Congenital Diaphragmatic Hernia Requiring ECMO: Current Evidence, Controversies, and Unresolved Questions

Abstract

Background/Objectives: Congenital diaphragmatic hernia (CDH) requiring extracorporeal membrane oxygenation (ECMO) remains one of the most challenging conditions in neonatal critical care. Despite advances in prenatal assessment and neonatal management, substantial uncertainty persists regarding ECMO selection, mode and duration of support, and timing of diaphragmatic repair. This review summarizes the available evidence and critically examines the major unresolved questions in the management of infants with CDH requiring ECMO. Methods: A narrative review was conducted based on a literature search in PubMed/MEDLINE, Scopus, Web of Science, and the Cochrane Library for studies published between June 2005 and June 2026. Additional relevant publications were identified through reference screening, and selected foundational studies predating the search period were included when appropriate. Prospective and retrospective studies, registry-based analyses, systematic reviews, meta-analyses, and relevant consensus statements were considered. Evidence was narratively synthesized, with particular attention to disease severity, methodological limitations, and survivor bias. Results: Available evidence highlights substantial heterogeneity in disease severity and management. Prenatal and postnatal markers aid risk stratification, but no single parameter reliably predicts ECMO requirement or defines the optimal management strategy. To date, no clear survival advantage has been demonstrated for either veno-arterial or veno-venous ECMO, although veno-arterial ECMO remains the predominant mode of support in most centers; mode selection should therefore be individualized according to cardiopulmonary physiology, patient characteristics, anatomy, and institutional expertise. Surgical timing remains particularly controversial. While repair after ECMO decannulation has been associated with favorable outcomes, recent observational analyses suggest that early repair during ECMO may be beneficial in selected patients, although the optimal timing remains uncertain. Prolonged ECMO may provide additional time for pulmonary and cardiovascular recovery, although this potential benefit must be weighed against the cumulative morbidity of prolonged support. Conclusions: Management should be individualized according to disease severity, clinical trajectory, cardiopulmonary recovery, and institutional expertise. Current evidence does not support a universal strategy for ECMO initiation, mode, surgical timing, or duration. Prospective multicenter studies using standardized definitions and adequately accounting for disease severity are needed to reduce practice variation and better define optimal management.