Effect of nifedipine and atosiban on perinatal brain injury: secondary analysis of the APOSTEL-III trial

Abstract

Objective: Brain injury in neonates born prematurely is associated strongly with poor neurodevelopmental outcome. The aim of this study was to evaluate whether tocolysis with nifedipine or atosiban in women with threatened preterm birth can reduce the incidence of overall brain injury in neonates born prematurely. Methods: This was a secondary analysis of the APOSTEL-III trial (Dutch Clinical Trial Registry, no. NTR2947), a randomized clinical trial in which women with threatened preterm labor between 25 and 34 weeks of gestation were allocated to treatment with nifedipine or atosiban. In this secondary analysis, women delivered at ≤ 32 weeks of gestational age in the two main contributing centers were included. Primary outcome was the presence of neonatal brain injury, which was defined as presence of abnormalities on ultrasound investigation and classified into mild and severe. To evaluate type and severity of brain injury, all neonatal ultrasounds performed during neonatal intensive and medium care admission were analyzed. To test the robustness of our results, a sensitivity analysis was performed assessing differences in baseline or known risk factors for brain injury. Results: A total of 117 neonates (from 102 women) were studied, of which 51 had been exposed to nifedipine and 66 to atosiban. Brain injury was observed in 22 (43.1%) neonates in the nifedipine group compared with 37 (56.1%) in the atosiban group (OR, 0.60; 95% CI, 0.29–1.24). Presence of mild brain injury was comparable between the nifedipine (33.3%) and atosiban (48.5%) groups (OR, 0.53; 95% CI, 0.25–1.13). Severe brain injury was also comparable between the groups, observed in 9.8% of neonates in the nifedipine vs 7.6% of those in the atosiban group (OR, 1.33; 95% CI, 0.36–4.85). Intraventricular hemorrhage (≥ Grade I) was the most frequently seen ultrasound abnormality, observed in 18 (35.3%) neonates in the nifedipine group vs 25 (37.9%) in the atosiban group (OR, 0.90; 95% CI, 0.42–1.91). The sensitivity analysis, with adjustment for maternal age and gestational age at randomization, showed no statistical difference between the groups for presence of brain injury (OR, 0.58; 95% CI, 0.27–1.27). Conclusion: In children born before 32 weeks of gestation after the use of tocolytics, the prevalence of brain injury was high. No significant differences were found with respect to overall brain injury between neonates exposed to nifedipine and those exposed to atosiban. However, as this study was a secondary analysis of the APOSTEL III trial, it was underpowered for brain injury.