PC1085: Use of Demographic, Biochemical, and Echocardiographic Data to Help Optimize Screening of Bronchopulmonary Dysplasia

Background: With improvement in extreme premature infant survival rates, secondary to improved ventilation strategies, there has been a rise in the development of bronchopulmonary dysplasia (BPD). Current recommendations include screening infants with BPD to evaluate for BPD-associated pulmonary hypertension (BPD-PH) so that prompt medical management can occur. In addition, BPD-PH can develop at any time and thus, those who do not have pulmonary hypertension, but are at risk for it, require ongoing surveillance. Per guideline recommendations, infants at our institution are screened for the presence of pumonary hypertension if diagnosed with BPD at 36 weeks post-menstrual age (PMA). Whether incorporating demographic, clinical and laboratory data can help predict which infants are most appropriate for BPD-PH screening remains unclear.

Methods: Retrospective cohort study of preterm infants with BPD who underwent echocardiographic screening at 36 weeks PMA, discharged from the University of Virginia neonatal intensive care unit (NICU) between 01/2016 and 12/2020. Infants were divided into 2 groups, those with normal echocardiographic findings and those with hemodynamically significant findings, including right ventricular (RV) dysfunction, RV dilation, atrial septal defect, patent ductus arteriosus, and pulmonary hypertension. We compared demographic data, NICU course, biochemical results, and 36 weeks PMA echocardiogram results between the two cohorts.

Results: 215 infants with BPD were enrolled in this study (mean gestational age (GA) 26.7 ± 2.7 weeks, mean birth weight (BW) 940 ± 270 g, 39% male, 61% female). Cohort demographics are shown in table 1. There were no statistically significant differences with regards to GA, BW, PCO2, serum BNP, weight, or duration of mechanical ventilation at time of echocardiogram (Table1). Infants with BPD-PH had longer lengths of stay (p=0.01) and were discharged at later PMA (42.1 ± 6.42 vs. 44.8 ± 8.40, p=0.01). In this cohort the presence of BPD-PH was low (1% overall). The most common echocardiogram findings were the presence of an ASD (n=40, 19%) and PDA (n=46, 21%) (Table 2). In addition, common NICU comorbidities or the use of pulmonary medications were not different between the 2 groups (Table 2).

Conclusion: These results show a low overall incidence of BPD-PH in a large cohort of infants with BPD. Clinical and laboratory data did not discriminate between those with normal echocardiogram versus those with abnormal echocardiogram. As could be expected, infants with hemodynamically significant findings on echocardiogram are more likely to require longer hospitalization. Further research is needed to identify if other associations with BPD are relevant and should be factored into screening protocols.