The prevalence of proteinuria in patients with type 1 DM is between 15% and 40% (9,13).
of the D allele (35). A similar increased risk for ESRD due to diabetic nephropathy was found in a second meta-analysis in Asian patients with type 2 DM and the DD genotype (36). An early study analyzed 115 candidate genes using a transmission/disequilibrium test in patients with type 1 DM and identified 20 genes with polymorphism (37). However, in spite of a large number of studies that tested the association of candidate genes with diabetic nephropathy, only three analyses have found convincing candidate genes with replicate studies, including PKC-β (38), erythropoietin gene promoter (39), and endothelial nitric oxide synthase (40). However, other investigators (41) using a large case-control meta-analysis were unable to confirm an association of the erythropoietin gene promoter as well as several other previously published associations to diabetic nephropathy.
TABLE 21.1 Selected possible susceptibility loci for diabetic nephropathy
provided in a study in which an inhibitor of histone deacetylase was administered to rats with streptozotocin (STZ)-induced DM (61). Kidney growth is one of the first changes seen in this model. The inhibitor reduced epithelial growth factor receptor mRNA and protein, resulting in decreased tubular proliferation. As we learn more about the epigenome, we will be able to determine the role of particular mechanisms.
FIGURE 21.3 Glomeruli showing progression from mild mesangial matrix increase (A) to moderate increase in matrix with early nodule formation at 12 o’clock (B). (PAS, A: ×260, B: ×260.)
advanced mesangial lesions with an increase only of type V collagen in late nodules. Other investigators have demonstrated increases in collagen type 1 as well as the proteoglycans decorin and biglycan (109). More recently, aberrant laminins have been demonstrated in the matrix as the expansion progresses (110). These modifications in the composition and amount of mesangial matrix are due to both increased synthesis and decreased degradation as discussed in the “Pathogenesis” section (109).
FIGURE 21.4 Solidified glomerulus filling the entire Bowman space. Note hyalinosis lesion with lipid droplets (arrow). (PAS, ×350.)
FIGURE 21.5 Ischemic change in the glomerulus, with collagen forming internal to the Bowman capsule. This change is commonly found when arterial disease is advanced. (PAS, ×380.)
FIGURE 21.6 Glomerulus with a single well-developed Kimmelstiel-Wilson nodule at 12 o’clock. Also note uniform thickening of the GBM. (PAS, ×390.)
FIGURE 21.7 Glomerulus from a patient with well-developed nodular diabetic glomerulosclerosis. Several well-formed nodules with cells arranged around the periphery are present with smaller developing nodules present in other lobules. (PAS, ×350.)
FIGURE 21.8 Glomerulus with multiple nodules with laminations. (Periodic acid-methenamine silver, ×260.)
progression of changes has been described wherein the GBM is loosened from its anchoring points as it reflects back over the mesangium (114). This change is accompanied by disintegration of the mesangial matrix (Fig. 21.10), the appearance of fibrillar material, and increasing compaction of the mesangial material resulting in the formation of several layers and a large nodule (Fig. 21.11). Agents implicated in the development of microaneurysms include platelet factors, hemodynamic factors, and possible changes in elasticity of the GBM (107,112,113). Stout et al. (115) suggested a different pathogenetic mechanism for the Kimmelstiel-Wilson nodule. Focal mesangiolysis appears first and progresses from an edematous to a proliferative stage characterized by a loose but organized fibrillar matrix. As this matrix condenses, the lesion changes from focal nodular mesangial expansion to a simple Kimmelstiel-Wilson nodule. Repeated injury then results in a similar progression of changes and the eventual formation of the laminated lesion called the complicated nodule. Paueksakon et al. (112) found increased plasminogen activator inhibitor-1 (PAI-1) in such lesions, particularly those in which they found fragmented red blood cells, an indicator of microvascular injury. They suggested this repeated microvascular injury caused the mesangiolysis.
FIGURE 21.9 Glomerulus with a small microaneurysm in center containing blood elements (arrow) above hilar vessels. Capillary walls are thickened throughout the entire glomerulus. (PAS, ×330.)
FIGURE 21.10 Glomerulus showing disintegration of the mesangial matrix (white arrow) and loss of anchoring points of the GBM (black arrow) from the mesangium. (Periodic acid-methenamine silver, ×450.)
FIGURE 21.11 Portion of the glomerulus with a microaneurysm showing compaction of fibrillar material and mesangial matrix to form a large mesangial nodule. (Periodic acid-methenamine silver, ×400.)
distribution expected for vascular disease; that is, they occur in stripes perpendicular to the capsular surface within the distribution of the affected vessel (126,127).
FIGURE 21.14 Glomerulus that has likely lost its connection to its proximal tubule, although serial sections would be needed to confirm this change. Note marked atrophy of surrounding tubules and wrinkling of glomerular capillary loops. (PAS, ×320.)
to the increase in interstitial volume as well as to decrease in estimated GFR (eGFR) (134). It is now considered that these inflammatory cells play an active role in the pathogenesis of the interstitial fibrosis. The increase in interstitial volume is due largely to increase in cells at early stages of diabetic nephropathy associated with only mild glomerular changes (135). Increases in collagen and other matrix components occur later in the disease (135). The presence of interstitial fibrosis, particularly when accompanied by inflammatory infiltrate, correlates inversely with renal survival (133,136). Lane et al. (137) demonstrated that mesangial expansion, arteriolar hyalinosis, global glomerular sclerosis, and interstitial expansion are interrelated. However, the progression in each compartment is not stereotypic for all patients. Thus, in some patients, the severity of interstitial disease may be greater than that of glomerular lesions, whereas in others, the reverse may be true.
FIGURE 21.16 Light micrograph showing tubules with Armanni-Ebstein lesion characterized by a subnuclear vacuole representing glycogen. (H&E, ×380.) (Courtesy of Dr. Melvin Schwartz.)
varies among individual patients and does not correspond to the severity of the glomerular lesions (144). Linear staining of capillary walls has also been reported with IgM, the third component of complement (C3), fibrinogen, and albumin (141,144). Staining of the mesangium and Kimmelstiel-Wilson nodules is rarer and usually shows fainter staining than that reported for capillary walls (141,144).
FIGURE 21.19 Immunofluorescence micrograph of a glomerulus with a diabetic nodule at 6 o’clock using antiserum to IgG. One sees positive linear staining along capillary walls. (×320.)
FIGURE 21.20 Immunofluorescence micrograph showing linear staining for albumin along the tubular basement membrane. (×400.)
nephropathy concurrent with mesangial expansion. Podocyte depletion has been confirmed in patients with either type 1 or type 2 DM by other investigators (116,118,159,160). Toyoda et al. (161) studied foot process detachment in patients with type 1 DM and found that 22% of GBM was not covered by intact foot processes in diabetic patients with proteinuria as compared to 11% in such patients with microalbuminuria and 4% of diabetics with normoalbuminuria. Weil et al. (159,160) using a more conservative technique found a smaller percentage of foot process detachment in type 2 diabetic patients, but the differences were significant between patients with proteinuria (1.48% detachment) and those with normoalbuminuria (0.41% detachment) or microalbuminuria (0.37% detachment). Such foot process detachment has been associated with glomerular sclerosis in a number of different glomerular lesions and may be important in the progression of diabetic nephropathy as well (161). Furthermore, the visceral epithelial cells play an important role in maintaining endothelial cells so that loss of podocytes may influence these cells as discussed below.
FIGURE 21.21 A: Drawing of normal capillary. B: Drawing of capillary with diabetic changes. C: EM of normal capillary. D: EM of capillary with diabetic changes.
25 years. Toyoda et al. (161) in a study of patients with type 1 DM found that the fractional surface of fenestrated endothelium was reduced from 41% in controls to 25% in patients with proteinuria. Patients with normoalbuminuria or microalbuminuria showed 32% fractional surface of fenestrated endothelium. Weil et al. (159) found similar results in patients with type 2 DM. We now recognize that vascular endothelial growth factor (VEGF) secreted by the podocyte has an important role in control of endothelial structure (163). However, Weil’s study did not find any difference in the degree of fenestration near areas of foot process detachment (159). Satchell suggests several possible explanations, including the finding that actual detachment of foot processes was quite rare (164). Furthermore, he suggests that other factors such as the endothelial glycocalyx may play a role in glomerular permeability in this setting (164,165).
on pathogenesis. Dachs et al. (169) noted that the first change consisted of widening of the usually delicate strands of mesangial matrix with increase of the numbers of mesangial cells. The number of mesangial cells and cellular processes may be slightly increased within the expanded mesangial matrix. These observations have been confirmed by morphometric analysis. Steffes et al. (170) showed increase in volume fraction of mesangial matrix per glomerulus and in mesangial cells per glomerulus in patients with type 1 diabetes compared with controls. The volume fraction of mesangium increases over the duration of the diabetes (147). Cell debris manifested as small calcific deposits, remnants of cell membranes, and scattered organelles are often present (166). On occasion, dense collections of fibrils measuring between 10 and 25 nm are present in the mesangium (171,172) (Fig. 21.25). The presence of such diabetic fibrillosis does not affect the course or prognosis (171,173). The fibrils can be distinguished from amyloid by their lack of staining by Congo red (172). Mesangiolysis has been described in association with the formation of microaneurysms (111,114). Some nodules develop loosening of the matrix, resulting in detachment of endothelial cells and loss of the anchoring points of the GBM to the mesangium (166). These alterations are thought to precede the exaggerated mesangial expansion of the single large nodules with laminated texture that are associated with microaneurysms (113).
FIGURE 21.23 Electron micrograph of a portion of a glomerulus showing a hyalinosis lesion in a capillary loop and marked foot process effacement. (Uranyl acetate and lead citrate, ×4700.)
FIGURE 21.24 Electron micrograph of a Kimmelstiel-Wilson nodule showing increased matrix and relationship to capillary loops with thickened GBM and effacement of foot processes. (Uranyl acetate and lead citrate, ×3000.)
the other membranous glomerulonephritis). Seventeen patients had typical nodular disease with Kimmelstiel-Wilson nodules, more marked GBM thickening, and more pronounced hyaline arteriolosclerosis. The remaining 15 patients had diffuse mesangial sclerosis consistent with diabetic nephropathy but did not have nodules. Mazzucco et al. (182) examined 393 renal biopsies from patients with type 2 DM in a multicenter study. They defined four classes of histologic findings. Class 1 was defined as typical diabetic nephropathy and was present in 40% of the biopsies. Within this group, 13% had predominantly nodular change, 34% had no nodules, and 53% showed a mixed nodular and diffuse pattern of mesangial expansion. Class 2 was characterized by ischemic changes with collapsed sclerotic glomeruli, tubular atrophy, and arterio- and arteriolosclerosis and was seen in 15% of the biopsies. Diabetic changes were not present in these cases. Class 3 was defined by the presence of another glomerular disease and was observed in 45% of the biopsies. Within this group, the glomerular disease was superimposed on diabetic changes in 38% of cases (class 3a). The remaining 62% did not have typical diabetic renal disease (class 3b).
defining feature of diabetic nephropathy and the use of a single Kimmelstiel-Wilson nodule to indicate a more severe lesion as such nodules may be present in the absence of clinical evidence of renal disease (195).
no association between hyperfiltration and albuminuria in patients with type 2 diabetes during the early stages of diabetic nephropathy. This study also showed that the decline of GFR and albuminuria followed a parallel course during the later stages of the disease raising the possibility that they share common pathophysiologic mechanisms. The beneficial effects of good metabolic control to reduce GFR toward normal levels within days to weeks have also been shown in several studies in both type 1 and type 2 diabetes (199,219). Some data seem to indicate strict control of blood glucose levels inhibits the progression of albuminuria; however, it does not offer any difference in clinical kidney disease outcome (220,221,222).
and partially treated with insulin to maintain moderate levels of hyperglycemia develop increased whole-kidney and single-nephron GFR (256). The increment in single-nephron GFR is due to increases in transcapillary hydraulic pressure difference and initial glomerular plasma flow rate (256). When glomerular pressure is lowered by the use of ACE inhibitors, glomerular injury is decreased in this model (259,260), a finding suggesting a role for increased glomerular pressure in this form of injury. ACE inhibitors also reduce the morphologic alterations seen in this model (261). Treatment with insulin normalized glomerular pressure and renal hypertrophy (262,263). However, Bank et al. (264) showed that increased glomerular pressure alone is not sufficient to cause glomerular lesions in diabetic rats. Some of the glomerular hemodynamic changes have been attributed to the effects of diabetes on afferent arteriolar resistance resulting in vasodilation. However, the precise mechanisms for the changes in glomerular hemodynamics remain undefined. Reports on nitric oxide (NO) as a potential mediator of vasodilation and hyperfiltration in diabetes are conflicting (265,266,267). Additional suggested mechanisms include increase in glomerular synthesis of prostaglandins (268,269), alterations in the intrarenal renin-angiotensin axis (270), changes in the polyol pathway (271), diminished bioavailability of proinsulin connecting peptide (C-peptide) (272), open KATP channels affecting preglomerular microvascular smooth muscle function (273), altered tubuloglomerular feedback (274,275), and increased production of IGF-1 (258). Hypertrophy and hyperfiltration are reduced in STZ-induced diabetes via modulation of IGF-1 bioactivity by inhibiting nitric oxide synthase supporting the role of IGF-1 mediating hyperfiltration (258). Some data also suggest that these events might be further modulated by proinflammatory mediators and cytokines (276).