Prognostic factors and outcomes in anti-neutrophil cytoplasmic antibody-associated glomerulonephritis: A retrospective single-center study from southern China

Objective: To investigate the prognostic factors and outcomes in patients with anti-neutrophil cytoplasmic antibody-associated glomerulonephritis (ANCA-GN) in Southern China Methods: A retrospective analysis of medical records of patients with ANCA-GN admitted to Shenzhen Hospital of Southern Medical University and Nanfang Hospital of Southern Medical University between September 2011 and September 2021 was performed. The clinical presentation, biological, and renal pathology were collected. In addition, the risk factors for end-stage renal disease (ESRD) and short-term overall survival in patients with ANCA-GN were analyzed. Results: A total of 93 patients with ANCA-GN were included in the study. Of them, 91.4%, were perinuclear anti-neutrophil cytoplasmic antibodies (MPO-pANCA)-positive. Approximately one-quarter (24.7%) of patients had progressed to ESRD, and 7.5% died within six months. Most patients presented with hematuria (94.6%), proteinuria (78.5%), elevated serum creatinine (86.0%), anemia (90.3%), and increased erythrocyte sedimentation rate (ESR) (44.1%). The majority (94.6%) of patients presented with crescent formations at histopathological examination. Serum creatinine, hemoglobin, and Birmingham vasculitis activity score (BVAS) were all independent factors for ESRD (P<0.05). Moreover, while ANCA renal risk score (ARRS) has an impact on prognosis of nephropathy, it did not influence ESRD independently (P>0.05). The effect of Berden’s histopathologic classification on ESRD has not been confirmed. Age at onset, ESR and cardiovascular involvement were all independent factors affecting short-term overall survival of patients with ANCA-GN (P<0.05). Conclusions: Serum creatinine, hemoglobin, and BVAS were all independent risk factors of ESRD, while ARRS and Berden’s histopathologic classification were not. Age at onset, ESR, and cardiovascular involvement were independent risk factors for the overall six-month survival rate in patients with ANCA-GN.


INTRODUCTION
Anti-neutrophil cytoplasmic antibody (ANCA)associated vasculitis (AAV) is an autoimmune disorder and characterized by relapsing and remitting necrotizing pauci-immune vasculitis that mainly affects kidneys and lungs. 1,2In case of kidney involvement, ANCAassociated glomerulonephritis (ANCA-GN) refers to AAV-induced kidney injury, which can be divided into myeloperoxidase (MPO)-ANCA-GN and proteinase-3 (PR3)-ANCA-GN. 3 The treatment of patients with AAV is improving.However, the mortality rate for patients with AAV is still 2.7 times higher than that of the general population, with most patients progressing to end-stage renal disease (ESRD). 46][7] Birmingham vasculitis activity score (BVAS) is often employed by clinicians to assess disease activity and prognosis. 8dditionally, vasculitis damage index (VDI) is used to evaluate chronic lesions with organ involvement. 9The predictive value of renal pathology for the prognosis of patients with ANCA-GN, particularly for ESRD, has been recognized by most experts. 10Nevertheless, its value in assessing the overall prognosis of the disease is still controversial.
Multiple studies have demonstrated that Berden's histopathologic classification is a reliable predictor of ESRD risk in patients with ANCA-GN. 10,11However, few data are available about how ANCA renal risk score (ARRS), proposed in 2018, 12 correlates with outcomes of ANCA-GN.According to a recent study, 13 ARRS has a better predictive value for ESRD than the histopathological classification, and a higher score is associated with a significantly higher risk of progression to ESRD.However, the study had no personalized treatment data to allow for a modeling of the association with renal outcomes.
This retrospective study aimed to investigate the predictive value of renal pathology and clinical presentation in evaluating nephropathy outcomes and mortality in patients with ANCA-GN in southern China who were assessed using the ARRS, BVAS and VDI scoring systems.

METHODS
This retrospective cohort study included patients with ANCA-GN from Shenzhen Hospital of Southern Medical University or Nanfang Hospital of Southern Medical University between September 1 st , 2011 and September 1 st , 2021.

Inclusion and Exclusion Criteria:
All AAV patients met the diagnostic criteria of Chapel Hill Consensus Conference 2012 (CHCC 2012) and had renal involvement, including abnormal urine test or elevated serum creatinine. 14And patients with AAV pauci-immune glomerulonephritis diagnosis confirmed by kidney biopsy were eligible.Patients with secondary vasculitis (such as vasculitis secondary to drugs, Henoch-Schonlein purpura, neoplasms, systemic lupus erythematosus, rheumatoid arthritis, and infection) or combined with other renal diseases (such as membranous nephropathy, diabetic nephropathy, hypertensive nephropathy) were excluded.Ethical Approval: The study received ethical approval from the Joint Research Ethics Committee, Shenzhen Hospital, Southern Medical University (Approval No: NYSZYYEC20210030 October 23, 2021) Data collection: Clinical, biological, and histological data were retrospectively collected from medical records.Patient characteristics such as age, sex, serotype of ANCA, serum creatinine (Scr) levels (µmol/L), height (m), weight (kg), proteinuria, hematuria, uric acid (UA) (μmol/L), hemoglobin (g/L), serum albumin (Alb) (g/L), ESR (mm/h), CRP (mg/L), complement, Berden's histopathologic classification, BVAS, VDI, ARRS, renal pathology, combined disease and extrarenal clinical presentation.There was a six months observation period between the time of diagnosis and ESRD or death.

Grouping and partial definitions:
The outcome of ANCA-GN were death or ESRD.Patients were grouped based on the mortality outcome (death or survival), based on the disease progression (ESRD-and a non-ESRD groups).Berden's histopathologic classification results classified patients into focal, crescentic, mixed and sclerotic classes according to the ratio of normal glomerulus, crescentic glomerulus and spherical glomerular sclerosis. 11ARRS was graded based on the conditions of normal glomeruli, renal tubular atrophy and interstitial fibrosis, and glomerular filtration rate at the time of diagnosis and divided into low-, mediumand high-risk groups according to the scores. 12ESRD was defined as chronic dialysis or kidney transplantation (>90 days).Statistical analysis: SPSS 25.0 was used for statistical analysis.The normally distributed continuous data were presented as mean ± standard deviation (SD) and compared using the Student's t test.The non-normally distributed continuous data were presented as median [interquartile range (IQR)], and compared using the Mann-Whitney U test.Categorical variables were presented as number (percentage) and compared with the χ 2 test.Comparisons between groups were performed using log-rank tests, and odds ratios (ORs) and 95% confidence intervals (CIs) were calculated.Cox regression analysis was used to calculate the hazard ratio (HR) of clinical presentation and histopathology in overall survival in the short term after adjustment.P values <0.05 were considered statistically significant.showed that age at onset, ESR and cardiovascular involvement were the risk factors for overall survival of patients with ANCA-GN within six months (P<0.05), and were all independent risk factors.Higher patient age, faster ESR, or the presence of cardiovascular involvement correlated with higher the risk of death.Surprisingly, ARRS and Berden's histopathologic classification were not independent factors for overall six months survival (P>0.05).

DISCUSSION
The results of our study showed that serum creatinine, hemoglobin, and Birmingham vasculitis activity score (BVAS) were all independent factors for ESRD (P<0.05) in patients with ANCA-GN.ARRS was not independently associated with ESRD.Age at onset, ESR and cardiovascular involvement were all independent factors affecting short-term (six months) overall survival of patients with ANCA-GN.Moreover, we found that the majority of patients with ANCA-GN were MPO-pANCA positive, and 70.1% of patients had respiratory system injury, which was similar to other studies. 2,3urthermore, patients with ANCA-GN predominantly exhibited hematuria with impaired renal function, which could be accompanied by proteinuria and increased blood pressure.Renal pathology revealed that 94.6% of patients had crescent formation, along with varying degrees of renal interstitial inflammatory cell infiltration, fibrosis, and tubular atrophy.Furthermore, 91.45% of patients had renal arteriole stenosis, while 5% had arteriolar necrosis.These renal manifestations and renal pathology are basically consistent with previous reports 1,11 that show that patients with ANCA-GN typically exhibit pauci-immune glomerulonephritis with crescent formation, which may be accompanied by inflammation and cellulose necrosis of the small vascular wall.Immunofluorescence staining typically reveals no or only minor immune complex deposition.
Previous studies have showed that infection is the main cause of mortality in AAV patients within a year, followed by vasculitis activity. 1,2Our study revealed that most patients with ANCA-GN had anemia, increased ESR and increased CRP during the initial diagnosis, indicating that the majority of patients had disease activity at diagnosis.As 61.3% of patients experience infection within six months after the diagnosis, it is crucial to ensure improved assessment procedures, infection prevention methods and individualized treatment during the course of the disease, especially during the epidemic of influenza and SARS-CoV-2. 152][13] While some studies showed the association of Berden's histopathologic classification with the overall survival of patients, [16][17][18] our logistic regression analysis found that Berden's histopathologic classification was not an independent factor for ESRD (P>0.5).The evaluation value of renal interstitial fibrosis (IF) or tubular atrophy (TA) can be used to forecast the prognosis of nephropathy. 19Brix et al. 12 proposed that the ARRS might be employed to predict the risk of ESRD among patients with ANCA-GN.Interestingly, our results indicated that ARRS may have an impact on the prognosis of ESRD (P<0.05).
Furthermore, logistic regression analysis of the ESRD and non-ESRD groups showed that serum creatinine level was an independent risk factor for ESRD in patients with ANCA-GN (OR=1.004,95% CI: 1.001-1.007,P<0.05).The GFR related to serum creatinine is one of the ARRS score components.Therefore, our results further demonstrated the effect of AARS on the prognosis of nephropathy.Compared with Berden's histopathologic classification, ARRS may have a higher predictive value for a poor prognosis in nephropathy.However, the multivariate logistic regression analysis indicated that there was no independent relationship between ARRS and ESRD (P>0.05).Interestingly, our results indicated that the abundance of fuchsinophilic protein depositions in renal pathology is associated with higher probability of progression to ESRD.However, there are currently no similar reports available, and more studies are needed to validate our results before possible clinical application.
Recent studies have also revealed that the prognosis of ANCA-GN is partially influenced by anemia, ESR, CRP, age, BVAS, and VDI. 20In our study, hemoglobin levels   and BVAS scores were both independent risk factors leading to ESRD.Patients with more severe anemia and higher BVAS scores were more likely to progress to the ESRD stage.This suggests the predictive value of these indexes for ESRD progression.

Infiltration of renal interstitial
Our results revealed that age at onset, ESR and cardiovascular involvement were all independent risk factors of overall survival of patients with ANCA-GN at six months after the diagnosis.In our study, the average onset age of patients with ANCA-GN was 58 years, which was similar to other studies 3 and higher the age of onset correlated with shorter survival time.We may speculate that this correlation is probably related to the age-related factors, such as underlying diseases, declining body condition, poor response to treatment, worse tolerance, etc.There was a correlation between the ESR and the risk of death, which might be related to the level of disease activity.Patients with cardiovascular involvement at diagnosis had a higher risk of death, which also indicates the impact of vital organ injuries on the disease prognosis.
According to this study, factors affecting ESRD prognosis and overall survival in patients with ANCA-GN were different.Berden's classification, ARRS, BVAS, and VDI score were not independent risk factors for overall survival at six months after diagnosis.In addition, sex, urine red blood cells, and CRP had no predictive value for survival and nephropathy outcomes in patients with ANCA-GN.Limitations: It includes the short follow-up time and insufficient number of enrolled patients.This may potentially affect our results and could partially explain the observed lack of predictive value of sex, urine red blood cells, and CRP.However, AAV is a multisystemic disease, and ANCA-GN is only one of the involved systems.Therefore, the extrarenal involvement system may partially affect the prognosis of patients with ANCA-GN.In addition, the disease itself, comorbidities, and side effects of treatment may also affect the overall prognosis of patients.Further studies that account for these potentially risk factors are needed.

CONCLUSION
We identified serum creatinine, hemoglobin, and BVAS as independent risk factors of ESRD.There was no correlation between Berden's histopathologic classification and ARRS, and ESRD progression.Age at onset, ESR, and cardiovascular involvement were independent risk factors for the overall six-month survival rate in patients with ANCA-GN.Further multicenter, large-sample studies with long follow-up should be conducted to explore additional risk factors for the prognosis of overall survival and ESRD in patients with ANCA-GN.

Table -
ESR, CRP, and hematuria, there was no significant difference between the ESRD and non-ESRD groups (Table-II).BVAS, VDI, and ARRS scores were higher in ESRD patients (p<0.01).ESRD patients had worse renal function at baseline and more frequently Table-I: General characteristics of patients with ANCA-GN.hyperuricemia(p=0.024).ESRD was associated with less frequent chronic nephritis syndrome (p=0.009).Moreover, the incidence of rapidly progressive nephritis was lower in the ESRD patients.ESRD patients had more serious anemia and lower C3 (p<0.05).We detected a trend for a higher ARRS in patients with ESRD, as shown in Table-III (P<0.05).Furthermore, Berden's histopathologic classification was comparable in non-ESRD and ESRD patients (P>0.05).Histopathological analysis showed significantly more fuchsinophilic protein depositions in the ESRD group of patients compared to the non-ESRD group (P=0.01).Multivariate logistic regression analysis results are shown in Table-IV and Table-V.Both blood creatinine and hemoglobin remained significant risk factors for ESRD occurrence.Higher serum creatinine levels or lower hemoglobin levels were associated with higher likelihood of disease progression to ESRD.In addition, only multivariate logistic regression analysis with a full entry model showed that BVAS score (OR=1.244,95% CI: 1.005-1.540)was an independent risk factor for ESRD (Table-IV).Likewise, only the multivariate logistic regression model using a stepwise regression method showed that fuchsinophilic protein deposition [HR 26.313 (95% CI 1.695-408.49)]was an independent risk factor for ESRD: higher rate of fuchsinophilic protein depositions correlated with more likely progression to ESRD (Table-V).Predictors of overall outcome: The outcome of renal survival in patients was not affected by ESR and age.However, Cox regression analysis (Table-VI) Values were expressed as mean ± SD for normally distributed data and median with interquartile range for non-normally distributed data, or n (%) for categorical variables.BMI, body mass index; Scr, serum creatinine; Alb, albumin; ESR, erythrocyte sedimentation rate; CRP, C-reactive protein; BVAS, Birmingham vasculitis activity score; VDI, vasculitis damage index; ARRS, ANCA renal risk score; CKD: chronic kidney disease.experienced

Table -
II: Clinical presentation of patients with ANCA-GN between ESRD and non-ESRD groups [n(%)]Table-III: Renal Pathology in ESRD and non-ESRD groups [n(%)].
distributed data, or n (%).Differences among the groups were analyzed by Student's t test for normally distributed values and by the Mann-Whitney U test for nonparametric values.Pearson's χ 2 test was employed to analyze categorical data.AAGN, anti-neutrophil cytoplasmic antibody-associated glomerulonephritis; BMI, Body mass index; BVAS: Birmingham Vasculitis Activity Score; VDI, Vasculitis damage index; ARRS, ANCA renal risk score; CKD, chronic kidney disease; ESR, erythrocyte sedimentation rate; CRP, C-reactive protein; Scr, serum creatinine; UA, uric acid.

Table -
IV: Multivariate logistic regression analysis with full entry model in patients with ANCA-GN between ESRD and non-ESRD groups.

Table -
VI: Cox regression analysis of risk factors for overall survival in patients with ANCA-GN within six months.