Independent of ischemic heart disease, RA patients are at increased risk of CHF.14-16 RA patients are at particular risk for CHF with preserved ejection fraction,17 which may be a result of systemic inflammation causing left ventricular stiffening.18,19 Similar to CAD, patients with RA are less likely to present with typical CHF symptoms, are less likely to receive guideline-concordant care, and have higher mortality rates following presentation with CHF.17
Although accounting for a lower proportion of the excess CVD morbidity and mortality in RA, the risk of noncardiac vascular disease is also increased in RA patients. Large meta-analyses have identified positive associations between RA with both ischemic (odds ratio [OR], 1.64 [95% confidence interval {CI}, 1.32-2.05]) and hemorrhagic (OR, 1.68 [95% CI, 1.11-2.53]) stroke.20 Similarly, RA patients appear to have an approximately twofold higher risk of venous thromboembolic events.21 Less frequently studied than other forms of CVD, peripheral arterial disease may be increased in RA patients independent of other CVD and CVD risk factors.22,23
Assessing CVD Risk in RA
CVD Risk Scores
In order to identify patients who may benefit from primary prevention interventions, such as lipid-lowering therapy, CVD risk estimation typically centers on the use of well-established CVD risk calculators (Table). CVD risk scores such as the Framingham Risk Score (FRS), Systematic Coronary Risk Evaluation (SCORE), and American College of Cardiology/ American Heart Association (ACC/AHA) Pooled Cohort Equation incorporate traditional CVD risk factors, including age, sex, smoking status, blood pressure, lipid levels, and presence of diabetes mellitus.24,25 However, CVD risk in RA patients appears to be inadequately explained by traditional CVD risk factors,26 with disease activity and inflammation being associated with higher CVD risk. Recognizing that inflammation may contribute to CVD risk even among non-RA patients, the Reynolds Risk Score includes high-sensitivity C-reactive protein (hsCRP) in its calculation.27 In contrast to more robust performance in the general population, these well-established CVD risk scores have had variable predictive potential of incident CVD in RA patients.28-30
Several models, or adaptations to existing models, have been proposed to improve CVD risk assessment in RA populations (Table). In 2009, the European League Against Rheumatism (EULAR) task force suggested using a correction factor of 1.5 with traditional CVD risk models in RA patients with 2 of the following criteria: disease duration exceeding 10 years, rheumatoid factor or anti-cyclic citrullinated peptide (CCP) antibody positivity, or extra-articular manifestations of RA.31 An update to these recommendations in 2015 continued to propose the use of a 1.5 correction factor, but suggested applying this to all RA patients.32 QRISK2, a modification to QRISK1 which was developed to predict CVD in the UK general population, includes the diagnosis of RA as a risk factor, and in early validation efforts more accurately discriminated patients in the general population at increased risk of CVD compared to the FRS.33 Additional disease-specific risk factors such as systemic lupus, steroid use, severe mental illness, and steroid and atypical antipsychotic use were incorporated in the QRISK3 algorithm, with model performance similar to the QRISK2.34 The Expanded Cardiovascular Risk Prediction Score for RA (ERS-RA) was specifically developed to assess CVD risk in RA patients by including RA disease activity, level of physical disability, RA disease duration, and prednisone use.35 Despite efforts to develop “RA-specific” risk scores, these have not consistently outperformed traditional CVD risk calculators.36-38 In one study involving more than 1700 RA patients, the ERS-RA performed similarly to the FRS and Reynolds Risk Score, with a net reclassification index of just 2.3% versus the FRS.36
Imaging Modalities
Imaging modalities may assist in characterizing the increased risk of CVD in RA and the subclinical CVD manifestations that occur. For example, RA patients were shown to have more prevalent and unstable coronary plaque, higher carotid intima media thickness, and impaired myocardial function with computed tomography (CT) angiography and carotid ultrasound.39,40 However, studies harnessing noninvasive imaging to augment CVD risk assessment in RA patients are limited.
