Malnutrition Inflammation Score cut-off predicting mortality in maintenance hemodialysis patients

Clinical Nutrition ESPEN xxx (2016) 1e5

Contents lists available at ScienceDirect

Clinical Nutrition ESPEN journal homepage: http://www.clinicalnutritionespen.com

Original article

Malnutrition Inflammation Score cut-off predicting mortality in maintenance hemodialysis patients Mariana Clementoni Costa Borges, Barbara Perez Vogt, Luis Cuadrado Martin, Jacqueline Costa Teixeira Caramori* ~o Paulo, Brazil Faculdade de Medicina de Botucatu, UNESP Univ Estadual Paulista, Department of Clinical Medicine, Botucatu, Sa

a r t i c l e i n f o

s u m m a r y

Article history: Received 14 July 2016 Accepted 29 October 2016

Background: Malnutrition is a strong predictor of mortality on hemodialysis patients, especially when it is associated with inflammation. Malnutrition Inflammation Score (MIS) is a simple and low cost tool which assesses the presence of malnutrition associated with inflammation. Therefore, the aim is to evaluate if MIS is associated with mortality in patients on maintenance hemodialysis and establish a cutoff to predict mortality at different follow-up periods. Methods: Observational retrospective cohort study including 215 patients on hemodialysis between July 2012 and June 2014, censored until November 2015. MIS was used to assess patient's nutritional status at the moment they were enrolled in the study. They were followed for at least 18 months. Results: At the end of 18 months, 38 (17.7%) deaths, 20 renal transplants (9.3%), four facilities transference (1.9%), three dialysis method change (1.4%) and one renal function recovery (0.5%) were observed. One hundred seventy one patients completed at least 24 months of follow-up, and during this additional period, there were five deaths and one renal transplant more. Score higher than 7 points was able to predict mortality for both follow-up periods using sensitivity and specificity analysis and ROC curves. Using this cut-off on KaplaneMeier survival curve, it was possible to confirm the association of MIS with all-cause mortality at 18 months and 24 or more months of follow-up. Finally, Cox multivariate analysis adjusted for demographic, clinical and nutritional variables showed MIS as the only significant predictor of mortality. Conclusion: MIS is an independent predictor of mortality in hemodialysis patients. © 2016 European Society for Clinical Nutrition and Metabolism. Published by Elsevier Ltd. All rights reserved.

Keywords: Hemodialysis Malnutrition Inflammation Score Mortality Nutritional assessment

1. Introduction Malnutrition is a strong predictor of mortality in the population with chronic kidney disease (CKD) on hemodialysis (HD). It is characterized by reduction of body protein and energy stores [1], and it is usually associated with decreased functional capacity. It is caused not only by decreased intake of nutrients, but also by the inflammatory state and other conditions associated with CKD and dialysis/renal replacement therapy [1]. As malnutrition and inflammation are predictors of mortality in HD patients [1], represented by protein energy wasting, it is

* Corresponding author. Faculdade de Medicina de Botucatu, UNESP, Univ rio Rubens Guimara ~es Montenegro, s/n, 18618687, Estadual Paulista, Av. Prof. Ma ~o Paulo, Brazil. Fax: þ55 14 3882 2238. Botucatu, Sa E-mail address: [email protected] (J.C.T. Caramori).

important to diagnose this condition early, using the best available tool to predict outcomes and, moreover, to allow the resort of specific nutritional strategies to avoid a more severe deterioration of nutritional status [2]. A clinically useful nutritional tool should be able to identify the problems, to predict risk of morbidity and mortality, to identify patients who should receive intervention and to evaluate responses to the therapy [2]. However, there is no gold standard single method able to diagnose the nutritional status of CKD patients. Thus, the use of composite methods is recommended [3]. Composite methods based on a combination of both objective and subjective elements were developed to classify nutritional status. In the context of CKD, Malnutrition Inflammation Score (MIS) [4], which was originally developed by Kalantar-Zadeh et al., is a quantitative score assessing the presence and degree of nutritional deficit. MIS incorporates ten components, 70% of them are

http://dx.doi.org/10.1016/j.clnesp.2016.10.006 2405-4577/© 2016 European Society for Clinical Nutrition and Metabolism. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Borges MCC, et al., Malnutrition Inflammation Score cut-off predicting mortality in maintenance hemodialysis patients, Clinical Nutrition ESPEN (2016), http://dx.doi.org/10.1016/j.clnesp.2016.10.006

2

M.C.C. Borges et al. / Clinical Nutrition ESPEN xxx (2016) 1e5

subjectively evaluated, and 30% are objectively evaluated. Studies have shown a strong association between MIS and morbidity and mortality in CKD non-dialysis [5], HD [4,6,8,9], peritoneal dialysis and [5] renal transplanted patients [7]. However, the cut-off that predicts this relationship has not been established yet. Therefore, the aim of this study is to assess if MIS is associated with all-cause mortality and to identify the best cut-off to predict mortality in a cohort of maintenance HD patients at different follow-up times. 2. Materials and methods 2.1. Study design and subjects It was a retrospective cohort study which included 215 patients on maintenance hemodialysis at the Clinics Hospital of UNESP ~o Paulo, Brazil, in the Universidade Estadual Paulista, Botucatu, Sa period between July 2012 and August 2014. Inclusion criteria were patients older than 18 years and prevalent in HD for at least one month. Patients who did not have MIS on nutritional assessment were excluded. The study protocol was approved by the local research ethics committee. Patient's assessments were obtained through analysis of medical and nutritional records. The following demographic and clinical data were collected: sex, age, dialysis vintage, main cause of endstage-renal-disease (ESRD) and presence of diabetes. Serum creatinine, albumin, C-reactive protein, bicarbonate, and hemoglobin were collected from routine examinations of patients, which are performed monthly. Total iron binding capacity (TIBC) was calculated from serum values of iron and transferrin [11]. Dialysis adequacy was measured by Kt/V, and it was calculated using second generation Daugirdas formula [12]. All measurements were performed at the specialized chemistry laboratory of the Clinics Hospital of UNESP. 2.2. Nutritional assessment Anthropometric assessment was performed after HD session. Body weight, height, mid-arm circumference, and triceps skinfold thickness were measured according to standard techniques [13]. From these measurements body mass index (BMI), mid-arm muscle circumference (MAMC) and their adjustments to the 50th percentile for age and sex were calculated [14,15]. MIS was applied during anthropometric assessment. It is a tool divided in four sections: I) nutritional history (change in dry weight, dietary intake, gastrointestinal symptoms, functional capacity, comorbidities and dialysis vintage), II) physical examination (decrease of fat stores or loss of subcutaneous fat and signals of muscle wasting), III) BMI, and IV) laboratory parameters (serum albumin and TIBC). Each component has four severity levels, which are scored from 0 (normal) to 3 (very severe), directly proportional with the severity of the disease [4]. 2.3. Follow-up and censored Patients were followed until November 2015 and censored in death, kidney transplantation, transference to another facility, change of dialysis modality, or renal function recovery. 2.4. Statistical analysis Data were expressed as mean ± standard deviation or median and first and third quartiles. Frequencies were expressed as percentage. Comparisons between baseline characteristics of survival and non-survival groups were performed using t Student's test or Mann Whitney. Frequencies were compared by chi-square test. All

patients included in the study completed at least 18 months of follow-up. Therefore, a first analysis was performed including all patients considering the outcomes until 18 months of follow-up. Subsequently, patients who completed at least 24 months of follow-up were selected to repeat the analysis. ROC curves were fitted to verify the area under the curve (AUC) and the significance of MIS predicting mortality. Analysis of sensitivity and specificity has been done to verify the best MIS cut-off able to predict mortality. KaplaneMeier survival curves were fitted to compare the groups above and below the cut-off previously established by the sensitivity and specificity analysis, and the difference between the curves was assessed by log-rank test. Proportional hazards Cox analysis was used to assess which are the independent predictors of mortality in both follow-up periods, adjusted for variables significantly different between the survival and non-survival groups. Variables that are already included on MIS were not included in the models. The statistical significance criterion was p < 0.05. Statistical analysis was performed using SPSS 22.0.

3. Results Two hundred and fifteen patients were enrolled in the study, most of them males (56%) and age ranged from 19 to 91 years. Diabetic nephropathy was the most prevalent cause of end-stage renal disease, followed by hypertensive nephrosclerosis. MIS median was 5, minimum 0 and maximum 26. Demographical, clinical and nutritional data of entire cohort is presented on Table 1. During 18 months of follow-up, there were 38 deaths (17.7%), 20 transplants (9.3%), four transfers to another facility (1.9%), three changes of dialysis modality (1.4%) and one renal function recovery (0.5%). The baseline characteristics of survival and non-survival after 18 months of follow-up were compared. The non-survival group showed higher prevalence of DM (p ¼ 0.04), higher serum CRP (p ¼ 0.02), and lower serum creatinine (p ¼ 0.03). As expected, the follow-up of non-survival was shorter (p < 0.001). Regarding nutritional parameters, neither of them were significantly different between the groups, except MIS (p < 0.01). Table 1 Baseline demographical, clinical, and nutritional data of 215 patients on maintenance hemodialysis. Characteristic

Total population (n ¼ 215)

Age (years) Gender [Male (%)] Diabetes [n (%)] Dialysis Vintage (months) Cause of end-stage renal disease [n (%)] Diabetic Nephropathy Hypertensive nephrosclerosis Unknown Chronic glomerulonephritis ADPKD Acute kidney injury Others Malnutrition inflammation score BMI (kg/m2) Percent standard of MAMC (%) Percent standard of TST (%) Serum Creatinine (mg/dl) Serum Albumin (g/dl) C reactive protein (mg/dl) Serum Bicarbonate (mEq/L) Hemoglobin (g/dl) Kt/V

58.4 ± 14.6 121 (56.3) 95 (44.4) 16.6 (6.8; 40.8) 60 (27.9) 42 (19.5) 43 (20) 20 (9.3) 8 (3.7) 5 (2.3) 37 (17.2) 5 (4; 8) 26.0 ± 6.0 99.4 ± 17 107 ± 60.9 8.7 ± 2.9 3.8 ± 0.6 1 (0.5; 1.9) 22.1 ± 2.7 11.4 ± 1.8 1.44 ± 0.3

Abbreviations: ADPKD: autosomal dominant polycystic kidney disease; BMI: body mass index; MAMC: mid-arm muscle circumference; TST: tricipital skinfold thickness.

Please cite this article in press as: Borges MCC, et al., Malnutrition Inflammation Score cut-off predicting mortality in maintenance hemodialysis patients, Clinical Nutrition ESPEN (2016), http://dx.doi.org/10.1016/j.clnesp.2016.10.006

M.C.C. Borges et al. / Clinical Nutrition ESPEN xxx (2016) 1e5

Another analysis was performed including 171 patients who completed follow-up of at least 24 months (median 29.8 months, maximum 36.1 months). During this period, there were five deaths and one kidney transplant, counting 43 deaths, 21 transplants, four facilities transfers, three dialysis modality changes and one renal function recovery. Ninety-nine patients (57.9%) kept on HD after 24 months or more of follow-up. Same differences in the comparison between survival and non-survival with follow-up of 18 months were observed in the follow-up of 24 months or more (Table 2). In addition to the differences already observed, serum albumin was significantly lower in the non-survival group. Among nutritional parameters, MIS was the only one that showed a significant difference in the comparison between survival and non-survival in both follow-up periods. 3.1. Receiver operating characteristic curve analysis e ROC curve In the ROC curve, MIS higher than 7 was predictor of mortality in both analysis. The follow-up of 18 months showed 52.6% sensitivity and 77.4% specificity (AUC 0.696, 95% CI 0.604e0.789, p < 0.001) (Fig. 1A), and for 24 months or more of follow-up showed 53.5% sensitivity and 82% specificity (AUC 0.707; 95% CI 0.615e0.799, p < 0.001) (Fig. 1B). 3.2. KaplaneMeier survival analysis In the analysis of survival by KaplaneMeier method, it was possible to confirm the association of MIS with all-cause mortality in either follow-up of 18 months (Fig. 2A) and for 24 months or more (Fig. 2B). 3.3. Proportional hazards analysis cox In unadjusted analyzes, MIS was an independent predictor of mortality after 18 months of follow-up (HR 1.12; 95% CI 1.06 to 1.18; p < 0.01) and also after 24 months or more (HR 1.12; 95% CI 1.06 to 1.17; p < 0.01). Variables that were significant different between survival and non-survival in univariate analysis were considered for the adjustment of the models, except those that already are part of

3

the MIS (serum albumin and dialysis vintage). MIS remained an independent predictor of mortality even after adjustments (Table 3).

4. Discussion This study showed that MIS, proposed as a tool for evaluation of nutritional status in HD patients, is an independent predictor of mortality in a cohort of 215 patients both at 18 months of follow-up (n ¼ 215) and in 24 months (n ¼ 171). The cut-off of 7 showed high specificity to predict mortality in both periods evaluated. To the best of our knowledge, this is the first study evaluating a cut-off for MIS able to predict all-cause mortality in HD patients using analysis of sensitivity and specificity. Multivariate analysis showed mortality risk increases 10.6% with each point of MIS. Ho et al. [10] observed that MIS of 3, 4 and 5 increased 1-year mortality risk of Asian HD patients in 10%, 40% and 80%, respectively. Rambod et al. [9] showed that the increase of every 2 units of MIS is associated with a 2-fold increased risk of death in 5 years in HD patients in USA. Although these studies were performed in ethnically diverse populations, it can be observed the same trend of increased risk of mortality with the increase of MIS. Kalantar-Zadeh et al. [16] compared MIS, anthropometric and biochemical markers of malnutrition and inflammation predicting outcomes and concluded that MIS was the best tool to assess mortality of maintenance HD patients. In this study, MIS average was 6.3, and the median was 5.5, while in the present study the median was 5. Rambod et al. [9] also showed the superiority of MIS to assess the risk of mortality compared to inflammatory markers, such as C-reactive protein and interleukin 6, which are not always available in clinical practice due to their high cost [9]. Inflammation is a strong predictor of poor quality of life, morbidity, hospitalizations and mortality [4,16e18]. The ability of MIS to reflect inflammatory state increases the relevance of its application in clinical practice. Association between increased MIS and inflammation, sleep quality and depression were shown in HD patients [7,9,19]. These results are of great impact, because among CKD patients, quality of life per se is an important predictor of hospitalization and mortality [9]. In this study, it was possible to

Table 2 Comparison between baseline demographical, clinical, and nutritional data of survival and non-survival patients after follow-up of 24 months or more. Characteristic

Survivals (n ¼ 128)

Non survivals (n ¼ 43)

P

Age (years) Gender [Male (%)] Diabetes [n (%)] Follow up time (months) Dialysis Vintage (months) Cause of end-stage renal disease [n (%)] Diabetic Nephropathy Hypertensive nephrosclerosis Unknown Chronic glomerulonephritis ADPKD AKI Others BMI (kg/m2) Percent standard of MAMC (%) Percent standard of TST (%) MIS Serum Creatinine (mg/dl) Serum Albumin (g/dl) C reactive protein (mg/dl) Serum Bicarbonate (mEq/L) Hemoglobin (g/dl) Kt/V

56.7 ± 15.5 69 (53.9) 49 (38.3) 34.5 (25.7; 36.8) 22.1 (7.4; 43.5)

61.9 ± 13.5 22 (51.2) 25 (59.5) 8.8 (5.6; 13.5) 20.7 (3.5; 41.1)

0.06 0.75 0.02 <0.01 0.69

30 (23.4) 24 (18.8) 25 (19.5) 14 (11) 6 (4.7) 3 (2.3) 26 (20.3) 26.2 ± 5.8 100.4 ± 16.7 110.9 ± 65.1 5 (3; 7) 8.8 ± 2.9 3.9 ± 0.5 0.9 (0.5; 1.7) 21.8 ± 2.4 11.3 ± 1.8 1.43 ± 0.25

16 (37.2) 9 (20.9) 7 (16.3) 2 (4.7) 1 (2.3) 1 (2.3) 7 (16.3) 25.8 ± 7.4 97.2 ± 19 99.6 ± 47.7 8 (5; 12) 7.8 ± 2.5 3.7 ± 0.5 1.5 (0.8; 2.1) 22.7 ± 3.5 11 ± 1.7 1.47 ± 0.41

0.6

0.16 0.21 0.26 <0.01 0.03 0.02 0.01 0.14 0.27 0.6

Abbreviations: ADPKD: autosomal dominant polycystic kidney disease; BMI: body mass index; MAMC: mid-arm muscle circumference; TST: tricipital skinfold thickness.

Please cite this article in press as: Borges MCC, et al., Malnutrition Inflammation Score cut-off predicting mortality in maintenance hemodialysis patients, Clinical Nutrition ESPEN (2016), http://dx.doi.org/10.1016/j.clnesp.2016.10.006

4

M.C.C. Borges et al. / Clinical Nutrition ESPEN xxx (2016) 1e5

Fig. 1. Receiver operating characteristic (ROC) curve of Malnutrition-Inflammation Score predicting mortality in hemodialysis patients: (A) 18 months of follow-up; (B) 24 months or more of follow-up.

Fig. 2. Survival plots by follow-up time (A) Cumulative survival in 18 months; (B) cumulative survival in 24 months or more.

realization. Therefore, these methods constitute a valid alternative with an extensive applicability for the diagnosis of protein energy malnutrition (PEW) [20]. In this study, the MIS cut-off of 7 was able to predict all-cause mortality. This suggests that a score below 10 points already indicates greater risk of death. Ho et al. [10] also found that low cutoffs are strong predictors of mortality. This reinforces the importance of MIS elements, not requiring high values to indicate poor prognosis. Limitations of this study include: the fact that MIS was applied for more than one professional, which is known to be a factor that may decrease the sensitivity of the tool. As MIS is a subjective tool for nutritional assessment, it is dependent on the judgment of each evaluator. Moreover, it was an observational study, in which there are confounders. Therefore, other factors may be also related to

observe significantly higher PCR among patients who died, showing how close it is linked to malnutrition in this population. Among the available tools to assess the nutritional status of hemodialysis patients, MIS has gained attention in clinical practice. It has been used in over 100,000 patients in USA due to the ease of application, requiring only a well-trained professional [9]. It can be applied quickly and does not require patient's memory, or their physical and mental conditions. Also, it does not depend on complex and expensive laboratory measurements and performing anthropometric measurements [4,16]. It is well established that the use of a single evaluation method is not capable to determine reliably the nutritional status of patients with CKD. So, the use of composite methods has gained attention for its advantages, such as to generate a global assessment of nutritional status using a small number of equipment for its

Table 3 Hazard ratio for mortality according to multivariate Cox proportional hazards model. Variables

MIS Age Gender Diabetes Serum Creatinine

Model 1 (18 months)

Model 2 (24 months)

HR (95% CI)

P

HR (95% CI)

P

1.107 e 0.844 1.563 0.903

<0.001 e 0.62 0.19 0.13

1.107 1.005 0.850 1.665 0.922

<0.001 0.69 0.61 0.12 0.23

(1.042e1.175) (0.428e1.665) (0.800e3.056) (0.792e1.030)

(1.051e1.167) (0.981e1.029) (0.453e1.595) (0.881e3.147) (0.808e1.053)

Please cite this article in press as: Borges MCC, et al., Malnutrition Inflammation Score cut-off predicting mortality in maintenance hemodialysis patients, Clinical Nutrition ESPEN (2016), http://dx.doi.org/10.1016/j.clnesp.2016.10.006

M.C.C. Borges et al. / Clinical Nutrition ESPEN xxx (2016) 1e5

mortality. In addition, the sample was from a single center, and longitudinal measures of MIS have not been evaluated during follow-up. Despite these limitations, it was shown an association between MIS and mortality in this cohort, which was evaluated at two different moments, despite the presence of other factors classically associated with poor outcomes. Moreover, it was the first study to establish a MIS cut-off predicting mortality using sensitivity and specificity analysis. With these two analyzes, the consistency of MIS was confirmed and it didn't lose predictive power during follow-up. In conclusion, MIS is an independent predictor of mortality in maintenance HD patients followed-up for 18 months and 24 months or more. The MIS cut-off able to predict mortality was 7, in both periods evaluated. Therefore, there is a great utility of application of MIS in clinical practice. These results are important and may provide rationale to further studies with longitudinal assessments that can evaluate the importance of protein energy wasting in long-term, providing early medical and nutritional interventions and thus, preventing worse prognosis in HD patients. Statement of authorship M. C. C. Borges and B. P. Vogt contributed to the conception and design of the research; M. C. C. Borges, B. P. Vogt, L. C. Martin and J. C. T. Caramori contributed to the acquisition, analysis, or interpretation of the data; M. C. C. Borges and B. P. Vogt drafted the manuscript; M. C. C. Borges, B. P. Vogt, L. C. Martin and J. C. T. Caramori critically revised the manuscript; and M. C. C. Borges, B. P. Vogt, L. C. Martin and J. C. T. Caramori agree to be fully accountable for ensuring the integrity and accuracy of the work. All authors read and approved the final manuscript. Conflicts of interest The authors declare they have no conflicts of interest. The authors declare that the results presented in this paper have not been published previously in whole or part, except in abstract format. Acknowledgements A master's degree scholarship was provided to MCCB, and a doctorate scholarship was provided to BPV by Coordination of ~o de Improvement of Higher Education Personnel (Coordenaça Aperfeiçoamento de Pessoal de Nível Superior), an organization of the Brazilian federal government under the Ministry of Education. Our acknowledgements to all members of the health care team from the Dialysis Unit of Clinics Hospital of Botucatu Medical School who dedicated efforts to the treatment program for patients on hemodialysis.

5

References [1] Fouque D, Kalantar-Zadeh K, Kopple J, Cano N, Chauveau P, Cuppari L, et al. A proposed nomenclature and diagnostic criteria for protein-energy wasting in acute and chronic kidney disease. Kidney Int 2008;73(4):391e8. [2] Jeejeebhoy KN. Nutritional assessment. Gastroenterol Clin North Am 1998;27(2):347e69. [3] Clinical practice guidelines for nutrition in chronic renal failure. K/DOQI, National kidney foundation. Am J Kidney Dis 2000;35(6 Suppl. 2). S1e140. [4] 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. [5] Ho L-C, Wang H-H, Chiang C-K, Hung K-Y, Wu K-D. Malnutrition-inflammation score independently determined cardiovascular and infection risk in peritoneal dialysis patients. Blood Purif 2010;29(3):308e16. [6] Amparo FC, Kamimura MA, Molnar MZ, Cuppari L, Lindholm B, Amodeo C, et al. Diagnostic validation and prognostic significance of the MalnutritionInflammation Score in nondialyzed chronic kidney disease patients. Nephrol Dial Transpl 2015;30(5):821e8. [7] Sohrabi Z, Eftekhari MH, Eskandari MH, Rezaeianzadeh A, Sagheb MM. Malnutrition-inflammation score and quality of life in hemodialysis patients: is there any correlation? Nephro-Urol Mon 2015;7(3):e27445. [8] Molnar MZ, Czira ME, Rudas A, Ujszaszi A, Lindner A, Fornadi K, et al. Association of the malnutrition-inflammation score with clinical outcomes in kidney transplant recipients. Am J Kidney Dis 2011;58(1):101e8. [9] 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. [10] Ho L, Wang H-H, Peng Y-S, Chiang C-K, Huang J-W, Hung K-Y, et al. Clinical utility of malnutrition-inflammation score in maintenance hemodialysis patients: focus on identifying the best cut-off point. Am J Nephrol 2008;28(5): 840e6. [11] Kalantar-Zadeh K, Kleiner M, Dunne E, Ahern K, Nelson M, Koslowe R, et al. Total iron-binding capacity-estimated transferrin correlates with the nutritional subjective global assessment in hemodialysis patients. Am J Kidney Dis 1998;31(2):263e72. [12] Daugirdas JT. Second generation logarithmic estimates of single-pool variable volume Kt/V: an analysis of error. J Am Soc Nephrol 1993;4(5):1205e13. ^neas. In: Avaliaca ~o da composic~ [13] Heyward VH, Stolarczyk LM. Dobras cuta ao corporal aplicada [Applied body composition assessment]. Barueri: Manole; 2000. p. 23e46. [14] Frisancho AR. New norms of upper limb fat and muscle areas for assessment of nutritional status. Am J Clin Nutr 1981;34(11):2540e5. [15] Frisancho AR. Anthropometric standards for the assessment of growth and nutritional status. University of Michigan Press; 1990. p. 208. [16] Kalantar-Zadeh K, Ikizler TA, Block G, Avram MM, Kopple JD. Malnutritioninflammation complex syndrome in dialysis patients: causes and consequences. Am J Kidney Dis 2003;42(5):864e81. [17] Bologa RM, Levine DM, Parker TS, Cheigh JS, Serur D, Stenzel KH, et al. Interleukin-6 predicts hypoalbuminemia, hypocholesterolemia, and mortality in hemodialysis patients. Am J Kidney Dis 1998;32(1):107e14. [18] Yeun JY, Levine RA, Mantadilok V, Kaysen GA. C-Reactive protein predicts allcause and cardiovascular mortality in hemodialysis patients. Am J Kidney Dis 2000;35(3):469e76. [19] Bilgic A, Akgul A, Sezer S, Arat Z, Ozdemir FN, Haberal M. Nutritional status and depression, sleep disorder, and quality of life in hemodialysis patients. J Ren Nutr 2007;17(6):381e8. [20] Bigogno FG, Fetter RL, Avesani CM. Applicability of subjective global assessment and malnutrition inflammation score in the assessment of nutritional status on chronic kidney disease. J Bras Nefrol 2014;36(2):236e40.

Please cite this article in press as: Borges MCC, et al., Malnutrition Inflammation Score cut-off predicting mortality in maintenance hemodialysis patients, Clinical Nutrition ESPEN (2016), http://dx.doi.org/10.1016/j.clnesp.2016.10.006