Published: August 2010
Diabetic nephropathy (DN) is typically defined by macroalbuminuria—that is, a urinary albumin excretion of more than 300 mg in a 24-hour collection—or macroalbuminuria and abnormal renal function as represented by an abnormality in serum creatinine, calculated creatinine clearance, or glomerular filtration rate (GFR). Clinically, diabetic nephropathy is characterized by a progressive increase in proteinuria and decline in GFR, hypertension, and a high risk of cardiovascular morbidity and mortality.
Diabetes has become the primary cause of end-stage renal disease (ESRD) in the United States, and the incidence of type 2 diabetes mellitus continues to grow in the United States and worldwide. Approximately 44% of new patients entering dialysis in the United States are diabetics. Early diagnosis of diabetes and early intervention are critical in preventing the normal progression to renal failure seen in many type 1 and a significant percentage of type 2 diabetics.
In the United States, approximately 20.8 million people, or 7.0% of the population, are estimated to have diabetes, with a growing incidence. Roughly one third of this population, 6.2 million, is estimated to be undiagnosed with type 2 diabetes. The prevalence of diabetes is higher in certain racial and ethnic groups, affecting approximately 13% of African Americans, 9.5% of Hispanics, and 15% of Native Americans, primarily with type 2 diabetes.1, 2 Approximately 20% to 30% of all diabetics will develop evidence of nephropathy, although a higher percentage of type 1 patients progress to ESRD.
The common progression from microalbuminuria to overt nephropathy has led many to consider microalbuminuria to define early or incipient nephropathy. Renal disease is suspected to be secondary to diabetes in the clinical setting of long-standing diabetes. This is supported by the history of diabetic retinopathy, particularly in type 1 diabetics, in whom there is a strong correlation. The natural history of diabetic nephropathy is a process that progresses gradually over years.
Early diabetes is heralded by glomerular hyperfiltration and an increase in GFR. This is believed to be related to increased cell growth and expansion in the kidneys, possibly mediated by hyperglycemia itself. Microalbuminuria typically occurs after 5 years in type 1 diabetes. Overt nephropathy, with urinary protein excretion higher than 300 mg/day, often develops after 10 to 15 years. ESRD develops in 50% of type 1 diabetics, with overt nephropathy within 10 years.
Type 2 diabetes has a more variable course. Patients often present at diagnosis with microalbuminuria because of delays in diagnosis and other factors affecting protein excretion. Fewer patients with microalbuminuria progress to advanced renal disease. Without intervention, approximately 30% progress to overt nephropathy and, after 20 years of nephropathy, approximately 20% develop ESRD. Because of the high prevalence of type 2 compared with type 1 diabetes, however, most diabetics on dialysis are type 2 diabetics.
Long-standing hyperglycemia is known to be a significant risk factor for the development of diabetic nephropathy. Hyperglycemia may directly result in mesangial expansion and injury by an increase in the mesangial cell glucose concentration. The glomerular mesangium expands initially by cell proliferation and then by cell hypertrophy. Increased mesangial stretch and pressure can stimulate this expansion, as can high glucose levels. Transforming growth factor β (TGF-β) is particularly important in the mediation of expansion and later fibrosis via the stimulation of collagen and fibronectin. Glucose can also bind reversibly and eventually irreversibly to proteins in the kidneys and circulation to form advanced glycosylation end products (AGEs). AGEs can form complex cross-links over years of hyperglycemia and can contribute to renal damage by stimulation of growth and fibrotic factors via receptors for AGEs. In addition, mediators of proliferation and expansion, including platelet-derived growth factor, TGF-β, and vascular endothelial growth factor (VEGF) that are elevated in diabetic nephropathy can contribute to further renal and microvascular complications.
Proteinuria, a marker and potential contributor to renal injury, accompanies diabetic nephropathy. Increased glomerular permeability will allow plasma proteins to escape into the urine. Some of these proteins will be taken up by the proximal tubular cells, which can initiate an inflammatory response that contributes to interstitial scarring eventually leading to fibrosis. Tubulointerstitial fibrosis is seen in advanced stages of diabetic nephropathy and is a better predictor of renal failure than glomerular sclerosis. Hyperglycemia, angiotensin II, TGF-β, and likely proteinuria itself all play roles in stimulating this fibrosis. There is an epithelial-mesenchymal transition that takes place in the tubules, with proximal tubular cell conversion to fibroblast-like cells. These cells can then migrate into the interstitium and produce collagen and fibronectin.
In diabetic nephropathy, the activation of the local renin- angiotensin system occurs in the proximal tubular epithelial cells, mesangial cells, and podocytes. Angiotensin II (ATII) itself contributes to the progression of diabetic nephropathy. ATII is stimulated in diabetes despite the high-volume state typically seen with the disease, and the intrarenal level of ATII is typically high, even in the face of lower systemic concentrations. ATII preferentially constricts the efferent arteriole in the glomerulus, leading to higher glomerular capillary pressures. In addition to its hemodynamic effects, ATII also stimulates renal growth and fibrosis through ATII type 1 receptors, which secondarily upregulate TGF-β and other growth factors.
Control of hypertension has clearly shown to be an important and powerful intervention in decreasing the progression of diabetic nephropathy. In diabetics who have disordered autoregulation at the level of the kidney, systemic hypertension can contribute to endothelial injury. Human studies of type 2 diabetics have shown that blood pressure lowering, regardless of the agent used, retards the onset and progression of diabetic nephropathy. In animal studies, the degree and severity of the diabetic nephropathy were strongly linked to systemic blood pressure.
The fact that most types 1 and 2 diabetics do not develop diabetic nephropathy (DN) suggests that other factors may be involved. Genetic factors clearly play a role in the predisposition to diabetic nephropathy in family members who have DN, and linkage to specific areas on the human genome is evolving. The theory of a reduction in nephron number at birth indicates that individuals born with a reduced number of glomeruli may be predisposed to subsequent renal injury and progressive nephropathy. This has been shown in animal studies in which the mother was exposed to hyperglycemia at the time of pregnancy. If this linkage is true in humans, that would have important implications concerning the role of maternal factors in the eventual development of kidney disease.3
Early signs and symptoms of kidney disease in patients with diabetes are typically unusual. However, a vast array of signs and symptoms listed below may manifest when kidney disease has progressed:4
The differential diagnosis of diabetic nephropathy is vast, but it includes the following in a patient with known diabetes mellitus:
In screening for diabetic nephropathy, we recommend early testing for glucose intolerance and diabetes to identify patients who are at risk for developing microalbuminuria, particularly if they have other risks for type 2 diabetes, such as hypertension, lipid abnormalities, or central obesity. As noted, approximately one third of type 2 diabetics are believed to be undiagnosed. Once the diagnosis of diabetes has been made, we routinely check urinary protein levels only to guide therapy and prognosis.
It is not uncommon to find microalbuminuria or macroalbuminuria in a type 2 diabetic at or soon after the initial diagnosis of diabetes. This may be because the patient has had undiagnosed diabetes for many years, or it may relate to the contributions of hypertension or other processes that may cause proteinuria independently of diabetes, such as small-vessel atherosclerosis. Microalbuminuria is now recognized as an independent cardiac risk factor, even in the absence of diabetes. Screening for microalbuminuria in nondiabetics may have important implications for cardiac risk, and should lead to instituting some of the same therapies as those used for diabetic nephropathy (Table 1).
|Method||Normal Value||Abnormal Values
|24-hour urine collection||<150 mg/day||150-300 mg/day||>300 mg/day|
|Spot protein-to-creatinine ratio (mg/mg)*||Estimate 1 g creatinine excretion/1.73 m2 (ratio of 0.15 =150 mg protein/24 hr/1.73 m2)
Or estimate per 20 mg/kg creatinine excretion for men, 15 mg/kg for women
|Ratio: 0.18-0.36 for 60-kg male, 0.135-0.27 for 60-kg female||Ratio: >0.36 for 60-kg male, >0.27 for 60-kg female|
|Albumin-to-creatinine ratio (μg/mg)†||<30 μg/mg||30-300 μg/mg||>300 μg/mg|
* Good for screening; does not correlate well with 24-hr urinary protein at higher (nephrotic) levels of proteinuria.
† Good for screening; does not account for patient weight or muscle mass; some recommend ratio of 17-249Â μg/mg for women and 25-354Â μg/mg for men to define microalbuminuria.
A renal ultrasound is typically obtained to observe for kidney size. In the early stages of diabetic nephropathy, kidney size may be enlarged from hyperfiltration. With progressive kidney disease from diabetes, the kidneys diminish in size from glomerulosclerosis. In addition, a renal ultrasound can assess for hyperechogenicity that suggests chronic kidney disease and can assist in ruling out obstruction.
Frequently, the question is raised as to whether proteinuria is from diabetes or from a primary renal disease. Suspicion may arise in patients with significant proteinuria without a long history of diabetes or without other signs of end-organ damage, such as retinopathy or neuropathy. Although the presence of retinopathy supports a diabetic source of proteinuria, the lack of diabetic retinopathy does not rule out diabetic nephropathy, particularly in type 2 diabetics.
Patients with diabetes can develop nephrotic-range proteinuria (higher than 3.5 g/24 hr), but typically only after long-standing diabetes. A bland urine sediment supports the diagnosis of diabetes, although it is not uncommon to have some microscopic hematuria with advanced diabetic nephropathy. The presentation of an acute nephrotic syndrome, rapidly rising urinary protein level, or rapidly declining GFR should lead to consideration of renal biopsy. The finding of red cell or white cell casts in the urine should also suggest a biopsy. Renal biopsy findings consistent with diabetic nephropathy in the early stages are mesangial expansion and glomerular basement membrane thickening. Eventual progression of diabetic nephropathy can lead to nodular glomerulosclerosis, also referred to as Kimmelstiel-Wilson disease.
Often, a patient will present without available history and may be frankly nephrotic in the face of long-standing diabetes. In cases of uncertainty such as this, it is not wrong to consider a renal biopsy, because the finding of a primary glomerular disease could potentially change the course of management.
See Table 2.
|1||Kidney damage with normal or raised GFR||≥90|
|2||Kidney damage with mild decrease in GFR||60-89|
|3||Moderate decrease in GFR||30-59|
|4||Severe decrease in GFR||15-29|
From National Kidney Foundation: GFR, 2008. Available at www.kidney.org/professionals/KLS/gfr.cfm#20.
GFR, glomerular filtration rate.
One keystone in the prevention and management of diabetic nephropathy is tight glycemic control. In the Diabetes Control and Complications Trial, type 1 diabetics were randomized to intensive or conventional insulin treatments and followed for an average of 6.5 years.5 Average hemoglobin A1c (HbA1c) values were 7.2% versus 9.2%. There was a 39% risk reduction in the development of microalbuminuria and a 54% reduction in the development of macroalbuminuria in the intensive treatment group.
In the UK Prospective Diabetes Study (UKPDS), 3867 patients with newly diagnosed type 2 diabetes were randomized to oral or insulin therapy versus dietary control and followed for 11 years.6 The difference in HbA1c was 7.0% versus 7.9%. After 9 years, there was a significant risk reduction in the intensive group, with a relative risk of 0.76 for the development of microalbuminuria.
The complete correction of hyperglycemia with pancreatic transplantation in type 1 diabetics has led to a dramatic resolution in glomerular and tubular expansion and fibrosis over time.7 With a drop in HbA1c from an average of 8.7% to 5.5% in eight transplanted patients, there was a significant reduction in basement membrane thickening and mesangial expansion on repeat biopsies over time. Even glomerular sclerosis appeared to resolve, showing that renal fibrosis may be reversible, although it took 10 years after transplantation to see these significant changes.
Control of blood pressure is the other keystone in prevention and treatment. Blood pressure control is critical in slowing the natural history of diabetic nephropathy in types 1 and 2 diabetics. Parving and colleagues studied 12 type 1 diabetics and treated hypertension before angiotensin inhibitors were available.8 Using metoprolol, hydralazine, and diuretics, patients with macroalbuminuria and declining GFR were treated to reduce mean arterial pressure from 120 to lower than 105 mm Hg. An initial decline in GFR was noted in most patients immediately after blood pressure lowering. However, a 2-year subsequent follow-up showed a slowing in the rate of reduction of GFR from an average of 0.91 to 0.39 mL/min/month, with significant preservation of GFR above that expected from pretreatment measurements. The initial drop in GFR illustrated a functional or hemodynamic effect of antihypertensive treatment, which does not lead to permanent renal damage but rather to renal preservation. One review has recommended maintaining antihypertensive therapy in renal disease, even if therapy causes some hemodynamic drop in GFR, provided there is a stabilization in the creatinine level, with an increase of less than 30% of baseline.9
The UKPDS studied 1148 hypertensive patients who had recently been diagnosed with type 2 diabetes.10 Patients were assigned to tight blood pressure control, achieving a mean level of 144/82 mm Hg, or less tight control, with a mean pressure of 154/81 mm Hg. After a median follow-up of 8.4 years, there was significant reduction in risk of death related to diabetes, stroke, and microvascular disease in the tight control group. There was no difference in outcomes within the tight control group in patients assigned to atenolol or captopril, although captopril was given only twice daily at 25 to 50 mg/dose.
The importance of blood pressure control, no matter what agent is used, cannot be emphasized enough in diabetes, both for slowing progression of nephropathy and for preventing cardiovascular morbidity and mortality. Currently, the recommendation from the most recent Joint National Committee guidelines is that blood pressure in diabetics be reduced to lower than 130/80 mm Hg.11 Some studies of hypertension treatment in diabetics have shown benefit with lowering blood pressure even further. For example, in the Hypertension Optimal Treatment trial, diabetics with a target diastolic pressure of 80 mm Hg had 50% of the cardiovascular events seen in patients with a target of 90 mm Hg.12
Furthermore, the Modification of Diet in Renal Disease Study has found value in further reduction in a diverse group of patients with renal disease and proteinuria.13 In patients with 24-hour urinary protein of more than 1 g, a blood pressure level at or below 125/75 mm Hg was shown to slow the decline in GFR. Based on these data, we recommend a target blood pressure of 130/80 mm Hg or lower in all diabetics, with a further goal of 125 to 130/75 to 80 mm Hg in diabetics with macroalbuminuria (Table 3). These blood pressure goals may seem difficult to accomplish in some patients but, in clinical trials, preset goals have been consistently achieved. It is important for clinicians and patients to be aware early on that three or more agents may be required to achieve the blood pressure goal, and that these agents will likely be needed long term.
|Patient||Goal Blood Pressure (mm Hg)||Recommended Blood Pressure Agents|
|Type 1 diabetic||≤130/80||ACEI; ARB if ACEI not tolerated; diuretic as second-line agent|
|Type 2 diabetic||≤130/80||ARB; ACEI as alternative; diuretic as second-line agent|
|Diabetic with macroalbuminuria||≤130/80||Angiotensin blockade in conjunction with diuretic; beta blocker or calcium channel blocker as third-line agent|
|Diabetic with congestive heart failure||≤130/80||ACEI remains a first-line agent, followed by beta blocker|
|Diabetic with coronary artery disease||≤130/80||Beta blocker is first-line agent, followed by ACEI; diuretic as third-line agent|
ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor blocker.
It is important to emphasize that diuretics are extremely beneficial adjuncts for blood pressure control and are often missing in patients who are not close to meeting blood pressure goals. Diuretics are first-line agents for many hypertensives, and we routinely add a diuretic as a second-line agent after angiotensin blockade in diabetics. Thiazide diuretics work well, even at low dosages for patients with normal renal function. For example, just 12.5 to 25 mg/day of hydrochlorothiazide is often effective. When the GFR is below 60 mL/min, we often institute loop diuretics for a better natriuretic effect. Short-acting loop diuretics such as furosemide work better when given at least twice daily to avoid rebound sodium retention.
Specific use of agents that block the renin-angiotensin system appears to be particularly beneficial in the prevention or slowing of progression of diabetic nephropathy. ATII can increase glomerular capillary pressure by preferentially constricting the renal efferent arteriole. ATII also stimulates renal cell growth and fibrosis, independently of hemodynamic effects. Even with blood pressure at goal and without microalbuminuria, we consider initiation of low dosages of angiotensin-blocking agents, with titration up as tolerated. Other antihypertensives may not offer this antiproliferative effect and may cause adverse hemodynamic effects. For example, the calcium channel blockers, in particular the dihydropyridines, cause afferent dilation and thus may increase glomerular capillary pressure. Dihydropyridines may best be reserved as third- or fourth-line agents in patients with diabetes, only after angiotensin blockade and diuretics have already been instituted.
The benefits of angiotensin-converting enzyme inhibitors (ACEIs) have been shown by the Collaborative Study Group using captopril compared with placebo in 409 patients with type 1 diabetes and macroalbuminuria.14 Patients with a creatinine level higher than 1.5 mg/dL had a significant reduction in risk of doubling serum creatinine and of the composite of death, dialysis, or transplantation when treated with captopril, even after controlling for changes in blood pressure.
A substudy of the Heart Outcomes Prevention Evaluation (HOPE) looked at 3577 patients ages 55 years or older with diabetes and randomized to ramipril, 10 mg/day, or placebo.15 All patients had at least one other cardiac risk factor in addition to diabetes, including lipid abnormalities, hypertension, microalbuminuria, and current smoking. Patients with dipstick-positive proteinuria and overt nephropathy were excluded, as were patients with congestive heart failure. The ramipril group had a reduction in blood pressure of 2.5 mm Hg systolic and 1 mm Hg diastolic. When corrected for blood pressure, there were significant reductions in risk of stroke and cardiovascular morbidity and mortality. There was also a relative risk reduction of 24% in the progression to diabetic nephropathy in all patients, although this was not controlled for the minor change in blood pressure. In subgroup analysis, the diabetics who had significant benefit were those who already had coronary disease or microalbuminuria.
Angiotensin receptor blockers (ARBs) have also shown renal protection in type 2 diabetics. Two studies of the prevention of progression in patients with microalbuminuria have been published.16, 17 Both irbesartan and valsartan were shown to lower albumin excretion and prevent development of diabetic nephropathy, even when controlling for changes in blood pressure. The effects of irbesartan appeared to be dosage dependent, with greater protection at 300 mg/day versus 150 mg/day. The study using valsartan did not stratify for dosage, and the average dosage was 122 mg/day. Two trials of ARBs in patients with overt diabetic nephropathy have also shown a decline in rate of progression of disease.18, 19 Both studies have shown a slowing in the rate of progression to doubling of creatinine or to ESRD using irbesartan titrated to 300 mg/day or losartan at an average dosage of 85.5 mg/day. The benefit in both studies was impressive, although more importantly the treatment only decreased the rate of progression of disease but did not halt it. The average patient had a delay in progression to ESRD of about 2 years.
Angiotensin blockade should be started early and may have its greatest benefit in prevention or reversal of early kidney disease. The benefits of angiotensin blockade have been shown in both trials of irbesartan and valsartan in type 2 diabetics with microalbuminuria in which a significant percentage of treated patients had albumin excretion reduction into the normal range. Administration of irbesartan, 300 mg/day, led to a 34% incidence of normalization of protein excretion compared with 21% in the placebo group. Similarly, valsartan treatment led to an approximately 30% normalization compared with 14.5% in the placebo group. Early treatment is important to prevent and possibly even reverse diabetic nephropathy.
The combination of an ACEI and an ARB in diabetic nephropathy has been studied in several small randomized trials.20 The largest analysis, the Candesartan and Lisinopril Microalbuminuria (CALM) study, involved lisinopril, 20 mg/day, and candesartan, 16 mg/day, given individually and then in combination in patients with type 2 diabetes, hypertension, and microalbuminuria. Any patient with a diastolic blood pressure lower than 80 mm Hg was not eligible to receive the combination, thus excluding 25% of the original participants. In the remaining patients receiving the combination, there was a further lowering in blood pressure and in urinary protein excretion compared with either agent alone, and the treatment was well tolerated.
We recommend attempting to maximize angiotensin blockade as much as patients will tolerate, especially given the dosage-related benefits shown in some studies. The benefit of using an ACEI-ARB combination over a higher dosage of a single agent alone appears to provide a greater reduction in protein excretion, although long-term studies are needed to prove the efficacy of the ACEI-ARB combination. If the maximal dosage of one agent has been achieved, it may be reasonable to introduce the second agent. Some clinicians have recommended adding an ARB to an ACEI once a moderate dosage of the ACEI is achieved. There may be ATII formation independent of ACE, and blocking the receptor should block this ATII activity. Angiotensin blockade may reach a threshold above which further benefit will not be gained, and animal studies at supramaximal dosages of ACEIs and ARBs have shown incomplete blockage of TGF-β. However, many patients are underdosed and could gain more benefit from maximal therapy.
At this point, it is difficult to recommend one type of renin-angiotensin blockade over another for initial treatment in diabetics. There are ample data supporting ACEIs for type 1 diabetics. Based on data from the HOPE trial, it would seem acceptable also to choose an ACEI in type 2 diabetics with coronary artery disease, although beta blockers remain first-line agents. Similarly, in patients with congestive heart failure, ARBs may be used in patients who are intolerant of ACEIs, but ACEIs remain first-line agents. Patients with type 2 diabetes and microalbuminuria or overt nephropathy but no known cardiac disease should probably receive an ARB as first-line treatment based on data reviewed earlier (Table 3).
It is important to treat with angiotensin blockade, even in patients with baseline renal insufficiency. Patients with a high serum creatinine level are still candidates for angiotensin blockade, although they need close follow-up, with cautious titration. As noted, GFR often drops somewhat at the initiation of therapy, and close monitoring of renal function should take place within 1 or 2 weeks of starting medication. Provided there is a bump of no more than 30% in the serum creatinine level, and it remains stable, therapy may be continued.9
ACEIs and ARBs are generally both well tolerated. However, attempting to institute and titrate either agent must be done while monitoring for side effects, such as symptomatically low blood pressure or hyperkalemia. Patients should be monitored for hyperkalemia and an increase in creatinine level within 1 week of starting or changing the dosage of either agent, and should be counseled on limiting potassium in the diet. Diuretics such as thiazides or loop diuretics are often useful and necessary adjunctive treatments for blood pressure control, and can have the added benefit of offsetting a rise in serum potassium level. The side effect of cough secondary to bradykinin accumulation with an ACEI may necessitate conversion to an ARB.
Rarely, severe, life-threatening angioedema can occur with ACEIs or ARBs. This side effect has been reported even months or years after starting an ACEI. The incidence of angioedema appears to be less frequent with ARBs. In a patient with a history of ACEI-induced angioedema, an ARB may be started cautiously at a low dosage if there is a strong indication, such as diabetes and microalbuminuria, but should probably be held if there is no abnormal urinary albumin excretion.21 ACEIs and ARBs are also both contraindicated in pregnancy, and should be used with caution in women of childbearing potential.
There are other classes of agents for diabetic nephropathy that may show hope for the future. Aldosterone receptor antagonists decrease proteinuria further in patients with diabetes on ACEIs.22 Eplerenone, a new aldosterone blocking agent, has been approved by the U.S. Food and Drug Administration. It is similar to spironolactone but lacks the sex steroid side effects, such as impotence and gynecomastia. In a small randomized study of type 2 diabetics already treated with an ACEI, eplerenone caused a significant reduction in urinary albumin excretion.23 One side effect noted in the study that needs to be monitored is hyperkalemia, which may be increased with this combination. Further long-term studies will be needed to decipher the benefit of this combination.
3-Hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors, or statins, are also being studied. There are already recommendations for tight lipid control in diabetics because of the high cardiac risk in these patients. Statins may have additional unique benefits, independent of lipid lowering. In animal models of diabetic nephropathy, statin treatment was found to block intracellular signaling and decreases the mRNA expression of TGF-β.24
Agents that directly inhibit or degrade TGF-β or inhibit AGEs have shown success in animals and are being developed for clinical human trials. One such agent, known as ALT-711, breaks cross- links in AGE complexes. It has been shown to improve vascular endothelial function in diabetes and may have promise for diabetic nephropathy.25
A common question from patients and primary care providers is whether the reduction of dietary protein is beneficial in diabetic nephropathy. Many animal studies of glomerulopathies have shown reduction in decline in renal function by restricting dietary protein. The data in humans remain inconclusive, however. We currently counsel patients to avoid protein supplements such as protein shakes or powders, but do not otherwise restrict protein from the diet. Others recommend a restriction of 0.8 g/kg/day in patients with overt nephropathy, or even 0.6 g/kg/day in the presence of a falling GFR.
It is important to maintain a low-sodium diet in diabetic nephropathy. Many diabetics with renal disease are salt-sensitive and minimizing salt intake can help in reaching blood pressure goals, with secondary benefits of decreased stroke risk, regression of left ventricular hypertrophy, and reduction in proteinuria. We advocate a low-sodium diet of 2.3 g or lower (5.8 g of NaCl) or 100 mEq/day in patients with diabetes and hypertension or any degree of proteinuria.
It is important to avoid nephrotoxic agents, if possible, in patients with diabetic nephropathy. Nonsteroidal anti-inflammatory drugs (NSAIDs) can cause a significant drop in GFR in patients with diabetic nephropathy, particularly when used with angiotensin-blocking agents. Daily low-dosage aspirin is safe in diabetics, and the cardiac benefits greatly outweigh any risk. However, aspirin at higher dosages and other NSAIDs should be avoided if possible. Cyclooxygenase 2 (COX-2) inhibitors are similar to other NSAIDs in their potential for renal toxicity.
Radiocontrast media are also particularly nephrotoxic for diabetics. Even with a normal serum creatinine level, patients with diabetes and proteinuria should be volume-loaded 12 hours before and after exposure to contrast, if possible. Diuretics should be temporarily discontinued, and hyperglycemia should be controlled. Other agents such as dopamine-like agonists and acetylcysteine may help prevent contrast nephropathy in diabetics but require further study.
Referral to a nephrologist should be considered if the GFR is steadily declining or is already below 60 to 70 mL/min. Challenges in blood pressure control, hyperkalemia, or rising creatinine level on angiotensin blockade may also prompt a referral. We consider it appropriate to refer a patient in any situation in which the primary physician feels that he or she needs additional input or assistance with the management of diabetic nephropathy.
Surgical care in diabetic nephropathy typically entails complications that can arise from peripheral vascular disease or diabetic foot ulcers. Surgical issues also arise in the context of the creation of a native or artificial access in preparation for dialysis. Lastly, transplantation options may be pursued in the form of kidney or pancreas transplantation, or both, in select individuals.
Special attention should be given to women of reproductive age who become pregnant and who are taking ACEIs or ARBs. Pregnancy in a patient with diabetic nephropathy does not seem to accelerate functional loss, however this issue is controversial. A more increased range of proteinuria and hypertension may occur after midgestation.