Original Articles

A Simple Scoring System to Predict Survival after Venoarterial Extracorporeal Membrane Oxygenation

Address correspondence to: Berhane Worku, MD, Department of Cardiothoracic Surgery, New York Presbyterian-Brooklyn Methodist Hospital, 263 7th Avenue, Suite 4H, Brooklyn, NY 11215. E-mail: bwm2002@med.cornell.edu

Received: 24 April 2019
Accepted: 2 August 2019

Abstract

Patients undergoing consideration for venoarterial extracorporeal membrane oxygenation (VA ECMO) require an immediate risk profile assessment in the setting of incomplete or no information. A retrospective cohort study of 100 patients undergoing VA ECMO placement at three institutions was carried out. Variables strongly associated with survival to discharge were used to calculate a risk stratification score. Indications for VA ECMO support included postcardiotomy shock (24%), ischemic etiologies (33%), nonischemic cardiomyopathy (32%), and other etiologies (11%). Pre–VA ECMO arrest occurred in 69%, and 30% of patients underwent cannulation during arrest. Survival to discharge was 38%. Three variables demonstrated a strong trend toward predicting survival to discharge: lactate >10 mmol/L (p = .054), albumin <3 g/dL (p = .062), and platelet count <180 K/uL (p = .064), and these variables were included in a scoring system. The extremes of age and duration of pre–VA ECMO ventilation were associated with a dismal prognosis and were also included. These five variables were used to construct a mortality prediction score. A score of 0 was associated with 10% expected mortality, whereas a score of 4+ was associated with 100% expected mortality. Mortality increased in a stepwise fashion with increasing scores. The expected mortality closely paralleled the observed mortality. A simple scoring system composed of easily collected variables may help predict mortality. However, it is not intended to replace an experienced clinician’s judgment, but to enhance it.