Outcome of patients with congenital diaphragmatic hernia requiring ECMO: can we predict?

Introduction

In spite of the advances in pre-neonatal and neonatal intensive care including the use of antenatal steroids, foetal surgery, nitric oxide (NO), surfactant, high-frequency oscillation (HFOV), liquid ventilation and extracorporeal membrane oxygenation (ECMO), the mortality of patients with congenital diagrammatic hernia (CDH) remains high. The use of ECMO is generally reserved for patients who cannot be managed by other treatments such as HFOV, NO and surfactant. The aim of this study was to identify factors that could predict the outcome in patients with CDH requiring ECMO.

Methods

This study is a retrospective review of case records, microfilms and our database between September 1991 and December 2004. All patients with CDH managed in our unit using ECMO were included. 'Pre ECMO' variables such as gestational age, sex of the patient, birth weight, age at the time of ECMO cannulation, acid-base status and blood gasses and ventilator settings before commencing ECMO, and variables during the ECMO course such as the mode of ECMO (VA vs VV), use of NO, surfactant, liquid ventilation, vaso-active agents, timing of repair of CDH, use of blood products, and complications (mechanical and patient related) on ECMO were analysed to identify predictors of outcome.

Results

A total of 52 patients were managed using ECMO in our unit during the study period. All these patients were referred to our unit following the failure of maximal conventional treatment. The mean gestational age at birth was 38.9 (SD 2.74) weeks and the birth weight was 2.92 (SD 0.44) kg. The APGARS scores were 4.35 (SD 2.9) at 1 min and 5.93 (SD 2.9) at 5 min of birth. The mean age at the time of referral for ECMO was 2 (SD 3.1) days. The mean PaO₂/FiO₂ ratio was 49.7 (SD 49.8). The overall survival to hospital discharge was 58%. The mode of ECMO was venovenous in 27 patients and venoarterial in 25 patients. On univariate analysis, the mean duration of ECMO (181 ± 120 vs 317 ± 156 hours; P = 0.001), use of NO (6 vs 10; P = 0.049) and renal complications (4 vs 14; P < 0.001) were significantly different between the survivors and nonsurvivors. The pre ECMO variables were comparable between the survivors and non-survivors. Multiple logistic regression analysis of the variables included in the analysis revealed the duration of ECMO and renal complications developing on ECMO to be independently associated with mortality.

Conclusion

Prolonged duration of ECMO and renal complications on ECMO were independently associated with mortality. None of the pre ECMO variables could be identified as predictors of mortality in patients with CDH requiring ECMO.