You are here

  • Home
  • Archive
  • Volume 69, Issue 4
  • Statin use in rheumatoid arthritis in relation to actual cardiovascular risk: evidence for substantial undertreatment of lipid-associated cardiovascular risk?

Article Text

Article menu

Download PDFPDF

Clinical and epidemiological research
Extended report
Statin use in rheumatoid arthritis in relation to actual cardiovascular risk: evidence for substantial undertreatment of lipid-associated cardiovascular risk?
  1. Tracey E Toms1,
  2. Vasileios F Panoulas1,
  3. Karen M J Douglas1,
  4. Helen Griffiths2,
  5. Naveed Sattar3,
  6. Jaqueline P Smith4,
  7. Deborah P M Symmons5,
  8. Peter Nightingale6,
  9. George S Metsios1,
  10. George D Kitas1,5
  1. 1Department of Rheumatology, Dudley Group of Hospitals NHS Trust, Russells Hall Hospital, Dudley, West Midlands, UK
  2. 2Life and Health Sciences, Aston University, Birmingham, UK
  3. 3University of Glasgow, Glasgow, Scotland, UK
  4. 4Department of Clinical Biochemistry, Dudley Group of Hospitals NHS Trust, Russells Hall Hospital, Dudley, West Midlands, UK
  5. 5arc Epidemiology Unit, Manchester University, Manchester, UK
  6. 6Wolfson Computer Laboratory, University Hospital Birmingham NHS Foundation Trust, Birmingham, UK
  1. Correspondence to Professor George D Kitas, Department of Rheumatology, Dudley Group of Hospitals NHS Trust, Russells Hall Hospital, Pensnett Road, Dudley, West Midlands DY1 2HQ, UK; gd.kitas{at}dgoh.nhs.uk or g.d.kitas{at}bham.ac.uk

Abstract

Background Cardiovascular disease (CVD) is partially attributed to traditional cardiovascular risk factors, which can be identified and managed based on risk stratification algorithms (Framingham Risk Score, National Cholesterol Education Program, Systematic Cardiovascular Risk Evaluation and Reynolds Risk Score). We aimed to (a) identify the proportion of at risk patients with rheumatoid arthritis (RA) requiring statin therapy identified by conventional risk calculators, and (b) assess whether patients at risk were receiving statins.

Methods Patients at high CVD risk (excluding patients with established CVD or diabetes) were identified from a cohort of 400 well characterised patients with RA, by applying risk calculators with or without a ×1.5 multiplier in specific patient subgroups. Actual statin use versus numbers eligible for statins was also calculated.

Results The percentage of patients identified as being at risk ranged significantly depending on the method, from 1.6% (for 20% threshold global CVD risk) to 15.5% (for CVD and cerebrovascular morbidity and mortality) to 21.8% (for 10% global CVD risk) and 25.9% (for 5% CVD mortality), with the majority of them (58.1% to 94.8%) not receiving statins. The application of a 1.5 multiplier identified 17% to 78% more at risk patients.

Conclusions Depending on the risk stratification method, 2% to 26% of patients with RA without CVD have sufficiently high risk to require statin therapy, yet most of them remain untreated. To address this issue, we would recommend annual systematic screening using the nationally applicable risk calculator, combined with regular audit of whether treatment targets have been achieved.

https://doi.org/10.1136/ard.2009.115717

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    The impact of cardiovascular disease (CVD) in patients with rheumatoid arthritis (RA) is significant, as it accounts for approximately half of all deaths.1 Patients with RA have been shown to be twice as likely to experience a myocardial infarction (MI) than the general population, and have higher rates of mortality post MI.2 These findings may partially explain the reduced life expectancy of 5–10 years observed in RA.3 CVD risk is increased at an early stage of the disease course,4 and this risk excess beyond traditional risk parameters appears driven by novel risk factors (eg, systemic inflammation and a prothrombotic state, among others). Nevertheless, modification of traditional risk factors through lifestyle changes and appropriate drug prescription (ie, statins and antihypertensives) may reduce this risk and is an obvious target for intervention.

    Cardiovascular risk can be assessed using several specially designed algorithms. The earliest risk prediction tool was developed over 30 years ago by the Framingham Heart study investigators, the Framingham Risk Score (FRS),5 and allowed an individual person's 10-year CVD event probability to be expressed as a percentage. This risk stratification system has formed the basis of quantifying risk and guiding treatment for many years, and is still used widely throughout the world. However, progression of our understanding of CVD has prompted the development of newer risk scoring systems.6,–,8 The National Cholesterol Education Program (NCEP) adult treatment guidelines 6 developed in 2002, perhaps offer a more comprehensive method of risk assessment as they have been designed to incorporate the Framingham 10-year Risk Score as well as additional factors such as high-risk conditions (eg, type 2 diabetes) and the use of drugs (eg, antihypertensives and lipid-lowering agents). The NCEP guidelines have been formulated specifically to guide the management of dyslipidaemia via lifestyle modification and drug intervention. A more recently developed cardiovascular risk tool, which was produced and is used in most of Europe, the Systematic Coronary Risk Evaluation (SCORE),9 involves the use of risk tables with the main focus being total cholesterol (TC) levels or the TC:high-density lipoprotein (HDL) ratio, and again allows a person's 10-year risk of CVD to be expressed as a percentage. However this method, unlike the FRS, estimates the 10-year risk of any first fatal atherosclerotic event and not just death from coronary heart disease, and it also only estimates CVD mortality and not morbidity. Further advances in our understanding of the pathogenesis of CVD, including the recognition of C reactive protein (CRP) as an independent CVD risk factor in the general population,10 11 have lead to the development of CVD risk algorithms.12 13 The Reynolds Risk Score (RRS) is based on the FRS but incorporates additional risk factors including levels of high sensitivity CRP and a parental history of myocardial infarction before the age of 60. Therefore, it offers the potential to account for the excess CVD risk seen in RA as a result of systemic inflammation. The RRS calculates a person's 10-year risk of cerebrovascular events in addition to CVD morbidity and mortality.

    A recent meta-analysis has demonstrated that patients with RA have an increased cardiovascular standardised mortality ratio of between 1.6 and 1.7.14 As a result of this observation, it has been suggested that risk as calculated by conventional risk stratification methods should incorporate a multiplier.15 16 In particular, the European League Against Rheumatism (EULAR) task force have suggested that the use of a 1.5 risk multiplier should be reserved for patients with RA fulfilling two of the following three criteria: (1) a disease duration greater than 10 years, (2) seropositive for rheumatoid factor or anti-cyclic citrullinated peptide (anti-CCP) antibody and (3) presence of extra-articular manifestations.16

    In this study, we aimed (a) to assess the prevalence of dyslipidaemia according to the NCEP definition; (b) to formally evaluate cardiovascular risk among an RA population using the FRS, NCEP, SCORE and RRS guidelines (in their original format and incorporating a 1.5 multiplier in applicable patients, according to EULAR task force recommendations to account for excess RA risk); and (c) to establish whether lipid-lowering therapy was used appropriately or not.

    Methods

    Prior to commencement of this study full local ethical approval was obtained and all participating patients gave informed written consent.

    In all, 400 consecutive patients with RA, fulfilling the American College of Rheumatology (ACR) criteria,17 were recruited from rheumatology outpatient clinics held at the Dudley Group of Hospitals NHS Foundation Trust, West Midlands, UK between 1 August 2004 and the 31 July 2006 (the Dudley Rheumatoid Arthritis Comorbidity Cohort (DRACCO); details of the cohort have been described extensively in other papers by our group).18,–,20 Of these 400 patients, 299 did not have prevalent CVD (current incidence or history of coronary heart disease, coronary artery bypass grafting, percutaneous coronary intervention, cerebrovascular disease or peripheral vascular disease) or diabetes mellitus (DM) and were formally risk assessed in this study as potential candidates for primary prevention.

    At baseline, all participants underwent a thorough assessment, including a one to one interview, physical examination and case note analysis performed by a rheumatologist. Details recorded included a personal and family history, clinical characteristics and demographics (see table 1). All medications (previous and current) were recorded with their indication, including statins, fibrates, antihypertensives, oral daily prednisolone, disease-modifying antirheumatic drugs (DMARDS), non-steroidal anti-inflammatory drugs (NSAIDs) and cyclo-oxygenase 2 specific inhibitors. Physical examination incorporated measurements of height and weight. Contemporary measurements of disease activity and physical function were assessed using the 28-joint Disease Activity Score (DAS28)21 and the Health Assessment Questionnaire (HAQ),22 respectively.

    View this table:
    Table 1

    NCEP risk categories/LDL levels requiring treatment with lipid-lowering therapy

    Fasting blood samples taken at baseline included a full lipid profile (TC, triglycerides (TG), HDL, low-density lipoproteins (LDL)), apolipoproteins A and B (ApoA and B), glucose, erythrocyte sedimentation rate (ESR) and rheumatoid factor. All blood samples were analysed by a single laboratory. The Vitros 5,1FS chemistry system (Ortho-Clinical Diagnostics, High Wycombe, UK) was used to measure all lipid components, however, TC, HDL-cholesterol and TG were measured using multilayered slides, whereas measurement of LDL-cholesterol (direct measurement) and ApoA and B required a dual chamber package. All lipids measured within the laboratory were found to adhere to external quality control standards.

    Prevalence of dyslipidaemia

    The presence of dyslipidaemia was assessed according to the NCEP criteria,6 which defines dyslipidaemia as one or more of the following: TC ≥6.2 mmol/litre or LDL ≥4.13 mmol/litre or HDL<1.03 mmol/litre or TG≥1.7 mmol/litre, or undergoing lipid-lowering therapy.

    Risk stratification

    CVD risk assessment was carried out on all patients via the application of the FRS,5 NCEP6 and SCORE9 and RRS23 24 algorithms.

    FRS

    Calculation of the FRS involved scoring patients according to their age, TC level, HDL-cholesterol level, systolic and diastolic blood pressures and smoking status. The scores from each of these components were then added together to give a total score and a corresponding 10-year CVD risk prediction. A 20% 10-year risk of global CVD events is generally accepted as a cut-off for the implementation of statin therapy for primary prevention in the UK.25 This tool is only validated for use in patients below the age of 75 and therefore was only applied to patients less than 75 without a history of CVD or diabetes.

    NCEP

    The five-step NCEP risk assessment involved classifying fasting lipid levels, identifying major risk factors for coronary heart disease (CHD; eg, age, cigarette smoking, hypertension and family history of CVD in first degree relatives) and calculating 10-year CHD risk. This evaluation enables the determination of the risk category that establishes need for lipid-lowering therapy and the LDL goal. In general, lipid-lowering therapy should be given in all patients found to have a 10-year risk >10% for global CHD events and an LDL-cholesterol >130 mg/dl (further details regarding risk categories and LDL treatment thresholds are summarised in table 1).

    View this table:
    Table 2

    Demographics of the study population

    SCORE

    Risk stratification using the SCORE criteria involves the use of specially designed and validated risk tables. The risk tables include data on age, gender, smoking, blood pressure, TC and HDL level. Individual tables have been specifically designed for low-risk and high-risk patients (based on TC levels) and for risk based on the TC:HDL ratio. For the purposes of this paper, we used the high-risk tables, as this study was carried out on a population based in the UK, a country classified as ‘high risk’.9 In addition, analysis was restricted to tables based on the TC:HDL ratio as this is in line with the EULAR task force recommendations.26 Patients were classified as at risk of CVD if their 10-year risk was ≥5% for CVD deaths. Patients are deemed eligible for statin therapy as part of CVD risk reduction if they have a 10-year risk ≥5% and an LDL ≥3 mmol/litre or TC ≥5 mmol/litre.26 Patients were excluded from SCORE risk stratification if they were already deemed at high risk of CVD, for example, a history of CVD, diabetes mellitus, familial hypercholestrolaemia (TC ≥8 mmol/litre or LDL ≥6 mmol/litre), or a blood pressure ≥180/110 mm Hg.

    RRS

    The RRS was calculated using an online electronic tool.23 Patients over the age of 80 and diabetic patients were deemed to be at high risk and thus excluded from risk stratification. Details regarding the patients age, gender, systolic blood pressure, smoking status, TC level, HDL level, high sensitivity CRP level and parental family history of CVD were used in the calculation of the 10-year risk. Patients with a 10-year risk ≥20% were classified at risk CVD and were eligible for statin therapy.

    The application of a risk multiplier to conventional calculators

    In accordance with the EULAR task force specific recommendations,16 patients with two out of three of the following: (1) a disease duration ≥10 years, (2) seropositive (RF positive or anti-CCP positive) and (3) evidence of extra-articular disease, had their CVD risk according to each definition multiplied by 1.5. For the FRS this required a straightforward multiplication of the 10-year risk. Adaptation of the NCEP risk stratification criteria was performed in a similar manner by incorporating the multiplied FRS into the five-stage process. For SCORE and RRS, the final 10-year risk was multiplied by 1.5. The cut-off levels for implementing lipid-lowering therapy remained unchanged, therefore allowing excess patients to be identified as at risk. The appropriateness of lipid-lowering prescriptions was then examined.

    Statistical analysis

    Data analysis was carried out using SPSS V.16.0 (SPSS, Chicago, Illinois, USA). The Kolmogorov–Smirnov test was used to evaluate the distribution of each parameter. Values were expressed as mean± SD, median (IQR) or percentages, as appropriate.

    Results

    Basic demographics and clinical characteristics of the study population

    The median age of the total RA population studied was 61.5 years (IQR 52.5–68.1) and 231/299 (77.3%) were women. Patients had a median disease duration of 10 (IQR 4–17) years, and a mean DAS score of 4.2. Statins were prescribed in 24 (8%) and antihypertensives in 97 (32.4%) patients (table 2).

    Prevalence of dyslipidaemia

    NCEP criteria identified 150/299 patients (50.2%) as dyslipidaemic, of whom 82.7% were women. The prevalence of NCEP dyslipidaemia steadily increased up until the age of 70, where it appeared to peak and then to decline.

    CVD risk and statin use

    A total of 266 patients could be assessed according to FRS (limited to patients over the age of 75, without DM or history of CVD), 294 patients were eligible for NCEP risk stratification following the exclusion of patients with a history of CVD and diabetes, 166 patients were assessed by SCORE (limited to those who are <65 years of age, without history of CVD, DM, familial hypercholestrolaemia and severe hypertension), and 291 patients were available to be assessed by the RRS once patients with DM, CVD and those over the age of 80 were excluded (see table 3).

    View this table:
    Table 3

    Statin use among patients without prior CVD identified as being at risk

    FRS

    In all, 5 of the 266 participants (1.6%) had a 10-year risk of >20% and required primary prevention as per current UK guidelines. Of them, only 1 (20%) patient was receiving lipid-lowering therapy (statins/fibrates), leaving a total of four untreated at risk patients (80% of the at risk patients or 1.5% of the total population).

    NCEP

    A total of 64 out of 294 (21.8%) were at high risk. Of those, 58 were eligible for statin therapy on the basis of their LDL level, but only 3 (5.2%) were receiving lipid-lowering therapy, leaving 55 untreated at risk patients (94.8% of the at risk patients and 18.7% of the total population).

    SCORE

    Based on the TC:HDL ratio, 43/166 (25.9%) patients with a 10-year risk ≥5% and an LDL ≥3 mmol/litre or TC ≥5 mmol/litre were identified. Of these, 25 (58.1% of the at risk population or 15.1% of the total population) were untreated and thus remained at risk.

    RRS

    In all, 45 of 291 (15.5%) patients had a 10-year risk ≥20%. Of these, only 5 (11.1%) were receiving statin therapy, thus leaving 40 (88.9% of the at risk population or 13.7% of the total population) patients untreated and at risk.

    Figure 1 shows the distribution of patients at high risk of CVD with each definition according to age and gender.

    Figure 1
    Figure 1

    Percentage of patients at high risk of cardiovascular disease (CVD) according to each definition, age and gender.

    Differences between high-risk patients treated with statins and those untreated

    Patients identified as being at ‘high risk’ of CVD according to one or more methods of risk stratification (FRS/NCEP/SCORE (TC:HDL)/RRS) were grouped together to form a ‘high-risk population’. This identified a total of 115/299 (38.5%) at risk patients without a history of CVD or DM. Of these only 8 were receiving statins, leaving 107 (93.0% of the at risk population or 46.7% of the total population) untreated and at risk. The only factor that was found to significantly associate with statin prescription was older age (statin users vs non-statin users, 66.2 (62.5 to 69.5) vs 64.2 (59.6 to 67.7), p=0.043).

    Modification of risk stratification algorithms to account for additional RA-associated risk

    The EULAR task force recommendations were applicable to 140/299 patients who had two of the following: disease duration ≥10 years, were seropositive (rheumatoid factor and/or anti-CCP) or had extra-articular disease. The number of additional patients identified following the application of a 1.5 multiplier are summarised in table 4. In patients whom the EULAR task force recommendations were applicable in the under 65 population (thus allowing comparison of all risk algorithms), just under half (32/84) were identified as high risk according to 1 or more definitions. A total of 18 patients (56.3%) were identified by a single risk stratification method, 7 (21.8%) by 2 methods, 4 (12.5%) by 3 methods and 3 (9.3%) by all 4 methods.

    View this table:
    Table 4

    Comparison of high-risk patients identified by the original risk stratification methods and by incorporating a 1.5 multiplier according to European League Against Rheumatism (EULAR) task force recommendations

    Discussion

    The results of this study suggest that 2% to 26% of patients with RA without CVD in secondary care are at sufficiently high risk of developing CVD, as calculated by the FRS, NCEP, SCORE or RRS algorithms, to require primary prevention therapy with lipid-lowering agents. This figure rises to 7% to 30% if a multiplier of 1.5 is applied to applicable patients, to reflect the additional risk conferred by having RA. Despite this, use of statins was grossly suboptimal and the reasons for this need to be addressed by the medical community.

    In this RA population, the highest prevalence of at risk patients (25.9%) was found when applying the SCORE (TC:HDL) criteria. The other conventional methods of risk stratification (NCEP, RRS and FRS) identified a lower prevalence of 21.8%, 15.5% and 1.6%, respectively; the latter however, have a higher threshold for requiring treatment. A recent study27 adopted a similar comparative approach in a Spanish primary care non-RA population. This reported conflicting results, with the FRS detecting the highest prevalence rates (13.5%), followed by the SCORE (11.4%) and NCEP (7.1%). Such large differences may be explained by disease-specific phenomena occurring as part of RA (eg, activity, severity, duration or therapy), as well as differences in other baseline demographic or anthropometric characteristics of the populations studied, particularly age and sex. In the present RA population, we have also seen a considerable lack of overlap between the different methods of risk stratification, with the majority of at risk patients only being identified by one or two out of the four methods. Reasons underlying this may include: (a) differences in the components of each risk stratification system (eg, the SCORE risk stratification method is only applicable to patients under the age of 65, where as FRS is applicable up to the age of 75), (b) differences in the sensitivity and specificity of each of the risk stratification methods, (c) differences in the objective of each risk stratification method (eg, SCORE focuses on the 10-year risk of any first fatal atherosclerotic event, whereas the FRS focuses on the 10-year risk of any cardiovascular event, fatal or non-fatal) and (d) differences in the application of lipid parameters for statin eligibility. This may be particularly important in an inflammatory condition such as RA, where lipid levels are often suppressed as a consequence of inflammation. Irrespective of this, these findings may have significant implications for clinical practice: most rheumatologists will choose to adopt just one method of risk stratification and thus large numbers of potentially at risk patients may remain unidentified and untreated.

    There are many potential explanations for under treatment of CVD in RA. These include lack of ownership for the management of CVD risk in RA (is it the role of the primary care physician, rheumatologist or cardiologist?); the wrong perception that CVD risk is low among most women (who constitute the majority of patients with RA); ambiguity, lack of clarity, or indeed knowledge, among RA specialists about risk stratification and its implications; or a perceived or actual reluctance of patients to adhere to further polypharmacy alongside their standard RA drug therapy.28 Interestingly, lipid-lowering therapy prescriptions were significantly higher in patients with shorter disease duration. This may reflect our evolving perceptions and management strategies, with patients with a relatively new onset of disease experiencing a more aggressive treatment approach for both, their RA and associated comorbidities.

    Overall, conventional risk calculators such as the FRS, NCEP, SCORE (TC:HDL) and RRS are reliable and have a good degree of accuracy in the general population. However, such tools have never been properly validated in chronic inflammatory conditions such as RA, where CVD risk is elevated. The validity of some of the conventional CVD risk assessment tools may also be questionable in the older people (eg, >75 for the FRS), while the FRS and NCEP algorithms have been shown to underestimate risk among women and may miss approximately a third of at-risk women.29 These problems may be particularly relevant in a condition such as RA where there is a strong (3:1) female preponderance and many patients are older. New, gender-specific prediction tools have recently been developed by the Framingham heart study researchers but they require further validation, particularly within specific populations such as RA.30

    In RA, there is likely to be a further underestimation of CVD risk by conventional calculators (with the exception of RRS), as they do not take into account the impact of inflammation.31 It has been hypothesised that systemic inflammation may play a more significant role in the development of CVD than traditional risk factors in RA.32 Multiple studies and recent meta-analyses suggest that the added risk RA confers is in the region of 1.5–1.7-fold,14 thus expert bodies, such as the EULAR task force16 have suggested the application of a 1.5 multiplier to each risk stratification method, at least to specific patient subgroups, to account for this. In the present population, this approach led to the identification of considerably more at risk patients, with increases ranging from 4% to 49%; despite this however, over half of the total population was still not at high risk, and this may argue against the ‘blanket’ usage of statins in all patients with RA. These findings clearly indicate the need for widely acceptable guidelines while RA-specific risk calculators are developed. One previous study proposed an algorithm for risk prediction among patients with chronic inflammatory diseases,33 but although it recommended minimising disease activity and glucocorticoid use among those found to be at risk of developing CVD, it did not include inflammation as a parameter when calculating risk. While the field evolves, a pragmatic approach may be to systematically screen all patients with RA using the nationally recommended risk stratification system with a ×1.5 multiplier in applicable patients, as suggested by expert bodies. In addition, audit must be implemented to ensure that predetermined treatment targets are reached and adjust therapy as necessary.

    This study is the first to assess in great detail cardiovascular risk stratification in the context of a large, well characterised, RA population with established disease. It has highlighted the need for a more dynamic approach to managing the burden of CVD in RA. However, the study has limitations. Its cross-sectional nature and absence of a non-RA control group makes it impossible to show how ‘stable’ risk stratification is, using these calculators, in the context of the changing inflammatory activity of the disease over time in a given individual. In addition, we did not have sufficient data regarding previous statin use to account for this in our analyses. Thus, some high-risk patients who appear to be untreated with statin therapy may have previously been treated with statins but have not been able to tolerate them due to an adverse reaction. However, the number of patients falling in to this category are likely to be small, particularly in light of our recently published findings regarding statin myopathy in RA.34 The study was performed in a single UK centre, so the findings regarding statin underutilisation may be location or system specific. Most importantly, this study does not provide any evidence that systematic risk stratification and primary prevention strategies encompassing statin use would actually reduce future CVD events in RA. This needs to be addressed prospectively in studies developed specifically for the purpose.

    References

    1. ↵
      1. Pincus T,
      2. Callahan LF
      . Taking mortality in rheumatoid arthritis seriously – predictive markers, socioeconomic status and comorbidity. J Rheumatol 1986;13:841–5.
      OpenUrlPubMedWeb of Science
    2. ↵
      1. Maradit-Kremers H,
      2. Crowson CS,
      3. Nicola PJ,
      4. et al
      . Increased unrecognized coronary heart disease and sudden deaths in rheumatoid arthritis: a population-based cohort study. Arthritis Rheum 2005;52:402–11.
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Kvien TK
      . Epidemiology and burden of illness of rheumatoid arthritis. Pharmacoeconomics 2004;22(2 Suppl 1):1–12.
      OpenUrl
    4. ↵
      1. Kremers HM,
      2. Crowson CS,
      3. Therneau TM,
      4. et al
      . High ten-year risk of cardiovascular disease in newly diagnosed rheumatoid arthritis patients: a population-based cohort study. Arthritis Rheum 2008;58:2268–74.
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      1. Wilson PW,
      2. D'Agostino RB,
      3. Levy D,
      4. et al
      . Prediction of coronary heart disease using risk factor categories. Circulation 1998;97:1837–47.
      OpenUrlAbstract/FREE Full Text
    6. ↵
      1. Expert Panel on Detection,
      2. Evaluation,
      3. and Treatment of High Blood Cholesterol in Adults
      . Executive Summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III). JAMA 2001;285:2486–97.
      OpenUrlCrossRefPubMedWeb of Science
    7. ↵
      1. Hippisley-Cox J,
      2. Coupland C,
      3. Vinogradova Y,
      4. et al
      . Performance of the QRISK cardiovascular risk prediction algorithm in an independent UK sample of patients from general practice: a validation study. Heart 2008;94:34–9.
      OpenUrlAbstract/FREE Full Text
    8. ↵
      1. Woodward M,
      2. Brindle P,
      3. Tunstall-Pedoe H
      ; SIGN group on risk estimation. Adding social deprivation and family history to cardiovascular risk assessment: the ASSIGN score from the Scottish Heart Health Extended Cohort (SHHEC). Heart 2007;93:172–6.
      OpenUrlAbstract/FREE Full Text
    9. ↵
      1. Conroy RM,
      2. Pyörälä K,
      3. Fitzgerald AP,
      4. et al.
      ; SCORE project group. Estimation of ten-year risk of fatal cardiovascular disease in Europe: the SCORE project. Eur Heart J 2003;24:987–1003.
      OpenUrlAbstract/FREE Full Text
    10. ↵
      1. Gomes MC
      . C-reactive protein: a new golden marker of cardiovascular risk. Rev Port Cardiol 2002;21:1329–1346.
      OpenUrlPubMed
    11. ↵
      1. Shah SH,
      2. Newby LK
      . C-reactive protein: a novel marker of cardiovascular risk. Cardiol Rev 2003;11:169–79.
      OpenUrlCrossRefPubMed
    12. ↵
      1. Ridker PM,
      2. Buring JE,
      3. Rifai N,
      4. et al
      . Development and validation of improved algorithms for the assessment of global cardiovascular risk in women: the Reynolds Risk Score. JAMA 2007;297:611–19.
      OpenUrlCrossRefPubMedWeb of Science
    13. ↵
      1. Ridker PM,
      2. Paynter NP,
      3. Rifai N,
      4. et al
      . C-reactive protein and parental history improve global cardiovascular risk prediction. The Reynolds Risk Score for men. Circulation 2008;118:2243–51.
      OpenUrlAbstract/FREE Full Text
    14. ↵
      1. Aviña-Zubieta JA,
      2. Choi HK,
      3. Sadatsafavi M,
      4. et al
      . Risk of cardiovascular mortality in patients with rheumatoid arthritis: a meta-analysis of observational studies. Arthritis Rheum 2008;59:1690–7.
      OpenUrlCrossRefPubMedWeb of Science
    15. ↵
      1. Nurmohamed MT
      . Cardiovascular risk in rheumatoid arthritis. Autoimmun Rev 2009;8:663–7.
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      1. Peters MJ,
      2. Symmons DP,
      3. McCarey D,
      4. 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.
      OpenUrlAbstract/FREE Full Text
    17. ↵
      1. Arnett FC,
      2. Edworthy SM,
      3. Bloch DA,
      4. et al
      . The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31:315–24.
      OpenUrlCrossRefPubMedWeb of Science
    18. ↵
      1. Panoulas VF,
      2. Douglas KM,
      3. Milionis HJ,
      4. et al
      . Prevalence and associations of hypertension and its control in patients with rheumatoid arthritis. Rheumatology (Oxford) 2007;46:1477–82.
      OpenUrlAbstract/FREE Full Text
    19. ↵
      1. Panoulas VF,
      2. Douglas KM,
      3. Stavropoulos-Kalinoglou A,
      4. et al
      . Long-term exposure to medium-dose glucocorticoid therapy associates with hypertension in patients with rheumatoid arthritis. Rheumatology (Oxford) 2008;47:72–5.
      OpenUrlAbstract/FREE Full Text
    20. ↵
      1. Stavropoulos-Kalinoglou A,
      2. Metsios GS,
      3. Panoulas VF,
      4. et al
      . Cigarette smoking associates with body weight and muscle mass of patients with rheumatoid arthritis: a cross-sectional, observational study. Arthritis Res Ther 2008;10:R59.
      OpenUrlCrossRefPubMed
    21. ↵
      1. Prevoo ML,
      2. van't Hof MA,
      3. Kuper HH,
      4. et al
      . Modified disease activity scores that include twenty-eight-joint counts. Development and validation in a prospective longitudinal study of patients with rheumatoid arthritis. Arthritis Rheum 1995;38:44–8.
      OpenUrlCrossRefPubMedWeb of Science
    22. ↵
      1. Kirwan JR,
      2. Reeback JS
      . Stanford Health Assessment Questionnaire modified to assess disability in British patients with rheumatoid arthritis. Br J Rheumatol 1986;25:206–9.
      OpenUrlAbstract/FREE Full Text
    23. ↵
      Dudley group of hospitals MHD foundation trust. Trial of atorvastatin in rheumatoid arthritis. http://www.dgoh.nhs.uk/tracera/ 2009 (accessed January 2010).
    24. ↵
      1. Ridker PM,
      2. Paynter NP,
      3. Rifai N,
      4. et al
      . C-reactive protein and parental history improve global cardiovascular risk prediction: the Reynolds Risk Score for men. Circulation 2008;118:2243–51.
      OpenUrlAbstract/FREE Full Text
    25. ↵
      British Cardiac Society; British Hypertension Society; Diabetes UK; HEART UK; Primary Care Cardiovascular Society; Stroke Association. JBS 2: Joint British Societies' guidelines on prevention of cardiovascular disease in clinical practice. Heart 2005;91:v1–52.
      OpenUrlFREE Full Text
    26. ↵
      1. De Backer G,
      2. Ambrosioni E,
      3. Borch-Johnsen K,
      4. et al.
      ; Third Joint Task Force of European Other Societies on Cardiovascular Disease Prevention in Clinical Practice. European guidelines on cardiovascular disease prevention in clinical practice. Third Joint Task Force of European and Other Societies on Cardiovascular Disease Prevention in Clinical Practice. Eur Heart J 2003;24:1601–10.
      OpenUrlFREE Full Text
    27. ↵
      1. Fornasini M,
      2. Brotons C,
      3. Sellarès J,
      4. et al
      . Consequences of using different methods to assess cardiovascular risk in primary care. Fam Pract 2006;23:28–33.
      OpenUrlAbstract/FREE Full Text
    28. ↵
      1. Treharne GJ,
      2. Douglas KM,
      3. Iwaszko J,
      4. et al
      . Polypharmacy among people with rheumatoid arthritis: the role of age, disease duration and comorbidity. Musculoskeletal Care 2007;5:175–90.
      OpenUrlCrossRefPubMed
    29. ↵
      1. Hemann BA,
      2. Bimson WF,
      3. Taylor AJ
      . The Framingham Risk Score: an appraisal of its benefits and limitations. Am Heart Hosp J 2007;5:91–6.
      OpenUrlPubMed
    30. ↵
      1. D'Agostino RB Sr,
      2. Vasan RS,
      3. Pencina MJ,
      4. et al
      . General cardiovascular risk profile for use in primary care: the Framingham Heart Study. Circulation 2008;117:743–53.
      OpenUrlAbstract/FREE Full Text
    31. ↵
      1. Wilson PW,
      2. Nam BH,
      3. Pencina M,
      4. et al
      . C-reactive protein and risk of cardiovascular disease in men and women from the Framingham Heart Study. Arch Intern Med 2005;165:2473–8.
      OpenUrlCrossRefPubMedWeb of Science
    32. ↵
      1. Roman MJ,
      2. Moeller E,
      3. Davis A,
      4. et al
      . Preclinical carotid atherosclerosis in patients with rheumatoid arthritis. Ann Intern Med 2006;144:249–56.
      OpenUrlCrossRefPubMedWeb of Science
    33. ↵
      1. Hall FC,
      2. Dalbeth N
      . Disease modification and cardiovascular risk reduction: two sides of the same coin? Rheumatology (Oxford) 2005;44:1473–82.
      OpenUrlAbstract/FREE Full Text
    34. ↵
      1. Toms TE,
      2. Smith JP,
      3. Panoulas VF,
      4. et al
      . Prevalence of risk factors for statin-induced myopathy in rheumatoid arthritis patients. Musculoskeletal Care 2010;In press.

    Footnotes

    • Funding This work is supported by an Arthritis Research Campaign Clinical Fellowship grant (grant number 18848 to TET) and an Arthritis Research Campaign infrastructure support grant (grant number 17682, given to the Dudley Group of Hospitals NHS Foundation Trust, Department of Rheumatology). VFP is supported by a PhD scholarship from the Empirikion Institute, Athens, Greece.

    • Competing interests None.

    • Patient consent Obtained.

    • Ethics approval This study was conducted with the approval of the Black Country ethics committee.

    • Provenance and peer review Not commissioned; externally peer reviewed.