Components of systemic inflammatory response as predictors of mortality in children presenting with fever at the emergency medical department, Muhimbili National Hospital.

Abstract

Introduction: Sepsis is defined as a known or suspected infection in a patient with elements of the systemic inflammatory response syndrome (SIRS). Septic patients present with a variety of clinical manifestations, but temperature dysregulation, tachycardia, tachypnea, and an abnormal white blood cell (WBC) count are considered cardinal components of SIRS. We investigated the predictive value of SIRS criteria for in-hospital mortality among febrile children under 5 years old presenting to the Emergency Department (ED) at Muhimbili National Hospital in Dar-es-Salam, Tanzania. Methods: This was a descriptive cohort study of febrile children under 5 years, presenting to our ED. Providers prospectively completed a standardized data sheet. Outcome data was obtained from hospital records and telephone follow-up. Study data were entered into Excel (Microsoft, Redmond, WA, USA) and analyzed in SAS 9.3 (Cary, North Carolina, USA). Results: We enrolled 105 patients between August and November 2012. The median age was 14 months, with 80% over 6 months old, and 63.8% were male. 57 (54.3%) children were referred from outside facilities. The overall mortality rate was 19%, and 90% of children who died had ≥2 SIRS criteria. Mortality in children with ≥2 SIRS criteria (in addition to fever) was significantly higher (27.7% versus 5%) than in those with 0-1 SIRS criteria, and children with fever and >2 SIRS criteria were seven times more likely to die (OR 7.05, p=0.01). 85 children were discharged from the hospital, and of the 64 (75.3%) children we were able to reach after discharge, all were alive at 14 day xi telephone follow-up. 19/85 children who survived to hospital discharge were lost to follow up. Conclusion: SIRS criteria may be helpful to predict febrile children at high risk of mortality. Further studies are needed to validate these findings in larger cohorts