0015: Validation and recalibration of the Systematic coronary risk evaluation (SCORE) in a coronary Tunisian population

Abstract

Development of a validated risk prediction model for cardiovascular death in coronary patients is a high priority for strategies of therapy. We sought to validate and recalibrate of the SCORE (Systematic coronary risk evaluation) risk chart based on Tunisian national mortality data and average major cardiovascular risk factor levels. Baseline data were collected between 1997 and 2004 in 146 male patients aged 52.4 ± 9 years hospitalized with STEMI treated with fibrinolysis in 68 % of patients and not revascularized early in 32% of patients. Vital status was checked and causes of death were obtained in 2011 after a mean follow up of 9.6 years. The expected cardiovascular mortality was calculated by applying the SCORE equation for high risk populations on the basis of the level of risk factors in the total population, in the diabetic and non-diabetic population and was compared with the observed mortality in each group. Correction factor was calculated for each group. Univariate analysis was used for statistical analysis. The optimum threshold of SCORE , allowing for an optimal sensitivity and specificity, was determined by the ROC curve (receiver operating characteristic). For risk thresholds 5% and that determined from the ROC curve of the European SCORE sensitivity, specificity values were calculated. The total number of cardiovascular death at 10 years is 18 with a mortality of 12.3%. The average of SCORE in our population was 8.73+/-5.12% with extremes ranging from 1 to 32%. The European SCORE was strongly correlated in our cohort to the occurrence of cardiovascular death at 10 years (p <0.0001). The correction factor of SCORE is calculated to 2.7 in diabetic group, 1 in non-diabetic group and 1.4 for the total population. The ROC curve has a c index ( AUC ) = 0.73 corresponding to the risk threshold of 9.12%. The relative risk of cardiovascular death at 10 years of SCORE for the 9.12% threshold is calculated at 3.6. For the risk threshold for 9.12%, sensitivity was calculated at 66.7% and specificity at 68.8%. For the risk threshold for 5%, sensitivity was calculated at 94.9% and specificity at 17.96%. SCORE is validated in coronary male Tunisian patients with and recalibrated using correction factors. Validation on a larger population and multi-ethnic remains our future desire.