Download - Diabetic nephropathy

COMPLICATIONS OF DIABETES

© 2006 Elsevier Ltd83MEDICINE 34:3

Diabetic nephropathy is a major cause of premature death in patients with diabetes, principally from cardiovascular disease, the incidence of which is about 15-fold greater in patients with diabetic nephropathy.

Epidemiology

In type 1 diabetes, the most common cause of kidney damage is classical diabetic nephropathy. Nephropathy is rare in the first 5–10 years of type 1 diabetes, but the incidence increases rap-idly over the next 10 years, to a peak of about 3% per year after 15 years. It then declines to about 1% per year in patients with diabetes of 40 years’ duration or more. Those with diabetes for more than 35 years who have not yet developed nephropathy are at low risk of doing so. This pattern strongly suggests that only some patients are susceptible to renal disease in diabetes, probably because both genetic susceptibility and the cumulative effect of hyperglycaemia are necessary. In type 2 diabetes, both the classical form of diabetic nephropathy and an atypical type may occur.• The classical form often occurs in younger individu-als, and retinopathy and proteinuria with progression from microalbuminuria to overt proteinuria may be seen.• In the atypical form, retinopathy may be absent, and proteinuria may be absent or minimal. Although other renal disease should be considered and excluded, the kidney lesion is often related to hypertension/glomerular ischaemia and its treatment is largely the same. The cumulative risk of nephropathy in type 2 diabetes varies with ethnic origin, ranging from 25% in those of European origin to about 50% in other ethnic groups (e.g. Afro-Caribbeans, Asian Indians, Japanese). Asian Indians develop type 2 diabetes more commonly and at a younger age than Europeans, and the preva-lence of hypertension is significantly greater in Afro-Caribbeans; these factors are believed to contribute to the higher risk. Type 2 diabetes is more common than type 1, and the number of patients with type 2 diabetes requiring renal replacement therapy (dialysis or transplantation) exceeds that in type 1. Worse glycaemic control, higher blood pressure, smoking and adverse

lipid profile are risk factors for diabetic nephropathy in both type 1 and type 2 disease. In Europe, diabetes is the only increasing cause of end-stage renal failure.

Definition and detection

By convention, diabetic nephropathy is defined as the appearance of persistent ‘clinical’ albuminuria (albumin excretion rate, AER > 300 mg/24 hours) in an individual with diabetes for more than 5 years and concomitant retinopathy, in the absence of urinary tract infection (UTI), other renal diseases and heart failure. This process is often associated with increasing blood pressure. After initial stabilization of metabolic control, all patients, should be screened for albuminuria at least once per year. Uri-nary AER in a timed urine collection should be determined only in those with an albumin concentration of 20 mg/litre, or an albumin:creatinine ratio (ACR) of 2.5 mg/mmol or more in men or 3.5 mg/mmol or more in women. An elevated AER (≥ 20 µg/minute) should be confirmed in two further collections.

Clinical course

Diabetic nephropathy is a multi-stage condition that takes several years to become clinically overt (Figure 1). At the onset of diabetes, there are usually changes in renal function such as glomerular hyperfiltration, increased renal blood flow and hvpertrophy of the kidney. Most of these changes can be reversed at an early stage by good glycaemic control, but they persist in many patients and may be important in the later development of clinical nephropathy.

Stephen Thomas is Consultant Physician at Guy’s and St Thomas’

Foundation Trust, London, UK. Conflicts of interest: none declared.

GianCarlo Viberti is Professor of Diabetes in the Department of Diabetes,

Endocrinology and Internal medicine at GKT School of Medicine, King’s

College London, UK. Conflicts of interest: none declared.

Diabetic nephropathyStephen Thomas

GianCarlo Viberti

1

Natural history of diabetic nephropathy

Duration of diabetes (years)

0 5 10 15 20

Glomerular filtration rate

Serum creatinine

End-stage renal failureProteinuria ≥ 200 µg/minute

Microalbuminuria ≥ 20–< 200 µg/minute

What’s new ?

• Anaemia is increasingly recognized as occurring earlier in diabetic nephropathy than in other kidney diseases

• Statin therapy should be considered for all patients with diabetic nephropathy



Microalbuminuria: in the past, the definition of diabetic nephropathy was dictated by the lower limit of detection of avail-able urinary albumin assays. Development of sensitive assays enabled detection of previously subclinical increases in urinary albumin excretion, which were termed ‘microalbuminuria’. This is the first indication of diabetic nephropathy, and is defined as a persistent increase (at least two of three consecutive sterile, timed urine specimens) in urinary AER to 20−200 µg/minute (30−300 mg/day). It may be detected 1 year after the onset of diabetes in post-pubertal patients with type 1 disease, and at diagnosis in type 2. Microalbuminuria has also been termed ‘incipient nephropathy’. However, there is significant structural glomerular disease even at this early phase, and the glomerular filtration rate (GFR) starts to decline during the phase of micro-albuminuria, though it may remain within the normal range until the AER approaches 200 µg/minute (300 mg/day). Microalbuminuria is also strongly predictive of death from cardiovascular disease, particularly in older patients with type 2 diabetes (Figure 2). Furthermore, in those with type 1 diabetes who develop, microalbuminuria after a very long duration of disease, it is more a predictor of cardiovascular than progressive renal disease. Microalbuminuria is also associated with retinopathy (particularly in type 1 diabetes), peripheral vascular disease and neuropathy. Blood pressure increases in this phase (Figure 3), and lipid abnormalities develop including increased low-density lipoprotein cholesterol, total triglycerides and apoliprotein B, and reduced high-density lipoprotein (subclass 2) cholesterol. These progressive abnormalities are seen in both type 1 and type 2 diabetes, though in the latter they occur on a background of often pre-existing hypertension and dyslipidaemia. Microalbuminuria is also associ-ated with generalized endothelial dysfunction in both type 1 and type 2 diabetes.

Persistent albuminuria: an increase in AER to persistently more than 200 µg/min (> 300 mg/day) marks the onset of clinically defined ‘overt diabetic nephropathy’ and is a harbinger of renal failure and cardiovascular complications in both types of diabetes. Arterial hypertension usually occurs with persistent proteinuria, and over time the protein loss may increase to cause nephrotic syndrome with hypoalbuminaemia and peripheral oedema. Lipid disturbances and atherosclerotic complications are prominent in this phase. Diabetic retinopathy usually accompanies persistent proteinuria, and its absence should alert health professionals to the possibility of a non-diabetic cause of the proteinuria. It should be remem-bered that, in up to 10% of patients with type 1 diabetes and (in some series) 30% of those with type 2, proteinuria may have a non-diabetic cause.

Uraemia: persistent albuminuria is accompanied by a gradual decline in GFR. If untreated, this eventually leads to uraemia and death after an average of 7−10 years.2

3

Diabetic nephropathy is associated with cardiovascular mortality in type 2 diabetes

Source: Adler et al. Kidney Int 2000; 63: 225–32.

Normal albumin excretion

Microalbuminuria

Overt proteinuria

Elevated plasma creatinineor renal replacement therapy

Death

2.0%

2.8%

1.4%

3.0%

4.6%

19.2%

2.3%

Source: Microalbuminuria Collaborative Study Group. Diabet Med 1999; 16: 918.

HbA1c and mean arterial pressure are raised in microalbuminuria in type 1 diabetes

Months

Albu

min

exc

reti

on ra

te(µ

g/m

inut

e)M

ean

arte

rial

pre

ssur

e(m

m H

g)H

bA1c

(%)

Progressors Non-progressors

0

60

30

20

0

8

50

40

120

110

100

90

80

13

12

11

10

9

10

12 24 36 48 60 72 84 96



Diagnosis

A normal individual has:• urinary albumin concentration < 20 mg/litre• ACR < 2.5 mg/mmol in men, < 3.5 mg/mmol in women• AER < 20 µg/minute.

Screening tests (Figure 4): all patients over the age of 12 years should have their AER and blood pressure measured at least once per year. Measurement of the ACR in a mid-stream, first-morning urine sample is a reliable screening method. Alternatively, semi-quantitative dipstick tests have been developed to determine albumin concentration in the microalbuminuria range and may be used in the doctor’s office or by patients themselves as first-line screening. It should be remembered that heavy exercise, UTI, acute illness and cardiac failure can transiently increase the AER.

Confirmation: if the ACR is raised, the test should be repeated and, if confirmed, a timed urine collection overnight or for 24 hours should ideally be performed to determine the AER. An AER of 20−200 µg/minute (30−300 mg/day) in two of three urine samples tested within 6−12 weeks warrants the diagnosis of microalbuminuria, and more than 200 µg/minute (> 300 mg/day) diabetic nephropathy. Timed urine collections are unnecessary in those with a normal ACR, who should just be re-screened annu-ally. Progression of the albuminuria should be checked by timed urine collection (where facilities are available) or by repeated determination of early-morning ACRs, because the greater the degree of albuminuria the higher the renal and cardiovascular risk. In these patients, regular and more frequent check-ups should be undertaken to assess blood pressure, glycaemic control, serum lipids and serum creatinine.

Management

Glycaemic control: there is no doubt that good glycaemic control can prevent diabetic nephropathy in both type 1 and type 2 diabetes.

There is less evidence that, once microalbuminuria has developed, good glycaemic control prevents progression of the kidney lesion. Nevertheless, in type 1 diabetes, improved blood glucose control and intensified insulin treatment seem to prevent histological worsening of glomerulopathy in those with microalbuminuria, and worse glycaemic control is associated with a faster decline in GFR in those with established diabetic nephropathy. There remain no conclusive data in type 2 diabetes, but insulin treatment should be considered in those with microalbuminuria whose diabetes is poorly controlled on oral agents. In particular, once renal function is impaired, biguanides and renally excreted sulphonylureas such as glibenclamide must not be used.

Blood pressure: raised blood pressure has particularly damaging consequences for the kidney, heart and retina. Current recom-mendations suggest that blood pressure should be lowered to 130/80 mm Hg or less (some authorities recommend 125/75 mm Hg or less, particularly in younger patients) in those with micro-albuminuria or overt clinical nephropathy. Non-pharmacological interventions such as dietary and lifestyle changes (e.g. restriction of salt and alcohol intake, weight reduction, increased exercise) are important, though most patients require antihypertensives (often more than one agent) to achieve target blood pressures. First-line treatment is usually with an inhibitor of the renin−angiotensin system. These agents are effective blood pressure-lowering agents in those with diabetic nephropathy and are more antiproteinuric than other classes of agents. In type 1 diabetes, angiotensin-converting enzyme (ACE) inhibi-tors reduce progression from microalbuminuria to overt diabetic nephropathy and, in those who have already developed overt nephropathy, reduce progression to end-stage renal failure and premature mortality. In type 2 diabetes, ACE inhibitors reduce mortality in those with microalbuminuria, and angiotensin II receptor blockers reduce progression from microalbuminuria to overt nephropathy and progression of renal failure. There is no evidence that angiotensin II receptor blockers reduce mortality in patients with type 2 diabetes and renal failure, though they seem to

4

Screening for renal complications

Annual dipstick urinalysisAnnual albumin:creatinine ratio

Plasma creatinine > 150 µmol/litreEstimated glomerular filtration rate < 50 ml/minute

Refer to diabetes/renal clinic

Positive dipstick protein Negative dipstick protein

Positive dipstick proteinconfirmed x 2 (proteinuria)

Urine for albumin:creatinine ratio

< 2.5 in men, < 3.5 in womenAnnual albumin:creatinine ratio

> 2.5 in men, > 3.5 in womenConfirm within 3–6 months x 2 (microalbuminuria)

Start renin–angiotensin inhibitoror

Refer to diabetes clinic as appropriate



reduce heart failure. First-line treatment is usually an ACE inhibitor in those with microalbuminuria or diabetic nephropathy in both type 1 and type 2 diabetes, though given the greater amount of data, some authorities favour angiotensin II receptor antagonists in those with type 2 diabetes. There is evidence that a combination of ACE inhibitor and angi-otensin II receptor antagonist therapy may be beneficial in terms of both blood pressure lowering and perhaps particularly reduction of proteinuria, though large studies are lacking. After starting ACE inhibitor or angiotensin II antagonist therapy, serum creatinine and potassium should be closely monitored, because of the possible coexistence of renal artery stenosis. This is a particular problem in patients with peripheral vascular disease, or ‘hyporeninaemic pseudohypoaldosteronism – type IV renal tubular acidosis’, with resulting hyperkalaemia, particularly in those with type 1 diabetes. In addition, postural hypotension may be troublesome in patients who have concomitant autonomic neuropathy, and may limit antihypertensive treatment. More than one antihypertensive agent is likely to be required. In type 2 diabetes with microalbuminuria, there is evidence that the combination of a low-dose thiazide diuretic with a renin−angiotensin system inhibitor may be beneficial.

Serum lipids: from the phase of microalbuminuria, cholesterol and triglycerides may be elevated in both type 1 and type 2 dia-betes. There are no long-term data on the effect of lipid-reducing interventions; however, dietary restriction, weight reduction and improved metabolic control should be considered in all cases. In type 2 diabetes, recent data suggest that statin therapy should be considered in those with any complications of diabetes, including microalbuminuria, and it seems reasonable to extrapolate this to type 1 diabetes with microalbuminuria/overt nephropathy, given the association with premature cardiovascular disease.

Smoking is associated with the development and progression of diabetic nephropathy and with cardiovascular disease, and should be discouraged in all patients.

Protein restriction is controversial, but has been shown to have a beneficial effect in type 1 diabetes patients with overt nephropathy. Reduction in animal protein intake to 0.6−0.7 g/kg/day should be considered. Replacement of animal by vegetable protein sources may also be considered, because vegetable protein seems to be less damaging to the kidney. An expert nutritionist should supervise all such treatment, and care should be taken to minimize any potential detrimental effects of low-protein diets. No prospective data are available for type 2 diabetes.

Associated complications: microvascular and macrovascular complications may progress rapidly in patients with clinical albuminuria. Retinopathy, neuropathy and atherosclerotic com-plications should be monitored more often and any abnormalities treated promptly. Anaemia is common in diabetic nephropathy and may occur at an earlier stage than in other kidney diseases; this is particularly the case in type 1 diabetes, and in women.

Nephrological referral: current guidelines suggest joint treatment by diabetes and renal specialists in patients whose serum creatinine is more than 150 µmol/litre. Serum creatinine measurements

may underestimate the degree of renal impairment; an estimated GFR should be calculated and patients with values of 30−50 ml/minute referred to a joint diabetes/renal clinic, depending on local arrangements. Referral to a joint clinic may be necessary earlier for management of anaemia. Late referral (within 6 months of the need for renal replacement therapy) is associated with higher mortality.

End-stage renal failure − the uraemia phase: diabetes should not exclude patients from renal replacement therapy programmes. The mode of therapy (dialysis or transplantation) depends on clini-cal judgement and local facilities and resources. Good glycaemic control is important for patients’ well-being before and during renal replacement therapy and is associated with lower mortality. In specialized centres, combined kidney and pancreas transplan-tation may be considered, to maintain euglycaemia and prevent recurrence of glomerulopathy in the transplanted kidney. During the azotaemic phase, interpretation of HbA1c, and particularly fructosamine (and thereby their use as monitoring indices for diabetes control), may be less reliable. HbA1c is prob-ably preferable, but should be interpreted with caution. Recent data suggest that careful cardiovascular assessment (including correction of significant coronary lesions) may improve life expectancy, particularly in patients undergoing transplantation. However, the outcome of renal replacement therapy remains poorer in patients with diabetic chronic renal failure than in those with non-diabetic disease. Furthermore, the associated vascular, neu-ropathic and infective complications increase the socioeconomic costs of renal replacement therapy in diabetes.

Pregnancy and proteinuria

Development of microalbuminuria or macroalbuminuria during pregnancy in a woman with diabetes should alert the physician to the risk of pre-eclampsia. Pregnancy is no longer contraindi-cated in women with diabetic proteinuria; however, proteinuria may increase and the risk of eclampsia is increased. Diabetes in pregnancy is discussed further on page 111.

FURTHER READINGCooper M E. Pathogenesis, prevention, and treatment of diabetic

nephropathy. Lancet 1998; 352: 213−19.

Gaede, P, Vedel, P, Larsen, N et al. Multifactorial intervention and

cardiovascular disease in patients with type 2 diabetes. N Engl J Med

2003; 348: 383−93.

Heart Outcomes Prevention Evaluation (HOPE) Study Investigators.

Effects of ramipril on cardiovascular and microvascular outcomes in

people with diabetes mellitus: results of the HOPE study and MICRO-

HOPE substudy. Lancet 2000; 355: 253−9.

Lewis E J, Hunsicker L G, Bain R P et al. The effect of

angiotensin-converting-enzyme inhibition on diabetic nephropathy.

The Collaborative Study Group. N Engl J Med 1993; 329: 1456−62.

Microalbuminuria Collaborative Study Group. Predictors of the

development of microalbuminuria in patients with type 1 diabetes

mellitus: a seven-year prospective study. Diabet Med 1999; 16: 918.