Is mode of delivery associated with the risk of necrotizing enterocolitis?

BACKGROUND

The pathogenesis of necrotizing enterocolitis remains poorly understood. Different newborn bacterial colonization due to cesarean delivery as opposed to vaginal delivery has been implicated as one potential contributing factor.

OBJECTIVE

We sought to determine whether mode of delivery is associated with the risk of necrotizing enterocolitis in neonates of women who were at imminent risk of delivery <32 weeks' gestational age.

STUDY

DESIGN

This is a secondary analysis of data from a randomized controlled trial of magnesium sulfate for the prevention of cerebral palsy. The parent trial included women with pregnancies at 24 to 31 6/7 weeks of gestation who were considered at imminent risk for preterm delivery. Women with a viable singleton gestation and data available on mode of delivery and development of necrotizing enterocolitis were included. Neonates delivered by vaginal delivery were compared to those delivered by cesarean delivery in bivariable analyses. Multivariable analysis was used to adjust for potential confounders.

RESULTS

A total of 2012 mother-neonate pairs were analyzed. Of these, 731 (36%) women delivered by cesarean delivery and 170 neonates (8.4%) developed necrotizing enterocolitis. In bivariable analyses, women who delivered by cesarean delivery were older (27 [interquartile range, 22-32] vs 24 [interquartile range, 20-29] years, P < .001) and had a higher frequency of chorioamnionitis (14.0% vs 10.5%, P = .021) compared to those who delivered vaginally. Neonates delivered by cesarean delivery were more premature (29.3 [interquartile range, 27.1-31.4] vs 30.3 [interquartile range, 27.9-31.9] weeks, P < .001), were smaller (1266 [interquartile range, 920-1643] vs 1465 [interquartile range, 1067-1850] g, P < .001), were more likely to be small for gestational age (4.4% vs 1.9%, P = .001), and had a higher frequency of proven sepsis (20.1% vs 14.7%, P = .002) compared to those who delivered vaginally. Rates of necrotizing enterocolitis (8.1% vs 8.7%, P = .65) and stage 2 or 3 necrotizing enterocolitis (4.4% vs 4.7%, P = .75) did not differ by mode of delivery. After adjusting for potential confounders, cesarean delivery continued to have no association with the frequency of necrotizing enterocolitis (adjusted odds ratio, 0.74; 95% confidence interval, 0.52-1.04) or stage 2 or 3 necrotizing enterocolitis (adjusted odds ratio, 0.73; 95% confidence interval, 0.46-1.16). This study was powered to detect a minimum relative risk of 1.5 for necrotizing enterocolitis associated with mode of delivery.

CONCLUSION

Mode of delivery was not significantly associated with necrotizing enterocolitis in neonates born to a cohort of women who were considered at imminent risk of extreme preterm delivery.