[Mortality and morbidity of very premature infants in Baden-Württemberg depending on hospital size. Is the current degree of regionalization adequate?].

Background: Regionalization of perinatal and neonatal care improves outcome. The aim of this study was to compare outcome in preterm infants with a gestational age (GA) < 32 weeks in relation to patient volume.

Methods: Outcome data from the state-wide neonatal quality assurance system from 2003 - 2004 from all infants treated in one of the five largest perinatal centers in Baden-Wuerttemberg were subtracted from the total dataset. Data derived from these five centers was compared with data from all other remaining NICU's in the state.

Results: Mortality was 33.3 % vs. 15.0 % (other NICU's vs. five perinatal centers; p < 0.001) for infants < 26 weeks GA, 11.4 % vs. 8.9 % (n. s.) for infants 26 - 27 weeks GA, and 2.5 % vs. 3.5 % (n. s.) for infants 28 - 31 weeks GA. When analyzed as one group of infants < 28 weeks GA, mortality was 20.1 vs. 12.1 % (p = 0.003). The rate of intraventricular hemorrhage degrees III-IV was 30.2 % vs. 18.6 % (p = 0.015) for infants < 26 weeks GA, 14.5 % vs. 10.2 % (n. s.) for infants 26 - 27 weeks GA, and 2.9 % vs. 2.5 % (n. s.) for infants 28 - 31 weeks GA. The rate of periventricular leukomalacia was 11.3 % vs. 6.7 % (p = 0.18) for infants < 26 weeks GA, 6.1 % vs. 2.8 % (n. s.) for infants 26 - 27 weeks GA, and 2.8 % vs. 2.3 %; (n. s.) for infants 28 - 31 weeks GA.

Conclusion: This study supports the hypothesis, that regionalization of neonatal care for very immature infants to few perinatal centers with a large case load may improve survival of these infants, and may reduce morbidity, associated with long-term sequelae.