ORIGINAL ARTICLE

Wegener’s granulomatosis: experience from a Braziliantertiary center

Fernando Henrique Carlos de Souza & Ari Stiel Radu Halpern &

Carmen Silvia Valente Barbas & Samuel Katsuyuki Shinjo

Received: 13 December 2009 /Revised: 5 February 2010 /Accepted: 12 February 2010 /Published online: 1 March 2010# Clinical Rheumatology 2010

Abstract Most epidemiological studies with Wegener’sgranulomatosis (WG) patients are based on populationsfrom the Northern hemisphere, whereas very few studieshave been conducted in Southern hemisphere populations,particularly from South America. The authors performed alarge retrospective, demographic study including clinicaland laboratory profiles of 134 consecutive WG patientsseen at one Brazilian center from 1999 to 2009. Mean ageat initial WG diagnosis was 43.4±15.5 years, and meandisease duration was 8.6±6.6 years. Sixty-four (47.8%)patients were male and a total of 113 (84.3%) subjects werewhite. Ear/nose/throat involvement occurred in 85.8%. Theclassic lung and renal involvement were observed in 77.6%and 75.4%, respectively, followed by ocular (35.8%),musculoskeletal (33.4%), cutaneous (29.1%), neurological(20.1%), cardiac (11.2%), and genitourinary involvement in2.2% of cases. Cytoplasmic pattern–antineutrophil cyto-plasmic antibody was detected in 83 (61.9%) cases. Ten(7.5%) individuals presented limited forms of WG. Classictherapy with corticosteroids and cyclophosphamide wasused in 97 cases (72.4%). There were no cases oftuberculosis or Pneumocystis jiroveci pneumonia, butcutaneous herpes zoster occurred in eight (6.0%) individ-uals. There were 29 deaths (21.6%). Eighteen patients diedof septic shock (mainly bacterial pneumonia), whereas four

died of alveolar hemorrhage, four of myocardial infarction,and three of other causes. In summary, our data from a verylarge retrospective and descriptive study mirrored the mainclinical features of WG described in other countries,demonstrating that they may serve as a reference for SouthAmerican populations.

Keywords Clinical features . Epidemiological studies .

Southern hemisphere . Systemic vasculitis .Wegener’sgranulomatosis

Introduction

Wegener’s granulomatosis (WG) is a rare systemic diseasecharacterized by necrotizing granulomatous vasculitis,affecting mainly the upper airways, lungs, and kidneys [1].

Most studies in the literature are based on patients fromthe Northern hemisphere [1–13]. There is very limited datain the literature concerning WG in the Southern hemisphere[14–16], specifically in South America. Only two smallstudies, one from Argentina and one from Chile, have beenpublished [15, 16]. Due to insufficient clinical dataconcerning GW in South America, we conducted a largeretrospective, descriptive study of Brazilian WG cases.

Materials and methods

The present work is a single center retrospective study.From January 1999 to January 2009, 134 consecutive WGpatients were followed at the Division of Rheumatologyand Pulmonary at the Hospital das Clínicas da Faculdade deMedicina da Universidade de São Paulo, São Paulo, Brazil.The follow-up period was 3 years (after the diagnosis of

F. H. C. de Souza :A. S. Radu Halpern : S. K. Shinjo (*)Division of Rheumatology, Faculdade de Medicina daUniversidade de São Paulo,Av. Dr. Arnaldo, 455, 3° andar, Sala 3190,CEP 01246-903 São Paulo, Brazile-mail: samuel.shinjo@gmail.com

C. S. Valente BarbasDivision of Pulmonary, Faculdade de Medicina da Universidadede São Paulo,São Paulo, Brazil

Clin Rheumatol (2010) 29:855–860DOI 10.1007/s10067-010-1408-4

WG) for each patient analyzed, although this reviewcovered a 10-year period altogether. All patients fulfilledat least two of the five modified American College ofRheumatology (ACR) criteria for the classification of WG(in the modified ACR criteria, a positive serum enzymeimmunoassay for antibodies against proteinase 3 (ELISA)was added to the original four criteria) [17, 18].

Diagnosis of a limited form of WG was established incases with symptoms limited to one organ system, andwithout immediate threat to a vital organ or to life [18].

The following clinical parameters were retrospectivelyextracted from the patients’ medical records in the first3 years of follow-up after initial diagnosis of WG:

– Ear/nose/throat (ENT) involvement (as defined by thepresence of rhinorrhea, nasal crusting, epistaxis, oral ornasopharyngeal ulceration, chronic sinusitis, otitismedia, saddle nose deformity, mastoiditis, hearing loss,and subglottic stenosis)

– Pulmonary involvement (hemoptysis, alveolar hemor-rhage, and respiratory failure, or confirmation ofabnormal thoracic imaging in the absence of concom-itant infection, pulmonary infiltrates and/or nodules, orcavitations)

– Renal involvement (serum creatinine > 1.8 mg/dL,hematuria, or red cell casts in the urinary sediment orproteinuria >0.5 mg/day)

– Articular involvement (arthralgias or arthritis)– Cutaneous involvement (petechiae or purpura, cutane-

ous vasculitis, and skin ulcers)– Ophthalmologic involvement (episcleritis, scleritis,

pseudotumor, amaurosis)– Neurologic involvement (peripheral neuritis, cranial

neuropathy, or central nervous system vasculitis)– Cardiac, gastrointestinal involvement, and genitouri-

nary involvement

An organ or system was diagnosed as positive when abiopsy contained granulomatous inflammation or vasculitis,or in the case of the kidneys, segmental glomerular necrosisor pauci-immune extracapillary glomerulonephritis.

The extent of organ damage secondary to WG wasassessed by means of the Vasculitis Damage Index (VDI)[19], with a score ranging from 0 to 64 (higher scoresindicating more severe damage).

Past history or new development of systemic arterialhypertension, dialysis-dependent end-stage renal failure,diabetes mellitus, myocardial infarction, thyroiditis, stroke,and deep venous thrombosis were also reported. We alsolooked for infectious events such as tuberculosis, herpeszoster, and Pneumocystis jiroveci pneumonia.

Statistical analysis Continuous variables were expressed asmean and standard deviations (SD) or as percentages.

Results

One hundred thirty-four consecutive WG patients wereevaluated. Mean age ± SD at initial diagnosis was43.4±15.5 years (range 8–78 years). Mean disease durationwas 8.6±6.6 years (range 1–19 years). Sixty-four (47.8%)patients were male and 113 (84.3%) were white.

The clinical and laboratory features of these patients areshown in Table 1.

Constitutional symptoms such as fever and/or weightloss at initial presentation were observed in 64.9% of cases.ENT involvement was reported in 85.8% while the otherclassic lung and renal involvement were observed in 77.6%and 75.4% of cases, respectively. Less frequently, ocular(35.8%), articular (33.4%), cutaneous (29.1%), neurological(20.1%), cardiac (11.2%), and genitourinary (2.2%) in-volvement were also reported.

Antineutrophil cytoplasmic antibody (ANCA) wasdetected in 100 patients (68.6%). Eighty-three patients(61.9%) had a cytoplasmatic staining pattern (c-ANCA),whereas 17 (12.9%) had a perinuclear (p-ANCA) pattern.

VDI performed retrospectively showed a mean score of7.9±3.4 (range 2 to 15).

Ten patients (7.5%) were diagnosed as limited WG. Themajority of them (five cases) had the disease limited to theupper airways. Three patients had ocular disease (pseudotu-mor) while one case of isolated lung involvement (pulmonarynodules) and one case of granulomatous vasculitis limited tothe genitourinary tract were also observed.

Organ systems involved over the study period are shownin Table 2 and compared with their reported involvement inother studies.

The classic therapeutic regimen with simultaneouscorticosteroid and cyclophosphamide treatment was usedin 97 (72.4%) cases. Cyclophosphamide was used either inan oral daily schedule of 2 mg/kg or as monthlyintravenous pulse therapy, with doses ranging between 0.5and 1.0 g/m2. The choice between oral and parenteraltreatment was based on access to medication and patientadherence to treatment. Oral prednisone, starting with1 mg/kg/day, was maintained for at least 2 months beforegradual reduction for the next 6 months. In cases ofimmediate threat to either a critical organ or to the patient’slife, corticosteroid was given as intravenous pulse therapy(methyl prednisolone 1 g+saline solution 0.9% 500 mL in4 h, once day, three consecutive days) followed by the oralregimen described above. Other immunosuppressive agentssuch as methotrexate, azathioprine, and mycophenolatemofetil were used in selected cases due to intolerance tocyclophosphamide. Eighteen (13.4%) refractory patientsreceived additional intravenous human immunoglobulin(2 g/kg). Cotrimoxazol (800 + 160 mg, twice per day, oral)was used in 82 patients (61.2%).

856 Clin Rheumatol (2010) 29:855–860

Seventy patients (52.2%) had arterial hypertension, 26(19.4%) had dialysis-dependent end-stage renal failure, and14 (10.4%) had diabetes mellitus.

Over the follow up period, eight patients (6.0%)developed myocardial infarction, seven (5.2%) new strokes,and three (2.2%) had deep venous thrombosis. Eight (6.0%)new cases of thyroid disease were also detected.

There were no cases of P. jiroveci pneumonia ortuberculosis, whereas cutaneous herpes zoster occurred ineight (6.0%) individuals. Neoplasm was reported in threecases (a melanoma, an esophageal, and a breast cancer) butwere all detected prior to WG diagnosis.

Twenty-nine (21.6%) patients died within 3 years afterWG diagnosis. Almost half (44.8%) died in less than 1 yearof follow-up. Eighteen patients died due to septic shock(mainly bacterial pneumonia), four died as a consequenceof diffuse alveolar hemorrhage, and four more due tomyocardial infarction.

Discussion

To our knowledge, this study is the first large series of WGfrom South America.

Different studies published in the literature have differentfollow-up periods resulting in some clinical discrepancybetween them. Some investigators included limited forms ofWG (with variable definitions for limited WG) in theiranalysis while others evaluated only the generalized form ofdisease. Nevertheless, compared to other studies, our patientspresented similar demographic and clinical manifestations. [1,6, 8, 9, 12–16], with some exceptions as discussed below.

In this study, there were significantly high rates of eventsof arterial thromboembolic events when compared with theWGET cohort study [20] or the meta-analysis of 349literary case reports [21]. Our high rates of events ofmyocardial infarction and stroke (ten times the reportedprevalence in the WGET study) was similar to that foundby Faurschou et al. [22] who also showed significantincreased risk for myocardial infarction in WG. Conversely,we found a lower prevalence of venous thrombosis [20, 23,24]. In the WGET trial, venous thrombosis occurred earlyafter diagnosis suggesting that deep venous thrombosis inthese patients is in fact a manifestation of disease activity.

End-stage renal disease has been associated as apotential marker of mortality [25–31]. This was found inapproximately 20% of our patients, in accordance withmost studies.

The concomitant use of cotrimoxazol may explain thefact that we did not observe a single case of P. jirovecipneumonia, in contrast to several reports that emphasizedthe high P. jiroveci pneumonia burden in WG patients [1,32–34]. There were no cases of tuberculosis, even though itis a very prevalent disease in our country.

Although the mortality rate in our study is comparable tothat of others [1, 14, 15], we noticed a higher infection(sepsis) rate [1] but in conformity with Reinhold-Keller etal. [9]. Moreover, the following up was 3 years in our study,whereas Hoffman et al. [1] and Reinhold-Keller [9]analyzing 8 and 7 years, respectively.

Table 1 Clinical and laboratorial features over the entire diseasecourse in 134 cases of Wegener’s granulomatosis

Constitutional symptoms (%) 87 (64.9)

ENT involvement (%) 115 (85.8)

Nasal crusting or epistaxis (%) 111 (82.8)

Chronic sinusitis (%) 97 (72.4)

Rhinorrhea (%) 72 (53.7)

Hearing loss (%) 33 (24.6)

Nasopharyngeal or oral ulcers (%) 32 (23.9)

Otitis media (%) 23 (17.2)

Saddle nose deformity (%) 18 (13.4)

Mastoiditis (%) 15 (11.2)

Subglottic stenosis (%) 15 (11.2)

Pulmonary involvement (%) 104 (77.6)

Hemoptysis (%) 54 (40.3)

Pulmonary nodules (%) 53 (39.6)

Pulmonary infiltrate (%) 43 (32.1)

Cavitations (%) 23 (17.3)

Alveolar hemorrhage (%) 20 (14.9)

Respiratory failure (%) 19 (14.2)

Renal involvement (%) 101 (75.4)

Hematuria or red cell casts in the urinary sediment (%) 72 (53.7)

Proteinuria>0.5 mg/day (%) 49 (36.6)

Initial creatinine (SD) mg/dL 2.7 (3.3)

Ophthalmologic involvement (%) 48 (35.8)

Episcleritis (%) 16 (11.9)

Pseudotumor (%) 14 (10.4)

Amaurosis (%) 11 (8.2)

Scleritis (%) 7 (5.2)

Joint involvement (%) 42 (33.4)

Cutaneous involvement (%) 39 (29.1)

Petechiae or purpura (%) 27 (20.1)

Cutaneous vasculitis (%) 13 (9.7)

Skin ulcers (%) 12 (9.0)

Gastrointestinal tract involvement (%) 11 (20.4)

Intestinal perforation (%) 8 (6.0)

Intestinal bleeding (%) 3 (2.2)

Neurological involvement (%) 27 (20.1)

Peripheral (%) 19 (14.2)

Central (%) 8 (5.9)

Cardiac involvement (%) 15 (11.2)

Genitourinary involvement (%) 3 (2.2)

ENT ear, nose, throat, SD standard deviation

Clin Rheumatol (2010) 29:855–860 857

Tab

le2

Characteristicsof

our13

4patientswith

Wegener’sgranulom

atosiscomparedto

otherWG

stud

ies

Present

stud

yHoffm

anet

al.

And

ersonet

al.

Mattesonet

al.

Reinh

old-Kelleret

al.

Cisternas

etal.

Gam

ronet

al.

Gibsonet

al.

Brazil

USA

UK

USA

Germany

Chile

Argentin

aAustralia

2009

1992

1992

1996

2000

2005

2006

2006

No.

ofpatients(n)

134

258

265

7715

558

1673

Study

type

Mon

ocenter

Multicenter

Multicenter

Multicenter

Mon

ocenter

Mon

ocenter

Mon

ocenter

Mon

ocenter

Male/female(%

)47

.8:52.2

50:50

55:45

64:36

49:51

43:57

56:44

55:45

Age

atdiagno

sis(years)

43.4

4150

4550

.852

66

Follow-up(years)

38

7.1

71.7

105

“Initialph

ase”

(localized

WG)at

diagno

sis(%

)10

2215

Organ

system

invo

lved

Ear/nose/throat

(%)

85.8

9273

9978

93.7

86

Pulmon

aryinvo

lvem

ent(%

)77

.685

7066

5052

Renal

invo

lvem

ent(%

)75

.477

8366

6280

78

Oph

thalmolog

ical

invo

lvem

ent(%

)35

.852

1461

43.7

Articular

invo

lvem

ent(%

)33

.467

2560

.2

Cutaneous

invo

lvem

ent(%

)29

.146

40

Gastrointestin

altract(%

)20

.411

10

Peripheralnervou

ssystem

(%)

14.2

1533

Cardiac

invo

lvem

ent(%

)11.2

86

1714

Central

nervou

ssystem

(%)

5.9

819

7720

c-ANCA

positiv

e(%

)61

.988

8479

68.7

82

Treatment

Cycloph

osph

amidetherapy(%

)94

.889

2960

92

Corticosteroid

therapy(%

)76

.170

70

Mortalityrate

(%)

21.6

2056

3614

2326

Death

dueto

infection(%

)13

.43

12

Death

from

WG

and/or

itstherapy(%

)8.2

1312

103

WG

Wegener’sgranulom

atosis,c-ANCAcytoplasmatic

staining

pattern–antineutrop

hilic

cytoplasmatic

antib

ody

858 Clin Rheumatol (2010) 29:855–860

The pathophysiology of WG is unknown. A possibleassociation with environmental and genetic factors has beenconsistently highlighted in the literature, as in a recentpaper from Gatenby correlating WG with ultravioletradiation [35]. As in other autoimmune diseases, theclinical spectrum of WG may be very heterogeneous. Itwould be useful to know if WG has a distinct phenotypeunder different environmental conditions such as thosefound in the tropics as compared to higher latitudes. Ourstudy did not assess the rates of events of WG in Brazil;however, it provided an overview of WG features in ourgenetically distinct population which is probably exposedto diverse environmental triggers.

In summary, our data are the first large series of WGpatients in South America. Except for a higher incidence ofthromboembolic events, we found a similar spectrum ofclinical features of WG to those described in othercountries, demonstrating that these studies may serve as areference for South American populations.

Disclosures None.

References

1. Hoffman GS, Kerr GS, Leavitt RY et al (1992) Wegenergranulomatosis: an analysis of 158 patients. Ann Intern Med116:488–498

2. Koldingsnes W, Nossent H (2000) Epidemiology of Wegener’sgranulomatosis in northern Norway. Arthritis Rheum 43:2481–2487

3. Watts RA, Scott DG (1997) Classification and epidemiology ofthe vasculitides. Baillieres Clin Rheumatol 11:191–217

4. Watts R, Gonzales-Gay M, Garcia-Porrua C et al (2000) ANCA-associated vasculitis in two European regions. Clin Exp Immunol120(Suppl 1):60

5. Gonzalez-Gay MA, Garcia-Porrua C, Guerrero J (2003) Theepidemiology of the primary systemic vasculitides in northwestSpain: implications of the Chapel Hill Consensus Conferencedefinitions. Arthritis Rheum 49:388–393

6. Reinhold-Keller E, Zeidler A, Mock C et al (1996) Epidemiologyof primary systemic vasculitis in North and South Germany[abstract]. Sarcoidosis 13(Suppl 3):272

7. Bligny D, Mahr A, Toumelin PL et al (2004) Predicting mortalityin systemic Wegener’s granulomatosis: a survival analysis basedon 93 patients. Arthritis Rheum 51:83–91

8. Anderson G, Coles ET, Crane M et al (1992) Wegener’sgranuloma. A series of 265 British cases seen between 1975 and1985. A report by a sub-committee of the British Thoracic SocietyResearch Committee. Q J Med 83:427–438

9. Reinhold-Keller E, Beuge N, Latza U et al (2000) An interdisci-plinary approach to the care of patients with Wegener's granulo-matosis: long-term outcome in 155 patients. Arthritis Rheum43:1021–1032

10. Jennings CR, Jones NS, Dugar J et al (1998) Wegener’sgranulomatosis—a review of diagnosis and treatment in 53subjects. Rhinology 36:188–191

11. Fauci AS, Woff SM (1994) Wegener’s granulomatosis: studies ineighteen patients and a review of the literature. Medicine[Baltimore] 73:315–324

12. Matteson EL, Gold KN, Bloch DA (1996) Long-term survival ofpatients with Wegener’s granulomatosis from the AmericanCollege of Rheumatology Wegener’s granulomatosis classificationcriteria cohort. Am J Med 101:129–134

13. Westman KW, Bygren PG, Olsson H et al (1998) Relapse rate,renal survival and cancer morbidity in patients with Wegener’sgranulomatosis or microscopic polyangiitis with renal involve-ment. J Am Soc Nephrol 9:842–852

14. Gibson A, Stamp LK, Chapman PT, O’Donnell JL (2006) Theepidemiology of Wegener’s granulomatosis and microscopicpolyangiitis in a Southern Hemisphere region. Rheumatology45:624–628

15. Cisternas ML, Soto L, Jacobelli S et al (2005) Característicasclínicas de granulomatosis de Wegener y poliangeítis micro-scópica em pacientes chilenos. Rev Med Chil 133:273–278

16. Gamron S, Eugenia Muscellini M, Onetti L et al (2006) Wegener'sgranulomatosis: its prevalence in a ten-year period in therheumatology service of the Clinic Hospital, Cordoba, Argentina.Rev Fac Cien Med Univ Nasc Cordoba 63:53–56

17. Leavitt RY, Fauci AS, Bloch DA et al (1990) The AmericanCollege of Rheumatology 1990 criteria for the classification ofWegener’s granulomatosis. Arthritis Rheum 33:1101–1107

18. The WGET Research Group (2002) Design of the Wegener’sGranulomatosis Etanercept Trial (WGET). Control Clin Trials23:450–468

19. Exley AR, Bacon PA, Luqmani RA et al (1998) Examination ofdisease severity in systemic vasculitis from the novel perspectiveof damage using the vasculitis damage index (VDI). Br JRheumatol 37:57–63

20. Seo P, Min YI, Holbrook JT et al (2005) Damage caused byWegener's granulomatosis and its treatment: prospective data fromthe Wegener's Granulomatosis Etanercept Trial (WGET). ArthritisRheum 52:2168–2178

21. Bajema IM, Hagen EC, van der Woude FJ et al (1997) Wegener'sgranulomatosis: a meta-analysis of 349 literary case reports. J LabClin Med 129:17–22

22. Faurschou M, Mellemkjaer L, Sorensen IJ, Thomsen BS, DreyerL, Baslund B (2009) Increased morbidity from ischemic heartdisease in patients with Wegener's granulomatosis. ArthritisRheum 60:1187–1189

23. Merkel PA, Lo GH, Holbrook JT et al (2005) Brief communication:high incidence of venous thrombotic events among patients withWegener granulomatosis: The Wegener’s Clinical Occurrence ofThrombosis (WeCLOT) study. Ann Intern Med 142:620–626

24. Allenbach Y, Seror R, Pagnox C et al (2009) High frequency ofvenous thromboembolic events in Churg–Strauss syndrome,Wegener's granulomatosis and microscopic polyangiitis but notpolyarteritis nodosa: a systematic retrospective study on 1130patients. Ann Rheum Dis 68:564–567

25. Aasarod K, Inversen BM, Hammerstrom J et al (2000) Wegener'sgranulomatosis: clinical course in 108 patients with renalinvolvement. Nephrol Dial Transplant 15:611–618

26. Krafcik SS, Covin RB, Lynch JP et al (2006) Wegener’sgranulomatosis in the elderly. Chest 109:430–437

27. Le Thi Huong DU, Wechsler B, Cabane J et al (1988) Wegener’sgranulomatosis: clinical aspects, nosologic problems: review ofthe literature a propos of 30 cases. Ann Med Interne 139:169–182

28. Vassallo M, Shepherd RJ, Iqbal P et al (1997) Age-relatedvariations in presentation and outcome in Wegener’s granuloma-tosis. J R Coll Physicians 31:396–400

29. Fauchais AL, Michon-Pasturel M, Rugale C et al (2001)Wegener’s granulomatosis in the elderly patient [in French]. RevMed Interne 22:127–131

Clin Rheumatol (2010) 29:855–860 859

30. Romas E, Murphy BF, d’Apice Aj et al (1993) Wegener’sgranulomatosis: clinical features and prognosis in 37 patients.Aust NZ J Med 23:168–175

31. Zycinska K, Wardyn KA, Tyszko P et al (2007) Analysis of earlydeath based on the prediction model in Wegener's granulomatosiswith pulmonary and renal involvement. J Physiol Pharmacol 58(Suppl 5):829–837

32. Jarrousse B, Guillevin L, Bindi P et al (1993) Increased risk ofPneumocystis carinii pneumonia in patients with Wegener'sgranulomatosis. Clin Exp Rheumatol 11:615–621

33. Godeau B, Coutant-Perronne V, Du Huong LT et al (1994)Pneumocystis carinii pneumonia in the course of connective tissuedisease: report of 34 cases. J Rheumatol 21:246–251

34. Godeau B, Mainardi JL, Roudot-Thoraval F et al (1995) Factorsassociated with Pneumocystis carinii pneumonia in Wegener'sgranulomatosis. Ann Rheum Dis 54:991–994

35. Gatenby PA, Lucas RM, Engelsen O et al (2009) Antineutrophilcytoplasmic antibody-associated vasculitides: could geographicpatterns be explained by ambient ultraviolet radiation? ArthritisRheum 61:1417–1424

860 Clin Rheumatol (2010) 29:855–860