Surgical Repair of Congenital Diaphragmatic Hernia After Extracorporeal Membrane Oxygenation Cannulation: Early Repair Improves Survival

Ann Surg. 2021 Jul 1;274(1):186-194. doi: 10.1097/SLA.0000000000003386.

Abstract

Objective: To determine the optimal timing of congenital diaphragmatic hernia (CDH) repair after extracorporeal membrane oxygenation (ECMO) cannulation.

Summary background data: The timing of CDH repair after ECMO cannulation remains a controversial topic due to studies with low power or strong selection bias.

Methods: This is a 2-aim retrospective cohort study based on the CDH Study Group registry for the period of 2007-2017. Aim 1-Compare On versus After ECMO repair. Aim 2-Compare Early versus Late repair on ECMO. In order to minimize selection bias and account for non-repairs, subjects in each aim were stratified into study groups based on their treatment center's characteristics. In each aim, the study groups were matched based on propensity score (PS). The main outcomes included mortality rate and incidence of non-repair.

Results: In aim 1, 136 patients remained in each group after PS matching. Compared to the After ECMO group, patients in the On ECMO group demonstrated a lower mortality rate, hazard ratio (HR) 0.54 (0.38, 0.77) (P < 0.001), and lower incidence of non-repair, 5.9% versus 33.8% (P < 0.001). In aim 2, 77 patients remained in each group after PS matching. Compared to the Late group, Early repair of CDH on ECMO was associated with a lower mortality rate, HR 0.51 (0.33, 0.77) (P = 0.002), and lower incidence of non-repair, 9.1% versus 44.2% (P < 0.001).

Conclusions: The approach of early repair after ECMO cannulation is associated with improved survival compared to delayed surgical correction.

Publication types

  • Research Support, N.I.H., Extramural

MeSH terms

  • Extracorporeal Membrane Oxygenation*
  • Female
  • Hernias, Diaphragmatic, Congenital / surgery*
  • Herniorrhaphy*
  • Humans
  • Infant, Newborn
  • Male
  • Propensity Score
  • Proportional Hazards Models
  • Retrospective Studies
  • Survival Analysis
  • Time-to-Treatment*