Validation of a clinical decision aid to discontinue in-hospital cardiac arrest resuscitations

Area of Science:

• Emergency Medicine
• Cardiology
• Clinical Decision Support

Background:

• In-hospital cardiac resuscitation survival rates remain low.
• A prior study developed a clinical decision aid to predict survival to discharge.
• This study aimed to validate that previously derived decision aid.

Purpose of the Study:

• To validate a previously developed clinical decision aid for predicting survival to hospital discharge after in-hospital cardiac resuscitation.
• To assess the accuracy of the decision aid in identifying patients unlikely to benefit from continued resuscitation.

Main Methods:

• Analysis of a large registry of in-hospital cardiac resuscitation attempts (n=2181) from 1987-1996.
• The decision aid predicted survival based on witnessed arrest, initial cardiac rhythm (ventricular tachycardia/fibrillation), or pulse regained within 10 minutes.
• Predictions were compared against actual hospital discharge outcomes.

Main Results:

• The decision aid correctly predicted survival to discharge for 99.1% of survivors (sensitivity).
• For patients predicted to have no chance of survival (n=269), only 1.1% were discharged, and none independently.
• The negative predictive value of the aid was 98.9%.

Conclusions:

• The validated clinical decision aid accurately identifies patients unlikely to survive in-hospital cardiac resuscitation.
• This tool can assist physicians in making informed decisions about continuing resuscitative efforts.
• The aid supports better resource allocation by identifying patients with extremely low likelihood of beneficial outcomes.