Abstract
Background
The Oxford Classification was proposed as an independent prognostic indicator in IgA nephropathy (IgAN). However, most studies on the subject focus on adults instead of children.
Objectives
Using a meta-analysis to appraise the predictive roles of the Oxford classification for the prognosis of pediatric patients with IgAN.
Methods
All cohort studies regarding the analysis of the association between poor kidney-related prognosis (GFR categories G2-G5) according to the Kidney Disease Improving Global Outcomes (KDIGO) Guideline in pediatric patients with IgAN and five pathologic lesions in the Oxford Classification were included. Hazard ratios (HRs) regarding the association between the Oxford classification and prognosis of pediatric patients with IgAN were synthesized using random effect models. The risk of bias in studies was assessed based on the Newcastle-Ottawa scale.
Results
Fourteen articles were included with 5679 IgAN patients and 710 endpoint outcome events occurred. M1 was associated with a higher risk of poor kidney-related prognosis compared with M0, pooled HR (1.79; 95%CI, 1.46–2.19; p < 0.001, random effect model). S1 and T1 or T2 increased the risk of poor kidney-related prognosis (pooled HR, 2.13; 95%CI, 1.68–2.70; p < 0.001; pooled HR, 2.64; 95%CI, 1.81–3.86; p < 0.001, respectively, estimated by random effect model). Compared with C0, C1, or C2 was also associated with an increased risk of poor kidney-related prognosis in the subgroup analysis of Asian and other populations. Evidence to indicate that E1 increased the risk of poor kidney-related prognosis was marginal.
Keywords: IgA nephropathy, Oxford Classification, MEST-C, pediatric
Introduction
IgA nephropathy (IgAN) is the most common glomerulonephritis worldwide. IgAN is responsible for a significant proportion of cases of kidney failure among glomerulopathies. Therefore, it is important to summarize the evidence regarding predictive factors of IgAN prognosis. The rates of 50% decline in estimated glomerular filtration rate (eGFR) or kidney failure were reported as 19% among children and adolescents, and 16% among children and adults, based on a European cohort (VALIGA) [1,2]. Similarly, the rate of eGFR or kidney failure in Chinese children was 14% [3]. Pediatric patients with IgAN had a higher proportion of active lesions than adults, and treatment options for adults were not entirely suitable for children [3–5]. Corticosteroid or immunosuppressive treatments were more common in children with active lesions than in adults and can lead to a favorable prognosis in the short term in children [6]. In the Oxford Classification, M, E, S, T, and C lesions were proposed as independent prognostic markers in IgAN: mesangial hypercellularity (M), endocapillary hypercellularity (E), segmental glomerulosclerosis (S), tubular atrophy and interstitial fibrosis (T), and crescent (C), i.e. MEST-C [7]. Some studies indicated that the Oxford classification had a predictive ability in IgAN patients [8–11]. As prognostic predictors, M, E, S, and T are included in the prognosis evaluation formula of IgA nephropathy [12]. However, most studies focused on adults or populations with a small proportion of children. Moreover, the results are not completely consistent, especially in pediatric patients. The contradictory findings were related to the differences in population, treatment and outcome criteria in studies. Therefore, this systematic review aims to summarize the predictive roles of prognosis by Oxford classification in pediatric patients with IgAN.
Methods
Information sources, search strategy, and eligibility criteria
The framed question, data extraction, reporting, and appraisal in this review were based on published guidance (Meta-analyses Of Observational Studies in Epidemiology [MOOSE] Checklist) to ensure a standard approach for transparent and complete reporting of systematic reviews and meta-analysis [13]. The search included controlled vocabulary and keywords for glomerulonephritis or IgA, if the patient is a child or adolescent, and the pathology or Oxford Classification/MEST-C score/MEST score. Due to no consensus definition of the Oxford Classification of IgAN before 2009, the Chinese and English databases were systematically searched from January 1, 2009, to July 3, 2024, including four English databases (Medline, Embase, the Cochrane Library, and Web of Science) and four Chinese databases (Wan Fang, VIP, CNKI, and Sinomed) in Chinese or English, with no limitation in publication type. The detailed search strategy is shown in Table S1. The literature was screen by two investigators (CW and NZ) based on titles and abstracts, and full articles were reviewed if eligible, with disagreements resolved by a third reviewer (XL). All studies regarding the association between five pathologic lesions in the Oxford Classification (M, E, S, T, and C) and the prognosis of pediatric patients with IgAN were included. Further, original studies including case reports, case control studies and studies regarding unclear lesions in the Oxford Classification were excluded. The detailed inclusion and exclusion criteria are shown in Table S2.
Data extraction, outcome, and the risk of bias assessment
The data extraction checklist is based on the previous reviews [14], including baseline patient characteristics (ethnicity, sex, and age), mean arterial pressure (MAP), eGFR, 24-h urine protein excretion, follow-up duration, percentage of patients receiving anti-hypertension medications, percentage receiving renin-angiotensin-aldosterone system inhibition treatment, percentage receiving immunosuppressive therapy, number of endpoint events, definition of kidney disease outcome, pathologic methodology, and characteristics of the Oxford Classification pathologic lesions (M, E, S, T, and C). Two investigators (NZ and CW) performed data extraction with disagreements resolved by a third reviewer (XL). Because the progression of IgAN to kidney failure is slower in children than in adults and children had a longer survival time with IgAN. Therefore, the outcome in our research was defined arbitrarily as poor kidney-related prognosis, i.e. GFR < 90 mL/min/1.73 m2 (GFR categories G2–G5) according to the Kidney Disease Improving Global Outcomes (KDIGO) Guideline in IgAN patients [15]. Based on the Newcastle-Ottawa scale (NOS) the risk of bias of studies was evaluated independently by two investigators (NZ and CW).
Statistical analysis
Individual-study including cohort studies which reported hazard ratios (HRs), with 95% confidence intervals (CIs), were extracted from each study for dichotomous outcomes before data pooling. The heterogeneity among the studies was estimated using the I2 statistics. However, considering studies on IgAN in children are heterogeneous, a random effect model was used for all analyses. Some studies reported that the clinical presentations were different in different ethnicities, and the E lesion and C lesion were more common in Asians than Europeans, so the subgroup analysis by ethnicity was performed to evaluate the predictive value of E lesion and C lesion among different ethnicities. The robustness of the synthesized results of associations between five pathologic lesions and poor kidney-related prognosis was assessed by sensitivity analysis after studies with adult patients were excluded. Moreover, publication bias was assessed by funnel plots. A two-sided p < 0.05 was regarded as significant for all analyses. All statistical analyses were done using Review Manager (version 5.4).
Results
Trial flow and study characteristics
A total of 2389 articles were searched in eight databases, and then 14 articles fulfilled the inclusion criterion with 5679 IgAN patients in which 710 poor kidney-related prognosis events happened [2–4,7,8,16–24]. Kappa for the selection and data extraction was 97.5%. The flow chart of study selection is shown in Figure 1. Fourteen studies with follow-up of 39.6–156.0 months in which HRs were used to estimate the association between the Oxford Classification Pathologic Lesions and poor kidney-related prognosis. Nine studies only included pediatric patients and five studies included both children and adults. Nine studies only included Asian populations (including Chinese, Japanese, and Korean), and five studies enrolled North Americans, South Americans, or European countries (including Poland, Spain, Sweden, and France). The endpoint events (i.e. poor kidney-related prognosis) of fourteen studies included the following: kidney failure, a GFR that decreased by more than 50%, chronic kidney diseases (CKD) stages 3a-5 (G3a-G5), GFR < 90 mL/min, SCr increasing more than 50%, and eGFR decreasing more than 10% and death. Twelve studies reported the proportion of patients who received immunosuppressive therapy, ranging from 11% to 65.5%. Initial eGFR, MAP, and proteinuria were common covariates to adjust the HRs in Cox regression analyses. Characteristics of the included studies are shown in Table 1 and the clinical, pathologic, and statistical characteristics of included studies are shown in Table S3 and Table S4. According to the risk of bias assessment based on the NOS, all included articles had a quality score ranging from 6 to 8. A total of 12 original studies were classified as high-quality evidence, and 2 original studies were classified as moderate-quality evidence (Table 2).
Figure 1.
Flow chart of study selection.
Table 1.
Clinical and pathologic of the included studies.
| Author, year [ref] | Shima, 2024 [16] | Wei Y, 2023 [18] | Xu L, 2023 [17] | Wu H, 2020 [3] | Shima Y, 2020 [19] |
|---|---|---|---|---|---|
| Country | Japan | China | China | China | Japan |
| Ethnicity (%) | |||||
| White | NA | NA | NA | NA | NA |
| Asian | 100% | 100% | 100% | 100% | 100% |
| African | NA | NA | NA | NA | NA |
| Other | NA | NA | NA | NA | NA |
| Design | Single | Single | Multiple | Single | Single |
| No. pts | 538 | 204 | 1168 | 1243 | 515 |
| Age (y) | 8 (4,11) | 16 (14,17) | 35.10 ± 11.73 | 13.7 | 10.7 (C-IgAN) 10.8 (Non-C-IgAN) |
| Age< 18 (y) | 100% | 100% | 2.40% | 100% | 100% |
| M: F | 298/240 | 132/72 | 583/585 | NA | 1.1:1 (C-IgAN) 1.2:1 (Non-C-IgAN) |
| F/U (mo) | NA | 90.7 (71.7,114.8) | 67.5 | 86.8 | 72 (Non-C-IgAN) 71 (Non-C-IgAN) |
| Proteinuria (g/d) | NA | 1.2 (0.7,2.5) | 1.27 (0.66–2.45) | 0.98 | NA |
| MAP (mmHg) | NA | NA | 93.59 ± 11.42 | 89.4 | NA |
| eGFR (mL/min/1.73m2) | NA | 132.9 | 85.91 | 102.1 | NA |
| CKD1/2/3/4/5 (%) | NA | NA | NA | NA | 17 (C3a-C5) |
| HTN (%) | NA | 78 (38.2%) | NA | 21.8 | NA |
| RAAS blockade | 20.8% | 56.4% | 95.1% | 69.8% | 0 (C-IgAN) 16.3%(Non-C-IgAN) |
| Immunosuppressive therapy | 27.1% | 52.0% | NA | 63.4% | 64.0% (C-IgAN) 17.3%(Non-C-IgAN) |
| Endpoint definition | ≥G3a CKD | ESKD or eGFR↓≥50% | ESKD | eGFR↓50% or kidney failure or death | CKD (G3a-G5) or kidney failure |
| Formula for determining eGFR | Schwartz formula | Schwartz formula(≤16 years); CKD-EPI(>16 years) | CKD-EPI | Schwartz formula(≤16 years); CKD-EPI(>16 years) | Schwartz formula and the updated Schwartz formula |
| No. endpoint events | 18 | 147 | 158 | 171 | 4 (C-IgAN) 13 (Non-C-IgAN) |
| No. kidney failure events | NA | NA | 158 | 82 | NA |
| Lesion (%) | |||||
| M0/M1 | 234//300 | 56/148 | 707/461 | 71/29 | 12/88 (C-IgAN)a 45/55 (Non-C-IgAN)a |
| E0/E1 | 180/222 | 150/54 | 768/400 | 65/35 | 17/83 (C-IgAN)a 47/53 (Non-C-IgAN)a |
| S0/S1 | 204/233 | 178/26 | 435/733 | 63/37 | 44/56 (C-IgAN)a 58/42 (Non-C-IgAN)a |
| T0/T1/T2 | 512/0/0 | 151/45/8 | 776/392 | 73/27a | 8/92 (C-IgAN)a 47/53 (Non-C-IgAN)a |
| C0/C1/C2 | 227/241/69 | 69/127/8 | NA | 52/48b | 0/0/100 (C-IgAN)a 45/47/8 (Non-C-IgAN)a |
| NOS | 8 | 7 | 7 | 8 | 7 |
Note: Values for continuous variables are given as mean.
“a” refers to T1 or T1 and T2; “b” refers to C1 or C1 and C2.
Abbreviations: CKD, chronic kidney disease; E, endocapillary hypercellularity; eGFR, estimated glomerular filtration rate; CKD (G3a-G5), chronic kidney disease stage 3a-5; F/U, follow-up; GFR, glomerular filtration rate; HTN, hypertension; M, mesangial hypercellularity; MAP, mean arterial pressure; NA, not available; pts, patients; RAAS, renin-angiotensin-aldosterone system; S, segmental glomerulosclerosis; T, tubular atrophy/interstitial fibrosis; No. number of; SCr, serum creatinine; NOS, Newcastle-Ottawa scale; eGFR↓50%, 50% decline in eGFR; CKD-EPI, the Chronic Kidney Disease Epidemiology Collaboration equation.
Table 2.
Risk of bias assessment based on the Newcastle-Ottawa scale.
| Author, year [ref] | Selection |
Comparability | Outcome/exposure |
Quality of evidencea | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Representativeness of the exposed cohort | Selection of the non-exposed cohort | Ascertainment of exposure | Demonstration that outcome interest was present at of start of the study not | Comparability of cohorts based on the design or analysis | Assessment of outcome | Was follow-up long enough for the outcome to occur | Adequacy of follow-up of cohorts | Quality score | ||
| Shima, 2024 [16] | * | * | * | * | * | * | * | * | 8 | High |
| Wei Y, 2023 [18] | * | * | * | * | * | * | * | 7 | High | |
| Xu L,2023 [17] | * | * | * | * | * | * | * | 7 | High | |
| Wu H, 2020 [3] | * | * | * | * | * | * | * | * | 8 | High |
| Shima Y, 2020 [19] | * | * | * | * | * | * | * | 7 | High | |
| Coppo R, 2017 [2] | * | * | * | * | * | * | * | 7 | High | |
| Fabiano R. C.G, 2017 [4] | * | * | * | * | * | * | * | * | 8 | High |
| Halling S.E, 2012 [22] | * | * | * | * | * | * | * | 7 | High | |
| Shima Y, 2012 [20] | * | * | * | * | * | * | * | 7 | High | |
| Kang, S.H., 2012 [21] | * | * | * | * | * | * | 6 | Moderate | ||
| Gutiérrez E, 2012 [23] | * | * | * | * | * | * | * | 7 | High | |
| Le W.B, 2012 [8] | * | * | * | * | * | * | * | * | 8 | High |
| Katafuchi R, 2011 [24] | * | * | * | * | * | * | * | 7 | High | |
| Cattran D.C, 2009 [7] | * | * | * | * | * | * | 6 | Moderate | ||
: According to quality score of the Newcastle-Ottawa scale, quality of evidence was classified into three types: high (quality score > 6), moderate (6 ≤ quality score < 4), and low (quality score ≤ 4).
| Author, year [ref] | Coppo R, 2017 [2] | Fabiano R, 2017 [4] | Halling S.E, 2012 [22] | Shima Y, 2012 [20] | Kang S.H, 2012 [21] |
|---|---|---|---|---|---|
| Country | Multi-country | Brazil | Sweden | Japan | Korea |
| Ethnicity (%) | |||||
| White | 97.6% | 51.9% | NA | NA | NA |
| Asian | 0.6% | NA | NA | 100% | 100% |
| African | 0.6% | 11.1% | NA | NA | NA |
| other | 1.2% | 37% | NA | NA | NA |
| Design | Multiple | Single | Single | Single | Single |
| No. pts | 174 | 54 | 99 | 161 | 197 |
| Age (y) | 12.7 | 9.7 | 12.0 | 11.7 | 32.4 |
| Age< 18 (y) | 100% | 100% | 100% | 100%* | NA |
| M: F | 2.6:1 | 1.3:1 | 1.4:1 | 1.7:1 | 1.36:1 |
| F/U (mo) | 55.6 | 90 | 156 | 54 | 56.8 |
| Proteinuria (g/d) | 0.84 | 0.97 | 2 | 0.7 | 2.07 |
| MAP (mmHg) | NA | NA | 85.4 | 79 | 91.4 |
| eGFR (mL/min/1.73m2) | 117 | 143.9 | 100 | 103 | 87 |
| CKD1/2/3/4/5 (%) | NA | NA | 75/16/4/3/2 | NA | NA |
| HTN (%) | NA | 22.2 | 30 | NA | NA |
| RAAS blockade | 66.7% | 43.4% | 86% | NA | 18.3% |
| Immunosuppressive therapy | 50.57% | 24.5% | 43% | 32% | 38.1% |
| Endpoint definition | kidney failure or eGFR↓50% | eGFR↓50% or kidney failure or death | eGFR↓50% or kidney failure | CKD (G3a-G5) | eGFR↓50% or kidney failure |
| Formula for determining eGFR | the modified Schwartz formula(≤18 years); MDRD (18–23 years) | Schwartz formula | clearances of inulin during water diuresis and plasma clearance of iohexol | Schwartz formula | MDRD |
| No. endpoint events | 11 | 10 | 18 | 7 | 16 |
| No. kidney failure events | 7 | 5 | 15 | NA | NA |
| Lesion (%) | |||||
| M0/M1 | 78/22 | 67/33 | 70/30 | 64/36 | 26/74 |
| E0/E1 | 86/14 | NA | 90/10 | 42/58 | 89/11 |
| S0/S1 | 57/43 | 76/24 | 76/24 | 91/9 | 44/56 |
| T0/T1/T2 | 94/6a | 96/4a | 84/12/4 | 99/1a | 66/26/8 |
| C0/C1/C2 | 85/15b | 93/7b | 82/18b | 48/52b | NA |
| NOS | 7 | 8 | 7 | 7 | 6 |
Note: Values for continuous variables are given as mean.
“a” refers to T1 or T1 and T2; “b” refers to C1 or C1 and C2.
*The children in this Japanese study were treated at children’s hospitals under the age of 21, not under 18 as we normally think.
Abbreviations: CKD, chronic kidney disease; E, endocapillary hypercellularity; eGFR, estimated glomerular filtration rate; CKD (G3a-G5), chronic kidney disease stage 3a-5; F/U, follow-up; GFR, glomerular filtration rate; HTN, hypertension; M, mesangial hypercellularity; MAP, mean arterial pressure; NA, not available; pts, patients; RAAS, renin-angiotensin-aldosterone system; S, segmental glomerulosclerosis; T, tubular atrophy/interstitial fibrosis; No. number of; SCr, serum creatinine; NOS, Newcastle-Ottawa scale; eGFR↓50%, 50% decline in eGFR; SCr↑50% or 100%, 50% or 100% increase in SCr.; MDRD, Modification of Diet in Renal Disease.
| Author, year [ref] | Gutiérrez E, 2012 [23] | Le W.B, 2012 [8] | Katafuchi R, 2011 [24] | Cattran D.C, 2009 [7] |
|---|---|---|---|---|
| Country | Spain | China | Japan | Multi-country |
| Ethnicity (%) | ||||
| White | 100% | NA | NA | 66% |
| Asian | NA | 100% | 100% | 27% |
| African | NA | NA | NA | 3% |
| other | NA | NA | NA | 4% |
| Design | Multiple | Single | Single | Multi |
| No. pts | 141 | 218 | 702 | 265 |
| Age (y) | 23.7 | 14.0 | 30 (8–82) | 36 |
| Age< 18 (y) | NA | 100% | NA | 22.3% |
| M: F | 1.76:1 | 1.9:1 | 1.0:1.2 | 2.6:1 |
| F/U (mo) | 108 | 56 | 62 | 65 |
| Proteinuria (g/d) | 0.2 | 1.5 | 0.85 | 1.7 |
| MAP (mmHg) | NA | 88 | 92 | 98 |
| eGFR (mL/min/1.73m2) | 111 | 134 | 82 | 83 |
| CKD1/2/3/4/5 (%) | NA | NA | 37/37/21/3/2 | 36/38/26/0/0 |
| HTN (%) | 16.3% | 6.5 | NA | 31% |
| RAAS blockade | 41.8% | 61.5% | 37% | 74% |
| Immunosuppressive therapy | NA | 56% | 32% | 29% |
| Endpoint definition | SCr↑50% or 100% or kidney failure | eGFR↓50% or kidney failure | kidney failure | eGFR↓50% or kidney failure or the slope of eGFR |
| Formula for determining eGFR | MDRD | Schwartz formula (≤16 years); MDRD (>16 years) | 194*Cr-1.094*Age0.287 (If female, *0.739) | Schwartz formula (children); MDRD (adults) |
| No. endpoint events | 6 | 24 | 84 | 58 |
| No. kidney failure events | 0 | NA | 84 | 34 |
| Lesion (%) | ||||
| M0/M1 | 67/33 | 55/45 | 88/12 | 20/80 |
| E0/E1 | 92/8 | 77/23 | 58/42 | 58/42 |
| S0/S1 | 84/16 | 38/62 | 21/79 | 65/200 |
| T0/T1/T2 | 95/5a | 93/7a | 70/18/12 | 207/58a |
| C0/C1/C2 | NA | 56/44b | 37/63b | 55/45b |
| NOS | 7 | 8 | 7 | 6 |
Note: Values for continuous variables are given as mean.
“a” refers to T1 or T1 and T2; “b” refers to C1 or C1 and C2.
Abbreviations: CKD, chronic kidney disease; E, endocapillary hypercellularity; eGFR, estimated glomerular filtration rate; CKD (G3a-G5), chronic kidney disease stage 3a-5; F/U, follow-up; GFR, glomerular filtration rate; HTN, hypertension; M, mesangial hypercellularity; MAP, mean arterial pressure; NA, not available; pts, patients; RAAS, renin-angiotensin-aldosterone system; S, segmental glomerulosclerosis; T, tubular atrophy/interstitial fibrosis; No. number of; SCr, serum creatinine; NOS, Newcastle-Ottawa scale; eGFR↓50%, 50% decline in eGFR; MDRD, Modification of Diet in Renal Disease.
Application of the Oxford Classification pathologic lesions in IgAN
Mesangial hypercellularity
IgAN was classified into M0 and M1 according to mesangial hypercellularity lesions; M0 was defined as a mesangial hypercellularity score ≤0.5; and M1 was defined as a mesangial hypercellularity score > 0.5 (equal to >50% of the glomeruli with >3 mesangial cells in the Periodic Acid-Schiff staining). Thirteen original studies, with 632 poor kidney-related prognosis events of 5538 IgAN patients, reported the predictive value of the M lesions [2–4,7,8,16–22,24]. Figure 2 showed that M1 was associated with a higher risk of poor outcome in IgAN patients compared with M0, and pooled HR was 1.79 (95%CI, 1.46–2.19) by the random effect model. In sensitivity analysis for studies with children only, the synthesized results still showed that M1 had a higher risk of poor kidney-related prognosis in pediatric patients with IgAN compared with M0, and pooled HR was estimated by random effect model 1.65 (95%CI, 1.35–2.02; p < 0.001) (Figure S1).
Figure 2.
HRs Of poor outcome of IgAN with M1 vs M0 (mesangial hypercellularity score ≤ 0.5 vs >0.5) and E1 vs E0.
Endocapillary hypercellularity
According to whether the E lesion was present or absent, IgAN patients were classified into E1 and E0, and IgAN patients with any E lesions were defined as E1. Eleven studies with 563 poor kidney-related prognosis prognosis events of 5099 IgAN patients reported the association between E lesion and prognosis of IgAN [3,4,8,16–22,24]. Overall, the evidence was marginal to promote that E1 increased the risk of poor kidney-related prognosis of IgAN, compared with E0 (pooled HR, 1.45; 95%CI, 0.99–2.11; p = 0.05, estimated by random effect model). However, as part of the subgroup analysis related to ethnicity (Asian population subgroups included Chinese, Japanese, and Korean, while the other ethnic populations included Swedish and Brazilian people), E1 increased the risk of poor kidney-related prognosis events of IgAN compared with E0 in other ethnic populations (pooled HR, 11.88; 95%CI, 2.40–58.89; p < 0.001), while in Asian populations the association was with no statistical significance (pooled HR, 1.17; 95%CI, 0.94–1.45; p = 0.16). Among eleven studies, three studies enrolled adults. The results of the sensitivity analysis were consistent with the subgroup analysis (see Figure S2).
Segmental glomerulosclerosis/adhesion
IgAN patients with reported segmental glomerulosclerosis or adhesions were classified as S1, and without segmental glomerulosclerosis or adhesion were defined as S0 (S0 group as reference). Fourteen studies included segmental glomerulosclerosis/adhesion lesion (710 cases from 5679 patients) [2–4,7,8,16–24]. S1 was associated with an increased risk of poor kidney-related prognosis events (pooled HR, 2.13; 95%CI, 1.68–2.70; p < 0.001, random effect model). Shown in Figure 3. The results of sensitivity analysis (studies included children only) also showed an increased association between S1 and poor kidney-related prognosis events (see Figure S3).
Figure 3.
HRs Of poor outcome for IgAN with S1 vs S0 and C1 or C2 vs C0.
Tubular atrophy and interstitial fibrosis
According to the level of tubular atrophy and interstitial fibrosis, T lesion was classified into three types: T0 was defined as tubular atrophy or interstitial fibrosis less than 25%, T1 was defined as >25% and ⩽50% tubular atrophy or interstitial fibrosis and T2 was defined as tubular atrophy or interstitial fibrosis more than 50%. The meta-analysis was performed in three groups: T1 vs T0, T2 vs T0, T1, or T2 vs T0.
T1 vs T0: Four studies with 149 poor kidney-related prognosis events of 1643 IgAN patients reported the association between T1 and the poor kidney-related prognosis events of IgAN [7,19,20,24]. Figure 4 showed that T1 increased the risk of poor kidney-related prognosis of IgAN compared with T0 (pooled HR, 7.28; 95%CI, 2.33–22.78; p < 0.001) by random effect model, with the evidence of heterogeneity (I2 = 64%, p = 0.04). As for the sensitivity analysis, after two studies that involved adult patients were excluded, the result still showed that T1 increased the risk of poor kidney-related prognosis in IgAN compared with T0 (Figure S4).
Figure 4.
HRs Of poor outcome for IgAN with T1 vs T0, T2 vs T0, and T1 or T2 vs T0, respectively.
T2 vs T0: Two studies with 142 poor kidney-related prognosis events of 967 IgAN patients reported the association between T2 with poor kidney-related prognosis of IgAN [7,24]. Figure 4 showed that T2 increased the risk of poor kidney-related prognosis outcome of IgAN compared with T0 (pooled HR, 10.99; 95%CI, 5.85–20.62; p < 0.001) according to the random effect model.
T1 or T2 vs T0: Nine studies with 472 poor kidney-related prognosis events of 4236 IgAN patients reported the association between T1 and T2 with poor kidney-related prognosis of IgAN [3,4,8,16–19,21,22]. Figure 4 showed that T1 or T2 increased the risk of poor kidney-related prognosis of IgAN compared with T0 (pooled HR, 2.64; 95%CI, 1.81–3.86; p < 0.001) according to the random effect model, which showed evidence of heterogeneity (I2 = 63%, p = 0.009). In sensitivity analysis, after studies with adults were excluded, the result still showed that T1 or T2 increased the risk of poor kidney-related prognosis of IgAN compared with T0 (Figure S5).
Cellular/fibrocellular crescents
C1 is defined as crescents in ⩽25% of glomeruli, C2 is defined as crescents in >25% of glomeruli, and C0 is defined as the absence of crescents. Eight studies including 453 poor kidney-related prognosis events of 4146 IgAN patients reported the association between C lesion and poor kidney-related prognosis of IgAN [3,8,16–20,22]. Because the clinical presentations varied among different ethnicities, this meta-analysis was conducted in different subgroups (Asian and other populations). Overall, Figure 3 showed that C1 or C2 increased the risk of poor kidney-related prognosis of IgAN compared with C0 and pooled HR was 1.64 (95%CI: 1.29–2.08) according to the random effect model with the evidence of heterogeneity (I2 = 22%, p = 0.26). Further, we did also observed similar results in the subgroup analyses by ethnicity: Asian population (pooled HR, 1.55; 95%CI, 1.29–1.85; p < 0.001) by using the random effect model. The result of the sensitivity analysis was consistent with that of the subgroup analysis (see Figure S6).
Publication bias
The funnel plot was used to estimate the publication bias of original studies which evaluated the association between five lesions of Oxford Classification and the prognosis of IgAN (Figure S7).
Discussion
The Oxford Classification was proposed as an independent prognostic indicator in IgAN. Compared with adults, fewer pediatric patients with IgAN were included and the endpoints event were difficult to be achieved. So, the predictive value of Oxford Classification in children with IgAN needs to be further verified.
The systematic review and meta-analysis is a strategy to overcome these obstacles. A total of 14 studies, encompassing 5679 IgAN patients and 710 poor kidney-related prognosis events were identified. This research showed that M, S, T, and C were associated with the increased risk of poor kidney-related prognosis in pediatric patients with IgAN.
M is an essential pathological change of IgAN. Majority of studies suggested that M1 was related to poor kidney-related prognosis, whether in adults or in children [20,21]. In this research, M1 was also associated with a higher risk of poor kidney-related prognosis. Furthermore, M score was also associated with poor renal outcomes after immunosuppressive therapy [25]. Children under 16 years of age with M0 and eGFR > 90 mL/min/1.73 m2 at presentation had a higher proteinuria remission rate than those with M1. Combined with proteinuria, M can predict the prognosis of pediatric patients with IgAN more accurately [26]. However, so far, no treatment strategy has been applied only according to the presentation of mesangial hypercellularity.
In our results, E1 did not increase the risk of poor kidney-related prognosis of pediatric patients with IgAN in the Asian population subgroup but increased the risk of poor kidney-related in the non-Asian population subgroup significantly. However, researchers proposed that two reasons perhaps resulted in inconsistent associations between E1 and poor kidney-related prognosis among Asian and non-Asian populations: Firstly, the number of original studies and sample size are inadequate in non-Asian population subgroup and the significant association between E1 and poor kidney-related prognosis in non-Asian needed a cautious explanation. Secondly, the different covariates of lesion E in Asian and non-Asina populations may resulted in the varied HRs. As active lesions, E1 was associated with daily proteinuria rising from under 0.5 g/day to 1, or 2 g/day or more, in children [1,27]. The independent predictive value of E is reduced by glucocorticoid/immunosuppressive therapy [10,28]. Nearly half of the original studies based on the Asian population included the glucocorticoid/immunosuppressive therapy situation as a covariate to adjust the association between E lesion and poor kidney-related prognosis, but the glucocorticoid/immunosuppressive therapy situation was not considered in the original studies of non-Asian population subgroup. The predictive value of E1 remains controversial. The VALIGA study examined 1147 IgAN patients from European countries including 174 children, and found that E1 did not independently predict poor kidney prognoses, even in the subgroup of 174 children aged less than 18 years [2]. In a meta-analysis with IgAN cohorts in adults, E1 was unrelated to the outcome [29]. In pediatric studies from Sweden, E1 was a predictor of IgAN prognosis only in models constructed by combining proteinuria [22]. Fabiano R indicated that only baseline proteinuria and endocapillary hypercellularity were independent predictors of kidney dysfunction in pediatric patients from South America [4].
Compared with previous studies, it is consistent that chronic pathological lesions such as T1 and T2 or S1 (especially T) can predict poor kidney-related prognosis in most studies including adults or children [1,3,8,22,30,31]. Some studies indicated that only T lesions were associated with the rate of eGFR loss in the entire cohort, including children and adults [8,11]. Moreover, in the study from China which enrolled 281 pediatric patients with IgAN, T was the only feature independently associated with kidney outcomes [8]. Repeat biopsies confirmed that only tubular atrophy/interstitial fibrosis was associated with the kidney outcome in IgAN adults [32]. However, compared to adults, pediatric patients more often than not lack S1 lesion, so more studies with large pediatric patients should be carried out to support evidence of the association between S lesion and poor kidney-related prognosis.
C was a significant active lesion that was associated with the prognosis of the disease [33]. IgAN patients with crescents had more severe clinicopathological features and a worse prognosis [34]. C lesion showed a significant association with the poor kidney-related prognosis of IgAN in pediatric and adult patients [11,35]. It reported that a greater number of crescents created a worse prognosis. C1 was mainly associated with the increased risk of halving of eGFR [36]. Haas et al. reported that patients with C1 were at an increased risk of a poor prognosis without immunosuppression and patients with C2 were at a greater risk of progression, even with immunosuppression [33]. Having ≥30%–50% crescents identified in a biopsy is related to the poor prognosis [34,37]. ≥25% Of crescents was an independent risk factor for a ≥ 50% reduction in eGFR, or kidney failure in treated and untreated IgAN patients, and Crescents ≥43.7% was an independent risk factor for kidney failure in those with immunosuppressants [34]. However, the predictive role of C1 is still controversial. Based on our study, we found a significant association between C1 and poor kidney-related prognosis in pediatric patients with IgAN regardless of the Asian or non-Asian population. Ossareh et al. found that the risk of poor kidney-related prognosis was strongly increased in the combined presence of at least one crescent and T1 ≥ 25% [37]. The median kidney survival was 78.0 months in patients with T0 + C0 and 32.3 months in patients with T1 + C1. However, Zhang X et al. failed to replicate the association between crescents and kidney function progression in 1152 Chinese adult patients with IgAN followed for more than 1 year. In patients with nephrotic-range proteinuria, patients in either the C1 or C2 group had a higher rate of decline equal to or greater than 50% in eGFR or to kidney failure [38,39]. In the future, the combined presence of different lesions, with symptomatic manifestation, maybe considered to predict a poor kidney-related prognosis.
The strengths of this meta-analysis are as follows: To our knowledge, this research is the first time to clarify the predictive value of Oxford Classification in pediatric patients with IgAN by meta-analysis. Furthermore, the stability of the results is evaluated through sensitivity analyses according to the population of original studies. However, there are also some limitations in the present study: Firstly, although this meta-analysis provided evidence of the predictive role of Oxford Classification, there is no specific treatment for pediatric IgAN patients or those lesions with high risk currently. Secondly, several studies have suggested that immunosuppressant therapy can affect the predictive value of lesions C or E [33,40]. However, specific effect values between histologic scores and IgA prognosis in treated and untreated patients were difficult to extract from original studies. We will conduct in-depth research to explore the effect of immunosuppressant therapy treatment on the association between Oxford Classification and poor kidney-related prognosis. Last but not least, the small sample size of original studies in the subgroup analysis of E lesions might contribute to the lack of power to prove the predictive role in pediatric patients with IgAN.
This meta-analysis identified the predictive role of Oxford classification for poor kidney-related prognosis in pediatric patients with IgAN. It provided evidence for clinicians to identify IgAN children with a high risk of poor prognosis.
Supplementary Material
Acknowledgments
Thank you for the methodological support from Xiaoxia Peng in the Center for Clinical Epidemiology and Evidence-based Medicine, Beijing Children’s Hospital, Capital Medical University.
Funding Statement
This research was supported by Baoding Scientific and Technological Program (Grant No. 2141ZF280), the Medical Scientific Research Project of Health Commission of Hebei Province (Grant No. 20230255), and Baoding Scientific and Technological Program (Grant No. 2241ZF077).
Authors’ contributions
Xiaohang Liu: Data curation, Formal analysis, Methodology, Visualization, Writing—original draft, Writing—review & editing. Chen Wang: Data curation, Formal analysis, Methodology, Visualization, Writing—original draft, Writing—review & editing. Zimo Sun: Data curation, Methodology, Writing—review & editing. Man Liu: Data curation, Writing—review & editing. Nan Zhou: Conceptualization, Data curation, Writing—original draft, Funding acquisition, Supervision. All authors read and approved the final manuscript.
Data sharing statement
Data sharing does not apply to this article as no datasets were generated or analyzed during the current study. Our research is a systematic review, which is a secondary use of original articles. The data in the meta-analysis was extracted from the included articles [2–4,7,8,16–24], which can be available from in the included researches in the systematic review. Thus, data sharing is not applicable to our research.
Disclosure statement
The author reports no conflicts of interest in this work.
Registrations of systematic review
This systematic review was registered on the PROSPERO platform with registration number “CRD42022353177”, but the protocol was not prepared.
Reporting statement
This research was reported according to the MOOSE, and the detailed MOOSE checklist is shown in Table S5.
References
- 1.Coppo 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(4):828–836. doi: 10.1038/ki.2014.63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Coppo R, Lofaro D, Camilla RR, et al. Risk factors for progression in children and young adults with IgA nephropathy: an analysis of 261 cases from the VALIGA European cohort. Pediatr Nephrol. 2017;32(1):139–150. doi: 10.1007/s00467-016-3469-3. [DOI] [PubMed] [Google Scholar]
- 3.Wu HY, Fang X, Xia ZK, et al. Long-term renal survival and undetected risk factors of IgA nephropathy in Chinese children-a retrospective 1243 cases analysis from single centre experience. J Nephrol. 2020;33(6):1263–1273. doi: 10.1007/s40620-020-00767-4. [DOI] [PubMed] [Google Scholar]
- 4.Fabiano RCG, Araújo SA, Bambirra EA, et al. The Oxford Classification predictors of chronic kidney disease in pediatric patients with IgA nephropathy. J Pediatr (Rio J). 2017;93(4):389–397. doi: 10.1016/j.jped.2016.09.003. [DOI] [PubMed] [Google Scholar]
- 5.Cambier A, Rabant M, El Karoui K, et al. Clinical and histological differences between adults and children in new onset IgA nephropathy. Pediatr Nephrol. 2020;35(10):1897–1905. doi: 10.1007/s00467-020-04614-3. [DOI] [PubMed] [Google Scholar]
- 6.Coppo R. Pediatric IgA nephropathy in Europe. Kidney Dis (Basel). 2019;5(3):182–188. doi: 10.1159/000495751. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Cattran DC, Coppo R, Cook HT, et al. The Oxford classification of IgA nephropathy: rationale, clinicopathological correlations, and classification. Kidney Int. 2009;76(5):534–545. doi: 10.1038/ki.2009.243. [DOI] [PubMed] [Google Scholar]
- 8.Le W, Zeng CH, Liu Z, et al. Validation of the Oxford classification of IgA nephropathy for pediatric patients from China. BMC Nephrol. 2012;13(1):158. doi: 10.1186/1471-2369-13-158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Coppo R, Robert T.. IgA nephropathy in children and in adults: two separate entities or the same disease? J Nephrol. 2020;33(6):1219–1229. doi: 10.1007/s40620-020-00725-0. [DOI] [PubMed] [Google Scholar]
- 10.Coppo R, Troyanov S, Camilla R, et al. The Oxford IgA nephropathy clinicopathological classification is valid for children as well as adults. Kidney Int. 2010;77(10):921–927. doi: 10.1038/ki.2010.43. [DOI] [PubMed] [Google Scholar]
- 11.Coppo R, D’Arrigo G, Tripepi G, et al. Is there long-term value of pathology scoring in immunoglobulin A nephropathy? A validation study of the Oxford Classification for IgA Nephropathy (VALIGA) update. Nephrol Dial Transplant. 2020;35(6):1002–1009. doi: 10.1093/ndt/gfy302. [DOI] [PubMed] [Google Scholar]
- 12.Barbour SJ, Coppo R, Zhang H, et al. Evaluating a new international risk-prediction tool in IgA nephropathy. JAMA Intern Med. 2019;179(7):942–952. doi: 10.1001/jamainternmed.2019.0600. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Brooke BS, Schwartz TA, Pawlik TM.. MOOSE reporting guidelines for meta-analyses of observational studies. JAMA Surg. 2021;156(8):787–788. doi: 10.1001/jamasurg.2021.0522. [DOI] [PubMed] [Google Scholar]
- 14.D’Amico G. Natural history of idiopathic IgA nephropathy and factors predictive of disease outcome. Semin Nephrol. 2004;24(3):179–196. doi: 10.1016/j.semnephrol.2004.01.001. [DOI] [PubMed] [Google Scholar]
- 15.KDIGO 2024 clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney International. 2024;105(4):S117–S314. doi: 10.1016/j.kint.2023.10.018. [DOI] [PubMed] [Google Scholar]
- 16.Shima Y, Mukaiyama H, Tanaka Y, et al. Factors related to recurrence of proteinuria in childhood IgA nephropathy. Pediatr Nephrol. 2024;39(2):463–471. doi: 10.1007/s00467-023-06116-4. [DOI] [PubMed] [Google Scholar]
- 17.Xu LL, Zhang D, Weng HY, et al. Machine learning in predicting T-score in the Oxford classification system of IgA nephropathy. Front Immunol. 2023;14:1224631. doi: 10.3389/fimmu.2023.1224631. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Wei YQ, Li XJ, Wu HY, et al. Long-term prognostic analysis of IgA nephropathy with focal segmental glomerulosclerosis in children. Chin J Nephrol. 2023;39(6):428–437. doi: 10.3760/cma.j.cn441217-20221212-01219. [DOI] [Google Scholar]
- 19.Shima Y, Nakanishi K, Hama T, et al. Crescentic IgA nephropathy in children. Pediatr Nephrol. 2020;35(6):1005–1014. doi: 10.1007/s00467-020-04483-w. [DOI] [PubMed] [Google Scholar]
- 20.Shima Y, Nakanishi K, Hama T, et al. Validity of the Oxford classification of IgA nephropathy in children. Pediatr Nephrol. 2012;27(5):783–792. doi: 10.1007/s00467-011-2061-0. [DOI] [PubMed] [Google Scholar]
- 21.Kang 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(1):252–258. doi: 10.1093/ndt/gfr295. [DOI] [PubMed] [Google Scholar]
- 22.Edström Halling S, Söderberg MP, Berg UB.. Predictors of outcome in paediatric IgA nephropathy with regard to clinical and histopathological variables (Oxford classification). Nephrol Dial Transplant. 2012;27(2):715–722. doi: 10.1093/ndt/gfr339. [DOI] [PubMed] [Google Scholar]
- 23.Gutiérrez E, Zamora I, Ballarín JA, et al. Long-term outcomes of IgA nephropathy presenting with minimal or no proteinuria. J Am Soc Nephrol. 2012;23(10):1753–1760. doi: 10.1681/ASN.2012010063. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Katafuchi R, Ninomiya T, Nagata M, et al. Validation study of oxford classification of IgA nephropathy: the significance of extracapillary proliferation. Clin J Am Soc Nephrol. 2011;6(12):2806–2813. doi: 10.2215/cjn.02890311. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Rui Y, Yang Z, Zhai Z, et al. The predictive value of Oxford MEST-C classification to immunosuppressive therapy of IgA nephropathy. Int Urol Nephrol. 2022;54(4):959–967. doi: 10.1007/s11255-021-02974-9. [DOI] [PubMed] [Google Scholar]
- 26.Barbour SJ, Espino-Hernandez G, Reich HN, et al. The MEST score provides earlier risk prediction in IgA nephropathy. Kidney Int. 2016;89(1):167–175. doi: 10.1038/ki.2015.322. [DOI] [PubMed] [Google Scholar]
- 27.Kamano C, Shimizu A, Joh K, et al. A cross-sectional study in patients with IgA nephropathy of correlations between clinical data and pathological findings at the time of renal biopsy: a Japanese prospective cohort study. Clin Exp Nephrol. 2021;25(5):509–521. doi: 10.1007/s10157-021-02022-x. [DOI] [PubMed] [Google Scholar]
- 28.Trimarchi 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(5):1014–1021. doi: 10.1016/j.kint.2017.02.003. [DOI] [PubMed] [Google Scholar]
- 29.Lv J, Shi S, Xu D, et al. Evaluation of the Oxford Classification of IgA nephropathy: a systematic review and meta-analysis. Am J Kidney Dis. 2013;62(5):891–899. doi: 10.1053/j.ajkd.2013.04.021. [DOI] [PubMed] [Google Scholar]
- 30.Xu K, Zhang L, Ding J, et al. Value of the Oxford classification of IgA nephropathy in children with Henoch-Schönlein purpura nephritis. J Nephrol. 2018;31(2):279–286. doi: 10.1007/s40620-017-0457-z. [DOI] [PubMed] [Google Scholar]
- 31.Wu H, Xia Z, Gao C, et al. The correlation analysis between the Oxford classification of Chinese IgA nephropathy children and renal outcome: a retrospective cohort study. BMC Nephrol. 2020;21(1):247. doi: 10.1186/s12882-020-01913-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Shen XH, Liang SS, Chen HM, et al. Reversal of active glomerular lesions after immunosuppressive therapy in patients with IgA nephropathy: a repeat-biopsy based observation. J Nephrol. 2015;28(4):441–449. doi: 10.1007/s40620-014-0165-x. [DOI] [PubMed] [Google Scholar]
- 33.Haas M, Verhave JC, Liu Z-H, et al. A multicenter study of the predictive value of crescents in IgA nephropathy. J Am Soc Nephrol. 2017;28(2):691–701. doi: 10.1681/ASN.2016040433. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Du Y, Chen S, Wang F, et al. The significance of crescents on the clinical features and outcomes of primary immunoglobin a nephropathy. Front Med (Lausanne). 2022;9:864667. doi: 10.3389/fmed.2022.864667. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Shao X, Li B, Cao L, et al. Evaluation of crescent formation as a predictive marker in immunoglobulin A nephropathy: a systematic review and meta-analysis. Oncotarget. 2017;8(28):46436–46448. doi: 10.18632/oncotarget.17502. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Park S, Baek CH, Park SK, et al. Clinical significance of crescent formation in IgA nephropathy: a multicenter validation study. Kidney Blood Press Res. 2019;44(1):22–32. doi: 10.1159/000497808. [DOI] [PubMed] [Google Scholar]
- 37.Ossareh S, Nazemzadeh N, Asgari M, et al. The added value of crescents on Oxford Classification score in risk stratification of end-stage kidney disease in patients with IgA nephropathy. Iran J Kidney Dis. 2022;16(2):115–124. [PubMed] [Google Scholar]
- 38.Zhang X, Shi S, Ouyang Y, et al. A validation study of crescents in predicting ESRD in patients with IgA nephropathy. J Transl Med. 2018;16(1):115. doi: 10.1186/s12967-018-1488-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Trimarchi H, Haas M, Coppo R.. Crescents and IgA nephropathy: a delicate marriage. J Clin Med. 2022;11(13):3569. doi: 10.3390/jcm11133569. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Hotta O, Furuta T, Chiba S, et al. Regression of IgA nephropathy: a repeat biopsy study. Am J Kidney Dis. 2002;39(3):493–502. doi: 10.1053/ajkd.2002.31399. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data sharing does not apply to this article as no datasets were generated or analyzed during the current study. Our research is a systematic review, which is a secondary use of original articles. The data in the meta-analysis was extracted from the included articles [2–4,7,8,16–24], which can be available from in the included researches in the systematic review. Thus, data sharing is not applicable to our research.