- Email: [email protected]
Cardiovascular risk induced by low-dose corticosteroids in rheumatoid arthritis: A systematic literature review
Joint Bone Spine 78 (2011) 23–30
Review
Cardiovascular risk induced by low-dose corticosteroids in rheumatoid arthritis: A systematic literature review Adeline Ruyssen-Witrand a,∗ , Bruno Fautrel b , Alain Saraux c , Xavier Le Loët d , Thao Pham e a
Service de rhumatologie B, hôpital Cochin, AP–HP, université Paris-V René-Descartes, 27, rue du Faubourg-Saint-Jacques, 75014 Paris, France Hôpital Pitié-Salpêtrière, AP–HP, université Paris-VI Pierre-et-Marie-Curie, 47-83, boulevard de l’Hôpital, 75651 Paris cedex 13, France c Hôpital de la Cavale Blanche, CHU de Brest, boulevard Tanguy-Prigent, 29609 Brest, France d Rouen university Hospital and Inserm U905 (IFRMP23), University of Rouen, 1, rue de Germont, 76031 Rouen cedex, France e Hôpital de la Conception, AP–HM, 147, boulevard Baille, 13385 Marseille cedex 05, France b
a r t i c l e
i n f o
Article history: Accepted 3 February 2010 Available online 14 May 2010 Keywords: Corticosteroids Cardiovascular disease Rheumatoid arthritis Treatment Epidemiology
a b s t r a c t Objectives: To assess the association between cardiovascular (CV) risk and low-dose corticosteroids (LDCT, defined as a daily dose < 10 mg/day of prednisone) in rheumatoid arthritis (RA) patients. Methods: Data source: A systematic review of the literature up to June 2009 was performed. Data extraction: (1) cardiovascular risk factors: high blood pressure, glycemia and lipid profile, carotid intima-media thickness, pulse-wave velocity, ventricular function; (2) “hard” outcomes: heart failure (HF), stroke, myocardial infarction (MI) or mortality. Data analysis: descriptive, comparing CV risk between LD-CTtreated RA patients and LD-CT-non-treated RA patients. Results: Of the 1138 screened reports, the literature search identified 37 assessing CV risk in LD-CT treated RA. The analysis showed a protective effect on serum lipid profile, an increase of insulin resistance or glycemia, probably no effect on blood pressure, no effect on atherosclerosis, discrepancies regarding arterial stiffness and no effect on ventricular function or heart rate variability. An association of LD-CT with major CV events was found in 4/6 studies. This included MI (HR = 1.7 [1.2–2.3]), stroke (OR = 4.36 [1.60–11.90] for LDC between 6 and 10 mg/day), mortality (HR = 2.03 [1.25-3.32]) and a composite index of CV events (in the group of rheumatoid factor positive RA, HR = 2.21 [1.22–4.00]). Two studies did not find any significant association between LD-CT exposure and mortality (OR = 2.25 [0.29–102.5]) or a composite index of CV events (OR = 1.3 [0.8–2.0]). Conclusion: Although the literature review showed poor association between LDC exposure and CV risk factors, a trend of increasing major CV events was identified. © 2010 Société franc¸aise de rhumatologie. Published by Elsevier Masson SAS. All rights reserved.
Rheumatoid arthritis (RA) is a chronic inflammatory disease that leads to severe joint destruction. Beside structural evolution, patients with RA are at risk of developing cardiovascular (CV) diseases including myocardial infarction (MI), heart failure (HF), as well as increased CV mortality [1–3]. Chronic inflammation that occurs in this population is believed to contribute to the increased risk [4–6]. Corticosteroids (CT) are powerful anti-inflammatory agents and have been used since the 1950s in the symptomatic treatment of RA. However, it is known that long-term CT use at high dose have considerable toxicity including effects on blood pressure, insulin resistance, lipid profile, body weight and fat distribution [7,8] that might significantly increase CV risk [9,10]. The risk of CV
∗ Corresponding author. Tel.: +33158412562; fax: +335843549256. E-mail address: [email protected] (A. Ruyssen-Witrand).
events associated with CT has already been demonstrated and is dose-dependant [11]. Therefore, alternative treatment approaches have been developed in RA, using conventional disease-modifying antirheumatic drugs (DMARDs) and more recently biologic agents, which can control RA activity and thus enable dose reduction of CT. CT may also have cardioprotective effects mediated by their anti-inflammatory and antiproliferative actions in the vessel wall [12,13]. However the CV risk with CT use < 10 mg per day has not been clearly demonstrated and based on its beneficial effects of CT in controlling chronic inflammation, low-dose CT (LD-CT) on CV risk is still a subject of controversy. The aim of this study is to assess through a systematic literature review the potential association between CV risk and the use of long-term LD-CT defined as an average daily dose below 10 mg per day, in RA.
1297-319X/$ – see front matter © 2010 Société franc¸aise de rhumatologie. Published by Elsevier Masson SAS. All rights reserved. doi:10.1016/j.jbspin.2010.02.040
24
A. Ruyssen-Witrand et al. / Joint Bone Spine 78 (2011) 23–30
1. Methods 1.1. Literature search A systematic literature search was performed using databases of Medline, Embase and Cochrane Library. The keywords used for the Medline search were: (“Arthritis, Rheumatoid/epidemiology”[Mesh:NoExp] or “Arthritis, Rheumatoid/ultrasonography” [Mesh:NoExp] or “Arthritis, Rheumatoid/complications” [Mesh:NoExp] or “Arthritis, Rheumatoid/drug therapy”[Mesh:NoExp] or “Arthritis, Rheumatoid/mortality”[Mesh:NoExp]) and (“Cardiovascular Diseases”[Mesh:NoExp] or “Heart Diseases”[Mesh:NoExp] or “Cardiomegaly”[Mesh] or “Cardiomyopathies”[Mesh] or “Heart Arrest”[Mesh] or “Heart Failure”[Mesh] or “Myocardial Ischemia”[Mesh] or “Ventricular Dysfunction”[Mesh] or “Cardiac Output, Low”[Mesh] or “Vascular Diseases” [Mesh:NoExp] or “Aneurysm”[Mesh] or “Aortic Diseases”[Mesh:NoExp] or “Aortic Aneurysm”[Mesh] or “Aortic Arch Syndromes”[Mesh] or “Arterial Occlusive Diseases” [Mesh] or “Cerebrovascular Disorders”[Mesh] or “Diabetic Angiopathies”[Mesh] or “Embolism and Thrombosis”[Mesh] or “Hypertension”[Mesh] or “Ischemia”[Mesh] or “Peripheral Vascular Diseases”[Mesh] or “Diabetes Mellitus”[Mesh]) NOT “Case Reports”[ptyp]. The search was limited to English, French, Spanish and German languages, humans and adults above 19 years old with no limit of date. For the Embase search, the keywords used were “Cardiovascular” and “rheumatoid arthritis” limited to English language and humans with no limit of date. For the Cochrane Library search, the keywords used were “Cardiovascular” and “rheumatoid arthritis” with no limits. The literature search was performed on December 1st 2007 and the database search was regularly updated until June 15th 2009. Abstracts from international congresses including American College of Rheumatology (ACR), European League Against Rheumatism (EULAR) and French national congress société franc¸aise de rhumatologie (SFR) from 2006 to 2009 were also investigated. Manual search was performed when reading references from studies included or from reviews from expert’s opinion when they were recently published (after 2000s) or using related links tool of Medline.
of the populations studied including disease activity and severity, rheumatoid factor and anti-CCP antibodies, cardiovascular pathologies, smoking status, diabetes mellitus, body mass index or high blood pressure were collected. The proportion of patients receiving LD-CT in the study was assessed with the current daily dose used and if mentioned, the cumulative dose and the duration of LDC therapy. The dose of CT used was expressed in milligrams (daily dose) or grams (cumulative dose) of prednisone. Concomitant therapies including non-steroidal anti-inflammatory drugs (NSAIDs), DMARDs or biologic agents when taken were noted. 1.4. Data analysis All comparisons between RA patients with or without LD-CT were analysed. Comparison could be given in percentage (binary outcome) or in mean (continuous outcome) in each group or could be assessed through a logistic regression or a Cox model with odds ratios (OR) or hazard ratios (HR). The type of comparison performed was analysed including LD-CT as a binary variable (Yes/No) or a continuous variable (current daily dose or cumulative dose) and the adjustment on potential confounding factors such as cardiovascular diseases, co-morbidities, cardiovascular risk factors, disease activity or disease severity. 2. Results The literature search on electronic databases identified 1138 titles (Fig. 1) and 277 articles were selected after reviewing the titles and abstracts. Finally, 34 studies were included in the analysis and three abstracts from international congresses were added. 2.1. Impact on the main cardiovascular risk factors Seven articles studied the risk of dyslipidaemia. One [14] of these articles was a randomised controlled trial (RCT) comparing oral prednisone (7.5 mg/day) versus placebo associated with a DMARD chosen by the physician. The authors did not show any difference of lipid serum levels between groups but 63% of patients in the prednisone arm had stopped the treatment before the end of the follow-up. Four studies [15–18] showed an increase of high-density
1.2. Articles selection Only reports assessing RA were selected. To be included in the selection, the articles had to compare two groups of RA patients: one receiving LD-CT defined as a daily dose below 10 mg per day or if the daily dose was not available, a cumulative dose under 5 g and another group without CT. The outcomes studied were divided in three: the “hard outcomes” or major CV events including mortality, myocardial infarction, stroke or heart failure, the “CV surrogate markers” such as atherosclerosis, arterial stiffness, subclinical heart failure, heart rate variability and the cardiovascular traditional risk factors such as high blood pressure, diabetes mellitus, lipids concentrations, homocystein levels or metabolic syndrome. Reviews and letters were used for the manual search and then excluded of the analysis. Articles with no study of effect of CT on CV risk, where the daily dose was not mentioned or the dose was above 10 mg per day were excluded. 1.3. Data extraction The data were extracted using a standardised form by one author (ARW). Beside general information about the article, the design of the study was extracted including randomised controlled trials, prospective observational studies, retrospective studies or studies using national databanks. The outcome assessed, the characteristics
Fig. 1. Flow-chart of inclusion procedure. RA: rheumatoid arthritis; CT: corticosteroids.
A. Ruyssen-Witrand et al. / Joint Bone Spine 78 (2011) 23–30
lipoproteins and one study [19] reported decrease in total cholesterol, low-density lipoproteins and triglycerides. One article [19] described fat distribution assessed by Dual X-Ray Absorptiometry in 43 RA patients and found a significant increase in abdominal fat distribution in men treated with LD-CT compared to men without LD-CT (data not shown). Seven articles studied the risk of diabetes mellitus induced by LD-CT. One study [20] found an association with exposure to 1 to 5 mg per day of prednisone with a hazard ratio (HR) of 2.1 [1.6–2.9] for commencement of diabetes treatment. However, this risk was not confirmed with higher dose (6 to 10 mg/day: HR = 1.5 [0.8–2.7]). When assessing the risk of diabetes according to the World Health Organization (WHO) criteria, one study [21] found no significant differences between patients treated with LD-CT or those without. Of the two RCT assessing glycemia, one [22] found an increase of glycemia in LD-CT treated patients (5.9 versus 5.1 mmol/L, p = 0.001) whereas the other one [14] showed no difference (but some patients had stopped treatment, as previously described). Three articles [23–25] found an increase of insulin resistance induced by LD-CT below 5 mg per day. Seven articles assessed high blood pressure risk induced by LD-CT. Of these studies, four RCT [14,22,26,27] did not show any significant variation in blood pressure induced by LD-CT. One retrospective study [28] on 400 RA patients followed-up for 10 years, showed the risk of developing high blood pressure for a RA patient treated with CT < 7.5 mg per day for more than 6 months was not significantly increased (OR = 1.26 [95% CI: 0.59–2.71]), after adjustment on age, sex, body mass index (BMI), smoking status, insulin resistance, kidney function, disease duration, health assessment questionnaire (HAQ) and orthopaedic surgery. Two observational studies [20,21] found an association with high blood pressure. One retrospective study [21] found an increase of high blood pressure in LD-CT treated RA patients but the details of the method used were not available in the abstract. The other one [20] showed an association with LD-CT (1–5 mg per day: HR = 1.33 [1.0–1.5], 6–10 mg per day: HR = 1.5 [1.1–2.1]). One report [29] studied the risk effects of metabolic syndrome and found no increase in risk. Two reports [15,30] assessed homocysteine levels, an independent CV risk factor and showed no variation of serum homocysteine induced by LD-CT (data not shown). One article [31] studied levels of circulating ␣2-heremans-schmid glycoprotein/fetuin-A (AHSG), an independent risk factor for CV events and found no variation of the level of this protein in RA patients treated with LD-CT (data not shown). The descriptive analysis of the effect of LD-CT on cardiovascular risk factors is presented in Table 1 .
2.2. Impact of LD-CT on cardiovascular surrogate markers Nine articles assessed atherosclerosis measurement in RA. Of these articles, eight studies [15,32–38] showed no effect of LD-CT on carotid or femoral intima-media thickness measured by ultrasonography, whereas one study [24] showed after adjustment on blood pressure, insulin resistance and disease activity score 28 (DAS28), a significant correlation between the measurement of carotid intima-media thickness and the dose of CT in patients, all treated with LD-CT (r = 0.544, p < 0.001). Three articles assessed arterial stiffness with indirect measures such as pulse wave velocimetry (PWV). Of the three articles, two studies [39,40] showed no variation of the measurements induced by LD-CT whereas one article [41] showed an increase of arterial stiffness assessed by PWV, endothelial-dependant dilatation and forearm resistances. In this study the authors neither adjusted their results on CV risk factors, nor on RA activity or severity.
25
Two studies [42,43] assessed cardiac function with cardiac ultrasonography and showed no dysfunction induced by LD-CT. One study [44] assessed heart rate variability, an independent risk factor of CV event and showed no influence of LD-CT in RA. The descriptive analysis of the effect of LD-CT on cardiovascular risk factors is presented in Table 2. 2.3. Impact of LD-CT on major cardiovascular events The literature search identified few articles assessing the role of LD-CT in the development of CV events in RA. Of the six studies, four identified an association between LD-CT exposure and major CV events. Wolfe and Michaud [20] showed an association with MI after adjustments on co-morbidities, CV risk factors, RA activity and severity (exposure to 1 to 5 mg/day: HR = 1.7 [1.2–2.3]; exposure to 6–10 mg/day: HR = 1.9 [1.2–3.1]). Nadareishvili et al. [45] found an association with strokes only for RA patients exposed to 6 to 10 mg per day of prednisone (OR = 4.36 [1.60–11.90]) and not with exposure to 1–5 mg per day of prednisone (OR = 1.68 [0.76–3.73]), after adjustments on disease severity and co-morbidities. Two articles [46,47] used a composite index for assessing CV events. Davis et al. [46] studied MI, HF and CV mortality combined in a composite index. The authors showed that for patients using < 7.5 mg per day of CT the risk of developing a CV event was not significant (HR = 1.26 [95% CI: 0.80–2.01]). But for patients with a positive rheumatoid factor (RF), the risk was significantly increased (HR = 2.21 [95% CI: 1.22–4.00]) whereas it was not increased in RF negative patients (HR = 0.69 [95% CI: 0.27–1.74]). One case-control study [47] assessed MI and stroke risk combined in a composite index and showed that when comparing RA patients with LD-CT (< 10 mg/day) given as monotherapy methotrexate (MTX) monotherapy, the risk of developing a CV event was significantly increased after adjustment on CV risk factors, RA activity and severity (OR = 1.5 [95% CI: 1.1–2.1]). When the authors compared RA patients receiving LD-CT associated with another DMARD to RA with MTX, the risk was not significant (OR = 1.3 [95% CI: 0.8–2.0]). In this study there was no comparison between RA patients treated with MTX and LD-CT to MTX treatment alone. Two studies assessed cardiovascular mortality induced by LD-CT in RA [49,50]. In the first prospective study of 182 RA patients [49], most patients (95%) were treated with LD-CT. In this study, after adjusting on CV risk factors and RA severity, the authors showed no significant increase of mortality risk induced by LD-CT but the confidence interval was very wide so the results have to be interpreted with caution. The second study, an abstract [48] not published yet was a prospective study of 779 RA recruited between 1996 and 2001 and followed until 2004. After adjustment to age, sex, comorbidities, CV risk factors, RA activity, the authors showed that the mortality risk was increased by CT (HR = 2.03 [1.25–3.32]) and the threshold of significance was 10 mg per day for the daily dose and 44 g for the cumulative dose. Using CT as a continuous criterion, the risk of CV mortality was also increased to 9% for each increase of 1 mg per day of prednisone (1 mg/day: HR = 1.09 [1.04–1.15]). The risks identified by each study are presented in Table 3 and Fig. 2. Data could not be plotted in a meta-analysis since the number of events in CT-treated and non-treated patients was not available. 3. Discussion This systematic literature review showed a poor association between LD-CT and CV risk factors and identified a trend of increasing major CV events. LDC exposure might have a beneficial effect on lipid profile, no effect on blood pressure and an increase in insulin
26
A. Ruyssen-Witrand et al. / Joint Bone Spine 78 (2011) 23–30
Table 1 Studies assessing cardiovascular risk factor and corticosteroid therapy. Study
Design, number of RA duration of the study
Outcome
CT therapy Patients receiving CT, daily dose, cumulative dose
Adjusted factors in the analysis
Results
RCT, 67 RA, 5.3 years
TC, LDL, HDL, TG, Lp(a)
50%, 7.5 mg/day
RCT*
No difference
Wolfe and Michaud [20] Arthritis 2008
Prospective, 6035 RA, 3 years
Antilipemic therapy
39.1%, 8.1 mg/j
Age, sex
No difference
Park et al. [38] Am J Med 2002
Prospective, 42 RA, 1 year
TC, LDL, HDL, TGL, Apo(A1), Apo(B100)
57%, < 10 mg/day
None
No difference
Dahlqvist et al. [16] Scand J R 2006
Prospective, 55 RA, 2 years
TC, LDL, HDL, Lp(a)
51%, < 10 mg/day
Age, sex, activity of RA
Increase of HDL
Asanuma et al. [15] Arthritis 1999
Retrospective, 131 RA
TC, LDL, HDL, TGL, Lp(a), Apo(A)
35%, 6.5 mg/day
None
Increase of HDL
Heldenberg et al. [17] Clin Rheumatol 1983
Retrospective, 66 RA
TC, LDL, HDL, TGL
18%, < 7.5 mg/day
None
Decrease of TC (p < 0.01), LDL (p < 0.025) and TGL (p < 0.01) Increase of HDL (p < 0.05)
Garcia-Gomez et al. [18] Eur J Clin Invest 2008 Diabetes mellitus Wolfe and Michaud [20] Arthritis 2008
Retrospective, 78 RA
TC, LDL, HDL, TGL, Lp(a), Apo(A), Apo(B)
83%, 5.2 mg/day
BMI, activity, severity, treatment
Increase of HDL
Prospective, 6035 RA, 3 years
Diabetes treatment
39.1%, 8.1 mg/j
Age, sex
Global HR = 1.7 [1.3–2.2]
Dyslipidaemia Hafstrom et al. [14] J Rheumatol 2007
1–5 mg/day: HR = 2.1 [1.6–2.9] 6–10 mg/day: HR = 1.5 [0.8–2.7] Mazzantini et al. [21] ACR 2007
Retrospective, 365 RA
Diabetes (WHO)
81%, 4.2 mg/day
None
No difference
van Everdingen et al. [22] Ann Int Med 2002
RCT, 81 RA, 2 years
Glycemia
50%, 5 mg/day
RCT*
Increase of glycemia in CT treated patients: 5.9 vs. 5.1 mmol/L, p = 0.001
Hafstrom et al. [14] J Rheumatol 2007
RCT, 67 RA, 5.3 years
Glycemia
50%, 7.5 mg:day
RCT*
No difference
Pappas et al. [25] ACR 2007
Retrospective, 169 RA
HOMA-IR
39%, 5 mg/day, 5.5 g
CV risk factors, treatment
1–5 mg/day: Increase of 0.25 IU [95% CI: 0.006–0.440] 5–0 mg/day: Increase of 0.32 IU [0.080–0.555]
La Montagna et al. [24] Diab Vasc Dis Res 2007
Retrospective 45 RA
HOMA-IR
51%, < 7.5 mg/day
None
Significative correlation r = 0.372, p = 0.012
Dessein et al. [23] J Rheumatol 2004
Retrospective 92 RA
QUICKI
40%, 0.6 mg/day, 4.8 g
CV risk factors
Inverse correlation r = −0.207, p = 0.047
High Blood Pressure van Everdingen et al., Hafstrom et al., Hickling et al., Capell et al. [14,22,26,27] Panoulas and Douglas [28] Rheumatology 2008
4 RCT, 67-167 RA, 2–5.3 years
Blood Pressure
50%, 5–7.5 mg/day
RCT*
No difference in the 4 RCT
Retrospective, 400 RA
Blood Pressure
31%, 3 doses with one group < 7.5 mg/day
CV risk factors treatment, RA severity
OR = 1.26
[95% CI: 0.59–2.71] Wolfe and Michaud 2008 [20]
Prospective, 6035 RA, 3 years
Blood pressure treatment
39.1%, 8.1 mg/j
Age, sex
Global HR = 1.2 [1.1–1.4] 1–5 mg/day HR = 1.33 [1.0–1.5] 6–10 mg/jday HR = 1.5 [1.1–2.1]
A. Ruyssen-Witrand et al. / Joint Bone Spine 78 (2011) 23–30
27
Table 1 (Continued.) Study
Design, number of RA duration of the study
Outcome
CT therapy Patients receiving CT, daily dose, cumulative dose
Adjusted factors in the analysis
Results
Mazzantini et al. [21] ACR 2007 Metabolic syndrome Chung et al. [29] Atherosclerosis 2008
Retrospective 365 RA
Blood Pressure
81%, 4.2 mg/day
None
32.5% versus 10.4%, p < 0.0005
Retrospective, 154 RA
Metabolic syndrome (WHO, NCEP)
50 %, current dose na, 2.7 g
None
No difference
RA: rheumatoid arthritis; CT: corticosteroid; RCT: randomized controlled trial; RCT*: results are considered as adjusted on confounding factors since patients are randomized; CV: cardiovascular; TC: total cholesterol; LDL: low-density lipoprotein; HDL: high-density lipoprotein; TGL: triglycerides; Lp(a): lipoprotein a; Apo(A1):Apolipoprotein (a) A1; Apo(B100): Apolipoprotein(a) B-100; HOMA-IR: homeostasis model assessment of insulin resistance; QUICKI: quantitative insulin sensitivity check index; WHO: World Health Organization; NCEP: National Cholesterol Education Program Adult Treatment Panel III; na: non available; OR: odds ratio; vs.: versus; BMI: body mass index.
Table 2 Studies assessing surrogate markers of cardiovascular risk and comparing patients with or without corticosteroids. Study
Design, number of RA, duration of the study
Outcome
CT therapy Patients receiving CT, daily dose, cumulative dose
Adjusted factors in the analysis
Results
RCT, 67 RA, 5.3 years
Carotid IMT
50%, 7.5 mg/day
RCT*
No difference
del Rincon et al. [32] Arthritis Rheum 2004
Retrospective 647 RA
Carotid and femoral IMT
66%, 6.4 mg/day, 16.3 g
Age, sex, CV risk factors
No difference
Dessein et al. [33] J Rheumatol 2005
Retrospective, 74 RA
Carotid IMT
na%, na mg/day, 0.58 g
None
No difference
Pahor [37] Rheumatol Int 2006
Retrospective, 70 RA
Carotid IMT
63%, < 10 mg/day
CV risk factors, RA activity
No difference
Dessein et al. [34] J Rheumatol 2007
Retrospective, 91 RA
Carotid IMT
15%, < 10 mg/day
None
No difference
Gonzalez-Juanatey et al. [35] Medicine 2003
Retrospective 47 RA
Carotid IMT
98%, 5 mg/day, 16 g
CV risk factor, RA activity and severity
No difference
Kumeda et al. [36] Arthritis Rheum 2002
Retrospective, 138 RA
Carotid and femoral IMT
18%, < 10 mg/day
CV risk factors, RA activity
No difference
Park et al. [38] Arthritis Rheum 2002
Retrospective, 53 RA
Carotid IMT
60%, 5.3mg/day
None
No difference
La Montagna et al. [24] Diab Vasc Dis Res 2007
Retrospective 45 RA
Carotid IMT
51%, < 7.5 mg/day
Blood pressure, insulin resistance, DAS 28
Significative correlation, r = 0.544, p < 0.001
Retrospective, 47 RA
PWV
60%, 4.7 mg/day
Age, smoke, blood pressure, NSAID, MTX
No difference
Maki-Petaja et al. [39] Circulation 2006
Retrospective 77 RA
PWV
39%, 7 mg/day
CV risk factor, RA activity and severity
No difference
Arosio et al. [41] J Hypertens 2007
Retrospective 60 RA
PWV, EDD, Forearm resistances
41%, < 10 mg/day
None
Increase of PWV, EDD and forearms resistance compared to MTX group
Retrospective, 54 RA
Cardiac function/US
50%, < 5 mg/day
None
No difference
Gonzalez-Juanatey et al. [42] Sem Arthritis Rheum 2004
Retrospective, 47 RA
Cardiac function/US
98%, 5 mg/day, 16 g
None
No difference
Anichkov et al. [44] Int J Clin Pract 2007
Retrospective 23 RA
Heart rate variability
30%, 7.1 mg/day
Cardiovascular risk factors, RA activity
No difference
Atherosclerosis Hafstrom et al. [14] J Rheumatol 2007
Arterial Stiffness Tanaka et al. [40] J Rheumatol 2006
Cardiac function Montecucco et al. [43] Clin Exp Rheum 1999
RA: rheumatoid arthritis; CT: corticosteroid; CV cardiovascular; IMT: intima-media thickness assessed by ultrasonography; PWV: pulse wave velocity; EDD: endotheliumdependant, dilatation; MTX: methotrexate; NSAID: non-steroid anti-inflammatory drugs; US: ultrasonography.
28
A. Ruyssen-Witrand et al. / Joint Bone Spine 78 (2011) 23–30
Table 3 Studies assessing major cardiovascular events and corticosteroids therapy. Study
MI Wolfe and Michaud 2008 [20]
Stroke Nadareishvili et al. 2008 [45]
Composite index Davis et al. [46] Arthritis Rheum 2007 Solomon et al. [47] Arthritis Rheum 2006
Mortality Gonzalez-Gay et al. [49] Arthritis Rheum 2007 del Rincon et al. [48] ACR 2007
Design, number of RA, duration of the study
Outcome
CT therapy Patients receiving CT, daily dose, cumulative dose, mean duration
Adjusted factors in the analysis
Results
Prospective, 6035 RA, 3 years
MI
39.1%, 8.1 mg/j
Co-morbidities, CV risk factors, activity, severity, age, sex
If 1–5 mg/day: HR = 1.7 [1.2–2.3] If 6–10 mg/day: HR = 1.9 [1.2–3.1]
Retrospective, 832 RA, between 1999 and 2006
Stroke
29.5% within 94% receiving < 10 mg/day
Severity, co-morbidities, age, sex
If 1–5 mg/day: HR = 1.68 [0.76–3.73] If 6–10 mg/day: HR = 4.36 [1.60–11.90]
Prospective, 558 RA, 9066 PY between 1955–1995 to 2003 Retrospective 3501 RA
MI, HF, CV Mortality
58%, 7.7 mg/day, 4.29 g, 2.1 year
CV risk factors, RA activity, RF, age, sex
MI, stroke
24.7%, with one group < 10 mg/day
CV risk factors, RA activity and severity, age, sex
HR = 1.26 [95% CI: 0.80–2.01] RF+:HR = 2.21 [1.22–4.00] RF− :HR = 0.69 [0.27–1.74] CT versus MTX: OR = 1.5 [95 % CI: 1.1–2.1] CT+DMARD versus MTX OR = 1.3 [95 % CI: 0.8–2.0]
Prospective 182 RA, 9 years
CV mortality
95%, 5 mg/day, 13.5 g
CV risk factors, RA severity, age sex
OR = 2.25 [95 % CI: 0.29–102.5]
Prospective, 779 RA, 3971 PY, between 1996–2001 to 2004
CV mortality
67%, 5 mg/day, 29.3 g
CV risk factors, RA severity, age sex
HR = 2.03 [95% CI: 1.25–3.32] Daily dose HR/mg = 1.09 [1.04–1.15] Daily dose mortality ratio > 10 mg/day Cumulative dose HR/10 g = 1.03 [1.01–1.04] Cumulative dose threshold > 44 g
RA: rheumatoid arthritis; CV: cardiovascular; CT: corticosteroid; OR: odds ratio; HR: hazard ratio; MI: myocardial infarction; HF: heart failure; PY: person-years; RF: rheumatoid factor; DMARD: disease modifying activity of rheumatic diseases; MTX: methotrexate.
resistance. The analysis of surrogate markers revealed no risk of atherosclerosis with LD-CT and no effect on cardiac function. The effect of LD-CT on arterial stiffness is not clear since two studies did not find any change [39,40] whereas one study [41] identified an
Fig. 2. Cardiovascular events associated with exposure to low-dose glucocorticoids.
increase of arterial stiffness. When analysing the risk of CV events, an association was found with LD-CT in RA in 4/6 articles. Thus, LD-CT seems to increase insulin resistance but have no effect on surrogate markers such as atherosclerosis. Nevertheless, a trend to an increased risk of major CV events with exposure to LD-CT was identified. The association between CV risk and CT use may be difficult to appreciate because patients with a high level of activity often require CT to control their disease. Therefore CV risk could be increased either by the higher dose of CT used or by the high level of RA activity. In order to control such a bias, the analysis of literature paid attention to studies that adjusted their results to RA activity or severity. The systematic literature review only selected six studies assessing major CV events in RA and the potential effect of CT. Thus, a lot of studies were excluded because the population studied was not RA or the analysis of the subgroup of RA was not performed, CT dose data was missing, dose was > 10 mg per day or the articles did not study CV risk of CT specifically. The systematic review did not select studies based on large administrative databases, although they could assess such an association, because in these studies, the current daily dose of CT could not be identified and the association with LD-CT could not be evaluated. Wei et al. [11] compared 68,781 CT users to 82,202 non-users with a record linkage database. In this study, 1165 patients had inflammatory arthritis and in this subgroup of patients, the rate of CV events was not significantly increased if the dose was below 7.5 mg per day. This study was not included in our analysis because the diagnosis of RA was not established and the population studied could also include
A. Ruyssen-Witrand et al. / Joint Bone Spine 78 (2011) 23–30
polymyalgia rheumatica, psoriatic arthritis or other connective tissue diseases. Even if a trend of an increased risk was identified, a metaanalysis could not be performed since the number of events in exposed and non-exposed to LD-CT patients was not available. The authors who calculated the OR and HR through a logistic regression did not provide this information. Most of the studies gave the dose of CT in current daily-dose. So we had no information about the doses previously administered and very few data about the time to exposure. A more interesting analysis could have been performed with the cumulative dose of CT that would better reflect the long-term impregnation of CT. However, some parameters such as blood pressure or insulin resistance should be directly related to the current daily dose since they usually decrease after stopping exposure. No studies presented the dose of CT adjusted to the body weight; however, it is commonly known that efficacy of CT depend on the dose related to the body weight. Therefore, safety should be appreciated in such a way. This literature review illustrates the need of detecting CV risk factors in RA even in patients treated with LD-CT. Recent working groups in the annual EULAR meeting 2009 has recommended the use of GC only in early arthritis with a dose between 7 and 10 mg per day for < 6 month. Regarding the results of our literature search, the exposure to 1 to 3 years with LD-CT does not increase CV risk significantly in several RCT but longer exposure could increase the risk of major CV events. The risk is increased by two- to threefolds if prednisone is given with dosage > 10 mg per day. Since RA is already a risk factor for major CV events, the final objective of each physician should be stopping CT as soon as possible and give CT only for a short period at the very beginning of arthritis. The dose of prednisone should be also tailored according to the weight of patients. The systematic literature search showed little association between LD-CT and CV risk factors in RA except an increase in insulin resistance. However, a trend to an increase in risk of major CV events was identified and must be confirmed by further studies. Because of heterogeneity of the studies included, cofunding factors such as activity or severity of the disease, the absence of data about cumulating dose of glucocorticoids in most of the studies, these results have to be interpreted with caution. This systematic review has been exposed in May 30th and 31st 2008 in a French meeting and was used to discuss French recommendation about glucocorticoids use in RA [50]. Conflict of interest statement The authors received a grant support from Abbott. Acknowledgments The authors thank Abbott for an institutional support for this study, Philippe Gaudin, Daniel Wendling, Alain Cantagrel, Pascal Claudepierre, Jean-Francis Maillefert, Thierry Schaeverbeke and Jacques Tebib for their help in elaborating the study and interpretation of the data. References [1] del Rincon ID, Williams K, Stern MP, et al. High incidence of cardiovascular events in a rheumatoid arthritis cohort not explained by traditional cardiac risk factors. Arthritis Rheum 2001;44:2737–45. [2] Solomon DH, Karlson EW, Rimm EB, et al. Cardiovascular morbidity and mortality in women diagnosed with rheumatoid arthritis. Circulation 2003;107:1303–7. [3] Wolfe F, Freundlich B, Straus WL. Increase in cardiovascular and cerebrovascular disease prevalence in rheumatoid arthritis. J Rheumatol 2003;30:36–40. [4] Sattar N, McCarey DW, Capell H, et al. Explaining how “high-grade” systemic inflammation accelerates vascular risk in rheumatoid arthritis. Circulation 2003;108:2957–63.
29
[5] Van Doornum S, McColl G, Wicks IP. Accelerated atherosclerosis: an extraarticular feature of rheumatoid arthritis? Arthritis Rheum 2002;46:862–73. [6] Pham T, Gossec L, Constantin A, et al. Cardiovascular risk and rheumatoid arthritis: clinical practice guidelines based on published evidence and expert opinion. Joint Bone Spine 2006;73:379–87. [7] Andrews RC, Walker BR. Glucocorticoids and insulin resistance: old hormones, new targets. Clin Sci (Lond) 1999;96:513–23. [8] Whitworth JA, Mangos GJ, Kelly JJ. Cushing, cortisol, and cardiovascular disease. Hypertension 2000;36:912–6. [9] Girod JP, Brotman DJ. Does altered glucocorticoid homeostasis increase cardiovascular risk? Cardiovasc Res 2004;64:217–26. [10] Nashel DJ. Is atherosclerosis a complication of long-term corticosteroid treatment? Am J Med 1986;80:925–9. [11] Wei L, MacDonald TM, Walker BR. Taking glucocorticoids by prescription is associated with subsequent cardiovascular disease. Ann Intern Med 2004;141:764–70. [12] Rab ST, King 3rd SB, Roubin GS, et al. Coronary aneurysms after stent placement: a suggestion of altered vessel wall healing in the presence of anti-inflammatory agents. J Am Coll Cardiol 1991;18:1524–8. [13] Versaci F, Gaspardone A, Tomai F, et al. Immunosuppressive therapy for the prevention of restenosis after coronary artery stent implantation (IMPRESS Study). J Am Coll Cardiol 2002;40:1935–42. [14] Hafstrom I, Rohani M, Deneberg S, et al. Effects of low-dose prednisolone on endothelial function, atherosclerosis, and traditional risk factors for atherosclerosis in patients with rheumatoid arthritis–a randomized study. J Rheumatol 2007;34:1810–6. [15] Asanuma Y, Kawai S, Aoshima H, et al. Serum lipoprotein(a) and apolipoprotein(a) phenotypes in patients with rheumatoid arthritis. Arthritis Rheum 1999;42:443–7. [16] Dahlqvist SR, Engstrand S, Berglin E, et al. Conversion towards an atherogenic lipid profile in rheumatoid arthritis patients during long-term infliximab therapy. Scand J Rheumatol 2006;35:107–11. [17] Heldenberg D, Caspi D, Levtov O, et al. Serum lipids and lipoprotein concentrations in women with rheumatoid arthritis. Clin Rheumatol 1983;2:387–91. [18] Garcia-Gomez C, Nolla JM, Valverde J, et al. High HDL cholesterol in women with rhematoid arthritis on low-dose glucocorticoid therapy. Eur J Clin Invest 2008;38:686–92. [19] Westhovens R, Nijs J, Taelman V, et al. Body composition in rheumatoid arthritis. Br J Rheumatol 1997;36:444–8. [20] Wolfe F, Michaud K. The risk of myocardial infarction and pharmacologic and nonpharmacologic myocardial infarction predictors in rheumatoid arthritis: A cohort and nested case-control analysis. Arthritis Rheum 2008;58:2612–21. [21] Mazzantini M, Doveri M, Consensi A, et al. Incidence of adverse effects during long-term, low-dose glucocorticoid treatment of rheumatoid arthritis. Arthritis Rheum 2007;56 [Abstract [991]]. [22] van Everdingen AA, Jacobs JW, Siewertsz Van Reesema DR, et al. Low-dose prednisone therapy for patients with early active rheumatoid arthritis: clinical efficacy, disease-modifying properties, and side effects: a randomized, doubleblind, placebo-controlled clinical trial. Ann Intern Med 2002;136:1–12. [23] Dessein PH, Joffe BI, Stanwix AE, et al. Glucocorticoids and insulin sensitivity in rheumatoid arthritis. J Rheumatol 2004;31:867–74. [24] La Montagna G, Cacciapuoti F, Buono R, et al. Insulin resistance is an independent risk factor for atherosclerosis in rheumatoid arthritis. Diab Vasc Dis Res 2007;4:130–5. [25] Pappas DA, Rodriguez-Pla A, Bathon JM, et al. Glucocorticoid treatment is associated with increased insulin resistance in subjects with rheumatoid arthritis. Arthritis Rheum 2007 Apr 44;56 [Abstract [992]]. [26] Capell HA, Madhok R, Hunter JA, et al. Lack of radiological and clinical benefit over two years of low-dose prednisolone for rheumatoid arthritis: results of a randomised controlled trial. Ann Rheum Dis 2004;63:797–803. [27] Hickling P, Jacoby RK, Kirwan JR, Arthritis and Rheumatism Council Low-Dose Glucocorticoid Study Group. Joint destruction after glucocorticoids are withdrawn in early rheumatoid arthritis. Br J Rheumatol 1998;37:930–6. [28] Panoulas VF, Douglas KM, Stavropoulos-Kalinoglou A, et al. Long-term exposure to medium-dose glucocorticoid therapy associates with hypertension in patients with rheumatoid arthritis. Rheumatology (Oxford) 2008;47:72–5. [29] Chung CP, Oeser A, Solus JF, et al. Prevalence of the metabolic syndrome is increased in rheumatoid arthritis and is associated with coronary atherosclerosis. Atherosclerosis 2008;196:756–63. [30] Roubenoff R, Dellaripa P, Nadeau MR, et al. Abnormal homocysteine metabolism in rheumatoid arthritis. Arthritis Rheum 1997:718–22. [31] Sato H, Kazama JJ, Wada Y, et al. Decreased levels of circulating alpha2heremans-schmid glycoprotein/fetuin-A (AHSG) in patients with rheumatoid arthritis. Intern Med 2007;46:1685–91. [32] del Rincon I, O’Leary DH, Haas RW, et al. Effect of glucocorticoids on the arteries in rheumatoid arthritis. Arthritis Rheum 2004;50:3813–22. [33] Dessein PH, Joffe BI, Veller MG, et al. Traditional and nontraditional cardiovascular risk factors are associated with atherosclerosis in rheumatoid arthritis. J Rheumatol 2005;32:435–42. [34] Dessein PH, Norton GR, Woodiwiss AJ, et al. Influence of nonclassical cardiovascular risk factors on the accuracy of predicting subclinical atherosclerosis in rheumatoid arthritis. J Rheumatol 2007;34:943–51. [35] Gonzalez-Juanatey C, Llorca J, Testa A, et al. Increased prevalence of severe subclinical atherosclerotic findings in long-term treated rheumatoid arthritis patients without clinically evident atherosclerotic disease. Medicine (Baltimore) 2003;82:407–13.
30
A. Ruyssen-Witrand et al. / Joint Bone Spine 78 (2011) 23–30
[36] Kumeda Y, Inaba M, Goto H, et al. Increased thickness of the arterial intimamedia detected by ultrasonography in patients with rheumatoid arthritis. Arthritis Rheum 2002;46:1489–97. [37] Pahor A, Hojs R, Gorenjak M, et al. Accelerated atherosclerosis in premenopausal female patients with rheumatoid arthritis. Rheumatol Int 2006;27:119–23. [38] Park YB, Ahn CW, Choi HK, et al. Atherosclerosis in rheumatoid arthritis: morphologic evidence obtained by carotid ultrasound. Arthritis Rheum 2002;46:1714–9. [39] Maki-Petaja KM, Hall FC, Booth AD, et al. Rheumatoid arthritis is associated with increased aortic pulse-wave velocity, which is reduced by anti-tumor necrosis factor-alpha therapy. Circulation 2006;114:1185–92. [40] Tanaka K, Inaba M, Goto H, et al. Paraarticular trabecular bone loss at the ultradistal radius and increased arterial stiffening in postmenopausal patients with rheumatoid arthritis. J Rheumatol 2006;33:652–8. [41] Arosio E, De Marchi S, Rigoni A, et al. Forearm haemodynamics, arterial stiffness and microcirculatory reactivity in rheumatoid arthritis. J Hypertens 2007;25:1273–8. [42] Gonzalez-Juanatey C, Testa A, Garcia-Castelo A, et al. Echocardiographic and Doppler findings in long-term treated rheumatoid arthritis patients without clinically evident cardiovascular disease. Semin Arthritis Rheum 2004;33:231–8. [43] Montecucco C, Gobbi G, Perlini S, et al. Impaired diastolic function in active rheumatoid arthritis. Relationship with disease duration. Clin Exp Rheumatol 1999;17:407–12.
[44] Anichkov DA, Shostak NA, Ivanov DS. Heart rate variability is related to disease activity and smoking in rheumatoid arthritis patients. Int J Clin Pract 2007:777–83. [45] Nadareishvili Z, Michaud K, Hallenbeck JM, et al. Cardiovascular, rheumatologic, and pharmacologic predictors of stroke in patients with rheumatoid arthritis: a nested, case-control study. Arthritis Rheum 2008;59:1090–6. [46] Davis 3rd JM, Maradit Kremers H, Crowson CS, et al. Glucocorticoids and cardiovascular events in rheumatoid arthritis: a population-based cohort study. Arthritis Rheum 2007;56:820–30. [47] Solomon DH, Avorn J, Katz JN, et al. Immunosuppressive medications and hospitalization for cardiovascular events in patients with rheumatoid arthritis. Arthritis Rheum 2006;54:3790–8. [48] del Rincon I, Pogosian S, Escalante A. Daily and Cumulative threshold glucocorticoid dose associated with cardiovascular and all-cause mortality in rheumatoid arthritis. Arthritis Rheum 2007;56 [Abstract [289]]. [49] Gonzalez-Gay MA, Gonzalez-Juanatey C, Lopez-Diaz MJ, et al. HLA-DRB1 and persistent chronic inflammation contribute to cardiovascular events and cardiovascular mortality in patients with rheumatoid arthritis. Arthritis Rheum 2007;57:125–32. [50] Ruyssen-Witrand A, Mouterde G, Dernis E, et al. Tolérance des glucocorticoïdes dans la polyarthrite rhumatoïde: élaboration de recommandations pour la pratique clinique, à partir d’une analyse systématique de la littérature et de l’opinion d’experts. Rev Rhum 2008;75:S22–8 (Édition spéciale rencontres d’experts en rhumatologie).
Review
Cardiovascular risk induced by low-dose corticosteroids in rheumatoid arthritis: A systematic literature review Adeline Ruyssen-Witrand a,∗ , Bruno Fautrel b , Alain Saraux c , Xavier Le Loët d , Thao Pham e a
Service de rhumatologie B, hôpital Cochin, AP–HP, université Paris-V René-Descartes, 27, rue du Faubourg-Saint-Jacques, 75014 Paris, France Hôpital Pitié-Salpêtrière, AP–HP, université Paris-VI Pierre-et-Marie-Curie, 47-83, boulevard de l’Hôpital, 75651 Paris cedex 13, France c Hôpital de la Cavale Blanche, CHU de Brest, boulevard Tanguy-Prigent, 29609 Brest, France d Rouen university Hospital and Inserm U905 (IFRMP23), University of Rouen, 1, rue de Germont, 76031 Rouen cedex, France e Hôpital de la Conception, AP–HM, 147, boulevard Baille, 13385 Marseille cedex 05, France b
a r t i c l e
i n f o
Article history: Accepted 3 February 2010 Available online 14 May 2010 Keywords: Corticosteroids Cardiovascular disease Rheumatoid arthritis Treatment Epidemiology
a b s t r a c t Objectives: To assess the association between cardiovascular (CV) risk and low-dose corticosteroids (LDCT, defined as a daily dose < 10 mg/day of prednisone) in rheumatoid arthritis (RA) patients. Methods: Data source: A systematic review of the literature up to June 2009 was performed. Data extraction: (1) cardiovascular risk factors: high blood pressure, glycemia and lipid profile, carotid intima-media thickness, pulse-wave velocity, ventricular function; (2) “hard” outcomes: heart failure (HF), stroke, myocardial infarction (MI) or mortality. Data analysis: descriptive, comparing CV risk between LD-CTtreated RA patients and LD-CT-non-treated RA patients. Results: Of the 1138 screened reports, the literature search identified 37 assessing CV risk in LD-CT treated RA. The analysis showed a protective effect on serum lipid profile, an increase of insulin resistance or glycemia, probably no effect on blood pressure, no effect on atherosclerosis, discrepancies regarding arterial stiffness and no effect on ventricular function or heart rate variability. An association of LD-CT with major CV events was found in 4/6 studies. This included MI (HR = 1.7 [1.2–2.3]), stroke (OR = 4.36 [1.60–11.90] for LDC between 6 and 10 mg/day), mortality (HR = 2.03 [1.25-3.32]) and a composite index of CV events (in the group of rheumatoid factor positive RA, HR = 2.21 [1.22–4.00]). Two studies did not find any significant association between LD-CT exposure and mortality (OR = 2.25 [0.29–102.5]) or a composite index of CV events (OR = 1.3 [0.8–2.0]). Conclusion: Although the literature review showed poor association between LDC exposure and CV risk factors, a trend of increasing major CV events was identified. © 2010 Société franc¸aise de rhumatologie. Published by Elsevier Masson SAS. All rights reserved.
Rheumatoid arthritis (RA) is a chronic inflammatory disease that leads to severe joint destruction. Beside structural evolution, patients with RA are at risk of developing cardiovascular (CV) diseases including myocardial infarction (MI), heart failure (HF), as well as increased CV mortality [1–3]. Chronic inflammation that occurs in this population is believed to contribute to the increased risk [4–6]. Corticosteroids (CT) are powerful anti-inflammatory agents and have been used since the 1950s in the symptomatic treatment of RA. However, it is known that long-term CT use at high dose have considerable toxicity including effects on blood pressure, insulin resistance, lipid profile, body weight and fat distribution [7,8] that might significantly increase CV risk [9,10]. The risk of CV
∗ Corresponding author. Tel.: +33158412562; fax: +335843549256. E-mail address: [email protected] (A. Ruyssen-Witrand).
events associated with CT has already been demonstrated and is dose-dependant [11]. Therefore, alternative treatment approaches have been developed in RA, using conventional disease-modifying antirheumatic drugs (DMARDs) and more recently biologic agents, which can control RA activity and thus enable dose reduction of CT. CT may also have cardioprotective effects mediated by their anti-inflammatory and antiproliferative actions in the vessel wall [12,13]. However the CV risk with CT use < 10 mg per day has not been clearly demonstrated and based on its beneficial effects of CT in controlling chronic inflammation, low-dose CT (LD-CT) on CV risk is still a subject of controversy. The aim of this study is to assess through a systematic literature review the potential association between CV risk and the use of long-term LD-CT defined as an average daily dose below 10 mg per day, in RA.
1297-319X/$ – see front matter © 2010 Société franc¸aise de rhumatologie. Published by Elsevier Masson SAS. All rights reserved. doi:10.1016/j.jbspin.2010.02.040
24
A. Ruyssen-Witrand et al. / Joint Bone Spine 78 (2011) 23–30
1. Methods 1.1. Literature search A systematic literature search was performed using databases of Medline, Embase and Cochrane Library. The keywords used for the Medline search were: (“Arthritis, Rheumatoid/epidemiology”[Mesh:NoExp] or “Arthritis, Rheumatoid/ultrasonography” [Mesh:NoExp] or “Arthritis, Rheumatoid/complications” [Mesh:NoExp] or “Arthritis, Rheumatoid/drug therapy”[Mesh:NoExp] or “Arthritis, Rheumatoid/mortality”[Mesh:NoExp]) and (“Cardiovascular Diseases”[Mesh:NoExp] or “Heart Diseases”[Mesh:NoExp] or “Cardiomegaly”[Mesh] or “Cardiomyopathies”[Mesh] or “Heart Arrest”[Mesh] or “Heart Failure”[Mesh] or “Myocardial Ischemia”[Mesh] or “Ventricular Dysfunction”[Mesh] or “Cardiac Output, Low”[Mesh] or “Vascular Diseases” [Mesh:NoExp] or “Aneurysm”[Mesh] or “Aortic Diseases”[Mesh:NoExp] or “Aortic Aneurysm”[Mesh] or “Aortic Arch Syndromes”[Mesh] or “Arterial Occlusive Diseases” [Mesh] or “Cerebrovascular Disorders”[Mesh] or “Diabetic Angiopathies”[Mesh] or “Embolism and Thrombosis”[Mesh] or “Hypertension”[Mesh] or “Ischemia”[Mesh] or “Peripheral Vascular Diseases”[Mesh] or “Diabetes Mellitus”[Mesh]) NOT “Case Reports”[ptyp]. The search was limited to English, French, Spanish and German languages, humans and adults above 19 years old with no limit of date. For the Embase search, the keywords used were “Cardiovascular” and “rheumatoid arthritis” limited to English language and humans with no limit of date. For the Cochrane Library search, the keywords used were “Cardiovascular” and “rheumatoid arthritis” with no limits. The literature search was performed on December 1st 2007 and the database search was regularly updated until June 15th 2009. Abstracts from international congresses including American College of Rheumatology (ACR), European League Against Rheumatism (EULAR) and French national congress société franc¸aise de rhumatologie (SFR) from 2006 to 2009 were also investigated. Manual search was performed when reading references from studies included or from reviews from expert’s opinion when they were recently published (after 2000s) or using related links tool of Medline.
of the populations studied including disease activity and severity, rheumatoid factor and anti-CCP antibodies, cardiovascular pathologies, smoking status, diabetes mellitus, body mass index or high blood pressure were collected. The proportion of patients receiving LD-CT in the study was assessed with the current daily dose used and if mentioned, the cumulative dose and the duration of LDC therapy. The dose of CT used was expressed in milligrams (daily dose) or grams (cumulative dose) of prednisone. Concomitant therapies including non-steroidal anti-inflammatory drugs (NSAIDs), DMARDs or biologic agents when taken were noted. 1.4. Data analysis All comparisons between RA patients with or without LD-CT were analysed. Comparison could be given in percentage (binary outcome) or in mean (continuous outcome) in each group or could be assessed through a logistic regression or a Cox model with odds ratios (OR) or hazard ratios (HR). The type of comparison performed was analysed including LD-CT as a binary variable (Yes/No) or a continuous variable (current daily dose or cumulative dose) and the adjustment on potential confounding factors such as cardiovascular diseases, co-morbidities, cardiovascular risk factors, disease activity or disease severity. 2. Results The literature search on electronic databases identified 1138 titles (Fig. 1) and 277 articles were selected after reviewing the titles and abstracts. Finally, 34 studies were included in the analysis and three abstracts from international congresses were added. 2.1. Impact on the main cardiovascular risk factors Seven articles studied the risk of dyslipidaemia. One [14] of these articles was a randomised controlled trial (RCT) comparing oral prednisone (7.5 mg/day) versus placebo associated with a DMARD chosen by the physician. The authors did not show any difference of lipid serum levels between groups but 63% of patients in the prednisone arm had stopped the treatment before the end of the follow-up. Four studies [15–18] showed an increase of high-density
1.2. Articles selection Only reports assessing RA were selected. To be included in the selection, the articles had to compare two groups of RA patients: one receiving LD-CT defined as a daily dose below 10 mg per day or if the daily dose was not available, a cumulative dose under 5 g and another group without CT. The outcomes studied were divided in three: the “hard outcomes” or major CV events including mortality, myocardial infarction, stroke or heart failure, the “CV surrogate markers” such as atherosclerosis, arterial stiffness, subclinical heart failure, heart rate variability and the cardiovascular traditional risk factors such as high blood pressure, diabetes mellitus, lipids concentrations, homocystein levels or metabolic syndrome. Reviews and letters were used for the manual search and then excluded of the analysis. Articles with no study of effect of CT on CV risk, where the daily dose was not mentioned or the dose was above 10 mg per day were excluded. 1.3. Data extraction The data were extracted using a standardised form by one author (ARW). Beside general information about the article, the design of the study was extracted including randomised controlled trials, prospective observational studies, retrospective studies or studies using national databanks. The outcome assessed, the characteristics
Fig. 1. Flow-chart of inclusion procedure. RA: rheumatoid arthritis; CT: corticosteroids.
A. Ruyssen-Witrand et al. / Joint Bone Spine 78 (2011) 23–30
lipoproteins and one study [19] reported decrease in total cholesterol, low-density lipoproteins and triglycerides. One article [19] described fat distribution assessed by Dual X-Ray Absorptiometry in 43 RA patients and found a significant increase in abdominal fat distribution in men treated with LD-CT compared to men without LD-CT (data not shown). Seven articles studied the risk of diabetes mellitus induced by LD-CT. One study [20] found an association with exposure to 1 to 5 mg per day of prednisone with a hazard ratio (HR) of 2.1 [1.6–2.9] for commencement of diabetes treatment. However, this risk was not confirmed with higher dose (6 to 10 mg/day: HR = 1.5 [0.8–2.7]). When assessing the risk of diabetes according to the World Health Organization (WHO) criteria, one study [21] found no significant differences between patients treated with LD-CT or those without. Of the two RCT assessing glycemia, one [22] found an increase of glycemia in LD-CT treated patients (5.9 versus 5.1 mmol/L, p = 0.001) whereas the other one [14] showed no difference (but some patients had stopped treatment, as previously described). Three articles [23–25] found an increase of insulin resistance induced by LD-CT below 5 mg per day. Seven articles assessed high blood pressure risk induced by LD-CT. Of these studies, four RCT [14,22,26,27] did not show any significant variation in blood pressure induced by LD-CT. One retrospective study [28] on 400 RA patients followed-up for 10 years, showed the risk of developing high blood pressure for a RA patient treated with CT < 7.5 mg per day for more than 6 months was not significantly increased (OR = 1.26 [95% CI: 0.59–2.71]), after adjustment on age, sex, body mass index (BMI), smoking status, insulin resistance, kidney function, disease duration, health assessment questionnaire (HAQ) and orthopaedic surgery. Two observational studies [20,21] found an association with high blood pressure. One retrospective study [21] found an increase of high blood pressure in LD-CT treated RA patients but the details of the method used were not available in the abstract. The other one [20] showed an association with LD-CT (1–5 mg per day: HR = 1.33 [1.0–1.5], 6–10 mg per day: HR = 1.5 [1.1–2.1]). One report [29] studied the risk effects of metabolic syndrome and found no increase in risk. Two reports [15,30] assessed homocysteine levels, an independent CV risk factor and showed no variation of serum homocysteine induced by LD-CT (data not shown). One article [31] studied levels of circulating ␣2-heremans-schmid glycoprotein/fetuin-A (AHSG), an independent risk factor for CV events and found no variation of the level of this protein in RA patients treated with LD-CT (data not shown). The descriptive analysis of the effect of LD-CT on cardiovascular risk factors is presented in Table 1 .
2.2. Impact of LD-CT on cardiovascular surrogate markers Nine articles assessed atherosclerosis measurement in RA. Of these articles, eight studies [15,32–38] showed no effect of LD-CT on carotid or femoral intima-media thickness measured by ultrasonography, whereas one study [24] showed after adjustment on blood pressure, insulin resistance and disease activity score 28 (DAS28), a significant correlation between the measurement of carotid intima-media thickness and the dose of CT in patients, all treated with LD-CT (r = 0.544, p < 0.001). Three articles assessed arterial stiffness with indirect measures such as pulse wave velocimetry (PWV). Of the three articles, two studies [39,40] showed no variation of the measurements induced by LD-CT whereas one article [41] showed an increase of arterial stiffness assessed by PWV, endothelial-dependant dilatation and forearm resistances. In this study the authors neither adjusted their results on CV risk factors, nor on RA activity or severity.
25
Two studies [42,43] assessed cardiac function with cardiac ultrasonography and showed no dysfunction induced by LD-CT. One study [44] assessed heart rate variability, an independent risk factor of CV event and showed no influence of LD-CT in RA. The descriptive analysis of the effect of LD-CT on cardiovascular risk factors is presented in Table 2. 2.3. Impact of LD-CT on major cardiovascular events The literature search identified few articles assessing the role of LD-CT in the development of CV events in RA. Of the six studies, four identified an association between LD-CT exposure and major CV events. Wolfe and Michaud [20] showed an association with MI after adjustments on co-morbidities, CV risk factors, RA activity and severity (exposure to 1 to 5 mg/day: HR = 1.7 [1.2–2.3]; exposure to 6–10 mg/day: HR = 1.9 [1.2–3.1]). Nadareishvili et al. [45] found an association with strokes only for RA patients exposed to 6 to 10 mg per day of prednisone (OR = 4.36 [1.60–11.90]) and not with exposure to 1–5 mg per day of prednisone (OR = 1.68 [0.76–3.73]), after adjustments on disease severity and co-morbidities. Two articles [46,47] used a composite index for assessing CV events. Davis et al. [46] studied MI, HF and CV mortality combined in a composite index. The authors showed that for patients using < 7.5 mg per day of CT the risk of developing a CV event was not significant (HR = 1.26 [95% CI: 0.80–2.01]). But for patients with a positive rheumatoid factor (RF), the risk was significantly increased (HR = 2.21 [95% CI: 1.22–4.00]) whereas it was not increased in RF negative patients (HR = 0.69 [95% CI: 0.27–1.74]). One case-control study [47] assessed MI and stroke risk combined in a composite index and showed that when comparing RA patients with LD-CT (< 10 mg/day) given as monotherapy methotrexate (MTX) monotherapy, the risk of developing a CV event was significantly increased after adjustment on CV risk factors, RA activity and severity (OR = 1.5 [95% CI: 1.1–2.1]). When the authors compared RA patients receiving LD-CT associated with another DMARD to RA with MTX, the risk was not significant (OR = 1.3 [95% CI: 0.8–2.0]). In this study there was no comparison between RA patients treated with MTX and LD-CT to MTX treatment alone. Two studies assessed cardiovascular mortality induced by LD-CT in RA [49,50]. In the first prospective study of 182 RA patients [49], most patients (95%) were treated with LD-CT. In this study, after adjusting on CV risk factors and RA severity, the authors showed no significant increase of mortality risk induced by LD-CT but the confidence interval was very wide so the results have to be interpreted with caution. The second study, an abstract [48] not published yet was a prospective study of 779 RA recruited between 1996 and 2001 and followed until 2004. After adjustment to age, sex, comorbidities, CV risk factors, RA activity, the authors showed that the mortality risk was increased by CT (HR = 2.03 [1.25–3.32]) and the threshold of significance was 10 mg per day for the daily dose and 44 g for the cumulative dose. Using CT as a continuous criterion, the risk of CV mortality was also increased to 9% for each increase of 1 mg per day of prednisone (1 mg/day: HR = 1.09 [1.04–1.15]). The risks identified by each study are presented in Table 3 and Fig. 2. Data could not be plotted in a meta-analysis since the number of events in CT-treated and non-treated patients was not available. 3. Discussion This systematic literature review showed a poor association between LD-CT and CV risk factors and identified a trend of increasing major CV events. LDC exposure might have a beneficial effect on lipid profile, no effect on blood pressure and an increase in insulin
26
A. Ruyssen-Witrand et al. / Joint Bone Spine 78 (2011) 23–30
Table 1 Studies assessing cardiovascular risk factor and corticosteroid therapy. Study
Design, number of RA duration of the study
Outcome
CT therapy Patients receiving CT, daily dose, cumulative dose
Adjusted factors in the analysis
Results
RCT, 67 RA, 5.3 years
TC, LDL, HDL, TG, Lp(a)
50%, 7.5 mg/day
RCT*
No difference
Wolfe and Michaud [20] Arthritis 2008
Prospective, 6035 RA, 3 years
Antilipemic therapy
39.1%, 8.1 mg/j
Age, sex
No difference
Park et al. [38] Am J Med 2002
Prospective, 42 RA, 1 year
TC, LDL, HDL, TGL, Apo(A1), Apo(B100)
57%, < 10 mg/day
None
No difference
Dahlqvist et al. [16] Scand J R 2006
Prospective, 55 RA, 2 years
TC, LDL, HDL, Lp(a)
51%, < 10 mg/day
Age, sex, activity of RA
Increase of HDL
Asanuma et al. [15] Arthritis 1999
Retrospective, 131 RA
TC, LDL, HDL, TGL, Lp(a), Apo(A)
35%, 6.5 mg/day
None
Increase of HDL
Heldenberg et al. [17] Clin Rheumatol 1983
Retrospective, 66 RA
TC, LDL, HDL, TGL
18%, < 7.5 mg/day
None
Decrease of TC (p < 0.01), LDL (p < 0.025) and TGL (p < 0.01) Increase of HDL (p < 0.05)
Garcia-Gomez et al. [18] Eur J Clin Invest 2008 Diabetes mellitus Wolfe and Michaud [20] Arthritis 2008
Retrospective, 78 RA
TC, LDL, HDL, TGL, Lp(a), Apo(A), Apo(B)
83%, 5.2 mg/day
BMI, activity, severity, treatment
Increase of HDL
Prospective, 6035 RA, 3 years
Diabetes treatment
39.1%, 8.1 mg/j
Age, sex
Global HR = 1.7 [1.3–2.2]
Dyslipidaemia Hafstrom et al. [14] J Rheumatol 2007
1–5 mg/day: HR = 2.1 [1.6–2.9] 6–10 mg/day: HR = 1.5 [0.8–2.7] Mazzantini et al. [21] ACR 2007
Retrospective, 365 RA
Diabetes (WHO)
81%, 4.2 mg/day
None
No difference
van Everdingen et al. [22] Ann Int Med 2002
RCT, 81 RA, 2 years
Glycemia
50%, 5 mg/day
RCT*
Increase of glycemia in CT treated patients: 5.9 vs. 5.1 mmol/L, p = 0.001
Hafstrom et al. [14] J Rheumatol 2007
RCT, 67 RA, 5.3 years
Glycemia
50%, 7.5 mg:day
RCT*
No difference
Pappas et al. [25] ACR 2007
Retrospective, 169 RA
HOMA-IR
39%, 5 mg/day, 5.5 g
CV risk factors, treatment
1–5 mg/day: Increase of 0.25 IU [95% CI: 0.006–0.440] 5–0 mg/day: Increase of 0.32 IU [0.080–0.555]
La Montagna et al. [24] Diab Vasc Dis Res 2007
Retrospective 45 RA
HOMA-IR
51%, < 7.5 mg/day
None
Significative correlation r = 0.372, p = 0.012
Dessein et al. [23] J Rheumatol 2004
Retrospective 92 RA
QUICKI
40%, 0.6 mg/day, 4.8 g
CV risk factors
Inverse correlation r = −0.207, p = 0.047
High Blood Pressure van Everdingen et al., Hafstrom et al., Hickling et al., Capell et al. [14,22,26,27] Panoulas and Douglas [28] Rheumatology 2008
4 RCT, 67-167 RA, 2–5.3 years
Blood Pressure
50%, 5–7.5 mg/day
RCT*
No difference in the 4 RCT
Retrospective, 400 RA
Blood Pressure
31%, 3 doses with one group < 7.5 mg/day
CV risk factors treatment, RA severity
OR = 1.26
[95% CI: 0.59–2.71] Wolfe and Michaud 2008 [20]
Prospective, 6035 RA, 3 years
Blood pressure treatment
39.1%, 8.1 mg/j
Age, sex
Global HR = 1.2 [1.1–1.4] 1–5 mg/day HR = 1.33 [1.0–1.5] 6–10 mg/jday HR = 1.5 [1.1–2.1]
A. Ruyssen-Witrand et al. / Joint Bone Spine 78 (2011) 23–30
27
Table 1 (Continued.) Study
Design, number of RA duration of the study
Outcome
CT therapy Patients receiving CT, daily dose, cumulative dose
Adjusted factors in the analysis
Results
Mazzantini et al. [21] ACR 2007 Metabolic syndrome Chung et al. [29] Atherosclerosis 2008
Retrospective 365 RA
Blood Pressure
81%, 4.2 mg/day
None
32.5% versus 10.4%, p < 0.0005
Retrospective, 154 RA
Metabolic syndrome (WHO, NCEP)
50 %, current dose na, 2.7 g
None
No difference
RA: rheumatoid arthritis; CT: corticosteroid; RCT: randomized controlled trial; RCT*: results are considered as adjusted on confounding factors since patients are randomized; CV: cardiovascular; TC: total cholesterol; LDL: low-density lipoprotein; HDL: high-density lipoprotein; TGL: triglycerides; Lp(a): lipoprotein a; Apo(A1):Apolipoprotein (a) A1; Apo(B100): Apolipoprotein(a) B-100; HOMA-IR: homeostasis model assessment of insulin resistance; QUICKI: quantitative insulin sensitivity check index; WHO: World Health Organization; NCEP: National Cholesterol Education Program Adult Treatment Panel III; na: non available; OR: odds ratio; vs.: versus; BMI: body mass index.
Table 2 Studies assessing surrogate markers of cardiovascular risk and comparing patients with or without corticosteroids. Study
Design, number of RA, duration of the study
Outcome
CT therapy Patients receiving CT, daily dose, cumulative dose
Adjusted factors in the analysis
Results
RCT, 67 RA, 5.3 years
Carotid IMT
50%, 7.5 mg/day
RCT*
No difference
del Rincon et al. [32] Arthritis Rheum 2004
Retrospective 647 RA
Carotid and femoral IMT
66%, 6.4 mg/day, 16.3 g
Age, sex, CV risk factors
No difference
Dessein et al. [33] J Rheumatol 2005
Retrospective, 74 RA
Carotid IMT
na%, na mg/day, 0.58 g
None
No difference
Pahor [37] Rheumatol Int 2006
Retrospective, 70 RA
Carotid IMT
63%, < 10 mg/day
CV risk factors, RA activity
No difference
Dessein et al. [34] J Rheumatol 2007
Retrospective, 91 RA
Carotid IMT
15%, < 10 mg/day
None
No difference
Gonzalez-Juanatey et al. [35] Medicine 2003
Retrospective 47 RA
Carotid IMT
98%, 5 mg/day, 16 g
CV risk factor, RA activity and severity
No difference
Kumeda et al. [36] Arthritis Rheum 2002
Retrospective, 138 RA
Carotid and femoral IMT
18%, < 10 mg/day
CV risk factors, RA activity
No difference
Park et al. [38] Arthritis Rheum 2002
Retrospective, 53 RA
Carotid IMT
60%, 5.3mg/day
None
No difference
La Montagna et al. [24] Diab Vasc Dis Res 2007
Retrospective 45 RA
Carotid IMT
51%, < 7.5 mg/day
Blood pressure, insulin resistance, DAS 28
Significative correlation, r = 0.544, p < 0.001
Retrospective, 47 RA
PWV
60%, 4.7 mg/day
Age, smoke, blood pressure, NSAID, MTX
No difference
Maki-Petaja et al. [39] Circulation 2006
Retrospective 77 RA
PWV
39%, 7 mg/day
CV risk factor, RA activity and severity
No difference
Arosio et al. [41] J Hypertens 2007
Retrospective 60 RA
PWV, EDD, Forearm resistances
41%, < 10 mg/day
None
Increase of PWV, EDD and forearms resistance compared to MTX group
Retrospective, 54 RA
Cardiac function/US
50%, < 5 mg/day
None
No difference
Gonzalez-Juanatey et al. [42] Sem Arthritis Rheum 2004
Retrospective, 47 RA
Cardiac function/US
98%, 5 mg/day, 16 g
None
No difference
Anichkov et al. [44] Int J Clin Pract 2007
Retrospective 23 RA
Heart rate variability
30%, 7.1 mg/day
Cardiovascular risk factors, RA activity
No difference
Atherosclerosis Hafstrom et al. [14] J Rheumatol 2007
Arterial Stiffness Tanaka et al. [40] J Rheumatol 2006
Cardiac function Montecucco et al. [43] Clin Exp Rheum 1999
RA: rheumatoid arthritis; CT: corticosteroid; CV cardiovascular; IMT: intima-media thickness assessed by ultrasonography; PWV: pulse wave velocity; EDD: endotheliumdependant, dilatation; MTX: methotrexate; NSAID: non-steroid anti-inflammatory drugs; US: ultrasonography.
28
A. Ruyssen-Witrand et al. / Joint Bone Spine 78 (2011) 23–30
Table 3 Studies assessing major cardiovascular events and corticosteroids therapy. Study
MI Wolfe and Michaud 2008 [20]
Stroke Nadareishvili et al. 2008 [45]
Composite index Davis et al. [46] Arthritis Rheum 2007 Solomon et al. [47] Arthritis Rheum 2006
Mortality Gonzalez-Gay et al. [49] Arthritis Rheum 2007 del Rincon et al. [48] ACR 2007
Design, number of RA, duration of the study
Outcome
CT therapy Patients receiving CT, daily dose, cumulative dose, mean duration
Adjusted factors in the analysis
Results
Prospective, 6035 RA, 3 years
MI
39.1%, 8.1 mg/j
Co-morbidities, CV risk factors, activity, severity, age, sex
If 1–5 mg/day: HR = 1.7 [1.2–2.3] If 6–10 mg/day: HR = 1.9 [1.2–3.1]
Retrospective, 832 RA, between 1999 and 2006
Stroke
29.5% within 94% receiving < 10 mg/day
Severity, co-morbidities, age, sex
If 1–5 mg/day: HR = 1.68 [0.76–3.73] If 6–10 mg/day: HR = 4.36 [1.60–11.90]
Prospective, 558 RA, 9066 PY between 1955–1995 to 2003 Retrospective 3501 RA
MI, HF, CV Mortality
58%, 7.7 mg/day, 4.29 g, 2.1 year
CV risk factors, RA activity, RF, age, sex
MI, stroke
24.7%, with one group < 10 mg/day
CV risk factors, RA activity and severity, age, sex
HR = 1.26 [95% CI: 0.80–2.01] RF+:HR = 2.21 [1.22–4.00] RF− :HR = 0.69 [0.27–1.74] CT versus MTX: OR = 1.5 [95 % CI: 1.1–2.1] CT+DMARD versus MTX OR = 1.3 [95 % CI: 0.8–2.0]
Prospective 182 RA, 9 years
CV mortality
95%, 5 mg/day, 13.5 g
CV risk factors, RA severity, age sex
OR = 2.25 [95 % CI: 0.29–102.5]
Prospective, 779 RA, 3971 PY, between 1996–2001 to 2004
CV mortality
67%, 5 mg/day, 29.3 g
CV risk factors, RA severity, age sex
HR = 2.03 [95% CI: 1.25–3.32] Daily dose HR/mg = 1.09 [1.04–1.15] Daily dose mortality ratio > 10 mg/day Cumulative dose HR/10 g = 1.03 [1.01–1.04] Cumulative dose threshold > 44 g
RA: rheumatoid arthritis; CV: cardiovascular; CT: corticosteroid; OR: odds ratio; HR: hazard ratio; MI: myocardial infarction; HF: heart failure; PY: person-years; RF: rheumatoid factor; DMARD: disease modifying activity of rheumatic diseases; MTX: methotrexate.
resistance. The analysis of surrogate markers revealed no risk of atherosclerosis with LD-CT and no effect on cardiac function. The effect of LD-CT on arterial stiffness is not clear since two studies did not find any change [39,40] whereas one study [41] identified an
Fig. 2. Cardiovascular events associated with exposure to low-dose glucocorticoids.
increase of arterial stiffness. When analysing the risk of CV events, an association was found with LD-CT in RA in 4/6 articles. Thus, LD-CT seems to increase insulin resistance but have no effect on surrogate markers such as atherosclerosis. Nevertheless, a trend to an increased risk of major CV events with exposure to LD-CT was identified. The association between CV risk and CT use may be difficult to appreciate because patients with a high level of activity often require CT to control their disease. Therefore CV risk could be increased either by the higher dose of CT used or by the high level of RA activity. In order to control such a bias, the analysis of literature paid attention to studies that adjusted their results to RA activity or severity. The systematic literature review only selected six studies assessing major CV events in RA and the potential effect of CT. Thus, a lot of studies were excluded because the population studied was not RA or the analysis of the subgroup of RA was not performed, CT dose data was missing, dose was > 10 mg per day or the articles did not study CV risk of CT specifically. The systematic review did not select studies based on large administrative databases, although they could assess such an association, because in these studies, the current daily dose of CT could not be identified and the association with LD-CT could not be evaluated. Wei et al. [11] compared 68,781 CT users to 82,202 non-users with a record linkage database. In this study, 1165 patients had inflammatory arthritis and in this subgroup of patients, the rate of CV events was not significantly increased if the dose was below 7.5 mg per day. This study was not included in our analysis because the diagnosis of RA was not established and the population studied could also include
A. Ruyssen-Witrand et al. / Joint Bone Spine 78 (2011) 23–30
polymyalgia rheumatica, psoriatic arthritis or other connective tissue diseases. Even if a trend of an increased risk was identified, a metaanalysis could not be performed since the number of events in exposed and non-exposed to LD-CT patients was not available. The authors who calculated the OR and HR through a logistic regression did not provide this information. Most of the studies gave the dose of CT in current daily-dose. So we had no information about the doses previously administered and very few data about the time to exposure. A more interesting analysis could have been performed with the cumulative dose of CT that would better reflect the long-term impregnation of CT. However, some parameters such as blood pressure or insulin resistance should be directly related to the current daily dose since they usually decrease after stopping exposure. No studies presented the dose of CT adjusted to the body weight; however, it is commonly known that efficacy of CT depend on the dose related to the body weight. Therefore, safety should be appreciated in such a way. This literature review illustrates the need of detecting CV risk factors in RA even in patients treated with LD-CT. Recent working groups in the annual EULAR meeting 2009 has recommended the use of GC only in early arthritis with a dose between 7 and 10 mg per day for < 6 month. Regarding the results of our literature search, the exposure to 1 to 3 years with LD-CT does not increase CV risk significantly in several RCT but longer exposure could increase the risk of major CV events. The risk is increased by two- to threefolds if prednisone is given with dosage > 10 mg per day. Since RA is already a risk factor for major CV events, the final objective of each physician should be stopping CT as soon as possible and give CT only for a short period at the very beginning of arthritis. The dose of prednisone should be also tailored according to the weight of patients. The systematic literature search showed little association between LD-CT and CV risk factors in RA except an increase in insulin resistance. However, a trend to an increase in risk of major CV events was identified and must be confirmed by further studies. Because of heterogeneity of the studies included, cofunding factors such as activity or severity of the disease, the absence of data about cumulating dose of glucocorticoids in most of the studies, these results have to be interpreted with caution. This systematic review has been exposed in May 30th and 31st 2008 in a French meeting and was used to discuss French recommendation about glucocorticoids use in RA [50]. Conflict of interest statement The authors received a grant support from Abbott. Acknowledgments The authors thank Abbott for an institutional support for this study, Philippe Gaudin, Daniel Wendling, Alain Cantagrel, Pascal Claudepierre, Jean-Francis Maillefert, Thierry Schaeverbeke and Jacques Tebib for their help in elaborating the study and interpretation of the data. References [1] del Rincon ID, Williams K, Stern MP, et al. High incidence of cardiovascular events in a rheumatoid arthritis cohort not explained by traditional cardiac risk factors. Arthritis Rheum 2001;44:2737–45. [2] Solomon DH, Karlson EW, Rimm EB, et al. Cardiovascular morbidity and mortality in women diagnosed with rheumatoid arthritis. Circulation 2003;107:1303–7. [3] Wolfe F, Freundlich B, Straus WL. Increase in cardiovascular and cerebrovascular disease prevalence in rheumatoid arthritis. J Rheumatol 2003;30:36–40. [4] Sattar N, McCarey DW, Capell H, et al. Explaining how “high-grade” systemic inflammation accelerates vascular risk in rheumatoid arthritis. Circulation 2003;108:2957–63.
29
[5] Van Doornum S, McColl G, Wicks IP. Accelerated atherosclerosis: an extraarticular feature of rheumatoid arthritis? Arthritis Rheum 2002;46:862–73. [6] Pham T, Gossec L, Constantin A, et al. Cardiovascular risk and rheumatoid arthritis: clinical practice guidelines based on published evidence and expert opinion. Joint Bone Spine 2006;73:379–87. [7] Andrews RC, Walker BR. Glucocorticoids and insulin resistance: old hormones, new targets. Clin Sci (Lond) 1999;96:513–23. [8] Whitworth JA, Mangos GJ, Kelly JJ. Cushing, cortisol, and cardiovascular disease. Hypertension 2000;36:912–6. [9] Girod JP, Brotman DJ. Does altered glucocorticoid homeostasis increase cardiovascular risk? Cardiovasc Res 2004;64:217–26. [10] Nashel DJ. Is atherosclerosis a complication of long-term corticosteroid treatment? Am J Med 1986;80:925–9. [11] Wei L, MacDonald TM, Walker BR. Taking glucocorticoids by prescription is associated with subsequent cardiovascular disease. Ann Intern Med 2004;141:764–70. [12] Rab ST, King 3rd SB, Roubin GS, et al. Coronary aneurysms after stent placement: a suggestion of altered vessel wall healing in the presence of anti-inflammatory agents. J Am Coll Cardiol 1991;18:1524–8. [13] Versaci F, Gaspardone A, Tomai F, et al. Immunosuppressive therapy for the prevention of restenosis after coronary artery stent implantation (IMPRESS Study). J Am Coll Cardiol 2002;40:1935–42. [14] Hafstrom I, Rohani M, Deneberg S, et al. Effects of low-dose prednisolone on endothelial function, atherosclerosis, and traditional risk factors for atherosclerosis in patients with rheumatoid arthritis–a randomized study. J Rheumatol 2007;34:1810–6. [15] Asanuma Y, Kawai S, Aoshima H, et al. Serum lipoprotein(a) and apolipoprotein(a) phenotypes in patients with rheumatoid arthritis. Arthritis Rheum 1999;42:443–7. [16] Dahlqvist SR, Engstrand S, Berglin E, et al. Conversion towards an atherogenic lipid profile in rheumatoid arthritis patients during long-term infliximab therapy. Scand J Rheumatol 2006;35:107–11. [17] Heldenberg D, Caspi D, Levtov O, et al. Serum lipids and lipoprotein concentrations in women with rheumatoid arthritis. Clin Rheumatol 1983;2:387–91. [18] Garcia-Gomez C, Nolla JM, Valverde J, et al. High HDL cholesterol in women with rhematoid arthritis on low-dose glucocorticoid therapy. Eur J Clin Invest 2008;38:686–92. [19] Westhovens R, Nijs J, Taelman V, et al. Body composition in rheumatoid arthritis. Br J Rheumatol 1997;36:444–8. [20] Wolfe F, Michaud K. The risk of myocardial infarction and pharmacologic and nonpharmacologic myocardial infarction predictors in rheumatoid arthritis: A cohort and nested case-control analysis. Arthritis Rheum 2008;58:2612–21. [21] Mazzantini M, Doveri M, Consensi A, et al. Incidence of adverse effects during long-term, low-dose glucocorticoid treatment of rheumatoid arthritis. Arthritis Rheum 2007;56 [Abstract [991]]. [22] van Everdingen AA, Jacobs JW, Siewertsz Van Reesema DR, et al. Low-dose prednisone therapy for patients with early active rheumatoid arthritis: clinical efficacy, disease-modifying properties, and side effects: a randomized, doubleblind, placebo-controlled clinical trial. Ann Intern Med 2002;136:1–12. [23] Dessein PH, Joffe BI, Stanwix AE, et al. Glucocorticoids and insulin sensitivity in rheumatoid arthritis. J Rheumatol 2004;31:867–74. [24] La Montagna G, Cacciapuoti F, Buono R, et al. Insulin resistance is an independent risk factor for atherosclerosis in rheumatoid arthritis. Diab Vasc Dis Res 2007;4:130–5. [25] Pappas DA, Rodriguez-Pla A, Bathon JM, et al. Glucocorticoid treatment is associated with increased insulin resistance in subjects with rheumatoid arthritis. Arthritis Rheum 2007 Apr 44;56 [Abstract [992]]. [26] Capell HA, Madhok R, Hunter JA, et al. Lack of radiological and clinical benefit over two years of low-dose prednisolone for rheumatoid arthritis: results of a randomised controlled trial. Ann Rheum Dis 2004;63:797–803. [27] Hickling P, Jacoby RK, Kirwan JR, Arthritis and Rheumatism Council Low-Dose Glucocorticoid Study Group. Joint destruction after glucocorticoids are withdrawn in early rheumatoid arthritis. Br J Rheumatol 1998;37:930–6. [28] Panoulas VF, Douglas KM, Stavropoulos-Kalinoglou A, et al. Long-term exposure to medium-dose glucocorticoid therapy associates with hypertension in patients with rheumatoid arthritis. Rheumatology (Oxford) 2008;47:72–5. [29] Chung CP, Oeser A, Solus JF, et al. Prevalence of the metabolic syndrome is increased in rheumatoid arthritis and is associated with coronary atherosclerosis. Atherosclerosis 2008;196:756–63. [30] Roubenoff R, Dellaripa P, Nadeau MR, et al. Abnormal homocysteine metabolism in rheumatoid arthritis. Arthritis Rheum 1997:718–22. [31] Sato H, Kazama JJ, Wada Y, et al. Decreased levels of circulating alpha2heremans-schmid glycoprotein/fetuin-A (AHSG) in patients with rheumatoid arthritis. Intern Med 2007;46:1685–91. [32] del Rincon I, O’Leary DH, Haas RW, et al. Effect of glucocorticoids on the arteries in rheumatoid arthritis. Arthritis Rheum 2004;50:3813–22. [33] Dessein PH, Joffe BI, Veller MG, et al. Traditional and nontraditional cardiovascular risk factors are associated with atherosclerosis in rheumatoid arthritis. J Rheumatol 2005;32:435–42. [34] Dessein PH, Norton GR, Woodiwiss AJ, et al. Influence of nonclassical cardiovascular risk factors on the accuracy of predicting subclinical atherosclerosis in rheumatoid arthritis. J Rheumatol 2007;34:943–51. [35] Gonzalez-Juanatey C, Llorca J, Testa A, et al. Increased prevalence of severe subclinical atherosclerotic findings in long-term treated rheumatoid arthritis patients without clinically evident atherosclerotic disease. Medicine (Baltimore) 2003;82:407–13.
30
A. Ruyssen-Witrand et al. / Joint Bone Spine 78 (2011) 23–30
[36] Kumeda Y, Inaba M, Goto H, et al. Increased thickness of the arterial intimamedia detected by ultrasonography in patients with rheumatoid arthritis. Arthritis Rheum 2002;46:1489–97. [37] Pahor A, Hojs R, Gorenjak M, et al. Accelerated atherosclerosis in premenopausal female patients with rheumatoid arthritis. Rheumatol Int 2006;27:119–23. [38] Park YB, Ahn CW, Choi HK, et al. Atherosclerosis in rheumatoid arthritis: morphologic evidence obtained by carotid ultrasound. Arthritis Rheum 2002;46:1714–9. [39] Maki-Petaja KM, Hall FC, Booth AD, et al. Rheumatoid arthritis is associated with increased aortic pulse-wave velocity, which is reduced by anti-tumor necrosis factor-alpha therapy. Circulation 2006;114:1185–92. [40] Tanaka K, Inaba M, Goto H, et al. Paraarticular trabecular bone loss at the ultradistal radius and increased arterial stiffening in postmenopausal patients with rheumatoid arthritis. J Rheumatol 2006;33:652–8. [41] Arosio E, De Marchi S, Rigoni A, et al. Forearm haemodynamics, arterial stiffness and microcirculatory reactivity in rheumatoid arthritis. J Hypertens 2007;25:1273–8. [42] Gonzalez-Juanatey C, Testa A, Garcia-Castelo A, et al. Echocardiographic and Doppler findings in long-term treated rheumatoid arthritis patients without clinically evident cardiovascular disease. Semin Arthritis Rheum 2004;33:231–8. [43] Montecucco C, Gobbi G, Perlini S, et al. Impaired diastolic function in active rheumatoid arthritis. Relationship with disease duration. Clin Exp Rheumatol 1999;17:407–12.
[44] Anichkov DA, Shostak NA, Ivanov DS. Heart rate variability is related to disease activity and smoking in rheumatoid arthritis patients. Int J Clin Pract 2007:777–83. [45] Nadareishvili Z, Michaud K, Hallenbeck JM, et al. Cardiovascular, rheumatologic, and pharmacologic predictors of stroke in patients with rheumatoid arthritis: a nested, case-control study. Arthritis Rheum 2008;59:1090–6. [46] Davis 3rd JM, Maradit Kremers H, Crowson CS, et al. Glucocorticoids and cardiovascular events in rheumatoid arthritis: a population-based cohort study. Arthritis Rheum 2007;56:820–30. [47] Solomon DH, Avorn J, Katz JN, et al. Immunosuppressive medications and hospitalization for cardiovascular events in patients with rheumatoid arthritis. Arthritis Rheum 2006;54:3790–8. [48] del Rincon I, Pogosian S, Escalante A. Daily and Cumulative threshold glucocorticoid dose associated with cardiovascular and all-cause mortality in rheumatoid arthritis. Arthritis Rheum 2007;56 [Abstract [289]]. [49] Gonzalez-Gay MA, Gonzalez-Juanatey C, Lopez-Diaz MJ, et al. HLA-DRB1 and persistent chronic inflammation contribute to cardiovascular events and cardiovascular mortality in patients with rheumatoid arthritis. Arthritis Rheum 2007;57:125–32. [50] Ruyssen-Witrand A, Mouterde G, Dernis E, et al. Tolérance des glucocorticoïdes dans la polyarthrite rhumatoïde: élaboration de recommandations pour la pratique clinique, à partir d’une analyse systématique de la littérature et de l’opinion d’experts. Rev Rhum 2008;75:S22–8 (Édition spéciale rencontres d’experts en rhumatologie).