Saudi Journal of Kidney Diseases and Transplantation

: 2020  |  Volume : 31  |  Issue : 6  |  Page : 1366--1375

Prognostic value of the Oxford classification and the Oxford score in IgA nephropathy: A Tunisian study

Hela Jebali1, Hiba Ghabi2, Ikram Mami2, Badr Ben Kaab2, Lilia ben Fatma2, Wided Smaoui2, Slim Haouet3, Hend Ben Mahjouba3, Mohanad Hassan4, Fethi Ben Hmida4, Lamia Raies2, Mohamed Karim Zouaghi2,  
1 Department of Nephrology, Rabta Hospital; Faculty of Medicine of Tunis, Tunis, Tunisia
2 Department of Nephrology, Rabta Hospital, Tunis, Tunisia
3 Department of Pathology, Rabta Hospital, Tunis, Tunisia
4 Department Laboratory Research of Kidney Pathology (LR00SP01), Charles Nicolle Hospital, Tunis, Tunisia

Correspondence Address:
Hela Jebali
Department of Nephrology, Rabta Hospital, Jabbary 1007 Tunis


Immunoglobulin A nephropathy is the most common primary glomerular disease. The main challenge in this disease is the evaluation of prognostic factors for end-stage renal disease. The aim of our study was to describe the characteristics of immunoglobulin A nephropathy, to evaluate the histological data according to Oxford classification, and to identify factors associated with renal survival. This was a retrospective study, including adults with primary immunoglobulin A nephropathy. The study was conducted over a period of 10 years. Renal biopsies were scored according to Oxford classification. Oxford score, based on the sum of the different histological lesions of Oxford classification, was calculated for each patient. We included 50 patients with a gender ratio (male:female) of 2.8. The average age was 35.6 ± 10.6 years. Fifty-eight percent of the patients had hypertension (HTN). The median proteinuria was 1.9 g/day. The median of the glomerular filtration rate was 47.6 mL/min/1.73 m2. According to Oxford classification, mesangial proliferation, endocapillary proliferation, glomerulosclerosis, interstitial fibrosis, and/or tubular atrophy and crescents were present in 40%, 38%, 88%, 36%, and 22% of the cases, respectively. The median Oxford score was 2. The median follow-up duration was 30 months. Ten patients (20%) reached end-stage renal disease. At univariate analysis, HTN, glomerular filtration rate, proteinuria, tubular involvement, and Oxford score >3 were associated with progression to end-stage renal disease (ESRD). Tubular involvement was an independent risk factor for ESRD. Our study confirms the prognostic value of the Oxford classification in immunoglobulin A nephropathy.

How to cite this article:
Jebali H, Ghabi H, Mami I, Kaab BB, Fatma Lb, Smaoui W, Haouet S, Mahjouba HB, Hassan M, Hmida FB, Raies L, Zouaghi MK. Prognostic value of the Oxford classification and the Oxford score in IgA nephropathy: A Tunisian study.Saudi J Kidney Dis Transpl 2020;31:1366-1375

How to cite this URL:
Jebali H, Ghabi H, Mami I, Kaab BB, Fatma Lb, Smaoui W, Haouet S, Mahjouba HB, Hassan M, Hmida FB, Raies L, Zouaghi MK. Prognostic value of the Oxford classification and the Oxford score in IgA nephropathy: A Tunisian study. Saudi J Kidney Dis Transpl [serial online] 2020 [cited 2021 Jun 19 ];31:1366-1375
Available from:

Full Text


Immunoglobulin A nephropathy (IgAN) is the most common primary glomerulonephritis in the world.[1] It is characterized by a highly variable course ranging from a totally benign condition to end-stage renal disease (ESRD) and its histological presentation is extremely heterogeneous, making it difficult to assess lesions that are associated with renal survival.

Oxford classification was developed in 2009 and revised in 2016 to evaluate the IgAN outcome according to histological parameters.[2],[3] This pathological classification was based on five scores mesangial hypercellularity, endo-capillary hypercellularity, segmental glomerulosclerosis, tubular atrophy/interstitial fibrosis, and cellular or fibrocellular crescents (MEST-C).

Many studies analyzed the prognostic value of these variables with different results. Thus, prognostication is still imperfect.

Development of a pathological risk model that combines the different variables of Oxford classification could have a more accurate risk prediction.

The aim of our study was to evaluate the histological data according to Oxford classification in Tunisian patients with primary IgAN, to develop a pathological risk model based on MEST-C scores, and to identify factors associated with renal survival.


This was a retrospective study carried out at the Department of Nephrology at La Rabta Hospital. It was conducted over a period of 10 years from May 2007 to December 2017.

Patients who were 18 years or older, with biopsy-proven IgAN, considered as idiopathic, were included. IgA vasculitis, secondary causes of mesangial IgA deposits, systemic disease, or diabetes and patients with <8 nonsclerotic glomeruli for light microscopy study were excluded.

Past medical history, clinical and biological parameters at the time of the renal biopsy, and treatment modalities were recorded. For each patient, the date of renal biopsy was established as the baseline point.

Histopathologic study

All renal biopsy specimens were stained with hematoxylin and eosin, periodic acid–Schiff, and Masson’s trichrome for light microscopy study. Direct immunofluorescence microscopy was used in all cases to detect the deposition of IgG, IgA, IgM, C3, and Clq.

Included renal biopsies were reviewed and analyzed according to Oxford classification[2],[3] by an experienced pathologist.

Oxford score, based on the sum of the different histological lesions of Oxford classification, was calculated for each patient as follow:

  • M = 0 if M0 and M = 1 if M1
  • E = 0 if E0 and E = 1 if E1
  • S = 0 if S0 and S = 1 if S1
  • T = 0 if T0, T = 1 if T1, and T = 2 if T2
  • C = 0 if C0, C = 1 if C1, and C = 2 if C2


Hypertension (HTN) was defined as blood pressure ≥140/90 mm Hg or the use of antihypertensive agents.[4]

Glomerular filtration rate (eGFR) was estimated using the Modification of Diet in Renal Disease formula.

Mean arterial pressure (MAP) was defined as diastolic pressure plus one-third of systolic pressure.

Renin-angiotensin system blockade (RASB) indicates any exposure to either angiotensin-converting enzyme inhibitor or angiotensin receptor blocker.

ESRD was defined as GFR <15 mL/min/1.73 m2 or the need for chronic dialysis.

Study endpoint

The kidney disease outcome was analyzed using the cumulative percentage of patients who developed ESRD in the course of the study. Follow-up time was considered as the interval between renal biopsy and the last outpatient visit or ESRD. The studied parameters were as follows: HTN, proteinuria, eGFR, pathological lesions of Oxford classification, and the Oxford score. In this retrospective study, therapy and follow-up duration could confound correlations between pathology and clinical outcome. Therefore, these parameters were assessed in relation to Oxford classification.

 Statistical Analysis

Statistical analyses were performed with the IBM SPSS Statistics for Windows, version 20.0 (IBM Corp., Armonk, NY, USA). Quantitative variables were expressed as means ± standard deviations or medians. Qualitative variables were presented by frequency and percentages. The data were assessed with regard to the distribution using the Shapiro-Wilk normality test. To assess differences between quantitative variables, we used the t-test or Mann–Whitney U-test when appropriate. To compare categorical variables, we used the Chi-square or Fisher’s exact test. Correlation coefficients and significance were calculated by Pearson’s test or Spearman’s test when appropriate. Renal survival was studied using Kaplan Meier in univariate analysis. Multivariate Cox regression was performed to determine the independent factors of ESRD.

Receiver operating characteristic curves were analyzed to calculate the best cutoff value of quantitative variables associated with the studied endpoint.

A P-value under 0.05 was considered statistically significant.


Between May 2007 and December 2017, 56 patients were diagnosed with primary IgAN. Oxford classification was adopted on 50 patients: 37 men (26%) and 13 women (26%) with a gender ratio (male: female) of 2.8.

At the time of renal biopsy, the mean age was 35.6 ± 10.6 years (18–60 years). The median MAP was 97 mm Hg (range from 76 to 130 mm Hg). Twenty-nine patients (57%) had hypertension. The median of proteinuria was 1.9 g/d (0–8.5). The median eGFR was 47.7 mL/min/1.73 m2 (6–153 mL/min/1.73 m2). Fifteen patients (30%) had an eGFR <60 mL/min/1.73 m2. Fourteen patients (28%) were receiving RASB before renal biopsy, and none of them were under immunosuppressive treatment.

According to Oxford classification, mesan-gial hypercellularity M1, endocapillary hyper-cellularity E1, and glomerulosclerosis S1 were noted in 20 (40%), 19 (38%), and 44 (88%) cases, respectively. For tubular atrophy/ interstitial fibrosis (T), eight cases (16%) were classified as T1 and 10 patients (20%) were classified as T2. Concerning cellular or fibro-cellular crescent (C), we observed C1 in 10 patients (20%) and C2 in one case (2%). The median of the calculated Oxford score was 2 (0–6).

Podocyte hypertrophy was noted in nine cases (18%). It was associated with S1 lesion in seven cases (14%). We did note the observed tip lesion.

Correlations between pathological features and clinical features at baseline are summarized in [Table 1]. Mesangial hypercellularity (M) was associated with MAP. Segmental sclerosis was associated with the presence of HTN, while tubular atrophy/interstitial fibrosis (T) was correlated to MAP, hypertension, and baseline eGFR. We found a strong positive correlation between MAP and Oxford score (r = 0.50; P < 0.001) [Figure 1] and a strong negative correlation between this score and baseline eGFR (r = -0.43; P = 0.002) [Figure 2].{Figure 1}{Figure 2}{Table 1}

Forty patients (80%) received RAB, and 21 patients (42 %) received corticosteroids. The use of RAB and corticosteroid treatment as well as the follow-up period was assessed in relation to pathologic lesions. The severity of tubulointerstitial fibrosis (T1, T2) was significantly associated with the prescription of corticosteroid therapy. Patients were followed up for median duration of 30.5 months. We did not find any significant difference concerning follow-up duration between each pathology variable of Oxford classification.

Ten patients (20%) reached ESRD. Renal survival from ESRD was correlated to T lesion and did not differ significantly between the other lesions [mesangial hypercellularity (M), endocapillary hypercellularity (E), segmental glomerulosclerosis (S), and cellular or fibrocellular crescent (C)] [Table 2].{Table 2}

Kaplan–Meier analyses showed that HTN (P = 0.005), proteinuria >2 g/day (P = 0.016), eGFR <60 mL/min/1.73 m2 (P = 0.008), T lesion of Oxford classification (P <0.001), and Oxford score >3 (P = 0.001) were associated with renal survival from ESRD [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7]. Multivariate Cox regression analysis was used to address whether pathological findings: T lesion and Oxford score can predict renal outcome independently of baseline clinical data (HTN, proteinuria, and eGFR) and corticosteroid treatment. Two Cox regression models (A and B) were designed. In the Cox regression model A, T lesion was evaluated. We merged T1 and T2 lesions in order to maximize statistical power. As shown in [Table 3], the presence of T1 or T2 lesions was predictive of ESRD independently of baseline clinical findings and corticosteroid treatment. In the Cox regression model B, an Oxford score >3 failed to attain independent significance [Table 4].{Figure 3}{Figure 4}{Figure 5}{Figure 6}{Figure 7}{Table 3}{Table 4}


The frequency of histological lesions according to the Oxford classification and their association with clinical findings varies between the different reported studies. In this work, segmental glomerulosclerosis (S) and mesangial hypercellularity (M) were the most observed lesions. These disparities may be explained by differences in scoring MEST-C lesions. As reported in the VALIGA cohort,[5] when the authors compared the evaluation of pathology lesions between local and central pathologists, they concluded that reprodu-cibility was poor for M, E, and C. Added to that, received treatment before renal biopsy may affect pathological findings. A repeat-biopsy study has shown that active lesions including endocapillary hypercellularity and cellular or fibrocellular crescent decreased significantly after an immunosuppressive therapy.[6] On the other hand, the widespread use of RASB before renal biopsy may also influence the severity of mesangial hyper-cellularity; hence, the prevalence of this lesion may in part differ according to the percentage of patients receiving RASB. Indeed, mesangial proliferation can be stimulated by local production of angiotensin II.[7],[8] Recently, Zhang et al[9] found that patients classified as M1 had a higher mesangial expression of angiotensin II receptors compared to those classified as M0. Twenty-eight percent of the patients were receiving RASB at renal biopsy in our study versus 39% in the VALIGA cohort.[10]

The high prevalence of glomerulosclerosis (S) and tubular atrophy/interstitial fibrosis (T) in our population can be explained by the inclusion of patients with advanced stage of chronic renal disease. Consistent with our finding, Yau et al[11] noted T1 or T2 lesions in 35% of the cases. In this study, 47% of the included patients had renal failure.

We found that MAP was significantly higher in the M1 group. It has been demonstrated that pro-inflammatory cytokines and angiotensin secreted by mesangial cells may induce blood pressure elevation.[12],[13]

Concerning correlations between E lesion and clinical parameters, results are controversial between the different studies. In this study, none of the studied parameters was associated with E lesion. In the Oxford study,[2] patients with E0 lesion had a lower eGFR and a higher MAP. On the other hand, proteinuria was higher in those with E1 lesions. Endocapillary hypercellularity (E1) was associated with lower eGFR and increased proteinuria in the study conducted by Zeng et al.[14] In the present study, S lesion was only correlated to HTN. Bellur et al[15] showed that the presence of podocytopathy features associated with glomerulosclerosis was correlated with proteinuria, unlike glomerulosclerosis alone. The Oxford Working Group proposed to specify whether S1 is associated with podocyte hypertrophy or sclerosis at the tubular pole (tip lesion).[3] S1 associated with podocyte hypertrophy was noted in only 14% of the cases in our study, which would explain the lack of correlation between S and proteinuria in our study.

In the VALIGA study[10] and in the first study of the Oxford Working Group,[2] T lesion was associated with MAP, proteinuria, and eGFR. In our study, T was significantly correlated with MAP, HTN, and eGFR.

The presence of cellular or fibrocellular crescents (C1 + C2) was not associated with clinical-biological data in our study. Our results may be due to the small percentage of patients who presented this lesion.

Otherwise, we found a strong positive correlation between MAP and Oxford score and a strong negative correlation between this score and baseline eGFR. Thus, the higher the Oxford score was, the more severe the IgAN clinical presentation was. Our findings support the utility of the evaluation of the Oxford score.

The contribution of additional prognostic information of Oxford classification beyond that provided by clinical features remains debatable. The prognostic impact of the different lesions of Oxford classification varies between studies. Several factors can explain these differences: First, inclusion criteria; second, treatment protocols; third, the follow-up duration; and fourth, the studied endpoint.

The Oxford Working Group[2] found in multivariate analysis that S and T lesions were independently associated with rate of eGFR decrease while M and T lesions were associated with survival from 50% reduction in renal function or ESRD.

The presence of C1 was associated with a poor renal outcome only in patients who did not receive immunosuppressive therapy, whereas C2 was associated with poor renal survival independently of received treatment. Kang et al[16] found that only T lesion was an independent predictor of the combined event (survival from 50% reduction in renal function or ESRD). In our study, only T was independently associated with renal survival from ESRD.

At univariate analysis, an Oxford score >3 was associated with renal outcome. We failed to confirm the prognostic value of this score independently of baseline clinical findings. We postulate that the small sample size of included patients and the relatively short follow-up duration may reduce the prognostic value of Oxford score in multivariate analysis.

We provided the first reported study of Oxford classification in primary IgAN among Tunisian patients. We proposed to calculate an Oxford score based on the sum of the different pathology variables of Oxford classification to optimize the evaluation of IgAN severity.

We did not find any significant difference in follow-up duration according to Oxford classification which provides a better endorsement for our statistical results, as follow-up duration could confound correlations between pathological variables and clinical outcome in retrospective studies.

Our study has some limitations. It was a retrospective and monocentric study including a small sample size of patients with a short follow-up period. Added to that, a single pathologist reviewed all the included renal biopsy; hence, reproducibility of Oxford classification was not evaluated in our study.


In this study, we evaluated the utility of the Oxford classification in Tunisian patients with primary IgAN. Only the prognostic value of T lesion was validated independently of clinical parameters and corticosteroid treatment. Further prospective studies are needed to validate the Oxford score as an independent prognostic tool.

Conflict of interest: None declared.


1Barbour SJ, Coppo R, Zhang H, et al. Evaluating a new International Risk-Prediction Tool in IgA Nephropathy. JAMA Intern Med 2019;179:942-52.
2Working Group of the International IgA Nephropathy Network and the Renal Pathology Society, Cattran DC, Coppo R, Cook HT, et al. The oxford classification of IgA nephropathy: Rationale, clinicopatho-logical correlations, and classification. Kidney Int 2009;76:534-45.
3Trimarchi H, Barratt J, Cattran DC, et al. Oxford Classification of IgA nephropathy 2016: an update from the IgA Nephropathy Classification Working Group. Kidney Int 2017;91:1014-21.
4Mancia G, Rosei EA, Azizi M, et al. 2018 ESC/ESH Guidelines for the management of arterial hypertension. Eur Heart J 2018;39: 3021-104.
5Bellur SS, Roberts ISD, Troyanov S, et al. Reproducibility of the Oxford classification of immunoglobulin A nephropathy, impact of biopsy scoring on treatment allocation and clinical relevance of disagreements: Evidence from the validation of IgA study cohort. Nephrol Dial Transplant 2019;34:1681-90.
6Shen XH, Liang SS, Chen HM, et al. Reversal of active glomerular lesions after immunosup-pressive therapy in patients with IgA nephro-pathy: A repeat-biopsy based observation. J Nephrol 2015;28:441-9.
7Miyake-Ogawa C, Miyazaki M, Abe K, et al. Tissue-specific expression of renin-angiotensin system components in IgA nephropathy. Am J Nephrol 2005;25:1-2.
8Gómez-Garre D, Ruiz-Ortega M, Ortego M, et al. Effects and interactions of endothelin-1 and angiotensin II on matrix protein expression and synthesis and mesangial cell growth. Hypertension 1996;27:885-92.
9Zhang Z, Jiang SM, Ma YP, et al. Expression of the intrarenal angiotensin receptor and the role of renin-angiotensin system inhibitors in igA nephropathy. Mol Cell Biochem 2019;453: 103-10.
10Coppo R, Troyanov S, Bellur S, et al. Validation of the oxford classification of igA nephropathy in cohorts with different presentations and treatments. Kidney Int 2014;86: 828-36.
11Yau T, Korbet SM, Schwartz MM, Cimbaluk DJ. The oxford classification of igA nephro-pathy: A retrospective analysis. Am J Nephrol 2011;34:435-44.
12Wen Y, Crowley SD. Renal effects of cytokines in hypertension. Curr Opin Nephrol Hypertens 2018;27:70-6.
13Zhang J, Rudemiller NP, Patel MB, et al. Interleukin-1 receptor activation potentiates salt reabsorption in angiotensin II-induced hypertension via the NKCC2 co-transporter in the nephron. Cell Metab 2016;23:360-8.
14Zeng CH, Le W, Ni Z, et al. A multicenter application and evaluation of the oxford application and evaluation of the oxford classification of IgA nephropathy in adult chinese patients. Am J Kidney Dis 2012;60: 812-20.
15Bellur SS, Lepeytre F, Vorobyeva O, Troyanov S, Cook HT, Roberts IS. Evidence from the Oxford Classification cohort supports the clinical value of subclassification of focal segmental glomerulosclerosis in IgA nephropathy. Kidney Int 2017;91:235-43.
16Kang SH, Choi SR, Park HS, et al. The Oxford classification as a predictor of prognosis in patients with igA nephropathy. Nephrol Dial Transplant 2012;27:252-8.