Extra-corporeal life support following cardiac surgery in children: analysis of risk factors and survival in a single institution

Objective: Application of extra-corporeal life support (ECLS) following pediatric cardiac surgery varies between different institutions based on manpower availability and philosophy towards ECLS utilization. We examined a large single institution experience with postoperative ECLS in children aiming to identify outcome predictors. Methods: Hospital records of all children who required postoperative ECLS at our institution were reviewed. Patients' demographics, cardiac anatomy, surgical and ECLS support details were entered into a multivariable regression analysis to determine factors associated with survival. Results: Between 1990 and 2007, 180 consecutive children, median age 109 days (range: 1 day-16.9 years), required postoperative ECLS. Sixty-nine children (38%) had undergone palliative treatment for single ventricle pathology. ECLS support was required for failure to separate from cardiopulmonary bypass (n = 83) or for postoperative low cardiac output state (n = 97). Forty-eight patients (27%) received rescue extra-corporeal membrane oxygenation (ECMO) support during active chest compression for refractory cardiac arrest. Under ECLS support, 37 patients required surgical revision and 20 received orthotopic heart transplantation. One hundred and nine patients (61%) survived >24 h following ECLS discontinuation and 68 (38%) were discharged alive. Hospital survivors required shorter ECLS support duration compared to non-survivors (median 3 vs 5 days, respectively, p = 0.05) however survival occurred after up to 16 days of ECLS support. ECLS indication (OR: 0.85 for failure to separate from bypass vs postoperative tow cardiac output 95% CI (0.47-1.56), p = 0.62) and rescue ECMO (OR: 0.63 for rescue ECMO vs not 95%CI (0.32-1.24), p = 0.18) were not associated with risk of mortality. In a multivariable logistic regression model, neurological complications (p = 0.0007), prolonged ECLS duration (p = 0.003), repeat ECLS requirement (p = 0.02), renal dysfunction (p = 0.04) and not performing heart transplantation (p = 0.04) were significant factors for hospital death. Conclusion: ECLS plays a valuable role in children with tow cardiac output state following cardiac surgery. More than one third of those patients, including young neonates, older children, patients with single ventricle, or those requiring rescue ECMO can be salvaged. Although prognosis worsens with prolonged ECLS duration, survival can be noted up to 16 days of support. Heart transplantation is often an important ECLS exit strategy and should be considered early in selected children. Patients' survival could improve if renal and neurological complications are avoided. (C) 2009 European Association for Cardio-Thoracic Surgery. Published by Elsevier B.V. All rights reserved.