Anti-neutrophil cytoplasmic antibody associated glomerulonephritis complicating treatment with hydralazine

Hydralazine, a widely used therapy for hypertension and heart failure, can elicit autoimmune disease, including anti-neutrophil cytoplasmic antibody associated glomerulonephritis (ANCA-GN). We identified 80 cases of ANCA-GN complicating treatment with hydralazine, accounting 4.3% (80/1858 biopsies) diagnosed between 2006 2019. Over three-fourths patients were on hydralazine at least one year, mean daily dose approximately 250 mg/day. ANCA testing revealed p-ANCA/myeloperoxidase-ANCA seropositivity in 98%, 39% dual cANCA/anti-protinase 3-ANCA positivity, often accompanied by anti-nuclear (89%), anti-histone (98%), hypocomplementemia (58%). Kidney biopsy necrotizing crescentic glomerulonephritis, similar to primary ANCA-GN, but significantly less frequently pauci-immune (77 vs. 100%) more commonly mesangial hypercellularity (30 5%), electron dense deposits (62 20%), endothelial tubuloreticular inclusions (11 0%); all significant differences. On follow-up, 42 51 received induction immunosuppression: 19 reached the combined end-points kidney failure or death 32 had creatinine 1.49 mg/dL last follow-up. Thus, hydralazine-associated exhibits overlapping clinical pathologic features mild immune complex resembling lupus nephritis. With discontinuation immunosuppression, outcomes are ANCA-GN. Treatment exposure certain drugs generation autoantibodies disease affecting kidneys, complex–mediated (GN) membranous glomerulopathy.1Hogan J.J. Markowitz G.S. Radhakrishnan J. Drug-induced glomerular disease: immune-mediated injury.Clin J Am Soc Nephrol. 2015; 10: 1300-1310Crossref PubMed Scopus (46) Google Scholar Clinical awareness drug-induced is important given need discontinue offending agent management these conditions. Drugs anti–neutrophil (ANCA)–associated vasculitis (AAV) include propylthiouracil, cocaine adulterated levamisole, tumor necrosis factor-α inhibitors, hydralazine.1Hogan Scholar, 2Stokes M.B. Foster K. et al.Development during anti-TNF-alpha rheumatoid arthritis.Nephrol Dial Transplant. 2005; 20: 1400-1406Crossref (187) 3Pendergraft W.F. Niles J.L. Trojan horses: drug culprits antineutrophil autoantibody (ANCA) vasculitis.Curr Opin Rheumatol. 2014; 26: 42-49Crossref (88) ScholarEditor’s NoteThis study perhaps largest most detailed investigation histopathologic antibody–associated occurring setting use. Hydralazine has also been lupus-like syndromes, several described here seemed have both AAV lupus. affects DNA methylation B-cell tolerance, fosters NETosis, which could contribute autoimmunity susceptible patients. Given array effective, possibly toxic, anti-hypertensive medications available today, Editorial Board wondered whether was very much anymore. An informal poll among nephrology colleagues showed considerable variation use worldwide. In Canada, Unites States, Australia, it frequently, cardiologists when renin-angiotensin system blockers contraindicated. It rarely Mexico, Argentina, mainland China, Hong Kong, Singapore, Japan. India, mainly congestive chronic disease. not France, Czech Republic, Germany, only Netherlands, United Kingdom, Russia. suspect that will continue be found, maybe so cardiovascular nephrologists keep this mind as we consult our cardiology colleagues. This an arterial vasodilator, failure. autoimmunity, lupus, long-recognized potential complication occurs 5% 10% receiving characterized development (ANAs) antibodies symptoms myalgias, fever, serositis.4Hess E. Drug-related lupus.N Engl Med. 1988; 318: 1460-1462Crossref (171) Renal involvement uncommon.5Yokogawa N. Vivino F.B. Hydralazine-induced comparison idiopathic ANCA-positive vasculitis.Mod 2009; 19: 338-347Crossref appears rare complication, <100 reported date.6Kumar B. Strouse Swee M. al.Hydralazine-associated vasculitis: vasculitis.Semin Arthritis Rheum. 2018; 48: 283-287Crossref (16) However, true incidence unknown. series, 10 30 highest titers anti-myeloperoxidase (MPO) exposed hydralazine.7Choi H.K. Merkel P.A. Walker A.M. Drug-associated antibody-positive prevalence high antimyeloperoxidase antibodies.Arthritis 2000; 43: 405-413Crossref (282) may show overlap.6Kumar The course dominated rapidly progressive GN. Extrarenal pulmonary involvement, rare. addition high-titer MPO-ANCA, serologic evaluation typically reveals positive ANA hypocomplementemia.7Choi Descriptions biopsy-proven largely limited case reports small reveal GN.5Yokogawa systematic descriptions histologic, immunofluorescence (IF), ultrastructural findings, well comparisons long-term lacking. Herein, report while series date, providing clinical-pathologic findings compare key control group All native biopsies diagnosis ANCA-associated GN accessioned Columbia University Pathology Laboratory from 2019 reviewed history preceding renal biopsy. Inclusion criteria included: (i) predominant light microscopic GN; (ii) seropositivity; (iii) hydralazine. Exclusion included known AAV, systemic erythematosus, other autoimmune/connective tissue before commencement time initial presentation acute injury, (including cocaine, anti–tumor therapy). Seventy-five consecutive 2016 showing absence documented utilized select demographic findings. Biopsies excluded group, IF evidence anti–glomerular basement membrane processed according standard techniques microscopy, IF, microscopy interpreted 1 6 pathologists. For 3-μm frozen sections stained fluorescein isothiocyanate–conjugated rabbit anti-human IgG, IgM, IgA, C3, C1q, κ λ chain (Dako). submitting provided following information: patient demographics, medical history, extrarenal duration use, biopsy, requirement replacement presentation, serum urine protein-to-creatinine ratio 24-hour protein level, presence hematuria (>5 cells/high-power field), antibody, antibodies, double-stranded complements, anticoagulant, anti-cardiolipin antibodies. addition, follow-up information regarding creatinine, repeated testing, type obtained. parameters collected: pattern injury; number total globally sclerotic glomeruli; percentage glomeruli cellular fibrocellular crescents; (iv) fibrous (v) cortex interstitial fibrosis tubular atrophy; (vi) degree inflammation (scale, 0–3; none indicates 0; (0%–25%), 1; moderate (26%–50%). 2; severe (>50%), 3); (vii) vascular sclerosis 0–3); (viii) coexisting process; (ix) intensity staining 0–3+); (x) location electron-dense deposits; (xi) inclusions. Histopathologic classifications assigned scheme Berden al.8Berden A.E. Ferrario F. Hagen E.C. al.Histopathologic classification glomerulonephritis.J 2010; 21: 1628-1636Crossref (463) Pauci-immune defined ≤1+ reactants, scale graded 0 3+. Baseline characteristics presented descriptively, ± SD continuous variables categorical variables. compared using 2-sample Wilcoxon rank-sum (Mann-Whitney) test Fisher exact Analyses performed STATA version 12.1 (StataCorp). approved Institutional Review Irving Medical Center. who developed whom taking practice. data Table 1. age 69 years, 61% female. cohort self-identified 69% White, 16% Black, 6% Latinx. Duration 47 (59%) length <12 months 11 (23%; minimal duration, 5 months), 12 60 29 (62%), >60 7 (15%). Daily doses patients, 254 88 mg/d.Table 1Clinical glomerulonephritisDemographic parametersResultsDemographics Age, yr69.3 10.7 Female sex49 (61.3) Race/ethnicityWhite55 (68.8)Black13 (16.3)Latino5 (6.3)Asian0 (0)Other1 (1.3)Unknown6 (7.5)Laboratory serologies Serum mg/dl4.3 2.6 Urinary protein, g/d g/g2.6 3.1 Hematuria80 (100) (n = 74)66 (89.2) dsDNA Ab 49)22 (44.9) Anti-histone 46)45 (97.8) Hypocomplementemia 66)38 (57.6) ANCAp-ANCA MPO only47 (58.8)c-ANCA PR3 only2 (2.5)Dual (+)31 (38.8)Duration therapy, mo (N 47) <1211 (23.4) 12–6029 (61.7) >607 (14.9)Ab, antibody; ANA, ANCA, c-, cytoplasmic; dsDNA, DNA; MPO, myeloperoxidase; p-, perinuclear; PR3, proteinase 3.Data n (%). Open table new tab Ab, 3. Data Mean 4.3 mg/dl. Thirteen (16%) dialysis dependent presentation. proteinuria g/g collection spot ratio. detected 89% (66 74) 45% (22 49) tested respectively. nearly (45 46 [98%]). present 58% (38 66) 28 depressed C3 C4, 8 only, 2 C4 only. Lupus anticoagulant (50%) 16 (69%) positive. Specifically, p-ANCA and/or 31 (39%) positive, (2.5%) c-ANCA 3 (PR3) negative antibody. At 14 (18%) radiologic hemoptysis), (3.8%) purpuric skin rash, upper airway tract vasculitis. (Table 28Berden Scholar), dominant injury (Figure 1a), global 24%, cellular/fibrocellular crescents 23%, 4.5%. “intact glomeruli” (devoid crescents) 48%. According al.,8Berden 6.3% class, 51.3% focal 11.3% 31.3% mixed class Scholar). fibrinoid 16.5%, generally involving crescents. Mesangial 24 (30%) 1b). Obliterative endocapillary proliferation (7.5%) atrophy (0%–25%) 33 (41%), (26%–50%) 36 (45%), (>50%) (14%). overall 32%, 1.6 (graded 0–3), arteriosclerosis 1.9 0–3). Necrotizing arteritis seen (3.8%). Findings superimposed background changes diabetic glomerulosclerosis 15 (19%).Table 2Pathology subjects glomerulonephritisPathologic parametersResultsLight Globally sclerotic, %24.3 16.0 Normal glomeruli, %47.8 22.3 Glomeruli crescents, %23.3 20.5 %4.5 8.7 arteritis3 (3.8) Interstitial (score, 0–3)1.6 0.8 hypercellularity24 (30) Endocapillary proliferation6 (7.5) Membranous2 (2.5) IFTA, %32.1 19.4 Arteriosclerosis 0–3)1.9 0.9 Underlying NDGS15 (18.8)ImmunofluorescencePauci-immune62 (77.5) Positive immunofluorescence18 (22.5)IgG, no. (mean intensity)16 (1.9)IgM, (1.6)IgA, intensity)4 (1.8)C3, intensity)18 (1.7)C1, intensity)5 (1.4) Full house0 (0) Extraglomerular deposits5 (6.3)Electron % deposits46 (62.2)Mesangial45 (60.8)Subendothelial21 (28.4)Subepithelial13 (17.6) TRIs8 (10.8) classification8Berden ScholarSclerotic5 (6.3)Focal41 (51.3)Crescentic9 (11.3)Mixed25 (31.3)ANCA, NDGS, nodular glomerulosclerosis; TRI, inclusion.Data (%), unless otherwise indicated. inclusion. By 62 (78%) classified pauci-immune, 40 trace 1+ reactant, IgG (65% [26 40]), IgM (83% [33 IgA (20% [8 (70% [28 C1 (23% [9 40]). remaining 18 >1+ 0–3+) (16 [89%]; intensity, 0.9), 0.7), (4 [22%]; 1.8 1.1), (18 [100%]; 1.7 (5 [28%]; 1.4 0.6) 1c). Most codominant. None full-house linear typical Five (6.3%) displayed granular immune-type complement tubulointerstitial compartments, pauci-immune. Electron 74 biopsies, (62%) detectable deposits. frequent (61% [45 74]) ranged segmental 1d). Subendothelial (28% [21 subepithelial (18% [13 common always segmental. Among 55 23 (42%). Endothelial (i.e., “interferon footprints”) (11%). 75 single year There no differences sex distribution 3). Compared higher rate positivity (89% 19%; P < 0.001) (58% 9.3%; 0.001), likely (39% 3%; c-ANCA/PR3 (2.5% 29%; 0.001). al. system.8Berden (30% 5.3%; infrequent common; however, did reach statistical significance (7.5% 1.3% [P 0.06] 2.5% 0% 0.2], respectively). (78% 100%; (62% 20%; Tubuloreticular inclusions, although infrequent, (11% 0%; 0.005). 62) those 18). Non–pauci-immune rates (90% 89%; 1.0), (50% 35%; 0.2), (59% 57%; 0.9) met non–pauci-immune demonstrated (55% 22%; 0.005) than cases.Table 3Hydralazine versus ANCA-GNVariableHydralazine 80)ANCA-GN 75)P valueAge, 10.766.5 16.10.2Female (61.3)46 (61.3)1.0Race/ethnicity<0.001 White55 (68.8)34 (45.3) Black13 (16.3)6 (8.0) Latino5 (6.3)8 (10.7) Asian0 (0)1 (1.3) Other1 (1.3)0 Unknown6 (7.5)26 (34.7)Serum 2.64.2 3.10.9ANA66 (89.2)14 (18.7)<0.001Hypocomplementemia38 (57.6)7 (9.3)<0.001ANCA<0.001 Ab47 (58.8)51 (68.0) Ab2 (2.5)22 (29.3) Dual (38.7)2 (2.7)Globally 16.025.0 22.40.9Normal 22.340.2 25.20.05Glomeruli 20.526.9 21.10.3Glomeruli 8.78.2 11.50.03Necrotizing (3.7)11 (14.7)0.02Interstitial 0.82.0 0.90.01Mesangial (30)4 (5.3)<0.001Endocapillary (7.5)1 (1.3)0.06Membranous2 (2.5)0 (0)0.2IFTA, 19.433.5 25.40.7Arteriosclerosis 0.91.4 0.90.003Pauci-immune IF62 (77.5)75 (100)<0.001Deposits EM46 (62.2)15 (20)<0.001TRIs EM8 (10.8)0 (0)0.005ANCA Scholar<0.001 Sclerotic5 (6.3)11 (14.7) Focal41 (51.3)29 (38.7) Crescentic9 (11.3)12 (16.0) Mixed25 (31.3)23 (30.7)Ab, EM, microscopy; GN, glomerulonephritis; immunofluorescence; 3; Follow-up 4) (64%), median 15.2 months. discontinued Forty-two immunosuppressive corticosteroids 41 (80%), cyclophosphamide 13 (26%), rituximab (57%), plasmapheresis 9 (18%), mycophenolate mofetil (3.9%). 27 rituximab, cyclophosphamide, agents, neither agents. Eight maintenance azathioprine (12%) During period (27%) progressed end-stage (ESKD), (21%) expired, ESKD death. (37%) end point ESKD. (range, 1.18–1.97) followed (22%) hypocomplementemia. Sixteen (53%) remained (range) 237 (25–1144) days, positivity. (33%) 1.5 mg/dl 1.3–5.0 mg/dl), di