A risk model derived from the National Registry of Myocardial Infarction 2 database for predicting mortality after coronary artery bypass grafting during acute myocardial infarction

Abstract

The mortality risk associated with coronary artery bypass grafting (CABG) after acute myocardial infarction (AMI) remains controversial. Although elective CABG is quite safe, the effects of recent myocardial infarction, gender, and other clinical factors on perioperative mortality rates are not completely understood. The objective of this study was to determine in-hospital mortality rates for patients with AMI receiving CABG and to generate a model to predict the risk for any individual patient with specific risk factors. Using the National Registry of Myocardial Infarction 2 database, we identified 71,774 subjects (21,270 women) with AMI who underwent CABG; we excluded those subjects who received immediate surgery as reperfusion therapy. Multivariate logistic regression was used to quantify the independent effects of age, recent myocardial infarction, gender, and other covariates on mortality. A risk score was then generated from the regression model to quantify the mortality risk. The results of logistic regression modeling determined that age was an independent predictor of in-hospital death (adjusted odds ratio [OR] 3.05, 95% confidence interval [CI] 2.76 to 3.37 for age >75), as were previous CABG (OR 2.84, 95% CI 2.55 to 3.16), heart failure on presentation (OR 1.73, 95% CI 1.57 to 1.91 for Killip class II), and female gender (OR 1.58, 95% CI 1.45 to 1.71). The mortality risk score showed that 55% of patients had risk scores of 2 to 5 and mortality rates of 4% to 13%. This moderate risk group experienced 76% of the total predicted mortality. Thus, in-hospital CABG mortality rates after AMI are high compared with elective surgery. Using the described risk score, clinicians can quantify the impact of patient risk factors in making decisions about referral for and timing of CABG.