Clinical indicators of Pneumocystis jiroveci pneumonia (PCP) in South African children infected with the human immunodeficiency virus

SummaryBackgroundPneumocystis pneumonia (PCP) caused by Pneumocystis jiroveci is common in HIV-infected children, producing substantial morbidity and mortality. Initiation of timely, effective therapy depends on clinical identification of children with PCP.ObjectiveTo develop a clinical decision rule to diagnose PCP in HIV-infected children for use where diagnostic resources are limited.MethodsAnalysis of data collected during a prospective incidence study of the etiology, features, and outcome of HIV-infected children hospitalized with pneumonia.ResultsFour clinical variables were independently associated with a diagnosis of PCP in multivariate analysis: age <6 months (OR 15.6; 95% CI 2.4–99.8; p = 0.004), respiratory rate >59 breaths/min (OR 8.1; 95% CI 1.5–53.2; p = 0.018), arterial percentage hemoglobin oxygen saturation (SaO2) ≤92% (OR 5.1; 95% CI 1.0–26.1; p = 0.052) and absence of history of vomiting (OR 11.2; 95% CI 1.9–68.0; p = 0.008). The sensitivity and specificity of diagnosing PCP with any two or more of these variables were 1.00 (95% CI 0.74–1.00) and 0.49 (95% CI 0.39–0.59), respectively. Diagnosing PCP with three or more of the indicators had a decreased sensitivity of 0.75 (95% CI 0.43–0.95) and increased specificity of 0.90 (95% CI 0.83–0.95).ConclusionEmpirical anti-pneumocystis therapy should be considered in HIV-infected infants presenting with tachypnea, hypoxia and absence of vomiting.