Survival rates of extremely preterm infants (gestational age <26 weeks) in Switzerland: impact of the Swiss guidelines for the care of infants born at the limit of viability

Background: Because ethical decision making in the care of extremely preterm infants varies widely across Europe, the Swiss Society of Neonatology decided to publish its own guidelines on the care of infants born at the limit of viability in 2002. Objective: To examine the potential impact of the guidelines on survival rates, short-term complication rates and centre-to-centre outcome differences of extremely preterm infants (22-25 completed weeks). Design: Population-based, retrospective cohort study. Setting: All nine level III neonatal intensive care units (NICU) and affiliated paediatric hospitals in Switzerland. Patients: 516 extremely preterm infants born alive between 1 January 2000 and 31 December 2004. Main outcome measures: Delivery room and NICU mortality rates, survival to hospital discharge and incidence of short-term complications in survivors were assessed. To study the impact of the guidelines, two cohorts from two different time periods were compared (years 2000/2001, n = 220; years 2003/2004, n = 204) whereas patients born in the year of the publication (2002, n = 92) were excluded. For centre-to-centre comparisons, the entire population in = 516) was analysed. Results: There was a significant increase in survival rates of extremely preterm infants from 31% to 40% (RR 1.24, 95% Cl 1.02, 1.50) after the publication of the Swiss guidelines. This improvement was largely explained by significantly improved survival from 42% to 60% (p = 0.01) among infants born at 25 completed weeks because of decreased NICU mortality. Improved survival was not associated with statistically significant changes in the incidence of short-term complications. Despite national guidelines, considerable centre-to-centre outcome differences have persisted. Conclusions: The publication of the Swiss guidelines was followed by significantly improved survival of extremely preterm infants but had no impact on centre-to-centre differences.