An Australian risk prediction model for 30-day mortality after isolated coronary artery bypass: the AusSCORE

Abstract

Objective: Our objective was to identify risk factors associated with 30-day mortality after isolated coronary artery bypass grafting in the Australian context and to develop a preoperative model for 30-day mortality risk prediction.

Summary background data: Preoperative risk associated with cardiac surgery can be ascertained through a variety of risk prediction models, none of which is specific to the Australian population. Recently, it was shown that the widely used EuroSCORE model validated poorly for an Australian cohort. Hence, a valid model is required to appropriately guide surgeons and patients in assessing preoperative risk.

Methods: Data from the Australasian Society of Cardiac and Thoracic Surgeons database project was used. All patients undergoing isolated coronary artery bypass grafting between July 2001 and June 2005 were included for analysis. The data were divided into creation and validation sets. The data in the creation set was used to develop the model and then the model was validated in the validation set. Preoperative variables with a P value of less than .25 in chi(2) analysis were entered into multiple logistic regression analysis to develop a preoperative predictive model. Bootstrap and backward elimination methods were used to identify variables that are truly independent predictors of mortality, and 6 candidate models were identified. The Akaike Information Criteria (AIC) and prediction mean square error were used to select the final model (AusSCORE) from this group of candidate models. The AusSCORE model was then validated by average receiver operating characteristic, the P value for the Hosmer-Lemeshow goodness-of-fit test, and prediction mean square error obtained from n-fold validation.

Results: Over the 4-year period, 11,823 patients underwent cardiac surgery, of whom 65.9% (7709) had isolated coronary bypass procedures. The 30-day mortality rate for this group was 1.74% (134/7709). Factors selected as independent predictors in the preoperative isolated coronary bypass AusSCORE model were as follows: age, New York Heart Association class, ejection fraction estimate, urgency of procedure, previous cardiac surgery, hypercholesterolemia (lipid-lowering treatment), peripheral vascular disease, and cardiogenic shock. The average area under the receiver operating characteristic was 0.834, the P value for the Hosmer-Lemeshow chi(2) test statistic was 0.2415, and the prediction mean square error was 0.01869.

Conclusion: We have developed a preoperative 30-day mortality risk prediction model for isolated coronary artery bypass grafting for the Australian cohort.