Spectrum of biopsy proven renal disease in South Asian children: Two decades at a tropical tertiary care centre

Abstract

Aim

We report findings from a large single centre paediatric renal biopsy cohort in South Asia.

Methods

We analyzed all renal biopsies performed on children aged ≤18 years between 1996 and 2015 at our centre. The clinical characteristics and histological diagnosis pertaining to each case, distribution of renal diseases in children with various clinical presentations, and changes in the pattern of kidney disease during the study period were analyzed.

Results

A total of 1740 paediatric kidney biopsies were performed during the study period. The mean age was 12.8±4.9 years (8 months to 18 years) and the male: female ratio was 1.5:1. The most common indication for renal biopsy was nephrotic syndrome (63.2%) followed by acute nephritic syndrome (13%). Minimal change disease was the most common cause of nephrotic syndrome while endocapillary proliferative glomerulonephritis (65.7% infection related), remained the commonest cause of acute nephritic syndrome. IgA nephropathy was the commonest cause of chronic kidney disease. Contrary to trends in European paediatric cohorts, the frequency of lupus nephritis increased over the two decades of the study, while that of endocapillary proliferative glomerulonephritis did not show any appreciable decline.

Conclusions

This study provides the largest data on biopsy proven renal disease in children from South Asia published till date and highlights important differences in the spectrum and trends of kidney disease compared to data from other regions.

Introduction

Paediatric renal biopsy registries provide information regarding the region specific pathological entities one is likely to encounter in children of different age groups. This is of special importance in South Asia, where the biopsy policy is conservative due to lack of expertise in performing paediatric kidney biopsies, and where children often present late in the course of their kidney disease.¹

This study presents data from a large single centre paediatric renal biopsy cohort in South Asia, catering to children from India, and the neighbouring countries of Nepal, Bangladesh and Bhutan.

Methods

We performed a retrospective review of consecutive children (≤18 years) who underwent native kidney biopsy at our centre between January 1996 and December 2015. Clinical and demographic details including age, sex, clinical presentation, indication for renal biopsy and histological diagnoses were retrieved from the hospital information system and treatment records.

Renal biopsies were performed after adequate counseling and written consent from the patient’s guardians in each instance. Prior to renal biopsy, basic investigations which included hemoglobin, prothrombin time, activated partial thromboplastin time, platelet count, blood urea, serum creatinine, 24 hour urine protein or urine protein/creatinine ratio and serum albumin were ordered. Blood borne virus screening (Hepatitis B,C and HIV) was routinely performed for all patients from 2006 onwards. Prior to that, screening was carried out where indicated, based on biopsy reports. All biopsies were performed by a consultant or registrar using an automated gun under ultrasound guidance. At least two cores of kidney tissue were obtained in each case and sent for histopathological examination, which was performed by two well-trained pathologists from the same laboratory, the senior pathologist remaining unchanged throughout the study period. For light microscopy (LM), 3 μm thick paraffin sections were stained with haematoxylineosin and periodic acid Schiff reagent in all cases, and other stains where indicated. Immunofluorescence (IF) sections were stained with anti-human IgG, IgM, IgA, C3, C4 in all cases and other antisera where indicated. Electron microscopy (EM) was performed only in selected cases because of financial constraints, either at the time of biopsy or on a separate core of tissue, obtained after LM and IF reports were available.

The indications for biopsy included nephrotic syndrome (NS), acute nephritic syndrome (ANS), acute kidney injury (AKI) of unknown etiology, chronic kidney disease of unknown etiology (CKD), rapidly progressive renal failure, subnephrotic proteinuria with or without haematuria, isolated haematuria, systemic diseases with renal manifestation, and children with family history of kidney disease. The indications for biopsy in NS were steroid dependent, steroid resistant and frequently relapsing NS, atypical features such as age < 1 year and > 10 years at presentation, gross haematuria, hypocomplementemia, hypertension or renal impairment. The indications for biopsy in ANS were physician determined.

Renal disease was classified into the following groups: Primary glomerulonephritis (GN), secondary GN, hereditary disease, tubulointerstitial disease and CKD.

Mesangioproliferative GN (MesPGN) was diagnosed when there were > 4 cells per mesangial region in > 50% of glomeruli with negative IF staining² except those with post infectious etiology, in whom C3 and/or IgG staining was noted. Crescentic GN (CrGN) was defined as presence of crescents in > 50% of glomeruli. Diffuse mesangial hypercellularity (DMH) was defined as > 75% of peripheral mesangial areas showing hypercellularity in > 80% of glomeruli.³ IgM nephropathy (IgMN) was defined as diffuse, global mesangial IgM staining (> 2+) in non-sclerotic glomeruli. Post infectious glomerulonephritis was diagnosed when there was a history of antecedent skin or throat infection, low (or in remote cases, normal) C3, mesangial or ‘exudative’ focal/diffuse endocapillary proliferation, mesangial or capillary wall C3 with or without IgG, and subepithelial humps in those with EM reports. Adequate biopsy specimen was defined as presence of at least 5 glomeruli in the specimen on LM, similar to previously published studies.^4,5

NS was defined as edema, proteinuria > 40 mg/m²/hour on timed sample and hypoalbuminemia < 2.5 g/dL. ANS was defined as haematuria, oliguria, edema, hypertension (≥ 95^th centile for age) and GFR < 75 ml/min/1.73m² by Schwartz’s formula.⁶ Rapidly progressive renal failure was defined as rapid worsening of kidney function (GFR < 75 ml/min/1.73m² by Schwartz’s formula at presentation) over 1-12 weeks. Isolated haematuria was defined as microscopic or gross persistent haematuria without proteinuria and with normal renal function (GFR >75 ml/min/1.73m² by Schwartz’s formula). AKI was defined as rapid worsening of kidney function (GFR < 75 ml/min/1.73m² by Schwartz’s formula at presentation) over < 1 week. Subnephrotic proteinuria was defined as proteinuria < 40 mg/m²/hour on timed sample with or without haematuria. CKD was defined as GFR persistently < 75 ml/min/1.73m² for > 3 months.

The study was approved by the Institutional Review Board and Ethics Committee of our institution.

Statistical analysis

Data were analyzed using the Statistic Package for Social Sciences version 15(SPSS Inc. Chicago, USA). Continuous variables were expressed as mean or median with interquartile range, categorical variables were expressed as frequency and percentage. Clinical and biochemical features between various primary GNs, and between the minimal change disease (MCD) and non-MCD group in patients who presented with NS, were compared using Chi square test for categorical variables and Student’s t test for continuous variables. The frequency of glomerular diseases in the two successive decades of the study was compared using Chi square test. A p value < 0.05 was considered significant.

Results

A total of 1740 children underwent renal biopsy during the 20 year study period. Among them, 858(49.3%) biopsies were obtained in the first decade (1996-2005) and 882(50.7%) in the second decade (2006-2015). The mean age was 12.8 ± 4.9 years (range 8 months – 18 years) and the male: female ratio was 1.5:1. Children aged above 15 years contributed 41.8% (n=728) of the total number. EM was done in 116 patients (6.6%) overall and 25 out of 64 patients (39%) with isolated haematuria.

Indication for renal biopsy

The indications for renal biopsy are shown in Table 1. The commonest indications were NS (n = 1100, 63.2 %), ANS (n = 226, 13%) and systemic disease with renal manifestation (n = 130, 7.5%).

Table 1. Indications for renal biopsy in 1740 paediatric patients aged < 18 years.

Syndrome	n(%)
Nephrotic syndrome (NS)	1100(63.2)
Acute nephritic syndrome (ANS)	226(13)
Systemic disease with renal involvement (SD)	130 (7.5)
Chronic kidney disease (CKD)	72 (4.1)
Subnephrotic proteinuria (SNP)	66 (3.8)
Isolated Haematuria (IH)	64 (3.7)
Rapidly progressive renal failure (RPRF)	57 (3.2)
Acute kidney injury (AKI)	24(1.4)
Family history of CKD	1 (0.05)

Renal biopsy diagnoses: Overall

The histological diagnoses obtained following kidney biopsy are shown in Table 2. Overall, MCD (n = 369, 21.2%) and lupus nephritis (LN, n = 282, 16.2%) were the most common primary and secondary GNs respectively. Alport syndrome (n = 22, 1.2%) was the most common hereditary disease, and chronic glomerulonephritis (CGN, n = 31, 1.8%) the most common cause of CKD.

Table 2. Histological diagnoses of paediatric kidney biopsies in this study and comparison with other published paediatric registries.

Region	Asia	Asia	Asia	Africa	Europe
Country	India, present study	Pakistan	China	Egypt	Italy
Study period	1996-2015	1997-2013	1989-2012	1998-2012	1992-1994
Number of patients	1740	423	1579	1226	432
Age group	≤ 18 years	< 18 years	≤ 18 years	≤ 18 years	≤ 15 years
Mean age (years)	12.8± 4.9	10.4 ± 4.5	9.3± 3.2	9.2 ± 3.7	8.9 ± 3.7
PRIMARY GN	1321 (75.9)	346 (81.8)	949 (60.1)	826 (67.4)	289 (66.9)
MCD	369(21.2)	128 (30.3)	235 (14.8)	267 (21.8)	61 (14.1)
MesPGN	294(16.9)	20 (4.7)	181 (11.4)		41 (9.5)
Undifferentiated	256 14.7)
Infection related	38 (2.2)
FSGS	191(10.9)	109 (25.8)	51 (3.2)	129 (10.5)	36 (8.5)
EndPGN	172(9.8)	21 (4.9)	62 (3.9)	352 (28.7)
Infection related	113 (6.5)	20 (4.7)			9 (2)
Undifferentiated	57 (3.3)	1 (0.2)
C3GN	2 (0.1)
IgAN	113 (6.5)	9 (2.1)	199 (12.6)	10 (0.8)	81 (18.8)
MN	47 (2.7)	27 (6.4)	27 (1.7)	4 (3.2)	4 (0.9)
DMH	48 (2.7)
IgMN	40 (2.3)	2 (0.5)	77 (4.8)
MPGN	19 (1.1)	18 (4.3)	40 (2.5)	42 (3.4)	24 (5.5)
CrGN	27(1.5)	15 (3.5)	20 (1.2)	20 (1.6)	8 (1.9)
ANCA negative	19 (1.1)
ANCA positive	2 (0.1)		10 (0.6)
Infection related	5 (0.3)
Anti GBM disease	1(0.05)
Fibrillary GN	1(0.05)
SECONDARY GN	323 (18.5)	27 (6.4)	493 (31.2)	238 (19.4)	106 (24.5)
LN	282 (16.2)	20 (4.7)	91 (5.7)	209 (17)	22 (5)
HSP	16 (0.9)		273 (17.3)	3 (0.2)	50 (11.6)
HBVGN	15 (0.8)		113 (7.1)
HUS	9 (0.5)	5 (1.2)	6 (0.4)	16 (1.3)	3 (0.7)
Diabetic nephropathy	1 (0.05)
TUBULOINTERSTITIAL DISEASE	12(0.7)	21 (5)	16 (1.0)	47 (3.8)	14 (1.1)
AIN	8 (0.5)	9 (2.1)			5 (1.2)
ATN	4(0.2)	1 (0.2)			1 (0.2)
HEREDITARY DISEASE	27 (1.5)	7 (1.6)	131 (8.3)	32 (2.6)	52 (12)
Alport syndrome	22 (1.2)	1 (0.2)	60 (3.8)	8 (0.6)	17 (3.9)
TBMD	3 (0.2)		65 (4.1)		22 (5)
Fabry’s	1 (0.05)
Congenital NS	1 (0.05)	6 (1.4)		21 (1.7)	1 (0.2)
CKD	51(2.9)	17 (4.1)	26 (2.7)	52 (4.3)	11 (2.6)
ESKD	6 (0.3)	7 (1.7)		50 (4.1)	3 (0.7)
CIN	14 (0.8)	10 (2.4)	26 (2.7)		8(1.9)
CGN	31 (1.8)			2 (0.2)
OTHERS
HTN	3 (0.2)
CN	1 (0.05)

Diagnoses mentioned in bold italics correspond to major histological diagnoses, which have been further sub-classified. Figures in brackets refer to the frequency of each histological diagnosis expressed as percentage of total number of biopsies.

Primary GN – Primary glomerular disease; MCD-minimal change disease; MesPGN-mesangioproliferative glomerulonephritis; FSGS-focal segmental glomerulosclerosis; EndPGN – Endocapillary proliferative glomerulonephritis; C3GN – C3 glomerulopathy; IgAN – IgA nephropathy; MN – membranous nephropathy; DMH-diffuse mesangial hypercellularity; IgMN-IgM nephropathy; MPGN-membranoproliferativeglomerulonephritis; CrGN – Crescentic glomerulonephritis; ANCA – antineutrophil cytoplasmic antibody; Anti GBM disease – Anti-glomerular basement membrane disease; Secondary GN – secondary glomerular disease; LN – Lupus nephritis; HSP – Henoch Schonlein purpura; HBVGN – Hepatitis B virus related glomerulonephritis; HUS – Haemolytic uraemic syndrome; AIN – acute interstitial nephritis; ATN – acute tubular necrosis; TBMD – Thin basement membrane disease; Congenital NS – Congenital nephrotic syndrome; CKD – chronic kidney disease; ESKD – end stage kidney disease of unknown etiology; CIN – chronic interstitial nephritis; CGN – chronic glomerulonephritis; HTN – Hypertensive nephrosclerosis; CN – Cortical necrosis

Post infectious etiology accounted for 12.9% of patients with MesPGN and 65.7% of those with endoproliferative glomerulonephritis (EndPGN). In the EndPGN group, 65% had diffuse proliferation and the remaining had focal proliferation. In patients with LN, the frequency of various ISN RPS classes were as follows: Class II (n = 51, 18.1%), Class III (n = 27, 9.6%), Class IV (n = 180, 63.8%), Class V (n = 22, 7.4%) and Class VI (n = 2, 0.7%). The histological diagnoses in children with Hepatitis B virus related GN (HBV related GN) were membranous nephropathy (MN, n = 8, 53.3%), MesPGN (n = 3, 20%), focal segmental glomerulosclerosis (FSGS, n = 2, 13.3%), IgA nephropathy (IgAN, n = 1, 6.6%), and MCD (n = 1, 6.6%). Haemolytic uraemic syndrome (HUS) was most commonly atypical (n = 7, 77.8%), 2 cases being reported post bone marrow transplant.

Renal biopsy diagnoses by clinical syndrome

The frequency of various biopsy proven renal diseases in children presenting with different clinical syndromes is given in figure 1. Figure 2 shows the frequency of biopsy proven renal diseases in children with different clinical syndromes stratified by age at presentation. MCD was the most common cause of NS overall in all age groups, except in children aged > 12 years, in whom MesPGN was the most common etiology. In children with ANS, EndPGN was the most common histological diagnosis overall in all age groups, with post infectious GN being the commonest etiology. In children whose clinical presentation was suggestive of AKI of unknown etiology, HUS was the most common etiology overall, and upto 12 years of age. In older children, acute interstitial nephritis was the most common etiology. In children with rapidly progressive renal failure, CrGN was the most common etiology overall, except in children aged 9-12 years, in whom EndPGN was the most common etiology. In children with isolated haematuria, MCD was the most common diagnosis overall and in all age groups, while IgAN was the most common cause of subnephrotic proteinuria. In those biopsied for CKD, IgAN was the most common diagnosis, followed by EndPGN. LN was the most common diagnosis in children with systemic disease in all age groups.

Fig. 1.

Pattern of renal diseases in children presenting with various clinical syndromes. Abbreviations used for the histological diagnosis are reported in Table 2

Fig. 2.

Pattern of renal diseases in children presenting with various clinical syndromes, stratified by age. Abbreviations used for the histological diagnosis are reported in Table 2

Comparison between primary glomerular diseases

At the time of diagnosis, children with MCD were younger (p< 0.001), had lower prevalence of haematuria (p = 0.027) and hypertension (p<0.001), higher prevalence of generalized edema (p < 0.0001) and lower mean serum creatinine (p<0.0001) compared to non-MCD pathologies. Children with DMH and IgMN had a similar clinical presentation to children with MCD, except that children with IgMN were older (mean age 11.9 years compared to 8.7 years for children with MCD). Children with IgA nephropathy (IgAN) had the highest prevalence of hypertension and haematuria at presentation, with a mean serum creatinine of 2.1 ± 2.6 mg/dL.

Changing trend of histopathological lesions

The frequency of various primary and secondary GNs at our institution over a period of 20 years is presented in Figure 3. During this period, there was a significant decrease in frequency of MCD and DMH and increase in FSGS and LN. There was no change in the proportion of various classes of LN over the two decades. The frequency of EndPGN did not change during the study period.

Fig. 3.

Change in pattern of renal disease over two decades of the study; *p < 0.05, 1996 – 2005 vs. 2006 – 2015. Abbreviations for the histological diagnosis are reported in Table 2

Discussion

In our study cohort, NS was the most common indication for biopsy, followed by ANS. The most common primary GN was MCD, followed by MesPGN, while LN was the most common secondary GN. MCD, EndPGN, HUS, CrGN, MCD and LN were the most common causes of NS, ANS, AKI, rapidly progressive renal failure, isolated haematuria and systemic disease with renal manifestations, respectively. IgAN was the most common cause of subnephrotic proteinuria and CKD. Over the course of two decades, the frequency of MCD declined, that of FSGS and LN increased, and all other GNs, including EndPGN, showed no significant change.

Our cohort was older than that described in other paediatric registries.^7–11 This is likely because we are a tertiary centre, with a large referred patient population drawn from neighbouring states and countries.

Comparison with registry data from Europe, North America and Africa

As in our study, NS is the most common indication for renal biopsy in Africa,¹⁰ followed by ANS.⁹ This is in stark contrast to Europe, where subnephrotic proteinuria and isolated haematuria remain the most common indications, and where ANS is a very uncommon indication for renal biopsy.^7,11 The most common primary GN in Africa too, is MCD,^9,10 unlike Europe, where it is IgAN,^7,11 and the United States, where it is FSGS.¹² Both EndPGN and membranous nephropathy had a higher prevalence in our cohort than in Europe,^7,11 but not Africa.^9,13 The prevalence of idiopathic membranoproliferative GN (MPGN) was low compared to both Europe and Africa.^7,9,11,13 Idiopathic MPGN is now attributed to abnormalities in the alternate complement pathway, and genetic factors play an important role in children with this histological diagnosis.¹⁴

While LN was the most common diagnosis among secondary glomerular diseases in our cohort, HBV related GN is the most common secondary GN in Sub-Saharan Africa, due to high HBV endemicity,¹⁰ while in Europe, Henoch Schönlein Purpura (HSP) predominates.⁷ There is compelling evidence that LN has a higher population prevalence in Asian Indians compared to Caucasians.¹⁵ Also, while MPGN is the most common histology among children with HBV related GN in Africa,⁹ in our cohort, it was membranous nephropathy.

In patients with a chronic kidney disease histology, where no definite histological diagnosis could be made, 60.7% had a chronic glomerular disease, while in Europe chronic interstitial nephritis was more common.⁷

Among children with a clinical presentation of NS, MCD was the most common diagnosis overall, similar to reports from Europe,^7,11 Africa ^9,13 and North America.¹² The second most common cause of NS overall, and the most common etiology in children aged > 12 years, was MesPGN, whereas in Europe^7,11 and North America,¹² it is FSGS. The high proportion of MesPGN in children with NS is notable, as MesPGN is associated with steroid resistance, frequent relapses and a higher rate of progression to CKD.¹⁶

EndPGN was the most common etiology for ANS across all age groups, unlike European and North American cohorts, in whom IgAN is most common overall,^7,11 and in children < 10 years,¹² respectively. It is also important to note that LN was the second most common cause of ANS in children > 12 years of age in our cohort, since most children with an acute nephritic presentation in South Asia are presumed to have a post infectious GN, and often not biopsied. SLE in Asian Indians has a higher systemic involvement, greater disease activity and mortality, compared to Caucasians¹⁵ and this is also seen among Indian children with lupus nephritis.¹⁷

In children biopsied on suspicion of AKI, HUS was the most common etiology, as in Europe. The majority were atypical HUS, which is consistent with data from a large Indian paediatric HUS cohort in whom at least 58% had anti-Factor H autoantibody associated HUS.¹⁸ Apart from HUS, acute interstitial nephritis is also an important cause for AKI in South Asian children, especially those above 12 years, as has been described by other investigators.⁸ This is a by-product of infectious disease and the widespread use of herbal and other over-the-counter medications. Only children in whom no obvious cause for AKI could be found were subjected to biopsy, hence our description of AKI etiology may not correlate with those clinicians in this region commonly encounter in clinical practice, where acute tubular necrosis due to diarrhoeal illness, pneumonia, malaria, dengue and scrub typhus predominates.¹⁹

Unlike Europe,^7,11 where children with isolated haematuria most commonly have IgAN, the most common pathologies on biopsy in our study were MCD and MesPGN, which are very likely to be TBMD or Alport syndrome, since only 39% of these patients underwent an EM biopsy. Although there is much debate about whether children with isolated urinary abnormalities should undergo a kidney biopsy, in our cohort, LN also presented with isolated haematuria, especially in children aged > 12 years, thus suggesting that a more liberal kidney biopsy policy for isolated haematuria may be indicated in this region.

IgAN was the most common cause of subnephrotic proteinuria with or without haematuria, similar to European paediatric cohorts.^7,11 It was also the most common etiology diagnosed in children presenting with CKD, while in Europe, nephronophthisis and chronic interstitial nephritis were the most common etiologies.⁷

Comparison with registry data from West and East Asia

The most common primary GN in West²⁰ and East Asia^5,21 is MCD, while in Japan, where children are screened for urinary abnormalities at school entry, it is IgAN.²² The high prevalence of MesPGN, however, has been reported only from South Asia.²³ EndPGN was also more common in our cohort than in West²⁰ and East Asia,^5,21,24 whose countries belong to the middle and high income bracket. Around 31% of South Asian children are left with residual renal injury at discharge,²⁵ which is corroborated by the fact that EndPGN was also the second most common cause of CKD.

LN is the most common secondary GN in West Asia,²⁰ whereas in East Asia,^5,21 HSP predominates. The prevalence of HBV related GN was lower than a concurrent Chinese cohort,⁵ and may be due to the fact that India has lower HBV endemicity.²⁶

MCD is the most common cause of NS in West²⁰ and East Asia.^5,21 The most common cause of ANS in East Asia,⁵ remains EndPGN, while in West Asia, similar to Europe, it is HSP.²⁰

IgAN was the most common etiology identified in children with subnephrotic proteinuria with or without haematuria in both our cohort and East Asia,⁵ while it is LN in West Asia.²⁰ That IgAN was also the most common etiology for CKD is unique to South Asia. This highlights an important difference in the way IgAN presents in this region. In East Asian children, IgAN presents with ANS and macroscopic haematuria,⁵ or in countries like Japan and Taiwan where children undergo urine screening at school entry, as isolated haematuria or subnephrotic proteinuria.^22,27 In our cohort, among all primary GNs, IgAN had the highest prevalence of hypertension, haematuria and renal dysfunction at presentation. This is at least partly explained by poor access to health care and delayed diagnosis, though genetic or environmental factors that contribute to a more aggressive or ‘malignant’ form of IgAN cannot be ruled out. A similar picture is seen in South Asian adults with IgAN, more than half of whom have renal failure at diagnosis, and progress more rapidly to end stage kidney disease compared to other races.²⁸

Changing trends in the pattern of histopathological diseases

Over the two decades of this study, there was a decline in MCD and increase in FSGS although the indications for renal biopsy in children with NS did not change. While this could be due to a genuine decline in the incidence of MCD as has been reported in Europe⁷ and North America,²⁹ we cannot exclude a change in referral pattern, leading to fewer number of children with MCD being referred to our centre, as a cause for this trend. The unique finding was the increase in LN, since European registries have reported the opposite trend.⁷ This is a function of race: Asian girls and boys have the highest prevalence rate ratio for LN compared to all other races.³⁰ Since our biopsy policy pertaining to LN, and the relative proportions of various ISN RPS classes diagnosed on biopsy also remained unchanged during the study period, we believe this indicates a true increase in the incidence of LN, though population based studies to support this finding are lacking. While the frequency of EndPGN has declined dramatically in Europe⁷ and East Asia,^5,21,31 whose countries belong to the high/very high development categories, this trend has not been observed in our cohort. A conservative estimate of incidence of acute post streptococcal GN in children varies from 6 to 24.3 cases per 100,000 person years in more and less developed countries respectively.³² South Asia has witnessed large scale migration from rural to urban areas, and the cramped living spaces, air pollution, suboptimal water and sanitation facilities in this region, provide a fertile ground for infectious disease.

Comparison between MCD and non-MCD primary GNs

Although there were important differences in the clinical presentation of MCD and non-MCD primary GNs, as has also been demonstrated by other authors,⁴ the presentation of IgMN and DMH, considered variants of MCD, did not differ greatly (apart from children with IgMN being older), even though their steroid responsiveness and prognosis have been shown to be worse.^33,34 This highlights the need for a more aggressive biopsy policy in children with NS in South Asia, especially in those with atypical features.

Strengths and Limitations

Our study has certain limitations. C1q staining was not routinely performed on all biopsies and EM could be done only for a few patients due to financial constraints. C1q nephropathy, TBMD, Alport syndrome and the C3 glomerulopathies may have been underreported as a result. Routine screening for blood borne viruses was introduced only from 2006, hence it is likely that the proportion of secondary GNs may have been higher than we have reported.

Lastly, because of poor access to health care and late referral to a nephrologist, the type and tempo of disease encountered in this region may vary significantly from that encountered by clinicians in other regions. The strengths of this study are the large number of biopsies reported from a region with limited access to nephrology care and high burden of kidney disease, and the consistent, high quality pathology services available over the study period. Knowledge of the pattern of kidney disease that might be encountered in children with varied clinical presentations would be of great benefit to practicing nephrologists in this region.

Conclusion

In South Asia, MCD remains the most common histopathological diagnosis in children with NS. EndPGN, predominantly infection related, is the commonest cause of ANS and the second most common cause of CKD. LN is the commonest secondary GN and its increasing frequency as a biopsy diagnosis may point to an increase in incidence, which needs to be confirmed by population based studies. Chronic glomerulonephritis is the most common cause of biopsy proven CKD, and IgAN the most common etiology identified.