Cardiovascular risk in inflammatory rheumatic diseases: loose ends and common threads.
T. Farragher,I. Bruce
2006-11-01
Abstract:It is well established that rheumatoid arthritis (RA) is associated with reduced life expectancy1-3. While the absolute increase in mortality risk varies between studies4, most found that the majority of these excess deaths result from cardiovascular disease (CVD)3,5,6, in particular, atherosclerotic CVD7,8. The risk of myocardial infarction in RA patients is, for example, increased by a factor of 2 to 36-8. Premature CVD mortality and morbidity in RA has been the subject of much investigation, and both traditional and nontraditional CVD risk factors have been implicated9-12. Traditional CVD risk factors found to be associated with atherosclerosis in RA include age, smoking, hypertension, and dyslipidemia10-12. However, traditional risk factors alone do not fully explain the excess CVD risk in RA. Other nontraditional factors are hypothesized to play a role, in particular the burden of inflammation as indicated by the C-reactive protein (CRP) and/or erythrocyte sedimentation rate (ESR)9,13,14. In a community-based cohort of patients with inflammatory polyarthritis, Goodson, et al also noted that excess CVD mortality was confined to patients who were rheumatoid factorpositive5. These markers of inflammation and inflammatory burden confer additional risk of CVD death in those with RA after adjusting for traditional CVD risk factors and comorbidities15. If RA is associated with increased CVD risk compared to the general population and this is in part due to chronic inflammation, then other inflammatory rheumatic diseases, such as ankylosing spondylitis (AS) and psoriatic arthritis (PsA) might also be hypothesized to have increased CVD risk. There are, however, a limited number of studies that have examined these conditions in relation to both mortality and the contribution of CVD to any observed excess deaths. In a community-based cohort of patients with PsA, for example, no excess mortality was noted16. In contrast, Wong, et al found in a hospital-based PsA cohort that mortality was increased, principally due to respiratory rather than cardiovascular deaths17. More recently, Peters, et al reviewed studies that investigated CVD mortality and morbidity and/or CVD risk factors in AS and PsA. They concluded that while there were a limited number of studies, there was evidence of increased CVD risk in AS and PsA18. CVD mortality in both conditions was associated with disease duration and severity, as well as traditional CVD risk factors, e.g., smoking, lipid profile, and hypertension. In addition, there was evidence that HLA-B27 status, increased platelet count, and fibrinogen may also play a role. It is clear, however, that compared to RA, the evidence base to support the hypothesis of increased CVD risk in PsA and AS is much less definitive, and many questions remain unresolved. Han and colleagues’ cross-sectional comparison of cardiovascular diseases and risk factors in RA, PsA, and AS patients, presented in this issue of The Journal, is an interesting contribution to this debate19. From a large health plan database in the US, the prevalence ratios of CVD diagnoses and risk factors in these 3 conditions were compared with unaffected population-matched controls (1:4 ratio). From almost 3 million individuals in the analysis, prevalence ratios of the cardiovascular diagnoses (ischemic heart disease, atherosclerosis, peripheral vascular disease, congestive heart failure, and cerebrovascular disease) and cardiovascular risk factors (type II diabetes, hyperlipidemia, and hypertension) were higher in the patients with RA, PsA, and AS. In addition, the use of cardiovascular drugs was also higher in all 3 patient groups. There are, however, issues relating to these types of data and subsequent analyses that may limit their generalizability to the population as a whole. Health plan data do not necessarily represent the population within which they are based, due to the selection criteria of the particular plan20. In addition, selection criteria for the study may further