Peking union medical college hospital Beijing, Beijing, China
Disclosure(s): No financial relationships with ineligible companies to disclose
Background/Purpose: Patients with rheumatoid arthritis (RA) are at increased risk of developing cardiovascular (CV) events compared to the general population. However, this risk is often underestimated by current CV event risk prediction algorithms that do not take the characteristics of RA into account. The expanded cardiovascular risk prediction score for rheumatoid arthritis (the ERS-RA) has been shown to be superior to models that only incorporate traditional risk factors in predicting the risk of CV in patients with RA. This study aims to externally validate the ERS-RA for predicting 10-year CV event risk in Chinese patients with RA. Methods: Patients with RA from a Chinese nationwide, multicenter, prospective cohort, enrolled between October 2008 and May 2024 were included as the external validation cohort. Discrimination was evaluated using C-index, and calibration was examined using comparisons of observed versus predicted risks. Results: A total of 10,721 patients with RA were included. The incidence rate of CV events was 5.65 per 1,000 person-years. The C-index of the ERS-RA in this Chinese cohort was 0.831 (95% confidence interval [CI] 0.801– 0.861). The overall calibration slope was 1.012 (95% CI 0.873– 1.152), and the observed-to-expected ratio was 0.796 (95% CI 0.676– 0.937). Across clinically relevant risk intervals, the ERS-RA slightly overestimated CV event risk by 1.41%–2.18% for patients with the predicted risks of < 5%, 5–7.5%, and 7.5–10%, but underestimated CV risk by 10.46% for patients with the predicted risk ≥10%. Conclusion: The ERS-RA model demonstrated excellentdiscrimination and calibration for 10-year CV risk prediction in Chinese patients with RA. These results add to growing evidence for the clinical utility of ERS-RA in US, European, and now Chinese populations.
