Clinical and laboratory predictors of survival for pediatric patients on non-postcardiotomy extracorporeal membrane oxygenation (ECMO).

ABSTRACT

Extracorporeal membrane oxygenation (ECMO) is used in pediatric patients with severe cardiopulmonary failure who do not respond to conventional therapy; only a few studies have been conducted in Turkey. We present the experience of pediatric ECMO with the aim of showing factors affecting mortality. We retrospectively reviewed our ECMO database to identify patients who received ECMO from October 2015 to March 2018. Our population comprised 30 pediatric patients. The mean patient age was 41.31±53.35 months and 17 (56.7%) patients were male. The median duration of ECMO support was 8.9 (6.6-10.8) days. The rates of successful ECMO weaning and survival to discharge were 70.0% (n=21) and 66.7% (n=20), respectively. Indications for ECMO were respiratory failure (40.0%), cardiac failure (33.3%), and sepsis (26.7%). We found that pre-cannulation values of pH (p=0.034), leukocytes (p=0.029), C-reactive protein (p=0.045), creatinine (p=0.047), chloride (p=0.001) and post-cannulation pH (p=0.0001), bicarbonate (p=0.014), lactate (p=0.002), chloride (p=0.0001) were associated with mortality. The results showed that preexisting sepsis and renal conditions contributed to poor outcomes. Indications, ECMO onset time, and pre- and post-cannulation laboratory values such as leukocytes, CRP, creatinine, bicarbonate, lactate, and chloride are factors that affect outcomes.