Multicenter comparison of conventional venoarterial access versus venovenous double-lumen catheter access in newborn infants undergoing extracorporeal membrane oxygenation.

A multicenter trial was designed to compare standard venoarterial (VA) access versus single-catheter, venovenous access using the double-lumen catheter (VV-DLC) for newborns with respiratory failure undergoing extracorporeal membrane oxygenation (ECMO). Twenty-seven ECMO centers participated, each submitting data from the center's most recent VA cases, and data from VV-DLC cases completed upon entering the study. Data from 135 VA ECMO cases and 108 VV-DLC cases were submitted. All diagnoses resulting in neonatal respiratory failure were submitted, including patients with congenital diaphragmatic hernia (24 patients VA bypass, 11 patients VV-DLC bypass). Overall survival in patients undergoing VA bypass was 87%, while survival in patients undergoing VV-DLC bypass was 95%. Eleven patients required conversion from VV-DLC bypass to VA bypass because of insufficient support--10 of these patients survived (91% survival). Average bypass time for newborns undergoing VA bypass was 132 +/- 7.4 hours versus 100 +/- 5.1 hours for those undergoing VV-DLC bypass. Neurologic complications were more common in the VA bypass patients, although the VV patients were more stable. Hemorrhagic, cardiopulmonary, and mechanical complications, other than kinking of the DLC, occurred with approximately equal frequency in each group. In conclusion, in newborns with adequate cardiac function, venovenous ECMO using the DLC can provide the same level of support as conventional VA ECMO, without ligation of the carotid artery.