Assessment of the prognostic value of coronary angiography in patients with non-ST segment elevation myocardial infarction.

BACKGROUND Management of patients with acute non-ST segment elevation myocardial infarction (NSTEMI) depends on risk evaluation. The recommended approach involves the use of risk stratification tools such as TIMI and GRACE risk scores. However, these clinical scores do not include variables derived from coronary angiography which is currently performed in most patients. AIM To evaluate the prognostic value of adding selected coronary angiographic parameters to the established TIMI and GRACE risk scores. METHODS We studied consecutive patients with NSTEMI who underwent coronary angiography. We evaluated selected vascular variables (vessel score, lesion location, percent stenosis, presence of thrombus, lesion length, vessel size, TIMI flow, lesion type according to the ACA/AHA classification, and extent score) and estimated risk using the TIMI and GRACE scores. We assessed total mortality at 30 days, 180 days, and 3 years. To determine the prognostic value of vascular variables and risk scores, we used a logit model and the Hosmer-Lemeshow test. Diagnostic utility of the models was measured by the area under receiver operating characteristic (ROC) curves. To determine usefulness of selected vascular variables as outcome predictors in addition to the GRACE and TIMI scores, we used Net Reclassification Improvement (NRI) and Integrated Discrimination Improvement (IDI) indices. RESULTS The study included 237 patients (mean age 65.5 years, 62% men). The TIMI and GRACE risk scores were good predictors of mortality in the evaluated periods. Among vascular variables, independent prognostic factors included the extent score which predicted mortality at 30 days (odds ratio [OR] 12.7, 95% confidence interval [CI] 1.6-99, p = 0.016), 180 days (OR 8.8, 95% CI 2.3-33.7, p = 0.002), and 3 years (OR 3.5, 95% CI 1.6-8.0, p = 0.003), and distal lesion location which predicted mortality at 180 days (OR 3.1, 95% CI 1.0-9.4). Addition of the extent score to the TIMI risk score improved the prognostic value of the latter at all time points, as confirmed by NRI and IDI indices. The GRACE risk score itself had good prognostic value which was not significantly improved by any of the evaluated vascular variables. CONCLUSIONS The extent score added to the TIMI risk score improves the prognostic value of the latter in patients with NSTEMI. Angiographic variables should be more widely used in risk stratification models in patients with acute coronary syndromes.