Abstract
OBJECTIVE: To investigate the association between the secular decrease in treatment of patent ductus arteriosus (PDA) and trends in neonatal mortality and morbidity in infants born at 26 0/7-28 6/7 weeks' gestation. STUDY DESIGN: A retrospective cohort study including infants born between 2012 and 2021 in continually participating hospitals in the National Institute of Child Health and Human Development Neonatal Research Network. The primary composite outcome was defined as surgical necrotizing enterocolitis, grade 2-3 bronchopulmonary dysplasia (BPD), severe intraventricular hemorrhage, or death. Relationships of temporal trends in PDA treatment with the primary composite outcome and its components were analyzed using a multilevel model accounting for patient-level factors. A separate analysis assessed these relationships stratified by hospital changes in PDA treatment. RESULTS: The study included 7864 infants. There was a decrease in any PDA treatment from 21% to 16% (P < .01) and an increase in the primary composite outcome from 24% to 36% (P < .01). Change in the primary outcome was driven by increased grade 2-3 BPD (13%-26%, P < .01), with grade 2 BPD accounting for most of this increase (10%-22%, P < .01). Temporal decreases in PDA treatment were associated with increases in the primary outcome and grade 2-3 BPD after adjusting for patient-level factors (P < .01). However, stratified analyses showed that grade 2-3 BPD increased in all hospital groups, regardless of changes in PDA management. CONCLUSIONS: From 2012 to 2021, temporal decreases in PDA treatment for infants 26-28 weeks were associated with an increase in grade 2-3 BPD. However, caution is warranted in determining causality. Reasons for increased grade 2-3 BPD during the past decade warrant investigation. TRIAL REGISTRATION: Generic Database: NCT00063063.