- Email: [email protected]
The predictive value of composite methods of nutritional assessment on mortality among haemodialysis patients
e-SPEN Journal xxx (2014) e1ee5
Contents lists available at ScienceDirect
e-SPEN Journal journal homepage: http://www.elsevier.com/locate/clnu
Original article
The predictive value of composite methods of nutritional assessment on mortality among haemodialysis patients Letícia Maria Führ a, Elisabeth Wazlawik b, *, Monique Ferreira Garcia a a b
polis, SC, Brazil Post-Graduate Program in Nutrition, Federal University of Santa Catarina, Floriano ~o em Nutriça ~o, Centro de Ci^ rio, Trindade, CEP 88040-900, Floriano s-Graduaça polis, SC, Brazil Programa de Po encias da Saúde, Campus Universita
a r t i c l e i n f o
s u m m a r y
Article history: Received 25 March 2014 Accepted 3 November 2014
Background & aims: Several parameters might indicate protein-energy wasting in patients undergoing haemodialysis (HD), and such depletion has been associated with the survival of these patients. Our aim was to identify the parameters that are associated with an increased risk of death among HD patients. Methods: This was a prospective study with at least 13 months follow-up three times per week of 138 HD patients; 61.6% of the patients were men, 28.9% had diabetes mellitus, and 81.9% had hypertension. The associations of the survival rates based on by KaplaneMeier analysis with the following nutritional parameters were verified: albumin, lymphocytes, % fat mass (% FM), mid-arm muscle circumference (MAMC), subjective global assessment (SGA), malnutrition-inflammation score (MIS), and nutritional risk screening 2002 (NRS 2002). Cox proportional hazard analysis was used to identify the patients' risk of death (hazard proportional ratio e HR). Results: The nutritional parameters of lymphocytes and % FM were not associated with the risk of patient death. The patients who were classified as malnourished based on MAMC had a greater risk of death than did those considered nourished, but this difference was not statistically significant. The parameters of serum albumin, SGA, MIS, and NRS 2002 were associated with the risk of patient death (HR ¼ 2.77 P ¼ 0.042, HR ¼ 1.88 P ¼ 0.202, HR ¼ 4.47 P ¼ 0.011, HR ¼ 3.13 P ¼ 0.022, respectively), and the latter two parameters were significantly associated with a high risk among malnourished. Conclusions: The scores for the MIS and NRS 2002 composite methods of nutritional assessment were associated with the highest mortality risk values; thus, in conditions similar to those of our study, we suggest that the use of these parameters should be preferred. © 2014 Published by Elsevier Ltd on behalf of European Society for Clinical Nutrition and Metabolism.
Keywords: Albumin Haemodialysis Malnutrition-inflammation score Mortality Subjective global assessment Nutritional risk screening
Introduction Several parameters can be indicative of protein-energy wasting in patients undergoing haemodialysis (HD), which can be observed based on a group of criteria that includes biochemical factors, weight loss, reduced total body fat, decreases in muscle mass, low protein or energy intake and appetite. Moreover, questionnaires and nutritional scores are also potential assessment tools [1]. The use of different parameters, such as subjective, anthropometric and laboratory measures [1], is recommended for the renal patients evaluation, and examinations of the associations of simple
* Corresponding author. Tel.: þ55 48 3721 5138; fax: þ55 48 3721 9542. E-mail addresses: [email protected] (L.M. Führ), [email protected], [email protected] (E. Wazlawik), [email protected] (M.F. Garcia).
and composite methods with survival among HD patients is also suggested [2]. Although serum albumin is influenced by other factors, reductions in this nutritional parameter have been shown to be associated with the risk of death in HD patients [3e8]. Decreases in the production of lymphocytes and the consequent damage to the immune system might increase the risks of infections, morbidity and mortality [9], and malnutrition as measured by reductions in lymphocytes in HD patients is significantly associated with the risk of mortality [10]. In patients undergoing HD, nutritional status deficits can lead to changes in body composition [11]. Some studies have shown that reductions in body fat as assessed by fat mass percentages (%FM) are associated with poor survival [11e13]. The mid-arm muscle circumference (MAMC) is a measure of the depletion of lean body mass and is significantly lower among individuals who have undergone HD and died within one year of
http://dx.doi.org/10.1016/j.clnme.2014.11.002 2212-8263/© 2014 Published by Elsevier Ltd on behalf of European Society for Clinical Nutrition and Metabolism.
Please cite this article in press as: Führ LM, et al., The predictive value of composite methods of nutritional assessment on mortality among haemodialysis patients, e-SPEN Journal (2014), http://dx.doi.org/10.1016/j.clnme.2014.11.002
e2
L.M. Führ et al. / e-SPEN Journal xxx (2014) e1ee5
follow-up [10]. Additionally, MAMC has previously been considered an independent predictor of mortality [5,13]. The subjective global assessment (SGA) was developed for surgical patients [14], can identify malnutrition among HD patients [8,15] and is also associated with mortality among patients undergoing HD [16e18]. Various adaptations of the SGA have been established for kidney patients, the use of these adaptations in this population has increased [15]. The malnutrition-inflammation score (MIS) was developed for dialysis patients and is an adaptation of the SGA [17]. The MIS is more sensitive in the detection of malnutrition than is the original adaptation of the SGA [19], and MIS outcomes are associated with the survival of HD patients [2,16,20,21]. The nutritional risk screening 2002 (NRS 2002) is a tool that categorizes individuals as having or not having a nutritional risk and assesses nutritional status and disease severity for the nutritional screening purposes [22]. The only relevant study of HD patients we found reported an association between the presence of nutritional risk and the risk of death [3]. The aim of this study was to identify the nutritional parameters that are associated with an increased risk of death among patients undergoing HD. Subjects and methods Patients This was a prospective cohort study with HD patients that began in 2011. Patients who had been undergoing HD for at least three polis, Southern Brazil months at two clinics in the region of Floriano and were 18 years or older were enrolled in the study. The patients were dialysed three times per week for 3e5 h per day. Patients with body mass indices (BMIs) >34 kg/m2, amputated or atrophied limbs, pacemakers, cancer, stroke or acquired immunodeficiency syndrome and those who were unable to respond or were hospitalized were excluded. The patients were evaluated between April and August of 2011 and followed for at least 13 months. The monitoring ended in September of 2012. Demographic data, the duration of HD, the presence of other diseases (e.g., diabetes mellitus, hypertension, and heart failure) and laboratory tests were obtained from the patient records. The study was approved by the Ethics Committee on Human Research of the Federal University of Santa Catarina, and each participant signed a consent form. Nutritional assessment These following composite methods were used for the nutritional assessments: SGA, MIS, and NRS 2002. The SGA evaluated the clinical histories and physical examinations of the patients, which were subjectively rated as follows: A e well nourished, B e moderately nourished or suspected of being malnourished, and C e severely malnourished. The patients in categories B and C were grouped for the statistical analyses [14]. In addition to the components of the SGA [14], the MIS [16] considered the time on HD, the presence of comorbidities, BMI, serum albumin, and total iron binding capacity. The sum of all components results in a score from 0 (normal) to 30 (severe malnutrition). The patients were classified as follows: wellnourished (<6) and malnourished (6) [23]. The NRS 2002 assessed the patients' nutritional statuses and disease severities. Each component score was summed, and one point was added to the total for patients 70 years of age. The patients were classified based on total scores as follow: without nutritional risk (<3), or with nutritional risk (3) [22].
In addition to these composite methods, the following isolate data points were analysed: serum albumin, total lymphocyte, % fat mass (% FM), the sum of four skinfolds, and mid-arm muscle circumference (MAMC). The cut-off points for malnutrition were as follows: serum albumin 3.8 g/dL [1]; lymphocytes <2000 cells/ mm3 [24]; % FM <10% [1]; and 90% MAMC adequacy [24]. All anthropometric measurements were collected by the same researcher who was trained in the procedures of the standardized body measurements that were performed. The anthropometric evaluation was performed after the end of an HD session. The following equipment was used for these measurements: Cescorf® inelastic tape (Cescorf Equipamentos para Esporte Ltda. e Porto Alegre, Rio Grande do Sul, Brazil), and Lange callipers (Beta Technology Incorporated, Cambridge, Maryland, USA). Statistical analyses The data were analysed using with data analysis and statistical software (STATA, version 11 for Windows; Stata Corporation, College Station, TX, USA). The sample is described with the absolute and relative frequencies, means and standard deviations or medians and interquartile ranges of the variables. T tests, ManneWhitney U tests, or chi-square tests were used for the bivariate analyses of the clinical characteristics between the patients who survived and died. The gross analyses of the associations between the nutritional parameters and mortality were performed with KaplaneMeier analysis from which we obtained the survival graphs. Cox proportional hazard analysis was used to identify the patients' risks of death (proportional hazard ratios, HRs), and race, sex, marital status, time on HD, age, education, and the presence of diabetes mellitus and hypertension were considered as potential confounders. Only adjusted variables with P values <0.20 were included in the crude analysis. This analysis was used to obtain the values of the density ratio of the incidence of death, i.e., the risk of mortality according to each nutritional parameter. P < 0.05 was considered statistically significant. Results As shown in the flowchart in Fig. 1, the sample consisted of 138 patients who were undergoing HD. Twenty-five patients refused to participate; these patients did not differ from the rest of the sample in terms of mean age or sex distribution (P ¼ 0.62 and 0.20, respectively). The causes of chronic kidney disease were as follows: 36.2% hypertension, 15.9% diabetes mellitus, 13.8% glomerulonephritis, 8% polycystic kidney disease, and 26.1% other or unknown causes. The main clinical characteristics of the patients are shown in Table 1 (i.e., nutritional parameters, comorbidities, and adequacy of dialysis). The prevalence of malnutrition according to the nutritional parameters and the incidence of death among the malnourished according to each parameter are shown in Table 2. Of all of the subjects included in the study, the final outcomes of 19.6% were not investigated; six patients switched from haemodialysis to peritoneal dialysis, six were transferred to other treatment centres, and fifteen underwent renal transplantation. The cumulative incidence of death during the study period was 12.3% (95% CI 6.7 to 17.9%); these 17 deaths corresponded to an incidence density of 10.8 per 100 persons at risk per year. The causes of death were as follows: 23.5% (n ¼ 4) cardiovascular diseases, 17.6% (n ¼ 3) pulmonary diseases, 17.6% uncertain causes that were related to the disease, 17.6% external causes unrelated to the disease, 11.8% (n ¼ 2) hyperkalemia, and 11.8% infection. The KaplaneMeier survival analysis (Fig. 2) revealed that regardless of the composite method of nutritional assessment that
Please cite this article in press as: Führ LM, et al., The predictive value of composite methods of nutritional assessment on mortality among haemodialysis patients, e-SPEN Journal (2014), http://dx.doi.org/10.1016/j.clnme.2014.11.002
L.M. Führ et al. / e-SPEN Journal xxx (2014) e1ee5
e3
Table 2 Prevalence of malnutrition and the mortality rate among the malnourished patients according to each nutritional parameter. Nutritional parameter
Prevalence of malnutrition (%)
Incidence of death among malnourished (%)
Albumin (<3.8 g/dL) Lymphocytes (<2000 cells/mm3) MAMC (
32.9 70.8 51.5 3.7 35.5 44.1 26.8
22.2 11.3 14.1 20.0 22.5 27.7 32.1
MAMC, mid-arm muscle circumference; FM, fat mass; SGA, subjective global assessment; MIS, malnutrition inflammation score; NRS 2002, nutritional risk screening 2002.
The analyses of the isolated nutritional parameters of total lymphocytes and % FM revealed no associations with survival. There was an association between the risk of death and the MAMC (HR 1.14), but this relationship was not statistically significant. The only isolated nutritional parameter that was significantly associated with death was serum albumin. Fig. 1. Patient selection flowchart.
Discussion
was employed, the haemodialysis patients who were classified as malnourished died more frequently and earlier than did those who were classified as well nourished. However, this association was only significant for the MIS and NRS 2002 (P ¼ 0.012 and 0.007, respectively); this association was not significant based on the SGA classifications (P ¼ 0.108). In the adjusted analysis (Table 3), the confounding factors of time in months that the individual had undergone HD and patient age in years were considered. The patients who were malnourished according to the MIS classification had an HR 4.47, the HR according to the NRS 2002 was 3.13, and the HR according to the SGA was 1.88; the latter value was not statistically significant. Table 1 Clinical characteristics and nutritional parameters of the HD patients at baseline. Characteristics
Mean ± SD (n ¼ 138)
Survivors (n ¼ 121)
Dead (n ¼ 17)
Age (years) Nutritional parameters Weight (kg) BMI (kg/m2) FM (%) Arm circumference (cm) MAMC (cm) Albumin (g/dL) Lymphocytes (cells/mm3) Phosphorus (mg/dL) Potassium (mEq/L) Haematocrit (%) Haemoglobin (g/100 mL) Comorbiditiesc Diabetes mellitus (%) Hypertension (%) Heart disease (%) Renal function Dialysis dose (Kt/V) Time on HD (months)d
55.4 ± 15.2
53.5 ± 14.8
68,6 ± 11.4b
65.5 ± 12.5 24.9 ± 3.8 26.9 ± 9.0 27.8 ± 3.8 23.3 ± 2.8 3.96 ± 0.28 1719 ± 619 6 ± 1.7 5.5 ± 0.8 34 ± 4.1 11 ± 1.4
65.4 ± 12.5 24.7 ± 3.7 26.7 ± 8.9 27.7 ± 3.7 23.2 ± 2.7 3.98 ± 0.28 1708 ± 610 6 ± 1.7 5.5 ± 0.8 34 ± 4 11 ± 1.4
65.5 ± 12.5 26.0 ± 4.6 28.4 ± 9.7 28.7 ± 4.4 24.1 ± 3.2 3.82 ± 0.24a 1793 ± 696 6.5 ± 1.9 5.7 ± 0.9 35 ± 5 11.2 ± 1.6
41 (29.7) 113 (81.9) 41 (29.7)
35 (28.9) 100 (82.6) 32 (26.5)
6 (35.3) 13 (76.5) 9 (52.9)a
1.37 ± 0.22 36 (13;77)
1.37 ± 0.22 36 (14;80)
1.32 ± 0.19 23 (10;43)
BMI, body mass index; FM, fat mass; MAMC, mid-arm muscle circumference; HD, haemodialysis. a P < 0.05. b P < 0.001. c Absolute and relative frequencies for categorical variables. d Median and interquartile range.
The mortality of HD patients is high, and malnutrition is associated with death in these patients [2]. Due to the variety of parameters that can be used to assess malnutrition, our aim was to identify whether composite methods of nutritional assessment were associated with increased risks of death. Despite the prior training of the researcher in data collection and nutritional assessment, possible limitations of this study include the facts that some of the parameters depended on the reports and memory of individuals, and some of the parameters were subjective. Despite the relevance of this study, particularly for the evaluation of HD patients in our region, this nutritional assessment that was conducted in 2011 did not include the sevenpoint SGA or items assessing food intake and the loss of lean body mass, and these factors form a portion of the criteria proposed by Fouque et al. (2007). The cumulative mortality rate in our sample was below the average percentage of deaths observed in the last census of the Brazilian Society of Nephrology but higher than those reported in studies in other countries [4,16,25]. Although some studies have reported higher values [2,18], the percentage observed in this study is still high and should call attention to the high risk of death in this population. Other surveys have also reported different prevalences of malnutrition that varied according to the methods and the cut-off points used in the diagnosis [26e28]. Our results also showed these differences, and such variation reinforces the importance of studies that demonstrate the associations of parameters with clinical events such as death with the goal of identifying the best parameters and cut-off points for clinical practice. We also found an association between serum albumin with death; other studies with longer follow-ups have reported similar observations [3e8]. In contrast to the results of other studies [10,18,27], in our sample, we observed a higher prevalence of diabetes mellitus [10], longer times in HD [18], and a higher mean age [18,27]; these factors might have compromised the nutritional statuses our participants. Moreover, the cut-off points employed by each study vary, which justifies our choice to use cut-offs recommended by a group of experts on HD [1]. Despite the association with outcomes, the isolated application of serum albumin as a nutritional marker is not recommended because it can show
Please cite this article in press as: Führ LM, et al., The predictive value of composite methods of nutritional assessment on mortality among haemodialysis patients, e-SPEN Journal (2014), http://dx.doi.org/10.1016/j.clnme.2014.11.002
e4
L.M. Führ et al. / e-SPEN Journal xxx (2014) e1ee5 Table 3 Hazard ratio for death according to the composite methods of nutritional assessment, albumin, and MAMC. Parameter (category)
SGA (B þ C) MIS (6) NRS 2002 (with risk) Albumin (<3.8 g/dL) MAMC (
Hazard ratio of death Gross analysis HR (95%CI)a
Adjusted analysisb HR (95%CI)a
P value
2.2 4.2 3.7 2.9 1.3
1.88 4.47 3.13 2.77 1.14
0.202 0.011 0.023 0.042 0.798
(0.8;5.7) (1.4;12.9) (1.4;9.6) (1.1;7.6) (0.5;3.5)
(0.7;4.9) (1.4;14.2) (1.2;8.4) (1.0;7.4) (0.4;3.0)
SGA, subjective global assessment; MIS, malnutrition inflammation score; MAMC, mid-arm muscle circumference; HR, hazard ratio. a The presented hazard ratio values are relative to the individuals who were classified as well-nourished (reference groups) according to each parameter. b Adjusted for variables with P < 0.20 in gross analysis: patient age and duration of HD.
Fig. 2. KaplaneMeier survival graphs according to the nutritional status classifications based on the composite methods of nutritional assessment.
changes due to disease [8] and shows a weak correlation with several nutritional indicators [29]. We found a single study that linked total lymphocytes to the survival of haemodialysis patients [10], and this result contrasts with our finding of an association between low levels of lymphocytes and survival; this discrepancy might be the result of the large
number of patients with long duration HD and the higher percentage of death. The % FM obtained as the sum of skinfold measurements has been associated with the risk of death in patients on haemodialysis, but the cut-offs employed by different studies have varied. One study divided the sample into tertiles [13], and the other studies used a cut-off of <12% FM [11]. Notably, the cut-off point chosen for this study has been specifically recommended HD patients [1]. Only one study found that MAMC is associated with survival using a Cox regression [13]. Other authors have reported results that are similar to ours, which suggests that this parameter alone is not associated with the risk of death [5,16,18] within the monitoring timeframe used. Although developed for surgical patients [14], the SGA has been applied to various clinical conditions [15,28] and is considered to be capable of identifying malnutrition in HD patients [8,15]. Similar to the observations of Pisetkul et al. [25], we found no association between SGA score and death; however, a high percentage of the malnourished patients in our study died. In other studies with longer [2,18] or equal [16,17] monitoring times, the SGA score has been found to be associated with a significant risk of death. These differences might be attributable to other characteristics of the present study, such as the lower average age, the shorter time on HD and the higher lean mass according to MAMC. The MIS was developed specifically for HD patients and is a composite method that covers aspects that are important for these patients [17]. This tool is useful for the nutritional and inflammatory evaluations of patients on HD and performs better than the SGA in the prediction of short-term complications [25]. Indeed, in the present study, the MIS was the parameter that exhibited the strongest association with death; the risk of death was nearly fivefold higher for the patients with scores greater than or equal to six compared with those with lower scores. All studies that have examined the relationship of this parameter with survival have found the MIS to be an independent predictor of death. This parameter has been analysed using the same cut-off point by Pisetkul et al. [25], using a higher cut-off point [2,20], using a sample divided into quartiles [21], and using a division based on the score for each of the 10 units [17]. The NRS 2002 was developed for hospitalized patients [22], and there is little information about its use with HD patients in the literature. This parameter has been associated with the clinical complications of hospitalized patients [30] and has been used as a standard to validate the hand grip strength of HD patients [28].
Please cite this article in press as: Führ LM, et al., The predictive value of composite methods of nutritional assessment on mortality among haemodialysis patients, e-SPEN Journal (2014), http://dx.doi.org/10.1016/j.clnme.2014.11.002
L.M. Führ et al. / e-SPEN Journal xxx (2014) e1ee5
Based on our results, the NRS 2002 can be considered to be an important tool because it was associated with a high risk of death in our sample. The study by Fiedler et al. (2009), which is the only study we found that performed this analysis, found an even higher risk, which might be attributable to the shorter time in HD, lower average age, and better dialysis adequacy of our sample. Our study revealed each of the three composite methods was a better nutritional parameter for the prediction of patient death than the isolated parameters. The MIS and NRS 2002 were significantly associated with a high risk of death in our sample. Although a greater percentage of those classified as malnourished according to the NRS 2002 died, the highest risk of death was among those who were classified as malnourished according to the MIS as observed in the Cox regression. In conclusion, we suggest that preference be given to the use of composite methods for the nutritional assessment of HD patients because these composite methods include different aspects of nutritional status and detect greater numbers of malnourished patients. Furthermore, we suggested that further studies should be performed with different outcomes measures and longer durations. Acknowledgements We would like to thank the Post-Graduate Program in Nutrition for supporting this study, the clinics and the patients for their cooperation and Coordination of Improvement of Higher Education Personnel (Coordenaç~ ao de Aperfeiçoamento de Pessoal de Nível Superior e CAPES) for granting the scholarship. References [1] Fouque D, Kalantar-Zadeh K, Kopple J, Cano N, Chauveau P, Cuppari L, et al. A proposed nomenclature and diagnostic criteria for proteineenergy wasting in acute and chronic kidney disease. Kidney Int 2007;73(4):391e8. [2] Fiedler R, Jehle PM, Osten B, Dorligschaw O, Girndt M. Clinical nutrition scores are superior for the prognosis of hemodialysis patients compared to lab markers and bioelectrical impedance. Nephrol Dial Transplant 2009;24: 3812e7. [3] Chan M, Kelly J, Batterham M, Tapsell L. Malnutrition (subjective global assessment) scores and serum albumin levels, but not body mass index values, at initiation of dialysis are independent predictors of mortality: a 10-year clinical cohort study. J Ren Nutr 2012;22(6):547e57. s F, Lozano J, et al. Risk factors [4] Cuevas X, García F, Martín-Malo A, Fort J, Llado associated with cardiovascular morbidity and mortality in spanish incident hemodialysis patients: two-year results from the answer study. Blood Purif 2012;33:22e9. [5] De Araújo IC, Kamimura MA, Draibei SA, Canziani MFE, Manfredi SR, Avesani CM, et al. Nutritional parameters and mortality in incident hemodialysis patients. J Ren Nutr 2006;16(1):27e35. [6] Mafra D, Farage NE, Azevedo DL, Viana GG, Mattos JP, Velarde LGC, et al. Impact of serum albumin and body-mass index on survival in hemodialysis patients. Int Urol Nephrol 2007;39:619e24. [7] Mazairac AHA, Wit GA, Grooteman MPC, Penne EL, van der Weerd NC, van den Dorpel MA, et al. A composite score of protein-energy nutritional status predicts mortality in haemodialysis patients no better than its individual components. Nephrol Dial Transplant 2011;26:1962e7. [8] Mutsert R, Grootendorst DC, Indemans F, Boeschoten EW, Krediet RT, Dekker FW, et al. Association between serum albumin and mortality in dialysis patients is partly explained by inflammation, and not by malnutrition. J Ren Nutr 2009;19(2):127e35.
e5
[9] Kuwae N, Kopple JD, Kalantar-Zadeh K. A low lymphocyte percentage is a predictor of mortality and hospitalization in hemodialysis patients. Clin Nephrol 2005;63(1):22e34. n R, Teruel JL, de la Cal MA, G [10] Marce amez C. The impact of malnutrition in morbidity and mortality in stable haemodialysis patients. Nephrol Dial Transplant 1997;12:2324e31. [11] Kalantar-Zadeh K, Kuwae N, Wu DY, Shantouf RS, Fouque D, Anker SD, et al. Associations of body fat and its changes over time with quality of life and prospective mortality in hemodialysis patients. The Am J Clin Nutr 2006;83: 202e10. [12] Mafra D, Guebre-Egziabher D, Fouque D. Body mass index, muscle and fat in chronic kidney disease: questions about survival. Nephrol Dial Transplant 2008;23:2461e6. [13] Stosovic M, Stanojevic M, Simic-Ogrizovic S, Jovanovic D, Djukanovic L. The predictive value of anthropometric parameters on mortality in haemodialysis patients. Nephrol Dial Transplant 2011;26:1367e74. [14] Detsky AS, McLaughlin JR, Baker JP, Johnston N, Wittaker S, Mendelson RA, et al. What is subjective global assessment of nutritional status? JPEN 1987;11:8e13. [15] Steiber AL, Kalantar-Zadeh K, Secker D, McCarthy M, Sehgal A, McCann L. Subjective global assessment in chronic kidney disease: a review. J Ren Nutr 2004;14(4):191e200. [16] Segall L, Mardare NG, Ungureanu S, Busuioc M, Nistor I, Enache R, et al. Nutritional status evaluation and survival in haemodialysis patients in one centre from Romania. Nephrol Dial Transplant 2009;24:2536e40. [17] Kalantar-Zadeh K, Kopple JD, Block G, Humphreys MH. A malnutritioninflammation score is correlated with morbidity and mortality in maintenance hemodialysis patients. Am J Kidney Dis 2001;38(6):1251e63. [18] Qureshi AR, Alvestrand A, Divino-Filho JC, Gutierrez A, Heimbürger O, Lindholm B, et al. Inflammation, malnutrition, and cardiac disease as predictors of mortality in hemodialysis patients. J Am Soc Nephrol 2002;13(1): S28e36. [19] Hou Y, Li X, Hong D, Zou H, Yang L, Chen Y, et al. Comparison of different assessments for evaluating malnutrition in Chinese patients with end-stage renal disease with maintenance hemodialysis. Nutr Res 2012;32(4):266e71. n e [20] Carreras RB, Mengarelli MC, Najun-Zarazaga CJ. El score de desnutricio n como predictor de mortalidad en pacientes en hemodi inflamacio alisis. Dial Traspl 2008;29(2):55e61. [21] Rambod M, Bross R, Zitterkoph J, Benner D, Pithia J, Colman S, et al. Association of malnutrition-inflammation score with quality of life and mortality in hemodialysis patients: a 5-year prospective cohort study. Am J Kidney Dis 2009;53(2):298e309. [22] Kondrup J, Rasmussen HH, Hamberg O, Stanga Z. Ad Hoc ESPEN working group. Nutritional risk screening (NRS 2002): a new method based on an analysis of controlled clinical trials. Clin Nutr 2003;22(3):321e36. [23] Yamada K, Furuya R, Takita T, Maruyama Y, Yamaguchi Y, Ohkawa S, et al. Simplified nutritional screening tools for patients on maintenance hemodialysis. Am J Clin Nutr 2008;87:106e13. [24] Blackburn GL, Thornton PA. Nutritional assessment of the hospitalized patient. Med Clin North Am 1979;63:1103e15. [25] Pisetkul C, Chanchairujira K, Chotipanvittayakul N, Ong-Ajyooth L, Chanchairujira T. Malnutrition-inflammation score associated with atherosclerosis, inflammation and short-term outcome in hemodialysis patients. J Med Assoc Thai 2010;93:S147e156. [26] Garcia MF, Meireles MS, Fuhr LM, Donini AB, Wazlawik E. Relationship between hand grip strength and nutritional assessment methods used of hospitalized patients. Rev Nutr 2013;26:49e57. [27] Amemiya N, Ogawa T, Otsuka K, Ando Y, Nitta K. Comparison of serum albumin, serum c-reactive protein, and pulse wave velocity as predictors of the 4-year mortality of chronic hemodialysis patients. J Atheroscler Thromb 2011;18(12):1071e9. [28] Garcia MF, Wazlawik E, Moreno YMF, Führ LM, Gonzalez-Chica DA. Diagnostic accuracy of handgrip strength in the assessment of malnutrition in hemodialyzed patients. e-SPEN (Oxford) 2013;8:e181e6. r [29] Gama-Axelsson T, Heimbürger O, Stenvinkel P, Ba any P, Lindholm B, Qureshi AR. Serum albumin as predictor of nutritional status in patients with ESRD. Clin J Am Soc Nephrol 2012;7(9):1446e53. [30] Raslan M, Gonzalez MC, Dias MCG, Nascimento M, Castro M, Marques P, et al. Comparison of nutritional risk screening tools for predicting clinical outcomes in hospitalized patients. Nutr 2010;26:721e6.
Please cite this article in press as: Führ LM, et al., The predictive value of composite methods of nutritional assessment on mortality among haemodialysis patients, e-SPEN Journal (2014), http://dx.doi.org/10.1016/j.clnme.2014.11.002
Contents lists available at ScienceDirect
e-SPEN Journal journal homepage: http://www.elsevier.com/locate/clnu
Original article
The predictive value of composite methods of nutritional assessment on mortality among haemodialysis patients Letícia Maria Führ a, Elisabeth Wazlawik b, *, Monique Ferreira Garcia a a b
polis, SC, Brazil Post-Graduate Program in Nutrition, Federal University of Santa Catarina, Floriano ~o em Nutriça ~o, Centro de Ci^ rio, Trindade, CEP 88040-900, Floriano s-Graduaça polis, SC, Brazil Programa de Po encias da Saúde, Campus Universita
a r t i c l e i n f o
s u m m a r y
Article history: Received 25 March 2014 Accepted 3 November 2014
Background & aims: Several parameters might indicate protein-energy wasting in patients undergoing haemodialysis (HD), and such depletion has been associated with the survival of these patients. Our aim was to identify the parameters that are associated with an increased risk of death among HD patients. Methods: This was a prospective study with at least 13 months follow-up three times per week of 138 HD patients; 61.6% of the patients were men, 28.9% had diabetes mellitus, and 81.9% had hypertension. The associations of the survival rates based on by KaplaneMeier analysis with the following nutritional parameters were verified: albumin, lymphocytes, % fat mass (% FM), mid-arm muscle circumference (MAMC), subjective global assessment (SGA), malnutrition-inflammation score (MIS), and nutritional risk screening 2002 (NRS 2002). Cox proportional hazard analysis was used to identify the patients' risk of death (hazard proportional ratio e HR). Results: The nutritional parameters of lymphocytes and % FM were not associated with the risk of patient death. The patients who were classified as malnourished based on MAMC had a greater risk of death than did those considered nourished, but this difference was not statistically significant. The parameters of serum albumin, SGA, MIS, and NRS 2002 were associated with the risk of patient death (HR ¼ 2.77 P ¼ 0.042, HR ¼ 1.88 P ¼ 0.202, HR ¼ 4.47 P ¼ 0.011, HR ¼ 3.13 P ¼ 0.022, respectively), and the latter two parameters were significantly associated with a high risk among malnourished. Conclusions: The scores for the MIS and NRS 2002 composite methods of nutritional assessment were associated with the highest mortality risk values; thus, in conditions similar to those of our study, we suggest that the use of these parameters should be preferred. © 2014 Published by Elsevier Ltd on behalf of European Society for Clinical Nutrition and Metabolism.
Keywords: Albumin Haemodialysis Malnutrition-inflammation score Mortality Subjective global assessment Nutritional risk screening
Introduction Several parameters can be indicative of protein-energy wasting in patients undergoing haemodialysis (HD), which can be observed based on a group of criteria that includes biochemical factors, weight loss, reduced total body fat, decreases in muscle mass, low protein or energy intake and appetite. Moreover, questionnaires and nutritional scores are also potential assessment tools [1]. The use of different parameters, such as subjective, anthropometric and laboratory measures [1], is recommended for the renal patients evaluation, and examinations of the associations of simple
* Corresponding author. Tel.: þ55 48 3721 5138; fax: þ55 48 3721 9542. E-mail addresses: [email protected] (L.M. Führ), [email protected], [email protected] (E. Wazlawik), [email protected] (M.F. Garcia).
and composite methods with survival among HD patients is also suggested [2]. Although serum albumin is influenced by other factors, reductions in this nutritional parameter have been shown to be associated with the risk of death in HD patients [3e8]. Decreases in the production of lymphocytes and the consequent damage to the immune system might increase the risks of infections, morbidity and mortality [9], and malnutrition as measured by reductions in lymphocytes in HD patients is significantly associated with the risk of mortality [10]. In patients undergoing HD, nutritional status deficits can lead to changes in body composition [11]. Some studies have shown that reductions in body fat as assessed by fat mass percentages (%FM) are associated with poor survival [11e13]. The mid-arm muscle circumference (MAMC) is a measure of the depletion of lean body mass and is significantly lower among individuals who have undergone HD and died within one year of
http://dx.doi.org/10.1016/j.clnme.2014.11.002 2212-8263/© 2014 Published by Elsevier Ltd on behalf of European Society for Clinical Nutrition and Metabolism.
Please cite this article in press as: Führ LM, et al., The predictive value of composite methods of nutritional assessment on mortality among haemodialysis patients, e-SPEN Journal (2014), http://dx.doi.org/10.1016/j.clnme.2014.11.002
e2
L.M. Führ et al. / e-SPEN Journal xxx (2014) e1ee5
follow-up [10]. Additionally, MAMC has previously been considered an independent predictor of mortality [5,13]. The subjective global assessment (SGA) was developed for surgical patients [14], can identify malnutrition among HD patients [8,15] and is also associated with mortality among patients undergoing HD [16e18]. Various adaptations of the SGA have been established for kidney patients, the use of these adaptations in this population has increased [15]. The malnutrition-inflammation score (MIS) was developed for dialysis patients and is an adaptation of the SGA [17]. The MIS is more sensitive in the detection of malnutrition than is the original adaptation of the SGA [19], and MIS outcomes are associated with the survival of HD patients [2,16,20,21]. The nutritional risk screening 2002 (NRS 2002) is a tool that categorizes individuals as having or not having a nutritional risk and assesses nutritional status and disease severity for the nutritional screening purposes [22]. The only relevant study of HD patients we found reported an association between the presence of nutritional risk and the risk of death [3]. The aim of this study was to identify the nutritional parameters that are associated with an increased risk of death among patients undergoing HD. Subjects and methods Patients This was a prospective cohort study with HD patients that began in 2011. Patients who had been undergoing HD for at least three polis, Southern Brazil months at two clinics in the region of Floriano and were 18 years or older were enrolled in the study. The patients were dialysed three times per week for 3e5 h per day. Patients with body mass indices (BMIs) >34 kg/m2, amputated or atrophied limbs, pacemakers, cancer, stroke or acquired immunodeficiency syndrome and those who were unable to respond or were hospitalized were excluded. The patients were evaluated between April and August of 2011 and followed for at least 13 months. The monitoring ended in September of 2012. Demographic data, the duration of HD, the presence of other diseases (e.g., diabetes mellitus, hypertension, and heart failure) and laboratory tests were obtained from the patient records. The study was approved by the Ethics Committee on Human Research of the Federal University of Santa Catarina, and each participant signed a consent form. Nutritional assessment These following composite methods were used for the nutritional assessments: SGA, MIS, and NRS 2002. The SGA evaluated the clinical histories and physical examinations of the patients, which were subjectively rated as follows: A e well nourished, B e moderately nourished or suspected of being malnourished, and C e severely malnourished. The patients in categories B and C were grouped for the statistical analyses [14]. In addition to the components of the SGA [14], the MIS [16] considered the time on HD, the presence of comorbidities, BMI, serum albumin, and total iron binding capacity. The sum of all components results in a score from 0 (normal) to 30 (severe malnutrition). The patients were classified as follows: wellnourished (<6) and malnourished (6) [23]. The NRS 2002 assessed the patients' nutritional statuses and disease severities. Each component score was summed, and one point was added to the total for patients 70 years of age. The patients were classified based on total scores as follow: without nutritional risk (<3), or with nutritional risk (3) [22].
In addition to these composite methods, the following isolate data points were analysed: serum albumin, total lymphocyte, % fat mass (% FM), the sum of four skinfolds, and mid-arm muscle circumference (MAMC). The cut-off points for malnutrition were as follows: serum albumin 3.8 g/dL [1]; lymphocytes <2000 cells/ mm3 [24]; % FM <10% [1]; and 90% MAMC adequacy [24]. All anthropometric measurements were collected by the same researcher who was trained in the procedures of the standardized body measurements that were performed. The anthropometric evaluation was performed after the end of an HD session. The following equipment was used for these measurements: Cescorf® inelastic tape (Cescorf Equipamentos para Esporte Ltda. e Porto Alegre, Rio Grande do Sul, Brazil), and Lange callipers (Beta Technology Incorporated, Cambridge, Maryland, USA). Statistical analyses The data were analysed using with data analysis and statistical software (STATA, version 11 for Windows; Stata Corporation, College Station, TX, USA). The sample is described with the absolute and relative frequencies, means and standard deviations or medians and interquartile ranges of the variables. T tests, ManneWhitney U tests, or chi-square tests were used for the bivariate analyses of the clinical characteristics between the patients who survived and died. The gross analyses of the associations between the nutritional parameters and mortality were performed with KaplaneMeier analysis from which we obtained the survival graphs. Cox proportional hazard analysis was used to identify the patients' risks of death (proportional hazard ratios, HRs), and race, sex, marital status, time on HD, age, education, and the presence of diabetes mellitus and hypertension were considered as potential confounders. Only adjusted variables with P values <0.20 were included in the crude analysis. This analysis was used to obtain the values of the density ratio of the incidence of death, i.e., the risk of mortality according to each nutritional parameter. P < 0.05 was considered statistically significant. Results As shown in the flowchart in Fig. 1, the sample consisted of 138 patients who were undergoing HD. Twenty-five patients refused to participate; these patients did not differ from the rest of the sample in terms of mean age or sex distribution (P ¼ 0.62 and 0.20, respectively). The causes of chronic kidney disease were as follows: 36.2% hypertension, 15.9% diabetes mellitus, 13.8% glomerulonephritis, 8% polycystic kidney disease, and 26.1% other or unknown causes. The main clinical characteristics of the patients are shown in Table 1 (i.e., nutritional parameters, comorbidities, and adequacy of dialysis). The prevalence of malnutrition according to the nutritional parameters and the incidence of death among the malnourished according to each parameter are shown in Table 2. Of all of the subjects included in the study, the final outcomes of 19.6% were not investigated; six patients switched from haemodialysis to peritoneal dialysis, six were transferred to other treatment centres, and fifteen underwent renal transplantation. The cumulative incidence of death during the study period was 12.3% (95% CI 6.7 to 17.9%); these 17 deaths corresponded to an incidence density of 10.8 per 100 persons at risk per year. The causes of death were as follows: 23.5% (n ¼ 4) cardiovascular diseases, 17.6% (n ¼ 3) pulmonary diseases, 17.6% uncertain causes that were related to the disease, 17.6% external causes unrelated to the disease, 11.8% (n ¼ 2) hyperkalemia, and 11.8% infection. The KaplaneMeier survival analysis (Fig. 2) revealed that regardless of the composite method of nutritional assessment that
Please cite this article in press as: Führ LM, et al., The predictive value of composite methods of nutritional assessment on mortality among haemodialysis patients, e-SPEN Journal (2014), http://dx.doi.org/10.1016/j.clnme.2014.11.002
L.M. Führ et al. / e-SPEN Journal xxx (2014) e1ee5
e3
Table 2 Prevalence of malnutrition and the mortality rate among the malnourished patients according to each nutritional parameter. Nutritional parameter
Prevalence of malnutrition (%)
Incidence of death among malnourished (%)
Albumin (<3.8 g/dL) Lymphocytes (<2000 cells/mm3) MAMC (
32.9 70.8 51.5 3.7 35.5 44.1 26.8
22.2 11.3 14.1 20.0 22.5 27.7 32.1
MAMC, mid-arm muscle circumference; FM, fat mass; SGA, subjective global assessment; MIS, malnutrition inflammation score; NRS 2002, nutritional risk screening 2002.
The analyses of the isolated nutritional parameters of total lymphocytes and % FM revealed no associations with survival. There was an association between the risk of death and the MAMC (HR 1.14), but this relationship was not statistically significant. The only isolated nutritional parameter that was significantly associated with death was serum albumin. Fig. 1. Patient selection flowchart.
Discussion
was employed, the haemodialysis patients who were classified as malnourished died more frequently and earlier than did those who were classified as well nourished. However, this association was only significant for the MIS and NRS 2002 (P ¼ 0.012 and 0.007, respectively); this association was not significant based on the SGA classifications (P ¼ 0.108). In the adjusted analysis (Table 3), the confounding factors of time in months that the individual had undergone HD and patient age in years were considered. The patients who were malnourished according to the MIS classification had an HR 4.47, the HR according to the NRS 2002 was 3.13, and the HR according to the SGA was 1.88; the latter value was not statistically significant. Table 1 Clinical characteristics and nutritional parameters of the HD patients at baseline. Characteristics
Mean ± SD (n ¼ 138)
Survivors (n ¼ 121)
Dead (n ¼ 17)
Age (years) Nutritional parameters Weight (kg) BMI (kg/m2) FM (%) Arm circumference (cm) MAMC (cm) Albumin (g/dL) Lymphocytes (cells/mm3) Phosphorus (mg/dL) Potassium (mEq/L) Haematocrit (%) Haemoglobin (g/100 mL) Comorbiditiesc Diabetes mellitus (%) Hypertension (%) Heart disease (%) Renal function Dialysis dose (Kt/V) Time on HD (months)d
55.4 ± 15.2
53.5 ± 14.8
68,6 ± 11.4b
65.5 ± 12.5 24.9 ± 3.8 26.9 ± 9.0 27.8 ± 3.8 23.3 ± 2.8 3.96 ± 0.28 1719 ± 619 6 ± 1.7 5.5 ± 0.8 34 ± 4.1 11 ± 1.4
65.4 ± 12.5 24.7 ± 3.7 26.7 ± 8.9 27.7 ± 3.7 23.2 ± 2.7 3.98 ± 0.28 1708 ± 610 6 ± 1.7 5.5 ± 0.8 34 ± 4 11 ± 1.4
65.5 ± 12.5 26.0 ± 4.6 28.4 ± 9.7 28.7 ± 4.4 24.1 ± 3.2 3.82 ± 0.24a 1793 ± 696 6.5 ± 1.9 5.7 ± 0.9 35 ± 5 11.2 ± 1.6
41 (29.7) 113 (81.9) 41 (29.7)
35 (28.9) 100 (82.6) 32 (26.5)
6 (35.3) 13 (76.5) 9 (52.9)a
1.37 ± 0.22 36 (13;77)
1.37 ± 0.22 36 (14;80)
1.32 ± 0.19 23 (10;43)
BMI, body mass index; FM, fat mass; MAMC, mid-arm muscle circumference; HD, haemodialysis. a P < 0.05. b P < 0.001. c Absolute and relative frequencies for categorical variables. d Median and interquartile range.
The mortality of HD patients is high, and malnutrition is associated with death in these patients [2]. Due to the variety of parameters that can be used to assess malnutrition, our aim was to identify whether composite methods of nutritional assessment were associated with increased risks of death. Despite the prior training of the researcher in data collection and nutritional assessment, possible limitations of this study include the facts that some of the parameters depended on the reports and memory of individuals, and some of the parameters were subjective. Despite the relevance of this study, particularly for the evaluation of HD patients in our region, this nutritional assessment that was conducted in 2011 did not include the sevenpoint SGA or items assessing food intake and the loss of lean body mass, and these factors form a portion of the criteria proposed by Fouque et al. (2007). The cumulative mortality rate in our sample was below the average percentage of deaths observed in the last census of the Brazilian Society of Nephrology but higher than those reported in studies in other countries [4,16,25]. Although some studies have reported higher values [2,18], the percentage observed in this study is still high and should call attention to the high risk of death in this population. Other surveys have also reported different prevalences of malnutrition that varied according to the methods and the cut-off points used in the diagnosis [26e28]. Our results also showed these differences, and such variation reinforces the importance of studies that demonstrate the associations of parameters with clinical events such as death with the goal of identifying the best parameters and cut-off points for clinical practice. We also found an association between serum albumin with death; other studies with longer follow-ups have reported similar observations [3e8]. In contrast to the results of other studies [10,18,27], in our sample, we observed a higher prevalence of diabetes mellitus [10], longer times in HD [18], and a higher mean age [18,27]; these factors might have compromised the nutritional statuses our participants. Moreover, the cut-off points employed by each study vary, which justifies our choice to use cut-offs recommended by a group of experts on HD [1]. Despite the association with outcomes, the isolated application of serum albumin as a nutritional marker is not recommended because it can show
Please cite this article in press as: Führ LM, et al., The predictive value of composite methods of nutritional assessment on mortality among haemodialysis patients, e-SPEN Journal (2014), http://dx.doi.org/10.1016/j.clnme.2014.11.002
e4
L.M. Führ et al. / e-SPEN Journal xxx (2014) e1ee5 Table 3 Hazard ratio for death according to the composite methods of nutritional assessment, albumin, and MAMC. Parameter (category)
SGA (B þ C) MIS (6) NRS 2002 (with risk) Albumin (<3.8 g/dL) MAMC (
Hazard ratio of death Gross analysis HR (95%CI)a
Adjusted analysisb HR (95%CI)a
P value
2.2 4.2 3.7 2.9 1.3
1.88 4.47 3.13 2.77 1.14
0.202 0.011 0.023 0.042 0.798
(0.8;5.7) (1.4;12.9) (1.4;9.6) (1.1;7.6) (0.5;3.5)
(0.7;4.9) (1.4;14.2) (1.2;8.4) (1.0;7.4) (0.4;3.0)
SGA, subjective global assessment; MIS, malnutrition inflammation score; MAMC, mid-arm muscle circumference; HR, hazard ratio. a The presented hazard ratio values are relative to the individuals who were classified as well-nourished (reference groups) according to each parameter. b Adjusted for variables with P < 0.20 in gross analysis: patient age and duration of HD.
Fig. 2. KaplaneMeier survival graphs according to the nutritional status classifications based on the composite methods of nutritional assessment.
changes due to disease [8] and shows a weak correlation with several nutritional indicators [29]. We found a single study that linked total lymphocytes to the survival of haemodialysis patients [10], and this result contrasts with our finding of an association between low levels of lymphocytes and survival; this discrepancy might be the result of the large
number of patients with long duration HD and the higher percentage of death. The % FM obtained as the sum of skinfold measurements has been associated with the risk of death in patients on haemodialysis, but the cut-offs employed by different studies have varied. One study divided the sample into tertiles [13], and the other studies used a cut-off of <12% FM [11]. Notably, the cut-off point chosen for this study has been specifically recommended HD patients [1]. Only one study found that MAMC is associated with survival using a Cox regression [13]. Other authors have reported results that are similar to ours, which suggests that this parameter alone is not associated with the risk of death [5,16,18] within the monitoring timeframe used. Although developed for surgical patients [14], the SGA has been applied to various clinical conditions [15,28] and is considered to be capable of identifying malnutrition in HD patients [8,15]. Similar to the observations of Pisetkul et al. [25], we found no association between SGA score and death; however, a high percentage of the malnourished patients in our study died. In other studies with longer [2,18] or equal [16,17] monitoring times, the SGA score has been found to be associated with a significant risk of death. These differences might be attributable to other characteristics of the present study, such as the lower average age, the shorter time on HD and the higher lean mass according to MAMC. The MIS was developed specifically for HD patients and is a composite method that covers aspects that are important for these patients [17]. This tool is useful for the nutritional and inflammatory evaluations of patients on HD and performs better than the SGA in the prediction of short-term complications [25]. Indeed, in the present study, the MIS was the parameter that exhibited the strongest association with death; the risk of death was nearly fivefold higher for the patients with scores greater than or equal to six compared with those with lower scores. All studies that have examined the relationship of this parameter with survival have found the MIS to be an independent predictor of death. This parameter has been analysed using the same cut-off point by Pisetkul et al. [25], using a higher cut-off point [2,20], using a sample divided into quartiles [21], and using a division based on the score for each of the 10 units [17]. The NRS 2002 was developed for hospitalized patients [22], and there is little information about its use with HD patients in the literature. This parameter has been associated with the clinical complications of hospitalized patients [30] and has been used as a standard to validate the hand grip strength of HD patients [28].
Please cite this article in press as: Führ LM, et al., The predictive value of composite methods of nutritional assessment on mortality among haemodialysis patients, e-SPEN Journal (2014), http://dx.doi.org/10.1016/j.clnme.2014.11.002
L.M. Führ et al. / e-SPEN Journal xxx (2014) e1ee5
Based on our results, the NRS 2002 can be considered to be an important tool because it was associated with a high risk of death in our sample. The study by Fiedler et al. (2009), which is the only study we found that performed this analysis, found an even higher risk, which might be attributable to the shorter time in HD, lower average age, and better dialysis adequacy of our sample. Our study revealed each of the three composite methods was a better nutritional parameter for the prediction of patient death than the isolated parameters. The MIS and NRS 2002 were significantly associated with a high risk of death in our sample. Although a greater percentage of those classified as malnourished according to the NRS 2002 died, the highest risk of death was among those who were classified as malnourished according to the MIS as observed in the Cox regression. In conclusion, we suggest that preference be given to the use of composite methods for the nutritional assessment of HD patients because these composite methods include different aspects of nutritional status and detect greater numbers of malnourished patients. Furthermore, we suggested that further studies should be performed with different outcomes measures and longer durations. Acknowledgements We would like to thank the Post-Graduate Program in Nutrition for supporting this study, the clinics and the patients for their cooperation and Coordination of Improvement of Higher Education Personnel (Coordenaç~ ao de Aperfeiçoamento de Pessoal de Nível Superior e CAPES) for granting the scholarship. References [1] Fouque D, Kalantar-Zadeh K, Kopple J, Cano N, Chauveau P, Cuppari L, et al. A proposed nomenclature and diagnostic criteria for proteineenergy wasting in acute and chronic kidney disease. Kidney Int 2007;73(4):391e8. [2] Fiedler R, Jehle PM, Osten B, Dorligschaw O, Girndt M. Clinical nutrition scores are superior for the prognosis of hemodialysis patients compared to lab markers and bioelectrical impedance. Nephrol Dial Transplant 2009;24: 3812e7. [3] Chan M, Kelly J, Batterham M, Tapsell L. Malnutrition (subjective global assessment) scores and serum albumin levels, but not body mass index values, at initiation of dialysis are independent predictors of mortality: a 10-year clinical cohort study. J Ren Nutr 2012;22(6):547e57. s F, Lozano J, et al. Risk factors [4] Cuevas X, García F, Martín-Malo A, Fort J, Llado associated with cardiovascular morbidity and mortality in spanish incident hemodialysis patients: two-year results from the answer study. Blood Purif 2012;33:22e9. [5] De Araújo IC, Kamimura MA, Draibei SA, Canziani MFE, Manfredi SR, Avesani CM, et al. Nutritional parameters and mortality in incident hemodialysis patients. J Ren Nutr 2006;16(1):27e35. [6] Mafra D, Farage NE, Azevedo DL, Viana GG, Mattos JP, Velarde LGC, et al. Impact of serum albumin and body-mass index on survival in hemodialysis patients. Int Urol Nephrol 2007;39:619e24. [7] Mazairac AHA, Wit GA, Grooteman MPC, Penne EL, van der Weerd NC, van den Dorpel MA, et al. A composite score of protein-energy nutritional status predicts mortality in haemodialysis patients no better than its individual components. Nephrol Dial Transplant 2011;26:1962e7. [8] Mutsert R, Grootendorst DC, Indemans F, Boeschoten EW, Krediet RT, Dekker FW, et al. Association between serum albumin and mortality in dialysis patients is partly explained by inflammation, and not by malnutrition. J Ren Nutr 2009;19(2):127e35.
e5
[9] Kuwae N, Kopple JD, Kalantar-Zadeh K. A low lymphocyte percentage is a predictor of mortality and hospitalization in hemodialysis patients. Clin Nephrol 2005;63(1):22e34. n R, Teruel JL, de la Cal MA, G [10] Marce amez C. The impact of malnutrition in morbidity and mortality in stable haemodialysis patients. Nephrol Dial Transplant 1997;12:2324e31. [11] Kalantar-Zadeh K, Kuwae N, Wu DY, Shantouf RS, Fouque D, Anker SD, et al. Associations of body fat and its changes over time with quality of life and prospective mortality in hemodialysis patients. The Am J Clin Nutr 2006;83: 202e10. [12] Mafra D, Guebre-Egziabher D, Fouque D. Body mass index, muscle and fat in chronic kidney disease: questions about survival. Nephrol Dial Transplant 2008;23:2461e6. [13] Stosovic M, Stanojevic M, Simic-Ogrizovic S, Jovanovic D, Djukanovic L. The predictive value of anthropometric parameters on mortality in haemodialysis patients. Nephrol Dial Transplant 2011;26:1367e74. [14] Detsky AS, McLaughlin JR, Baker JP, Johnston N, Wittaker S, Mendelson RA, et al. What is subjective global assessment of nutritional status? JPEN 1987;11:8e13. [15] Steiber AL, Kalantar-Zadeh K, Secker D, McCarthy M, Sehgal A, McCann L. Subjective global assessment in chronic kidney disease: a review. J Ren Nutr 2004;14(4):191e200. [16] Segall L, Mardare NG, Ungureanu S, Busuioc M, Nistor I, Enache R, et al. Nutritional status evaluation and survival in haemodialysis patients in one centre from Romania. Nephrol Dial Transplant 2009;24:2536e40. [17] Kalantar-Zadeh K, Kopple JD, Block G, Humphreys MH. A malnutritioninflammation score is correlated with morbidity and mortality in maintenance hemodialysis patients. Am J Kidney Dis 2001;38(6):1251e63. [18] Qureshi AR, Alvestrand A, Divino-Filho JC, Gutierrez A, Heimbürger O, Lindholm B, et al. Inflammation, malnutrition, and cardiac disease as predictors of mortality in hemodialysis patients. J Am Soc Nephrol 2002;13(1): S28e36. [19] Hou Y, Li X, Hong D, Zou H, Yang L, Chen Y, et al. Comparison of different assessments for evaluating malnutrition in Chinese patients with end-stage renal disease with maintenance hemodialysis. Nutr Res 2012;32(4):266e71. n e [20] Carreras RB, Mengarelli MC, Najun-Zarazaga CJ. El score de desnutricio n como predictor de mortalidad en pacientes en hemodi inflamacio alisis. Dial Traspl 2008;29(2):55e61. [21] Rambod M, Bross R, Zitterkoph J, Benner D, Pithia J, Colman S, et al. Association of malnutrition-inflammation score with quality of life and mortality in hemodialysis patients: a 5-year prospective cohort study. Am J Kidney Dis 2009;53(2):298e309. [22] Kondrup J, Rasmussen HH, Hamberg O, Stanga Z. Ad Hoc ESPEN working group. Nutritional risk screening (NRS 2002): a new method based on an analysis of controlled clinical trials. Clin Nutr 2003;22(3):321e36. [23] Yamada K, Furuya R, Takita T, Maruyama Y, Yamaguchi Y, Ohkawa S, et al. Simplified nutritional screening tools for patients on maintenance hemodialysis. Am J Clin Nutr 2008;87:106e13. [24] Blackburn GL, Thornton PA. Nutritional assessment of the hospitalized patient. Med Clin North Am 1979;63:1103e15. [25] Pisetkul C, Chanchairujira K, Chotipanvittayakul N, Ong-Ajyooth L, Chanchairujira T. Malnutrition-inflammation score associated with atherosclerosis, inflammation and short-term outcome in hemodialysis patients. J Med Assoc Thai 2010;93:S147e156. [26] Garcia MF, Meireles MS, Fuhr LM, Donini AB, Wazlawik E. Relationship between hand grip strength and nutritional assessment methods used of hospitalized patients. Rev Nutr 2013;26:49e57. [27] Amemiya N, Ogawa T, Otsuka K, Ando Y, Nitta K. Comparison of serum albumin, serum c-reactive protein, and pulse wave velocity as predictors of the 4-year mortality of chronic hemodialysis patients. J Atheroscler Thromb 2011;18(12):1071e9. [28] Garcia MF, Wazlawik E, Moreno YMF, Führ LM, Gonzalez-Chica DA. Diagnostic accuracy of handgrip strength in the assessment of malnutrition in hemodialyzed patients. e-SPEN (Oxford) 2013;8:e181e6. r [29] Gama-Axelsson T, Heimbürger O, Stenvinkel P, Ba any P, Lindholm B, Qureshi AR. Serum albumin as predictor of nutritional status in patients with ESRD. Clin J Am Soc Nephrol 2012;7(9):1446e53. [30] Raslan M, Gonzalez MC, Dias MCG, Nascimento M, Castro M, Marques P, et al. Comparison of nutritional risk screening tools for predicting clinical outcomes in hospitalized patients. Nutr 2010;26:721e6.
Please cite this article in press as: Führ LM, et al., The predictive value of composite methods of nutritional assessment on mortality among haemodialysis patients, e-SPEN Journal (2014), http://dx.doi.org/10.1016/j.clnme.2014.11.002