In order to determine the risk factors and outcomes of ventilator-associated pneumonia (VAP) in the neonatal intensive care unit (NICU), a retrospective cohort study was conducted on 259 patients who were ventilated >48 h. Clinical characteristics and risk factors were compared and non-conditional logistic regression analysis was performed to determine independent predictors for VAP. There were 52 episodes of VAP (20.1%). The main pathogens were G(-) bacterium (82.1%, 23/28). Hospital stay in the VAP group was 19.9+/-5.9 vs. 16.7+/-7.2 days in controls (P<0.01). The mortality rate of the VAP group was 13.5% (7/52) vs. 12.1% in controls (P>0.05). By logistic regression analysis the following independently predicted VAP: re-intubation (OR 5.3, 95% CI 2.0, 14.0), duration of mechanical ventilation (OR 4.8, 95% CI 2.2, 10.4), treatment with opiates (OR 3.8, 95% CI 1.8, 8.5) and endotracheal suctioning (OR 3.5, 95% CI 1.6, 7.4). VAP occurred at significant rates among mechanically ventilated NICU patients and is associated with care procedures. The risk factors of neonatal VAP were re-intubation, duration of mechanical ventilation, treatment with opiates and endotracheal suctioning. Additional studies are necessary to develop interventions to prevent neonatal VAP.