Trends in Neonatal Extracorporeal Membrane Oxygenation During a Venovenous Cannula Shortage

• Published in: Pediatric critical care medicine Vol. 24; no. 3; p. 245
• Main Authors: Hatch, L Dupree, Bridges, Brian C, Chapman, Rachel L, Danko, Melissa E, Schumacher, Robert E, Patrick, Stephen W
• Format: Journal Article
• Language: English
• Published: United States 01.03.2023
• Subjects: Cannula; Extracorporeal Membrane Oxygenation - adverse effects; Female; Humans; Infant; Infant, Newborn; Infant, Newborn, Diseases; Meconium Aspiration Syndrome - etiology; Meconium Aspiration Syndrome - therapy; Respiratory Insufficiency; Retrospective Studies
• ISSN: 1529-7535
• DOI: 10.1097/PCC.0000000000003126

To report temporal trends in venovenous extracorporeal membrane oxygenation (ECMO) use for neonatal respiratory failure in U.S. centers before and after functional venovenous cannula shortage due to withdrawal of one dual lumen venovenous cannula from the market in 2018. Retrospective cohort study. ECMO registry of the Extracorporeal Life Support Organization. Infants who received neonatal (cannulated prior to 29 d of age) respiratory ECMO at a U.S. center and had a record available in the ECMO registry from January 1, 2010 to July 20, 2021. None. Primary outcome was receipt of venovenous ECMO (vs venoarterial or other), and secondary outcomes were survival to hospital discharge and adverse neurologic outcomes. Using an interrupted time series design, we fit multivariable mixed effects logistic regression models with receipt of venovenous ECMO as the dependent variable, treatment year modeled as a piecewise linear variable using three linear splines (pre shortage: 2010-2014, 2014-2018; shortage: 2018-2021), and adjusted for center clustering and multiple covariates. We evaluated trends in venovenous ECMO use by primary diagnosis including congenital diaphragmatic hernia, meconium aspiration, pulmonary hypertension, and other. Annual neonatal venovenous ECMO rates decreased after 2018: from 2010 to 2014, adjusted odds ratio (aOR) for yearly trend 0.98 (95% CI 0.92-1.04), from 2014 to 2018, aOR for yearly trend 0.90 (95% CI 0.80-1.01), and after 2018, aOR for yearly trend 0.46 (95% CI 0.37-0.57). We identified decreased venovenous ECMO use after 2018 in all diagnoses evaluated, and we failed to identify differences in temporal trends between diagnoses. Survival and adverse neurologic outcomes were unchanged across the study periods. Venovenous ECMO for neonatal respiratory failure in U.S. centers decreased after 2018 even after accounting for temporal trends, coincident with withdrawal of one of two venovenous cannulas from the market.