SUMMARYThe aim of the sampling design for stereology is to obtain the maximal amount of quantitative structural information at a given total cost or effort. Principles of such optimal designs are discussed and methods for generating them are illustrated by a biological example. In general, the variation between different individuals—the biological variation—is the major determinant of overall efficiency, whereas the variation between single microscopic features is unimportant. It follows that the expenditure of time and/or money in order to increase the precision of the individual measurements is irrational in almost all studies where the emphasis is on the biological results.

This report deals with the reconstruction of the distribution of membrane thickness T from that of orthogonal length Lo, measured in random section planes. In such planes the membrane appears as a band and the linear distance from one of its boundaries perpendicular to the opposite one is the length of the orthogonal intercept. Using a membrane model, an integral equation relating the probability density functions of orthogonal intercept length f(lo) and membrane thickness g(tau) is derived. Relations between moments are derived and the analytic solution to the problem of reconstructing g(tau) from f(lo) is given. The parametric approach by which it assumed that g(tau) has some known analytic form with unknown parameters is considered, and the use of a suggested analytic form for describing the thickness distribution of the human glomerular basement membrane is discussed.

Exact parameters for relevant glomerular structures in the course of streptozotocin diabetes in rats with 1 to 18 months' duration were obtained with stereological methods. Renal cortical tissue from diabetic (D) and control animals (C) was processed for light- and electron microscopy and measurements were performed on systematically sampled glomeruli. The thickness of the basement membrane (BM) increased with age in both groups, but the rate of increase was 50% higher in D: 19 +/- 1.2 nm/month (mean +/- SD) vs. 13 +/- 0.9 nm/month, P = 0.0003. The time course of other structural quantities was characterized by the acute changes constituting the glomerular hypertrophy, earlier shown to develop within the first few days of diabetes. All these changes were confirmed in the present study: In the earliest phase the diabetic rats showed an increased total volume of glomeruli, mesangium, and mesangial BM material, as well as an increased surface of the capillary walls. However, none of these differences between the groups showed progression with increasing duration. Mesangial changes corresponding to those of the glomerulopathy in long-term diabetes were not demonstrable within the experimental period. The streptozotocin diabetic rat, therefore, is not useful as a model of advanced diabetic glomerulopathy. But the BM thickness follows the same predictable time course as in human diabetes insofar as moderately advanced cases are concerned. BM thickness is the parameter of choice when a potential effect of different variables on the development of diabetic glomerulopathy is under study.

Diabetic nephropathy leading to kidney failure is a major complication of both type I (insulin-dependent) and type II (non-insulin-dependent) diabetes mellitus, and glomerular structural lesions (especially expansion of the mesangium) may constitute the principal cause of decline in kidney function experienced by a significant fraction of diabetic patients. Although the biochemical bases of these mesangial abnormalities remain unknown, an understanding of the natural history of diabetic nephropathy from a combined structural and functional approach can lead to greater pathophysiological insight. Work in animals has supported the concept that the metabolic disturbances of diabetes mellitus cause diabetic nephropathy, with structural and functional lesions prevented or reversed with improved or normalized glycemic control. Additional research must address this fundamental issue in humans, especially the response of advancing mesangial lesions to improved glycemic control. Factors not directly related to the metabolic perturbations of diabetes may serve to accelerate or diminish the pathophysiological processes of diabetic nephropathy. The elucidation and management of these factors, when coupled with improved glycemic control, may moderate the development or progression of diabetic kidney lesions in humans.

SummaryWe investigated in a randomized, prospective study the influence of improved blood glucose control during 2-3 years in young insulin-dependent diabetic (IDDM) patients with microalbuminuria, which is indicative of early nephropathy. Patients were randomized either to intensive treatment by continuous subcutaneous insulin infusion (CSII) (n = 9) or CT (n = 9). Kidney biopsies were taken at baseline and after 26-34months. End points were structural changes in the glomeruli. Sensitive, quantitative, morphometric methods were used. The blood glucose control improved significantly (p ---0.01) during the study in the CSII-group as glycated haemoglobin (HbAlc) fell from 10.1% ([95 % CI] 8.9-11.3) to 8.6 % (7.9-9.2), but not in the . Mean HbAIr during the study period was significantly lower in the CSII-group than in the CTgroup, 8.7% (8.1-9.3) vs 9.9% (8.5-11.3), p =0.04. Basement membrane thickness (BMT) increased in both groups, most (CT vs CSII, p = 0.03) in the CTgroup: 140 nm (50-230) vs CSII: 56 nm (27-86). In the CT-group only an increase was seen in matrix/mesangial volume fraction (p = 0.006) and matrix star volume (p =0.04). Furthermore, a positive correlation between mean HbAlc during the study and change from baseline in BMT (r=0.70, p =0.001) and matrix/glomerular volume fraction (r --0.33,p = 0.09, NS) was demonstrated. Albumin excretion rate correlated significantly to BMT and most of the matrix parameters. The present study shows that during a period of only 2.5 years, a close relationship between the level of mean blood glucose and progression of glomerular morphological changes in early diabetic nephropathy can be demonstrated. [Diabetologia (1994) 37: 483-490] Key words Diabetic glomerulopathy, microalbuminuria, basement membrane thickness, mesangial expansion, mesangial matrix, stereology, hyperglycaemia.Even though hyperglycaemia is a prerequisite for the development of diabetic nephropathy, the impact of long-term hyperglycaemia on the progression of early diabetic nephropathy is not well understood. Abbreviations: IDDM, insulin-dependent diabetes mellitus; CSII, continuous subcutaneous insulin infusion; CT, conventional treatment (2, 3 or multiple injections daily); BMT, basement membrane thickness; AER, urinary albumin excretion rate; CI, confidence interval.Pirart [1] showed in a study comprising both IDDM and NIDDM patients, that the risk of development of serious complications including nephropathy, is associated with more severe hyperglycaemia. In cross-sectional [2--6] and long-term retrospective studies [7,8] an association between the level of glycated haemoglobin and microalbuminuria, which is an early sign of diabetic nephropathy [9,10], has been found. Improved blood glucose control obtained in prospective randomized studies has retarded the progression of AER [11] and the risk of developing clinical nephropathy [12,13]. However, it is not known if reducing mean blood glucose affects the progression of morphological changes at a very early stage of diabetic nephropathy. There...