Small-vessel vasculitis surrounding an uninflamed temporal arteryA new diagnostic criterion for polymyalgia rheumatica.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 58, No. 8, August 2008, pp 2565–2573 DOI 10.1002/art.23700 © 2008, American College of Rheumatology Small-Vessel Vasculitis Surrounding an Uninflamed Temporal Artery A New Diagnostic Criterion for Polymyalgia Rheumatica? Denis Chatelain,1 Pierre Duhaut,1 Robert Loire,2 Sylvie Bosshard,2 Hélène Pellet,2 Jean-Charles Piette,3 Henri Sevestre,1 and Jean-Pierre Ducroix1 Objective. To assess the prevalence and clinical significance of small-vessel vasculitis (SVV) surrounding an uninflamed temporal artery (TA) in patients diagnosed as having giant cell (temporal) arteritis (GCA) and/or polymyalgia rheumatica (PMR). Methods. Patients with GCA and/or PMR (n ⴝ 490) were included in this multicenter prospective study. Slides of TA biopsy specimens were reviewed by 2 pathologists who were blinded with regard to clinical information. SVV was defined as aggregates of mononuclear inflammatory cells surrounding a capillary, distant from an uninflamed temporal artery. Clinical and biologic data of patients in the SVV group (n ⴝ 35) were compared with data of patients with biopsy-proven GCA (n ⴝ 280) and with negative TA biopsy findings (n ⴝ 175). Results. SVV was diagnosed in 18 women and 17 men (mean ⴞ SD age 74.5 ⴞ 9.4 years). The group of patients with SVV had a higher proportion of men than in the entire GCA series, had systemic symptoms, headache, jaw claudication, and an abnormal temporal artery less frequently at clinical examination, but had symptoms of PMR more often than patients in the biopsy-proven GCA group (P ⴝ 2.6 ⴛ 10ⴚ7, odds ratio 9.17 [95% confidence interval 3.44–24.4]). Levels of inflammation markers were significantly lower in the SVV group. Patients in the SVV group had fever less frequently than patients in the group with negative TA biopsy findings, but otherwise shared the same clinical (including PMR symptoms) and biologic features. Eighteen of the 94 patients with pure PMR (19%) had SVV. Conclusion. SVV is often neglected by pathologists, and appears to be strongly associated with PMR symptoms in patients with a clinical diagnosis of GCA and/or PMR. However, SVV as a new diagnostic criterion for PMR must be assessed in prospective studies. Microscopic examination of temporal artery (TA) biopsy specimens is still the gold standard for the diagnosis of giant cell (temporal) arteritis (GCA) (1–4). Patients with GCA are usually ⬎50 years, and present with headache, jaw claudication, and an increased erythrocyte sedimentation rate (ESR) (1,2). Microscopic examination of TA biopsy specimens shows infiltration of mononuclear inflammatory cells, often with multinucleated giant cells in the arterial wall, and interruption of the internal elastic laminae (3,4). Polymyalgia rheumatica (PMR) is observed in 40–60% of patients with GCA (4,5). The incidence of temporal arteritis in patients with pure PMR is still controversial, ranging between 0% and 41% in clinical series (6). Small-vessel vasculitis (SVV) surrounding an uninflamed temporal artery is an infrequent, but perhaps underestimated, finding during microscopic examination of TA biopsy specimens (7–10). SVV can be defined as aggregates of mononuclear inflammatory cells, without polynuclear neutrophils and eosinophils and without fibrinoid necrosis of the vessel wall, surrounding a small capillary distant from an uninflamed temporal artery. 1 Denis Chatelain, MD, Pierre Duhaut, MD, PhD, Henri Sevestre, MD, PhD, Jean-Pierre Ducroix, MD: Centre Hospitalier Universitaire du Nord and RECIF, Amiens, France; 2Robert Loire, MD, Sylvie Bosshard, MD, Hélène Pellet, MD, PhD: Lyon-GrangeBlanche and RECIF, Claude Bernard University, Lyon, France; 3 Jean-Charles Piette, MD: Hôpital de la Pitié Salpêtrière and RECIF, Paris, France. Address correspondence and reprint requests to Denis Chatelain, MD, Department of Pathology, Centre Hospitalier Universitaire du Nord, Place Victor Pauchet, 80054 Amiens Cedex 01, France. E-mail: email@example.com. Submitted for publication October 11, 2007; accepted in revised form April 18, 2008. 2565 2566 CHATELAIN ET AL SVV has been reported in various inflammatory, infectious, or neoplastic diseases but its clinical significance is still unclear (7–10). The aims of our study were to assess the prevalence of SVV on TA biopsy specimens in patients with a clinical diagnosis of GCA or PMR and to evaluate its clinical significance in these patients. PATIENTS AND METHODS Patients. The recruitment of patients with GCA and PMR and the inclusion and exclusion criteria used in this study have been previously described (11,12). Members of every internal medicine, rheumatology, geriatrics, neurology, and ophthalmology department at university hospitals in France, and those at internal medicine departments affiliated with the French Society of Internal Medicine, were contacted by mail in January 1991. (See Appendix A for members of the Groupe de Recherche sur l’Artérite à Cellules Géantes [GRACG] and their locations.) We only included newly diagnosed patients, using preestablished criteria in order to best avoid misclassification and recall bias. Each new patient completed a questionnaire to provide medical history and clinical examination data, and a blood sample was obtained before corticosteroid treatment was started or, at most, 48 hours after treatment began. The questionnaire had to be completed directly by the patient, and not using data recorded in the medical chart. Diagnostic criteria. Patients had to meet the following preestablished diagnostic criteria: 1) ⬎50 years of age; 2) ESR (Westergren) ⬎40 mm/hour (except for a few cases with typical symptoms, for whom TA biopsy findings were positive on microscopic examination); 3) clinical response within 72 hours to corticosteroid therapy (disappearance of fever and pain); 4) positive TA biopsy findings; 5) clinically abnormal temporal artery (tenderness, swelling, redness, nodular artery); 6) visual disturbances (blindness, diplopia, blurred vision), including those occurring during the first week of treatment; 7) jaw claudication; 8) headache, temporal headache, facial pain, or sensation of facial swelling; 9) systemic symptoms, such as fever, loss of ⬎10% of total weight, anorexia, malaise, asthenia; and 10) PMR symptoms with persistent proximal muscle pain, tenderness, or morning stiffness lasting ⬎1 hour, involving the neck, shoulders, and/or pelvic girdle (duration ⬎2 weeks). All patients had to fulfill the first 3 criteria. In addition to meeting these 3 criteria, patients included in the biopsy-proven GCA group had to fulfill criterion 4. Patients included in the negative-biopsy group had to fulfill 2 criteria among criteria 5–9. The symptoms in criterion 10 could be present or not. Patients diagnosed as having GCA met the American College of Rheumatology classification criteria (13). Pure GCA was defined as GCA with no sign or symptom of PMR. Patients diagnosed as having PMR fulfilled the criteria described by Chuang et al (14). Pure PMR was defined as PMR with no sign or symptom of GCA. Exclusion criteria consisted of current malignant diseases, current infectious diseases, or other inflammatory or vasculitis diseases, notably a history of rheumatoid arthritis (RA), systemic lupus erythematosus, or polyarteritis nodosa. All patients were followed up over a 5-year period, and the mean followup in this series was 4 years. The patients were classified into subgroups once the inclusion criteria were reviewed by physicians at the coordinating center (PD, SB), and the TA biopsies were reviewed by the pathologists (DC, RL). Variables at the time of diagnosis. There were 3 types of data: initial clinical signs and symptoms, biologic data, and time of diagnosis and the delay between the onset of symptoms and the diagnosis. Initial clinical signs and symptoms included systemic symptoms, such as fever, anorexia, malaise, weight loss, and asthenia; recent headache, either temporal, diffuse, or facial; malaise; jaw claudication; visual problems, such as blindness, diplopia, or blurred vision; abnormal temporal artery on clinical examination; or symptoms of PMR. Biologic data included ESR (Westergren), C-reactive protein (CRP), haptoglobin, fibrinogen, hemoglobin level, mean globular volume, and platelet count. Pathologic examination of TA biopsy specimens. TA biopsy has been the standard technique for diagnosis of GCA. This was a multicenter study, and the surgeons performed the biopsies according to the usual standards. TA biopsy specimens were fixed in 10% buffered formalin and processed routinely. In most of the cases they were cut transversely in small, 3–4-mm artery segments. TA biopsy specimens were embedded in paraffin, and 4-m–thick paraffin-embedded tissue segments were cut and stained with hematoxylin–eosin– saffron. The number of levels examined per arterial segment varied from 1 (1 case) to 38 (1 case) (median 6), depending on the participating center. A copy of the initial pathologic report on the TA biopsy specimens was requested, and the slides were reviewed by 2 senior pathologists (DC, RL) without clinical information. Histologic criteria were assessed and recorded on a preestablished form. Quality of the biopsy (excellent, good, medium, poor), orientation of the arterial segments (transverse or longitudinal), number of arterial segments, and number of levels generated from the paraffin blocks were assessed. During microscopic examination, there is always some connective tissue surrounding the adventitia, and this connective tissue contains capillaries, arterioles, small nerves, and sometimes small veins. SVV surrounding a spared temporal artery was defined as aggregates of mononuclear inflammatory cells around capillaries and not arterioles (ⱖ15 inflammatory cells) surrounding an uninflamed temporal artery. There was neither fibrinoid necrosis of the capillary wall nor leukocytoclasia. Small capillaries were 0.5–1.5 mm from the arterial wall. We did not call them vasa vasorum because the lesions we described never involved arterioles and venules, only capillaries, located outside the arterial wall. Control group of patients without GCA and/or PMR. We studied a control group of patients who had TA biopsy specimens but had not been diagnosed as having GCA and/or PMR. We extracted all information on TA biopsies performed in 2004 from the archives of the Department of Pathology of Amiens University Hospital. Statistical analysis. We compared the 3 groups of patients: patients with SVV were compared with patients with biopsy-proven GCA and with patients with negative-biopsy PMR and/or GCA. Data analysis was performed with SAS SVV AS A DIAGNOSTIC CRITERION FOR PMR Table 1. 2567 Characteristics of the patients with SVV versus the patients with biopsy-proven GCA* No. male/female (male:female ratio) Age, mean ⫾ SD years Clinical diagnosis Pure GCA GCA with PMR symptoms Pure PMR Duration of symptoms, median (range) days Corticosteroid response Systemic symptoms Fever Asthenia Anorexia Weight loss Headache Malaise Jaw claudication Visual impairment Blindness PMR symptoms Abnormal TA on clinical examination ESR, median (range) mm/hour CRP, median (range) mg/liter Haptoglobin, median (range) gm/liter Fibrinogen, mean ⫾ SD gm/liter Hemoglobin, median (range) gm/liter Mean globular volume, median (range) 3 Platelet count, median (range) ⫻ 109/liter Ferritin, median (range) g/liter Patients with SVV (n ⫽ 35) Patients with biopsyproven GCA (n ⫽ 280) 17/18 (0.94) 74.5 ⫾ 9.4 73/207 (0.35) 75.4 ⫾ 8 5 (14) 12 (34) 18 (51) 36 (1–730) 30 (86) 27 (77) 6 (17) 22 (63) 11 (31) 4 (11) 12 (34) 5 (14) 8 (23) 7 (20) 1 (3) 30 (86) 6 (17) 72 (21–130) 62 (2–156) 3.8 (1.7–6.8) 5.5 ⫾ 1.4 122 (85–143) 91 (78–111) 335 (159–700) 250 (3–786) 165 (59) 108 (39) 7 (3) 52 (0–2,113) 259 (95) 255 (91) 122 (44) 227 (81) 129 (46) 83 (30) 229 (82) 22 (8) 116 (42) 91 (32) 25 (9) 111 (40) 168 (60) 88 (11–150) 83 (3–590) 4.6 (1–12.7) 6.7 ⫾ 1.8 113 (65–158) 87 (70–104) 417 (92–954) 327 (23–4,987) P 0.005 0.47 ⬍0.001 0.282 0.144 0.018 0.003 0.013 0.1 0.023 1.16 ⫻ 10⫺8 0.203 0.033 0.175 0.333 2.6 ⫻ 10⫺7 1.3 ⫻ 10⫺6 0.004 0.075 0.008 0.0001 0.001 0.0001 0.005 0.06 * Except where indicated otherwise, values are the number (%). SVV ⫽ small-vessel vasculitis surrounding an uninflamed temporal artery (TA); GCA ⫽ giant cell arteritis; PMR ⫽ polymyalgia rheumatica; ESR ⫽ erythrocyte sedimentation rate; CRP ⫽ C-reactive protein. software (SAS Institute, Cary, NC). The chi-square test or Fisher’s exact test, when appropriate, was applied to dichotomous variables. Odds ratios (ORs) and 95% confidence intervals (95% CIs) were computed. For continuous or ordinal variables, Wilcoxon’s rank sum test was performed. All tests were 2-sided, and P values less than 0.05 were considered significant. RESULTS Population. TA biopsy specimens from 142 men and 348 women (n ⫽ 490) were reviewed (83% of requested biopsies). Among them, 197 patients had pure GCA, 192 had GCA with PMR symptoms, and 101 had pure PMR (Tables 1 and 2). Two hundred eighty patients were diagnosed as having biopsy-proven GCA: patients in the group with a positive TA biopsy finding could have clinical symptoms of GCA without PMR symptoms (clinically pure GCA; n ⫽ 165), mixed symptoms of GCA and PMR (n ⫽ 108), or PMR symptoms with no GCA symptoms (n ⫽ 7) (Table 1). One hundred seventy-five patients had a negative TA biopsy finding and presented with GCA or PMR symptoms (27 clini- cally pure GCA, 76 clinically pure PMR, 72 mixed clinical symptoms of GCA and PMR) (Table 2). Thirtyfive patients had a pathologic diagnosis of SVV: 5 patients with GCA symptoms, 12 with both GCA and PMR symptoms, and 18 with PMR symptoms (Tables 1 and 2). SVV subgroup. Isolated SVV was diagnosed in 35 of 490 patients (7%) (18 women and 17 men, mean ⫾ SD age 74.5 ⫾ 9.4 years) (Tables 1 and 2). Thirty of the 35 patients had PMR symptoms (86%). Five patients had a clinical diagnosis of GCA, 12 had a clinical diagnosis of GCA with PMR symptoms, and 18 had a clinical diagnosis of PMR (Tables 1 and 2). None of these patients had symptoms of another inflammatory, infectious, or neoplastic disease on the followup examination. On microscopic examination, SVV consisted of mononuclear inflammatory cells, essentially small lymphocytes, surrounding a capillary distant from the temporal artery (Figures 1 and 2). The number of mononuclear inflammatory cells ranged from 15 to 90 2568 CHATELAIN ET AL Table 2. PMR* Characteristics of the patients with SVV versus the patients with biopsy-negative GCA and/or No. male/female (male:female ratio) Age, mean ⫾ SD years Clinical diagnosis Pure GCA GCA with PMR symptoms Pure PMR Duration of symptoms, median (range) weeks Corticosteroid response Systemic symptoms Fever Asthenia Anorexia Weight loss Headache Malaise Jaw claudication Visual impairment Blindness PMR symptoms Abnormal TA on clinical examination ESR, median (range) mm/hour CRP, median (range) mg/liter Haptoglobin, median (range) gm/liter Fibrinogen, mean ⫾ SD gm/liter Hemoglobin, median (range) gm/liter Mean globular volume, median (range) 3 Platelet count, median (range) ⫻ 109/liter Ferritin, median (range) g/liter Patients with SVV (n ⫽ 35) Patients with biopsy-negative GCA and/or PMR (n ⫽ 175) 17/18 (0.94) 74.5 ⫾ 9.4 52/123 (0.42)† 73.9 ⫾ 7.4 5 (14) 12 (34) 18 (51) 36 (1–730) 30 (86) 27 (77) 6 (17) 22 (63) 11 (31) 4 (11) 12 (34) 5 (14) 8 (23) 7 (20) 1 (3) 30 (86) 6 (17) 72 (21–130) 62 (2–156) 3.8 (1.7–6.8) 5.5 ⫾ 1.4 122 (85–143) 91 (78–111) 335 (159–700) 250 (3–786) 27 (15) 72 (41) 76 (43) 53 (0–767) 164 (94) 148 (85) 65 (37)‡ 132 (75) 63 (36) 44 (25) 87 (50) 19 (11) 23 (13) 28 (16) 3 (2) 148 (85) 34 (19) 77 (40–145) 64 (3–271) 3.8 (0.8–9.8) 6 ⫾ 1.7 118 (71–165) 90 (77–104) 363 (130–617) 259 (18–1,330) * Except where indicated otherwise, values are the number (%). See Table 1 for definitions. † P ⫽ 0.03 versus patients with SVV. ‡ P ⫽ 0.02 versus patients with SVV. Figure 1. Small-vessel vasculitis surrounding an uninflamed temporal artery (arrow) (hematoxylin–eosin–saffron stained; original magnification ⫻ 4). Figure 2. Mononuclear inflammatory cells around a capillary distant from the temporal artery (arrow) (hematoxylin–eosin–saffron stained; original magnification ⫻ 25). SVV AS A DIAGNOSTIC CRITERION FOR PMR Table 3. 2569 Quality of TA biopsy specimens in patients with SVV versus patients with biopsy-proven GCA* Quality of the biopsy specimen Excellent Good Medium Bad Orientation of the arterial segment Transverse Longitudinal No. of arterial segments, median (range) No. of levels, median (range) Patients with SVV (n ⫽ 35) Patients with biopsy-proven GCA (n ⫽ 280) 1 (3) 23 (66) 4 (11) 7 (20) 8 (3) 137 (49) 63 (22) 72 (26) 31 (89) 4 (11) 3 (1–22) 6 (2–33) 227 (81) 53 (19) 3 (1–26) 6 (1–38) * Except where indicated otherwise, values are the number (%). See Table 1 for definitions. (median 20). Inflammatory cells extended into the connective tissue surrounding capillaries in 6 cases and surrounded small nerves in 4 cases. There was no necrosis of the capillary wall or leukocytoclastic vasculitis. SVV lesions were focal and were seen in only 1 arterial segment in 32 cases (91%) and in 2 arterial segments in 3 cases (9%). SVV was seen in 1–10 levels (median 3), and in all levels in only 9 cases (26%). The temporal artery showed mild to moderate arteriosclerosis without inflammatory infiltrates. Calcifications were seen in the media in 13 cases (37%). The initial pathology reports noted the presence of SVV in 3 cases (9%) and described the TA biopsy findings as normal (6 cases) or showing evidence of arteriosclerosis (29 cases). Comparison of the 3 subgroups of patients. Biopsy quality. The quality of the TA biopsy specimens, number of arterial segments (average of 3 arterial segments in the 3 subgroups of patients), and number of levels generated from the paraffin blocks were similar in the SVV group, in the biopsy-proven GCA group, and in the biopsy-negative group (Tables 3 and 4). Clinical features. The proportion of men was higher in the SVV group than in the biopsy-proven GCA group (Table 1) and the negative TA biopsy group (Table 2). The age of the patients and the time to diagnosis were similar among the 3 groups (Tables 1 and 2). Patients in the SVV group had fewer systemic symptoms and fewer occurrences of asthenia, fever, weight loss, headache, jaw claudication, and abnormal temporal artery at clinical examination than patients in the biopsy-proven GCA group (Table 1). Patients in the SVV group had PMR symptoms more often than patients in the biopsy-proven GCA group (P ⫽ 2.6 ⫻ 10⫺7, OR 9.17 [95% confidence interval 3.44–24.4]). SVV with a spared temporal artery was seen more frequently in patients with PMR symptoms than in patients with GCA symptoms (Table 1). There was only 1 case of blindness in the SVV group, compared with 25 cases in the biopsy-proven GCA group (P ⫽ 0.33) (Table 1) and 3 cases in the biopsynegative group (P ⫽ 0.51) (Table 2). Patients in the SVV group had fever less often than patients in the negative TA biopsy group, but otherwise had the same clinical signs and symptoms Table 4. Quality of TA biopsy specimens in patients with SVV versus patients with biopsy-negative GCA and/or PMR* Quality of the biopsy specimen Excellent Good Medium Poor Orientation of the arterial segment Transverse Longitudinal No. of arterial segments, median (range) No. of levels, median (range) Patients with SVV (n ⫽ 35) Patients with biopsy-negative GCA and/or PMR (n ⫽ 175) 1 (3) 23 (66) 4 (11) 7 (20) 6 (3) 116 (66) 40 (23) 13 (7) 31 (89) 4 (11) 3 (1–22) 6 (2–33) 161 (92) 14 (8) 3 (1–21) 6 (2–18) * Except where indicated otherwise, values are the number (%). See Table 1 for definitions. 2570 CHATELAIN ET AL (including PMR symptoms) (Table 2). Seventy-six patients (43%) in the negative TA biopsy group had a clinical diagnosis of PMR (Table 2). Biologic features. ESR, haptoglobin level, fibrinogen level, and platelet count were higher in the biopsyproven GCA group than in the SVV group (Table 1). Patients in the SVV and negative TA biopsy groups shared the same biologic features (Table 2). In our study, 18 of the 94 patients (19%) diagnosed as having pure PMR and with an uninflamed TA on microscopic examination showed signs of SVV on the TA biopsy, whereas the biopsy findings were negative in the remaining 76 patients. Control group. Seventy-one patients had a TA biopsy performed in 2004. Twenty-two patients had a final diagnosis of GCA and/or PMR, and were excluded from the control group. The ESR in all but 6 of the remaining 49 patients was ⬎50 mm/hour at the time of the TA biopsy. The 49 patients included 31 women (mean ⫾ SD age 76 ⫾ 9.8 years) and 18 men (mean ⫾ SD age 69.5 ⫾ 10 years). Various final diagnoses were made, including cancers (12 patients), headache (4 patients), fibromyalgia (1 patient), stroke (1 patient), thrombosis of the central retinal artery (3 patients), degenerative joint disease (4 patients), myeloproliferative disease (2 patients), non-Hodgkin’s lymphoma (3 patients), myelodysplasia (1 patient), infectious diseases (12 patients), hemochromatosis (1 patient), polymyositis (1 patient), and RA (4 patients). TA biopsy specimens consisted, on average, of 3 arterial segments, and 6 levels were cut. All patients had a negative TA biopsy finding during microscopic examination, except 1 patient (2%), whose specimen showed SVV. SVV lesions were seen on 1 arterial section on the first 2 levels. This patient was a 76-year-old woman with a severe inflammatory syndrome who had a final clinical diagnosis of pyelonephritis. DISCUSSION Isolated SVV surrounding an uninflamed temporal artery is a rare finding in TA biopsy specimens. In this study, it was found in 7% of patients with a clinical diagnosis of GCA and/or PMR. In comparison, Chakrabarty and Franks (15) found SVV in 22 of 172 TA biopsies (13%) performed for symptoms suggestive of GCA, regardless of the final diagnosis, and Schmidt and Löffler (16) found an isolated perivascular location of inflammatory cells in 6 of 85 consecutive patients with GCA (7%). In our study, SVV lesions were focal, usually seen in only 1 arterial segment during microscopic examination and only on a few levels. Chakrabarty and Franks (15) have already shown the importance of examining multiple levels, because in their series, SVV was found in 64% of the 22 cases (14 cases) on the first level and in 36% of the 22 cases (8 cases) on 2 additional levels. Our study shows that SVV is a microscopic finding often neglected by pathologists, and is described in only 9% of the cases on the initial pathologic report. Immunohistochemical study was performed only by Disdier et al, who showed that the inflammatory cells surrounding capillaries in SVV were mainly T lymphocytes associated with B lymphocytes and macrophages (7). Paraffin blocks of the TA biopsy specimens were not available in our study, and we could not perform immunohistochemistry or examine new levels. The clinical significance of SVV is still unclear. SVV has been described in patients presenting with vascular diseases (GCA, polyarteritis nodosa, cholesterol emboli, cryoglobulinemia, and Wegener’s granulomatosis), inflammatory diseases (PMR, RA, relapsing polychondritis, crescentic glomerulonephritis, uveitis, and erythema nodosum), infectious diseases (hepatitis C with mixed cryoglobulinemia, bacterial endocarditis, and acute cytomegalovirus infection), and neoplasia (retroperitoneal schwannoma, acute necrosis of hypophyseal tumor, and Waldenström’s macroglobulinemia) (7,8,10). Our study is the first to show a strong association between SVV and PMR symptoms within a large series of patients with GCA and/or PMR diagnosed using conventional pathologic and clinical criteria. There are only 3 studies of SVV published in the literature, but none showed a correlation with PMR. Disdier et al studied a series of 28 patients with isolated vasculitis of vasa vasorum on the TA biopsy specimen (7). Patients had various clinical diagnoses, but only 4 patients (14%) had a clinical diagnosis of PMR. The authors concluded that the finding of isolated vasculitis of vasa vasorum on the TA biopsy specimen frequently indicates a systemic vasculitis disorder (7). Esteban et al studied a series of 28 patients with a finding of SVV surrounding a spared temporal artery on TA biopsy specimens (8). Twelve patients had a clinical diagnosis of GCA, 3 patients had a diagnosis of systemic necrotizing vasculitis, and 13 patients had no precise diagnosis. Interestingly, 14 patients (50%) had PMR symptoms, but this association was not assessed in the study. The authors concluded that, among patients presenting with SVV surrounding a spared temporal artery on TA biopsy specimens, a diagnosis of GCA could SVV AS A DIAGNOSTIC CRITERION FOR PMR be reasonably established in most, when there was no apparent evidence of additional organ involvement (8). Corcoran et al compared 81 TA biopsy specimens that showed isolated SVV surrounding a spared temporal artery (among which 27 had GCA and 13 had PMR) with 76 normal or arteriosclerotic TA biopsy specimens with no inflammation (9). Patients had various clinical diagnoses in both groups. Thirteen patients (16%) in the SVV group and 11 patients (14%) in the group with no inflammation on TA biopsy specimens had a clinical diagnosis of PMR. The authors found no difference between the groups and considered that SVV was probably related to aging (9). In our study we did not review TA biopsy specimens from patients with other diseases, only from patients who fulfilled the clinical diagnostic criteria for GCA and/or PMR (in the GRACG study, TA biopsies were performed even in patients with pure PMR with no clinical symptoms of GCA). In this specific population, in which all clinical and biologic data have been prospectively collected using preestablished questionnaires, the association between SVV and PMR is statistically significant (OR 9.17 [95% CI 3.44–24.4]). Patients with SVV have distinctive biologic and clinical features. They are more often men, have fewer systemic and cranial symptoms, and more frequently have a normal temporal artery on clinical examination than patients with biopsy-proven GCA. They also have lower ESR, fibrinogen levels, haptoglobin levels, and platelet counts. Patients with SVV share the same clinical and biologic features, including PMR symptoms, as patients with negative TA biopsy findings (43% in this latter subgroup were diagnosed as having PMR). In our study, 19% of patients with a clinical diagnosis of PMR had signs of SVV during microscopic examination of TA biopsy specimens. The pathogenesis of PMR is still unclear, but our findings emphasize the possible vasculitic nature of the disease. Meliconi and colleagues’ study showed that microscopic examination of synovial biopsy specimens from patients with PMR indicated inflammatory infiltration in the synovial tissue in 12 of 19 patients (63%), and 6 of these 12 patients had synovial vasculitis (17). Synovial vasculitis, described in their studies (2,17), shared the same microscopic features as the SVV found in the TA biopsy specimens in our studies. Weyand et al suggested that a subclinical vasculitis may be present in the temporal arteries of patients with PMR (18). Those authors found in situ production of messenger RNA for cytokines indicative of macrophage activation 2571 (interleukin-1␤ [IL-1␤], IL-6, and transforming growth factor ␤1) and T cell activation (IL-2) in temporal artery specimens in patients with a clinical diagnosis of PMR, despite the lack of microscopic evidence of arteritis (18). They suggested that patients with PMR had subclinical vascular involvement that could correspond to the SVV found in our study on microscopic examination of TA biopsy specimens. In conclusion, our results indicate that SVV is an underestimated microscopic finding strongly associated with PMR symptoms in patients with a clinical diagnosis of GCA and/or PMR. SVV should not be considered as a positive biopsy finding because it is not associated with temporal arteritis. On the other hand, it is difficult to consider it as an entirely negative biopsy finding because of the presence of inflammatory infiltrates surrounding small capillaries. No pathognomonic test is currently available for the diagnosis of PMR. Our findings raise the question of whether SVV could be a pathologic marker for PMR. At this stage, our study does not provide the answer to that question, just a hypothesis to be tested. One patient in the control group without GCA and/or PMR had SVV on the TA biopsy specimen. The specificity and positive predictive value of SVV for PMR must be assessed in prospective studies. ACKNOWLEDGMENT We thank Valerie Petigny for expert technical assistance. AUTHOR CONTRIBUTIONS Dr. Chatelain had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study design. Chatelain, Duhaut, Bosshard, Pellet. Acquisition of data. Chatelain, Duhaut, Loire, Bosshard, Sevestre. Analysis and interpretation of data. Chatelain, Duhaut, Piette, Ducroix. Manuscript preparation. Chatelain, Duhaut, Pellet, Ducroix. Statistical analysis. Chatelain, Duhaut. REFERENCES 1. Weyand CM, Goronzy JJ. Giant-cell arteritis and polymyalgia rheumatica. Ann Intern Med 2003;139:505–16. 2. Salvarani C, Cantini F, Boiardi L, Hunder GG. Polymyalgia rheumatica and giant-cell arteritis. N Engl J Med 2002;347:261–71. 3. McDonnel PJ, Moore GW, Miller NR, Hutchins GM, Grenn WR. Temporal arteritis: a clinico-pathologic study. Ophthalmology 1986;93:518–30. 4. Lie JT. When is arteritis of the temporal arteries not temporal arteritis? J Rheumatol 1994;21:186–9. 5. Salvarini C, Gabriel SE, O’Fallon WM, Hunder GG. The inci- 2572 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. dence of giant cell arteritis in Olmsted County, Minnesota: apparent fluctuation in a cyclic pattern. Ann Intern Med 1995;123: 192–4. Schmidt WA, Gromnica-Ihle E. Incidence of temporal arteritis in patients with polymyalgia rheumatica: a prospective study using colour Doppler ultrasonography of the temporal arteries. Rheumatology (Oxford) 2002;41:46–52. Disdier P, Pelissier JF, Harle JR, Figarella-Branger D, Bolla G, Weiller PJ. Significance of isolated vasculitis of the vasa vasorum on temporal artery biopsy. J Rheumatol 1994;21:258–60. Esteban MJ, Font C, Hernandez-Rodriguez J, Valls-Sole J, Sanmarti R, Cardellach F, et al. Small-vessel vasculitis surrounding a spared temporal artery: clinical and pathologic findings in a series of twenty-eight patients. Arthritis Rheum 2001;44:1387–95. Corcoran GM, Prayson RA, Herzog KM. The significance of perivascular inflammation in the absence of arteritis in temporal artery biopsy specimens. Am J Clin Pathol 2001;115:342–7. Genereau T, Lortholary O, Pottier MA, Michon-Pasturel U, Ponge T, de Wazieres B, et al, for the French Vasculitis Study Group. Temporal artery biopsy: a diagnostic tool for systemic necrotizing vasculitis. Arthritis Rheum 1999;42:2674–81. Duhaut P, Berruyer M, Pinede L, Demolombe-Rague S, Loire R, Seydoux D, et al, for the Groupe de Recherche sur l’Artérite à Cellules Géantes. Anticardiolipin antibodies and giant cell arteritis: a prospective, multicenter case–control study. Arthritis Rheum 1998;41:4:701–9. Duhaut P, Pinede L, Demolombe-Rague S, Loire R, Seydoux D, Ninet J, et al, on behalf of the Groupe de Recherche sur l’Artérite à Cellules Géantes. Giant cell arteritis and cardiovascular risk factors: a multicenter, prospective case–control study. Arthritis Rheum 1998;41:1960–5. Hunder GG, Bloch DA, Michel BA, Stevens MB, Arend WP, Calabrase LH, et al. The American College of Rheumatology 1990 criteria for the classification of giant cell arteritis. Arthritis Rheum 1990;33:1122–8. Chuang TY, Hunder GG, Ilstrup DM, Kurland LT. Polymyalgia rheumatica: a 10-year epidemiologic and clinical study. Ann Intern Med 1982;97:672–80. Chakrabarty A, Franks AJ. Temporal artery biopsy: is there any value in examining biopsies at multiple levels? J Clin Pathol 2000;53:131–6. Schmidt G, Loffler KU. Temporal arteritis: comparison of histological and clinical findings. Acta Ophthalmol (Copenh) 1994;72: 319–25. Meliconi R, Pulsatelli L, Uguccioni M, Salvarani C, Macchioni P, Melchiorri C, et al. Leukocyte infiltration in synovial tissue from the shoulder of patients with polymyalgia rheumatica: quantification analysis and influence of corticosteroid treatment. Arthritis Rheum 1996;39:1199–207. Weyand CM, Hicok KC, Hunder GG, Goronzy JJ. Tissue cytokine patterns in patients with polymyalgia rheumatica and giant cell arteritis. Ann Intern Med 1994;121:484–91. APPENDIX A: MEMBERS OF THE GROUPE DE RECHERCHE SUR L’ARTÉRITE À CELLULES GÉANTES The GRACG study was conducted at the following institutions (all in France): Departments of pathology. CHU Nord, Amiens: M. Bensaid, MD, A. Bruniau, MD, D. Chatelain, MD, PhD, C. Cordonnier, MD, M. F. Gontier, MD, F. Leclercq, MD, F. Nagorniewcz, MD, A. M. Marie, MD, H. Sevestre, MD, PhD; Centre Hospitalier, Beauvais: S. Bendjaballah, MD; Centre Hospitalier, Belfort: C. Lassabe, MD, C. Monnin, MD; CHU, Besançon: B. Kantelip, MD, G. Marquet, MD; CHU, Bordeaux: J. M. Meunier, MD, C. Segretin, MD; CHU Am- CHATELAIN ET AL broise Paré, Boulogne-Billancourt: A. M. Bergemer, MD; CHG, Bourg-en-Bresse: R. Angonin, MD, C. Cavailles, MD, J. H. Gasse, MD; CHG, Bourgoin-Jallieu: B. Bringeon, MD; CHU, Brest: J. P. Leroy, MD, P. Rivière, MD, R. Roué, MD, A. Volant, MD; CHU, Caen: J. C. Mandart, MD, P. Rousselot, MD; CHU ClermontFerrand: A. Clemenson, MD, P. Dechelotte, MD, J. L. Kemeny, MD; Hôpital Louis Mourier, Colombes: J. Barge, MD; CHU, Dijon: M. Bordes, MD, L. Arnould, MD, P. Chalabreysse, MD, E. Justrabo, MD, R. Michiels, MD, F. Piard, MD; CHG, Firminy: M. Pialat, MD; CHU, Grenoble: H. Dufour, MD, F. Le Marc Hadour, MD, M. H. Panh, MD, B. Pasquier, MD, N. Pinel, MD; CHG, Lagny: G. Wajcner, MD; CHG, Libourne: M. Julien, MD, M. Augrand, MD; CHU, Lyon: C. Agard, MD, F. Allias, MD, J. Beurlet, MD, F. Berger, MD, PhD, N. Berger, MD, C. Crozes, MD, C. David, MD, M. Devouassoux, MD, B. Frappart, MD, A. Jouvet, MD, M. Guillaud, MD, S. Isaac, MD, R. Loire, MD, A. Milox, MD, M. Patricot, MD, J. Y. Scoazec, MD, PhD, A. Tabib, MD, F. Thivolet, MD, P. Tuaillon, MD, A. Vauzelle, MD, D. Vitrey, MD; Institut Pasteur, Lyon: M. Chevallier, MD, F. Gérard, MD, M. Maisonneuve, MD, M. Neyra, MD; HIA Desgenettes, Lyon: M. C. Mouden, MD, M. C. Saccharin, MD; CHG, Montanban: M. H. Hounieu-Ritoux, MD; CHU, Montpellier, Hôpital Lapeyronie: P. Baldet, MD, I. Serre, MD; CHU, Nancy: F. Baylac, MD, A. Duprez, MD, J. C. Lemaire, MD, F. Plenat, MD; Hôpital Saint-Joseph, Paris: E. Baviera, MD, P. Vaury, MD; Hôpital Tenon, Paris: P. Callard, MD; Hôpital Cochin, Paris: A. Carlioz, MD, M. Forest, MD, B. Terris, MD, PhD; Hôpital de la Pitié-Salpêtrière, Paris: F. Charlotte, MD; Hôpital du Val de Grâce, Paris: H. Schill, MD, P. Camparo, MD; Hôtel-Dieu, Paris: P. Texier, MD; Laboratoire d’Anatomie Pathologique, Perpignan: M. Guiou, MD; CHG, Roanne: S. Vancina, MD; CHG, Roanne: P. Terdjman, MD; CHU, Saint-Etienne: S. Boucheron, MD, G. Poulard, MD; CHG, Saint-Quentin: S. Carton, MD; Laboratoire Anatomie Pathologique, Saint-Savine: F. Nollez, MD; CHU, Strasbourg: G. Averous, MD, J. P. Bellocq, MD, M. P. Chenard, MD, A. Engelmann, MD, L. Marcellin, MD, C. Marin, MD, A. Neuville, MD, A. Onea, MD, N. Rouyer, MD, E. Vaneenoge, MD, A. Warter, MD; Hôpital Bretonneau, Tours: F. Fetissof, MD, R. Khallouf, MD, L. Renjard, MD; CHG, Troyes: C. Hopfner, MD, S. Mehaut, MD; CHG, Valence: N. Bottero, MD, V. André, Y. Suignard; Laboratoire d’Anatomie Pathologique, Villeurbanne: C. Donne, MD. Clinical departments. CHU Nord, Amiens: S. Canaple, MD, R. Cevallos, MD, J. P. Ducroix, MD, P. Duhaut, MD, F. Grados, MD, PhD, L. Le Page, MD, MPH, Z. Oukachbi, MD, A. Smail, MD, V. Salle, MD, J. Schmidt, MD; CHG, Belfort: V. Brezovski, MD, F. Duchêne, MD, O. Labrune-Bobey, MD, G. Simon, MD; CHU, Besançon: M. Becker-Schneider, MD, B. De Wazières, MD, T. Fest, MD, PhD, A. Maskani, MD, H. Gil, MD, J. L Dupont, MD; CHU, Bordeaux: C. Mestre, MD; Hôpital Ambroise Paré, Boulogne: T. Hanslik, MD, PhD, J. C. Boulard, MD; CHG, Bourg-en-Bresse: B. Bocquet, MD, P. Debat, MD, P. Granier, MD, R. Philipot, MD, M. Rousille, MD, N. Sprunck, MD; CHG, Bourgoin-Jallieu: M. Roux, MD; CHU, Brest: A. Cenac, MD, P. Le Goff, MD, A. Saraux, MD; CHU, Caen: L. Agron, MD; CHG, Chambéry: C. Cugnet, MD; CHG, Chartres: R. Damade, MD; CHU, Clermont-Ferrand: M. André, MD, O. Aumaı̂tre, MD, D. Meloux, MD, B. Sauvezie, MD, J. Schmidt, MD; Hôpital Louis Mourier, Colombes: J. Pouchot, MD, PhD, J. Vinceneux, MD; CHG, Coulommiers: M. Gatfosse, MD; CHU, Dijon: P. Pfitenmeyer, MD, S. Baudry-Brunet, MD; CHG, Firminy: P. Sagnol, MD; CHU, Grenoble: T. Auday, MD, B. Imbert, MD, C. Massot, MD, F. Sarrot-Reynauld, MD; Cabinet de Rhumatologie, La Tour du Pin: I. Gozzo, MD, E. Sornay-Rendu, MD; Hôtel-Dieu, Le Creusot: C. Boussuge, MD; CH Robert Boulin, Libourne: V. Meunier, MD; CHG, Lons-le-Saunier: B. Saugier, MD; Clinique Protestante, Lyon: S. Demolombe-Ragué, MD, C. Dolmazon, MD, L. Pinède, MD, PhD; CHU, Lyon: P. Bourrat, MD, J. F. Brantus, MD, R. Chapurlat, MD, L. Delattre, MD, B. Demiaux, MD, PhD, F. Duvert, MD, M. Filbet, MD, E. Fontanges, M. H. Girard-Madoux, MD, P. Haond, MD, D. Liens, MD, S. Martinon, MD, P. Miossec, MD, PhD, E. Noël, MD, C. Savy, SVV AS A DIAGNOSTIC CRITERION FOR PMR MD; Hôpital Saint-Joseph, Lyon: P. Bachet, MD, D. ManièreConstantin, MD; HIA Desgenettes, Lyon: B. Brunot, MD, P. Debouordeau, MD, A. Fléchaire, MD; CHG Montauban: C. Seigneuric, MD; CHU Saint-Eloi: E. Braun, MD, E. Quere, MD, PhD; CHU, Nancy: B. Mouget, MD, G. Vaillant, MD, D. Wahl, MD, PhD; Hôpital Saint-Joseph, Paris: N. Postel-Vinay, MD; Hôpital Cochin, Paris: P. Hilliquin, MD, C. Job-Deslandre, MD, A. Marin, MD, S. Perrot, MD, M. Renoux, MD; Hôpital de la Pitié-Salpêtrière, Paris: Z. Amoura, MD, PhD, D. Boutin, MD, PhD, C. Chapelon-Abric, MD, N. Costedoat-Chalumeau, MD, PhD, J. Haroche, MD, J. C. Piette, MD; 2573 Hôpital du Val-de-Grâce, Paris: J. P. Algaxres, MD; CHG, Roanne: P. Amaranto, MD, J. Roche, MD; CHG, Romans: H. Colomb, MD; Cabinet de Rhumatologie, Saint-Priest: A. Lespine; CHU, SaintEtienne: K. Bouchou, MD, P. Cathebras, MD, PhD, H. Rousset, MD; CHU, Strasbourg: S. Friess, MD, B. Goichot, MD, PhD, F. Grunenberger, MD, M. Imler, MD, R. M. Javier, MD, G. Kaltenbach, MD, PhD, A. Pradignac, MD, J. L. Schlienger, MD, S. Vinzio, MD, C. Wicky-Strich, MD; CHU, Tours: P. Ménage, MD; CHG, Troyes: A. Fur, MD; CHG, Valence: B. Drouin, MD; CHG, Viriat: A. Pirolet, MD.