Factors associated with neurodevelopmental outcome at 2 years after very preterm birth: the population-based Nord-Pas-de-Calais EPIPAGE cohort.

OBJECTIVE We sought to (1) evaluate at 2 years the postsurfactant era developmental outcome of children who were born before 33 weeks of gestational age (GA) in the Nord-Pas-de-Calais area in France in 1997 and (2) identify risk factors of poor developmental quotient (DQ). Children were part of the EPIPAGE study, which included all of these births in 9 French regions. METHODS A prospective observational study was conducted of all births before 33 weeks in 1997. Risk factors of poor DQ were obtained from a multiple linear regression, and results were expressed as DQ differences with 95% confidence intervals. RESULTS A total of 546 births were included in the study. A total of 461 (84.4%) had a clinical evaluation at 2 years of age, and 380 (69.6%) had an assessment with the use of the Brunet-Lezine scale of infant development. Their mean GA was 29.9 weeks (29.7-30.1 weeks), and mean birth weight was 1378 g (1338-1418 g). A total of 9% had a recognizable pattern of cerebral palsy, 0.2% were blind, and 0.8% required hearing aids. The mean DQ was 94 +/- 11 and decreased from 97 at 32 weeks to 86 at 24 to 25 weeks. After multivariate analysis, children who were born at 24 to 25 weeks had a mean DQ reduction of 11 points (-20 to -1) compared with those who were born at 32 weeks, but minor differences were found from 26 to 32 weeks. Boys had a DQ 4 points lower than girls (-7 to -1). CONCLUSION In this study, the outcome of extremely preterm infants was poor. After 25 weeks, outcome was related mainly to the sociocultural level of the family and to the presence of severe cerebral ultrasound abnormalities. Consequently, in the postsurfactant era, we have to propose follow-up programs to children who are born extremely preterm and to concentrate our efforts on children with less-than-optimal social and family setting.