Outcome of ANCA-Associated Renal Vasculitis:A 5-Year Retrospective Study

Anthony D. Booth, MD, Mike K. Almond, MD, Aine Burns, MD, Peter Ellis, MA, Gill Gaskin, MD,Guy H. Neild, MD, Martin Plaisance, MD, Charles D. Pusey, MD, and David R.W. Jayne, MD,

for the Pan-Thames Renal Research Group

● Background: Renal involvement is frequently present in antineutrophil cytoplasmic autoantibody (ANCA)-associated systemic vasculitis and is an important cause of end-stage renal failure (ESRF). Methods: Thisretrospective, multicenter, sequential cohort study reports presenting features and outcome of 246 new patientsdiagnosed in London, UK, between 1995 and 2000. Results: Diagnostic subgroups were microscopic polyangiitis,120 patients (49%); Wegener’s granulomatosis (WG), 82 patients (33%); renal-limited vasculitis, 33 patients (13.5%);and Churg-Strauss angiitis, 11 patients (4.5%). Median age was 66 years, 57% were men, and median creatinine levelat presentation was 3.87 mg/dL (342 �mol/L). ANCA was present in 92%. Cumulative patient survival at 1 and 5 yearswas 82% and 76%, respectively. Mortality was associated with age older than 60 years (P < 0.001), development ofESRF (P < 0.001), initial creatinine level greater than 2.26 mg/dL (200 �mol/L; P � 0.01), and sepsis (P < 0.048).ESRF occurred in 68 patients (28%), of whom 47% died. Fifty-six patients who presented with a creatinine levelgreater than 5.65 mg/dL (500 �mol/L) survived, and 31 patients (55%) achieved dialysis independence. Relapseoccurred in 34% after a median of 13 months and was more common in patients with WG (P � 0.048) and proteinase3-ANCA (P � 0.034). Leukopenia occurred in 41% and was associated with sepsis (P < 0.001). Conclusion: Mortalityand morbidity of ANCA-associated systemic vasculitis are improving compared with previous series, but remainhigh. Renal vasculitis often affects older patients, who have a particularly poor outcome. Early diagnosis improvesoutcome. Leukopenia, caused by immunosuppressive therapy, should be avoided because of the close associationwith sepsis and death. Am J Kidney Dis 41:776-784.© 2003 by the National Kidney Foundation, Inc.

INDEX WORDS: Antineutrophil cytoplasmic autoantibody (ANCA); renal vasculitis; outcome.

THE PRIMARY small-vessel vasculitides area group of life-threatening diseases classi-

fied according to their clinical and pathologicalfeatures. The largest subgroup is associated withantineutrophil cytoplasmic autoantibodies(ANCAs). This includes Wegener’s granuloma-tosis (WG), microscopic polyangiitis (MPA),Churg-Strauss angiitis (CSA), and renal-lim-ited vasculitis (RLV). ANCAs most often aredirected against proteinase 3 (PR3) or myelo-peroxidase (MPO) and are present in morethan 90% of ANCA-associated systemic vascu-litis with renal involvement. Several experimen-tal studies point to their role in pathogenesis.1-3

Renal vasculitis is the most common severemanifestation of the primary small-vessel vascu-litides, occurring in more than 50% at presenta-tion, but in 70% to 85% of patients with WG,MPA, or RLV during the course of their disease.The hallmark of renal injury is a focal necrotiz-ing glomerulonephritis, usually associated withcrescent formation and renal impairment, butwith few or absent immune deposits.

The annual incidence of renal vasculitis inEurope is 10 to 20 per million per year.4 There isan increased incidence with age, peaking in thoseaged 65 to 74 years. Geographic and ethnicdifferences in disease pattern and incidence havebeen suggested, but lack confirmation.5,6

The cause of ANCA-associated systemic vas-culitis remains unknown, but likely is multifacto-rial. A number of familial cases have been de-scribed, and suggested candidate genes includecytotoxic T-lymphocyte–associated protein 4(CTLA-4), interleukin-1 receptor antagonist (IL-1ra), IL-10, and Fc�RII/Fc�RIII.7-16 Environmen-tal associations include silica exposure and or-ganic solvents.17-21 Some reports suggested agreater rate of ANCA-associated systemic vascu-litis in winter months, but no clear infectious linkhas been made.22-25 Certain drugs, including pro-

From the Department of Medicine, Addenbrookes’s Hospi-tal, Cambs; Southend Hospital, Essex; Centre for Nephrol-ogy, Royal Free and University College Medical School,University College London; Kings College Hospital; Divi-sion of Medicine, Imperial College, Hammersmith Hospital;Middlesex Hospital; and Guy’s Hospital, London, UK.

Received August 12, 2002; accepted in revised formDecember 3, 2002.

Address reprint requests to Anthony D. Booth, MD, Deptof Medicine, Box 157, Addenbrookes’s Hospital, Cambs CB22QQ, UK. E-mail: tonybooth1@yahoo.co.uk

© 2003 by the National Kidney Foundation, Inc.0272-6386/03/4104-0006$30.00/0

American Journal of Kidney Diseases, Vol 41, No 4 (April), 2003: pp 776-784776

pylthiouracil, penicillamine, and hydralazine, areassociated with the disease.26

Untreated, these conditions have a 1-year mor-tality rate of 80%.27 Immunosuppressive thera-pies generally are effective in reversing mortal-ity, controlling disease activity, rescuing renalfunction, and preventing relapse.28 Treatmenthas changed the outcome of ANCA-associatedsystemic vasculitis to that of a chronic relapsingdisorder with accumulating morbidity and mor-tality related to slow response to standard regi-mens, high relapse rates,29,30 cyclophosphamideand steroid toxicity,31 and accumulating damagefrom disease scars.32 Combination therapy withcyclophosphamide and prednisolone is the cur-rent standard of treatment, although optimaldoses, routes of administration, and duration andthe role of additional plasma exchange remainpoorly defined. The European Vasculitis StudyGroup (EUVAS) has published consensus proto-cols and sought to harmonize therapy throughrandomized controlled trials.33

This study arose out of a need to surveycurrent practice and outcomes outside clinicaltrials, establish standards of care, and serve as abaseline for future studies. Demographics, pre-senting features, and outcomes of a sequentialcohort of 246 patients with ANCA-associatedsystemic vasculitis and renal involvement aredescribed, and the frequency of major treatmenttoxicities is assessed.

METHODS

Participating CentersRetrospective patient data were submitted from seven

London hospitals: Guy’s and St Thomas’s, Hammersmith,Kings, Royal Free, Southend, St Helier (Carshalton), andUniversity College/Middlesex Hospital.

Inclusion CriteriaPatients were included if newly diagnosed with renal

vasculitis between January 1, 1995, and January 1, 2000.Diagnoses were based on the Chapel-Hill disease definitionsand included WG, MPA, RLV, CSA, polyarteritis nodosa,Henoch-Schonlein purpura, cryoglobulinemic vasculitis, andanti–glomerular basement membrane disease.34 Renal in-volvement was defined by the presence of hematuria (�30red blood cells/high-power field) with or without red bloodcell casts, an elevated creatinine level attributable to thedisease, or histological evidence of pauci-immune necrotiz-ing glomerulonephritis. Renal biopsy specimens were avail-able in 91% of patients with ANCA-associated systemicvasculitis.

DefinitionsRemission was defined as the absence of clinical signs or

symptoms or laboratory evidence of vasculitis activity. Re-lapse was defined as the return of clinical signs or symptomsor laboratory evidence of disease activity sufficient to war-rant a sustained increase in immunosuppressive therapy.

Adverse EffectsIn view of the likely underreporting of adverse effects in a

retrospective study, this was limited to the occurrence ofleukopenia (total white cell count � 4.0 � 109/L), sepsisrequiring hospital admission, cardiovascular events, osteopo-rosis and bone fractures, and malignancy.

Statistical Methods and Data CollectionData were combined, validated, and analyzed using the

SPSS personal computer statistical package (version 9; SSPSInc, Chicago, IL). Standardized mortality rate was calcu-lated using age-specific standard mortality rates for Englandand Wales (1998) as follows:

Ratio � (observed/expected deaths) � 100

where expected deaths equal pkMk, Pk is population in agegroup at risk, and Mk is death rate for age group (k) for aparticular year.

Patient survival and relapse were analyzed further usingKaplan-Meier life survival analysis. Times to relapse ordeath were taken as outcome variables. Subject withdrawalor loss to follow-up was censored. Continued survival andremission were taken as end points of the study. Between-group comparisons were performed using log-rank test.Demographic analysis and further analysis of these variableswere undertaken using Fisher’s exact test for categoricalvariables, whereas continuous variables were analyzedusing independent-sample t-test for equality of means. The5% level of confidence was taken to indicate statisticalsignificance.

RESULTS

Demographics

Three hundred thirteen patients with a newdiagnosis of renal vasculitis were registered inthe study. Diagnoses included ANCA-associatedsystemic vasculitis (246 patients), Henoch-Schonlein purpura (25 patients), cryoglobuline-mic vasculitis (7 patients), polyarteritis nodosa(17 patients), and anti–glomerular basementmembrane (18 patients). The 246 patients withANCA-associated systemic vasculitis were sub-sequently analyzed in detail and included MPA,120 patients (49%); WG, 82 patients (33%);RLV, 33 patients (13.5%); and CSA, 11 patients(4.5%). Median age at presentation was 66 years.Those with WG were younger (median, 62 years;95% confidence interval [CI], 54 to 61) than

ANCA-ASSOCIATED RENAL VASCULITIS: 1995-2000 777

those with RLV (median, 71 years; 95% CI, 63 to73; P � 0.002).

Fifty-seven percent were men (odds ratio [OR],1.28; 95% CI, 0.93 to 1.78), and 83% wereCaucasian. Other ethnic groups included His-panic (5%), African or Afro-Caribbean (4%), andAsian (4%). Mean duration of prodromal symp-toms was 4 months (range, 0.5 to 20 months).There was no seasonal variation at presentation.

ANCA was present in 92% at diagnosis: RLV,MPA, and CSA had 65% perinuclear ANCA(P-ANCA) and MPO-ANCA and 25% cytoplas-mic ANCA (C-ANCA) and PR3-ANCA; WG,83% C-ANCA and PR3-ANCA and 12% P-ANCA and MPO-ANCA. One patient wasC-ANCA/PR3-ANCA and P-ANCA/MPO-ANCA positive. Initial creatinine levels werehigher in those with RLV and MPA (P � 0.002)and greater than 2.26 mg/dL (200 �mol/L) in72% with MPA, 90% with RLV, and 58% withWG (Table 1). No significant difference wasnoted between ANCA serological results andcreatinine level at presentation.

Treatment

All patients were administered oral pred-nisolone for 18 to 48 months. Cyclophospha-mide was administered to 214 of 241 patients(88%), with a mean cumulative dose of 10.3 g.Two centers administered cyclophosphamide byintravenous bolus, and the remainder used oraltherapy. Azathioprine was used as initial therapyby two centers, but otherwise was introduced at 3months or at the time of remission. Azathioprinetherapy was continued in 75% for a median of 9months and, in one center, continued indefinitely.Forty patients underwent plasma exchange, withno consistent indications between centers. Othertherapies included co-trimoxazole (12 patients),

methotrexate (4 patients), mycophenolate mofetil(11 patients), intravenous immunoglobulin (11patients), tacrolimus (4 patients), and immunoad-sorption (1 patient). Five patients received renaltransplants, and all continue to have independentrenal function.

Survival

Cumulative actuarial patient survival rates at 1and 5 years were 84% and 76%, respectively.Median time of follow-up in survivors was 3.1years. Using data calculated from the UK Na-tional Statistics Series,35 death rates were ana-lyzed according to age at diagnosis (Table 2).The standardized mortality ratio was 284% com-pared with the local population, and OR was 3.56(95% CI, 1.56 to 7.12).

The Kaplan-Meier survival curve (Fig 1) im-plied two phases of disease: an early steep de-crease in survival was associated with diseaseseverity at presentation and initial therapy. Be-yond 6 months, the mortality gradient approachednormal rates. Survival was worse in those olderthan 60 years (P � 0.0002; Fig 2). Presentingcreatinine level was associated with increasedmortality (P � 0.01) and risk for end-stage renalfailure (ESRF; P � 0.01; Fig 3). Cumulativesurvival rates at 1 and 5 years for those withESRF were 64% and 53%, respectively; meansurvival with ESRF was 3 years (95% CI, 30.2 to42.9 months), which was considerably worsethan for those who remained dialysis indepen-dent (P � 0.001), of whom 88% were alive at 5years (Fig 4).

Compared with all causes of ESRF, UK RenalRegistry data (RR),36 90-day survival of patientswith vasculitis was worse (P � 0.004); but at 1and 2 years, there was no difference (Fig 5).There were no significant associations with sex

Table 1. Mean Presenting Creatinine Level and ANCA Serological Results According to ANCA-AssociatedSystemic Vasculitis Subtypes

DiagnosisMean Creatinine

(mg/dL)Mean Creatinine for

ANCA-Positive (mg/dL)Mean Creatinine for

ANCA-Negative (mg/dL)

MPA 5.33 � 119 5.45 � 108 4.20 � 11RLV 7.44 � 27 7.86 � 23 5.02 � 4CSA 1.36 � 10 1.38 � 9 1.13 � 1WG 4.32 � 75 4.20 � 73 5.81 � 2Total 5.08 � 231 5.14 � 213 4.39 � 18

NOTE. To convert serum creatinine in mg/dL to �mol/L, multiply by 88.4.

BOOTH ET AL778

or the presence or ANCA subtype. AlthoughCSA was associated with greater survival (P �0.004), numbers are small (n � 11) comparedwith other subgroups (Fig 6). There was nodifference in survival between WG, RLV, andMPA (P � 0.24).

ESRF

Sixty-eight patients (28%) developed ESRF.ESRF was more common among surviving pa-tients who had presented with a creatinine levelgreater than 5.65 mg/dL (500 �mol/L; n � 56;P � 0.001), of whom 23 of 56 patients (45%)developed ESRF and 31 of 56 patients (55%)recovered independent renal function. There wereno other significant laboratory or clinical predic-tors of renal survival. Plasma exchange was usedin 40 of the 77 patients presenting with a creati-nine level greater than 5.65 mg/dL (500 �mol/L). There was no reduction in rate of progression

to ESRF; however, indications were variable,and it was often used as second-line therapy inrefractory patients.

Relapse

Relapse occurred in 34% at a mean of 13months (95% CI, 7.56 to 18.29) and was morecommon in those with WG (P � 0.048) andC-ANCA/PR3-ANCA specificity (P � 0.034;Fig 7). Relapse was not associated with otherdemographic features.

Adverse Events

Leukopenia and infection were the most com-mon adverse events (Table 3; Fig 8). They wereclosely related (P � 0.0005), and the presence ofsepsis increased the risk for death (P � 0.048;Fig 9). Cardiovascular events occurred in 23patients (9.5%) and were associated with ESRF(P � 0.005) and a diagnosis of WG (P � 0.018).

Fig 2. Survival and age. P value <0.001.

Table 2. Relationship Between Age at Presentation and Death Rate Using the UK National Statistics Series(Southeast Area)

Age Group(y)

Annual Death Rate(/1,000 Population)

Total PatientYears at Risk

PatientDeaths

Patient AnnualDeath Rate (/1,000)

15-24 0.45 20 0 025-34 0.6 52 1 25.535-44 1.1 40 1 19.345-54 2.8 96 3 31.355-64 7.55 173 8 46.265-74 22.25 337 23 68.275-84 62.8 265 23 86.8Total 97.55 893 59 277.3

NOTE. Calculated standardized mortality ratio, 284%; odds ratio, 3.58; 95% CI, 1.56 to 7.12.

Fig 1. Patient survival.

ANCA-ASSOCIATED RENAL VASCULITIS: 1995-2000 779

Recording of osteoporosis was unreliable be-cause of variable use of bone densitometry; how-ever, bone fractures were recorded in 11 patients(4.5%). Eleven patients developed a malignancy(4.5%), with no consistent tumor type or associa-tion with cumulative exposure to cyclophospha-mide (Table 4).

DISCUSSION

Since the association of ANCA with WG andMPA in the mid-1980s,37 many smaller studieshave reported outcome data on primary small-vessel vasculitis. This multicenter study aims toreport the major outcomes of ANCA-associatedsystemic vasculitis with renal involvement andassess predictors of outcome at presentation.

Demographic data in our study highlight theolder age (median, 66 years) at presentation ofrenal vasculitis in this cohort compared withprevious studies, particularly in RLV, presum-ably reflecting its more indolent course.29,38-43

This raises the possibility of historical underdiag-nosis in the elderly. Consistent with previousreports, there was no significant sex difference.No firm conclusions could be made from theethnic distribution, although Africans and Afro-Caribbeans may be underrepresented.

The rate of ANCA positivity, 93%, was similarto previous studies; the significance of ANCAnegativity remains unclear, but false-negativeassay results may be important.29,42-46 A highercreatinine level at presentation in patients with

Fig 3. Survival and presenting creatinine (Cr) level.P value � 0.01.To convert mg/dL to �mol/L, multiplyby 88.4.

Fig 4. Survival and ESRF. P value <0.001.

Fig 5. ANCA-associated systemic vasculitis andESRF (comparison with the Renal Registry).

Fig 6. Survival and diagnosis. P value � 0.004.

BOOTH ET AL780

RLV and MPA also was found in the recentCyclophosphamide Versus Agathiopine DuringRemission (CYCAZAREM) trial.47 This is notentirely caused by less extrarenal disease andtherefore a later diagnosis because renal biopsyspecimens from patients with WG have moresevere active glomerular and tubular lesions.39-40

The duration of prodromal symptoms beforediagnosis, 4 months, compares with a range of 3to 15 months in other reports.29,43 Because diag-nostic delay is associated with a greater creati-nine level at presentation and a poor outcome,reduction of diagnostic delay is an importantambition for the future. No seasonal variation indiagnosis was shown, but this may reflect differ-ences in methods of reporting the time betweenfirst reported symptoms and final diagnosis ofdisease.22-25

Cumulative 1- and 5-year survival rates were84% and 76%, respectively; this is consistentwith the 2-year survival rate of approximately80% reported elsewhere.38,41-43,45,48,49 Mortalityin the first 6 months was very steep, reflectingthe aggressive course of the disease and of the

early treatment. The standardized mortality rateof 284% emphasizes the impact of renal vasculi-tis on patient survival. The major prognosticdeterminants of survival were age older than 60years and renal impairment at presentation, inline with other work.30,41,43,47,50-52 Treatment-induced leukopenia occurred in 41% of patientsand was strongly associated with sepsis. Further-more, sepsis was a determinant of survival. Itwould appear that avoidance of cyclophospha-mide-associated leukopenia is an important aimfor protocols in the future. Interestingly, ANCAsubtypes and disease category were not prognos-tic survival indicators.41,45,48,50

ESRF developed in 28% of the cohort and hada high mortality rate of 47%. In other series,

Fig 7. Relapse and ANCA status. P value � 0.034.Fig 8. Adverse events: infections. Abbreviations:

HZV, herpes zoster; UTI, urinary tract infection.

Fig 9. Adverse events: survival and sepsis. Pvalue � 0.048.

Table 3. Adverse Events: Leukopenia and Sepsis

Leukopenia

Sepsis

TotalYes No

Yes 46 53 99No 27 117 144Total 73 170 243

NOTE. P � 0.001.

ANCA-ASSOCIATED RENAL VASCULITIS: 1995-2000 781

ESRF has been reported in 9% to 20% of patientsat 1 year.40-42,44,49 ESRF mortality in vasculitiswas greater than for other causes of ESRF, andwe suggest this is caused by extrarenal vasculitisand the toxicity of the therapy. Age and present-ing creatinine level correlated with the develop-ment of ESRF. Renal recovery was found in 31of 56 patients (55%) who presented with a creat-inine level greater than 5.65 mg/dL (500 �mol/L), similar to rates of 22% to 70% in otherseries.40,42,53 Five of our patients with ESRFreceived successful renal transplants; this hasbeen shown by others to be successful in vasculi-tis and to reduce the vasculitis recurrence rate.54

Remission rates were 81% and 94% in survi-vors at 6 months; thus, few survivors failed toachieve remission. This figure is greater than thatin studies restricted to WG, particularly thoseincluding nonrenal disease and previously treatedpatients.29 Relapses occurred in 34% of patientsat a mean of 13 months. The increased relapserate in WG or those with PR3-ANCA againpoints to this subgroup having disease that ismore difficult to control.29,39,42,48,55 Reasons forthis are unclear. Relapse rates in the CYCAZA-REM trial were lower (16% at 18 months),possibly because of prolonged immunosuppres-

sion in all patients, but relapse also was associ-ated with a diagnosis of WG.47

The current standard of treatment for activerenal vasculitis includes steroids and cyclophos-phamide, and known relapse rates argue forprolonged immunosuppression with a safer im-munosuppressive agent, such as azathioprine ormethotrexate.47 In our study, 88% were adminis-tered cyclophosphamide initially, and 75% ofthose who achieved remission then were admin-istered azathioprine. Treatment practices variedbetween centers, with 15 patients administeredsteroids alone. The different indications forplasma exchange among centers did not permitassessment of the role of plasma exchange. Thedemonstrated 20% mortality rate at 1 year and30% to 50% relapse rate indicate the need formore effective better targeted therapy. Newerapproaches under evaluation include blockade oftumor necrosis factor, lymphocyte depletion, andalternative immunosuppressive drugs, such asmycophenolate mofetil and deoxyspergualin.56-59

Predictors of outcome in ANCA-associatedsystemic vasculitis with renal involvement arehighlighted in this study. Systemic vasculitis wasmore common in older age groups. Disease sub-groups and ANCA subtypes were not importantin terms of major outcomes, but were associatedwith subsequent disease course. Outcomes wererelated to age and presenting creatinine level;thus, diagnostic delay may have a major influ-ence on outcome. Increased awareness of ANCA-associated systemic vasculitis, recognition of thepresence of renal involvement at diagnosis, andthe strong correlation with ANCA should shortenthe delay in diagnosis. Urine dipstick analysisprovides a simple bedside test to identify earlyrenal pathological states. Furthermore, the com-bination of microscopic hematuria and a positiveANCA result has up to a 95% positive predictivevalue for the presence of necrotizing glomerulo-nephritis.45

Consensus statements, such as those from EU-VAS, will help standardize existing treatmentoptions and develop better therapeutic regimensto address the toxicity and partial efficacy ofcurrent treatment. Leukopenia should be avoidedbecause of the close association with sepsis anddeath. Careful dosing of cyclophosphamide, tak-ing into consideration patient age, is required.The current study provides the basis for future

Table 4. Adverse Events: Cardiovascular, Bone,and Malignancy

Adverse EventNo. of

PatientsTotal

Events (%)

Cardiovascular 23 9.5Myocardial infarction 9 39Cardiac failure 4 17Aortic valve stenosis or heart block 3 13Pulmonary embolus 1 4Abdominal aortic aneurysm 2 8Cerebrovascular event 2 8Other 2 8

Malignancy 11 4.5Colon 2 18.2Lung 2 18.2Prostate 2 18.2Melanoma 1 9.1Gall bladder 1 9.1Hematological 1 9.1Other 2 18.2

Bone 11 4.5Vertebral fractures 4 36.4Hip fractures 3 27.3Avascular necrosis 1 9.0Osteoporosis 3 27.3

BOOTH ET AL782

clinical trials and standards in the treatment ofsystemic vasculitis.

ACKNOWLEDGMENT

The authors thank Dr F. Compton for generous support andguidance in the statistical analysis of the work presented.

REFERENCES

1. Savage CO, Harper L, Holland M: New findings inpathogenesis of antineutrophil cytoplasm antibody-associ-ated vasculitis. Curr Opin Rheumatol 14:15-22, 2002

2. Heeringa P, Brouwer E, Cohen Tervaert JW, WeeningJJ, Kallenberg CG: Animal models of anti-neutrophil cyto-plasmic antibody associated vasculitis. Kidney Int 53:253-263, 1998

3. Jennette JC, Xiao H, Heeringa P, et al: Induction ofpauci-immune necrotizing and crescentic glomerulonephri-tis (NCGN) by intravenous administration of anti-myeloper-oxidase (anti-MPO) antibodies to recombinase activatinggene-2 deficient (Rag-2�/�) mice. Cleve Clin J Med 69:SII-13, 2002 (abstr)

4. Watts RA, Lane SE, Bentham G, Scott DG: Epidemiol-ogy of systemic vasculitis: A ten-year study in the UnitedKingdom. Arthritis Rheum 43:414-419, 2000

5. Watts RA, Lane SE, Scott DG, et al: Epidemiology ofvasculitis in Europe. Ann Rheum Dis 60:1156-1157, 2001

6. el-Reshaid K, Kapoor MM, el-Reshaid W, Madda JP,Varro J: The spectrum of renal disease associated withmicroscopic polyangiitis and classic polyarteritis nodosa inKuwait. Nephrol Dial Transplant 12:1874-1882, 1997

7. Franssen CF, ter Maaten JC, Hoorntje SJ: Brother andsister with myeloperoxidase associated autoimmune disease.Ann Rheum Dis 53:213, 1994

8. Hull CM, Couser WG, Knostman JD: A familial caseof P-ANCA glomerulonephritis presenting in a father anddaughter. Am J Kidney Dis 35:E23, 2000

9. Nowack R, Lehmann H, Flores-Suarez LF, Nanhou A,van der Woude FJ: Familial occurrence of systemic vasculi-tis and rapidly progressive glomerulonephritis. Am J KidneyDis 34:364-373, 1999

10. Brener Z, Cohen L, Goldberg SJ, Kaufman AM:ANCA-associated vasculitis in Greek siblings with chronicexposure to silica. Am J Kidney Dis 38:E28, 2001

11. Dijstelbloem HM, Scheepers RH, Oost WW, et al:Fcgamma receptor polymorphisms in Wegener’s granuloma-tosis: Risk factors for disease relapse. Arthritis Rheum42:1823-1827, 1999

12. Tse WY, Abadeh S, Jefferis R, Savage CO, Adu D:Neutrophil FcgammaRIIIb allelic polymorphism in anti-neutrophil cytoplasmic antibody (ANCA)-positive systemicvasculitis. Clin Exp Immunol 119:574-577, 2000

13. Tse WY, Abadeh S, McTiernan A, Jefferis R, SavageCO, Adu D: No association between neutrophil Fcgam-maRIIa allelic polymorphism and anti-neutrophil cytoplas-mic antibody (ANCA)-positive systemic vasculitis. ClinExp Immunol 117:198-205, 1999

14. Zhou Y, Giscombe R, Huang D, Lefvert AK: Novelgenetic association of Wegener’s granulomatosis with theinterleukin 10 gene. J Rheumatol 29:317-320, 2002

15. Huang D, Giscombe R, Zhou Y, Lefvert AK: Polymor-

phisms in CTLA-4 but not tumor necrosis factor-alpha orinterleukin 1beta genes are associated with Wegener’s granu-lomatosis. J Rheumatol 27:397-401, 2000

16. Specks U, Bartfai Z, Gaede KI, Russell KA, Mura-kozy G, Muller-Quernheim J: Frequency of functional IL-10and TGF-� gene polymorphisms in ANCA-associated vascu-litis(AAV). Cleve Clin J Med 69:157A(abstr), 2002

17. Stratta P, Messuerotti A, Canavese C, et al: The roleof metals in autoimmune vasculitis: Epidemiological andpathogenic study. Sci Total Environ 270:179-190, 2001

18. Hogan SL, Satterly KK, Dooley MA, Nachman PH,Jennette JC, Falk RJ: Silica exposure in anti-neutrophilcytoplasmic autoantibody-associated glomerulonephritis andlupus nephritis. J Am Soc Nephrol 12:134-142, 2001

19. Tervaert JW, Stegeman CA, Kallenberg CG: Siliconexposure and vasculitis. Curr Opin Rheumatol 10:12-17,1998

20. Gregorini G, Tira P, Frizza J, et al: ANCA-associateddiseases and silica exposure. Clin Rev Allergy Immunol15:21-40, 1997

21. Nuyts GD, Van Vlem E, De Vos A, et al: Wegenergranulomatosis is associated to exposure to silicon com-pounds: A case-control study. Nephrol Dial Transplant 10:1162-1165, 1995

22. Cotch MF, Hoffman GS, Yerg DE, Kaufman GI,Targonski P, Kaslow RA: The epidemiology of Wegener’sgranulomatosis. Estimates of the five-year period preva-lence, annual mortality, and geographic disease distributionfrom population-based data sources. Arthritis Rheum 39:87-92, 1996

23. Tidman M, Olander R, Svalander C, Danielsson D:Patients hospitalized because of small vessel vasculitideswith renal involvement in the period 1975-95: Organ involve-ment, anti-neutrophil cytoplasmic antibodies patterns, sea-sonal attack rates and fluctuation of annual frequencies.J Intern Med 244:133-141, 1998

24. Raynauld JP, Bloch DA, Fries JF: Seasonal variationin the onset of Wegener’s granulomatosis, polyarteritis no-dosa and giant cell arteritis. J Rheumatol 20:1524-1526,1993

25. Falk RJ, Hogan S, Carey TS, Jennette JC: Clinicalcourse of anti-neutrophil cytoplasmic autoantibody-associ-ated glomerulonephritis and systemic vasculitis. The Glomer-ular Disease Collaborative Network. Ann Intern Med 113:656-663, 1990

26. Choi HK, Merkel PA, Walker AM, Niles JL: Drug-associated antineutrophil cytoplasmic antibody-positive vas-culitis: Prevalence among patients with high titers of antimy-eloperoxidase antibodies. Arthritis Rheum 43:405-413, 2000

27. Walton EW: Giant-cell granuloma of the respiratorytract (Wegener’s granulomatosis). Br Med J 2:265-270,1958

28. Adu D, Pall A, Luqmani RA, et al: Controlled trial ofpulse versus continuous prednisolone and cyclophospha-mide in the treatment of systemic vasculitis. Q J Med90:401-409, 1997

29. Hoffman GS, Kerr GS, Leavitt RY, et al: Wegener’sgranulomatosis: An analysis of 158 patients. Ann Intern Med116:488-498, 1992

30. Gordon M, Luqmani RA, Adu D, et al: Relapses in

ANCA-ASSOCIATED RENAL VASCULITIS: 1995-2000 783

patients with a systemic vasculitis. Q J Med 86:779-789,1993

31. Talar-Williams C, Hijazi YM, Walther MM, et al:Cyclophosphamide-induced cystitis and bladder cancer inpatients with Wegener granulomatosis. Ann Intern Med124:477-484, 1996

32. Exley AR, Bacon PA, Luqmani RA, et al: Examina-tion of disease severity in systemic vasculitis from the novelperspective of damage using the vasculitis damage index(VDI). Br J Rheumatol 37:57-63, 1998

33. Jayne DR, Rasmussen N: Treatment of antineutrophilcytoplasm autoantibody-associated systemic vasculitis: Ini-tiatives of the European Community Systemic VasculitisClinical Trials Study Group. Mayo Clin Proc 72:737-747,1997

34. Jennette JC, Falk RJ, Andrassy K, et al: Nomencla-ture of systemic vasculitides. Proposal of an internationalconsensus conference. Arthritis Rheum 37:187-192, 1994

35. UK National Statistics Series (mortality data). SeriesDH 31:1469-2805, 1999

36. Ansell D, Feest T, Will E, et al: The UK RenalRegistry Second Annual Report. UK Renal Registry, 1999

37. Van der Woude FJ, Rasmussen N, Lobatto S, et al:Autoantibodies against neutrophils and monocytes: Tool fordiagnosis and marker of disease activity in Wegener’s granu-lomatosis. Lancet 1:425-429, 1985

38. Pettersson EE, Sundelin B, Heigl Z: Incidence andoutcome of pauci-immune necrotizing and crescentic glomer-ulonephritis in adults. Clin Nephrol 43:141-149, 1995

39. Geffriaud-Ricouard C, Noel LH, Chauveau D, Hou-hou S, Grunfeld JP, Lesavre P: Clinical spectrum associatedwith ANCA of defined antigen specificities in 98 selectedpatients. Clin Nephrol 39:125-136, 1993

40. Andrassy K, Erb A, Koderisch J, Waldherr R, Ritz E:Wegener’s granulomatosis with renal involvement: Patientsurvival and correlations between initial renal function,renal histology, therapy and renal outcome. Clin Nephrol35:139-147, 1991

41. Westman KW, Bygren PG, Olsson H, Ranstam J,Wieslander J: Relapse rate, renal survival, and cancer mor-bidity in patients with Wegener’s granulomatosis or micro-scopic polyangiitis with renal involvement. J Am Soc Neph-rol 9:842-852, 1998

42. Nachman PH, Hogan SL, Jennette JC, Falk RJ: Treat-ment response and relapse in antineutrophil cytoplasmicautoantibody-associated microscopic polyangiitis and glo-merulonephritis. J Am Soc Nephrol 7:33-39, 1996

43. Hogan SL, Nachman PH, Wilkman AS, Jennette JC,Falk RJ: Prognostic markers in patients with antineutrophilcytoplasmic autoantibody-associated microscopic polyangi-itis and glomerulonephritis. J Am Soc Nephrol 7:23-32,1996

44. Bajema IM, Hagen EC, Hermans J, et al: Kidneybiopsy as a predictor for renal outcome in ANCA-associated

necrotizing glomerulonephritis. Kidney Int 56:1751-1758,1999

45. Falk RJ, Nachman PH, Hogan SL, Jennette JC: ANCAglomerulonephritis and vasculitis: A Chapel Hill perspec-tive. Semin Nephrol 20:233-243, 2000

46. Guillevin L, Cohen P, Gayraud M, Lhote F, JarrousseB, Casassus P: Churg-Strauss syndrome. Clinical study andlong-term follow-up of 96 patients. Medicine (Baltimore)78:26-37, 1999

47. Jayne DRW, Gaskin G, for the European VasculitisGroup (EUVAS): Randomised trial of CyclophosphamideVersus Azathioprine During Remission in ANCA-Associ-ated Vasculitis (CYCAZAREM). J Am Soc Nephrol 10:105A(abstr), 1999

48. Franssen CF, Stegeman CA, Kallenberg CG, et al:Antiproteinase 3- and antimyeloperoxidase-associated vas-culitis. Kidney Int 57:2195-2206, 2000

49. Aasarod K, Iversen BM, Hammerstrom J, et al:Wegener’s granulomatosis: Clinical course of 108 patientswith renal involvement. Nephrol Dial Transplant 15:611-618,2000

50. Bindi P, Mougenot B, Mentre F, et al. Necrotizingcrescentic glomerulonephritis without significant immunedeposits: A clinical and serological study. Q J Med 55-68,1993

51. Krafcik SS, Covin RB, Lynch JP III, Sitrin RG:Wegener’s granulomatosis in the elderly. Chest 109:430-437, 1996

52. Savige J, Davies D, Falk RJ, Jennette JC, Wiik A:Antineutrophil cytoplasmic antibodies and associated dis-eases: A review of the clinical and laboratory features.Kidney Int 57:846-862, 2000

53. Mekhail TM, Hoffman GS: Long-term outcome ofWegener’s granulomatosis in patients with renal diseaserequiring dialysis. J Rheumatol 27:1237-1240, 2000

54. Nachman PH, Segelmark M, Westman K, et al: Recur-rent ANCA-associated small vessel vasculitis after transplan-tation: A pooled analysis. Kidney Int 56:1544-1550, 1999

55. Jayne D: Evidence-based treatment of systemic vas-culitis. Rheumatology (Oxford) 39:585-595, 2000

56. Booth AD, Jefferson HJ, Ayliffe W, Andrews PA,Jayne DR: Safety and efficacy of TNFalpha blockade inrelapsing vasculitis. Ann Rheum Dis 61:559, 2002 (letter)

57. Hagen EC, de Keizer RJ, Andrassy K, et al: Compas-sionate treatment of Wegener’s granulomatosis with rabbitanti-thymocyte globulin. Clin Nephrol 43:351-359, 1995

58. Lockwood CM, Thiru S, Isaacs JD, Hale G, Wald-mann H: Long-term remission of intractable systemic vascu-litis with monoclonal antibody therapy. Lancet 341:1620-1622, 1993

59. Nowack R, Gobel U, Klooker P, Hergesell O, An-drassy K, van der Woude FJ: Mycophenolate mofetil formaintenance therapy of Wegener’s granulomatosis and mi-croscopic polyangiitis: A pilot study in 11 patients with renalinvolvement. J Am Soc Nephrol 10:1965-1971, 1999

BOOTH ET AL784