A meta-analysis of neurodevelopmental outcomes at 4-10 years in children born at 22-25 weeks gestation

Abstract

Aim: To update our meta-analysis on neurodevelopmental disability rates in children born at 22-25 weeks gestation. The main outcome measure was rates of neurodevelopmental disability in survivors at age 4-10 years.

Methods: We used a peer-reviewed electronic and grey search to identify articles. Two authors independently reviewed cohorts published after May 2012 with: born ≥1995 in a developed nation; assessed at 4-10 years; prospective; >65% follow-up; definitions for neurodevelopmental disability as per the EPICure cohort; results reported by gestation. We contacted authors for clarification. Random effects meta-analysis was used to estimate pooled proportions of neurodevelopmental disability. Within each study, the absolute change in proportions with each week was estimated and then pooled.

Results: We reviewed 3980 records; 21 articles were assessed and six were included. With the previous 9 cohorts, the meta-analysis now contains 15. Rates of moderate-to-severe neurodevelopmental disability were as follows: 42% (95% CI 23,64%; I² 0%) at 22; 41% (95% CI 31,52%; I² 20%) at 23; 32% (95% CI 25,39%; I² 45%) at 24; 23% (95% CI 18,29%; I² 60%) at 25 weeks. The analysis shows a significant decrease in risk of moderate-to-severe neurodevelopmental disability between each week (8.1% (95% CI -11.8, -4.5%); I² 0%; p < 0.001).

Conclusion: Physicians can use this high-quality data to support parents during decision-making.

Keywords: Child development; Developmental disabilities; Extremely premature; Gestational age; Infant.