Discussion
This retrospective study of prospectively collected data cannot establish causality but merely suggest plausible associations. However, this cohort is unique in several ways. We evaluated mortality and neurodevelopmental outcomes at 18-22 months' corrected age in children who were born at