Background/Purpose: ANCA-associated vasculitides (AAVs) are the most frequent cause of rapidly progressive glomerulonephritis (RPGN), for which the major prognostic issue is the risk of developing end-stage renal disease (ESRD) or dying. The large, multicenter, prospective PEXIVAS trial, including patients with severe AAVs, i.e. lung hemorrhage and/or renal failure (estimated glomerular filtration rate (eGFR) < 50 ml/min/1.73 m2),showed that plasma exchanges (PLEX) did not lower the ESRD or death risk for AAV patients, but the prognostic impact of renal histology on the renal outcome and response to PLEX was not assessed.We aimed to evaluate whether histopathological findings could predict the renal outcome of PLEX-treated AAV patients.

Methods: This retrospective, multicenter study included patients with microscopic polyangiitis, granulomatosis with polyangiitis or renal-limited vasculitis, fulfilling ACR criteria or Chapel Hill Consensus Conference definitions. All patients had a renal biopsy and were treated with PLEX. Two histopathological classifications were evaluated from renal biopsy reports: Berden’s, distinguishing 4 categories (focal, crescentic, mixed and sclerotic), using the percentages of different glomeruli lesions; and Brix’s, discerning 3 risk groups (low, medium or high), using the percentages of normal glomeruli, tubular atrophy and interstitial fibrosis, and baseline eGFR. The primary endpoint was dialysis independence and survival at month (M) 12. Secondary endpoints were eGFR >30 ml/min/1.73 m²at M12, and delta eGFR >15 ml/min/1.73 m²from baseline to M12. 

Results: We included 163 patients from 19 centers: 99 (61%) men; 92 (56%) MPO-ANCA+, 65 (40%) PR3-ANCA+; mean±SD baseline serum creatinine 555±274 µmol/L, 62 (38%) with alveolar hemorrhage; 139 (85%) given cyclophosphamide and 30 (18%) received rituximab; 74 (45%) required dialysis at baseline. Mean number of PLEX was 7.0±2.3.
Berden categories were focal (19%), crescentic (39%), mixed (20%) and sclerotic (21%). Brix risk groups were low (11%), medium (48%) and high (41%). Both Berden and Brix histopathological classifications on baseline renal biopsies were associated with the primary endpoint, i.e. dialysis independence and survival at M12 (χ2 24.9; P< 0.0001, and χ2 11.7; P=0.003, respectively).In contrast, baseline serum creatinine levels did not differ betweenfavorable and poor-prognosis groups (respectively:526±252 µmol/L vs. 597±300 µmol/L; P=0.11). Secondary endpoints were also associated with Berden and Brix classifications, respectively: achievement of eGFR >30 ml/min/1.73 m2at M12 (χ2 14.0; P=0.003, and χ2 17.3; P=0.0002); andimproved renal function with delta eGFR >15 ml/min/1.73 m2from baseline to M12 (χ2 18.1; P=0.0004, and χ2 17.7; P=0.0001).

Conclusion: In a cohort of AAV patients treated with PLEX for RPGN, our findings suggest that renal biopsy classification at diagnosis strongly predicted renal outcome at M12. AAV patients with RPGN should undergo renal biopsy to help physicians identify who among them would benefit the most from PLEX.

« Back to ACR/ARP Annual Meeting 2019

ACR Meeting Abstracts - https://acrabstracts.org/abstract/renal-histopathological-classifications-predict-the-renal-outcomes-of-plasma-exchange-treated-anca-associated-vasculitides-with-renal-failure/