Japanese Society for Dialysis Therapy Guidelines for Management of Cardiovascular Diseases in Patients on Chronic Hemodialysis

The annual all-cause mortality in chronic dialysis patients in our country is within 10%, indicating that the outcome of dialysis therapy in Japan is one of the best in the world. It is nothing short of extraordinary to maintain favorable survival like this despite challenging conditions such as aging of the patients and increase in the proportion of patients on long-term dialysis and with diabetes mellitus. We can be proud of our achievement. Novel therapeutic strategies for dialysis patients have been developed, such as antihypertensive drugs (e.g. angiotensin II receptor blockers, calcium channel blockers and beta blockers), treatment of anemia (e.g. erythropoiesis stimulating agents), and management of chronic kidney disease-mineral and bone disorder (CKD-MBD) (e.g. activated vitamin D, calcimimetics, and new phosphate binders). While the beneficial effect of these new approaches is well acknowledged, we must not forget that the favorable outcome is also due to the considerable efforts and excellence in management of all the medical staff, including physicians, nurses, and clinical engineers, who are engaged in dialysis therapy in Japan. While mortality due to infectious diseases is increasing at present, about half of dialysis patients die from cardiovascular disease (CVD). Thus, the management of CVD has become the most challenging clinical issue in dialysis patients. With regard to CVD, the main focus has so far been on blood vessel diseases of the heart and brain; however, peripheral artery disease (PAD) is now also attracting attention because the number of patients with atherosclerotic obstruction of the peripheral arteries in the lower extremities has increased in recent years and endovascular catheter therapy has been introduced and developed. The number of specialists in the field of PAD has increased along with the development of new biomedical technology and expansion of their use. Endovascular catheter therapy is currently offered to patients with chronic dialysis and we expected an increase in the number of patients benefiting from this therapy. Evidence suggests that the pathological process of CVD is also involved in the aggravation of systemic atherosclerosis associated with renal dysfunction, prompting the use of potent anti-atherogenic agents, such as statins in dialysis patients similar to the general population. With regard to CVD in dialysis patients, unfortunately, there is little clinical evidence to justify the compilation of clinical guidelines. For example, the appropriate blood pressure level in such patients remains unknown, and the target blood pressure level for management of hypertension has not yet been defined even in the guidelines issued by Western countries. Although we discussed this issue in detail in several committee meetings, we only agreed on setting the target blood pressure though we presented this as an opinion rather than guideline by the committee. There is no doubt that we need to validate in the future whether the level is appropriate or not. In fact, we do not know whether any statement on the clinical guideline is right or not especially when evidence is insufficient, and any statement is nothing but "themes of clinical questions". We need to validate this issue by prospective high-evidence grade studies. The Japanese Society for Dialysis Therapy (JSDT) maintains a patient registry database kept with the standing committee responsible for statistics and investigation. We used the data stored in this database to generate the present guideline. We stress that we should continue to maintain this important registry system in order to revise the clinical guidelines in the future. The chapters on cardiac failure, ischemic heart disease, arrhythmia, valvular heart disease, cerebrovascular disease, and peripheral artery disease in the guideline are separated into those for "renal dialysis physicians" and "cardiologists (or strokologists)" in order to demonstrate the importance of cooperation between these two specialties. We think that the excellent outcome of dialysis therapy in Japan is in part attributed to the implementation of excellent daily clinical procedures, which are based on "evidence" and/or "experience" in each dialysis facility. We have to validate the daily procedures and present them as treatment guidelines. We hope this guideline is useful in daily clinical practice. We determined the grading evidence and recommendation levels according to the position statement from Kidney Disease: Improving Global Outcomes (1,2). In dialysis patients, dyslipidemia is an independent risk factor for cardiovascular diseases, particularly incident myocardial infarction (B). We recommend measurement of low-density lipoprotein cholesterol (LDL-C), non-high-density lipoprotein cholesterol (non-HDL-C), HDL-C, and triglyceride (TG) before dialysis (casual blood sampling) for routine evaluation (1B). We suggest the control target levels should be LDL-C < 120 mg/dL or non-HDL-C < 150 mg/dL for the primary prevention, and LDL-C < 100 mg/dL or non-HDL-C < 130 mg/dL for the secondary prevention of ischemic heart disease (2C). We suggest that administration of statin should be considered if lipid control cannot be achieved by dietary/exercise therapy (2B). We suggest that the evaluation and intervention of undernutrition should be considered if hypolipidemia is present. Observational studies in Japan have demonstrated a close relationship between dyslipidemia (hyper-LDL-cholesterolemia, hypo-HDL-cholesterolemia, hypertriglyceridemia, and/or hyper-non-HDL-cholesterolemia) and the severity of atherosclerosis (1,2) and also the risk of myocardial infarction (3) in dialysis patients. In addition, dyslipidemia is more closely related to coronary artery disease than cerebrovascular disorders. However, observational cohort studies of dialysis patients showed a higher risk of death due to all causes (4) or death due to cardiovascular disease (5) in patients with low total cholesterol (TC) level, reflecting a reverse tendency compared with epidemiological data in the general population. Such relationship is, however, not observed in dialysis patients who are free of inflammation or are not undernourished (as defined by the levels of C-reactive protein [CRP] and serum albumin, respectively) (4,6). In Western countries, the survival curve of dialysis patients who develop acute coronary syndrome is poorer in patients with low body mass index (BMI) than in those with high BMI (7). Similarly, in Japanese dialysis patients, old age, low BMI, and high CRP are reported to be factors that enhance the risk of death after a cardiovascular event (3). These reports suggest that undernutrition, represented by hypoalbuminemia, low BMI, and hypocholesterolemia, correlates with increased risk of death by increasing the risk of death after an event (fatality rate), although hypocholesterolemia per se is not considered to promote atherosclerosis (8). Dyslipidemia can be classified into primary and secondary dyslipidemia, depending on the cause. Primary dyslipidemia includes familial hypercholesterolemia (FH) and familial combined hyperlipidemia (FCHL), with a reported respective prevalence of each type of 1:500 and 1:100. Secondary dyslipidemia is caused by various conditions such as diabetes, endocrine (thyroid, adrenal) disorders, liver diseases, kidney diseases, and drugs. Hypercholesterolemia associated with nephrotic syndrome and hypertriglyceridemia and hypo-HDL-cholesterolemia associated with chronic kidney failure are well-known dyslipidemias caused by kidney diseases. Low lipoprotein lipase activity (high apo C-III levels), low hepatic lipase level, and low lecithin cholesterol acyltransferase (LCAT) activity contribute to dyslipidemia in patients with chronic renal failure. According to the Guidelines for Prevention of Atherosclerotic Cardiovascular Diseases by the Japan Atherosclerosis Society (9), hyper-LDL-cholesterolemia is defined as LDL-C ≥140mg/dL, hypo-HDL-cholesterolemia as HDL-C <40 mg/dL, and hypertriglyceridemia as TG ≥150 mg/dL in fasting blood samples. However, fasting blood samples are often difficult to obtain from dialysis patients. In general, post-prandial changes in TC or HDL-C level are very small, compared with the increase in TG levels. Thus, LDL-C level calculated by the Friedewald equation decreases while little change is observed in non-HDL-C level (TC minus HDL-C). Also, since non-HDL-C level is the sum of LDL-C and cholesterol present in TG-rich lipoproteins (Fig. 1), it is regarded as an integrated index of the atherogenic lipoprotein level. Therefore, for routine evaluation in dialysis patients, non-HDL-C level in a casual blood sample is considered acceptable in addition to the standard fasting LDL-C level. Serum total cholesterol and its components. Serum contains a mixture of lipoproteins of different densities (specific gravity), and the total sum of cholesterol in the various lipoproteins represents serum total cholesterol. Several methods are used to fractionate lipoproteins. HDL has an anti-atherosclerotic properties, and all other fractions apart from HDL (collectively called non-HDL) are atherogenic. The cholesterol present in non-HDL is expressed as non-HDL-C. Thus, non-HDL-C is the sum of cholesterol in atherogenic lipoproteins. In subjects with dyslipidemia in general, secondary dyslipidemia is usually excluded first, followed by recommendations for long-term dietary/exercise therapy. Drug treatment is also considered if the target level cannot be achieved. However, in patients with coronary artery disease, the first option should be drug treatment. A strict target lipid level is set in patients that have not developed coronary artery disease but are at high risk, while a stricter target is set for patients with established coronary artery disease (secondary prevention group). According to the recent epidemiological study of the Japanese Society for Dialysis Therapy (3), the risk of occurrence of acute myocardial infarction increases 1.24 times (95% confidence interval: 1.14–1.35) with every increase in non-HDL-C level of 1 mmol/L (38.7 mg/dL). Based on the results of this observational study, the present guidelines propose a target level of LDL-C <120 mg/dL or non-HDL-C <150 mg/dL for primary prevention, and LDL-C < 100 mg/dL or non-HDL-C level <130 mg/dL for secondary prevention. There are only a few randomized controlled trials in dialysis patients regarding whether lipid lowering therapy significantly reduces the risk of cardiovascular events. The 4D (Die Deutsche Diabetes Dialyse) study using atorvastatin (10) and AURORA Study using rosuvastatin (11) suggested that the risk of all cardiovascular diseases (including those not directly related to atherosclerosis such as heart failure and cerebral hemorrhage) can only be reduced slightly even by lipid lowering therapy using statins. However, the risk of ischemic cardiac accidents decreased significantly by 18% in the 4D Study. Taking these results and the results of the observational cohort study in Japan into consideration, it would be reasonable to treat dialysis patients with high LDL-C or non-HDL-C levels with statins to reduce the risk of ischemic heart disease. Furthermore, there is little medical basis for discontinuation of statins therapy, since statin use is reported to associate with better survival in both incident (12) and prevalent dialysis patients (13). In conducting drug therapy, statins are the first choice. Statins reduce LDL-C level by 25–40% although this effect varies with the drug and dose. In the above 4D (10) and AURORA (11) studies, the frequency of adverse effects were comparable between the statins and placebo groups, suggesting no safety problems with the use of statins. Excluding clinofibrate, fibrates available in Japan are contraindicated in patients with renal failure due to the high risk of rhabdomyolysis based on their excretion via the kidney. Bile acid-binding resins, eicosapentaenoic acid preparations, and intestinal cholesterol transporter inhibitors can also be used in dialysis patients. Niceritrol, a nicotinic acid derivative, reduces serum phosphate levels but could cause anemia and thrombocytopenia and must be administered with caution in dialysis patients. Many patients are treated with more than one drug. For safe treatment, one should monitor symptoms and laboratory tests including serum creatine kinase, aspartate aminotransferase and alanine aminotransferase, and also pay attention to drug interactions. If the patient develops hypolipidemia, a nutritional disorder should be suspected, and measures to improve the nutritional state should be considered. The following issues are proposed as topics of future studies; whether a very high TG level is a risk factor of acute pancreatitis in dialysis patients, and whether patients undergoing peritoneal dialysis and children with renal failure should be treated in a manner similar to that of adult hemodialysis patients. We expect further data from sub-analyses and meta-analysis of the 4D, AURORA, and Study of Heart and Renal Protection (SHARP) studies.* Papers on subanalyses of the 4D Study (14) and AURORA Study (15), and the original report of SHARP (16) appeared during the publication of the guidelines, on which the present simplified guidelines are based, and the preparation of this simplified version. The subanalyses of 4D and AURORA studies suggested that lipid lowering therapy prevents atherosclerotic cardiovascular events in diabetic patients on dialysis, and that it significantly prevents such events more effectively in patients with higher LDL-C levels before the treatment. SHARP also showed that lipid lowering therapy using the combination of simvastatin and ezetimibe significantly reduced the risk of atherosclerotic cardiovascular events and that such reduction showed no significant heterogeneity between the patient groups before and after the initiation of dialysis therapy. To evaluate the risk of cardiovascular death in dialysis patients, we recommend the inclusion of risk factors specific to renal failure (e.g. anemia, inflammation, undernutrition, abnormal mineral metabolism), in addition to classic risk factors (1C). The extent of arterial wall thickening, arterial wall stiffening, and vascular calcification may be used for the evaluation of cardiovascular risk (Opinion). In dialysis patients, the risk of death due to cardiovascular disease (CVD) such as ischemic heart disease, cerebrovascular diseases, and heart failure is markedly increased, and the relative risk compared to the general population is reported to be 10–30 (1). Dialysis patients are characterized by a high risk of CVD events and low survival rate after the onset (high fatality rate). Compared to the general population, dialysis patients show 2–5 times higher risk of incident acute myocardial infarction and poorer survival rate after acute myocardial infarction (2). This is also true for cerebrovascular diseases (3). The high incidence and high fatality rate are considered to synergistically increase the risk of death due to CVD (4). One of the reasons for the high risk of CVD in dialysis patients is advanced atherosclerosis before the initiation of dialysis. About half of the patients have significant coronary artery stenosis at the initiation of dialysis (5,6), and the presence or absence of coronary artery disease at the initiation of dialysis is a strong predictor of cardiovascular events after the initiation of dialysis (7). Vascular calcification is classified into atherosclerotic calcification affecting the intimal layer of the artery and Mönckeberg's sclerosis affecting the medial layer of the artery, especially the latter is more frequently observed in dialysis patients. Both types of calcification are significant predictors of death in dialysis patients. Abnormal mineral and bone metabolism including vascular calcification associated with chronic kidney diseases (CKD) has been integrated as a new concept named CKD-mineral and bone disorder (CKD-MBD) (8), and it is considered important in clinical practice of dialysis patients. Because the risk of CVD in dialysis patients is significantly high even after correction for classic risk factors such as old age, hypertension, dyslipidemia, and diabetes, factors specific to CKD are considered to be involved in the elevated risk of CVD (9). Sarnak et al. (10) noted many factors including anemia, inflammation, undernutrition, and abnormal mineral metabolism as non-classic risk factors. Among them, undernutrition (wasting) is diagnosed in daily clinical practice based on the presence of hypoalbuminemia or low BMI. According to reports from Japan, low BMI is a predictor of all-cause death (11,12) and CVD death (11), but not a predictor of future myocardial infarction (12). A report from the United States (13) observed that the survival curve after the onset of acute coronary syndrome was poorer in the low BMI group. In Japan, also, low BMI is independently related to the risk of death after CVD including myocardial infarction, cerebral infarction, and cerebral hemorrhage (12). Thus, certain non-classic risk factors are considered factors that enhance the fatality rate after the onset of CVD. Clinically, atherosclerosis-arteriosclerosis can be evaluated quantitatively and qualitatively by examination of the thickness and stiffness of the arterial wall and vascular calcification (Table 1). These measurements may serve as surrogate markers between risk factors and CVD events. Carotid artery intima-medial thickness is measured by B mode ultrasonography, which provides quantitative evaluation of arterial wall thickening, and is a predictor of the risk of CVD death and total death in dialysis patients (14). Aortic pulse wave velocity (cfPWV, hfPWV) is a representative index of arterial stiffness and a predictor of CVD death and total death in dialysis patients (15). While a high baPWV measured in the brachium and ankle is also a prognostic factor in dialysis patients (16), its value falsely decreases in patients with obstructive arteriosclerosis in the lower limbs. Therefore, caution is needed and simultaneous measurement of the ankle brachial pressure index (ABI) may be helpful. The Cardio-ankle vascular index (CAVI) and augmentation index (AI) have also been used as new indices of arterial stiffness. Various methods are available to evaluate vascular calcification. Among these, electron beam computed tomography (EBCT) has excellent temporal resolution and provides specific assessment of the heart and large blood vessels. Coronary artery calcification is usually evaluated using the coronary artery calcification score (CACS) calculated by Agatston's method. CACS has been reported to be a predictor of cardiovascular events (cardiac death, non-fatal myocardial infarction) in non-dialysis patients with coronary artery disease (17). However, while dialysis patients with high CACS have poor survival, CACS is not necessarily related to cardiovascular events (18). The sensitivity of multi-detector computed tomography (MDCT) has improved in recent years, and this modality has become the mainstay of coronary artery computed tomography (CT). Abdominal plain CT is used to measure the area of aortic calcification, using the aortic calcification index (ACI), which is determined in 10 slices at 1-cm intervals above the origin of the common iliac artery. In dialysis patients, there is a strong correlation between ACI and coronary artery calcification (19). The presence or absence of vascular calcification examined by thoracoabdominal CT (20,21) has also been shown to be an independent predictor of all-cause death and CVD death in dialysis patients and is considered to be useful in daily clinical practice. Although evaluation of these non-invasive surrogate indices may help estimate the individual CVD risk, the criteria used for their evaluation or appropriate frequency of their use have not been established. Longitudinal changes in these measures are not well known in dialysis patients. We propose that the evaluation method(s) should be selected taking into consideration the characteristics of individual patients and availability in the medical facilities, and to the measurement every if We do not the treatment for each risk factors and their effects on atherosclerosis-arteriosclerosis because are discussed in detail in for other including and at an appropriate for each patient are considered Furthermore, if treatment of CVD are particularly important in dialysis patients. In dialysis patients, we recommend blood pressure should be evaluated not only in the dialysis but also at (1B). In patients long-term dialysis with no of the cardiac we suggest the target of antihypertensive treatment should be blood pressure before dialysis at the of the (Opinion). We recommend should be set in the target blood pressure (1B). We recommend antihypertensive should be administered when the reduction in blood pressure is even after of (1B). According to reports on the present state of chronic dialysis therapy in Japan at the of of all dialysis patients were based on blood pressure measured at the initiation of dialysis and the criteria of the Japanese Society of (1). hypertension is a cause of ischemic heart disease, heart failure, and death, and the control of hypertension is important in dialysis patients (2). However, a in blood pressure during dialysis not only have effects on outcome (3), it may also the of or In dialysis patients, aortic calcification also has a on as pulse and and The outcome is poorer in patients with low pressure but pressure and also in patients with high pressure with pressure Also, the mortality rate is reported to significantly if the blood pressure of the pulse pressure measured before and after dialysis times a and at daily at and before to (7). This is also true for various other including the and blood blood pressure and blood pressure it is important to blood pressure in the evaluation It is more important to any clinical or therapeutic on the of measurements than a casual blood pressure measurement Many cohort studies demonstrated poorer of patients with high blood pressure compared with patients. This is due to the inclusion the of patients with or those with chronic heart failure reverse prospective studies are needed to further evaluate patients Many factors are to be responsible for hypertension in dialysis patients. blood pressure is reported to in more than of patients following strict management of body of is and the present guidelines propose should be of blood pressure measurement is for the of blood pressure of blood pressure pressure should be measured conditions although it can be measured in the or position depending on the at each facility. the of dialysis therapy, blood pressure should be measured after a of of at before the of dialysis. should from before the measurement and from during the pressure should be measured with the heart rate at every the of dialysis, the blood pressure should be measured in a similar manner before of blood and within after the of of and pressure measured before of blood at the of dialysis is called pressure at of setting or blood pressure should be measured also in the standing position at the of dialysis. blood pressure should be measured as by the guidelines of the Japanese Society of before to at and at in the are While has also been reported to be useful it cannot be due to the use of one for In dialysis patients with changes in body it is important to evaluate blood pressure not only in the dialysis but also at In dialysis patients, evaluation on a basis is because dialysis is after pressure decreases from the to the of the on the blood pressure should be evaluated or used are very The blood pressure represents the blood pressure measured before and after dialysis times a and daily blood in the and observational studies demonstrated that is a more significant predictor of and cardiovascular than casual or blood pressure measured at the of the pressure measured at on a non-dialysis in the of the could be also used since it correlates with In dialysis patients, a relationship is observed between blood pressure and However, this relationship that is, it is important to the subjects and when the target blood pressure of antihypertensive treatment. The of antihypertensive treatment is to reduce the long-term risk of cardiovascular diseases in patients on chronic dialysis, rather than reduce all-cause mortality Therefore, patients with cardiac dysfunction, for example, are The target level should be set after evaluation of cardiac in both patients with markedly reduced and those with reduced due to In since the outcome is reported to in patients with increased aortic stiffness due to aortic calcification, by in blood pressure and increase in pulse caution reduction in blood pressure is in consideration of the effects of the blood pressure on various conditions such as chronic heart failure and coronary blood For these the criteria for blood pressure control should not be to all patients but by patients with reduced cardiac for example, and further evaluated at While it is difficult to propose specific target blood pressure in the present guidelines due to the of a dialysis blood pressure than at the of the is as a However, a in blood pressure or in during dialysis and after dialysis are reported to and further studies are the other observational studies that blood pressure not with the effect of in blood pressure during dialysis The target blood pressure is set to reduce the long-term risk of cardiovascular diseases in patients on dialysis, and should not be to patients with cardiovascular disorders. In patients, the blood pressure during dialysis correlates significantly with the risk of death within years (3). One of antihypertensive treatment is appropriate of dialysis, and the conditions of dialysis such as blood and dialysis need to be should an appropriate be and This should be followed by administration of appropriate antihypertensive drugs when If a in blood pressure is observed during dialysis, antihypertensive should be or its the should be set the patient should be followed and antihypertensive if control of and of are the most important of during dialysis. To the target of antihypertensive treatment, the must be set first issue is discussed in a different It is to control changes in body between to any in blood pressure during