Background/Purpose: Cardiovascular risk in rheumatoid arthritis (RA) is increased. The cardiovascular (CV) risk algorithms used in the general population may underestimate the risk of cardiovascular disease (CVD) in the RA population¹. This study was undertaken to assess the predictive ability of 5 established CV risk models for the 10-year risk of fatal and non-fatal CVD in European patients with Rheumatoid Arthritis.

Methods: Included were the Framingham risk score (FRS), the Systematic Coronary Risk Evaluation score (SCORE), the modified SCORE according to the EULAR recommendations for CV risk² (M-SCORE), the Reynolds risk score (RRS) and the QRisk II risk score. Prospectively collected data from the Nijmegen early RA inception cohort were used. Patients with CVD prior to enrollment were excluded. Discriminatory ability for CV risk prediction was estimated by the area under the receiver operating characteristic (ROC) curve. Calibration was assessed by comparing the observed versus expected number of events using Hosmer-Lemeshov tests and calibration plots. Sensitivity and specificity were calculated for the cut-off values of 10% and 20% predicted risk.

Results: In total, n=1050 patients were included. During follow-up, 145 patients developed a CV event. The mean±SD age was 54±13.8 years, 66% were female and 74%  were rheumatoid factor positive. Areas under the ROC curve were 0.75 (95% CI; 0.71-0.80), 0.75 (95% CI;0.71-0.80), 0.76 (95% CI; 0.73-0.80), 0.75 (95% CI; 0.71-0.79) and 0.77(95% CI; 0.73-0.80) for the SCORE, M-SCORE, FRS, Reynolds and QRisk II respectively, indicating moderate to good discrimination between patients with and without a CV event. All five models underestimated CV risk at low and middle observed risk levels, and overestimated CV risk at high observed risk levels (fig.1). For the 10% and 20% cut-off values used as indicators for CV preventive treatment, sensitivity ranged from 66-87% and 43-67% respectively and specificity ranged from 44-57% and 67-81% respectively. Depending on the model, 13% to 35% of observed CV events occurred in RA patients who were classified as low risk (<10%) for CVD.

Conclusion: Established risk models generally underestimate CV risk in RA patients. There is an unmet need for development of a RA-specific CV risk model.

Figure 2. Observed (closed bars) versus predicted (open bars) CV event rate (%) in deciles of predicted risk, for the SCORE (A), M-SCORE (B), FRS (C),Reynolds (D) and QRisk II (E) risk algorithms.

References

1       Crowson CS, Matteson EL, Roger VL, et al. Usefulness of Risk Scores to Estimate the Risk of Cardiovascular Disease in Patients With Rheumatoid Arthritis. Am J Cardiol. 2012;10:420-24.

2        Peters MJ, Symmons DP, McCarey D, et al. EULAR evidence-based recommendations for cardiovascular risk management in patients with rheumatoid arthritis and other forms of inflammatory arthritis. Ann Rheum Dis. 2010;69:325-31.