Unresolved Questions in Rheumatology: Motion for Debate: The Data Support Evidence‐Based Management Recommendations for Cardiovascular Disease in Rheumatoid Arthritis

Cardiovascular disease (CVD) is a major source of morbidity and mortality in rheumatoid arthritis (RA). While we know much about the epidemiology and biology of this association, less is known about the treatment. This debate focuses on the management of CVD in RA: Do we know enough to set forth evidence-based recommendations? Underpinning this debate are many challenging scientific and clinical issues. First, should management focus on traditional cardiovascular risk factors or on inflammation? Second, should management of traditional cardiovascular risk factors be tailored to patients with RA, modifying treatment thresholds and targets, or are general population recommendations adequate? Third, should RA treatment be tailored to patients at risk of CVD? While these questions cannot be decided by a debate, our guest debaters have brought evidence to bear on many aspects of these topics. We are sure that you will enjoy the spirited back and forth and will come away with your own opinion of these issues. Cardiovascular disease is a major source of morbidity and mortality in RA and may equal the (contemporary) CVD burden in diabetes mellitus, a well-established risk factor for CVD (2-5). Currently, cardiovascular risk in diabetes mellitus is substantially lower than during previous decades as a result of effective implementation of strategies to accomplish good glycemic control and, in particular, optimal cardiovascular risk management, with statin treatment and blood pressure reduction being key drivers of this effect (6). In RA, the magnitude of CVD has not appreciably changed over the last decades (7, 8). Despite this well-established higher cardiovascular risk, a significant proportion of RA patients still receive no or suboptimal cardiovascular risk management (9-11). Based on this evidence, we can no longer bury our heads in the sand and pretend that cardiovascular risk management should not be part of our agenda. Cardiovascular risk prevention (i.e., targeting preventive strategies toward high-risk individuals) should constitute a key goal of management in RA. A task force of the European League Against Rheumatism (EULAR) has provided evidence- and expert opinion–based recommendations to help clinicians assess and control cardiovascular risk in RA (12). But, who should be screened, how often, and by whom? Which CVD risk prediction chart should be used? And, what should be the targets or thresholds for treatment with statins and antihypertensive agents? In RA, well-established cardiovascular risk factors, such as smoking, dyslipidemia, hypertension, diabetes mellitus, and decreased physical activity, occur more frequently (13-15). With information on age, sex, smoking status, lipid levels, and blood pressure, the 10-year absolute risk of a (fatal) cardiovascular event can be calculated with the use of established CVD risk prediction charts (e.g., the Systematic Coronary Risk Evaluation [SCORE] or the Framingham Risk Score) (16, 17). Cardiovascular risk management can be easily incorporated into routine visits by adding the measurement of blood pressure and nonfasting total cholesterol and high-density lipoprotein (HDL) cholesterol levels to routine blood testing (18). Of note, recent observations support the use of the total cholesterol–to–HDL cholesterol ratio as the most stable prognostic cardiovascular indicator in RA (19). All RA patients should receive evidence-based advice with regard to smoking, physical activity, and weight control, and if required based on the absolute cardiovascular risk, adequate management of that risk. Similarly, we underscore the need to assess cardiovascular risk factors in all RA patients, and we encourage clinicians to initiate statin treatment and blood pressure reduction according to national guidelines to reduce the cardiovascular risk. One could argue that we should await intervention trials with statins and/or antihypertensive agents and CVD end points in RA before conclusions about their efficacy can be reached. However, the efficacy of statins and their reduction of clinical end points in RA are probably at least equivalent to their effects in the general population (20, 21). Indeed, the effects of statins as well as some antihypertensive agents (i.e., angiotensin-converting enzyme inhibitors and angiotensin blockers) might be more pronounced in RA, since their additional beneficial effects include antiinflammatory properties (22, 23). Withholding treatment of proven efficacy (i.e., statins and/or antihypertensive agents) from RA patients, who are already known to be at increased cardiovascular risk, is unethical. So far, there is no substantial evidence for the use of lower treatment targets for statins and/or antihypertensive agents in RA as compared with those used in the general population. There is no indication for the use of aspirin for primary prevention of CVD in RA. To ensure sufficient uptake of cardiovascular risk prevention, we have recommended that patients receive a yearly cardiovascular risk assessment (12), but we recognize that in patients with low cardiovascular risk who have low levels of disease activity, a lower frequency of assessment could be adopted. Traditional cardiovascular risk factors, however, account for only part of the excess cardiovascular risk in RA (23). Inflammation is also important, as it accelerates atherosclerosis, either directly or via effects on cardiovascular risk factors (24). Hence, tight disease control and adequate suppression of the inflammatory process is crucial for lowering CVD risk in RA. Early and effective antirheumatic treatment has been shown to be associated with a lower cardiovascular risk, with methotrexate and tumor necrosis factor α (TNFα)–blocking therapy having the best available data, accepting that even here, we lack clinical trials showing definitive proof (25, 26). Interestingly, it is possible that cardiovascular risk is reduced only in patients who respond to TNFα-blocking therapy (27). One could argue that the introduction of the current powerful biologic therapies will neutralize the excess cardiovascular risk in RA, making cardiovascular risk management redundant. Data to support this notion are lacking, however, and recent risk estimates still demonstrate an ∼50% increased risk of CVD. Moreover, in sharp contrast to the much cheaper and more widely tested generic lipid- and blood pressure–lowering agents (e.g., in Europe, simvastatin costs at little as 荤5 per year, compared to more than 荤10,000 per year for TNFα-blocking agents), these new agents are very expensive and are not effective in ∼30% of RA patients (28). In addition, despite the ability to strongly suppress disease activity, systemic levels of cytokines sometimes remain high relative to those in non-RA patients. Hence, even RA patients receiving such treatment may still be at elevated risk of CVD. Nonsteroidal antiinflammatory drugs (NSAIDs), coxibs, and corticosteroids may influence cardiovascular risk in two competing ways. NSAIDs and coxibs exert prothrombotic effects, while corticosteroids have deleterious effects on lipids, insulin resistance, blood pressure, and obesity (29, 30). In contrast, these drugs may actually reduce cardiovascular risk by suppressing inflammation, which may improve mobility. Yet, conclusive evidence is not available. With these agents, we recommend the use of the lowest dose possible for the shortest period possible. Given the complex interplay between disease activity, cardiovascular risk, and the effect of antirheumatic treatment, rheumatologists are well positioned to perform cardiovascular risk assessments, considering that blood samples need to be obtained anyway and blood pressure measurement is noninvasive. Alternatively, cardiovascular risk management could be done in coordination with primary care physicians, who are experts in this process. Which risk scoring models should be used? We acknowledge that a CVD risk prediction chart is lacking for RA and that calibration of existing risk prediction charts for RA has not been done. There are several options for dealing with this. First, we can use existing CVD risk prediction charts until a model has been developed that is appropriate for RA. This will take many years, however, and in the meantime, many patients will receive suboptimal cardiovascular risk management because existing CVD risk prediction charts underestimate the true cardiovascular risk in RA (31). Second, we can use lower treatment targets for all RA patients, but evidence to support this notion is lacking. Moreover, the use of lower treatment targets is expensive and will increase the likelihood of patients' being harmed by the therapy. Third, the existing risk prediction charts are adapted in such a way that patients with an increased cardiovascular risk receive management. The EULAR task force thought this last option was the most appropriate and therefore recommended a multiplication factor. The task force selected 3 prognostic CVD markers to identify patients at highest risk of developing CVD: disease duration >10 years, presence of rheumatoid factor (RF) and/or anti–cyclic citrullinated peptide antibodies, and presence of severe disease (e.g., extraarticular manifestations). When patients meet at least 2 of these criteria, the risk estimate calculated with an existing CVD risk prediction chart should be multiplied by a factor of 1.5. For example, consider a 60-year-old female patient with RA who smokes and has a total cholesterol level of 6.8 mmoles/liter, an HDL cholesterol level of 1.1 mmoles/liter, and a systolic blood pressure of 170 mm Hg. According to the SCORE model (https://escol.escardio.org/Heartscore/), this patient has an 8% risk of experiencing a fatal cardiovascular event within 10 years. If this patient is IgM-RF positive and has an RA duration of >10 years, we need to multiply this risk by 1.5, which results in a 12% chance of a fatal cardiovascular event within 10 years. According to several European guidelines, treatment with statins and/or antihypertensive agents will be started when the CVD risk score is >10%, provided that the systolic blood pressure is >140 mm Hg and/or the low-density lipoprotein (LDL) cholesterol level is >2.5 mmoles/liter (96.7 mg/dl). We acknowledge that the multiplication factor of 1.5, which was derived from relevant standardized mortality ratios, is a matter of debate, since there is a paucity of prospective cohorts to enable the more usual approach to defining multiplication factors. A recent study of general practitioners in the UK provided additional support for a multiplication factor of ∼1.5 (32). Finally, Crowson and colleagues (31) recently demonstrated that cardiovascular risk appears to be most pronounced among RA patients who are RF positive, older, and have severe disease. Their findings are consistent with those upon which the EULAR recommendations were based. In summary, the increased CVD risk in RA is indisputable and is due to both an excess of “traditional” cardiovascular risk factors and the underlying chronic inflammatory process. We acknowledge the need for CVD prevention trials and RA-specific CVD risk prediction charts, but we cannot wait the many years it would take before these are available. We should incorporate cardiovascular risk management now—thereby reducing the CVD risk burden in our patients—and continue to evaluate its effectiveness over time. Evidence-based medicine has been defined as “the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients” (37). This definition incorporates the use of clinical expertise and external evidence, in particular from patient-centered clinical research, in order to come to optimal decisions about interventions for our patients (37). Where there is a burden of disease, we need to know with confidence how best to improve health. It has been established that in patients with RA, CVD rates are increased as compared to those in the general population. This is manifested as an increased incidence of myocardial infarction and stroke (38), and increased rates of from cardiovascular and the are new have been on this over the last years The current EULAR guidelines which were 2 years have already been over the of the evidence into the evidence from all available is but if we are to we have a effective intervention for reducing cardiovascular risk that is in It can be to the debate into a of First, is cardiovascular disease Second, by what is CVD Third, what is the evidence that intervention the risk of and to what is it is that intervention the evidence as we this of of these questions will be in Cardiovascular risk in patients with RA is increased by a of both traditional hypertension, and risk factors. This has been can be to many of increased risk. However, the from RA to increased CVD risk that it cannot be that in has a clinical There are many interventions have or even effects, for example, increased with the use of with and with therapy The debate therefore focus on we have sufficient evidence that an intervention at of the many risk the burden of CVD. end points are but the clinical end points are the that is the evidence that intervention the incidence of CVD and to what is it This can be with to interventions at traditional cardiovascular risk factors and risk factors. there are no trials the efficacy of statins or antihypertensive agents on clinical cardiovascular end points in RA Similarly, there are no of weight reduction or smoking such end points in RA. There are therefore no from in this particular disease population to support the use of these primary prevention This not that we have no as we can the results from study and of smoking in patients with disease a reduction in the relative risk of in those who as compared to those who continue to All rheumatologists will guidelines that smoking in patients with RA, even the was in the general population. of statins and antihypertensive agents have shown a reduction of cardiovascular end points in with existing CVD of a positive among with no CVD have been shown to have a positive effect in high-risk patients but is in patients with low cardiovascular risk prevention of CVD in the general population is not because the and the absolute risk of cardiovascular disease is patients would need to be the associated effects and order to of CVD. The in RA here, among patients with a lower risk of CVD. the evidence that RA is to an additional risk of CVD that by the traditional risk factors, the EULAR guidelines an cardiovascular risk prediction score by additional risk can an individual the risk, and the treatment, The in the recommendations a female patient with traditional risk factors that would a risk of a fatal cardiovascular event within 10 years in the general population. is multiplied by because of RA, a 10-year risk of experiencing a fatal cardiovascular event of This into a risk that the treatment at this is to the traditional risk factors. But, is this increased risk by RA by interventions to or blood statins less well in RA patients than in the general population CVD in RA more than lipid If statins have a positive is this at a particular of it be the increased CVD is established in RA, reducing lipid levels can but cannot the questions remain and the in this option for reducing CVD in patients with RA is to the risk factors, such as There is evidence to support this. All in a recent a reduced risk of cardiovascular in patients with methotrexate relative risk confidence The magnitude of the reduced risk is is possible as from a cardiovascular risk reduction would be a of the this as having the best evidence for risk While this may be we be not to that methotrexate to a risk reduction than or There have been the cardiovascular effects of these and no If we are to the optimal treatment that disease and cardiovascular risk, we need to this the cardiovascular effects of therapy is because is at many of the It may even have effects, for example, by reducing but lipid Interestingly, of cardiovascular therapy have to results from a lower to a increased of cardiovascular One for the as these are all the influence of therapy on CVD in patients with RA, no two have the the definition of cardiovascular and for the there are additional between in of the of the disease and we still not know the effects of therapy on cardiovascular in this might be but need to to questions The last and most is intervention is it is a of and This is not an since is in and information can be despite evidence for a of questions in RA we can still be in reducing cardiovascular risk. such as weight reduction and smoking a of positive and we continue to such control of disease has already been established as the of care for patients with RA in order to the reduction of disease in their Cardiovascular can be as an However, to traditional risk factors in patients with RA at a at which intervention would not be in the general population not have a evidence of RA patients to such we need to have the and that RA patients are at increased cardiovascular risk. However, we are of the increased cardiovascular risk by RA. A recent study that there is cardiovascular risk management by primary care as well as This the need for to improve of cardiovascular risk in RA by the that cardiovascular risk in RA is not only to risk factors, but also to risk factors, systemic and antirheumatic treatment. The about may our of the of RA-specific treatment in cardiovascular risk in RA. While we to some this to all and But, should because of our lack of about its true effect on the CVD burden in RA, statin therapy in RA patients at high risk of CVD? We is in that the of intervention effects from non-RA patients to those with RA is not by the findings of However, the can be about the use of many in that have been in such as and many We should not treatment of cardiovascular risk factors from these patients if of the of CVD of the findings of intervention are to the of and the of to support is part of evidence-based medicine The only patient at this of cardiovascular intervention trials have are those with disease, who to be in of cardiovascular There is no for chronic or factors would statins or antihypertensive agents This not that trials should not be but it that RA patients should be the of the In addition, from several interventions in which there were effects, for example, increased with therapy. this that we should not patients who are Similarly, should we statin therapy in RA patients at high risk of CVD because of its possible We with this of from on the effects and/or effects of statin therapy or drugs in RA should not from to the higher risk of CVD in patients with RA. there are no that would to that statin therapy is less effective in RA, and the of statin therapy to be in to the in many (20, We that in RA patients will improve the evidence on this relevant such trials have been however, drugs that are very to in the RA population unethical. debate focuses on the lack of evidence to support treatment of risk factors in patients with RA at a intervention would not be in the general population. we would to that cardiovascular risk management in RA should be done according to national guidelines, no from that in the general that only when the 10-year risk of a fatal cardiovascular event is is treatment at the when the blood pressure and/or lipid levels are recently underscore the that well-established CVD risk prediction charts, such as the Framingham Risk and the Risk underestimate the true cardiovascular risk in patients with RA, that the is in order to an appropriate of the true cardiovascular risk in RA (31). To with this and because assessment of cardiovascular risk on the of the RA, the EULAR task force risk In approach is the of RA as a cardiovascular risk factor is of CVD risk prediction charts will to implementation of strategies in routine clinical The of the effects of statins and antihypertensive agents to patients with RA is and and is in with what as an appropriate and definition of evidence-based medicine in the very as There is significant in the and in the of the might even be for the of To with most of the of and We that there is an increased risk of CVD in patients with RA, this risk is due to both an excess of “traditional” cardiovascular risk factors and the underlying chronic inflammatory and while there is a need for CVD prevention trials in RA as well as RA-specific CVD risk prediction charts, we the treatment we Where we is over which cardiovascular risk management interventions should be in and the focus of this debate, on what our cardiovascular risk management. with RA who general population for cardiovascular risk reduction should receive interventions weight blood pressure control, of their RA. with RA and obesity should be advice about weight management. While their arthritis might more weight reduction should remain a should be to smoking, of have RA. or should also be if based on absolute cardiovascular risk, prediction models such as the Framingham Risk We all that the of evidence for the reduction of cardiovascular end points in the RA population should not from cardiovascular risk factors as we would in the general population. We also that we should inflammation, which will not only the disease, but has also been proven to have a beneficial effect on cardiovascular end Which treatment has the most for this to be and should be The in our is we should cardiovascular risk factors in patients who would not have the a multiplication factor of to a risk score of 8% would some the treatment but is this multiplication factor RA its excess cardiovascular risk only by the risk by the traditional cardiovascular risk factors, as the If targeting the cholesterol would be appropriate and would the cardiovascular risk of RA. This model that RA via the of traditional risk factors an effect of these established We have already all that this is not the to the model is that RA is an risk factor. In a cholesterol level additional per patients, and the presence of RA would a the cholesterol level in this would have the as it in the general population. such a treatment was in general population it would be no more appropriate in patients with RA, since the and costs are The is to between these about appropriate treatment in this This is the evidence is lacking, to patients who would not meet traditional treatment thresholds for cardiovascular preventive but by of their RA, are to be at higher risk. We need to know how best to to their cardiovascular risk. how can this be primary prevention in RA statins and to cardiovascular end points would to this Indeed, the of for the of Cardiovascular in with was established in the UK to this However, despite being the of RA in the the study was more than patients of the population been due to a low incidence of the primary cardiovascular end to a Despite its the that patients of a low absolute risk of cardiovascular end points is about the of interventions are based on absolute risk reduction or the to The low incidence that between the is less to be If are how we this in can the of statins in patients with RA, but of results from by The treatment is statins for a particular and or known and is to have a higher risk of CVD compared to the who not have or known CVD. One possible is to trials within patients to treatment, information on as part of routine clinical This will take a the of but it is a goal This debate was the evidence to support treatment Despite some about the multiplication factor for risk prediction and the treatment we on most The of patients with RA who need to be to of patients with RA have a of and perform no physical to a a and not smoking could account for an reduction in cardiovascular risk, and into this management of cardiovascular risk factors that intervention in the general population can a in RA. should continue to the effectiveness of treatment with antirheumatic drugs for cardiovascular end points and evaluate the of and for primary prevention in RA. All their individual of the it for and the to be and acknowledge and for their advice and of their