RESUMO
OBJECTIVE: The aim of this study was to determine the accuracy of the procalcitonin (PCT) test for diagnosis of bacterial sepsis in pediatric cancer patients with febrile neutropenia. METHODS: Three major databases, MEDLINE, EMBASE and the Cochrane Library were searched for studies that evaluated the diagnostic value of PCT alone or compared with other laboratory markers such as C-reactive protein (CRP) to identify bacterial sepsis in children with fever and neutropenia. A bivariate model was used to derive summary sensitivity and specificity of the diagnostic tests. RESULTS: A total of 10 studies looking into PCT tests and 8 studies looking into CRP tests were included in the final analysis. The prevalence of bacterial sepsis was 304 of 1031 (29.5%) in PCT studies and 741 of 1316 (56.3%) in CRP studies. In terms of area under the receiver operating characteristic curve, PCT had comparable discrimination to CRP (area under the curve: 0.75 versus 0.74). PCT was not as sensitive as the CRP test. The pooled sensitivity of PCT was 0.59 (95% confidence interval [CI]: 0.42-0.74) as compared with 0.75 (95% CI: 0.61-0.85) for CRP. PCT was more specific than sensitive whereas CRP was more sensitive than specific in this population. The pooled specificity was 0.76 (95% CI: 0.64-0.85) for PCT and 0.62 (95% CI: 0.49-0.73) for CRP. PCT had greater likelihood ratio positive (2.50; 95% CI: 1.64-3.81), making it the better rule-in test. CONCLUSIONS: Of three markers potentially useful for diagnosing bacterial sepsis in children with fever and neutropenia, PCT had comparable diagnostic accuracy to CRP.