Gastrointestinal involvement in granulomatosis ... - Wiley Online Library

International Journal of Rheumatic Diseases 2014; 17: 412–419

ORIGINAL ARTICLE

Gastrointestinal involvement in granulomatosis with polyangiitis and microscopic polyangiitis: histological features and outcome Joerg LATUS,1 Ina KOETTER,1 Peter FRITZ,2 Martin KIMMEL,1 Dagmar BIEGGER,2 German OTT,3 Eduard F. STANGE,4 Kerstin AMANN,5 Dominik M. ALSCHER,1 and Niko BRAUN1 1 Division of Nephrology, Department of Internal Medicine, Robert-Bosch Hospital, 2Margarete-Fischer–Bosch Institute of Clinical Pharmacology, University of Tuebingen, 3Division of Pathology, Department of Diagnostic Medicine, 4Division of Gastroenterology and Hepatology, Department of Internal Medicine, Robert-Bosch Hospital, Stuttgart, and 5Department of Pathology, Nephropathology, University of Erlangen-Nuernberg, Erlangen, Germany

Abstract Aim: Gastrointestinal (GI) involvement in patients with granulomatosis with polyangiitis (GPA) or microscopic polyangiitis (MPA) is rare. Method: Medical charts of seven patients with GPA and MPA and GI involvement were reviewed regarding clinical presentation, outcome, diagnostic tools and therapy. Second, the cellular composition of the inflammatory infiltrate associated with the vascular lesions in histological samples (ileum, colon, rectum, duodenum) were investigated to identify possible treatment targets. Immunohistochemistry was done with antibodies against CD20, CD3 and CD34. Samples from a healthy control group (n = 15) were used for comparison. Results: Mean age at onset of the first symptoms of vasculitis was 48 21.3 years. At time of diagnosis GI symptoms were present in five out of seven patients (71%) and occurred during relapse of the vasculitis in two patients (29%). All patients had abdominal pain, four of seven (57%) had an acute kidney injury and three patients required renal replacement therapy. At the time of diagnosis five of seven patients (71%) required surgery and mean Birmingham Vasculitis Activity Score (BVAS) on admission was high (26.3 7.7). Regarding outcome, one patient died due to gastrointestinal bleeding. Histological analysis showed significantly higher expression of CD3 in this patient compared to the control group (P = 0.02). Analysis of expression of CD20 and CD34 showed no statistically significant differences between patients with GPA and MPA with GI involvement compared to the control group. Conclusions: GI involvement in GPA and MPA is rare. Therapy directed at T cells might be an alternative treatment option. Key words: gastrointestinal involvement, GPA, MPA, target-related therapy.

INTRODUCTION

Correspondence: Dr Joerg Latus, Division of Nephrology, Department of Internal Medicine, Robert-Bosch Hospital, Auerbachstrasse 110, Stuttgart 70376, Germany. Email: [email protected]

Antineutrophil cytoplasmatic antibody (ANCA)-associated vasculitides (AAV) are potentially life-threatening diseases, and include microscopic polyangiitis (MPA), granulomatosis with polyangiitis (former Wegener’s, GPA), eosinophilic granulomatosis with polyangiitis

© 2013 Asia Pacific League of Associations for Rheumatology and Wiley Publishing Asia Pty Ltd

Gastrointestinal involvement in GPA/MPA

(EGPA, formerly Churg–Strauss syndrome) and renal limited vasculitis.1,2 Long-term survival in patients with GPA and MPA has improved dramatically since the addition of cyclophosphamide to the therapeutic regimen. Notwithstanding, mortality is still higher than in the general population and some patients have relapsing disease and a poor prognosis. At time of diagnosis of AAV or during relapse of the disease, leading clinical symptoms vary widely between patients. The symptoms and signs of systemic vasculitis involving the gastrointestinal (GI) tract mostly result from mesenteric ischemia. In contrast to patients with chronic mesenteric ischemia due to arteriosclerosis, patients with vasculitis present mostly with acute onset of abdominal pain. Small bowel obstruction secondary to strictures, bowel perforations or massive GI bleeding could be the first clinical manifestations of systemic vasculitis. In contrast to patients with GPA and MPA, patients with polyarteritis nodosa (PAN) more often present with symptoms related to GI ischemia (reported rates 40–60%3–6) caused by the involvement of larger-size vessels.7 Only few data exist in in the current literature about patients with GPA and GI involvement.8–14 Several studies about treatment with rituximab in GPA exist in the current literature.15,16 CD20 is expressed in almost all stages of B cell development and in all peripheral B cells. Immunohistochemistry can be used to determine the presence of CD20 on cells in different histological tissue sections. Inflammatory cells are involved in the pathogenesis of systemic vasculitis. CD34 is an important adhesion molecule and is required for T cells to enter lymph nodes and is expressed in vascular endothelial cells. Therefore, it has been shown to facilitate cell migration,17,18 and cell migration is thus involved in the inflammatory processes. The CD3 T cell co-receptor is part of the T cell receptor complex in mature T lymphocytes. Recently, Berden et al.19 showed that tubular T cell infiltrates predict the outcome of glomerulonephritis in patients with AAV. The study objective here was to analyze retrospectively patients with GPA and MPA with GI involvement regarding clinical presentation, outcome, diagnostic tools and therapy. Second, we studied the cellular composition of the inflammatory infiltrate in histopathological samples associated with vascular lesions to detect possible treatment targets in patients presenting with GI involvement.


MATERIALS AND METHODS Data from seven patients admitted to our hospital between 1995 and 2010 with GPA and MPA and GI involvement were compiled. Patient inclusion required: 1 Fulfilling the 1990 American College of Rheumatology classification criteria for GPA20 and using the Chapel Hill Consensus Conference (CHCC) nomenclature1 for MPA; and 2 GI involvement considered to be associated with GPA and MPA: i GI symptoms that were present at the time of vasculitis diagnosis, ii GI symptoms that were present at the time of a relapse of the vasculitis. In August 2010 seven patients fulfilled these criteria. We retrospectively reviewed the medical charts of these seven patients with GPA and MPA and GI involvement. Informed consent to publish clinical data and images was obtained from all patients. The ethics committee of the University of Tuebingen approved the study. We focused on clinical presentation, demographics, laboratory work-up, radiological and histological findings, therapy and outcome. We categorized GI involvement by the following items: abdominal pain, diarrhea, gastroduodenal ulcers, colorectal ulcers, upper GI bleeding, lower GI bleeding, bowel perforations, intestinal occlusion, surgical abdomen, acute pancreatits, gastritis, abnormal angiography, abnormal abdominal computed tomography (CT) scan. According to recent studies,2,21–23 we differentiated these patients as having or not having a surgical abdomen. Surgical abdomen was defined as GI manifestations of the vasculitis requiring surgery. Furthermore, the following extra-intestinal manifestations were recorded: myalgia, arthralgia, ENT involvement, renal involvement, hemodialysis requirement, lung involvement and results of gastroscopy, colonoscopy and CT scan of the abdomen. Furthermore, we investigated the histological samples (n = 11: ileum, colon, rectum, doudenum) of patients with GI involvement in GPA and MPA concerning expression of CD20, CD34 and CD3 with immunostaining and compared these with histological samples of a control group of 15 patients (n = 15 samples). The control group samples were taken during preventive medical check-up and included 15 patients without

413

J. Latus et al.

signs of vasculitis and systemic inflammation. Mean age of the control group was 54.7 15.1 years (11 male, four female). Biopsies were taken from sigmoid (n = 3), rectum (n = 3), colon (n = 4), ileum (n = 4) and duodenum (n = 1).

performed the semiquantitative scoring were blinded to the diagnosis. Comparisons between the two groups were made using the Fisher test. The level of significance was defined as P < 0.05.

Immunohistochemistry

RESULTS

Tissues from the samples were fixed in buffered formalin for up to 72 h. Then, after 3-lm sections had been cut, they underwent hematoxylin and eosin staining by standard methods. Immunophenotyping of CD20 used a monoclonal antibody anti-CD20cy (M 0755, DAKO, Clone L26, mouse IgG2a kappa), diluted 1 : 400 in antibody diluent (S2022, DAKO, Glostrup, Denmark). Immunophenotyping of CD34 used a monoclonal antibody CD34 (M 7165, DAKO, Clone QBEnd-10, mouse IgG1 kappa), diluted 1 : 25 in antibody diluent (S2022, DAKO, Glostrup, Denmark). Immunophenotyping of CD3 used a polyclonal antibody anti-CD3 (A 0452, DAKO, rabbit), diluted 1 : 100 in antibody diluent (S2022, DAKO, Glostrup, Denmark). For the immunostaining, we used a TechMate system (DAKO) and a dextran-coated peroxidase-coupled polymer system (REAL EnVision detection kit, K 5007, DAKO). Endogenous peroxidase blocking was done using hydrogen peroxide in phosphate buffer (peroxidase blocking solution, S 2023, DAKO). For epitope retrieval, the slides for all antibodies were heated in pH6 puffer (target retrieval solution, S 1699, DAKO). The final reaction product was produced by incubation in diaminobenzidine–hydrogen peroxide for 10 min, followed by counterstaining of nuclei in hemalaun. For assessment of staining results we used a magnification of 9 40. The expression pattern of CD20 and CD3 immunostaining was analyzed as nodal, diffuse and intraepithelial and analyzed individually (CD20 score: nodal: 0 no, 1 few, 2 moderate, 3 frequent; diffuse: 0 no, 1 1–2 cells/high power field [HPF], 2 3–5 cells/HPF, 3 >5 cells/HPF; intraepithelial: related to 100 enterocytes: 1 1–10 cells, 2 11–40 cells, 3 > 40 cells; CD3 score: nodal: 0 no, 1 few, 2 moderate, 3 frequent; diffuse: 0 no, 1 1–10 cells/HPF, 11–20 cells/HPF, 3 > 21 cells/HPF; CD34 score was analyzed as vessel density [Score 0–3 with 1: 1–5 per visual field, 2:5–10 per visual field and 3: > 10 per visual field]).

Patient data

Statistical analysis Data are reported as mean standard deviation (SD) unless otherwise specified. The two observers who

414

Between 1995 und 2010 six patients with GPA and one patient with MPA and GI involvement were admitted to our hospital. Mean age at onset of first symptoms of vasculitis was 48 21.3 years (see Table 1).

Gastrointestinal symptoms At the time of diagnosis GI symptoms were present in five out of seven patients (71%) and occurred at a relapse of the vasculitis in two patients (29%) with an interval of 7 and 18 months after time of initial diagnosis. All patients suffered from abdominal pain on admission. Four patients (57%) had diarrhea and one patient (14%) had upper GI bleeding on admission. The spectrum of GI symptoms is summarized in Table 1.

Surgery At the time of diagnosis five of seven patients (71%) required surgery (two small bowel resection, one hemicolectomy, one sigmoid resection, one ileocaecal resection) (Table 2). In patients with GPA, five of six patients (83%) required surgery, five patients at time of diagnosis and one patient 1 year after time of diagnosis during relapse of the disease. One patient (14%) with GPA had an intestinal occlusion, and one patient (14%) presented with a bowel perforation. All patients who had GI symptoms at the time of diagnosis (5/7 patients, 71%) required surgery immediately after admission.

Birmingham Vasculitis Activity Score (BVAS) score24 Mean BVAS on admission was 26.3 7.7. After 6–9 months of therapy BVAS dropped but remains still high (9.4 1.9).

Angiographic findings In total, two of seven patients (29%) underwent angiography with signs of vasculitis in both cases. Case 1 showed an active bleeding in the region of the superior mesenteric artery and was treated with a coil (see Fig. 1b,c). Case 7 showed an arterial occlusion and was immediately transferred to the operation room and a small bowel resection was done.



Table 1 Gastrointestinal (GI) manifestations, radiologic findings, anti-PR3 concentrations and anti-myeloperoxidase concentrations in seven patients with AAV and GI involvement

Case Age (years) Sex Vasculitis Abdominal pain Diarrhea Gastrodoudenal ulcers Colorectal ulcers LGI bleeding UGI bleeding Bowel perforation Intestinal occlusion Surgical abdomen Acute pancreatitis Gastritis Abnormal angiography Abnormal abdominal CT Anti-PR3 concentrations (normal < 1) Anti-MPO concentrations (normal < 1)

1

2

3

4

5

6

7

42 Male GPA Yes No Yes No Yes No No No Yes No Yes Yes Yes 38

85 Female GPA Yes No No No No Yes No No Yes No Yes ND Yes >200

75 Female MPA Yes Yes Yes Yes No Yes No No No No Yes ND Yes –

24 Male GPA Yes Yes No Yes No No No No No Yes No ND No >200

30 Female GPA Yes Yes No No No No No No No No No ND No >200

41 Male GPA Yes No No No No No Yes No Yes No No ND Yes 89

39 Male GPA Yes Yes No No No No No Yes Yes No No Yes Yes 26

–

–

143

–

–

–

–

AAV, antineutrophil cytoplasmatic antibody (ANCA)-associated vasculitides; CT, computed tomography; GPA, granulomatosis with polyangiitis; LGI, lower GI bleeding; MPA, microscopic polyangiitis; ND, not done; UGI, upper GI bleeding.

Table 2 Clinical findings in seven patients with AAV and GI manifestations Case

Age

GI manifestations present at diagnosis

Surgery

Diagnosis

Treatment

1 2 3 4 5 6 7

42 85 75 24 30 41 39

Yes Yes No No Yes Yes Yes

Small bowel resection Hemicolectomy No No Sigmoid resection Ileocaecal resection Small bowel resection

GPA GPA MPA GPA GPA GPA GPA

Steroids + cyclophosphamide + azathioprine Steroids + cyclophosphamide Steroids + cyclophosphamide + azathioprine Steroids + cyclophosphamide + azathioprine Steroids + cyclophosphamide + azathioprine + MTX Steroids + cyclophosphamide + azathioprine Steroids + Cyclophosphamide, MMF, MTX, immunoglobulins, etanercept

AAV, antineutrophil cytoplasmatic antibody (ANCA)-associated vasculitides; GI, gastrointestinal; GPA, granulomatosis with polyangiitis; MMF, mycophenolat mofetil; MPA, microscopic polyangiitis; MTX methotrexate.

Endoscopic findings Upper and/or lower GI endoscopy were performed in all patients. Two of seven (29%) patients had GI ulcers, whereas three of seven patients (43%) had gastritis only. Case 1 showed an erosive, partly ulcerative jejunitis (see Fig. 1a).

Renal function Among the seven patients, four (57%) had an acute kidney injury and three required renal replacement therapy (range 6–34 days). Kidney biopsies were done in three


patients. Histologic examinations showed a rapid crescentic glomerulonephritis (RPGN) in all kidney biopsies (see Fig. 2c,d).

Histological analysis Gastrointestinal samples were taken from all patients (5/7 patients during surgery, two patients during endoscopy) and subjected to histologic examination. Examination of small intestine or colon specimens removed from five patients during surgery detected signs of ischemic and/or thrombotic necrosis unequivocally due to

415

J. Latus et al.

(a)

(b)

(c)

Figure 1 (a) Erosive, partly ulcerative jejunitis; (b) active bleeding (arrow) in the region of the superior mesenteric artery; (c) after coiling (arrow).

(a)

(b)

(c)

(d)

Figure 2 (a) Histological sample of the ileum showing arteriolitis and perivasculitis; (b) Ileum: fibrinoid necrosis and arteriolitis; (c and d) rapid progressive glomerulonephritis.

vasculitis in three of five cases (60%) (see Fig. 2a,b). Granulomas were not observed throughout the intestinal wall.

group (P > 0.05) (see Table 3). Mean values and P-values for all parameters are given in Table 3.

Gut immunohistochemistry

Overall, five of seven patients (71%) required surgery, two patients were conservatively treated. All patients were treated with intravenous (i.v.) steroids and oral or i.v. cyclophosphamide. Five of seven patients (71%) were treated with azathioprine after induction therapy. Two patients who had GI symptoms at time of relapse

The expression of CD3 was significantly higher the patient group compared to the control group (P = 0.02) (see Table 3). The analysis of the expression of CD20 and CD34 showed no statistically significant differences between the patient compared to the control

416

Treatment



Table 3 CD20, CD3 and CD34 immunostaining in patients with GPA and MPA with GI involvement compared to control group Variable

GPA/MPA (n = 11)

Control (n = 15)

P

Score CD 20† Diffuse Nodal CD 3‡ Intraepithelial Diffuse Nodal CD 34§ Vessel densitiy

0/1,2,3

0/1,2,3

3/8 4/7

6/9 8/7

0.88 0.33

4/7 0/11 3/8

10/5 0/15 5/10

0.02 0.11 0.77

2/9

4/11

0.95

†Nodal: 0 no, 1 few, 2 moderate, 3 frequent; diffuse: 0 no, 1 1–2 cells/ HPF, 2 3–5 cells/HPF, 3 > 5 cells/HPF; intraepithelial: related to 100 enterocytes: 1 1–10 cells, 2 11–40 cells, 3 > 40 cells; HPF, high power field; ‡Nodal: 0 no, 1 few, 2 moderate, 3 frequent; diffuse: 0 no, 1 1–10 cells/HPF, 11–20 cells/HPF, 3 > 21 cells/HPF; §Vessel density: 0 no, 1 1–5, 2 5–10, 3 > 10.

of the vasculitis were previously treated with steroids and cyclophosphamide. Under treatment with azathioprine case 4 developed elevated liver enzymes and therapy was changed to methotrexate. Case 6 was treated with steroids and azathioprine. During therapy with oral cyclophosphamide case 7 developed thrombocytopenia and leucopenia. Therapy was stopped and changed to mycophenolate mofetil. During the course of the disease, he developed a relapse of the vascultis. Therapy was changed to methotrexate and prednisolone and he received i.v. immunoglobulin once.

DISCUSSION Involvement of the GI tract in GPA and MPA is rare, difficult to diagnose and associated with poor outcome. Therefore, early diagnosis is mandatory. It is noteworthy that patients with other systemic vasculitides, especially PAN, often had GI involvement.7,25 Guillevin et al.22 and Le Thi Huong et al.4 reported in their studies that primary GI symptoms in patients with different kinds of systemic necrotizing vasculitides are present in up to 13–16% of cases. In three series (total of 352 patients) one-third had GI manifestations during the course of the disease, in the majority of cases patients with PAN.7,25,26 In patients with GPA or MPA, data about GI involvement are sparse.14 Rates of involvement of the GI tract range between 5% and 11% and data about the course of the disease and outcome data are lacking.27


In our study, mean age at the time of diagnosis was 48 21.3 years, which is in line with the findings in other cohorts, including patients with different types of systemic vasculitides. All patients in our study suffered from abdominal pain, a finding that is also in agreement with previous studies.6,21,22,28,29 Interestingly, four of seven patients (57%) in our study had surgical abdomen on admission and all of them required surgery. The high rate of surgical abdomen in our study is controversially compared to the observations made by Pagnoux et al.21 None of the patients in this study with GPA and GI involvement had surgical abdomen. In our patient cohort, GI endoscopy with biopsies was done in all patients. Remarkably, neither biopsies from the upper GI tract nor from the lower GI tract showed histopathological signs of systemic necrotizing vasculitides. As described by others,30 endoscopic biopsies have a low sensitivity to diagnose GI involvement in systemic vasculitis, possibly due to their superficial character. Therefore, upper or lower GI endoscopy could not be used to exclude GI involvement in AAV patients. Examination of small intestine or colon resection specimens removed from five patients during surgery detected signs of ischemic and/or thrombotic necrosis due unequivocally to vasculitis in three of five cases. This is equivalent to the reported results from Pagnoux et al.21 (vasculitis of the GI tract was histologically proven in 50%). GI involvement in systemic vasculitis is associated with a high morbidity and mortality.5,21,22,31–33 Therefore, Pagnoux et al.21 mentioned that alternative therapeutics should be considered to control the disease rapidly, especially in these patients. In our study BVAS on admission was remarkably high (26.3 7.7) and treatment response was poor (BVAS after induction therapy 9.4 1.9). All patients were treated with i.v. pulse steroid therapy initially, oral steroids and oral cyclophosphamide and/or i.v. cyclophosphamide. Up to now, one patient died at the age of 85 years due to gastrointestinal bleeding. Four of seven patients (57%) had an acute kidney injury and three patients required renal replacement therapy (range 6–34 days). Due to the fact that response to induction therapy is poor, targeted related therapy might be an alternative in patients with GPA and severe GI manifestations to control disease and reduce morbidity and mortality.15,16 In our study we investigated CD20, CD3 and CD34 expression in histological samples from patients with GPA and MPA with GI involvement in comparison

417

J. Latus et al.

to a healthy control group. In our study CD20+ B lymphocytes were observed in the histological samples of the patient and the control group. Their occurrence was related to nodal mucosal infiltrations of B lymphocytes. No significant differences could be detected between the patient and the control group concerning CD20 expression. It is noteworthy that these findings do not exclude the important role of B cells in the course of the disease because B cells might be found in the circulation and not in the tissue. Secreting ANCA in the circulation might then cause vasculitis by activation of neutrophils in the GI vessel walls. In contrast to CD20 and CD34 expression, expression of CD3+ cells was significantly higher in samples of the patient compared to the control group (P = 0.02). Remarkably, there is some data regarding anti-CD3 or antilymphocyte T cell treatment in patients with refractory GPA.34 Recently, Berden et al.19 suggested that in addition to anti-B cell therapy, therapy directed at T cells might improve renal outcomes in patients with AAV. Further research is mandatory in this area. In conclusion, GPA and MPA with GI involvement are rare, often severe and need rapid recognition and an early treatment. Due to the poor response to established therapy, high morbidity and mortality, alternative therapeutics should be considered. Therapy directed at T cells might be an alternative treatment option. Further clinical studies with more samples are needed to underline this hypothesis.

ACKNOWLEDGEMENTS None.

COMPETING INTERESTS None.

FUNDING None.

AUTHOR CONTRIBUTIONS All authors have made substantive contributions to the study, and all authors endorse the data and conclusions. Specific author contributions: JL, MDA, NB, MK, EFS and IK played the leading role in collecting the data and drafting of the manuscript. PF, GO and DB were involved in the investigation of the histological

418

samples. KA investigated the kidney biopsies and played an important role in drafting of the manuscript.

REFERENCES 1 Jennette JC, Falk RJ, Andrassy K et al. (1994) Nomenclature of systemic vasculitides. Proposal of an international consensus conference. Arthritis Rheum 37, 187–192. 2 Hogan SL, Falk RJ, Chin H et al. (2005) Predictors of relapse and treatment resistance in antineutrophil cytoplasmic antibody-associated small-vessel vasculitis. Ann Intern Med 143 (9), 621–631. 3 Genereau T, Lortholary O, Royer I, Lhote F, Darras-Joly C, Guillevin L (1997) Digestive manifestations of periarteritis nodosa. Gastroenterol Clin Biol 21 (6–7), 503–510. Les manifestations digestives de la periarterite noueuse. 4 Le Thi Huong D, Wechsler B, Grand d’Esnon A et al. (1988) Severe digestive vasculitis: value of bolus of cyclophosphamide. Gastroenterol Clin Biol 12, 402–404. Vascularites digestives graves: interet des bolus de cyclophosphamide. 5 Levine SM, Hellmann DB, Stone JH (2002) Gastrointestinal involvement in polyarteritis nodosa (1986–2000): presentation and outcomes in 24 patients. Am J Med 112 (5), 386–391. 6 Muller-Ladner U (2001) Vasculitides of the gastrointestinal tract. Best Pract Res Clin Gastroenterol 15 (1), 59–82. 7 Guillevin L, Le Thi Huong D, Godeau P, Jais P, Wechsler B (1988) Clinical findings and prognosis of polyarteritis nodosa and Churg-Strauss angiitis: a study in 165 patients. Br J Rheumatol 27 (4), 258–264. 8 Deniz K, Ozseker HS, Balas S, Akpynar E, Sokmensuer C (2007) Intestinal involvement in Wegener’s granulomatosis. J Gastrointestin Liver Dis 16, 329–331. 9 Yildirim AC, Kocak E, Yildiz P et al. (2010) Multiple intestinal perforation in a patient with Wegener’s granulomatosis: a case report and review of the literature. Gastroenterol Clin Biol 34 (12), 712–715. 10 Tokuda M, Kurata N, Daikuhara H et al. (1989) Small intestinal perforation in Wegener’s granulomatosis. J Rheumatol 16 (4), 547–549. 11 Geraghty J, Mackay IR, Smith DC (1986) Intestinal perforation in Wegener’s granulomatosis. Gut 27, 450–451. 12 McNabb WR, Lennox MS, Wedzicha JA (1982) Small intestinal perforation in Wegener’s granulomatosis. Postgrad Med J 58 (676), 123–125. 13 Storesund B, Gran JT, Koldingsnes W (1998) Severe intestinal involvement in Wegener’s granulomatosis: report of two cases and review of the literature. Br J Rheumatol 37, 387–390. 14 Watts RA, Lane SE, Bentham G, Scott DG (2000) Epidemiology of systemic vasculitis: a ten-year study in the United Kingdom. Arthritis Rheum 43 (2), 414–419. 15 Cartin-Ceba R, Golbin JM, Keogh KA et al. (2012) Rituximab for remission induction and maintenance in



16

17

18

19

20

21

22

23

24

refractory granulomatosis with polyangiitis (Wegener’s): a single-center ten-year experience. Arthritis Rheum 64, 3770–3778. Roubaud-Baudron C, Pagnoux C, Meaux-Ruault N et al. (2012) Rituximab maintenance therapy for granulomatosis with polyangiitis and microscopic polyangiitis. J Rheumatol 39 (1), 125–130. Blanchet MR, Maltby S, Haddon DJ, Merkens H, Zbytnuik L, McNagny KM (2007) CD34 facilitates the development of allergic asthma. Blood 110 (6), 2005–2012. Maltby S, Wohlfarth C, Gold M, Zbytnuik L, Hughes MR, McNagny KM (2010) CD34 is required for infiltration of eosinophils into the colon and pathology associated with DSS-induced ulcerative colitis. Am J Pathol 177 (3), 1244– 1254. Berden AE, Jones RB, Erasmus DD et al. (2012) Tubular lesions predict renal outcome in antineutrophil cytoplasmic antibody-associated glomerulonephritis after rituximab therapy. J Am Soc Nephrol 23 (2), 313–321. Leavitt RY, Fauci AS, Bloch DA et al. (1990) The American College of Rheumatology 1990 criteria for the classification of Wegener’s granulomatosis. Arthritis Rheum 33 (8), 1101–1107. Pagnoux C, Mahr A, Cohen P, Guillevin L (2005) Presentation and outcome of gastrointestinal involvement in systemic necrotizing vasculitides: analysis of 62 patients with polyarteritis nodosa, microscopic polyangiitis, Wegener granulomatosis, Churg-Strauss syndrome, or rheumatoid arthritis-associated vasculitis. Medicine (Baltimore) 84, 115–128. Guillevin L, Lhote F, Gayraud M et al. (1996) Prognostic factors in polyarteritis nodosa and Churg-Strauss syndrome. A prospective study in 342 patients. Medicine (Baltimore) 75 (1), 17–28. Guillevin L, Cohen P, Gayraud M, Lhote F, Jarrousse B, Casassus P (1999) Churg-Strauss syndrome. Clinical study and long-term follow-up of 96 patients. Medicine (Baltimore) 78 (1), 26–37. Mukhtyar C, Lee R, Brown D et al. (2009) Modification and validation of the Birmingham Vasculitis Activity Score (version 3). Ann Rheum Dis 68 (12), 1827–1832.


25 Lopez LR, Schocket AL, Stanford RE, Claman HN, Kohler PF (1980) Gastrointestinal involvement in leukocytoclastic vasculitis and polyarteritis nodosa. J Rheumatol 7 (5), 677–684. 26 Scott DG, Bacon PA, Elliott PJ, Tribe CR, Wallington TB (1982) Systemic vasculitis in a district general hospital 1972–1980: clinical and laboratory features, classification and prognosis of 80 cases. Q J Med 51 (203), 292–311. 27 Pinkney JH, Clarke G, Fairclough PD (1991) Gastrointestinal involvement in Wegener’s granulomatosis. Gastrointest Endosc 37 (3), 411–412. 28 Bailey M, Chapin W, Licht H, Reynolds JC (1998) The effects of vasculitis on the gastrointestinal tract and liver. Gastroenterol Clin North Am 27 (4), 747–782, v–vi. 29 Camilleri M, Pusey CD, Chadwick VS, Rees AJ (1983) Gastrointestinal manifestations of systemic vasculitis. Q J Med 52 (206), 141–149. 30 Vallat M, Vallat JM, Julien J, Leleu JP, Pesme D, Leger H (1977) Periarteritis nodosa with fatal liver and gallbladder involvement. Arteriographic aspects. Sem Hop 53 (36), 1953–1954. Periarterite noueuse a manifestation hepatovesiculaire mortelle. Aspects arteriographiques. 31 Desbazeille F, Soule JC (1986) Digestive manifestations of vasculitis. Gastroenterol Clin Biol 10, 405–414. Manifestations digestives des vascularites. 32 Travers RL, Allison DJ, Brettle RP, Hughes GR (1979) Polyarteritis nodosa: a clinical and angiographic analysis of 17 cases. Semin Arthritis Rheum 8 (3), 184–199. 33 Cacoub P, Le TH, Guillevin L, Godeau P (1988) Causes of death in systemic vasculitis of polyarteritis nodosa. Analysis of a series of 165 patients. Ann Med Interne (Paris) 139 (6), 381–390. Les causes de mortalite au cours des vascularites systemiques du groupe de la periarterite noueuse. Analyse d’une serie de 165 malades. 34 Schmitt WH, Hagen EC, Neumann I, Nowack R, FloresSuarez LF, van der Woude FJ (2004) Treatment of refractory Wegener’s granulomatosis with antithymocyte globulin (ATG): an open study in 15 patients. Kidney Int 65 (4), 1440–1448.

419