(286) 1-Year neonatal survival in extremely preterm and extremely low birth weight infants

Objective:
Counseling and decision-making in the periviable period are often based on rates of adverse neonatal outcomes at a given gestational age (GA) and/or birthweight. Viability thresholds have been lowered but remain inconsistent between institutions. We aimed to characterize infant one-year mortality rates by GA and birthweight among extremely preterm and extremely low birth weight infants, to help provide counseling guidance.

Study Design:
We performed a retrospective cohort study of all infants born at 22 0/7-27 6/7 weeks GA and birthweight < 1000g who delivered at a single center from 1998 to 2023. We excluded infants who died within 12 hours of birth without active postnatal treatment to reduce bias in mortality rates from restriction of care. Infant one-year mortality rates were categorized by completed weeks of gestation and by birthweight into predefined 100-gram (g) ranges (Table). Receiver-operator curves (ROC) with associated area under the curve (AUC) were used to characterize ability to predict 1-year survival.

Results:
3,221 infants met study criteria and were analyzed. One infant resuscitated at ≤300g (n=1) did not survive. Infants born ≤500g had ≥50% chance of mortality at all extremely preterm GA categories; mortality decreased with increasing birthweight and gestational age. Among infants at 301-400g and 401-500g, the mortality rate was high across all GA (range 67-100% for 301-400g, 56-77% for 401-500g), with reduction by GA only noted in the 401-500g group. At 24 weeks and after, infants >500 grams had ≤50% mortality. Birthweight alone (AUC = 0.75) and GA alone (AUC = 0.72) had similar but only fair prediction for survival at 1 year, but when combined, prediction was excellent (AUC = 0.86). 

Conclusion:
Infants born ≤500g had ≥50% chance of mortality at all extremely preterm GA categories; whereas mortality decreased as both GA and birthweight increased among these infants. Neither birthweight nor GA alone predict 1-year survival; but both together resulted in excellent prediction and should be taken into consideration when counseling patients on obstetric and neonatal interventions.