The SMART-REACH2 model predicted recurrent cardiovascular events with a pooled C-statistic of 0.68 (95% CI 0.66-0.69) and adequate calibration across global risk regions.
Cohort (n=2,094,488)
Yes
The SMART-REACH2 model provides a validated tool to estimate lifetime risk of recurrent CV events and potential treatment benefits across different geographical risk regions, facilitating shared decision-making.
Effect estimate: C-statistic 0.68 (95% CI 0.66-0.69)
BACKGROUND AND AIMS: The 2021 ESC guidelines on cardiovascular (CV) disease prevention recommend the SMART-REACH lifetime risk model to guide treatment decisions in patients with established atherosclerotic CV disease. The aim was to develop the SMART-REACH2 model for estimating lifetime risk of recurrent CV events and treatment benefits in patients with established atherosclerotic CV disease, with systematic recalibration to the four European and other global risk regions. METHODS: SMART-REACH2 was derived in 8708 individuals aged 40-90 years with coronary, cerebrovascular, peripheral artery disease and/or abdominal aortic aneurysm from the UCC-SMART cohort. Sex-stratified, cause-specific Cox models for recurrent CV events and non-CV death were fitted using age as timescale and routinely available predictors. Recurrent CV events were defined as a composite of myocardial infarction, stroke, or CV death. Recalibration was based on representative cohorts per risk region. External validation was performed in 2 085 780 patients from 54 countries; model performance was assessed by calibration plots and Harrell's C-statistic. RESULTS: In the derivation cohort, 2057 recurrent CV events occurred over a median follow-up of 8.5 years (25th-75th: 4.3-13.0). In external validation, 307 706 events occurred. The pooled C-statistic was 0.68 (95% confidence interval 0.66-0.69) and ranged from 0.66 (0.64-0.69) for European low-risk region up to 0.72 (0.66-0.78) for Latin America, with adequate calibration across risk regions. Performance was consistent across sexes and CV disease subtypes. Using SMART-REACH2, estimated potential gains in CV disease-free life expectancy for a 50-year-old example patient receiving intensified preventive treatment (15 mmHg systolic blood pressure and 1.0 mmol/L low-density lipoprotein cholesterol reduction) ranged from 2 years in the low-risk region to 4.4 years in the very-high-risk region. CONCLUSIONS: The updated SMART-REACH2 model accounts for geographical and sex-specific variations and allows estimation of short-term and lifetime risk of recurrent CV events and treatment benefits, facilitating shared decision-making as recommended by guidelines.
Published in the European Heart Journal, this paper introduces an updated lifetime risk prediction model for patients with established ASCVD. It's a topic of discussion among preventive cardiologists for its potential to improve shared decision-making regarding long-term treatment strategies.
Holtrop et al. (Mon,) conducted a cohort in established atherosclerotic cardiovascular disease (n=2,094,488). SMART-REACH2 model was evaluated on Recurrent CV events (composite of myocardial infarction, stroke, or CV death) (C-statistic 0.68, 95% CI 0.66-0.69). The SMART-REACH2 model predicted recurrent cardiovascular events with a pooled C-statistic of 0.68 (95% CI 0.66-0.69) and adequate calibration across global risk regions.