PNEUMOCYSTIS-CARINII PNEUMONIA IN ZIMBABWE

Citation
As. Malin et al., PNEUMOCYSTIS-CARINII PNEUMONIA IN ZIMBABWE, Lancet, 346(8985), 1995, pp. 1258-1261
Citations number
25
Categorie Soggetti
Medicine, General & Internal
Journal title
LancetACNP
ISSN journal
01406736
Volume
346
Issue
8985
Year of publication
1995
Pages
1258 - 1261
Database
ISI
SICI code
0140-6736(1995)346:8985<1258:PPIZ>2.0.ZU;2-0
Abstract
Pneumocystis carinii pneumonia (PCP) is said to be rare in Africa, wit h reported rates of 0-22% in human-immunodeficiency-virus (HIV) infect ed individuals with respiratory symptoms. Over one year in a central h ospital in southern Africa, 64 HIV-infected patients with acute diffus e pneumonia unresponsive to penicillin and sputum smear-negative for a cid-fast bacilli underwent fibreoptic bronchoscopy. Bronchoalveolar la vage fluid was assessed for bacteria, fungi, Pneumocystis carinii, and mycobacteria. 21 patients (33%) had PCP and 24 (39%) had tuberculosis ; 6 of these had both infections. 5 patients had Kaposi's sarcoma (KS) associated with PCP, tuberculosis, or another infection, in 1 patient KS was the only finding, and in 21 no pathogen was identified. A logi stic regression model was used to assess clinical, radiographic, and a rterial blood gas predictors of PCP and tuberculosis, Fine reticulonod ular shadowing on the chest radiograph (nodular component <1 mm) was t he strongest independent predictor of PCP (odds ratio 8.5 [95% CI 6.1- 10.9]). A respiratory rate of more than 40/min was the best clinical p redictor of PCP (odds ratio 11.2 [95% CI 8.8-13.6]). Median CD4(+) T c ell count for all cases of PCP was 134/mu L (range 5-355) and for tube rculosis without PCP 206/mu L (range 61-787). In resource-limited coun tries, a regionally appropriate management algorithm is required.