External validation of SYNTAX score II in a real-world cohort undergoing coronary artery bypass grafting.

Abstract

Background: Our aim was to evaluate SYNTAX Score II for its ability to predict mortality in all-comers undergoing isolated coronary artery bypass grafting.

Methods: External validation of SYNTAX Score II was performed in a retrospective analysis of 2 tertiary care centers. Mortality at 4 years after surgery was defined as the primary outcome variable. External validation included assessment of calibration (calibration-in-the-large, observed-expected ratio, calibration slope) and discrimination (concordance statistic, Receiver Operating Characteristic curve). Additionally, SYNTAX Score II's performance was compared with the performance of EuroSCORE II, the logistic EuroSCORE, and ACEF Score.

Results: The study cohort included 1454 patients (Medical University of Vienna, n = 782; Medical University of Innsbruck, n = 672). Kaplan-Meier survival curves showed that tertiles of SYNTAX Score II were significantly associated with mortality (log-rank test, p < 0.001). In a stratified multivariable Cox proportional-hazards regression model, the following score components were independently associated with mortality: age (hazard ratio 1.03, 95% confidence interval 1.00 to 1.06), creatinine clearance (hazard ratio 0.98, 95% confidence interval 0.97 to 0.99), left-ventricular ejection fraction (hazard ratio 0.97, 95% confidence interval 0.95 to 0.98), and chronic obstructive pulmonary disease (hazard ratio 2.02, 95% confidence interval 1.34 to 3.05). The anatomical SYNTAX Score was not independently associated with mortality (hazard ratio 1.00, 95% confidence interval 0.98 to 1.02). Assessment of SYNTAX Score II calibration revealed an observed-expected ratio of 0.61 and a calibration slope of 0.62 (p < 0.001 for comparison with slope = 1.0), indicating general overestimation of 4-year mortality. The c-statistic amounted to 0.73. Performance of SYNTAX Score II was comparable with the performance of EuroSCORE II (c-statistic 0.73), the logistic EuroSCORE (c-statistic 0.74) and ACEF Score (c-statistic 0.72).

Conclusions: Our analysis shows that SYNTAX Score II has acceptable discriminative strength with respect to 4-year mortality in all-comers undergoing isolated coronary artery bypass grafting. Notably, mortality is over-estimated in patients with higher SYNTAX Score II values. SYNTAX Score II, EuroSCORE II, the logistic EuroSCORE, and ACEF Score offer comparable predictive value towards 4-year mortality.