Epidemiology of Nephropathy
Kidney disease is a common complication of types 1 and 2 diabetes. For clinical care and epidemiological studies, diabetic kidney disease (DKD) is defined by elevated urine albumin excretion or reduced glomerular filtration rate (GFR), or both. Using this definition, DKD develops in a third to a half of all patients with diabetes. As the prevalence of obesity and type 2 diabetes has grown over the last two decades, the overall number of people with DKD has also increased.
Hyperglycaemia is a primary cause of DKD, and intensive glucose control reduces the risk of developing DKD. Other environmental factors as well as genetic predisposition also influence the development and progression of DKD. End stage renal disease (ESRD) requiring maintenance dialysis or kidney transplantation is the most severe direct clinical consequence of DKD, but DKD also markedly increases the risks of cardiovascular diseases and premature mortality, even without progression to ESRD.
The clinical manifestations, timing, and evolution of DKD vary comparing types 1 and 2 diabetes. This probably relates to incompletely overlapping underlying disease mechanisms as well as differences in age and comorbidities, such as hypertension and vascular disease. Therefore, the clinical course of DKD is described below separately for type 1 diabetes and type 2 diabetes, followed by the health consequences of DKD in general.
Kidney disease in type 1 diabetes
Paradigms of kidney disease in type 1 diabetes
Landmark studies of type 1 diabetes in the 1970s and 1980s characterized the natural history of DKD as progressively increasing urine albumin excretion followed by GFR loss and the development of ESRD. In this paradigm, microalbuminuria, defined as albumin excretion rate (AER) 30-299 mg/24 h and historically referred to as “incipient nephropathy,” progressed steadily to macroalbuminuria, defined as AER ≥300 mg/24 h and historically referred to as “diabetic nephropathy.” Patients with microalbuminuria were commonly noted to have elevated GFR (“hyperfiltration”), while patients with macroalbuminuria were observed to have rapid GFR loss leading consistently to ESRD. These studies provided a critical framework for classifying DKD and guiding clinical diagnosis and treatment. Over time, further observation has revealed frequent exceptions to this paradigm. Specifically, albuminuria (microalbuminuria or macroalbuminuria) has been observed to regress, while GFR loss has been observed without macroalbuminuria and is not always progressive. As a result, an updated paradigm holds that albuminuria and impaired GFR are complementary, overlapping manifestations of DKD. In this context, albuminuria clearly increases the risk of GFR loss and there are common risk factors for both albuminuria and GFR loss, but these phenotypes may not evolve together.
Incidence of kidney disease in type 1 diabetes
Accumulated experience suggests that about half of patients with type 1 diabetes (or slightly less) develop DKD over the course of their lifetime. Albuminuria and reduced GFR (<60 mL/min/1.73m2) are rare during the first decade after type 1 diabetes diagnosis. Microalbuminuria commonly develops during the second decade after diabetes diagnosis, with macroalbuminuria and reduced GFR generally developing after 20 years’ diabetes duration. The cumulative incidence of DKD appears to be decreasing over time. Early studies suggested that up to 45% of patients with type 1 diabetes developed macroalbuminuria. In more recent studies, the lifetime cumulative incidence of macroalbuminuria has been described as 15-25%, and the cumulative incidence of microalbuminuria has been reported as 25-40%. In early studies, up to 35% of participants developed end stage renal disease (ESRD). Now, in Finland and in the Pittsburgh Epidemiology of Diabetes Cohort (Pittsburgh, PA, USA), the long-term cumulative incidence of ESRD has fallen to less than 10%, though the rate of ESRD has remained higher in the Joslin type 1 diabetes cohort (Boston, MA, USA).
Risk factors for kidney disease in type 1 diabetes
Hyperglycaemia is an established risk factor for DKD. Moreover, the Diabetes Control and Complications Study (DCCT) and its observational follow-up study, the Epidemiology of Diabetes Interventions and Complications (EDIC) Study, clearly demonstrated that intensive glucose control reduces the risk of DKD. Specifically, during the DCCT, intensive therapy aimed at lowering glucose concentrations as close as safely possible to the normal range reduced the risks of incident microalbuminuria (defined for the DCCT as AER ≥40 mg/24 h) and macroalbuminuria (AER ≥300 mg/24 h) by 39% (95% CI 21%-52%) and 54% (95% CI 29%-74%), respectively, compared with conventional therapy. During
Figure 1. Cumulative incidence of reduced glomerular filtration rate (GFR) in the DCCT/EDIC Study. Reduced GFR is defined as GFR estimated from serum creatinine <60 mL/min/1.73m2 on two consecutive study visits at least one year apart and is described by DCCT treatment assignment and time from DCCT randomization according to intention-to-treat. Intensive diabetes therapy reduced the risk of reduced GFR by 50%, with cumulative incidence curves diverging only 10 years after randomization. Reproduced with permission from de Boer IH et al, N Engl J Med. 2011 Dec 22;365(25):2366-76.the subsequent 18 years of the observational EDIC Study, participants previously assigned to DCCT intensive therapy continued to experience lower rates of incident microalbuminuria and macroalbuminuria despite a loss of separation in HbA1c by treatment assignment, with risk reductions of 45% (95% CI 26%-59%) and 61% (95% CI 41%-74%), respectively.
The term “metabolic memory” has been used to describe these differences in outcomes that persist or even expand after the original separation in glycaemia has disappeared. Beneficial effects of intensive therapy on the development of reduced GFR (sustained estimated GFR <60 mL/min/1.73m2) became evident during long-term combined DCCT/EDIC follow-up, with a risk reduction of 50% (95% CI 18%-69%), (Figure 1). Together, these findings demonstrate that intensive glucose control results in clinically important, durable reductions in the risk of DKD in type 1 diabetes.
Other environmental factors and genetic predisposition also influence which patients develop DKD. Environmental risk factors for DKD include male sex, obesity, blood pressure, inflammation, insulin resistance, vitamin D deficiency, and dyslipidaemia. In addition, a hereditary component to DKD has long been recognized. While genetic loci and polymorphisms in specific genes have been associated with DKD, the genetic polymorphisms known today fail to explain a large proportion of the risk of DKD. In part, difficulty identifying the genetic underpinnings of DKD may be due to difficulties characterizing heterogenous DKD phenotypes, and research in this area is currently very active.
Progression of kidney disease in type 1 diabetes
The progression of DKD in type 1 diabetes is variable. In the Joslin type 1 diabetes cohort, 29% of participants who developed microalbuminuria went on to develop reduced GFR within 12 years’ average follow-up. In the EURODIAB type 1 diabetes study, 14% of microalbuminuric participants progressed to macroalbuminuria over 7.3 years’ average follow-up. In the Steno type 1 diabetes inception cohort, 34% of participants who developed persistent microalbuminuria went on to develop macroalbuminuria over 7.5 years’ average follow-up. In the DCCT/EDIC cohort, among participants who developed incident microalbuminuria, the 10-year cumulative incidence of macroalbuminuria was 28%.
The fall in GFR tends to be most rapid among patients with macroalbuminuria. In the DCCT/EDIC cohort, participants who developed incident macroalbuminuria lost estimated GFR at a mean rate of 5.7% per year, and the 10-year cumulative incidence of impaired GFR (sustained estimated GFR <60 mL/min/1.73m2) was 32%. Among participants with incident microalbuminuria, mean rate of estimated GFR loss was 1.2% per year, and the 10-year cumulative incidence of reduced GFR (sustained estimated GFR <60 mL/min/1.73m2) was 15%.
However, substantial GFR loss can occur without developing macroalbuminuria. In the Joslin type 1 diabetes cohort, “early renal function decline” occurred in 31% of participants with microalbuminuria and also occurred occasionally in participants with persistent normoalbuminuria (AER <30 mg/24 h). In this cohort, “early renal function decline” was defined as loss of estimated GFR exceeding 3.3% per year, starting within the “normal range” of estimated GFR (≥60 mL/min/1.73m2). The estimation of GFR using serum cystatin C may have facilitated observation of changes in GFR that go unobserved using serum creatinine. In the DCCT/EDIC cohort, 39% of DCCT/EDIC participants who developed sustained reduction in GFR (<60 mL/min/1.73m2 for at least one year) did so without previously manifesting macroalbuminuria. Together, these observations suggest that albuminuria and GFR loss are linked but are not necessarily reflective of a single, homogenous underlying disease process.
Regression of kidney disease in type 1 diabetes
Not all patients who develop DKD progress to advanced stages. Microalbuminuria commonly regresses to normoalbuminuria. In the Joslin type 1 diabetes cohort, 58% of participants with persistent microalbuminuria during calendar years 1991-1992 regressed to persistent normoalbuminuria over the following 6 years, mostly without using inhibitors of the renin-angiotensin-aldosterone system (RAAS). Similar results were observed in the DCCT/EDIC cohort, in which 40% of participants who developed incident persistent microalbuminuria regressed to persistent normoalbuminuria within 10 years, and in other type 1 diabetes cohorts. In these cohort studies, low HbA1c, low blood pressure, and favorable lipid profiles were associated with a greater likelihood of microalbuminuria regression. This suggests that microalbuminuria is sensitive to the overall metabolic milieu.
Of DCCT/EDIC participants who developed incident macroalbuminuria, 52% regressed to sustained microalbuminuria or normoalbuminuria within 10 years, many but not all under treatment with RAAS inhibitors. In addition, regression of macroalbuminuria to microalbuminuria or normoalbuminuria was associated with an 89% lower risk of progressing to reduced GFR. These observations suggest that even macroalbuminuria may represent a dynamic and modifiable state. Whether regression of microalbuminuria or macroalbuminuria is associated with histologic changes to the renal parenchyma is unknown. However, longitudinal studies of pancreas transplantation demonstrate that the pathological lesions of diabetic glomerulopathy can regress with euglycaemia.
Kidney disease in type 2 diabetes
Incidence and prevalence of kidney disease in type 2 diabetes
The incidence of DKD and rates of DKD progression are less clear in type 2 compared with type 1 diabetes, largely due to the highly variable age of onset, difficulty defining the precise time of diabetes onset, and the relative scarcity of long-term type 2 diabetes cohorts. In this regard, two of the best characterized type 2 diabetes cohorts are the United Kingdom Prospective Diabetes Study (UKPDS) and the Pima Indian population. The UKPDS enrolled 5,102 participants with new-onset type 2 diabetes. After a median 15 years of follow-up, microalbuminuria (defined as persistent urine albumin concentration ≥50 mg/L) occurred in 38% of participants, and reduced GFR (defined as persistent estimated creatinine clearance ≤60 mL/min/1.73m2) occurred in 29% of participants. Among Pima Indians, for whom the onset and duration of diabetes are more precisely determined due to systematic diabetes screening, the cumulative incidence of heavy proteinuria (≥1 gram per gram creatinine) was 50% at 20 years’ duration, prior to widespread use of RAAS inhibitors. The high rate of proteinuria in the Pima population has remained stable over time, though the incidence of ESRD has declined.
Figure 2B* *Figure 2A & 2B: Prevalence of diabetic kidney disease (DKD) among adults in the United States, 1988-2008. (A) Among people with diabetes (mostly type 2 diabetes), the prevalence of DKD has remained stable near 35%, with a shift from an albuminuria phenotype to a phenotype of impaired glomerular filtration rate. (B) The number of people with DKD in the United States has expanded markedly since 1988, which is attributable to the growth of obesity and type 2 diabetes. Reproduced with permission from de Boer IH et al, JAMA. 2011 Jun 22;305(24):2532-9.
In most type 2 diabetes populations, the prevalence of DKD at any point in time is approximately 30-50%. The prevalence and manifestations of DKD vary by duration of diabetes and by age. Among US adults with diabetes (>90% type 2), the prevalence of DKD is approximately 35% overall (Figure 2A & 2B), ranging from approximately 25% with age <65 years to approximately 50% with age ≥65 years. At younger ages, the most common manifestation of DKD is microalbuminuria. With older age, reduced GFR is increasingly prevalent. From 1988 to 2008, the overall prevalence of DKD has remained stable near 35%, but there was a shift in clinical manifestations, with a modestly decreased prevalence of albuminuria and a significantly increased prevalence of impaired GFR. Though the increased prevalence of reduced GFR is accentuated among older adults with diabetes, aging of the population does not account for these trends. Increased use of medications that reduce albuminuria, such as glucose-lowering medications and RAS inhibitors, may explain the modest reductions in the prevalence of albuminuria. Explanations for the increasing prevalence of reduced GFR are not clear.
Clinical manifestations of kidney disease in type 2 diabetes
As with type 1 diabetes, albuminuria has traditionally been considered a hallmark of DKD in type 2 diabetes. However, the phenotype of reduced GFR with normoalbuminuria has been increasingly recognized in type 2 diabetes. In population-based studies of diabetes in the United States and Australia, 36-55% of individuals with reduced GFR did not have concurrent microalbuminuria or macroalbuminuria. Frequently, non-albuminuric reduced GFR was observed in the absence of diabetic retinopathy, suggesting underlying processes other than diabetic glomerulopathy. In the UKPDS, female gender, increased age, and insulin resistance were risk factors for reduced GFR but not microalbuminuria, while male gender, adiposity, hyperglycaemia, and dyslipidaemia were risk factors for microalbuminuria but not reduced GFR. Higher blood pressure was a risk factor for both reduced GFR and microalbuminuria.
Progression of kidney disease in type 2 diabetes
The progression and regression of established DKD is highly variable in type 2 diabetes. In the UKPDS, transition from microalbuminuria to macroablbuminuria occurred at a rate of 2.8% per year, and transition from macroalbuminuria to elevated plasma creatinine or ESRD occurred at a rate of 2.3% per year. These rates are high but clearly leave many participants without progression. In addition, regression of microalbuminuria to normoalbuminuria also appears to be common in type 2 diabetes. As with type 1 diabetes, loss of GFR can occur at any level of urine albumin excretion but tends to be more rapid with greater urine albumin excretion. Large clinical trials testing agents to slow the progression of DKD have therefore focused on patients with macroalbuminuria, often including many participants who also have impaired GFR at baseline. In this high-risk population, RAAS antagonists slow the progression of established DKD, but a high rate of DKD progression remains.
Health consequences of diabetic kidney disease
An unfortunately common outcome of DKD is premature death. Microalbuminuria, impaired GFR, or both are consistently associated with increased risks of cardiovascular disease and death. In fact, the excess mortality risk observed among people with both types 1 and 2 diabetes is largely confined
Figure 3. 10-year cumulative all-cause mortality among adults with type 2 diabetes in the United States, by presence or absence of kidney disease. The Y axis demonstrates absolute 10-year cumulative mortality, standardized to the age, gender, and racial/ethnic distribution of US adults, with 95% confidence intervals. The dotted line depicts 10-year cumulative incidence for US adults with neither diabetes nor kidney disease. Mortality is only slightly increased for adults with type 2 diabetes but no evidence of kidney disease but markedly increased for adults with albuminuria (≥30 mg urine albumin per gram creatinine), reduced glomerular filtration rate (<60 mL/min/1.73m2), or both. Reproduced with permission from Afkarian M et al, J Am Soc Nephrol. 2013 Feb;24(2):302-8.to those with evidence of DKD (Figure 3). DKD is associated with a number of interrelated cardiovascular diseases, including atherosclerosis of the coronary arteries, cerebral arteries, and aorta leading to myocardial infarction and stroke; left ventricular hypertrophy leading to congestive heart failure; medial artery calcification leading to arterial stiffness and peripheral vascular disease; and sudden cardiac death. In part, DKD may be a marker of diffuse vascular injury. In addition, DKD may contribute causally to cardiovascular disease by worsening traditional cardiovascular risk factors (such as hypertension and dyslipidaemia) and activating disease pathways more specific to kidney disease (such as oxidative stress, inflammation, anaemia, and altered mineral metabolism). Older patients with type 2 diabetes are more likely to die than to progress to ESRD, while progression to ESRD is more likely among patients with more severe DKD at baseline and patients who are younger with less cardiovascular comorbidity.
^ Andersen AR, Christiansen JS, Andersen JK, Kreiner S, Deckert T. Diabetic nephropathy in Type 1 (insulin-dependent) diabetes: an epidemiological study. Diabetologia. Dec 1983;25(6):496-501.
^ Hovind P, Tarnow L, Rossing P, et al. Predictors for the development of microalbuminuria and macroalbuminuria in patients with type 1 diabetes: inception cohort study. Bmj. May 8 2004;328(7448):1105.
^ Krolewski AS, Warram JH, Christlieb AR, Busick EJ, Kahn CR. The changing natural history of nephropathy in type I diabetes. The American journal of medicine. May 1985;78(5):785-794.
^ de Boer IH. Kidney disease and related findings in the diabetes control and complications trial/epidemiology of diabetes interventions and complications study. Diabetes Care. Jan 2014;37(1):24-30.
^ de Boer IH, Sun W, Gao X, et al. Effect of intensive diabetes treatment on albuminuria in type 1 diabetes: long-term follow-up of the Diabetes Control and Complications Trial and Epidemiology of Diabetes Interventions and Complications study. The lancet. Diabetes & endocrinology. Jul 17 2014.
^ De Boer IH, Afkarian M, Rue TC, et al. Renal outcomes in patients with type 1 diabetes and macroalbuminuria. J Am Soc Nephrol. 2014 Oct;25(10):2342-50.
^ Perkins BA, Ficociello LH, Silva KH, Finkelstein DM, Warram JH, Krolewski AS. Regression of microalbuminuria in type 1 diabetes. N Engl J Med. Jun 5 2003;348(23):2285-2293.
^ Fioretto P, Steffes MW, Sutherland DE, Goetz FC, Mauer M. Reversal of lesions of diabetic nephropathy after pancreas transplantation. The New England journal of medicine. Jul 9 1998;339(2):69-75.
^ Retnakaran R, Cull CA, Thorne KI, Adler AI, Holman RR. Risk factors for renal dysfunction in type 2 diabetes: u.k. Prospective diabetes study 74. Diabetes. Jun 2006;55(6):1832-1839.
^ Pavkov ME, Knowler WC, Bennett PH, Looker HC, Krakoff J, Nelson RG. Increasing incidence of proteinuria and declining incidence of end-stage renal disease in diabetic Pima Indians. Kidney Int. Nov 2006;70(10):1840-1846.
^ de Boer IH, Rue TC, Hall YN, Heagerty PJ, Weiss NS, Himmelfarb J. Temporal trends in the prevalence of diabetic kidney disease in the United States. JAMA : the journal of the American Medical Association. Jun 22 2011;305(24):2532-2539.
^ Adler AI, Stevens RJ, Manley SE, Bilous RW, Cull CA, Holman RR. Development and progression of nephropathy in type 2 diabetes: the United Kingdom Prospective Diabetes Study (UKPDS 64). Kidney Int. Jan 2003;63(1):225-232.
^ Tuttle KR, Bakris GL, Bilous RW, et al. Diabetic Kidney Disease: A Report From an ADA Consensus Conference. Diabetes Care. Oct 2014;37(10):2864-2883.