Termination of Resuscitation Rules for In-Hospital Cardiac Arrest.

Summary

Using national registries from Denmark (derivation) and Sweden/Norway (validation), five in-hospital termination-of-resuscitation (TOR) rules were developed. The best rule (unwitnessed, unmonitored, asystole, ≥10 min resuscitation) identified 11% of arrests for termination with a 0.6% false-positive rate for 30-day mortality across all cohorts.

Key Findings

• Among 53,864 rule combinations, five TOR rules met prespecified performance criteria and were clinically usable.
• The best rule (unwitnessed, unmonitored, asystole, resuscitation ≥10 minutes) achieved a positive rate of 11% and a false-positive rate of 0.6% for predicting 30-day mortality.
• Performance was consistent across external validation cohorts from Sweden and Norway, supporting generalizability.

Clinical Implications

Clinicians (including anesthesiologists/ICU teams) can consider implementing the best-performing TOR rule to guide real-time decisions, reduce futile resuscitation, and standardize practice while safeguarding against premature termination.

Limitations

• Observational registry data may be subject to unmeasured confounding and misclassification
• Ethical, legal, and institutional variations may affect implementation and clinician adherence

Future Directions

Prospective implementation studies to assess workflow integration, clinician adherence, patient/family communication, and outcomes; exploration of integration with real-time EHR/CPR dashboards.