Abstract Body: Background: Out-of-hospital cardiac arrest (OHCA) survivors who regain spontaneous circulation may progress to brain death, but early prediction remains challenging. Accurate prognostication can inform treatment decisions and facilitate organ donation, yet robust models are lacking.
Objectives: To identify predictors of brain death progression in OHCA patients treated with targeted temperature management (TTM) and to develop and validate an early prediction model, also assessing subsequent organ donation rates.
Methods: We retrospectively analyzed data from the Korean Hypothermia Network registry, including OHCA patients treated with TTM between October 2015 and September 2024. Patients with acute brain hemorrhage or stroke, poor pre-arrest cerebral status (CPC 3 or 4), terminal illness, severe hypothermia (<30°C), or those not receiving post-arrest care were excluded. The development cohort (2015–2020) included 1,966 patients; the validation cohort (2020–2024) included 927 patients.
Results: Of 23,160 registered patients, 3,067 underwent TTM with consent; 2,893 with complete data were analyzed. In the development cohort, 273 patients (13.9%) were suspected of brain death. Multivariate analysis identified age, witnessed arrest, respiratory cause, heart rate and GCS motor score after ROSC, NSE at 72 hours, and post-rewarming fever as significant predictors. The model achieved an AUC of 0.860 (95% CI: 0.825–0.892), sensitivity 89.6%, and specificity 72.1% in the development cohort; in the validation cohort, AUC was 0.893 (95% CI: 0.836–0.918), sensitivity 89.4%, specificity 83.4%. Among 273 suspected brain death patients, only 113 (41.4%) proceeded to organ donation, indicating that about half of potential donors did not become actual donors. Six-month survival and good neurological outcome rates were 41.7% and 31.7% in the development cohort.
Conclusions: We developed and validated a robust model for early prediction of brain death in OHCA survivors treated with TTM. Despite accurate identification, only about half of brain death cases led to organ donation, highlighting a gap between potential and actual donors. This model can support clinical decision-making and may help increase organ donation rates through timely identification.