The performance of SYNTAX score versus the coronary angiogram standard evaluation in the prediction of cardiovascular events in a cohort of patients with stable coronary heart disease

BACKGROUND: Scores for prediction of cardiovascular events in patients with stable coronary artery disease (CAD) submitted to coronary artery bypass grafting (CABG), percutaneous coronary intervention (PCI) or medical-therapy (MT), such as the SYNTAX score (SXscore), have been proposed, but there is no comparative assessment of their performance with the coronary angiogram standard evaluation (CASE). This study aimed to evaluate the performance of the SXscore versus the CASE in the prediction of major cardiovascular outcomes (MACCE) in patients with chronic CAD who were treated with MT or additionally submitted to CABG or PCI. METHODS: Prospective cohort study of 454 patients with CAD referred for elective diagnostic coronary angiography in Hospital de Clínicas de Porto Alegre, Brazil, with 40 years of age or over, which were followed on average for 6±2.0 years. Patients with acute coronary syndromes, valvular heart disease, aortic diseases, previous coronary revascularization, heart failure, chronic renal disease, history of cancer, or severe psychiatric illness were excluded. Agreement between the scores was evaluated by Kappa statistics. The performance of the scores to predict MACCE was evaluated by Cox proportional hazard models. Areas under the ROC curves were compared by the DeLong test. RESULTS: Patients with moderate to high SXscores or with left main or multivessel CAD (LMMCAD) in the CASE evaluation had higher rates of all-cause death and MACCE than those with low SXscore or without LMMCAD. After adjusting for confounding, only LMMCAD remained associated with the incidence of all-cause death in the total sample (HR =2.81;95% CI: 1.17–6.74) and for MACCE in patients undergoing MT (HR =8.72; 95% CI: 1.73–44.10). The ROC curves for all treatments were similar. Kappa statistics was not significant in patients submitted to MT, poor for patients treated by PCI and fair for the whole sample and patients treated with CABG. CONCLUSIONS: The severity of CAD defined by CASE or the SXscore provides similar prediction of the occurrence of cardiovascular events in patients submitted to clinical, PCI or CABG therapies. CASE is easier to do and may be the preferential method in the stratification of risk of patients with stable CAD.