Objective: To compare the clinical characteristics and survival of elderly and younger hospitalized patients who sustain cardiac arrest and receive cardiopulmonary resuscitation (CPR) in the 1990's and to assess predictors of survival.
Design: Retrospective survey of cardiac arrest database and hospital charts, plus telephone follow-up.
Setting: 450-bed acute care teaching hospital.
Study population: Seventy-eight hospitalized patients 70 years or older and 73 hospitalized patients under 70 years of age.
Measurements: Survey of cardiac arrest data base, hospital charts, and telephonic follow-up to allow (1) comparison of clinical characteristics, survival, and long-term follow-up between two age groups and (2) univariate and multivariate analysis of predictors of mortality.
Main results: Pre-arrest clinical characteristics were not significantly different between the age groups. Prior to arrest the majority of patients were functionally active, and over one-third were hospitalized for acute coronary artery syndromes. In approximately 85% of the patients, the arrest was witnessed, and 70% of the patients had their cardiac rhythm monitored at onset of the arrest. Survival was not significantly different between the age groups; 26% of the total 151 patients were discharged. No significant difference was noted in pre-post arrest functional status of survivors. Survival at 1, 2, and 3 years in elderly and younger survivors was 86% versus 80%, 76% versus 67%, and 71% versus 61%, respectively. Multivariate analysis identified the presence of coronary artery disease, admission systolic blood pressure, and functional level to be independent pre-arrest predictors of mortality. At the time of the arrest, the initial cardiac rhythm and duration of CPR were found to be independent predictors of mortality.
Conclusions: Elderly patients hospitalized in the 1990's who receive CPR have outcomes similar to younger patients who receive CPR. The favorable outcome in the elderly patients may reflect patient selection: the majority of our patients were functionally active prior to hospitalization, without multiple serious illnesses; many were hospitalized for acute coronary artery syndromes; and, in most cases, the arrest was witnessed with the patient's cardiac rhythm monitored at onset of the arrest.