Estimation of individual lifetime cardiovascular disease risk and treatment benefit in patients with established atherosclerotic cardiovascular disease: the updated SMART-REACH2 model

Abstract Background The 2021 ESC Guidelines on Cardiovascular Disease (CVD) Prevention in Clinical Practice recommend using (lifetime) risk to guide preventive treatment intensification. Purpose Updating and systematically recalibrating the SMART-REACH lifetime prediction model for the estimation of lifetime risk of recurrent CVD in patients with established atherosclerotic CVD (ASCVD) to various European and international risk regions. Methods The updated SMART-REACH (i.e., SMART-REACH2) model was derived using data from 9,477 individuals between 40-90 years with coronary artery disease, cerebrovascular disease, peripheral artery disease, or abdominal aortic aneurysms included in the UCC-SMART cohort in whom 2,090 ASCVD events were observed during a median follow-up of 8.9 years (interquartile range 4.5–14.1). Coefficients for type of CVD, estimated glomerular filtration rate, non-high density lipoprotein cholesterol, systolic blood pressure, duration of CVD, C-reactive protein, current smoking, and presence of diabetes were derived using sex-stratified cause-specific Cox models with age as the time axis, adjusted for the competing risk of non-CVD mortality. Lifetables were used to estimate 10-year and lifetime CVD-risk and CVD-free life expectancy. The SMART-REACH2 model was externally validated in 121,701 patients from the NOR-COR, Nor-COAST, HUNT3, SWEDEHEART, Bialystok PLUS/Polaspire, REACH-EU, and REACH non-EU (Asia, Japan, Latin America, North America) covering 43 countries. The model was recalibrated to specific risk regions based on the expected-observed ratios from representative cohorts. Performance was assessed by calibration of predicted and observed risk and Harrel’s C-statistic for discrimination. Results 16,689ASCVD events were observed in the external validation cohorts. The model was visually well calibrated for the various risk regions. The pooled C-statistic across risk regions was 0.665 [95% confidence interval 0.640-0.691], with adequate performance in the separate regions (Figure 1). Using SMART-REACH2 to estimate potential gain in CVD-free life expectancy upon treatment initiation showed varying projections. For example, two comparable 55-year-old males from the low-risk region and very high-risk region would gain an estimated 2.2 and 3.9 CVD-free life years as a result of smoking cessation (Figure 2). Conclusion(s) By using larger and more contemporary data sources, considering geographical differences, and extending the age range, the updated SMART-REACH2 model provides an improved algorithm for the estimation of lifetime risk and CVD-free life expectancy for individuals with established ASCVD, facilitating shared decision-making for cardiovascular prevention as recommended by the 2021 ESC Prevention of CVD guidelines.C-statistics across global risk regionsCase vignette