Predicting the probability of good neurological outcome after in-hospital cardiac arrest based on prearrest factors: validation of the good outcome following attempted resuscitation 2 (GO-FAR 2) score

Abstract

The Good Outcome Following Attempted Resuscitation (GO-FAR) 2 score is a prognostic tool developed to support decision-making for do-not-attempt-resuscitation (DNAR) orders by predicting neurological outcomes after in-hospital cardiac arrest (IHCA) based on prearrest variables. However, this scoring system requires further validation. We aimed to validate the GO-FAR 2 score for predicting good neurological outcome in Korean patients with IHCA. A single-centre registry of adult patients with IHCA from 2013 to 2017 was analysed. The primary outcome was discharge with good neurological outcome (Cerebral Performance Category score of 1 or 2). The patients were divided into four categories according to the GO-FAR 2 score: very poor (≥ 5), poor (2-4), average (- 3 to 1), and above-average (< - 3) likelihood of good neurological outcome. Of 1,011 patients (median age, 65 years), 63.1% were men. The rate of good neurological outcome was 16.0%. The proportions of patients categorised as having very poor, poor, average, and above-average probability of good neurological outcome were 3.9%, 18.3%, 70.2%, and 7.6%, respectively. In each category, good neurological outcome was observed in 0%, 1.1%, 16.8%, and 53.2%, respectively. Among patients in below-average categories (very poor + poor, GO-FAR 2 score ≥ 2), only 0.9% had good outcome. GO-FAR 2 score ≥ 2 showed a sensitivity of 98.8% and a negative predictive value of 99.1% in predicting good neurological outcome. The GO-FAR 2 score can predict neurological outcome after IHCA. In particular, GO-FAR 2 score ≥ 2 may support decision-making for DNAR orders.