Increased blood glucose concentration is the clinical hallmark of diabetes and is a cardinal determinant of its complications. Given the prominent role of hyperglycemia in the development of diabetic complications, it is not surprising that the duration of diabetes is one of the most important risk factors for DKD, and its influence is far greater than that of age, sex, or type of diabetes. ^{, ,} Furthermore, most of the large observational studies indicate that hyperglycemia is an important predictor of progression to advanced CKD in both type 1 and type 2 diabetes. For instance, in type 1 diabetes, poor glycemic control independently predicts progression to development of albuminuria and kidney failure, and in type 2 diabetes higher hemoglobin A1c (HbA1c) levels are one of the strongest determinants of DKD progression independent of albuminuria.

Early evidence for the role of hyperglycemia in the development of DKD comes from biopsy studies in identical twins discordant for type 1 diabetes and from morphologic studies before and after pancreas transplantation. ^, Glomerular changes including widened glomerular and tubular basement membranes and increased mesangial fraction are identified only in the diabetic member of twin pairs, suggesting that metabolic status, and not genetic predisposition, is responsible for the development of these diabetic kidney lesions. In addition, prolonged normoglycemia following pancreas transplant in persons with type 1 diabetes and established DKD promotes virtually complete reversal of glomerular and tubular basement membrane thickness and increases in mesangial and interstitial volumes. Nevertheless, the relative contributions of hyperglycemia, insulin resistance (IR), and alterations in endogenous insulin production to the development and progression of DKD remain unclear.

Further evidence that supports the role of early control of hyperglycemia comes from the DCCT/ Epidemiology of Diabetes Interventions and Complications (EDIC) study and UKPDS. Both studies demonstrated a beneficial effect of more rigorous glycemic control early in the course of diabetes, using well-delineated sets of glucose-lowering medications and interventions available at the time of the studies, respectively, for both T1DM (EDIC) and T2DM (UKPDS). Remarkably, the greater preservation of GFR and reduction in DKD in the intervention groups in both studies remained statistically significant despite dissipation and loss through much of the subsequent duration of the studies of the impressive glycemic control differences (with HbA1c as one of the important measures of metabolic control) achieved in the early years after enrollment. In light of these results, it was hypothesized that the observed beneficial effect of early strict glycemic control, termed the “legacy effect” or “metabolic memory,” could be explained by prevention of metabolism-driven epigenetic modifications. The mechanistic role of epigenetic alterations in DKD is supported by the observation that periods of uncontrolled hyperglycemia are associated with an elevated risk of developing DKD that does not reflect real-time HbA1c levels. ^, The “legacy effect” may, at least in part, explain discrepant results from different observational studies evaluating the predictive value of HbA1c for DKD.

Persistent hyperglycemia causes dysregulation of several effector molecules through various biochemical pathways in the kidney, including generation and accumulation of advanced glycation end products, increased activity of the polyol pathway, and activation of vasoactive hormones, such as angiotensin II and endothelin. Activated prosclerotic cytokines, such as transforming growth factor beta (TGF-β) and vascular endothelial growth factor (VEGF), are important mediators between the metabolic and hemodynamic pathways affected by hyperglycemia that ultimately lead to the pathologic changes of DKD.

A cross-sectional study of 100 patients with type 2 diabetes found that albuminuric patients had lower glucose disposal rates than nonalbuminuric patients. The clinical evidence that IR precedes DKD came from prospective studies in the early 1990s. A prospective analysis of 108 patients with type 2 diabetes and normoalbuminuria who underwent euglycemic clamps at enrollment demonstrated that hyperinsulinemia at baseline was strongly associated with albuminuria at the 5-year examination. A larger prospective analysis of 582 nondiabetic siblings of patients with type 2 diabetes also demonstrated that IR at baseline (assessed with glucose tolerance tests and euglycemic clamps) correlated with the development of albuminuria. Interestingly, in patients with type 1 diabetes or their family members, glucose disposal rates are also reduced in the presence of albuminuria, ^, and in a small longitudinal study of patients with type 1 diabetes and normoalbuminuria, IR predicted development of albuminuria after 3 years. One of the proposed mechanisms is hyperinsulinemia-mediated vasodilation of the afferent arteriole contributing to glomerular hyperperfusion and hyperfiltration ( Fig. 41.4 ).

Normal and diabetic nephron with altered hemodynamics.

(A) Afferent vasodilation is promoted by hyperglycemia, hyperinsulinemia, elevated levels of circulating amino acids, COX-2 prostanoids, and reduction of tubule glomerular feedback. Tubuloglomerular feedback is a kidney intrinsic autoregulatory mechanism, which helps regulate the rate of glomerular filtration rate. Because of increased reabsorption of glucose and sodium via SGLTs in diabetes, sodium chloride delivery to macula densa cells of juxtaglomerular apparatus is decreased, resulting in lower production of adenosine and consequent relative vasodilation of afferent arteriolar. (B) Efferent vasoconstriction is promoted by high local angiotensin II levels, endothelin I, reactive oxygen species, and thromboxane A2.

Hypertension is common in patients with DKD. In type 1 diabetes, elevation of blood pressure is typically coincident with the appearance of albuminuria, reflecting higher blood pressure as a consequence of DKD. Conversely, other studies demonstrate the predictive value of high blood pressure for DKD in type 1 diabetes. Familial aggregation of blood pressure is widely reported and may represent a predisposition to hypertension. Higher blood pressure in the nondiabetic parents of patients with type 1 diabetes and albuminuria suggests that this predisposition to hypertension is also a risk factor for DKD in type 1 diabetes. In one study, maternal blood pressure was more strongly related to albuminuria in the offspring than paternal blood pressure, suggesting a dominant effect of maternal genes or an effect of the intrauterine environment on the risk of microalbuminuria. In addition, increased nocturnal blood pressure predicts the development of microalbuminuria in adolescents and young adults with type 1 diabetes.

In type 2 diabetes, the onset of hypertension generally precedes the first clinical evidence of DKD or even the diagnosis of diabetes. In Pima Indians, blood pressure is measured at sequential research examinations that begin long before the onset of diabetes, and higher blood pressure before the onset of type 2 diabetes was strongly associated with albuminuria after the onset of diabetes. In addition, the prevalence of proteinuria was significantly higher if both parents had hypertension than if only one or neither parent had hypertension. Increased nocturnal blood pressure is associated with microalbuminuria in type 2 diabetes, as it is in type 1 diabetes, and a baseline systolic blood pressure >140 mm Hg has been shown to increase the risk for kidney failure and death. In patients with newly diagnosed type 2 diabetes, each 10 mm Hg increase in mean systolic blood pressure was associated with a 13% ( P < 0.0001) increase in the hazard ratio for development of microalbuminuria and macroalbuminuria, eGFR <60 mL/min/1.73 m ² , or doubling of the plasma creatinine level after a median follow-up of 15 years. Treatment to a target blood pressure of <150/85 mm Hg compared with treatment to a target blood pressure <180/105 mm Hg resulted in a significant reduction in overall risk of microvascular complications (37%, P = 0.009) in the UKPDS. In contrast to the legacy effects observed with intensive glycemic control, benefits of blood pressure control are transient and are lost if blood pressure rises again.

Many of the abnormalities in plasma lipoproteins associated with CKD are sequelae of kidney dysfunction, but dyslipidemia may also play a role in the pathogenesis of glomerular injury. Persons with diabetes and CKD typically have significant serum hypertriglyceridemia, elevated low-density lipoprotein (LDL), and low high-density lipoprotein (HDL) cholesterol concentrations. These abnormalities are also more pronounced in persons with macroalbuminuria than in those with microalbuminuria. Besides quantitative changes, lipid particles in persons with diabetes change qualitatively, as LDL and HDL particles tend to be smaller and denser with advancing CKD.

Among 439 patients with type 1 diabetes and proteinuria who were followed at the Joslin Clinic, elevated serum cholesterol concentration was a strong predictor of a rapid loss of kidney function. Specific lipids or lipid profiles may also predict progression of DKD. In 152 patients with type 1 diabetes from the United Kingdom and Finland, elevated serum LDL cholesterol concentration predicted progression of DKD during 8 to 9 years of follow-up, as defined by a doubling of albuminuria or a decline in creatinine clearance >3 mL/min/year. Higher triglyceride content of very low-density lipoprotein (VLDL) and intermediate-density lipoprotein (IDL) particles predicted worsening of microalbuminuria, whereas smaller LDL size was associated with declining kidney function in persons with macroalbuminuria at baseline. The predictive value of these lipid profiles, however, has not been consistently observed in other studies and may conceivably reflect overall reduced access or adherence to beneficial lifestyle, dietary and medical intervention, and follow-up. ^,

In type 2 diabetes, the UKPDS reported that higher serum LDL cholesterol concentration increased the risk for macroalbuminuria. In a post hoc analysis of 1061 persons with type 2 diabetes in the Reduction of Endpoints in Non-insulin dependent diabetes with the Angiotensin II Antagonist Losartan (RENAAL) trial, the risk of kidney failure was 32% higher for each 50 mg/dL increase in LDL cholesterol concentration and 67% higher for each 100 mg/dL increase in total cholesterol concentration. Lowering LDL cholesterol concentrations associated with the 1-year risk of kidney failure, although concurrent treatment with losartan likely contributed to this improvement. Elevated plasma triglycerides are also associated with increased risk for both albuminuria and kidney failure in type 2 diabetes. Triglyceride content within lipoprotein subclasses, particularly VLDL, correlates with albuminuria in youth with T1D. In addition, low concentrations of HDL cholesterol predict increased risk of albuminuria progression in persons with type 2 diabetes and microalbuminuria. In the UKPDS, each mmol/L decline in HDL cholesterol increased the risk of doubling of serum creatinine concentration nearly threefold during a median follow-up of 15 years. Interestingly, HDL proteomic analysis in patients with kidney failure and CKD has revealed a unique proteomic profile. ^,

Dyslipidemia may contribute to onset and progression of DKD through mechanisms like those responsible for atherogenesis. ^, Hypercholesterolemia may impair the kidney’s hemodynamic responses and tubular function by decreasing nitric oxide production and/or increasing superoxide activity in the kidney, with resulting antidiuretic and antinatriuretic effects. Although not directly affecting GFR, these actions may play a role in the development of systemic hypertension associated with diabetes. Oxidized LDL and free fatty acids can cause structural and functional damage to podocytes by inducing mitochondrial dysfunction and accumulation of reactive oxygen species, suggesting a causal role in the development and progression of proteinuria. Nevertheless, despite much investigation and experimental evidence supporting various mechanisms, a definitive role for dyslipidemia in the development and progression of DKD in humans remains to be established.

Besides the role of circulating lipids, the diabetic environment facilitates glomerular production of triglycerides and cholesterol, which appear to cause kidney injury both directly by accumulating in the cellular and extracellular structures and indirectly by stimulating the expression of prosclerotic, proliferative, and proinflammatory cytokines. ^{, ,} Accumulation of lipid droplets containing triglycerides and cholesterol has been described in kidney cells in experimental diabetes. These lipid droplets are associated primarily with increased expression of sterol regulatory element-binding protein (SREBP)-1c, decreased expression of peroxisome proliferator-activated receptor (PPAR)-α, and decreased expression of liver X receptor (LXR)-α, LXR-β, and ATP-binding cassette transporter-1 ( ABCA1 ). Under normal conditions, ABCA1 mediates the efflux of cholesterol to lipid-poor apolipoproteins (primarily Apo A1) to form HDLs. Down regulation of glomerular expression of ABCA1 was observed in glomerular transcripts from persons with DKD when compared with normal controls and was correlated with lower eGFR. The downregulation of ABCA1 was also associated with lipid droplet accumulation in podocytes, ^, which in turn was found to cause cardiolipin-dependent mitochondrial dysfunction. Recent studies have also challenged the relative contribution of free or esterified cholesterol to podocyte injury and proteinuria in experimental DKD. A major effort was made to identify lipid content by magnetic resonance imaging (MRI), as fat content may inversely relate with eGFR and predict response to treatment strategies. If and how prevention of lipid accumulation in the kidney parenchyma may affect DKD development and progression remains to be established. In this regard, the development of drugs that target Oxysterol Binding Protein Like 7 to increase cholesterol efflux, which have been shown to slow experimental CKD progression, may hold promise for clinical research.

In experimental models, excessive protein intake causes kidney vasodilation and glomerular hyperperfusion with a resulting increase in intraglomerular pressure that leads to glomerular damage. ^, Long-term high-protein intake accelerates structural and functional injury in experimental models of DKD, whereas low-protein diets offer kidney protection. Physiologic studies in humans confirm a greatly augmented glomerular hyperfiltration in response to an amino acid infusion mimicking a high-protein diet among persons with type 1 diabetes or type 2 diabetes, but only in the presence of chronic hyperglycemia ( Fig. 41.4 ). ^, Dietary protein of animal origin may also be a significant source of advanced glycation end products, which could accelerate DKD progression. Meta-analyses of clinical trials that examined effects of dietary protein restriction on DKD have shown reduction in albuminuria but were inconclusive about impacts on kidney function with highly variable results between studies.

The U.S. Centers for Disease Control (CDC) categorizes obesity in adults according to body mass index (BMI as kg body weight/ height m ² ) into class 1: BMI of 30 to <35; class 2 BMI of 35 to <40; class 3: BMI of ≥40. More than one-third of adults and 17% of youth in the United States are obese. Nearly 6% of youth have class III obesity, exceeding 120% of the sex-specific 95th percentile on the CDC BMI-for-age growth charts in youth. Obesity is a risk factor for CKD, independent of diabetes and hypertension. Indeed, the presence of severe obesity increases the risk of kidney failure sevenfold relative to normal weight. ^, Obesity-associated hemodynamic changes in GFR and renal plasma flow appear to be responsible for the kidney damage. ^, The effects of adiposity may also impact the kidneys directly through the production of various adipokines, including leptin, adiponectin, and resistin, which activate profibrotic, proliferative, and proinflammatory mechanisms. A retrospective clinical and histopathologic study of 6818 native kidney biopsies found diabetes-like lesions in 45% of persons with obesity-related glomerulopathy. The prevalence of obesity-related glomerulopathy, defined as glomerulomegaly with or without focal segmental glomerulosclerosis, increased 10-fold between 1986 and 2000, attributed to the rising prevalence of obesity. Although obesity-related glomerulopathy may represent a distinct entity from DKD, obesity is largely responsible for the increase in diabetes prevalence worldwide, and similarities in underlying mechanisms of obesity-related CKD and DKD suggest that treatments targeting obesity may slow the progression of both forms of kidney disease.

Among women with normal kidney function, regardless of the presence or absence of diabetes, pregnancy is associated with a rise in GFR of about 50% that persists through the 37th week of gestation and is accompanied by a moderate increase in urinary protein excretion. Among pregnant women in Austria with type 1 diabetes, those with normoalbuminuria experience a 3.8-fold increase in albuminuria, whereas those with microalbuminuria (>15 mcg/min) have a 6.7-fold increase. Nevertheless, the albuminuria levels in both groups typically normalize within 12 weeks after delivery. Much less is known about changes in albuminuria in pregnant women with type 2 diabetes, but the prevalence of albuminuria in both types of diabetes in a Danish cohort is equivalent, and pregnancy outcomes are comparable.

Although neither pregnancy nor parity adversely affects the course of early kidney disease in women with diabetes, ^, those with poor glycemic control, hypertension, or preexisting CKD are at increased risk of pregnancy complications and subsequent deterioration in kidney function. ^, In addition, mothers with CKD are more likely to develop preeclampsia and have offspring with low birth weight. In Finnish women with type 1 diabetes, a prior history of preeclampsia was associated with 7.7-fold higher odds of subsequent DKD than in those with normotensive pregnancies. Preeclampsia is also associated with podocyte injury and detachment and could conceivably accelerate diabetic glomerular injury in a pregnant woman with diabetes who develops preeclampsia.

Fetal exposure to an abnormal intrauterine environment has lasting effects on anthropomorphic and metabolic development that lead to increased risk of CKD later in life. Certain adverse intrauterine exposures, such as a shortage of maternal fuels, various drugs, vitamin A deficiency, and maternal diabetes, directly affect development of the fetal kidneys and reduce nephron number in the offspring, an effect that may be mediated in part via epigenetic changes from the perturbed metabolic milieu with durable effects on fetal kidney development and integrity. With the resulting reduction in filtration surface area, the kidney’s adaptive capacity in response to insults, such as diabetes, is reduced, thereby enhancing the risk of CKD. Impaired nephrogenesis may be passed on to subsequent generations through changes in epigenetic gene regulation.

Low birth weight appears to affect DKD risk in both type 1 and type 2 diabetes. In a case-control study of 184 Danish patients with type 1 diabetes, 75% of the women below the 10th percentile of birth weight had persistent macroalbuminuria (≥300 mg/24 hours) compared with 35% of those above the 90th percentile. No relationship was found, however, between birth weight and urine albumin excretion in the men. In Pima Indians with type 2 diabetes, the prevalence of elevated urine albumin excretion (albumin-to-creatinine ratio ≥30 mg/g) was twice as high in both men and women of low birth weight than in those of normal birth weight and was three times as high in those of high birth weight. Two-thirds of those in the high birth weight category were the offspring of diabetic mothers. Irrespective of birth weight, preterm birth also appears to impose an additive risk for progression of kidney disease and such information should be collected as part of a patient’s health history.

Fuel-mediated alterations in maternal metabolism in pregnancies complicated by diabetes may preferentially harm poorly replicating, terminally differentiated cells, such as those found in the nephron. A proposed mechanism for this differential apoptosis during nephrogenesis is activation of the nuclear factor NF-κB signaling pathway. The adverse effect of hyperglycemia on nephrogenesis was demonstrated experimentally by exposing pregnant rats to diabetes and counting nephrons in the offspring. The number of nephrons was reduced by up to 35% in those exposed to diabetes, and even minor elevations of glucose concentration were associated with impaired metanephros development. In adult offspring of women with type 1 diabetes who were exposed to maternal diabetes in utero, kidney functional reserve was significantly reduced relative to those who were not exposed, likely reflecting reduced nephron mass because of this exposure. In Pima Indians with type 2 diabetes, the odds of elevated urine albumin excretion were nearly fourfold higher in those exposed to diabetes in utero than in those who were not exposed. Moreover, exposure to diabetes in utero among Pima Indian offspring increased nearly fourfold over a 30-year period, paralleled by a doubling in the prevalence of childhood-onset diabetes attributable to this exposure. These data point to in utero diabetes exposure as a contributor to the rising prevalence of childhood and youth-onset type 2 diabetes. Intrauterine exposure to diabetes was also associated with a fourfold increase in the age-sex-adjusted incidence of kidney failure in young adults with type 2 diabetes, mediated largely by the younger age at onset of diabetes. These observations have potential relevance in the developing world and in disadvantaged groups in developed countries, where a higher proportion of type 2 diabetes develops during the childbearing years. The greater frequency of diabetes during the childbearing years in these populations means that the offspring are more likely to be exposed to a diabetic intrauterine environment and to suffer consequences of that exposure.

The role of smoking in promoting DKD is unclear, but substantial evidence supports an effect on albuminuria, although a relationship with progressive loss of GFR is less clear. ^, For instance, smoking increased the risk of albuminuria after 5 years of follow-up in a population-based cohort of 3667 Swedish persons with type 2 diabetes who had no kidney disease at baseline, but it was not associated with GFR decline to <60 mL/min/1.73 m ² during follow-up. In the Finnish Diabetic Nephropathy Study, the 12-year cumulative risk of macroalbuminuria and kidney failure in 3613 patients with type 1 diabetes were 14.4% and 10.3% for current smokers, 6.1% and 10.0% for ex-smokers, and 4.7% and 5.6% for nonsmokers.

Nicotine induces podocyte apoptosis through increasing oxidative stress, promotes mesangial cell proliferation and extracellular matrix production, and increases transforming growth factor (TGF)-β and fibronectin expression in experimental models. Given that persons with diabetes already have widespread vascular damage because of their diabetes, smoking only accelerates this damage. The long-term risks of electronic cigarettes are also of concern, as possible nephrotoxicity of e-cigarette refill liquid was observed in rat kidneys.

Periodontal disease is an inflammatory condition, which often occurs in the absence of diabetes, but is also a frequent complication of diabetes, contributing to poor glycemic control, low-grade chronic systemic inflammation, and increased risk of macrovascular and microvascular complications. ^, Among patients with CKD, periodontal disease substantially increases the risk of premature mortality. In a Swedish case-control study of 78 patients with type 1 or type 2 diabetes, those with severe periodontitis, based on alveolar bone loss, had a higher frequency of dipstick-positive proteinuria and cardiovascular complications after 6 years than those with mild periodontal disease. Severity of periodontitis by alveolar bone loss and being edentulous predicted both albuminuria and kidney failure in a dose-dependent manner among 529 Pima Indian adults with type 2 diabetes who were followed for a median of 9 years. A study investigating the relationships among diabetes, periodontal disease, and CKD in 11,211 adults from the NHANES 1988–1994 population suggested a bidirectional relationship between CKD and periodontal disease. Periodontitis increased the risk of CKD, while CKD also increased risk of periodontitis. Although the mechanisms linking periodontal disease to kidney damage remain to be established, serum lipopolysaccharide activity, which is induced by bacterial infections, is associated with the progression of DKD in Finnish patients with type 1 diabetes.

Control of periodontal infection in diabetic adults improves A1c level and reduces the concentration of various markers of inflammation, coagulation, and adhesion. Whether such control also reduces the onset or progression of DKD is not known.

Abnormalities in heart rate control and vascular dynamics associated with diabetes are referred to as cardiac autonomic neuropathy and are the consequence of damage to and loss of the small unmyelinated nerve fibers that innervate the heart and blood vessels. Advanced cardiac autonomic neuropathy may present as resting tachycardia and orthostatic hypotension. The presence of cardiac autonomic neuropathy is a risk factor for DKD in several longitudinal studies of type 1 and type 2 diabetes. Among 35 Swedish patients with type 1 diabetes, those with cardiac autonomic neuropathy had a significant decline in GFR measured by 51Cr-EDTA clearance over a 10-year period, whereas those without cardiac autonomic neuropathy experienced almost no change. Cardiac autonomic neuropathy was strongly associated with both early GFR loss and progression to CKD, defined by estimated GFR <60 L/min/1.73 m ² , in a subset of the First Joslin Kidney Study, which included 204 normoalbuminuric patients with type 1 diabetes and 166 with microalbuminuria who were followed for a median of 14 years. Similar associations were reported in a multiethnic cohort of 204 adults with type 2 diabetes from the United Kingdom, in whom estimated GFR declined more rapidly over 2.5 years in those with cardiac autonomic neuropathy (9.0% decline) than in those without (3.3% decline) and in a cohort of 1117 Korean patients with type 2 diabetes and estimated GFR ≥60 mL/min/1.73 m ² at baseline, in whom the presence of cardiac autonomic neuropathy increased the risk of developing CKD by over 2.6-fold during nearly 10 years of follow-up. A cross-sectional study in 63 Pima Indians who underwent research kidney biopsies demonstrated an association between cardiac autonomic neuropathy and the structural lesions of early progressive DKD.

Familial clustering of DKD, as well as population ancestry and ethnic differences in disease susceptibility, suggest a genetic predisposition with a 2.1- to 2.3-fold risk for DKD in type 1 diabetes siblings with DKD. Narrow-sense DKD heritability, defined as the proportion of phenotypic variance explained by additive effects of genotyped SNPs, ranged between 24% and 59% in unrelated individuals with type 1 diabetes from the FinnDianne Cohort. ^, Similar analyses of individuals with type 2 diabetes suggested only 8% to 25% heritability, which may potentially reflect that more heterogeneous mechanisms contribute to DKD in these patients and possibly a more important contribution of environmental factors.

Several candidate genes have been identified that may be related to DKD. One of the most intensively studied candidate genes is the insertion/deletion (I/D) polymorphism of the ACE gene ( ACE /ID). Among 168 Japanese patients with type 2 diabetes who were followed for 10 years, analysis of the time course of the three ACE genotypes indicated that patients with the DD genotype were more likely to progress to kidney failure during follow-up than those with the other genotypes, and they had higher mortality once dialysis was initiated. The deleterious effect of the D allele on kidney function was subsequently confirmed in patients from other patient groups with type 2 diabetes and in type 1 diabetes. ^, Presence of the D allele was also associated with more severe structural lesions in type 2 diabetes and with greater progression of structural lesions in type 1 diabetes. In the RENAAL trial, the presence of the D allele increased the likelihood of reaching the primary endpoint of doubling of baseline serum creatinine concentration, kidney failure, or death in the placebo group, but treatment with losartan mitigated that risk, suggesting that losartan had its greatest effect in patients with the D allele. Another candidate gene related to the renin-angiotensin system was identified by the Bergamo NEphrologic Diabetes Complications Trial (BENEDICT), which found that the Pro618Ala polymorphism of the ADAMTS13 gene was associated with a higher risk of CKD progression in patients with type 2 diabetes, but patients with this polymorphism also had a better response to ACE inhibitors.

Kidney failure is considered by some investigators to be the optimal phenotype of DKD in genetic association studies, ^, so the present discussion focuses on this phenotype. Genome-wide association studies (GWASs) aim to identify associations between a genotype and a phenotype. While there is evidence of heritability of DKD, only a few loci associated with DKD, albuminuria, or eGFR in subjects with diabetes have been identified. In a GWAS study of pooled genomic data, the plasmacytoma variant 1 (PVT1) gene was identified as a potential susceptibility locus for kidney failure in the Pima Indians and this association was subsequently confirmed in the Genetics of Kidneys in Diabetes (GoKinD) study among persons of European descent with type 1 diabetes. A GWAS study also involving the GoKinD cohort and confirmed in the DCCT/EDIC cohort, found SNPs in the FERM domain-containing protein 3 (FRMD3) gene and near the cysteinyl-tRNA synthetase (CARS) gene associated with DKD, defined by overt proteinuria or kidney failure. A GWAS meta-analysis of 6691 patients with type 1 diabetes and DKD performed by the GENIE consortium (GEnetics of Nephropathy: an International Effort) identified two SNPs in the AFF3 and an SNP between the RGMA and MCTP2 genes associated with kidney failure. Thereby, AFF3 seems to contribute to renal tubule fibrosis via the TGF-β1 pathway. The strongest association with DKD as a primary phenotype was seen for an intronic SNP in the ERBB4 gene (rs7588550). A subsequent analysis in the Joslin Study of Genetics of Nephropathy in Type 2 Diabetes found that susceptibility loci near CARS were common to both types of diabetes. The Family Investigation of Nephropathy and Diabetes (FIND) conducted a genome-wide association study in 6197 European American, African American, American Indian, and Hispanic American families in whom type 2 diabetes was the predominant cause of kidney disease. Index cases all had diabetes for more than 5 years or diabetic retinopathy, with a UCAR >1 g/g or kidney failure; unrelated controls had diabetes for at least 9 years and normal UCARs. A replication cohort included 7539 additional European American, African American, and American Indian cases and nonnephropathy controls. The FIND investigators found and replicated a DKD-associated genetic locus on chromosome 6q25.2 (rs955333) between the SCAF8 and CNKSR genes across persons of different ancestries. Findings were supported by significantly different gene expression patterns in this region from kidney tissue in persons with DKD versus those without. Extensive work in this field has identified several genomic areas of interest, but results thus far only support a role for multiple susceptibility genes, each with weak effects.

In a meta-analysis of 19,406 subjects with type 1 diabetes by the Diabetic Nephropathy Collaborative Research Initiative (DNCRI) consortium, only four genome-wide loci were significantly associated with DKD (i.e., COL4a3, BMP7, COLEC11, and DDR1). Interestingly, the COL4a3 common missense mutation rs55703767 (Asp326Tyr) was associated with a 21% lower risk for DKD.

Phenome-Wide Association Studies (PheWAS) is a novel approach where multiple phenotypes are analyzed in comparison with a single genetic variant. Unlike a GWAS, in which a genotype is associated with a phenotype, this approach tries to associate phenotypes with genotypes. PheWAS was originally designed to use phenotype records obtained from electronic health records as a tool for genomic investigations. While PheWAS has been used to confirm the associations, for example, of XOR variants in patients at high risk of developing DKD, more research is needed to unambiguously establish phenotype-genotype associations in DKD. Nevertheless, this approach is promising and may also provide a tool to determine long-term consequences of drug target manipulation.

The Type 2 Diabetes Knowledge Portal (T2DKP) is a public resource of genetic datasets and genomic annotations focused on type 2 diabetes–related traits, which aims to provide supporting evidence for the role of certain genes in disease progression, thereby increasing their viability as therapeutic targets.

In addition to genetic factors, multifaceted crosstalk between genes and environmental factors can induce tissue-specific epigenetic changes and mechanisms. These changes include DNA cytosine methylation, histone posttranslational modifications in chromatin, and noncoding (nc) ribonucleic acids (RNAs), all of which can modulate diabetes complications through alterations in gene expression. Epigenome-wide association studies (EWAS) revealed differentially methylated regions localized in a large number of genes in patients with type 1 and type 2 diabetes that are associated with DKD. Through epigenetic mechanisms, cells acquire metabolic memory of prior hyperglycemic exposure that may modulate development and progression of DKD. ^, Histone post translational modification studies, in particular at key metabolic genes suggest a crucial role for this process in the pathogenesis of DKD. Hyperglycemia-induced epigenetic changes activate transcription factors involved in the expression of genes mediating the pathogenesis of DKD. In addition, cytosine methylation changes in genes related to kidney fibrosis in human kidney tissue correlate with downstream transcript levels and provide further evidence that epigenetic dysregulation plays a role in the development of CKD. Similarly, posttranslational modifications of ncRNAs, certain long noncoding RNAs, and miRNAs also have regulatory roles in DKD including promoting/modulating fibrotic gene expression in kidney cells by targeting transcription repressors. Although the mechanisms of such cellular memory are not entirely known, its presence is supported by the regression of morphologic lesions in diabetic kidneys after a prolonged period of normoglycemia following pancreas transplantation. Similarly, the long-lasting effects of previous strict glycemic control observed in persons with type 1 diabetes in the DCCT or with type 2 diabetes in the UKPDS may be due to metabolic memory.

Recent studies report the possible role of the gut microbiota in development and progression of DKD. Interactions between gut microbiota and the kidney, constituting the so-called “gut‐kidney axis,” is maintained via complex cellular and molecular signaling. Metabolomic studies have demonstrated significant differences in gut microbiota diversity among the patients with diabetes and DKD compared with healthy controls. ^, Furthermore, in patients with biopsy-confirmed DKD, a recent study reported the presence of microorganisms positively correlated with glomerulosclerosis, glomerular basement membrane thickening, and albuminuria. The exact mechanism by which gut microbiota exert the influence on DKD risk remains incompletely understood. One of the proposed mechanisms is microbiome disruption associated with a deficiency of short-chain fatty acids, which play a vital role in reducing inflammation.

The excretion of albumin in the urine is determined by the amount of albumin filtered across the glomerular capillary barrier and the amount reabsorbed by the tubular cells. The glomerular capillary wall serves as a filter that discriminates among molecules on the basis of size, electrical charge, and configuration. Studies of glomerular filtrate collected by micropuncture or narrow size fractioning of exogenous polymers, such as dextran, indicate that albuminuria is primarily the result of impairment of the electrostatic barrier within the glomerulus, consequent to a decrease in endothelial cell glycocalyx and heparan sulfate content of the glomerular basement membrane, as well as by changes in the size-selective properties of the glomerular capillary barrier. The size-selective defect in the glomerular capillary appears to become more prominent with higher levels of albumin excretion among patients with either type 1 or type 2 diabetes. ^{, ,} Morphometric data in kidney tissue from Pima Indians with macroalbuminuria demonstrate a significant correlation between the magnitude of this permselective defect and podocyte foot-process width, which is not observed in those with microalbuminuria. These findings are consistent with the view that permselective defects responsible for increased albumin excretion may be focal and are likely due to podocyte foot-process effacement and simplification, as well as to defective intercellular junctions.

Multiphoton fluorescence techniques permit direct imaging of the structure and function of living kidney tissue. In some studies, this technique revealed what appeared to be a much higher albumin glomerular sieving coefficient than was calculated or measured by micropuncture, prompting some investigators to propose alternative explanations for the facilitated urine clearance of albumin in DKD, ^, including the idea that albuminuria is the result of proximal tubular cell dysfunction in retrieving and degrading albumin. ^, Other studies using improved imaging techniques do not support this concept and confirm that glomerular filtration barrier permeability to macromolecules is largely restricted to areas of podocyte damage.

In healthy adults, the GFR ranges from 90 to 120 mL/min/1.73 m ² , is stable through midadult life, and declines by approximately 1 mL/min per year after 50 years of age. The onset of diabetes is associated with hemodynamic changes in the glomerular circulation that lead to increased renal plasma flow, glomerular capillary hyperperfusion, and an increased glomerular transcapillary hydraulic pressure gradient, which together contribute to an increase in GFR. The prevalence of glomerular hyperfiltration, frequently defined as a GFR of at least two standard deviations above the mean GFR in persons with normal glucose tolerance, varies from 10% to 67% in persons with type 1 diabetes and from 6% to 73% in those with type 2 diabetes ( Fig. 41.12 ). Large variations in these estimates are attributed to differences in age, race and ethnicity, glycemic control, duration of diabetes, absence of diet standardization, the definition of hyperfiltration, and methodologies used to measure and report GFR among different populations. Glomerular capillary hypertension and the ensuing increase in filtration pressure are partly responsible for the elevation of GFR, but various other glomerular and tubular factors also influence the magnitude of the hyperfiltration.

Schematic (net) effect of factors implicated in the pathogenesis of glomerular hyperfiltration in diabetes.

Several vascular and tubular factors ^{, ,} are suggested to result in a net reduction in afferent arteriolar resistance, thereby increasing (single-nephron) glomerular filtration rate (GFR). Effects of insulin per se seem to depend on insulin sensitivity. ^, A net increase in efferent arteriolar resistance—leading to increased GFR—is proposed for other vascular factors. ^{, , , ,} Growth hormone and insulin-like growth factor likely increase filtration by augmenting total renal blood flow, without specific arteriolar preference. Glucagon and vasopressin seem to (principally) act through transforming growth factor (TGF). Intrinsic defects of electromechanical coupling or alterations in signal transduction in afferent arterioles may impair vasoactive responses to renal hemodynamic (auto) regulation. Augmented filtration by increases in the ultrafiltration coefficient, and net filtration pressure via reduction in intratubular volume and subsequent hydraulic pressure in the Bowman space are not depicted. Several vascular factors may be released or activated after a (high-protein) meal (e.g., nitric oxide, cyclooxygenase-2 prostanoids, and angiotensin II), ^{, ,} whereas TGF becomes (further) inhibited, through increased amino acid- (and glucose) coupled sodium reabsorption in the proximal tubule ^, and/or increased glucagon/vasopressin-dependent sodium reabsorption in the thick ascending limb. These changes may collectively play a part in postprandial hyperfiltration. COX-2, Cyclooxygenase-2; ETA, endothelin A receptor.

After the initial elevation at the onset of diabetes, GFR decreases in response to glycemic control in both type 1 and type 2 diabetes but usually not to levels found in nondiabetic persons. ^, In some patients, this reversal could reflect the initiation of progressive kidney disease, as suggested by the appearance of global glomerular sclerosis and the fall in single-nephron filtration coefficient. In others, it could represent a purely functional change in kidney vasomotion associated with improvement in glycemic control or simply the intrinsic variability of GFR in the absence of significant histopathologic changes. Distinguishing between these two potential causes of GFR decline requires either observation of GFR over an extended period to determine whether it plateaus, reverses direction, or continues to decline to pathologic levels or attention to other biomarkers including albuminuria. Nevertheless, differentiating between these possibilities is crucial because progressive kidney function decline indicative of underlying pathology may precede the onset of albuminuria in both type 1 and type 2 diabetes. This progressive decline is predominantly linear, but the rate of decline varies widely among individuals. Importantly, half of those in a type 1 diabetes cohort who developed kidney functional decline progressed from a normal eGFR to kidney failure within 2 to 10 years.

Guidelines for the management of DKD emphasize foundational lifestyle interventions, patient-centered self-management education that emphasizes CKM risk reduction, and optimized management of traditional risk factors (blood glucose, blood pressure, and lipids) for all patients with DKD (e.g., type 1 and type 2 diabetes). ^, In addition to consuming a healthy diet and engaging in physical activity, encouraging and helping facilitate smoking cessation in tobacco smokers is recommended, as is achieving and maintaining weight management goals. ^,