JN 2000; Vol.13 N°4: 275-281

Original Investigation

JNEPHROL 2000; 13: 275-281

Clinical significance and long-term evolution of minimal change histopathologic variants and of IGM nephropathy among Egyptians

Ahmed F. Donia, Mohamed A. Sobh, Fatma E. Moustafa, Mohamed A. Bakr, Mohamed A. Foda - Urology and Nephrology Center, University of Mansoura - Egypt

ABSTRACT: In children, the most frequent idiopathic nephrotic syndrome is minimal change nephrotic syndrome (MCNS). Typically, MCNS shows no abnormalities by light microscopy: "nil disease". Beside this classic picture, there are other minor light microscopic abnormalities which are considered as MCNS variants. Our 172 MCNS patients were divided into a nil disease group, two groups of MCNS variants (mild mesangial hypercellularity and mild mesangial thickening) and a fourth group with normal light microscopy and diffuse IgM deposition (IgM nephropathy group). The relation of this fourth group to MCNS is controversial in the literature. Age and serum creatinine were significantly different in the four histologic groups (P=0.03 for age and 0.047 for serum creatinine). Comparing the groups in pairs, it appeared that these significant differences were due to significantly higher age and serum creatinine in the mild mesangial hypercellularity group than in the IgM nephropathy group (P = 0.02 for age and 0.01 for serum creatinine). The groups were similar as regards follow-up creatinine clearance and early and late steroid response. We concluded that mild mesangial hypercellularity may differ from other MCNS forms as regards age at presentation and renal function. We also suggest that IgM nephropathy with normal light microscopy is similar to MCNS.

Keywords: Evolution, IgM nephropathy, Minimal change nephrotic sydrome variants

Introduction

Minimal-change nephrotic syndrome (MCNS) is defined morphologically on the basis of normal glomeruli by light microscopy, absent or non-specific immunoglobulin deposits by immunofluorescence, and diffuse epithelial cell foot-process effacement by electron microscopy (1).
Based on electron microscopic and other findings, new data suggest that the podocyte phenotype may help define MCNS and focal segmental glomerulosclerosis (FSGS) (2). Additionally, podocyte vacuolization and effacement may have a diagnostic and prognostic role in MCNS (3). The absence or presence of podocytes in urine may differentiate MCNS from mesangial proliferation (4). In 1981, an ISKDC report (5) stated that although the glomeruli in MCNS are customarily normal by light microscopy (nil disease), the histopathologic concept of minimal changes does, however, include minor abnormalities. Accordingly, MCNS was divided into five subcategories (variants): (a) Nil disease, (b) Focal glomerular obsolescence, (c) Mild mesangial thickening, (d) Focal tubular changes, and (e) Mild mesangial hypercellularity. The ISKDC report described the histopathologic changes seen in these five subgroups and assessed their clinical significance.
Since this report, no other studies have described MCNS variants according to their light microscopic features, especially in developing, subtropical and mediterranean countries where race, socioeconomic status and environmental factors may greatly affect the patterns.
According to the ISKDC classification for MCNS, we reviewed renal biopsies of our nephrotic patients diagnosed by a single histopathologist as MCNS. Out of a total of 172 biopsies, 85 were nil disease, 52 were mild mesangial hypercellularity and 23 were mild mesangial thickening. None of our biopsies was diagnosed as focal glomerular obsolescence or focal tubular atrophy. So we had only three variants of MCNS. The remaining 12 biopsies were normal by light microscopy. However, an account of their very specific immunofluorescent picture, they were diagnosed as IgM nephropathy. It is still controversial whether this last group constitutes a separate entity or is simply a part of the spectrum of MCNS-FSGS-mesangial proliferative glomerulonephritis (6).
We conducted this study to correlate three of the histopathologic variants of MCNS, based on light microscopic findings, and their clinical features among Egyptians. The histopathologic diagnosis of IgM nephropathy with a normal light microscopic picture was added to the study and was correlated with its clinical data, trying to clarify its similarity or dissimilarity from MCNS variants. We also assessed the long-term evolution of the histopathologic variants of MCNS and of IgM nephropathy as regards steroid response and renal function.

Materials and methods

Patients

All idiopathic, biopsy-proven MCNS patients were included in this study. They were collected from a total of 1000 biopsy-proven, glomerulonephritis patients diagnosed in the Urology and Nephrology Center, University of Mansoura, Egypt, between 1983 and 1993. Their total number was 172 patients, 108 males and 64 females. Their age ranged from 2 to 50 years (median 10 years). They were classified into four subgroups:
1) Nil disease: 85 patients
2) Mild mesangial hypercellularity: 52 patients
3) Mild mesangial thickening: 23 patients
4) IgM nephropathy (no abnormalities by light microscopy): 12 patients.

Clinical methods

We reviewed the history of all patients at the time of biopsy and yearly for four years thereafter, with special attention to the duration of disease, number of relapses/year and degree of edema. Any with a history of disease or drug use that could cause secondary glomerulonephritis were excluded. From examination data, hypertension and edema were carefully looked for. Investigations to exclude secondary glomerulonephritis were reviewed. Intestinal bilharziasis, one of the major causes of secondary glomerulonephritis in our country, was excluded by stool analysis and rectal mucosal biopsy for those with a history of exposure to the parasite.
Results of urinalyis, 24-hour urinary protein and serum albumin, cholesterol and creatinine at time of biopsy and at each follow-up were reported.

Histology

One core of the renal tissue was fixed in 10% neutral buffered formalin and processed as paraffin sections which were stained with hematoxylin and eosin, periodic acid Schiff, Masson trichrome, silver methenamine and Congo red stains and then examined by light microscopy.
The second core was immediately deep-frozen (-70 °C). Frozen kidney sections were stained by fluorescein, labeled antihuman IgG, IgA, IgM, C₃, C₄, C_1q and fibrinogen at dilutions of 1:8 to 1:10 preparing them for immunofluorescent examination.
A third core (taken in some patients only) was used for electron microscopic examination.

Clinical definitions

Established definitions are used to define prednisone regimens, steroid responder, infrequent relapser, frequent relapser, steroid-dependent and non-responder (7). Till the age of 17, hypertension was diagnosed if systolic and/or diastolic blood pressure was above the 95^th percentile (8). Patients over 17 years were considered hypertensive if blood pressure exceeded 140/90 (9).

Histologic definitions

Nil disease was diagnosed when glomeruli were entirely normal by light microscopy with neither mesangial thickening nor mesangial hypercellularity (5). Mild mesangial thickening was diagnosed when there were slight increases in the amount of PAS-positive mesangial matrix without increased cellularity (5, 10, 11). In mild mesangial hypercellularity there was a slight increase of mesangial cellularity. It was differentiated from nil disease on the one hand and from diffuse mesangial hypercellularity on the other on the basis of the number of cells per peripheral mesangial area. One or two cells per peripheral mesangial area indicated nil disease, whereas three were taken as mild mesangial hypercellularity. Four or more cells were reagarded as diffuse mesangial hypercellulariy (5, 11). The IgM nephropathy subgroup included biopsies with no changes by light microscopy but with diffuse IgM mesangial deposition. IgM nephropathy with light microscopic pictures other than minimal changes was excluded.

Statistical analysis

To test for normal distribution, the frequency of cases was plotted against the normal distribution curve for the group mean and standard deviation. As most of the data showed obvious deviation from the normal distribution, non-parametric statistical methods were used.Non-parametric Kruskal-Wallis analysis of variance (ANOVA) was used to compare the four groups. To test for differences between pairs of groups, the Mann-Whitney test was used. Cross-tabulation and the X2 test were used to test for associations between different pathologic types and other categorial variables. The tests were run on an IBM-compatible computer using SPSS/PC+ statistical package (SPSS Inc Chicago, IL). A P value less than 0.05 was considered significant.

Results

The 172 patients included 85 with nil disease, 52 with mild mesangial hypercellularity (Fig. 1), 23 with mild mesangial thickening (Fig. 2)

Donia F1

Fig. 1

Donia F2

Fig. 2

and 12 with IgM nephropathy. The main demographic and clinical characteristics of patients in the four groups at time of biopsy are shown in Table I.

TABLE I - DEMOGRAPHIC AND CLINICAL CHARACTERISTICS^a OF DIFFERENT SUBGROUPS^b

.	Nil	NMH	MMT	IgM	p Value
Age (years)	10	14	8	3.5	0.03^c
Age <6 years	42/85 (49.1%)	35/52 (67.2%)	20/33 (87.9%)	11/12 (91.7%)	0.0007^d
Sex (female)	34/85 (40%)	17/52 (32.7%)	7/23 (30.4%)	6/12 (50%)	0.566
Disease duration (months)	5	5	4	6	0.8
Numberof relapses/year	0	0	1.2	2	0.0001^e
Microscopic hematuria	17/85 (20%)	15/52 (28.8%)	6/23 (26.1%)	3/12 (25%)	0.687
Hypertension	19/85 (22.4%)	14/52 (26.9%)	4/23 (17.8%)	3/12 (25%)	0.826

a. Median is used to describe age, disease duration and number of relapses/year, and number/total number and percentage to describe the other data.
b. Abbreviations used are Nil, nil disease; MMH, mild mesangial hypercellularity; MMT, mild mesangial thickening; IgM, lgM nephropathy.
c. Significant (Kruskal-Wallis 1-way ANOVA)
MMH versus IgM: p= 0.02 (using Mann-Whitney U test).
d. Significant (Kruskal Wallis 1-way ANOVA)
Nil versus MMT: p= 0.002
Nil versus IgM: p= 0.014
e. Significant (Kruskal-Wallis 1-way ANOVA)
Nil versus MMT: p= 0.009
Nil versus IgM: p= 0.0466
MMH versus MMT: p= 0.0001
MMH versus lgM: D= 0.0067

Significant differences were observed only for age and number of relapses/year. The difference in age among the four groups (p=0.03) was due to higher age in the mild mesangial hypercellularity group than in the IgM nephropathy group (P=0.02). The number of relapses/year was significantly different among groups (p=0.0001). This was due to a lower number of relapses/year in the nil disease and mild mesangial hypercellularity groups than in the mild mesangial thickening (p=0.0009 and 0.04) and IgM nephropathy groups (p = 0.0001 and 0.006).
Laboratory characteristics of patients in the different groups are shown in Table II.

TABLE II - LABORATORY FINDINGS^a IN DIFFERENT SUBGROUPS^b

.	Nil	NMH	MMT	IgM	p Value
Serum creatinine (mg/dl)	0.6	0.7	0.5	0.5	0.047^c
Creatinine cleareance^d (ml/min./1.73²)	123.2	117.7	129.25	135.45	0.54
Serum albumin (g/dl)	1.7	1.9	1.2	2	0.2
Serum cholesterol (mg/dl)	359	338	414	360	0.31

a. Medians are given
b. Abbreviations used are Nil, nil disease; MMH, mild mesangial hypercellularity; MMT, mild mesangial thickening; IgM, lgM nephropathy.
c. Significant (Kruskal Wallis 1-way ANOVA)
MMH versus IgM: p=0.0112
d. Calculated from serum creatinine using Schwartz's formula for patients under 21 years and Cockcroft and Gault's formula for patients over 21 years.

Only serum creatinine showed a significant difference among groups (p=0.04). This was due to a significantly higher value in mild mesangial hypercellularity than in IgM nephropathy (p = 0.01). However, normalization of serum creatinine to creatinine clearance/1.73 m² surface area using Schwartz's formula for patients under 21 years of age (12) and Cockcroft and Gault's formula for those over 21 years of age (13) resulted in the disappearance of this difference.
At time of biopsy, there was no significant difference between the groups as regards the incidence of permanent remission, steroid dependence, steroid resistance, infrequent relapses or frequent relapses (Tab. III).

TABLE III - STEROID RESPONSES^a IN DIFFERENT SUBGROUPS^b

.	Nil	NMH	MMT	IgM	p Value
Permanent remission	26/60 (43.3%)	13/35 (37.1%)	4/21 (19%)	1/7 (14.3%)	.
Steroid-dependent	25/60 (41.7%)	19/35(54.3%)	13/21 (61.9%)	6/7 (85.7%)	0.3
Steroid-resistant	8/60 (13.3%)	3/35 (8.6%)	1/21 (4.8%)	0/7 (0%)	.
Infrequent relapses	1/60 (1.7%)	0/35 (0%)	1/21(4.8%)	0/7 (0%)	.
Frequent relapses	0/60 (0%)	0/35 (0%)	2/21 (9.5%)	0/7 (0%)	.

a. Number/total number and percentages are used to describe data.
b. Abbreviations used are Nil, nil disease; MMH, mild mesangial hypercellularity; MMT; mild mesangial thickening; IgM, IgM nephropathy. Total number in each group is reduced because of exclusion of patients who received adjuvant therapy.

Patients who received adjuvant treatment (azathioprine, cyclosphosphamide or cyclosporine A) were excluded from this analysis. Table IV shows that this similarity in steroid response continued among subgroups over the four years after renal biopsy. Patients who did not attend four years of follow-up after biopsy or who received adjuvant therapy after biopsy were excluded from this analysis.
Serial creatinine clearance done yearly for four years showed insignificant differences among subgroups.

TABLE IV - STEROID RESPONSES^a IN DIFFERENT SUBGROUPS^b AT LAST FOLLOW-UP^c

.	Nil	NMH	MMT	IgM	p Value
Permanent remission	3/20 (15%)	2/10 (20%)	1/14 (7.1%)	0/7 (0%)	.
Steroid-dependent	9/20 (45%)	3/10(30%)	7/14 (50%)	6/7 (85.7%)	0.67
Steroid-resistant	6/20 (30%)	4/10 (40%)	4/14 (28.6%)	1/7 (14.3%)	.
Infrequent relapses	2/20 (10%)	0/10 (0%)	1/14(7.1%)	0/7 (0%)	.
Frequent relapses	2/20 (0%)	0/10 (0%)	1/14 (7.1%)	0/7 (0%)	.

a. Number/total number and percentages are used to describe data.
b. Abbreviations used are Nil, nil disease; MMH, mild mesangial hypercellularity; MMT; mild mesangial thickening; IgM, lgM nephropathy.
c. Total number of patients is far less than the initial population, because patients who did not continue the follow-up or who received adjuvant therapy were excluded.

TABLE V -SERIAL CREATININE CLEAREANCE^a,b IN DIFFERENT SUBGROUPS^c,d

.	Nil	NMH	MMT	IgM	p Value
One year	123.2	117.7	129.3	135.5	0.73
Two years	116.3	134.2	126.5	121	0.42
Three years	111.8	141.6	119.6	121	0.21
Four years	103.8	134.4	113.3	103.4	0.23

a. Medians are given
b. Calculated from serum creatinine using Schwartz's formula for patients under 21 years and Cockcroft and Gault's formula for patients over 21 years.
c. Abbreviations used are Nil, nil disease; MMH, mild mesangial hypercellularity; MMT, mild mesangial thickening; IgM, lgM nephropathy.
d. Total number of patients (51) is far less than the initial population (172 patients) because patients who did not continue the follow-up or who received adjuvant therapy were excluded.

Discussion

The clinical significance of histopathologic variants of minimal change disease was thoroughly evaluated in the ISKDO report (5). However, similar studies in developing, subtropical or mediterranean countries are lacking. Our study was done in Egyptian patients who are good representatives of such countries where race, socioeconomic status and environmental factors are expected to produce different results from those seen in western populations. In addition, our study included the subgroup of IgM nephropathy with normal light microscopic picture. The relation of this group to minimal change disease is still debitable and needs to be clarified. Some authors consider that patients with IgM in the mesangium should be considered as a separate entity "IgM-associated nephropathy" and have a greater tendency to steroid unresponsiveness and evolution to progressive renal failure (14-16). Others found that such deposits, as long as the light microscopic and ultrastructural features are consistent with minimal change disease, did not seem to indicate any tendency to steroid unresponsiveness or to progressive disease (17-19).
Our study was done in one center and all renal biopsies were examined by one histopathologist. This was in contrast with the ISKDC report (5) where data were collected from 24 participating clinics and renal biopsies were examined by four pathologists.
This retrospective study included 172 patients with
idiopathic MCNS, divided into four groups; nil disease, mild mesangial thickening, mild mesangial hypercellularity and IgM nephropathy (with no abnormalities by light microscopy). Unlike in the ISKDC report (5) we had no patients with focal glomerular obsolescence or focal tubular changes.
Sixty-two percent of MCNS patients were aged from 1 to 6 years in our study. This is in agreement with the fact that the peak incidence of MCNS is between 1 and 6 years (20). Interestingly, only 49.1% of nil disease group patients were below the age of 6, compared with 78.5% in the ISKDC report (5). This difference may be due to a difference in sample size (85 in our study and 219 in the ISKDC report).
his study, like the ISKDC report (5) found a significant difference in age in the different subgroups. In our study, this difference (p=0.03) was due to significantly higher age (p=0.02) in mild mesangial hypercellularity than in IgM nephropathy. Again the number of patients aged less than six years is significantly lower in the nil group than in the mild mesangial thickening and IgM nephropathy groups (P=0.002 and 0.014). These findings suggest that IgM nephropathy may present at younger age and this needs to be further evaluated with bigger studies.
The male-to-female ratio in our study was 1.75 : 1, slightly lower than the ratio reported by Glassock et al (6) in childhood MCNS (2 to 2.5 : 1). This may be attributed to the relatively higher age range in our group (2-50 years) than the ISKDC group (12 weeks-16 years), as the male: female ratio comes closer to unity in adult MCNS (6). We found no significant sex difference in the different subgroups. This coincides with the results of ISKDC (5).
We did find significant differences in the numbers of relapses/year in different subgroups (p=0.001). This was significantly higher in IgM nephropathy than in nil disease (p=0.046). This correlates with the study by Trachtman et al (21) who proved the paucity of nil disease in children with frequent relapses; among 16 frequent relapser nephrotic syndrome patients, only four were nil disease but seven were IgM nephropathy.
Our data indicated no significant differences among subgroups regarding the incidence of microscopic hematuria or bypertension. This agrees with the ISKDC report (5). The order of frequency in nil disease, mild mesangial thickening and mild mesangial hypercellularity was similar in our study and the ISKDC report. In both studies, incidence was the highest in mild mesangial hypercellularity (hematuria occurred in 28.8% in our study and 36% in the ISKDC report, and hypertension in 26.9% and 20.8% respectively). This supports the ISKDC suggestion (5) that mild mesangial hypercellularity is different from other MCNS subgroups. The overall incidence of microscopic hematuria among Egyptian MCNS patients (23.8%) correlates with that reported by the ISKDC in 1983 (23%). Similarly, the incidence of hypertension among our patients (23.4%) agrees with the 20% reported by the ISKDC (5).
The significantly higher serum creatinine in mild mesangial hypercellularity than in IgM nephropathy in our study may again support the ISKDC suggestion (5) that mild mesangial hypercellularity is different from other MCNS subgroups, or may be due to differences in age in the two groups, as suggested by the disappearance of this difference after normalization of serum creatinine to creatinine clearance/surface area.
Serum cholesterol and albumin were insignificantly different among groups, suggesting that the degree of nephrotic syndrome was similar in all groups.
The pattern of response to steroid therapy was similar among subgroups. The incidence of permanent remission, steroid dependence, steroid resistance, infrequent relapses and frequent relapses showed no real difference among subgroups at biopsy. This agrees with the ISKDC report (5). These observations in our study continued throughout a four-year follow-up.
Long-term GFR showed insignificant difference among subgroups. This was evidenced by comparable creatinine clearance values each year for four years of follow-up.
We concluded that, among MCNS variants, mild mesangial hypercellularity has a significantly higher age at presentation and hence higher serum creatinine. It also has the highest incidence of hypertension and microscopic hematuria. Although these findings are likely to affect the outcome of this variant, the three MCNS variants were similar as regards early and late steroid response and long-term function. We also concluded that the presence of IgM deposits in MCNS biopsy, the so-called IgM nephropathy, did not seem to affect steroid response or long-term outcome

Acknowledgements

The authors express their gratitude to Ayman EI-Magawry and Fathya Gado for their technical assistance. The authors thank Hend Sharaby for her secretarial assistance in preparing the manuscript and are grateful to Sahar Abdel-Rahman for statistical assistance.

Reprint requests to: M.A. Sobh, M.D., FACP - Professor and Head of Nephrology Department Urology and Nephrology Center Mansoura University Egypt

References (when available, each reference has been linked to PubMed)

1. Glassock RJ. The glomerulopathies. In: Schrier RW ed. Renal and electrolyte disorders, ed 4. Boston, Toronto, London: Little Brown and Company 1993; 727-805.

2. Sharif K, Goyal M, Kershaw D, Kunkel R, Wiggins R. Podocyte phenotypes as defined by expression and distribution of GLEPP 1 in the developing glomerulus and in nephrotic glomeruli from MCD and FSGS. A dedifferentiation hypothesis for the nephrotic syndrome. Exp Nephrol 1998; 6: 234-44.

3. Chiang ML, Hawkins EP, Berry PL, Barrish J, Hill LL. Diagnostic and prognostic significance of glomerular epithelial cell vacuolization and podocyte effacement in children with minimal lesion nephrotic syndrome and focal segmental glomerulosclerosis: an ultrastructural study. Clin Nephrol 1988; 30: 1: 8-14.

4. Hara M, Yanagihara T, Takada T, Itoh M, Matsuno M, Yamamato T, Kihara I. Urinary excretion of podocytes reflects disease activity in children with glomerulonephritis. Am J Nephrol 1998; 18: 1: 35-41.

5. International Study of Kidney Disease in Children: Primary nephrotic syndrome in children. Clinical significance of histopathologic variants of minimal change and of diffuse mesangial hypercellularity. Kidney Int 1981; 20: 765-71.

6. Glassock RJ, Cohen AH, Adler SG. Primary glomerular diseases. In: Brenner BM and Rector FC Jr eds. The kidney, ed 5. Philadelphia: E.B. Saunders 1996; 1392-497.

7. International Study of Kidney Disease in Children. Prospective, controlled trial of cyclophosphamide therapy in children with the nephrotic syndrome. Lancet 1974; 2; 423-7.

8. Report of the Second Task Force on Blood Pressure Control in Children. Pediatrics 1987; 79: 1-25.

9. Joint National Committee. The fifth report of the Joint National Committee on the Detection, Evaluation and Treatment of High Blood Pressure. Arch Intern Med 1993; 153: 154-83.

10. Churg J, Habib R, White, RHR. Pathology of the nephrotic syndrome in children. Lancet 1970; 1: 1299.

11. Churg J, Berstein J, Glassock RJ. Minor glomerular abnormalities. In: Churg J, Berstein J, Glassock RJ eds. Renal disease: Classification and atlas of glomerular diseases, 2nd ed. Tokyo, New York: Igaki-Shoin. 1995: 45.

12. Schwartz GJ, Gauthier BA. Simple estimate of glomerular filtration rate in adolescent boys. J Pediatr 1985; 106: 3: 522-6.

13. Cockroft D. Gault MK. Prediction of creatinine clearance from serum creatinine. Nephron 1976; 16: 31-41.

14. Cohen AH, Border WA, Glassock RJ. Nephrotic syndrome with glomerular mesangial IgM deposits. Lab Invest 1978; 38: 610-9.

15. Kopopolvic J, Shvil Y, Pomeranz A, Ron N, Rubinger D, Oren R. IgM nephropathy: morphological study related to clinical findings. Am J Nephrol 1987; 7: 275-80.

16. Saha H, Mustoneu J, Pasternak A, Halin H. Clinical follow up of 54 patients with IgM nephropathy. Am J Nephrol 1989; 9: 124.

17. Hui A, Poucell S, Thorner P, Rance CP, Baumal R. Clinical features and glomerular immunofluorescence of renal biopsies from children with nephrotic syndrome due to minimal change disease and two variants of mesangial proliferative glomerulonephritis. Int J Pediatr Nephrol 1984; 1: 5-10.

18. Habib R, Girardin E, Gagnadoux MF, Hinglais N, Levy M, Broyer M. Immunopathological findings in idiopathic nephrosis: clinical significance of glomerular "immune deposits". Pediatr Nephrol 1988; 2: 204-8.

19. Andematten F, Bianchetti MG, Gerber HA, Zimmermann A, Meregalli P, Luthy C, Oetliker OH. Outcome of idiopathic childhood nephrotic syndrome. A 20 year experience. Scand J Urol Nephrol 1995; 92: 1: 15-9.

20. Thomson NM, Charlesworth F. Classification, pathology and clinical features of glomerulonephritis. In: Whitworth JA and Lawrance JR eds. Textbook of renal disease, ed 2. Edinburgh: Churchill Livingstone 1994; 129-73.

21. Trachtman H, Carroll F, Phadke K, Khawar M, Nicastri A, Chen CK, Tejani A. Paucity of minimal-change lesion in children with early frequently relapsing steroid-responsive nephrotic syndrome. Am J Nephrol 1987; 7: 13-7.

Received: April 20, 1999 Revised: January 27, 2000 Accepted: May 19, 2000

Back to Home Page