External validation of the R-EDByUS score for neuroprognostication in out-of-hospital cardiac arrest

The R-EDByUS score predicts 1-month neurological outcomes for Out-of-hospital cardiac arrest (OHCA) patients, considering prehospital return of spontaneous circulation (ROSC) and ongoing cardiopulmonary resuscitation (CPR). We aimed to externally validate the R-EDByUS scoring system in Taiwanese patients who experienced OHCA. This multicenter retrospective cohort study was conducted at the National Taiwan University Hospital and its affiliated branches, including adult patients who experienced non-traumatic OHCA, from January 2016 to December 2023. We assessed the performance of the R-EDByUS scoring system in predicting unfavorable neurological outcomes (Cerebral Performance Category scores: 3–5) at hospital discharge. The area under the receiver operating characteristic curve (AUROC), sensitivity, specificity, and calibration curves were used to evaluate the model’s performance in the two cohorts. A total of 3,417 patients were included and divided into prehospital ROSC (n = 170) and ongoing CPR (n = 3,247) cohorts. In the prehospital ROSC cohort, the AUROC values were 0.79 (95% confidence interval (CI): 0.72–0.86) for the regression-based model and 0.73 (95% CI: 0.65–0.80) for the simplified model, indicating good predictive accuracy. In the ongoing CPR cohort, the AUROC values were 0.78 (95% CI: 0.74–0.81) and 0.77 (95% CI: 0.73–0.82), respectively. For ongoing CPR cohort, the calibration curve showed underestimation at low predicted probabilities but overestimation at high predicted probabilities. With only prehospital variables, the R-EDByUS scoring system showed good performance in Taiwanese OHCA patients for prehospital ROSC group. However, further studies are warranted before the score can be applied clinically in the ongoing CPR cohort.