Oxford-MEST classification in IgA nephropathy patients : A report from Iran

Hamid Nasri1,*, Mojgan Mortazavi2, Ali Ghorbani3, Heshmatollah Shahbazian3, Soleiman Kheiri4, Azar Baradaran5, Afsoon Emami-Naieni2, Maryam Saffari1, Saeed Mardani6, Ali Momeni6, Yahya Madihi2, Milad Baradaran-Ghahfarokhi7, Mahmoud Rafieian-Kopaie8, Parin Hedayati2, Shahzad Baradaran9, Mohammadreza Ardalan10, Shahram Sajjadieh2, Naziheh Assarzadegan11, Seyed Mohammad Ahmadi Soleimani12, Mohamad Reza Tamadon13

Indeed there has been a lack of international consensus on the classification and the predictive value of the histopathology findings in IgAN.Previously Lee and Haas, have published a classifications method for IgAN (15)(16)(17)(18)(19)(20).The recent Oxford classification of IgAN assesses the presence of hypercellularity, proliferation in mesangial and endocapillary regions, segmental glomerulosclerosis and tubular atrophy/interstitial (MEST).Also it has been reported as having prognostic value (14,(21)(22)(23)(24).This classification method looked for reproducible pathological features that were correlated to renal survival independent of clinical features, rather than grouping patients into categories (24).

Objectives
The prevalence and clinical picture of IgAN varies from region to region.This study was undertaken, firstly to study the morphologic changes in a group of IgAN patients, to assess the Oxford classification method and other morphologic variables, and secondly to compare pathological findings with published studies.

After publication of Oxford classification of
IgAN on 1 July 2009 (25)(26)(27), we applied it in this study for the classification of IgAN.

Definition of IgAN
The pathologic diagnosis of IgAN requires the demonstration of IgA-dominant mesangial or mesangio-capillary immune deposits through immune fluorescence (IF) microscopy.The immune deposits were semiquantitied from 0 to 3+ positive bright.The definition of IgAN needs the presence of diffuse and global IgA deposits that were graded ≥ 2+ and the absence of C1q deposition (3,(27)(28)(29).All renal biopsies performed at various medical centers (university hospitals or private centers) from July 2009 to February 2012, were sent to a reference laboratory.Except for two Afghan patients, the rest of cases were Iranian.None of the patients was treated before the biopsy.In one hundred ten (110) biopsies, prominent IgA depositis were observed on IF study.

I
Of 110 biopsies with dominant IgA deposition, one was due to a kidney transplantation (reccurrence of IgAN), one patient with diagnosis of Henoch Schonelin purpura , two of the patients with the dominant IgA deposits in association with concurrent diabetes mellitus and one patient with IgAN, whose biopsy had been taken from medulla for light microscopy (without glomerules), which were excluded from the study.Biopsies with less than 8 glomeroli were also excluded from the study.Two of the patients underwent re-biopsy, due to absence of glomeruli for IF or light microscopic study.None of the patients was diagnosed as IgAN, having history of collagen vascular diseases and liver cirrhosis based on questionaire filled at the time of biopsy admission, laboratory data in patients' records and a breif history provided by referee physicians at the time of biopsy admission.

Histologic data
All renal biopsies were prepared for light and direct immunofluorescence microscopy.Tissue was fixed in 10% formalin for histologic sectioning.Each kidney biopsy was prepared by cutting paraffin blocks into 3-μm sections and staining 2 slides with periodic acid Schiff, 2 slides for hematoxylin and eosin, 1 slide for Jones methenamine silver and 1 slide for trichrome.Each slide contained 2-3 sections.Materials usedfor IF was snapfrozen in liquid nitrogen.Sections (Six-micron in thickness) were stained for immunofluorescence study with fluorescein isothiocyanate-conjugated antibodies specific for human IgG, IgM, IgA, C1q, C3 and fibrin (DAKO, Produktionsvej 42, DK-2600 Glostrup, Denmark).IF slides were reported in a scale of 0, to 3+ positive bright.IF slides were reported by a nephropathologist (HN) and a general pathologist (SB).IF study was performed unaware of patients' data and before reviewing the slides for light microscopy.After IF diagnosis of IgAN, histopathology glass slides were reviewed to assess the morphologic variables, which were applied in Oxford-MEST classification method.Other important pathology findings were also assessed including glomerular fibrinoid necrosis, extracapillary proliferation (cellular, fibro-cellular or fibrous crescents), mesansial interposition and mesangiolysis.Also, the presence of foamy cells and tubular thyroidization in interestitium were assessed.Vascular lesions were scored and attention was made toward the presence of thrombotic microangiopathy in cases.After selection of biopsies having dominant deposition IgA by IF study and in the absence of exclusion criteria, glass slides were independently reported for the classification of Oxford-MEST.Each reviewer completed a score sheet for every biopsy and these were used to compile the pathological data set.If there were any differences among the score sheets, the histologic sections would be reviewed concurrently by them at a two headed microscope and the discrepancy reconciled.Therefore, a consensus was established to confirm each parameter.

Vascular definitions
Arterial lesions are scored based on the most severe lesions.Intimal thickening is scored by comparing the thickness of the intima to that of the media in the same segment of vessel.Intima is scored variously as as normal (score: 0), and thickened to more or less than the thickness of the media (score: 1).Arteriolar hyaline is noted as the proportion of arterioles affected and is scored according to the following categorization; normal arterioles (score: 0), 1-25% affected arterioles (score: 1), 26-50% affected arterioles (score: 2) and>50% affected arterioles (score: 3) (25).

Clinical Studies and Laboratory Data
The medical records of patients were reviewed to obtain various demographic, clinical and laboratory information at the time of biopsy and for follow-up activities.Data gathered at the time of biopsy included race, gender, age, serum creatinine(Cr) and proteinuria (based on a 24hour urine collection).

Statistical analysis
Mean values and standard deviations were calculated, and statistical significance of the differences between groups were calculated using T, Chi-square and Likelihood Ratio tests.Because of high positive skewness of data, the Spearman's coefficient of correlation was used to check the correlation.A computer program (SPSS version 16.0, Chicago, IL, USA) was used for statistical analysis.P < 0.05 was considered statistically significant.

Population Characteristics
This is an observational study, conducted on IgAN patients.A total of 102 biopsies were enrolled to the study.

Prevalence
In this survey, of 102 patients 71.6 % were male, 2 % were below 20 years, 60.8 % were between 20-40 years and 37.2 % were more than 40 years of age.The age of patients was from 18 to 73 years with mean of 37.7 ± 13.6 years (medain: 35 years).Mean±SD of proteinuria and serum creatinine were 1795 ± 1369 mg/day and 1.651 ± 1.61 mg/dL, respectively.The medain level of proteinuria and serum creatinine were 1500 mg/ day and 1.2 mg/dL respectively.The mean serum creatinine in females was 1.14±0.73mg/dL and in males was 1.85±1.80mg/dL.Nephrotic range proteinuria was in 8.8% of patients, all of which were male.Also, the mean of proteinuria were 1323±735 mg/day and 1982±1515 mg/ day,in females and males repectively.
After stratification of patients according to their proteinuria, we found that, 23.5% of the patients had proteinuria below 1 g/day, 65.7% had proteinuria between 1-3 g/day and 10.8% of the patients had proteinuria more than 3 g/day.
In this study, of total 15 ± 7.3 glomeruli (medi-an=14 glomeruli) in all biopsies, 2.5 ± 3 were totally sclerotic (median=2 glomeruli) and 0.46±1.11 of them were crescentic.In general, extracapillary proliferation occurred in 23 % of the patients and types of crescents were as follows; cellular crescent in 10.8%, fibro-cellular crescent in 7.8% and fibrous crescents in 8.8% of biopsies.

Oxford classification (MEST)
Four variables of Oxford classification were analyzed in our patients; M1; we found , 66% of the patients with mesangial hypercellularity in >50% of glomeruli.However, mesangial hypercellularity occurred in 90.2% of cases, similarly mesangial widening in 92.2% of patient was observed.Also The mean (SD) age of all patients whohad M0 and M1 was 42.27 ± 14.09 and 35.10 ± 12.69 years respectively.E1; endocapillary lesions; were found in 32 % of patients.S1; segmental glomerulosclerosis whichincludes mesangial sclerosis with capsular adhesions occurred in 67 % of the patients.However simple tuft-capsule adhesion, without mesangial sclerosis, was observedin 61% of patients.The percentage of segmental glomerulosclerosis were 41.1% and 72.6 %, in females and males respectively.
T; tubular atrophy/interstitial fibrosis (scores I and II) were 30% and 19% respectively, while 51% showed zero score.Furthermore, mean (SD) percent of interstitial fibrosis was 14.6 ± 21.1% with a median of 5%.T1 and T2 in females were as follows; 17.2% and 13.8% , while these frequencies were 37% and 20.5% for males, respectively.

Vascular lesions
Score I of intimal thickening of interlobular arteries was 55.9%, while zero score was observed in 44.1% of the patients.Scores of arteriolar hyaline were as follows; score I and II in 31.4 % and 24.5%, respectively.Also, score 0 was observed in 44.1 % of the patients.We also found thrombotic microangiopathy in 2% of our patients.

Other lesions
Glomerular fibrinoid necrosis were present in 6.9% of the patients.Mesangial interposition and mesangiolysis each occurred in 1% of the patients, respectively.Foamy cells and tubular thyroidization of the interstitial area were found in 3.9% and 11.8% of patients, respectively.

Gender and group differences
The independent samples t-test showed a higher serum creatinine (P=0.044) and proteinuria (P=0.028) in males.The Chi-square test showed significant difference in segmental glomerulosclerosis between males and females with higher values in male geneder (P= 0.003).The Likelihood Ratio test showed significant differences in frequency distribution of interstitial fibrosis/tubular atrophy in males and females, with higher values in male geneder (P= 0.045).Also, M1 was more prevalent in younger patients (P= 0.01).
In this study there was also a significant correlation between serum Cr and segmental glomerulosclerosis (Spearman's rho = 0.521, P<0.001) (fig.2).
In the present study, mesangial proliferation and mesangial widening were observed in 90.2% and 92.2%, respectively.However only 66% of biopsies showed score M1, while M1 was more prevalent in younger patients.Furthermore, variables of E1 and S1were found in 32% and 67% of patients, respectively.In our survey, of 102 patients, 71.6 % were male, and mean age of patients was 37.7 ± 13.6 years.This finding is comparable with findings in a study conducted by Mittal et al. in India in which 66 patients with IgAN were identified.The mean age of their patients was 29.9 years with a male: female ratio of 4.4:1.They found that, most patients presented with renal failure and a significant percentage (23%) also had nephrotic range proteinuria (35).This finding is in contrast to our results because in our study, the mean of proteinuria and serum creatinine were 1795 ± 1369 mg/day and 1.651 ± 1.61 mg/dL, respectively.Also, nephrotic range proteinuria was observed in 8.8% of patients.hypercellularity, glomerular sclerosis, endocapillary hypercellularity and IFTA were associated with the renal outcome likelihood ratio (22).To evaluate the predictive value of the Oxford classification on renal survival defined by doubling creatinine or end-stage renal disease in patients with IgA nephropathy, 183 patients with IgA nephropathy were investigeted by Alamartine et al., in France.They found that tubular atrophy/interstitial fibrosis, segmental glomerulosclerosis, and endocapillary hypercellularity strongly impacted doubling creatinine or end-stage renal disease.On the contrary, mesangial hypercellularity was not associated with renal outcome.They concluded that the baseline renal function seems to be of a greater importance than pathologic lesions (38).Shi et al., studiedthe Likelihood Ratio of 410 patients with IgAN, from China.They evaluated the predictive value of the Oxford classification and found that tubular atrophy and interstitial fibrosis are the independent predictive factors of ESRD.Patients who had >25% of glomeruli with endocapillary hypercellularity showed higher proteinuria, lower glomerular filtration rate and higher mean arterial pressure (MAP) than patients with less endocapillary hypercellularity (39).Similarly data obtained from 146 cases with biopsy proven IgAN were analyzed by Walsh et al., from Canada.Their data revealed that >40% glomerular sclerosis and a systolic blood pres-sure>150 mmHg were risk thresholds.Moreover, they found that baseline creatinine, proteinuria, systolic blood pressure, glomerular sclerosis, interstitial fibrosis and crescentic disease were the predictors of primary outcome (40).Furthermore, they showed that interstitial fibrosis, glomerular sclerosis and crescents remained as the independent predictors of primary outcome and significantly improved model fit compared to clinical characteristics alone.They concluded that baseline histopathologic parameters are the predictors of adverse outcomes in IgAN even after taking into consideration clinical characteristics.Also, relatively small degrees of interstitial fibrosis confer an increased risk for progressive IgAN (40).In a study conducted by El Karoui et al., on 128 IgAN cases with mean age of 38.7 years (range18-78) in France , they found that mean proteinuria was 2.43 g/day(33), while we found 1795 ± 1369 mg/day proteinuria in our cases.The medain level of proteinuria in our patients was 1.5g/day, implies that patients studied by El Karoui et al., were more proteinuric.We also found that nephrotic range proteinuria was observed in 8.8% of our patients, where all were male.Similarly, the age of our patients was ranging from 18 to 73 years with a mean of 37.7 ± 13.6 years.Except of four variables of MEST system, there was a concern, whether other pathology variables in the glomeruli, interstitial or vessels may have prognostic outcome.Among them, more importantly, however, is extracapillary proliferation.Indeed, this classification is entirely limited to four above mentioned morphological features (19) and did not include crescent in the analysis (25-27, 29, 31).Likewise, there are other pathology variables, which may be associated with the renal outcome and need further attention.These are IF and electron microscopic findings, which were not included in this classification (19).Various studies, however, showed that crescentic disease is a predictor of the primary outcome and had been reported as one of the most important prognosticators in IgAN (15,28,41).In our study, extracapillary proliferation occurred in 23% of patients, which was lower than what had been observed in the Oxford study (45%).This lower percentage of extracapillary proliferation was also found in the study conducted by El Karoui et al., in pares.They found that extracapillary proliferation occurs in 31 (24.2%)out of 128 IgAN cases, while mean age of their patients was similar to ours (33).Likewise, Bitencourt-Dias et al., analyzed 144 patients with IgAN.Crescents were found in 18% of their patients.They concluded that the presence of crescents was associated with higher levels of initial serum creatinine and urinary protein excretion and a higher frequency of hypertension and progression to end-stage renal disease (42).Simiraly we found a significant correlation between serum Cr and total crescents.Edström Halling et al., investigeted ninety-nine children with IgAN and showed that mesangial hypercellullarity score >0.5, presence of endocapillary hypercellularity or IFTA of >25% were each associated with a poor outcome.They found that three of the four histology lesions in the Oxford classification, as well as the presence of crescents, were valid in predicting a poor outcome in their patient (10,43,44).Similarly, Katafuchi et al., conducted a study on the likelihood ratio of 702 patients with IgAN.They suggested extracapillary proliferation to be included in the next version of the Oxford classification of IgAN to widen the scope of classification (31).Hence, it seems that data on prognostic significance of extracapillary proliferation,as a variable which need to accommodate on MEST, is increasing.
Thrombotic microangiopathy (TMA) occurs in IgA nephropathy, but is uncommon in the setting of IgA nephropathy and its significance, as a concomitant histologic finding, is unclear (45).In a study on 10 patients with an established diagnosis of IgA nephropathy and concurrent findings of TMA, based on their renal biopsies, Chang et al., found six patients presented with malignant hypertension, while three others had severe hypertension and five patients had nephrotic-range proteinuria (45).They also found that seven patients had occasional arteriolar thrombi identified by light microscopy and prominent glomerular subendothelial space widening by electron microscopy, while three patients demonstrated only ultrastructural features of thrombotic microangi-opathy.They suggested that a thrombotic microangiopathy injury, when present, is usually found in advanced stages of IgA nephropathy and can be associated with severe proteinuria (45).In our study, of 102 patients with IgAN, 2% of them were found to have TMA, while all of them had malignant hypertension.In contrast, a high frequency of thrombotic microangiopathy (TMA) was observed by some investigators in the cases of IgAN.El Karoui et al., retrospectively examined a series of 128 patients diagnosed with IgA nephropathy in France.In their series, 53% presented with lesions of TMA, acute or organized, in arteries and/or arterioles.Among patients with TMA, 4% were normotensive, 25% had controlled hypertension, and 71% had uncontrolled hypertension.Of those with uncontrolled hypertension, 26% had malignant hypertension (46).Histologically, the group with TMA had a significantly greater percentage of sclerotic glomeruli and worse tubulointerstitial fibrosis than those of the group without TMA.However, a significant minority of patients had near-normal histology, with minimal tubular atrophy (20%) and/ or <20% interstitial fibrosis (24%).Their study showed that, TMA rarely occurs in the absence of significant proteinuria.They concluded that lesions of TMA are frequent in IgA nephropathy and may occur in normotensive patients with near-normal renal histology.However, this needs further investigation regarding its association with IgAN and its prognostic value.Their studies rule out severe hypertension or advanced renal disease as the sole causes of TMA (46).

Conclusions
Higher prevalence of segmental glomerulosclerosis and interstitial fibrosis/tubular atrophy, as the two of four variables of Oxford-MEST classification in male patients with IgAN, further attests that male gender is a risk factor in this disease.Significant correlation between se-rum creatinine and crescent in this study, was in an agreement with previous studies and suggests the probable accomodation of crescent as a new variable in the MEST system.

Figure 1 :
Figure 1: Scatter plot of total crescent based on serum creatinine Study of Mittal et al., showed advanced disease in much younger individuals(35), which was not supported by our study.Khawajah et al. studied a total of 42 patients, who were diagnosed to have primary IgAN in Saudi Arabia.Majority of the study cases were males, in the second, third and fourth decades of life.There were varying degrees of mesangial proliferation(36).To analyze the findings of 343 patients with IgAN, Maixnerova et al. have assessed the main demographic, clinical and histological data(37).Similarly, they found that, male gender is predominated over female gender.The median age of IgAN patients was 32.3 years, and the median level of proteinuria was 1.8 g/day.These findings were in agreement with our study, in which patients had a median age of 35 years and median proteinuria of 1.5g/day.To clarify the prognostic factors and the renal survival rates of 223 patients with IgAN, Kang et al. found that glomerular sclerosis, is a risk factor for developing ESRD.We also found a significant correlation between serum Cr and segmental glomerulosclerosis, a finding which correlates with Kang's study.To find the relationship of four variables applied in the MEST classification with the outcome of the disease, Grcevska et al., conducted a study in 40 adult patients with IgAN.They found that mesangial

Figure 2 :
Figure 2: Scatter plot of segmental glomerulosclerosis based on serum creatinine