of the glomerulus (diabetic glomerulosclerosis). It begins with hyperfiltration followed by proteinuria, hypertension, and progressive renal failure. The pathogenesis and clinical course of diabetic nephropathy are better known in type 1 than in type 2 diabetes. They appear to be similar but not identical.
Epidemiology
The prevalence of microalbuminuria or any more advanced stage of nephropathy in IDDM increases during the first 20 years after diagnosis (or, in children, after puberty) to > 50% and levels off thereafter. During this time only about half of the microalbuminuric subjects will develop macroproteinuria and only a minority will develop end-stage renal disease [176]. The cumulative incidence of persistent macroproteinuria is about 35% in both IDDM and NIDDM. However, endstage renal disease after 30 years of diabetes is more often present in IDDM (>20%) than in NIDDM (10%) [177–179]. About 30–50% of all patients on chronic dialysis have diabetes (which does not imply that the reason is always the diabetes). Since type 2 diabetes is much more frequent than type 1 diabetes, it contributes the majority of these subjects.
The incidence and progression of diabetic nephropathy are related to metabolic control [127,130–133] and blood pressure [158,159]. A threshold phenomenon was postulated [180], but was rejected [181,182]. Mortality is high in subjects with diabetic nephropathy. Microalbuminuria has been identified as a strong predictor of cardiovascular disease.
In proteinuric subjects coronary artery disease is about 15 times more frequent than in diabetic subjects without proteinuria [154], and cardiovascular mortality is increased nine-fold [183,184]. Ten-year survival of subjects with persistent proteinuria used to be only 20-50% [179,185]. Antihypertensive treatment and renal replacement therapy have effectively improved the prognosis. In recent studies, eight-year survival after the onset of persistent proteinuria rose to 70-87% [176,186,187].
Before renal replacement therapy was available, the main cause of death in subjects with proteinuria was uremia. About 25% died from myocardial infarction or stroke. Renal replacement therapy reduced deaths from uremia but increased deaths from cardiovascular causes [188].
Pathology
As in diabetic retinopathy, there is no doubt about the influence of hyperglycemia on the development of diabetic nephropathy. However, the causal relationship is less clear. There is no linear relation between the cumulative incidence of any sign of diabetic nephropathy and the duration of diabetes, and more than half of diabetic subjects never develop such nephropathy [176,189]. Normal kidneys transplanted into diabetic recipients may develop typical lesions, but the rate of development varies and is independent of metabolic control [155]. These observations suggest that hyperglycemia is necessary but not sufficient for the development of diabetic nephropathy. Other important pathogenetic factors are hypertension, protein intake, renal hemodynamics, smoking, and genetics.
The hyperglycemia-related pathogenetic effects discussed on pages 11 and 12 are also found in diabetic nephropathy. In addition, synthesis of the glucosaminoglycan heparan sulfate and glycoproteins is impaired [136]. These molecules contribute to the negative charge of the glomerular capillary membranes and are involved in the selectivity of glomerular filtration, which consequently may be reduced. Hemodynamics are another important pathogenetic factor. The impact of hypertension on diabetic nephropathy has been shown in epidemiological and antihypertensive treatment studies [156–159,190,191]. A significant example of direct jeopardizing of the kidney by hypertension is seen in people with unilateral renal artery stenosis, where only the kidney with the patent artery develops glomerulosclerosis [155].
Evidence for a genetic influence comes from family studies. Siblings of probands with nephropathy develop signs of nephropathy several times more often than do siblings of probands without nephropathy [192,193]. Recently epidemiological studies showed an increased incidence of diabetic nephropathy at level of a protein intake exceeding 20% of total energy [194], suggesting a pathogenetic role of nutritional protein.
The clinical picture of diabetic nephropathy is dominated by functional disorders, which may be classified according to Mogensen [195] (Table 1.12). The functional disorders correspond to morphological changes. The early increase in glomerular filtration rate has been explained by the ubiquitously increased blood flow and peripheral vasodilation. It correlates to increased kidney size, glomerular volume, and capillary filtration surface area.
Microalbuminuria develops without apparent morphological changes. It seems to be caused by increased glomerular capillary pressure and a loss of negative charge of the glomerular basement membrane. When the pores of this membrane enlarge, filtration selectivity is lost, and (macro-)proteinuria develops. With mesangial expansion due to continuous deposition of indigestible matrix proteins (formation of AGE on collagen, laminin, fibronectin) and thickening of the endothelial layer, vascular obstruction will occur, which results in a decrease of the filtering area. Histological studies show diffuse or nodular glomerulosclerosis [196–198]. In this situation blood pressure increases, glomerular filtration rate decreases, and progressive renal failure with end stage renal disease will develop.
Management
It is essential to detect diabetic nephropathy at a reversible stage. At Mogensen's stages 1-3 (Table 1.12) the disorders are reversible and renal function may be kept normal if effectively treated. In stages
