatest degree of hypertrophy. Fibrosis was most severe in doxorubicin-treated patients and was least severe in those with idiopathic congestive heart failure. However, there was an enormous amount of overlap between the various groups, and there was relatively poor correlation between histologic findings and the extent of left ventricular dysfunction.
Olsen found morphologic changes of myocarditis in 45 of 500 patients with dilated congestive cardiomyopathy. He suggested that it may be useful to combine endomyocardial biopsy with antibody tests for coxsackievirus type B. Some patients with congestive cardiomyopathy have elevated titers of antibody to this virus, even though there is no evidence of myocarditis at biopsy. Olsen concludes that histologic and/or serologic evidence of myocarditis is present in an appreciable number of patients with suspected dilated cardiomyopathy. Other patients demonstrate heart-reactive antibodies and antinuclear antibodies, as well as the presence of {gamma}-globulin bound to the myocardium. Daly et al. studied 12 patients with congestive cardiomyopathy after a recent influenza-like illness; 11 were found to have histologic changes of myocarditis. Most responded to treatment with the immunosuppressive drugs prednisone and azathioprine. For further discussion of the relation between myocarditis and cardiomyopathy, see article 32-2.{--}E.B.
Clinically Unrecognized Ventricular Dysfunction in Young Diabetic Patients
Mildenberger, Richard R.; Bar-Shlomo, Ben; Druck, Maurice N.; Jablonsky, George; Morch, John E.; Hilton, J. David; Kenshole, Anne B.; Forbath, Nicholas; McLaughlin, Peter R.
J. Am. Coll. Cardiol., August 1984, Vol. 4, Pg. 234-238, Ch. 32-5
Affiliation: Toronto
Abstract: The Framingham study showed an increased risk of congestive heart failure in diabetic patients without clinical coronary artery disease or rheumatic heart disease. Subclinical impairment of left ventricular function has been observed in asymptomatic diabetics. Function was assessed at rest and on supine bicycle exercise in 20 diabetic patients without clinical evidence of heart disease and in 18 normal controls. The diabetic patients were aged 21-44 years, and all but 1 were insulin dependent. Nuclide angiography was performed with pertechnetate-labeled red blood cells in conjunction with staged exercise.
No patient experienced chest pain or had an abnormality on ECG during exercise; both groups had similar resting and exercise heart rates and blood pressures. Similar workloads were achieved by the two groups. No regional wall motion abnormalities were noted at rest or on exercise. Ejection fractions at rest were similar in the two groups, but the response to peak exercise was significantly lower in the diabetic group. Responses to exercise were extremely variable in the diabetic patients. Ten of 19 diabetic patients and 3 of 18 controls failed to have a 5% or greater increase in ejection fraction on exercise. Seven diabetics and only 1 control had a decrease in ejection fraction on exercise.
About one third of diabetic patients may have subclinical left ventricular dysfunction not correlated with risk factors for atherosclerosis or other complications of diabetes. Whether the dysfunction results from inapparent coronary artery disease or primary myocardial disease remains uncertain. Patients with ventricular dysfunction should be followed up to detect the possible development of clinical cardiac abnormalities.
Post-comment: Diabetic cardiomyopathy has been recognized with increasing frequency in the last few years, and noninvasive techniques such as echocardiography, as well as cardiac catheterization and left ventricular biopsy, have shown myocardial abnormalities in diabetics without obstructive coronary disease. When more sensitive methods of measuring left ventricular function are employed, the frequency of diabetic cardiomyopathy turns out to be even higher than has been suspected. As summarized above, Mildenberger et al. found the incidence of left ventricular dysfunction to occur in young diabetic patients.
Kuo et al. found that both lysosomal and nonlysosomal proteolytic enzymes may mediate the accelerated cardiac muscle degradation that occurs in the late stage of diabetic cardiomyopathy in the genetically diabetic mouse.--E.B.
Plasma Norepinephrine as Guide to Prognosis in Patients With Chronic Congestive Heart Failure
Cohn, Jay N.; Levine, T. Barry; Olivari, Maria Teresa; Garberg, Victoria; Lura, Dennis; Francis, Gary S.; Simon, Ada B.; Rector, Thomas
N. Engl. J. Med., Sept. 27, 1984, Vol. 311, Pg. 819-823, Ch. 32-6
Affiliation: Univ. of Minnesota
Abstract: The plasma norepinephrine level is elevated in patients in heart failure, reflecting sympathetic activation; the degree of increase might be a more sensitive indicator of cardiac function than are traditional measures of cardiac performance. Hemodynamic and hormonal measurements were made in 106 patients who had severe congestive failure refractory to digitalis-diuretic therapy; the follow-up period was 1-62 months. Sixty patients were considered to have severe coronary artery disease. The 83 men and 23 women, whose mean age was 55 years, were clinically stable at the time of study.
The mean initial pulmonary wedge pressure was 24.5 mm Hg, and the mean stroke-work index was 21 g-m/sq m. The plasma norepinephrine level averaged 700 pg/ml. The average plasma renin activity was 15 ng/ml per hour. Mortality during follow-up was 57%, and the mean survival of patients who died was 9 months. Survivors were followed for an average of 18 months. Patients who died of cardiovascular causes had considerably higher plasma norepinephrine values than the others had. They also had slightly higher plasma renin activities, and lower stroke-work indices. Only the plasma norepinephrine level was a significant prognostic factor. This level was higher in patients who died of pump failure than in those who died suddenly.
Measurement of the plasma norepinephrine level in a stable patient with chronic heart failure might provide an effective guide to prognosis and to the need for more aggressive medical treatment or cardiac transplantation. Acute cardiac decompensation may lead to a rise in the plasma norepinephrine value that does not necessarily indicate the long-term prognosis.
Post-comment: Now that a variety of new medical measures (new inotropic agents) and surgical measures (cardiac transplantation and assisted circulation) are available for the treatment of severe heart failure, it is important to established accurate prognostic indices. In the past, these have been based on New York Heart Association classification and a variety of hemodynamic variables, all of which provide a clue to the prognosis. The resting supine plasma norepinephrine level appears to be a significant predictor of mortality. Presumably it reflects the activity of the sympathetic nervous system, which may be a better measure of the severity of the syndrome of left heart failure than are hemodynamic indices of left ventricular performance. On the other hand, the possibility cannot be excluded that the elevated norepinephrine level is itself responsible for early demise and could be a contributor to mortality from heart failure.--E.B.
Sustained Improvement of Cardiac Function in Patients With Congestive Heart Failure After Short-term Infusion of Dobutamine
Liang, Chang-seng; Sherman, Lon G.; Doherty, John U.; Wellington, Karen; Lee, Victor W.; Hood, William B., Jr.
Circulation, January 1984, Vol. 69, Pg. 113-119, Ch. 32-7
Abstract: Continuous short-term infusion of dobutamine has led to sustained improvement in function of the failing heart. The authors' double-blind study examined the changes in cardiovascular function produced by 72 hours of continuous dobutamine infusion in 15 patients with symptomatic congestive cardiomyopathy of the idiopathic or alcoholic type. All patients were in New York Heart Association functional classes III and IV. In 8 patients, dobutamine was infused for 72 hours at an initial rate of 5 mu g/kg per minute, which was increased until the cardiac output was twice baseline, the heart rate rose by 25 beats per minute, or the mean arterial pressure rose 15 mm Hg above baseline. The average final dose was 25 mu g/kg per minute. The other 7 patients received 5% dextrose in water. The two groups were well matched clinically.
Resting left ventricular ejection fraction improved significantly after dobutamine infusion. Exercise tolerance also improved significantly in this group; baseline values were similar in the two groups. The postexercise arteriovenous oxygen difference was less than the preinfusion value in dobutamine-treated patients. The dobutamine and control groups had similar blood lactate and norepinephrine concentrations at similar workloads.
Long-term improvement in left ventricular performance was produced by dobutamine infusion over 72 hours in these patients with symptomatic congestive cardiomyopathy. The changes cannot be attributed to a conditioning effect. The approach appears to be promising for long-term management of severe heart failure, but the ultimate effect on survival remains to be determined. The clinical improvement may be related to reparative processes effected by dobutamine.
Post-comment: On first reading, this paper is startling. How could a drug that is active for only minutes have an effect persisting for weeks after it is discontinued? Far-fetched as this may seem, there are other data to support it. Thus, Liang et al. found that intermittent infusions of dobutamine in dogs produced cardiovascular and metabolic effects similar to those of exercise training. Leier et al. and Unverferth et al. have shown that short-term infusions of dobutamine in patients cause a sustained improvement in left ventricular function. Unverferth et al. also reported that a 72-hour dobutamine infusion caused an improvement of myocardial mitochondrial structure and increased the adenosine triphosphate-creatine ratios of endocardial biopsy specimens in patients with heart failure.
Frankly, I am still puzzled about what is going on here, but I do believe that Liang et al., as well as Unverferth's group, may have uncovered something rather important about the pathophysiology of heart failure that may have important therapeutic implications. I understand that at the present time a multicenter randomized trial is being carried out to study the clinical response to intermittent infusions of dobutamine in a large number of patients with chronic congestive heart failure. The results will be awaited with interest.
In the 1984 YEAR BOOK the initial hemodynamic studies with milrinone, a bipyridine inotropic-vasodilating agent, were described. Since then several imidazolone compounds have also been shown to be positive inotropes and vasodilators. Among these compounds, MDL 17,043 has been demonstrated to exert a positive inotropic effect and balanced (arterial and venous) vasodilation in patients with reduced left ventricular performance. MDL 19,205, piroximone, has similar hemodynamic effects. Both drugs are orally active and together with amrinone and milrinone constitute a new series of noncatecholamine, nonglycoside positive inotropic vasodilator agents. The role, if any, of these drugs in the chronic management of heart failure remains to be determined. Their mechanism of action apparently is based on inhibition of membrane-bound phosphodiesterase, resulting in an increase in intracytoplasmic cyclic AMP that stimulates contractility of myocardial cells and produces dilatation in vascular smooth muscle. Oral levodopa, which is decarboxylated to form dopamine, has also been shown to raise cardiac output when given orally, and in several patients oral levodopa produced sustained clinical improvement after more than 6 months of therapy.--E.B.
Prostaglandins in Severe Congestive Heart Failure
Dzau, Victor J.; Packer, Milton; Lilly, Leonard S.; Swartz, Stephen L.; Hollenberg, Norman K.; Williams, Gordon H.
N. Engl. J. Med., Feb. 9, 1984, Vol. 310, Pg. 347-352, Ch. 32-8
Affiliation: Harvard Univ.
Abstract: Plasma levels of the metabolites of the vasodilator prostaglandins I$sub{2} and E$sub{2} were measured in 15 patients with severe chronic heart failure to determine whether prostaglandins are involved in circulatory homeostasis in congestive heart failure.
Mean plasma renin activity values (15.0 +- 3.8 ng of angiotensin I per ml per hour; range 0.2-53.0 ng), plasma prostaglandin E$sub{2}-M (843 +- 20 pg/ml; range, 13-3,000 pg/ml), and 6-keto-prostaglandin F$sub{1}$sub{alpha} (128 +- 31 pg/ml; range 23-490 pg/ml) were markedly elevated in patients with severe heart failure. Compared with normal subjects, mean circulating levels of both metabolites were 3-10 times higher. Plasma concentrations of prostaglandin E$sub{2} metabolites varied linearly and directly with plasma renin activity (r = .72; P < .005) and with plasma angiotensin II concentrations (r = .84; P < .001). A linear correlation also was observed between plasma concentrations of 6-keto-prostaglandin F$sub{1}$sub{alpha} and plasma concentrations of angiotensin II (r = .64; P < .02). Significant inverse correlations were found between serum sodium concentrations and plasma levels of prostaglandin E$sub{2}-M (r = -.92; P < .001) and plasma renin activity (r = -.69; P < .02). Comparison of patients with normal and low serum sodium concentrations showed that hyponatremic patients had increased levels of prostaglandin E$sub{2}-M, plasma angiotensin II, plasma 6-keto-prostaglandin F$sub{1}$sub{alpha}, and plasma renin activity. Hyponatremic patients also had significantly higher levels of blood urea nitrogen and serum creatinine (P < .01). Of 23 patients with heart failure who were challenged with indomethacin, an inhibitor of prostaglandin synthesis, those with hyponatremia showed significant decreases in cardiac index (P < .001) and increases in mean arterial pressure (P < .01), pulmonary capillary wedge pressure (P < .001), and systemic vascular resistance (P < .001), without significant changes in mean right atrial pressure or heart rate. In contrast, no significant hemodynamic changes were observed after the administration of indomethacin in patients with normal serum sodium concentrations.
The findings indicate the vasodilator prostaglandins are increased in patients with severe heart failure in accordance with the degree of activation of the neurohumoral vasoconstrictor systems. These increases may offset the circulatory effects of systemic and regional vasoconstriction. The close inverse relationship between hormonal levels and serum sodium concentrations suggests that the latter may be a useful index of hormonal status in patients with heart failure. Patients with hyponatremia may be particularly susceptible to clinical deterioration if given drugs that inhibit prostaglandin synthesis, but may be likely to improve when given drugs that enhance the production of endogenous prostaglandins.
Post-comment: As noted in article 32-1, congestive heart failure is associated with elevated circulating norepinephrine, a vasoconstrictor. The paper by Dzau et al. abstracted above indicates that in addition to norepinephrine, plasma renin activity and angiotensin II levels are also elevated in patients with congestive heart failure, providing further vasoconstrictor stimulation. However, several prostaglandin derivatives that produce vasodilatation were also markedly elevated in patients with severe heart failure, perhaps serving as a compensatory mechanism to maintain local flow in the face of reduced cardiac output and vasoconstriction. Patients with heart failure and hyponatremia (serum sodium levels of less than 135 mEq) differed from patients with normonatremia in that they had more extreme changes in both plasma angiotensin and prostaglandins.
The importance of these circulating prostaglandin vasodilators is reflected in the observation that administration of indomethacin, which inhibits the production of prostaglandins, resulted in deterioration of left ventricular function. Deterioration occurred as a consequence of the marked increase in systemic vascular resistance that indomethacin caused in these patients, indicating that they were dependent on prostaglandin synthesis to counteract the vasoconstrictor influences of heart failure. Converting-enzyme inhibitors, such as captopril and enalapril, may be effective in heart failure in part because they enhance the production of endogenous prostaglandins.
The beneficial effects of captopril in chronic congestive heart failure have recently been confirmed. Recently a new oral converting-enzyme inhibitor, enalapril, has been introduced (see comments following article 30-4). This drug has a gradual onset and prolonged duration of action and since it lacks sulfhydryl groups it may have fewer side effects. It appears to be well tolerated and quite effective in the treatment of chronic congestive heart failure.--E.B.
Effect of Esophageal Emptying and Saliva on Clearance of Acid From the Esophagus
Helm, James F.; Dodds, Wylie J.; Pelc, Lorie R.; Palmer, David W.; Hogan, Walter J.; Teeter, Bruce C.
N. Engl. J. Med., Feb. 2, 1984, Vol. 310, Pg. 284-288, Ch. 33-1
Affiliation: Med. College of Wisconsin, Milwaukee
Abstract: Clearance of acid from the esophagus after acid reflux is an important defense against esophagitis. The authors examined the role of esophageal emptying and its interaction with swallowed saliva in esophageal acid clearance in 19 male and 3 female healthy subjects, aged 20-30 years. Acid clearance studies were done using 0.1 N hydrochloric acid after an overnight fast. Esophageal emptying was assessed using $sup{9}$sup{9}$sup{m}Tc-sulfur colloid.
Injection of acid into the esophagus produced a transient increase in pressure and a secondary peristaltic sequence that nearly completely cleared the 15-ml bolus of acid. The esophageal pH began to increase at the first swallow, 30 seconds after bolus injection; it then returned to normal in a stepwise manner in association with swallow-induced peristaltic sequences. The overall clearance time to a pH of 4 was 284 seconds. Saliva stimulation by an oral lozenge reduced the acid clearance time by about half in the 10 subjects studied. Emptying of the bolus volume was unchanged by saliva stimulation. Oral aspiration of saliva abolished the stepwise increases in esophageal pH in the same subjects. Esophageal emptying remained incomplete in the absence of peristalsis after propantheline administration. Propantheline also eliminated salivation and esophageal acid clearance.
Esophageal acid clearance normally is a two-step process consisting of esophageal emptying followed by acid neutralization. One or two peristaltic sequences reduced esophageal acid to a minimum residual amount that sustains a low pH, and swallowed saliva then restores the esophageal pH in a stepwise manner. Abnormal peristalsis can lead to impaired esophageal emptying and a large residual acid volume that is not rapidly neutralized by saliva.
Post-comment: The emptying of an acid volume from the esophagus and restoration of a normal intraluminal pH are related but distinctly different events. After virtually complete emptying of a 15-ml acid volume by an immediate peristaltic sequence, esophageal pH did not begin to rise until the first swallow 30 seconds later. In the past, it was believed that neutralization of acid by saliva contributed little to esophageal acid clearance. This conclusion was based on the observation that a large volume of saliva was needed to neutralize a 15-ml aliquot of 0.1 N hydrochloric acid. However, the volume of swallowed saliva entering the esophagus during several minutes is sufficient to neutralize a 1.0-ml aliquot of 0.1 N hydrochloric acid. Thus, when peristalsis reduces the amount of esophageal acid to a small volume, swallowed saliva is capable of restoring the intraluminal pH to normal within minutes.
The key message in this article is that esophageal acid clearance may be delayed by either impaired esophageal motility or abnormality in acid neutralization by swallowed saliva, or both.--N.J.G.
Cimetidine, Cigarette Smoking, and Recurrence of Duodenal Ulcer
N. Engl. J. Med., Sept. 13, 1984, Vol. 311, Pg. 689-692, Ch. 34-2
Abstract: A 1-year, double-blind, multicenter trial was conducted to compare three dose schedules of cimetidine with placebo in preventing recurrent duodenal ulcer in 370 patients with recently healed duodenal ulcer. All experienced at least one clinical recurrence in the year preceding the study. The double-dummy technique was used. Cimetidine was given in a dose of 200 mg twice daily, 300 mg twice daily, or 400 mg at bedtime. Endoscopy was done after 6 and 12 months and whenever symptoms of dyspepsia occurred or there was evidence of bleeding. The treatment groups were clinically and demographically comparable.
The proportion of patients with symptomatic ulcer recurrence, similar in all cimetidine-treated groups after 1 year, ranged from 13% to 19% and was significantly lower than in the group given placebo. Asymptomatic recurrences also were more frequent in the placebo-treated patients. Cimetidine appeared to protect both smokers and nonsmokers from symptomatic ulcer recurrence (Fig 34-2). Nonsmokers given placebo had a lower recurrence rate than did cimetidine-treated smokers. The duration of ulcer disease before the study did not influence recurrence significantly, and age and sex were not factors. No major complications occurred in either cimetidine-treated or placebo-treated patients.
Smoking appears to be a major factor causing recurrence of duodenal ulcer, and its cessation may be a more effective preventive measure than administration of cimetidine is. The decision of whether or not to use maintenance therapy depends on the frequency and severity of ulcer attacks, which in turn may be influenced by the patient's smoking history.
Post-comment: McCarthy has written a timely editorial (N. Engl. J. Med. 311:726, 1984) on smoking and ulcers, and in the three paragraphs below I have abstracted some of the key concepts from that editorial.
``It has been estimated that about 2.5% of ulcers recur each month with maintenance therapy, as compared with 8.5% per month with placebo; about 43% of these recurrent ulcers will heal spontaneously. Thus, the number of recurrences detected depends on the frequency of endoscopy in asymptomatic subjects: those without ulcers at 6 or 12 months may have had ulcers at other times. This effect of sampling is reflected in the apparent increases in recurrence rates that Sontag et al. observed at scheduled times of endoscopy in patients without sufficient symptoms or bleeding to prompt early endoscopy. The study was not stratified for smoking habits. Caution is therefore indicated in comparing smokers and nonsmokers, since other important predictive variables may have had different distributions between the groups, sex and compliance being particularly important.
``However, in support of the conclusions drawn by Sontag et al., impaired healing of duodenal ulcers and increased recurrence rates in smokers, as compared with nonsmokers, have been observed in more than 20 trials, and the findings apply with statistical significance not only to treatment with cimetidine but also to therapy with oxmetidine, ranitidine, pirenzepine, high- and low-dose antacids, trimipramine, and placebo. In additional trials, similar trends, though not statistically significant, have been observed for sucralfate, carbenoxolone, arbaprostil, sulpiride, and tripotassium dicitrato-bismuthate. Studies showing counter trends are exceptional and involve only small samples.
``The best evidence suggesting causation comes from data linking ulcers with the quantity of cigarettes smoked. Numerous studies have shown positive correlations between the number of cigarettes smoked per day and the prevalence of peptic ulcer, the degree of impaired healing, and rates and rapidity of ulcer recurrence. In several studies, smoking fewer than 10 cigarettes per day was not associated with an increased susceptibility to ulcers, delayed healing, or risk of recurrence. Above this rate of smoking, the risks rose significantly: smoking more than 30 cigarettes per day was associated with 100% recurrence within 3 months after discontinuation of therapy. A long duration of disease and smoke inhalation, but strangely, not alleged high nicotine or tar content, also increase the risks. `Low tar' cigarettes will not solve the problem.''
All of the above studies support the view that cigarette smoking may be the most important factor in determining ulcer healing or recurrence.--N.J.G.
Ranitidine 150 mg Twice Daily Versus 300 mg Nightly in Treatment of Duodenal Ulcers
Ireland, A.; Colin-Jones, D. G.; Gear, P.; Golding, P. L.; Ramage, J. K.; Williams, J. G.; Leicester, R. J.; Smith, C. L.; Ross, G.; Bamforth, J.; DeGara, C. J.; Gledhill, T.; Hunt, R. H.
Lancet, Aug. 14, 1984, Vol. 2, Pg. 274-276, Ch. 34-4
Abstract: Ranitidine has been as effective as cimetidine in preventing relapse of duodenal ulcer disease when used in a dose of 150 mg twice daily. The effect of this regimen was compared with that of a nightly dose of 300 mg of ranitidine administered for 4 weeks to 102 patients with endoscopically proved duodenal ulcers. Those with pyloric stenosis or recent perforation were excluded. The median patient age was 48 years. The patients treated twice daily received doses in the morning and at night. The two treatment groups were similar in terms of background data.
Healing rates were comparable in the patients treated twice daily and those treated at night only. There were no signi significant differences in mean days or nights with pain or in the use of antacids. No significant biochemical or hematologic abnormalities were observed. Six patients in each group experienced adverse side effects. Nine of the 11 patients whose ulcers failed to heal were regular smokers. Smoking was associated with poorer ulcer healing only in the twice-daily treatment group (table).
Ranitidine in a nightly dose of 300 mg is at least as effective as treatment with 150 mg twice daily in patients with duodenal ulcer disease. The same may be true for cimetidine. The adverse effect of smoking on ulcer healing may be due to a reduction in inhibition of nocturnal acid secretion by an H$sub{2}-receptor antagonist. Administration of 300 mg of ranitidine at night may overcome this effect.
Post-comment: Several studies have established that ranitidine is an effective agent in the treatment of duodenal ulcer disease. Earlier investigations suggested that treatment with ranitidine is associated with fewer side effects than are associated with cimetidine, especially side effects related to drug-drug interactions. However, more recent studies have indicated that several drug-drug interactions may occur with ranitidine as well. Interactions with warfarin, metoprolol, and theophylline have been reported. In particular, it should be noted that ranitidine may impair theophylline clearance. Finally, there are two recent reports of ranitidine-associated hepatitis. With ranitidine use increasing, further side effects similar to those reported with cimetidine may be anticipated.--N.J.G.
Omeprazole in Zollinger-Ellison Syndrome: Effects of a Single Dose and of Long-Term Treatment in Patients Resistant to Histamine H{sub-2}-Receptor Antagonists
Lamers, Cornelis B. H. W.; Lind, Tore; Moberg, Sven; Jansen, Jan B. M. J.; Olbe, Lars
N. Engl. J. Med., Mar. 22, 1984, Vol. 310, Pg. 758-761, Ch. 34-5
Abstract: Omeprazole is a substituted benzimidazole that inhibits gastric acid secretion stimulated by various means in vitro and in vivo. The authors examined the effect of a single oral dose of omeprazole on gastric acid secretion in 5 patients with Zollinger-Ellison syndrome. Six patients with Zollinger-Ellison syndrome whose symptoms were resistant to high doses of histamine H$sub{2}-receptor antagonists were treated with omeprazole, as was a patient whose peptic ulcer was resistant to low doses of cimetidine. Treatment has continued for an average of 14 months. The initial dose was 60 mg daily; subsequently the dose was adjusted on the basis of gastric acid measurements.
An 80-mg dose of omeprazole inhibited gastric acid secretion in all instances. Gastric acid secretion was inhibited by 76%-100% after 24 hours, when no plasma omeprazole was detectable. Both acid concentration and gastric volume were reduced. No side effects were observed. The long-term study showed omeprazole to be a potent, long-acting antisecretory agent (Fig 34-3). Symptoms improved within 1-2 weeks of the start of treatment. Peptic ulcers and esophagitis were healed after a month of treatment in all cases. Absorption improved markedly in 1 case. The serum creatinine level returned to normal in a patient previously given cimetidine. Four patients were maintained on daily doses of 30-40 mg.
Omeprazole is a very potent, long-acting antisecretory drug that is effective in patients with Zollinger-Ellison syndrome whose symptoms are refractory to treatment with histamine H$sub{2}-receptor antagonists. Further study of the long-term safety of omeprazole therapy is required, but the drug has been well tolerated by the present patients.
Post-comment: McArthur et al. have provided additional data indicating that omeprazole, given as a single daily dose, is effective therapy in Zollinger-Ellison syndrome.
Subjects for their study were 8 patients with Zollinger-Ellison syndrome (basal acid output, 42 mEq/hour; maximal acid output, 49 mEq/hour). After an initial single oral dose of omeprazole (60 mg), maximal inhibition of gastric acid secretion occurred at a mean of 4 hours (range, 3-7 hours). Inhibition of gastric acid secretion persisted for at least 48 hours, in that 48 hours after the patients received a single oral dose of omeprazole, gastric acid secretion was still inhibited by 87%. To establish the dose requirement of omeprazole, patients took the drug once in the morning every 24 hours, and gastric acid secretion was measured during the last hour before the next dose of medication. The dose requirement was the minimal dose of omeprazole that reduced gastric acid secretion to below 10 mEq/hour during the last hour before the next dose of omeprazole. The mean dose requirement for omeprazole in the 8 patients was 65 mg/day (range, 20-100 mg/day). Patients were treated with a single daily dose of omeprazole for as long as 6 months. The efficacy of the drug persisted during this time, and there were no detectable side effects or changes in laboratory test results. These findings indicate that omeprazole given as a single daily dose is safe and effective in the long-term management of patients with gastric acid hypersecretion and promises to become the drug of choice for treatment of Zollinger-Ellison syndrome.
For an up-to-date authoritative brief review of pump blockers and ulcer disease, see the editorial by Sachs.--N.J.G.
A Randomized Controlled Study of Propranolol for Prevention of Recurrent Gastrointestinal Bleeding in Patients With Cirrhosis: A Final Report
Abstract: The authors previously reported the results of a controlled trial that showed continuous oral administration of propranolol reduced the risk of recurrent gastrointestinal (GI) bleeding in patients with cirrhosis; only a part of the patients had been followed up for 1 year. The present controlled trial was continued for an additional year. Therefore, all patients have now been followed up for at least 2 years.
Seventy-four patients with cirrhosis who were admitted for an episode of GI bleeding were included in the study. Ascites, jaundice, and encephalopathy were absent or mild and transient. The patients were randomly assigned to two groups: One group of 38 received propranolol twice daily at doses that reduced the resting heart rate by 25%; the other group of 36 received a placebo twice daily. Cumulative percentages of patients free of recurrent GI bleeding 1 and 2 years after inclusion were 87% and 79% in the propranolol group and 42% and 32% in the placebo group (Fig 34-8). Both differences were highly significant. Cumulative percentages of surviving patients 1 and 2 years after inclusion were 94% and 90% in the propranolol group and 84% and 57% in the placebo group. The difference between the two groups was not significant at 1 year but was statistically significant at 2 years (Fig 34-9).
Prolonged follow-up of patients confirms and extends the previous conclusion that continuous oral administration of propranolol reduces the risk of rebleeding in patients with cirrhosis who are in good condition. Most rebleeding episodes took place within 3 months of the original episode of GI hemorrhage. These times of rebleeding may explain the difference in the number of patients lost to follow-up between the two groups. It is not surprising, therefore, that a prolonged follow-up does not markedly alter the conclusion based on the shorter period of observation.
The conclusion that propranolol reduces the risk of rebleeding is reinforced by the observation that the percentage of patients free of rebleeding was not different in the propranolol group and in the placebo group, which received the agent after rebleeding. The prolonged follow-up likewise confirms that continuous oral administration of propranolol reduces the risk of rebleeding not only in those who bled from ruptured esophageal or gastric varices (Fig 34-10) but also in those who bled from acute gastric erosions. Gastrointestinal bleeding is a significant factor in the mortality of patients with cirrhosis but in otherwise good condition who were selected a few days after an episode of GI bleeding, and prevention of rebleeding in such patients may increase the survival rate.
Propranolol was well tolerated by most patients. Significant changes in liver function tests, which might have resulted from the slight diminution in hepatic blood flow by propranolol, were not observed in patients who received the drug. Although reduction in renal blood flow may have resulted from a decrease in cardiac output induced by the beta-blocker, no significant increase in level of serum creatinine or in the occurrence of the hepatorenal syndrome was noted. This suggests that despite the decrease in cardiac output, renal blood flow is maintained in patients with cirrhosis who are receiving propranolol.
Inability to increase cardiac output is a potential risk in patients who are receiving propranolol and rebleed; this complication was not observed in any of the patients who rebled. Antagonists of beta-blockers may, however, be useful in management of patients who are receiving propranolol and have inappropriate cardiac response to bleeding.
Post-comment: Conn has written a thoughtful editorial on propranolol in portal hypertension and much of what follows has been abstracted from that editorial (Hepatology 4:560, 1984). This landmark investigation has revolutionized the treatment of portal hypertension and has raised a number of theoretic and practical questions.
(1) Is propranolol effective in preventing variceal hemorrhage in patients with advanced decompensated cirrhosis? Lebrec et al. selected a group of cirrhotic patients, predominantly alcoholic, who were in good condition, i.e., jaundice and ascites were ``absent or mild and transient.'' The patients have been considered, therefore, to be equivalent to class A of the Child Turcotte classification. Lebrec et al. report that 72% of their patients were class A and 28% were class B using a modified classification that excludes the nutritional component. In general, their patients are not dissimilar to those reported in a number of controlled clinical trials of other forms of elective therapy of variceal hemorrhage. Differences in application and scoring of the Child Turcotte classification might easily account for the different results reported. Differences in the accuracy of diagnosis, the site of hemorrhage, and the time of initiation of therapy may be equally important outcome factors.
Burroughs and co-workers in London recently reported the results of their randomized controlled trial of propranolol in 48 patients who survived an episode of variceal hemorrhage. They noted no decrease in the frequency of hemorrhage from varices in the propranolol-treated patients. Presumably, Burroughs' patients, who were unselected, had more advanced cirrhosis than did the carefully chosen patients in Lebrec's study. Actually 60% were class A, 30% were B, and 10% were C. Although these differences are not as great as assumed, they indicate that Burroughs' patients tended to have more severe cirrhosis than did Lebrec's. The etiology of the cirrhosis may actually have been more important than its severity, since the majority of Burroughs' patients had nonalcoholic cirrhosis. How this tale of two cities will be resolved is not known.
(2) Does propranolol improve survival? Lebrec et al. concluded that propranolol reduced the mortality during a 2-year period of continuous oral administration. Analyzing their data with the Kaplan-Meier method of cumulative survival and the log-rank test, they found a statistically significant reduction in mortality (P < .02). They point out, however, that the mortality was not significantly lower if the number of deaths in the two groups is analyzed using the chi{sup-2} test in relation to the numbers of patients allocated to the two groups or if the patients who are lost to follow-up are considered dead.
(3) Is propranolol safe when administered over long periods? In Lebrec's study, a single patient developed heart failure that required discontinuation of the drug, and several complained of sexual depression or asthenia. Even more impressive is the observation that the frequency of deaths from liver failure and from the hepatorenal syndrome was low and similar in both groups. They found no significant differences in results of blood tests for liver function or in serum creatinine concentrations during 2 years of propranolol ingestion. These observations are reassuring for those who feared the effects of reduced hepatic and renal blood flow, since other studies have suggested that propranolol diminishes the efficacy of diuretic agents in cirrhotic patients with ascites.
(4) Are cardioselective {beta}-adrenergic blockers superior to nonselective blockers? Negative answers appear from several sources.
(5) By what mechanism does propranolol prevent bleeding? Despite the reduction in the wedged to free hepatic vein pressure gradient (HVPG) observed by Lebrec et al. in virtually every patient, other observers have found that 25%-40% of cirrhotic patients fail to show a significant decrease in portal pressure with beta-adrenergic blockade. It is not clear whether a decrease in portal pressure is a prerequisite for successful therapy, although this is the working hypothesis. Butzow et al. noted bleeding in a patient who had not exhibited a decrease in portal venous pressure with metoprolol. Burroughs et al. noted bleeding in patients whose HVPG had been reduced with propranolol well below the traditional 12-mm Hg gradient.
``For almost a third of a century we have been in the therapeutic rut of esophageal tamponade followed by a portal-systemic shunt to treat patients with esophageal hemorrhage. The emergence of vasoconstrictive agents, fiberoptic therapeutic endoscopy, and new and better portal-systemic shunts indicate that we were making progress. The introduction by Lebrec and his colleagues of an important, innovative pharmacotherapeutic approach to management promised a bright new world. So far, however, this therapy has become a little tarnished. When we have objectively determined which of these new therapies is most effective--when, how, and in whom--the world may yet be a little shinier.''--.J.G.
Gastroscopic Screening in 80 Patients With Pernicious Anemia
Stockbrugger, R. W.; Menon, G. G.; Beilby, J. O. W.; Mason, R. R.; Cotton, P. B.
Gut, December 1983, Vol. 24, Pg. 1141-1147, Ch. 34-12
Affiliation: London
Abstract: Patients with pernicious anemia are thought to have a 3-4 times greater risk than control patients of developing gastric cancer, which has a better prognosis if detected at an early stage. The authors performed upper gastrointestinal endoscopy with biopsy and cytologic study in 80 patients with pernicious anemia younger than 70 years of age. Their mean age was 62 years. Twenty-nine patients had dyspeptic symptoms at the time of endoscopy. The overall mean duration of disease was 4.9 years, but 13 patients were examined at the time pernicious anemia was diagnosed. Half the patients had had neurologic symptoms at the time of presentation.
The endoscopic appearances suggested antral inflammation in 36 cases, but histologic signs of inflammation were present in only 17 patients, and they never were marked. Intestinal metaplasia was present in the antral mucosa in 24 cases. Atrophic gastritis in the body of the stomach was a universal finding. Intestinal metaplasia was present in the body mucosa in 78 of 80 cases. Mucosal dysplasia was found in 33 cases and was severe in 3 of them (table). Histologic examination was more sensitive than cytology in diagnosing dysplasia. Gastric carcinoma was diagnosed in a patient with a 4-year history of pernicious anemia but no gastrointestinal symptoms. No remaining cancer was present in the gastrectomy specimen following endoscopic biopsies. In addition, 18 patients had gastric polyps (often multiple), and 1 had multiple polypoid cardinoid tumors. Barium meal studies missed some abnormalities and did not reveal any lesion not previously diagnosed at endoscopy in the 17 patients examined.
These findings warrant further endoscopic studies in patients with pernicious anemia, and sequential studies to establish the natural course of gastric dysplasia in these patients. The results suggested that all patients except the most elderly and frail should have endoscopy with multiple biopsies at least once at the time of diagnosis of pernicious anemia.
Post-comment: Carney et al. have described the syndrome of gastric argyrophil carcinoid tumors and nonantral gastric atrophy. Interestingly, several of their patients had pernicious anemia. Records of the 30 cases of gastric carcinoid at the Mayo Clinic were reviewed and showed that the gastric mucosa was normal, hyperplastic, or atrophic (nonantral) in 12, 2, or 16 patients, respectively. In the atrophic group, the tumors were in the gastric body and fundus; they were small, polypoid, and multicentric, and they were associated with fundal argyrophil cell hyperplasia. In immunocytochemical studies, minor tumor cell populations stained positively for 5-hydroxytryptamine, gastrin, and somatostatin in 1 case and for 5-hydroxytryptamine in 3 others. Metastasis occurred in 3 patients. Twelve patients had pernicious anemia. Parietal cell or intrinsic factor antibodies or both were present in all 12 patients tested. Each of the 7 patients with an intact antrum had massive hypergastrinemia. No common HLA-A, -B, or -DR antigen pattern was detected among the 10 patients tested. In summary, these results suggest that nonantral gastric atrophy predisposes to gastric carcinoid as well as to gastric carcinoma.--N.J.G.
Narcotic Bowel Syndrome Treated With Clonidine: Resolution of Abdominal Pain and Intestinal Pseudo-Obstruction
Sandgren, John E.; McPhee, Mark S.; Greenberger, Norton J.
Ann. Intern. Med., September 1984, Vol. 101, Pg. 331-334, Ch. 35-4
Affiliation: Univ. of Kansas
Abstract: Five patients taking large doses of opiate analgesics for abdominal pain experienced a syndrome of chronic pain, vomiting, weight loss, and features of intestinal pseudo-obstruction. The symptoms resolved in all 5 patients after the use of opiate analgesics was discontinued (table). The patients were treated with clonidine during the withdrawal period.
Man, 62, patient 1, underwent resection of sigmoid colon cancer at age 40 years, and at age 54 years mechanical small bowel obstruction was relieved by lysis of adhesions. At age 61 years, the patient had anteroposterior colon resection and colostomy because of rectal cancer; radiotherapy was delivered to the lower abdomen. Lysis of adhesions was again performed 4 months before the current hospitalization. Intramuscular morphine was given for crampy right lower quadrant pain. The pain, as well as signs of distention and borborygmi, was partly relieved by vomiting. The current dose of morphine was 10 mg every 6-8 hours; orally administered oxycodone and pentazocine also were used to control abdominal pain. The patient was undernourished and had a tender right midabdomen with active bowel sounds. Abdominal distention and increased tenderness occurred during episodes of pain, along with high-pitched bowel sounds and rushes. Roentgenography showed mildly dilated small bowel loops and several air-fluid levels. Treatment with narcotic analgesics was discontinued. Clonidine was given orally in a dose of 0.1 mg twice daily, and later 0.1 mg 4 times daily. Abdominal symptoms resolved within 36 hours, and the clonidine dosage was tapered over 10 days. The patient remained well for the subsequent 18 months, with good weight gain and normal bowel function.
These 5 patients took narcotics daily in moderate or large amounts, but pain relief was brief and inadequate. Ileus or pseudo-obstruction sometimes was severe enough to suggest mechanical bowel obstruction. A direct action of narcotics on the intestine may be responsible. Other possibilities are chronic constriction of the sphincter of Oddi and early sympathetic withdrawal activity in the intestine after narcotic administration. The mechanism of action of clonidine in this setting is unclear, but the drug is of benefit to patients with opiate withdrawal symptoms primarily involving the gastrointestinal tract.
Post-comment: The narcotic bowel syndrome is a clinically important and frequently unrecognized cause of chronic abdominal pain. We have studied more than a dozen such patients, and virtually all of them have responded to treatment with clonidine. The usual regimen is clonidine, 0.1 mg four times daily for 3 days, then 0.1 mg three times daily for 3 days, then 0.1 mg twice daily for 3 days, then 0.1 mg once daily for 3 days. If patients develop hypotension the dose is reduced by one third to one half and tapered over 9 or 10 days. However, some patients have required more prolonged therapy with lower dosages of clonidine, i.e., 0.05 mg twice daily.
There is a provocative preliminary report on the reversal of chronic idiopathic constipation by the opioid antagonist naloxone. Two patients with long-standing idiopathic chronic constipation, which responded only to large daily doses of laxatives and additional suppositories and enemas, were treated with the specific opioid antagonist naloxone on a single-blind crossover basis. Both patients responded to naloxone treatment, with increased passage of feces and increased wet and dry fecal weight. Though naloxone is poorly absorbed after oral administration, there was a positive response during oral as well as intravenous treatment, suggesting that the primary effect of naloxone is at specific opiate receptor sites in the myenteric plexus and other neural and endocrine cells of the intestinal wall.--N.J.G.
Rectal Biopsy Helps to Distinguish Acute Self-Limited Colitis From Idiopathic Inflammatory Bowel Disease
Surawicz, Christina M.; Belic, Lanie
Gastroenterology, January 1984, Vol. 86, Pg. 104-113, Ch. 36-2
Affiliation: Seattle
Abstract: The authors undertook a retrospective blind evaluation of rectal biopsy specimens from 44 patients having acute self-limited colitis (ASLC) and 104 with idiopathic inflammatory bowel disease (IIBD) in order to determine whether the two states can be distinguished histologically. When a patient is first examined, the question arises of whether the illness is self-limited or is the initial episode of IIBD. The patients with ASLC has been ill for less than a month, and usually for less than 2 weeks at the time of sigmoidoscopy. The illness resolved completely at short-term follow-up in all cases. Eighty-two patients had a definite diagnosis of IIBD at the time of biopsy. Twenty-two others having symptoms for less than 3 months had definite IIBD after at least 1 year of follow-up.
Seven histologic features proved to be highly discriminating by their frequent occurrence in IIBD and rare occurrence in ASLC (table). At least one of these features was present in 79% of the cases of IIBD. The 3 biopsy features favoring ASLC were less helpful in differential diagnosis. Distorted crypt architecture and mixed lamina propria inflammation discriminated well between ASLC and both ulcerative colitis and Crohn's disease. Crypt atrophy, a villous surface, basal lymphoid aggregates, and surface erosions all strongly favored the diagnosis of ulcerative colitis, whereas granuloma strongly favored the diagnosis of Crohn's disease over ASLC. Crypt abscesses were more frequent in ulcerative colitis than in ulcerative proctitis.
Rectal biopsy is helpful in distinguishing ASLC from IIBD, even when the latter presents with acute symptoms. Distortion of the crypt architecture is the single most useful finding. Discriminant features were present in nearly 80% of the specimens from patients with IIBD in the present study.
Post-comment: Two entities that need to be considered in patients with an apparent self-limited colitis are (1) collagenous colitis and (2) microscopic colitis.
Teglbjaerg et al. have described the morphologic findings in sequential colorectal biopsy specimens obtained from 2 patients who within 1.5-2.5 years developed collagenous colitis. The initial biopsy specimens revealed an acute nonspecific inflammatory reaction in the colorectal mucosa. An intermediate stage, characterized by edema and slight fibrosis in the subepithelial region of the colorectal mucosa, preceded by final development of collagenous colitis. It is suggested that the mechanism that leads to the formation of a thick, bandlike subepithelial collagenous deposit in the colorectal mucosa may be triggered by some inflammatory or toxic stimulus.
Bo-Linn et al. have presented important new information on microscopic colitis as a cause of chronic diarrhea. They previously reported several patients with chronic idiopathic diarrhea in whom biopsies of apparently normal colon (by barium enema and colonoscopy) revealed nonspecific inflammation. Although they designated these patients as having ``microscopic colitis,'' it was uncertain whether the changes were truly abnormal and whether they were associated with abnormal colonic absorption. Bo-Linn et al. have now reported their studies in 5 patients (all women, 42-69 years of age) with microscopic colitis in whom an extensive evaluation failed to reveal any other cause of diarrhea.
``Results: First, diarrhea was present for 1/2 to 13 years prior to diagnosis. Sedimentation rates were normal. Mean daily stool output was 705 gm while eating a normal diet; during a fast, diarrhea persisted in 1 patient and disappeared in the other 4. Steatorrhea and blood were absent. Second, colon biopsies were examined in a blinded fashion. Biopsies from age-matched, healthy subjects were free of inflammation, while all biopsies from patients contained mild-to-moderate infiltrates of acute and chronic inflammatory cells. No ulcers, crypt abscesses, or granulomas were seen. The degree of inflammation was greater in proximal than distal colon. Third, by segmental intestinal perfusion studies, small bowel absorption was abnormal in 2 patients and normal in the other 3. Colon absorption was markedly reduced in all patients, as were P {right-arrow} L and L {right-arrow} P fluxes of $sup{3}$sup{6}Cl$sup{-} and $sup{2}$sup{4}Na$sup{+}; PD was normal.
``Conclusion: Microscopic colitis is a distinct cause of chronic diarrhea, characterized by (1) normal-appearing colon, (2) moderate acute and chronic inflammation of colonic mucosa, and (3) impaired colonic absorption. Colonic flux and PD data suggest impaired active ion transport with reduced permeability. Abnormal small bowel absorption in 2 patients indicates the small bowel may be involved in some patients, although jejunal biopsies revealed no definite abnormalities. Two patients improved on sulfasalazine, but 3 did not. None have developed ulcerative colitis or Crohn's disease after 1/2 to 3 years of follow-up.''--N.J.G.
Adjuvant Therapy of Colon Cancer: Results of a Prospectively Randomized Trial
N. Engl. J. Med., Mar. 22, 1984, Vol. 310, Pg. 737-743, Ch. 36-5
Abstract: In 1975, the Gastrointestinal Tumor Study Group began a prospective trial of chemotherapy with fluorouracil and semustine, immunotherapy with methanol extraction residue of bacillus Calmette-Guerin (MER), and chemoimmunotherapy in 621 patients who had undergone curative resection of modified Dukes' stage B$sub{2}, C$sub{1}, or C$sub{2} colon carcinoma. Treatment continued for 70 weeks, and the median follow-up period was 51/2 years. The same regimens of chemotherapy and immunotherapy that were used alone were used in patients randomly assigned to chemoimmunotherapy. The treatment groups were comparable in age, sex, and extent of organ invasion.
Disease-free survival is shown in Figure 36-3. Recurrences developed in 34% of patients, with no significant group differences. Comparison of the two chemotherapy regimens with the two immunotherapy regimens failed to indicate any advantage for either modality. Disease-free survival could not be related to the tumor site in the colon, age, or tumor size, but the Dukes' stage was a significant prognostic factor. Leukemia developed in 7 patients, all of whom had received chemotherapy. The median interval from the start of chemotherapy to the appearance of leukemia was 44 months. One patient died of bone marrow depression early in the course of chemotherapy. Severe skin reactions occurred in two thirds of the patients given immunotherapy.
These results fail to support the use of chemotherapy with fluorouracil and semustine, immunotherapy with MER, or combined treatment as an adjuvant program for patients at high risk of recurrent colon cancer. The ineffectiveness of the treatment was recognized only because of the availability of a prospectively selected untreated control group. More effective treatments for colon carcinoma are needed. The apparently improved prognosis for patients with Dukes' stage B$sub{2} disease raises questions regarding their inclusion in future adjuvant trials.
Post-comment: It should be noted that the Gastrointestinal Tumor Study Group (GITSC) confused Dukes' staging system by redefining the criteria for stage C$sub{1} and stage C$sub{2} lesions. In the Astler and Coller 1954 modification of Dukes' system, a C$sub{1} lesion is defined as one limited to the wall of the colon with positive nodes; a C$sub{2} lesion is defined as one through all layers of the wall with positive nodes. The GITSC defined a C$sub{1} lesion as one with one to four lymph nodes positive and a C$sub{2} lesion as one with more than four lymph nodes positive. However good this new staging system is, it is different from previous Dukes' C subclassifications, and the opportunity for comparison with previous studies is lost.
Another important issue relates to the presence of occult hepatic metastases detected by computed tomographic scanning. A recent study on the effect of occult hepatic metastases on survival after curative resection for colorectal carcinoma suggested that with patients classified as Dukes' stage C but without occult hepatic metastases, 3-year survival approximated that of patients classified as Dukes' stage B. These two points indicated that selection criteria, appropriate at the initiation of a study, may influence survival rates.
DeCosse has provided a balanced perspective of the GITSC. What follows has been abstracted from that editorial.
``Who should receive adjuvant chemotherapy for colon cancer? If you are cured by surgery, you don't need chemotherapy. The results of the GITSC study are a strong argument against adjuvant treatment of patients with stage B$sub{2} colon cancer as defined by the authors' criteria. Aside from the good results of surgery alone and the lack of benefit from adjuvant treatment, the physical, emotional, and economic costs of adjuvant chemotherapy must be factored into the equation. Here, the seven deaths from acute leukemia related to alkylating agents are especially alarming.
``Should adjuvant chemotherapy be recommended for patients with regional lymph-node involvement from colon cancer? The results of the GITSC study suggest that it should not. However, in the data presented, differences are tabulated according to the type of treatment; the possibility of benefit within subgroups delineated by stage is not examined. Only 102 patients had five or more positive lymph nodes (stage C$sub{2}), presumably placing about 25 patients in each subgroup. In a Central Oncology Group trial of adjuvant postoperative fluorouracil, benefit was seen only in patients with positive lymph nodes. In the Veterans Administration study, benefit from treatment with adjuvant fluorouracil and semustine was noted only in the patients who had from one to four positive lymph nodes. It is in this setting of more advanced regional spread from colon cancer that adjuvant chemotherapy should be tested in randomized trials.
``We need better prognostic criteria, whether in the form of staging or other histopathologic indicators, to identify the subgroups of patients who will benefit from adjuvant treatment. Some patients with measurable, recurrent large-bowel cancer surely respond to multiple-agent chemotherapy; chemotherapy should work in the adjuvant setting as well.''--N.J.G.
Perirectal Infections in Acute Leukemia: Improved Survival After Incision and Debridement
Barnes, Scott G.; Sattler, Fred R.; Ballard, James O.
Ann. Intern. Med., April 1984, Vol. 100, Pg. 515-518, Ch. 36-6
Affiliation: Pennsylvania State Univ.
Abstract: Perirectal infection in patients with acute leukemia carries a high mortality, but physicians often are reluctant to drain the lesions because of the risks of bleeding and poor wound healing. The authors evaluated surgical incision and debridement in this setting. Of 202 patients seen in a 6-year period with acute leukemia, 16 (7.9%) had perirectal infection. Thirteen had acute nonlymphocytic leukemia, and 3 had acute lymphocytic leukemia. Carbenicillin or ticarcillin and an aminoglycoside were given empirically to all febrile patients with an absolute neutrophil count below 1,000/cu mm. Specific antimicrobial therapy was with at least 2 bactericidal drugs having in vitro activity against the infecting isolates.
Eleven of the 16 patients had a history of rectal problems. Neutropenia was present for an average of 10 days before the appearance of perirectal infection (table). The earliest and most consistent physical findings were point tenderness and poorly demarcated induration. Eleven patients had positive blood cultures. On average, two enteric organisms were isolated in each patient. Spontaneous drainage by fistula formation occurred in 5 cases; 10 other patients had surgical incision and debridement of their lesions an average of 6 days after swelling and induration appeared. These 15 patients became free of pain within 48 hours and were afebrile after an average of 4 days. Two patients required a diverting colostomy. Two patients with drained lesions remained hospitalized and died in an early stage of hematologic remission. An undrained patient died in septic shock.
Patients in this study with acute leukemia and perirectal infection appeared very ill. Incision and drainage were carried out because of persistent fever and failure of the lesion to drain spontaneously. Appreciable operative bleeding did not occur; there were no intraoperative or postoperative complications. Pain and bacteremia promptly subsided after surgical drainage, and fever rapidly resolved. Early surgery may be lifesaving in septic, granulocytopenic patients, in whom perirectal infection progresses rapidly, particularly if there is genital or peritoneal involvement. Surgery may not be necessary if spontaneous fistula formation occurs promptly.
Post-comment: Two key points are evident from this article: (1) perirectal infections are fairly common in patients with acute leukemia; and (2) early incision and debridement can be lifesaving in those with rapidly progressive perirectal lesions, especially those with genital involvement or peritoneal signs.--N.J.G.
Clinical and Prognostic Differences in Fulminant Hepatitis Type A, B, and Non-A Non-B
Gimson, A. E. S.; White, Y. S.; Eddleston, A. L. W. F.; Williams, Roger
Gut, December 1984, Vol. 24, Pg. 1194-1198, Ch. 37-4
Affiliation: London
Abstract: The authors reviewed the serologic findings in 73 consecutive patients seen in 1977-1981 with fulminant viral hepatitis. Grade IV encephalopathy was present at admission or developed subsequently in each case, and the first signs of encephalopathy appeared within 8 weeks of the onset of illness. Forty-four patients underwent hemodialysis, and 19 had charcoal hemoperfusion with prostacyclin infusion.
Serologic evidence of hepatitis A was found in 31.5% of the patients, and of hepatitis B in 24.7%, leaving 43.8% with presumed non-A, non-B hepatitis. The three groups were similar in age and sex distribution. Five patients with hepatitis A had had recent contact with jaundice or had recently traveled abroad. Five with hepatitis B were known drug addicts, 3 may have had contact with hepatitis B at work, and 2 had possible sexual exposure. In only 1 non-A, non-B case was a possible source of infection identified. Fever was more common in the hepatitis A group than in the other groups. The duration of illness before encephalopathy developed was longest in the non-A, non-B group. More severe hepatic dysfunction was present in the hepatitis B group. Cerebral edema was least frequent in the hepatitis A group. Survival data are given in the table. The low survival rate of patients with a relatively long duration of illness before encephalopathy developed was related to a higher proportion of patients having non-A, non-B hepatitis.
Non-A, non-B infection was the most common serologic type of fulminant viral hepatitis in this series. Patients with hepatitis A had the least severe illness. Maximum prolongation of the prothrombin time (58 seconds vs. 150 seconds) and the incidence of cerebral edema (16.6% vs. 43.4%) were significantly less in the hepatitis A group than in the hepatitis B group. Clinical differences between the various serologic types of disease may reflect differences in the mechanisms underlying the initiation and progression of hepatic necrosis, as well as the potential for hepatic regeneration. The duration of symptoms before encephalopathy appears does not appear to be a significant prognostic factor within each serologic group of cases.
Post-comment: In this series, non-A, non-B infection was the most frequent type of fulminant viral hepatitis accounting for approximately 44% of the cases. The differences in presentation and clinical course between fulminant hepatitis types A, B, and non-A, non-B are of considerable interest. Thus, patients in the hepatitis A group had a less severe illness as shown by the lower prothrombin time, reduced occurrence of cerebral edema, and better survival than either the hepatitis B or non-A, non-B hepatitis groups. The median duration of illness before the development of encephalopathy was significantly longer in the non-A, non-B group compared with the hepatitis A and hepatitis B groups (21 days vs. 10 days vs. 7 days). The severity of liver dysfunction, judged by the prolongation of prothrombin time, was significantly greater in the hepatitis B group. Overall survival was highest in the patients with hepatitis A and lowest in the non-A, non-B hepatitis group.--N.J.G.
Spontaneous Reactivation of Chronic Hepatitis B Virus Infection
Davis, Gary L.; Hoofnagle, Jay H.; Waggoner, Jeanne G.
Gastroenterology, February 1984, Vol. 86, Pg. 230-235, Ch. 37-5
Affiliation: Natl. Inst. of Health
Abstract: The authors observed recrudescence of chronic hepatitis disease activity in 8 of 25 patients with chronic type B hepatitis who lost hepatitis B e antigen (HBeAg). The patients were initially symptomatic and had biopsy-documented chronic type B hepatitis. They were followed for a mean of 25 months after the disappearance of HBeAg, hepatitis B virus DNA, and DNA polymerase activity from the serum. The 23 men and 2 women had a mean age of 40 years. Infection was attributed to homosexual contact in 14 cases, heterosexual contact in 4, and drug use, transfusion, and medical exposure in 1 case each. Fourteen patients had received steroids, and 6 had received adenine arabinoside.
Twenty-four patients developed antibody to HBeAg (anti-HBe) during follow-up, and all had a decrease in serum aminotransferase levels. Eight patients (table) had reactivation of hepatitis with an abrupt rise in serum aminotransferase level and the reappearance of serum markers. In 7 of the 8 cases, HBeAg reappeared. Three patients had been minimally symptomatic before reactivation. Reactivation occurred a mean of 6.7 months after the loss of HBeAg. All episodes of reactivation were clinically severe. Three patients became jaundiced, 3 developed ascites, and 2 had variceal bleeding. The episodes abated after 4-12 months in 5 instances, but persisted for more than a year in 2 others. One patient died after 13 months of reactivated disease. The patients with reactivation were older than the others, had a longer duration of chronic hepatitis, and more often had cirrhosis. No differences in treatment or presumed source of infection were apparent.
The loss of HBeAg and seroconversion to its antibody do not necessarily indicate a permanent remission of chronic type B hepatitis. Spontaneous reactivation of chronic hepatitis B virus infection may be a significant cause of progressive liver injury. Reactivation should be considered whenever an acute exacerbation of disease activity occurs in a previously stable HBsAg carrier who shows seroconversion to anti-HBe.
Post-comment: This article, as well as article 37-6, calls attention to the problem of spontaneous reactivation of chronic hepatitis B virus (HBV) infection. The reactivation of chronic type B hepatitis by various forms of chemotherapy and immunosuppressive agents has been well documented. Reappearance of HBV replication in these patients seems to be related to depression of host immunity. Subsequent discontinuation of the immunosuppressive agent(s) allows the return of immunocompetence and may precipitate an acute hepatitis, which is associated with clearance of markers of HBV replication. In contrast, in spontaneous reactivation, as seen in the two studies summarized here, acute elevations of serum aminotransferase levels occurred in conjunction with the appearance of markers of HBV replication (HBeAg). The occurrence of spontaneous reactivation of chronic type B hepatitis suggests that actively replicating HBV, like some other DNA viruses (herpes simplex, varicella-zoster, cytomegalovirus), may become latent. The occurrence of spontaneous reactivation implies the continued potential for recrudescence of active disease.
Exacerbation of disease activity could not be attributed to concurrent infection with HBV of a different subtype, infection with the hepatitis A virus, or infection with the delta-agent. Antibodies to the latter two agents did not develop during the episodes of reactivation. Furthermore, detailed questioning of all patients regarding use of medication, concurrent illness, alterations of diet, unusual emotional or physical stress, and exposure to toxins or potential sources of hepatitis viruses was unrevealing. Thus, spontaneous reactivation should be considered whenever an acute exacerbation of hepatitis disease activity occurs in a previously stable anti-HBe-positive, chronic HBsAg carrier.--N.J.G.
Postexposure Prophylaxis of Hepatitis B: Recommendations of the Immunization Practices Advisory Committee
Centers for Disease Control
Ann. Intern. Med., September 1984, Vol. 101, Pg. 351-354, Ch. 37-8
Abstract: Prophylaxis against hepatitis B infection after exposure is properly considered in several settings. The finding that hepatitis B vaccine combined with hepatitis B immune globulin (HBIG) is highly effective in preventing chronic infection in infants of hepatitis B surface antigen (HBsAg)-positive mothers has necessitated revision of recommendations for postexposure prophylaxis (table). A combination of hepatitis vaccine and a single dose of HBIG produces immediate, sustained high levels of protective antibody to HBsAg. Administration of both vaccine and HBIG to the exposed infant is about 90% effective, but about 5% of perinatal infections may occur in utero. The concurrent use of HBIG and vaccine does not seem to impair the efficacy of the vaccine. Because HBIG must be given on the day of birth, identification of HBsAg-positive mothers is necessary before delivery. Injection of HBIG at birth should not interfere with vaccines given at age 2 months.
It is also reasonable to recommend administration of both hepatitis B vaccine and HBIG after percutaneous or mucosal exposure to hepatitis B virus in health-care workers and others. A second dose of HBIG is not considered necessary if vaccine is administered. Further, HBIG should be given within 24 hours of exposure if possible, and vaccine should be given within a week of exposure. Persons who refuse vaccine may be given two doses of HBIG.
Sexual contacts of acutely infected persons are at an increased risk of acquiring hepatitis B infection. Sex partners should be prescreened for susceptibility before HBIG is recommended if prescreening does not delay treatment for more than 2 weeks after the last exposure. A single dose of HBIG is adequate, with a second dose given if the index patient remains HBsAg-positive 3 months after detection. If the index patient is a known carrier or remains HBsAg-positive for 6 months, regular sexual contacts should be offered hepatitis B vaccine. Homosexual men who are exposed should receive vaccine when HBIG is given after exposure to the disease.
Post-comment: An excellent review of postexposure prophylaxis of hepatitis B appeared in Morbidity and Mortality Weekly Report. What follows are some highlights from that review.
``With the development of [hepatitis B] HB vaccine, the possibility arose that HB vaccine, alone or in combination with HBIG, might be useful for postexposure prophylaxis. Studies have shown that response to HB vaccine is not impaired by concurrent administration of HBIG and that the combination of HB vaccine and one dose of HBIG produces immediate and sustained high levels of protective antibody to the hepatitis B surface antigen (anti-HBs).
``There are no prospective studies directly testing the efficacy of a combination of HBIG and HB vaccine in preventing clinical HB following percutaneous or mucous membrane exposure to HBV. However, since health-care workers at risk to such accidents are candidates for HB vaccine and since combined HBIG plus vaccine is more effective than HBIG alone in perinatal exposures, it is reasonable to recommend both HB vaccine and HBIG after such exposure. This combination will provide prolonged immunity to subsequent exposures and may also increase efficacy in preventing HB in such postexposure situations. In addition, because the second dose of HBIG is not considered necessary if the vaccine is used, the cost of combination treatment is usually less than that of two HBIG doses alone. If exposure to blood occurs in situations where the HBsAg status of the blood is unknown, refer to `Immune Globulins for Protection against Viral Hepatitis.' If HBsAg testing reveals the source of the blood to be positive, the following treatment schedule should be instituted as soon as possible.
"For percutaneous (needlestick), ocular, or mucous membrane exposure to blood known to contain HBsAg and for human bites from HBsAg carriers that penetrate the skin, a single dose of HBIG (0.06 ml/kg or 5.0 ml for adults) should be given as soon as possible after exposure and within 24 hours if possible. Hepatitis B vaccine 1 ml (20 mu g) should be given intramuscularly at a separate site as soon as possible, but within 7 days of exposure, with the second and third doses given 1 month and 6 months, respectively, after the first. If HBIG is unavailable, immunoglobulin (IG [formerly ISG or ``gamma globulin]) may be given in an equivalent dosage (0.06 ml/kg or 5.0 ml for adults). If an individual has received at least two doses of HB vaccine before an accidental exposure, no treatment is necessary if serologic tests show adequate levels (>10 S/N by RIA) of anti-HBs. For persons who choose not to receive HB vaccine, the previously recommended two-dose HBIG regimen may be used.
``Sexual contacts of persons with acute HB infection are at increased risk of acquiring HB infection. Two published studies have assessed the value of post-exposure prophylaxis for regular sexual contacts of persons with acute HB infection. One showed that HBIG was significantly more effective than IG that contained no measurable anti-HBs in preventing both HB infection and clinical illness. The second study, however, showed comparable disease rates in persons receiving HBIG and IG containing the increased levels of anti-HBs found in currently available lots. Because data are limited, the period after sexual exposure during which HBIG is effective is unknown, but extrapolation from other settings makes it unlikely that this period would exceed 14 days. The value of HB vaccine alone in this setting is unknown. However, since about 90% of persons with acute HB infections become HBsAg-negative within 15 weeks of diagnosis, the potential for repeated exposure is usually self-limited. Hepatitis B vaccine is not routinely recommended for such exposures.
``Prescreening sexual partners for susce susceptibility before HBIG treatment is recommended if it does not delay HBIG administration beyond 14 days after last exposure. In one study, 27% of regular sexual partners (heterosexual) were positive for HBsAg or anti-HBs at the time they presented for evaluation. Among homosexually active males, over 50% have markers indicating prior infection, and 5%-6% are HBsAg positive. Testing for anti-HBc is the most efficient prescreening test to use in this population group.
``A single dose of HBIG (0.06 ml/kg or 5 ml for adults) is recommended for susceptible individuals who have had sexual contact with an HBsAg-positive person if HBIG can be given within 14 days of the last sexual contact, and for persons who will continue to have sexual contact with an individual with acute HB before loss of HBsAg in that individual. In exposures between heterosexuals, a second HBIG dose should be given if the index patient remains HBsAg-positive 3 months after detection. If the index patient is a known hepatitis B virus carrier or remains HBsAg-positive for 6 months, HB vaccine should be offered to regular sexual contacts. For exposures among homosexual men, the HB vaccine series should be initiated at the time HBIG is given following a sexual exposure, since HB vaccine is recommended for all susceptible homosexual men. Additional doses of HBIG are unnecessary if vaccine is given. Because current lots of IG contain anti-HBs, it remains an important alternative to HBIG when HBIG is unavailable.''--N.J.G.
Hepatitis B Viral Nucleotide Sequences in Non-A, Non-B or Hepatitis B Virus-Related Chronic Liver Disease
Abstract: The Southern blot hybridization assay for hepatitis B virus (HBV) nucleotide sequences in hepatocellular DNA was used to examine liver biopsy specimens from 19 hepatitis B surface antigen (HBsAg)-negative patients with chronic liver disease. Hepatotoxic drug use and metabolic causes of chronic liver disease were excluded. Six patients had severe beta-thalassemia requiring frequent packed red blood cell transfusions. Five patients were addicted to parenteral drugs. Three had liver disease associated with alcohol abuse. Four patients had documented chronic hepatitis of unknown origin, and 1 had thalassemia intermedia. Free forms of HBV DNA were detected by analyzing the hepatocellular DNA without hydrolysis by restriction endonucleases.
The findings are given in the table. Four patients who were seronegative for antibodies to HBsAg, to core antigen, and to e antigen were classed as having non-A, non-B hepatitis. The other patients were considered to have hepatitis B. The histologic diagnosis was chronic persistent hepatitis in most patients. Five patients had chronic active hepatitis, 2 had micronodular cirrhosis, and 1 had fatty liver. A single band corresponding with the monomeric form of HBV DNA was found on autoradiography in 5 cases, several bands of larger forms in 3, and both the monomeric form and the larger forms in 8. No HBV DNA was found in 3 cases. Hepatitis B virus DNA was found in the hepatocellular DNA of 6 patients who underwent splenectomy, but hybridization did not occur with DNA extracted from the spleens.
Hepatitis B virus-like agents are found in the livers of HBsAg-negative patients with chronic liver disease. Immunologically distinct but genetically related viral agents may be involved in hepatitis B and chronic non-A, non-B hepatitis.
Post-comment: Intranuclear aggregates of spherical and tubular particles, first described by Shimizu et al., have been repeatedly demonstrated in acute and chronic non-A, non-B hepatitis in man. A key issue concerns the specificity of these intranuclear particles as a marker of parenterally acquired non-A, non-B hepatitis infection. The reader is referred to articles both supporting and not supporting this concept.--N.J.G.
Clinical Study of 86 Cases of Idiopathic Portal Hypertension and Comparison With Cirrhosis With Splenomegaly
Gastroenterology, April 1984, Vol. 86, Pg. 600-610, Ch. 37-14
Abstract: The authors reviewed the findings in 86 cases of idiopathic portal hypertension (IPH), the equivalent of hepatoportal sclerosis in the United States and of noncirrhotic portal fibrosis in India. It is characterized by splenomegaly, anemia, and portal hypertension in the absence of hepatic cirrhosis and portal vein or hepatic vein occlusion. The patients, seen in 1965-1982, were compared with 63 patients with cirrhosis and splenomegaly and 80 patients with cirrhosis without splenomegaly.
The study group included 70 women and 16 men with average ages of 46 and 36 years, respectively. Anemia was the most common presentation. The spleen exceeded 1 kg in weight in three fourths of the 57 evaluable cases. The liver usually was not enlarged. Most patients had moderate normochromic normocytic anemia. Liver enzyme values were normal in most cases. Esophageal varices were freqently present. Scintigraphy generally showed a large spleen, a somewhat small liver, and minimal uptake of activity by the bone marrow. Portographic findings were consistently abnormal, showing a paucity of middle-size portal vein branches in the liver, absent or cut-off peripheral branches, and an avascular area in the immediate subcapsular area. Venography showed frequent vein-to-vein anastomoses. Pressure measurements in IPH are compared with those in cirrhosis in the table. Histologic studies showed portal fibrosis of varying degrees in all livers. Survival at 10 years was 77% for patients with IPH and 29% for all patients with cirrhosis. The mean survival from the time of onset of disease was 25 years for patients with IPH and 6 years for patients with cirrhosis. Alcohol abuse was more frequent in the cirrhotic groups. There were some hematologic similarities between IPH and cirrhosis with splenomegaly, but the basic disease processes appeared to be distinctly different.
Idiopathic portal hypertension can be distinguished from hepatic cirrhosis clinically and histopathologically. The prognosis of IPH is much better than that of cirrhosis. The cause of marked splenomegaly, which does not seem to be simply congestion, remains an enigma. The relation of IPH to extrahepatic portal obstruction remains to be elucidated.
Post-comment: This is a useful review of idiopathic portal hypertension. It should be noted that several different disorders can mimic the clinical picture of idiopathic portal hypertension. These include (1) reduction in intrahepatic portal venous radicles, i.e., intrahepatic portal venopathy; (2) sclerosis of portal venules, or hepatoportal sclerosis; (3) partial nodular transformation of the liver; (4) vitamin A intoxication; and (5) portal hypertension associated with myeloproliferative disorders such as myelofibrosis.--N.J.G.
Human Liver Transplantation: Analysis of Data on 540 Patients From Four Centers
Abstract: In this study of liver transplants performed at 4 centers in the United States and Western Europe it was found that 25% (139) were performed for neoplastic disease; 1- and 3-year survival in patients with neoplasms were about 26% and and 12%, and survival differed little for transplants performed before or after January 1, 1980 (Fig 37-6). For the 401 patients who received transplants for nonneoplastic disease, a median survival of 3.0 months was calculated, with the probability of survival being about 48% at 3 months, 39% at 1 year, and 30% at 3 years. However, overall survival for these patients was better if they received a transplant after January 1, 1980 (Fig 37-7).
Among the 44% of patients who received a transplant for end-stage cirrhosis, 3-year survival was lowest for those with alcoholic cirrhosis (20%). Three-year survival was greater for patients with nonalcoholic cirrhosis (29% overall), but it did not differ markedly among the various subtypes and was greater for patients who received a transplant after January 1, 1980 (42%) compared with 22% for those who had an earlier transplant. Patients with biliary atresia, sclerosing cholangitis, and metabolic and miscellaneous disorders constituted the remaining 30% of patients; 3-year survival varied from 20% to 44% overall for the various subgroups, with a consistent trend toward improved survival among patients who received a transplant after January 1, 1980.
A total of 33 different diagnoses were represented among the 540 patients. For these analyses, only the primary diagnoses were considered. Tumors were disproportionately represented in 2 of the 4 centers; neonatal cholestatic disorders, sclerosing cholangitis, and metabolic disorders were disproportionately represented at one center. Nonalcoholic cirrhosis was a frequent diagnosis at all centers and was the most common diagnosis overall (39% of all patients). Fifty-two patients (nearly 10%) received a second liver transplant owing to failure of the first. Survival in this group was less than for patients receiving only one transplant, with the 1-year survival probability being 38% and there being no survivors beyond 3 years.
Among 23 patients who survived 4-12 months after transplantation, the mean total number of weeks of hospitalization after the procedure was 12.6; among 22 patients who survived at least 1 year after transplantation, the mean number of weeks of hospitalization after the procedure was 9.2. Among 184 patients at 2 centers who were still living at least 4 months after transplantation and for whom data were available, 179 were listed in good condition (fully rehabilitated); only 3 were listed in fair condition (able to care for self) and 2 were in poor condition. A recent survey of transplant recipients at one center indicated that more than 80% of patients who survived longer than 1 year had resumed their former occupation or activities. Thus, as assessed by hospitalization time and functional status, the quality of life for transplant recipients who survived at least 4 months appears to be good.
The survival of patients who receive a transplant because of neoplastic disease is less than that of patients who receive a transplant because of nonneoplastic disease. Among patients with nonneoplastic disease, survival is better for those who received a transplant after January 1, 1980, as compared with before, and this improvement may not be wholly attributable to the more recent use of cyclosporine.
Post-comment: The National Institutes of Health on June 20-23, 1983, convened a Consensus Development Conference on Liver Transplantation. After 2 days of expert presentation of the available data, a Consensus Panel consisting of hepatologists, surgeons, internists, pediatricians, immunologists, biostatisticians, ethicists, and public representatives considered the offered evidence to arrive at answers to the following key questions: (1) Are there groups of patients for whom transplantation of the liver should be considered appropriate therapy? (2) What is the outcome (current survival rates and complications) in different groups? (3) In a potential candidate for transplantation, what are the principles guiding selection of the appropriate time for surgery? (4) What are the skills, resources, and institutional support needed for liver transplantation? (5) What are the directions for future research?
With regard to question 1, the panel noted that prolongation of life of good quality for patients who would otherwise have died has been reported in the following conditions: extrahepatic biliary atresia, chronic active hepatitis, primary biliary cirrhosis, inborn errors of metabolism, hepatic vein thrombosis, sclerosing cholangitis, primary hepatic malignancy confined to the liver, and alcohol-related liver cirrhosis and alcoholic hepatitis.
In considering question 2, the panel found that although sufficient data for thorough assessment of liver transplantation are not available to date, certain trends appear to emerge. Patients currently being accepted for transplantation have a high probability of imminent death and a low quality of life in the absence of transplantation. Patients undergoing transplantation have an operative mortality (within 1 month) of 20%-40%. One-year survival among transplant recipients since 1980 is favorable when compared with their expected course in the absence of transplantation. Since 1980, 1-year survival appears improved over the earlier transplant experience. Individual patients have survived for many years with good quality of life after transplantation. Data are insufficient to evaluate survival rates beyond 1 year following transplantation with current technologies. Short-term quality of life is probably enhanced in many transplant survivors. We lack systematically gathered information on quality of life among long-term survivors.
As to the principles guiding selection of the appropriate time for surgery, the panel stated that although no single best time for surgery can be specified, transplantation should be reserved for patients in any of the following phases of disease: when death is imminent, when irreversible damage to the central nervous system is inevitable, or when quality of life has deteriorated to unacceptable levels. The exact choice of the time for liver transplantation in an individual requires the judgment of a qualified medical team and a well-informed patient.
After extensive review and consideration of all available data, the panel concluded that liver transplantation is a therapeutic modality for end-stage liver disease that deserves broader application. However, in order for liver transplantation to gain its full therapeutic potential, the indications for and results of the procedure must be the object of comprehensive, coordinated, and ongoing evaluation in the years ahead. This can best be achieved by expansion of this technology to a limited number of centers where performance of liver transplantations can be carried out under optimal conditions.--N.J.G.
Fecal Fat Concentration in Patients With Steatorrhea
Bo-Linn, George W.; Fordtran, John S.
Gastroenterology, August 1984, Vol. 87, Pg. 319-322, Ch. 39-1
Affiliation: Baylor Univ.
Abstract: Patients with gastrointestinal tract disease would be expected to have more severe diarrhea from a given degree of steatorrhea than those with pancreatic insufficiency because of the gastrointestinal disorder itself and because of higher intraluminal concentrations of hydrolyzed lipid, leading to greater fecal losses of water. This hypothesis was examined in 19 patients with steatorrhea caused by exocrine pancreatic insufficiency, usually chronic pancreatitis, and 31 with steatorrhea caused by gastrointestinal tract disease. Ten of the latter patients had gastric resection, 8 had celiac sprue, and 6 had Crohn's disease. Steatorrhea was considered to be present when more than 5 gm of fat was excreted in 24 hours.
When the severity of diarrhea, in terms of stool wet weight, was related to the severity of steatorrhea, there was much overlap in the severity of both conditions in the two patient groups. For the same degree of steatorrhea, however, the patients with gastrointestinal tract disease usually had more severe diarrhea than those with pancreatic insufficiency had.
Fecal fat concentration may be a useful clue to the etiology of steatorrhea in adults. Values exceeding 9.5% in a patient with steatorrhea of 21 gm or more daily suggest pancreatic insufficiency. Patients with steatorrhea due to low levels of conjugated bile salts in the duodenum but without bile salt malabsorption, as in cholestasis and intestinal bacterial overgrowth syndromes, may also have high fecal fat concentrations. The fecal fat level may be misleadingly elevated by the use of opiates and other antidiarrheal agents that increase water absorption in the bowel without improving fat absorption.
Post-comment: A useful new application of an old test. Fecal fat concentration may be a useful clue in distinguishing pancreatic steatorrhea from steatorrhea due to gastrointestinal disease.--N.J.G.
Course and Outcome of Chronic Pancreatitis: Longitudinal Study of a Mixed Medical-Surgical Series of 245 Patients
Ammann, R. W.; Akovbiantz, A.; Largiader, F.; Schueler, G.
Gastroenterology, May 1984, Vol. 86, Pg. 820-828, Ch. 39-4
Affiliation: Zurich
Abstract: The natural history of chronic pancreatitis (CP) is ill-defined. A number of longitudinal studies have been reported, but most are based on retrospective analysis of predominantly small series. Some recent studies are surgical series, which only partially reflect the natural history, in that many patients with CP need no pancreatic surgery. From July 1963 to June 1981, 245 patients with chronic pancreatitis were prospectively studied at regular intervals with particular regard to pain, pancreatic functions, calcifications, pancreatic surgery, and survival. Of these patients, 163 had relapsing alcohol-induced CP (AICP), and 145 of these 163 had calcific pancreatitis. The median period of observation in the group with alcoholic relapsing calcific pancreatitis was 10.4 years. In this group of 145 patients, 85% experienced lasting pain relief within a median time of 4.5 years from onset. A gradual increase of pancreatic calcifications and dysfunction was observed with increasing duration of disease. Pain relief was accompanied by a marked increase in pancreatic dysfunction and calcification. Of 163 patients with AICP, 87 (53%) needed no pancreatic surgery. Seventy-six patients (47%) had recurrent or persistent severe pain, mainly due to pseudocysts (56 patients). Of these 76, 22 underwent a cyst drainage procedure; 4 had papillotomy; 24, distal pancreatectomy; and 26, Wirsungo-jejunostomy.
The proportion of patients experiencing lasting pain relief was similar in both patients who were operated on and those who were not. In both groups, lasting relief from pain was correlated with duration of disease and was associated with marked pancreatic dysfunction (Fig 39-2). The 50% survival time in AICP (with or without pancreatic surgery) was 20-24 years after onset, thus markedly shorter than in nonalcoholic pancreatitis. Of the 245 patients, 86 died. About 20% of deaths were related to pancreatitis and its complications. Most extrapancreatic causes of death were malignancies, cardiovascular diseases, severe infections, and nonpancreatic surgery.
Indication of surgery in uncomplicated AICP (without pseudocysts or cholestasis) is debated. The selection of patients and choice of surgical procedure differ in various series for lack of definite criteria. The present group of uncomplicated AICP with surgery is too small for proper evaluation of results. (Only 7 of the 20 patients with uncomplicated AICP retrospectively fulfilled criteria for surgery.) However, the results of surgery in the entire series regarding lasting pain relief seem to be relevant. The timing of surgery seems to influence the results in AICP. About 50% of all operations were performed within 3 years of onset. The rate of prompt and lasting pain relief after surgery was about 65% in patients operated late (more than 3 years from onset) as compared with 35% in early operations. The data suggest that the probability of obtaining lasting pain relief after surgery tends to increase with the duration of AICP.
A problem in pain-relieving surgery of CP is the failure rate of 25%-35% regardless of the type of surgery. In the present series, 36 of 69 patients had pain relapses postoperatively. However, 26 had spontaneous permanent pain relief within an average of 2.9 years after surgery associated with marked deterioration of panceratic function (and increased calcification) in most instances. Reduction of alcohol intake was achieved in less than 25% of these patients. A close parallel gradual increase in the relative number of patients obtaining lasting pain relief is seen in surgical and nonsurgical groups. These data add substantial evidence to the hypothesis that lasting pain relief after surgery is not solely due to the surgical intervention per se.
Post-comment: An interesting and important study which emphasizes that surgery, as a means of pain elimination in patients with chronic pancreatitis, has to be evaluated critically by additional prospective long-term studies.--N.J.G.
Carcinoma of the Head of the Pancreas: Therapeutic Implications of Endoscopic Retrograde Cholangiopancreatography Findings
Nix, G. A. J. J.; Schmitz, P. I. M.; Wilson, J. H. P.; Van Blankenstein, M.; Groeneveld, C. F. M.; Hofwijk, R.
Gastroenterology, July 1984, Vol. 87, Pg. 37-43, Ch. 39-9
Affiliation: Erasmus Univ., Rotterdam, The Netherlands
Abstract: An attempt was made to determine whether endoscopic retrograde cholangiopancreatography (ERCP) can improve the selection of patients with cancer of the head of the pancreas for curative surgery. The 123 patients on whom data were reviewed included 69 men and 54 women whose mean age was 65 years. Several patients also had echocardiography, angiography, percutaneous transhepatic cholangiography, or a combination of these. Weight loss was the most common presenting symptom. Jaundice was a late feature. The first diagnostic procedure was ERCP in about three fourths of cases. Disease was localized to the head of the pancreas in 41% of cases. Metastases were found in 48%. Patients were grouped according to maximal tumor diameters of 2.5-4.0 cm, 4.5-6.0 cm, and 7.0-15.0 cm.
Fifty-five percent of cases were primarily inoperable, and another 23% of patients did not have curative resection, most often because of advanced age, poor general condition, or cardiovascular disease. The mean interval from onset of symptoms to treatment was 41/2 months. All 13 patients who were not operated on and all 13 who underwent exploration only died within 16 months. Three of 70 patients treated palliatively were alive at 19 months. Survival after the Whipple operation in 15 patients was 23% at 35 months, and that after total pancreatectomy in 11 patients was 22% at 4 years. Smaller tumors were more often operable (Fig 39-7). Mean tumor diameter decreased during the 10-year study. Tumors less than 3 cm in diameter were regularly resectable, whereas those more than 8 cm in diameter were seldom resectable.
The overall prognosis for patients with cancer of the head of the pancreas is poor, but encouraging trends are evident. Further improvement may be achieved by more intensive use of ERCP in symptomatic patients, but a less invasive yet equally sensitive means of detecting small cancers is urgently needed.
Post-comment: Savarino et al. undertook a study to determine whether new diagnostic tools (angiography, endoscopic retrograde cholangiopancreatography, ultrasonography, and computed tomography) have improved the detection of pancreatic cancer at a resectable stage. Two groups of patients with histologically ascertained cancer of the pancreas were examined. The first consisted of 54 cases investigated before the clinical application of the four above-mentioned techniques and the second of 74 cases investigated thereafter. Cancer resectability was judged by a surgical team at the time of laparotomy according to preestablished criteria. No significant difference was found between the number of resections and other procedures in the two groups. On the other hand exploratory laparotomies performed exclusively for diagnostic purposes were significantly (P<.01) reduced in the second group. On the basis of these results one may conclude that new diagnostic aids are useful and reliable in the preoperative detection of pancreatic cancer, but their introduction into clinical practice has not permitted an earlier diagnosis of the disease, and therefore there has been no important increase in resectability.--N.J.G.
How Effective Is External Pituitary Irradiation for Growth Hormone Secreting Pituitary Tumors?
Feek, C. M.; McLelland, J.; Seth, J.; Toft, A. D.; Irvine, W. J.; Padfield, P. L.; Edwards, C. R. W.
Abstract: Conventional external pituitary irradiation was evaluated in 46 patients with GH-secreting tumors. Radiotherapy was delivered after pituitary surgery in 16 cases, and primarily in 30 cases. A 4-MeV linear accelerator was used to deliver a total dose of 3,750 rad in 15 fractions through two opposed fields. Suprasellar extension was present in 9 of the patients who were operated on, and it caused visual field defects in 8 of them. The mean serum GH level was higher in these patients, and they were younger than the others at presentation.
The mean serum GH level in the patients having primary radiotherapy was 74 ng/ml at baseline and fell by 28% per year over 0-5 years after treatment, and by 16% per year over 0-20 years. In the surgical group, the mean serum GH level was 265 ng/ml at baseline, fell by 76% in the first year, and then declined by 30% per year over 1-5 years, and by 16% per year over 1-20 years of follow-up. A life-table analysis is shown in Figure 40-4. Anterior pituitary function decreased progressively over 10 years after radiotherapy. Gonadotropin function was lost in nearly half the patients, corticotropin in 30%, and thyrotropin in 16% in patients having radiotherapy alone. The respective figures for the patients having combined treatment were 70%, 54%, and 38%. No extrasellar extension was observed after radiotherapy. One patient had marked deterioration in visual acuity from presumed tobacco amblyopia. No major side effects or deaths were attributed to radiotherapy.
Large GH-secreting tumors probably require surgery, radiotherapy, and dopamine agonist therapy. Microsurgery can rapidly reduce the serum GH level in patients with microadenomas while normal anterior pituitary function is preserved. Radiotherapy, however, appears to be quite safe and can be recommended if skilled neurosurgery is unavailable. It also can be used as an adjunct where surgery has failed, and perhaps in older patients in whom the need to reduce the serum GH may be less urgent.
Factors in Predicting Outcome From Operation in Patients With Prolactin-Secreting Pituitary Adenomas
Nelson, Paul B.; Goodman, Michael; Maroon, Joseph C.; Martinez, A. Julio; Moossy, John; Robinson, Alan G.
Neurosurgery, December 1981, Vol. 13, Pg. 634-641, Ch. 40-7
Affiliation: Univ. of Pittsburgh
Abstract: It has been found that patients with PRL-secreting pituitary tumors classified as Hardy grades I and II (tumors less than 1 cm) and preoperative levels of PRL lower than 100-200 ng/ml are those most likely to be cured by operation. To determine what additional factors might be predictive of surgical outcome, the authors retrospectively analyzed the findings in 40 consecutive patients who underwent operation for presumed PRL-secreting pituitary adenomas.
In the first 3 months after operation, 25 patients had a normal level of PRL (30 ng/ml or less), and 15 had persistent disease. Nine of the 25 who had normal postoperative levels were found to have elevated levels more than 3 months after operation. However, postoperative levels of PRL were considerably lower than preoperative levels in patients with persistent and recurrent disease.
Patients with tumors classified as Hardy grades I and II had a significantly better outcome than those with tumors larger than 1 cm. Preoperative levels of PRL also were significantly lower in patients with successful surgical outcomes (mean, 204 ng/ml) than in those with operative failures (mean, 524 ng/ml). Patients aged 26 years or younger and those who were symptomatic for 6 years or less were most likely to be cured.
Of 23 patients with amenorrhea for 6 years or less, 14 had long-term cures, compared with only 2 of 21 patients with amenorrhea for more than 6 years. Ten of 15 patients aged 26 years or younger at time of operation had long-term cures, compared with only 6 of 29 older than age 26. Surgical outcome was significantly better in patients with normal responses to stimulation of GH than in those with a blunted response before operation.
The findings suggest that PRL-secreting pituitary tumors may cause a progressive disorder for which an operative cure may be realized only during the early phase of the disease.
Post-comment: Although these authors believe that prolactinomas are naturally progressive, this opinion is not universally shared. In a group of 25 women who had had symptoms and signs of prolactinomas for an average of 11.5 years, the disease usually showed no evidence of progression. Of 22 presenting with amenorrhea, 7 resumed menses spontaneously. Galactorrhea resolved spontaneously in 6 of 19 patients originally complaining of this disorder. Only 1 patient had minor evidence of progressive impairment of visual fields. The mean plasma PRL level fell spontaneously from 225 to 155 ng/ml. Apparently, untreated hyperprolactinemia in women with tumors less than 1.0 cm in diameter is usually a benign, nonprogressive disease even without treatment. On the other hand, surgical removal of the tumor does not guarantee cure. Of 29 patients with microadenomas followed for 4 years after their plasma PRL concentrations had returned to normal postoperatively, 5 developed recurrent hyperprolactinemia within 6-16 months. This record is not appreciably better than that of 5 patients with macroadenomas whose plasma PRL levels were normal postoperatively. Only 1 of this group had a relapse. It does not seem to me that these data present irrefutable arguments for the surgical treatment of the disease.
Men with prolactinomas have reduced sexual activity, manifested by reduced nocturnal penile tumescence, oligospermia, asthenospermia, teratospermia, elevated semen fructose, and decreased serum LH and testosterone concentrations. In 6 of 8 patients followed after treatment of the tumor with surgery or dopaminergic drugs, serum PRL levels returned to normal within 6 months. The serum LH concentration increased as early as 2 months, testosterone levels rose significantly as early as 3 months, and normal levels were found after 6-8 months. In spite of this, only 2 patients demonstrated a normal semen analysis. It is not clear whether further recovery of semen parameters will occur after longer periods of normal PRL secretion.
It is important to avoid unnecessary and expensive tests in patients in whom the answer is clear without them. In a study of 95 women with hyperprolactinemia, it was found that no patient with a serum baseline level of PRL in excess of 60 ng/ml had a normal response to TRH administration. Moreover, of those with baseline levels between 20 and 60 ng/ml, only those with an abnormal response to TRH proved to have tumors on subsequent computed tomographic or polytomographic scanning of the pituitary. The authors therefore believe that there is no point in doing a TRH test in patients with a baseline serum PRL level greater than 60 ng/ml or in doing a pituitary scan in those who respond to TRH and whose baseline serum level lies between 20 and 60 ng/ml.--P.K.B.
Treatment of Prolactinomas With Megavoltage Radiotherapy
Grossman, A.; Cohen, B. L.; Charlesworth, M.; Plowman, P. N.; Rees, Lesley H.; Wass, J. A. H.; Jones, A. E.; Besser, G. M.
Abstract: The best means of treating prolactinomas remains uncertain. Surgical cure is reported in as few as 40% of patients, and recurrent hyperprolactinemia is not infrequent. The effect of megavoltage radiotherapy was evaluated in 36 women aged 18-48 years with prolactinomas. The mean basal serum PRL level exceeded 1,300 ng/ml. Dopamine agonist therapy was begun just before or during radiotherapy and was withheld periodically once the desired number of pregnancies was completed. The mean length of follow-up after radiotherapy was 4.2 years. Eighteen patients were retested for pituitary reserve while taking bromocriptine. With the use of 4-MeV or 15-MeV linear accelerator, a lesion dose of 4,500 rad was delivered in 25 fractions over 35 days.
No serious side effects of radiotherapy were observed. The serum PRL level fell to normal after treatment with dopamine agonists in all patients but 1. All but 1 of 27 patients who stopped taking dopamine agonist therapy after radiotherapy had serum PRL levels below baseline, and 8 had normal values (Fig 40-6). No patient with a normal PRL level had recurrent hyperprolactinemia during follow-up without bromocriptine treatment. Eight patients had GH deficiency before treatment, and many more developed deficient GH levels after radiotherapy. Abnormal responses to ACTH became normal. Most patients had no change in the free thyroxine index after treatment. Four patients had late menstrual irregularities after treatment. All but 4 of 29 women who wished to conceive did so after radiotherapy and during bromocriptine therapy, and all of them delivered normally. No patient had evidence of tumor expansion during pregnancy. Radiography indicated remineralization or return to a normal sellar appearance in 45% of the patients assessed.
Definitive radiotherapy for prolactinoma reduces the risk of visual field defect or headache occurring during pregnancy. It should be deferred until plans for conception are definite. Combined radiotherapy and dopamine agonist therapy is a safe and effective approach to patients with prolactinoma.
Post-comment: There seems to be no compelling reason to recommend surgery for prolactinomas unless there is evidence that the tumor is very large and fails to shrink with bromocriptine treatment. Indeed, an argument can be made that it is not even cost-effective to perform a computed tomographic scan to diagnose prolactinoma, since the presence of a tumor is unimportant unless it is large enough to be seen easily with an ordinary skull x-ray film. When clinical and laboratory evidence indicates that a prolactinoma is present, bromocriptine is the treatment of choice. If this fails, radiation therapy is the next reasonable step.
Mild hypopituitarism is compatible with pregnancy. Clinical evidence of pituitary insufficiency may sometimes appear dramatically, as in the case of a 32-year-old pregnant woman who developed hypoglycemia with seizures and coma in the 32nd week of an uncomplicated pregnancy. The hypoglycemia was easily controlled. Study showed her to have hypopituitarism with deficiencies of ACTH, GH, FSH, LH, and PRL and probably of TSH as well.
Pituitary apoplexy has been reviewed in some detail. The syndrome occurs almost exclusively in patients with a preexisting adenoma. Many manifestations usually clear up spontaneously. These include mild ophthalmoplegia and some of the endocrine deficiencies, although recovery in this situation may take many years. Computed tomographic scan is helpful in diagnosis, whereas plain skull film is useful only if the hemorrhage occurs in a tumor large enough to produce changes in the configuration of the sella turcica. Angiography is rarely necessary. Treatment at the time of the acute episode should include immediate administration of corticosteroids. In many cases surgical evacuation of the hemorrhage is recommended, but in view of the high incidence of spontaneous improvement it is not always clear when surgery has been helpful.
Some pituitary tumors that are clinically evaluated as ``nonfunctioning'' in fact may be secreting incomplete forms of hormones. For example, 6 patients have been described with nonfunctioning tumors that in fact produced the alpha subunit of TSH. Because the alpha subunit is inactive unless linked to the beta subunit, the secretory product of these tumors had no biologic effect.--P.K.B.
Plasma Adrenocorticotropin and Cortisol Responses to Ovine Corticotropin-Releasing Factor in Patients With Adrenocortical Insufficiency Due to Hypothalamic and Pituitary Disorders
J. Clin. Endocrinol. Metab., April 1984, Vol. 58, Pg. 758-760, Ch. 41-2
Abstract: Synthetic ovine corticotropin-releasing factor (CRF) has proved useful in the differential diagnosis of Cushing's syndrome. Responses to CRF were examined in 8 patients having adrenocortical insufficiency of varying etiology. Four had hypothalamic hypopituitarism associated with suprasellar tumors, 2 had Sheehan's syndrome, and 2 had isolated ACTH deficiency. Two of the patients with hypothalamic hypopituitarism had germinomas, 1 had a craniopharyngioma, and 1 had a meningioma. Glucocorticoid replacement was stopped at least 24 hours before the study. Ten normal subjects also were examined. Tests of CRF were done after overnight fasting; 100 mu g of synthetic ovine CRF was used. Plasma ACTH was determined by radioimmunoassay.
All but 2 patients had basal plasma ACTH levels below the limit of detection (10 pg/ml). All those with hypothalamic hypopituitarism had marked, prolonged increases in plasma ACTH after CRF injection, as well as gradual but definite increases in plasma cortisol levels (Fig 41-1). One of the 2 patients with Sheehan's syndrome had slight increases in plasma ACTH and cortisol levels after receiving CRF. Neither patient with isolated ACTH deficiency had detectable ACTH or cortisol after CRF administration.
The CRF test would appear to be useful in evaluating pituitary ACTH reserve and in diagnosing adrenocortical insufficiency due to hypothalamic and pituitary disorders. No adverse effects of CRF administration occurred in the present cases.
Post-comment: This is consistent with the idea that response of the pituitary gland to CRF is modulated by the ambient level of cortisol. Further support for this idea is provided by the observation that the response of plasma ACTH concentration after administration of CRF to human subjects is negatively correlated with the preinjection concentration of cortisol in plasma. This negative relationship between plasma cortisol concentration and ACTH secretory response is also consonant with the observation that the pituitary glands of patients dying with untreated adrenocortical insufficiency have both hyperplasia and nodular tumor formation of the corticotropin-secreting cells. Thus the increased secretory response is related to an increased mass of corticotropes.
Adrenocortical insufficiency can be caused by many different factors, although autoimmune damage is probably the most common explanation at present. It has been recognized for many years that ``idiopathic'' adrenal insufficiency is an autoimmune disease and that many patients with other types of autoimmune disease have circulating antiadrenal antibodies without evidence of adrenal failure. By analogy with the thyroid, however, one might postulate the presence in some patients of compensated hypoadrenalism, manifested by normal levels of adrenal function associated with elevated levels of plasma ACTH. This situation does indeed occur. Among 1,675 patients with insulin-dependent diabetes mellitus and 2,032 of their relatives, the incidence of antiadrenal antibodies was significantly higher than in a control group of 2,543 normal subjects. The mean plasma ACTH concentration and renin activity were both significantly higher in the group with positive antibodies (30 tested) than in controls without antibodies. Static tests of adrenal function were otherwise normal, but ACTH responses and suppression tests were not performed.$sup{1}$sup{5} Carcinomatous metastases to the adrenal are common, but they rarely destroy enough of the gland to cause failure. Two patients in whom this cause of failure was observed have been described recently. Patients with acquired immunodeficiency syndrome (AIDS) frequently have necrosis of the adrenal glands at autopsy. Ten such patients were described recently, in 7 of whom cytomegalovirus infiltration of the adrenal cortical cells was demonstrated. Among 20 patients with AIDS demonstrating symptoms of adrenocortical insufficiency, 4 had chemical evidence of defective adrenocortical function. Two also had elevated levels of ACTH. All 4 responded clinically to replacement treatment with adrenal corticosteroids.
The usual dose of fludrocortisone in patients with Addison's disease is 0.05-0.1 mg daily. There is some evidence that this amount of the steroid is insufficient to restore physiologic control of electrolyte metabolism. In a group of 10 patients with adrenocortical insufficiency treated with this dosage, 9 had elevated levels of plasma renin and decreased total body sodium and water. At a daily dose of 0.3 mg, all patients gained weight and retained salt and water, and all had a fall of plasma levels of renin, urea, and potassium. The dose of fludrocortisone required to maintain the plasma renin at a normal level was between 0.15 and 0.20 mg daily. Of course, higher doses would be contraindicated if they caused arterial hypertension.
Hypoadrenocorticism is sometimes induced as a last-ditch method of treating advanced carcinoma of the breast. Surgical adrenalectomy is one method of approach; but aminoglutethimide treatment has become increasingly popular as a way of inducing chemical adrenalectomy. The response is favorable in about a third of patients, but all eventually relapse. Immediately before clinical relapse is apparent, increased plasma concentrations of estrogens, dehydroepiandrosterone sulfate and androstenedione can be seen. Statistically, the increment in estrogens is the most significant. This suggests that escape of the adrenal from suppression by the drug may contribute to the relapse. Escape from this medication is not unexpected in patients with carcinoma of the adrenal, but this observation suggests that it is impossible to suppress indefinitely even the relatively normal adrenal of the patient with advanced breast cancer.
Corticosteroids are sometimes injected rapidly in large doses in order to achieve immunosuppression, especially during episodes of transplant rejection. This is not an entirely safe procedure. Two patients have been reported who died suddenly after injection of a 1-gm bolus of methylprednisolone intravenously in one case and 0.5 gm over 10 minutes in the other. The first patient died 30 minutes after the injection, and the second within 30 seconds after completing the injection. In both cases there was nausea, wheezing, cyanosis, and respiratory arrest. Autopsy showed pulmonary small airways changes consistent with an acute asthmatic attack, although the exudate was less dense.--P.K.B.
Treatment of Cushing's Disease in Childhood and Adolescence by Transsphenoidal Microadenomectomy
Styne, Dennis M.; Grumbach, Melvin M.; Kaplan, Selna L.; Wilson, Charles B.; Conte, Felix A.
N. Engl. J. Med., Apr. 5, 1984, Vol. 310, Pg. 889-893, Ch. 41-6
Affiliation: Univ. of California, San Francisco
Abstract: Prompt correction of hypercortisolemia is important in children with Cushing's disease because of its severe inhibition of growth. Fifteen children and adolescents aged 7-13 years with Cushing's disease underwent transsphenoidal microadenomectomy, and were followed up for 10 months to 9 years. The patients presented with weight gain and slowed linear growth. Several were fatigued and had symptoms of persistent depression. Pubertal development was delayed or arrested in 6 of 10 evaluable patients. Nine patients were hypertensive.
Ten patients had mild, transient diabetes insipidus after transsphenoidal surgery. Two others with an adenoma in or near the posterior pituitary were affected for longer periods. Fourteen patients had an adenoma or seemingly pathologic tissue removed from the pituitary, and 13 had ACTH deficiency postoperatively. One patients had a second adenoma removed after 6 months. Four of 11 patients had adequate cortisol secretion at follow-up a median of 9 months after operation, whereas 7 remained partially deficient in ACTH. Twelve patients treated before epiphyseal fusion had normal growth or catch-up growth starting within 3 months after operation. Secondary sex characteristics developed normally in patients of pubertal age after they were operated on. Hypertension resolved in all cases. One patient had a recurrence after 4 years and remitted after a repeat microadenomectomy.
Transsphenoidal adenomectomy appears to be the best approach to Cushing's disease in childhood at present. Cushing's disease must be considered in children with rapid weight gain and growth failure. Only if treatment is instituted before epiphyseal fusion occurs can some of the lost growth be regained. Catch-up growth has minimized the predicted loss of mature height in patients treated early enough.
Post-comment: The advantage of transsphenoidal adenomectomy in children is similar to that in adults. Some surgeons, however, recommend radical hypophysectomy by the transsphenoidal route. This is probably a little less radical than it sounds, since complete removal of the pituitary is very difficult, especially by the transnasal route. In a series of 16 patients treated by this method, for example, normal menstruation was restored in 5 women, and 2 became pregnant and had normal deliveries. Nevertheless, only 3 of the patients were healthy without endocrine supplementation. In this series, 14 had complete and permanent remission and 1 partial remission of Cushing's disease when followed for 5-14 years (mean, 8 years). This approach to treatment would certainly not be appropriate for children, however, because of the problems of growth and sexual maturation, which would be difficult to surmount after hypophysectomy.
Now that the amino acid composition of CRF is known, the organic chemists are busily manipulating its structure to try to produce analogues that will block the effects of the native material. Several such antagonists have been described which are effective in the rat in blocking the activity of CRF itself, both in vitro and in vivo. In addition, an analogue has been developed that is much more potent than CRF itself in stimulating corticotropic cells to release ACTH. It seems likely to me that a CRF antagonist might prove quite useful clinically in treating patients whose pituitary tumor cannot be controlled by local means. I wonder whether such a material would inhibit ACTH synthesis or release from a tumor producing ectopic ACTH. It is not clear whether such tumors are entirely free from controls, though the controlling material might be produced by the tumor itself. For example, consider the possibility--as yet unexplored, I believe--that the reason ACTH is produced ectopically is that the tumor produces something like CRF which acts locally on the tumor cells. In that case, a CRF blocker might have an effect on the tumoral production of ACTH. Pretty hypothetical, but not entirely without precedent. At any rate, I suspect that it will be some years before such analogues are ready for testing in human patients.--P.K.B.
Long-term Results of Adrenal Surgery in Patients With Cushing's Syndrome due to Adrenocortical Adenoma
Valimaki, M.; Pelkonen, R.; Porkka, Leena; Sivula, A.; Kahri, A.
Clin. Endocrinol., February 1984, Vol. 20, Pg. 229-236, Ch. 41-8
Affiliation: Univ. of Helsinki
Abstract: The authors followed up 14 of 16 patients who had Cushing's syndrome due to adrenal adenoma and who had undergone adrenal surgery in 1967-1981. The patients were reexamined a mean of 4 1/2 years after the operation. The 14 women and 2 men were aged 22-60 years. All had typical clinical findings and documented hypercortisolism, and the tumors were benign adrenocortical adenomas. Hydrocortisone was gradually tapered postoperatively in accord with the basal and ACTH-stimulated plasma cortisol level. Two patients died during follow-up, 1 of myocardial infarction and 1 of suicide.
None of the patients had cushingoid features at follow-up. Five patients still were obese, but 11 weighed less than before surgery. Blood pressures had declined significantly since operation, although 4 patients still had elevated pressure, and 1 used antihypertensive drugs. Seven women were amenorrheic before the operation, and 1 afterwards. Muscle weakness had disappeared in 8 of the 9 affected patients. Diabetes was cured in 1 of 2 patients; the other was still receiving insulin at the time of death 2 weeks after surgery. All patients had normal fasting blood glucose levels. Only 1 patient was still using hydrocortisone regularly at the time of follow-up. Plasma cortisol was normal in all but 2 patients, and 1 of the exceptions was using oral contraception. Urinary cortisol levels were not elevated in any patient, and dexamethasone testing gave normal results in all cases. The plasma cortisol response to ACTH was blunted in 3 cases. Ten of 12 patients had normal plasma ACTH levels. Bone mineral density was reduced in the female patients studied.
Patients surviving adrenal surgery for adrenocortical adenomas causing Cushing's syndrome usually have a very good clinical response. Function of the pituitary-adrenal axis is often, but not always, completely restored after this surgery.
Post-comment: It is often difficult to distinguish clinically between a functioning adrenal adenoma and a carcinoma. The differentiation may be aided by measuring the plasma levels of 17-hydroxyprogesterone, testosterone, androstenedione, and dehydroepiandrosterone. These steroids are present in much higher concentrations in patients with carcinoma than in those with adenoma. Moreover, elevated levels of 11-deoxycortisol (``compound S'') are present in cancer patients. Since this substance may cross-react in the binding analysis of cortisol, it may not be recognized, and the estimate of plasma cortisol level maybe higher than the true cortisol level. It is therefore recommended that whenthe possibility of cancer is suspected chromatographic methods be usedto separate these two steroids before analysis by binding displacement methods.
The cure rate of carcinoma of the adrenal is poor, because most tumors ultimately metastasize. Moreover, mitotane, commonly tried in treating this disease, seldom provides more than a brief respite. It is heartening therefore to read about 1 patient who was cured by use of the drug. High doses were used (6 gm/day) and titrated to produce a plasma level of 15-20 mu g/ml. Oral glucocorticoid replacement was required. After a year of receiving this dosage of DDD, the patient developed slurred speech, ataxia, and emotional depression. At this time the plasma level of mitotane was as high as 34 mu g/ml. These symptoms cleared when DDD was discontinued. A laparotomy showed complete disappearance of the tumor, and the patient remains clinically tumor free 3 years after the drug has been stopped.--P.K.B.
Simultaneous Assay of Noradrenaline and Its Deaminated Metabolite, Dihydroxyphenylglycol, in Plasma: A Simplified Approach to the Exclusion of Pheochromocytoma in Patients With Borderline Elevation of Plasma Noradrenaline Concentration
Brown, M. J.
Eur. J. Clin. Invest., February 1984, Vol. 14, Pg. 67-72, Ch. 41-10
Affiliation: Royal Postgrad. Med. School, London
Abstract: An attempt was made to ascertain that dihydroxyphenylglycol (DHPG) is formed mainly from neuronally released norepinephrine and to evaluate use of a plasma norepinephrine: DHPG ratio in the diagnosis of pheochromocytoma. The source of DHPG was investigated in 8 hypertensive patients and 4 normal volunteers. To determine whether net extraction of DHPG by the liver occurs, DHPG concentration was measured in hepatic vein and arterial samples from the 8 patients. In the volunteers the effect of norepinephrine infused directly into the circulation at the rate of 0.07 mu g/kg per minute over 1 hour was compared with that of neuronal norepinephrine release stimulated by an oral dose of 800 mg of tyramine. The diagnostic value of a norepinephrine: DHPG ratio was estimated in 17 patients with proved pheochromocytoma and 19 patients without the disease.
Among 8 patients tested, mean hepatic extraction of norepinephrine was 64.3% and that of DHPG was 67.3%; 4 patients proved to have pheochromocytoma, and in the other 4 no tumor was found. The ratio of norepinephrine to DHPG in plasma was several times greater in the 17 patients with pheochromocytoma (Fig 41-2) than in patients in whom no tumor was found. The ratio was greater than 2 (range, 2.05-3.57). In the 19 patients without pheochromocytoma, the reverse was found (Fig 41-3). The ratio of DHPG to norepinephrine was greater than 2 (range, 2.08-2.74).
Exposure to elevated catecholamine concentrations causes down-regulation of adrenoceptors in pheochromocytoma patients. This probably explains the occurrence of moderate or severe hypertension only in ``large'' pheochromocytoma patients with a severalfold elevation of norepinephrine secretion. The opposite diagnostic problem is posed by ``small'' pheochromocytoma patients, for they have only slightly elevated (less than twofold) catecholamine secretion and are not readily distinguishable from patients with elevated sympathoadrenal activity due to anxiety, stress, or drugs.
The likely value of the test is its exclusion of the diagnosis of pheochromocytoma in the group of patients depicted in Figure 41-3. Pheochro mocytoma was found in less than 1 in 10 patients with elevated vanillylmandelic acid (VMA) excretion, who had been referred because measurement of VMA excretion or plasma norepinephrine concentration had failed to exclude pheochromocytoma. In the early stages of a recurrence, the plasma norepinephrine concentration may be within normal range; but even in this instance, the reversed norepinephrine: DHPG ratio betrays the tumor origin of the norepinephrine.
Post-comment: A new modality has been added to computed tomographic (CT) scanning as a useful way of diagnosing pheochromocytoma. Meta-iodobenzylguanidine localizes to adrenergic tissue. When the compound is labeled with $sup{131}I, it can be used to locate tumors of the adrenergic system. Computed tomographic scanning and $sup{131}I-MIBG are equally effective in localizing tumors that are primary in the adrenal area and their metastases; but the radioisotopic method is better for locating extraadrenal tumors. Also, because it does not require tomography, it is easier to use it to scan the entire body than to use the CT scan.
Malignant pheochromocytomas are by no means unusual, but their treatment with chemotherapy is difficult. A patient has been reported who responded favorably to 2 gm of streptozotocin monthly. His hepatic metastasis shrank, and urinary excretion of dopamine, homovanillic acid and vanillylmandelic acid also fell to the normal range. There was some tendency for the excretion of these products to rise after a while, and it is not certain that the patient's condition would be kept entirely controlled in the future; but for 18 months a good result was obtained.--P.K.B.
Propranolol-Induced Pulmonary Edema and Shock in a Patient With Pheochromocytoma
Abstract: Propranolol has been recommended for use in patients with epinephrine-secreting pheochromocytomas, but pulmonary edema has been reported in patients given this drug who have no history of heart disease.
Woman, 43, presented with acute substernal chest discomfort radiating to the left arm, facial flushing, and palpitations. Supraventricular tachycardia was noted, with a blood pressure of 172/90 mm Hg, and the patient appeared flushed and dyspneic. Cardiac examination was negative. Nitroglycerin provided no relief. Premature ventricular beats were controlled by intravenous lidocaine. The patient became cyanotic and had pulmonary rales shortly after receiving 2 mg of propranolol intravenously for persistent tachycardia. The blood pressure was 200/120 mm Hg. Chest roentgenography showed fluffy alveolar infiltrates bilaterally and vascular redistribution consistent with pulmonary edema. Respiratory and cardiac arrest ensued, and dopamine was necessary after resuscitation. Norepinephrine was given after resuscitation from a second episode of arrest. Oliguria persisted after furosemide therapy, and peritoneal dialysis was instituted. A transvenous pacemaker was inserted because of third-degree atrioventricular block. Electromechanical dissociation developed, and the patient died after failing to respond to bicarbonate, cal calcium, epinephrine, and pericardiocentesis.
Autopsy showed a 2X2-cm pheochromocytoma of the right adrenal. No cardiac or coronary artery lesions were found. The tumor contained 4.8 mg/gm of epinephrine and no norepinephrine. Blood cultures were negative.
Pulmonary edema rapidly followed intravenous administration of propranolol in this case. The respiratory distress appeared to be cardiogenic, and the pulmonary edema cleared rapidly when the wedge pressure declined on peritoneal dialysis. The tumor produced primarily epinephrine. Propranolol therapy presumably caused both beta{sub-1}- and beta{sub-2}-blockade, leaving alpha effects unopposed and resulting in hypertension and increased afterload. Refractory hypertension may have resulted from a loss of vasomotor control secondary to hypoxic damage after a period of intense vasoconstriction.
It is potentially hazardous to use propranolol in patients with pheochromocytoma without first inducing alpha-adrenergic blockade. Use of a cardioselective beta{sub-1}-blocker such as metoprolol might be preferable.
Post-comment: Patients with pheochromocytoma characteristically are less sensitive to a given dose of adrenergic medication than are normal individuals. Indeed, this reduced sensitivity has been used as a (rather dangerous) physiologic test for the tumor. The mechanism has been explored in rats that have a congenital tendency to develop norepinephrine-secreting pheochromocytoma. Down-regulation of beta{sub-1}-receptors was found in the heart and adipocytes; and the cyclic AMP response to isoproterenol in adipocytes was diminished. There was no change in the beta-receptors of lung or mesenteric arteries. The receptors in these tissues are beta{sub-2} receptors.--P.K.B.
Ratio of Serum Triiodothyronine to Thyroxine and the Prognosis of Triiodothyronine-Predominant Graves' Disease
Ann. Intern. Med., March 1984, Vol. 100, Pg. 372-375, Ch. 42-4
Abstract: Triiodothyronine (T$sub{3})-predominant Graves' disease is characterized by persistently high levels of serum T$sub{3}, normal levels of serum thyroxine (T$sub{4}), and a high ratio (greater than 20) of serum T$sub{3} (measured in nanograms) to serum T$sub{4} (measured in micrograms). The authors studied patients with Graves' disease under treatment with thionamide drugs, and they compared the clinical course and pathogenesis of the group that developed a T$sub{3}-predominant pattern with the group of patients with normal levels of serum T$sub{3} and T$sub{4} and a normal T$sub{3}:T$sub{4} ratio during treatment (``controls'').
All patients were seen at Osaka Medical College Hospital, Osaka, or Kuma Hospital, Kobe, Japan and were observed for at least 18 months after 1-4 years of drug therapy. Twenty-one of the 30 patients with normal T$sub{3}:T$sub{4} ratios had long-term remissions after discontinuation of thionamides; the other 9 had relapses of hyperthyroidism. Of the 30 patients with T$sub{3}-predominant Graves' disease, 9 had a mean decrease in the serum T$sub{3}:T$sub{4} ratio to 18.1 during treatment. Four of these 9 had remissions and 5 had relapses. Of the remaining 21 patients with T$sub{3}-predominant Graves' disease, only 2 had remissions and 19 had relapses. The T$sub{3}-predominant group had greater activity of TSH-like antibody in the serum than the control group (50% vs. 20%), and their glands were significantly heavier (mean, 115 vs. 44 gm). The 2 patients with T$sub{3}-predominant disease who had remissions had lower activity of serum TSH-like antibody during treatment. Their remissions have persisted without recurrence for as long as 28 months. The patients with T$sub{3}-predominant disease also had significantly lower contents of T$sub{3} and T$sub{4} in their thyroglobulin than the control patients. Serum TSH-like antibody correlated positively with the serum T$sub{3}:T$sub{4} ratio in the patients with T$sub{3}-dominant disease. Activity of thyroid T$sub{4}5'-deiodinase was significantly greater in those with T$sub{3}-predominant disease than in control patients.
The results seem to indicate that patients with T$sub{3}-predominant Graves' disease are not likely to have a long-term remission after drug treatment is discontinued. The high T$sub{3}:T$sub{4} ratio is attributable, in part, to greater activity of thyroid T$sub{4}5'-deiodinase, which may be mediated by high levels of Graves' immunoglobulin.
Post-comment: The paper abstracted above implies that patients with Graves' disease fall into at least two categories: those who have long-standing remissions (who can be recognized by the presence of normal T$sub{3}:T$sub{4} ratios) and those whose ratio of T$sub{3} to T$sub{4} is high and whose probability of sustained remission is low. The paper summarized in article 42-3 implied the same type of division but used T$sub{3} suppression to separate the patients with a good prognosis from those with a poor prognosis for prolonged remission. But there remains an important question: Does treatment actually alter the long-term prognosis for cure of patients with Graves' disease? In other words, when treatment with antithyroid medications is withdrawn from patients whose Graves' disease has been controlled for a year, about half remain euthyroid. Does this represent a higher percentage than would have attained the euthyroid status merely by natural evolution of the disease? Do the antithyroid agents actually change the course of the disease? The arguments supporting such an effect have been marshalled by Weetman et al., who believe that the immunosuppressive activities of the thiourylene drugs increase the probability of controlling the disease permanently. Thus it is clinically important to be able to distinguish the good-prognosis group from those likely to relapse, and application of these tests (T$sub{3}:T$sub{4} ratio and suppressibility with T$sub{3}) is useful.
Thyrotoxic crisis remains a serious complication of hyperthyroidism, carrying a high potential mortality. The best way of treating this disease is not clear. One unusual approach, which was successful in a very sick woman, was to undertake a plasma exchange with an IBM cell separator. After the first exchange, the free serum T$sub{4} dropped from 6.4 to 4.0 ng/dl, while the total T$sub{4} was unchanged. The T$sub{4}-binding globulin concentration in the plasma increased by 50%, which probably explained the fall of free T$sub{4} concentration. The woman, who had been moribund before treatment, improved rapidly after the exchange and ultimately recovered completely.--P.K.B.
Long-Term Follow-Up Study of Compensated Low-Dose {sup-1}{sup-3}{sup-1}I Therapy for Graves' Disease
N. Engl. J. Med., Aug. 16, 1984, Vol. 311, Pg. 426-432, Ch. 42-5
Abstract: Low-dose radioiodine therapy was given to 187 patients with Graves' disease using a protocol that included compensation for thyroid size. All were seen at the University of Chicago. Twenty-nine patients received antithyroid drugs and potassium iodide starting 24 and 48 hours, respectively, after $sup{131}I treatment. Another 122 patients were treated with antithyroid drugs for 6-24 months, and 166 others were treated surgically. The initial therapeutic dose of $sup{131}I (in millicuries) was calculated as the estimated thyroid weight in grams multiplied by the scheduled mu Ci/gm divided by the fractional $sup{131}I uptake at 24 hours multiplied by 1,000. The dose ranged from 40 mu Ci/gm of thyroid tissue for normal-sized glands to 85 mu Ci/gm for those weighing more than 81 gm. Gland weight was the average of estimates made by three to six observers. The mean initial total dose of $sup{131}I was 4.9 mCi.
Of the study patients, 57 required retreatment with radioiodine. Eleven retreated patients remained hyperthyroid when last tested. The cumulative incidence of hypothyroidism was 76% after 11 years. The rate in the surgical group was 27%. The disease remitted in only 40% of medically treated patients, and hypothyroidism developed in 2% of this group. The only serious side effect of radioiodine therapy was an episode of cardiac arrhythmia in 1 patient. Total morbidity in the surgical group was 7%; there were no deaths.
It appears difficult to modify radioiodine therapy in order to achieve long-term euthyroidism in many patients with Graves' disease, and there is no assurance that long-term hypothyroidism will not develop. The effect of low-dose $sup{131}I administration followed by potassium iodide and antithyroid drug treatment is currently under study. Surgery and antithyroid drug therapy both are useful alternatives to treatment with radioiodine in selected patients.
Post-comment: It is no secret that patients whose Graves' disease has been treated with radioactive iodine have a high incidence of hypothyroidism, which increases progressively during follow-up. Unfortunately, this complication cannot be avoided by adjusting the dosage of radioactive iodine. Since unrecognized hypothyroidism can lead to complications, one must follow patients closely after treatment. To make follow-up less difficult, one group in Newcastle upon Tyne has tried deliberately destroying the thyroid with a high dose of 15 mCi of $sup{131}I and accepting the virtual certainty that replacement will be required immediately. In their group of 225 patients followed for 1-6 years, hypothyroidism was well established in 50% of the patients within 3 months and within 1 year in 64% of patients. Hyperthyroidism was controlled in all but 5.6% within 1 year. Treatment with replacement was begun as soon as evidence of hypothyroidism was found. The referring general practitioners found this method of treatment simpler and more satisfactory than earlier methods requiring constant titration of the status of thyroid function.
Although we ordinarily think of thyroid tumors as the chief late complication of radiation to the thyroid area, hypothyroidism can also occur. Among 95 children receiving more than 2,600 rad to the cervical region for treatment of Hodgkin's disease, 74 (78%) developed abnormal thyroid function, almost always consisting of hypothyroidism and elevated TSH levels. Two developed hyperthyroidism. Among 24 children receiving less than 2,600 rad, the incidence was only 4 (17%). The average time to appearance of hypothyroidism was 18 months in the low-dose group and 31 months in the group that received the high dose. Hypothyroidism almost always occurred within 36 months of treatment. Evidence of hypothyroidism disappeared in 2 of the patients receiving the lower dose range and 18 in the high range. Only 1 patient has developed a thyroid cancer so far. Apparently, therefore, children with Hodgkin's disease receiving cervical radiation are more prone to late hypothyroidism than other radiated groups, in whom late tumors are the major problem and thyroid function otherwise remains relatively normal. Perhaps the iodine in the dye used for lymphangiograms in all of the patients with lymphomas plays a part in rendering their thyroid glands especially susceptible to radiation.
Similar damage to the thyroid occurs in patients with head and neck cancer treated with chemotherapy (cisplatin, bleomycin, and methotrexate), surgery, and radiation therapy. In a group of 43 previously untreated patients studied prospectively at the Dana Farber Cancer Center, 37% developed elevated plasma TSH levels during a mean follow-up of 9 months after treatment.
Let me return for a moment to hyperthyroidism. The disease is almost always a result of thyroid-stimulating autoantibodies, but rare cases continue to appear in which the cause is hypersecretion of TSH, often by a pituitary tumor. In the even rarer situation in which no such tumor can be demonstrated, one group has found that the disease could be controlled with bromocriptine. During the first 4 months of treatment, serum TSH, T$sub{3}, and T$sub{4} levels all returned to normal, as did the PRL concentration, which had also been elevated. The patient returned to clinical normalcy.--P.K.B.
Follow-Up of Solitary Autonomous Thyroid Nodules Treated With {sup-1}{sup-3}{sup-1}I
Goldstein, Rose; Hart, Ian R.
N. Engl. J. Med., Dec. 15, 1983, Vol. 309, Pg. 1473-1476, Ch. 42-8
Affiliation: Ottawa Civic Hosp., Ottawa, Ont.
Abstract: The authors studied the long-term effects of $sup{131}I therapy for solitary autonomous thyroid nodules on nodule size and thyroid function in 23 patients with autonomous thyroid adenomas that had been treated with $sup{131}I 4-16.5 years earlier (mean, 8.5 years).
One of the patients was excluded because she had undergone a partial thyroidectomy immediately after treatment with $sup{131}I. Of the remaining 22 patients, 8 no longer had a palpable nodule, 10 had a palpable nodule that was smaller than before treatment, 2 had a nodule that had not changed size, and 2 had an enlarged nodule. One of the enlarged nodules remained hot on thyroid scans, and the other nodule was cold. Follow-up scans of 13 of the 14 patients with nodules remaining palpable showed that only 2 remained hot, 5 were warm, and 6 were cold. Seven of the 22 patients had developed symptomatic and biochemically proven hypothyroidism. The incidence of hypothyroidism increased from 25% 5 years after treatment to 50% 15 years after treatment. The incidence of hypothyroidism was not related to nodule size or level of thyroid function prior to $sup{131}I therapy or the cumulative dose received. Three of 10 patients who had complete suppression of extranodular thyroid tissue on their original scan became hypothyroid. However, 6 of 12 patients with incomplete suppression of extranodular tissue on their original scan became hypothyroid. When either exogenous triiodothyronine (T$sub{3}) or thyroxine (T$sub{4}) was administered before and after the dose of $sup{131}I to patients with incomplete suppression of extranodular tissue, only 1 of 5 became hypothyroid, compared with 5 of 7 who were not given T$sub{3} or T$sub{4} in this situation.
The high incidence of hypothyroidism in this series and the limited success on clinical ablation of the nodules support the view that $sup{131}I therapy should be reserved for the older patient or the patient who is not a good candidate for surgery.
Post-comment: It is surprising that $sup{131}I treatment of a solitary nodule that has apparently suppressed the entire remaining gland should produce hypothyroidism. This suggests that suppression was not complete. I don't say this to cast doubt on the observations, since I have seen the same phenomenon myself; but I don't understand it.
When should substernal goiters be removed surgically? The surgeons are pretty aggressive about wanting to remove all substernal goiters because (1) there is no other treatment for long-standing nodular goiter; (2) $sup{131}I treatment can precipitate acute reactions in the elderly resulting in respiratory distress; (3) a long history of nontoxic goiter does not rule out the possibility that the enlargement is malignant; (4) malignancy occurs in a substantial percentage of substernal goiters and is not accessible to needle biopsy; and (5) the goiters can almost always be removed through a cervical incision. These recommendations are based on 50 substernal goiters found in 872 patients subjected to thyroidectomy at the University of Michigan. I am much less aggressive about recommending surgery because (1) I have seen an appreciable mortality in elderly patients subjected to this operation; and (2) I have never seen a reaction to $sup{131}I that could not be readily controlled. Nonetheless, I accept reasons 1, 3, and 5 as reasonable. The recommendation that ``the presence of a substernal goiter is in itself an indication for operation'' is too sweeping for my taste. Patients deserve to have their treatment more individualized than this statement suggests.--P.K.B.
Evaluation of Various Types of Needle Biopsies of the Thyroid
Esselstyn, Caldwell B., Jr.; Crile, George, Jr.
World J. Surg., August 1984, Vol. 8, Pg. 452-457, Ch. 42-9
Affiliation: Cleveland Clinic
Abstract: More than 2,000 needle biopsies of the thyroid have been done at the Cleveland Clinic since the 1940s, and after 1970 biopsies were performed routinely in all thyroid nodules and diffuse nontoxic goiters. Only 1 instance of a cancer implant in the needle track has occurred. One patient had extrathyroidal bleeding requiring observation, and 1 had paralysis of the recurrent nerve after biopsy by an inexperienced person. No complications have occurred in biopsies done with 18-gauge or fine needles. Of 481 needle biopsies done between 1979 and 1980, 114 were done with an 18-gauge needle, 119 with the fine needle, and 248 with the Tru-cut(R) needle. Two types of biopsy were done at the same time in 114 patients, usually with the Tru-cut and fine needle methods.
The Tru-cut biopsy was most frequently successful. The failure rate was only 4% in the hands of the surgeon who used it most often. Aspiration with an 18-gauge needle and fixation of the aspirate to form a cell block was a satisfactory means of evaluating cysts and small nodules. The fine-needle biopsy technique was a satisfactory means of diagnosing all types of thyroid cancer except for the low-grade follicular type. Even with Tru-cut core biopsies it may not be possible to distinguish between cellular adenomas and low-grade adenocarcinomas. There were only 2 errors in Tru-cut biopsies in this series, and both were corrected on clinical grounds.
Routine needle biopsy of thyroid nodules led to a diagnosis of malignancy in more than 5% of patients. Surgery was avoided in 87% of all patients who had thyroid nodules. Further, 47% of 46 patients having thyroid surgery during the review period had malignant tumors.
Post-comment: The high percentage of malignancy in the operated group indicates success of the procedure in separating benign from malignant lesions. The fact that low-grade adenocarcinomas are not well recognized is not surprising, since even when the entire specimen is available pathologists may sometimes have difficulty in distinguishing benign from malignant processes in this group.
Other authors have discussed the continuing controversy that surrounds the question of needle biopsy in diagnosing the nature of thyroid nodules. One compared the use of fine-needle aspiration (providing a cytologic approach) and Tru-cut needle biopsy (which permits histologic examination) in 405 previously untreated patients. Both techniques achieve a very high overall accuracy, and false-positive errors occur only among nonfollicular lesions labeled as ``possibly malignant.'' Neither method distinguishes reliably between a benign and a malignant follicular neoplasm. If follicular neoplasms are all included as ``suspicious,'' the sensitivity rates for identifying cancers are 86.5% for fine needle and 91.9% for Tru-cut. This is probably not a significant difference; and since the larger needle increases the danger of bleeding and of dissemination of cancer cells along the needle track, I feel fine-needle aspiration is preferable. Another way to study the value of fine-needle aspiration is to follow the course of those patients whose findings were ``suspicious.'' Among 1,970 patients in whom the procedure was carried out, 333 had suspicous cytologic findings, and 253 could be studied prospectively. Malignant lesions were found in 60 (24%). Since thyroid scanning did not distinguish malignant from benign lesions in this group of patients, excisional biopsy is probably the proper course when fine-needle aspiration cytology leaves the diagnosis in doubt. Another group studied 304 patients by fine-needle aspiration in the Vancouver General Hospital. Of those in whom biopsies were performed, 79 were operated on and 37 malignancies were found, an incidence of 47%. The clinic's previous record was only 14% malignancy among patients whose thyroid nodules were excised. The authors feel that needle aspiration greatly improved their diagnostic ability and made it possible, on the basis of previous experience, to avoid 231 unnecessary operations. Moreover, the procedure is suitable for outpatient use and is therefore economical.--P.K.B.
The Effect of Estrogen Dose on Postmenopausal Bone Loss
Horsman, Anthony; Jones, Mary; Francis, Roger; Nordin, Christopher
N. Engl. J. Med., Dec. 8, 1983, Vol. 309, Pg. 1405-1407, Ch. 43-1
Abstract: Estrogen therapy clearly limits bone loss in postmenopausal women, but the minimal effective dose is unknown. Sequential changes in metacarpal cortical diameter were measured by radiographic morphometry in 120 normal postmenopausal women given ethinyl estradiol alone in daily doses of 5-50 mu g. The mean age was 52 years, and the mean postmenopausal period 6 years. The mean interval between hand roentgenograms in individual subjects was about 2 years.
Changes in metacarpal width and cortical width are related to the dose of ethinyl estradiol in Figure 43-1. Net endosteal resorption was inhibited at doses greater than 25 mu g daily. A progressive increase in mean total width was observed, at least with doses exceeding 15 mu g daily, representing expansion of the bones. The changes in women taking less than 15 mu g daily did not differ significantly from those in untreated postmenopausal women. Bone was neither gained nor lost at doses of 15-25 mu g daily, but a net gain was associated with the use of higher doses.
Bone loss with low doses of ethinyl estradiol is related to expansion of the medullary cavity without any change in total bone width. With intermediate doses, endosteal resorption of bone is countered by periosteal apposition. A net gain of bone occurs with doses exceeding 25 mu g daily because endosteal resorption is totally inhibited while periosteal bone apposition continues. The effective dose probably is greater than the minimal dose needed to control menopausal symptoms. The gap might be bridged by combining a small dose of estrogen with a calcium supplement.
Post-comment: If I read this correctly, it says that osteoporosis can be reversed at doses of ethinyl estradiol higher than 25 mu g daily. In contrast to these findings, it is generally held that estrogens are probably useless for treating established postmenopausal osteoporosis, although they can prevent its development.
There is enough concern about the side effects of estrogen that one would like to use the minimal effective dose. The abstract indicates that 25 mu g daily of ethinyl estradiol will afford protection. Conjugated estrogens are less potent; a single-blind study to determine the lowest dose protecting against bone loss found that 0.625 and 1.25 mg/day of conjugated equine estrogens were equally effective in both postmenopausal and ovariectomized women when the rate of bone loss was estimated by single photon absorptiometry or radiogrammetry. Lower doses were less effective. The 50% response level was about 0.45 mg/day. Direct observations were confirmed by measurements of calcium and hydroxyproline excretion in the urine. Thus it appears that 0.625 mg daily is the optimal oral dose of conjugated estrogens.
Although the prescription of estrogens is useful in minimizing hot flushes and preventing osteoporosis, many physicians resist its use because they fear the alleged increase in danger of uterine and breast cancer. In an attempt to find out whether there is a concensus among physicians in this regard, 717 physicians, family practitioners and gynecologists practicing in the 13 counties around Syracuse and Albany, New York, were asked to answer a questionnaire that presented a case history of a symptomatic menopausal woman. The physicians were asked to indicate how they would have treated this patient in 1974 and how they would treat her now. Among gynecologists, 87% would have prescribed estrogen at both times, but the recommended dosages, which were widely scattered in 1974, tended to aggregate around 0.625 mg of conjugated estrogens in 1984. Internists would have prescribed in only 65% of instances in 1984, though 82% would have prescribed in 1974. They recommended using higher doses than the gynecologists. My prejudice is to agree with the gynecologists.
In contrast, a similar study of the present attitude of 25 community physicians and 25 gynecologists in private practice in five mid-Michigan cities toward giving estrogens in menopausal women showed agreement between the two groups, with about 40% of each recommending estrogen treatment. In this group of physicians, the risk of osteoporosis apparently played little part in making the decision. Severe vasomotor symptoms were the major clinical indication for treatment, and risk of endometrial cancer was the major deterrent.--P.K.B.
The Treatment of Hypogonadotrophic Hypogonadism in Men by Pulsatile Infusion of Luteinizing Hormone-Releasing Hormone
Morris, D. V.; Adeniyi-Jones, R.; Wheeler, M.; Sonksen, P.; Jacobs, H. S.
Abstract: Chronic pulsatile infusion of luteinizing hormone-releasing hormone (LRH) is effective in inducing ovulation in women with hypogonadotropic hypogonadism (HH) and in inducing puberty in both males and females with this disorder. A trial of chronic subcutaneous infusion of LRH was undertaken in 17 males with HH. Ten patients had primary HH, including 4 with Kallman's syndrome of anosmia and HH. Seven patients had secondary HH, 5 after surgical treatment of pituitary lesions. In 2 patients HH developed after they passed through normal puberty; these 2 had no evidence of other abnormality of pituitary function. All but 4 patients previously received human menopausal gonadotropin and chorionic gonadotropin therapy. Most patients were given pulses of 15 mu g of LRH by infusion via a subcutaneous cannula in the anterior abdominal wall.
The treatment was well accepted, and its mean duration was 1 year. In the first 3 months, 3 patients with primary HH had continuing increases in gonadotropin and testosterone levels as well as an increase in testicular volume and became fertile. All but 4 patients had increased gonadotropin levels within a week of the start of treatment, and 8 also had increased serum testosterone levels. Increased spermatogenesis was demonstrated in 7 patients. Pituitary desensitization to LRH was identified in 5 patients with primary HH who failed to have increased testosterone secretion despite initial stimulation of gonadotropin release.
Pulsatile infusion of LRH is an effective treatment for infertility in some men with HH. However, further study is needed to determine why the normal integration of gonadotropin secretion and testicular response is lost in some of these patients during such treatment.
Post-comment: Apparently what works for the female also works for the male--at least in this instance. Replacement of testosterone in patients with male hypogonadism is rather a problem. The oral preparations sometimes cause jaundice, and injectable preparations are a bother. The situation has been reviewed, with the conclusion that implantation of testosterone pellets is the most satisfactory method of replacing the hormone. Injected preparations maintained serum testosterone levels above control values for only about 2-3 weeks, whereas a single implantation maintained normal levels for 4-5 months.
In order to interpret the levels of sex hormones in blood it is necessary to take into account the levels of sex hormone-binding protein of plasma. The concentration of this protein increases gradually with increasing age, from preadolescence until the ninth decade, being about twice as high in the mid-80s as in the early 20s. At any given age, the levels for women are about twice as high as for men.
Returning to the subject of the releasing hormones, an unlikely use has been made of the ability of sustained administration of LRH or its analogues to suppress secretion of testosterone. This effect can be useful in treating testosterone-dependent cancers; but it is also potentially useful in treating less dangerous testosterone-dependent disease. For example, it has been reported to be helpful in treating acne conglobata. I don't mean to minimize the distress of patients with this problem, but it seems a bit radical, at least until we know more about long-term side effects. The patient who was reported to have responded was receiving Buserelin as treatment of prostatic cancer. I'm doubtful about using it as yet in adolescents whose only problem is a skin disease.
It is now well known that LH is released in a pulsatile fashion, but the frequency of pulsations is not entirely settled. Most clinical studies draw plasma samples every 15-20 minutes. However, when samples are drawn every 1 or 2 minutes, pulsations are unmasked that are not recognized with the longer sampling intervals. The significance of these shorter pulsations is not known, but perhaps the pattern they represent may be more physiologic and therefore preferable to the slower pulsations now used in delivering stimulating doses of LRF. But it must be remembered that the slower pulses seem to work pretty well.
Luteal phase deficiency usually reflects inadequate progesterone production because of poor follicle development. Failure of complete development of the follicle is associated in some patients with more frequent LH pulses than normal. Four deficient women had 8.5 pulses in 8 hours, as compared with only 5.2 pulses in 8 hours in normal women. There was no significant difference in the mean level of FSH in the deficient women compared with normal women. Thus increased frequency of pulsation can be associated with partial ovarian failure. In this study, blood samples were drawn every 10 minutes. One wonders whether the same difference would have been recognized with more frequent samples; but, in any case, it appears that the timing of pulses may be critical when infusion of LRH deficiency is being used to correct defective LH secretion.
The exact cause of the vasomotor flushes that plague menopausal women is not known, but apparently the mean level of plasma LRH is much higher in symptomatic menopausal women than in asymptomatic women; and the spikes of LRH concentration synchronize accurately with spikes of skin temperature during flushes.--P.K.B.
Effects of Cyclosporine Immunosuppression in Insulin-Dependent Diabetes Mellitus of Recent Onset
Stiller, C. R.; Dupre, J.; Gent, M.; Jenner, M. R.; Graffenried, P. A. Ont.); Wolfe, B. M. J.
Science, Mar. 30, 1984, Vol. 223, Pg. 1362-1367, Ch. 45-1
Affiliation: University Hosp., London, Ontario, Canada
Abstract: The evidence for type I diabetes mellitus as an autoimmune disease is largely circumstantial. After diagnosis, most patients achieve a partial remission, but only about 3% become insulin independent. The function of beta cells, as reflected by plasma concentrations of C-peptide, becomes absent after 6-30 months. To assess the therapeutic effect of the immunosuppressive agent cyclosporine (Cy), the authors studied 41 insulin-dependent diabetic patients treated with Cy for 2-12 months. All patients were also treated with purified pork insulin.
Sixteen of 30 patients treated within 6 weeks of the diagnosis of diabetes became insulin independent, with normal concentrations of plasma C-peptide and decreasing titers of islet cell antibodies. Average time on insulin for these 16 patients was 14.6 days. In contrast, only 2 of 11 patients entered into the study at 8 to 44 weeks after diagnosis achieved insulin independence. Average time on insulin for these 2 patients was 102 days. Patients in both groups showed significant increases in concentrations of plasma C-peptide after receiving Cy for 30 days. However, mean values were higher in patients entered early in the study, and the differences were significant at 195 and 285 days after the start of insulin therapy.
The results support the theory of an autoimmune pathogenesis for type I diabetes. A controlled trial of Cy should be conducted to confirm and quantify the observed effects and to determine whether the benefits of Cy therapy are greater than the risks.
Post-comment: An excellent brief review has been published of the evidence implicating both immune and genetic mechanisms in the pathogenesis of insulin-dependent diabetes mellitus and outlining early trials of cyclosporine in preventing its expression (Science 225:1381, 1984).
Since it is critical that treatment be initiated as soon as possible, the earlier a diagnosis is made the better the results should be. It would be best if the disease could be diagnosed during its incubation period. One possible early indicator of the impending development of diabetes might be changes in the circulating lymphocytes. Activated T lymphocytes are apparently the mediators of the beta-cell damage that leads to insulin deficiency. Soon after the development of the disease, raised levels of activated T lymphocytes were found in 14 of 15 patients. In patients who had had the disease for a long time, only 7 of 28 had high levels. Moreover, 5 of 7 nondiabetic twins of insulin-dependent diabetics had high levels of T-cell activity. In contrast, the level of activated T cells was normal in patients with type II diabetes. These data support strongly the idea that cell-mediated immunity is associated with the onset of type I diabetes. They also suggest that monitoring T-lymphocyte activity in patients believed to be candidates for developing the disease might lead to very early diagnosis. But it may be a bit premature to come to this conclusion. Although it is exciting to contemplate predicting the future onset of insulin-dependent diabetes mellitus and preventing the disease by immune suppression, one must be aware that the presence of antibodies to islet cells (as distinct from activated T lymphocytes) does not prove that diabetes will inevitably or even probably develop in the future. In a group of 685 unaffected first-degree relatives of children with type I diabetes, 20 had complement-fixing autoant autoantibodies to islet cells. During a 5-year follow-up, 7 became diabetic, 1 continued to have antibodies but did not develop clinical disease, and the remaining 12 lost their antibodies without developing diabetes. Thus the presence of antibodies is not in itself justification for initiating immunosuppression. Whether the same is true for T lymphocytes remains to be determined.--P.K.B.
Mortality Among Diabetic Patients Using Continuous SubcutaneousInsulin-Infusion Pumps
Teutsch, Steven M.; Herman, William H.; Dwyer, Diane M.; Lane, J. Michael
N. Engl. J. Med., Feb. 9, 1984, Vol. 310, Pg. 361-368, Ch. 45-2
Affiliation: Centers for Disease Control, Atlanta
Abstract: The authors reviewed data on 35 patients who died in late 1982 among an estimated 3,500 diabetic patients in the United States who used continuous subcutaneous insulin infusion pumps. The estimated ascertainment rate was 77%. Three other deaths occurring abroad were noted, and 14 fetal deaths were reported in women using pumps, 12 of them in the first trimester. The number of pumps initially used and discontinued over time in 10 states is shown in Figure 45-1. The observed number of deaths was not greater than the expected number calculated from age-specific death rates for conventionally treated patients having type I diabetes. The causes of death were not unusual apart from 1 death associated with malfunction of the device and another attributed to acute bacterial endocarditis that arose from an abscess at the site of catheter insertion. The patients who died had high rates of both autonomic neuropathy (66%) and renal disease (39%).
The total number of deaths occurring among users of continuous subcutaneous insulin infusion pumps appears not to be excessive in relation to the expected number for comparable groups of conventionally treated patients with type I diabetes. Certain types of patients, however, may be at an increased risk for death, and morbidity and mortality from pump treatment may be reduced by using more strict selection criteria and by guarding against the specific complications known to be associated with this treatment. Patient motivation and compliance are important considerations. Physicians should be realistic in choosing a desired level of glycemic control. Patients should be encouraged to seek help for intercurrent illnesses. Patients should be willing to monitor their blood glucose levels regularly and should maintain close contact with a friend or relative.
Post-comment: Continuous subcutaneous insulin infusion results in improved control of diabetes and also in an increased sensitivity to insulin with a consequent reduction of insulin requirement. Although the demonstration of increased insulin sensitivity requires special techniques, such as the insulin clamp, clinical evidence of increased insulin effectiveness in one series took the form of a mean reduction of insulin requirement of 23%. I suppose one possible explanation for early deaths of diabetics using the pump might be unrecognized need to reduce the insulin dosage.
I have always been pretty skeptical of the importance placed on diet in the treatment of diabetes by some practitioners. The fact that the American Diabetes Association itself has dodged the question by indicating merely that a ``diabetic diet'' should be one which maintains normal weight and nutrition suggests that I have some company in my doubts. Now a very interesting study has evaluated the effects of diets on patients treated with the insulin pump. In type I lean diabetics controlled by the pump, there was no advantage in using exchanges, special meal planning, or any of the apparatus of the expert dietician. The ``liberalized'' diet used in these patients maintained their weight constant, did not cause any increase in plasma lipids and reduced the HbA$sub{1}$sub{c} to within essentially normal levels. In other words, with use of the insulin pump, lean insulin-dependent patients can eat a normal unrestricted diet. Although some still feel that there is a difference in the effects of various types of carbohydrates on blood glucose, even this belief seems erroneous. A widespread article of faith holds that sucrose causes hyperglycemia whereas starches do not. It turns out, however, that in well-controlled type I diabetics the effects of equicaloric starch and glucose loads on the blood glucose and insulin concentrations are indistinguishable.--P.K.B.
The ``Dawn Phenomenon''{--}A Common Occurrence in Both Non-Insulin-Dependent and Insulin-Dependent Diabetes Mellitus
Bolli, Geremia B.; Gerich, John E.
N. Engl. J. Med., Mar. 22, 1984, Vol. 310, Pg. 746-750, Ch. 45-4
Affiliation: Mayo Med. School
Abstract: The ``dawn'' phenomenon, described in patients with insulin-dependent diabetes mellitus (IDDM), is an abrupt increase in plasma glucose between 5 and 9 A.M. after relative stability through the night. This phenomenon was examined in 20 patients with IDDM and 13 with non-insulin-dependent diabetes (NIDDM) during overnight closed-loop intravenous insulin infusion. Eight of the patients with NIDDM were receiving insulin, and 1 was receiving sulfonylurea therapy; 4 were controlled with diet alone. In patients who were receiving intermediate-acting insulin, the injections were discontinued 48-72 hours before the study and three to four daily subcutaneous injections of regular insulin were given.
The plasma glucose increased from 89 mg/dl in the period from midnight to 6A.M. to 98 mg/dl at 6-9 A.M. in the patients with NIDDM. A comparable increase occurred in insulin-treated patients and those not receiving insulin. Patients with IDDM had an increase from 80 mg/dl at 5:30 A.M. to a peak of 94 mg/dl by 8 A.M. Insulin requirements increased by at least 50% for 11/2 hours in 77% of the patients with NIDDM and 75% of those with IDDM. The phenomenon occurred in 17 of 20 observation periods in 5 patients with IDDM who were studied on four occasions. Their insulin requirements increased 22% after 6 A.M.
The dawn phenomenon is frequent in both patients with IDDM and those with NIDDM. Its potential variability must be taken into account when attempts are made to adjust insulin doses. The mechanism of the dawn phenomenon is not clear, but acceleration of the clearance of insulin early in the morning is a possibility. Studies are needed to determine whether an analogous effect of less magnitude can be observed in nondiabetic persons.
Post-comment: In the past, the ``Somogyi effect'' has been cited as the explanation of early morning hyperglycemia. The idea was that nocturnal hypoglycemia, unrecognized because the patient was asleep, would call forth hyperglycemic anti-insulin reactions such as secretion of epinephrine, growth hormone, and glucagon with resulting temporary insulin resistance. This paper gives no support for such an interpretation. Another study confirms this, showing that the insulin level during the ``dawn phenomenon'' is low and that most so-called rebound hyperglycemia reflects increased carbohydrate intake in response to hypoglycemic symptoms.--P.K.B.
Use of Controlled Insulin Infusion System for Diagnosis and During Surgery on an Insulinoma Patient
Muller-Esch, G.; Valesky, A.; Weber, U.; Wood, G.
Klin. Wochenschr., January 1984, Vol. 62, Pg. 81-86, Ch. 45-12
Affiliation: Med. Univ. Lubeck, Federal Republic of Germany
Abstract: Successful surgical treatment of organic hyperinsulinism depends on the removal of all tumor tissue with autonomous insulin production. In spite of modern procedures, the insulin producing tumor, which is often small, cannot always be diagnosed preoperatively. Even intraoperatively the tumor becomes evident in only 70%-80% of cases, and multiple and partially extrapancreatic tumors can be detected in only 10%.
Woman, 71, with insulinoma was studied. Preoperatively, use of the glucose-controlled insulin infusion system (GCIIS) for glucose clamping at various blood glucose levels demonstrated autonomous insulin production; intravenous glucose needs for maintenance of normoglycemia were evaluated. The results of a somatostatin suppression test, guided by the GCIIS, supported the hypothesis of a well-differentiated beta-cell adenoma with reduced storage capacity. These assumptions were later confirmed by histochemical and ultrastructural investigations. Hypoglycemia during surgery was avoided by means of the GCIIS. Upon clamping of the plasma glucose at 90 mg/dl, 15.5 gm of dextrose had to be given until the tumor was resected. Immediately thereafter, a sharp rise in plasma glucose of 140 mg/dl as well as a need for 4.1 units of insulin showed that the insulinoma tissue had been removed completely.
The development of the GCIIS makes the complete removal of a nondetectable insulinoma possible. One author, using the GCIIS, found that the level of intravenous glucose needs coincides with the hepatic glucose production of normal individuals. This supports the hypothesis that hypoglycemia in insulinoma patients is primarily caused not by increased glucose utilization but by suppression of hepatic glucose production. Even when surgery is contraindicated, the intravenous suppression with the GCIIS gives valuable information about the possible effectiveness of an oral diazoxide long-term therapy.
The results of this study justify the emphasis on two advantages of the GCIIS. Preoperatively; it contributes to avoidance of neuroglucopenia during diagnosis and preoperative determination of the grade of the tumor. Intraoperatively; it prevents hypoglycemia through feedback-controlled dextrose infusion and demonstrates the complete removal of the hormone-secreting tumor.
Post-comment: The hypoglycemic clamp technique can also be used diagnostically. Patients with nontumoral hypoglycemia require more glucose and have higher insulin levels at a given glucose level (e.g., 80 mg/dl) than normal, but the amount of glucose required and the insulin level is higher still at the same blood glucose in patients with hypoglycemia caused by insulinoma. The 24-hour glucose clamp is both a safer and faster way of studying the cause of hypoglycemia than the standard 72-hour fast.--P.K.B.
Prevalence of Diffuse Pancreatic Beta Islet Cell Disease With Hyperinsulinism: Problems in Recognition and Management
Harrison, Timothy S.; Fajans, Stefan S.; Floyd, John C., Jr.; Thompson, Norman W.; Rasbach, Dennis A.; Santen, Richard J.; Cohen, Cynthia
World J. Surg., August 1984, Vol. 8, Pg. 583-589, Ch. 45-13
Abstract: Causes of endogenous hyperinsulinism were reviewed in a series of 77 patients seen at the University of Michigan and Pennsylvania State University since 1960, excluding children younger than 9 years of age. A solitary hyperfunctioning islet cell adenoma was present in 73% of the patients. One had multiple endocrine neoplasia (MEN) type I. Eleven patients had adenomatosis, and in 4 of these the disease was diagnosed intraoperatively. Four patients with adenomatosis were from 3 MEN I kindreds and had associated lesions. Four patients without evidence of MEN I had nesidioblastosis. The results of portal venous sampling in 1 of these patients were compatible with a solitary adenoma in the pancreatic head (Fig 45-5). Two patients had islet cell hyperplasia. Six patients in all, from 3 kindreds, had MEN I.
Multifocal islet cell disease was present in 22% of this series of patients with endogenous hyperinsulinemia. Preoperative identification of such patients is helpful in planning surgery. Localized and multifocal hyperinsulinism can be distinguished by percutaneous transhepatic portal venous sampling. Nesidioblastosis can be either diffuse or focal. Because hypersecretion of insulin is not constant, concomitant systemic arterial or venous samples must be obtained at the time of portal venous sampling. Other islet cell abnormalities may be present in hyperinsulinemic patients with MEN I. Even if hyperinsulinism persists after subtotal pancreatic resection for functioning multifocal islet cell disease, it is more manageable postoperatively with administration of previously ineffective drug regimens.
Post-comment: Multiple endocrine neoplasia type I (also sometimes called multiple endocrine adenomatosis or MEA type I, or Wermer's syndrome) consists of adenomas of the parathyroid glands, the pancreatic islets, and the pituitary. Pancreatic islet cell adenomas may secrete insulin or gastrin, in which case they cause the aggressive peptic ulceration known as the Zollinger-Ellison syndrome. Recognition of diffuse beta-cell hyperplasia is facilitated both by the hypoglycemic clamp technique described in the previous abstract and by the absence of insulin-secreting ``hot spots'' on catheterization of the pancreatic venous drainage, as described in the paper from China presented in article 45-11.--P.K.B.
Membranoproliferative Glomerulonephritis: Prospective Clinical Trial of Platelet Inhibitor Therapy
Donadio, James V., Jr.; Anderson, Carl F.; Mitchell, John C., III; Holley, Keith E.; Ilstrup, Duane M.; Fuster, Valentin; Chesebro, James H.
N. Engl. J. Med., May 31, 1984, Vol. 310, Pg. 1421-1426, Ch. 46-2
Affiliation: Mayo Clinic and Found.
Abstract: The authors report a double-blind, placebo-controlled trial in 50 patients with membranoproliferative glomerulonephritis. Type I disease (mesangiocapillary glomerulonephritis) was present in 46 patients, and type II disease (intramembranous dense-deposit) was present in 4.
Dipyridamole, 75 mg, and aspirin, 325 mg, or identical placebo tablets were administered three times daily before meals for 1 year. Observations were made at 3-month intervals. Long-term follow-up extended up to 7 years. Treatment failure due to loss of renal function occurred before the end of 1 year in 7 patients in the placebo group and 3 in the treatment group (table); if complications of treatment were also designated as failures, there were 8 in the placebo group and 5 in the drug group.
Average iothalamate clearance was estimated in 36 patients without treatment complications (Fig 46-1); it was 1.3 ml/minute per 1.73 sq m per 12 months in the treatment group and 19.6 ml/minute per 1.73 sq m per 12 months in the placebo group. No differences between the groups were observed in proteinuria, amount of hematuria or blood casts in the urine sediment, or serum complement concentrations (C3, C4, and CH$sub{5}$sub{0}).
At baseline, the half-life of $sup{5}$sup{1}Cr-labeled platelets was reduced in 12 of 17 patients. The half-life became longer and renal function stabilized in the treated group, as compared with the placebo group, suggesting a relation between platelet consumption and the glomerulopathy.
Nine of the 19 patients in the placebo group acquired end-stage renal disease (requiring dialysis) 10-63 months into the study. Their pretreatment iothalamate clearance rates ranged from 15 to 81 ml/minute per 1.73 sq m. In the treatment group, 20 of the 21 patients continued to take dipyridamole and aspirin after completion of masked therapy, and 3 of the 21 acquired end-stage renal disease 37-70 months after entry into the study. Their pretreatment iothalamate clearance rates ranged from 43 to 102 ml/minute per 1.73 sq m.
The differences between the two groups in the numbers of patients with end-stage renal disease and the times from entry into the study to its development suggest that stabilization of the glomerular disease may be an achievable goal with dipyridamole and aspirin.
Post-comment: The hallmark of membranoproliferative glomerulonephritis is a proliferation of mesangial cells with a marked increase in the matrix of the mesangium, which extends along the inner aspect of the glomerular capillary to form a characteristic ``double-contour'' pattern. Two distinct subtypes can be distinguished by the pathologist. In type I, electron-dense deposits, C3, and properdin are distributed along the endothelial side of the basement membrane and in the mesangium in a granular pattern. In type II, the dense material occurs in a unique ribbon-like deposit along the basement membrane. Patients often present with both hematuria and the nephrotic syndrome, and the serum complement level is characteristically low. The prognosis is serious; in one series of 84 patients, 25% had lost all renal function after 6 years, 11% had chronic renal failure, 29% had the nephrotic syndrome, 38% had asymptomatic proteinuria, and only 5% appeared to be in a spontaneous remission.
The rationale for the use of platelet-inhibitor drugs stems from demonstrations of platelet activation in glomerulonephritis and of the stimulation of smooth-muscle proliferation by platelet factors. Thromboxane, released by platelets, now appears to play an important role in experimental glomerulonephritis (see article 46-1), contributing to inflammatory damage and reducing glomerular functions.
This paper makes a major contribution, since it provides evidence that the natural course of membranoproliferative glomerulonephritis can be altered by therapeutic intervention with a regimen that is well tolerated. The proposed treatment with dipyridamole and aspirin is inexpensive and, except for occasional and reversible bleeding problems, relatively innocuous. In a similar prospective controlled trial, treatment with warfarin and dipyridamole also stabilized renal function and decreased protein excretion in patients with membranoproliferative glomerulonephritis, though bleeding was a frequent complication. Though these results are encouraging, it should be emphasized that platelet inhibiting drugs are not a cure-all for glomerulonephritis and they should be given at present only to patients in whom the specific diagnosis of membranoproliferative glomerulonephritis has been confirmed by an adequate renal biopsy.--F.H.E.
Associated Focal and Segmental Glomerulosclerosis in the Acquired Immunodeficiency Syndrome
Rao, T. K. Sreepada; Filippone, Edward J.; Nicastri, Anthony D.; Landesman, Sheldon H.; Frank, Elliot; Chen, C. K.; Friedman, Eli A.
N. Engl. J. Med., Mar. 15, 1984, Vol. 310, Pg. 669-673, Ch. 46-3
Affiliation: SUNY, Downstate Med. Center
Abstract: Of 92 patients with acquired immunodeficiency syndrome (AIDS) seen between 1981 and September 1983, 36 were addicted to intravenous drugs, 22 were bisexual or homosexual, 30 were Haitian immigrants, and 5 had no known risk factors. Nine patients had the nephrotic syndrome (urinary protein excretion of more than 3.5 gm/24 hours), and 2 had azotemia with lesser amounts of urinary protein. Five of these 11 patients gave a history of intravenous heroin addiction for 2-6 years. Five of the other 6 patients, in whom there were no known predisposing factors, had a distinct glomerulopathy characterized by focal and segmental glomerulosclerosis with segmental deposit of IgM and C3. In the sixth patient a mild immune complex glomerulonephritis with IgG and C3 and interstitial nephritis was found.
The observations point strongly to an association of a distinct nephropathy with AIDS. In idiopathic focal and segmental glomerulosclerosis the lesions are present at onset of the disease and generally progress to renal failure in 2-3 years. Among the authors' patients, even in those in whom the creatinine clearance was greater than 20 ml/minute at diagnosis, the mean duration to onset of uremia was 7.1 months. In the authors' experience with AIDS, associated renal disease caused rapid deterioration, leading to severe renal insufficiency in 8-16 weeks, a decline not previously observed in other patients with focal and segmental glomerulosclerosis.
Patients with AIDS typically have altered cell-mediated immunity characterized by lymphopenia with reversal of the normal ratio of the helper to suppressor lymphocytes, cutaneous anergy, and suppressed lymphocyte responsiveness to mitogens and antigens in vitro. Some cases of nephrotic syndrome may be related to altered T-lymphocyte function.
Post-comment: It's not surprising that focal sclerosing glomerulonephritis, often progressing rapidly to renal failure, should occur in AIDS, since recurrent and severe infections, with invasion of the bloodstream, characteristically complicate the disorder. The same circumstances presumably predispose to focal glomerulosclerosis in intravenous drug addiction. As in other patients with focal segmental glomerulosclerosis, heavy proteinuria portends an accelerated downhill course.--F.H.E.
Controlled Trial of Methylprednisolone and Chlorambucil in Idiopathic Membranous Nephropathy
N. Engl. J. Med., Apr. 12, 1984, Vol. 310, Pg. 946-950, Ch. 46-4
Abstract: Despite emerging evidence that some patients respond to corticosteroid or cytotoxic agents, treatment of idiopathic membranous nephropathy is controversial. In a randomized, prospective study, the authors assessed the effects of a new regimen based on oral administration of methylprednisolone (0.4 mg/kg per day) or prednisone (0.5 mg/kg per day) and chlorambucil (0.2 mg/kg per day), each given in alternate months for 6 months. The treated group comprised 32 patients and the control group, 30. All had nephrotic syndrome (urinary protein excretion exceeding 3.5 gm/day for at least two consecutive measurements, plasma albumin concentration of less than 2.5 gm/100 ml, and variable edema) and membranous nephropathy confirmed by renal biopsy.
At the end of follow-up (mean, 31.4 months for the treated group), 12 patients were in complete remission, 11 were in partial remission, 9 were unchanged, and none was worse (Fig 46-2). Among the 30 controls (mean follow-up, 37 months), 2 were in complete remission, 7 were in partial remission, 13 were unchanged, 7 were worse, and 1 with impaired renal function had died of cardiac infarction (Fig 46-3). Thus, 30% of the control group were in complete or partial remission, whereas 72% of the treated group were in complete or partial remission. Complete remission was defined as a reduction of urinary protein excretion to 0.2 gm/day or less, partial remission as protein excretion between 0.21 and 2 gm/day with normal plasma creatinine concentrations, unchanged condition as protein excretion of more than 2 gm/day and plasma creatinine concentrations that were stable or increased less than 50% over basal values, and worsened condition as plasma creatinine increases of at least 50% over baseline. No patient in the treated group had any increase in plasma creatinine concentration after follow-up, whereas 8 controls had increases of at least 50% over basal values.
Side effects were mild and rare. Decreases in white blood cell counts to 3,000 and 3,600/cu mm in 2 patients were reversed when the dose of chlorambucil was reduced by half. Tremor (2 patients), anxiety (1), and cramps (2) reported during corticosteroid administration disappeared after it was discontinued.
The patients in this study were adults, and chlorambucil was alternated with corticosteroid, which seems to offer protection from the gonadal toxicity of the alkylating agent reported by others. The total doses of chlorambucil, ranging from 0.9 to 1.4 gm for only 3 months, were considerably below those that have reportedly caused acute leukemia or lymphoma.
The better results demonstrated in this study may be attributable not only to the prolonged period of therapy but also to the addition of chlorambucil to corticosteroid.
Post-comment: You may recall that 5 years ago Cecil Coggins and a collaborative group of American nephrologists reported that prednisone treatment of adults with membranous nephropathy and the nephrotic syndrome was beneficial over the long term. Renal function deteriorated more slowly in treated patients compared with untreated controls, though there was no marked effect on proteinuria. The present paper from a consortium of Italian nephrologists suggests a much more impressive therapeutic effect when chlorambucil is alternated with prednisone each month for 6 months. In almost half of their patients, proteinuria disappeared within 6 months, and over the span of 2 years, glomerular function was well maintained. This form of treatment deserves further controlled trials in patients with idiopathic membranous glomerulonephritis, and it should be seriously considered--as the best treatment currently available--in patients with this diagnosis.--F.H.E.
Late Recurrence of Minimal Lesion Nephrotic Syndrome
Pru, Cesar; Kjellstrand, Karl M.; Cohn, Richard A.; Vernier, Robert L.
Ann. Intern. Med., January 1984, Vol. 100, Pg. 69-72, Ch. 46-5
Affiliation: Univ. of Minnesota
Abstract: The authors studied data in 16 patients who had the typical nephrotic syndrome of childhood, had been free of disease for 4-25 years, and had a late recurrence. Clinical criteria for selection of patients for this study were: development of nephrotic syndrome (edema, protein excretion of more than 3 gm/day, and serum albumin concentration of less than 2.5 gm/dl); age younger than 10 years at onset; rapid response to corticosteroid treatment during the initial episode and recurrences; and interval of more than 4 years free from nephrotic syndrome without treatment followed by recurrence during follow-up. All patients initially responded to treatment with adrenocorticosteroids, but 8 also received either nitrogen mustard or cyclophosphamide because of frequent relapses, corticosteroid intoxication, or both.
After remission lasting for a mean of 10.3 years, nephrotic syndrome recurred in each patient. All patients had at least one further relapse after the late recurrence. Late recurrences were associated with and seemed to be precipitated by upper respiratory tract infection in 5 patients and by pregnancy in 1. The nephrotic syndrome did not respond to treatment with high doses of corticosteroids in 2 patients until the infection was under control.
Renal biopsy specimens taken from 8 patients during a late recurrence showed few abnormalities in 7 and focal segmental glomerulosclerosis in 1. Renal biopsy is considered to be unnecessary, for experience suggests that the response of the late relapse to corticosteroid therapy should confirm the diagnosis of minimal change disease as it does that of nephrotic syndrome in the typical age group (2-9 years).
At the end of the follow-up period (mean, 20.6 years), all patients had normal renal function as shown by normal serum creatinine, blood urea, or creatinine clearance values.
Post-comment: In the patient with the longest interval since his childhood nephrosis, 25 years elapsed until the nephrotic syndrome recurred in adulthood! The main point is that, even after so many years, an earlier remission generally signifies a good prognosis, and promises a prompt therapeutic response to prednisone. The exceptions to this rule are likely to have developed focal glomerular sclerosis.--F.H.E.
Predicting Diabetic Nephropathy in Insulin-Dependent Patients
Mogensen, C. E.; Christensen, C. K.
N. Engl. J. Med., July 12, 1984, Vol. 311, Pg. 89-93, Ch. 47-1
Affiliation: Aarhus, Denmark
Abstract: Diabetic nephropathy develops in as many as 45% of insulin-dependent diabetics. A means of predicting the likelihood of renal disease developing would help in selecting patients for trials of intensified insulin therapy. The value of microalbuminuria, with a urinary albumin excretion rate of 15-150 mu g/minute, in predicting the development of increased proteinuria was assessed in 43 male diabetics who were aged 20 years and younger when diabetes was diagnosed and had had diabetes for 7 to 19 years when first examined. Renal function was reevaluated at least 7 years after the initial examination. The patients initially excreted less than 0.5 gm of protein per 24 hours. The mean follow-up was 10.4 years.
Four patients had increased urinary albumin excretion at follow-up, and 12 developed diabetic nephropathy. Ten of the latter patients had overt proteinuria at follow-up. Renal function tended to fall when albumin excretion rose above 70 mu g/minute. Patients with initial excretion rates below 15 mu g/minute generally did not have proteinuria at follow-up. Glomerular filtration rates generally were quite elevated in patients with initial albumin excretion rates below 70 mu g/minute but in whom nephropathy progressed, averaging 157 +- 7 ml/minute compared with 134 +- 9 ml/minute in patients whose disease did not progress. The renal plasma flow fell in patients without progression, in whom glomerular filtration remained stable. Blood pressure rose significantly in patients with progression. Those with increased albumin excretion rates were at high risk of developing proliferative retinopathy. Two patients with progression died, 1 of uremia and 1 of an unknown cause.
Incipient diabetic nephropathy, characterized by increased urinary albumin excretion during standard diabetic control and by a high risk of diabetic nephropathy, is observed in some young insulin-dependent diabetics. Clinical renal disease is not likely to develop for at least a decade when the initial albumin excretion rate is less than 15 mu g/minute.
Post-comment: Although morphological lesions develop in the glomeruli of all diabetics after a few years, only about half of all diabetic patients eventually develop clinical signs of diabetic nephropathy including proteinuria, hypertension, and declining renal function. It would therefore be of great interest to determine what factors predict or predispose to progressive nephropathy. Two early findings appeared to predict later renal disease in this 7-year study of juvenile diabetes: the degree of microalbuminuria (measured by radioimmunoassay, since these low levels are not detectable by the usual tests) and an elevated glomerular filtration rate. Other studies confirm the suggestion that microproteinuria forecasts macroproteinuria (and the idea seems intuitively reasonable). This is, however, the first prospective study of human diabetics that supports the hypothesis that hyperfiltration plays an important role in the pathogenesis of diabetic nephropathy.--F.H.E.
Drug-Associated Acute Interstitial Nephritis: Clinical and Pathologic Features and the Response to High-Dose Steroid Therapy
Pusey, Charles D.; Saltissi, David; Bloodworth, Lionel; Rainford, David J.; Christie, John L.
Q. J. Med. (New Series), Spring, 1983, Vol. 52, Pg. 194-211, Ch. 48-1
Abstract: The authors treated nine episodes of histologically proved acute interstitial nephritis in 7 patients (4 men), aged 19-55 years. Despite the variety of clinical features and the absence of other allergic phenomena, all patients treated with high dosages of corticosteroids had dramatic responses.
The drugs involved were cotrimoxazole, ampicillin, penicillin, gentamicin, paracetamol, and bendrofluazide. The intervals between administration of the drug and presentation ranged from 24 hours to 30 days. Rash and fever were present in four episodes each; they were antibiotic induced. Loin pain was a significant feature in 2 cases. Blood pressure was normal on presentation in 6 patients.
An intravenous urogram showed the kidneys to be enlarged in 5 patients and relatively enlarged in 2. The white blood cell count was elevated in 3 patients, and mild eosinophilia was present in 1 patient. Concentrations of IgG, IgM, and IgA were normal in most instances.
All episodes led to acute renal failure; five were nonoliguric. Seven episodes of acute interstitial nephritis were treated with high intravenous doses of methylprednisolone; all responded by onset of diuresis or a spontaneous fall in serum creatinine concentration within 72 hours. Renal function returned to near normal, with creatinine clearance rates ranging from 68 to 90 ml/minute. Of 2 patients admitted before the present regimen of high-dose methylprednisolone, 1 recovered renal function slowly, and disease in the other progressed to chronic renal impairment.
An inflammatory infiltrate was always present on light microscopic examination. It tended to be lighter and more diffuse in the cortex and heavier and more focal in the medulla. Immunofluorescence was absent in the 4 cases studied in the acute phase and showed scattered deposits of IgG, IgM, IgA, and C3 on the tubular basement membrane in 1 patient.
The drug most frequently implicated in acute interstitial nephritis is methicillin; ampicillin, sulfonamides, rifampin, phenindione, and glafenine are also frequently implicated. Presenting symptoms frequently include systemic manifestations of an allergic response, such as rash, fever, and arthralgia. Proteinuria is common. Microscopic examination of the urine usually yields abnormal findings, and excess red blood cells (66%), white blood cells (78%), and casts (33%) were seen in this study.
Drug-induced acute interstitial nephritis should always be suspected as a cause of acute renal failure, particularly in the absence of an obvious precipitant of acute tubular necrosis. A history of recent drug ingestion is essential. Renal biopsy confirms the diagnosis and can be safely performed. High-dose corticosteroid therapy appears to shorten the course of illness and insure full recovery of renal function.
Post-comment: Treatment of acute interstitial nephritis with steroids, employing a regimen similar to that used for an episode of transplant rejection, often produces dramatic improvement; in this series a response was usually seen within 72 hours. The most direct and effective way to make the diagnosis is by renal biopsy; interstitial nephritis can be virtually excluded if an inflammatory infiltrate is not seen. Heavy proteinuria in the range of 2-3 gm/day and hematuria are not uncommon, often suggesting the mistaken diagnosis of glomerulonephritis (for example, in patients treated with antibiotics for endocarditis). Fever and rash need not be present. The predominant cells in the infiltrate are T lymphocytes, consistent with the hypothesis of a delayed hypersensitivity mechanism. The gallium-67 scan, earlier suggested as an aid in diagnosis, is of no practical value. An interesting feature of the present report is that 1 patient was thought to have a reaction to gentamicin, which usually causes toxic but not inflammatory renal injury.--F.H.E.
Hypertension and Renal Failure (Scleroderma Renal Crisis) in Progressive Systemic Sclerosis: Review of a 25-Year Experience With 68 Cases
Traub, Yehuda M.; Shapiro, Alvin P.; Rodnan, Gerald P.; Medsger, Thomas A.; McDonald, Robert H., Jr.; Steen, Virginia D.; Osial, Thaddeus A., Jr.; Tolchin, Sanford F.
Medicine (Baltimore), November 1983, Vol. 62, Pg. 335-352, Ch. 48-2
Affiliation: Univ. of Pittsburgh
Abstract: Development of scleroderma renal crisis (SRC) was defined by the following criteria: (1) onset or aggravation, usually abrupt, of arterial hypertension (blood pressure higher than 160/90 mm Hg); (2) appearance of grade III retinopathy (flame-shaped hemorrhages, cotton-wool exudates, or both) and grade IV retinopathy (papilledema); (3) elevation of plasma renin activity (PRA) to twice the upper limit of normal or higher; and (4) rapid deterioration of renal function (less than 1 month).
In the patient population were 48 women (71%) and 11 blacks (16%). Of blacks with progressive systemic sclerosis (PSS), 20.8% had SRC, in contrast with 7.4% of whites, a highly significant difference. Scleroderma renal crisis occurred at the average age of 49.4 years and 3.0 years after the initial symptoms of PSS. The mean of last blood pressure readings recorded before SRC was 133/80 mm Hg, and the mean of highest values during SRC was 208/124 mm Hg in 48 patients. Funduscopic examinations in 59 patients revealed retinopathy consistent with accelerated or malignant arterial hypertension in 86%.
This report covers cases seen over 26 years when types of antihypertensive drugs, treatment of renal failure, and general supportive efforts were in flux. Before 1971, no patient lived longer than 1 year; usual survivals ranged from 1 to 3 months. Renal dialysis, more effective treatment of severe hypertension, and bilateral nephrectomy in selected anuric patients improved longevity. Only recently, a group of 11 patients has survived for longer than 1 year. Use of minoxidil and other potent antihypertensive therapy, including PRA suppression with beta-blockers, prolonged the average length of survival, but rarely prevented progress of renal failure and need for dialysis. A new era in the management of SRC seems to have been opened with the advent of angiotensin-converting enzyme inhibitors. The malignant course of SRC was reversed in 4 of 6 patients who received captopril. This agent has a number of adverse effects among which the most threatening are severe neutropenia and agranulocytosis. The urgency of starting treatment early in these patients must be emphasized.
The clinical characteristics of onset and progress of SRC suggest the sudden imposition of severe stress such as cold and autoimmune insult affecting vulnerable arteries and arterioles. Renal damage becomes self-perpetuating, with extremely high PRA causing a further rise in blood pressure and more renal and systemic damage. Pharmacologic agents that specifically block the effect of angiotensin II by inhibiting angiotensin I-converting enzyme have brightened the outlook for patients with SRC.
Post-comment: Though hypertension is not prominent in all cases of scleroderma renal crisis, the most important feature of successful treatment of scleroderma kidney appears to be the early and effective control of malignant hypertension. Although captopril is a logical drug to use, because renin levels are characteristically very high, it is not entirely clear that it is more effective in prolonging life and reversing renal failure than other antihypertensive agents, notably the calcium-channel blockers. Early prediction of renal involvement in scleroderma would be useful if it permitted preventive treatment that would avoid hypertensive exacerbations. Dr. Rodnan's group in Pittsburgh examined this question in another publication. Surprisingly, they found that isolated mild hypertension, urinary abnormalities, or hyperreninemia were not reliable predictors of subsequent renal failure (scleroderma crisis). However, a subset of patients with widespread and increasing skin thickening, who developed anemia, pericardial effusion, or congestive heart failure, commonly later developed scleroderma renal crises.--F.H.E.
Polycystic Kidney Disease: Prospective Analysis of Nonazotemic Patients and Family Members
Gabow, Patricia A.; Ikle, David W.; Holmes, Joseph H.
Ann. Intern. Med., August 1984, Vol. 101, Pg. 238-247, Ch. 48-3
Affiliation: Univ. of Colorado
Abstract: Review was made of 495 members of families having at least 1 adult affected with autosomal dominant polycystic renal disease and a serum creatinine level below 3 mg/dl. Fifty-nine children less than 12 years old were included. In all, 148 families were represented. There was no evidence of polycystic renal disease in 250 subjects; however, 81 others had suspected disease and 164 (33%) had definite disease.
The mean ages at the time of diagnosis of suspected or definite disease were 28.5 years and 30.4 years, respectively. Symptoms were absent in 32% of the patients with disease and present in 30% of those without disease. Hypertension, palpable kidneys, and a palpable liver were significantly more common in patients with disease, but a systolic murmur unrelated to hypertension and peripheral edema also were frequent findings. Normal laboratory values did not exclude polycystic renal disease. Ultrasonography was more sensitive than excretory urography in detecting disease, and it also detected hepatic cysts. Liver cysts were present in 38% of patients with definite disease, in 7% with suspected disease, and in 2% of those without disease.
Symptoms and routine laboratory data are not helpful in identifying family members with polycystic renal disease. Normal-sized kidneys and normal renal architecture seen on urography do not exclude the diagnosis. Ultrasonography is a reliable means of diagnosing polycystic renal disease, although some patients may have equivocal findings. The liver, ovaries, and perhaps the heart should also be examined if ultrasonography is done. Berry aneurysms are an important cause of morbidity and mortality in nonuremic patients with polycystic kidney disease.
Post-comment: Although the prevalence of cerebral aneurysm in patients with polycystic kidneys is estimated at 10%-30%, a decision analysis of risks and benefits indicates that cerebral arteriography should not be carried out routinely. In contrast to Dalgaard's initial assertion that first-degree relatives develop renal failure at about the same age, many observations have now invalidated that ``rule.'' Physicians tend, naturally, to remember patients who are sick and require medical attention rather than those who are well and don't consult their doctor. The gloomy prognosis often attached to the diagnosis of polycystic disease needs to be reevaluated in this light. Three out of 4 unselected polycystic patients will not develop end-stage renal disease by age 50; one half will not develop it by age 73. Obligatory screening by computed tomographic scan of children in families at risk is not recommended by Dr. Jared Grantham because there is no cure for the disease and because the early diagnosis may hamper insurability and employability (and, I would add, may impose unwarranted psychological restrictions).--F.H.E.
Effects of Sulindac and Ibuprofen in Patients With Chronic Glomerular Disease: Evidence for the Dependence of Renal Function on Prostacyclin
N. Engl. J. Med., Feb. 2, 1984, Vol. 310, Pg. 279-283, Ch. 48-4
Affiliation: Rome
Abstract: The authors examined the renal vasodilator metabolites of arachidonic acid in 20 consecutive hospitalized women with chronic glomerular disease by measuring their urinary excretion of prostaglandin E$sub{2} and of 6-keto-prostaglandin F$sub{1}$sub{alpha}. The functional importance of renal prostaglandin E$sub{2} and prostacyclin synthesis was determined by randomization of patients to short-term treatment with ibuprofen (1.2 gm/day) previously shown to inhibit renal prostaglandin E$sub{2} production, or sulindac (0.4 gm/day) which does not inhibit prostaglandin E$sub{2} and prostacyclin synthesis. The 20 patients were compared with 19 healthy women.
Mean serum creatinine concentration was 1.1 mg/dl, mean creatinine clearance rate was 90 ml/minute per 1.73 sq m, and mean para-aminohippurate clearance was 464 ml/minute per 1.73 sq m in the 20 patients with glomerulonephritis. Mean urinary excretion rate of prostaglandin E$sub{2} was 8.9 ng/hour compared with 7.6 ng/hour in the 19 healthy women.
The two groups of patients randomly assigned to 1-week treatment with ibuprofen or sulindac had similar baseline values. However, 4 patients receiving ibuprofen but only 1 receiving sulindac had systemic lupus erythematosus. Sulindac treatment did not alter significantly urinary excretion of prostaglandin E$sub{2} or 6-keto-prostaglandin F$sub{1}$sub{alpha}, despite an 85% reduction of platelet thromboxane B$sub{2} production that returned to control values when the drug was discontinued (Fig 48-1). Serum creatinine concentrations and creatinine and para-aminohipurate clearance rates were unchanged during treatment (Fig 48-2). Ibuprofen, which produced similar inhibition of platelet cyclo-oxygenase activity, also reduced mean urinary excretion rates of 6-keto-prostaglandin F$sub{1}$sub{alpha} and prostaglandin E$sub{2} by 79% and 81%, respectively. After the drug was stopped, platelet and renal cyclo-oxygenase activities were completely reversed. The serum creatinine concentration increased by about 40% during ibuprofen treatment, but mean creatinine and para-aminohippurate clearance rates were decreased by 28% and 35%, respectively. Mean sodium clearance and plasma renin activity were reduced by 29% and 59%, respectively, during ibuprofen treatment.
In 5 healthy women treated for 1 week with ibuprofen, mean baseline values for creatinine clearance, urinary excretion of prostaglandin E$sub{2}, urinary excretion of 6-keto-prostaglandin F$sub{1}$sub{alpha}, and serum thromboxane B$sub{2} concentration were 112 ml/minute, 7.6 ng/hour, 3.7 ng/hour, and 351 ng/ml, respectively. During treatment, creatinine clearance was virtually unchanged, despite profound suppression of urinary prostaglandin E$sub{2} and 6-keto-prostaglandin F$sub{1}$sub{alpha} excretion (means of 78% and 76%, respectively) and of platelet thromboxane B$sub{2} production (mean of 93%).
Dependence of renal function on cyclo-oxygenase may reflect intrarenal as well as extrarenal mechanisms. Reduced urinary excretion of 6-keto-prostaglandin F$sub{1}$sub{alpha} may be of diagnostic value in ident identifying patients at risk. Sulindac may be a suitable substitute for other nonsteroidal anti-inflammatory drugs in patients with mild impairment of renal function at risk of having adverse renal effects.
Post-comment: The adverse consequences of prostaglandin inhibitors in patients with renal impairment have received a lot of attention, and I have commented on them in previous editions of the YEAR BOOK. Sulindac is a unique anti-inflammatory drug in that it is a pro-drug (sulindac sulfoxide), inactive in its own right, that requires reduction in the liver to the active metabolite, sulindac sulfide. Sulindac sulfide, on the other hand, is oxidized by the kidney back to an inactive form so that under normal circumstances renal prostaglandin synthesis is not inhibited by a therapeutic dose of the drug given to suppress inflammation elsewhere. That makes sulindac sound like the drug of choice in heart failure or renal disease. Note, however, that sulindac sulfide is excreted in the bile. If biliary obstruction or parenchymal liver failure is present, sulindac sulfide may accumulate in the bloodstream and exceed the kidney's ability to oxidize it. So beware of sulindac if your patient has both liver and kidney disease.--F.H.E.
Acute Renal Failure Caused by Extreme Hyperphosphatemia After Chemotherapy of an Acute Lymphoblastic Leukemia
Cancer, July 1984, Vol. 53, Pg. 2425-2429, Ch. 49-2
Affiliation: Brest, France
Abstract: Lymphoproliferative syndromes can be responsible for acute renal failure (ARF). Hyperphosphatemia-induced nephrocalcinosis developed in a patient after he received chemotherapy for acute lymphoblastic leukemia; the pathologic findings suggested a direct relationship between hyperphosphatemia and the onset of ARF.
Boy, 15, with general deterioration in health, was found to have petechiae, enlarged lymph nodes, hepatosplenomegaly, and a mediastinal mass. Acute lymphoblastic leukemia was diagnosed by marrow aspiration. The chemotherapeutic drugs used included vincristine, daunorubicin, cyclophosphamide, cytosine arabinoside, and prednisone. Oligoanuric ARF developed 2 days after the start of chemotherapy. There was no evidence of disseminated intravascular coagulation. Hyperphosphatemia of 6.3 mmole/L (19 mg/dl) was found, as well as hypocalcemia of 1.3 mmole/L (5.2 mg/dl), hyperkalemia, and severe metabolic acidosis. Hemodialysis was undertaken, but the patient became febrile on day 8 and septic shock developed; he died of hemodynamic failure. Blood cultures yielded Escherichia coli. Autopsy showed pericardial effusion and enlarged kidneys with microlithiasis in all calices. The calcium deposits present in the renal tubules had a concentric structure and were found on electronic microsound analysis to consist of calcium phosphate.
Hyperphosphatemia-induced ARF in this patient resulted from acute tumor lysis associated with chemotherapy. Preventive measures include control of hyperuricemia, maintenance of correct hydration with abundant diuresis at neutral pH, and close monitoring of the levels of serum electrolytes, calcium, and phosphorus. The phosphorus concentration of the dialysate also should be monitored. Dialysis may be necessary if hyperphosphatemia persists. The role of phosphorus chelators remains to be established.
Post-comment: Although hyperuricemia is a notorious cause of acute renal failure following chemotherapy of lymphoma or leukemia, hyperphosphatemia is less well recognized. Several similar patients have been reported, most with acute or chronic lymphatic leukemia or Burkitt's lymphoma. In one patient, hemodialysis acutely reduced the serum phosphate level, and the patient regained normal renal function.
Another cause of renal failure during antineoplastic chemotherapy is thrombotic thrombocytopenic purpura with microangiopathy, causing the hemolytic uremic syndrome. Five patients with this unfortunate combination were reported; they had been treated with cisplatin, bleomycin, and a vinca alkaloid. Two recovered renal function, but the others died.--F.H.E.
Prevention of Acute Renal Failure in Traumatic Rhabdomyolysis
Ron, David; Taitelman, Uri; Michaelson, Moshe; Bar-Joseph, Gad; Bursztein, Simon; Better, Ori S.
Abstract: Rhabdomyolysis is a potential cause of myoglobinuric acute renal failure in victims of crush injury who sustain prolonged compression of the limbs or trunk. Early induction of an alkaline solute diuresis was tried in 7 men with traumatic rhabdomyolysis in an attempt to prevent renal failure. The average age was 25 years. All 7 had spent an average of more than 10 hours under the ruins of a collapsed building and sustained crush injuries to the lower extremities. All had neurologic abnormalities. Myoglobin was present in the urine, and all 7 had hyperphosphatemia. Hyperkalemia was noted in 5.
The goals were rapid attainment of a urinary pH above 6.5 and maintenance of diuresis of at least 300 ml/hour. A solution containing sodium chloride, 75 mEq/L, in 5% dextrose was rapidly infused via a central venous line. Sodium bicarbonate, 44 mEq/L, was added to alternating 500-ml bottles. If a response was not obtained but the central venous pressure rose by more than 5 cm H$sub{2}O, a 20% mannitol solution was given in a dose of 1 gm/kg. An average of 568 ml/hour of fluid was needed to maintain adequate diuresis. A markedly positive fluid balance resulted despite the use of an average of 160 gm of mannitol daily. The injured extremity became swollen, but pulses remained palpable and new neurologic deficits did not develop. An average of 685 mEq of bicarbonate was given over 60 hours, along with an average of 1.5 doses of 250 mg of acetazolamide to prevent alkalemia. Hyperkalemia resolved rapidly. The serum creatinine level never rose above 1.5 mg/dl, and the creatinine clearance rate never fell below 100 ml/minute. No patient required dialysis. Renal function was normal on day 5 or at discharge.
Renal failure did not occur in these patients with severe ``crush syndrome.'' A lesser diuresis might be effective and not require catheterization of the bladder, but the minimal rate of urine flow required to prevent renal failure is unknown.
Post-comment: This remarkable success in preventing the `'crush kidney'' syndrome of acute renal shutdown was achieved by applying principles learned from the study of animals with experimental myoglobinuric kidney failure. The key was early establishment of a brisk alkaline diuresis and fearless administration of salt and water to expand blood volume and protect the circulation from massive ``third spacing'' owing to the sequestration of fluid in crushed muscles. The average positive balance of fluid after 21/2 days of treatment was 12 L! Creatine phosphokinase exceeded 30,000 units/ml in all patients, but none developed renal failure. The treatment was effective in 2 patients who had been trapped under the collapsed building for more than 24 hours before rescuers arrived, suggesting that the events leading to renal failure, including release of myoglobin into the circulation and trapping of fluid in injured muscle, may not gain momentum until release from compression.--F.H.E.
Low Fractional Excretion of Sodium: Occurrence With Hemoglobinuric- and Myoglobinuric-Induced Acute Renal Failure
Corwin, Howard L.; Schreiber, Martin J.; Fang, Leslie S. T.
Abstract: The authors reviewed the records of 10 patients, whose mean age was 50 years and who had myoglobinuric and hemoglobinuric acute renal failure. They had recorded urinary diagnostic indices allowing for calculation of fractional excretion of sodium (FE$sub{N}$sub{a}): FE$sub{N}$sub{a} = [(urine sodium/plasma sodium) divided by (urine creatinine/plasma creatinine)] X 100.
The diagnosis of acute renal failure was made when there was a 50% or greater increase in the baseline serum creatinine concentration. Acute hemoglobinuric renal failure developed in 5 patients, and 5 had acute deterioration of function secondary to myoglobinuria. Antigen transfusion mismatches were detected in the 5 patients with hemoglobinuria. Traumatic (2 patients) and nontraumatic (3 patients) rhabdomyolysis accounted for the myoglobinuria.
The mean baseline serum creatinine concentration was 1.1 mg/dl, and the value rose to a mean maximum of 6.9 mg/dl. The serum urea nitrogen concentration increased from a mean of 16 mg/dl to a mean maximum of 81 mg/dl. Although 3 patients required peritoneal dialysis or hemodialysis, all recovered renal function. Mean time to recovery was 18 days.
An FE$sub{N}$sub{a} value of less than 1% was consistently observed during the oliguric phase of myoglobinuric and renal failure. Patients with serial FE$sub{N}$sub{a} measurements demonstrated an increase in the FE$sub{N}$sub{a} value when they entered the nonoliguric phase. A low FE$sub{N}$sub{a} value was not indicative of more mild injury. The finding of a low FE$sub{N}$sub{a} value in hemoglobinuric and myoglobinuric acute renal failure suggests that a common mechanism may account for the renal injury in both types of pigment toxicity.
Post-comment:
The toxic agent in both myoglobinuria and hemoglobinuria is ferrihemate, which is found in acid urine (pH less than 5.6), so it's not surprising that these forms of pigment nephropathy should resemble each other clinically. What is peculiar is that a low urinary sodium should occur so frequently in the oliguric stages of renal failure induced by myoglobin or hemoglobin. Tubular obstruction doesn't seem like a good explanation, since acute uric acid nephropathy (which is caused by tubular obstruction) isn't characterized by a low concentration of sodium in the urine. A low concentration of sodium in the urine is, however, typically encountered in patients with tubular necrosis who have diminished glomerular perfusion (prerenal failure) at the same time. Examples are patients with circulatory shock, congestive heart failure, or the hepatorenal syndrome. We are not told by the authors what the circulatory status of their patients was. If their blood pressure was normal, it is possible that ferrihemate elicited a particularly pronounced tubuloglomerular feedback reaction, resulting in diffuse glomerular constriction.--F.H.E.
Effect of a Keto Acid-Amino Acid Supplement to a Restricted Diet on the Progression of Chronic Renal Failure
Mitch, William E.; Walser, Mackenzie; Steinman, Theodore I.; Hill, Sylvia; Zeger, Scott; Tungsanga, Kriang
N. Engl. J. Med., Sept. 6, 1984, Vol. 311, Pg. 623-629, Ch. 50-1
Abstract: Improved renal function has been observed in some patients with chronic renal failure given a diet containing 20-30 gm of protein supplemented with essential amino acids and the calcium salts of their keto analogues. This formulation was evaluated in 24 patients with established chronic renal failure and no correctable factors. The average age was 56 years. The most common diagnoses were chronic glomerulonephritis and hypertensive nephropathy. The mean initial glomerular filtration rate was 7.6 ml/minute. Serial serum creatinine measurements provided a well-defined rate of progression before treatment in 17 patients. Two mixtures containing varying amounts of serine, ketovaline, ketoleucine, and threonine were administered; both mixtures supplied 1.8 gm of nitrogen daily. Eight patients are continuing to receive this treatment.
Ten of 17 evaluable patients (59%) had clinically significant slowing of the rise in the serum creatinine level during treatment for an average of 20 months, compared with the predicted rate. None had a faster rise than predicted. In 6 of 7 patients in whom treatment was begun before the creatinine level reached 8 mg/dl, the level remained at or below baseline during an average follow-up period of 22 months. Nutrition was well maintained, as judged from the levels of nitrogen balance, serum albumin, and serum transferrin, and from body weight. Five of the 8 patients continuing treatment have been employed full-time, and 2 others do housework; the remaining patient is retired.
Progression of chronic renal failure was slowed or halted in most patients in this study. The mechanism remains to be established; however, if the findings can be generalized, the need for dialysis may be reduced substantially. Patients treated before the serum creatinine level reaches 8 mg/dl fare particularly well with this treatment.
Post-comment: In roughly half of all patients with moderate renal insufficiency who can be persuaded to adhere to this severely restricted diet, the formerly progressive fall in creatinine clearance rate seems to be slowed. In general, patients feel more energetic and less nauseated, and their blood urea nitrogen level is, of course, lower. What puzzles me is the early fall in serum creatinine level that is so frequently observed within the first month on a low-protein diet. I suspect that this will turn out to be an artifact that does not reflect a true rise in glomerular filtration rate. Creatinine reaches the final urine via tubular secretion as well as glomerular filtration, and it seems likely to me that nitrogenous products of protein metabolism compete with creatinine for secretion. Hence, creatinine would be more easily secreted, increasing its clearance and decreasing its plasma concentration, on a low-protein diet. Serial measurements of glomerular filtration rate by standard (noncreatinine) methods are necessary to settle the issue.--F.H.E.
Effect of Dietary Phosphorus on Circulating Concentrations of 1,25-Dihydroxyvitamin D and Immunoreactive Parathyroid Hormone in Children With Moderate Renal Insufficiency
Portale, Anthony A.; Booth, Beverley E.; Halloran, Bernard P.; Morris, R. Curtis, Jr.
J. Clin. Invest., June 1984, Vol. 73, Pg. 1580-1589, Ch. 50-2
Affiliation: Univ. of California at San Francisco
Abstract: Hyperparathyroidism in patients with moderate renal insufficiency could result from a reduced level of plasma ionized calcium induced by recurrent increases in levels of plasma inorganic phosphorus or by phosphorus-mediated suppression of renal 25-hydroxyvitamin D-1alpha-hydroxylase, leading to a reduced plasma level of 1,25-dihydroxyvitamin D (1,25-[OH]$sub{2}D) as well as decreased intestinal absorption and bone resorption of ionized calcium. The effects of short-term restriction and supplementation of dietary phosphorus were assessed in 8 children aged 6-17 years with moderate renal insufficiency. Glomerular filtration rates were about 25-50 ml/minute per 1.73 sq m. Six children had primarily tubulointerstitial renal disease and 2 had glomerular disorders. Seventeen healthy children aged 6-16 years were also studied. Dietary phosphorus was restricted to 350 mg daily for 5 days and was increased to 2,400 mg daily for a similar period.
The children with renal insufficiency had lower plasma levels of 1,25-(OH)$sub{2}D than did controls at baseline. An increase occurred in association with phosphorus restriction (Fig 50-1), with no significant change in the plasma 25-hydroxyvitamin D level. The serum parathyroid hormone level declined in most of the children. Fasting serum calcium and phosphorus levels were not significantly altered, but the fractional urinary excretion of phosphorus fell rapidly and progressively. The urinary calcium level increased significantly within 3 days of the start of phosphorus restriction. Phosphorus supplementation was associated with a decline in the plasma 1,25-(OH)$sub{2}D level, a rise in the serum parathyroid hormone level, and an increase in phosphorus excretion. Total urinary calcium excretion did not change significantly.
Induced changes in phosphorus intake lead to altered plasma levels of 1,25-(OH)$sub{2}D and resultant changes in serum parathyroid hormone levels. Even short-term dietary phosphorus restriction can increase the production of 1,25-(OH)$sub{2}D to normal in children with moderate renal insufficiency. The rate of phosphorus excretion per nephron is reduced to a normal level.
Post-comment: The classic version of the pathogenesis of changes in calcium and phosphorus in renal failure goes like this: With progressively decreasing excretory function, the phosphorus excretion rate falls and the serum phosphate levels rises, inducing a fall in the serum calcium level, which stimulates the secretion of parathyroid hormone, which restores serum calcium and phosphorus levels to normal by its action on bone and kidney. These direct and elegant studies of children with renal disease point to quite another sequence, as follows: Early phosphate retention depresses renal formation of 1,25(OH)$sub{2}D, which decreases intestinal absorption of calcium and increases parathyroid hormone (PTH) secretion via a feedback mechanism of vitamin D itself on the parathyroid gland. Restriction of phosphorus, on the other hand, as illustrated in these experiments, stimulates vitamin D formation by the kidney, which depresses PTH formation without altering the level of serum calcium. The new emphasis is on the primary role played by renal manufacture of the active form of vitamin D. An important question, not yet answered, is whether normal levels of 1,25(OH)$sub{2}D, attained in uremic subjects by exogenous administration, can suppress excessive PTH secretion without producing hypercalcemia.--F.H.E.
Predicting Success of Intensive Dialysis in the Treatment of Uremic Pericarditis
De Pace, Nicholas L.; Nestico, Pasquale F.; Schwartz, Allan B.; Mintz, Gary S.; Schwartz, J. Sanford; Kotler, Morris N.; Swartz, Charles
Am. J. Med., January 1984, Vol. 76, Pg. 38-46, Ch. 52-2
Affiliation: Philadelphia
Abstract: To identify factors predicting the response of patients with uremic pericarditis to intensive dialysis, a review was made of data on 97 episodes in patients with end-stage renal disease. The patients had chronic renal failure and a typical pericardial friction rub or, alternatively, hemodynamic evidence of cardiac tamponade. Sixty-seven patients responded to intensive hemodialysis or continuous peritoneal dialysis, whereas treatment failed in 30. Of the latter patients, 8 died of hemodynamic complications of pericarditis before surgery and 22 underwent operation. The average duration of intensive dialysis before surgery was 2 weeks. All 4 surgically treated patients who died had cardiac tamponade.
All patients with a temperature of more than 102 F or a white blood cell count of more than 15,000/cu mm, or who received only peritoneal dialysis because of severe hemodynamic instability, failed to respond to intensive dialysis. Other factors correlating with failure to respond were blood pressure below 100 mm Hg on hospital admission, jugular venous distention, a large effusion, and the presence of both anterior and posterior effusion on echography. Echographic left ventricular size and function were not useful predictors of the response to intensive dialysis. The time of development of pericarditis in relation to dialysis was not a factor. A seven-variable discriminant function classified all 12 patients evaluated prospectively in terms of response to intensive dialysis.
Patients with uremic pericarditis who are likely to respond to daily intensive dialysis can be selected fairly reliably, with alternative management considered for those unlikely to respond.
Post-comment: The authors have carefully evaluated 55 independent variables retrospectively to assess the prognosis of patients presenting with uremic pericarditis either before or during dialysis treatment. Not surprisingly, those patients with high temperature, leukocytosis and left shift of the white blood cell differential, large pericardial effusion, rales, or jugular venous distention fared worse than the others when treated with intensive dialysis alone. Many such patients required surgical pericardiotomy or pericardiectomy, and several died. Two questions must be raised about this study. First, is intensive dialysis an efficacious therapy, or would such patients fare better if they continued their regular dialysis schedule? Assuming that no other evidence of underdialysis is found, it is possible that increasing the frequency of hemodialysis with its attendant anticoagulation from thrice weekly to every day may actually have predisposed to more hemorrhagic pericarditis. Second, would simple pericardiocentesis, a procedure that was not utilized in this study, offer a better prognosis than more aggressive surgery, as suggested by the lower mortality rate with pericardiocentesis in a review by Silverberg et al.? Another technique for pericardial drainage using the subxiphoid placement of a large-bore tube under local anesthesia with instillation of triamcinolone has been recently proposed, but the death of 2 of 11 patients treated in this manner in a recent study raises a note of caution.--Robert Brown, M.D.
Evidence That Patients With Addison's Disease Are Undertreated With Fludrocortisone
Smith, Stephen J.; MacGregor, Graham A.; Markandu, Nirmala D.; Bayliss, John; Banks, Richard A.; Prentice, Malcolm G.; Dorrington-Ward, Peter; Wise, Peter
Lancet, Jan. 7, 1984, Vol. 1, Pg. 11-14, Ch. 54-5
Affiliation: Charing Cross Hosp. Med. School
Abstract: The authors studied data in 6 men and 4 women (mean age, 41 years) with primary adrenocortical failure. The patients were receiving standard maintenance therapy with either 30 mg of hydrocortisone per day or 37.5-50 mg of cortisone acetate per day in divided doses; all were taking 0.5 or 1.0 mg of fludrocortisone per day, except 1 patient with detectable aldosterone. Plasma renin activity was high (mean, 16.0 ng/ml per hour) for sodium intake (mean, 165 mmole/24 hours) in all patients at their initial visit (Fig 54-2). Four of 5 patients with symptoms (lack of energy, tiredness) had the highest plasma renin activities.
Fludrocortisone was withdrawn while the patients were in the hospital, resulting in an immediate significant increase in urinary sodium excretion and an average cumulative sodium loss of 121 mmole per patient, which was accompanied by a significant fall in mean weight of 1.2 kg by the sixth day of withdrawal. When a daily dose of 0.3 mg of fludrocortisone was given, there was a significant reduction in sodium excretion, leading to an average cumulative sodium retention of 350 mmole and a mean weight gain of 1.2 kg by the sixth day. With 0.3 mg of fludrocortisone per day there was a highly significant fall in plasma renin activity and an associated fall in blood urea and plasma potassium concentrations. Mean plasma volume increased to 107% (P < .005). When treatment with fludrocortisone was discontinued, there was a significant fall in standing systolic blood pressure and a narrowing of standing pulse pressure by 8 mm Hg.
At outpatient follow-up, fludrocortisone was reduced by 0.05-mg decrements as necessary every 2-4 weeks to a median dose of 0.2 mg/day. By the third month, 7 of the 10 patients had plasma renin activity at or just below the normal range for their sodium intake.
Measurement of plasma renin activity in conjunction with 24-hour urinary sodium excretion appeared to provide the most helpful guide to both undertreatment and overtreatment with fludrocortisone, provided that when the dose was increased, sufficient time was allowed for juxtaglomerular hyperplasia to resolve. Most patients appear to require about 0.2 mg of fludrocortisone per day to maintain adequate sodium balance on the current western sodium intake (about 150 mmole/day). Before the dose of fludrocortisone is increased for patients with Addison's disease, the beneficial effects of restoring sodium balance adequately should be weighed against the possible long-term consequences on blood pressure.
Post-comment: This is an interesting paper because it applies a well-established principle of endocrinology to salt and water physiology, to wit, ``That dose of a replacement hormone is adequate, which reduces the circulating level of its trophic hormone to normal.'' Renin (or more correctly, angiotensin II), is the chief known trophic hormone for the zona glomerulosa of the adrenal cortex, (though the abstract below provides speculation about other, unknown trophic hormones), so the proper replacement dose of mineralocorticoid is that which restores plasma renin activity to normal. The correct amount of fluorohydrocortisone needed to return plasma renin to normal in most of these 10 patients with Addison's disease proved to be 0.2 mg daily, about twice that previously prescribed. Another notable observation was that the high levels of plasma renin required 3-6 weeks to return to the normal range, after mineralocorticoid replacement, even though sodium balance leveled off after 1 week. The authors ascribe this to the time needed for juxtaglomerular hyperplasia to regress. A similar phenomenon may be responsible for the transient excessive salt retention observed when administration of diuretics is discontinued in chronically salt-depleted patients who are addicted to taking diuretics for ``idiopathic edema.''--F.H.E.
Impaired Vitamin D Metabolism With Aging in Women: Possible Role in Pathogenesis of Senile Osteoporosis
Tsai, Keh-Sung; Heath, Hunter, III; Kumar, Rajiv; Lawrence Riggs, B.
J. Clin. Invest., June 1984, Vol. 73, Pg. 1668-1672, Ch. 55-1
Affiliation: Mayo Clinic and Found.
Abstract: Aging is associated with decreased intestinal calcium absorption that can explain the rise in parathyroid function seen in elderly men and women. The metabolism of 25-hydroxyvitamin D [25(OH)D] to 1,25(OH)$sub{2}D, a major active form of the vitamin, was assessed by measuring the increase in serum 1,25(OH)$sub{2}D concentration after infusion of parathyroid hormone. Subjects were 10 normal premenopausal women whose mean age was 37 years, as well as in 8 normal women with a mean age of 61 years who were within 20 years of menopause, 10 normal elderly women with a mean age of 78 years, and 8 elderly women of similar age with proximal femoral fractures. No vertebral fractures were present. The hip fractures resulted from minor falls occurring more than 6 months before the study. Studies were done on a metabolic ward.
The elderly women had lower glomerular filtration rates, basal serum 1,25(OH)$sub{2}D levels, and increases in 1,25(OH)$sub{2}D levels after parathyroid hormone infusion. Increases in serum 1,25(OH)$sub{2}D activity were less in the group with hip fracture than in the normal elderly women, even after adjustment for glomerular filtration rate or age alone, and for both factors together. Urinary cyclic AMP responses to bovine parathyroid hormone were similar in all groups. In all women, increasing age was associated with reductions in the glomerular filtration rate and in basal serum 1,25(OH)$sub{2}D activity. The serum 1,25(OH)$sub{2}D response to parathyroid hormone infusion correlated inversely with age.
Elderly women appear to have abnormal vitamin D metabolism, particularly impaired conversion of 25(OH)D to 1,25(OH)$sub{2}D. The findings apply to patients with senile osteoporosis but not to those with postmenopausal osteoporosis. Senile osteoporosis has generally been attributed to bone-cell senescence and thus considered untreatable, but the present findings suggest that impaired calcium metabolism in elderly women may be correctable by treatment with a small dose of 1,25(OH)$sub{2}D, possibly reducing bone loss.
Post-comment: Riggs and his colleagues make a case for two kinds of osteoporosis in elderly women. In the first, or postmenopausal form, in women 51-65 years of age, fractures chiefly involve trabicular bone such as the vertebrae and radius. In these women the response of serum 1,25(OH)$sub{2} vitamin D to parathyroid infusion is usually normal, and the osteoporosis is ascribed primarily to estrogen withdrawal. In the second form, occurring predominantly in women older than 75, fractures usually involve the hip and other cortical bone. Serum levels of the active form of vitamin D are depressed and do not increase normally when parathyroid hormone is given. It is not yet clear whether the impaired renal hydroxylation of 25(OH) vitamin D is the result of the gradual reduction in functional renal mass that accompanies aging or of another process. It is this form of truly senile osteoporosis that, the authors speculate, might be prevented by small doses of 1,25(OH)$sub{2} vitamin D. In addition, there is some evidence of a primary deficit in calcium absorption in some elderly women who have had vertebral fractures. Inadequate dietary intake of calcium undoubtedly contributes to both forms of osteoporosis in aging women.--F.H.E.
Extracorporeal Shock Wave Lithotripsy (ESWL) for Kidney Stones: An Alternative to Surgery?
Chaussy, Christian; Schmiedt, Egbert
Urol. Radiol., April 1984, Vol. 6, Pg. 80-87, Ch. 56-1
Affiliation: Munich
Abstract: Four years of experience with extracorporeal shock wave lithotripsy (ESWL) for renal and ureteral calculi were reviewed. Shock waves offer the possibility of destroying calculi without contact. Stones of varying chemical composition are reproducibly disintegrated experimentally and in rats. The lithotripter uses shock waves produced outside the body that are focused at the stone and transmitted by body tissues. The waves are generated by an underwater spark discharge triggered by the electrocardiogram to avoid arrhythmias. A biplane x-ray system is used to localize kidney stones to be treated. Because the shock waves are painful, epidural anesthesia is administered to most patients. The average treatment lasts 54 minutes and involves application of about 1,000 shock waves. Forced diuresis is carried out after treatment. The average hospitalization has been about a week.
Of 1,012 treatments given to 896 patients, results were satisfactory in 99%, and 90% were rendered free of stones. There were no significant complications. Thirteen percent of patients required a second or third treatment to become free of stones. Only 0.5% of patients required surgical removal of residual stone fragments when inadequately disintegrated fragments produced obstruction. Clearance studies showed significantly improved function in the treated kidney 3 and 18 months after treatment.
Indications for ESWL have expanded significantly. Infected stones can be treated without a risk of pyelonephritis if antibiotics are given for 1-2 days before the procedure is undertaken. Passage of stone debris is less complicated than expected. Patients with severe medical disease can be treated more safely by ESWL than by surgery if careful anesthesiologic monitoring is assured. Ureteral stones present for less than 4-6 weeks can be treated successfully. The sole remaining important contraindication to ESWL is obstruction distal to the stone, and even here the procedure is feasible if preceded by percutaneous nephrostomy to facilitate discharge of stone fragments. It is expected that about 85% of patients with kidney stones can be cured nonsurgically.
Post-comment: I'd nominate the development of extracorporeal shock wave dissolution of kidney stones as the most important practical innovation in therapy of kidney diseases in the past several years. The equipment is very expensive and has thus far been installed in only a few medical centers in the United States and around the world. In each of these, there is currently a long waiting list for noninvasive lithotripsy. You can see why!
Not generally appreciated is the need for anesthesia because the shock waves are painful and because of the necessity to lie still and supine for an hour or more. The treatment turns the stone into a kind of paste that is passed in the urine. If the ureter is obstructed, the stone remnants will not pass. Because the shock waves will disrupt anything solid they are focused on, the presence of calcified arterial plaques close to the stone, which carry the risk of hemorrhage if disturbed, is thought to be a contraindication. Stones that have been impacted for 2-4 weeks or more may resist ultrasonic fragmentation, and they should be dislodged first. Ureteral stones can be treated easily, but not if they are located below the pelvic brim, which shields them from the sonic waves.
Ultrasonic energy can also be applied directly to stones that resist passage. This is done via a probe placed on the stone through a rigid fiberoptic nephroscope. The procedure, of course, requires a small skin incision but is far less incapacitating than a full-scale operation on the kidney. The average period of hospitalization is only 5 days, and patients can usually return to work in a week. Used together with appropriate antibiotic therapy and, in some cases, with inhibitors of urease activity, these procedures can eliminate infected staghorn calculi in almost all patients and prevent recurrence in most.--F.H.E.
Raised Transmembrane Oxalate Flux in Red Blood Cells in Idiopathic Calcium Oxalate Nephrolithiasis
Baggio, B.; Gambaro, G.; Marchini, F.; Cicerello, E.; Borsatti, A.
Lancet, July 7, 1984, Vol. 2, Pg. 12-13, Ch. 56-3
Affiliation: Univ. of Padua, Italy
Abstract: In idiopathic calcium oxalate nephrolithiasis, the high frequency of a family history of the disease and the findings of increased intestinal absorption and urinary excretion of oxalate suggest that it is a metabolic disease caused by defective cellular oxalate transport. This hypothesis was examined by studying the flux of $sup{1}$sup{4}C-labeled oxalate through red blood cell membranes in 28 patients with recurrent idiopathic calcium oxalate stone formation; the 18 men and 10 women were aged 25-46 years. All had passed at least one stone in the preceding year, and all had normal renal function, normal results on pyelograms, sterile urine, and normal biochemical findings. Twenty healthy individuals served as controls.
The mean flux constant was 0.29 L/minute in controls and 0.93/minute in the stone-forming patients, a significant difference. Of the 28 stone formers, 22 had a rate of oxalate flux more than 2 SD above the control mean. Study of 4 patients with abnormal transmembrane oxalate flux showed that the rate constant for flux returned to normal in the presence of inhibitors of band 3, a protein that transports anions across the red cell membrane.
The finding that patients with idiopathic calcium oxalate nephrolithiasis have more rapid red blood cell transmembrane oxalate flux than normal strengthens the view that the disorder is a metabolic disease characterized by a cellular defect in oxalate transport. The defect may reside in an alteration of band 3 protein. Knowledge of this defect may help identify individuals at risk so that stone formation can be prevented. In addition, red blood cells may be useful in evaluating drugs for preventing and treating nephrolithiasis.
Post-comment: While much attention has been lavished on possible abnormalities in the excretion of calcium in patients with calcium oxalate stones, systemic investigations of the role of oxalate are less common, possibly because oxalate is harder to measure. But you need both to form stones! In fact, as many as half of all recurrent stone formers excrete a quantity of oxalate in the urine that is more than 2 SD above the normal mean of 26 +- 10 mg/day (i.e., more than 46 mg). Against this background, the finding of a raised transmembrane flux of oxalate in erythrocytes of patients with calcium oxalate nephrolithiasis is especially interesting. If oxalate is transferred more easily into the red blood cells of such patients, it might also be absorbed more easily from the intestine or secreted more readily into the urine. All of these possibilities now need to be investigated, especially in that population of stone formers in whom urinary calcium levels seem to be normal and oxalate secretion seems high.--F.H.E.
Low-Dose Prednisone Therapy in Rheumatoid Arthritis: A Double-Blind Study
Harris, Edward D., Jr.; Emkey, Ronald D.; Nichols, Julia E.; Newberg, Arthur
J. Rheumatol., 1983, Vol. 10, Pg. 713-721, Ch. 58-2
Affiliation: Dartmouth-Hitchcock Med. Center, Hanover, N.H.
Abstract: The role of glucocorticoids in the treatment of rheumatoid arthritis (RA) remains controversial. The efficacy of 5 mg of prednisone, given daily to patients with active RA, was assessed in a double-blind study in 34 patients aged 18-75 years whose disease was diagnosed at least a year previously. Eighteen patients received prednisone, and 16 patients received placebo. After 24 weeks, all patients were given placebo for another 8 weeks (Fig 58-1). Patients with contraindications to steroid therapy were excluded. No patient previously received orally administered steroids. The 18 prednisone-treated patients and the 16 given only placebo were similar in age and in the duration of illness.
Slight functional improvement was observed in prednisone-treated patients during the 24-week treatment period, but patients' condition deteriorated during the 8 weeks when placebo was given. All study patients felt worse after the final placebo period than after 24 weeks of active treatment. Hand erosions progressed in 1 prednisone-treated patient and in 4 controls. Two steroid-treated patients sustained asymptomatic vertebral compression fractures. No other steroid toxicity was evident. Sedimentation rates declined during steroid therapy, but subsequently were similar to baseline values.
Low-dose prednisone therapy may be helpful as a link between nonsteroidal anti-inflammatory drug therapy and the use of disease-modifying agents in patients with RA. Even low-dose steroid therapy can accelerate osteopenia, and supplementary vitamin D therapy might be considered. A larger double-blind study is needed to confirm the efficacy of low-dose glucocorticoid therapy in RA. Treatment should be gradually tapered to prevent a clinical flare-up of arthritis.
Post-comment: Rheumatologists continue to wrestle with the problem of low-dose corticosteroids in rheumatoid arthritis. A dose of 5 mg of prednisone each morning (or sometimes on alternate days) seems to improve function and well-being in many patients beyond the benefit of their baseline regimens, without disturbing the hypothalamic-pituitary-adrenal axis or circadian rhythms of glucocorticoid excretion; however, even these low doses can accelerate osteopenia. This problem has no solution that is entirely satisfying. Many of us use low-dose regimens (generally with supplemental vitamin D and calcium) which we taper, often by as little as 1 mg at a time, when conditions improve. One might not expect a flare in inflammatory activity on abrupt withdrawal of only 5 mg of prednisone--i.e., less corticosteroid equivalent than is ordinarly made spontaneously--but this phenomenon is well known to practitioners and is confirmed in the current study.--S.E.M.
Chlamydia trachomatis Infections in Men with Reiter's Syndrome
Martin, David H.; Pollock, Scott; Kuo, Cho-Chou; Wang, San-Pin; Brunham, Robert C.; Holmes, King K.
Ann. Intern. Med., February 1984, Vol. 100, Pg. 207-213, Ch. 60-1
Affiliation: Univ. of Washington
Abstract: Chlamydia trachomatis has been implicated as a triggering agent in Reiter's syndrome. The prevalence of urethral infection due to C. trachomatis and Ureaplasma urealyticum was determined in 48 patients with Reiter's syndrome, defined as rheumatoid factor-negative peripheral arthritis present for more than 1 month in conjunction with urethritis. Two thirds of the men had other manifestations of the syndrome, usually conjunctivitis and skin or mucosal lesions. Twenty-eight of 34 patients had HLA-B27. Some patients had an episode of diarrhea before onset of disease.
All but 3 of 29 patients with acute nondiarrheal Reiter's syndrome had had a new sex partner or had been exposed to two or more partners within 3 months before onset of joint disease. Urethritis often was asymptomatic when these men were evaluated. The microbiologic findings are given in the table. Of 19 men, 9 with acute nondiarrheal Reiter's syndrome who had not recently taken antibiotics had C. trachomatis infection; C. trachomatis-specific antibody and cellular immune responses were found more often in these patients than in those with diarrhea-associated Reiter's syndrome or in men with other forms of arthritis. Peak antichlamydial titers and lymphocyte transformation indices were higher in the C. trachomatis-infected men with Reiter's syndrome than in C. trachomatis-infected men with uncomplicated nongonococcal urethritis.
Nondiarrheal Reiter's syndrome often appears to be secondary to sexually transmitted disease caused by C. trachomatis. If diagnosis is not feasible, men having objective evidence of urethritis should receive tetracycline or erythromycin. All patients with clinical findings known to be closely associated with C. trachomatis infection should be treated, and regular sexual partners of men with Reiter's syndrome also should receive appropriate antibiotic treatment.
Post-comment: Susceptible individuals--especially those with the human histocompatibility leukocyte antigen HLA-B27--develop reactive arthritis or full-blown Reiter's syndrome triggered in some way by enteric pathogens (Shigella flexneri, Salmonella, Yersinia, Campylobacter) or following sexual intercourse with new partners; chlamydia has been implicated in some of the latter patients. The current authors present excellent further circumstantial evidence for the initiating role of Chlamydia trachomatis (but not of Ureaplasma urealyticum). It is of note that by the time the patients in the present study presented with arthritis, they frequently denied symptoms of urethritis, despite microscopic evidence of its presence. In sexually active men with undiagnosed arthritis, the authors reasonably recommend a careful examination for urethritis, preferably before the first morning voiding. That means stripping of the urethra for exudate, and Gram-staining of an endourethral swab specimen for evidence of asymptomatic urethritis.
Given evidence of urethritis, their recommendation for antibiotic treatment of the patient and of his regular sexual partners is reasonable, if only because C. trachomatis is a significant genital pathogen, whatever its role in inducing arthritis in a few susceptible patients. The further hope is of course that removal of the putative trigger will lead to the subsidence of arthritis. Rheumatologists often given their patients with Reiter's syndrome a course of antibiotics routinely. In this disorder, in which recurrent attacks are frequent and disability significant, we can look forward to controlled trials of the efficacy of antibiotic therapy.--S.E.M.
Peripheral Arthritis in Ankylosing Spondylitis: Review of 209 Patients Followed Up for More Than 20 Years
Ginsburg, William W.; Cohen, Marc D.
Mayo Clin. Proc., September 1983, Vol. 58, Pg. 593-596, Ch. 60-2
Affiliation: Mayo Clinic and Found.
Abstract: Many patients with ankylosing spondylitis have involvement of the peripheral joints at some time, especially of the hips and shoulders. In some patients a seronegative polyarthritis resembling rheumatoid disease develops when the ankylosing spondylitis is quiescent. Review was made of findings in 209 patients with ankylosing spondylitis aged at least 50 years who were followed up for at least 20 years. All had low back pain and x-ray evidence of bilateral sacroiliac involvement and spinal disease. Men constituted 84% of the series. All patients had negative rheumatoid factor test results; HLA-B27 was identified in 56 of 63 patients.
Seronegative peripheral arthritis, excluding shoulder or hip involvement alone, developed in 32% of patients and lasted for at least 1 month. The relationship between onset of peripheral arthritis and onset of back pain is shown in the table. In many patients with later peripheral arthritis, the spondylitis was in remission at the time peripheral involvement developed. All but 2 patients had involvement of the knee, ankle, or metatarsophalangeal joints at some time. Nineteen patients had recurrent exacerbations of peripheral arthritis lasting for a month or longer. Evidence of erosive change or joint-space narrowing was found in 22 patients (33%).
Peripheral joint involvement in ankylosing spondylitis is common and can resemble rheumatoid arthritis clinically. Most patients have episodic involvement without residual deformity; they usually respond to treatment with nonsteroidal anti-inflammatory agents.
Post-comment: Ankylosing spondylitis is a chronic inflammatory disease affecting primarily the axial skeleton. It generally begins before age 40 with insidious low back pain and morning stiffness that improve with exercise. Symptoms persisting at least 3 months and x-ray evidence of symmetrical sacroiliitis make the diagnosis. Confusion arises when peripheral joint involvement is prominent, especially when the latter comes first (see the table). Compared with seropositive rheumatoid arthritis, the peripheral involvement associated with ankylosing spondylitis tends to be asymmetric, and on x-ray one finds less periarticular demineralization, smaller erosive changes, a propensity for bony ankylosis, and marginal periostitis. Most workers consider peripheral involvement to be part of the clinical spectrum of ankylosing spondylitis--albeit a temporally capricious part--rather than relegating it to the wastebasket category of ``seronegative rheumatoid arthritis.'' However, the following article (article 60-3) suggests that peripheral involvement represents a second HLA-B27-associated disease.--S.E.M.
Physiologic Abnormalities of Cardiac Function in Progressive Systemic Sclerosis With Diffuse Scleroderma
Follansbee, William P.; Curtiss, Edward I.; Medsger, Thomas A., Jr.; Steen, Virginia D.; Uretsky, Barry F.; Owens, Gregory R.; Rodnan, Gerald P.
N. Engl. J. Med., Jan. 19, 1984, Vol. 310, Pg. 142-148, Ch. 61-1
Affiliation: Univ. of Pittsburgh
Abstract: It is not clear whether myocardial fibrosis in progressive systemic sclerosis is a primary process involving the heart or is secondary to disease elsewhere. Rest and exercise radionuclide ventriculography and exercise scintigraphy were performed in 26 patients with progressive systemic sclerosis involving the torso as well as the upper extremities. Symptom-limited maximal treadmill exercise testing and lung function testing also were performed. Radionuclide ventriculography was done using $sup{9}$sup{9}$sup{m}Tc-labeled red blood cells; scintigraphy with $sup{2}$sup{0}$sup{1}Tl. The 18 women and 8 men in the study had a mean age of 50 years and had been ill for a mean of 41/2 years. Six patients had evidence of cardiac involvement, most often congestive failure and arrhythmias.
No patient had physical findings of severe pulmonary hypertension or ECG evidence of right ventricular hypertrophy. Thallium scans were abnormal in 20 cases. Eighteen patients had fixed defects, and 10 had reversible exercise-induced defects; 8 had both types of defect. Seven of the 10 patients with exercise-induced defects had normal coronary angiograms. Both the mean resting left ventricular EF and the mean resting right ventricular EF were lower in patients with postexercise left ventricular thallium defect scores above the median (Fig 61-1). The 6 patients with systemic hypertension did not differ from the others with regard to left ventricular EF or perfusion.
Most patients with diffuse scleroderma appear to have clinically detectable myocardial involvement, suggesting that myocardial dysfunction in this disease is secondary to ischemic injury resulting from abnormal intramyocardial circulation, an injury that ultimately leads to fibrosis. Whether this is a vasospastic or a structural process remains unknown. Overt myocardial dysfunction is associated with a poor prognosis, and earlier detection of myocardial abnormalitites might permit the evaluation of various preventive or therapeutic measures.
Post-comment: The authors used exercise and radionuclide techniques to study 26 patients with progressive systemic sclerosis with widespread (diffuse) scleroderma in order to determine whether the histopathologic lesions seen so frequently in the heart at autopsy have pathophysiologic counterparts during life. In a word, they do. Moreover, the defect appears to be in the myocardial microcirculation. Among the 10 patients with new, reversible thallium-perfusion abnormalities during exercise, all 7 who underwent subsequent cardiac catheterization had normal coronary angiograms. None of these patients had either a history of Prinzmetal's variant angina or findings of ST-segment elevation during exercise that suggested vasospasm of the large, extramural coronary arteries.
Cor pulmonale, often invoked as a cause of right ventricular dysfunction in patients with diffuse scleroderma, was not a principal cause in the patients presented here. Right ventricular ejection fraction was independent of indexes of pulmonary function or of pulmonary artery pressure when it was measured. Rather, right ventricular dysfunction appeared to be a manifestation of the same process that was occurring in the left ventricle, in that it correlated with both left ventricular dysfunction and the size of the left ventricular thallium-perfusion defects. Although the authors cannot choose between a vasospastic or a structural process producing ischemic injury and ultimately fibrosis, their findings are compatible with the attractive earlier hypothesis that fibrosis results from intermittent spasm of small coronary arteries, a form of ``myocardial Raynaud's phenomenon.'
As stressed in the accompanying editorial, the study abstracted above was selective; it was not only confined to the 20% of systemic sclerosis patients with diffuse involvement but also contained a majority of patients with clinical features (dyspnea, 62%; palpitation, 54%) strongly suggestive of cardiopulmonary dysfunction. The results need not relate to the large majority of patients with systemic sclerosis, who have more limited cutaneous involvement (sclerodactyly alone [40%] or acrosclerosis [hands, forearms, arms, and face, but not truncal skin [40%]).--S.E.M.
Factors Predicting Development of Renal Involvement in Progressive Systemic Sclerosis
Steen, Virginia D.; Medsger, Thomas A., Jr.; Osial, Thaddeus A., Jr.; Ziegler, Gayle L.; Shapiro, Alvin P.; Rodnan, Gerald P.
Am. J. Med., May 1984, Vol. 76, Pg. 779-786, Ch. 61-2
Affiliation: Univ. of Pittsburgh
Abstract: Progressive systemic sclerosis (PSS) frequently is associated with renal involvement, but little is known of the factors that precede and predispose to the development of ``scleroderma renal crisis.'' Sixty of 596 new patients with PSS seen between 1972 and 1982 had ``scleroderma renal crisis,'' defined as malignant arterial hypertension or rapidly progressive oliguric renal failure without other apparent cause. The diastolic blood pressure was more than 120 mm Hg, and grade III or IV hypertensive funduscopic changes were present. Forty-seven of the 60 study patients had diffuse scleroderma when renal crisis developed. Ten had unclassified disease, and 3 had the CREST syndrome. All but 2 patients met American Rheumatism Association criteria for definite systemic sclerosis.
Renal crisis developed within a mean of 3.2 years after the onset of illness. Patients with renal involvement had less extensive and severe skin thickening than nonrenal patients had, and less frequent peripheral vascular involvement. Six patients experienced cardiac abnormalities within 6 months before onset of renal disease. Overall, 42% of patients received steroid therapy in this period, as did 24% of those without renal involvement. Only 1 patient with renal disease had an unexplained increase in the plasma level. Changes in skin scores are compared in Figure 61-2. Many more patients without renal involvement received D-penicillamine for 6 months or longer.
Patients with PSS who have diffuse sclerodermatous skin changes that progress rapidly early in the course of illness appear to be at increased risk for the development of ``scleroderma kidney.'' Other warning signs may be the onset of anemia, pericardial effusion, or congestive heart failure. Early control of malignant hypertension markedly improves the survival of patients with ``scleroderma renal crisis.''
Post-comment: The salient finding here is that persons with diffuse sclerodermatous skin changes who have rapid progression of the extent and degree of skin thickening early in their illness seem particularly vulnerable to the development of ``scleroderma renal crisis.'' The onset of anemia, pericardial effusion, or congestive heart failure should also be considered warning signs for the onset of renal involvement. A few years ago this complication, which may appear explosively, was almost uniformly fatal. With the institution of early and aggressive antihypertensive therapy, control of hypertension and stabilization of renal function have become possible. A clinical early-warning system is welcome.--S.E.M.
Neurologic Abnormalities of Lyme Disease: Successful Treatment With High-Dose Intravenous Penicillin
Steere, Allen C.; Pachner, Andrew R.; Malawista, Stephen E.
Ann. Intern. Med., December 1983, Vol. 99, Pg. 767-772, Ch. 64-3
Affiliation: Yale Univ.
Abstract: Lyme disease is characterized by development of erythema chronicum migrans at the site of a tick bite, followed in some cases by neurologic, cardiac, or joint abnormalities. The authors have successfully used high-dose intravenous penicillin to treat patients in whom neurologic abnormalities developed despite previous oral antibiotic therapy. The results with high-dose penicillin fit with the recovery of the causative spirochete from the cerebrospinal fluid of 1 patient with neurologic involvement. Twelve patients with meningitis due to Lyme disease who were referred for neurologic problems received 3.3 million units of sodium penicillin G intravenously every 4 hours for 10 days.
Headache and stiff neck developed a mean of 5 weeks after onset of the initial illness. Four patients also had cranial neuritis, most often facial palsy; 2 had motor or sensory radiculoneuropathy; and 4 had both cranial and peripheral radiculoneuropathies. All 12 patients had abnormal cerebrospinal fluid findings. Three patients had more intense pain and low-grade fever in the early phase of treatment (a Jarisch-Herxheimer-like reaction). Neurologic pain usually resolved during treatment, and the motor deficits began to improve. No patient relapsed after treatment was stopped. Fifteen previous patients who were treated with prednisone in daily doses of 40-60 mg required treatment for a mean of 29 weeks. Meningitic symptoms resolved more rapidly in the penicillin-treated patients (table). Frank arthritis developed subsequently in none of the latter patients.
High-dose intravenous penicillin is effective for patients with neurologic abnormalities of Lyme disease. Recurrent meningitic symptoms are a helpful clue to Lyme disease, and serologic testing is often helpful. Oral tetracycline might be considered for patients allergic to penicillin. The addition of corticosteroid therapy to treatment with antibiotics does not appear to hasten the resolution of neurologic abnormalities.
Post-comment: When we found that oral tetracycline and penicillin shorten the duration of erythema chronicum migrans (ECM) and prevent or attenuate subsequent disease, the clear implication--since confirmed--was that an agent sensitive to these antibiotics is often still present in this early stage of Lyme disease. This finding ruled out a viral etiology and helped narrow the possible contenders to a list on which some kind of spirochete was most likely.
It also provided a general method for examining whether the host is still harboring a live causative agent at a given stage of disease: if penicillin cures the disease, then the agent was there. And penicillin in particular can be used in high parenteral dosage with relative impunity. This study, then, shows that Lyme meningitis is not an ``autoimmune'' disorder; it is the result of persistent infection and is curable. Similar efforts are currently being directed at Lyme arthritis.
Now one should ask: If this illness is bacterial, why was the earlier, protracted treatment with high-dose corticosteroids successful instead of perhaps disastrous? The likely answer is that although this disorder is not autoimmune, it is immune-mediated; and although the Lyme spirochete is immunostimulatory, it is otherwise apparently rather unaggressive. Formerly, we suppressed the host's response to the agent (with corticosteroids, over several months); now, we have addressed the agent directly (with antibiotics, in days). Moral: Successful treatment with corticosteroids does not rule out bacterial infection.--S.E.M.
