Article Text

Download PDFPDF

Extended report
Increased expression of the endothelin system in arterial lesions from patients with giant-cell arteritis: association between elevated plasma endothelin levels and the development of ischaemic events
  1. E Lozano,
  2. M Segarra,
  3. M Corbera-Bellalta,
  4. A García-Martínez,
  5. G Espígol-Frigolé,
  6. A Plà-Campo,
  7. J Hernández-Rodríguez,
  8. M C Cid
  1. Vasculitis Research Unit, Department of Systemic Autoimmune Diseases, Hospital Clínic, University of Barcelona, Institut d’Investigacions Biomèdiques August Pi Sunyer (IDIBAPS), Barcelona, Spain
  1. Correspondence to Dr M C Cid, Vasculitis Research Unit, Department of Systemic Autoimmune Diseases, Hospital Clínic, Villarroel 170, 08036 Barcelona, Spain; mccid{at}


Objective: Approximately 15–20% of patients with giant-cell arteritis (GCA) develop ischaemic complications often preceded by transient ischaemia. The expression of the endothelin (ET) system in GCA lesions was investigated to assess its relationship with the development of ischaemic complications.

Methods: Plasma ET-1 was quantified by immunoassay in 61 patients with biopsy-confirmed GCA and 16 healthy donors. ET-1, endothelin-converting enzyme (ECE-1) and endothelin receptor (ETAR and ETBR) messenger RNA were measured by real-time quantitative reverse transcriptase–PCR in temporal arteries from 35 of these patients and 19 control arteries. Proteins were measured by immunoassay and Western blot.

Results: ET-1 concentration was increased at the protein level in temporal artery samples from GCA patients compared with controls (0.98 (SEM 0.32) vs 0.28 (SEM 0.098) fmol/mg, p = 0.028). ECE-1, ETAR and ETBR/actin ratios (Western blot) were also significantly higher in GCA patients. Intriguingly, mRNA expression of ET-1, ECE-1 and both receptors was significantly reduced in GCA lesions compared with control arteries. When investigating mechanisms underlying these results, platelet-derived growth factor and IL-1β, present in GCA lesions, were found to downregulate ET-1 mRNA in cultured human temporal artery-derived smooth muscle cells. Glucocorticoid treatment for 8 days did not result in significantly decreased endothelin tissue concentration (0.87 (SEM 0.2) vs 0.52 (SEM 0.08); p = 0.6). Plasma endothelin concentrations were higher in patients with ischaemic complications (1.049 (SEM 0.48) vs 1.205 (SEM 0.63) pg/ml, p = 0.032).

Conclusions: The endothelin system is increased at the protein level in GCA lesions creating a microenvironment prone to the development of ischaemic complications. Recovery induced by glucocorticoids is delayed, indicating persistent exposure to endothelin during initial treatment.

Statistics from

Giant-cell arteritis (GCA) is a chronic granulomatous vasculitis preferentially targeting large and medium-sized arteries.1 Inflammatory involvement triggers a vascular remodelling response resulting in lumen reduction with the potential for ischaemic complications.2 3 4 5 6 7 8 9

Partial or complete visual loss, usually due to anterior ischaemic optic neuropathy (AION), occurs in approximately 15–20% of patients. Visual impairment is the most common ischaemic complication in GCA and has a deep impact on patients’ quality of life.10 11 12 13 14 15 16 Scarce necropsy studies focusing on the anatomical substrate of AION have shown that posterior ciliary and cilioretinal arteries supplying the optic nerve and retina may be involved by GCA lesions17 18 19 Irreversible visual loss is often preceded by recurrent episodes of transient blindness (amaurosis fugax), a medical emergency highly predictive of subsequent permanent visual defects in most series.10 11 12 13 14 15 The transient and often repetitive nature of amaurosis fugax suggests the contribution of inflammation-induced vasospastic phenomena in the pathogenesis of ischaemic complications.

The endothelin (ET) family of 21 amino acid peptides includes three members, ET-1, ET-2, and ET-3. ET-1 is, by far, the main isoform produced in the cardiovascular system and is one of the most powerful vasoconstrictors identified.20 21 22 23 24 The ET-1 encoding gene generates a transcript, which translates into a protein, pre-pro ET-1, intracellularly processed by a furin-like endopeptidase yielding an ET-1 precursor, big-ET-1, with low biological activity.20 24 Cleavage of big-ET-1 by the metalloprotease endothelin-converting enzyme (ECE), generates the active peptide ET-1.24 ET-1 is constitutively produced in blood vessels by endothelial cells and vascular smooth muscle cells (VSMC) and is also expressed by activated macrophages.25 Endothelin peptides exert their vasoactive activity on VSMC through endothelin receptors A (ETAR) and B (ETBR). Binding to ETAR elicits VSMC contraction, whereas, in some instances, binding to ETBR may trigger vasodilatation. ETBR is also involved in clearing circulating ET-1 by endocytosis, particularly in the lung vasculature.21 23

Based on the characteristics of the ischaemic complications of GCA, we hypothesised that the endothelin axis might play a role in inflammation-induced vasospastic phenomena leading to amaurosis fugax and contributing to visual loss in GCA. The aim of the study was to investigate the expression and regulation of the endothelin system in GCA lesions and to assess whether increased endothelin production is associated with the development of disease-related ischaemic events. This is the first study quantitatively exploring the expression of the endothelin system (ET-1, ECE-1, ETAR and ETBR) at both the messenger RNA and protein level in a human inflammatory vasculopathy.

Patients and methods


Between 1997 and 2006, 158 patients were diagnosed with biopsy-confirmed GCA in our institution (Hospital Clinic, Barcelona). Clinical data were collected at the time of diagnosis and recorded in a database. Plasma-EDTA could be obtained from 61 of these patients during active disease (before starting therapy (43 patients) or after a single prednisone dose (18 patients)) and stored at −80°C. From 35 of them, total RNA could be extracted from frozen fragments of the temporal artery biopsies excised before starting treatment (30 patients) or after a single prednisone dose (five patients). Protein extracts could be obtained from 24 of these patients. Preliminary experiments showed no significant differences in any of the components of the endothelin axis between treatment-naive patients and patients who had received a single prednisone dose.

Clinical findings of the study group (table 1) were similar to previously published series except for a higher frequency of ischaemic complications, which served well the purpose of this study.10 12 13 14 Twenty-six out of these 61 patients developed GCA-related ischaemic complications. Nineteen suffered permanent visual loss (four binocular and 15 monocular) in 14 due to AION, in two due to retinal ischaemia, in one due to central retinal artery occlusion and in two fundoscopy was normal. In four out of the 19 patients, visual loss was preceded by amaurosis fugax and in two by transient diplopia. Two additional patients had stroke, two reversible amaurosis fugax, one transient diplopia, one a transient ischaemic attack and one tongue ischaemia. Patients were considered to have a weak systemic inflammatory response when they had up to two of the following: fever greater than 37°C; weight loss greater than 3 kg; haemoglobin less than 110 g/l; erythrocyte sedimentation rate of 85 mm or greater. Patients with three to four of these findings were considered to have a strong systemic inflammatory response, as reported3 10 26 27

Table 1

Clinical findings in the study cohort patients with GCA

For control purposes, plasma-EDTA was obtained from 16 volunteers with similar age (median 73 years, range 60–87) and gender distribution (10 women and six men). Controls had no past history of cancer, chronic inflammatory disease or recent (⩽3 months) infection. As controls for biopsy studies, we included consecutive temporal artery biopsies from 19 patients (13 women and six men) with a median of 77 years (range 64–91) in whom GCA was initially considered but were subsequently diagnosed as having other diseases. The prevalence of cardiovascular risk factors (smoking habit, hypertension, diabetes or hypercholesterolaemia) was similar between patients and controls.

To assess the effect of glucocorticoids on the endothelin system in lesions we consecutively selected temporal artery samples from 18 additional patients (labelled as treated patients) diagnosed within the same period of time with biopsy-confirmed GCA who had received prednisone (1 mg/kg per day) for an average of 8 days (range 3–20) when the temporal artery biopsy was excised.

The study was approved by the Ethics Committee of our institution (Hospital Clínic). Patients signed informed consent for the collection and storage of biological material.


Human ET-1 was purchased from MP Biomedicals (Aurora, Ohio, USA). Recombinant human transforming growth factor beta (TGFβ), tumour necrosis factor alpha (TNFα), IL-1β, IL-6 and purified platelet-derived growth factor (PDGF)-AB were obtained from R&D Systems (Minneapolis, Minnesota, USA).

Measurement of circulating endothelin

Plasma was concentrated fourfold by speed-vacuum lyophilisation, and the ET-1 concentration was determined in the entire cohort of patients and controls by using the Parameter ELISA kit (R&D Systems) according to the manufacturer’s protocol. This kit exhibits cross-reactivity with other endothelin peptides as follows: ET-2 45%; ET-3 14%; and big-ET less than 1%.

Measurement of the endothelin system mRNA by quantitative real-time RT–PCR

Temporal artery biopsies were embedded in optimal cutting temperature (OCT, Sakura, The Netherlands), snap-frozen in liquid nitrogen and stored at −80°C until used. Sections consecutive to those that provided the histopathological diagnosis were processed for RNA isolation using TRIzol reagent (Invitrogen, Carlsbad, California, USA).

Total RNA (1 μg) was reverse transcribed to complementary DNA using the Archive kit (Applied Biosystems, Foster City, California, USA) in a final volume of 100 μl, employing random hexamer priming. cDNA was measured by quantitative real-time PCR using specific Pre-Developed TaqMan gene expression assays (ET-1 Hs00174961_m1, ET-2 Hs00266516_m1, ECE-1 Hs00154837_m1, ETAR Hs00609865_m1 and ETBR Hs00240747_m1) from Applied Biosystems, as previously described.9

Endothelin system protein detection in tissue


Temporal artery protein extracts were obtained from the phenol phase during RNA isolation. The ET-1 concentration was determined by immunoassay (Biomedica Medizinprodukte GmbH, Vienna, Austria) and normalised for total protein content. This kit exhibits cross-reactivity with other endothelin peptides as follows: ET-2 100%; ET-3 less than 5%; and big-ET less than 1%.

Western blot analysis

Twenty-five micrograms of protein per condition were resolved on 10% reducing sodium dodecyl sulphate–polyacrylamide gel electrophoresis gels and blotted onto nitrocellulose membranes (Invitrogen). Immunodetection was performed with the following antibodies: goat anti-human ECE-1 (R&D Systems), mouse anti-human ETAR (BD Biosciences Pharmingen, Franklin Lakes, New Jersey, USA) and rabbit polyclonal anti-human ETBR (Abcam, Cambridge, UK) at 1 : 1000 dilution. Chemiluminescence signals were measured with the LAS-3000 imaging system (Fujifilm Corporation, Tokyo, Japan).

Cell culture

Primary culture of HTASMC

Human temporal artery-derived vascular smooth muscle cells (HTASMC) were isolated from fresh remnant segments of temporal artery biopsies from patients with GCA, as previously described.9 Cells obtained by this method had a myointimal phenotype as confirmed by the expression of α-smooth muscle actin by flow cytometry and type I collagen expression by reverse transcriptase (RT)–PCR, as reported.9

HUVEC isolation and culture

Human umbilical vein endothelial cells (HUVEC) were isolated from freshly delivered cords as reported.28 The growth medium consisted of M199 (Invitrogen) supplemented with 20% iron-fortified bovine calf serum (Hyclone Laboratories, Logan, Utah, USA), 200 μg/ml endothelial cell growth supplement (BD Biosciences), 100 U/ml penicillin/streptomycin, 50 μg/ml gentamycin, 2.5 μg/ml amphotericin B, 2 mM glutamine and 50 U/ml sodium heparin.


The Mann–Whitney test and Spearman’s rho correlation coefficient were used for statistical analysis.


ET-1, ECE-1 and ETAR and ETBR are overexpressed in GCA lesions

As displayed in fig 1A, temporal artery biopsies from patients with GCA contained significantly higher concentrations of ET-1 than normal arteries (0.979 (SEM 0.315) vs 0.280 (SEM 0.098) fmol/mg total protein; p = 0.028). ECE-1, ETAR and ETBR expression was also significantly higher in GCA lesions (fig 1B,C). No differences were found in temporal artery concentrations of the endothelin system components between patients with or without ischaemic complications.

Figure 1

Endothelin (ET) system in temporal artery biopsies from giant-cell arteritis (GCA) patients. (A) ET-1 concentration in GCA arteries from 24 active patients (eight with and 16 without ischaemic complications) and controls (N  =  19). (B) Representative immunoblot disclosing endothelin-converting enzyme (ECE-1) and endothelin receptors A and B (ETAR and ETBR) expression. Actin immunodetection was used as a control for loading. (C) ECE-1, ETAR and ETBR protein/actin ratio determined by densitometric analysis of Western blots of temporal artery biopsy protein extracts from 24 active patients and 19 controls.

Endothelin system mRNA expression is downregulated in GCA lesions

Unexpectedly, a substantial decrease in ET-1, ECE-1, ETAR and ETBR mRNA was observed in GCA lesions compared with temporal arteries from controls (fig 2A–D). Consequently, ET-1 protein in lesions negatively correlated with its own mRNA (r  =  –0.612, p = 0.015). To exclude the possibility that cross-reactive ET-2 may account for the increase in ET-1 peptide found in lesions, ET-2 mRNA was subsequently measured by real-time RT–PCR. ET-2 mRNA concentrations were virtually negligible, 1000 times less abundant than ET-1 mRNA with no differences between patients and controls (0.001866 vs 0.001291 relative units, p = 0.926). A positive correlation was found between the expression of ET-1 and ECE-1 mRNA (r  =  0.673, p<0.001), between ET-1 and ETBR mRNA (r  =  0.675, p<0.001) and, to a lesser extent, between ET-1 and ETAR mRNA (r  =  0.322, p = 0.052), suggesting coordinated regulation.

Figure 2

Endothelin (ET) system mRNA expression in temporal artery biopsies from giant-cell arteritis (GCA) patients. (A) ET-1; (B) endothelin-converting enzyme (ECE-1); (C) endothelin receptor A (ETAR) and (D) endothelin receptor B (ETBR) mRNA expression in temporal arteries from 35 active GCA patients and 19 controls.

PDGF and IL-1β downregulate in vitro ET-1 mRNA expression by VSMC from GCA arteries (HTASMC)

Dissociation between ET-1 mRNA and protein may suggest a negative regulation by ET-1 itself or by other mediators present in GCA lesions. As VSMC are quantitatively the major source of endothelin in medium-sized arteries, we chose this model to investigate the regulation of ET-1 production by ET-1 itself or by other factors produced in the GCA inflammatory microenvironment, including TGFβ, PDGF, IL-1β, IL-6 and TNFα.2 3 4 8 9 ET-1 regulation in response to inflammatory mediators was also investigated in endothelial cells, using HUVEC as a model.

Overall, HUVEC produced significantly more ET-1 mRNA than HTASMC (fig 3). ET-1 did not downregulate its own expression either in HUVEC or in HTASMC (fig 3A). Among other factors tested, TGFβ and TNFα significantly increased ET-1 expression by HTASMC (fig 3B,C) and TNFα and IL-1β increased ET-1 expression by HUVEC. Interestingly, PDGF and IL-1β remarkably reduced ET-1 expression by HTASMC only, in a dose-dependent manner (fig 3C). At the range of concentrations tested, IL-6 did not elicit significant changes in ET-1 expression by HUVEC nor by HTASMC (fig 3B,C). These results indicate that some inflammatory mediators present in GCA lesions are able to downregulate ET-1 mRNA expression markedly and that ET-1 expression is differently regulated in endothelial cells compared with HTASMC.

Figure 3

Regulation of the endothelin (ET) system in cultured human umbilical vein endothelial cells (HUVEC) and human temporal artery-derived vascular smooth muscle cells (HTASMC). Subconfluent cells were incubated with increasing concentrations of depicted factors for 24 h. ET-1 mRNA expression was measured by real-time PCR as described in the Methods section. Experiments were repeated three times with consistent results and a representative experiment is shown. (A) Effect of ET-1 on its own expression. (B) Effect of cytokines and growth factors on ET-1 expression by HUVEC and (C) by HTASMC. PDGF, platelet-derived growth factor; TGFβ, transforming growth factor beta; TNFα, tumour necrosis factor alpha.

Glucocorticoid therapy partly modulates endothelin system expression

To analyse the effect of glucocorticoid therapy on the endothelin system, we cross-sectionally compared the protein concentration in arteries from active versus treated GCA patients. We did not observe significant differences in the ET-1 concentration between temporal arteries from active patients and temporal arteries from patients treated for a median of 8 days; in both groups ET-1 remained elevated compared with control arteries (fig 4A). However, ECE-1 and ETAR concentrations were lower in treated compared with active patients, reaching values found in normal temporal arteries. The decrease in ETBR levels in treated patients was not significant (fig 4B).

Figure 4

Effect of glucocorticoid therapy on the endothelin (ET) system. (A) Temporal artery biopsy ET-1 concentrations in active versus treated patients as described in the Methods section. (B) Densitometric analysis of endothelin-converting enzyme (ECE-1), endothelin receptor A (ETAR) and endothelin receptor B (ETBR) protein expression in temporal arteries from active versus treated patients assessed by Western blot. (C) ET-1 mRNA (– – –) and protein (––––) concentration in paired temporal artery biopsies obtained from three giant-cell arteritis (GCA) patients before treatment (1) and 46–50 weeks after glucocorticoid treatment (2). (D) ECE-1, ETAR and ETBR protein expression in the paired temporal artery biopsies from the same patients as in C. Number 1 refers to the first biopsy and number 2 refers to the second biopsy.

These findings suggest that glucocorticoid treatment for a median of 8 days decreases some components of the endothelin system, but ET-1 levels persist elevated. Three of our patients agreed to a second biopsy, which was performed within 46–50 weeks after the initiation of glucocorticoid treatment. As shown in fig 4(C–D), ET-1 and ECE-1 levels decreased remarkably in paired biopsies obtained before and after prolonged treatment. Changes in endothelin receptors were less conclusive.

We found again dissociation between mRNA and protein with regard to the effects of glucocorticoid treatment on ET-1 expression (fig 5). Temporal arteries from treated patients contained significantly higher concentrations of ET-1, ECE-1 and ETBR mRNA than temporal arteries from active patients, although they did not reach values found in normal temporal arteries. ETAR expression was similar in active versus treated patients but, in both groups, remained inferior to controls (fig 5A–D).

Figure 5

Changes on endothelin (ET) system mRNA expression in temporal artery biopsies according to glucorticosteroid treatment. (A) ET-1; (B) endothelin-converting enzyme (ECE-1); (C) endothelin receptor A (ETAR) and (D) endothelin receptor B (ETBR). GCA, giant-cell arteritis.

Circulating plasma ET-1 concentrations are elevated in GCA patients with ischaemic complications

Circulating ET-1 levels were similar in GCA patients as in healthy donors (1.112 (SEM 0.04) pg/ml vs 1.119 (SEM 0.06) pg/ml, p = 0.642). However, among GCA patients, circulating ET-1 concentrations were significantly higher in patients with ischaemic complications (1.205 (SEM 0.63) pg/ml) compared with patients without ischaemic events (1.048 (SEM 0.48) pg/ml, p = 0.032; fig 6A). ET-1 was also higher in patients with a weak systemic inflammatory response than in those with a strong systemic inflammatory reaction (1.120 (SEM 0.06) vs 0.990 (SEM 0.05) pg/ml, p = 0.002) who, as previously published, are at lower risk of ischaemic events10 12 13 (fig 6B).

Figure 6

Circulating endothelin (ET-1) concentration in patients with giant-cell arteritis. (A) ET-1 concentration in plasma from patients with ischaemic complications (N  =  25) compared with patients without ischaemic events (N  =  36). (B) ET-1 concentration in plasma from patients with a weak systemic inflammatory response (N  =  39) compared with those with a strong systemic inflammatory reaction (N  =  22).


Increased endothelin expression and function contributes to the pathogenesis of a number of diseases including primary pulmonary hypertension, portal hypertension and systemic sclerosis.29 30 Recognition of the role of the endothelin system in these disorders has led to efficient new therapeutic approaches.23

In this study we found that ET-1, ECE-1 and endothelin receptors are increased in vascular lesions from patients with GCA. Although our data do not definitely demonstrate a direct participation of the endothelin system in the generation of vascular occlusion in GCA, these findings certainly configure a scenario prone to endothelin-induced vasospastic and vasocclusive responses.

Plasma concentrations of ET-1 were similar between patients and controls. Elevated plasma ET-1 was previously reported in four patients with GCA compared with reference values obtained from the general population.31 The discrepancy may be due to the fact that we studied a much larger cohort of patients and used control individuals within a similar age range. As with other mediators involved in vascular remodelling,5 ET-1 may increase with age. However, the plasma ET-1 concentration was significantly higher in patients with ischaemic events, suggesting that elevated ET-1 may render patients more prone to develop vasospasm in small arteries supplying the optic nerve. Conversely, ischaemia itself may contribute to elevated ET-1 because ET-1 expression and release are under control by hypoxia inducible factor.32 Plasma ET-1 concentrations have been found, indeed, to correlate with disease severity in other conditions such as congestive heart failure,33 pulmonary artery hypertension29 34 and atherosclerotic disease.35

The measurement of ET-1 in tissue extracts disclosed increased ET-1 concentrations in temporal arteries from GCA patients. However, contrary to what was observed for circulating endothelin, no differences were found between patients with or without ischaemic complications. Several possibilities may account for this apparent discrepancy. Tissue endothelin measurement was performed in a smaller cohort and in a vascular tissue distant from the vascular beds where ischaemic complications usually occur. In addition, as with other potent mediators, ET-1 bioavailability is not only transcriptionally regulated, but it is regulated at multiple levels including proteolytic cleavage by ECE-1, receptor expression and control of ET-1 secretion. Secretion at a given time point may be more relevant than the overall ET-1 content in cells and tissues.36 37 The observation that endothelin receptors are remarkably upregulated in target vascular lesions increases the potential biological relevance of increased circulating ET-1 found in patients with ischaemic complications.37

Unexpectedly, and further supporting the complexity of endothelin regulation, all the components of the endothelin system were downregulated at the mRNA level. Destruction or dysfunction of the main cellular component of the artery wall, VSMC, which constitutively express ET-1, might account for reduced active transcription of the endothelin system at a given time point, while ET-1 peptide produced by infiltrating macrophages may still remain.25 In addition, counterregulatory mechanisms may also operate and may influence mRNA expression or stability. Contrary to other systems in which accumulation of the final product exerts a negative feedback loop on its own de novo expression, we could not demonstrate ET-1 downregulation by ET-1 itself, either in endothelial cells or in HTASMC. Among several factors expressed in temporal artery lesions, IL-1β and, in particular, PDGF substantially decreased ET-1 mRNA in HTASMC but not in endothelial cells. Interestingly, PDGF is a potent stimulator of the proliferating, migrating and secretory myointimal phenoptype of HTASMC. Myointimal cells may no longer be responsive to the vasoconstrictor effects of ET-1, which, indeed, may be inefficient in advanced lesions with intimal hyperplasia. Therefore, it is not surprising that PDGF downregulates ET-1 production by HTASMC.

Approximately 10–17% of GCA patients presenting with visual symptoms continue to have deteriorating vision during the first 1–2 weeks after the initiation of glucocorticoid treatment, indicating that glucocorticoids do not immediately/completely prevent visual loss.38 39 40 Treatment for a median of 8 days did not efficiently result in decreased ET-1 concentration in tissue, although some of the components of the system such as ECE-1 and ETAR were reduced. Although after long-term glucocorticoid treatment ET-1 concentration eventually decreased in paired biopsies, our findings indicate that the endothelin system may persist increased in GCA lesions for at least one week after the start of therapy.

Based on these results it is attractive to hypothesise that the endothelin system may play a role in the development of transient or irreversible visual loss in patients with GCA. Incomplete regulation of the endothelin system with glucocorticoid treatment may at least partly explain why some patients continue to lose sight during the first days after glucocorticoid therapy.38 39 40 Interference with the endothelin system may, then, be a therapeutic option for patients with GCA, particularly those who continue to present with visual symptoms after the beginning of glucocorticoids and for whom no additional options exist. The number of patients who continue to lose vision after glucocorticoid therapy is fortunately small, but this situation, for which there are no established alternatives, is one of the unsolved issues in the management of GCA.11 38 39 40 Endothelin receptor antagonists have clinical efficacy and are approved for the treatment of patients with pulmonary hypertension41 and for patients with digital ulcers secondary to systemic sclerosis.42 However, the potential efficacy of endothelin receptor blockade in reducing the risk of visual loss in patients with visual symptoms remains hypothetical but would be worth considering in multicentre clinical trials.



  • Funding This study was supported by the Ministerio de Educación y Ciencia and Fondo Europeo de Desarrollo Regional (FEDER) (SAF 05/06250 and SAF 08/04328) and Marató-TV3 (06/0710). GE-F and MCC are supported by the Instituto de Salud Carlos III.

  • Competing interests None.

  • Ethics approval The study was approved by the Ethics Committee of the Hospital Clínic.

  • Patient consent Obtained.

  • The results were partly presented at the 71st Annual Meeting of the American College of Rheumatology. Boston, USA, November 2007.

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

Request Permissions

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