Objective: To identify factors associated with survival in patients with congenital diaphragmatic hernia (CDH) treated with extracorporeal membrane oxygenation (ECMO).
Methods: We retrospectively analyzed the data on 3100 patients with CDH in the Congenital Diaphragmatic Hernia Study Group from 82 participating pediatric surgical centers (1995-2004). Covariates considered included prenatal and perinatal clinical information, specifics of surgical repair, and the duration of extracorporeal support.
Result: Nine hundred seven patients from the registry were identified as having been both managed with ECMO and undergone attempted surgical repair. The survival rate for the entire Congenital Diaphragmatic Hernia Study Group registry was 67% and 61% for those receiving ECMO in whom repair was attempted (P < .001). Among ECMO-treated children, survivors had a greater estimated gestational age (38 +/- 2 vs 37 +/- 2 weeks; P < .01), greater birth weights (3.2 +/- 0.5 vs 2.9 +/- 0.5 kg; P < .001), were less often prenatally diagnosed (53% vs 63%; P < .01), and were on ECMO for a shorter period of time (9 +/- 5 vs 12 +/- 5 days; P < .001). In logistic regression models, therapy-related variables, including the duration of ECMO, the nature of diaphragmatic repair, and the type of abdominal closure and certain comorbidities, particularly the presence of a concomitant severe cardiac abnormality, were independently associated with outcome.
Conclusion: Our model identifies a group of pre-surgical and postsurgical parameters that predict survival rate in patients with CDH on ECMO support. This model was derived from the retrospective data from a large database and will need to be prospectively tested.