Morbidity of conversion from venovenous to venoarterial ECMO in neonates with meconium aspiration or persistent pulmonary hypertension

Background Outcomes in neonates receiving extracorporeal membrane oxygenation (ECMO) for meconium aspiration syndrome (MAS) and/or persistent pulmonary hypertension (PPHN) are favorable. Infants with preserved perfusion are often offered venovenous (VV) support to spare morbidities of venoarterial (VA) ECMO. Worsening perfusion or circuit complications can prompt conversion from VV-to-VA support. We examined whether outcomes in infants requiring VA ECMO for MAS/PPHN differed if they underwent VA support initially versus converting to VA after a VV trial, and what factors predicted conversion. Methods We reviewed the Extracorporeal Life Support Organization registry from 2007 to 2017 for neonates with primary diagnoses of MAS/PPHN. Propensity score analysis matched VA single-runs (controls) 4:1 against VV-to-VA conversions based on age, pre-ECMO pH, and precannulation arrests. Primary outcomes were complications and survival. Data were analyzed using Mann–Whitney U and Fisher's exact testing. Multivariate regression identified independent predictors of conversion for VV patients. Results 3831 neonates underwent ECMO for MAS/PPHN, including 2129 (55%) initially requiring VA support. Of 1702 patients placed on VV ECMO, 98 (5.8%) required VV-to-VA conversion. Compared with 364 propensity-matched isolated VA controls, conversion runs were longer (190 vs. 127 h, P < 0.001), were associated with more complications, and decreased survival to discharge (70% vs. 83%, P = 0.01). On multivariate regression, conversion was more likely if neonates on VV ECMO did not receive surfactant (OR = 1.7;95%CI = 1.1–2.7;P = 0.03) or required high-frequency ventilation (OR = 1.9;95%CI = 1.2–3.3;P = 0.01) before ECMO. Conclusion Conversion from VV-to-VA ECMO in infants with MAS/PPHN conveys increased morbidity and mortality compared to similar patients placed initially onto VA ECMO. VV patients not receiving surfactant or requiring high-frequency ventilation before cannulation may have increased risk of conversion. While conversions remain rare, decisions to offer VV ECMO for MAS/PPHN must be informed by inferior outcomes observed should conversion be required. Level of evidence Level of evidence 3Retrospective comparative study.