Selective Application of Cardiopulmonary Resuscitation Improves Survival Rates

This study is a retrospective review of all patients who died without cardiopulmonary resuscitation (CPR) or who sustained a sudden cardiopulmonary arrest in the hospital and received CPR during a 2-yr period at a large medical center. Based on a review of Current Procedural Terminology codes, patients were classified into one of the ten disease categoriesmultiple medical problems, acute disease, procedure-related, congenital disease, neoplasm, metastatic neoplasm, trauma, burn, acquired immunodeficiency syndrome, and dementia. A total of 1206 patient deaths without a CPR effort were identified. CPR was administered to another 550 patients who had a sudden cardiopulmonary arrest, of which 71 % survived the resuscitative attempt initially, but only 25% survived CPR until discharge from the hospital. CPR was applied less frequently than the mean in the metastatic neoplasm (P < 0.0001), trauma (P = 0.013), and dementia (P = 0.0003) groups and more frequently in the acute disease (P < 0.0001) and procedure-related (P < 0.0001) groups. Survival to discharge from the hospital was more frequent than the mean in the congenital disease group (P = 0.0004) and less frequent in the neoplasm group (P = 0.0425). The other groups had survival rates comparable to the mean. Patients 70 yr of age and older were less likely to receive CPR than those younger than 70 (P < 0.0001). However, if they did receive CPR, they were just as likely to survive to discharge from the hospital as the younger patients (P = 0.3404). In conclusion, the high survival rate demonstrated in most diagnostic groups reflects prior decisions to withhold CPR in many patients in whom it would be ineffective. Additionally, CPR in selected patients with multiple medical problems and in patients of advanced age was associated with reasonable chance of survival to discharge.