Outcomes of cardiopulmonary resuscitation in nursing homes: can we predict who will benefit?

Abstract

Purpose: To determine the benefits of cardiopulmonary resuscitation (CPR) in nursing home patients and assess possible prearrest and arrest predictors of survival.

Patients and methods: During a 4-year period (1986 to 1989), consecutive nursing home patients from Milwaukee, Wisconsin, who sustained cardiac arrest and received CPR by paramedics were studied. The patients' prearrest clinical characteristics were determined including age, length of stay in nursing home, medical diagnoses, medications, circumstances surrounding the arrest, laboratory studies, and baseline functional status. Cardiac arrest data were obtained from a paramedic computer data base and included whether the arrest was witnessed, initial cardiac rhythm, and success of CPR. Survival was defined as the discharge of the patient alive from the hospital, and the patient's pre- and post-arrest functional status was compared. Possible predictors of survival were analyzed from the patient's prearrest characteristics and arrest characteristics.

Results: Of the total 196 patients who received CPR, 37 (19%) were successfully resuscitated and hospitalized, and 10 (5%) survived to be discharged. However, 27% of patients survived whose arrests were witnessed and who demonstrated ventricular fibrillation at the time of the arrest. In comparison, only 2.3% of all other nursing home patients who received CPR survived (p < 0.0002). Age, mental or functional status, hematocrit, renal dysfunction, pulmonary disease, cancer, and cardiovascular disease were not significant predictors of survival. At the time of hospital discharge, the functional status of the majority (80%) of the survivors was comparable to their prearrest status and 40% of the survivors lived for greater than 12 months.

Conclusion: We conclude that only a small percentage of nursing home patients who sustain cardiac arrest will benefit from CPR. However, greater than 25% of nursing home patients whose arrest is witnessed and who demonstrate ventricular fibrillation will survive. This is comparable to the survival rate of elderly community-dwelling persons who sustain cardiac arrest. Our data suggest that CPR should be initiated only in nursing home patients whose cardiac arrest is witnessed and should only be continued in patients whose initial documented cardiac rhythm is ventricular fibrillation or ventricular tachycardia.