Article Text

Download PDFPDF

The prevalence and clinicopathological profile of IgM nephropathy in children with steroid-resistant nephrotic syndrome at a single centre in Pakistan
  1. Shaheera Shakeel1,
  2. Muhammed Mubarak1,
  3. Javed I Kazi1,
  4. Ali Lanewala2
  1. 1Department of Histopathology, Sindh Institute of Urology and Transplantation (SIUT), Karachi, Sindh, Pakistan
  2. 2Department of Pediatric Nephrology, Sindh Institute of Urology and Transplantation (SIUT), Karachi, Sindh, Pakistan
  1. Correspondence to Dr Muhammed Mubarak, Department of Histopathology, Sindh Institute of Urology and Transplantation, Karachi 74200, Pakistan; drmubaraksiut{at}yahoo.com

Abstract

Background There is little information on the clinicopathological characteristics of IgM nephropathy (IgMN) in paediatric steroid-resistant nephrotic syndrome (SRNS) and its response to calcineurin inhibitors (CNI).

Material and Methods This study was conducted at Sindh Institute of Urology and Transplantation, from January 2009 to August 2011. All SRNS children who received renal biopsies were included. Relevant data were compared among minimal change disease (MCD) and IgMN. The response to CNI was analysed in detail in IgMN by groups (group 1: complete or partial remission; group 2: no response).

Results The frequency of IgMN in 147 children with SRNS was 13.6%. Compared with MCD, there was a male preponderance in IgMN. Blood urea and serum creatinine both at presentation and at last follow-up were significantly higher in IgMN. Regarding subgroups of IgMN, systolic blood pressure (SBP), blood urea and serum creatinine were significantly higher in group 2 at presentation, while at last follow-up, SBP, diastolic blood pressure and proteinuria were higher in group 2. The prevalence and degree of mesangial proliferation, global glomerulosclerosis, interstitial fibrosis and tubular atrophy were significantly higher in group 2.

Conclusions IgMN is a common cause of paediatric SRNS and is significantly different from MCD. There is also a significant difference in clinical and laboratory parameters among responders and non-responders to CNI in IgMN.

  • Histopathology
  • Immunofluorescence
  • Nephrology
  • Paediatric Pathology

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

Steroid-resistant nephrotic syndrome (SRNS) is a common problem in paediatric nephrology practice and is a difficult therapeutic challenges for the paediatric nephrologist.1 The frequency of SRNS varies throughout the world ranging from 10% to 20%.1 We have earlier found steroid-resistant pattern in 30% of our children with idiopathic nephrotic syndrome (INS), who underwent renal biopsy.2 There is a relative paucity of published data on the histopathological spectrum of the glomerulopathies underlying SRNS in children.1 ,3–6 Few published studies on the subject show that there is a wide range of histopathological lesions underlying SRNS, varying according to age, regional and racial factors.6 ,7

IgM nephropathy (IgMN) is a relatively common, and still debatable entity in the list of glomerulopathies, presenting with INS, isolated haematuria or combined haematuria and asymptomatic proteinuria in both adults and children.8 Its extreme forms resemble two previously well described entities of minimal change disease (MCD) and focal segmental glomerulosclerosis (FSGS). Some authors have suggested that these lesions represent the spectrum of the same disease, with MCD at one end, IgMN in the middle and FSGS at the other end.9 Other investigators are of the view that these diseases including IgMN are distinct from each other.8 ,9 Later authors have demonstrated increased steroid resistance and less favourable outcomes in IgMN compared to MCD, which it most closely resembles histologically.8 ,9

We have earlier compared a subset of IgMN and MCD patients for a number of demographic, clinical and outcome parameters with a follow-up period of at least 6 months. It was apparent from our results that the two diseases differ significantly in some aspects, especially the long-term renal outcome.8 The previous analysis was, however, a retrospective exercise. There are only a few long-term longitudinal studies of IgMN with an analysis of responses to calcineurin inhibitors (CNI).7 ,10–12 No such study has been reported from this region of the world to date.

In this study, we aimed to determine the frequency of IgMN in children presenting with SRNS at our centre and to compare the demographic, clinical and laboratory parameters at the time of presentation and at last follow-up among MCD and IgMN patients in a prospective manner, and to compare the above parameters as well as the immunopathological features and outcome in terms of initial clinical response to CNI therapy among the subgroups of IgMN patients.

Materials and methods

This prospective and observational study was conducted from January 2009 to August 2011 at the Departments of Histopathology and Paediatric Nephrology, Sindh Institute of Urology and Transplantation, Pakistan. All children (≤18 years) presenting with SRNS at the paediatric nephrology outpatient clinic at the Sindh Institute of Urology and Transplantation and in whom renal biopsies were performed during the above-mentioned period were included. Their demographic, clinical and laboratory data at the time of presentation and on last follow-up, treatment given after biopsy and outcome were retrieved. The light microscopy and immunofluorescence findings and the final pathological diagnoses were recorded from the original biopsy forms. The detailed methodology of these procedures has been described in our previous studies.1 ,2 Clinical and laboratory details were recorded from case files of the patients. MCD was defined as normal findings on light microscopy, negative or trace positivity of IgM on immunofluorescence, and diffuse effacement of foot processes on electron microscopy.1 IgMN was defined on the basis of diffuse, granular, mesangial positivity (≥2+) of IgM as the sole or predominant immunoglobulin on immunofluorescence microscopy irrespective of the light microscopy findings.8 Written informed consent was taken from the parents before obtaining real-time ultrasound-guided percutaneous renal biopsies by automated biopsy gun in all cases.

According to the treatment protocol followed at our institute, all children presenting with INS are given corticosteroids at a dosage of 60 mg/m2 per day until they achieve remission. Patients unresponsive to prednisolone therapy for four consecutive weeks are considered resistant, and are kept on the same dose of prednisolone therapy on every alternative day for an additional 4 weeks and renal biopsy is performed. Once the pathological diagnosis is made on biopsy, the glomerular filtration rate (GFR) is calculated using the Schwartz formula (Height in cm/serum creatinine×constant (0.5 for children)) before starting CNI. All children with MCD or IgMN are then treated with one of the following CNI drugs along with low-dose steroids: (1) ciclosporin A (CsA) in a dose of 5 mg/kg per day for 12 weeks (if GFR is ≥100 ml/min); (2) tacrolimus in a dose of 0.1 mg/kg per day for 12 weeks (if GFR is <100 ml/min).13 The initial response to these therapies was assessed after an interval of 12 weeks. The above-mentioned protocol and the number of children following each path are shown in a schematic flow diagram (figure 1). If patients achieved complete remission (CR), steroids were slowly tapered off and the lowest dose of CsA or tacrolimus was continued for the long term. In patients who achieved partial remission (PR), in addition to tapering off the steroids and lowering the dosage of CsA, an ACE inhibitor was added at a dose of 10 mg in two divided doses. The patients not responding to CsA were switched to tacrolimus in the above-mentioned dose, until they achieved PR or CR. The following definitions of CR and PR were used: CR was defined as negative or trace proteinuria by dipstick with normalisation of serum albumin and no oedema; PR was defined as significant improvement in proteinuria (at least 50% drop) without complete normalisation of serum albumin and oedema.12

Figure 1

Schematic diagram showing multiple stages of study protocol with the number of patients and the rationale of decision making. ACEi, ACE inhibitor; CR, complete response; CsA, ciclosporin A; GFR, glomerular infiltration rate; IgMN, IgM nephropathy; NR, no response; PR, partial response; SRNS, steroid-resistant nephrotic syndrome; TAC, tacrolimus.

Children with IgMN were further subdivided into two groups depending on their initial responses to CNI agents. Group 1 consisted of 15 (75%) children who responded to CNI therapy, either with CR or PR. Group 2 contained five (25%) patients who did not respond to the CNI treatment at 12 weeks.

The main demographic, clinical and laboratory features were compared among MCD and IgMN cohorts and among the two subgroups of IgMN. Treatment responses to CNI agents were also compared among the two subgroups of IgMN children after 3 months of treatment and correlated with immunopathological features on renal biopsies.

Statistical analysis

Statistical analysis was carried out using IBM-compatible SPSS for Windows V.10.0. The median (IQR) values were used for descriptive variables such as age, clinical and laboratory features. Numbers (percentages) were used for categorical data, such as response to CNI drugs and prevalence of immunopathological variables. Mann–Whitney, Fisher's exact and χ2 tests were applied as appropriate, and a p value less than 0.05 was considered significant.

Results

One hundred and forty-seven children with SRNS who received renal biopsies before the initiation of adjuvant immunosuppressants were included in this study. Their demographic and clinicopathological parameters have already been described in detail in our previous study on the spectrum of histopathological lesions underlying SRNS in children.1 The present prospective study describes in detail the demographic, clinical, laboratory and histopathological features and the response to initial CNI treatment of one of the important lesions underlying SRNS, ie, IgMN, as well as a comparison of some of the above parameters of this disease with MCD.

The frequency of IgMN in the current series was 13.6%, and it constituted the third most common cause of SRNS in children in the series. The main demographic, clinical, laboratory and pathological features of IgMN cases in comparison with MCD are shown in table 1. The median age (IQE) in the MCD group was 4.0 years (2–6.1) and in the IgMN group 4.5 years (3–8.0), and proved to be statistically insignificant (p=0.413). Both groups showed male predominance, but boys were more common than girls in the IgMN cohort (p=0.244). Among the laboratory parameters at the time of presentation, blood urea in mg/dl (16.50 (14.2–18.0) vs 12.0 (9.0–18.0); p=0.015) and serum creatinine in mg/dl (0.62 (0.32–0.68) vs 0.42 (0.33–0.49); p=0.037) were significantly higher in IgMN children than in the MCD group. Among the various parameters analysed at the time of last follow-up, only serum creatinine, in mg/dl, was significantly higher in the IgMN group than in the MCD group (0.62 (0.46–0.81) vs 0.39 (0.33–0.52); p<0.001).

Table 1

Demographic, clinical, and laboratory characteristics of IgMN and MCD groups

As far as treatment outcome is concerned, CsA in addition to low-dose steroids on alternative days was given to all 34 patients in the MCD group, and 26 of 34 (76.4%) patients achieved CR, six of 34 (17.6%) patients achieved PR, while two of 34 (5.8%) patients showed no response (NR) to treatment. In six patients who were in PR, an ACE inhibitor was added and after 3 months, two of six achieved CR and the remaining four of six patients were still in PR (figure 2).

Figure 2

Comparison of clinical responses to first-line newer immunosuppressive agents (either ciclosporin A or tacrolimus) among minimal change disease (MCD) and IgM nephropathy (IgMN) children. All figures are in percentages.

In the IgMN group, CsA was given in addition to alternative day steroid therapy in 17 of 20 (85%) patients. Of these, seven (41.17%) patients achieved CR and seven (41.17%) were in PR, while three (17.64%) patients showed NR to therapy. In seven patients, who were in PR, an ACE inhibitor was added and after 3 months, four of seven achieved CR and the remaining three of seven patients were still in PR. The three patients who did not respond to CsA were switched to tacrolimus and after 3 months all three achieved PR. Only three of 20 (15%) patients were initially treated with tacrolimus as a first-line therapy in addition to steroids, and only one of them achieved PR, while the remaining two showed NR to tacrolimus. In these two patients, an ACE inhibitor was added and both of them achieved PR (figure 1).

Regarding comparison among the subgroups of IgMN (table 2), the median age of group 1 was 5 years (3.5–10) and of group 2 it was 9 years (7.5–12), and there was no significant difference between the two groups. In group 1, there were 11 boys (73.3%) and four girls (26.6%). The male to female ratio was 3:1. In group 2, there were four boys (80%) and only one girl (20%). The male to female ratio was 4:1. There was no significant gender difference between the two groups. Our results showed that among seven parameters recorded at the time of presentation, systolic blood pressure (SBP) in mmHg (median (IQR) 138 (124–140) vs 112.0 (110–130; p=0.019), blood urea in mg/dl (34.0 (26–42) vs 18.0 (16–20; p=0.004) and serum creatinine in mg/dl (0.91±0.29 vs 0.47±0.19, 0.80 (0.70–1.19) vs 0.42 (0.33–0.62); p=0.002) were significantly higher in IgMN group 2. However, among parameters compared at last follow-up, SBP in mm Hg (128.4±3.85 vs 118.0±7.71, 130.0 (125–131) vs 120.0 (112–120); p=0.004), diastolic blood pressure in mm Hg (86.0±7.71 vs 76.27±6.41, 88.0 (78.5–92.5) vs 74.0 (70–80); p=0.019) and proteinuria (p=0.001) were significantly higher in IgMN group 2.

Table 2

Demographic, clinical, and laboratory characteristics of IgMN subgroups

As far as the comparison of morphological features is concerned, we compared glomerular changes, tubulointerstitial changes and the intensity of IgM on immunofluorescence among the two subgroups (table 3). Among glomerular changes, the prevalence and degree of mesangial proliferation and global glomerulosclerosis were significantly more common in group 2 compared with group 1 (p=0.02 and 0.006, respectively). No vasculopathy was observed in either of the groups. In the tubulointerstitial compartment, the prevalence and degree of interstitial fibrosis and tubular atrophy were also significantly higher in group 2 (p=0.002).

Table 3

Comparison of some immunopathological variables in subgroups of IgMN patients

Similar to the morphological features, the intensity of IgM positivity on immunofluorescence was also higher in group 2 but was non-significant (p=0.38). There was no significant difference in the positivity of C3 and C1q immunoreactants among the two subgroups.

Discussion

There is a relative paucity of data on the prevalence and clinicopathological characteristics of IgMN in one of the important clinical subtypes of INS, ie, SRNS in children. There are also very few studies on the long-term evolution and the responses to CNI drugs in this disease.14 To our knowledge, this is first study on the detailed histopathological and clinical characterisation of IgMN patients in SRNS children and their response to CNI drugs in the region and the world. However, we acknowledge that there are a number of limitations in the study. It is a single centre study, the number of children with IgMN is small and the follow-up is relatively short. We plan to continue active follow-up of this cohort and report the long-term outcome of these children in future.

IgMN is an ill-understood and still largely controversial glomerulopathy that is a fairly common cause of INS among all age groups and both genders.1 ,8 ,15 Since its first description, the distinctness of this entity has been the subject of heated debate.16–18 We have earlier shown in a retrospective study that MCD and IgMN are distinct entities at least in some clinical parameters.8 In the present prospective work, we aimed to verify the findings of our earlier observations.

There is a wide variation in the reported frequency of IgMN among the renal biopsy series throughout the world.1 The exact cause of this wide variation in the prevalence of IgMN reported from different parts of the world is not well understood, but varying criteria of biopsy indications, varying diagnostic criteria used and racial and genetic factors might be responsible.8 ,9 We have earlier reported a frequency of IgMN of 18.5% in children with INS in the largest published series on IgMN in the world to date.8 A very high prevalence of 41.8% of IgM positivity was observed in a study on children with steroid-dependent nephrotic syndrome or SRNS at the Texas Childrens Hospital, USA.14 It is of note that the authors of the above study did not mention the minimum threshold of IgM positivity used for the diagnosis of IgMN. Similarly, many other studies on IgMN in the past have included cases with trace or minimal positivity of IgM on immunofluorescence.19 In the present study, we have found the frequency of IgMN to be 13.6% in children presenting with SRNS, and in this subtype of INS it constituted the third most common cause of nephrotic syndrome. The markedly low rate in our study may partly be due to the higher threshold of IgM positivity (≥2+) used in our study.

There is evidence that deposits of IgM in the mesangium may be found in cases of MCD, FSGS and mesangial proliferative glomerulonephritis. The clinical significance of mesangial deposits of IgM in these diseases is controversial.19 Part of the controversy stems from the subjectivity of the interpretation of immunofluorescence results and the variability in the criteria used for the diagnosis of the above diseases as well as IgMN. However, the majority of the studies have concluded that the presence of IgM deposits in the mesangium is a marker of disease severity irrespective of the morphological appearance of the underlying lesion.14

In this study, IgMN was compared with MCD with respect to a variety of demographic, clinical and laboratory parameters in children presenting with SRNS at our centre. Our results indicate that age is not significantly different among the two groups. Although a male predominance was noted in both groups, boys showed a marked but non-significant predominance in the IgMN group (p=0.24). Blood urea and serum creatinine at presentation were significantly higher in the IgMN group. On the other hand, blood pressure, total proteins and serum albumin showed no significant difference between the groups. These results are different from those reported in the study by Al Eisa et al,18 in which hypertension was significantly greater in the IgMN cohort. We have reported earlier that haematuria and progression to renal failure were significantly different among the two groups. This is in concordance with the present study results because serum creatinine at the last follow-up is significantly different in the two groups. Although no patient progressed to end-stage renal disease in this series, it might be due to the short follow-up duration in this study.

In this study, we have compared the outcome of MCD and IgMN in terms of response to CNI drugs, which included CsA and tacrolimus, given in accordance with the calculated GFR. Our results revealed that overall 94% of patients in the MCD group and 75% in the IgMN group showed response to CNI drugs. NR was observed in 5.8% of patients in the MCD group and 25% of patients in the IgMN group. To our knowledge, there is a paucity of published data on the comparison of these two entities with respect to the response to CNI. Our results indicate that in the MCD group response to CNI was much better than the IgMN group, and these results are different from others reported in the literature. Al Eisa et al18 found no significant difference between MCD and IgMN in response to CNI. Gregory et al11 reported a better response to CsA in the IgMN group compared with the MCD group.

There is little information in the literature on the response to CNI among patients with IgMN, especially in the SRNS subtype of nephrotic syndrome.19–24 Our results indicate that overall 75% of patients in the IgMN group responded to CNI therapy at 3 months and these results are almost similar to those reported from Jordan by Hamed,12 who reported a response rate of 70% in SRNS children with IgMN. Swartz et al14 found an 88% rate of CR or PR in those children treated initially with CsA.

We also compared the clinical and pathological parameters among subgroups of IgMN with respect to the initial response to CNI agents, and found that there was no age and gender predilection among responders and non-responders. However, blood pressure, blood urea and serum creatinine were significantly higher in non-responders, and persistent proteinuria was also found in all non-responders to CNI. These results are in concordance with those reported by Hamed,12 who found a higher prevalence of hypertension, persistent proteinuria and progression to renal failure in non-responders to CsA among IgMN patients.

There is no information in the literature on the correlation of initial clinical response to CNI and the immunopathological features on pre-CNI therapy biopsy. A few studies have analysed the response to CsA in relation to the histological category of the nephrotic syndrome, but those studies have not analysed in detail the extent of glomerular, tubular and interstitial changes in relation to the response to therapy.10 Our comparative analysis of morphological and immunohistochemical features among responders and non-responders of the IgMN cohort shows that the degree of glomerular mesangial proliferation, global glomerulosclerosis, tubular atrophy and interstitial fibrosis are significantly higher in the non-responder group. These observations are similar to those observed by Smoyer et al,10 who reported that more severe histological lesions on biopsy correlate with poor response to CsA. Segmental glomerulosclerosis was also more prevalent in non-responders but it did not achieve statistical significance. This is the most controversial lesion among the morphological spectrum of IgMN lesions. Many authors have excluded this lesion in their analysis of IgMN, while others have included this lesion in the category of IgMN when immunofluorescence showed diffuse and global positivity of IgM. We have also included cases of FSGS showing diffuse and generalised positivity of IgM (≥2+) in the category of IgMN, as in our previous study.8 Further studies are needed to clarify the status of this lesion in IgMN. In our results, the intensity of IgM on immunofluorescence was also higher in non-responders to CsA.

In conclusion, our data support the view that IgMN should be considered a distinct entity. Moreover, there is a significant difference in histological, clinical and laboratory parameters among responders and non-responders to CNI drugs in IgMN, and these parameters may be helpful in predicting the long-term outcome of IgMN patients.

Take-home message

  • IgM nephropathy is not such an uncommon disease. It merits the attention of researchers throughout the world.

References

Footnotes

  • Competing interests None.

  • Ethics approval Ethics approval was obtained from the CBEC, Sindh Institute of Urology and Transplantation.

  • Patient consent Obtained.

  • Provenance and peer review Not commissioned; externally peer reviewed.