RESUMO
INTRODUCTION: The Charlson Comorbidity Index (CCI) is widely utilized for risk stratification by providers, payors, and administrative database researchers for non-cardiac surgical patients. CCI scores have not been validated in cardiac surgical patients. We hypothesize that the CCI will predict mid-term mortality and re-admissions, but performance may be different than purpose-built cardiac surgery risk calculators. METHODS: Patients undergoing isolated CABG between 2011 and 2017 were reviewed. Age-adjusted CCI scores were calculated based on clinical status at a time of index operation using prospectively captured data from institutional databases. Primary endpoint was 5-year mortality and 1-year re-admissions. The CCI, STS predicted mortality, and ASCERT 5-year mortality scores were compared in a sub-cohort of 500 patients. Patients underwent analysis using Cox Proportional Hazard ratios, Kaplan-Meier analysis, and ROC comparisons. RESULTS: Average CCI score for the overall population (n = 6064) was 3.40 ± 1.75. Kaplan-Meier analysis revealed significant difference in mortality stratified by CCI. Hazard ratio for 5-year mortality increased with each interval increase in CCI score value (HR 1.38 [1.33-1.43], P < 0.001), as did the risk of 1-year re-admission (HR 1.19 [1.15-1.22], P < 0.001). ROC curves for CCI, STS mortality, and ASCERT 5-year mortality risk demonstrate that all three scores are predictive at 5 y, but the ASCERT score performs best (ROC 0.76 versus 0.69, P = 0.004). CONCLUSIONS: The CCI can serve as a useful mid-term risk stratification tool in patients undergoing CABG when variables for the purpose-built STS and ASCERT scores are unavailable. However, the ASCERT score performs better at 5-year mortality calculation.