Janice Drake

Increased Susceptibility to Pulmonary Emphysema among HIV-Seropositive Smokers

Several reports (1-3) have suggested that before recognized AIDS-related respiratory complications develop, some HIV-seropositive persons may develop an accelerated form of lung injury that has physiologic features or findings on computed tomography of the chest that are consistent with pulmonary emphysema. Emphysema, defined as enlargement of the terminal air spaces and destruction of the alveolar walls (4), is uniquely characterized by disappearance of lung tissue (4). Insight into the cellular mechanisms critical to pathogenesis of this condition in humans has been limited by the lengthy course over which the disease develops and by the fact that only a minority of smokers develops significant disease (5). The possibility that HIV-seropositive persons may have a greater risk for developing accelerated lung destruction may have broad biological relevance regarding emphysema pathogenesis. With this in mind, we sought to extend previous anecdotal studies and attempted to prospectively characterize a group of HIV-infected persons exhibiting evidence of emphysema-like destructive lung disease. Methods The study sample consisted of 114 consecutive HIV-seropositive persons who underwent high-resolution computed tomography of the chest (1994 through 1997). This group was a sample of a larger cohort of 321 HIV-seropositive persons who had undergone a detailed assessment of respiratory symptoms and pulmonary function. Most participants were from Columbus, Ohio, and were recruited through advertisements, by word of mouth, and by the Ohio State University Medical Center AIDS-Clinical Trials Unit. Less than 10% of the participants were receiving protease inhibitors at the time of this study. Persons with a history of Pneumocystis carinii pneumonia and other pulmonary complications of AIDS were excluded. Controls consisted of 44 HIV - seronegative volunteers matched for age, sex, and smoking history; they were recruited from the general population by advertisements. All participants completed a modified American Thoracic Society Questionnaire (6) for symptoms, smoking history, drug use, history of pneumonia, and medications. Nonsmokers were identified as having a history of cumulative cigarette smoking of 1 pack-year or less. Pulmonary function studies were performed according to American Thoracic Society standards. The study was approved by the Ohio State University human subjects review board. Informed consent was obtained from all participants. Computed tomography was performed with a GE 9800 CT scanner (GE Medical Systems, Milwaukee, Wisconsin) or a Picker PQ 2000 CT scanner (Picker International, Solon, Ohio) with 1.5-mm collimation at 10-mm intervals through the chest. Scans were obtained at total lung capacity in the supine position. Images were reconstructed by using the high spatial frequency algorithm and were photographed at a lung window width of 1500 Hounsfield units (brightness level, 700 Hounsfield units). Emphysema was considered present if the scans showed evidence of bullae, thin-walled cystic spaces, or abnormal decreases in attenuation, accompanied by vascular disruption. Emphysema severity was estimated by assigning an emphysema score (0 to 10) for each lobe according to the percentage of the lobe that was affected. The lingula was considered a separate lobe. The total score represented the sum for all lobes. Scans were interpreted by two experienced chest radiologists, who were blinded to the patients HIV status and physiologic data. Participants were considered to have clear evidence of at least early emphysema if 1) the computed tomography emphysema score was 6 or higher [for example, 25% of two lobes involved with emphysema] or 2) pulmonary function tests demonstrated a total lung capacity greater than 120% of the predicted value, a diffusing capacity less than 60% of predicted, and computed tomographic evidence of emphysema in more than two lobes. Bronchoalveolar lavage and differential cell counts were performed according to standard techniques (7) on consenting participants. This included 46 HIV-seropositive smokers and 14 control smokers. A sample containing more than 1 000 000 cells was analyzed by a fluorescence-activated cell sorter for T-lymphocyte subtyping. An experienced microbiologist, who was blinded to HIV status, examined all lavage samples for the presence of fungi, acid-fast bacilli, P. carinii, and bacteria. Statistical analyses were performed by using the SAS JMP package (8). Analysis of variance was used to compare groups, and post hoc analysis was performed by using the Dunnett procedure (8). The Pearson chi-square test was used for equality of proportions; exact P values were calculated for small samples (9). Results With the exception of a lower CD4 count and diffusing capacity in the HIV-seropositive group, the two study groups had similar baseline characteristics (Table). Of note, 25% of the HIV-seropositive group had a history of oral thrush; however, other HIV-related opportunistic infections were rarely reported (<2%). The percentage of HIV-seropositive persons who reported any use of intravenous drugs was low (approximately 10%), and no significant differences in the use of intravenous drugs, crack cocaine, or marijuana were found between the HIV-seropositive group and the control group. Table. Clinical and Demographic Characteristics of the Study Groups Clinical variables did not significantly differ between the HIV-seropositive group of this study and the 321 persons in the overall HIV cohort (Table). For example, the median age of the overall cohort was 33 years (range, 20 to 66 years), the median CD4 count was 348 cells/mm3 (range, 0 to 1188 cells/mm3), and the median diffusing capacity was 83.7% of predicted (range, 38.8% to 153%). Sixty-three percent of the overall cohort smoked cigarettes. Emphysema was identified in 17 of 114 HIV-seropositive participants compared with 1 of 44 HIV-seronegative controls (P=0.025). Only one HIV-seronegative participant had a computed tomography emphysema score as high as 6, whereas 14 HIV-seropositive participants had scores between 6 and 23. The mean (SE) emphysema score for the HIV-seropositive group with emphysema was 10.5 1.7, the mean total lung capacity was 110% 6.9% of predicted, and the diffusing capacity was 60.1% 11.3% of predicted. Of note, persons with emphysema had only mild airflow obstruction (ratio of FEV1 to FVC, 69.2% 9.4%), which may relate to a higher than expected number of persons (approximately 30%) demonstrating subpleural or peripheral lesions (10). The Figure shows examples of precocious emphysema in HIV-seropositive persons, with accompanying emphysema scores. Figure. High-resolution computed tomographic scans of the chest.A. arrows B. Because a relatively high proportion of cigarette smokers appeared to be susceptible to early destructive changes, we specifically studied the role of smoking history. Thirty-seven percent (14 of 38) of HIV-seropositive smokers with a smoking history of 12 pack-years or more met criteria for emphysema, compared with 0% (0 of 14) HIV-seronegative controls (P=0.011). Furthermore, 46% (11 of 24) of HIV-seropositive participants with a smoking history of 25 pack-years or more met criteria for emphysema, compared with 0% (0 of 10) in the HIV-seronegative controls (P=0.013). Next, using current pack-year of cigarette smoking as a covariate, we compared lung-cell populations among three groups of smokers: HIV-seronegative smokers (n=14), HIV-seropositive smokers without emphysema (n=34), and HIV-seropositive smokers with emphysema (n=12). The numbers of alveolar macrophages and neutrophils in the lavage fluid were similar among the three groups. Although the two HIV-seropositive groups were found to have threefold more lymphocytes than the uninfected controls, no significant difference in lymphocyte numbers were noted between HIV-seropositive persons with and those without emphysema. However, when lymphocyte subtypes were examined, HIV-seropositive persons with emphysema were found to have the highest percentage of lavage lymphocytes bearing the cytotoxic phenotype; the mean (SE) value in this group was 58% 4.6%, compared with 46.6% 2.3% in HIV-seropositive smokers without emphysema (P<0.05) and 32.2% 4.6% in HIV-seronegative smokers (P<0.01). Of note, no pathogens were observed in microbiological stains of lavage fluid from study participants. Discussion This prospective study demonstrates the development of an accelerated form of pulmonary emphysema in a stable HIV-seropositive outpatient sample. Furthermore, the results suggest that the lesion is related to a heightened susceptibility to cigarette smoke. We hypothesize that HIV infection or secondary inflammatory abnormalities directly accelerate the process of smoking-induced parenchymal lung destruction. The cellular mechanisms predisposing HIV-seropositive smokers to emphysema are unclear. However, immunologic aspects of HIV disease may be relevant to understanding emphysema pathogenesis in the general population. Of particular interest are the many bronchoalveolar lavage and lung pathology studies that have demonstrated increased numbers of cytotoxic lymphocytes in the lungs of HIV-seropositive persons (11-13). Although prevailing theories for emphysema have focused on smoking-induced production of proteolytic enzymes by neutrophils and macrophages (4, 5), recent morphometric analyses of lung biopsy sections from non-HIV-infected smokers have demonstrated a high correlation between lung lymphocytes and the presence of emphysema (14, 15). Furthermore, experimental evidence suggests that viral activation of cytotoxic lymphocytes may contribute to parenchymal lung destruction (16). Another potential mechanism recently hypothesized is that latent viral infections may be an important cofactor in the development of chronic obstructive pulmonary disease (17). Adenoviral proteins, latently expressed in host epithelial cells, a

HIV Infection Increases Susceptibility to Smoking- Induced Emphysema*

2000;117;285S Chest Henk F. Kauffman and Dirkje S. Postma Steven R. Rutgers, Huib AM Kerstjens, Wim Timens, Nikolaos Tzanakis, * -Monophosphate in COPD ′ Hyperresponsiveness to Adenosine 5 Airway Inflammation and http://chestjournal.chestpubs.org/content/117/5_suppl_1/285S.2.full.html services can be found online on the World Wide Web at: The online version of this article, along with updated information and ISSN:0012-3692 ) http://chestjournal.chestpubs.org/site/misc/reprints.xhtml ( written permission of the copyright holder. this article or PDF may be reproduced or distributed without the prior Dundee Road, Northbrook, IL 60062. All rights reserved. No part of Copyright2000by the American College of Chest Physicians, 3300 Physicians. It has been published monthly since 1935. is the official journal of the American College of Chest Chest