Article

PDF
Access to the PDF text
Advertising


Free Article !

Archives of cardiovascular diseases
Volume 103, n° 4
pages 253-261 (avril 2010)
Doi : 10.1016/j.acvd.2010.03.007
Received : 3 December 2009 ;  accepted : 29 Mars 2010
High risk of clinical cardiovascular events in rheumatoid arthritis: Levels of associations of myocardial infarction and stroke through a systematic review and meta-analysis
Polyarthrite rhumatoïde et risque d’infarctus du myocarde, d’accident vasculaire cérébral : détermination du sur-risque à partir d’une méta-analyse
 

Christophe Meune a, , Emmanuel Touzé b, Ludovic Trinquart c, Yannick Allanore d
a Département de cardiologie, université Paris-Descartes, hôpital Cochin, Assistance publique–Hôpitaux de Paris, 27, rue du Faubourg St.-Jacques, 75014 Paris, France 
b Inserm UMR 894, département de neurologie, université Paris-Descartes, centre de psychiatrie et neurosciences, hôpital Sainte-Anne, Paris, France 
c Inserm CIE 4, unité de recherche clinique, université Paris-Descartes, hôpital européen Georges-Pompidou, Paris, France 
d Département de rhumatologie A, université Paris-Descartes, hôpital Cochin, Assistance publique–Hôpitaux de Paris, Paris, France 

Corresponding author.
Summary
Background

While there are convergent data suggesting that overall cardiovascular mortality is increased in patients with rheumatoid arthritis, the relative contributions of myocardial infarction and stroke remain unclear.

Aims

We sought to clarify this issue by conducting a meta-analysis of cohort studies on myocardial infarction and stroke in patients with rheumatoid arthritis.

Methods

A MEDLINE search from January 1960 to September 2009 and abstracts from international conferences from 2007 to 2009 were searched for relevant literature. All cohort studies reporting on standardized mortality ratio or incidence rate ratio of myocardial or stroke associated with rheumatoid arthritis, with available crude numbers, were included. STATA meta-analysis software was used to calculate pooled risk estimates.

Results

Seventeen papers fulfilled the inclusion criteria, corresponding to a total of 124,894 patients. Ten studies reported on standardized mortality ratio for fatal myocardial infarction, which ranged from 0.99 to 3.82. The overall pooled estimate was 1.77 (95% confidence interval [CI] 1.65–1.89). Incidence rate ratio for myocardial infarction was reported in five studies; the pooled estimate was 2.10 (95% CI 1.52–2.89). Nine studies reported on fatal stroke, with standardized mortality ratio ranging from 1.08 to 2.00; the pooled estimate was 1.46 (95% CI 1.31–1.63). The pooled incidence rate ratio for stroke (three studies) was 1.91 (95% CI 1.73–2.12).

Conclusion

Our results show that risks of myocardial infarction and stroke are increased in patients with rheumatoid arthritis. In addition, both account for the observed increased mortality in individuals with rheumatoid arthritis.

The full text of this article is available in PDF format.
Résumé
Introduction

La mortalité cardiaque globale est augmentée au cours de la polyarthrite rhumatoïde, mais le niveau de risque de survenue d’un infarctus du myocarde et d’un accident vasculaire cérébral demeurent incertains.

Objectif

Préciser ce niveau de risque à l’aide d’une méta-analyse.

Méthodes

La recherche a été effectuée sur Medline, de janvier 1960 à septembre 2009 et à partir des abstracts des conférences internationales de 2007 à 2009. Les études de cohorte rapportant les risques de mortalité (SMR) et les taux d’incidence (IRR) d’infarctus du myocarde ou d’accident vasculaire cérébral au cours de la polyarthrite rhumatoïde ont été inclues.

Résultats

Dix-sept articles et/ou abstracts ont été retenus, ce qui correspond à une population totale de 124 894 patients. Dix études rapportent le SMR pour l’infarctus du myocarde fatal; il varie entre 0,99 et 3,82 selon les études. Le SMR global calculé est de 1,77 (IC 95 % : 1,65–1,89). Le risque de survenue d’infarctus du myocarde total a été rapporté dans cinq études ; le risque global calculé est de 2,10 (IC 95 % : 1,52–2,89). Neuf études ont rapporté le risque d’accident vasculaire cérébral fatal avec des SMR compris entre 1,08 et 2,00 ; le SMR global est de 1,46 (IC 95 % : 1,31–1,63). Le risque d’accident vasculaire cérébral total est de 1,91 (IC 95 % : 1,73–2,12).

Conclusion

Nos résultats montrent qu’il existe un sur-risque d’infarctus du myocarde et d’accident vasculaire cérébral au cours de la polyarthrite rhumatoïde ; tous deux contribuent à la surmortalité observée au cours de la polyarthrite rhumatoïde.

The full text of this article is available in PDF format.

Keywords : Meta-analysis, Myocardial infarction, Rheumatoid arthritis, Stroke

Mots clés : Polyarthrite rhumatoïde, Infarctus du myocarde, Accident vasculaire cérébral, Méta-analyse

Abbreviations : CI, CV, IRR, MI, RA, SMR


Background

Rheumatoid arthritis (RA) is a chronic systemic inflammatory disease, of unknown origin, that affects 0.5–1% of the adult population [1, 2]. Over the past decade, several studies have shown that RA is associated with an increased cardiovascular (CV) mortality [3, 4, 5, 6]. This has been confirmed by some meta-analyses that documented a 50–60% increase in CV mortality in RA patients [7, 8, 9]. However, although some studies specifically reported the risk of myocardial infarction (MI), there are fewer data on stroke. Additionally, it has been suggested that the impact of RA on the risk of stroke may be weaker [10]. Two meta-analyses investigated concomitantly CV mortality, MI and stroke, but these studies had some limitations [7, 8]. One pooled standardized mortality ratio (SMR) and incidence rate ratio (IRR), and was not exhaustive [7]. The other meta-analysis did not include studies performed after 2005, a period of major advances in disease-modifying antirheumatic drugs [8]. Therefore, the relative impact of RA on MI and stroke remain unclear. To clarify this issue, we performed a systematic review of published studies to assess the risk of MI and stroke in RA patients.

Methods
Search methods for identification of studies

We sought studies published between January 1960 and September 2009, without restriction in settings or language. Electronic searches were performed using MEDLINE, with both keywords and free text words “rheumatoid arthritis” and (“mortality” or “myocardial infarction” or “cardiovascular disease” or “stroke” or “cerebrovascular accident” or “coronary artery disease”). We reviewed the reference lists of all included studies, any relevant guidelines, general reviews, and commentaries pertaining to the management of patients with rheumatoid arthritis. To identify recent studies not yet published as full papers, we also searched books of abstracts from 2007 to 2009 conferences (European League against Rheumatism [EULAR] and American College of Rheumatology [ACR]).

Inclusion criteria

Eligible studies were cohort studies of patients with RA diagnosed according to the recommendations available during the study period [11]. Articles had to report either enough data to compute a SMR or an IRR for at least one of the following outcomes: fatal or non-fatal MI, fatal or non-fatal stroke. Studies that reported death related to “ischemic heart disease”, “arteriosclerotic heart disease” or “coronary artery disease” were considered as relevant; their results were computed as fatal MI. In cases of multiple publications from the same cohort, we chose the study that provided the greatest amount of information (i.e., the largest number of patient-years).

Study selection

Two reviewers independently reviewed the studies to assess their eligibility for inclusion in the review. Disagreements were resolved by consensus. The two reviewers extracted data from the selected studies separately. Disagreements were resolved by consensus among all authors. To describe population characteristics, we abstracted the following: geographic area, mean age, female proportion, and mean duration of disease. To assess quality criteria, we abstracted the following: prospective study, consecutive enrolment, population-based study and completeness of follow-up.

Statistical analysis

SMR and IRR and their 95% confidence intervals (CIs) were calculated with crude numbers extracted from individual studies. Publication bias was assessed graphically, using a funnel plot [12].

If SMRs were reported without 95% CI, the interval was calculated using observed and expected number of deaths and assuming a Poisson distribution of observed cases. Pooled estimates of SMR for fatal MI and fatal stroke–IRR for MI and stroke–were calculated using the inverse variance method with the STATA meta-analysis program (STATA 10 software, StataCorp L, College Station, TX, USA). According to the heterogeneity test (significance at 5% level) either a fixed or a random-effects model was applied. To estimate the influence of individual studies on the summary effect estimate, meta-analysis estimates were re-computed omitting one study at a time.

Results

Among a total of 1939 references identified, 16 studies fulfilled our inclusion criteria. From abstract books of conferences we retrieved one abstract that fulfilled our inclusion criteria. Thus, we ultimately selected and included 17 studies, including one abstract, corresponding to a total population of 124,894 patients (Figure 1). The characteristics of the studies included in the review are given in Table 1, Table 2. SMRs stratified by specific cohort characteristics are presented in Table 3 as highest, lowest and pooled estimates.



Figure 1


Figure 1. 

Study flow chart. MI: myocardial infarction; SMR: standardized mortality ratio; IRR: incidence rate ratio.

Zoom

Funnel plots did not show evidence of publication bias (data not shown).

Fatal and non-fatal myocardial infarction

Ten studies reported on SMR for fatal MI, corresponding to a total of 85,323 patients [13, 14, 15, 16, 17, 18, 19, 20, 21, 22] (Table 1). SMR from fatal MI ranged from 0.99 to 3.82. Seven studies demonstrated a significant increase in fatal MI [13, 14, 16, 17, 19, 20, 22], whereas the three remaining did not. Data were pooled by use of the random-effects model, as the heterogeneity test was significant (χ2=31, P <0.001). The pooled SMR was 1.77 (95% CI 1.65–1.89) (Figure 2, Table 3). The omission of one study from a subsequent analysis had no significant impact on the meta-analysis estimates.



Figure 2


Figure 2. 

Forest plot of myocardial infarction standardized mortality ratio (SMR) for patients with rheumatoid arthritis. Each square represents an individual SMR estimate, the size of the square being proportional to the weight of the study. The lines represent the 95% confidence interval for the point estimate in each study.

Zoom

Three studies included patients from Sweden [18, 19] or Scotland [22], regions assumed to be at high risk of cardiovascular events. Mortality rates were similar in these studies conducted in Sweden/Scotland and in those performed in the USA/England (Table 3). Five studies reported mean disease duration in their cohort, which was less than 1year in three studies [15, 16, 17] and greater than 1year in the two remaining studies [18, 20]. Pooled SMRs disclosed similar mortality in patients with less than 1-year of disease duration than in the remaining patients (Table 3).

Five studies, including one study as yet only published as an abstract [23], reported on IRR for total MI [4, 23, 24, 25, 26], corresponding to a total of 34,705 patients. The pooled estimate of the IRR was 2.10 (1.52–2.89; P <0.0001) (Figure 3). The exclusion of the abstract did not alter the pooled estimate of IRR.



Figure 3


Figure 3. 

Incidence rate ratio (IRR) for myocardial infarction in rheumatoid arthritis with 95% confidence intervals and the combined IRR.

Zoom

Fatal and non-fatal stroke

The nine included studies (88,500 patients) reported SMR from fatal stroke ranging from 1.08 to 2.00 [13, 14, 15, 16, 18, 19, 22, 27, 28]. Data were pooled by use of the random-effects model, as the heterogeneity test was significant (χ2=34, P <0.001). The pooled SMR for stroke was 1.46 (95% CI 1.31–1.63) (Figure 4). The omission of one study in an alternative analysis had no impact on the meta-analysis estimates. Four studies reported the risk of stroke among patients from Scandinavia or Scotland [18, 19, 22, 28]; the pooled SMR for stroke was similar in these patients from Scandinavia/Scotland than in the remaining patients from the USA/England (Table 3). Only three studies reported mean disease duration in their cohort, which was less than 1year in two studies [15, 16] and greater than 1 year in a single cohort [18]. SMRs for stroke were similar among these studies (Table 3).



Figure 4


Figure 4. 

Forest plot of stroke standardized mortality ratio (SMR) for patients with rheumatoid arthritis. Each square represents an individual SMR estimate, the size of the square being proportional to the weight of the study. The lines represent the 95% confidence interval for the point estimate in each study.

Zoom

Lastly, three studies reported on IRR for total stroke (26,143 patients) [4, 24, 25]. The pooled estimate was 1.91 (95% CI 1.73–2.12; P <0.0001) (Figure 5).



Figure 5


Figure 5. 

Incidence rate ratio (IRR) for stroke in rheumatoid arthritis with 95% confidence intervals and the combined IRR.

Zoom

Discussion

Our study is an update of the risk of fatal and non-fatal MI and stroke in RA. Its main result is that the risks of MI and stroke are increased in RA patients.

The primary strengths of our study are the inclusion of recent studies, the application of strict search and inclusion criteria, and the assessment of population-based cohort studies only. In addition, all included studies reported data on patients with RA defined according to recommendations available at the moment of the study.

It has been suggested that the impact of RA on the risk of stroke may be weaker possibly because of the longer time to occurrence of cerebrovascular events [10]. Our study shows that both MI and stroke are more frequent in RA patients than in the general population, and suggest that both MI and stroke account for a similar extent in the observed mortality and CV events observed in RA. Our results are consistent with both our knowledge of atherosclerosis as a systemic disease [29], and the hypotheses that have been generated to explain such an increase in CV events in RA patients. Traditional risk factors for atherosclerosis should be considered as important contributors to CV events in RA patients [6, 30, 31, 32]. Indeed, atherogenic lipid profile may warrant special consideration, as previous studies have demonstrated that RA patients have raised low-density lipoprotein, and do have an atherogenic lipid profile when compared with controls, even before the onset of articular involvement [33, 34]. Inflammation by itself may promote atherosclerotic plaque development, increase their vulnerability and promote thrombosis [18, 35, 36, 37, 38]. As a consequence, one may expect that all forms of atherosclerotic disease should be increased in RA.

Age at inclusion and disease duration may be important parameters to consider when analysing the literature. However, age ranged from 51 to 61.8 years in the studies included in our analysis, with most patients aged 50–55 years at inclusion. Therefore, we cannot derive any conclusion concerning the influence of age in our study. Pooled SMR were similar in studies that focused on patients with disease duration less than 1 year than in studies that reported patients with more prolonged disease duration (Table 3). In their study, Gunnarsson et al. documented that RA patients were not reported to have increased MI by the time of the diagnosis whereas the risk of MI increased significantly during follow-up. This may be interpreted as a progressive risk over time, possibly due to prolonged disease, older age or both [23]. We must, however, acknowledge that disease duration may be hard to ascertain in some patients and was not reported in many studies.

In our analysis, we did not find any association between the time of the study and both the quality of the studies (i.e., adequate follow-up, validation of events) or the existence of increased risk of MI or stroke. This is in accordance with our previous report that mortality was not significantly altered during the past 50 years [9]. Such results could be regarded as in conflict with advances in both the management of RA and cardiovascular disease. In fact, there is strong evidence that TNF⍺ inhibitors may alter the CV risk of these patients. TNF⍺ inhibitors increase high-density lipoprotein (HDL) cholesterol, may reverse metabolic insulin resistance, reduce C-reactive protein concentration [10], and ultimately reduce the rate of first CV events according to a longitudinal observational study [39]. However, one should keep in mind that SMR is the ratio of observed versus expected mortality. Adequate interpretation should be that MI and stroke have decreased in both the general population and RA patients with a similar extra risk in RA patients [9].

Our study has some limitations. Our analysis showed heterogeneity among study results for MI and stroke. However, we analysed population-based cohort studies that allowed the determination of SMR and IRR and used strict inclusion criterion. In addition, we used a random-effect model to include an estimate of the between-studies variability [40] and therefore assume that our results should be considered as robust. Most of the studies evaluated the prognosis of patients with disease onset in the 1980s or early 1990s, a period when disease-modifying antirheumatic drugs were poorly prescribed and only minimal attention was paid to control of inflammation and prevention of cardiovascular events. Some studies do support the beneficial effect of these drugs on mortality; their impact upon MI and stroke remain to be demonstrated. Lastly, SMR and IRR evaluated the association between RA and measured outcomes (MI, stroke) adjusted for age and sex and not for traditional risk factors.

Conflict of interest statement

None.

References

Roux C.H., Saraux A., Le Bihan E., and al. Rheumatoid arthritis and spondyloarthropathies: geographical variations in prevalence in France J Rheumatol 2007 ;  34 : 117-122
Sangha O. Epidemiology of rheumatic diseases Rheumatology (Oxford) 2000 ;  39 (Suppl 2) : 3-12
Watson D.J., Rhodes T., Guess H.A. All-cause mortality and vascular events among patients with rheumatoid arthritis, osteoarthritis, or no arthritis in the UK General Practice Research Database J Rheumatol 2003 ;  30 : 1196-1202
Solomon D.H., Karlson E.W., Rimm E.B., and al. Cardiovascular morbidity and mortality in women diagnosed with rheumatoid arthritis Circulation 2003 ;  107 : 1303-1307 [cross-ref]
DeMaria A.N. Relative risk of cardiovascular events in patients with rheumatoid arthritis Am J Cardiol 2002 ;  89 : 33D-38D [cross-ref]
Assous N., Touzé E., Meune C., Kahan A., Allanore Y. Cardiovascular disease in rheumatoid arthritis: single-center hospital-based cohort study in France Joint Bone Spine 2007 ;  74 : 66-72 [cross-ref]
Levy L., Fautrel B., Barnetche T., Schaeverbeke T. Incidence and risk of fatal myocardial infarction and stroke events in rheumatoid arthritis patients. A systematic review of the literature Clin Exp Rheumatol 2008 ;  26 : 673-679
Aviña-Zubieta J.A., Choi H.K., Sadatsafavi M., Etminan M., Esdaile J.M., Lacaille D. Risk of cardiovascular mortality in patients with rheumatoid arthritis: a meta-analysis of observational studies Arthritis Rheum 2008 ;  59 : 1690-1697
Meune C., Touzé E., Trinquart L., Allanore Y. Trends in cardiovascular mortality in patients with rheumatoid arthritis over 50years: a systematic review and meta-analysis of cohort studies Rheumatology (Oxford) 2009 ;  48 : 1309-1313 [cross-ref]
Avouac J., Allanore Y. Cardiovascular risk in rheumatoid arthritis: effects of anti-TNF drugs Expert Opin Pharmacother 2008 ;  9 : 1121-1128 [cross-ref]
Arnett F.C., Edworthy S.M., Bloch D.A., and al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis Arthritis Rheum 1988 ;  31 : 315-324 [cross-ref]
Egger M., Davey Smith G., Schneider M., Minder C. Bias in meta-analysis detected by a simple, graphical test BMJ 1997 ;  315 : 629-634
Monson R.R., Hall A.P. Mortality among arthritics J Chronic Dis 1976 ;  29 : 459-467 [cross-ref]
Prior P., Symmons D.P., Scott D.L., Brown R., Hawkins C.F. Cause of death in rheumatoid arthritis Br J Rheumatol 1984 ;  23 : 92-99
Reilly P.A., Cosh J.A., Maddison P.J., Rasker J.J., Silman A.J. Mortality and survival in rheumatoid arthritis: a 25-year prospective study of 100patients Ann Rheum Dis 1990 ;  49 : 363-369 [cross-ref]
Young A., Koduri G., Batley M., and al. Mortality in rheumatoid arthritis. Increased in the early course of disease, in ischaemic heart disease and in pulmonary fibrosis Rheumatology (Oxford) 2007 ;  46 : 350-357
Goodson N., Marks J., Lunt M., Symmons D. Cardiovascular admissions and mortality in an inception cohort of patients with rheumatoid arthritis with onset in the 1980s and 1990s Ann Rheum Dis 2005 ;  64 : 1595-1601 [cross-ref]
Wallberg-Jonsson S., Ohman M.L., Dahlqvist S.R. Cardiovascular morbidity and mortality in patients with seropositive rheumatoid arthritis in Northern Sweden J Rheumatol 1997 ;  24 : 445-451
Björnådal L., Baecklund E., Yin L., Granath F., Klareskog L., Ekbom A. Decreasing mortality in patients with rheumatoid arthritis: results from a large population based cohort in Sweden, 1964–95 J Rheumatol 2002 ;  29 : 906-912
Erhardt C.C., Mumford P.A., Venables P.J., Maini R.N. Factors predicting a poor life prognosis in rheumatoid arthritis: an eight year prospective study Ann Rheum Dis 1989 ;  48 : 7-13 [cross-ref]
Lewis P., Hazleman B.L., Hanka R., Roberts S. Cause of death in patients with rheumatoid arthritis with particular reference to azathioprine Ann Rheum Dis 1980 ;  39 : 457-461 [cross-ref]
Thomas E., Symmons D.P., Brewster D.H., Black R.J., Macfarlane G.J. National study of cause-specific mortality in rheumatoid arthritis, juvenile chronic arthritis, and other rheumatic conditions: a 20-year follow up study J Rheumatol 2003 ;  30 : 958-965
Gunnarsson M, Alfredsson L, Jacobsson L, Klareskog L, Dahlqvist SR, Askling L. From no increase to doubled risk after ten years: morbidity and mortality from myocardial infarction in newly diagnosed rheumatoid arthritis. American College of Rheumatology Congress 2008 (abstract). 2008.
del Rincón I.D., Williams K., Stern M.P., Freeman G.L., Escalante A. High incidence of cardiovascular events in a rheumatoid arthritis cohort not explained by traditional cardiac risk factors Arthritis Rheum 2001 ;  44 : 2737-2745
Solomon D.H., Goodson N.J., Katz J.N., and al. Patterns of cardiovascular risk in rheumatoid arthritis Ann Rheum Dis 2006 ;  65 : 1608-1612 [cross-ref]
Nicola P.J., Crowson C.S., Maradit-Kremers H., and al. Contribution of congestive heart failure and ischemic heart disease to excess mortality in rheumatoid arthritis Arthritis Rheum 2006 ;  54 : 60-67 [cross-ref]
Wolfe F., Mitchell D.M., Sibley J.T., and al. The mortality of rheumatoid arthritis Arthritis Rheum 1994 ;  37 : 481-494 [cross-ref]
Allebeck P. Increased mortality in rheumatoid arthritis Scand J Rheumatol 1982 ;  11 : 81-86 [cross-ref]
Aronow W.S., Ahn C. Prevalence of coexistence of coronary artery disease, peripheral arterial disease, and atherothrombotic brain infarction in men and women>or=62 years of age Am J Cardiol 1994 ;  74 : 64-65 [cross-ref]
McEntegart A., Capell H.A., Creran D., Rumley A., Woodward M., Lowe G.D. Cardiovascular risk factors, including thrombotic variables, in a population with rheumatoid arthritis Rheumatology (Oxford) 2001 ;  40 : 640-644 [cross-ref]
Choi H.K., Seeger J.D. Lipid profiles among US elderly with untreated rheumatoid arthritis–the Third National Health and Nutrition Examination Survey J Rheumatol 2005 ;  32 : 2311-2316 [cross-ref]
Manfredsdottir V.F., Vikingsdottir T., Jonsson T., and al. The effects of tobacco smoking and rheumatoid factor seropositivity on disease activity and joint damage in early rheumatoid arthritis Rheumatology (Oxford) 2006 ;  45 : 734-740 [cross-ref]
van Halm V.P., Nielen M.M., Nurmohamed M.T., and al. Lipids and inflammation: serial measurements of the lipid profile of blood donors who later developed rheumatoid arthritis Ann Rheum Dis 2007 ;  66 : 184-188
Hurt-Camejo E., Paredes S., Masana L., and al. Elevated levels of small, low-density lipoprotein with high affinity for arterial matrix components in patients with rheumatoid arthritis: possible contribution of phospholipase A2 to this atherogenic profile Arthritis Rheum 2001 ;  44 : 2761-2767 [cross-ref]
Warrington K.J., Kent P.D., Frye R.L., and al. Rheumatoid arthritis is an independent risk factor for multi-vessel coronary artery disease: a case control study Arthritis Res Ther 2005 ;  7 : R984-R991
Del Rincón I., Williams K., Stern M.P., Freeman G.L., O’Leary D.H., Escalante A. Association between carotid atherosclerosis and markers of inflammation in rheumatoid arthritis patients and healthy subjects Arthritis Rheum 2003 ;  48 : 1833-1840
Solomon D.H., Curhan G.C., Rimm E.B., Cannuscio C.C., Karlson E.W. Cardiovascular risk factors in women with and without rheumatoid arthritis Arthritis Rheum 2004 ;  50 : 3444-3449 [cross-ref]
Ross R. Atherosclerosis–an inflammatory disease N Engl J Med 1999 ;  340 : 115-126 [cross-ref]
Jacobsson L.T., Turesson C., Gülfe A., and al. Treatment with tumor necrosis factor blockers is associated with a lower incidence of first cardiovascular events in patients with rheumatoid arthritis J Rheumatol 2005 ;  32 : 1213-1218
Higgins J.P., Thompson S.G., Deeks J.J., Altman D.G. Measuring inconsistency in meta-analyses BMJ 2003 ;  327 : 557-560 [cross-ref]



© 2010  Elsevier Masson SAS. All Rights Reserved.
EM-CONSULTE.COM is registrered at the CNIL, déclaration n° 1286925.
As per the Law relating to information storage and personal integrity, you have the right to oppose (art 26 of that law), access (art 34 of that law) and rectify (art 36 of that law) your personal data. You may thus request that your data, should it be inaccurate, incomplete, unclear, outdated, not be used or stored, be corrected, clarified, updated or deleted.
Personal information regarding our website's visitors, including their identity, is confidential.
The owners of this website hereby guarantee to respect the legal confidentiality conditions, applicable in France, and not to disclose this data to third parties.
Close
Article Outline