Training bronchoscopists for the new era.

Gustav Killian introduced bronchoscopy a little more than a century ago. At that time, the only way others could learn to perform bronchoscopy was by one-on-one tutoring, using a rigid bronchoscope with no side portals and no imaging devices such as a television camera and monitor. One-on-one teaching remains an integral part of learning how to perform bronchoscopy well, but many new technologies have emerged that make it far less labor intensive to train bronchoscopists. This article focuses on the training of bronchoscopists for the new era.

Permanent declines in pulmonary function following pneumonia in human immunodeficiency virus-infected persons. The Pulmonary Complications of HIV Infection Study Group.

Human immunodeficiency virus (HIV)-associated respiratory infections, most notably Pneumocystis carinii pneumonia (PCP), but also bacterial pneumonia (BP), result in reductions in lung function that have been studied mainly during the course of acute infection. Whether HIV-associated pneumonias also cause permanent changes in pulmonary function is unknown. In this study we investigated the long-term effects of PCP and BP on pulmonary function in a cohort of HIV-infected persons. One thousand, one hundred forty-nine HIV-infected persons were followed in a prospective, observational cohort study at six centers in the United States. Study participants had pulmonary function testing performed at regular preset intervals. PCP and BP diagnoses were verified with defined criteria. Longitudinal multivariate analysis was used to model pulmonary function in terms of demographic data and occurrence of PCP or BP. We found that PCP or BP was associated with permanent decreases in FEV(1), FVC, FEV(1)/FVC, and the diffusing capacity of carbon monoxide. Neither infection resulted in statistically significant changes in TLC. We conclude that PCP and BP result in expiratory airflow reductions that persist after the acute infection resolves. The clinical implications of these changes are unknown, but they may contribute to prolonged respiratory complaints in HIV-infected patients who have had pneumonia.

Impact of bacterial pneumonia and Pneumocystis carinii pneumonia on human immunodeficiency virus disease progression. Pulmonary Complications of HIV Study Group.

The course of pneumonia caused by pyogenic bacteria and Pneumocystis carinii was examined in a multicity cohort study of HIV infection. The median duration of survival among 150 individuals following initial bacterial pneumonia was 24 months, compared with 37 months among 299 human immunodeficiency virus (HIV)-infected control subjects matched by study site and CD4 lymphocyte count (P<.001). For 152 subjects with P. carinii pneumonia, median survival was 23 months, compared with 30 months for 280 matched control subjects (P = .002). Median durations of survival associated with the two types of pneumonia differed by only 47 days, despite a higher median CD4 lymphocyte count associated with bacterial pneumonia. These results suggest that both P. carinii pneumonia and bacterial pneumonia are associated with a significantly worse subsequent HIV disease course. The similarity of prognosis after one episode of bacterial pneumonia vs. an AIDS-defining opportunistic infection and the proportion of cases occurring in association with a CD4 lymphocyte count of >200 suggest that measures to prevent bacterial pneumonia should be emphasized.

Risk factors and outcomes associated with identification of Aspergillus in respiratory specimens from persons with HIV disease. Pulmonary Complications of HIV Infection Study Group.

STUDY OBJECTIVES: To examine the significance of previously suggested risk factors and assess outcomes associated with Aspergillus identification in respiratory specimens from HIV-seropositive individuals. DESIGN: This was a nested case-control study. Patients who had Aspergillus species identified in respiratory specimens were matched at the time of study entry 1:2 with control subjects according to study center, age, gender, race, HIV transmission category, and CD4 count. SETTING: The multicenter Pulmonary Complications of HIV Infection Study. PARTICIPANTS: HIV-seropositive study participants. MEASUREMENTS AND RESULTS: Between November 1988 and March 1994, Aspergillus species were detected in respiratory specimens from 19 (1.6%) participants. The rate of Aspergillus identification among participants with CD4 counts <200 cells per cubic millimeter during years 2 through 5 after study entry ranged from 1.2 to 1.9%. Neutropenia, a CD4 count <30 cells per cubic millimeter, corticosteroid use, and Pneumocystis carinii infection were associated with subsequent identification of Aspergillus in respiratory specimens. Cigarette and marijuana use, previously suggested risk factors, were not associated with Aspergillus respiratory infection. A substantially greater proportion of patients with Aspergillus compared with control subjects died during the study (90% vs 21%). Excluding four cases first diagnosed at autopsy, 67% died within 60 days after Aspergillus was detected. CONCLUSIONS: Although Aspergillus is infrequently isolated from HIV-infected persons, the associated high mortality would support serious consideration of its clinical significance in those with advanced disease and risk factors.

Predictors of Pneumocystis carinii pneumonia in HIV-infected persons. Pulmonary Complications of HIV Infection Study Group.

The Pulmonary Complications of HIV Infection Study is a prospective, multicenter, observational study evaluating pulmonary disease among HIV-infected persons. For approximately 52 mo, 1,182 HIV-infected subjects were followed. All participants were evaluated for pulmonary disease on a predetermined schedule. There were 145 episodes of Pneumocystis carinii pneumonia (PCP). Low CD4 count correlated with risk of PCP (p < 0.0001); 79% had CD4 counts less than 100/microl and 95% had CD4 counts less than 200/microl. Subtle changes in diffusing capacity for carbon monoxide (DLCO) were associated with PCP. Univariate analysis identified recurrent undiagnosed fevers, night sweats, oropharyngeal thrush, and unintentional weight loss to be associated with risk among persons with CD4 counts above 200/microl. Subjects in whom CD4 counts declined to below 200/microl and who were not receiving preventive therapy were nine times more likely to develop PCP within 6 mo compared with subjects who received such therapy. A strong trend toward differences between the sexes was detected. Black subjects had less than one third the risk of developing PCP as did white subjects (p < 0.0001). There was no significant difference in risk by HIV transmission category, study site, frequency of follow-up, age, education, smoking history, or use of antiretroviral therapy. Multivariable analysis revealed low CD4 lymphocyte count (p < 0.0001), use of prophylaxis (p < 0.0001), racial differences (p < 0.0001), and declining DLCO (p = 0.015) to influence risk. Constitutional signs and symptoms indicate increased risk for PCP among HIV-infected persons with CD4 counts above 200/microl.

Intensive care of patients with HIV infection: utilization, critical illnesses, and outcomes. Pulmonary Complications of HIV Infection Study Group.

To examine intensive care unit (ICU) admission rates and diagnoses of patients with HIV infection, and to determine the outcomes of different critical illnesses, we analyzed data derived from the 63 patients who were admitted to an ICU from among the 1,130 adults with HIV infection who did not have AIDS at the time of enrollment in a multicenter prospective study. Patients were admitted and treated according to the judgment of their physicians. During 4,298 patient-years of follow-up for the entire cohort, there were 1,320 hospital admissions, of which 68 (5%) included admission to an ICU. Twenty-five (40%) of the patients admitted to the ICU died during that admission. Twenty-four patients (38%) were admitted with a principal diagnosis of lung disease; 11 had Pneumocystis carinii pneumonia (PCP), one of whom was coinfected with Aspergillus fumigatus and Legionella pneumophilia, and six of them (55%) died. Four had bacterial pneumonia, two had pulmonary edema caused by renal failure, and one each had pulmonary tuberculosis, pulmonary Kaposi's sarcoma, pneumothorax, adult respiratory distress syndrome, severe pulmonary fibrosis, cytomegalovirus pneumonitis, and metastatic adenocarcinoma to the lungs. Eleven of these 14 patients (79%) died. Thirty-nine patients had 44 admissions for nonpulmonary diagnoses, including gastrointestinal disorders (14 admissions), cardiovascular disorders (nine), sepsis syndrome (six), neurologic disorders (four), monitoring and ICU nursing care during or after a procedure (four), metabolic disorders (three), trauma (two), drug overdose (one), and unknown reasons (one). Nine (23%) of these patients died. Twenty-eight patients underwent mechanical ventilation, and 16 (57%) died. Seven (25%) had PCP (five died), seven had other primary pulmonary diseases (six died), and 14 were placed on mechanical ventilation for nonpulmonary disorders (five died). Survival did not correlate with CD4 count determined within 6 mo of admission to the ICU. In conclusion, the range of indications for critical care in patients with HIV infection is diverse. PCP accounted for only 16% of the ICU admissions, and mechanical ventilation for PCP and other pulmonary disorders was associated with a high mortality rate. In contrast, mechanical ventilation for nonpulmonary disorders, and admission to the ICU for nonpulmonary diagnoses was associated with a more favorable outcome.

Respiratory disease trends in the Pulmonary Complications of HIV Infection Study cohort. Pulmonary Complications of HIV Infection Study Group.

We examined trends in the incidence of specific respiratory disorders in a multicenter cohort with progressive human immunodeficiency virus (HIV) disease during a 5-yr period. Individuals with a wide range of HIV disease severity belonging to three transmission categories were evaluated at regular intervals and for episodic respiratory symptoms using standard diagnostic algorithms. Yearly incidence rates of respiratory diagnoses were assessed in the cohort as a whole and according to CD4 count or HIV transmission category. The most frequent respiratory disorders were upper respiratory tract infections, but the incidence of lower respiratory tract infections increased as CD4 counts declined. Specific lower respiratory infections followed distinctive patterns according to study-entry CD4 count and transmission category. Acute bronchitis was the predominant lower respiratory infection of cohort members with entry CD4 counts > or = 200 cells/mm3. In cohort members with entry CD4 counts of 200 to 499 cells/mm3, the incidence of bacterial and Pneumocystis carinii pneumonia each increased an average of 40% per year. In members with entry CD4 counts < 200 cells/mm3, acute bronchitis, bacterial pneumonia, and P. carinii pneumonia occurred at high rates without discernible time trends, despite chemoprophylaxis in more than 80% after Year 1, and the rate of other pulmonary opportunistic infections increased over time. Each year, injecting drug users had a higher incidence of bacterial pneumonia than did homosexual men. The yearly rate of tuberculosis was < 3 episodes/100 person-yr in each entry CD4 and HIV-transmission group. We conclude that the time trends of HIV-associated respiratory disorders are determined by HIV disease stage and influenced by transmission category. Whereas acute bronchitis is prevalent during all stages of HIV infection, incidence rates of bacterial pneumonia and P. carinii pneumonia rise continuously during progression to advanced disease. In advanced disease, the incidence of acute bronchitis, bacterial pneumonia and P. carinii pneumonia is high despite widespread chemoprophylaxis.

Nonspecific airway hyperresponsiveness in HIV disease. Pulmonary Complications of HIV Infection Study Group.

OBJECTIVES: HIV disease is frequently complicated by episodic acute bronchitis, suggesting the presence of chronic bronchial inflammation. To further examine this concept, we investigated the possible association of nonspecific airway hyperresponsiveness (AHR) and HIV disease. DESIGN: Methacholine inhalation challenge studies were performed on 66 HIV-seropositive and 8 HIV-seronegative members of the Pulmonary Complications of HIV Infection Study Cohort. AHR was defined as 20% or more decline in FEV1 from the postdiluent value after inhalation of 125 or less cumulative breath units. The prevalence of AHR in HIV-seropositive cohort members was compared with that in matched control subjects who had undergone methacholine challenge testing for two unrelated studies. Demographic, behavioral, and clinical features in HIV cohort members with and without AHR were contrasted. The relationship between AHR and the occurrence of episodic airway disease or symptoms suggestive of airway disease was examined. RESULTS: AHR was not more prevalent in HIV-seropositive cohort members than control subjects (19.3% vs 12.9%; p > 0.1). Within the cohort, AHR was detected more frequently in members with than without a history of asthma (60% vs 16%; p < 0.05). A greater proportion with than without AHR had 1 or more episode of pneumonia within 2 years (46% vs 9%; p < 0.01), 1 or more asthma episode during the study period (39% vs 1.9%; p < 0.001), or wheeze noted during clinic visits (62% vs 17%; p < 0.01). The proportion that experienced acute bronchitis did not differ in the two groups. CONCLUSIONS: This study suggest that HIV-infected persons do not have increased prevalence of nonspecific AHR. In HIV disease, AHR is associated asthma, but not episodic acute bronchitis. Thus, the possibility that airway injury without demonstrable AHR might complicate HIV disease remains.

Incidence of tuberculosis in the United States among HIV-infected persons. The Pulmonary Complications of HIV Infection Study Group.

BACKGROUND: The resurgence of tuberculosis in the United States is largely linked to the human immunodeficiency virus (HIV) epidemic. Despite this link, the epidemiology of tuberculosis and preventive strategies in patients infected with HIV are not completely understood. OBJECTIVES: To determine the incidence and predictors of tuberculosis in HIV-infected persons. DESIGN: Prospective, multicenter cohort study. SETTING: Community-based cohort of persons with and without HIV infection at centers in the eastern, midwestern, and western United States. PARTICIPANTS: 1130 HIV-seropositive patients without AIDS who were followed for a median of 53 months (814 homosexual men, 261 injection drug users, and 55 women who had acquired HIV through heterosexual contact). MEASUREMENTS: Delayed hypersensitivity response to purified protein derivative (PPD) tuberculin and mumps antigen, CD4 T-lymphocyte counts, and frequency of tuberculosis. RESULTS: 31 HIV-seropositive patients developed tuberculosis (0.7 cases per 100 person-years [95% CI, 0.5 to 1.0]). The most important demographic risk factor was location (adjusted risk ratio for eastern compared with midwestern and western United States, 4.1 [CI, 2.0 to 8.4]). Tuberculosis occurred more frequently in persons with CD4 counts of less than 200 cells/mm3 (1.2 cases per 100 person-years [CI, 0.7 to 1.9]) than in those with higher counts (0.5 cases per 100 person-years [CI, 0.3 to 0.8]). The rate of tuberculosis was highest among tuberculin converters (5.4 cases per 100 person-years [CI, 1.1 to 15.7]), lower among patients who were PPD positive at first testing (4.5 cases per 100 person-years [CI, 1.6 to 9.7]), and lowest among patients who remained PPD negative (0.4 cases per 100 person-years [CI, 0.2 to 0.7]). Tuberculosis was not reported among persons who had PPD reactions of 1 to 4 mm. Compared with that of patients who tested positive for mumps, the risk for tuberculosis of those who tested negative was increased about sevenfold if they were PPD positive (P < 0.03) and fourfold if they were PPD negative (P < 0.02). CONCLUSIONS: Incidence of tuberculosis was higher in the eastern United States, in patients with CD4 counts of less than 200 cells/mm3, and in PPD-positive patients. Analysis of tuberculin reaction size supports the current interpretive criteria of the Centers for Disease Control and Prevention. Nonreactivity to mumps antigen indicated increased risk for tuberculosis independent of PPD response.

Impact of a gonorrhea control program, including selective mass treatment, in female sex workers.

A gonorrhea control program initiated in 1967 in registered female sex workers (FSWs) in the Philippines involved weekly endocervical cultures for Neisseria gonorrhoeae, with treatment of FSWs found infected or named as contacts by US Navy servicemen. Gonorrhea prevalence in FSWs in Olongapo city fell from 11.9% to 4.0% within 4 months, and gonorrhea incidence in servicemen at nearby Subic Bay fell by half. Selective mass treatment (SMT) with oral ampicillin-probenecid or tetracycline was then given to registered FSWs in an attempt to further reduce gonorrhea rates. N. gonorrhoeae was isolated from 105 (4.0%) of 2640 FSWs before SMT and from 43 (1.6%) 1 week later (P < .001). However, gonorrhea incidence among servicemen fell no lower, and gonorrhea prevalence in FSWs quickly returned to higher levels. Thus, after implementation of weekly screening and treatment of FSWs found infected or named as contacts, SMT of FSWs (without increasing condom use or treating regular partners) contributed nothing further to gonorrhea control.

Reliability of anergy skin testing in persons with HIV infection. The pulmonary Complications of HIV Infection Study Group.

Testing with antigens that elicit delayed-type cutaneous hypersensitivity reactions is commonly used to evaluate immune competence in persons infected with the human immunodeficiency virus; however, the reliability of such testing has not been determined. We performed serial testing with tuberculin, mumps, and Candida antigens in 491 HIV-infected persons and found that 30% of persons who initially had no reaction (0 mm) to any of the three antigens, and, thus, were considered to be anergic, had reaction to the mumps or Candida antigen when they were retested 12 months later. We also examined the results of mumps antigen tests in 50 subjects who had a negative tuberculin tests after an initial positive test. The mumps antigen test was positive in 39% of the subjects when the tuberculin test was falsely negative. We conclude that tests commonly used to define anergy cannot reliably identify the anergic state. Moreover, using the mumps antigen to aid in the interpretation of the tuberculin test will often lead to erroneous conclusions. These data indicate that the results of anergy testing should not be used to make individual patient decisions concerning preventive therapy for tuberculosis.

Bronchoscopic biopsies and bronchoalveolar lavage.

Bronchoscopy with biopsies, bronchoalveolar lavage, and other sampling techniques are frequently needed to establish definitive diagnoses of pulmonary disorders. Combinations of specimens provide superior results to single specimens alone for lung cancer, including those which are endoscopically visible and peripheral in location. Transbronchial biopsy is useful to establish tissue diagnoses in certain diffuse parenchymal lung diseases with specific recognizable histologic patterns such as sarcoidosis or eosinophilic granuloma, but it is less useful for disorders such as interstitial pulmonary fibrosis. Patients with tuberculosis can be diagnosed by performing bronchoscopy, but other sampling techniques are equally good and safer for the bronchoscopist and other health care workers. Bronchoalveolar lavage is especially valuable for confirming infectious complications in immunocompromised hosts, and it also has great potential to elucidate basic mechanisms of pulmonary diseases in research applications.

Long-term survival following surgical treatment of solitary brain metastasis in non-small cell lung cancer.

Dissemination of lung cancer beyond the intrathoracic lymph nodes (stage IV disease) implies surgical unresectability. However, solitary brain metastases (SBMs) from non-small cell lung cancer (NSCLC) have often been treated by combined resection of the primary tumor and its metastasis. Such an aggressive approach appears to substantively improve patient outcome and provide better quality of life in selected cases. A search of the literature reveals extended survival (10 years or longer) in 16 patients following combined surgical excision. We report three patients with NSCLC and isolated central nervous system involvement who achieved exceptionally long survival. The existing literature on SBMs from NSCLC is reviewed.

Lack of usefulness of radiographic screening for pulmonary disease in asymptomatic HIV-infected adults. Pulmonary Complications of HIV Infection Study Group.

OBJECTIVE: To determine the use of chest radiographs in the screening of asymptomatic adults infected with the human immunodeficiency virus (HIV). METHODS: A prospective, multicenter study of the pulmonary complications of HIV infection in a community-based cohort of persons with and without HIV infection. The subjects included 1065 HIV-seropositive subjects without the acquired immunodeficiency syndrome at the time of enrollment: 790 homosexual men, 226 injection drug users, and 49 women with heterosexually acquired infection. Frontal and lateral chest radiographs were performed at 3-, 6-, and 12-month intervals, CD4 lymphocyte measurements at 3- and 6-month intervals, tuberculin and mumps skin tests at 12-month intervals, and medical histories and physical examinations at 3- and 6-month intervals. Pulmonary diagnoses that occurred within 2 months following each radiograph were analyzed and correlated with the radiographic results. RESULTS: Evaluable screening chest radiographs (5263) were performed in HIV-seropositive subjects while they were asymptomatic; of these, 5140 (98%) were classified as normal and 123 (2%) as abnormal. A new pulmonary diagnosis was identified within 2 months following a screening radiograph in 55 subjects. Only 11 of these subjects had abnormal radiographs; the sensitivity of the radiograph was 20%. The sensitivity was similarly low at baseline, within each transmission category, and in subjects whose CD4 lymphocyte counts were less than 0.2 x 10(9)/L (200/microL). The types of pulmonary diseases that occurred were similar in the subjects with normal and abnormal screening radiographs. CONCLUSION: Screening chest radiography in asymptomatic HIV-infected adults is unwarranted because the diagnostic yield is low.

Castleman's disease.

OBJECTIVE: To describe the two variants of Castleman's disease--the hyaline-vascular type and the plasma-cell type--and discuss the associated histologic features. DESIGN: We present a case of the hyaline-vascular type and review the literature. RESULTS: Castleman's disease was once thought to be localized and self-limited, but in recent years, reports have described a multicentric variety with severe systemic manifestations and, at times, an inexorable clinical course. Unlike the localized type for which surgical excision is curative regardless of the histologic type, multicentric disease often necessitates aggressive systemic therapy and portends a poor outcome. Little is known about the cause of this disorder, but the bulk of evidence points toward faulty immunoregulation that results in excessive proliferation of B lymphocytes and plasma cells in lymphoid organs. CONCLUSION: Castleman's disease is rare and poorly understood. The diagnosis is "contextual" and must be considered in the appropriate clinical setting and only after all other causes of lymphadenopathy have been investigated and excluded. The optimal therapeutic regimen is unknown.

Bacterial pneumonia in persons infected with the human immunodeficiency virus. Pulmonary Complications of HIV Infection Study Group.

BACKGROUND: Patients with human immunodeficiency virus (HIV) infection are at increased risk for bacterial pneumonia in addition to opportunistic infection. However, the risk factors for bacterial pneumonia and its incidence in this population are not well defined. METHODS: In a multicenter, prospective, observational study, we monitored 1130 HIV-positive and 167 HIV-negative participating adults for up to 64 months for pulmonary disease. The HIV-positive group comprised 814 homosexual or bisexual men, 261 injection-drug users, and 55 female partners of HIV-infected men. RESULTS: There were 237 episodes of bacterial pneumonia among the HIV-positive participants (rate, 5.5 per 100 person-years), as compared with 6 episodes among the HIV-negative participants (rate, 0.9 per 100 person-years; P < 0.001). The rate of bacterial pneumonia increased with decreasing CD4 lymphocyte counts (2.3, 6.8, and 10.8 episodes per 100 person-years in the strata with more than 500, 200 to 500, and fewer than 200 cells per cubic millimeter, respectively; P < or = 0.022 for each comparison). Injection-drug users had a higher rate of bacterial pneumonia than did homosexual or bisexual men or female partners. In the stratum with the fewest CD4 lymphocytes, cigarette smoking was associated with an increased rate of pneumonia. Mortality was almost four times higher among participants with an episode of pneumonia than among the others. Prophylaxis with trimethoprim-sulfamethoxazole was associated with a 67 percent reduction in confirmed episodes of bacterial pneumonia (P = 0.007). CONCLUSIONS: Bacterial pneumonia is more frequent in HIV-positive persons than in seronegative controls, and the risk is highest among those with CD4 lymphocyte counts below 200 per cubic millimeter and among injection-drug users.

Pulmonary function tests in HIV-infected patients without AIDS. Pulmonary Complications of HIV Infection Study Group.

To determine the prevalence, incidence, and types of lung diseases that occur in association with HIV infection, 1,353 subjects, including HIV-seropositive homosexual men, injection drug users, female sexual partners of HIV-positive men, and HIV-seronegative control subjects from the first two transmission categories were evaluated prospectively in a multicenter study. Patients with AIDS at the time of initial evaluation were excluded. One thousand two-hundred ninety-four subjects who had no AIDS-defining diagnosis within 3 mo of enrollment had measurements of FVC, FEV1 and DLCO at the time of enrollment. As a group, all subjects had mean values of FVC and FEV1 close to 100% predicted. Those with CD4 counts below 200/mm3 had slightly reduced DLCO compared with the others. Subjects with a history of HIV-associated symptoms (thrush, weight loss, herpes zoster) also had a reduced DLCO compared with those without symptoms. Injection drug users had reduced FVC, FEV1 and DLCO compared with homosexual men and female sexual partners of HIV-infected men, with DLCO more substantially reduced. Part of the reduction in DLCO in drug users was attributable to factors other than HIV infection, especially cigarette smoking and race. Using predicted values that take cigarette smoking into account, the prevalence of abnormality in DLCO was higher among injection drug users (33.3%) than among homosexual men (11.2%) and female sexual partners (12.7%). These results show that advanced HIV infection, characterized by CD4 count < 200/mm3 or HIV-associated symptoms, and factors unrelated to HIV infection, including race, cigarette smoking, and injection drug use, are all associated with reductions in DLCO measurements.

Routine analysis of induced sputum is not an effective strategy for screening persons infected with human immunodeficiency virus for Mycobacterium tuberculosis or Pneumocystis carinii. Pulmonary Complications of HIV Infection Study Group.

A prospective multicenter cohort study comprising 1,171 individuals who were seropositive for human immunodeficiency virus (HIV) but did not have AIDS at the time of enrollment and 182 HIV-seronegative controls, was studied by means of routine induced-sputum analysis in an attempt to detect occult tuberculosis or Pneumocystis carinii pneumonia. One occult case of tuberculosis was discovered upon the patient's enrollment (at baseline); none were discovered during follow-up. Two additional Mycobacterium tuberculosis isolates were recovered (one at baseline, one during follow-up) from subjects with symptoms or abnormalities evident on chest roentgenograms. Three specimens were false-positive (one for M. tuberculosis, two for P. carinii). Five pathogenic nontuberculous mycobacteria isolates were recovered during follow-up. Nonpathogenic, nontuberculous mycobacteria were recovered from 51 (4.6%) of 1,113 baseline specimens and 56 (3.7%) of 1,518 follow-up specimens, primarily at a center where the water supply was contaminated. We conclude that routine induced-sputum analysis is not an effective strategy for screening HIV-infected asymptomatic subjects for tuberculosis or P. carinii pneumonia before the onset of clinically recognizable disease activity.