Recent Advances in Serodiagnosis of Pneumocystis Carinii

Communications for this section will be published as space and priotities pennit. The comments should not exceed 350 words in Isngth, with tJ mtUimum offioe references;one figure or table can be prinIetl. Exception.s tn4fI occur under particular circtnn8tances. Contnbutiona mtI!I include comments on articles published in this petiodU;4l, or they tn4fI be reports ofunique educational character. Spedfic permission to publish should be cited in a cooering letter or tJPPeRded as a postscript.

Recent Advances In serodiagnosis of Pneumocystls carlnll To the Edltor: Several points in Dr. Hughes' response to my communication

(Chat 1985; 87:698-700) concerning the serodiagnosis of PneumDCf/8fU carin" require clarification. First, the blinded study by Meyers et all referenced in both letters did not include giemsa stains oflung tissue sections or "touch preps." This is highly significant because the Seattle investigators did not look for trophozoites of P carini;, but rather stained for the cyst phase only. According to Hughes et al, II as well as others experienced in tinctorial techniques applied to P carin", it is recommended that stains capable of detecting all stages of the life cycle be employed in examining these specimens to prevent false-negative biopsy results. In the absence of these critical data, we cannot positively exclude the existence of subclinical infection in the intensely immunocompromised patients in Meyers et als report. Studies in the animal model have suggested thatsubclinical infection with P carini; may be characterized by a paucity of cyst phase organisms and a predominance of trophozoites. 3 Secondly, all of the viral and idiopathic pneumonia patients were acutely immunocompromised marrow transplant recipients who received treabnent in 1977-79 without benefit of trimethoprimsulfamethoxazole (TMP-SMX) prophylaxis. At that time, the incidence of acute pcp in that population was substantially higher than it is today. That the incidence of P carinii antigenemia was high should surprise no one. Eighty to ninety percent of normal children have IgG antibody to P carini; by age 4;· it is seen post mortem in about 3.3 percent of unselected adults," in 13.3 percent of asymptomatic pediatric cancer patients" and in nearly 70 percent of newly-diagnosed AIDS patients. Both histologic and serologic evidence clearly points to the ubiquity of P CtJrinii. There is small wonder that intensely immunocompromised marrow recipients without benefit of TMP-SMX prophylaxis exhibited P carinii antigenemia. Acute pcp has been eliminated at St. Jude Children's Research Hospital only by routine TMP-SMX prophylads." Concerning the diagnostic utility of P carinii antigen tests, our group has stated on numerous occasions that clearly not all antigenpositive patients require treatment." No laboratory test result should be taken out of context. All data should be reviewed in proper perspective with the patients history, risk factors, clinical presentation and other information. As long as P carini; can be found in healthy individuals and in an increased percentage of compromised patients, the phenomenon of antigenemia will be observed. This is not a defect in the antigen test, but rather is a reflection of the opportunistic nature ofP carin" infection. It is up to the clinician to view all findings in perspective and to use antigen data to the patient's benefit. Confusion is largely eliminated when the clinician exercises sound judgement in concert with established facts about P

carinii and all available data concerning the patient in question. An algorithm designed to aid in these decisions is available. 5 The CIE tests developed by the Centers for Disease Control and by Dr. Hughes for P carinii antigenemia bear only a slight resemblance to the one developed and improved in our laboratories. Antisera used in their tests were not prepared from cell culturegrown P carin;i organisms. At the CDC's invitation, we tested their antiserum and determined that it was only minimally reactive in the CIE test when electrophoresed in parallel with our own antiserum. 8 Preparationof highest quality antiserum is difficult, however, and this most likely explains our differing results. Discussion concerning the usefulness of P carinii antigen tests may be finally resolved to everyones satisfaction by our recent report and forthcoming manuscript describing a new quantitative latex particle agglutination test for P carinii antigen. I) The test detected antigenemia in coded specimens as early as two months before the onset of acute pcp in compromised patients, and antigen titers have been shown to correlate well with clinical progress or lack thereof. Thus, quantitative antigen titers should provide the clinical physician with much more substantial evidence of the existence or development of PCP in the at-risk patient. With support by the National Cancer Institute, our group serves as a national serologic reference laboratory for P carinUinfections in AIDS, cancer and other immunocompromised patients. We presently perform the quantitative latex particle agglutination test for P carinii antigen and an enzyme-linked immunosorbent assay for IgG antibody to P carinii at no charge for any referring physician in the US and Canada. The submission of fresh, coded sera collected prior to the initiation of therapy for PCP is invited. The submission of coded sera and the results therefrom should settle the question concerning the utility of P carinii antigen tests on the basis of hard data rather than opinion. Unda L. Pifer, Ph.D. Associate Professor ofPediatrics Pediatric Besearch. Laboratorie« University ofTennessee Center for the Health Sciences, Memphis This work was supported by the National Institutes of Health (National Cancer Institute) Grant lUOI CA 34987-01. Reprint requests: Dr. Pifer, PediatricResearch, Rm F210, 965 Court

Street, Memphis 38163

REFERENCES 1 Meyers JD, Pifer LL, Sale GE, Thomas ED. The value ofPneumocystis carinii antibody and antigen detection for diagnosis of PneumDCf/stis carinii pneumonia aft:er marrow transplantation. Am Rev Respir Dis 1979; 120:1283-87 2 Hughes WI: PneumDCf/Btis carinii pneumonia. N Eng} J Med 1976; 295: 72J)-27 3 Pifer L, Pifer D, Freeman-Shade L, Woods D, Beattie J, Hanks ~ Subclinical PneumDCf/stiI carin" infection: implications for the immunocompromised patient. 20th Interscience Conference on Antimicrobial Agents and Chemotherapy 1980; abstract no. 340 4 Pifer L, Hughes WI: Stagno S, Woods D. Pneumocystis carini; infection: evidence for high prevalence in normal and immunosuppressed children. Pediatrics 1978; 61:35-41 5 Pifer L. Pneumocystis carinii: a diagnostic dilemma. Ped Infect Dis 1983; 2:177-83 6 Perera DR, Western KA, Johnson HD, Johnson WW, Schultz MG, Agers P\t: Pneumocystiscarinii pneumonia in a hospital for Communications to the EdItor

children. Epidemiologic aspects. JAMA 1970; 214:1074-81 7 Hughes WT. Five-year absence of Pneurrwcystis carinii pneumonitis in a pediatric oncology center. J Infect Dis 1984; 150:305-06 8 Pifer L, Nie11 HB, Morrison BJ, Counce JD, Freeman JM, Woods DR, et al. Pneurrwcystis carinii antigenemia in adults with malignancy, infection, or pulmonary disease. J Clin Microbiol 1984; 20:887-90 9 Jarowenko M, Pifer L, Kerman R, Kahan BD. Serologic methods for the early diagnosis of Pneurrwcystis carinii infection in renal allograft recipients. Proc Am Soc Transplant Surg 1985.

To the Editor: In my review article on P carinii pneumonitis (Chest 1984; 85:810-13)1 stated that "further studies are required before the CIE method for antigenemia can be used as an alternative to an invasive procedure for histologic confirmation of the diagnosis". In response to Dr. Pifer's first letter indicating dissatisfaction with this statement, 1 put forth arguments and cited references in support of this view. Dr. Pifer's second letter now challenges some of the studies 1 referenced. In the paper by Meyers et aI, 1 the authors (one of whom is Dr. Pifer) conclude that, "Detection of this antigen does not appear to establish the diagnosis of P carinii pneumonia in the absence of other clinical or histologic data:' Now, Dr. Pifer argues that since touch preps and Giemsa stains for trophozoites were not done, the diagnosis of P carinii pneumonitis cannot be excluded. 1 know of no evidence in the world literature, from clinical or basic investigators or from my own experience, to show that even a single case of P carinii pneumonitis has ever been encountered in man or lower animals in which only trophozoites were found. The organism cannot replicate without the cyst stage. Cyst forms are always present with the pneumonitis. Our reason for recommending the use of polychrome stains along with the toluidine blue 0 and Gomori-Grocott stains for diagnostic biopsy specimens is not because demonstration of all forms in the life cycle is important for diagnosis; rather, overstaining sometimes occurs with the cyst wall stains, and nonbudding yeasts may resemble P carinii cysts. Polychrome stains, such as that of Giemsa, may be helpful in resolving the problem. While subclinical infection could conceivably exist with trophozoites alone (no proof of this) or with a paucity of cysts (very likely the case), I must scream out to emphasize that we are talking about the diagnosis of pneumonitis-not asymptomatic, subclinical and latent infection. The point I want most earnestly to make is that this test does not precisely differentiate between pneumonitis and subclinical infection. There is no disagreement with Dr. Pifer's comments about the prevalence of subclinical infection in which no pneumonitis is evident. Dr. Pifer states that confusion about the antigen test is largely eliminated when the clinician exercises sound judgement. From the laboratory, she has provided an algorithm to aid clinicians in these decisions.! In her algorithm for the compromised host with diffuse pneumonitis who has sterile cultures for bacteria, fungi, viruses and chlamydia and in whom the antigen test is positive, the diagnosis of P carinii pneumonitis is "probable:' At this point, "if consistent with clinical presentations," the patient is treated for P carinii infection. However, the algorithm's alternate pathway shows that an invasive diagnostic procedure is indicated, "if justifiable:' At the bedside of a critically ill patient with pneumonia, waiting for conclusions on the cultures mentioned and unable to define terms like "clinical presentation" and "if justifiable," one has difficulty in exercising sound judgement. Hopefully, the new test to which Dr. Pifer refers will satisfy our needs. However, it is worrisome to learn that the new test detects antigen as early as two months before the onset of pneumonitis. This brings us back to restatement of our problem. The physician

confronted with an acutely ill immunosuppressed patient with diffuse pneumonitis needs a non-invasive test to determine whether or not the cause is P carinii. If the test also detects subclinical infection, its diagnostic validity for pneumonitis is inversely proportional to the prevalence of the subclinical infection. Dr. Pifer has commented on the prevalence of subclinical infection. Since my first response to Dr. Pifer, another report on the use of this antigen test has been published. Tanabe et al' studied 13 patients with P carinii pneumonia and found only three (23 percent) to be antigen-positive. Since the publication of my paper, I have seen no data to convince me to change the statement. Perhaps other readers have opposing views to mine. Since the test is being used fairly extensively according to Dr. Pifer, and its use is sponsored by the National Institutes of Health, it would be of interest to learn of the experience of others.

Walter T. Hughes, M.D. Chairman, Department of Child Health Sciences and Director, Division of Infectious Diseases St. Jude Children's Research Hospital Memphis

REFERENCES 1 Meyers JD, Pifer LL, Sale GE, Thomas ED. The value of Pneumocystis carinii antibody and antigen detection for diagnosis of Pneumocystis carinii pneumonia after marrow transplantation. Am Rev Respir Dis 1979; 120:1283-87 2 Pifer L. Pneumocystis carinii: a diagnostic dilemma. Ped Infect Dis 1983; 2:177-83 3 Tanabe K, Furuta T, Ueda K, Tanaka H, Shimoda K. Serological observation of Pneumocystis carinii infection in humans. J Clin Microbiol1985; 22:1058-60

The Ethics of Technology To the Editor: I just read Dr. Weinberg's editorial in the February, 1985 issue of Chest (1985; 87:141.) It was superb! The critical care nurses of the United States strongly support the realistic application of intensive care. Those of us practicing critical care daily need to apply liberal doses of humanity rather than believing that, because we have the technology, it must be used. Many years ago, in a conversation I had with Hans Selye bemoaning the problems of invasive technologic techniques used on patients who had no chance of survival, he told me something that I would like to pass along. He said that, because we have this technology that can, in some cases, save lives and in others, prolong dying, we have a greater responsibility to determine when that technology will be used. This conversation took place in the mid-I970s, when the country was so concerned with the "do not resuscitate" issue and withdrawal oflife support systems. He told me that we should change our focus so that we would only rarely be in the position of having to make decisions about withdrawal of treatment and life support systems. In other words, put our attention at the other end and make decisions early about when technology should be applied, when it is appropriate. Because of my concerns since the late-1960s, I have been studying ethics very intensely. It seems to me that our focus now needs to be on the patients we care for and we need to discuss with them what they would like done if certain circumstances occur. To wait until they are unable to make the decision because of a coma or severity of illness, and then to try to make the decision with families who are already over-stressed, somehow does not seem appropriate. Some of the physicians at the Hospital of the University of Pennsylvania are discussing this issue with their patients. This helps CHEST / 89 / 5 / MAY, 1986

765