In-hospital cardiopulmonary resuscitation: a 30-year review.

Abstract

We performed a meta-analysis to: (1) assess the disputed issue of in-hospital cardiopulmonary resuscitation (CPR) success rates among elderly patients, (2) investigate the possibility of a declining CPR success rate between 1960 and 1990, (3) provide an overview estimate of CPR effectiveness in specific patient groups, and (4) assess CPR risks. Ninety-eight reports providing in-hospital CPR survival-to-discharge rates were included in this overview. These reports were identified from MEDLINE searches, previous reviews, and reference citations. A pooled analysis revealed that 2994 (15 percent) of 19,955 patients were successfully resuscitated (survival to discharge). The rate of successful CPR has not changed in 30 years (r = -0.14, P > 0.05), but there has been a steady decline in the optimism regarding its value (r = -0.29, P < 0.01). Patients younger than 70 years of age had a success rate of 16.2 percent (odds ratio = 1.36; 95 percent confidence interval, 1.20 to 1.53) versus 12.4 percent for patients older than 70 years (P < 0.001). Community hospitals had a higher CPR success rate than teaching hospitals (18.5 percent versus 13.6 percent, P < 0.001). Although 72.9 percent of the post-CPR deaths were within 72 hours, prolonged in hospital survival in a vegetative state did occur; 1.6 percent of successfully resuscitated patients had a permanent neurological impairment. The increasing pessimism about the value of CPR, specifically, its futility in the elderly patient, is not supported by this review. The results of this meta-analysis should assist both the physician and the patient in determining the probable outcome of CPR.