Neonatal intensive care at borderline viability--is it worth it?

Background Very preterm infants at the borderline of viability, especially those Aims To determine incremental changes in long-term outcome and consumption of resources by very preterm infants in the 1990s. Design Cohort study. Patients Consecutive livebirths with gestational ages 23–27 weeks born in the state of Victoria in two discrete eras, 1991–1992 ( n =401) and 1997 ( n =208), and randomly selected contemporaneous normal birthweight (NBW, birthweight >2499 g) controls (1991–1992 n =265, 1997 n =198). Main Outcome Measures Survival, and neurosensory impairments, disabilities and utilities, and consumption of hospital resources to 2 years of age. Results Compared with 1991–1992, in 1997 more infants were offered intensive care and the survival rate was higher at each week of gestation, and overall (absolute increase in survival 16%; 95% confidence interval, 8%, 24%). The largest increases in the survival and quality-adjusted survival rates were in infants at 23 weeks (31% and 20%, respectively). The incremental resource costs of improving survival and quality-adjusted survival were similar in infants of 23–24 weeks compared with those of 25–27 weeks (e.g., 112 vs. 105 days of assisted ventilation per additional survivor, or 167 vs. 180 days of assisted ventilation per additional quality-adjusted survivor, respectively). Conclusions Increased intensive care in the late 1990s for infants at the borderline of viability was associated with improved outcomes, at incremental costs that were not excessive compared with slightly more mature infants.