glomerulosclerosis, which is reflected in the four classes, as discussed further below. Although this system has not been validated with clinical outcomes, it serves as an important clinical and research tool to classify the severity of diabetic nephropathy lesions. Class I Diabetic Nephropathy On ultrastructural evaluation of the kidney histology, among the earliest change that occurs in the kidney is thickening of the GBM; light microscopy shows minimal, non-specific, or no changes. Thickening of the GBM does not directly correlate with clinical injury. Patients may have such thickening but have no increase in urine albumin excretion rate or impairment of glomerular filtration rate (39,40). Although an increase in diastolic blood pressure (40) or nocturnal blood pressure (39) is correlated with GBM thickening, the causal relation is not established because of a lack of longitudinal data and interventional studies. GBM thickening occurs as a result of either an increased rate of deposition or a reduced rate of removal of connective tissue. Target molecules include collagen IV and VI, fibronectin, and laminin (35,41). Class II Diabetic Nephropathy Among the earliest manifestations on kidney histology that correlate with kidney damage is an increase in mesangial matrix, as seen in class II diabetic nephropathy. Class II is further subclassified based on the degree of mesangial expansion; class IIa is characterized by ≤25% mesangial expansion, and class IIb involves >25% of the mesangial expansion. An increase in mesangial matrix, glomeruli, and kidney volume is clinically manifested as kidney enlargement; kidneys are often 11 cm or larger on kidney ultrasound. Urine albumin excretion is often increased in these patients. Class III Diabetic Nephropathy An increase in mesangial matrix is followed by mesangial sclerosis. The hallmark lesion on a kidney biopsy is nodular glomerulosclerosis, or Kimmelstiel-Wilson nodules. The presence of Kimmelstiel-Wilson nodules on kidney biopsy correlates with the occurrence of diabetic retinopathy, suggesting activation of common pathogenetic pathways such as VEGF. Class IV Diabetic Nephropathy Advanced, or class IV, diabetic nephropathy is characterized by sclerosis in >50% of the glomeruli. These patients often have a loss of kidney function at the time of biopsy. An enlargement of glomeruli is often seen along with thickening of the walls of the glomerular capillaries. Arteriolar hyalinosis of both the afferent and efferent arteriole should alert health care professionals to the possibility of diabetic nephropathy. The proximal tubules can contain protein resorption droplets. In the setting of severe persistent hyperglycemia, glycogen deposits may be seen rarely in the proximal tubules (i.e., Armanni Ebstein lesion). Interstitial fibrosis and tubular atrophy (IFTA) and interstitial inflammation are often seen. Despite tubular atrophy, the basement membranes are often thickened in patients with diabetes. The Heterogeneity of Kidney Injury in Type 2 Diabetes: A Pathogenetic Explanation Although kidney disease is histologically similar in type 1 and type 2 diabetes, the relative contributions of causes of kidney damage differ in these two conditions. Compared to patients TABLE 2 Pathological Classification of Diabetic Nephropathy Class I ⊲ GMB thickening on electron microscopy; minimal, non-specific, or no changes on light microscopy Class II ⊲ Increase in mesangial matrix Class IIa ⊲ Mesangial expansion ≤25% Class IIb ⊲ Mesangial expansion >25% Class III ⊲ Nodular glomerulosclerosis: Kimmelstiel-Wilson lesion Class IV ⊲ Advanced glomerulosclerosis; >50% glomeruli sclerotic CHRONIC KIDNEY DISEASE AND TYPE 2 DIABETES 7 with type 1 diabetes, those with type 2 diabetes are older, have a greater BMI, and are more likely to have dyslipidemia, hypertension, and other cardiovascular risk factors and, consequently, atherosclerosis and arteriosclerosis. Thus, the nature of kidney injury in patients with type 2 diabetes may be modified by environmental factors and genetic background. This heterogeneity in environmental and genetic factors in patients with type 2 diabetes may explain the distinct kidney injury phenotypes. As an example, consideration of an animal experiment provides evidence for interplay between genetics and environment with regard to kidney injury phenotype (42). Progeny of rats with one parent with heart failure and another with obesity were fed a diet either high in carbohydrate or high in fat; all progeny had diabetes (42). Compared to animals fed a high-carbohydrate diet, animals fed a high-fat diet demonstrated a greater preponderance of tubulointerstitial injury and non-nodular glomerulosclerosis. There was evidence of lipid peroxidation and increased kidney TGFβ1 that correlated with kidney injury. Furthermore, injury in animals fed a high-fat diet was seen in the arterial wall and renal microcirculation. In contrast, animals fed a high-carbohydrate diet had increased glycoxidation stress biomarkers, but these did not correlate with kidney injury (42). Conclusion The pathogenesis of diabetic nephropathy is similar in type 1 and type 2 diabetes. Diabetic nephropathy is classified histologically by the appearance of the glomerulus on kidney biopsy. It progresses from GBM thickening, to mesangial expansion, nodular glomerulosclerosis, and global glomerulosclerosis.