a quantitative analysis of the mesangium in children with iga nephropathy: sequential study

JOURNAL OF PATHOLOGY, VOL. 161: 57-64 ( 1 990)

A QUANTITATIVE ANALYSIS OF THE MESANGIUM IN CHILDREN WITH IgA NEPHROPATHY: SEQUENTIAL STUDY

JUNK0 SUZUKI, NORISHIGE YOSHIKAWA AND HAJIME NAKAMURA

Department of Pediatrics, Kobe University Hospital, 7-5-1, Kunsunoki-rho, Chuo-ku, Kobe. 650, Japan

Received 1% Januarv 1989 Accepted 4 Januarj. 1990

SUMMARY

Quantitative analysis of the mesangial matrix and cells was performed on serial renal biopsies from 41 children with IgA nephropathy. In the repeat renal biopsy, nine patients showed a significant increase of mesangial matrix, 29 showed no change and in three there was a significant decrease. Eight of the nine patients (89 per cent) with a matrix increase had persistent proteinuria at the second biopsy. whereas only 14 of the 32 (44 per cent) without a matrix increase had persistent proteinuria (P < 0.05). Although the mesangial matrix increased in patients with persistent proteinuria, there was no decrease in patients with clinical remission. In contrast to the mesangial matrix, mesangial cells significantly decreased in 23 patients, did not change in 16, and significantly increased in only two in the second biopsy. These findings suggest that mesangial matrix increase is usually an irreversible change and that persistent proteinuria is associated with matrix increase with worsening in glomerular morphology and clinical outcome. This study indicates the importance of serial renal biopsy in children with IgA nephropathy with persistent proteinuria.

KEY WORDS-IgA nephropathy, repeat biopsy, mesangial matrix and cells. -

INTRODUCTION

Although IgA nephropathy has now become widely accepted as a distinct clinicopathologic

many reports have revealed variable clinical features and outcome.&* A considerable number of patients show a slow progression to renal failure. The precise mechanism responsible for this progression remains unknown.

On light microscopy, IgA nephropathy is charac- terized by global or segmental widening of the glomerular mesangial region produced by a variable combination of hypercellularity and increase in matrix. Our previous reportg suggested that predominant mesangial hypercellularity was charac- teristic of the early lesion of childhood IgA nephropathy, and that progression of the disease led to a gradual decrease of mesangial cellulanty and an increase of matrix with sclerosis. These findings prompted us to perform a quantitative analysis of

Addressee for correspondence: Dr Norishige Yoshikawa. Department of Pediatrics, Kobe University H O S ~ I I ~ . 7-5-1, Kusunoki-cho, Chuo-ku. Kobe. 650, Japan.

the mesangial matrix and cells on serial renal biopsies.

The purpose of this study was to examine the changes in the mesangial matrix and cells in each child who had had repeat renal biopsy and to eluci- date the relationship between the mesangial changes and clinical outcome.

MATERIALS AND METHODS

All children with IgA nephropathy who had had a repeat renal biopsy at Kobe University Hospital during the years 1981-1986 were reviewed. Initial renal biopsy specimens were obtained from children with persistent proteinuria with or without haema- tuna. During this period all patients with IgA nephropathy. with or without urinary abnormalities, were aksed to have a repeat renal biopsy 3 years from the first biopsy. Repeat biopsies were performed in 70 of the 154 patients who were biopsied before 1983.

Percutaneous renal biopsy was performed as previously described.lO." Biopsy specimens for light

0022-341 7/90/050057-08 $05.00 Q 1990 by John Wiley & Sons, Ltd.

58 J. SUZUKI ET AL.

microscopy were fixed in phosphate-buffered 10 per cent formalin, embedded in paraffin, sectioned at 2- 4 pm thickness, and stained with haematoxylin and eosin (H&E), periodic acid Schiff (PAS) and silver methenamine (PASM). Tissue for immunofluorescence was snap-frozen in dry ice and acetone, cut at 4 pm, and stained with fluorescein-tagged com- mercial antisera to human IgG, IgA, IgM, C4, C3, and fibrinogen (Hoechst, Darmstadt, F.R.G.).

The diagnosis of IgA nephropathy was based on the presence of IgA as the sole or predominant immunoglobulin in the glomerular mesangial region. Patients with systemic diseases such as systemic lupus erythematosus and Henoch-Schonlein purpura were excluded. Forty-one patients, 22 boys and I9 girls, were available for quantitative analysis of mesangial matrix and cells. The mean age at onset or discovery of nephropathy was 9.1 f 2.6 years. The mean ages at first and second renal biopsies were 10.1 & 2.9 and 13-0 & 3.1 years, respectively.

Twenty children with benign haematuria were examined as a control group. They had haematuria but light microscopy, immunofluorescence, and electron microscopy showed little or no glomerular changes. The age at biopsy ranged from 2 to 16 years, with a mean age of 9.7 k4.5 years.

Fig. I--Li&t of a glomemlus

Estimation of mesangial matrix H&E stained sections were examined with an

IBAS image analysis system (Zeiss). The method involves six procedures:

Figure 1 shows a H&E-stained glomerulus of IgA nephropathy on light microscopy. The microscopic image of the glomerulus of IgA nephropathy was put into the IBAS system through the video camera. The outer margin of the glomerular tuft was identified and encircled manually (Fig. 2). The image outside the encircled area was erased, and the area (glomerular area: C) and maximum diameter of the remaining area were measured automatically by the IBAS system. By manual operation, all nuclei of the glomerulus including mesangial, endothelial, and epithelial nuclei were identified according to density differences (Fig. 3). The mesangial, endothelial, and epithelial nuclei which were identified in procedure (3) and mesangial matrix were picked up by the same manual operation, taking care not to overestimate or underestimate the mesang*al matrix (Fig. 4).

Fig. 2-Image of a glomerulus by the IBAS analysis system. The image outside the glomerular tuft is erased

MESANGIAL MATRIX AND CELLS IN IgA NEPHROPATHY 59

Fig. %-Image of the glomerular nuclei of Fig. 1 by the IBAS image analysis system

Fig. &Image of the glomerular nuclei and mesangjum of Fig. 1 by the IBAS image analysis system

(5) The mesangial matrix area ( M ) was obtained as a subtraction image (Fig. 4- Fig. 3) and was measured automatically (Fig. 5).

( 6 ) The quantitative index of the mesangial matrix was expressed as the mesangial matrix area divided by the glomerular area (M/G per cent). The mean value and standard deviation were calculated for each specimen.

In each specimen. more than five glomeruli which had a diameter at least half of the largest glomerulus were examined. We excluded the glomeruli containing segmental sclerotic lesions, crescents, capsular adhesions. and crush artefact. The number of glomeruli available per specimen was 1 1.4 & 5.2.

PAS-stained sections were not used because there was almost no difference between the density of nuclei and mesangial matrix and distinction was dif- ficult by the IBAS system. PASM-stained sections were not used because the density of nuclei was not different from that of the basement membrane.

Measurements of mesangial cells PAS-stained sections were used to count

meSangia1 Cells Since the Various Cell types Were best identified in these sections.

~ i ~ , +Image of the mesankal matrix of F , ~ , 1 by the IBAS image analysis system

60 J. SUZUKI ETAL.

Table I-Duplicate measurements on three biopsies

Mesangial matrix (YO) Mesangial cells (matrix area/glomerular area) (no. of ceIis/104p2 glomerular area)

Biopsy First Second First Second

1 21.7 f 6.1 225 f 4.8 35.2f7-9 32.3 f 6.5 2 34.9 f 8.2 32-0 f 3.5 34.6 f 5.5 32.9 f 6.7 3 57.6 f 9.7 54.5 & 5.8 23.3 f6 .0 204 f 5.1 Mean 38.0 36.3 31.0 28.6

The sections were examined with a Zeiss photo- microscope at x 500 magnification and the number of mesangial cell nuclei in a glomerulus (N) was counted by manual operation. Mesangial cells were identified by the fact that their nuclei were surrounded by PAS-stained matrix fibrils. The microscopic image of the section was put into the IBAS system through the video camera. The outer margin of glomerular tufts was identified and encircled by manual operation. The image outside the encircled area was erased, and the total area (G) and maximum diameter of the remaining glomerular area were measured automatically. The quantitative index of the mesangial cells was expressed as the number of nuclei per G (N/G 104pm2). The mean value and standard deviation were calculated for each specimen.

In each specimen, more than five glomeruli which had a diameter at least half of the largest glomerulus were examined. We excluded glomeruli containing segmental sclerotic lesion, crescents, capsular adhesions, and crush artefact. The number of glomeruli available in the biopsies was 9.5 & 2-6.

To assess the reproducibility of this method, the same observer re-examined three cases which she had analysed 6 months previously. The three biopsy specimens contained 28, 27, and 13 glomeruli, respectively. The results are shown in Table I. The two sets of measurements showed close agreement in all three cases, indicating that the method has an acceptable level of reproducibility. Reproducibility was statistically examined using the coefficient of variation. Five glomeruli were measured ten times. Each value of the coefficient of variation was within 6 per cent (3.4-59 per cent).

Statistical analyses were performed using Student's z-test for comparison of the means and Fisher's exact test for association of the categorical variables.

RESULTS

Benign haematuria Figure 6 shows the correlation between mesangial

matrix and age at biopsy. There was a significant increase of the mesangial matrix with age. No correlation was found between mesangial cellularity and age (Fig. 7).

30 -I

0 1 0 5 10 15

age at biopsy (years)

Fig. bMesangial matrix and age at biopsy in children with benign haematuria. Mesangial matrix (J) increased with age (x) (v= 17.1 +0,6x). r=0.83 , P<0401

IgA nephropathj Mesangial matrix-In the repeat renal biopsy,

nine patients showed a significant increase of mesangial matrix, 29 no significant change, and three showed a significant decrease.

MESANGIAL MATRIX AND CELLS IN I@ NEPHROPATHY 61

m

m m m m

at onset, age at the first and second biopsies, incidence of macroscopic haematuria, degree of proteinuria at initial presentation, and the persistence of proteinuria at the time of the repeat biopsy. Four of the 23 patients with a decrease of mesangial cells and 5 of the 18 patients without a decrease showed a significant increase of the mesangial matrix.

Table V shows that the decrease in mesangial cellularity was independent of any change in the matrix.

0 5 10 15 age at biopsy (years)

Fig. 7-Mesangial cellularity and age at biopsy in children with benign haematuria. No correlation was found. r =0.18

For further analysis, the patients were divided into two groups on the basis of mesangial matrix change in the repeat biopsy: nine patients with a significant increase and 32 without an increase (Table 11). The mean mesangial matrix area in the initial biopsy was similar in both groups. These groups were comparable with respect to sex, age at onset, age at the first and second biopsies, incidence of macroscopic haematuria, and degree of proteinuria at initial presentation. Eight of the nine patients with a matrix increase had proteinuria at the time of the second biopsy, whereas only 14 of the 32 without a matrix increase had proteinuria at this time (P < 0.05).

Table I11 shows the association between persistent proteinuria and mesangial matrix change. Mesangial matrix was significantly increased in the 22 patients with persistent proteinuria at the time of the second biopsy (P<O.O5). However, the matrix did not change in the 19 patients without proteinuria at this time. These two groups of patients were comparable with respect to age at onset and age at the first and second biopsies.

All children who had lost their proteinuria by the time of the second biopsy were free from haematuria and showed complete clinical remission.

Mesangial cells-In the repeat renal biopsy, 23 patients showed a significant decrease of mesangial cells; in 16 there was no significant change; and only two had a significant increase.

For the purpose of analysis, the patients were divided into two groups (Table IV): 23 patients with a significant decrease of mesangial cells and 18 patients with no decrease. In the two group, there was no significant difference with respect to sex, age

DISCUSSION

Point count methods have been used in the uan- titative examination of mesangial matrix.''.' We used the IBAS analysis system (Zeiss) which analysed the image of specimens and evaluated the area of mesangial matrix by computer. Our method of evaluating the matrix area by the IBAS system is less laborious and it is possible to get a more precise evaluation than by point counting methods. Only with the H&E-stained sections could the IBAS system differentiate the mesangial matrix from the other regions of the glomerulus, but we used PAS- stained sections to count mesangial cells since the various cell types were best identified in these sections.

IgA nephropathy has now become widely accepted as a distinct clinicopathological entity.'-3 IgA nephropathy is the most common primary glomerulonephritis in children in our renal unit, detected in 25 per cent of biopsy spe~imens.''.'~ Although IgA nephropathy was initially considered as a benign disease with favourable outcome,4-6 recent long-term studies revealed patients who pro- gressed to renal In adult patients, increasing mesangial matrix with duration of the disease has been noted.'.15 However, there have been few reports in children. In Japan, all school children are screened every year for urinary abnormalities and the majority of our patients with IgA nephropathy were diagnosed early.".160urprevious report' suggested that the mesangial matrix increases and mesangial cellularity decreases with the duration of the disease in children. However, in that report, quantitative analysis was not performed.

Most of the patients showed an increase or no significant change of the mesangial matrix and only three patients showed a decrease in the follow-up biopsy. The mesangial matrix rarely decreased even in patients who had lost their proteinuria. These observations suggest that matrix increase is usually

62 J. SUZUKI ET AL.

Table II-Clinical findings in children with and without a mesangial matrix increase

Mesangial matrix at second biopsy Increase No increase (n = 9) (n = 32)

Mesangial matrix/glomerular area (%) First biopsy Second biopsy

Male/female Age at onset (years)

Age at first biopsy (years)

Age at second biopsy (years)

Macroscopic haematuria Proteinuria at initial presentation

(range)

(range)

(range)

Heavy proteinuria Nephrotic syndrome Non-nephrotic proteinuria 2 1 g/day/M'

Proteinuria < 1 g/day/M' Slight proteinuria

Proteinuria at second biopsy

Nephrotic syndrome Non-nephrotic proteinuria 2 1 g/day/M* Proteinuria < I g/day/M'

Present

Absent

34.5 f 10.9 47.8 f 10.4*

9.2 k 3.3 (3.8- 1 3.4)

10-2 f 3.8 (4.2 k 15.9) 13.853.7 (7.6f 19.9) 5 (56%)

514

4 (44%) 0 4

5 (56%)

8 (89Yo)t 0 3 5

1 (11Yo)t

39.9 * 7.4 39.2 f 6.8

17/15 9.0 2.5

(5.7-1 2.4) 10.0 f 2.6

(5.8 f 14.6) 12.8f3.0

16 (50%) (6.9-17.2)

12 (38%) 4 8

20 (62%)

14 (44%) 1 1

12 18 (56%)

*P<O.OOI ascompared with the first biopsy. t P c O . 0 5 as compared with the patients with no increase of the mesangial matrix at the second biopsy.

Table 111-Relation of persistent proteinuria and matrix changes

Proteinuria at second biopsy Present Absent (n = 22) (n= 19)

Mesangial matrix/glomerular area (%) First biopsy Second biopsy

Age at onset (years) (range)

Age at first biopsy (years) (range)

Age at second biopsy (years) (range)

40.6 f 8.7 45.2 _+ 9-2* 9.8 & 2.6

(3.8-13.4) 10.9 f 2.9 (4.2-1 5.9) 13.6 f 3.2 (6.9-1 9.9)

37.0 f 7.7 36.6 k4.7 8.3 f 2-4

(4.8-1 2.7) 9.1 22 .6

( 6 w 4.3) 12.4f2.9 (7.3-17.2)

*P<0.05 as compared with the first biopsy.

an irreversible change. Matrix increase was significantly associated with persistence of proteinuria. substances in experimental Increased permeability of the glomerular basement membrane has been shown to be associated with

enhanced mesangial uptake of biolo ically active

that such biologically active substances, including IgA, may stimulate the production of the mesangial

P It is possible

MESANGIAL MATRIX AND CELLS IN IgA NEPHROPATHY

Table IV-ClinicaI findings and mesangial matrix changes in patients with and without a mesangial cell decrease

63

Mesangial cell at second biopsy Decrease No decrease (n = 23) (n= 18)

Malelfemale Age at onset (years)

Age at first biopsy (years)

Age at second biopsy (years)

Macroscopic haematuria Proteinuria at initial presentation

(range)

(range)

(range)

Heavy proteinuria Nephrotic syndrome Non-nephrotic proteinuria 2 1 g/day/M2

Proteinuria < 1 g/day/M2 Slight proteinuria

Proteinuria at second biopsy

Nephrotic syndrome Non-nephrotic proteinuria 2 1 g/day/M2 Proteinuria < 1 g/daylM2

Present

Absent Mesangial matrix increase at second biopsy

1419 9.1 f2.8

(3.8-13.4) 9.7 & 3-0

(4.2-14.3) 12.8 k 3.0 (6.9-17.2) 11 (48%)

9 (39%) 2 7

14 (61%)

9 (39%) 0 1 8

14 (6 1 Yo) 4(17%)

919 9.0 +_ 2-5

(5.7-12.7) 10.6 & 2.8 (5.8-15.9) 13.3k3.3 (7.3-1 9.9) 10 (56%)

9 (39%) 2 5

11 (61%)

13 (72%) 1 3 9

5 (28%) 5 (27%)

Table V-Relationship between mesangial matrix and cell changes

Mesangial matrix Increase No increase at second biopsy (n = 9) (n = 32)

No. of mesangal cells/104 pm2 glomerular area First biopsy 29.7 f 4.4 32.3 k 8.3 Second biopsy 2 1.9 f 6.2* 21.6&4.3?

*P<0 .02 as compared with the first biopsy. tP<O.001 ascompared with the first biopsy

matrix material with a consequent worsening in glomerular morphology, function, and clinical outcome.

Children with benign haematuria were also examined as a control group. Although the mesangial matrix @) increased with age (x) (I>= 17.1 +0.6x), the increase of mesangial matrix with age was only 1.8 per cent over 3 years, the mean interval between the first and second biopsies in our study. Both initial and repeat biopsies showed a greater area of mesangial matrix in children with IgA nephropathy

than in those the benign haematuria. The age- associated increase of mesangial matrix is therefore negligible in this study.

In contrast to mesangial matrix, mesangial cells decreased in the second biopsy in patients with or without persistent proteinuria. This confirms our previous observations that prominent mesangial hypercellularity is almost exclusively seen in the initial biopsy and disappears in the follow-up biopsy.’

Mesangial cells decreased in the second biopsy irrespective of any change in the matrix. An increase in mesangial cells and monocytes is responsible for mesangial hypercellularity in mesangial proliferat- ive glomerulonephritis. Nolasco et a].” and Ferrario er dZ0 reported that in adult IgA nephropathy, few monocytes were found, and that the increase in mesangial cells presumably resulted from proliferation of resident mesangial cells. Mesangial matrix is produced by a mesangial cell, which is thought to have less ability than the monocyte to phagocytose and dispose of the matrix.2’ It follows that an increase of mesangial matrix is likely to be associated with an increase of mesangial cells rather than an increase of monocytes. It is therefore

64 J. SUZUKI ETAL.

possible that the mesangial matrix will increase if the initial mesangial hypercellularity is mesangial cell-dominant but not if it is monocyte-dominant.

Our results also showed that patients with persistent proteinuria showed a significant matrix increase in repeat biopsy, while those without persistent proteinuria had no change. At the first biopsy, there was no difference in mesangial matrix between the two groups. In recent studies of repeat renal biopsy, Berthoux et a1.22 and Levy et a1.6 reported that the histologic grade of an early biopsy was a reliable prognostic indicator. However, this study indicates that follow-up biopsy and examination of matrix change is important for evaluation of the prognosis in children with IgA nephropathy with persistent proteinuria.

ACKNOWLEDGEMENT

The authors are grateful to Dr A. H. Cameron, The Children’s Hospital, Birmingham, for helpful advice.

REFERENCES

I . Clarkson AR. Woodroffe AJ, Banister KM. Lomax-Smith JD. The

2. RodicioJL. IdiopathicIgAnephropathy. Kirlne,.Int 1984;25:717-729. 3. Emacipater SN, Gallo GR, Lamme ME. IgA nephropathy: perspective

on pathogenesis and classification. Clin Nephrol1985; tQ: 161-179. 4. Michalk D, Waldherr R. Seelig HP. Webcr HP, Scharer K. Idiopathic

mesangial IgA-glomeruloncphritis in childhood. Eur J Pediarr 1980; 134: 13-22.

syndrome of IgA nephropathy. C/in Nephrol1984; 21: 7-14.

5. Linne T, Aperia A, Brobcrgcr, 0, era/. Course of renal function in I& glomerulonephritis in children and adolescents. Acra fediarr Scand 1982; 71: 735-743.

6. Levy M, Gonzalrr-Buschard G. Broycr M, er a/. Berger’s disease in children. Medicine 1985; 64. 157-180.

7 . Clarkson AR. Seymour AE. Thompson AJ, era/. IgA nephropathy: a syndrome of uniform morphology. diverse clinical features and uncertain prognosis. Clin NephrolI977; 8: 459-471.

8. Droz D. Natural historyofprimaryglomerulonephntis withmesangial deposits of IgA. Contr Nephrol1976; 2 150-157.

9 . Yoshikawa N, Iijima K, Maehara K, et a/. Mesangial changes in IgA nephropathy in children. Kidney Inr 1987; 32: 585-589.

10. Yoshikawa N, Matsuo T. IgA nephropathy in children. Compr Therapy 1984; 10 35-4 I .

1 I , Yoshikawa N, Uehara S, Yamada K. ef a/. Clinicopathological study of persistent asymptomatic proteinuria in children. Nephron 1980; 25: 127-133.

12. Iidaka K, McCoy J, Kimmelstiel P. The glomerular mesangium: a quantitative analysis. Lob Invesr 1968; 19 573-579.

13. Dunnill MS, Halley W. Some observations on the quantitative anatomyofthekidney.Jfatho/ 1973; 110 113-121.

14. Yoshikawa N, Ito H, Nakahara C, er al. Glomerular electron-dense deposits in childhood IgA nephropathy. Virchows Arch [farhol Anal/ 1985;406:33-43.

15. HaraM,EndoY,NihciH,etal. IgAncphropathywithsubcndothelial deposits. Virchows Arch [farho/ A M I ] 1980; 386: 249-263.

16. Yoshikawa N, Ito H, Yoshiara S , er a/ . Clinical course of IgA nephropathy in children. J fediarr 1987; 110 555-560.

17. Michael AF, Keare WF, Raij L, Vernier RL, Mauer SM. The glomerular mesangium. Kidney Inr 1980; 17: 1-13.

18. Cower WG, Salant DJ. In siru immune complex formation and glomerular injury. Kidney In? 1980; 17: 1-13.

19. Nolasco FEB, Cameron JS. Hartley B. e r a / . Intraglomerular T cells and monocytes in nephritis: study with monoclonal antibodies. Kidney Inr 198% 31: 1160-1 166.

20. Femrio F, Castiglione A, Colasanti G, er a/. The detection of monocytesinhumanglomerulonephritis.KidneyInr 1985;28:513519.

21. Sinniah R, Ow CK, Chia LS. Tan TC. Mesangial handling of immune deposits in IgA nephropathy. In: Davison AM, Briggs JD, Green R, er a/. , eds. Nephrology; Xth International Congress of Nephrology. London: Bailliere Tindall. 1988; 558-573.

22. Ekrthoux FC, Sabatier JC, G n i n C, Gonthier R, Toulon J. Compara- tive renal status at time of initial and five-year-later renal biopsies in mesangial IgA glomerulonephritis. 9th International Congress of Nephrology, Los Angeles, 7 I A (Abstract).

a quantitative analysis of the mesangium in children with iga nephropathy: sequential study

Documents