Serum Albumin as Predictor of Nutritional Status in Patients with ESRD

Transcription

1 CJASN epress. Published on June 21, 2012 as doi: /CJN Article Serum Albumin as Predictor of Nutritional Status in Patients with ESRD Thiane Gama-Axelsson, Olof Heimbürger, Peter Stenvinkel, Peter Bárány, Bengt Lindholm, and Abdul Rashid Qureshi Summary Background and objectives Serum albumin is a widely used biomarker of nutritional status in patients with CKD; however, its usefulness is debated. This study investigated serum albumin and its correlation with several markers of nutritional status in incident and prevalent dialysis patients. Design, setting, participants, & measurements In a cross-sectional study, serum albumin (bromocresol purple), and other biochemical (serum creatinine), clinical (subjective global assessment [SGA]), anthropometric (handgrip strength; skinfold thicknesses), and densitometry (dual-energy x-ray absorptiometry) markers of nutritional status were assessed in 458 incident (61% male; mean age, years; GFR, ml/min per 1.73 m 2 ; recruited ) and 383 prevalent (56% male; mean age, years; recruited ) dialysis patients. Results In incident patients, serum albumin was correlated with age (b =20.15; P,0.001), diabetes (b=20.30; P,0.001), high-sensitivity C-reactive protein (b =20.37; P,0.001), and urinary albumin excretion (b=20.38; P,0.001) but less so with poor nutritional status (SGA score. 1; b=20.19; P,0.001). In prevalent patients, serum albumin was correlated with age (b=20.15; P,0.001), high-sensitivity C-reactive protein (b=20.30; P,0.001), diabetes (b=20.31; P,0.001), and SGA score. 1(b=20.16; P,0.001). In predicting nutritional status assessed by SGA and other markers, adding serum albumin to models that included age, sex, diabetes, and cardiovascular disease did not significantly increase explanatory power. Conclusions In incident and prevalent dialysis patients, serum albumin correlates poorly with several markers of nutritional status. Thus, its value as a reliable marker of nutritional status in patients with ESRD is limited. Clin J Am Soc Nephrol 7: ccc ccc, doi: /CJN Divisions of Baxter Novum and Renal Medicine, Department of Clinical Science, Intervention and Technology, Karolinska Institutet, Stockholm, Sweden Correspondence: Dr. Bengt Lindholm, Divisions of Baxter Novum and Renal Medicine, Department of Clinical Science, Intervention and Technology Karolinska Institute, K56 Karolinska University Hospital, Huddinge 14186, Stockholm, Sweden. Bengt. Lindholm@ki.se Introduction Biochemical analyses are routinely used to assess and monitor nutritional status in patients with CKD. However, none of the currently favored biochemical nutritional markers have been demonstrated to accurately reflect nutritional status in CKD (1 3). Nevertheless, serum albumin is still being widely used for research purposes and, in the clinical setting, as a biomarker of nutritional status (4). In patients with CKD, factors such as overhydration and protein loss into urine and dialysate reduce serum albumin concentrations. Also limiting the usefulness of serum albumin as a nutritional marker are counter-regulatory mechanisms. In the short term, protein deficiency decreases the rate of albumin synthesis (5), but over time compensation occurs through a decrease in albumin breakdown and a shift of albumin from the extravascular to the intravascular space. Among numerous complications of CKD, progressive loss of body protein mass and energy reserves is one of the most typical and detrimental. This loss has been termed protein-energy wasting (PEW) (6). It is morecommoninpatientswithadvancedstagesof CKD and may affect 18% 75% of patients with ESRD (6 8). PEW is to a large extent caused by inadequate nutritional intake leading to malnutrition; still, increased catabolism that is due to chronic low-grade inflammation and leads to wasting seems to be of equal importance (6,9). According to the definition of PEW (6), patients could have PEW and not necessarily be undernourished. There is considerable overlap among poor nutritional status, malnutrition (representing mainly chronic undernutrition), and PEW; however, these terms have different definitions and meanings and should not be used interchangeably (10). No single marker can be regarded as ideal to assess nutritional status (11), especially not in patients with CKD, in whom various metabolic alterations and other confounding factors, such as fluid overload, are common. However, subjective global assessment (SGA), anthropometry, and dual-energy x-ray absorptiometry (DEXA) are widely used methods to assess nutritional status in patients with CKD and also are recommended by the National Kidney Foundation Kidney Disease Outcomes Quality Initiative (11,12). Furthermore, SGA was found to be an appropriate tool for cross-sectional assessment of nutrition status in a study of patients with CKD (13). Vol 7 September, 2012 Copyright 2012 by the American Society of Nephrology 1

2 2 Clinical Journal of the American Society of Nephrology Because albumin has a fairly long half-life and is present in large quantities (14), the effect of a temporarily decreased protein intake on concentrations of albumin is also limited (15).Instead,serumalbuminismorelikelytobeinfluenced by its role as an acute phase reactant. Thus, low serum albumin levels in dialysis patients are strongly associated with inflammation (16). Because of the continuing controversy and the widespread use of serum albumin as a biomarker of nutritional status in patients with CKD, we investigated serum albumin and its correlation with several biomarkers of nutritional status in cross-sectional cohorts of incident and prevalent dialysis patients. Materials and Methods Patients and Study Design This is a cross-sectional study of data from two independent cohorts coordinated at the Department of Renal Medicine, Karolinska University Hospital Huddinge, Karolinska Institutet, Stockholm, Sweden: incident dialysis patients who were investigated just before or in conjunction with start of dialysis therapy (17) and prevalent patients who, at different time periods, had been undergoing peritoneal dialysis (18) or hemodialysis (19,20) for more than 3 months in the Stockholm/Uppsala area. Exclusion criteria were current hospitalization, clinical signs of infection, or acute vasculitis at the time of enrollment or within 3 weeks before that date, as well as unwillingness to participate in the study or inability to give informed consent. Patients were categorized according to CKD status on the day of inclusion as incident dialysis patients (n=458; 61% male; mean age, years; GFR, ml/min) or prevalent dialysis patients (n=383; hemodialysis, n=347; peritoneal dialysis, n=36; 56% male; mean age, years). Each patient s medical chart was reviewed to extract data on underlying cause of CKD, cardiovascular disease (CVD) history, and diabetes mellitus. The study protocols were approved by the Ethics Committee of Karolinska Institutet Hospital and Uppsala University Hospital. Informed consent was obtained from all patients before their inclusion in the study. Anthropometric Evaluation Body weight, body mass index (BMI; in kg/m 2 ), and anthropometric measurements were obtained on a dialysis day in the prevalent dialysis patients (immediately after the dialysis session for the prevalent hemodialysis patients), and, for the incident dialysis patients, at the time of or within 1 week of blood sample collection, and fat mass and lean body mass (LBM) were assessed from these data according to the method of Durnin et al. (21). Fat mass was assessed by using four skinfold thicknesses (biceps, triceps, subscapular, and suprailiac), measured with a conventional skinfold caliper (Cambridge Scientific Instruments,Cambridge,MD).In addition, LBM was measured in incident dialysis patients by DEXA using the DPX-L device (Lunar Corp., Madison, WI). Handgrip strength was measured in both the dominant and nondominant hands by using a Harpenden Handgrip Dynamometer (Yamar, Jackson, MI). Each measurement was repeated three times for each arm, and the highest value for each arm was noted. For our analysis, we used the dominant arm for handgrip strength because fistulas were usually placed in the nondominant arm. Laboratory Analysis Blood samples were collected in the morning before the dialysis session. The plasma was separated within 30 minutes, and samples were kept frozen at 270 C if not analyzed immediately. Concentrations of high-sensitivity C-reactive protein (hscrp; high-sensitivity nephelometry assay), serum creatinine, and serum albumin (bromocresol purple) were measured by routine methods at the Department of Laboratory Medicine, Karolinska University Hospital, Huddinge. Inflammation status was defined as CRP level #10 mg/l (noninflamed) and CRP.10 mg/l (inflamed). GFR was estimated in incident dialysis patients by the mean of urinary creatinine and urea clearances during a 24-hour urine collection. Nutritional Status Assessment Nutritional status was assessed using the SGA score (22). The SGA has been validated in patients with CKD and consists of six components: three subjective, patientperformed assessments on the rate of the history of weight loss, incidence of poor appetite, and incidence of vomiting and three evaluator-performed assessments that subjectively grade muscle wasting, the presence of edema, and the loss of subcutaneous fat. On the basis of these assessments, each patient received a nutritional status score: 1 = normal nutritional status, 2 = mild malnutrition, 3 = moderate malnutrition, and 4 = severe malnutrition. For this study, poor nutritional status was defined as an SGA score.1; a score of 1 indicates normal nutritional status. Statistical Analyses All variables are expressed as mean 6 SD or as median (10thand90thpercentiles),unless otherwise indicated. Statistical significance was set at the level of P,0.05. Multivariate logistic regression analyses were used to study the relative associations of selected markers with serum albumin. A determinant of serum albumin was analyzed using linear multivariate regression analysis. All statistical analyses were performed using SAS statistical software, version 9.2 (SAS Institute, Inc., Cary, NC). Results Baseline Characteristics The studied population consisted of 841 patients (458 incident and 383 prevalent dialysis patients). Demographic, clinical, and important biochemical characteristics are presented in Table 1. Briefly, the incident dialysis patients comprised 61% men (average age, years); 30% of patients had diabetes, and 35% had a history of CVD. The prevalent dialysis group comprised 56% men (average age, years); 22% of patients had diabetes and 63% had a history of CVD. Serum albumin level was significantly lower in incident than in prevalent dialysis patients ( versus g/l; P,0.001). Clinical Correlates of Serum Albumin Concentration The results of multiple regression models predicting serum albumin (g/l) in both incident dialysis and prevalent dialysis patients are shown in Tables 2 and 3. Briefly, in incident dialysis patients, multivariate analysis that included age, sex, hscrp, presence of diabetes mellitus,

3 Clin J Am Soc Nephrol 7: ccc ccc, September, 2012 Serum Albumin as Predictor of Nutritional Status, Gama-Axelsson et al. 3 Table 1. Clinical characteristics, anthropometric measurements, and biochemical markers of nutritional and inflammatory status in incident and prevalent patients undergoing dialysis Variable Incident Dialysis Group (n=458) Prevalent Dialysis Group (n=383) P Value General age (yr) ,0.001 men (%) diabetes mellitus (%) cardiovascular disease (%) GFR (ml/min per 1.73 m 2 ) 663 ND hemoglobin (g/l) ,0.001 urinary albumin excretion (mg/24 hr) 1981 ( ) ND Nutritional variables SGA score.1 (%) 31 42,0.001 serum albumin (g/l) ,0.001 body mass index (kg/m 2 ) ,0.01 handgrip strength (% of control) ,0.001 serum creatinine (mmol/l) ,0.05 Inflammation variable hscrp (mg/l) 9 ( ) 10 ( ),0.05 Data are presented as mean 6 SD, median (25th 75th percentile), or percentage. Handgrip strength was normalized with the measurements from healthy persons. ND, not determined; SGA, subjective global assessment; hscrp, high-sensitivity C-reactive protein. Table 2. Multiple regression models for serum albumin in incident dialysis patients (n=458) Variable Crude (r 2 =0.01) Model 1 (r 2 =0.09) Model 2 (r 2 =0.17) Model 3 (r 2 =0.16) Model 4 (r 2 =0.32) Age (0.02) a (0.27) a (0.61) (0.60) (0.08) Female sex 0.01 (0.78) 0.05 (0.27) 0.05 (0.26) (0.02) a hscrp (0.001) a (0.001) a (0.001) a (0.07) Diabetes (0.001) a (0.001) a (0.004) a CVD (0.70) (0.60) (0.79) SGA score (0.04) a (0.001) a Urinary albumin excretion rate (mg/24 hr) (n=261) (0.001) a hscrp, high-sensitivity C-reactive protein; CVD, cardiovascular disease; SGA, subjective global assessment. SGA.1 denotes poor nutritional status according to SGA. a Statistically significant result: b (P value). CVD, clinical signs and symptoms indicating a poor nutritional status (defined as SGA score.1), and urinary albumin excretion rate showed that serum albumin was significantly correlated with female sex (b=2 0.13; P=0.02), diabetes mellitus (b=20.18; P=0.004), SGA score.1 (b=20.19; P=0.001), and, especially, urinary albumin excretion (b=20.42; P=0.001). In prevalent dialysis patients, and after adjustment for age, sex, hscrp, presence of diabetes mellitus, CVD, and poor nutritional status (SGA score.1), serum albumin was associated with age (b=20.14; P=0.05), hscrp (b=20.34; P=0.001), diabetes mellitus (b=20.11; P=0.04), and SGA score.1 (b=20.14; P=0.003). Effect of Inflammation Figure 1 shows that serum albumin levels among patients with normal nutritional status did not differ between inflamed and noninflamed patients; this was true among both incident and prevalent patients. However, patients with poor nutritional status both incident and prevalent who were also inflamed had a significantly lower serum albumin level (P,0.001) than noninflamed patients (Figure 1). We separately analyzed the patients who were not inflamed and found that in the noninflamed incident dialysis group, both LBM and handgrip strength, but not albumin, were significantly lower in patients with a poor nutritional status (SGA score.1) (data not shown). In the noninflamed prevalent dialysis patients, those with a poor nutritional status (SGA score.1) were older and had significantly lower fat mass, lower LBM, and lower serum albumin levels ( versus g/l; P,0.05); handgrip strength did not differ. However, after adjustment for age, sex, and diabetes mellitus, the difference in serum albumin levels

4 4 Clinical Journal of the American Society of Nephrology Table 3. Multiple regression models for serum albumin in prevalent dialysis patients (n=383) Variable Crude (r 2 =0.03) Model 1 (r 2 =0.04) Model 2 (r 2 =0.19) Model 3 (r 2 =0.21) Age (0.009) a (0.009) a (0.002) a (0.005) a Female sex 0.12 (0.01) a 0.06 (0.2) 0.03 (0.48) hscrp (0.001) a (0.001) a (0.001) a Diabetes (0.009) a (0.04) a CVD (0.18) (0.19) SGA score (0.003) a hscrp, high-sensitivity C-reactive protein; CVD, cardiovascular disease; SGA, subjective global assessment. SGA score.1 denotes poor nutritional status according to SGA. a Statistically significant result: b (P value). Figure 1. Serum albumin levels were not affected by poor nutritional status in noninflamed patients but were markedly lower in patients with poor nutritional status who were also inflamed. Box plot of serum albumin by nutritional status in inflamed and noninflamed patients. (A) Incident (n=429) and (B) prevalent (n=380) dialysis patients. The whiskers represent 10th and 90th percentiles. hscrp, high-sensitivity C-reactive protein; NS, not significant. between patients with normal and those with poor nutritional status was no longer significant (data not shown). Effect of Urinary Albumin Loss In Figure 2, incident dialysis patients were divided by inflammation state and albuminuria, as well as by serum albumin levels. Albuminuria was a strong predictor of serum albumin levels in both inflamed and noninflamed patients. Bivariate Correlations between Serum Albumin and Other Markers of Nutritional Status Using receiver-operating characteristic curve (ROC) analysis, the areas under the curve for serum albumin (g/l) as a predictor of SGA score.1 were0.62intheincidentdialysis patients (albumin explained only 12% of the variation in SGA) and 0.64 in the prevalent dialysis patients (albumin explained 14% of the variation).

5 Clin J Am Soc Nephrol 7: ccc ccc, September, 2012 Serum Albumin as Predictor of Nutritional Status, Gama-Axelsson et al. 5 Figure 2. Albuminuria was a strong predictor of serum albumin levels in both inflamed and noninflamed patients. hscrp, high-sensitivity C-reactive protein. The crude unadjusted associations (in incident and prevalent patients, respectively) between albumin and handgrip strength (r 2 =0.02 and 0.08), LBM estimated by anthropometrics (r 2 =0.005 and 0.01), LBM estimated by DEXA (only in incident patients; r 2 =0.006), and BMI (r 2 =0.003 and ) were very weak. Relative Contributions of Factors Explaining the Variation of Nutritional Status In a separate analysis, we analyzed how well combinations of different factors (clinical and demographic data and different nutritional markers) could predict the variation of the different estimates of nutritional status (SGA, handgrip strength, LBM, and BMI), along with the added value of also using information on serum albumin (Figure 3). The traditional clinical factors (age, sex, and presence of diabetes mellitus and CVD) could explain of the variation in different estimates of nutritional status (SGA, handgrip strength, LBM, and BMI) among the incident dialysis patients; addition of serum albumin in the model did not increase the explanatory power (pseudo r/r 2 = ) (Figure 3A). When all the investigated measures of nutritional status (including serum albumin) were included in the model, the explanatory power increased to for the different estimates of nutritional status. Similarly, in the prevalent dialysis patients, the traditional clinical factors (age, sex, presence of diabetes mellitus, and CVD) could explain of the variation in the different estimates of nutritional status (SGA, handgrip strength, LBM calculated with anthropometrics, and BMI). Again, the addition of serum albumin in the model did not increase the explanatory power, which remained for the different nutritional markers (Figure 3B). When all the investigated measures of nutritional status (including serum albumin) were included in the model, the explanatory power increased to for the different estimates of nutritional status. Discussion Although serum albumin is still a widely used biomarker for nutritional status in CKD (4) and has a strong association with a poor outcome (23,24), its usefulness as a nutritional marker has been challenged (4). In a cohort study, we investigated serum albumin and its correlation with several markers of nutritional state in both incident and prevalent dialysis patients. In incident dialysis patients with residual renal function, urinary albumin was the strongest predictor of serum albumin levels. However, after adjustments for age, sex, hscrp, diabetes mellitus, and CVD, an SGA score.1 (indicating presence of poor nutritional status) was independently but, compared with other predictors, relatively weakly associated with low serum albumin levels in both groups. It has traditionally been assumed that serum albumin is an indicator of nutritional status, and serum albumin is predominantly low in patients with CKD. Thus, it has been considered a supported sign of malnourishment. However, the effect of a decreased protein intake on concentrations of serum albumin is limited by albumin s considerable halflife (up to 20 days) and abundance (14). Thus, even in extreme cases of malnutrition, such as marasmus and anorexia nervosa, serum albumin levels remain normal (25). Furthermore, results from the Minnesota study (15) show that after induced, prolonged starvation in healthy volunteers, with participants showing multiple signs of

6 6 Clinical Journal of the American Society of Nephrology Figure 3. Negligible added value of using serum albumin to predict variation of different estimates of nutritional status over and above prediction based on age, sex, diabetes, and CVD. (A) Predictive strength (expressed as pseudo r for subjective global assessment [SGA] score and r 2 for handgrip strength [HGS], lean body mass [LBM] per dual-energy x-ray absorptiometry, LBM per anthropometric testing, and body mass index [BMI]) of available information of traditional clinical factors (Model 1, M1), Model 1 + serum albumin (Model 2, M2), and use of all available nutritional information (Model 3, M3), for the prediction of the variability of different nutritional markers in incident dialysis patients. The results show that the added value of using serum albumin over and above prediction obtained by using only information about age, sex, and comorbid diabetes and CVD is negligible. Traditional factors were age, sex, diabetes, and CVD. Other factors were all additional available data on nutritional status: SGA score, HGS, LBM (estimated by two methods), BMI, high-sensitivity C-reactive protein, and smoking. (B) The predictive strength (expressed as pseudo r for SGA score and r 2 for HGS, LBM per anthropometric testing, and BMI) of clinically available information about traditional factors (Model 1, M1), Model 1 + serum albumin (Model 2, M2), and use of all available nutritional information (Model 3, M3), for the prediction of different nutritional markers (expressed as pseudo r for SGA score and r 2 for HGS, LBM per anthropometric testing, and BMI) in prevalent dialysis patients. The results show that the added value of using serum albumin over and above prediction obtained by using information on age, sex, and comorbid diabetes and CVD is negligible. Traditional factors were age, sex, diabetes, and CVD. Other factors were all additional data information about nutritional status: SGA, HGS, LBM (estimated by anthropometric testing), BMI, and high-sensitivity C-reactive protein. malnutrition, serum albumin levels changed only slightly. Given these findings, as well as the study showing that healthy individuals and patients with CKD have a similar plasma albumin degradation rate (26), it would be surprising if malnutrition was the main factor influencing serum albumin in CKD. Indeed, patients with CKD commonly have comorbid conditions, fluid overload, and low-grade inflammation, all of which are known to affect serum albumin concentrations (27). Supporting this interpretation, our study did not reveal strong association between nutritional status and serum albumin in incident or prevalent dialysis patients. Whereas serum albumin was a weak predictor of nutritional status defined by SGA score, associations between serum albumin and other markers of nutritional status, suchashandgripstrengthandlbm,wereevenweaker. Our data showing this clear lack of value of serum albumin as a predictor of nutritional status (Figure 3, A and B) correspond to and extend previous results reported by us (28) and others (29) showing that serum albumin is a poor predictor of nutritional status in dialysis patients. The current results support views expressed by

7 Clin J Am Soc Nephrol 7: ccc ccc, September, 2012 Serum Albumin as Predictor of Nutritional Status, Gama-Axelsson et al. 7 Friedman and Fadem (4) that there is a rationale for reconsidering albumin as a marker of illness rather than nutrition. In the current study, the lack of predictive power of nutritional status by serum albumin levels among the incident dialysis patients may be partly explained by the strong negative correlations between serum albumin with urinary albumin losses, and with hscrp. Inflammation, usually assessed with hscrp in patients with CKD, is well described as associated with PEW (9,17), and there are also putative causal pathways linking low serum albumin and inflammation (20). Indeed, studies (1,24) have shown that inflammation is consistently associated with low levels of serum albumin in uremia. Although our findings underline the limited value of serum albumin as a predictor of nutritional status in incident and prevalent dialysis patients, they also underscore the massive importance of urinary albumin loss for circulating albumin levels in incident dialysis patients with residual renal function. This may be of more general importance, as Warnock et al. (30) found that increased albuminuria is an independent risk factor for all-cause mortality in a prospective observational study consisting of 17,393 healthy participants. Our study has many limitations. First, patients in the prevalent dialysis group were older than the incident dialysis patients, which probably explains differences in hscrp and handgrip strength. Second, the study design by its nature precludes inferring causalities. Third, residual renal function and urinary albumin were not evaluated in prevalent dialysis patients; however, most of them were anuric. Fourth, the prevalent dialysis patients were included from different biomedical cohorts. Fifth, although overhydration may influencesomeofour findings, we did not include a marker of fluid status. Finally, although our findings did not support the value of serum albumin as a marker of nutritional status, we did not assess serum albumin as a marker of PEW; this, from a conceptual aspect, could be questionable considering that a low serum albumin level is a readily usable criterion for PEW classification. In summary, among incident and prevalent dialysis patients, serum albumin correlates poorly with several markers of nutritional status; thus, its value as a reliable marker of nutritional status in patients with ESRD is limited. Acknowledgments We would like to thank the patients and personnel at KBC (Annika Nilsson, Ann-Christin Emmoth, and Ulrika Jensen) and KFC (Björn Anderstam, Monica Eriksson, and Ann-Christin Bragfors-Helin) involved in the creation of these cohorts. We acknowledge the support from Martin Rind s Foundation (A.R.Q.), the Swedish Medical Research Council (P.S.), and the Wertman Foundation (P.S.). Baxter Novum is the result of a grant to the Karolinska Institutet from Baxter Healthcare Corporation. These data were presented in abstract form at the 33rd Congress of the European Society for Clinical Nutrition and Metabolism, September 3 6, 2011, Gothenburg, Sweden. Disclosures B.L. is employed by Baxter Healthcare Corporation. P.S. is a member of the scientific advisory board of Gambro AB. None of the other authors declare any conflicts of interest. References 1. de Mutsert R, Grootendorst DC, Indemans F, Boeschoten EW, Krediet RT, Dekker FW; Netherlands Cooperative Study on the Adequacy of Dialysis-II Study Group: Association between serum albumin and mortality in dialysis patients is partly explained by inflammation, and not by malnutrition. J Ren Nutr 19: , Stenvinkel P, Barany P, Chung SH, Lindholm B, Heimbürger O: A comparative analysis of nutritional parameters as predictors of outcome in male and female ESRD patients. Nephrol Dial Transplant 17: , Ikizler TA, Wingard RL, Harvell J, Shyr Y, Hakim RM: Association of morbidity with markers of nutrition and inflammation in chronic hemodialysis patients: A prospective study. Kidney Int 55: , Friedman AN, Fadem SZ: Reassessment of albumin as a nutritional marker in kidney disease. J Am Soc Nephrol 21: , Rothschild MA, Oratz M, Schreiber SS: Regulation of albumin metabolism. Annu Rev Med 26: , Fouque D, Kalantar-Zadeh K, Kopple J, Cano N, Chauveau P, Cuppari L, Franch H, Guarnieri G, Ikizler TA, Kaysen G, Lindholm B, Massy Z, Mitch W, Pineda E, Stenvinkel P, Trevi~no- Becerra A, Wanner C, Wanner C: A proposed nomenclature and diagnostic criteria for protein-energy wasting in acute and chronic kidney disease. Kidney Int 73: , Kopple JD: McCollum Award Lecture, 1996: Protein-energy malnutrition in maintenance dialysis patients. Am J Clin Nutr 65: , Kalantar-Zadeh K, Ikizler TA, Block G, Avram MM, Kopple JD: Malnutrition-inflammation complex syndrome in dialysis patients: Causes and consequences. Am J Kidney Dis 42: , Stenvinkel P, Heimbürger O, Lindholm B, Kaysen GA, Bergström J: Are there two types of malnutrition in chronic renal failure? Evidence for relationships between malnutrition, inflammation and atherosclerosis (MIA syndrome). Nephrol Dial Transplant 15: , Mak RH, Ikizler AT, Kovesdy CP, Raj DS, Stenvinkel P, Kalantar- Zadeh K: Wasting in chronic kidney disease. J Cachexia Sarcopenia Muscle 2: 9 25, Fouque D, Vennegoor M, ter Wee P, Wanner C, Basci A, Canaud B, Haage P, Konner K, Kooman J, Martin-Malo A, Pedrini L, Pizzarelli F, Tattersall J, Tordoir J, Vanholder R: EBPG guideline on nutrition. Nephrol Dial Transplant 22[Suppl 2]: ii45 ii87, Kopple JD: National kidney foundation K/DOQI clinical practice guidelines for nutrition in chronic renal failure. Am J Kidney Dis 37[Suppl 2]: S66 S70, Campbell KL, Ash S, Bauer JD, Davies PS: Evaluation of nutrition assessment tools compared with body cell mass for the assessment of malnutrition in chronic kidney disease. J Ren Nutr 17: , Kirsch R, Frith L, Black E, Hoffenberg R: Regulation of albumin synthesis and catabolism by alteration of dietary protein. Nature 217: , Keys ABJ, Henschel A, Mickelsen O, Taylor H: The Biology of Human Starvation, Minneapolis, University of Minnesota Press, Kaysen GA, Dubin JA, Müller HG, Rosales L, Levin NW, Mitch WE; HEMO Study Group NIDDK: Inflammation and reduced albumin synthesis associated with stable decline in serum albumin in hemodialysis patients. Kidney Int 65: , Stenvinkel P, Heimbürger O, Paultre F, Diczfalusy U, Wang T, Berglund L, Jogestrand T: Strong association between malnutrition, inflammation, and atherosclerosis in chronic renal failure. Kidney Int 55: , Qureshi AR: Malnutrition in patients with chronic renal failure. PhD thesis. ISBN: X. Karolinska Institutet, Stockholm, Sweden, Meuwese CL, Halbesma N, Stenvinkel P, Dekker FW, Molanaei H, Qureshi AR, Barany P, Heimburger O, Lindholm B,KredietRT,BoeschotenEW,CarreroJJ:Variationsin C-reactive protein during a single haemodialysis session do not associate with mortality. Nephrol Dial Transplant 25: , 2010

8 8 Clinical Journal of the American Society of Nephrology 20. Carrero JJ, Qureshi AR, Axelsson J, Avesani CM, Suliman ME, Kato S, Bárány P, Snaedal-Jonsdottir S, Alvestrand A, Heimbürger O, Lindholm B, Stenvinkel P: Comparison of nutritional and inflammatory markers in dialysis patients with reduced appetite. Am J Clin Nutr 85: , Durnin JV, Womersley J: Body fat assessed from total body density and its estimation from skinfold thickness: measurements on 481 menandwomenagedfrom16to72years.br J Nutr 32: 77 97, Qureshi AR, Alvestrand A, Danielsson A, Divino-Filho JC, Gutierrez A, Lindholm B, Bergström J: Factors predicting malnutrition in hemodialysis patients: a cross-sectional study. Kidney Int 53: , Cooper BA, Penne EL, Bartlett LH, Pollock CA: Protein malnutrition and hypoalbuminemia as predictors of vascular events and mortality in ESRD. Am J Kidney Dis 43: 61 66, Kaysen GA, Johansen KL, Cheng SC, Jin C, Chertow GM: Trends and outcomes associated with serum albumin concentration among incident dialysis patients in the United States. J Ren Nutr 18: , Whitehead RG, Alleyne GA: Pathophysiological factors of importance in protein-calorie malnutrition. Br Med Bull 28: 72 79, Coles GA, Peters DK, Jones JH: Albumin metabolism in chronic renal failure. Clin Sci 39: , Ballmer PE, McNurlan MA, Hulter HN, Anderson SE, Garlick PJ, Krapf R: Chronic metabolic acidosis decreases albumin synthesis and induces negative nitrogen balance in humans. J Clin Invest 95: 39 45, Heimbürger O, Qureshi AR, Blaner WS, Berglund L, Stenvinkel P: Hand-grip muscle strength, lean body mass, and plasma proteins as markers of nutritional status in patients with chronic renal failure close to start of dialysis therapy. Am J Kidney Dis 36: , Bossola M, La Torre G, Giungi S, Tazza L, Vulpio C, Luciani G: Serum albumin, body weight and inflammatory parameters in chronic hemodialysis patients: A three-year longitudinal study. Am J Nephrol 28: , Warnock DG, Muntner P, McCullough PA, Zhang X, McClure LA, Zakai N, Cushman M, Newsome BB, Kewalramani R, Steffes MW, Howard G, McClellan WM; REGARDS Investigators: Kidney function, albuminuria, and all-cause mortality in the RE- GARDS (Reasons for Geographic and Racial Differences in Stroke) study. Am J Kidney Dis 56: , 2010 Received: October 6, 2011 Accepted: May 22, 2012 Published online ahead of print. Publication date available at www. cjasn.org.