Secondary prevention of cardiovascular disease in end-stage kidney disease (dialysis)

INTRODUCTION — Cardiovascular disease (CVD) accounts for the majority of deaths in patients with end-stage kidney disease (ESKD) [1]. Of these deaths related to CVD, the majority are due to arrhythmia/sudden cardiac death, whereas a minority are attributable to coronary heart disease (CHD), stroke, and heart failure. (See "Patient survival and maintenance dialysis" and "Chronic kidney disease and coronary heart disease".)

Compared with the general population, patients undergoing maintenance dialysis have a significantly increased incidence of CVD. This is due to both an increased prevalence of traditional risk factors for CVD and risk factors due to the severe loss of kidney function. (See "Risk factors and epidemiology of coronary heart disease in end-stage kidney disease (dialysis)" and "Chronic kidney disease and coronary heart disease".)

A discussion of risk-factor modification for CVD by therapeutic lifestyle changes and drug therapies among patients undergoing maintenance dialysis will be presented in this topic review. Although many of the issues relating to CVD are presumed similar in patients with and without kidney failure, many traditional CVD risk factors and their management have not been adequately studied in the ESKD population. This topic review will emphasize those features that distinguish the patient on dialysis from those without kidney dysfunction. Discussions of this issue in patients with chronic kidney disease (CKD) who are not yet on dialysis as well as those without kidney disease are presented separately. (See "Chronic kidney disease and coronary heart disease" and "Prevention of cardiovascular disease events in those with established disease (secondary prevention) or at very high risk".)

CORONARY HEART DISEASE RISK EQUIVALENT — Some patients without known coronary heart disease (CHD), such as those with diabetes, have a risk of subsequent cardiovascular events that is comparable with that seen in patients with established coronary disease. Such patients are regarded as having a CHD risk equivalent.

Practice guidelines from several national organizations recommend that chronic kidney disease (CKD), including end-stage kidney disease (ESKD), be considered a CHD risk equivalent. This is based upon evidence that advanced CKD is associated with an increase in CHD risk [2]. However, whether CKD per se, in the absence of other traditional risk factors, accelerates atherogenesis is uncertain. Not all experts agree with the generalization that CKD is a CHD risk equivalent for all persons with CKD. (See "Chronic kidney disease and coronary heart disease".)

OVERVIEW OF SECONDARY PREVENTION — Although the efficacy of preventive measures for coronary heart disease (CHD) in patients on dialysis is often uncertain, we recommend risk-factor modification for patients on dialysis. In patients with established CHD, these measures are referred to as "secondary prevention" or prevention of complications due to established disease.

Evidence of benefit with cardiovascular risk factor modification has largely been shown in patients without chronic kidney disease (CKD). By comparison, few interventions in patients on dialysis have been examined in large, randomized trials, and even fewer have shown to be effective. Risk factor modification for secondary prevention in patients on dialysis will be reviewed here, with the evidence examining each of the possible modalities.

Detailed discussions of each risk factor in those without kidney disease and the efficacy of primary prevention of cardiovascular disease (CVD) are presented separately. (See "Prevention of cardiovascular disease events in those with established disease (secondary prevention) or at very high risk" and "Overview of primary prevention of cardiovascular disease".)

LIPID MODIFICATION — Among patients without chronic kidney disease (CKD), there is a graded relationship between the serum total cholesterol concentration and coronary risk, and the absolute risk is greater in patients with previous manifestations of coronary heart disease (CHD). During the past few decades, many large, randomized trials in patients with normal and/or modestly reduced kidney function have demonstrated the benefit of cholesterol lowering, largely with statin therapy and more recently with evolocumab, for secondary prevention. These issues are discussed in detail separately. (See "Management of low density lipoprotein cholesterol (LDL-C) in the secondary prevention of cardiovascular disease".)

Available evidence, however, calls into the question whether the strategy of using statin therapy for secondary cardiovascular disease (CVD) prevention is applicable to patients on maintenance dialysis. Specifically, two large, well-designed trials, Die Deutsche Diabetes Dialyse (4-D) and A Study to Evaluate the Use of Rosuvastatin in Subjects on Regular Hemodialysis: An Assessment of Survival and Cardiovascular Events (AURORA), evaluated the effect of statin therapy in patients on dialysis on the combined endpoint of death from cardiovascular causes, nonfatal myocardial infarction (MI), and stroke. Despite a substantial decrease in serum low-density lipoprotein (LDL) cholesterol levels, both trials found that the initiation of statin therapy provided no cardiovascular benefit in primary analyses. In apparent contrast, the Study of Heart and Renal Protection (SHARP) trial showed a benefit of pharmacologic therapy using the combination of statin and ezetimibe in the entire study cohort, which included patients with nondialysis CKD and patients on dialysis. The beneficial effect on atherosclerotic events was not statistically significant in the dialysis subgroup. However, the SHARP trial was not powered to detect an effect on subgroups, and there was no evidence that the effect differed according to the presence of end-stage kidney disease (ESKD) at baseline [3]. Thus, the SHARP trial may be interpreted as showing efficacy of LDL cholesterol lowering in cardiovascular benefit in patients on dialysis.

The effect of proprotein convertase subtilisin/kexin type 9 (eg, evolocumab) inhibitors on cardiovascular outcomes in patients with ESKD is unknown. (See "PCSK9 inhibitors: Pharmacology, adverse effects, and use".)

Trials of statin therapy — The issue of whether statins are effective in lowering the risk of a cardiovascular event in patients on dialysis was addressed in the 4-D, AURORA, and SHARP trials.

4-D trial — In the 4-D study, 1255 patients on hemodialysis (80 percent were not being treated with a statin) with type 2 diabetes and elevated serum LDL cholesterol levels were randomly assigned to placebo or atorvastatin (20 mg/day) [4]. The primary outcome measure was the composite of cardiovascular death, nonfatal MI, and stroke. After four weeks, atorvastatin successfully lowered LDL cholesterol (121 to 72 mg/dL [3.1 to 1.9 mmol/L]) versus no change with placebo (125 to 120 mg/dL [3.2 to 3.1 mmol/L]).

At a median follow-up of four years, however, there was no difference in the incidence of the primary outcome between both groups (relative risk [RR] 0.92, 95% CI 0.77-1.10). There was, however, a significant reduction in the rate of cardiac events (RR 0.82, 95% CI 0.68-0.99). On the other hand, there was also a significant and apparently paradoxical increase in the rate of fatal stroke (RR 2.03, 95% CI 1.05-3.93), but the absolute number of stroke events was small in both the statin and the placebo groups.

A post-hoc analysis that stratified patients by baseline LDL cholesterol concentration demonstrated a reduction in the primary outcome among those with a baseline LDL cholesterol concentration >145 mg/dL (3.76 mmol/L) (hazard ratio [HR] 0.69, 95% CI 0.48-1.00) [5]. Among such patients, there were also reductions in risk for cardiac death, sudden cardiac death, nonfatal MI, and for all cardiac deaths combined. There was no difference between groups among patients with baseline LDL cholesterol <145 mg/dL. Importantly, this analysis was performed in a subgroup that was not predefined, which limits its interpretation.

AURORA trial — In the AURORA trial, 2776 patients on hemodialysis who were not being treated with a statin were randomly assigned to rosuvastatin (10 mg/day) or placebo [6]. At three months, mean serum LDL cholesterol levels were lowered significantly with rosuvastatin (100 to 58 mg/dL [2.6 to 1.5 mmol/L]) versus no change with placebo (99 to 98 mg/dL [2.56 to 2.53 mmol/L]).

Nonetheless, at a median follow-up period of 3.8 years, the incidence of the primary composite endpoint (death from cardiovascular causes, nonfatal MI, or nonfatal stroke) was similar in the two groups (9.2 versus 9.5 events per 100 patient-years; HR 0.96, 95% CI 0.84-1.11). The individual components of the primary composite endpoint as well as all-cause mortality were also not significantly different between the two groups. Active therapy did not provide benefit for any prespecified subgroups, including those with diabetes or elevated C-reactive protein levels. Rosuvastatin was not associated with an increased incidence of adverse events.

A post-hoc analysis of 731 patients with diabetes from AURORA also showed no difference in the composite outcome or in the risk of stroke [7]. However, there was a 32 percent reduction in cardiac events (including cardiac death and nonfatal MI) among patients with diabetes assigned to receive rosuvastatin (n = 388) compared with placebo (n = 343; HR 0.68, 95% CI 0.51-0.90) [7]. The rosuvastatin group had a higher incidence of hemorrhagic stroke compared with placebo, although the number of events was small (12 versus 2, respectively). Similar to the 4-D study, these data suggest a possible benefit of statins in the reduction of cardiac events among patients with diabetes who are on dialysis, although this is only a subgroup analysis [8].

SHARP trial — The SHARP trial evaluated the efficacy of simvastatin plus ezetimibe compared with placebo in lowering cardiovascular morbidity in patients with CKD, approximately one-third of whom required maintenance dialysis [3]. The results dealing with the patients with CKD who were not on dialysis are discussed separately. (See 'Trials of statin therapy' above and "Lipid management in patients with nondialysis chronic kidney disease".)

Originally, patients were randomly allocated to three treatment arms: a combination pill containing 20 mg of simvastatin and 10 mg of ezetimibe, 20 mg of simvastatin only, or placebo. After one year of follow-up, however, the patients in the simvastatin-only arm were reassigned to receive either the combination pill (containing simvastatin and ezetimibe) or placebo; this rerandomization of the simvastatin-only group was prespecified to occur at one year if adverse events were no greater with the combination of simvastatin and ezetimibe than with simvastatin alone. (See "Low-density lipoprotein cholesterol lowering with drugs other than statins and PCSK9 inhibitors", section on 'Ezetimibe'.)

SHARP included 3023 patients treated with maintenance dialysis. By design, none had a history of MI or coronary revascularization. During a median follow-up of 4.9 years, patients taking the combination of simvastatin and ezetimibe had a lower incidence of atherosclerotic cardiovascular events that was not statistically significant (15.0 versus 16.5 percent [RR 0.90, 95% CI 0.75-1.08]). However, the trial was not powered to detect an effect specifically in the dialysis population. Results for the individual components of this composite endpoint were presented only for the entire study population (combining patients on dialysis and not on dialysis) and are discussed separately. (See "Lipid management in patients with nondialysis chronic kidney disease".)

Meta-analyses — Three meta-analyses have been performed of randomized, controlled trials that examined a benefit of lipid-lowering agents (including statins and statins plus ezetimibe) among patients with all stages of CKD, including those on dialysis [9-11]:

●In one meta-analysis, subgroup analyses of patients on dialysis showed reductions in the risk of cardiac mortality (two studies, n = 1986; RR 0.78, 95% CI 0.68–0.89) and MI (one study, n = 731; RR 0.72, 95% CI 0.56-0.92) [9]. Analysis of two studies showed an increase in the risk of stroke (n = 1986; RR 1.47, 95% CI 1.09-2.0); however, the studies that were included in this last analysis were considered of poor quality.

●Another meta-analysis reported that, among patients on dialysis, statin treatment had little or no effect on all-cause mortality (RR 0.96, 95% CI 0.88-1.04), cardiovascular mortality (RR 0.94, 95% CI 0.82-1.07), and major cardiovascular events (RR 0.95, 95% CI 0.87-1.03). The effects of treatment on MI and stroke were uncertain [10].

●A third meta-analysis showed that statins marginally lower the relative risks (RR 0.92, 95% CI 0.85-0.99) of cardiovascular events among patients on dialysis, although this reduction in relative risks was significantly lower than that in patients with CKD not on dialysis (RR 0.70, 95% CI 0.63-0.88). There were no differences in relative risk reduction associated with statins for coronary events (RR 0.91, 95% CI 0.81-1.02) or for stroke (RR 1.16, 95% CI 0.91-1.47) [11].

The results of these meta-analyses among patients with CKD who are not receiving dialysis and among transplant recipients are discussed separately. (See "Lipid management in patients with nondialysis chronic kidney disease" and "Lipid abnormalities after kidney transplantation", section on 'Treatment'.)

Recommendations for treatment — We agree with the 2013 Kidney Disease: Improving Global Outcomes (KDIGO) guidelines and with the National Kidney Foundation Kidney Disease Outcomes Quality Initiative (KDOQI) work group that statin therapy not be routinely initiated in patients on dialysis [12,13]. We include in this group patients on peritoneal dialysis, although the data on patients on peritoneal dialysis are more sparse than that for patients on hemodialysis. Despite significant reductions in mean serum LDL cholesterol levels, the 4-D, AURORA, and SHARP trials found no definite clinical benefit with statin therapy in patients on hemodialysis. However, as noted above, primary analysis of the SHARP trial showed a decrease in atherosclerotic events in the entire cohort that included both dialysis and nondialysis-dependent patients with CKD and a potential benefit in atherosclerotic events in the dialysis subgroup, and post-hoc analyses of AURORA and 4-D showed benefits among patients with diabetes and among those with an LDL cholesterol concentration >145 mg/dL (3.76 mmol/L), respectively. In addition, one meta-analysis including all three trials showed a modest benefit in atherosclerotic-related events, while another found no benefit of statin treatment in all-cause mortality, cardiovascular mortality, and major cardiovascular events. (See 'Meta-analyses' above.)

By contrast, randomized, prospective studies have reported consistent and robust improvements in cardiovascular outcomes with statin therapy in patients with normal kidney function at increased risk of adverse cardiovascular events. In addition, meta-analyses and post-hoc analyses of studies of patients with mild-to-moderate kidney dysfunction have noted benefits with statin therapy. (See "Management of low density lipoprotein cholesterol (LDL-C) in the secondary prevention of cardiovascular disease" and "Chronic kidney disease and coronary heart disease".)

The most likely explanation for these discordant results is that the pathogenic processes for adverse cardiovascular outcomes among patients with ESKD differ from those with either mild-to-moderate kidney dysfunction or normal kidney function. In the dialysis population, approximately 60 percent of all cardiac deaths are presumably due to sudden death or arrhythmias, and the underlying diseases that trigger the sudden death or arrhythmias are unclear. There is a markedly increased prevalence of myocardial abnormalities such as left ventricular hypertrophy (present in approximately 75 percent of patients on dialysis) and alterations in myocardial ultrastructure and function. Atheromatous coronary lesions are thought by some investigators to be less important than the nonatherosclerotic diseases, such as nonischemic cardiac fibrosis, in patients on maintenance dialysis. The supportive data for these conclusions are presented separately. (See "Evaluation of sudden cardiac arrest and sudden cardiac death in patients on dialysis".)

We also agree that statin therapy be continued in patients who are already receiving statins or a statin/ezetimibe combination at the time of initiation of dialysis [12,13], especially in those patients with significantly elevated serum levels of LDL cholesterol, largely due to the overwhelming evidence of benefit in the general population and limited data in the dialysis population. However, some clinicians would withdraw statin therapy in patients on dialysis based upon the lack of benefit in 4-D and AURORA. The question of whether to withdraw statin therapy in patients on dialysis already being treated with these agents has not been addressed in randomized trials. Both 4-D and AURORA trials had a large number of patients who discontinued the study medication, but this was most commonly due to achieving a study endpoint or undergoing kidney transplantation.

Our overall recommendation not to routinely administer statins to patients on dialysis does not necessarily apply to those with very high serum LDL cholesterol levels (such as >145 mg/dL [3.8 mmol/L]), as the post-hoc analysis of the 4-D trial showed an apparent beneficial effect in patients in that subgroup. This uncertainty is particularly poignant for patients with very high LDL cholesterol (eg, >190 mg/dL [4.9 mmol/L]) since those patients might not have been enrolled in AURORA, because of the potential reluctance of nephrologists to refer these patients to the randomized trial and be allocated to the placebo arm. Thus, in the uncommon patient on dialysis with very high LDL cholesterol levels, it is reasonable to initiate statin therapy. For these patients in whom statin therapy is prescribed, the following goals are reasonable, although they have not been validated in patients on dialysis:

●A serum LDL cholesterol of <100 mg/dL (<2.6 mmol/L)

●A serum non-high-density lipoprotein (HDL) cholesterol concentration (ie, total minus HDL cholesterol) of <130 mg/dL (<3.36 mmol/L) in patients who have already achieved the target LDL cholesterol level but have fasting triglycerides ≥200 mg/dL (≥2.26 mmol/L)

For patients not on dialysis, some experts have advocated the use of statins without follow-up serum cholesterol measurements. We do not support that approach in patients on dialysis, since statins have not been definitively shown to be beneficial in improving clinical outcomes, as described above. For patients on dialysis who have very high baseline LDL cholesterol levels, the use of statins should be justified based solely on their cholesterol-lowering effects, rather than on putative pleotropic effects.

For patients on dialysis with very high baseline LDL cholesterol levels who initiate pharmacologic treatment, ezetimibe is probably unnecessary as long as the LDL cholesterol and the non-HDL cholesterol targets can be achieved with statin therapy. Ezetimibe was not used in the 4-D and AURORA trials, in which subgroup analyses showed some clinical benefits, as described above.

Specific statins and doses that have been shown to be beneficial in the CKD population in clinical trials include fluvastatin 80 mg daily, atorvastatin 20 mg daily, rosuvastatin 10 mg daily, pravastatin 40 mg daily, and simvastatin 40 mg daily [12]. Whether other doses of these medications or use of similar agents are of equal or greater benefit remains to be proven in patients on dialysis, although there is no reason to suspect otherwise (table 1).

The results from 4-D, AURORA, and SHARP should also not be extrapolated to patients with mild-to-moderate CKD. As noted above, statin therapy is recommended in such patients given that meta-analyses and post-hoc analyses of randomized trials have reported decreased all-cause and cardiovascular mortality. (See "Chronic kidney disease and coronary heart disease" and "Lipid management in patients with nondialysis chronic kidney disease".)

Hypertriglyceridemia — Hypertriglyceridemia is the most common lipid disturbance in patients on dialysis and is attributed to an increased production and the retention of putatively atherogenic particles, such as lipoprotein remnants, in the plasma. Hypertriglyceridemia is not usually treated with pharmacologic therapy in patients on dialysis, partly because the relationship between serum triglyceride levels and clinical outcome is uncertain, the propensity of patients on dialysis to develop side effects from drugs, and the prevalence of polypharmacy in this population. We agree with the 2013 KDIGO guidelines suggesting that therapeutic lifestyle changes be advised among patients with fasting triglyceride levels >500 mg/dL (5.65 mmol/L) [12]. Such lifestyle changes include dietary modification, weight reduction, increased physical activity, reduction in alcohol intake, and treatment of hyperglycemia, if present. Caution should be exercised, however, in dietary modifications to avoid malnutrition, which is often present in patients on chronic dialysis. (See "Lipid management in patients with nondialysis chronic kidney disease".)

In selected patients without CKD who have proven CHD, a strong family history of CHD, or multiple coexisting cardiovascular risk factors, drug therapy for isolated hypertriglyceridemia may confer cardiovascular protective benefits. (See "Hypertriglyceridemia in adults: Management", section on 'Severe hypertriglyceridemia'.)

In addition, in the Veterans Affairs High-Density Lipoprotein Intervention Trial, gemfibrozil therapy lowered the risk of cardiovascular disease (CVD) in patients with nondialysis CKD who had established CHD and an HDL cholesterol <40 mg/dL (1.03 mmol/L). However, there was no effect on total mortality, and gemfibrozil caused a significant decline in kidney function. (See "Lipid management in patients with nondialysis chronic kidney disease".)

Whether these findings in patients without CKD or with nondialysis CKD translate to patients on dialysis is not known.

HYPERTENSION — The epidemiology, pathogenesis, and treatment of hypertension in patients on dialysis are discussed elsewhere. (See "Hypertension in patients on dialysis".)

TRADITIONAL DRUG THERAPY

Aspirin — In the general population, aspirin therapy has been widely used to decrease the risk of myocardial infarction (MI), stroke, and cardiovascular death among those with prior manifestations of cardiovascular disease (CVD). The evidence concerning these issues is presented separately. (See "Aspirin for the secondary prevention of atherosclerotic cardiovascular disease".)

By contrast, there is a paucity of definitive data concerning the efficacy of aspirin in patients on dialysis. Although patients on dialysis have a high risk of CVD, many CVD events in the dialysis population are not atherosclerotic or thrombotic in nature, and patients on dialysis are at increased risk of bleeding. (See "Uremic platelet dysfunction".)

Some larger, observational studies have suggested that aspirin use may be associated with increased cardiovascular mortality or adverse cardiovascular events [14-16]. In one study of 28,320 patients randomly selected from the Dialysis Outcomes and Practice Patterns Study, aspirin was associated with an increased risk of MI (relative risk [RR] 1.21, 95% CI 1.06-1.38) and any cardiac event (RR 1.08, 95% CI 1.02-1.14) [14]. By contrast, the risk of stroke was decreased with aspirin (RR 0.82, 95% CI 0.69-0.98). Similarly, an observational study of 41,425 patients on hemodialysis reported that aspirin was associated with an increased risk of mortality (hazard ratio [HR] 1.06, 95% CI 1.01-1.11) [15]. In these studies, confounding by indication is difficult to rule out.

A Cochrane meta-analysis that included randomized trials of nondialysis chronic kidney disease (CKD) and end-stage kidney disease (ESKD) patients found that antiplatelet agents, compared with no treatment or placebo, reduced the risk of MI (17 studies; RR 0.87, 95% CI 0.76-0.99) but not all-cause mortality (30 studies; RR 0.93, 95% CI 0.80-1.06), cardiovascular mortality (19 studies; RR 0.89, 95% CI 0.70-1.12), or stroke (11 studies; RR 1.00, 95% CI 0.58-1.72) [17]. At the same time, antiplatelet agents increased the risk of major bleeding (27 studies; RR 1.33, 95% CI 1.10-1.65) and minor bleeding (18 studies; RR 1.49, 95% CI 1.12-1.97). A more recent meta-analysis reported similar findings [18].

In the absence of definitive evidence in patients on dialysis, we suggest that individual treatment decisions be based upon consideration of patients' individual risks, potential benefits, and preferences. The prescription of low-dose aspirin (81 mg/day) is probably safe in most patients on chronic dialysis who do not have contraindications or an excess bleeding risk. In the absence of such risk, aspirin therapy for secondary cardiovascular prevention is reasonable. However, the avoidance of aspirin in patients on dialysis is also reasonable, given the unproven cardiovascular benefit and increased risk of bleeding compared with the general population.

Beta blockers — In the general population, beta blockers, unless contraindicated, should be part of routine therapy in patients with an acute MI or heart failure due to systolic dysfunction. (See "Acute myocardial infarction: Role of beta blocker therapy" and "Primary pharmacologic therapy for heart failure with reduced ejection fraction", section on 'Beta blocker'.)

Issues surrounding the use of beta blockers for secondary prevention in patients on dialysis with an acute MI or heart failure are presented separately. (See "Management of heart failure in patients on dialysis", section on 'Heart failure with reduced ejection fraction (HFrEF)'.)

ACE inhibitors, ARBs, and MRAs — Among those with normal kidney function, angiotensin-converting enzyme (ACE) inhibitors and angiotensin receptor blockers (ARBs) decrease cardiovascular mortality in post-MI patients with systolic dysfunction. In addition, ACE inhibitors provide cardiovascular benefit in most patients with an acute anterior MI. In patients with relatively preserved kidney function, mineralocorticoid receptor antagonists (MRAs) also are used for heart failure. The evidence supporting these therapies in patients not on dialysis is presented separately. (See "Angiotensin converting enzyme inhibitors and receptor blockers in acute myocardial infarction: Recommendations for use" and "Antihypertensive therapy and progression of nondiabetic chronic kidney disease in adults" and "Primary pharmacologic therapy for heart failure with reduced ejection fraction", section on 'Evidence'.)

There is a paucity of evidence concerning the cardiovascular benefits and risk/benefit ratios of ACE inhibitors, ARBs, and MRAs in patients on dialysis. These agents have the potential to cause hyperkalemia, especially in those with residual kidney function. The use of these agents in patients on dialysis is discussed separately. (See "Management of heart failure in patients on dialysis", section on 'Heart failure with reduced ejection fraction (HFrEF)'.)

Anti-hyperglycemic agents — Sodium-glucose cotransporter 2 inhibitors and glucagon-like peptide 1 receptor agonists provide cardiovascular and kidney benefits in patients with nondialysis CKD. (See "Overview of the management of chronic kidney disease in adults", section on 'Slowing the rate of progression' and "Treatment of diabetic kidney disease", section on 'Type 2 diabetes: Treat with additional kidney-protective therapy'.)

These agents have not been tested in patients with ESKD and are therefore not recommended in such patients.

ANEMIA — Many retrospective, population-based studies in patients on dialysis and some prospective studies have explored clinical outcomes associated with various hemoglobin (Hb) or hematocrit (Hct) levels. Although there is some variability in results among studies, the most consistent observation has been that better outcomes without an increase in adverse reactions are associated with Hb values that are approximately between 10 and 12 g/dL (Hct of 30 to 36 percent) compared with values below this level.

In patients undergoing maintenance dialysis, the clinical benefits and adverse effects associated with attaining normal or near-normal Hb values, compared with lower levels, have been evaluated in multiple randomized trials, meta-analyses, and systematic reviews. Evaluated outcomes have included overall mortality, cardiovascular mortality, and morbidity due to cardiovascular disease (CVD). These data and recommendations for treatment of patients on dialysis with erythropoietic agents are presented separately. (See "Treatment of anemia in patients on dialysis".)

DISORDERS OF MINERAL METABOLISM — Disorders of mineral metabolism are associated with an increased risk of all-cause and cardiovascular mortality in patients on dialysis (see "Patient survival and maintenance dialysis"). Cardiovascular disease (CVD) in patients on dialysis may be due in part to the presence of vascular calcification, independent of atherosclerosis, particularly in the form of extensive coronary artery calcification as well as peripheral vascular calcification, which increases peripheral vascular resistance and afterload. Two key questions are what drive the development and maintenance or progression of this calcification and whether interventions to decrease vascular medial calcification improve clinical outcomes.

Abnormalities of mineral metabolism and therapeutic maneuvers aimed at correcting these abnormalities have been implicated as primary underlying pathogenic factors. These include hyperphosphatemia, administration of calcium-containing oral phosphate binders, elevated parathyroid hormone levels, and other factors that favor vascular calcification in uremia.

No prospective, randomized studies have definitively demonstrated a clinical cardiovascular benefit and/or a survival advantage with any of the current therapeutic options, including limiting calcium intake, use of non-calcium-containing or calcium-containing phosphate binders, active vitamin D therapy, and administration of calcimimetics. Some, though not all, observational studies have shown improved survival in patients on hemodialysis treated with active vitamin D analogs. These issues are discussed in detail elsewhere. (See "Management of secondary hyperparathyroidism in adult nondialysis patients with chronic kidney disease" and "Management of secondary hyperparathyroidism in adult patients on dialysis" and "Management of hyperphosphatemia in adults with chronic kidney disease" and "Vascular calcification in chronic kidney disease".)

Another potential contributor of CVD in patients on dialysis is cardiomyopathy with subsequent left ventricular hypertrophy that is independent of hypertension, ischemia, or valvular diseases in pathogenesis. The cardiomyopathy may be due to myocardial fibrosis caused by uremic toxins, such as fibroblast growth factor-23. Other factors causing cardiomyopathy are discussed in detail separately. (See "Management of secondary hyperparathyroidism in adult nondialysis patients with chronic kidney disease" and "Management of secondary hyperparathyroidism in adult patients on dialysis" and "Management of hyperphosphatemia in adults with chronic kidney disease" and "Vascular calcification in chronic kidney disease".)

GLYCEMIC CONTROL IN PATIENTS WITH DIABETES — There are cardiovascular benefits of glycemic control in patients with type 1 and type 2 diabetes who are not on dialysis. This is discussed in detail separately. (See "Glycemic control and vascular complications in type 1 diabetes mellitus", section on 'Macrovascular disease' and "Glycemic control and vascular complications in type 2 diabetes mellitus".)

The hemoglobin (Hb) A1C target that is associated with the best clinical outcome in patients on dialysis with diabetes has not been established. This issue can be found separately. (See "Management of hyperglycemia in patients with type 2 diabetes and advanced chronic kidney disease or end-stage kidney disease".)

HYPERHOMOCYSTEINEMIA — In patients with normal kidney function, there is an increased risk of developing coronary heart disease (CHD) in those with elevated plasma homocysteine levels. Although folic acid lowers homocysteine levels, evidence from randomized trials does not support its use in secondary prevention in the nondialysis population.

Hyperhomocysteinemia is common in patients with end-stage kidney disease (ESKD). As observed in those with normal kidney function, lowering homocysteine levels does not improve cardiovascular outcomes in patients on dialysis. This was best shown in the double-blind, randomized, controlled trial, Homocysteinemia in Kidney and End-Stage Renal Disease, which compared the combination of folic acid (40 mg/day), pyridoxine (100 mg/day), and vitamin B12 (2 mg/day) versus placebo on vascular outcomes in 2056 patients with advanced chronic kidney disease (CKD), including ESKD [19]. At a median follow-up of 3.2 years, there was no difference between the two groups in terms of total mortality, myocardial infarction (MI), stroke, and amputations, despite a significant decrease in homocysteine concentrations.

Therefore, we do not administer folic acid, vitamin B6, and/or vitamin B12 for the purpose of lowering homocysteine levels and improving cardiovascular outcomes.

DECREASED OXIDATIVE STRESS — In patients without kidney failure, lipid peroxidation of cellular membrane polyunsaturated fatty acids by reactive oxygen species, which are increased during ischemia, is considered the major mechanism of ischemia-reperfusion injury. Although it had been hoped that antioxidant treatment would prevent or retard atherosclerosis, randomized evidence has not convincingly demonstrated clinical benefits with antioxidant vitamins or other putative antioxidant supplements or drugs. A discussion of this issue in patients with normal kidney function is provided separately. (See "Vitamin intake and disease prevention", section on 'Antioxidants'.)

A number of studies have demonstrated that patients on hemodialysis are, in general, at a state of increased oxidative stress, suggesting the possibility that antioxidant therapies may improve clinical outcomes. Limited evidence suggests that an attempt to lower oxidative stress using antioxidants may improve cardiovascular disease (CVD) outcomes in end-stage kidney disease (ESKD) [20,21]:

●In one trial, 196 patients on dialysis with known heart disease were randomly assigned to 800 international units/day of oral vitamin E or placebo [20]. At follow-up at a median of 519 days, the incidence of a primary endpoint (which consisted of myocardial infarction, ischemic stroke, peripheral vascular disease, and unstable angina) was significantly lower in the active therapy group (16 versus 33 percent; relative risk [RR] 0.46, 95% CI 0.27-0.78).

●In a second randomized, placebo-controlled trial of 134 patients, the oral administration of acetylcysteine (600 mg two times per day) was associated with a decrease in total and cardiovascular mortality at a median follow-up of 14 months (RR 0.60, 95% CI 0.38-0.95) [21].

●A Cochrane meta-analysis that included both studies cited above showed a significant benefit conferred by antioxidant therapy for prevention of CVD in patients on dialysis (RR 0.57, 95% CI 0.41-0.80) [22].

It must be emphasized that both of these trials are small; large, randomized trials on the effect of antioxidant therapy on clinical outcomes in patients on dialysis are lacking. One trial examined the effect of oral antioxidants on inflammatory biomarkers and response of hemoglobin to erythropoietin-stimulating agents (ESAs) [23]. Among 353 patients randomly assigned to receive a combination of mixed tocopherols (666 international units/day) plus alpha-lipoic acid (600 mg/day) or placebo, there was no difference between the two groups in plasma markers of inflammation (C-reactive protein, interleukin-6) or oxidation (F2 isoprostanes and isofurans) or in ESA responsiveness at three and six months. There was also no difference between groups in hospitalization rates and mortality during the study.

Another strategy to attempt to lower oxidative stress is hemodialysis with vitamin E-bonded dialysis membranes [24-26]. With this technique, vitamin E is incorporated by coating a cellulose-based membrane with alpha-tocopherol. A number of studies have found that the use of such membranes is associated with subclinical improvements, such as better endothelial dysfunction. In the only trial with clinical outcomes thus far, regression of carotid intimal thickness was noted in the arm with 17 patients exposed to vitamin E-coated dialyzers [27]. Further studies are needed to better understand the role for these dialyzers.

Pending further evidence of clinical benefit, we do not prescribe any agent with the aim of lowering oxidative stress to improve cardiovascular outcomes.

FISH OIL — In the general population, evidence from observational studies suggests that increased dietary fatty fish intake and fish oil supplementation are associated with a lower cardiac mortality. Based upon these results, subsequent randomized trials have evaluated the benefit of fish oil supplements containing omega-3 fatty acids in various high-risk populations. (See "Overview of sudden cardiac arrest and sudden cardiac death".)

In one small randomized, placebo-controlled study of omega-3 fatty acids for secondary prevention in 206 patients on hemodialysis, no significant effect on the primary outcome of cardiovascular events and death was observed with omega-3 fatty acids [28]. However, a significantly smaller number of myocardial infarctions, a secondary outcome, occurred with active therapy (4 versus 13 in the placebo group).

In another small randomized, placebo-controlled trial in which the effect of fish oil on arteriovenous (AV) graft thrombosis as the primary outcome was assessed among 201 patients on hemodialysis, patients in the fish oil group had better cardiovascular event-free survival (hazard ratio [HR] 0.43, 95% CI 0.19-0.96) and lower mean systolic blood pressure (-3.61 versus 4.49 mmHg) at 12 months of follow-up [29]. However, the number of patients with cardiovascular events in this trial was small (26 total). The effect of fish oil on graft thrombosis is discussed separately. (See "Overview of hemodialysis arteriovenous graft maintenance and thrombosis prevention", section on 'Fish oil'.)

The efficacy of fish oil to prevent coronary events in patients with end-stage kidney disease (ESKD) requires further study, particularly in larger trials.

At present, we do not administer fish oil for cardiovascular protection in patients on dialysis. The role of fish oil for prevention of AV graft thrombosis is discussed separately. (See "Overview of hemodialysis arteriovenous graft maintenance and thrombosis prevention", section on 'Fish oil'.)

THERAPEUTIC LIFESTYLE CHANGES — Therapeutic lifestyle changes are likely a part of an effective strategy to decrease the risk of adverse cardiovascular outcomes in patients on dialysis. Strong data from randomized trials in patients on dialysis are, however, lacking.

●Smoking – In patients on dialysis, observational studies suggest that smoking greatly increases the risk for cardiovascular morbidity and mortality [30-32]. However, unlike that found in the general population, there are few studies showing beneficial cardiovascular outcomes with smoking cessation in patients on dialysis. In a post-hoc analysis of the Hemodialysis (HEMO) study, including 1842 patients on hemodialysis, there was a trend toward an increase in risk of cardiovascular mortality among current smokers compared with former smokers, but this was not statistically significant [32]. In this analysis, the population-attributable fractions (ie, the fraction of observed deaths that might have been avoided) were 5.3 for current smokers versus patients who never smoked and 2.1 for current versus former smokers. Nonetheless, we recommend that patients on dialysis completely stop smoking, given the marked benefits of smoking cessation observed in the general population. (See "Cardiovascular risk of smoking and benefits of smoking cessation".)

●Weight reduction – In those with normal kidney function, obesity is associated with an increased risk for cardiovascular disease (CVD). The benefits of weight reduction in patients on dialysis are, however, unclear. In fact, there appears to be a survival advantage in patients on dialysis who have obesity. The higher mortality risks in patients on dialysis with lower body mass indices (BMIs) may be attributed to malnutrition. Limited data suggest that increased muscle mass confers greater survival advantage than increased fat in the dialysis population [33].

In general, we do not currently recommend weight reduction for all patients on dialysis with higher BMI. Weight reduction, however, should be considered in patients with severe obesity, especially in those who are planning kidney transplant surgery.

●Exercise – Regular exercise has a variety of cardiovascular benefits in those with normal kidney function. (See "Exercise and fitness in the prevention of atherosclerotic cardiovascular disease".)

By comparison, a paucity of data exists among patients on dialysis concerning the association between exercise and cardiovascular and overall survival. This was examined in an observational study using data for 2507 patients from the Dialysis Morbidity and Mortality Wave II Study [34]. Patients who exercised two to three and four to five times per week had lower mortality. However, for unclear reasons, daily exercise provided no survival benefit. Nonetheless, a gradual increase in physical activity is probably beneficial in most patients on dialysis and should be encouraged unless there are specific medical contraindications. (See "Uremic myopathy and deconditioning in patients with chronic kidney disease (including those on dialysis)".)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Dialysis".)

SUMMARY AND RECOMMENDATIONS

●Secondary prevention – Cardiovascular risk factor modification should generally be undertaken for patients on dialysis, given that they have very high risk for cardiovascular events. In patients with established cardiovascular disease (CVD), these measures are referred to as "secondary prevention". (See 'Coronary heart disease risk equivalent' above and 'Overview of secondary prevention' above.)

●Lipid modification – We recommend not initiating statin therapy in patients on dialysis with no, mild, or moderate increase in serum low-density lipoprotein (LDL) cholesterol levels, pending more definitive data (Grade 1B). We suggest continuing statin therapy in patients on dialysis currently receiving these agents unless side effects develop (Grade 2C). However, some clinicians would withdraw statin therapy in patients on dialysis based upon the lack of clearly demonstrable benefits. In patients on dialysis with very high serum LDL cholesterol levels, the initiation of statins may be reasonable. (See 'Lipid modification' above.)

●Hypertension – Target blood pressure in patients on dialysis should be individualized, partly depending upon symptoms of intradialytic hypotension. (See "Hypertension in patients on dialysis".)

●Aspirin – For most patients on dialysis who have established CVD, we suggest the administration of aspirin (Grade 2C). The dose of aspirin is 81 mg/day. However, patients on dialysis who would prefer to avoid the increased risk of bleeding for an unproven cardiovascular benefit may reasonably choose to forgo aspirin therapy. (See 'Aspirin' above.)

●Beta-blockers, ACE-inhibitors, ARBs, and MRAs – The use of beta blockers, angiotensin-converting enzyme (ACE) inhibitors, angiotensin II receptor blockers (ARBs) and mineralocorticoid receptor antagonists (MRAs) for secondary prevention in patients on dialysis is discussed in detail separately. (See "Management of heart failure in patients on dialysis", section on 'Heart failure with reduced ejection fraction (HFrEF)'.)

●Therapeutic lifestyle changes – We recommend complete smoking cessation (Grade 1A). Patients on dialysis may also benefit from a gradual increase in physical activity as tolerated for cardiovascular protection and increase in muscle mass. (See 'Therapeutic lifestyle changes' above.)