RESUMO
OBJECTIVES: Extracorporeal life support is a resource-intense treatment offered to the sickest patients. We aimed to investigate long-term survival rates and late deaths. DESIGN: Retrospective cohort study. SETTING: Tertiary referral center for extracorporeal life support. PATIENTS: All patients who required extracorporeal life support from 1992 to 2010 at our center. The U.K. National Health Service number was used to trace survival status of all patients who received extracorporeal life support at our center, grouped by diagnosis. Death more than 90 days after extracorporeal life support was defined as late, and these medical records were reviewed. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: A total of 741 children with 272 early deaths (36.7%) and 46 late deaths (6.2%) were included. Median follow-up time in survivors was 7.1 (interquartile range, 3.0-11.9) years. Five-year survival estimates were highest for meconium aspiration syndrome 88.0% (95% CI, 80.6-92.7%) and lowest for congenital heart disease 32.3% (95% CI, 25.1-39.8%). Five-year survival estimates conditional on being alive at 90 days were highest for meconium aspiration syndrome 97.9% (95% CI, 92.0-99.5%) and lowest for congenital diaphragmatic hernia 73.6% (52.3-86.5%). There was increased risk of late death in congenital diaphragmatic hernia, congenital heart disease, and acquired heart disease (p < 0.001, p < 0.01, p = 0.01) in comparison with the risk in meconium aspiration syndrome. For 46 late deaths, 17 had a cardiac cause, 16 had a respiratory cause, 10 had a comorbid cause, one died of sepsis, and in two, causation was unknown. CONCLUSIONS: Although the majority of deaths were early, late mortality was observed following extracorporeal life support. Late deaths were more prevalent in children with underlying complex long-term conditions, particularly heart disease and congenital diaphragmatic hernia. Evaluation of longer term survival is an important component of audit for extracorporeal life support outcomes.