100 - Sensitivity, Specificity, and Predictive Values for Pulmonary Hypertension in Infants with Bronchopulmonary Dysplasia

Poster Number: 100
Publication Number: 100.2116

Background: The American Heart Association and Pediatric Pulmonary Hypertension Network recommend echocardiographic screening for pulmonary hypertension (PH) in infants with severe BPD (sBPD), as PH is present in ≈ 25% of infants with sBPD. However, the sensitivity, specificity, and predictive values of echocardiographic parameters at 36 weeks’ postmenstrual age (PMA) for discharge parameters or PH diagnoses is unclear.

Objective: To determine if single echocardiographic parameters at 36 weeks’ PMA are predictive of echocardiographic findings or PH at discharge in infants with sBPD.

Design/Methods: Infants with sBPD in our referral neonatal unit (NICU) between 2019-2022 were evaluated retrospectively for PH outcomes and for echocardiographic parameters at 36 weeks’ PMA (36-wk echo) and at the final study before NICU discharge (discharge echo). PH was defined by vasodilator use or clinical diagnosis in the medical record. Clinical variables and sBPD diagnoses were abstracted from the Children’s Hospitals Neonatal Consortium (CHNC) database. Welch two sample t-tests and Fisher’s exact tests evaluated differences in outcomes between infants with/without PH. Sensitivities, specificities, and predictive values between single 36-wk echo parameters and discharge echo or PH diagnosis were calculated.

Results: Of 73 infants who met criteria for sBPD, 22 (30%) had PH. A higher proportion of infants with PH died or were transferred to other in-hospital units, compared to infants without PH (Table 1). Infants with PH (n=51) had no differences in parameters on 36-wk echo, but on discharge echo, infants with PH had more right ventricular structure & function abnormalities, right-to-left or bidirectional shunting, and septal flattening (Table 2, p< 0.05 for all). Measures such as tricuspid regurgitant jet velocity were often not attainable. Single 36-wk echo parameters for the outcome of discharge echo showed high specificities and negative predictive values for certain parameters, and values for PH outcomes were similar (Table 3). Sensitivities and positive predictive values for single 36-wk echo parameters were poor for discharge echo parameters and for PH diagnoses.

Conclusion(s): Our data suggests that single 36-wk echo parameters in infants with sBPD have poor sensitivity and positive predictive value for echocardiographic abnormalities on discharge echo or PH diagnosis, and that estimates of systolic pulmonary artery pressure are often not available. Studies to evaluate composite echocardiographic measures and improved discernment of PH should be considered for infants at high-risk.