Internal Medicine · Neurology · 1 h ago
2025 ERC/ESICM Algorithm Shows High Specificity After Cardiac Arrest
A retrospective study of 271 comatose out-of-hospital cardiac arrest survivors assessed the 2025 ERC/ESICM neuroprognostication algorithm using a restricted set of predictors. The algorithm had 97.1% specificity for poor six-month functional outcome, while favorable criteria reduced the proportion of patients left unclassified.
- Algorithm specificity for poor outcome was 97.1%.
- Favorable criteria reduced indeterminate classifications to 28.5%.
- Negative predictive value was 50.0%.
- Restricted predictors may have limited sensitivity.
This retrospective observational study evaluated the 2025 European Resuscitation Council/European Society of Intensive Care Medicine neuroprognostication algorithm in adults who remained comatose (Glasgow Coma Scale motor score <6) 72 hours after out-of-hospital cardiac arrest and targeted temperature management. Of 400 survivors, 271 met inclusion criteria; 233 (86.0%) had poor functional outcomes at six months. Withdrawal of life-sustaining therapy occurred in 5.5% of the cohort.
Using four unfavorable and two favorable predictors, the algorithm identified poor outcomes with 76.2% sensitivity and 97.1% specificity. Positive predictive value was 99.1%, while negative predictive value was 50.0%. Adding favorable criteria in the second step reclassified 68 of 139 patients who remained unclassified after the first step, reducing the final indeterminate proportion from 55.8% to 28.5%. Six of 22 patients with conflicting favorable and unfavorable predictors had good outcomes.
The findings support high specificity in this cohort, where withdrawal of life-sustaining therapy was uncommon, and show that the sequential approach can reduce unclassified results. However, the restricted predictor set may have reduced sensitivity, and the low negative predictive value and persistent indeterminate cases limit interpretation. Results apply to the selected population and should not be treated as standalone prognoses.
In practice
For adults who remain comatose 72 hours after out-of-hospital cardiac arrest, this cohort suggests the 2025 ERC/ESICM algorithm, applied with the study’s restricted predictor set, can identify poor six-month outcomes with high specificity. The second step using favorable criteria reduced the proportion left unclassified, but the negative predictive value was only 50%, and some patients with discordant predictors had good outcomes. These retrospective results support cautious interpretation rather than changing practice on the basis of a single predictor or this validation alone.
AI-generated clinical interpretation · Not editor-reviewed
Is this summary clinically accurate?
Help fellow clinicians: your rating sends inaccurate summaries straight to our editors.
Sign in to rate this summary →Source
Critical Care: Validating the 2025 ERC/ESICM neuroprognostication algorithm using a restricted predictor set in a minimal WLST cohort ↗This is an automated AI-condensed summary that has not yet been reviewed by an editor. Always consult the full item at the original source.