91 - Impact of patent ductus arteriosus treatment on respiratory metrics in very low birth weight infants.

Poster Number: 91
Publication Number: 91.1794

Background: Hypoxemia and respiratory compromise occur in VLBW ( < 1500 grams) infants and may be associated with patent ductus arteriosus (PDA) shunting. We previously showed that a respiratory acuity score (RAS) and measures of hypoxemia are predictive of bronchopulmonary dysplasia but have not analyzed these in relation to PDA treatment.

Objective: Determine whether pharmacologic PDA treatment improves RAS and other measures of respiratory compromise.

Design/Methods: This is a retrospective analysis of VLBW infants in a level IV NICU from 2012-2022 with echocardiographic evidence of a PDA. We excluded those with major congenital cardiac disease or multiple anomalies. Infants were classified as never treated in the NICU (PDA-NT) or treated pharmacologically (PDA-T). RAS was calculated as mean FiO2 times a number representing level of respiratory support (PMID 30374050). Every-2-second SpO2 values, when available, were obtained from a vital sign archive and every-1-hour FiO2 from the medical record. We calculated five respiratory metrics at two 24-hour time periods: day 0 (day of diagnosis for PDA-NT and day of first treatment for PDA-T) and 3 days later. The five metrics are: mean FiO2, RAS, and SpO2, skewness of SpO2, and SpO2/FiO2 ratio. Baseline metrics were compared between groups by Mann Whitney test and before/after metrics by Wilcoxon signed rank test.

Results: In total, 134 infants were included: 67 PDA-NT (median GA 27.4 weeks) and 67 PDA-T (median GA 25.4 weeks, median age at treatment 13 days; 31 indomethacin, 6 ibuprofen, 30 acetaminophen). Of those, 88 infants had SpO2 data available for analysis (44 PDA-NT and 44 PDA-T). On the day of diagnosis, the PDA-T cohort had significantly worse FiO2, RAS, and SpO2 compared to PDA-NT (Table 1). For PDA-NT, skewness of SpO2 decreased from d0 to d3 (p < 0.05) with no significant change in the other metrics. For PDA-T, RAS improved from d0 to d3, though FiO2 and S:F worsened (p < 0.05 for all). There was no change in SpO2 or skewness of SpO2 for PDA-T (Table 2).

Conclusion(s): VLBW infants with worse respiratory metrics (FiO2, RAS, and SpO2) at the time of PDA diagnosis were more likely to receive pharmacologic treatment. Treatment was associated with improved RAS but slightly worsened FiO2 and S:F ratio. We speculate that improvement in RAS is a better measure of response to PDA treatment than change in oxygenation parameters.