JOHN M. DAVIS, III, MD,

Download PDF View Table of Contents

Clinical Fellow in Rheumatology,
Division of Rheumatology;
HILAL MARADIT-KREMERS, MD, MSc,
Assistant Professor of Epidemiology,
Department of Health Sciences Research;
SHERINE E. GABRIEL, MD, MSc,
Professor of Epidemiology and Medicine,
Department of Health Sciences Research,
Mayo Clinic,
200 First Street SW
Rochester, MN 55905, USA.
 
Address reprint requests to Dr. S.E. Gabriel.

People with rheumatoid arthritis (RA) experience an increased burden of cardiovascular disease (CVD) and reduced survival compared to the general population¹. Myocardial infarction (MI) and heart failure appear to be more prevalent in RA compared to subjects without RA². In one study, the risk of congestive heart failure was 2-fold higher among RA subjects compared to non-RA controls³. Much of the best evidence suggests that systemic inflammation plays an important role in the pathogenesis of CVD in RA¹.

Still, a nagging question persists in the minds of many rheumatologists today: What is the role of glucocorticoids (GC)? A common view is that GC harm the cardiovascular system. We wish to examine an alternative hypothesis: that GC might actually reduce the risk of CVD in patients with RA. Our aim is not to provide a comprehensive review of the cardiovascular effects of GC but rather to bring balance to the debate of how use of GC might affect the development of CVD in patients with RA.

A discussion of GC is highly relevant given current rheumatology practice. GC have been used to treat people with RA for the last half-century, and recently, there has been renewed interest in these medications⁴. Use of GC is highly prevalent among people with RA; for example, these are taken by 30–50% of patients enrolled in recent therapeutic trials⁵. Rheumatologists commonly use GC as "bridge therapy" awaiting efficacy of slow-acting disease modifying antirheumatic drugs (DMARD), and also for treating disease flares. Frequently, delays occur in initiating biologic agents due to time required for insurance company authorization, tuberculin skin testing, and/or patient education visits. For these reasons, GC will likely continue to be an important component of the armamentarium for patients with RA, at least for the foreseeable future.

Despite years of use, relatively little is known regarding the longterm effects of GC, particularly on the development of CVD⁴. In light of the increased burden of CVD in RA, it is critical to better understand how GC affect the development of CVD in this population, whether beneficial or harmful.

Inflammation plays a fundamental role in the pathogenesis of CVD in RA. For example, abnormalities in traditional cardiovascular risk factors alone cannot explain the increased risk of CVD in this population⁶. Also, clinical features in RA that are associated with high inflammatory disease activity such as elevated acute phase markers and extraarticular manifestations are associated with cardiovascular events and mortality^3,7-10. Theoretically, GC could modulate the risk of CVD in RA in 2 competing ways. The risk could increase due to potentially deleterious effects of GC on lipids, glucose tolerance, hypertension, or obesity¹¹. Alternatively, by alleviating systemic inflammation, GC may actually decrease the risk of atherosclerosis and CVD^1,12. Evidence supporting both directions of effect will be reviewed. Furthermore, there is evidence that effective treatment of systemic inflammation with DMARD is associated with lower risk of cardiovascular mortality¹³. Treatment with antagonists of tumor necrosis factor-alpha (TNF-a) may lower the risk of heart failure¹⁴. Therefore, it is reasonable to consider the possibility that GC may also reduce the risk of cardiovascular events and mortality.

A number of novel findings in epidemiological studies suggest that "flares" of inflammation in RA may be involved in triggering cardiovascular events. First of all, the risk of cardiovascular events is independent of the duration of RA⁸. Second, the risk of sudden death and silent MI appears to be increased very early, even prior to patients fulfilling American College of Rheumatology (ACR) classification criteria⁷. Third, there is new evidence that the erythrocyte sedimentation rate (ESR) rises significantly in the 6-month period immediately prior to onset of heart failure in patients with RA⁹. Similarly, increases in systemic inflammation due to acute infections have been linked with a short term increased risk of MI and stroke in non-RA subjects¹⁵. These associations between episodic fluctuations in level of systemic inflammation and onset of cardiovascular events suggest the possibility that using GC to control flares of disease activity may actually lower the risk of cardiovascular events.

PROBLEMS WITH THE "EVIDENCE" AGAINST GC IN RA

Several authors who have written on this subject over the years have concluded that GC likely cause CVD^11,16-18. However, fair questions can be raised regarding the "evidence" that GC are associated with an increased risk of CVD, at least in the context of RA. Several of these review papers cite mechanistic studies that assess in vitro or animal models¹⁹, or studies of other disease states (i.e., Cushing's disease²⁰, systemic lupus erythematosus^21,22, and asthma²³), to support the hypothesis that GC have harmful cardiovascular effects. Frequently, these studies contain a heterogeneous sample of patients treated for many different rheumatic and non-rheumatic diseases^23,24 and often include otherwise healthy subjects. Therefore, it is unclear how relevant this literature is to RA. Nashel¹⁶ described a paper by Kalbak²⁵ in 1972 as "the strongest evidence that longterm corticosteroid therapy is important in the development of arteriosclerosis." This study found a higher rate of radiographic arterial calcification at the ankle among patients with RA treated with steroids compared to patients not treated²⁵. However, more clinically relevant events such as MI, heart failure, or death were not assessed. Additionally, no information regarding confounding variables was provided. While this sort of information is compelling and warrants further study, it is no substitute for longitudinal studies with objective clinical endpoints.

In the RA setting and in GC users overall, epidemiological studies conflict regarding the risk of CVD associated with GC exposure. Several studies have found potentially harmful effects of GC^26-29 whereas others have found either no effect or beneficial effects^8,12,30-32. Table 1 shows selected studies that have assessed the association of GC exposure and CVD in RA. It is apparent from these study findings that the magnitude and even the direction of the effect of GC on cardiovascular morbidity and mortality are unclear.

Table 1. Epidemiologic studies examining the association of GC exposure and cardiovascular outcomes in RA. Confidence intervals (95%) are shown in brackets.

The most significant limitation of observational study designs for this question is the potential for confounding by indication (also called channeling bias)^17,33. For patients with RA, this implies that the association between GC exposure and CVD is not causal but rather operates through a third factor: disease activity/severity. In other words, people with more active and/or severe RA receive more GC and are also more likely to develop CVD. To understand this, consider an example. A patient with longstanding RA with no cardiovascular history develops a flare of polyarthritis with a high ESR. If the patient has a MI a week after starting GC, then it might seem as though GC were the cause. However, in light of current evidence, it is also possible that the "flare" triggered the MI. A second example: a patient with seropositive, erosive RA has received a large cumulative dose of GC over many years. The patient has congestive heart failure. Could the large dose of GC be to blame, or the many disease flares over the years for which the GC were prescribed?

Although confounding by indication is widely acknowledged as a limitation of observational study designs for RA treatment effects, few studies have attempted to measure the extent of this bias. Wolfe and others showed that patients with RA or osteoarthritis who were switched to cyclooxygenase-2 inhibitors when they became available in 1999 had clinically important increases on several measures of disease severity compared to those who remained on conventional nonsteroidal antiinflammatory drugs (NSAID)³⁴. Erkan and colleagues surveyed US rheumatologists and found a strong preference to prescribe GC as first-line therapy to RA patients with higher disease activity/severity, irrespective of whether medication cost was considered a factor³⁵. These data provide support that confounding by indication is a clinically important phenomenon in observational studies of GC effects in RA.

Investigators have attempted to account for this confounding by statistically modeling the association between GC exposure and cardiovascular events, adjusting for a broad range of variables that are markers of RA severity: for example, rheumatoid factor (RF), tender/swollen joint counts, erosive disease, Health Assessment Questionnaire scores, and ESR. Despite this technique, studies have reported divergent effects of GC on cardiovascular risk albeit with varying degrees of adjustment (see Table 1 for model descriptions). Wolfe and Michaud recently reported that GC exposure predicted MI in the following 6 months in a large cohort of patients with RA²⁹. Wallberg-Jonsson and colleagues showed complex effects of GC exposure^12,32. Early use in the first 3 years was associated with a higher risk of cardiovascular events; in contrast, cumulative GC use more than one year before cardiovascular events was associated with a lower risk in a subset of patients who experienced cardiovascular events during followup.

Wei and others²⁶ recently examined GC exposure (defined by prescription records) and the risk of subsequent cardiovascular events. Their large sample was heterogeneous, including many different disease processes; a subset with inflammatory arthritis (n = 1165) was defined by taking subjects who had prescriptions for NSAID and DMARD or by hospitalization before study entry. High dose (> 7.5 mg/day) GC use was associated with increased risk of heart failure, MI, stroke, and overall mortality. This was also true for the subset of patients with inflammatory arthritis. These results were adjusted for a broad range of medications, including antirheumatic drugs.

These analyses were certainly interesting, but although Wei and colleagues argued otherwise, confounding by indication could have affected the results. Sensitivity analyses suggested that the prevalence of such a confounder would not be sufficiently different between GC exposed and unexposed groups. Additionally, it was suggested that the risk associated with an unmeasured confounder would not be sufficiently high to explain the results. However, this argument is precarious, particularly for those with inflammatory arthritides. For example, the presence of RF in RA has been associated with a 2-fold increased risk of congestive heart failure³. Significant differences in the prevalence of RF could exist between users of high dose GC and non-users of GC within a heterogeneous sample of patients with inflammatory arthritis. Moreover, the possibility of other unknown inflammatory mediators associated with the disease activity requiring GC treatment as well as cardiovascular outcomes was overlooked. Therefore, the potential for confounding by unmeasured factors remains a significant concern in this study.

In another study, del Rincon and colleagues recently used high-resolution ultrasonography to examine the effects of GC on arteries of patients with RA³⁶. These subjects were examined annually for a median of 39 months before the sonographic examination; joint counts were measured and averaged for these visits. They attempted to control for disease severity by adjusting for joint counts, rheumatoid nodules, RF positivity, and ESR. Among the group of patients in the highest tertile of cumulative GC exposure (> 16 g prednisone), the prevalence of both arterial plaque and incompressible arteries was higher than the unexposed group. Duration of GC exposure was not associated with arterial plaque, intima-media thickness, or incompressible vessels. However, markers of RA severity were measured during a few visits during the study, yet these large GC doses were accumulated over many years of followup (mean 7.5 years, range 4 months to 52 years). It is certainly possible that unmeasured factors of disease severity confounded the results of this study. Also, it is notable that another study reported no association between prednisolone use and carotid intima-media thickness³⁷. Ultimately, each of these studies mentioned herein does not adequately answer the question of whether the association between GC use and arterial atherosclerosis is causal or mediated through the association with disease activity/severity.

From these studies, several conclusions are evident. First, therapeutic factors such as the timing of GC exposure and/or patient factors such as personal history of CVD may significantly modulate the effect of GC on the cardiovascular system. Second, these studies exemplify the difficulties in disentangling the complex effects of GC from RA disease activity/severity in pharmacoepidemiological studies. Randomization to GC versus placebo in a longterm clinical trial powered to detect cardiovascular outcomes would be the most valid study design for this question. However, such a trial would be difficult to conduct and might be unfeasible. Therefore, creative study designs and/or statistical methodologies to address the issue of confounding by indication are necessary in order to better understand the true effect of GC on cardiovascular risk in RA.

BIOCHEMICAL EFFECTS OF GC: POTENTIALLY BENEFICIAL IN RA?

Much has been written of the potentially harmful effects of GC at the biochemical/molecular level. For example, in a comprehensive review by Girod, et al¹⁸ numerous deleterious effects of GC were presented, including exacerbations in blood pressure, body composition (i.e., decreased muscle mass, increased visceral adiposity), insulin resistance, plasma lipoproteins (i.e., increased low density lipoprotein, increased triglycerides), and hemostatic factors (i.e., increased plasminogen activator inhibitor 1). The authors concluded that the net effect of GC on the cardiovascular system is likely to be harmful.

However, the GC effects in the setting of RA are possibly much different. This is because inflammatory mediators and their downstream consequences are likely to be more important in the etiology of CVD in patients with RA compared to the general population. For example, Sattar, et al³⁸ noted the striking difference in high-sensitivity C-reactive protein elevations between patients with RA and members of the general population. These authors raised the theory, which is supported well by current evidence (discussed above), that the systemic inflammatory response in RA drives the development of CVD. Inflammatory mediators that have "pleiotropic effects" were suggested to act distantly from joints on the liver, skeletal muscle, adipose, and endothelium, leading to abnormalities in many important cardiovascular factors and ultimately to cardiovascular events. If inflammatory mediators mitigate part of the cardiovascular risk associated with RA, then antagonism of such mediators by GC might lead to reduced risk of CVD. The following are examples of how GC effects might be beneficial among patients with RA.

Traditional cardiovascular risk factors may respond to GC contrary to their predicted effects. For example, exacerbation of insulin resistance has been suggested as a mechanism by which GC harm the cardiovascular system. Actually, RA itself has been associated with peripheral insulin resistance, which seems to be linked to systemic inflammation^39,40. This may be cytokine-mediated; for example, TNF-a, an important proinflammatory cytokine in RA, has the ability to induce insulin resistance by inhibiting signal transduction through the insulin receptor⁴¹. In small short term studies, successful treatment with GC led to amelioration of systemic inflammation and normalization of glucose handling^39,40.

Another example is dyslipidemia in RA, which is similar to that observed in insulin resistance states. This has been described as an "atherogenic lipid profile," with lower than normal cholesterol in all lipoprotein categories, and a high ratio of total cholesterol to high density lipoprotein cholesterol^42,43. This profile has been associated with active systemic inflammation in RA⁴³. Therapy with regimens that included GC was shown to reduce systemic inflammation in RA and ameliorate the lipid abnormalities^42,43. Similarly, lipoprotein(a) was shown to be higher among patients with RA compared to controls, and lipoprotein(a) levels were lower in patients with RA receiving GC compared to those not, although this observation was not statistically significant⁴⁴.

Inflammatory mediators are included among the novel cardiovascular risk factors currently being investigated¹⁸. Serum levels of cytokines (i.e., interleukin-6 and TNF⁴⁵), C-reactive protein⁴⁶, adhesion molecules (i.e., vascular cell adhesion molecule-1⁴⁶), and matrix metalloproteinases⁴⁷ are much higher in patients with RA compared to controls in small studies. GC treatment is predicted to cause inhibition of most of these mediators; however, data for the effect of GC on some specific mediators are lacking. GC treatment has been shown to inhibit expression of adhesion molecules in the synovium⁴⁸ as well as serum cytokine levels such as TNF-a⁴⁹. Homocysteine is a novel cardiovascular risk factor and prothrombotic factor found to be elevated in patients with RA, and treatment with high-dose pulse GC rapidly decreases their homocysteine levels⁵⁰.

Nitric oxide is another novel mediator that is important in endothelial function. Nitric oxide activity has been found to be reduced in RA; for example, arterial vasodilatation in response to nitric oxide stimulation was inhibited in patients with RA, suggesting that they have impaired responsiveness to nitric oxide in the vessel wall⁵¹. This correlated with inflammatory markers, again suggesting the role of inflammation in RA. Therapy with an antirheumatic regimen that included low dose prednisone improved the nitric oxide responsiveness⁵².

These data emphasize the difficulty in understanding the complex interactions between GC, novel and traditional cardiovascular risk factors, and biochemical and molecular mediators of atherosclerosis. Indeed, from biochemical data, it appears very difficult to predict the overall impact of GC on CVD in RA. Therefore, the need for valid clinical outcome data for CVD in patients with RA cannot be over-stated.

SUMMARY AND FUTURE DIRECTIONS

The true effect of GC on cardiovascular morbidity and mortality in the RA population remains unclear. In fact, even the direction of the GC effect is controversial: these may either increase or decrease risk. It is likely that the relationship between GC and CVD in RA is complex and multifactorial, as is the case with sepsis where there is evidence for either favorable or unfavorable effects depending on the context⁵³. As shown by previous examples, the effect may depend on characteristics of GC treatment such as daily dosage, timing of therapy, and cumulative steroid dose. It is important to consider confounding by RA disease activity/severity when considering the effects of GC on CVD. Patient factors such as personal history of coronary heart disease or RA characteristics may influence the magnitude and direction of the effect. Moreover, there may be different effects depending on what cardiovascular outcome is considered, for example, MI versus heart failure.

Disentangling these complex relationships is of paramount importance given the high prevalence of GC exposure in RA and the excess burden of CVD in this patient group. Such information will also help us to more appropriately address the concerns of our patients regarding side effects of GC. Better understanding of the clinical, biochemical, and molecular effects of GC on the pathogenesis of cardiovascular disease in RA will lead to more appropriate use of GC. Additional data from population-based studies are needed to extend our knowledge of the epidemiology and pathogenesis of heart disease in RA.

In the meantime, eradicating inflammation in RA appears important not only for the joints, but also for the longevity of the cardiovascular system. Use of GC in early RA and targeted use to treat flares continues to be reasonable and possibly beneficial. However, limiting chronic exposure and employing careful measures to prevent osteoporosis, infection, and other steroid-induced sequelae are critical.

ACKNOWLEDGMENT

We thank Cynthia Crowson and Paulo Nicola for their thoughtful reviews of the manuscript.

REFERENCES

Search PubMed for:

2. Wolfe F. Increase in cardiovascular and cerebrovascular disease prevalence in rheumatoid arthritis. J Rheumatol 2003;30:36-40. [MEDLINE]

3. Nicola P, Maradit Kremers H, Roger V, et al. The risk of congestive heart failure in rheumatoid arthritis: a population-based study over 46 years. Arthritis Rheum 2005;52:412-20. [MEDLINE]

4. Bijlsma JW, Boers M, Saag KG, Furst DE. Glucocorticoids in the treatment of early and late RA. Ann Rheum Dis 2003;62:1033-7. [MEDLINE]

5. Keystone EC, Kavanaugh AF, Sharp JT, et al. Radiographic, clinical, and functional outcomes of treatment with adalimumab (a human anti-tumor necrosis factor monoclonal antibody) in patients with active rheumatoid arthritis receiving concomitant methotrexate therapy: a randomized, placebo-controlled, 52-week trial. Arthritis Rheum 2004;50:1400-11. [MEDLINE]

6. del Rincon I, Williams K, Stern MP, Freeman GL, Escalante A. High incidence of cardiovascular events in a rheumatoid arthritis cohort not explained by traditional cardiovascular risk factors. Arthritis Rheum 2001;44:2737-45. [MEDLINE]

7. Maradit-Kremers H, Crowson C, Nicola P, et al. Increased unrecognized coronary heart disease in rheumatoid arthritis: a population-based cohort study. Arthritis Rheum 2005;52:402-11. [MEDLINE]

8. Maradit Kremers H, Nicola P, Crowson C, Ballman K, Gabriel S. Cardiovascular death in rheumatoid arthritis: a population-based study. Arthritis Rheum 2005;52:412-20. [MEDLINE]

9. Maradit Kremers H, Crowson C, Nicola P, Ballman KV, Gabriel S. Erythrocyte sedimentation rate (ESR) and congestive heart failure (CHF) in rheumatoid arthritis (RA) [abstract]. Arthritis Rheum 2004;50 Suppl:S557-S58.

10. Turesson C, O'Fallon WM, Crowson CS, Gabriel SE, Matteson EL. Occurrence of extraarticular disease manifestations is associated with excess mortality in a community based cohort of patients with rheumatoid arthritis. J Rheumatol 2002;29:62-7. [MEDLINE]

11. Maxwell SR, Moots RJ, Kendall MJ. Corticosteroids: do they damage the cardiovascular system? Postgrad Med J 1994;70:863-70. [MEDLINE]

12. Wallberg-Jonsson S, Johansson H, Ohman ML, Rantapaa-Dahlqvist S. Extent of inflammation predicts cardiovascular disease and overall mortality in seropositive rheumatoid arthritis. A retrospective cohort study from disease onset. J Rheumatol 1999;26:2562-71. [MEDLINE]

13. Choi HK, Hernan MA, Seeger JD, Robins JM, Wolfe F. Methotrexate and mortality in patients with rheumatoid arthritis: a prospective study. Lancet 2002;359:1173-7. [MEDLINE]

14. Wolfe F, Michaud K. Heart failure in rheumatoid arthritis: rates, predictors, and the effect of anti-tumor necrosis factor therapy. Am J Med 2004;116:305-11. [MEDLINE]

15. Smeeth L, Thomas SL, Hall AJ, Hubbard R, Farrington P, Vallance P. Risk of myocardial infarction and stroke after acute infection or vaccination. N Engl J Med 2004;351:2611-8. [MEDLINE]

16. Nashel DJ. Is atherosclerosis a complication of long-term corticosteroid treatment? Am J Med 1986;80:925-9. [MEDLINE]

17. Raynauld JP. Cardiovascular mortality in rheumatoid arthritis: how harmful are corticosteroids [editorial]? J Rheumatol 1997;24:415-6. [MEDLINE]

18. Girod JP, Brotman DJ. Does altered glucocorticoid homeostasis increase cardiovascular risk? Cardiovasc Res 2004;64:217-26. [MEDLINE]

19. Stamler J, Pick R, Katz L. Effects of cortisone, hydrocortisone and corticotropin on lipemia, glycemia and atherogenesis in cholesterol-fed chicks. Circulation 1954;10:237-46.

20. Colao A, Pivonello R, Spiezia S, et al. Persistence of increased cardiovascular risk in patients with Cushing's disease after five years of successful cure. J Clin Endocrinol Metab 1999;84:2664-72. [MEDLINE]

21. Jonsson H, Nived O, Sturfelt G. Outcome in systemic lupus erythematosus: a prospective study of patients from a defined population. Medicine 1989;68:141-50. [MEDLINE]

22. Bulkley BH, Roberts WC. The heart in systemic lupus erythematosus and the changes induced in it by corticosteroid therapy. A study of 36 necropsy patients. Am J Med 1975;58:243-64.

23. Jefferys D, Lessof M, Mattock M. Corticosteroid treatment, serum lipids and coronary artery disease. Postgrad Med J 1980;56:491-3. [MEDLINE]

24. Stern MP, Kolterman OG, Fries JF, McDevitt HO, Reaven GM. Adrenocortical steroid treatment of rheumatic diseases. Arch Int Med 1973;132:97-101.

25. Kalbak K. Incidence of arteriosclerosis in patients with rheumatoid arthritis receiving long-term corticosteroid therapy. Ann Rheum Dis 1972;31:196-200.

26. Wei L, MacDonald T, Walker B. Taking glucocorticoids by prescription is associated with subsequent cardiovascular disease. Ann Int Med 2004;141:764-70. [MEDLINE]

27. Souverein PC, Berard A, Van Staa TP, et al. Use of oral glucocorticoids and risk of cardiovascular and cerebrovascular disease in a population based case-control study. Heart 2004;90:859-65. [MEDLINE]

28. Wolfe F, Michaud K. Corticosteroids increase the risk of diabetes mellitus in RA and contribute to the risk of myocardial infarction and heart failure [abstract]. Ann Rheum Dis 2004;63 Suppl:SAT0379.

29. Wolfe F, Michaud K. Prednisone but not biologics or DMARDs is associated with increased risk of myocardial infarction in persons with RA [abstract]. Ann Rheum Dis 2004;63 Suppl:OP0039.

30. Suissa S, Bernatsky S, Hudson M, Kezouh A. DMARD use and the risk of acute myocardial infarction in rheumatoid arthritis [abstract]. Ann Rheum Dis 2004;63 Suppl:OP0038.

31. Bernatsky S, Hudson M, Suissa S. Anti-rheumatic drug use and risk of hospitalization for congestive heart failure in rheumatoid arthritis. Rheumatology Oxford 2005;44:677-80. [MEDLINE]

32. Wallberg-Jonsson S, Ohman ML, Dahlqvist SR. Cardiovascular morbidity and mortality in patients with seropositive rheumatoid arthritis in Northern Sweden. J Rheumatol 1997;24:445-51. [MEDLINE]

33. McMahon A. Approaches to combat with confounding by indication in observational studies of intended drug effects. Pharmacoepidemiol Drug Saf 2003;12:551-8. [MEDLINE]

34. Wolfe F, Flowers N, Burke TA, Arguelles LM, Pettitt D. Increase in lifetime adverse drug reactions, service utilization, and disease severity among patients who will start COX-2 specific inhibitors: quantitative assessment of channeling bias and confounding by indication in 6689 patients with rheumatoid arthritis and osteoarthritis. J Rheumatol 2002;29:1015-22. [MEDLINE]

35. Erkan D, Yazici Y, Harrison MJ, Paget SA. Physician treatment preferences in rheumatoid arthritis of differing disease severity and activity: the impact of cost on first-line therapy. Arthritis Care Res 2002;47:285-90. [MEDLINE]

36. del Rincon I, O'Leary DH, Haas RW, Escalante A. Effect of glucocorticoids on the arteries in rheumatoid arthritis. Arthritis Rheum 2004;50:3813-22. [MEDLINE]

37. Park YB, Ahn CW, Choi HK, et al. Atherosclerosis in rheumatoid arthritis: morphologic evidence obtained by carotid ultrasound. Arthritis Rheum 2002;46:1714-9. [MEDLINE]

38. Sattar N, McCarey DW, Capell H, McInnes IB. Explaining how "high-grade" systemic inflammation accelerates vascular risk in rheumatoid arthritis. Circulation 2003;108:2957-63. [MEDLINE]

39. Svenson KL, Lundqvist G, Wide L, Hallgren R. Impaired glucose handling in active rheumatoid arthritis: effects of corticosteroids and antirheumatic treatment. Metabolism 1987;36:944-8. [MEDLINE]

40. Hallgren R, Berne C. Glucose intolerance in patients with chronic inflammatory diseases is normalized by glucocorticoids. Acta Med Scand 1983;213:351-5. [MEDLINE]

41. Hotamisligil GS, Peraldi P, Budavari A, Ellis R, White MF, Spiegelman BM. IRS-1-mediated inhibition of insulin receptor tyrosine kinase activity in TNF-a- and obesity-induced insulin resistance. Science 1996;271:665-8. [MEDLINE]

42. Svenson KL, Lithell H, Hallgren R, Vessby B. Serum lipoprotein in active rheumatoid arthritis and other chronic inflammatory arthritides. Arch Intern Med 1987;147:1917-20. [MEDLINE]

43. Boers M, Nurmohamed MT, Doelman CJA, et al. Influence of glucocorticoids and disease activity on total and high density lipoprotein cholesterol in patients with rheumatoid arthritis. Ann Rheum Dis 2003;62:842-5. [MEDLINE]

44. Asanuma Y, Kawai S, Aoshima H, Kaburaki J, Mizushima Y. Serum lipoprotein(a) and apolipoprotein(a) phenotypes in patients with rheumatoid arthritis. Arthritis Rheum 1999;42:443-7. [MEDLINE]

45. Manicourt DH, Triki R, Fukuda K, Devogelaer JP, Nagant de Deuxchaiznes C, Thonar EJ. Levels of circulating tumor necrosis factor alpha and interleukin 6 in patients with rheumatoid arthritis. Relationship to serum levels of hyaluronan and antigenic keratan sulfate. Arthritis Rheum 1993;36:490-9. [MEDLINE]

46. Wong M, Toh L, Wilson A, et al. Reduced arterial elasticity in rheumatoid arthritis and the relationship to vascular disease risk factors and inflammation. Arthritis Rheum 2003;48:81-9. [MEDLINE]

47. Tchetverikov I, Ronday HK, van El B, et al. MMP profile in paired serum and synovial fluid samples of patients with rheumatoid arthritis. Ann Rheum Dis 2004;63:881-3. [MEDLINE]

48. Youssef PP, Triantafillou S, Parker A, et al. Effects of pulse methylprednisolone on cell adhesion molecules in the synovial membrane in rheumatoid arthritis. Reduced E-selectin and intercellular adhesion molecule 1 expression. Arthritis Rheum 1996;39:1970-9. [MEDLINE]

49. Youssef PP, Haynes DR, Triantafillou S, et al. Effects of pulse methylprednisolone on inflammatory mediators in peripheral blood, synovial fluid, and synovial membrane in rheumatoid arthritis. Arthritis Rheum 1997;41:761-7. [MEDLINE]

50. Lazzerini P, Capecchi PL, Bisogno S, Galeazzi M, Marcolongo R, Pasini FL. Reduction in plasma homocysteine level in patients with rheumatoid arthritis given pulsed glucocorticoid treatment. Ann Rheum Dis 2003;62:694-5.

51. Yki-Jarvinen H, Bergholm R, Leirisalo M. Increased inflammatory activity parallels increased basal nitric oxide production and blunted response to nitric oxide in vivo in rheumatoid arthritis. Ann Rheum Dis 2003;62:630-4. [MEDLINE]

52. Bergholm R, Leirisalo-Repo M, Vehkavaara S, Makimattila S, Taskinen MR, Yki-Jarvinen H. Impaired responsiveness to NO in newly diagnosed patients with rheumatoid arthritis. Arterioscler Thromb Vasc Biol 2002;22:1637-41. [MEDLINE]

53. Minneci PC, Deans KJ, Banks SM, Eichacker PQ, Natanson C. Meta-analysis: the effect of steroids on survival and shock during sepsis depends on the dose. Ann Intern Med 2004;141:47-56. [MEDLINE]

54. Jacobsson LT, Knowler WC, Pillemer S, et al. Rheumatoid arthritis and mortality. A longitudinal study in Pima Indians. Arthritis Rheum 1993;36:1045-53. [MEDLINE]

55. Wolfe F, Mitchell DM, Sibley JT, et al. The mortality of rheumatoid arthritis. Arthritis Rheum 1994;37:481-94. [MEDLINE]