BACKGROUND: A high proportion of extremely preterm (EPT) infants are born by cesarean section (CS). However, whether the mode of delivery is related to long-term neurodevelopment in these infants is unclear. OBJECTIVES: This study aimed to determine whether the mode of delivery is associated with mortality and long-term outcomes in EPT infants. METHODS: We analyzed data of the Neonatal Research Network in Japan (NRNJ), a population-based, nationwide registry. Inclusion criteria were neonates who were born between 2003 and 2012 with a gestational age <26 weeks. The primary composite outcome was death before 3 years or neurodevelopmental impairment (NDI) at 3 years. Confounder-adjusted odds ratios (OR) were estimated by logistic generalized linear mixed models, which accounted for clustering within hospitals. RESULTS: 2,138 eligible infants (703 by vaginal delivery [VD] and 1,435 by CS) were identified for primary analysis. The composite outcome of death or NDI was not different between both groups (66.7% by VD and 62.7% by CS, p = 0.075). After multivariate analysis adjusting for confounders, we found that CS did not improve the composite outcome of death or NDI (OR = 0.839, 95% confidence interval = 0.816-1.328, p = 0.742). For secondary outcomes, mortality (OR = 0.824, p = 0.150), NDI (OR = 1.237, p = 0.165), and other neurodevelopmental outcomes were not different between the groups. CONCLUSIONS: Among neonates born at <26 weeks, CS does not improve mortality and neurodevelopmental outcomes at 3 years in the NRNJ cohort. However, because of several potential biases such as high rates of infants lost to follow-up, further evidence may be required.
BACKGROUND: A high proportion of extremely preterm (EPT) infants are born by cesarean section (CS). However, whether the mode of delivery is related to long-term neurodevelopment in these infants is unclear. OBJECTIVES: This study aimed to determine whether the mode of delivery is associated with mortality and long-term outcomes in EPT infants. METHODS: We analyzed data of the Neonatal Research Network in Japan (NRNJ), a population-based, nationwide registry. Inclusion criteria were neonates who were born between 2003 and 2012 with a gestational age <26 weeks. The primary composite outcome was death before 3 years or neurodevelopmental impairment (NDI) at 3 years. Confounder-adjusted odds ratios (OR) were estimated by logistic generalized linear mixed models, which accounted for clustering within hospitals. RESULTS: 2,138 eligible infants (703 by vaginal delivery [VD] and 1,435 by CS) were identified for primary analysis. The composite outcome of death or NDI was not different between both groups (66.7% by VD and 62.7% by CS, p = 0.075). After multivariate analysis adjusting for confounders, we found that CS did not improve the composite outcome of death or NDI (OR = 0.839, 95% confidence interval = 0.816-1.328, p = 0.742). For secondary outcomes, mortality (OR = 0.824, p = 0.150), NDI (OR = 1.237, p = 0.165), and other neurodevelopmental outcomes were not different between the groups. CONCLUSIONS: Among neonates born at <26 weeks, CS does not improve mortality and neurodevelopmental outcomes at 3 years in the NRNJ cohort. However, because of several potential biases such as high rates of infants lost to follow-up, further evidence may be required.