Survival of ventilated extremely premature neonates with severe intraventricular hemorrhage

BACKGROUND: Severe intraventricular hemorrhage (IVH) is a leading mortality risk factor among extremely premature neonates. Because other life-threatening conditions also occur in this population, it is unclear whether severe IVH is independently associated with death. The existence and potential implications of regional variation in severe IVH-associated mortality are unknown. METHODS: We performed a retrospective cohort study of mechanically ventilated neonates born at 22 to 29 weeks' gestation who received care in 242 American NICUs between 2000 and 2014. After building groups composed of propensity score-matched and center-matched pairs, we used the Cox proportional hazards analysis to test our hypothesis that severe IVH would be associated with greater all-cause in-hospital mortality, defined as death before transfer or discharge. We also performed propensity score-matched subgroup analyses, comparing severe IVH-associated mortality among 4 geographic regions of the United States. RESULTS: In our analysis cohort, we identified 4679 patients with severe IVH. Among 2848 matched pairs, those with severe IVH were more likely to die compared with those without severe IVH (hazard ratio 2.79; 95% confidence interval 2.49-3.11). Among 1527 matched pairs still hospitalized at 30 days, severe IVH was associated with greater risk of death (hazard ratio 2.03; 95% confidence interval 1.47-2.80). Mortality associated with severe IVH varied substantially between geographic regions. CONCLUSIONS: The early diagnosis of severe IVH is independently associated with all-cause in-hospital mortality in extremely premature neonates. Regional variation in severe IVH-associated mortality suggests that shared decision-making between parents and neonatologists is strongly influenced by ultrasound-based IVH assessment and classification.