MIGUEL A. GONZALEZ-GAY, JAVIER OLIVER, ELENA SANCHEZ, CARLOS GARCIA-PORRUA, LAURA PACO, MIGUEL A. LOPEZ-NEVOT, WILLIAM E.R. OLLIER, and JAVIER MARTIN

ABSTRACT.

Objective. To assess the contribution of 2 polymorphisms within the inducible nitric oxide (NOS2A) promoter region to susceptibility to giant cell arteritis (GCA).

Methods. One hundred three patients with biopsy-proven GCA and 198 ethnically matched controls from the Lugo region (Northwest Spain) were studied. Patients and controls were genotyped using polymerase chain reaction techniques for a multiallelic (CCTTT)_n and for the TAAA repeat polymorphism in the promoter region of the NOS2A gene.

Results. No significant differences in allele or genotype frequencies for the (CCTTT)_n repeat polymorphism in the NOS2A gene between patients with GCA and controls were observed. However, significant differences for the TAAA repeat polymorphism between patients and controls were found. The overall distribution of NOS2A TAAA genotypes in patients with biopsy-proven GCA was significantly different than controls (p = 0.026). Patients with GCA had an increased frequency of the NOS2A TAAA+ allele (16.5%) compared with controls (9.1%) (p = 0.007; OR 1.98; 95% CI 1.20–3.27). This was due to an increased frequency of both heterozygotes (27.2%) and homozygotes (2.9%) for NOS2A TAAA+ observed in patients compared to controls (15.2% and 1.5%, respectively) (p = 0.007; OR 2.15; 95% CI 1.23–3.78).

Conclusion. Our results suggest a potential implication for NOS2A TAAA gene polymorphism in GCA susceptibility. (J Rheumatol 2005;32:2178-82)

Key Indexing Terms:

GIANT CELL (TEMPORAL) ARTERITIS
TEMPORAL ARTERY BIOPSY
DISEASE SUSCEPTIBILITY
NITRIC OXIDE
NOS2A POLYMORPHISMS

From the Instituto de Parasitologia y Biomedicina Lopez-Neyra, CSIC; Division of Immunology, Hospital Virgen de las Nieves, Granada; Division of Rheumatology, Hospital Xeral-Calde, Lugo, Spain; and the Centre for Integrated Genomic Medical Research, Manchester, UK.

Supported by grant SAF03-3460 from Plan Nacional de I+D+I and in part by Junta de Andalucía, grupo CTS-180.

M.A. Gonzalez-Gay, MD, PhD; C. Garcia-Porrua, MD, PhD, Rheumatology Division, Hospital Xeral-Calde; J. Martin, MD, PhD; J. Oliver, PhD; E. Sanchez, PhD, Instituto de Parasitologia y Biomedicina Lopez-Neyra; M.A. Lopez-Nevot, MD, PhD; L. Paco, PhD, Division of Immunology, Hospital Virgen de las Nieves; W.E.R. Ollier, PhD, Centre for Integrated Genomic Medical Research.

Dr. Gonzalez-Gay and Dr. Martin share senior authorship of this report.

Address reprint requests to Dr. J. Martin, Instituto de Parasitología y Biomedicina, CSIC, Parque Tecnológico de Ciencias de la Salud, Avenida del Conocimiento s/n 18100-Armilla, Granada, Spain. E-mail: martin@ipb.csic.es

Accepted for publication June 20, 2005.

Giant cell arteritis (GCA) involves large and medium-size blood vessels with a predisposition to the involvement of cranial arteries¹. It is the most common systemic vasculitis in people over the age of 50 years in Western countries², in particular in those with Northern European ancestry^2-4. The main manifestations of GCA are due to vascular involvement.

Reports of well documented families of first-degree relatives with GCA support a genetic component in the pathogenesis of this vasculitis⁵. Associations between GCA and genes that lie within the HLA class II region have been described⁶. However, GCA appears to be a polygenic disease and different genes may influence the phenotype and the outcome of this condition⁷.

Nitric oxide (NO) is the product of conversion of L-arginine to L-citrulline by a class of enzymes called NO synthases (NOS). It plays a role in both prevention and development of atherosclerosis. NO is produced constitutively by endothelial (eNOS or NOS3) or neuronal (nNOS or NOS1) synthases or in higher concentrations by iNOS (or NOS2) after stimulation of a variety of proinflammatory cytokines⁸.

Several functionally relevant polymorphisms in the NOS2A and NOS3 genes have been identified. Association of these polymorphisms with different vascular⁹, autoimmune¹⁰, and infectious diseases¹¹ has been reported. Haplotype association between 3 polymorphisms of the NOS3 gene was found in patients with GCA from Northwest Spain¹². Also, an association with a Glu/Asp(298) polymorphism in exon 7 of the NOS3 gene has recently been reported in patients from Italy with GCA¹³.

NOS2A is located on chromosome 17q11.2-12, and 2 NOS2A microsatellites, the CCTTT repeat sequence in position –2662 to –2608, and the TAAA repeat polymorphism in position –754 to –739, have been described (Genebank accession number X97821).

A highly polymorphic pentanucleotide (CCTTT)_n repeat located at the NOS2A promoter region has been shown to be functionally important in the regulation of NOS2A transcription¹⁰. A trend toward association of (CCTTT)_n repeat variations with rheumatoid arthritis (RA) has been reported¹⁴. Of note, significant differences in this NOS2A promoter polymorphism genotype frequency between patients with RA from Northwest Spain and controls have been observed¹⁵. Also, a functional polymorphism in the proximal promoter involving the insertion or deletion of one unit of a TAAA repeat^16,17 has proved to be associated with increased risk of renal abnormalities and other complications of type 2 diabetes¹⁸.

We assessed the contribution of these 2 polymorphisms within the NOS2A promoter region to susceptibility to GCA in a series of patients with biopsy-proven GCA.

MATERIALS AND METHODS

Study population. We examined iNOS polymorphisms in a series of 103 consecutive patients diagnosed with biopsy-proven GCA in the Division of Rheumatology of the Hospital Xeral-Calde (Lugo, Spain) who gave informed consent for immunogenetic studies. All patients fulfilled the 1990 American College of Rheumatology criteria for the classification of GCA¹⁹. The mean age ± standard deviation (SD) of the patients was 74.5 ± 6.0 years, and the ratio of women/men was 1.3/1.0. Age (± 3 yrs) and sex matched controls were also studied. Since all patients assessed in this study were from the Lugo region in Northwest Spain, all controls (n = 198) were recruited from the same area. They were required to be healthy volunteers living in and around the city of Lugo, and we could trace their ancestry in the Lugo region for at least 3 generations. The main characteristics of the Lugo population have been reported²⁰. Only patients with GCA who had a positive temporal artery biopsy showing disruption of the internal elastic laminae with infiltration of mononuclear cells into the arterial wall with or without giant cells were included in this study.

Patients with GCA were considered to have an associated polymyalgia rheumatica (PMR) if they had severe bilateral aches and pains involving the neck, the shoulder, and/or the pelvic girdles, associated with morning stiffness^21,22. As reported²³, patients were considered to have severe ischemic manifestations if they suffered visual manifestations (transient visual loss including amaurosis fugax, permanent visual loss, or diplopia), cerebrovascular accidents (stroke and/or transient ischemic attacks), jaw claudication, or limb claudication of recent onset. With regard to smoking, as reported, we established 2 categories: (1) heavy smokers who still smoked at the time of disease diagnosis or who had smoked within 10 years before onset of GCA symptoms; and (2) patients who never smoked or stopped smoking at least 10 years before disease onset^24,25. Corrections were made for age and sex within the GCA groups with or without severe ischemic manifestations or PMR and for sex and history of heavy smoking.

Patients and controls gave written informed consent prior to participation in the genetic studies, which was approved by the local institutional committee.

TAAAn and (CCTTT)_n genotyping. DNA was isolated from anticoagulated peripheral blood mononuclear cells using standard methods. We determined the TAAA_n and (CCTTT)_n genotypes by a polymerase chain reaction (PCR) method as described^26,27. Forward and reverse primers were 5' TGC CAC TCC GCT CCAG 3' and 5' GGC CTC TGA GAT GTT GGT CTT 3' for TAAA_n, and 5' ACC CCT GGA AGC CTA CAA CTG CAT 3' and 5' GCC ACT GCA CCC TAG CCT GTC TCA 3' for (CCTTT)_n. The forward primers were 5' labeled with 6-FAM fluorescent dye. PCR aliquots of 0.5 µl were added to 3 µl of formamide and 0.5 µl of internal size standard. Samples were analyzed in denaturing gels (6% acrylamide/7 M urea) and sized using Genescan 672 software (Applied Biosystems, Foster City, CA, USA). With respect to TAAA_n repeats, sequencing revealed that the longer PCR fragment, 224 bp in length, contained 5 TAAA repeats (denoted NOS2A+ allele) whereas the shorter one, 220 bp, contained 4 repeats (denoted NOS2A– allele). Concerning (CCTTT)_n repeats, the size of the PCR products ranged from 171 bp to 216 bp, depending on the number of pentanucleotide repeat units.

Statistical analysis. Strength of association between patient groups and controls and alleles or genotypes of NOS2A polymorphisms was estimated using odds ratios (OR) and 95% confidence intervals (CI). Levels of significance were determined using contingency tables by either chi-square or Fisher's exact analysis. Statistical significance was defined as p ≤ 0.05. Calculations were performed with the Stata statistical package, V6. We used the UNPHASED software created for case-control analysis of haplotypes. The power of the study to detect an effect of a polymorphism in disease susceptibility was estimated using Quanto 0.5 software (Department of Preventive Medicine, University of Southern California, Los Angeles, CA, USA). The sample size used had 66% power to detect the effect of the TAAA polymorphism, conferring an OR of 1.9 at the 5% significance level.

RESULTS

Clinical characteristics of the GCA patients. The group of patients with biopsy-proven GCA comprised 59 women and 44 men (median age at disease diagnosis was 75 yrs, range 60–92 yrs). From onset of GCA symptoms until 1 month after onset of steroid therapy, 87 (84%) had headache, 77 (75%) abnormal temporal artery on physical examination, 41 (40%) PMR, and 53 (51%) developed severe ischemic manifestations. Among them, 41 (40%) had jaw claudication, 23 (22%) visual ischemic manifestations, and 1 (1%) stroke. Other clinical features of the 103 patients with biopsy-proven GCA are summarized in Table 1.

Table 1. Main clinical features of 103 patients with biopsy-proven GCA from Lugo (Northwest Spain). Numbers in brackets represent percentages.

NOS2A promoter CCTTT repeat microsatellite polymorphism in GCA. (CCTTT)_n allele and genotype frequencies were examined in patients with biopsy-proven GCA and controls. No evidence of departure from Hardy-Weinberg equilibrium was observed (p = 0.80) in controls. Table 2 shows the allelic frequencies in patients and controls. Overall (CCTTT)_n allelic distribution did not show statistically significant differences between patients with GCA and controls (p = 0.31) or any significant differences in allele or genotype frequencies for the (CCTTT)_n repeat polymorphism in NOS2A (Table 2). No significant differences in allele or genotype distribution were found when patients with GCA were stratified according to presence of PMR, severe ischemic manifestations, smoking history, or sex (data not shown).

Table 2. Allele frequencies of NOS2A CCTTTn gene polymorphism in patients with biopsy-proven giant cell arteritis (GCA) and controls. * The overall allelic distribution for the NOS2A CCTTTn gene polymorphism did not show statistically significant differences between patients and controls.

NOS2A promoter TAAA repeat polymorphism in GCA. When overall genotype distribution of patients with GCA was compared to controls, a significant difference was observed (p = 0.026, by chi-square test from 3 ´ 2 contingency table; Table 3). This was due to an increase in both heterozygotes +/– and homozygotes +/+ for NOS2A genotypes observed in patients with biopsy-proven GCA compared to controls (p = 0.007, OR 2.15, 95% CI 1.23–3.78). In addition, a significantly increased frequency of the NOS2A TAAA+ allele was observed in patients with biopsy-proven GCA compared with controls (p = 0.007, OR 1.98, 95% CI 1.20–3.27; Table 3). However, no allele or genotype differences were observed when patients with GCA were stratified according to presence of PMR, severe ischemic manifestations, smoking history, or sex (Table 4). It was also the case when patients with GCA were corrected by age, sex, and heavy smoking history.

Table 3. Allele frequencies and genotype distribution of NOS2A TAAA polymorphism in patients with GCA and controls.

Table 4. Clinical features according to NOS2A TAAA polymorphism in patients with GCA. No statistically significant differences among the different groups were observed.

NOS2A promoter CCTTT– TAAA haplotypes. The relationship between the 2 microsatellites repeats in the NOS2A promoter was investigated in our population. For the 2 TAAA alleles, there was a strikingly wide distribution of CCTTT microsatellite alleles, reflecting lack of linkage disequilibrium (LD) between the 2 markers analyzed (global Δ 0.27).

DISCUSSION

Our study is the first to assess the influence of NOS2A polymorphism in the development of GCA in a large series of patients whose diagnosis was confirmed by temporal artery biopsy. No association with the polymorphic pentanucleotide (CCTTT)_n repeat in the promoter region of NOS2A was found. However, significant differences were observed when a biallelic TAAA – repeat located 0.75 kb upstream of the gene were assessed. Unlike healthy controls, those with biopsy-proven GCA exhibited an increased frequency of NOS2A TAAA+ allele. The relative absence of LD between CCTTT and TAAA markers indicates that the association of TAAA repeats with GCA is independent of the CCTTT repeats. Of interest, a recent study has shown a lack of LD between the CCTTT microsatellites and dinucleotide repeats in the NOS2A promoter region²⁸. Also, the NOS2A TAAA association observed in our study was independent of the previous NOS3 gene haplotype association found in patients with GCA¹².

Our results reinforce the potential implication of NOS2A TAAA polymorphism in GCA susceptibility. NOS2A TAAA+ allele has been also associated with an elevated plasma glucose level and unstable angina²⁹. In addition, a recent study has proposed a role of the NOS2A TAAA repeat in more pronounced coronary artery disease³⁰.

Immunochemistry and in situ hybridization techniques have shown that macrophages and vascular smooth muscle cells express iNOS in both early and advanced atherosclerotic lesions^31-33. Production of NO by increased expression of iNOS may have a double role in the development of atherosclerotic lesions. As a pro-oxidant, NO may also promote the process of atherogenesis by increasing platelet adherence, cell death, and necrosis³⁴. Interestingly, patients with GCA present with clinical manifestations that are the result of vascular involvement. Since it is often difficult to differentiate healed arteritis from arteriosclerosis on morphological grounds, some investigators have suggested that atherosclerosis and GCA may have a common pathway^35,36. Machado, et al, in a retrospective case-control study of 88 patients with biopsy-proven GCA, reported an association between smoking and disease development³⁷. Duhaut, et al, in a prospective multicenter case-control study of 207 patients with biopsy-proven GCA, described a strong association between smoking and previous atheromatous disease in women and GCA³⁸. As a result of atherosclerosis and subclinical inflammation, as described in Rochester, USA, where GCA incidence is high³⁹, we also observed the development of aortic aneurysmal disease in followup of patients with GCA from Northwest Spain²⁴. Thus, abnormalities in NOS2A may play a potential role in the development of this vasculitis.

Since in transient transfection assays the presence of NOS2A TAAA+ allele has been associated with increased NOS2A promoter activity¹⁸, our results suggest that the higher production of NO by iNOS due to an increased frequency of NOS2A TAAA+ allele in patients with biopsy-proven GCA may be detrimental rather than beneficial to individuals for the development of this vasculitis. Additional studies are required to ascertain the mechanism behind the association between NOS2A TAAA polymorphism and GCA. In this regard, studies still are needed to evaluate whether the NOS2A 4 bp insertion/deletion affects iNOS expression in other cell lines or in temporal artery biopsy specimens. In addition, replication studies in other populations would confirm the validity of our observations.

ACKNOWLEDGMENT

We thank Sonia Morales and Maria Paz Ruiz for excellent technical assistance.

REFERENCES

Search PubMed for:

2. Gonzalez-Gay MA, Garcia-Porrua C. Epidemiology of the vasculitides. Rheum Dis Clin North Am 2001;27:729-49. [MEDLINE]

3. Gran JT, Myklebust G. The incidence of polymyalgia rheumatica and temporal arteritis in the county of Aust Agder, south Norway: a prospective study 1987-94. J Rheumatol 1997;24:1739-43. [MEDLINE]

4. Huston KA, Hunder GG, Lie JT, Kennedy RH, Elveback LR. Temporal arteritis: a 25-year epidemiologic, clinical, and pathologic study. Ann Intern Med 1978;88:162-7.

5. Wernick R, Davey M, Bonafede P. Familial giant cell arteritis: report of an HLA-typed sibling pair and a review of the literature. Clin Exp Rheumatol 1994;12:63-6. [MEDLINE]

6. Weyand CM, Hicok KC, Hunder GG, Goronzy JJ. The HLA-DRB1 locus as a genetic component in giant cell arteritis. Mapping of a disease-linked sequence motif to the antigen binding site of the HLA-DR molecule. J Clin Invest 1992;90:2355-61. [MEDLINE]

7. Gonzalez-Gay MA, Amoli MM, Garcia-Porrua C, Ollier WE. Genetic markers of disease susceptibility and severity in giant cell arteritis and polymyalgia rheumatica. Semin Arthritis Rheum 2003;33:38-48. [MEDLINE]

8. Moncada S. Nitric oxide in the vasculature: physiology and pathophysiology. Ann NY Acad Sci 1997;811:60-7. [MEDLINE]

9. Wang XL, Wang J. Endothelial nitric oxide synthase gene sequence variations and vascular disease. Mol Genet Metab 2000;70:241-51. [MEDLINE]

10. Warpeha KM, Xu W, Liu L, et al. Genotyping and functional analysis of a polymorphic (CCTTT)_n repeat of NOS2A in diabetic retinopathy. FASEB J 1999;13:1825-32. [MEDLINE]

11. Burgner D, Xu W, Rockett K, et al. Inducible nitric oxide synthase polymorphism and fatal cerebral malaria. Lancet 1998;352:1193-4. [MEDLINE]

12. Amoli MM, Garcia-Porrua C, Llorca J, Ollier WE, Gonzalez-Gay MA. Endothelial nitric oxide synthase haplotype associations in biopsy-proven giant cell arteritis. J Rheumatol 2003;30:2019-22. [MEDLINE]

13. Salvarani C, Casali B, Nicoli D, et al. Endothelial nitric oxide synthase gene polymorphisms in giant cell arteritis. Arthritis Rheum 2003;48:3219-23. [MEDLINE]

14. Pascual M, Lopez-Nevot MA, Caliz R, et al. Genetic determinants of rheumatoid arthritis: the inducible nitric oxide synthase (NOS2) gene promoter polymorphism. Genes Immun 2002;3:299-301. [MEDLINE]

15. Gonzalez-Gay MA, Llorca J, Sanchez E, et al. Inducible but not endothelial nitric oxide synthase polymorphism is associated with susceptibility to rheumatoid arthritis in northwest of Spain. Rheumatology Oxford 2004;43:1182-5. [MEDLINE]

16. Bellamy R, Hill AV. A bi-allelic tetranucleotide repeat in the promoter of the human inducible nitric oxide synthase gene. Clin Genet 1997;52:192-3. [MEDLINE]

17. Nunokawa Y, Ishida N, Tanaka S. Promoter analysis of human inducible nitric oxide synthase gene associated with cardiovascular homeostasis. Biochem Biophys Res Commun 1994;200:802-7. [MEDLINE]

18. Morris BJ, Markus A, Glenn CL, Adams DJ, Colagiuri S, Wang L. Association of a functional inducible nitric oxide synthase promoter variant with complications in type 2 diabetes. J Mol Med 2002;80:96-104. [MEDLINE]

19. Hunder GG, Bloch DA, Michel BA, et al. The American College of Rheumatology 1990 criteria for the classification of giant cell arteritis. Arthritis Rheum 1990;33:1122-8. [MEDLINE]

20. Gonzalez-Gay MA, Garcia-Porrua C, Llorca J, et al. Visual manifestations of giant cell arteritis. Trends and clinical spectrum in 161 patients. Medicine (Baltimore) 2000;79:283-92. [MEDLINE]

21. Chuang TY, Hunder GG, Ilstrup DM, Kurland LT. Polymyalgia rheumatica: a 10-year epidemiologic and clinical study. Ann Intern Med 1982;97:672-80. [MEDLINE]

22. Gonzalez-Gay MA. Giant cell arteritis and polymyalgia rheumatica: two different but often overlapping conditions. Semin Arthritis Rheum 2004;33:289-93. [MEDLINE]

23. Gonzalez-Gay MA, Garcia-Porrua C, Amor-Dorado JC, Llorca J. Fever in biopsy-proven giant cell arteritis: Clinical implications in a defined population. Arthritis Rheum 2004;551:652-5. [MEDLINE]

24. Gonzalez-Gay MA, Garcia-Porrua C, Piñeiro A, Pego-Reigosa A, Llorca J, Hunder GG. Aortic aneurysm and dissection in patients with biopsy-proven giant cell arteritis from Northwestern Spain: A population-based study. Medicine (Baltimore) 2004;83:335-41. [MEDLINE]

25. Gonzalez-Gay MA, Piñeiro A, Gomez-Gigirey A, et al. Influence of traditional risk factors of atherosclerosis in the development of severe ischemic complications in giant cell arteritis. Medicine (Baltimore) 2004;83:342-7. [MEDLINE]

26. Johannesen J, Pociot F, Kristiansen OP, et al. No evidence for linkage in the promoter region of the inducible nitric oxide synthase gene (NOS2) in a Danish type 1 diabetes population. Genes Immun 2000;1:362-6. [MEDLINE]

27. Xu W, Liu L, Emson PC, Harrington CR, Charles IG. Evolution of a homopurine-homopyrimidine pentanucleotide repeat sequence upstream of the human inducible nitric oxide synthase gene. Gene 1997;204:165-70. [MEDLINE]

28. Burger D, Rockett K, Ackerman H, et al. Haplotypic relationship between SNP and microsatellite markers at the NOS2A locus in two populations. Genes Immun 2003;4:506-14. [MEDLINE]

29. Morris BJ, Glenn CL, Wilcken DEL, Wang XL. Influence of an inducible nitric oxide synthase promoter variables in patients with coronary artery disease. Clin Sci 2001;100:551–6. [MEDLINE]

30. Kunnas TA, Mikkelsson J, Ilveskoski E, et al. A functional variant of the iNOS gene flanking region is associated with LAD coronary artery disease: an autopsy study. Eur J Clin Invest 2003;33:1032-7. [MEDLINE]

31. Luoma JS, Stralin P, Marklund SL, Hiltunen TP, Sarkioja T, Ylä-Herttuala S. Expression of extracellular SOD and iNOS in macrophages and smooth muscle cells in human and rabbit atherosclerotic lesions: colocalization with epitopes characteristic of oxidized LDL and peroxynitrite-modified proteins. Arterioscler Thromb Vasc Biol 1998;18:157–67. [MEDLINE]

32. Buttery LD, Springall DR, Chester AH, et al. Inducible nitric oxide synthase is present within human atherosclerotic lesions and promotes the formation and activity of peroxynitrite. Lab Invest 1996;75:77–85. [MEDLINE]

33. Behr-Roussel D, Rupin A, Sansilvestri-Morell P, Fabiani JN, Verbeuren TJ. Histochemical evidence for inducible nitric oxide synthase in advanced but non-ruptured atherosclerotic carotid arteries. Histochem J 2000;32:41–51. [MEDLINE]

34. Depre C, Havaux X, Renkin J, Vanoverschelde JL, Wijns W. Expression of inducible nitric oxide synthase in synthase in human coronary atherosclerotic plaque. Cardiovasc Res 1999;41:465–72. [MEDLINE]

35. O'Brien JP. A new risk factor and other actinic radiation in giant cell arteritis and atherosclerosis. Int J Dermatol 1987;26:345-8. [MEDLINE]

36. O'Brien JP. Vascular accident after actinic (solar) exposure: an aspect of the temporal arteritis/polymyalgia rheumatica syndrome. Int J Dermatol 1987;26:366-70. [MEDLINE]

37. Machado EBV, Gabriel SE, Beard CM, Michet CJ, O'Fallon WM, Ballard DJ. A population-based case-control study of temporal arteritis: evidence for an association between temporal arteritis and degenerative vascular disease? Int J Epidemiol 1989;18:836-41. [MEDLINE]

38. Duhaut P, Pinede L, Demolombe-Rague S, et al. Giant cell arteritis and cardiovascular risk factors. A multicenter, prospective case-control study. Arthritis Rheum 1998;41:1960-5. [MEDLINE]

39. Nuenninghoff DM, Hunder GG, Christianson TJ, McClelland RL, Matteson EL. Incidence and predictors of large-artery complication (aortic aneurysm, aortic dissection, and/or large-artery stenosis) in patients with giant cell arteritis: a population-based study over 50 years. Arthritis Rheum 2003;48:3522-31. [MEDLINE]