Differential Lobe Lavage for Diagnosis of Acute Pneumocystis carinii Pneumonia in Patients Receiving Prophylactic Aerosolized Pentamidine Therapy

Abstract

The diagnostic yield of bronchoalveolar lavage (BAL) for Pneumocystis carinii pneumonia (PCP) in patients infected with human immunodeficiency virus has been reported to be 95 percent, but falls to 62 percent in patients receiving aerosolized pentamidine. Because aerosolized pentamidine appears to be preferentially deposited in the middle and lower lobes, we postulated that an upper lobe lavage would have a higher diagnostic yield than the standard middle/lower lobe lavage in patients receiving aerosolized pentamidine. Twenty-five patients receiving aerosolized pentamidine suspected of having acute PCP underwent separate BAL of an upper lobe and lower lobe as well as transbronchial biopsy. Fifteen of the 25 (60 percent) were diagnosed as having PCP. Of the 15, one had the samples inadvertently combined. In the remaining 14, BAL was positive for P carinii organisms in 12 lavages of the lower lobe and 14 of the upper lobe. Upper lobe lavage had statistically significantly more P carinii organisms by semiquantitative technique than the lower lobe. In patients receiving aerosolized pentamidine, who develop acute PCP, an upper lobe lavage may have a higher diagnostic yield than the standard middle/lower lobe lavage. In addition, the transbronchial biopsy specimen offered no treatable diagnosis that was not made by lavage alone in the 25 patients. This raises the question of the utility of transbronchial biopsies in these patients. The diagnostic yield of bronchoalveolar lavage (BAL) for Pneumocystis carinii pneumonia (PCP) in patients infected with human immunodeficiency virus has been reported to be 95 percent, but falls to 62 percent in patients receiving aerosolized pentamidine. Because aerosolized pentamidine appears to be preferentially deposited in the middle and lower lobes, we postulated that an upper lobe lavage would have a higher diagnostic yield than the standard middle/lower lobe lavage in patients receiving aerosolized pentamidine. Twenty-five patients receiving aerosolized pentamidine suspected of having acute PCP underwent separate BAL of an upper lobe and lower lobe as well as transbronchial biopsy. Fifteen of the 25 (60 percent) were diagnosed as having PCP. Of the 15, one had the samples inadvertently combined. In the remaining 14, BAL was positive for P carinii organisms in 12 lavages of the lower lobe and 14 of the upper lobe. Upper lobe lavage had statistically significantly more P carinii organisms by semiquantitative technique than the lower lobe. In patients receiving aerosolized pentamidine, who develop acute PCP, an upper lobe lavage may have a higher diagnostic yield than the standard middle/lower lobe lavage. In addition, the transbronchial biopsy specimen offered no treatable diagnosis that was not made by lavage alone in the 25 patients. This raises the question of the utility of transbronchial biopsies in these patients. Mycobacterium avium intracellulare Pneumocystis carinii pneumonia Pneumocystis carinii pneumonia (PCP) is the most common cause of pneumonia in patients infected with the human immunodeficiency virus (HIV).1Hopewell PC Luce JM Pulmonary involvement in the acquired immunodeficiency syndrome.Chest. 1985; 87: 104-112Crossref PubMed Scopus (88) Google Scholar, 2Murray JF Felton CP Garay SM Gottlieb MS Hopewell PC Stover DE et al.Pulmonary complications in the acquired immunodeficiency syndrome.N Engl J Med. 1984; 310: 1682-1688Crossref PubMed Scopus (482) Google Scholar At times, however, PCP may be difficult to diagnose and may mimic other pathologic processes. Bronchoscopy has been shown to be a very sensitive technique in diagnosing PCP with diagnostic yield for lavage alone ranging from 94 to 97 percent.3Coleman DL Dodek PM Luce JM Golden JA Gold WM Murray JF Diagnostic utility of fiberoptic bronchoscopy in patients with Pneumocystis carinii pneumonia and the acquired immune deficiency syndrome.Am Rev Respir Dis. 1983; 128: 795-799PubMed Google Scholar, 4Broaddus C Dake MD Stulbarg MS et al.Bronchoalveolar lavage and transbronchial biopsy for the diagnosis of pulmonary infections in the acquired immunodeficiency syndrome.Ann Intern Med. 1985; 102: 747-752Crossref PubMed Scopus (261) Google Scholar, 5Golden JA Hollander H Stulbarg MS Gamsu G Bronchoalveolar lavage as the exclusive diagnostic modality for Pneumocystis carinii pneumonia.Chest. 1986; 90: 18-22Abstract Full Text Full Text PDF PubMed Scopus (122) Google Scholar, 6Stover DE White DA Romano PA Gellene RA Diagnosis of pulmonary disease in acquired immunodeficiency syndrome.Am Rev Respir Dis. 1984; 130: 659-662PubMed Google Scholar Since the recognition of PCP as a major complication of HIV disease, several prophylactic measures have been investigated. Currently the two most common means of prophylaxis against PCP are trimethoprimsulfamethoxazole and aerosolized pentamidine.7Centers for Disease Control. Guidelines for prophylaxis against Pneumocystis carinii pneumonia for persons infected with human immunodeficiency virus. JAMA 262:1989, 335-39Google Scholar With initiation of prophylactic measures, there has been concern over the ability to diagnose acute PCP in patients receiving aerosolized pentamidine with bronchoalveolar lavage (BAL) alone. Jules-Elysee et al8Jules-Elysee KM Stover DE Zaman MB Bernard EM White DA Aerosolized pentamidine: effect on diagnosis and presentation of Pneumocystis carinii pneumonia.Ann Intern Med. 1990; 112: 750-757Crossref PubMed Scopus (190) Google Scholar reported a decreased yield of P carinii organisms by BAL in patients receiving prophylactic aerosolized pentamidine. The reason for the decreased yield may well be related to the atypical characteristics of PCP in this population. Recurrent PCP in patients receiving aerosolized pentamidine prophylaxis appears to have a predilection for isolated upper lobe disease demonstrated both radiographically and by gallium scanning.9Bradburne RM Ettensohn DB Opal SM McCool FD Relapse of Pneumocystis carinii pneumonia in the upper lobes during aerosolized pentamidine prophylaxis.Thorax. 1989; 44: 591-593Crossref PubMed Scopus (32) Google Scholar, 10Abd AG Nierman DM Ilowite JS Pierson RN Bell AL Bilateral upper lobe Pneumocystis carinii pneumonia in a patient receiving inhaled pentamidine prophylaxis.Chest. 1988; 94: 329-331Abstract Full Text Full Text PDF PubMed Scopus (55) Google Scholar Therefore, in assessing PCP in patients receiving aerosolized pentamidine, performing a lavage in the lower and middle lobes may have a decreased yield as these lobes may have a lower P carinii burden when compared with the upper lobes. Jules-Elysee et al8Jules-Elysee KM Stover DE Zaman MB Bernard EM White DA Aerosolized pentamidine: effect on diagnosis and presentation of Pneumocystis carinii pneumonia.Ann Intern Med. 1990; 112: 750-757Crossref PubMed Scopus (190) Google Scholar preferentially lavaged the middle lobe or lingula regardless of the radiographic presentation. The purposes of this study are first to compare the diagnostic yield of (1) lavage of the upper lobe, (2) lavage of the middle/lower lobe, and (3) transbronchial biopsy specimens in the diagnosis of acute PCP in patients receiving aerosolized pentamidine prophylaxis. Secondly, through the use of semiquantitative techniques,11Baughman RP Strohofer S Colangelo G Frame PT Semiquantitative techniques for estimating Pneumocystis carinii burden in the lung.J Clin Microbiol. 1990; 28: 1425-1427Google Scholar we will determine if the number of P carinii organisms recovered is greater in the upper lobe than the middle/lower lobe. Patients were considered for enrollment if they were (1) 18 years of age or older, (2) HIV positive, (3) receiving monthly aerosolized pentamidine for either primary or secondary prophylaxis, (4) compliant for at least the previous two months, and (5) had an abnormal chest radiograph and at least one sign or symptoms suggestive of PCP, including fever, cough, dyspnea, and/or elevated lactate dehydrogenase (LDH) level. Patients were excluded if they had a history of asthma or any condition (ie, coagulopathy) that would preclude transbronchial biopsy. All patients enrolled were receiving their aerosolized pentamidine in the semirecumbent position. Patients underwent bronchoscopy in the standard fashion with the patient in the semirecumbent position. After inspecting all bronchial segments, the lung that was radiographically most involved was used for the study. In cases in which both lungs had equal involvement, the right lung was used. To prevent contamination from the upper lobe lavage, the lower lobe lavage was performed first. After the bronchoscope was wedged into a segmental bronchus, 100 ml of sterile saline solution in 20-ml aliquots was instilled and collected. Once the specimen was collected, the canister and suction tubing were changed and the same procedure was repeated in the upper lobe. Finally, four transbronchial biopsy specimens were obtained from the segment that appeared to be most involved radiographically. A small portion of the original lavage was submitted for bacteriology and virology studies. The remainder was submitted separately to cytology along with the upper lobe lavage. Each specimen was reviewed in random order by the cytopathologist who was blinded as to the location of the lavage. Each specimen was spun at 1,000 rpm for 7 min. The supernate was then placed onto glass slides and modified Wright-Giemsa stain (Diff-Quik) as well as Papanicolaou and Gomori methenamine silver staining were done. The modified Wright-Giemsa-stained slides were reviewed for the presence of trophozoites and were read as negative if none was seen after a thorough review averaging 30 min. In those patients who were found to have PCP by either of the lavages or by the transbronchial biopsy specimen, both the upper and the lower lobe modified Wright-Giemsa-stained smears were again reviewed by the cytopathologist who counted the number of clusters of P carinii organisms seen per 500 nucleated cells in the lavage specimen. This technique was previously reported by Baughman et al.11Baughman RP Strohofer S Colangelo G Frame PT Semiquantitative techniques for estimating Pneumocystis carinii burden in the lung.J Clin Microbiol. 1990; 28: 1425-1427Google Scholar Between November 1990 and May 1992, 25 patients were enrolled into the study. They were all male homosexuals between the ages of 25 and 48 years. Of the 25 patients, 15 were diagnosed as having acute PCP (Table 1). Eight of these patients had no prior episodes of PCP and therefore were receiving primary prophylaxis while the other seven each had one prior episode and were receiving secondary prophylaxis. Seven of the 15 patients (47 percent) had radiographic evidence of predominantly upper lobe disease while the remaining 8 patients (53 percent) had the more classic diffuse pattern. All patients tolerated the bilobar lavage. The mean lavage return for all patients for the 100-ml upper lobe lavage was 49 ml (range, 25 to 75 ml), while the mean lavage return for the 100-ml lower lavage was 53 ml (range, 40 to 80 ml). All but two patients had transbronchial biopsies. One patient had significant coughing and one had a brief decrease in oxygen saturation during the procedure that precluded performing transbronchial biopsies. Of the 23 patients who underwent transbronchial biopsies, three patients had inadequate tissue obtained for analysis. Of the 20 patients with adequate tissue, 10 were eventually diagnosed as having acute PCP and the transbronchial biopsy specimen was positive in all 10 (100 percent). In one patient, the lavage specimens were inadvertently mixed before separate analysis, but in the remaining 14 patients, all of the upper lobe lavages were positive for P carinii organisms by modified Wright-Giemsa (100 percent), whereas only 12 lower lobe lavages were positive (86 percent) (NS).Table 1Prior Episodes of PCP, Radiographic Manifestation, and Diagnostic Yield of Bilobar Lavage and Transbronchial Biopsy in Patients With Proven PCP (n = 15)*Bil = bilateral; Dz = disease; NA = not applicable; PTX = pneumothorax; TBBX = transbronchial biopsy; PCP = Pneumocystis carinii pneumonia.PatientPrior PCPUpper LavageLower LavageTBBXChest Radiograph11+++Bil apical Dz20++Not doneBil upper zone Dz31+-No tissueBil upper zone Dz40+++Bil diffuse Dz5†Samples inadvertently mixed.0NA +NA+Bil diffuse Dz61++Not doneBil apical Dz/PTX70+–+Bil diffuse Dz80+++Bil upper zone Dz91++No tissueBil diffuse Dz101++No tissueBil diffuse Dz110+++Bil apical Dz121+++Bil apical Dz130+++Bil diffuse Dz140+++Bil diffuse Dz151+14/14 (100%)+12/14 (86%)+10/10 (100%)Bil diffuse Dz* Bil = bilateral; Dz = disease; NA = not applicable; PTX = pneumothorax; TBBX = transbronchial biopsy; PCP = Pneumocystis carinii pneumonia.† Samples inadvertently mixed. Open table in a new tab In the 14 specimens in which semiquantitative technique was performed, there was a higher number of P carinii clusters in the upper lobe in all but two patients (Table 2). For all 14 patients combined, there was a mean of 23.2 ±26 clusters per 500 nucleated cells in the upper lobe lavage compared with a mean of 7.0 ±13 clusters in the lower lavage that was significantly different (p<0.05).Table 2Quantitatice Pneumocystis Load of Bilobar Lavage in Patients With Proven PCP (n=11)*N/A=not applicable; PCP=Pneumocystis carinii pneumonia.No. ml Returned/100 ml LavagePatientUpper Lavage (No./500 Cells)Lower Lavage (No./500 Cells)UpperLower18025752598455534050604604755455N/A50506263505572175808247570922605510342645551152505512721506013304540141014545152934545Mean23.2 ±267.0±13* N/A=not applicable; PCP=Pneumocystis carinii pneumonia. Open table in a new tab In the 10 patients with negative studies for PCP, there was no clinical evidence for the development of PCP in the follow-up period that ranged from 6 to 18 months. An alternative diagnosis was made in nine of the ten patients (Table 3). Three patients had a nonspecific inflammation and two had chronic fibrosis on the transbronchial biopsy specimens. Two patients had the diagnosis of disseminated Mycobacterium avium intracellulare (MAI) established on the lavage cultures and with concomitant blood cultures. Another patient had the diagnosis of MAI established from cultures of a mediastinal lymph node obtained by mediastinoscopy after having a negative lavage and transbronchial biopsy specimen. Finally, one patient grew Streptococcus pneumoniae from his blood cultures as well as his lavage and responded to antibiotic therapy.Table 3Alternative Diagnoses in Patients Without PCP (n=10)*PCP=Pneumocystis carinii pneumonia.DiagnosisNo.Chronic inflammation3Fibrosis2Mycobacterium avium intracellulare3Streptococcus pneumonia1Normal1* PCP=Pneumocystis carinii pneumonia. Open table in a new tab Pneumocystis carinii pneumonia is the leading cause of death from opportunistic infection in patients with HIV disease.12Devita VT Broder S Fauci AS Kovacs JA Chabner BA Developmental therapeutics and the acquired immunodeficiency syndrome.Ann Intern Med. 1987; 106: 568-581Crossref PubMed Scopus (69) Google Scholar Aerosolized pentamidine has been demonstrated to be an effective means of prophylaxis against PCP, reducing the relapse rate from 40 percent to a range of 5 to 12 percent.13Leoung GS Feigal DW Montgomery AB et al.Aerosolized pentamidine prophylaxis against Pneumocystis carinii pneumonia.N Engl J Med. 1990; 323: 770-775Crossref Scopus (271) Google Scholar, 14Hirschel B Lazzarin A Chopard M et al.A controlled study of inhaled pentamidine for primary prevention of Pneumocystis carinii pneumonia.N Engl J Med. 1991; 324: 1079-1083Crossref PubMed Scopus (165) Google Scholar Although aerosolized pentamidine is effective as prophylaxis, breakthrough cases of acute PCP are seen. Recent concern has been raised over the ability to diagnose acute PCP in patients receiving aerosolized pentamidine with BAL alone. Jules-Elysee et al8Jules-Elysee KM Stover DE Zaman MB Bernard EM White DA Aerosolized pentamidine: effect on diagnosis and presentation of Pneumocystis carinii pneumonia.Ann Intern Med. 1990; 112: 750-757Crossref PubMed Scopus (190) Google Scholar reported only a 62 percent yield of the standard middle lobe lavage in patients with acute PCP who were receiving aerosolized pentamidine compared with a 100 percent yield of lavage for P carinii organisms in patients not receiving aerosolized pentamidine. The yield of transbronchial biopsy specimens was similar in both groups (81 percent and 84 percent, respectively). These authors therefore advocated performing transbronchial biopsies in patients receiving aerosolized pentamidine in order to diagnose PCP. One reason for the decreased yield of the BAL in patients receiving aerosolized pentamidine may be the higher incidence of atypical presentation of acute PCP in these patients. Thirty-eight percent of the patients of Jules-Elysee et al8Jules-Elysee KM Stover DE Zaman MB Bernard EM White DA Aerosolized pentamidine: effect on diagnosis and presentation of Pneumocystis carinii pneumonia.Ann Intern Med. 1990; 112: 750-757Crossref PubMed Scopus (190) Google Scholar who were receiving aerosolized pentamidine had disease restricted radiographically to the upper lobes. The lavage, however, was performed in either the right middle lobe or the lingula. By performing both upper as well as lower lobe lavages, the diagnostic yield of lavage might be increased. In our study group, 47 percent had predominantly upper lobe disease. We found the upper lobe lavage to be diagnostic 100 percent of the time. The lower zone lavage was diagnostic in 86 percent of the patients, which is higher than that reported by Jules-Elysee et al.8Jules-Elysee KM Stover DE Zaman MB Bernard EM White DA Aerosolized pentamidine: effect on diagnosis and presentation of Pneumocystis carinii pneumonia.Ann Intern Med. 1990; 112: 750-757Crossref PubMed Scopus (190) Google Scholar This may be related to the smaller number of patients in our study. Also of interest was the demonstration of higher numbers of Pneumocystis clusters in 12 of the 14 upper lobes as compared with the lower lung zones. The predilection of upper lobe involvement may well be related to the distribution of the aerosolized pentamidine.15O'Doherty MJ Thomas SH Page CJ Bradbeer C Nunan TO Bateman NT Does inhalation of pentamidine in the supine position increase deposition in the upper part of the lung?.Chest. 1990; 97: 1343-1348Abstract Full Text Full Text PDF PubMed Scopus (20) Google Scholar Ventilation to the lower lobes is greater than to the upper lobes in the erect position. Deposition of aerosol within the lung is influenced not only by ventilation but also by the physical characteristics of the aerosol. Radionuclide studies demonstrate greater distribution of aerosolized pentamidine to the lower lung zones particularly in the sitting position.15O'Doherty MJ Thomas SH Page CJ Bradbeer C Nunan TO Bateman NT Does inhalation of pentamidine in the supine position increase deposition in the upper part of the lung?.Chest. 1990; 97: 1343-1348Abstract Full Text Full Text PDF PubMed Scopus (20) Google Scholar, 16Baskin MI Abd AG Ilowite JS Regional deposition of aerosolized pentamidine: effects of body position and breathing pattern.Ann Intern Med. 1990; 113: 677-683Crossref PubMed Scopus (46) Google Scholar Therefore, it would seem that the upper zones are the least protected by this mode of prophylactic therapy. Our study design did not include patients who were not receiving aerosolized pentamidine. We are therefore unable to comment whether the observed increased burden in the upper lobes is distinct in patients with acute PCP receiving aerosolized pentamidine. Another explanation may be that the organism itself is more prone to infect the upper lobes much in the same way that Mycobacterium tuberculosis does, and that this is not related to aerosolized pentamidine. The diffuse infiltrates on radiograph seen in up to 90 percent of cases of acute PCP, however, would suggest that the organism does not have an inherent upper lobe predominance. Another point of interest in our study was those diagnoses made in the patients with respiratory symptoms other than PCP who were receiving aerosolized pentamidine. In this group of ten patients, only four had treatable infectious diseases. Two had MAI diagnosed by lavage and blood cultures. One had pneumococcal pneumonia diagnosed by lavage and blood cultures. One had MAI infection diagnosed by mediastinal lymph node biopsy specimens. In none of the patients was a treatable diagnosis made by transbronchial biopsy that was not made by lavage. In fact, in all 23 patients in the study, the transbronchial biopsy specimen did not add to the diagnostic yield of bilobar lavage for treatable diseases. Although the number of subjects studied is small, the data suggest that transbronchial biopsy as an initial diagnostic procedure in HIV-infected patients receiving aerosolized pentamidine who are referred with respiratory complaints may not be worthwhile, particularly in centers with cytopathologists who are experienced in reading modified Wright-Giemsa-stained slides. By excluding the biopsy from the initial bronchoscopic examination, the patient would no longer be exposed to the increased risk of pneumothorax or hemorrhage. In summary, in our patients who developed acute PCP while receiving aerosolized pentamidine, 47 percent had predominantly upper lobe involvement radiographically. The diagnostic yield of BAL alone was increased from 86 percent to 100 percent with the addition of an upper lobe lavage. The infection appears to be more severe in the upper lobes as there is a higher number of Pneumocystis clusters seen in the upper lobe lavage specimen. In the study group, all treatable infectious diagnoses were made by lavage alone except for one case of MAI diagnosed from a mediastinal lymph node biopsy specimen. The transbronchial biopsy only exposed the patients to increased risk of complication without adding to the diagnostic yield. The authors would like to thank Kelly Bauer for expert technical assistance in the preparation of the smears and Sherry Magruder for her professional assistance with the bronchoscopies.