Primary preventive interventions to reduce ventilator-associated pneumonia in neonatal intensive care units: a meta-analysis

Background To assess preventive strategies for ventilator-associated pneumonia (VAP) in mechanically ventilated (MV) neonates. Methods PubMed, Embase, Web of Science, and Cochrane Library were searched in June 2025. Eligible studies evaluated primary VAP-prevention interventions in neonates receiving MV. Outcomes included odds ratios (ORs), incidence rate ratios (IRRs), and incidence rate differences (IRDs). Results Thirty-four studies involving 6,301 neonates were included. Bundle-based quality-improvement packages were associated with lower VAP occurrence (OR = 0.40, 95% CI 0.28–0.57; I² = 37%). MV-adjusted analyses showed lower incidence (IRR = 0.56, 95% CI 0.43–0.73) and 8.33 fewer VAP cases per 1000 MV-days (IRD = –0.0083, 95% CI –0.0127 to –0.0039). Oral care was associated with lower VAP odds (OR = 0.53, 95% CI 0.35–0.81; I² = 3%). Evidence for other interventions was limited and heterogeneous. Conclusions Bundle-based quality-improvement packages were associated with lower neonatal VAP rates, although heterogeneous bundle composition and non-randomized designs limit causal interpretation. Oral care was also associated with lower VAP odds, but confidence in the magnitude of effect is limited by small-study effects, inconsistent subgroup findings, and variable diagnostic criteria. Impact This study provides the most comprehensive synthesis to date focused exclusively on VAP in ventilated neonates, incorporating ventilator-day–adjusted outcomes. By using robust meta-analytic methods across diverse study designs, it clarifies which interventions have consistent evidence and which remain uncertain. Bundle-based quality-improvement packages were associated with meaningful relative and absolute reductions in neonatal VAP. The findings support further evaluation and implementation of bundle-based quality-improvement strategies. Uncertainty regarding the magnitude of benefit from oral care and the need for higher-quality controlled studies.