Is There a Relation with PR-3 Negative C-ANCA and Hypergammaglobulinaemia ?

Ann Rheum Dis 1998;57:141–145 141

Clinical spectrum associated with positive Ann Rheum Dis: first published as 10.1136/ard.57.3.141 on 1 March 1998. Downloaded from ANCA titres in 94 consecutive patients: is there a relation with PR-3 negative c-ANCA and hypergammaglobulinaemia ?

D Blockmans, E Stevens, G Mariën, H Bobbaers

Abstract In 1985, van der Woude et al1 first found the so Objective—To calculate the positive pre- called antineutrophil cytoplasmic antibodies dictive value (ppv) of cytoplasmic anti- (ANCA)—which were described three years neutrophil cytoplasmic antibodies (c- earlier by Davies et al2—to be connected with ANCAs) and anti-proteinase 3 (PR 3) Wegener’s granulomatosis (WG). They pro- antibodies for Wegener’s granulomatosis posed that ANCAs were useful for diagnosis (WG) and to evaluate their association and for follow up of disease activity in patients with other diseases. with this form of systemic vasculitis. Later, Methods—The clinical files of all 94 pa- cytoplasmic staining ANCAs (c-ANCAs, tients who had a positive c- or perinuclear which were sensitive and specific for WG) were (p)-ANCA test, or both, in the laboratory diVerentiated from perinuclear staining ANCAs (p-ANCAs), found in other vascu- of the University Hospital, Leuven be- 3 tween April 1995 and March 1996 and who litides such as microscopic polyangiitis. Using attended the Internal Medicine Depart- ELISA techniques, the responsible antigen for c-ANCA staining was proved to be ment of the hospital were retrospectively 4 studied. proteinase-3 (PR 3), while most p-ANCAs were directed against the myeloperoxidase Results—Of the 94 patients with ANCAs (MPO) enzyme.5 In more recent years, other (fluorescence titre > 1/40), 57 were diseases without vasculitis were reported to be c-ANCA positive and 45 p-ANCA positive associated with ANCAs, for example, ulcera- (eight were simultaneously c- and tive colitis, primary biliary cirrhosis, rheuma- p-ANCA positive). Of the 57 c-ANCA toid arthritis, human immunodeficiency virus positive patients, 23 had WG. The ppv for infection or Sjögren’s syndrome.67 Hence, it WG thus was 40%. This value did not has become less clear what diagnostic value increase by defining a higher threshold for should be attached to a positive ANCA http://ard.bmj.com/ a positive ANCA. There was not a good determination in clinical practice. relation between ANCA titres and disease We therefore calculated in a retrospective activity in the WG patients, nor was there way the positive predictive value (ppv) of a relation between anti-PR 3 antibody c-ANCA and of anti-PR 3 antibodies for WG levels and WG disease activity. The ppv of and analysed the association of c- and p-ANCA anti-PR 3 antibodies for WG however was with other diseases. very high (85%). There was a positive on September 28, 2021 by guest. Protected copyright. correlation between the level of (hyper) Methods gammaglobulinaemia and c-ANCA titres The clinical files of all 94 patients who had a in those patients with final diagnoses not positive c- or p-ANCA test, or both, in our known to be associated with c-ANCA. laboratory during a one year period (between Forty five patients had positive p-ANCAs. April 1995 and March 1996) and who The largest group were those with attended the Internal Medicine Department of inflammatory bowel disease (n = 20, of our hospital were retrospectively studied. Department of whom the majority had colitis ulcerosa Internal Medicine Vasculitis was bioptically confirmed in 20 of D Blockmans or primary sclerosing cholangitis, or 23 WG patients. A total of 14 renal biopsies, 12 H Bobbaers both); the great majority of these patients bronchoscopic biopsies, 10 nasal biopsies, five had no anti-myeloperoxidase antibodies. skin biopsies, and one open lung biopsy were and Laboratory of Vasculitis was present in eight patients, of performed. Among the 20 patients with Immunology whom two had WG (both were also histologically confirmed vasculitis, two patients E Stevens 8 G Mariën c-ANCA positive). fulfilled all four ACR criteria for WG, 11 Conclusion—There is a low ppv of patients fulfilled three ACR criteria, six pa- University Hospital, c-ANCAs for WG, caused by a high tients fulfilled two ACR criteria, and only one Leuven, Belgium percentage of PR 3 negative, positive patient fulfilled only one ACR criterium. Of the c-ANCA determinations, possibly related three patients without histological proof of vas- Correspondence to: Dr D Blockmans, General to hypergammaglobulinaemia. Anti-PR 3 culitis, one patient fulfilled three ACR criteria Internal Medicine, UZ antibodies have a high ppv for WG. and two patients fulfilled two criteria. Gasthuisberg, Herestraat 49, However, neither c-ANCA titre, nor the A WG patient was considered to be in B-3000 Leuven, Belgium. level of anti-PR 3 antibodies correlated remission (or to have inactive disease) when Accepted for publication with the activity of the disease. there were no clinical or biochemical signs after 20 January 1998 (Ann Rheum Dis 1998;57:141–145) three months of ongoing inflammation (normal 142 Blockmans, Stevens, Mariën, et al

Table 1 Overview of the 94 ANCA positive patients with anti-PR 3 and anti-MPO results (total number of Ann Rheum Dis: first published as 10.1136/ard.57.3.141 on 1 March 1998. Downloaded from determinations = 44)

Number of anti-PR 3 Number of anti-MPO Patients (n) positive (%) positive (%) Pure c-ANCA 49 21 WG 11/11 (100) 0/11 (0) 9 non-WG vasculitis 1/1 0/1 2 inflammatory bowel disease 0/0 0/0 (1 + non-WG vasculitis) 18 miscellaneous 1/6 (17) 0/6 (0) (borderline) Pure p-ANCA 37 18 inflammatory bowel disease 0/9 (0) 1/9 (11) 6 non-WG vasculitis 0/6 (0) 3/6 (50) 13 miscellaneous 0/8 (0) 0/8 (0) c- and p-ANCA 8 2 WG 0/1 1/1 2 inflammatory bowel disease 0/1 0/1 4 miscellaneous 0/1 1/1

sedimentation rate and C reactive protein) and STATISTICS when the urinary sediment had returned to Data are given as mean (SEM). Correlations normal. ANCA titres were not taken into were done with the Spearman rank correlation. account to determine if a patient had active A p value of <0.05 was considered significant. disease or not. ANCA tests were performed by indirect immunofluorescence using commercially avail- Results able slides (INOVA Diagnostics, San Diego, Of the 94 patients with positive (> 1/40) CA). Ethanol fixed neutrophils were exposed ANCA determinations, 49 patients were only to patient serum diluted 1:40 for screening. A c-ANCA positive, 37 patients were only positive c-ANCA serum, a positive p-ANCA p-ANCA positive, and eight patients had both serum, and a negative control were included in c-and p-ANCAs (table 1). Hence, a total of 57 every run. A specific anti-IgG fluorescent con- patients were c-ANCA positive and a total of jugate was used. Any positive samples were 45 patients were p-ANCA positive. titrated in doubling dilutions until end point. When a p-ANCA pattern was seen (peripheral PATIENTS WITH POSITIVE C-ANCA or diVuse nuclear staining), serum was tested DETERMINATIONS for the presence of antinuclear antibodies Twenty three of these 57 patients suVered from (ANA) on Hep-2 cells (Immunoconcepts, Sac- WG (21 were only c-ANCA positive, two ramento, CA). All positive ANA serum sam- patients had c- and p-ANCAs), which repre- ples were further tested in a dilution of 1/40 on sents a ppv of a positive c-ANCA determina- formalin fixed neutrophils (INOVA Diagnos- tion for WG of 40%. Of these 23 patients, 12 tics, San Diego, CA). Those samples that gave were in remission after treatment and 11 had

a cytoplasmic staining pattern on formalin active disease. http://ard.bmj.com/ fixed neutrophils, were reported to be true The ppv of c-ANCAs did not increase when p-ANCA samples despite the simultaneous taking a higher titre as cut oV point: if we con- presence of an ANA. sider only c-ANCAs at 1/80 or higher as Results of c- and p-ANCA were reported as indicative for WG, the predictive value was negative or positive 1/40 to 1/10240. 44%, at > 1/160, > 1/320, and > 1/640, it was ELISA tests for anti-PR 3 and anti-MPO 45%, 46%, and 38% respectively. The highest were purchased from INOVA Diagnostics (San c-ANCA titre observed (1/5120) was found in Diego, CA). A specific enzyme labelled anti- a non-WG patient. on September 28, 2021 by guest. Protected copyright. IgG conjugate was used. ELISA tests were car- Follow up determinations were carried out ried out from December 1995, but only on in 10 patients, three of them showing active serum samples positive in the immunofluores- disease. There was a considerable variation in cence test and according the manufacturer’s ANCA titres, both in patients with still active instructions (n = 44). The results were ex- disease and in patients who went into remis- pressed in arbitrary units per litre (AU/l). Sam- sion. In WG patients in remission, we found ples were considered to be positive for anti-PR ANCA titres fluctuating from negative to 1/320 3 when values were greater than 20 AU/l and without corresponding change in disease activ- for anti-MPO when values were greater than ity. Other patients in stable remission contin- 5 AU/l. ued to have high ANCA titres throughout fol- Total protein concentration was determined low up (titres from 1/640 to 1/2560). using Boehringer-Mannheim reagent kits and Nine other patients with positive c-ANCA application on a Hitachi-747 (Boehringer- determinations had a form of systemic vasculi- Mannheim, GNBH, Mannheim, Germany). tis diVerent from WG: one patient had renal Gammaglobulins were measured using limited vasculitis, one patient had Behçet’s dis- cepharose-cellulose-polyacetate electrophore- ease, three patients suVered from classic sis strips for Microsome System (Gelman Sci- periarteritis nodosa, and in four patients ences, Ann Arbor, MI, USA). The gels were vasculitis could not be further classified. equilibrated with di-ethyl-barbital buVer pH Combining all patients with a systemic 8.6. vasculitic disorder (23 WG patients and nine Serum immune complexes were measured non-WG vasculitis patients), our ppv of a posi- by C1q and monoclonal rheumatoid factor tive c-ANCA determination for systemic vas- binding solid phase radioimmunoassays. culitis reaches 56%. Clinical spectrum associated with positive ANCA titres 143

4.0 PATIENTS WITH POSITIVE P-ANCA Ann Rheum Dis: first published as 10.1136/ard.57.3.141 on 1 March 1998. Downloaded from DETERMINATIONS 3.8 P-ANCAs (titre > 1/40) were found in 45 3.6 patients: 37 patients had pure p-ANCAs, eight 3.4 patients had c- and p-ANCAs. 3.2 Twenty patients suVered from an inﬂamma- 3.0 tory bowel disease or primary sclerosing cholangitis, or both: 10 patients had isolated 2.8 colitis ulcerosa, seven patients had a combina- 2.6 tion of colitis ulcerosa and primary sclerosing 2.4 cholangitis (of whom two were also c-ANCA 2.2 positive), two patients suVered from Crohn’s 2.0 disease and in one patient, there was an isolated primary sclerosing cholangitis. The ppv of 1.8 p-ANCA was 44% for the group of inﬂamma- 1.6 tory bowel disease patients. 1.4 Systemic vasculitis was found in eight

Gammaglobulinaemia (g/dl) 1.2 patients (six patients with pure p-ANCAs, two 1.0 patients with c- and p-ANCAs): in two patients a diagnosis of microscopic polyangiitis was 0.8 Normal range made, two patients had WG (both were also 0.6 c-ANCA positive), one patient had renal 0.4 limited vasculitis, one had arteritis temporalis, 0.2 and in two patients vasculitis was secondary to 0 a connective tissue disease (scleroderma and

40 80 dermatomyositis respectively). Thus, the ppv 160 320 640 1280 2560 5120

10240 of p-ANCA was 18% for this group of systemic c-ANCA titre vasculitis patients. Figure 1 Gammaglobulinaemia compared with c-ANCA The 17 remaining patients with p-ANCAs titres in 16 patients without systemic vasculitis. Spearman (13 pure p-ANCAs, four combination of c- and rank correlation for these data givesapvalue < 0.02. p-ANCAs) suVered from a wide spectrum of diseases: systemic lupus erythematosus, pneu- monia (n = 2), emphysema, rheumatoid arthri- Four patients with positive c-ANCA deter- tis, Hashimoto thyroiditis, tuberculosis, bron- minations (two pure c-ANCA, two combina- chiectasias, extrinsic allergic alveolitis, tion of c- and p-ANCA, see table 1) had colitis autoimmune haemolytic anaemia, idiopathic ulcerosa; in three patients this was associated lung ﬁbrosis, chronic obstructive lung disease, with primary sclerosing cholangitis and in one ankylosing spondylitis, hepatitis B, spinal canal

with Behçet’s disease (see above). stenosis, unexplained isolated arthralgias (n = http://ard.bmj.com/ In the 22 remaining patients (18 patients 1). The ã globulin value was known in 13 of with pure c-ANCAs, four patients with c- and these 17 patients. Seven patients had normal p-ANCAs), no obvious reason could be found values of ã globulin, there was a borderline for their positive c-ANCA result. The following increase in one patient, and in ﬁve other diagnoses were made in these patients: idio- patients there was hypergammaglobulinaemia. pathic lung ﬁbrosis, bronchiectasias (n = 3), There was no correlation between p-ANCA titres and ã globulin values.

pulmonary emphysema (n = 2), interstitial on September 28, 2021 by guest. Protected copyright. lung disease caused by rheumatoid arthritis, haemoptysis caused by an abnormal bronchial ANTI-PR 3 AND ANTI-MPO ANTIBODY artery, polymyalgia rheumatica, idiopathic DETERMINATION thrombocytopenic purpura, sarcoidosis, Anti-PR 3 antibodies were only found in chronic rhinitis, IgA nephritis, urosepsis, c-ANCA positive serum samples, never in pure hypersensitivity vasculitis, chronic renal fail- p-ANCA serum samples. The opposite was ure, cirrhosis caused by hepatitis C, Sjögren’s true for anti-MPO antibodies. syndrome in combination with myelodysplasia, Figure 2 shows anti-PR 3 values obtained in four WG patients with active disease, seven lupus nephritis, and previous hepatitis B infec- WG patients in remission, and in seven patients tion (n = 1). In 16 of these 22 patients, with conditions other than WG (all pure gammaglobulinaemia was determined and for c-ANCA serum samples). As figure 2 shows, 11 of them, the value was above the normal there is no diVerence in anti-PR 3 concentra- range (fig 1). There was a positive Spearman tions between WG patients with still active dis- rank correlation (rs = 0.77, t = 4.56, p < 0.02) ease and patients in remission. Patients with between the c-ANCA titres of these patients conditions other than WG were mostly anti-PR and their gammaglobulinaemia. In WG pa- 3 negative. The only patient with a high tients, there was no positive correlation be- anti-PR 3 antibody value had classic periarteri- tween ANCA titres and (hyper)gammaglobuli- tis nodosa. The other six patients in this group naemia. ã Globulin concentrations in WG had no systemic vasculitis: five of six were patients were on the average rather low (mean negative, one was borderline positive (21 AU, a (SEM) ã globulin concentration in 19 WG patient with primary sclerosing cholangitis and patients: 0.93 (0.09) g/dl), probably because of sarcoidosis). The ppv of a positive anti-PR 3 immunosuppressive treatment. result for WG is thus 11 of 13 or 85%, for sys- 144 Blockmans, Stevens, Mariën, et al

210 and p-ANCA antibodies, immune complexes Ann Rheum Dis: first published as 10.1136/ard.57.3.141 on 1 March 1998. Downloaded from 200 could be demonstrated. 190 180 170 Discussion 160 Positive predictive value can be used to 150 evaluate the usefulness of ANCA determina- 9 140 tion in clinical practice. However, preselection 10 130 bias may considerably aVect its value. Fur- 120 thermore, the population studied should be 110 deﬁned unequivocally to save comparability 100 with other studies. Therefore, we deﬁned our 90 study population as all ANCA positive patients 80 who attended our department during one year.

anti-PR-3 ELISA (AU/l) ANCA determinations were carried out only 70 on clinical suspicion of ANCA related diseases. 60 Ninety four positive patients were found. In 50 eight of them (8.5%) both c- and p-ANCAs 40 were present, which contrasts with previous 30 statements that c- and p-ANCAs are mutually 20 11 Normal exclusive. It is remarkable that most of these 10 range patients (62%) were also ANA positive. 0 WG WG Others We calculated a ppv of 40% for c-ANCAs in patients patients in WG and of 56% in systemic vasculitis. This with remission ﬁgure is independent of the antibody titre, active while sensitivity of course dramatically drops disease when only considering higher c-ANCA titres as Figure 2 Highest anti-PR3 antibody values in four WG indicative for WG. This latter value should be patients with active disease, seven WG patients in remission, and in seven patients with conditions other than compared with the ppv of 40% obtained by WG. The one high value in this last group was a classic Davenport et al12 for a combination of WG, periarteritis nodosa patient, the borderline positive value microscopic polyangiitis, and the syndrome of was in a patient with primary sclerosing cholangitis and sarcoidosis. systemic vasculitis and with the 38% obtained by Edgar et al13 for any primary vasculitic disorder. In contrast with our experience, the temic vasculitis 92% (including the patient ppv obtained by these authors increased by with classic periarteritis nodosa). taking higher cut oV points. Anti-MPO antibodies were measured in 10 The ppv of a positive p-ANCA determina- patients with colitis ulcerosa (nine pure tion for a systemic vasculitic disorder was 18%

p-ANCAs and 1 c- and p-ANCAs); they were (eight of 45 positive p-ANCA patients had sys- http://ard.bmj.com/ negative in 9 of 10. Anti-MPO antibodies were temic vasculitis), very similar to the 20% determined in all six vasculitis patients with obtained by Edgar et al.13 Almost half of our pure p-ANCAs. In three of them, anti-MPO patients with positive p-ANCA determinations antibodies were present at a level higher than 5 suVered from an inflammatory bowel disease, AU/l. Anti-MPO antibodies were determined especially colitis ulcerosa, isolated or associated in nine patients with p-ANCAs without good with primary sclerosing cholangitis (predictive explanation: they were negative in eight (< 5 positive value for these disorders 44% ). AU/l) and positive in one (11AU/l, a patient In contrast with c-ANCA immunofluores- on September 28, 2021 by guest. Protected copyright. with combined c- and p-ANCAs) (table 1). ence results, a positive anti-PR 3 ELISA was very predictive of WG. Of 13 positive anti-PR 3 ANA AND IMMUNE COMPLEX DETERMINATION IN determinations, 11 were found in WG patients, THE 94 PATIENTS one in a classic periarteritis nodosa patient and ANA determinations were available in 82 only one (borderline) positive determination in patients. ANA was present in 17 patients: 11% a non-vasculitic disorder (fig 2). This repre- (n= 4) of pure c-ANCA serum samples were sents a ppv for WG of 85%, for systemic vascu- ANA positive (all homogenous patterns), 21% litis of 92%. However, there was no correlation (n = 8) of pure p-ANCA serum samples were between the height of a positive ELISA test and ANA positive (also all homogenous pattern) the activity of the vasculitis in WG patients (fig while five serum samples (62%) with combined 2). c-and p-ANCAs were positive (three with a During our study period, we had only a lim- homogenous pattern, one nucleolar, one finely ited number of positive anti-MPO ELISA speckled pattern). Anti-DNA antibodies were results: four of seven determinations were posi- negative in 12 of 13 ANA positive serum sam- tive in p-ANCA positive vasculitis patients ples tested. (including one WG patient with combined c- Serum immune complexes were determined and p-ANCA), one of 10 patients with inflam- in 38 patients, they were positive in 68% of matory bowel disease and in one patient with these patients (n = 26): 74% of positive bronchiectatic lung disease (with high c- and c-ANCA serum samples tested contained p-ANCAs). MPO is considered the main anti- immune complexes, 55% of p-ANCA serum gen responsible for p-ANCA staining in vascu- samples tested had immune complexes, and in litis patients, but other antigens may also be 75% of serum samples with a combination of c- involved, for example, elastase, lactoferrin, etc. Clinical spectrum associated with positive ANCA titres 145

P-ANCAs in inflammatory bowel disease are hypothesised that polyclonal B cell prolifera- Ann Rheum Dis: first published as 10.1136/ard.57.3.141 on 1 March 1998. Downloaded from generally directed against antigens other than tion may result in the presence of these anti-PR MPO.14 3 negative c-ANCAs (or atypical ANCAs), as For three WG patients with active disease was found in patients from the tropics with and seven WG patients in remission, ANCA malaria or tuberculosis.20 We are however determinations were carried out several times aware of the potential pitfalls of any retrospec- during the study period. We found substantial tive approach. For instance, diagnosis was variations in titres with time. These were not made in the knowledge of serological results correlated to disease activity, as these patients and this might potentially have influenced the were in constant remission. Some patients in ppv obtained for ANCAs. Therefore, the remission always had higher ANCA titres than results obtained in this study should be other patients with still active disease. In a confirmed (or denied) by ANCA determina- recently finished study in 50 Wegener patients, tions in patients with hypergammaglobulinae- followed up in our department,15 we also could mia and in diVerent types of vasculitis. not find a correlation between ANCA titres and disease activity. Therefore, we do not treat 1 van der Woude FJ, Rasmussen N, Lobatto S, et al. Autoan- patients based on variations in their ANCA tibodies against neutrophils and monocytes : tool for diag- titres, but rely on clinical and well documented nosis and marker of disease activity in Wegener’s granulomatosis. Lancet 1985;i:425–9. biochemical parameters of disease activity. 2 Davies DJ, Moran JE, Niall JF, Ryan GB. Segmental necro- Kerr 16 and Davenport 17 were also tising glomerulonephritis with antineutrophil antibody : et al et al possible arbovirus aetiology? BMJ 1982;285:606. unable to find a close relation between changes 3 Jenette JC, Falk RJ. Antineutrophil cytoplasmic auto- in ANCA titres and disease activity. ANCA antibodies and associated diseases : a review. Am J Kidney Dis 1990;15:517–29. assays should always be used in conjunction 4 Niles JL, McCluskey RT, Ahmad MF, Arnaout MA. Wege- with other parameters of disease activity and ner’s granulomatosis autoantigen is a novel proteinase. Blood 1989;74:1888–93. should not be the sole basis for changing 5 Falk RJ, Jennette JC. Anti-neutrophil cytoplasmic auto- treatment.18 antibodies with specificity for myeloperoxidase in patients with systemic vasculitis and idiopathic necrotizing and In 22 patients, we could not find an obvious crescentic glomerulonephritis. N Engl J Med 1988;318: cause for positive c-ANCAs. However, our 1651–7. 6 Rump JA, Scholmerich J, Gross V, et al. A new type of peri- attention was drawn to the fact that many of the nuclear antineutrophil cytoplasmic antibody (p-ANCA) in diVerent diseases found in these patients were active ulcerative colitis but not in Crohn’s disease. Immunobiology 1990;181:406–13. chronic diseases that are frequently associated 7 Peter HH, Metzger D, Rump A, Röther E. ANCA in with polyclonal hypergammaglobulinaemia diseases other than systemic vasculitis. Clin Exp Immunol 1993;91 (suppl 1):S12–14 (for example, bronchiectasias, cirrhosis, Sjö- 8 Leavitt RY, Fauci AS, Bloch DA, et al. The American Col- gren’s syndrome, etc). Hyperglobulinaemia lege of Rheumatology 1990 criteria for the classification of Wegener’s granulomatosis. Arthritis Rheum 1990;33: was present in 11 of 16 patients. We found a 1101–7 . significant correlation between the gamma- 9 Edgar JDM. The clinical utility of ANCA positivity. Ann Rheum Dis 1996;55:494–6. globulinaemia and the c-ANCA titre in these 10 Jennette JC. Anti-neutrophil cytoplasmic autoantibody- 16 patients. All except one were anti-PR 3 associated disease: a pathologist’s perspective. Am J Kidney Disease 1991;18:164–70. negative. It is possible that part of our 11 GeVriaud-Ricouard L, Noel LH, Chauveau D, Houhou S, http://ard.bmj.com/ c-ANCAs are what is called “atypical Grünfeld JP, Lesafre P. Clinical spectrum associated with ANCA of defined antigen specificities in 98 selected ANCAs”. Positive ANCA staining with a more patients. Clin Nephrol 1993;39:125–36. homogenous rather than a granular cytoplas- 12 Davenport A, Lock RJ, Wallington TB, Feest TG. Clinical significance of anti-neutrophil cytoplasm antibodies de- mic staining pattern was seen, for example, in tected by a standardised indirect immunofluorescence patients with HIV.19 These atypical patterns are assay. Q J Med 1994;87:291–9. 13 Edgar JDM, McMillan SA, Bruce I, Conlan SK. An audit of associated with antibodies to other antigens ANCA in routine clinical practice. Postgrad Med J such as lactoferrin and elastase. ELISAs with 1995;71:605–12. 14 Kallenberg CGM, Brouwer E, Weening JJ, Cohen Tervaert on September 28, 2021 by guest. Protected copyright. these antigens are not available in our labora- JW. Anti-neutrophil cytoplasmic antibodies: current diag- tory. In immunofluorescence tests, a clear dis- nostic and pathophysiological potential. Kidney Int 1994; 46:1–15. tinction between these atypical ANCAs and 15 Blockmans D, Vandenbon C, Macken T, Stevens E, true c-ANCAs is not always possible and relies Bobbaers H. Wegener’s granulomatosis : a study on 50 patients. European Journal of Internal Medicine 1997;8: highly on interpretation by the technician 19–26. (granular versus more homogenous cytoplas- 16 Kerr GS, Fleisher TA, Hallahan CW, et al. Limited prognostic value of changes in anti-neutrophil cytoplasmic mic fluorescence). In our laboratory and in antibody titer in patients with Wegener’s granulomatosis. many others, fine or more granular diVuse Arthritis Rheum 1993;36:365–71. 17 Davenport A, Lock RJ, Wallington T.Clinical significance of cytoplasmic staining patterns are all reported the serial measurement of autoantibodies to neutrophil as c-ANCAs. cytoplasm using a standard indirect immunofluorescence test. Am J Nephrol 1995;15:201–7. Hypergammaglobulinaemia, perhaps in 18 De’Oliveira J, Gaskin G, Dash A, Rees AJ, Pusey CD. Rela- combination with immune complex formation tionship between disease activity and anti-neutrophil cytoplasmic antibody concentration in long-term management (found in most our patients), as a cause of false of systemic vasculitis. Am J Kidney Dis 1995;25:380–9. positive c-ANCA determinations may explain 19 Klaassen RJL, Goldschmeding R, Dolman KM, et al. Anti- neutrophil cytoplasmic autoantibodies in patients with why patients with HIV or with Sjögren’s symptomatic HIV infection. Clin Exp Immunol 1992;87: disease, two disorders frequently associated 24–30. 20 Adebajo AO, Charles P, Maini RN, Hazleman BL. Autoan- with polyclonal hypergammaglobulinaemia, tibodies in malaria, tuberculosis and hepatitis B in a West are sometimes ANCA positive. It may be a African population. Clin Exp Immunol 1993;92:73–6.