Effect of race on survival following in-hospital cardiopulmonary resuscitation.

BACKGROUND

Race has been shown to be a significant predictive factor in a number of treatment decisions and outcomes, including survival following out-of-hospital cardiopulmonary resuscitation (CPR). The goal of this study was to determine whether race is associated with the rate of survival to discharge following in-hospital CPR.

METHODS

Consecutive adult patients undergoing attempted CPR at three teaching hospitals were identified. Demographic, clinical, and laboratory data from the time of admission, information about the resuscitation attempt, and the outcome of CPR were recorded for each patient. The characteristics of black and non-black patients were compared. Logistic regression was used to determine whether race was a significant independent predictor of CPR outcome.

RESULTS

A total of 656 patients were identified. Black patients had a higher mean severity of illness as measured by the Acute Physiology and Chronic Health Evaluation (APACHE) III score, were more likely to have an initial rhythm of electromechanical dissociation or asystole, were less likely to have an admitting diagnosis of myocardial infarction or a history of coronary artery disease, and had a higher serum creatinine level, lower serum albumin value, and lower 24-hour urine output for the first 24 hours. There was no difference between black and nonblack patients regarding the rate of survival of the resuscitative effort itself. However, black patients were significantly less likely than nonblack patients to survive to discharge following resuscitation (Mantel-Haenszel odds ratio, 0.31; 95% confidence interval, 0.15 to 0.68). This relationship persisted after adjusting for potential confounders such as age, sex, initial cardiac rhythm, diagnosis of pneumonia, serum creatinine level, hospital, and APACHE III score.

CONCLUSIONS

Black race is significantly associated with a lower rate of survival to discharge following in-hospital CPR. Further work is needed to explore this association in other settings; to examine the effect of other possible confounding variables, such as tobacco use, socioeconomic status, and marital status; and to further study the determinants of physician decision-making about resuscitation.