Induced sputum in HIV- infected patients: diagnosis of acute pulmonary diseases

OBJECTIVE: To make a narrative review of the accuracy of induced sputum for diagnosis of pulmonary disease in HIV-infected patients. Data sources: The MEDLINE, LILACS, EMBASE and the Cochrane Library were searched. reference lists, abstracts of conference proceedings and scientific meetings were hand searched. METHODS: Study selection: Fifteen articles that specifically addressed the stated purpose were selected. Data extraction: Yield of sputum induction and fiberoptic bronchoscopy with bronchoalveolar lavage were analyzed using explicit methodologic to evaluate the quality of clinical trials. RESULTS: Sputum induction demonstrated 55.5% sensitivity and 98.6% specificity to Pneumocystis pneumonia. Sensitivity of sputum induction was significantly higher with immunofluorescence than with cytochemical staining (67.1 versus 43.1%). Sputum induction for diagnosis of bacterial pneumonia demonstrated 60% sensitivity, 40% specificity, 80% positive predictive value, 20% negative predictive value and 56% accuracy. In relation to tuberculosis, sputum induction demonstrated 36% sensitivity, 100% specificity, 100% positive predictive value and 54.2% negative predictive value. CONCLUSION: Sputum induction seems to be effective and safe for diagnosis of pulmonary diseases in HIV-infected patients.