|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
In-Depth Review |


* Nephrology;
Endocrinology, Department of Biomedical and Surgical Sciences, University Hospital of Verona, Verona; and
Division of Nephrology, "Brotzu" Hospital, Cagliari, Italy
Correspondence: Prof. Giovanni Gambaro, Divisione di Nefrologia, Dipartimento di Scienze Biomediche e Chirurgiche, Università di Verona, Ospedale Maggiore, P.le Stefani 1, 37126 Verona, Italy. Phone: +39-045-8122521; Fax: +39-045-915176; E-mail giovanni.gambaro{at}univr.it
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Demographic changes in Western populations also have a significant impact on renal care. In many Western countries, the elderly now represent the most rapidly growing population initiating dialysis. In the United States, the median age of the incident population starting renal replacement therapy reached 64.6 yr in 2005, and the incidence of ESRD by age >75 yr was approximately 1500 pmp, 750 in Canada, and 500 in Italy (4). Because the prevalence of diabetes has been increasing among the elderly, it is hardly surprising that elderly individuals with diabetes contribute substantially to the overall burden of stage 5 chronic kidney disease (CKD) in Western countries. In 2005, in fact, one third of the incident cases of ESRD in the United States in people who were older than 75 yr had diabetic kidney disease (4).
Two very recently released publications are also of interest in the present contention, the Kidney Disease Outcomes Quality Initiative (KDOQI) Clinical Practice Guidelines and Clinical Practice Recommendations for Diabetes and Chronic Kidney Disease (5) and the Quality Indicators for the Care of Vulnerable Elders (the Assessing Care of Vulnerable Elders 3 [ACOVE-3] initiative) (6). Only the former marginally addresses the issue that elderly people with diabetes and CKD warrant special treatment considerations, but neither of them nor the EDAG and ADAG specifically addresses the topic of advanced CKD in older patients with diabetes. This review aims to make up for this shortcoming relating to the treatment of elderly people with diabetes and advanced renal impairment (i.e., in stages 3 and 4 CKD). As the lower age limit, we consider patients between 70 and 80 yr because such long-lasting patients with diabetes are most likely to have a considerably reduced life expectancy and a number of diabetes-related comorbidities. The relevant clinical issue here, however, is the patients vulnerability, not their age, so our considerations could also be extended to younger, frail patients with diabetes and stages 3 to 4 CKD.
An attempt has been made to quote and grade the level of evidence. There is a sizable amount of good-quality research on the prevention and management of the complications of type 2 diabetes and related aspects, but most of it is not specific to elderly patients, and none of these studies concerned elderly people with advanced CKD. Most of the suggestions have consequently been classified simply as clinical practice recommendations (CPR), at the level of "opinion-based" advice, whereas a clinical practice guideline (CPG) has been proposed in just one case.
| Treating Elderly Patients with Diabetes and Stages 3 to 4 CKD: Limits and Perspectives of Current Guidelines |
|---|
|
|
|---|
Although they do not address stages 3 to 4 CKD, the ADAG (2) and the EDAG (1) do consider these issues in general terms, following the assumption that treatment for elderly patients with diabetes needs to focus on particular issues and different priorities; the guidelines emphasize what can be reasonably treated and afforded in this particular age group. The ADAG also introduce the concept of the time horizon for the benefits of certain measures: It may take as long as 8 yr for glycemic control to have a positive fallout on microvascular complications or only 2 to 3 yr for the benefits of a better control of BP and lipids to become apparent (2).
Several of these elderly patients with diabetes are frail. According to the ACOVE project, a vulnerable elder is a person who is older than 65 yr and is at increased risk for death or functional decline in a 2-yr period (6). Patients with diabetes and stages 3 to 4 CKD considered in this review most likely belong to this category. This is because advanced CKD dramatically reduces life expectancy and is quite frequently associated with macrovascular clinical manifestations and neuropathy in patients with diabetes (11,12). Cardiovascular mortality and morbidity rates are extremely high in patients with diabetes and CKD (13). The EDAG emphasize that ischemic heart disease (IHD) is the most common cause of death in elderly people with type 2 diabetes, but peripheral artery disease, the strongest mortality predictor in patients with ESRD (14), and heart failure (HF) (12) are also very prevalent in these patients. It is obvious from these considerations that the treatment agenda for elderly patients with type 2 diabetes and stages 3 to 4 CKD needs to be reconsidered, and key issues in such a reassessment will concern treatment tailoring and simplification and prioritization.
| Treatment Agenda |
|---|
|
|
|---|
Metabolic Control
The recent KDOQI document (5) emphasizes that much of the support for strict metabolic control to improve kidney outcomes in patients with diabetes arises from the favorable effect on the onset of microalbuminuria, whereas there is only weak evidence of a beneficial effect on progression to more severe stages of CKD. The EDAG also emphasize that there is no clear evidence that strict glycemic control is effective in preventing macrovascular (cardiovascular) complications in middle-aged patients with type 2 diabetes. Although several studies involving older patients demonstrated high complication and mortality rate correlating with poor blood-sugar control (15), no randomized, controlled trials have assessed the impact of optimal glucose control on primary cardiovascular disease prevention for the older patients with diabetes, whereas in younger patients, it is unsuccessful (16) (Table 1) or may even increase cardiovascular mortality (17).
|
Treatment with an insulin secretagogue (normally a sulfonylurea) or metformin is considered the front-line therapy for older people with type 2 diabetes. Care is needed in prescribing oral hypoglycemic agents for patients with decreased renal function, however, and some should be avoided altogether because of the risk for persistent and severe hypoglycemic episodes (chlorpropamide, glyburide/glybenclamide) or life-threatening lactic acidosis (metformin). The current practice in such cases is to switch patients from oral therapy to insulin treatment because of the hypoglycemic risk, although some national guidelines, including the French ALFEDIAM-SFN report, suggested that oral antidiabetic agents may be considered on a case-by-case basis (9), and the recent KDOQI document (5) considered the possibility of using glipizide and gliclazide without any dosage adjustment in stages 3 and 4 CKD. Oral hypoglycemic agents are currently used to treat 8% of prevalent patients with ESRD in Canada and 30% in France (19,20). Insulin treatment should also be considered with caution, however, in terms of the hypoglycemic risk, because exogenous insulin is eliminated primarily by the kidney. In French patients with diabetes, the incidence of sulfonylurea-related severe hypoglycemic episodes was 1.2% patient/yr, whereas for sulfonylurea and insulin in combination, it was 2.7%, and for insulin alone, it was 3% (20). The vast majority of these episodes are associated with old age, errors in drug dosage, drug interactions, and the initial period of treatment, whereas renal impairment is less important (20). Death, which happens in 4.9 to 12% of cases of severe hypoglycemia, occurs only in patients with severe concomitant diseases (21).
Renal impairment has not been reported as notably predisposing patients to severe hypoglycemia induced by the sulfonylurea glipizide, the clearance and half-life of which are unaffected by renal function (22). Thus, if a sulfonylurea is used for patients with stages 3 to 4 CKD, the best choice is glipizide without dosage adjustment.
Unlike nateglinide, the nonsulfonylurea insulin secretagogue repaglinide can also be used in stages 3 to 4 CKD without any dosage adjustment (5,23). No relationship was found between the degree of renal impairment (GFR as low as 20 ml/min) and the risk for hypoglycemia in patients who were treated with repaglinide (23).
Thiazolidinediones are interesting oral agents that reduce insulin resistance. Adverse effects include fluid retention, edema, and congestive HF. The consensus is to exclude patients with New York Heart Association class 3 and 4 HF from this treatment. Two recent meta-analyses (24,25) suggested that rosiglitazone is associated with a significant increase in the risk for myocardial infarction (MI) and HF. A third meta-analysis on patients who were treated with pioglitazone reached partly different conclusions, because its use was associated with a significantly lower risk for death, MI, or stroke among patients with diabetes, but the risk for severe HF increased (26). Because both rosiglitazone and pioglitazone are almost entirely metabolized by the liver and the major metabolites do not accumulate in patients with stages 3 to 4 CKD, no dosage adjustment is necessary. Although pioglitazone was safe and well tolerated in a small group of patients with varying degrees of CKD (27), and according to a post hoc analysis of the PROactive trial patients who had diabetes and GFR <60 ml/min and were treated with pioglitazone were less likely to develop major cardiovascular events (28), and finally combined therapy with rosiglitazone and sulfonylureas was well tolerated in 301 patients with diabetes and GFR 30 to 80 ml/min (29), we believe that for the time being, the results of the three meta-analyses advise against their generalized use in patients with such high cardiovascular risk as those discussed here (CPR, opinion). As for acarbose, because experience is lacking on its long-term use in patients with advanced CKD, its generalized use is not recommended for the time being (5,20) (CPR, opinion) although no major adverse effect has been reported.
With regard to insulin, it is eliminated primarily by the kidney. The pharmacokinetics of various insulin preparations have not been studied sufficiently in patients with varying degrees of renal dysfunction, but the new long- and short-acting insulin analogs do not seem to have any better pharmacokinetics in patients with CKD. To reduce the risk for hypoglycemic episodes, insulin dosage reduction of 25% has been recommended in renal patients with GFR 10 to 50 ml/min, but, when the long-acting insulins glargine or detemir are used, it may be wise to start treatment with 50% of the usual initial dosage of 0.1 U/kg, titrating the dosage until target fasting glucose concentrations are reached. In "treat-to-target" studies, the addition of a basal insulin (glargine) to regimens with oral medication significantly improved metabolic control, resembling the effect of neutral protamine Hagedorn insulin, while reducing the number of nocturnal hypoglycemic episodes (30).
In our opinion, when oral treatment with glipizide, gliclazide, or repaglinide is not sufficient, the use of acarbose and, in selected cases, of pioglitazone should be considered to minimize the emotional burden associated with insulin therapy. If, notwithstanding the less stringent targets for metabolic control in these patients, oral treatment alone is not adequate, then the combination with the insulin analog glargine in the evening might be considered as the front-line treatment for elderly patients with diabetes and stages 3 to 4 CKD, rather than multiple insulin injections, given the need to simplify the treatment as much as possible (CPR, opinion).
Preventing Advanced CKD Progression to ESRD versus Preventing Cardiovascular Outcomes
Individually, both diabetes and CKD are cardiovascular risk factors, but together they synergistically increase cardiovascular mortality (13), so patients with type 2 diabetes and nephropathy are more likely to die than to progress to more severe CKD stages (11). What is peculiar to elderly patients with diabetes and CKD by comparison with younger patients with same clinical conditions is the time horizon, which may leave little scope for a worthwhile prevention of CKD progression, although there may be time enough for the cardiovascular prevention.
The EDAG suggest tailoring a multitarget approach to elderly patients with diabetes according to the severity of cardiovascular risk, but there is no doubt that elderly patients with diabetes and CKD belong to the very high-risk group (11,12). Very few trials, if any, have investigated the effect of the usual treatments for containing cardiovascular risks in elderly people or patients with CKD, and medications that are used in the general population for cardiovascular event prevention are underprescribed in patients with CKD, especially when they are elderly (31) and have diabetes (32), although observational studies suggested that elderly patients with stages 3 to 4 CKD or diabetes benefit from such standard treatments at least as much as patients with a preserved renal function. Furthermore, cumulative incidence curves from lipid- and BP-lowering trials indicated that cardiovascular benefits start after 2 to 4 yr of treatment, which makes these treatments worthwhile even for patients with a short life expectancy (33). These observations show that a multifaceted approach to cardiovascular risk factors is even more essential to optimize cardiovascular outcomes in elderly patients with diabetes and CKD than in other age groups (CPR, moderate).
BP Control
Controlling hypertension may, in theory, be worthwhile to contain both the progression of nephropathy and the cardiovascular risk in the patients with whom we are dealing. Judging from the KDOQI analysis (5), however, most of the evidence (classified as weak) supporting a favorable effect of BP control on CKD progression in patients with diabetes is supported by trials of patients with stages 1 to 2 CKD. Only angiotensin II receptor blockers (ARB) have been tested on patients with type 2 diabetes and more severe CKD, including stage 3, and shown to reduce GFR decline, but this seems to be partially independent of BP changes and due to a specific nephroprotective effect (34,35). Conversely, according to a meta-analysis of approximately 1900 patients who had diabetes (most of them with stages 1 and 2 CKD) and were treated with angiotensin-converting enzyme inhibitors (ACEI), the evidence that ACEI reduce the risk for stages 3 to 5 CKD is weak (36).
As regards elderly patients with diabetes in particular, unfortunately no evidence-based medical conclusions (EBM) can be drawn on the progression of renal disease from the available clinical trials. Because of its long follow-up, only the UK Prospective Diabetes Study (UKPDS) helps to some degree. The most significant results were obtained on cardiovascular complications rather than on microangiopathies, and strict BP control proved better for their prevention than strict metabolic control (16,37) (Table 1). Thus, in patients with type 2 diabetes and stages 3 to 4 CKD, the available evidence in favor of controlling hypertension seems to be particularly important to the prevention of cardiovascular risks.
The benefits of a lower BP on the cardiovascular risk in the very old is confirmed by a meta-analysis on trials conducted on antihypertensive agents in the subgroup of patients who were older than 80 yr, which concluded that treatment reduces the risk for strokes, major cardiovascular events, and HF by 34, 22, and 39%, respectively (38), hence the EDAG statement that in patients who are older than 75 yr and have type 2 diabetes BP control should never take second place vis à vis other therapeutic options (e.g., blood glucose control). We believe that this should be extended to elderly patients with diabetes and advanced CKD (CPR, opinion).
Both the ADAG and the EDAG suggest that the threshold for treating high BP in older people with type 2 diabetes should be 140/80 mmHg. This is based on the likelihood of reducing the cardiovascular risk in older individuals balanced against issues relating to tolerability, clinical factors and disease severity, and targets that are likely to be achievable. According to the EDAG, a lower BP should be aimed for in cases with concomitant CKD, but for vulnerable patients, when HF and stroke prevention is more important than slowing the progression of microvascular complications, BP <150/90 mmHg are proposed as acceptable; however, the recent ACOVE-3 guidelines suggested for vulnerable elderly patients different systolic targets according to different clinical conditions but anyway <140 mmHg (39). Thus, although most elderly patients with diabetes and stages 3 to 4 CKD likely belong to the category of frail patients, a goal of 140/90 seems to be reasonable with exceptions depending on tolerability and the patient's safety (CPR, opinion).
As for the choice of antihypertensive drugs for elderly patients with diabetes and nephropathy, the EDAG recommend ACEI. The use of short-acting calcium channel blockers and
-adrenoreceptor blockers is discouraged. The ADAG do not take a stand on the priority of certain classes of antihypertensive agents, whereas they do highlight that, according to EBM (36), diuretics, ACEI, β blockers, and calcium antagonists have comparable effectiveness in reducing cardiovascular morbidity and mortality. The ACOVE-3 discourages the use of first- or second-generation calcium antagonists (verapamil, diltiazem, and nifedipine) in patients with HF because of safety concerns (40).
Elderly patients with diabetes and stages 3 to 4 CKD are likely to need more than one agent to achieve the target BP, however. We suggest that diuretics be considered whenever possible (CPG, strong). The long-term results of the Systolic Hypertension in Europe (SHEP) trial are quite impressive (41); even a modest dosage of a mild diuretic was sufficient to reduce cardiovascular and all-cause mortality, particularly in patients with diabetes. When it comes to choosing additional agents, we should also consider β blockers for those with HF or IHD (see the section on β-blockers on the next page).
The EDAG do not rely on the EBM emerging from the Reduction of Endpoints in NIDDM with the Angiotensin II Antagonist Losartan (RENAAL) study (34) or the Irbesartan Diabetic Nephropathy Trial (IDNT) (35), because they recommended ACEI (consensus-based statement) for elderly patients with type 2 diabetes and nephropathy and ARB only when the former are not tolerated or are contraindicated. The ACOVE-3 suggested instead that either ACEI or ARB be used in patients with diabetes and CKD, especially when they have a history of heart disease. ARB have proved capable of slowing the progression of advanced CKD in patients with type 2 diabetes (34,35), although they were not powered to demonstrate an effect on cardiovascular end points. After the publication of some challenging meta-analyses, there is much debate on the superiority of ACEI over ARB in terms of preventing cardiovascular mortality (42), and even a rough comparison of the effect of different agents on cardiovascular risk in the subgroup of elderly patients with type 2 diabetes from several controlled trials on hypertension showed that lowering BP reduced the cardiovascular risk whatever drug was being used, but ARB were not the most effective (Table 2). We therefore agree with the EDAG statement that, in patients such as those discussed here, for whom preventing cardiovascular outcomes is more important than delaying progression to ESRD, ACEI should be first-line therapy and ARB should be the next best choice when the former cannot be used.
|
Controlling Cholesterol Levels
High blood lipid levels are a known independent cardiovascular risk factor. Previous statin trials indicate that an absolute reduction in LDL cholesterol produces proportionally comparable cardiovascular risk reductions in older and younger people, with or without CKD. Benefits may become apparent quite early. In the Heart Protection Study, which included patients who had diabetes and were between 40 and 80 yr and were treated with simvastatin 40 mg/d, a reduction in the risk for MI and stroke was observed within <12 mo, regardless of age or serum creatinine levels (48). Although they did not analyze the effect of age, two meta-analyses of pravastatin trials showed that the treatment was equally effective on mortality and cardiovascular end points in patients with and without CKD (at least up to stage 3 CKD), with and without diabetes (49,50). According to the KDOQI document (5), all patients with diabetes and stages 1 through 4 CKD should be treated, although there is no evidence to support treating stage 4 CKD, and the ACOVE-3 firmly recommended dyslipidemia treatment in all frail patients with diabetes (CPR, opinion).
Aspirin
In a systematic review of secondary prevention trials, the Antithrombotic Trialists Collaboration (51) determined the effects of antiplatelet therapy on cardiovascular death, MI, and stroke in various categories of patients: Taking all high-risk categories together, aspirin therapy reduced nonfatal MI, nonfatal stroke, and vascular death by approximately one third, irrespective of middle or old age or diabetic condition; the effect was much the same at dosages ranging from 75 to 325 mg/d. According to the Hypertension Optimal Treatment (HOT) trial, patients who have diabetes and were aged 50 to 80 yr may benefit, in terms of major cardiovascular events, even more than patients without diabetes from treatment with aspirin 75 mg/d (52). The EDAG, ADAG, and ACOVE-3 thus suggested that older adults who have type 2 diabetes and are not on anticoagulant therapy and have no contraindications should be offered low-dosage aspirin. This should be the rule for renal patients, too, although there is no EBM for elderly patients with diabetes and renal insufficiency (CPR, opinion). Nephrologists often tend to avoid the use of aspirin because they are worried about the risk for bleeding in patients with advanced CKD, given their uremic platelet dysfunction. One of the major causes of this dysfunction is anemia, because erythrocytes play a significant part in tracking platelets to the endothelium. Nowadays, anemia is no longer a problem in renal patients, so the uremic thrombopathy issue seems to have become less important. A number of observations confirm that bleeding complications are no longer a significant problem of antiaggregation in renal patients. A Spanish group (53) showed that aspirin treatment coincided with a three-fold risk for bleeding in patients who were on hemodialysis (versus a two-fold risk in the general population), but optimal anemia correction was associated with reduced risk for hemorrhage. Furthermore, the US Renal Data System Dialysis Morbidity and Mortality Studies paper on the follow-up of >30,000 patient-years in patients with ESRD reported that gastrointestinal bleeding was not associated with aspirin use (54). Moreover, according to the Antithrombotic Trialists Collaboration meta-analysis (51), patients who were on hemodialysis and were treated with low-dosage aspirin showed a statistically significant 41% reduction in major cardiovascular end points with only 2% of cases of major extracranial bleeding (very similar to the 2.3% in the general population).
β-Blockers
Large, controlled trials have shown that β blockers are highly effective in reducing the risk for cardiovascular events and death in patients who have diabetes and have sustained an MI (55). The UKPDS study on patients with type 2 diabetes and hypertension showed that β blockade was at least as effective as ACE inhibition in preventing all primary macrovascular end points (37). The use of β blockers in patients with diabetes has generally been restricted because they are believed to impair hypoglycemia counterregulatory mechanisms; available evidence suggests, however, that β blockers are not associated with any increase in the hypoglycemic risk (19,37,56).
According to the ACOVE-3, in vulnerable elderly patients with HF or IHD, selective β blockers, such as carvedilol, metoprolol, and bisoprolol, should be offered unless the patient has a documented contraindication. Peripheral artery disease may be a contraindication to the use of β blockers, but their use is nonetheless recommended in patients with concomitant coronary heart disease (57). We therefore suggest that elderly patients with diabetes and stages 3 to 4 CKD receive β blockers unless they are contraindicated (CPR, opinion). The dosage of β blockers needs to be adapted to GFR.
Controlling Anemia
Anemia in patients with stages 3 to 4 CKD is associated with left ventricular hypertrophy, cognitive impairment, and increased general hospitalization and mortality (58). Early, appropriate anemia treatment with erythropoietin and high hemoglobin targets (13 to 15 g/dl) were thus proposed to prevent cardiovascular and noncardiovascular morbidity and mortality in patients with diabetes and CKD (58,59); however, a meta-analysis of 5143 patients also disclosed a significantly higher risk for all-cause mortality (20% more) in the higher (>13.5) than in the lower hemoglobin (<11.5) target group (60). In fact, after reviewing the latest results from randomized, controlled trials, the US National Kidney Foundation recently revised the 2006 KDOQI guidelines (61) and released (62) an opinion-based statement that hemoglobin targets should generally be in the range of 11.0 to 12.0 g/dl, independent of age and diabetes status.
Cautious Use of Potentially Nephrotoxic Agents or Procedures
Frail patients have very vulnerable kidneys, and renal function may suddenly deteriorate after various medical treatments and maneuvers. For instance, a body of evidence shows that elderly patients with diabetes may benefit like the rest of the population from a full range of cardiovascular treatments (including surgery and minimally invasive techniques), but they are at greater risk for a decline in renal function after the use of contrast agents. Adequate patient preparation before administration of contrast agents and the use of limited quantities of contrast agent may reduce this risk. A watchful use or even the temporary suspension of ACEI and ARB in conditions that favor dehydration or with the concomitant use of nonsteroidal anti-inflammatory drugs is also very important to prevent acute renal decompensation (63) and/or life-threatening hyperkalemia. Finally, any use of antialdosterone agents and nonsteroidal anti-inflammatory drugs must be very prudent, because of the risk for hyperkalemia.
Low-Protein Diet?
Low-protein diet may delay the progression of diabetic CKD, but this is not the priority of treatment in elderly patients with diabetes, and its effect is too modest and takes too long to come about, going beyond the patient's time horizon. Even the general guidelines for patients with diabetic CKD do not recommend any restriction in protein intake (8) or advocate only a moderate reduction (0.7 to 0.9 g/kg body wt per d) provided that the caloric intake is adequate (5,7). It is also worth bearing in mind that elderly people commonly reduce their protein intake spontaneously, so it seems unreasonable to recommend any protein restriction to elderly patients with diabetes and CKD (CPR, opinion).
| Conclusions |
|---|
|
|
|---|
|
| Disclosures |
|---|
|
|
|---|
| Acknowledgments |
|---|
We thank Prof. Giuseppe Maschio for critical reading of this article.
| Footnotes |
|---|
| References |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |