For Out-of-Hospital Cardiac Arrest, Knowing When to Quit

TORONTO, Aug. 2 -- Most out-of-hospital cardiac arrests end on the scene, and investigators here have devised a simple clinical prediction rule to help emergency workers know when to stop.

Implementation of a clinical prediction rule based on three simple factors would reduce futile resuscitation efforts by almost 63%, reported the Termination of Resuscitation (TOR) trial investigators in the Aug. 3 issue of the New England Journal of Medicine.

"The prediction rule had a positive predictive value of 99.5% [for death] and a specificity of 90.2% [for death]," wrote Laurie J. Morrison, M.D., of the University of Toronto, and colleagues there and at Owen Sound (Ontario) Hospital.

The investigators analyzed data from 1,240 adult patients treated by 24 emergency medical systems in the province of Ontario from Jan. 1, 2002 to Jan. 30, 2004.

The predictive rule recommended termination of automated external defibrillator and basic life support efforts when there was no return of spontaneous circulation, no shocks were administered, and the arrest was not witnessed by emergency medical services personnel.

Using that predictive rule, 776 patients were recommended for termination of which four survived (0.05%), which was significantly lower (P=0.04) than the threshold of 1% that has been suggested as reflective of medical futility, they wrote. Three of the four patients survived with good cerebral performance and one had severe cerebral disability.

The prediction rule recommended transport and continued resuscitation for 464 patients, of whom 37 survived, resulting in a negative predictive value of 8.0% (95% CI, 6.6 to 9.7%).

The mean age of patients was 69.2 and 69% were men. Cardiac arrest was witnessed in 57.4% of cases, although EMS personnel witnessed less than 10% of cases. The median EMS response time was eight minutes.

There was no return of spontaneous circulation in 1,172 cases, and no shocks were delivered in 868 cases. Yet, all of these cases were transported for continued resuscitation. If the decision to transport was determined by the predictive rule, only 37.4% of cases would have been transported, the authors wrote.

A potential weakness of the study, wrote Gordon A. Ewy, M.D., of the University of Arizona Health Sciences Center in Tucson, in an accompanying editorial, was the fact that the EMS personnel followed CPR guidelines issued by the American Heart Association in 2000. Following those guidelines meant that only about half of needed chest compressions were performed "thus perfusing the heart and brain only half the time,"

The AHA issued new CPR and cardiovascular care guidelines in 2005. Dr. Ewy wrote that the newer techniques used now could affect the predictive power of the clinical rule.

The TOR trialists concede that better CPR is likely to result in an increased rate of spontaneous circulation and an increase in shockable rhythms, but that would only increase the number of patients recommended for continued resuscitation.

"Although such changes will alter the rate of transportation to the emergency department, the rule will continue to be helpful in identifying patients who are unlikely to survive despite optimized therapy," they wrote.

Finally Dr. Ewy cautioned that even the best clinical prediction rules will still require medical judgment to determine "when enough is enough."