A Quality Improvement Bundle to Reduce Central Line-Associated Bloodstream Infections in Neonatal Intensive Care Unit: An Observational Study.

Background: Dedicated bundles were proven to reduce CLABSIs in a neonatal intensive care unit (NICU). Methods: We performed an observational pre-post study to evaluate the impact of a bundle for CLABSI prevention in our NICU. All umbilical vein catheters (UVCs) and epicutaneo-caval catheters (ECCs) with dwell time > 2 days were included. The primary outcome was CLABSI rate/1000 central line days. Results: A total of 145 catheters (67 UVCs and 78 ECCs) and 142 catheters (65 UVCs and 77 ECCs) were inserted before and after bundle implementation, respectively. The duration of the UVC was significantly shorter before than after the bundle [4 (3-6) vs. 8 (6-11) days; p < 0.0001], while the duration of the ECC did not differ [10 (6-17) vs. 11 (6-19) days; p = 0.711]. CLABSI were less frequent after than before bundle (3.6 vs. 10.7/1000 CL days; p = 0.042); both UVC-related and ECC-related CLABSI were significantly reduced (0 vs. 7.2/1000 CL days, p = 0.015; and 4.4 vs. 12.3/1000 CL days, p = 0.044, respectively). The Kaplan-Meier curve for ECC-related CLABSIs showed no differences between the two periods (p = 0.255), but higher survival without CLABSIs after vs. before bundle was found if considering only ECC with dwell time < 14 days (p = 0.040). Gestational age (p = 0.004) and bundle (p = 0.026) were predictive factors for CLABSIs. Non-infective complications were significantly less frequent after than before bundle (11 vs. 20%, p = 0.033). Conclusions: Our bundle reduced the overall CLABSI rate, and both UVC- and ECC-related CLABSI occurrence. The efficacy for the reduction in ECC-related CLABSIs seems limited to the first 14 days of dwell time.