Clinicopathological Spectrum and Outcome of Crescentic Glomerulonephritis: A Retrospective Study from North-East India

Crescentic glomerulonephritis (CrGN) is characterized by the presence of crescents in more than 50% of glomeruli. This study aims to identify the etiology and clinicopathological features and outcomes of CrGN. In this observational study, 80 biopsy-proven CrGN were included. Patients’ demographic profile, clinical parameters, treatments, and outcomes were collected and analyzed. The mean age in our study popu- lation was 40.86 ± 16.5 years. Type II CrGN was the most common type of CrGN. Female predominance was observed in type I and type II CrGN. The highest percentage of glomeruli with crescents was seen in type I (87 ± 15.2%, P = 0.04), followed by type III and type II. At the last follow-up, mean estimated glomerular filtration rate was 25.8 ± 11.41 mL/min/1.73 m 2 and was significantly lower in type I CrGN (11.6 ± 4.8 mL/min/1.73 m 2 P = 0.001). The overall 5-year renal survival rate was 55% and was highest in type II (69.4%), followed by type III and type I (27.3%) CrGN (P = 0.0299). In our study, oliguria at the time of presentation, percentage of crescents, glomerular sclerosis, and moderate/severe IFTA were associated with poor renal outcomes. In conclusion, CrGN was seen in 5.7% of kidney biopsies in our study. Type II CrGN was the most common type of CrGN followed by type III CrGN. Renal survival was poor in type I CrGN patients compared to type II and type III CrGN. Also, oliguria, crescents, glomerular sclerosis, and moderate/severe IFTA were associated with poor renal outcomes. tubular atrophy. IFTA: interstitial fibrosis and tubular atrophy. follow-up were assessed for the outcome of our study population. At the last follow-up, the mean eGFR was 25.8 ± 11.41 mL/min/1.73 m 2 and was significantly lower in type I CrGN (11.6 ± 4.8 mL/min/1.73 m 2 ). Renal survival at the end of follow-up was seen in almost half of the patients.


Introduction
Crescentic glomerulonephritis (CrGN) is a histopathological entity characterized by the presence of crescents, usually in more than 50% of the glomeruli sampled (1). CrGN manifests clinically as a rapid decline in kidney function and is classified as rapidly proliferative glomerulonephritis (2). Disruption of glomerular basement membrane (GBM) due to the antibody or immune-complex-mediated injury leads to the accumulation of circulating leukocytes, inflammatory mediators, and coagulation factors in the Bowman's space. Fibrin exudation and proliferation of parietal epithelial cells, macrophages, and interstitial fibroblasts result in the obliteration of Bowman's space and crescent formation (3). The extent of crescent formation correlates with the severity of glomerular damage (4). Crescents with predominant cellular components represent acute glomerular injury which can resolve with treatment, whereas fibrous crescents, interstitial fibrosis, or tubular atrophy may not have favorable renal outcomes (5).
The etiology and outcomes of the CrGN are heterogeneous (2,6). CrGN is classified based on the immunofluorescence (IIF) pattern into anti-GBM disease, immune-complex-mediated glomerulonephritis, and pauciimmune glomerulonephritis (2). Pauci-immune glomerulonephritis represents the majority of cases in the adult population, especially amongst whites, males, and people aged more than 65 years (2,7), whereas CrGN in children is more commonly immune-complex-mediated (8). The clinical course of CrGN depends both on the percentage of glomeruli with crescents and the underlying disease (9,10). The renal outcome is also determined by the severity of renal insufficiency at the time of presentation (11,12). Although there is limited data from India (8,13,14), there is no published data on CrGN from North-East India. This study aims to identify the etiology and evaluate the clinical characteristics, histopathological features, and outcomes of CrGN.

Study design and study population
This was a retrospective, observational study conducted at the Department of Nephrology, Gauhati Medical College and Hospital, Assam, India. We included all adult patients (>18 years of age) who underwent native kidney biopsy in our centre and had CrGN in the biopsy. A total of 80 biopsy-proven CrGN diagnosed from January 2013 to January 2018 were included in this study. Patients with less than seven nonsclerotic glomeruli and no consent were excluded. This study was approved by the institutional ethical committee of Gauhati Medical College and Hospital, and informed consent was obtained from all patients.

Data collection
Patients' data regarding demographic profile, clinical features, laboratory parameters, treatments, and morbidity/ mortality were collected from the hospital records. Clinical presentations varied from acute kidney injury (AKI), CKD, and asymptomatic urinary abnormalities (AUA). Patients' outcome data with regard to serum creatinine, development of end-stage renal disease, and complications were collected at each follow-up visit until August 2019.

Definitions
Hematuria was defined as the presence of ≥5 red blood cells per high power field in urine microscopy. Proteinuria was measured in 24 h timed collection as 24 h urine protein. Serum creatinine was measured by the enzymatic method and estimated glomerular filtration rate was calculated using CKD Epidemiology Collaboration (CKD-EPI) equation (15). AKI was defined as an increase in creatinine level by 0.3 mg/dL (26.5 mmol/L) within 48 h, or a percentage increase in serum creatinine ≥50% (1.5-fold from baseline) within 7 days, or urine volume # 0.5 mL/kg/h for 6 h. CKD was defined as eGFR, 60 mL/min/1.73 m 2 , or markers of kidney damage for more than 3 months (including nephrotic syndrome, AUA, or nephritic syndrome).
All kidney biopsy tissue specimens were studied for light microscopy (Magnus opto systems India Pvt Ltd model number CH-20i) and IF microscopy (Zeiss axio zoom v16 model number 3312000226). CrGN was defined by light microscopy as the presence of crescents in more than half of the total glomeruli. IF microscopy was used to categorize the CrGN into types I, II, and III. Type I was characterized by linear deposition of antibodies along GBM and included anti-GBM disease. Type II had granular deposition of immunoglobulins and complement in the glomerulus, and included primary and secondary glomerular diseases. Type III CrGN was defined by the absence of immune deposits in kidney pathology. Interstitial fibrosis and tubular atrophy (IFTA) were graded semiquantitatively into categories none (0), mild (1), moderate (2), and severe (3). Interstitial inflammation was semiquantitatively graded into categories based on the degree of inflammation in tubulointerstitium: 0 (0%), 1 (<20%), 2 (20 to 50%), and 3 (>50%).
Patients were treated as per standard protocols by treating physicians (Table 1). Dialysis was given as indicated. from the study due to inadequate biopsy and two patients had no follow-up data; 80 patients were included in this study. Follow-up duration varied from 3 to 32 months, and more than 1 year of follow-up data were available in 47 patients (58.8%). The etiology of CrGN in our study population is described in Table 1. Type II CrGN was the most common type, followed by type III and type I CrGN. In type II CrGN, lupus nephritis and IgAN were the most common causes of CrGN. The mean age of patients was 40.86 ± 16.5 years. Patients in type III CrGN were older than patients in type I and type II CrGN. Of 80 patients with CrGN, 43 (53.7%) were female and 37 (46.3%) were male. Female predominance was observed in type I and type II CrGN, whereas there was male predominance in type III. Table 2 describes the clinical and laboratory features of our study population. Duration of symptoms was more in type II CrGN than in type I and type III CrGN (P = 0.011). Oliguria was seen in 53.7% of the study population. Skin rash was significantly (P = 0.015) seen more in type II CrGN (38.9%), compared to type I (9.1%) and type III (12.12%). Hemoptysis was seen significantly (0.0432) more in type I (27.3%) and type III (18.2%), compared to type II (2.7%).
Type II CrGN had the highest mean proteinuria (3.9 ± 2.1, P < 0.0001) and lowest mean serum albumin (2.7± 0.7, P = 0.006). The mean estimated glomerular filtration rate (eGFR) was lowest in type I CrGN (6.7 ± 4.9 mL/min/1.73 m 2 ), compared to type II CrGN (16.1 ± 8.6 mL/min/1.73 m 2 ) and type III CrGN (12.4 ± 8.2 mL/min/1.73 m 2 ), with a P value of 0.003. Serum complements were low in around half of the patients (low C3 in 38.7% and low C4 in 12.5%). Complement C3 (P = 0.0001) and C4 (P = 0.01) were significantly lower in type II CrGN, compared to type I and type III CrGN. ANA was positive in 25 cases (31.3%) out of which the majority (72%) belonged to type II CrGN. Anti-dsDNA was positive only in type II CrGN (16 cases, 50%). ANCA positivity was seen in 37 cases (46.3%), the majority of which had type III CrGN. ANCA positivity was seen in 87.9% cases of type III CrGN out of which 54.5% cases had anti-MPO and 39.4% had anti-PR3 autoantibodies. Anti-GBM antibodies were seen in 6% of patients with type III CrGN. In type I CrGN, anti-GBM antibodies were present in 90.1% and ANCAs were present in 18.2% of patients.

Histopathological features
Characteristic histopathological features noted in kidney biopsies have been summarized in Table 3. Crescents were seen in 75.1 ± 18.3 of glomeruli with the highest percentage seen in type I CrGN (87 ± 15.2, P = 0.04), followed by type III CrGN (76.4 ± 21.4%) and type II CrGN(70.2 ± 18.1%). Fibrocellular crescents were the most common type of crescent seen in all three groups of CrGN. Mesangial proliferation (69.4%, P = 0.0005) and endocapillary proliferation Plasmapheresis was given for patients having severe renal failure (requiring dialysis or serum creatinine >5.7 mg/dL). Patients were followed up with monthly serum creatinine, urine protein, urine microscopy, and blood counts. Patients' outcomes were collected at the end of follow-up for renal survival, mortality, and eGFR. Multivariate analysis was done to evaluate the risk factors for renal survival.

Statistical analysis
All data were analyzed using the Statistical software, Statistical Package for the Social Sciences (SPSS v 19.0 IBM Corporation, USA). Continuous data were expressed as mean ± standard deviation or median with interquartile range, and categorical data were expressed as frequencies and percentages. Categorical data were compared by using chi-square test or Fisher's exact test and continuous data were compared by student's independent t-test or one-way analysis of variance (ANOVA) test. A five percent level of significance was used to carry the statistical analysis, with a P value of <0.05 being statistically significant.

Demography, clinical and laboratory features
A total of 1498 kidney biopsies were done in our centre from January 2013 to December 2018, out of which 85 biopsies were having CrGN (5.7%). Three patients were excluded  (97.2%, P ≤ 0.0001) were mostly seen in type II CrGN. The highest percentage of glomeruli with neutrophilic infiltrates was seen in type I (81.2%, P = 0.00861), followed by type II (52.7%) and type III (30.3%) CrGN. Tuft necrosis was significantly seen more in type I CrGN and type III CrGN (45.5% and 24.2%, respectively, P = 0.01937), compared to type II CrGN (8.3%). Chronic lesions in the form of moderate/severe IFTA were seen more in type I CrGN (63.6%), although not statistically significant.

Treatment and outcomes
Treatments of the patients were done as per the standard guidelines (16). Table 4 describes the treatment received by  the patients in this study. Hemodialysis was needed in more than half of the cases at the time of presentation and was significantly more common in type I (72.7%) and type II (57.6%) CrGN (P = 0.0034). Immunosuppression was given in 88.7% of the patients with no difference between the three groups of CrGN. The majority of patients received steroids plus cyclophosphamide, while 21.2% of patients received steroids alone. Fourteen patients (21.25%) received therapeutic plasma exchange, which was significantly (P = 0.0046) more in type I CrGN (45.5%) and type III CrGN (30.3%), compared to type II CrGN (5.6%). Patients were followed up with a mean follow-up of 23.1 ± 12.4 months. Table 5 describes the outcomes of our study population. Death, renal survival, and eGFR at the end of   (Table 6). In our study, oliguria at the time of presentation, a higher percentage of crescents, glomerular sclerosis, and moderate/severe IFTA were associated with poor renal outcomes.

Discussion
CrGN is a severe form of glomerulonephritis and is the leading histopathological diagnosis associated with rapidly progressive glomerulonephritis. There is a paucity of data related to CrGN etiology and outcomes from India. We analyzed the clinical, laboratory, treatment, and outcome data of 80 patients with CrGN. In our study, CrGN was seen in 5.7% of kidney biopsies done from January 2013 to January 2018, which was comparable to the study by Rempelli et al. from South India (17). Although various studies from different parts of the world have reported type III as the most common type of CrGN (2,18,19), type II was the most common type of CrGN in our study followed by type III. However few studies from India, China, and Saudi Arabia have reported type II CrGN to be the most common (17,(20)(21)(22). The higher incidence of type II CrGN could be explained by the higher incidence of infections and IgA nephropathy in this part of the world. In our study, type I and II CrGN had a female preponderance. Multiple studies have established that type II has a female preponderance, but the gender distribution is variable in type I and type III CrGN (2,17,19,21,23). In our study, average serum creatinine at the time of presentation was 5.87 ± 3.9 mg/dL; this was more than in studies by Oudah et al. and chen et al. but comparable to Gupta et al. (6,14,21) High serum creatinine in our study can be due to late presentation of the patients as was seen in other studies from India. More patients with type I CrGN presented with oliguria, hemoptysis, anemia, and dialysis requirement compared to other types. Other studies have reported that type I CrGN has most severe renal involvement at the time of presentation (21,23).
Serum ANCA is deemed pathogenic for ANCA-associated vasculitis (AAV). ANCA positivity was seen in 87.9% cases of type III CrGN out of which 54.5% cases had anti-MPO and 39.4% had anti-PR3 autoantibodies. Although most studies have demonstrated similar rates of ANCA positivity (21,22), other studies from India have shown lower highest percentage of glomeruli with crescents, neutrophilic infiltrates, and tuft necrosis was seen in type I CrGN. In our study, renal survival was poor in type I CrGN patients, compared to type II and type III CrGN. In our study, oliguria, crescents, glomerular sclerosis, and moderate/severe IFTA were associated with poor renal outcomes.
rates (2,14). As reported in various reports, 90.9% of cases with type I CrGN had anti-GBM autoantibodies (21,24). In patients with type I CrGN, ANCA positivity was seen in two (18.0%) cases which was comparable to other studies (21,23,25). In critically ill patients, performing a kidney biopsy is not always possible, so ANCA positivity can be taken as a surrogate marker of type III CrGN and possibly treated empirically in such situations of rapidly progressive renal function decline.
On histopathological examination, similar to other studies, the highest percentage of glomeruli with crescents (87%), neutrophilic infiltrates (81.2%), and tuft necrosis (45.5%) was seen in type I CrGN. Also, moderate/severe IFTA was seen more commonly in type I CrGN (21). These histological findings might be an explanation of the severe clinical features seen in type I CrGN patients. Fibrocellular crescents were predominant in each type of CrGN and the proportion of cellular, fibrocellular, and fibrous crescents were similar among the three types. This was consistent with other studies as well (6,14,26).
In our study, renal survival was poor in type I CrGN patients, compared to type II and type III CrGN. This was consistent with other studies by Jennette et al., Gupta et al., and Chen S et al. (2,14,21) Poor survival in type I CrGN could be due to a higher percentage of crescents in type I CrGN on initial biopsy, compared to type II and type III CrGN. In our study, renal survival was better in type II CrGN, compared to type III CrGN. This was consistent with other studies by Gupta (14,21,23,27). More studies may be needed to evaluate the renal survival in type II CrGN and type III CrGN patients. In our study, no differences in deaths were seen among the three CrGN groups. Multiple predictors have been seen for poor renal outcome in CrGN and include glomerulosclerosis, acute tubular necrosis, fibrinoid necrosis, serum creatinine, age, oliguria, crescents, and interstitial inflammation (2,5,7,16). In our study, oliguria, crescents, glomerular sclerosis, and moderate/severe IFTA were associated with poor renal outcomes. This was consistent with other studies, although age and acute tubular necrosis were not predictive of poor renal outcome (7,16).

Conclusion
In our study, CrGN was seen in 5.7% of kidney biopsies, which was comparable to other studies from India. Type II CrGN was the most common type of CrGN, followed by type III. In type II CrGN, lupus nephritis followed by IgAN were the most common causes. More patients with type I CrGN presented with oliguria, hemoptysis, anemia, and dialysis requirement, compared to the other types. The