315 - Decrease in Mortality or Severe Neuro-Morbidity in Extremely Preterm Infants by Each Day of Gestation

Poster Number: 315
Publication Number: 315.1180

Background: A non-linear association may exist between gestational age in days and mortality and severe neuro-morbidity risk; analyzing outcomes by gestational age in days could help guide perinatal counseling/decisions and inform prognosis.

Objective: To test the hypothesis that there are days during pregnancies at dates around extra-uterine viability when additional days of gestation are most associated with lower risk of mortality and severe neuro-morbidity.

Design/Methods: We included extremely preterm infants (21 0/7-26 6/7 weeks gestation) admitted to the University of Alabama at Birmingham neonatal intensive care unit from 1998-2022 with a determination of mortality by 36 weeks and/or severe neuro-morbidity. Infants with lethal syndromes or malformations were excluded. The primary outcome was mortality and/or severe neuro-morbidity defined as ventricular hemorrhage resulting in ventricular enlargement, intraparenchymal cerebral hemorrhage, and/or cystic periventricular leukomalacia on head ultrasounds by 1-month after birth. We conducted multivariable logistic regression to examine the association between gestational age in days (based on the best obstetrical estimate) and outcomes controlling for sex, multiple gestation, and antenatal steroid exposure. We computed the absolute change in outcome probabilities, reflecting the differences as gestational age advanced each day. All statistical analyses were performed using R (Version 4.3.1), with the level of significance set at 0.05.

Results: We included 2432 extremely preterm infants. The baseline characteristics (Table 1) varied as expected among groups. The predicted probabilities of mortality and severe neuro-morbidity, and mortality decreased as gestational age in days increased (Figure 1). The highest absolute change in mortality and morbidity probability was observed around the gestational age of 24 3/7 weeks (more than 2% lower mortality and/or severe neuro-morbidity per additional day of gestation, Figure 2-A), and the highest absolute change in mortality probability was observed around the gestational age of 23 4/7 weeks gestation (more than 2.5% lower mortality per additional day of gestation, Figure 2-B).

Conclusion(s): There is a non-linear association between gestational age in days with outcomes and a resultant critical period when each gestational day is associated with the largest decreases in the probability of mortality and neuro-morbidity in extremely preterm infants, underscoring the vital role of considering specific gestational days in guiding perinatal care decision-making and prognosis in pregnancies at the threshold of viability.