68 Predicting 3-Year Mortality After PCI: Revised Clinical Syntax Model Based on Patient Level Data from 7 Contemporary Stent Trials

<h3>Introduction</h3> The Logistic Clinical Syntax score, developed by combining clinical variables with the Syntax score, has been shown to perform better than the Syntax score alone in predicting 1-year outcomes after PCI. However, the ability of this score to predict long-term survival is unknown. We aimed to update this score to predict 3-year survival after PCI and compare its performance with the Syntax score. <h3>Methods</h3> Patient level data (n = 6,304) from seven contemporary PCI trials (SIRTAX, ARTS-II, STRATEGY, MULTI-STRATEGY, LEADERS, SYNTAX, and RESOLUTE) were analysed. We revised the overall risk and the predictor effects in the core model (Syntax score, age, creatinine clearance and left ventricular ejection fraction) using Cox regression analysis to predict mortality at 3-year. We also developed a core-plus model by combining core model with additional independent predictors of 3-year mortality i.e. diabetes mellitus, peripheral vascular disease and body mass index. <h3>Results</h3> There were 399 deaths over 3-year follow-up (cumulative risk 6.3%).The revised Clinical Syntax models showed better discriminative ability than the anatomical Syntax score for prediction of 3-year mortality after PCI (c-index: Syntax score 0.61; core model 0.71; and core-plus model 0.73). The core-plus model particularly performed better in differentiating low and intermediate risk groups.Cross validation confirmed the good agreement between the observed and predicted risks.The final model is presented as a score chart and the sum scores can be used to predict risk of mortality (Figure 1). <h3>Conclusion</h3> The Clinical Syntax plus score outperforms the Syntax score to predict 3-year mortality.