Performance of four cardiovascular risk scales in chronic inflammatory rheumatic disease patients: 10-year follow-up data.

Abstract

To compare the performance of QRESEARCH risk estimator (QRISK)3, Systematic Coronary Risk Evaluation (SCORE)2, Predicting Risk of cardiovascular disease EVENTs (PREVENT)-cardiovascular disease (CVD) and PREVENT-atherosclerotic cardiovascular disease (ASCVD) equations in predicting cardiovascular (CV) risk among individuals with chronic inflammatory rheumatic diseases (CIRD) enrolled in the Spanish CARMA project. The study included CIRD patients from 67 Spanish hospitals (2010-2012) without prior CV events. After a 10-year follow-up, data were collected through medical records, patient/family contact or the Spanish National Statistics Institute. CV risk was analysed using Cox regression models adjusted for inflammatory disease. Model performance was evaluated using AIC, BIC, and Gönen and Heller's K. Among 2080 patients, 8.4% (174) experienced a first CV event. Those with CV events were older, had longer disease duration, higher body mass index and more frequent hypertension, diabetes or hypercholesterolemia. All four scales were strongly associated with CV events, with hazard ratios ranging from 1.04 (95% confidence interval [CI]: 1.03, 1.05) for QRISK3 to 1.13 (95% CI: 1.11, 1.16) for SCORE2. QRISK3 demonstrated better adjustment (lower AIC/BIC) but other scales showed higher concordance with actual events. QRISK3 and PREVENT-CVD accurately estimated 10-year CV risk for the overall cohort, while SCORE2 and PREVENT-ASCVD underestimated risk. Among rheumatoid arthritis patients, QRISK3 overestimated risk, whereas PREVENT-CVD was the only scale to perform reliably. All scales effectively distinguished between low and high CV risk. PREVENT-CVD emerges as a valuable tool for assessing CV risk in CIRD patients.