Prognostic performance of the ESC SCORE and its German recalibrated versions in primary and secondary prevention

Abstract

The systematic coronary risk evaluation (SCORE) chart of the European Society of Cardiology (ESC) published in 2003 is a valuable tool used in routine practice that permits individual cardiovascular risk stratification in primary prevention. Furthermore, a SCORE-based prevention strategy is able to improve cardiovascular prognosis. To adapt to the current regional risks of the German population, recalibrations were published in 2005/2016. A beneficial effect of such recalibrations in risk prediction has recently been discussed to cope better with the regional and present risks of a population. Although originally intended for primary prevention, the SCORE system is often used in various other settings to define cardiovascular disease (CVD) risk without an awareness of how accurately it performs in a context such as secondary prevention. The aim of the current investigation was to evaluate the predictive performance of these SCORE variants regarding primary and secondary prevention using a German cohort of stable patients with suspected chronic coronary syndrome (CCS). The study cohort was drawn from patients enrolled consecutively within the first 3 years in an ongoing, prospective, multicentre registry that was initiated in August 2010. In brief, patients were enrolled if they were indicated for invasive coronary angiography (ICA) due to suspected CCS irrespective of previously known coronary artery disease (CAD). To gain a stable population only patients with an ejection fraction greater than 55% (based on echocardiography or recent medical reports) and in whom no invasive or surgical treatment was needed after ICA were evaluated (n1⁄4 1253). Twenty-six patients were excluded due to missing data needed for the SCORE calculation. Data on all-cause mortality as an outcome measure were obtained in August 2018 with a median time frame of 6.53 (5.64–7.4) years since enrolment. The use of patient material complied with all the relevant national regulations and institutional policies and was in accordance the tenets of the Declaration of Helsinki. The study was approved by the ethics board of the University Giessen (AZ147/11) and all patients gave informed consent. Individual values for all three SCORE variants were calculated using the recommended low-risk tables based on serum total cholesterol. In addition, the SCORE 2003 formula without truncating age above 65 years was used. The number of patients in the predefined risk categories (< 1%, 1%, 2%, 3–4%, 5–9%, 10–14%, 15%) according to each SCORE variant were compared, and odds ratios (ORs) were calculated. The area under the receiver operator characteristic curve (AUROC) was calculated to describe the predictive value of the continuous scores. A total of 1227 patients (median age 68.3 (interquartile range (IQR) 59.0–74.8) years; 40.3% women) constituted the study cohort including 663 patients without