Early Cardiopulmonary Physiology, Patent Ductus Arteriosus Characteristics, and Outcomes at the Limits of Viability

To examine associations of center-specific, clinically assigned early clinical-hemodynamic patent ductus arteriosus (PDA) phenotypes and echocardiographic characteristics with in-hospital outcomes among infants born at 21-24 weeks' gestation.

Retrospective, single-center cohort of infants undergoing targeted neonatal echocardiography (TNE) within 36 postnatal hours. A retrospectively selected early TNE and contemporaneous clinical assessment classified infants as having no PDA, low-volume PDA, pharmacotherapy-recommended PDA, or ductal-supportive physiology. The selected examination was not necessarily the first and could incorporate clinical evolution and clinician response during the first 36 postnatal hours. Favorable outcome was survival to discharge without major prematurity-associated morbidity. Logistic and ordinal regression evaluated associations with outcome.

Among 189 infants, 88 (46.6%) had a favorable outcome. Outcomes differed across the 4 center-specific early clinical-hemodynamic phenotypes (p=0.003). Among infants with a PDA, unfavorable outcome occurred in 70.8% with ductal-supportive physiology vs 49.6% with low-volume or pharmacotherapy-recommended PDA (p=0.012). Unfavorable outcome occurred in 68.8%, 40.7%, and 70.3% of infants with indexed PDA diameter <1.5, 1.5-3, and >3 mm/kg, respectively (P<0.001). Each 1 mm/kg increase in indexed diameter was associated with lower odds of favorable outcome after adjustment (OR 0.76; 95% CI 0.59-0.98; P=0.034), with no diameter-by-ductal supportive interaction (P=0.945). In analyses excluding early deaths, both the ductal-supportive comparison (63.2% vs 45.6%; P=0.058) and the adjusted indexed-diameter association (OR 0.78; 95% CI 0.61-1.01; P=0.059) were weaker and no longer statistically significant.

Selected early clinical and echocardiographic findings were associated with in-hospital outcome, but the focused ductal-supportive comparison and adjusted indexed-diameter association were attenuated and no longer statistically significant after exclusion of early deaths. These center-specific, retrospectively derived findings should be considered exploratory risk markers requiring prospective validation; they do not establish temporality, causation, treatment responsiveness, or the effectiveness of a PDA-management strategy.