Long-term prognostic value of risk scores after drug-eluting stent implantation for unprotected left main coronary artery: A pooled analysis of the ISAR-LEFT-MAIN and ISAR-LEFT-MAIN 2 randomized clinical trials

Objectives: To evaluate the long-term prognostic value of risk scores in the setting of drug-eluting stent (DES) implantation for uLMCA. Background: Data on the prognostic value of novel risk scores developed to select the most appropriate revascularization strategy in patients undergoing DES implantation for uLMCA disease are relatively limited. Methods: The study represents a patient-level pooled analysis of the ISAR-LEFT-MAIN (607 patients randomized to paclitaxel-eluting or sirolimus-eluting stents) and the ISAR-LEFT-MAIN-2 (650 patients randomized to everolimus-eluting or zotarolimus-eluting stents) randomized trials. The Syntax Score (SxScore) as well the Syntax Score II (SS-II), the EuroSCORE and the Global Risk Classification (GRC) were calculated. The primary outcome was all-cause mortality. Results: At a mean follow-up of 3 years there were 160 deaths (12.7%). The death-incidence was significantly higher in the upper tertiles than in the intermediate or lower ones for all risk scores (log-rank test P < 0.01 for all comparisons). The discriminatory power of a multivariable model for prediction of 3-year mortality was significantly improved after the inclusion of EuroSCORE (adjusted area under the receiver operating characteristic (ROC) curve = 0.779, 95% confidence interval 0.747 to 0.810, P = 0.008), but not after the inclusion of SxScore, SS II, or GRC. Conclusions: In patients undergoing DES implantation for uLMCA disease, all evaluated risk scores were able to stratify the mortality risk at long-term follow-up. EuroSCORE was the only risk score that significantly improved the discriminatory power of a multivariable model to predict long-term mortality.