Attributable Mortality for Pediatric and Neonatal Central Line-Associated Bloodstream Infections in Greece

Abstract

Central line-associated bloodstream infections (CLABSIs) are the most frequent pediatric hospital-acquired infections and significantly impact outcomes. The aim of this study was to estimate the attributable mortality for CLABSIs in pediatric and neonatal patients in Greece. A retrospective matched-cohort study was performed, in two tertiary pediatric hospitals. Inpatients with a central line in neonatal and pediatric intensive care units (NICUs and PICUs), hematology/oncology units, and a bone marrow transplantation unit between June 2012 and June 2015 were eligible. Patients with confirmed CLABSI were enrolled on the day of the event and were matched (1:1) to non-CLABSI patients by hospital, hospitalization unit, and length of stay prior to study enrollment (188 children enrolled, 94 CLABSIs). Attributable mortality was estimated. During the study period, 22 CLABSIs and nine non-CLABSIs died (23.4 vs. 9.6%, respectively, p = 0.011), leading to an attributable mortality of 13.8% (95% confidence interval [CI] = 3.4–24.3%). Children in PICUs were more likely to die, presenting an attributable mortality of 20.2% (95% CI = − 1.4–41.8%), without reaching, however, statistical significance. After multiple logistic regression, CLABSIs were four times more likely to die (odds ratio [OR] = 4.29, 95% CI = 1.28–14.36, p = 0.018). Survival analysis showed no difference in time to death after study enrollment between CLABSIs and non-CLABSIs (log-rank p = 0.137, overall median survival time = 7.8 months). Greek pediatric mortality rates are increased by the CLABSI occurrence, highlighting the importance of infection prevention strategies.