Sputum induction as a diagnostic tool for community-acquired pneumonia in infants and young children from a high HIV prevalence area.

Sputum induction is a standard diagnostic procedure to identify pathogens in lower respiratory tract secretions in adults with pneumonia, but has rarely been studied or used in infants and young children. Our aim was to determine the usefulness of induced sputum (IS) as a diagnostic method for infants and children hospitalized with community-acquired pneumonia (CAP) in a high HIV prevalence area. Children hospitalized for CAP were prospectively enrolled over a year. IS was obtained by nebulization with hypertonic (5%) saline, physiotherapy, and suctioning. Sputum was submitted for bacterial and mycobacterial culture and P. carinii detection. Gastric lavages (GLs) were done for M. tuberculosis culture; a nasopharyngeal aspirate (NPA) was obtained for bacterial culture and P. carinii detection. IS was obtained in 210 children (median age, 7 (25th to 75th percentile, 3-18) months); 138 (66%) were HIV-infected; 148 (70%) were receiving supplemental oxygen. Bacteria were isolated from 101 (50%) IS and 141 (70%) NPA paired specimens (P < 0.001). A significantly higher rate of S. aureus, H. influenzae, M. catarrhalis, and S. pneumoniae was found in NPAs compared to IS; this pattern was particularly evident in HIV-infected children. M. tuberculosis was cultured from sputum in 19 patients (9%); GLs performed in 142 children were positive in only 9 (6%). The difference (95% confidence interval) between yields for M. tuberculosis from culture of IS compared to GL was 4.3% (95% CI, 0-5.6%; P = 0.08). P. carinii was identified from IS in 12 (5.7%) children; all corresponding NPAs were negative. Seven (3%) children could not tolerate sputum induction. Side effects included increased coughing in 4%, epistaxis in 3%, and wheezing responsive to bronchodilators in 1%. In conclusion, induced sputum is a useful and safe diagnostic procedure in infants and children with CAP from a high HIV prevalence area.