Pauci-immune glomerulonephritis: does negativity of antineutrophilic cytoplasmic antibodies matters?

Transcription

1 International Journal of Rheumatic Diseases 2016; 19: ORIGINAL ARTICLE Pauci-immune glomerulonephritis: does negativity of antineutrophilic cytoplasmic antibodies matters? Aman SHARMA, 1, * Ritambra NADA, 2 Godasi S. R. S. N. K. NAIDU, 3 Ranjana W. MINZ, 4 Harbir Singh KOHLI, 3 Vinay SAKHUJA, 3 Krishan Lal GUPTA 3 and Manish RATHI 3, * Departments of 1 Internal Medicine, 2 Histopathology, 3 Nephrology, and 4 Immunopathology, Postgraduate Institute of Medical Education & Research, Chandigarh, India Abstract Aim: A significant proportion of pauci-immune glomerulonephritis (PIGN) patients are reported to have absence of anti-neutrophilic cytoplasmic antibodies (ANCA). However, studies are controversial regarding their significance and there is limited data after the new prognostic classification of PIGN. Methods: Renal biopsy-proven cases of PIGN were included and their clinical details, ANCA status by immunofluorescence (IIF) and enzyme-linked immunosorbent assay (ELISA), Birmingham Vasculitis Activity Score (BVAS) and treatment outcomes at 6 months were noted. The renal biopsies were classified according to the proposed histopathological classification. Scoring was done from 0 3 for interstitial edema, interstitial fibrosis and tubular atrophy (IFTA), interstitial inflammation and arteriosclerosis. The percentage of glomeruli with sclerosis, cellular and fibrous crescents, and percentage of subjects with glomerulitis, tuft necrosis, interstitial granuloma and vasculitis were noted. Results: Out of the 84 subjects included in the study, 33 (39.3%) were negative for ANCA by both IIF and ELISA. These subjects had significantly higher renal involvement, less extra-renal manifestations and lower BVAS. On histology, they had significantly higher proportion of crescentic class (66.7% vs. 41.2%, P = 0.039), higher number of cellular crescents (66.12% vs. 53.3%, P = ), higher IFTA (1.53 vs. 1.02, P = 0.009) and less interstitial edema (1.44 vs. 1.96, P = 0.003). The treatment outcomes were worse in ANCA-negative PIGN subjects, with significantly less improvement (37.2% vs. 62.8%, P = 0.02), more deterioration (40.7% vs. 14%, P = 0.006), and reduced probability of becoming dialysis free (31.6% vs. 69.6% P = 0.009). Conclusions: A negative ANCA in PIGN is associated with crescentic class, more IFTA and poor treatment outcomes. Key words: anti-neutrophilic cytoplasmic antibodies, pauci-immune glomerulonephritis, rapidly progressive glomerulonephritis, renal biopsy. INTRODUCTION Among different vasculitides, anti-neutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV), Correspondence: Associate Professor Manish Rathi, Department of Nephrology, Postgraduate Institute of Medical Education & Research, Chandigarh , India. drmanishrathi2000@yahoo.co.in *These authors have contributed equally to the manuscript and should be considered as first authors. which includes microscopic polyangiitis (MPA), granulomatosis with polyangiitis (GPA, previously Wegener s granulomatosis), eosinophilic granulomatosis with polyangiitis (EGPA, previously Churg Strauss syndrome) and renal limited vasculitis, have greater propensity to involve kidneys. 1,2 Renal involvement varies from asymptomatic urinary abnormalities to rapidly progressive deterioration of renal function. 3 The glomeruli are the most frequently involved compartment, and the morphologic changes in the renal biopsy 2015 Asia Pacific League of Associations for Rheumatology and Wiley Publishing Asia Pty Ltd

2 ANCA-negative pauci-immune glomerulonephritis are the gold standard for establishing a diagnosis of ANCA-associated glomerulonephritis. Renal biopsy is characterized by necrotizing and crescentic glomerulonephritis on light microscopy and the classical pauci-immune staining on immunofluorescence. 4,5 This pattern of renal involvement has been characteristically described as pauci-immune glomerulonephritis (PIGN). AAV are characterized by the presence of ANCA in the sera, 5 which can be detected either by indirect immunofluorescence (IIF) as cytoplasmic (c-anca) or peri-nuclear (p-anca) staining patterns, or by enzymelinked immunosorbent assay (ELISA) as antiproteinase-3 and anti-myeloperoxidase antibodies. 6 Although the positivity rates of ANCA in AAV is in the range of 90 95%; studies recruiting subjects with PIGN have reported that up to % of such patients can be ANCA negative As these patients constitute a small fraction of AAV; they have not been studied well. Moreover, most of the previous studies have included patients with crescentic PIGN and thus the significance of ANCA negativity in the light of recent prognostic histological classification of PIGN 13 is unclear. The present study was performed to investigate the clinical and pathological characteristics, and treatment outcomes of patients with ANCA-negative PIGN, as compared to those with positive ANCA. SUBJECTS AND METHODS All subjects with a renal biopsy diagnosis of PIGN and a minimum follow-up of 6 months, between July 2006 and October 2012 were included in the study. An informed consent was obtained from all subjects in the follow-up; the procedures followed were in accordance with the ethics standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in The study was approved by the Institute Ethics Committee. Clinical and laboratory parameters The subjects details, including clinical evaluation, investigations and ANCA status were noted according to the preformed questionnaire. The disease activity at first presentation was assessed by the Birmingham Vasculitis Activity Score (BVAS), version All subjects with positive ANCA were included in the ANCA-positive group, while those with negative ANCA were included in the ANCA-negative group. Hemoglobin, blood urea, serum creatinine, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), hematocrit, urine routine examination and microscopy and 24-h urinary protein at presentation were noted. The subjects of both groups were followed up with hemogram, blood urea, serum creatinine, urine routine and microscopy examination, 24-h urinary protein at the first, third and sixth months of treatment. The estimated glomerular filtration rate (egfr) was calculated using Modification of Diet in Renal Diseases (MDRD) formula 15 at baseline and 6 months. The treatment outcomes, that is, improvement, deterioration or death were noted at 6 months. Histopathological parameters The kidney biopsy slides of subjects with PIGN were reviewed by the pathologist (RN), blinded to the ANCA status. Pauci-immune staining pattern on immunofluorescence microscopy along with at least one glomeruli showing presence of necrotizing or crescentic glomerulonephritis on light microscopy was required for diagnosis of PIGN. Those with other co-morbid diseases or overlap syndromes, such as PIGN in combination with anti-glomerular basement membrane nephritis were excluded. The biopsy was analyzed and classified into four classes as per the new classification system for ANCA-associated PIGN. 13 In addition, the total number of glomeruli and percentage of involved glomeruli for tuft necrosis (glomnecr), cellular crescents (cellcres), fibrous crescents (fibcres), glomerular sclerosis (glomscl) and glomerulitis were also noted. Note was made of interstitial granuloma and vasculitis, while interstitial fibrosis and tubular atrophy (IFTA), interstitial edema, and interstitial inflammation were assessed and graded on a score of 0 3, where 0 indicate absence, 1 indicate involvement of 1 25% biopsy area, 2 indicate 26 50% biopsy area and 3 indicate > 50% biopsy area. The degree of arteriosclerosis was also graded from 0 3, where 0 indicates absence, 1 indicates luminal compromise by 25 50%, 2 indicates luminal compromise of 51 80%, while 3 indicates > 80% luminal occlusion. Testing for ANCA The ANCA positivity and type was determined using IIF and ELISA methods. The IIF assay was performed using in house ethanol-fixed neutrophil preparations. The staining pattern was classified as c-anca or p-anca or negative. The ELISA was performed with an indirect noncompetitive enzyme immunoassay using commercially available Varelisa TM myeloperoxidase (MPO) and proteinase (PR3) immunoglobulin G ELISA kits (Phadia GmbH, Freiburg, Germany). After loading the plates International Journal of Rheumatic Diseases 2016; 19:

3 A. Sharma et al. with serum samples, they were incubated for 30 min at room temperature. Subsequently the wells were washed with phosphate-buffered saline and the serum samples were decanted from the wells. After this, 100 ll of the horseradish peroxidase conjugate was added to each well and incubated for 30 min. After washing, 100 ll of tetramethyl blue chromagen was added to each well and incubated in the dark at room temperature for 30 min. The reaction was stopped using sulphuric acid, and the plates were gently tapped to mix the wells. The results were analyzed on ELISA Reader and the absorbance (OD) of each well was measured within 1 h of stopping the reaction. Every plate contained positive and negative controls and the samples were reported as positive or negative according to the manufacturer s recommended cut-off value. Definitions Normalization of serum creatinine or < 30% rise of serum creatinine from baseline value (i.e., value at diagnosis) was considered as improvement. A rise in serum creatinine > 30% from baseline or requirement of renal replacement therapy was considered as deterioration in renal function. 16 End-stage renal disease (ESRD) was defined as the requirement of renal replacement therapy for more than 12 weeks. Statistical analysis The data was initially entered into Microsoft Office Excel and subsequently analyzed using Statistical Package for the Social Sciences (SPSS) software (IBM, New York, NY, USA). For the descriptive statistics, continuous variables were reported in terms of mean, standard deviation and range, and categorical variables were presented as percentages. For univariate analysis, categorical variables were analyzed using Chi-square test, while Student s t-test was used for continuous variables. The multivariable analysis to look for factors affecting treatment outcome in ANCA-positive and negative groups was performed using linear regression analysis. The survival analysis and probability of reaching end-stage renal disease was analyzed by Kaplan Meier curves. P-value of < 0.05 was considered as statistically significant. RESULTS Baseline characteristics During the study period, 84 subjects were included, of which 52.4% were female and the mean age of the study population was 42.3 years (range years). Fifty-one subjects (60.7%) were ANCA-positive, while the remaining 33 subjects (39.3%) were ANCA-negative. In the ANCA-positive group, ANCA was positive by IIF in all patients (c-anca, 20; p-anca, 31) and by ELISA in 27 patients. Of these 27 patients, nine were positive for anti-proteinase 3 antibodies, while the remaining 18 had anti-myeloperoxidase antibodies. The important baseline characteristics in the two groups are summarized in Table 1, while the presenting manifestations are summarized in Table 2. ANCA-negative subjects had significantly more renal manifestations at the time of presentation (97% vs. 74.5%, P = 0.002), had less arthralgia, upper respiratory tract symptoms and neurological symptoms. Due to the paucity of extra-renal symptoms, a higher number of subjects with negative ANCA were labelled as unclassified as compared to ANCA-positive subjects (45.4% vs. 0%, P < 0.05), whereas MPA was more common in ANCApositive subjects (56.9% vs. 18.6%, P < 0.05). The baseline laboratory parameters were similar in the two groups. Although ANCA-negative subjects had a higher serum creatinine at presentation (674.5 lmol/l vs lmol/l), the difference was not statistically Table 1 Baseline characteristics of the two groups Parameters, n (%) ANCA-positive N = 51 ANCA-negative N = 33 Age (years), mean Sex Male 22 (43.1) 18 (54.5) Female 29 (56.9) 15 (45.5) Clinical diagnosis GPA 21 (41.2) 12 (36.4) MPA 29 (56.9)* 6 (18.2) CSS 1 (1.9) 0 (0) Unclassified 0 (0) 15 (45.4)* Laboratory parameters: mean (SD) Hb (g/l) 85.6 (2.53) 85.4 (3.17) Blood urea 47.2 (3.28) 55.3 (4.06) (mmol/l) Serum creatinine (63.23) (80.39) (lmol/l) egfr 20.9 (2.92) 16.1 (3.78) (ml/min/1.73 m 2 ) Serum albumin (g/l) 32.9 (0.96) 28.8 (1.24) 24-h UP (g/tv) 1.75 (0.20) 1.87 (0.25) Hematuria (% of points) ANCA, anti-neutrophilic cytoplasmic antibodies; egfr, estimated glomerular filtration rate; UP, urinary protein; GPA, granulomatosis with polyangiitis; MPA, microsopic polyangiitis; CSS, Churg Strauss syndrome; TV, total volume. *P < International Journal of Rheumatic Diseases 2016; 19: 74 81

4 ANCA-negative pauci-immune glomerulonephritis Table 2 Clinical manifestations according to the ANCA status Symptoms (%) ANCA-positive (n = 51) ANCA-negative (n = 33) significant. The mean BVAS score at presentation in the ANCA-positive group was 19.9 (range 8 38), while in the ANCA-negative group, it was 15.4 (range 8 32, P = 0.001). Histopathological classification and scoring P- value Constitutional Renal Pulmonary Arthralgia Mucocutaneous URT Neurological Ocular BVAS ANCA, anti-neutrophilic cytoplasmic antibodies; URT, upper respiratory tract; BVAS, Birmingham Vasculitis Activity Score. Bold values indicate significant P values. The distribution of various histological classes and the important histological parameters in the two groups are shown in Table 3. Significantly higher number of subjects in the ANCA-negative group had crescentic class of PIGN on histology as compared to ANCA-positive subjects (66.7% vs. 41.2%, P < 0.039), while there was no difference in other classes. Both the groups had similar extent of glomnecr, fibcres, glomscl, glomerulitis and interstitial granuloma. The subjects with ANCA-negative PIGN had significantly more cellcres (66.12% vs. 53.3%, P = ), higher IFTA (1.53 vs. 1.02, P = 0.009) and less interstitial edema (1.44 vs. 1.96, P = 0.003), while there was no difference in mean interstitial inflammation or degree of arteriosclerosis in the two groups. None of the patients with ANCA-negative PIGN had vasculitis or granulomatous glomerulitis. Treatment and outcome In both the groups, treatment consisted of 3 5 intravenous methylprednisolone pulses along with oral steroids 1 mg/kg/day for at least 6 weeks, followed by slow taper. In addition, 64 subjects (77%) received cyclophosphamide, either intravenous (n = 43, 67%) or oral (n = 21, 33%). Plasmapheresis was performed in 12 subjects, of which nine were ANCA-positive. Twenty-three subjects (45.1%) in the ANCA-positive group and 19 subjects (57.6%) in the ANCA negative group required renal replacement therapy (P = not significant). Table 3 Histological class and other biopsy parameters in the two groups Parameters ANCApositive (n = 51) ANCAnegative (n = 33) P-value Histological class, n (%) Focal 10 (19.6) 3 (9.1) Crescentic 21 (41.2) 22 (66.7) Sclerotic 8 (15.7) 4 (12.1) Mixed 12 (23.5) 4 (12.1) Histological scoring % glomerulitis with glom scl % glomerulitis with cellcres % glomerulitis with fibcres % pts with tuft necrosis % pts with glomerulitis % pts with interstitial granuloma % pts with vasculitis Mean interstitial 1.96 (0.68) 1.44 (0.84) edema (SD) Mean interstitial 1.02 (0.72) 1.53 (0.98) fibrosis (SD) Mean interstitial 1.86 (0.79) 2 (0.67) inflammation (SD) Mean arteriosclerosis (SD) 0.98 (0.83) 1.22 (0.87) ANCA, anti-neutrophilic cytoplasmic antibodies; glom scl, glomerular sclerosis; cellcres, cellular crescents; fibcres, fibrous crescents; % pts, % of patients. Bold values indicate significant P values. At the end of 6 months, both the groups showed significant improvement in serum creatinine, egfr and serum albumin, while 24-h proteinuria improved significantly in the ANCA-negative group and hemoglobin improved significantly in the ANCA-positive group (Table 4). The treatment outcomes in the two groups are summarized in Table 5. A significantly higher proportion of ANCA-positive subjects improved as compared to the ANCA-negative group (62.8% vs. 37.1%, P = 0.02), while the subjects showing deterioration were higher in the ANCA-negative group (40.7% vs. 14%, P = 0.006). However, the mortality rate at the end of 6 months was similar in the two groups (Table 5). Out of six patients who died in the ANCA-negative International Journal of Rheumatic Diseases 2016; 19:

5 A. Sharma et al. Table 4 Investigations at baseline and 6 months in the two groups Variables ANCA-positive, mean (SD) ANCA-negative, mean (SD) Baseline 6 months P-value Baseline 6 months P-value Hemoglobin (g/l) 85.6 (2.5) (2.9) (3.2) 90.0 (2.9) Blood urea (mmol/l) 47.2 (3.3) 25.7 (3.3) (4.1) 30.8 (3.3) Serum creatinine (lmol/l) (63.2) (44.5) (80.4) (44.7) egfr (ml/min/1.73 m 2 ) 20.9 (2.9) 47.0 (6.1) (3.8) 42.5 (6.1) Serum albumin (g/l) 32.9 (0.9) 37.5 (1.1) (1.2) 34.5 (1.1) h urinary protein (g/tv) 1.75 (0.2) 1.32 (0.1) (0.3) 1.17 (0.1) Hematuria (%) ANCA, anti-neutrophilic cytoplasmic antibodies; egfr, estimated glomerular filtration rate; TV, total volume. Bold values indicate significant P values. group, three succumbed due to severe infection, one died due to diffuse alveolar hemorrhage, while cause of death was not clear in other two patients. Of the 23 subjects in the ANCA-positive group who required renal replacement therapy at presentation, 16 (69.6%) became dialysis-independent with treatment, while in the ANCA negative group, only six out of 19 subjects (31.6%), requiring renal replacement therapy at presentation became dialysis-free (P = 0.009). The cumulative survival was similar in the two groups (Fig. 1), while the probability of reaching end-stage renal disease was significantly higher in the ANCA-negative group (Fig. 2). On multivariable linear regression analysis, serum creatinine was the only independent predictor of improvement, as well as progression to ESRD in both ANCA-positive and negative subjects (odds ratio: overall, 1.32; ANCA-positive, 1.37; ANCA-negative, 1.24), while none of the histological class or other parameters were significant. Table 5 Treatment outcomes at 6 months N (%) ANCApositive ANCAnegative P- value Improved 27 (62.8) 10 (37.1) 0.02 Deteriorated 6 (14.0) 11 (40.7) Death 10 (23.2) 6 (22.2) 0.44 Renal replacement therapy 23 (45.1) 19 (57.6) 0.14 (RRT) at presentation Dialysis-independent 16 (69.6) 6 (31.6) Dialysis-dependent 7 (30.4) 13 (68.4) ANCA, anti-neutrophilic cytoplasmic antibodies. Bold values indicate significant P values. DISCUSSION Although ANCA positivity is noted in > 90% of AAV cases, studies focusing on PIGN have observed negative ANCA in about % cases Although the initial studies were hampered by lack of commercial availability of ELISA kits; even with improvement in technology and wider availability of ELISA kits, recent studies have demonstrated a negative rate of 32 37%. 8,9 Thus, the ANCA negativity rate of 39.3% observed in our study is comparable to previous studies. The exact reason of this high rate of negative ANCA in PIGN is not clear but may allow one to think of the alternative pathogenic mechanisms of the renal injury. Most studies on ANCA-negative PIGN have observed that the majority of these cases are limited to kidneys Figure 1 Kaplan Meier survival curves according to anti-neutrophilic cytoplasmic antibodies (ANCA) status (P = not significant). 78 International Journal of Rheumatic Diseases 2016; 19: 74 81

6 ANCA-negative pauci-immune glomerulonephritis Figure 2 Kaplan Meier curves showing probability of developing end-stage renal disease (ESRD) according to anti-neutrophilic cytoplasmic antibodies (ANCA) status (P = 0.009). with lower incidence of extra-renal symptoms, 8 11 a point which is further substantiated by our study. It may be possible that either the functional effects of ANCA epitopes are different or they are present in very low titers. However, other auto-antibodies have also been implicated in the pathogenesis of such subjects. 17 These include, but are not restricted to anti-moesin antibodies, 18 anti-endothelial cell antibodies 19 and antihuman lysosome-associated membrane protein-2 (hlamp-2) antibodies. 20 In animal studies, the antigens for these antibodies are shown to be expressed specifically on the glomerular endothelium, and binding of antibodies to these antigens leads to activation of neutrophils, apoptosis and renal vasculitis However, the exact role of these antibodies in humans or their cause and effect relationship has not been elucidated to date. Similar to uncertain pathogenesis, the significance of ANCA negativity in PIGN is also not clear. While some studies suggest that these subjects fare worse than the ANCA-positive subjects, 8,9 others do not agree. 11 Moreover, previous studies have predominantly focused only on crescentic PIGN. A new histological classification was proposed in 2010, 13 where PIGN was classified into four classes as per renal histology, that is, focal, crescentic, mixed and sclerotic. Various studies have validated this classification and shown it to be of help in prognosticating the patient and renal outcome, with focal class being the best and sclerotic being the worst However, there is no good data on ANCAnegative PIGN in the light of this prognostic classification. In a small retrospective study, Minz et al. did not observe any difference in the histological classes among ANCA-positive and negative patients. 10 However, in our study we observed that PIGN subjects with negative ANCA were more likely to have crescentic class on histopathology as compared to ANCA-positive subjects. We also noted a higher serum creatinine at presentation, significantly more cellular crescents and interstitial fibrosis and significantly less interstitial edema in ANCA-negative PIGN. All these features indicate a more severe and advanced disease. In a large study from China including crescentic PIGN, Chen et al. have shown a higher percentage of cellular crescents and severe interstitial fibrosis in those with negative ANCA, 8 while in another study, ANCA-negative PIGN was shown to have less crescent formation but more chronic glomerular lesions. 9 We hypothesize that due to relative lack of significant extra-renal features, these patients may present late to the clinicians, leading to delay in diagnosis and renal biopsy, thus allowing the disease to become more advanced and chronic. However, other potential reasons like differences in pathogenesis or other antibodies may also contribute to this increased chronicity. Studies from China and Taiwan observed that ANCAnegative PIGN were younger. 8,9 In our study, although the ANCA-negative subjects were younger than the ANCA-positive subjects, the difference was not statistically significant. The mean BVAS at presentation in the ANCA-negative group was significantly less compared to ANCA-positive subjects. This finding is similar to that by Chen et al. 8 and may be due to the paucity of extrarenal manifestations in ANCA-negative subjects. The baseline renal parameters were similar in the two groups, except for a non-significant higher serum creatinine in ANCA-negative PIGN. Previous studies are controversial in terms of differences in renal presentation of the ANCA-negative PIGN, with some observing a higher degree of proteinuria, 8 while others noted similar renal manifestations. 9,11 ANCA-negative PIGN subjects in our study had poor renal outcomes at the end of 6 months with only 37.1% of subjects showing improvement as compared to 62.8% in the ANCA-positive group. However, there was no difference in the overall survival of patients in the two groups. An advanced degree of renal involvement associated with significant chronicity may be International Journal of Rheumatic Diseases 2016; 19:

7 A. Sharma et al. responsible for poor renal outcome, but comparable patient survival in ANCA-negative patients. Although Eisenberger et al. in their small study have concluded that renal outcomes are not different according to ANCA status; 11 results of other studies were similar to ours in terms of poor renal survival and comparable patient survival in ANCA-negative patients. 8,9 In a recent study on AAV with severe renal failure, Lee et al. noted that both low egfr at baseline and high chronicity score on renal biopsy were associated with lower treatment response, 24 while in our study, serum creatinine at baseline was the only significant predictor of treatment outcome irrespective of ANCA status, histological class or any other histological parameter. To summarize, we noted a significant number of PIGN patients to be negative for ANCA. A high index of suspicion and an early renal biopsy is required to make a diagnosis in such patients. Most of these subjects have severe and chronic renal histology depicting a delay in presentation and diagnosis, have worse renal function and fare worse as compared to ANCA-positive subjects. However, these results should be noted in the light of several limitations of our study. This was a single-center study with all subjects belonging to the same race. The majority of subjects were evaluated retrospectively, leading to an error of bias and inadequate follow-up. Although the treatment consisted of steroids and cyclophosphamide, the protocols were not the same in all the patients. Similarly, plasmapheresis was used in only a few patients. The results of our study thus need to be validated by large multicentric studies involving subjects from various regions and races. Transparency declarations: none to declare. The results presented in this paper have not been published previously in whole or part, except in abstract format. AUTHORS AND CONTRIBUTORS All the authors meet the four criteria defined by ICMJE. They have made substantial contributions to the conception or design of the work, or the acquisition, analysis, or interpretation of data for the work. The authors have contributed to the draft preparation and revising it critically for important intellectual content. All the authors have approved the final version of the draft and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. In addition, MR conceptualized and designed the study; AS and GN recruited the patients; MR and GN performed the data acquisition and statistical analysis; MR and AS supervised the patient recruitment and contributed to patient management and end-point assessment; RN analyzed the biopsy; RWM performed the ANCA testing; MR, AS and VS drafted the manuscript; AS, RN and GN helped in data acquisition, data analysis, and literature search; and VS, KLG and HSK supervised patient management, edited the manuscript and provided intellectual support. Both AS and MR have contributed equally to the study and should be considered as first authors. REFERENCES 1 Falk RJ, Jennette JC (1997) ANCA small-vessel vasculitis. J Am Soc Nephrol 8, Mittal T, Rathi M (2014) Rheumatological diseases and kidneys: a nephrologist s perspective. Int J Rheum Dis 17, Jennette JC, Falk RJ (1990) Antineutrophil cytoplasmic autoantibodies and associated diseases: a review. Am J Kidney Dis 15, de Lind van Wijngaarden RA, Hauer HA, Wolterbeek R et al. (2006) Clinical and histologic determinants of renal outcome in ANCA-associated vasculitis: a prospective analysis of 100 patients with severe renal involvement. J Am Soc Nephrol 17, Hauer HA, Bajema IM, van Houwelingen HC et al. (2002) Renal histology in ANCA-associated vasculitis: differences between diagnostic and serologic subgroups. Kidney Int 61, van der Woude FJ, Rasmussen N, Lobatto S et al. (1985) Autoantibodies against neutrophils and monocytes: tool for diagnosis and marker of disease activity in Wegener s granulomatosis. Lancet 1, Hoffman GS, Specks U (1998) Antineutrophil cytoplasmic antibodies. Bull Rheum Dis 47, Chen M, Yu F, Wang S, Zou W, Zhao M, Wang H (2007) Antineutrophil cytoplasmic autoantibody-negative pauciimmune crescentic glomerulonephritis. J Am Soc Nephrol 18, Hung PH, Chiu YL, Lin WC et al. (2006) Poor renal outcome of antineutrophil cytoplasmic antibody negative pauci-immune glomerulonephritis in Taiwanese. J Formos Med Assoc 105, Minz RW, Chhabra S, Joshi K et al. (2012) Renal histology in pauci-immune rapidly progressive glomerulonephritis: 8-year retrospective study. Indian J Pathol Microbiol 55, Eisenberger U, Fakhouri F, Vanhille P et al. (2005) ANCAnegative pauci-immune renal vasculitis: histology and outcome. Nephrol Dial Transplant 20, Hedger N, Stevens J, Drey N et al. (2000) Incidence and Outcome of pauci-immune rapidly progressive glomeru- 80 International Journal of Rheumatic Diseases 2016; 19: 74 81

8 ANCA-negative pauci-immune glomerulonephritis lonephritis in Wessex, UK: a 10-year retrospective study. Nephrol Dial Transplant 15, Berden AE, Ferrario F, Hagen EC et al. (2010) Histopathologic classification of ANCA-associated glomerulonephritis. J Am Soc Nephrol 21, vasculitis.org [internet]. Birmingham Vasculitis Activity Score (version 3). Available from: 15 Levey AS, Coresh J, Greene T et al. (2006) Using standardized serum creatinine values in the modification of diet in renal disease study equation for estimating glomerular filtration rate. Ann Intern Med 145, Kidney Disease: Improving Global Outcomes (KDIGO) glomerulonephritis work group (2012) KDIGO clinical practice guidelines for glomerulonephritis. Kidney Int Suppl 2, Furuta S, Jayne DRW (2013) Antineutrophil cytoplasm antibody-associated vasculitis: recent developments. Kidney Int 84, Nagao T, Suzuki K, Utsunomiya K et al. (2011) Direct activation of glomerular endothelial cells by anti-moesin activity of anti-myeloperoxidase antibody. Nephrol Dial Transplant 26, Cong M, Chen M, Zhang JJ, Hu Z, Zhao MH (2008) Antiendothelial cell antibodies in antineutrophil cytoplasmic antibodies negative pauci-immune crescentic glomerulonephritis. Nephrology 13, Kain R, Tadema H, McKinney EF et al. (2012) High prevalence of autoantibodies to hlamp-2 in anti-neutrophil cytoplasmic antibody-associated vasculitis. J Am Soc Nephrol 23, Naidu GS, Sharma A, Nada R et al. (2014) Histopathological classification of pauci-immune glomerulonephritis and its impact on outcome. Rheumatol Int 34, Iwakiri T, Fujimoto S, Kitagawa K et al. (2013) Validation of a newly proposed histopathological classification in Japanese patients with anti-neutrophil cytoplasmic antibody-associated glomerulonephritis. BMC Nephrol 14, Chang D, Wu L, Liu G, Chen M, Kallenberg CGM, Zhao MH (2012) Re-evaluation of the histopathological classification of ANCA-associated glomerulonephritis: a study of 121 patients in a single centre. Nephrol Dial Transplant 27, Lee T, Gasim A, Derebail VK et al. (2014) Predictors of treatment outcomes in ANCA-associated vasculitis with severe kidney failure. Clin J Am Soc Nephrol 9, International Journal of Rheumatic Diseases 2016; 19: