Cachexia in Chronic Kidney Disease: Malnutrition- Inflammation Complex and Reverse Epidemiology

Transcription

1 Chapter 6.4 Cachexia in Chronic Kidney Disease: Malnutrition- Inflammation Complex and Reverse Epidemiology Kamyar Kalantar-Zadeh, Joel D. Kopple Introduction Chronic kidney disease (CKD) is an irreversible and progressive disease state leading to renal dysfunction and related morbidity [1]. According to the National Kidney Foundation (NKF) Kidney Disease Outcome Quality Initiative (K/DOQI) guidelines, CKD is defined as a chronic disease state in that irreversible, structural, or functional abnormalities of the kidney, with or without a decreased glomerular filtration rate (GFR), are present for at least three consecutive months [1]. The degree of renal insufficiency, based on the magnitude of the estimated GFR for 1.73 m 2 body surface, is used to classify the CKD into five stages: (1) GFR > 90 ml/min, (2) GFR ml/min, (3) GFR ml/min, (4) GFR ml/min, and (5) GFR < 15 ml/min [1]. If they survive long enough, CKD patients eventually reach stage 5 CKD, also known as end-stage renal disease (ESRD), in which life prolongation is exclusively dependent upon renal replacement therapy, i.e. maintenance haemodialysis or peritoneal dialysis treatment and/or kidney transplantation. However, the majority of CKD patients die before reaching ESRD [2]. Epidemiological data indicate that there are currently at least 20 million individuals with CKD in the US [3], including over ESRD patients who undergo maintenance dialysis. Diabetes mellitus accounts for half of all cases of ESRD in industrialised nations [4]. According to the estimates of the US Renal Data System (USRDS), the number of ESRD patients will surpass one-half million by 2010 and will be between 1.5 and 3.1 million by 2030, whereas the US population will grow only from 280 to 350 million over the same 30-year period [4]. This exponential growth has major public health implications, especially since ESRD patients consume a disproportionately large component of the US Medicare budget, due to their requirement for continuous renal replacement therapy and their frequent morbidity [5, 6]. Despite many years of efforts and improvement in dialysis technique and patient care, the mortality rate in maintenance dialysis patients in the US and most industrialized countries continues to be unacceptably high, currently still approximately 20% per year in the US and 10 15% in Europe and Japan [7 9]. CKD and ESRD patients also commonly have a high hospitalisation rate and a low self-reported quality of life [10 13]. Cardiovascular, cerebral-vascular, and peripheral vascular diseases comprise the bulk of the severe morbidity and mortality in CKD patients [14, 15]. Indeed even a slight increase in serum creatinine (an often clinically used marker of renal insufficiency) has been shown to be an independent risk factor for cardiovascular disease and atherosclerosis in the general population [16 18]. Among potential candidates to explain the high rate of morbidity and mortality and cardiovascular disease in CKD patients, wasting syndrome and cachexia continue to top the list. Epidemiological studies have repeatedly and consistently shown a strong association between clinical outcome and measures of both proteinenergy malnutrition [19 22] and inflammation in CKD patients [23, 24]. Hypoalbuminaemia, rather than such conventional risk factors as hypertension and hypercholesterolaemia, is one of the strongest risk factors for mortality among dialysis patients (Fig. 1). Almost half of all dialysis patients have a serum albumin < 3.8 g/dl, which has been shown to be associated with at least a two-fold increase in mortality [25]. However, it is not known whether hypoalbuminaemia is a reflection of malnutrition, inflammation, or both. Many

2 306 Kamyar Kalantar-Zadeh, Joel D. Kopple Unadjusted HR Case-mix adjusted HR Hazard ratio (HR) of death Death HR albumin 3.8 g/dl Unadjusted: 2.40 (95% CI: ) Adjusted: 2.19 (95% CI: ) <3.0 Serum Albumin albumin Groups groups Fig. 1. Association between serum albumin concentration and all-cause mortality in a 2-year cohort of maintenance haemodialysis patients [25]. After dichotomising albumin values based on the cutoff level of 3.8 g/dl, the unadjusted and case-mix adjusted HR for serum albumin 3.8 g/dl was 2.40 (95% CI: ) and 2.19 (95% CI: ), respectively [25] investigators have observed that these two conditions tend to occur concurrently and coexist in individuals with CKD, and many factors that engender one of these conditions also lead to the other [22, 23, 26, 27]. Therefore the term malnutrition-inflammation complex (or cachexia) syndrome (MICS) [22, 28] or malnutrition-inflammation-atherosclerosis (MIA) syndrome [29] has been proposed to indicate the combination of these two conditions in such patients and their associations with atherosclerotic cardiovascular disease and poor outcome (see below). Chronic Inflammation as a Cause of Cachexia in CKD Patients Inflammation is defined as a localised adaptive response elicited by injury or destruction of tissues that serves to destroy, dilute, or sequester both the injurious agent and the injured tissue [30, 31]. The acute-phase response (or reaction) is a major pathophysiological phenomenon that accompanies inflammation and is associated with increased activity of pro-inflammatory cytokines [32]. With this reaction, normal homeostatic mechanisms are replaced by new set points that presumably contribute to defensive or adaptive capabilities [33]. Hence, inflammation is a physiological response, and as an acute response to infections, trauma, or toxic injury, it helps the body to defend against pathophysiological insults [34, 35]. Inflammation can become more subtle and less organ-specific and may involve many body organs or the entire organism. If inflammation becomes prolonged in the form of the chronic acute-phase reaction, it may lead to adverse consequences, such as a chronic decline in appetite, increased rate of protein depletion in skeletal muscle and other tissues, muscle and fat

3 6.4 Cachexia in Chronic Kidney Disease: Malnutrition-Inflammation Complex and Reverse Epidemiology 307 wasting, hypercatabolism, endothelial dysfunction, and atherosclerosis [35]. Inflammatory processes are common in CKD patients. Approximately 30 60% of North American [36, 37] and European [24, 38] maintenance dialysis patients have increased inflammatory markers, whereas long-term dialysis patients in Asian countries have a lower prevalence of inflammation [39, 40]. A recent study of 331 maintenance haemodialysis patients showed a strong association between anorexia and high levels of pro-inflammatory cytokines [41]. In this study, an appetite questionnaire was used, and the subjectively reported appetite was scored from 1 to 4, corresponding to normal to poor appetite. Inflammatory markers including serum concentrations of high-sensitivity C-reactive protein (hs-crp), tumour necrosis factor (TNF)-α and interleukin-(il)-6 were measured. Markers of inflammation were progressively higher in association with declining grades of appetite. There were statistically significant negative correlations between appetite score and serum CRP and TNF-α, and these correlations remained significant after case-mix multivariate adjustment for age, sex, race, and diabetes. After dichotomising the appetite score, the odds ratio (OR) of anorexia, controlled for case mix and other pertinent covariates, for each 10 pg increase in serum TNF-α/ml was 1.75 (confidence interval [CI] , p = 0.01) and for each 10 mg increase in hs-crp/l the OR was 2.31 (CI: , p < 0.001) [41]. Hence, inflammation is strongly associated with anorexia in dialysis patients. In recent years, more attention has been focused on inflammatory processes as the possible cause of accelerated atherosclerosis as well as protein-energy malnutrition and concurrent wasting syndrome, all of which lead to a poor outcome in those with underlying kidney disease. As mentioned above, chronic renal insufficiency per se is now considered as an independent risk factor for cardiovascular diseases [16, 18, 42]. It is believed that inflammation may play an important role in the increased prevalence of cardiovascular disease and mortality associated with renal insufficiency [23, 24, 27, 28, 43]. Renal failure may lead to increased inflammatory responses through a Table 1. Possible causes of chronic inflammation and wasting syndrome in chronic kidney disease (CKD) patients A. Causes of inflammation due to CKD or decreased GFR 1. Decreased clearance of pro-inflammatory cytokines 2.Volume overload a 3. Oxidative stress (e.g. oxygen radicals) a 4. Carbonyl stress (e.g. pentosidine and advanced glycation end-products) 5. Decreased levels of antioxidants (e.g. vitamin E, vitamin C, carotenoids, selenium, glutathione) a 6. Deteriorating protein-energy nutritional state and food intake a B. Coexistence of comorbid conditions 1. Inflammatory diseases with kidney involvement (SLE, HIV, etc.) 2. Increased prevalence of comorbid conditions a C. Additional inflammatory factors related to dialysis treatment 1. Haemodialysis a. Exposure to dialysis tubing b. Dialysis membranes with decreased biocompatiblility (e.g. cuprophane) c. Impurities in dialysis water and/or dialysate d. Back-filtration or back-diffusion of contaminants e. Foreign bodies (such as PTFE) in dialysis access grafts f. Intravenous catheter D. Peritoneal dialysis 1. Episodes of overt or latent peritonitis a 2. PD-catheter as a foreign body and its related infections 3. Constant exposure to PD solution a These factors may also be associated with protein-energy malnutrition. GFR, glomerular filtration rate; SLE, systemic lupus erythematosus; HIV, human immune deficiency virus; PTFE, polytetrafluoroethylene; PD, peritoneal dialysis

4 308 Kamyar Kalantar-Zadeh, Joel D. Kopple number of mechanisms that are listed in Table 1 and reviewed comprehensively elsewhere [44 47]. As indicated in Table 1, some of these factors may also result in protein-energy malnutrition and cachexia and consequently cause an overlap between malnutrition and inflammation. Comorbid conditions may contribute considerably to the development and maintenance of inflammation in dialysis patients. Due to the very high prevalence of comorbid conditions in these individuals, it is difficult at present to ascertain the role of inflammation in the absence of preexisting comorbidity. There is no uniform approach for assessing the degree of severity of inflammation in individuals with kidney disease [48]. Positive acute-phase reactants, such as serum CRP or ferritin, are markers whose levels are elevated during an acute episode of inflammation. The serum levels of negative acute-phase reactants, such as albumin or transferrin, decrease during an inflammatory process [34, 35, 44, 47]. Many negative acute-phase reactants are also traditionally known as nutritional markers, since their serum levels also decrease when there is a decline in nutritional status. Hence, it is not clear whether these markers have any specificity in the detection of either of these two conditions. Among pro-inflammatory cytokines, IL-6 is reported to have a central role in the pathophysiology of the adverse effects of inflammation in patients with renal disease [49 51]. These pro-inflammatory cytokines also may be engendered during oxidative stress, which per se can happen in the setting of protein-energy malnutrition [52]. Protein-Energy Malnutrition and Cachexia in CKD Protein-energy malnutrition in CKD patients is a state of decreased body pools of protein with or without fat depletion, or a state of diminished functional capacity. It is caused at least partly by inadequate nutrient intake relative to nutrient demand and is improved by nutritional repletion [31]. Hence, protein-energy malnutrition is engendered when the body s need for protein or energy fuels or both cannot be satisfied by the individual s nutrient intake diet [53]. Proteinenergy malnutrition is a common phenomenon in CKD patients and intensifies progressively as CKD stages advance over time (Fig. 2) [54, 55]. Proteinenergy malnutrition is a risk factor for poor quality of life and increased morbidity and mortality, including cardiovascular death, in these individuals [56, 57]. Various studies with different criteria have been used to establish the presence of protein-energy malnutrition in the CKD population, which has been studied more thoroughly in maintenance dialysis patients than among other CKD groups. Its reported prevalence varies between 18 and 75% among these individuals according to the type of dialysis modality, nutritional assessment tools, and origin of patient population [22, 58, 59]. Although protein-energy malnutrition per se neither requires nor precludes micronutrient malnutrition, many malnourished dialysis patients may also have a deficiency of various vitamins or trace elements [60, 61]. The aetiology of protein-energy malnutrition in CKD patients is probably multifaceted. Some probable causes are listed in Table 2 and have been reviewed in detail elsewhere [62 66]. As it is evident from Table 2, some of these factors can also lead to inflammation. Hence, the known overlap between malnutrition and inflammation in CKD patients may have its root at the aetiological level. The origin of protein-energy malnutrition generally precedes the need for dialysis treatment, and it is engendered progressively as the GFR falls below 60 ml/min, i.e. CKD stage 3 and above [55, 67]. Hypoalbuminaemia and hypocholesterolaemia have been shown to develop along with the progression of CKD stages, as shown in the Modification of Diet in Renal Disease (MDRD) Study [55] (Fig. 2) and other studies [67]. Assessment of MICS Classically, three major lines of inquiry, i.e. dietary intake, biochemical measures, and body composition, are used to assess the protein-energy nutritional status; a fourth category of nutritional assessment, composite indices that include a com-

5 6.4 Cachexia in Chronic Kidney Disease: Malnutrition-Inflammation Complex and Reverse Epidemiology 309 a b c d Fig. 2. Mean levels of biochemical measures of nutritional status as a function of glomerular filtration rate (GFR) in MDRD Study. The estimated mean levels with 95% confidence limits of biochemical nutritional markers are shown as a function of GFR (males solid line, females dashed line) controlling for age, race, and use of protein and energy restricted diets. In men, the slope of the relationship was greater at GFR = 12 than GFR = 55 ml/min/1.73 m 2 for serum total cholesterol (p = 0.014). a Males, N = 1065 (p = 0.004); females, N = 698 (p < 0.001). b Males, N = 1065 (p < 0.001); females, N = 698 (p < 0.001). c Males, N = 1063 (p = 0.052); females, N = 694 (p = 0.63). d Males, N = 1017 (p < 0.001); females, N = 664 (p < 0.001). (Modified from [54, 55]) bination of assessment measures within these categories, are also utilised, especially the Subjective Global Assessment of Nutrition (SGA) [68, 69] and Malnutrition-Inflammation Score (MIS) [28, 70]. The four categories of nutritional assessment tools are described in Table 3 and have been reviewed in detail elsewhere [22, 71]. As indicated in Table 3, many of these nutritional assessment tools are also designed to detect a combination of protein-energy malnutrition and inflammation and to grade their severity. Hence, the overlap between malnutrition and inflammation also exists at the diagnostic level, in addition to their overlapping aetiologies. No uniform approach has been agreed upon for rating the overall severity of protein-energy malnutrition. Among all four categories, dietary assessment is probably the most nutrition-specific entity. Another, rather nutrition-focused measure is the normalised protein equivalent of total nitrogen appearance (npna), also known as protein catabolic rate (npcr); a low npna is associated with increased hospitalisation and mortality in haemodialysis patients even when the dose of

6 310 Kamyar Kalantar-Zadeh, Joel D. Kopple Table 2. Conditions related to protein-energy malnutrition as a cause of wasting syndrome in CKD patients A. Inadequate nutrient intake 1. Anorexia a a. Due to uraemic toxicity b. Due to impaired gastric emptying c. Due to inflammation with or without comorbid conditions a d. Due to emotional and/or psychological disorders 2. Dietary restrictions a. Prescribed restrictions: low-potassium, low-phosphate regimens b. Social constraints: poverty, inadequate dietary support c. Physical incapacity: inability to acquire or prepare food or to eat B. Nutrient losses during dialysis 1. Loss through haemodialysis membrane into haemodialysate 2. Adherence to haemodialysis membrane or tubing 3. Loss into peritoneal dialysate C. Hypercatabolism due to comorbid illnesses 1. Cardiovascular diseases a 2. Diabetic complications 3. Infection and/or sepsis a 4. Other comorbid conditions a D. Hypercatabolism associated with dialysis treatment 1. Negative protein balance 2. Negative energy balance E. Endocrine disorders of uraemia 1. Resistance to insulin 2. Resistance to growth hormone and/or IGF-1 3. Increased serum level of or sensitivity to glucagons 4. Hyperparathyroidism 5. Other endocrine disorders F. Acidaemia with metabolic acidosis G. Concurrent nutrient loss with frequent blood losses a These factors may also be associated with inflammation; IGF-1, insulin-like growth factor 1 dialysis is standard or high (K t /V sp > 1.20) [20]. Although, as discussed above, it has been argued that appetite can be suppressed by inflammation and particularly by two pro-inflammatory cytokines, IL-6 and TNF-α [41, 72], the reduced nutritional state is still expected to induce malnutrition and its consequences regardless of the cause of anorexia. Some more frequently studied indicators of malnutrition in dialysis patients that are associated with clinical outcome include decreased dietary protein and energy intake [20, 60]; reduced weight-for-height [57], body mass index (BMI) [73 75] and total body fat percentage [76, 77]; decreased total body nitrogen [78, 79] and total body potassium [80]; reduced mid-arm muscle mass and skinfold thicknesses [81]; low serum concentrations of albumin [82], prealbumin (transthyretin) [83, 84], transferrin (TIBC) [68, 85], cholesterol [86, 87], and creatinine [88]; and a more abnormal score by such nutritional assessment tools as the SGA [89, 90] and MIS [28]. Although the foregoing measures of nutritional status have practical value, it should be recognised that each of these methods has its limitations. For example, serum albumin, transferrin, and prealbumin are negative acute-phase reactants and may reflect inflammation [46, 68, 91]. The SGA may also be a marker of the degree of sickness and comorbidity in maintenance dialysis patients [68]. During acute catabolic states and hypercatabolism, the urea nitrogen appearance may transiently increase independently of food intake [92]. More elaborate nutritional measures that have

7 6.4 Cachexia in Chronic Kidney Disease: Malnutrition-Inflammation Complex and Reverse Epidemiology 311 Table 3. Assessment tools for the evaluation of protein-energy malnutrition in CKD patients. (Data from [22, 71]) A. Nutritional intake 1. Direct: diet recalls and diaries, food frequency questionnaires 2. Indirect: based on urea nitrogen appearance: npna (npcr) B. Body composition 1. Weight based measures: BMI, weight-for-height, oedema-free/fat-free weight 2. Skin and muscle anthropometry via caliper: skinfolds, extremity muscle mass 3. Total body elements: total body potassium 4. Energy-beam based methods: DEXA, BIA, NIR 5. Other energy-beam related methods: total body nitrogen 6. Other methods: underwater weighing C. Laboratory values 1. Visceral proteins (negative acute-phase reactants): albumin, prealbumin, transferrin a 2. Lipids: cholesterol, triglycerides, other lipids, and lipoproteins a 3. Somatic proteins and nitrogen surrogates: creatinine, SUN 4. Growth factors: IGF-1, leptin 5. Peripheral blood cell count: lymphocyte count D. Scoring systems 1. Conventional SGA and its modifications (e.g. DMS, MIS, and CANUSA) a 2. Other scores: HD-PNI, others (e.g. Wolfson, Merkus, Merckman) a a These tools may also detect inflammation npna, Normalised protein nitrogen appearance; npcr, normalised protein catabolic rate; BMI, body mass index; DEXA, dual energy X-ray absorptiometry; BIA, bioelectrical impedance analysis; NIR, near infra-red interactance; SGA, subjective global assessment of nutritional status; DMS, dialysis malnutrition score; MIS, malnutrition inflammation score; CANUSA, Canada-USA study based modification of the SGA; HD-PNI, haemodialysis prognostic nutritional index; SUN, serum urea nitrogen; IGF-1, insuline-like growth factor 1 been used in dialysis and CKD patients include dual energy X-ray absorptiometry (DEXA) [93, 94], total body nitrogen or potassium measurements [79, 80, 95], underwater weighing [96], bioelectrical impedance analysis (BIA) [94], and near infra-red interactance (NIR) [76, 97]. Malnutrition-Inflammation Complex Syndrome The foregoing discussion, which is summarised in Tables 1 3, indicates that there is a major overlap among both possible aetiologic factors and the assessment tools for protein-energy malnutrition and inflammation in CKD patients. The link between protein-energy malnutrition and inflammation in CKD population may be an explanation for malnutrition-associated mortality [22, 23, 91]. Indeed, several investigators suggest that proteinenergy malnutrition and subsequent wasting syndrome and cachexia are a consequence of chronic inflammatory processes in patients with renal insufficiency [26, ]. Thus, chronic inflammation may be the missing link that causally ties protein-energy malnutrition to morbidity and mortality in these individuals. The following arguments have been proposed to indicate that the development of protein-energy malnutrition is secondary to inflammation: (1) Pro-inflammatory cytokines such as TNF-α not only promote catabolic processes, engendering both protein degradation and suppression of protein synthesis, but also induce anorexia [ ]. Low appetite has been shown to be associated with increased inflammatory markers in haemodialysis patients [41, 72]. (2) Dialysis patients with signs of inflammation are reported to develop weight loss and a negative protein balance even with an intact appetite, since there may be a shift in protein synthesis from muscle to acute-phase proteins as renal function declines [100]. (3) In both CKD and ESRD patients, albumin synthesis is suppressed when serum CRP is elevated [91, 104]. (4) Inflammation may also

8 312 Kamyar Kalantar-Zadeh, Joel D. Kopple lead to hypocholesterolaemia, a strong mortality risk factor in ESRD patients and also a marker of poor nutritional status [49]. The following counter-arguments have questioned the role of inflammation as a primary cause of protein-energy malnutrition: (1) In several studies, serum albumin and other indicators of protein-energy nutritional status correlate with indicators of protein intake independently of inflammatory status [ ]. (2) In dialysis patients, the association of serum albumin and CRP is not precise, and the reported correlation coefficients are usually less than 0.50; hence, other factor(s) possibly unrelated to inflammation must affect albumin [104, 105]. (3) Serum albumin concentrations usually do not fluctuate on a monthto-month basis, whereas serum CRP and other inflammatory markers do [107]. (4) In some but not all controlled trials involving patients with acute or chronic illnesses, the provision of nutritional support without management of inflammation improves hypoalbuminaemia and clinical outcome [ ]. (5) Malnourished CKD patients may be deficient in antioxidants, such as vitamin C or carotenoids, which may lead to increased oxidative stress and thus to inflammation [60]. A study using food frequency questionnaires to compare food intake of dialysis patients with that of normal individuals detected such dietary inadequacies, which could be attributed to nutritional restrictions such as low-potassium, low-phosphorus diets [60]. Studies in malnourished children have shown that protein-energy malnutrition may lead to oxidative stress, which can lead to increased activity of pro-inflammatory cytokines [52]. Moreover, in dialysis patients a reverse association has been reported between serum vitamin C (ascorbate) and serum CRP levels [112]. (6) There is evidence that certain nutrients, such as arginine and glutamine, enhance the immune response [113]. Moreover, preliminary data suggest that levocarnitine protects against endotoxins and also suppresses elaboration of TNF-α from monocytes [114]. Thus, protein-energy malnutrition may decrease host resistance and predispose to latent or overt infection, which is an inflammatory disorder. In summary, given the fact that mortality is still very high in dialysis patients (approximately 10 20% per year in Westernised countries), inflammation, independent of clinically evident comorbid conditions or malnutrition, cannot fully explain this extremely poor clinical outcome, especially since in otherwise healthy individuals inflammation has been found to be associated with an annual mortality rate that is only at 2 3% [115]. The foregoing considerations indicate that there is a lack of conclusive consensus with regard to the nature and direction of the association between protein-energy malnutrition and inflammation in renal cachexia. Hypoalbuminaemia, a strong and reliable predictor of cardiovascular disease and mortality in patients with renal insufficiency, is probably caused by both inflammation and protein-energy malnutrition, and it is not clear which one of these two conditions has a larger influence on serum albumin concentration [22, 104, 116]. Consequences of the Wasting Syndrome in CKD The renal wasting syndrome, be it due to inflammation, malnutrition, or both as MICS, has been found to be associated with cardiovascular diseases and atherosclerosis in the CKD population, overwhelming and even reversing the effect of traditional cardiovascular risk factors especially in maintenance dialysis patients (Table 4) [117]. MICS is associated with adverse relevant clinical consequences, including refractory anaemia, increased rate of atherosclerotic cardiovascular disease, and poor outcome, including low quality of life and increased hospitalisation and mortality, and may be the cause of reverse epidemiology in patients with renal failure (Fig. 3). Refractory Anaemia Elements of MICS and subsequent cachexia may blunt the responsiveness of anaemia to recombinant human erythropoietin (EPO) in CKD patients. Refractory anaemia appears to be more common in those dialysis patients who suffer from protein-energy malnutrition and/or inflammation [68, 118, 119]. Several previous studies

9 6.4 Cachexia in Chronic Kidney Disease: Malnutrition-Inflammation Complex and Reverse Epidemiology 313 Table 4. Reverse epidemiology of cardiovascular (CV) risk factors in dialysis patients: the effect of CV risk factors in maintenance dialysis patients is the opposite of the general population. (Data from [123]) Risk factors of cardio- Direction of the associations between risk factors and outcomes vascular disease General population Maintenance dialysis patients BMI High BMI and obesity are generally deleterious. High BMI, or weight for height, and moderate obesity are protective. Underweight is deleterious Serum cholesterol Hypercholesterolemia, high LDL and low HDL are deleterious. Hypercholesterolaemia (and maybe high LDL) is protective. Low serum cholesterol is deleterious BP Serum creatinine Total plasma homocysteine Serum iron Hypertension and even borderline high BP are deleterious. A mild to moderate increase in serum creatinine is an independent risk factor of CVD. A high level is a risk factor for increased CVD in the general population and likely in dialysis patients A high serum iron level is associated with haemochromatosis and poor outcome. Pre-dialysis low BP may indicate a deleterious state An increased pre-dialysis serum creatinine level is associated with a better survival Several recent studies have found that a low level is associated with increased risk of cardiovascular disease and mortality A low iron and transferrin saturation level has been recently found to be associated with higher mortality and hospitalisation in dialysis patients AGEs Energy (calorie) and/or protein intake Patients with higher AGE levels, such as diabetic patients, have a poor outcome. A high energy and food intake may be associated with risk of obesity and increased mortality. A recent report indicates a paradoxically reverse association between lower AGE levels and higher mortality in dialysis patients Increased protein intake is associated with better survival CVD, cardiovascular disease; MD, maintenance dialysis; LDL, low-density lipoprotein; HDL, high-density lipoprotein; BMI, body mass index; BP, blood pressure; AGEs, advanced glycation end-products report an association between anaemia and inflammation, such as occurs in dialysis patients, which is reflected by a high serum concentration of CRP [118, 120] or of pro-inflammatory cytokines such as IL-6 and TNF-α [121, 122]. We recently reported that serum IL-6 levels had the strongest correlation with administered EPO dose in 339 haemodialysis patients, and that the association remained statistically significant in different statistical analyses and after multivariate adjustments [124]. Both serum CRP and TNF-α showed similar trends and their associations with EPO dose remained significant in some but not all analysis modalities conducted in that study [124]. An inverse association was reported between markers of nutritional status or inflammation, e.g. serum prealbumin, TIBC, and total cholesterol concentration, and blood lymphocyte count, and the EPO dose [124]. Such associations are less well-described than the association between EPO dose and inflammation. Improving nutritional status in CKD patients may improve anaemia and lead to a lower required EPO dose. A cross-sectional study of 59 dialysis patients showed that the required EPO dose was higher in the poorly nourished patients as per SGA scoring [68]. In a meta-analysis by Hurot et al., L-carnitine administration, which is used to improve nutritional state, was associated with improved haemoglobin and a decreased EPO dose and EPO resistance in

10 314 Kamyar Kalantar-Zadeh, Joel D. Kopple Comorbid conditions: DM, Cardiovascular disease GFR clearance of inflammatory cytokines Endocrine disorders Acidosis, uremic toxins Oxidative stress, carbonyl stress Nutrient loss via dialysis Low nutrient intake Protein-Energy Malnutrition Inflammation Malnutrition-Inflammation Complex Syndrome (MICS) Dialysis Rx related factors Hypervolemia, endotoxinemia Anorexia Hypercatabolism Hypoalbuminemia EPO resistance Cytokines/CRP Cholesterol Cachexia Athero- Quality of life Hospitalisation BMI Reverse Epidemiology (risk factor paradox) _Mortality Fig. 3. The causes and consequences of malnutrition-inflammation complex syndrome (MICS). Modified from [123] BMI, body mass index; DM, diabetes mellitus; GFR, glomerular filtration rate; EPO, erythropoietin anaemic dialysis patients [125]. Moreover, anabolic steroids have also been used successfully to simultaneously improve both nutritional status and anaemia in dialysis patients [126]. Insulinlike growth factor (IGF)-1 is reported to enhance bone marrow progenitor cell proliferation in uraemic mice [127]. Hence, CKD-associated anaemia may represent both an EPO and a functional IGF-1 deficient state [127]. It is still not completely clear how MICS is related to CKD-associated refractory anaemia pathophysiologically. It has long been known that anaemia is frequently observed in patients suffering from chronic inflammatory disorders even with a normal kidney function [128]. Several mechanisms for cytokine-induced anaemia have been proposed, including impaired iron metabolism, suppression of endogenous EPO production, and reduced erythropoiesis [129, 130]. Serum ferritin, a measure of iron stores and a positive acute-phase reactant, has been shown to be paradoxically high in ESRD patients with refractory anaemia [131, 132]. Increased ferritin production may prevent iron delivery to erythrocyte precursors [131]. Moreover, the uptake of iron from the intestine is reduced in inflammatory states [129]. Patients with inflammatory diseases have inappropriately low levels of blood endogenous erythropoietin [133]. IL-1 and TNF-α have been shown to inhibit endogenous erythropoietin production in vitro [134]. Furthermore, increased release or activation of inflammatory cytokines, such as IL-6 or TNF-α, has been shown to have a suppressive effect on erythropoiesis [135]. IL-6 and IL-1 have been found to

11 6.4 Cachexia in Chronic Kidney Disease: Malnutrition-Inflammation Complex and Reverse Epidemiology 315 antagonise EPO s ability to stimulate bone marrow proliferation in culture [136]. Finally, patients with inflammation may be more prone to gastrointestinal bleeding [129, 130]. Atherosclerotic Cardiovascular Disease Cachexia, by virtue of MICS, may predispose CKD patients to atherosclerotic cardiovascular disease [24, 49, 51]. Dialysis patients with coronary heart disease often have hypoalbuminaemia and elevated levels of acute-phase reactants [24]. Moreover, progression of carotid atherosclerosis during dialysis may be related to IL-6 levels [137]. It should be noted that the cascade of inflammatory factors leading to an acute-phase reaction is counter-regulated by various anti-inflammatory cytokines, such as IL-10. Recently, Girndt et al., in a study of 300 haemodialysis patients [138], showed that the 1082A allele, which is associated with low production of IL-10, is associated with an increased risk of cardiovascular events. Inflammatory processes may promote proliferation and infiltration of inflammatory cells into the tunica intima of small arteries, including the coronary arteries; these processes lead to atherosclerosis and stenosis of blood vessels and consequent coronary and other vascular diseases [137, 139]. Epidemiological evidence suggests that inflammation may be linked to cardiovascular disease via specific low-grade infections, such as caused by Chlamydia pneumoniae [137, 139]. C. pneumoniae infection is shown to predict adverse outcome in dialysis patients [140], and elevated C. pneumoniae IgA titres predict progression of carotid atherosclerosis in these individuals [141]. Myeloperoxidase, an abundant enzyme secreted by neutrophils, may also link inflammation to oxidative stress and atherosclerosis in dialysis patients [142]. Indeed, recent data have shown that a functional variant of the myeloperoxidase gene is associated with cardiovascular disease in CKD patients [143]. Inflammation might also directly cause endothelial dysfunction via stimulation of intercellular adhesion molecules in CKD patients [144]. The association between elements of MICS and atherosclerosis has been underscored by some investigators, who have chosen the term malnutrition-inflammation-atherosclerosis (MIA) syndrome for this entity [29, 145]. Clinical Outcome and Reverse Epidemiology Many recent studies have suggested that proteinenergy malnutrition and inflammation in maintenance dialysis patients are associated with a decreased quality of life and increased hospitalisation and mortality, especially from cardiovascular diseases [10, 27, 28, 123]. Epidemiological studies indicate that hypoalbuminaemia and increased serum CRP are strong predictors of poor clinical outcome in the CKD population [36, 37]. Compared to traditional risk factors, such as obesity, hypercholesterolaemia, and hypertension, hypoalbuminaemia per se, which is generally considered an indicator of MICS, has one of the most striking and consistent associations with the prediction of clinical outcome in these individuals [146]. In highly industrialised, affluent countries, protein-energy malnutrition is an uncommon cause of poor outcome in the general population, whereas over-nutrition is associated with a greater risk of cardiovascular disease and has an immense epidemiological impact on the burden of this disease and on shortened survival. In contrast, in maintenance dialysis patients, undernutrition is one of the most common risk factors for adverse cardiovascular events [22, 117, 147]. Hence, certain markers that predict a low likelihood of cardiovascular events and an improved survival in the general population, such as decreased body mass index (BMI) [73 75, 148, 149] (Fig. 4) or lower serum cholesterol levels [49, 87], are risk factors for increased cardiovascular morbidity and death in dialysis patients [117]. Obesity, hypercholesterolaemia, and hypertension appear paradoxically to be protective features that are associated with greater survival of dialysis patients. A similar protective role has been described for high serum creatinine and total homocysteine levels in these patients [150]. The association between under-nutrition and adverse cardiovascular outcome in dialysis

12 316 Kamyar Kalantar-Zadeh, Joel D. Kopple Cardiovascular Death Hazard Ratio (time-dependent) Body Mass Index Categories (kg/m 2 ) Fig. 4. Reverse epidemiology of obesity in maintenance haemodialysis patients. Association between changes in BMI over time and cardiovascular mortality in a 2-year cohort of maintenance haemodialysis patients [149] patients, in contrast to the case in non-dialysis individuals, has been referred to as reverse epidemiology [117]. The aetiology of this inverse association between conventional risk factors and clinical outcome in dialysis patients is not clear. Several possible causes have been hypothesised, including survival bias and time discrepancy between competing risk factors (under-nutrition vs over-nutrition). However, the presence of MICS in dialysis patients offers the most plausible explanation for the existence of reverse epidemiology. Protein-energy malnutrition, inflammation, or the combination of the two are much more common in dialysis patients than in the general population, and many elements of MICS, such as low weight-for-height or BMI, hypocholesterolaemia, or hypocreatininaemia, are known risk factors of poor outcome in dialysis patients [117]. The existence of reverse epidemiology may have a bearing on the management of dialysis patients. It is possible that new standards or goals for such traditional risk factors as BMI, serum cholesterol, and blood pressure should be considered for these individuals. The phenomenon of risk-factor paradox is caused or at least accentuated by MICS in several ways. First, patients who are underweight or who have a low serum cholesterol, creatinine, or homocysteine, may be suffering from the MICS and its poor outcome. Thus, MICS may both cause these alterations and also be associated with increased mortality, either caused by the illnesses that engender MICS or by atherosclerotic cardiovascular disease that seems to be promoted by MICS [27, 151, 152]. Second, the above paradoxical factors may indicate a state of under-nutrition, which may predispose to infection or other inflammatory processes [22]. Finally, it has been argued that when individuals are malnourished, they are more susceptible to the ravages of inflammatory diseases [153]. Hence, any condition that potentially attenuates the magnitude of protein-energy mal-

13 6.4 Cachexia in Chronic Kidney Disease: Malnutrition-Inflammation Complex and Reverse Epidemiology 317 nutrition or inflammation should be favourable to dialysis patients. Suliman et al. reported a more specific example of the contribution of MICS to risk-factor reversal concerning hyperhomocysteinaemia in dialysis patients [ ]. In their study, plasma total homocysteine levels were shown to be dependent on nutritional status, protein intake, and serum albumin in haemodialysis patients. Dialysis patients with cardiovascular disease had lower plasma homocysteine levels as well as a higher prevalence of malnutrition and hypoalbuminaemia than those without cardiovascular disease. Furthermore, in another study, plasma total homocysteine was shown to rise during treatment of malnourished peritoneal dialysis patients who were given a daily exchange of an amino-acid-containing peritoneal dialysate (containing 1.7 g methionine) [157]. The puzzling inverse relationship between low blood pressure and poor outcome in the dialysis population might also be accounted for by nutritional status and/or inflammation. Iseki et al. [158] showed a significant association between a low diastolic blood pressure, hypoalbuminaemia, and risk of death in a cohort of 1243 haemodialysis patients who were followed for up to 5 years. The death rate was inversely correlated with diastolic blood pressure, which per se was positively correlated with serum albumin and negatively correlated with age. Hence, hypotension may in some cases be a manifestation of MICS in dialysis patients. Diagnosis and Management of MICS and Wasting Syndrome in CKD Since various markers of nutritional state and inflammation may independently predict outcome and assess different aspects of nutritional status, several researchers have tried to develop composite scores to identify the wasting syndrome and MICS in CKD. Ideally, such a scoring system would not only reflect the overall nutritional and inflammatory status of a chronic dialysis patient but would predict outcome. Wolfson et al. [81] introduced a composite score based on body weight, mid-arm muscle circumference, and serum albumin and found that 70% of haemodialysis patients were malnourished. Marckman et al. [159] developed a nutritional scoring system based on serum transferrin, relative body weight, triceps skinfold, and mid-arm muscle circumference. The SGA of nutritional status was designed primarily to evaluate surgical patients with gastrointestinal diseases [68]. It has since been employed in a number of epidemiological studies and clinical trials in dialysis patients [69]. SGA is significantly correlated with morbidity and mortality among dialysis patients [160, 161]. The K/DOQI has recommended SGA as an appropriate nutritional assessment tool for dialysis patients [162]. CANUSA (Canada-USA) [163] and other studies [164] have led to improved, more quantitative versions of the SGA. The recently developed MIS is based on the SGA, but also includes BMI and serum albumin and transferrin concentrations in an incremental fashion [28]. In two independently conducted longitudinal studies in haemodialysis patients, the MIS was strongly correlated with 12-month hospitalisation rates and mortality [28, 70] and had superior outcomepredictability compared to measurements of serum albumin [70]. MIS is believed to reflect the degree of severity of MICS in dialysis patients. Protein-energy malnutrition and inflammation lead to wasting syndrome and cachexia and are powerful predictors of death risk for CKD patients; thus, if they are treatable, it is possible that nutritional and anti-inflammatory interventions will improve poor outcome in the CKD population. Experience with nutritional support of sick or malnourished individuals who do not have CKD may provide some insight into the independent role of protein-energy malnutrition on clinical outcome in dialysis patients. Ample evidence suggests that maintaining an adequate nutritional intake in patients with a number of acute or chronic catabolic illnesses improves their nutritional status irrespective of its aetiology [165, 166]. In some of these studies, such improvement was associated with reduced morbidity and mortality and improved quality of life [167]. However, evidence as to whether nutritional treatment improves morbidity and mortality in dialysis patients is quite limited. There are no large-scale, randomised, prospective interventional studies

14 318 Kamyar Kalantar-Zadeh, Joel D. Kopple that have examined these issues. Among studies based on the nutritional response to such interventions, Kuhlmann et al. reported that prescription of 45 Kcal/kg/day and 1.5 g protein/kg/day, as compared to no such prescription, induced weight gain and improved serum albumin and other measures of nutritional status in malnourished haemodialysis patients [109]. Leon et al. reported that tailored nutritional intervention increased serum albumin levels in 52 haemodialysis patients, and this effect was observed even among patients with high serum CRP levels [108]. Several retrospective studies demonstrated a beneficial effect of intradialytic parenteral nutrition (IDPN) on clinical outcome [ ]. Recently, Pupim et al. [110] demonstrated that IDPN promoted a large increase in whole-body protein synthesis and a significant decrease in whole-body proteolysis in seven haemodialysis patients without signs of inflammation. However, a number of other studies of IDPN failed to show improvement in nutritional status or clinical outcome in dialysis patients [172, 173]. Many of these studies used small sample sizes, failed to restrict study subjects to those with protein-energy malnutrition, did not control for concurrent food intake, did not define or adjust appropriately for comorbid conditions, performed nutritional interventions for only short periods of time, or had only a short period of follow-up. Thus, until large-scale, prospective, randomised interventional studies are conducted, it will be difficult to ascertain the potential benefits of increasing nutritional intake in malnourished dialysis patients [173]. A number of other techniques have been employed for the prevention or treatment of protein-energy malnutrition in dialysis patients. Routine methods include preventing protein-energy malnutrition before the onset of dialysis therapy, dietary counselling, maintenance of an adequate dose of dialysis, avoidance of acidaemia, and aggressive treatment of superimposed catabolic illnesses [56]. More novel, non-dietary interventions in addition to IDPN include an appetite stimulant such as megestrol acetate [174], L-carnitine [175, 176], and growth factors including recombinant human growth hormone (rhgh) [177], IGF-1 [178], and anabolic steroids [179]. Nonetheless, although L-carnitine and these hormones may cause increased nitrogen retention, with the exception of the probable effects of L-carnitine administration on quality of life, none of these treatments have yet been shown to improve quality of life, morbidity, or mortality in dialysis patients. Although epidemiological evidence strongly links inflammation to poor outcome in individuals with renal insufficiency, it must be recognised that as yet there are no randomised clinical trials to indicate improvement of cachexia and its outcome by inflammation-reducing approaches. However, some treatment modalities may target inflammation directly, or they may focus on oxidative and carbonyl stress or endothelial dysfunction. The following approaches may be considered: (1) Statins (HMG-CoA reductase inhibitors) have been shown to decrease CRP levels independently of their lipid-lowering effects and may be associated with reduced mortality in CKD patients [180, 181]. (2) Angiotensinconverting enzyme inhibitors may have antiinflammatory properties in both the general population and in CKD patients [182], and are associated with delayed progression of chronic renal failure and improved outcome in these individuals [18]. (3) Vitamin E may have anti-inflammatory effects, and its administration may be associated with a decreased risk for cardiovascular mortality in chronic dialysis patients [183]. In the general population, some epidemiological studies indicate that a vitamin-e-rich diet is associated with a better cardiovascular outcome [184], but large clinical trials, such as the HOPE study, did not confirm such results [185, 186]. There are several forms of vitamin E, and it is possible that purified supplements, particularly the commonly used DL α-tocopherol (tocopheral) form, may not show the benefits of natural (national) dietary vitamin E or γ-tocopherol (tocopheral) components. A number of preliminary studies indicate that vitamin-ecoated dialysers may have favourable effects and anti-oxidant properties [187]. 4) Optimisation of dialysis treatment may improve inflammatory status in dialysis patients, and the type of dialysis membrane may have a bearing [188]. Ultra-pure dialysate and biocompatible membranes have been shown to decrease serum CRP [189, 190].

15 6.4 Cachexia in Chronic Kidney Disease: Malnutrition-Inflammation Complex and Reverse Epidemiology 319 Conclusions and Future Steps Both chronic inflammation and protein-energy malnutrition, together also referred to as MICS, are involved in engendering the commonly encountered wasting syndrome in CKD population. Hypoalbuminaemia is a marker of MICS and a strong outcome-predictor in these patients. The wasting syndrome in CKD patients per se is a chronic and slowly progressive condition that worsens over time in both pre-dialysis [54, 55] and dialysis patients [191]. Hence, a state of cachexia in slow motion can be described in these individuals. The effect of MICS on overall clinical and psychosocial aspects of CKD patients is so overwhelming that it even reverses the conventional associations between risk factors and outcome, leading to a counter-intuitive state of reverse epidemiology. However, dialysis patients are not the only population with a reverse epidemiology. Individuals with chronic heart failure (CHF) and geriatric populations have risk-factor reversal as well [192]. Hence, a better understanding of the role of chronic cachexia in CKD patients may help improve clinical management of not only these patients but also CHF, geriatric, and other vulnerable populations. According to an epidemiological study of over haemodialysis patients, if an intervention could increase serum albumin above 3.8 g/dl and by doing so improve survival in dialysis patients, almost one-third of all deaths among these patients could be hypothetically prevented or delayed. Since approximately patients out of over haemodialysis patients in the USA die every year, a hypoalbuminaemia-correcting intervention might theoretically prevent deaths every year [25]. If this is correct, there is a great need to develop effective nutritional and/or antiinflammatory interventions and to carry out randomised, prospective, controlled clinical trials to demonstrate the benefits of such interventions. References 1. National Kidney Foundation (2002) K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Am J Kidney Dis 39:S1-S Keith DS, Nichols GA, Gullion CM et al (2004) Longitudinal follow-up and outcomes among a population with chronic kidney disease in a large managed care organization. Arch Intern Med 164: Jones CA, McQuillan GM, Kusek JW et al (1998) Serum creatinine levels in the US population: third National Health and Nutrition Examination Survey. Am J Kidney Dis 32: Anonymous (2003) System USRD: USRD 2003 Annual Data Report; Atlas of End Stage Renal Diseases in the United States. National Institute of Health, National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda 5. Hirth RA, Held PJ, Orzol SM, Dor A (1999) Practice patterns, case mix, Medicare payment policy, and dialysis facility costs. Health Serv Res 33: Garella S (1997) The costs of dialysis in the USA. Nephrol Dial Transplant 12(Suppl 1): Anonymous (2002) United States Renal Data System: US Department of Public Health and Human Services, Public Health Service, National Institutes of Health, Bethesda, Maryland 8. Devereaux PJ, Schunemann HJ, Ravindran N et al (2002) Comparison of mortality between private for-profit and private not-for-profit hemodialysis centers: a systematic review and meta-analysis. JAMA 288: Eggers PW, Frankenfield DL, Greer JW et al (2002) Comparison of mortality and intermediate outcomes between medicare dialysis patients in HMO and fee for service. Am J Kidney Dis 39: Kalantar-Zadeh K, Kopple JD, Block G, Humphreys MH (2001) Association among SF36 quality of life measures and nutrition, hospitalization, and mortality in hemodialysis. J Am Soc Nephrol 12: Carlson DM, Duncan DA, Naessens JM, Johnson WJ (1984) Hospitalization in dialysis patients. Mayo Clin Proc 59: Fried L, Abidi S, Bernardini J et al (1999) Hospitalization in peritoneal dialysis patients. Am J Kidney Dis 33: Habach G, Bloembergen WE, Mauger EA et al (1995) Hospitalization among United States dialysis patients: hemodialysis versus peritoneal dialysis. J Am Soc Nephrol 5: Foley RN, Parfrey PS, Sarnak MJ (1998) Epidemiology of cardiovascular disease in chronic renal disease. J Am Soc Nephrol 9:S16-S Foley RN, Parfrey PS, Sarnak MJ (1998) Clinical epidemiology of cardiovascular disease in chronic renal disease. Am J Kidney Dis 32:S112 S Shlipak MG, Heidenreich PA, Noguchi H et al (2002) Association of renal insufficiency with treatment and outcomes after myocardial infarction in elderly patients. Ann Intern Med 137: