Main Organiser

Julius Centre University of Malaya

Co-organiser

Department of Social and Preventive Medicine, Faculty of Medicine, University of Malay

Supported by

University of Malaya

IDENTIFYING HIGH RISK PATIENTS OF LONG-MORTALITY FOLLOWING CABG IN AUSTRALIA

Author

Billah Baki

Institution

Monash University

Abstract

Objectives: To develop a risk algorithm for identifying high risk patients for long-term mortality following isolated coronary artery bypass grafting (CABG) in Australian population.

Methods: Data were collected in Australia from 2001-2010. The Cox’s regression and bootstrapping were used to develop survival risk model which was validated using bootstrap discrimination (C- index) and calibration. An algorithm was derived to identify high risk patients and a survival calculator was developed.

Results: 21295 patients underwent isolated CABG from 2001-2010 with average follow-up of 3.6±2.5years. 16 variables including age, ejection fraction estimate, preoperative dialysis, new renal failure, peripheral vascular disease and stroke were identified as key risk factors. The bootstrap C-index (87.5%) and calibration of the model were very good. Additive score (AS) for each variable was calculated from beta coefficient and 3 risk categories were derived: low risk AS = 2, moderate risk AS 3-7 and high risk AS = 8. At the end of follow-up the mortality was 30.3% in high, 11.2% in moderate and 2.8% in low risk groups; the Kaplan- Meier survival was 45.9% in high, 73.0% in moderate and 92.0% in low risk groups. The mortality risk was 13 folds higher in the high risk group compared to low risk group.

Conclusion: The performance of the risk algorithm was excellent in identifying high risk patients. The model would be a guideline for surgeons, hospitals and patients to assess long-term survival and to determine post hospital-release care following isolated CABG.