Statistics from Altmetric.com
- autoimmune diseases
- granulomatosis with polyangiitis
- systemic vasculitis
- patient reported outcome measures
We read the published article by Gercik et al1 with great interest. In their retrospective study, they tested the existing classification systems to predict the progression to end-stage renal disease of patients with renal involvement by anti-neutrophil cytoplasmic antibody-associated vasculitis (AAV), demonstrating a better performance of the AAV renal risk score (ARRS) proposed by Brix et al2 as compared with the 4-tiered glomerulocentric histological Berden’s classification.3 They suggested that the employment of baseline glomerular filtration rate in the ARRS can partly represent a possible explanation for these results. However, the evaluation of extra-glomerular histological parameters that strongly correlate with the renal outcome4 (eg, interstitial fibrosis/tubular atrophy (IFTA)), can play a further role in the improvement of the ARRS performance. In this setting, many other classifications demonstrated the putative role of disparate histological features to predict the outcome of patients with primary (eg, IgA nephropathy5) and secondary (eg, lupus nephritis6) renal diseases, suggesting the possibility to further increase the prognostic role of the existing classification for AAV.
We retrospectively evaluated the performances of the currently used systems and investigated whether additional histological features can improve prognostic workflow of AAV. For this purpose, a retrospective, multicentric series of AAV cases have been reviewed. Each case has been independently evaluated by two renal pathologists, and classified according to the Berden’s scheme and ARRS. Additional glomerular, tubulointerstitial and vascular lesions have been recorded for each case, following the previously provided definitions.7 8 The outcome of interest was time to need for renal replacement therapy (RRT) or death, whatever occurred first. Cox proportional hazards regression models were constructed with time to composite event, loss to follow-up or censoring (30 June 2019). Time at risk started at the date of renal biopsy. The histological features, collapsed into binary variables (0 to 1=low; 2 to 3=high) and subdivided as ‘active’ (cellular/fibrocellular crescents, endocapillary hypercellularity, fibrinoid necrosis and Bowman’s capsule rupture (BCR)) and ‘chronic’ (global glomerulosclerosis, fibrous crescents, segmental glomerulosclerosis, IFTA and arteriosclerosis) have been evaluated individually and in association with the currently proposed systems to assess their ability to predict the outcome. Univariate and multivariate models (HRs and 95% CIs) have been used to assess the prognostic performance of Berden's class/ARRS alone and with additional predictors (Harrell’s c-statistic).
After the selection of cases with available renal biopsy, complete clinical data (at least 6 months of follow-up) and positive anti-neutrophil cytoplasmic antibody (ANCA), 52 patients have been analysed (30 (58%) males, median age 68 years (IQR 58 to 75)). ANCA showed myeloperoxidase (MPO) specificity in 31 (60%) and PR3 in 21 (31%), with a median titre of 101 U/mL (IQR 55 to 264). Six patients (12%) required dialysis at the time of the diagnosis. After the biopsy, 47 (90%) patients were treated with corticosteroids, 34 (66%) with additional immunosuppression and/or plasmapheresis (8 (15%)). During a median follow-up of 31 months (1828 person-months), 13 composite events developed (8 deaths, 5 RRT). Sixteen (31%) cases were classified as Focal, 8 (15%) as Crescentic, 24 (46%) as Mixed and 4 (8%) as Sclerotic. Patients were grouped as low (n=21, 40%), medium (n=24, 46%) and high risk (n=7, 13%) based on ARRS. Among the histological predictors tested (table 1), only BCR was significantly associated with the outcome at the univariate analysis (p=0.023). Its addition (figure 1) to the model, which included only Berden's class (c=0.67) or ARRS (c=0.62), significantly improved the prognostic performance (c=0.76 and 0.73, respectively). This has been confirmed in the multivariate model which includes Berden's class (HR 3.61, 95% CI 1.15 to 11.34; p=0.028) and ARRS (HR 5.25, 95% CI 1.53 to 18.08; p=0.009). The present study demonstrates an improved performance of prognostic systems in predicting AAV outcome after the implementation of BCR, the additional predictive role of which can partly lie in the irreversible loss of nephrons consequent to the segmental glomerulosclerosis caused by BCR.9 Further investigations on a larger prospective cohort are required to confirm these results.
Contributors VL ideated the study; VL and GV provided the retrospective histological assessment of renal biopsies; MF performed the statistical analysis; VL and FP drafted the paper; all the authors approved and revised the final version of the manuscript.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting or dissemination plans of this research.
Patient consent for publication Not required.
Provenance and peer review Not commissioned; internally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.