Although a prevalence rate for bronchospastic disorder in AIDS has been reported in an earlier study, the prevalence cited in that report (3 percent) was considerably lower than that found here, and, in fact, was close to the rate of airway hyperreactivity found in the general population. This discrepancy is not attributable to physiologic criteria used to assess airway dysfunction, since we employed similar criteria derived from closely related normal population studies. In the previous study only 27 of 130 patients (21 percent) underwent pulmonary function testing, and selection criteria for testing were not adequately described. Therefore, the discrepancy in reported prevalence of airway dysfunction between that study and the current one may be related to differences in patient selection.
We also considered whether the high prevalence of abnormal airway function we observed could have resulted from selection bias. For example, could it be that from among all those with AIDS who were treated at this hospital only those with suspected airway dysfunction were referred for testing? This possible bias in the selection of subjects can be addressed in the following manner. During the study period a total of 153 AIDS patients were treated at the New England Deaconess Hospital (NEDH). For the sake of argument, it could be assumed that all 48 patients not referred to the pulmonary laboratory had normal function. This is equivalent to using the total of 153 AIDS patients followed at the NEDH as the denominator for calculating prevalence. In this scenario, we still find that 29 percent of all those with AIDS who were treated as either inpatients or outpatients have airway dysfunction. Since this figure is several times greater than the prevalence of airway dysfunction among the general population, we believe that our results do not merely reflect a bias in sample selection.
Tags: AIDS, airway function, pulmonary, respiratory