Thickness of glomerular and tubular basement membranes in preclinical and clinical stages of diabetic nephropathy

Introduction

Diabetes Mellitus (DM) is a metabolic disease characterized by absolute or relative insulin deficiency due either to inadequate secretion or insulin resistance. Type 1 DM is usually associated with absolute insulin deficiency caused by autoimmune destruction of insulin-producing cells in the pancreas, while type 2 diabetes or non-insulin-dependent diabetes mellitus (NIDDM) is characterized by insulin resistance in peripheral tissue and an insulin secretory defect in the beta cells.

Diabetic nephropathy is a frequent complication of DM and occurs with either type 1 or type 2 diabetes, particularly in patients with poor glycemic control. However, the natural history and pathology are similar in both types. The various risk factors associated with development of diabetic nephropathy are the duration of the disease, gender, racial and ethnic factors, genetic factors and degree of metabolic control.1 Although India is termed the ‘Diabetes capital of the world’ the prevalence of microvascular complications such as retinopathy and nephropathy are comparatively lower. The prevalence of overt nephropathy has been reported to be 2.2% and that of microalbuminuria to be 26.9%.2

It is a well-accepted fact that metabolic control plays an important role in the development of diabetic nephropathy (DN). The early phase of nephropathy is characterized by microalbuminuria (MA, 30-300 µg/24 h), which can be recorded by sensitive laboratory measurements and this is followed by overt albuminuria (>300 µg/24 h). There is increased risk of microalbuminuria with a longer duration of diabetes and higher glycosylated hemoglobin value.3

Since the spectrum of morphological changes in DM patients with clinical proteinuria have been well characterized, renal changes associated with microalbuminuria have not been well delineated. Since light microscopic (LM) examination of renal biopsy does not reveal significant changes at this stage, we conducted ultrastructural study of glomerular changes and morphometric analysis of glomerular and tubular basement membranes (GBM and TBM, respectively) to characterize progressive renal changes.

Materials and Methods

In this study, 47 patients with type 2 DM referred from the Department of Nephrology, Safdarjung Hospital, New Delhi, from October 2003 to April 2005 were included. A detailed history was acquired for every patient. The onset of diabetes was defined as the time when the patient first visited with symptoms such as thirst and polyuria and when either urinary sugar or hyperglycemia was first detected. Approval for the study was granted by the ethics committee of Safdarjung Hospital and consent was obtained from each patient.

Diabetes was diagnosed according to the World Health Organization (WHO) criteria.4 Fasting and postprandial plasma glucose were measured by oxidase method. Urinary albumin and sugar were estimated by dipstick method. Of the 47 patients, 11 had albuminuria (A) and were included in the clinical group. The remaining 36 patients were subjected to microalbuminuria estimation by radioimmunoassay. Microalbuminuria (MA) was detected in a total of 14 patients who were included in the preclinical category. The 25 patients with albuminuria and/or microalbuminuria were investigated further for HbA1c, blood urea, serum creatinine and renal biopsy.

Three renal tissue cores were collected in 4% formaldehyde, 70% ethanol and forglu for LM, immunofluorescence and electron microscopy (EM), respectively. To facilitate histological diagnosis, special stains such as Periodic acid-Schiff (PAS), silver methenamine and Masson's trichrome were used besides hematoxylin and eosin (H & E). Direct immunofluorescence staining of frozen sections was performed for immunoglobulin (Ig)A, IgG, IgM, complement 3 (C3) and fibrinogen. For EM studies, ultra-thin sections were stained with aqueous saturated solution of uranyl acetate and lead citrate. These contrasted sections on the grid were examined under HITACHI H7500 TEM. The sections were studied for ultrastructural changes in the glomeruli, tubules and blood vessels.

A series of images covering the entire available glomeruli and tubules were obtained through Gatan charge-coupled device (CCD) camera at a magnification of 6000x; these images were then converted to JPEG format to perform morphometric analysis to measure the GBM and TBM thickness by using a Soft Imaging System GmbH (analysis 3). The images were calibrated using an inbuilt micron bar. A 200 × 200 nm grid was overlaid on the images [Fig.1]. GBM and TBM width was obtained by the orthogonal intercept method. The measurements were made at each point where a grid line intercepted an endothelial/peripheral GBM interface. The GBM width was measured on a line orthogonal to the edge of the peripheral GBM at the endothelial side of the intercept. TBM was measured at 50 orthogonal intercepts per patient and presented as mean ± SD.

Close

Fig. 1

Ultrastructure of the glomerular capillary loop with a grid overlay for morphometric assessment

Export to PPT

Seven patients of age-matched minimal change nephrotic syndrome (MCNS) were used as the control for comparison. The mean GBM and TBM thickness was then calculated for each patient in all the three patient groups [control, preclinical (MA) and clinical (A) groups]. Then, the mean GBM and TBM thickness was calculated in each of the three groups and the data was presented as mean values ± SD.

The data were analyzed for statistical differences among different groups by using Statgraphics Statistical Package. Student's t test was applied to estimate the statistical variation between the means and variances of two samples; chi-Square test was applied to determine the effect of different conditions on different parameters.

Results

The biochemical parameters in the preclinical and clinical groups are presented in Table 1 and show a significantly longer duration of disease, higher blood urea and serum creatinine levels in the clinical group than that in the preclinical group.

Discussion

Renal disease is the leading cause of death and disability in diabetes. GS is the most important cause of renal failure in diabetic patients since approximately 30% of the diabetic patients die due to renal failure.5 A high proportion of individuals with type 2 DM are found to have microalbuminuria and overt nephropathy shortly after diagnosis since diabetes is actually present for many years before being diagnosed.6 Without specific interventions, 20–40% of type 2 diabetic patients with microalbuminuria progress to overt nephropathy.7 If diabetic patients do not have proteinuria after 25–30 years of diabetes, they are unlikely to ever develop GS. Once proteinuria occurs, progression to end-stage renal disease (ESRD) is inevitable despite good metabolic control.89 It is therefore imperative to recognize the onset of renal damage to plan further management to reverse the progression of this disease. At present, MA is the most reliable feature to predict early renal damage in diabetic patients.

The evolution of diabetic GS is described in five clinical stages. The initial two stages include increased GFR with hypertrophic kidneys and a stage in which glomerular lesions develop in the absence of clinical evidence of renal disease. During these stages, the GBM begins to thicken and the mesangial matrix expands. These two stages are reversible with a good glycemic control. This is followed by stages 3–5 manifested by the development of microalbuminuria, overt diabetic nephropathy and ESRD, respectively.10 The incidence of diabetic GS is known to increase with the duration of diabetes, and our study has demonstrated a significant increase in BM thickness with increasing duration, irrespective of whether the patient had microalbuminuria or albuminuria. The GBM and TBM width were also found to increase with HbA1c levels but not with fasting and postprandial glucose levels, indicating the influence of consistent glycemic control in the occurrence of basement membrane thickness.

Drummond et al.11 had demonstrated that the mean GBM thickness was significantly higher in type I diabetic microalbuminuric patients (510 ± 93 nm) as compared to the controls (425 ± 78 nm), while Osterby et al.12 showed a similar increase in albuminuric patients (562 nm) compared to the controls (353 nm). In our study group comprising mainly type 2 diabetic patients with microalbuminuria (preclinical), a significantly large number of patients (78.6%) had normal renal morphology as observed by LM, and only one patient with overt nephropathy showed normal morphology. Ultrastructural examination revealed a large number of patients (48%) with increased GBM membrane thickness in both the preclinical and clinical groups even without LM change.

Caramori et al.13 demonstrated a progressively increasing GBM width in type 1 diabetic patients as protein excretion progressed from normoalbuminuria to microalbuminuria and further to proteinuria; these findings were consistent with those of the present study, indicating the importance of morphometry in evaluating GBM and TBM width for the possible diagnosis of early diabetic nephropathy especially in the event of no microscopic changes. Ultrastructural morphometry showed GBM and TBM thickness in 64.3% patients, while 14.3% did not show any changes in either GBM or TBM, suggesting that with a good glycemic control, the progression to renal damage could be prevented.

In this study, the TBM thickness was higher than that of the GBM. This could possibly be due to the presence of tubular atrophy. TBM thickening has previously been reported in diabetics compared to nondiabetics. Vascular changes were observed under LM but not measured on EM.

A population difference has been noted in the mean thickness of GBM and TBM in all groups of Indian patients compared to the findings in Caucasians; however, this finding needs to be confirmed using a larger set of samples. A systematic study on GBM thickness by Rayat et al. in the Indian population showed the mean GBM thickness to be 321 ± 28 nm.14 This value is comparable to that reported by us in this study. It is possible that the genes mediating the synthesis of capillary GBM and mesangial matrix components have ethnic variation accounting for variation in basement membrane thickness and also causing a delay in the progression to diabetic glomerulosclerosis. The significance of identification of early renal changes using morphometric techniques is that it may enable reversing of the renal damage by controlling microalbuminuria.15