Eileen N. Ellis

Renal biopsies in 45 patients with insulin-dependent diabetes mellitus (IDDM) were examined by semiquantitative light microscopy and quantitative electron microscopic stereologic morphometry. In these 14 males and 31 females, aged 13-52 yr, who had had IDDM for 2.5-29 yr there was no strong relationship between either glomerular basement membrane (GBM) thickness or mesangial expansion and duration of IDDM. There was only a weak relationship between the thickness of the GBM and expansion of the mesangium. Thus, GBM thickening and mesangial expansion in IDDM occur at rates that often differ from one another and that vary greatly among patients. The clinical manifestations of diabetic nephropathy, albuminuria, hypertension, and decreased glomerular filtration rate related poorly or not at all to GBM thickening. In contrast, all light and electron microscopic measures of mesangial expansion were strongly related to the clinical manifestations of diabetic nephropathy, although in the absence of these clinical findings, it was not possible to predict the severity of any of the diabetic glomerular lesions. Mesangial expansion had strong inverse correlations with capillary filtering surface area density. It is hypothesized that mesangial expansion could lead to glomerular functional deterioration in IDDM by restricting the glomerular capillary vasculature and its filtering surface. However, capillary closure, glomerular sclerosis, and interstitial fibrosis could also contribute to the clinical manifestations of this disorder.

Since several studies have suggested that a slight increase in urinary albumin excretion (microalbuminuria) is predictive of nephropathy in patients with diabetes mellitus, we studied the relation of albumin excretion to renal structure in patients with insulin-dependent (Type I) diabetes. Renal biopsy specimens were evaluated with light- and electron-microscopical morphometric techniques in 48 patients who had had diabetes for 5 to 40 years and who excreted less than 200 mg of urinary albumin per 24 hours. Patients in Group I (n = 26) had normal urinary albumin excretion, creatinine clearance, and blood pressure; those in Group II (n = 10) had increased urinary albumin excretion but normal creatinine clearance and blood pressure; those in Group III (n = 12) had increased urinary albumin excretion and hypertension, decreased creatinine clearance, or both. Glomerular structure varied similarly, ranging from normal to abnormal in Groups I and II, but was consistently abnormal in Group III. The thickness of the glomerular basement membrane, the fractional volume of the mesangium, and the mesangial volume per glomerulus in Group III exceeded the corresponding values in the other groups significantly. Thus, microalbuminuria, when present with hypertension, decreased creatinine clearance, or both, indicates established abnormalities of glomerular structure. Normal albumin excretion, or microalbuminuria without these other functional abnormalities, does not accurately predict the severity of the underlying glomerular lesions in patients with Type I diabetes.