An epidemiologic approach to predictors of elective coronary artery bypass mortality in a non-university hospital population.
This investigation was undertaken to determine whether perioperative variables present in patients scheduled for elective coronary artery bypass grafting (CABG) surgery, and documented in earlier studies at university programs, bear any significant relationship to mortality occurring in a population of prospectively monitored non-university hospital patients. The investigation sought to identify the risk factors that might be present in these patients. Toward this purpose, data from 1,095 patients having CABG surgery at four major non-university Pittsburgh hospitals were analyzed using univariate and multivariate analysis to develop a model for predicting survival. From this model, three variables, unstable angina or recent myocardial infarction (< 6 weeks); evidence of chronic heart failure at the time of the preoperative examination; and greater than 65 years of age were found to have high statistical significance (P < 0.02) in predicting mortality in this non-university-hospital population. In summary, the study demonstrated that three variables, previously confirmed as being valuable predictors of risk in university hospital populations, also have significant predictive value in elective non-university CABG survival.