Background/Purpose: Rheumatoid arthritis (RA) patients have an increased cardiovascular (CV) risk. Current algorithms generally underestimate the risk in these patients [1]. In a meta-analysis, we have recently shown that younger RA patients bare the highest excess risk as compared to age-matched counterparts from the general population [2]. Objective: To investigate whether gender and age are contributing to the misclassification of CV risk in RA patients, when current risk algorithms developed for the general population are used. 

Methods: Prospectively collected data on cardiovascular risk factors and incident events from the Nijmegen inception cohort were analyzed, with up to 10 years follow-up [1]. Original as well as the EULAR modified (M_) SCORE algorithms were used to calculate the CV risk. Patients were stratified in deciles according to increasing risk and Hosmer-Lemeshow test was used to check concordance between observed and predicted risk, in subgroups based on gender and age.

Results: 863 prospectively followed RA patients were included in the analysis, 566 females and 297 males. During the first 10 years of follow-up 128 incident cardiovascular events had been recorded. In the whole group there was evidence of discrepancy between the predicted and observed CV risk (H-L test p<0.003) when the SCORE algorithm was applied. This mismatch was also present in the female subgroup (H-L test p<0.001), but it was less pronounced in males (H-L test p=0.09). Among the group of women who developed an event 40% had a predicted risk lower than 10% (corresponding to the low-risk group), whereas this was just 14 % in the RA males. When analyzing sub groups based on age, the H-L test p-values were: <55 years p<0.001, between 55 – 65 years p =0.93 and  >65 years p =0.96. The discrepancy between observed and predicted CV events in the youngest RA patients consisted mainly of underestimation of the CV risk (5.3% predicted vs. 8.0% observed). Similar results were obtained when the M_SCORE was applied. 

Conclusion: We report here in prospectively followed RA patients that CV risk is especially underestimated in women and younger patients. This suggests that modifying the weight for female gender and/or younger age in currently used CV risk algorithms might improve their predictive value in RA, contributing to better CV risk management in this group of patients. 

References

1. Arts EE, Popa C et al. Performance of four current risk algorithms in predicting cardiovascular events in patients with early rheumatoid arthritis. Ann Rheum Dis 2015;74:668-74.
2. Fransen J, Seyyed M, Bredie SJ, Popa CD. Rheumatoid arthritis disadvantages younger patients for cardiovascular diseases: a meta-analysis. EULAR 2015 [manuscript submitted for publication]

« Back to 2015 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/younger-age-and-female-gender-are-the-main-determinants-of-underestimation-of-cardiovascular-risk-in-rheumatoid-arthritis-patients/