ANTHROPOMETRIC PARAMETERS OF NUTRITIONAL ASSESSMENT AS PREDICTIVE FACTORS OF THE MINI NUTRITIONAL ASSESSMENT (MNA) OF HOSPITALIZED ELDERLY PATIENTS

Transcription

1 08 MERHI/c/pppp_04 LORD_c 21/10/11 10:08 Page181 JNHA: NUTRITION AS PREDICTIVE FACTORS OF THE MINI NUTRITIONAL ASSESSMENT (MNA) OF HOSPITALIZED ELDERLY PATIENTS V.A. LEANDRO-MERHI 1, J.L. BRAGA DE AQUINO 2 Institution: School of Nutrition and Medicine, Puc-Campinas-SP, Brazil. 1. Prof Dr of the School of Nutrition, Puc-Campinas-SP; 2. Prof. Dr of the School of Medicine, Puc-Campinas- SP. Address correspondence to: Vânia Ap. Leandro-Merhi, Av. Carlos Grimaldi, 1171, Quadra D-13, Bairro: Jardim Madalena, Residencial Vila Verde, Cep.: Campinas-SP, Brazil. valm@dglnet.com.br Abstract: Objective: The objective of this study was to identify nutritional indicators that predict MNA (mini nutritional assessment) classification in hospitalized elderly patients. Method: This cross-sectional study assessed the nutritional status of 109 elderly patients at the beginning of their hospital stay with anthropometric and laboratory indicators and the MNA. Habitual energy intake (HEI) was also determined. The assessed nutritional indicators were investigated by univariate and multivariate logistic regression analysis to verify if they can predict MNA classification. The odds ratio (OR) and its respective confidence interval (CI) of 95% were also calculated, and the significance level was set at 5% (p<0.05). Results: The nutritional status of most patients (61.47%) was appropriate but 30.28% were at risk of malnourishment and 8.26% were malnourished. Statistical differences were found for those aged more than 70 years and for arm circumference, body mass index, calf circumference, triceps skinfold thickness and mid-arm muscle circumference. Initially, the predictive factors identified by univariate logistic regression were body mass index (BMI) (p=0.0001; OR=0.825), calf circumference (CC) (p=0.0026; OR=0.832), arm circumference (AC) (p<0.0001; OR=0.787), triceps skinfold thickness (TST) (p=0.0014; OR=0.920) and mid-arm muscle circumference (MAMC) (p=0.0003; OR=0.975); later, multiple logistic regression analyses revealed that first AC (p=0.0025; OR=0.731 ( )), then BMI (p=<0.0001; OR= ( )) and finally TST (p=0.0040; OR=0.924 ( )) and MAMC (p=0.0010; OR=0.976 ( )) were factors that predict MNA classification. Conclusion: In the conditions of this study, first AC, then BMI and finally TST and MAMC together were capable of predicting MNA classification. Key words: Elderly, mini nutritional assessment, nutritional indicators, predictive factors. Introduction The MNA (Mini Nutritional Assessment) (1) has been a largely used method to identify risk of malnutrition in the elderly and in those who can benefit from early intervention. MNA is a simple, inexpensive and non-invasive method that can be used in bedridden individuals (1). The sum of the MNA scores can differentiate elderly individuals who have an appropriate nutritional status from those who are at risk of malnutrition or are malnourished. It has been demonstrated that the sensitivity of this scale is 96%, the specificity is 98% and the prognostic value for malnutrition is 97% (1). This method has been extensively used in elderly patients (2-7) and a greater prevalence of malnutrition has been found among the elderly population in greatest need of care (8). There are at least 40 screening and assessment tools for subjective nutritional status assessments, and some are for the general population and others for specific populations (9). The Nutritional Risk Screening (NRS-2002), proposed more recently, has proven to be an important instrument to assess nutritional risk and predict length of hospital stay of elderly patients (10). Other parameters considered routine nutritional indicators for assessing the nutritional status of hospitalized patients, such as anthropometry, laboratory tests and food intake assessment, continue to be widely used in clinical Received January 11, 2010 Accepted for publication February 18, practice and in nutritional investigations and diagnosis in hospitals (11, 12). Thus, the objective of this study was to identify nutritional indicators that can predict MNA classification in hospitalized elderly patients. Casuistic and method This study was done from February to September 2009 with 109 elderly patients of both genders (68 males and 41 females), staying at the surgery service of Hospital e Maternidade Celso Pierro of the Pontifical Catholic University of Campinas, state of São Paulo, Brazil. The study began after the administration of the hospital and the Research Ethics Committee of the institution approved it (Protocol number 925/08). The inclusion criteria were: age equal to or greater than 60 years, having undergone a nutritional assessment in the first 48 hours of their stay at the hospital, and information regarding nutritional status, disease and length of hospital stay (LHS) documented in the medical records of the institution. The age group of the population was determined by the National Policy for the Elderly, Article 2, which states that individuals older than 60 years are considered elderly (13). The nutritional status of the patients was assessed right at the beginning of their hospital stay by different parameters, such as: anthropometric and laboratory indicators, habitual energy intake (HEI) assessment and mini nutritional assessment

2 08 MERHI/c/pppp_04 LORD_c 21/10/11 10:08 Page182 (MNA). This hospital does nutritional assessment routinely in at least 70% of the elderly patients admitted. Anthropometric indicators The following anthropometric indicators were measured: current weight (CW), height (A), arm circumference (AC), triceps skinfold thickness (TST), and calf circumference (CC). These measurements allowed the body mass index (BMI) and mid-arm muscle circumference (MAMC) to be calculated. A flexible and inelastic metric tape measure with 0.1 cm accuracy was used to measure AC and CC according to the WHO s standard technique (14). Skinfold thickness was measured with the skin caliper Lange Skinfold Caliper and MAMC was calculated by the formula: MAMC=AC (TST x 0.314). The reference values established by Frisancho (15) for individuals under 65 years of age and those established by Burr and Phillips (16) for individuals 65 years and over were used for nutritional status classification. Body mass index was calculated by dividing the weight by the height squared and classified according to the criteria established by Lipschitz (1994) (17), with the following cut-off points for the elderly: underweight when BMI<22; normal weight when 22<BMI<27; and overweight when BMI>27. Other data, such as lymphocyte count, hemoglobin and length of hospital stay were also obtained from the patients medical records. Investigation of the habitual energy intake (HEI) The HEI was based on the usual eating habits, determined upon admission. The centesimal composition of the foods listed in the food recalls was then calculated by the software NutWin (2002) (18) version 1.5. Next, the estimated caloric adequacy of the habitual energy intake in relation to the energy requirement was determined (%HEI/ER). The energy requirement (ER) represents the total energy expenditure of the individual and was calculated by the Harris-Benedict equation (19). Mini nutritional assessment (MNA) The MNA was administered to the patients as recommended by Guigoz et al, 1994 (1), and includes aspects that are specific for the elderly. MNA includes questions regarding weight change, dietary change, gastrointestinal symptoms that persist for more than two weeks, functional capacity, physical assessment and disease and its relationship with nutritional requirement (1). The patients were then classified as normal (N), malnourished (M) or at risk of malnourishment (RM) depending on their MNA score. Assessment of factors that may predict MNA classification Gender, age, anthropometric indicators such as BMI, CC, AC, TST, MAMC, HEI, ER, %HEI/ER, lymphocyte count and hemoglobin were investigated as factors that can predict MNA classification. Statistical analysis At first, a descriptive analysis of the patients was done by calculating the mean, standard deviation and proportion of the studied variables. The chi-square test was used to compare proportions and the Mann-Whitney test was used to compare continuous or ordinal measures between two groups. Univariate and multiple logistic regression analyses were used to identify the factors that can predict nutritional status determined by the MNA. The respective odds ratio (OR) and confidence interval (CI) of 95% were calculated (20, 21) and the significance level was set at 5% (p<0.05). The data were analyzed by the statistical software SAS (22) (Statistical Analysis System). Results The nutritional status of 61.47% of the 109 studied elderly was normal, 30.28% were at risk of malnourishment and 8.26% were malnourished. Most of the elderly at risk of malnourishment or malnourished were 70 years or older. Table 1 shows the characteristics of the studied population according to the nutritional indicators investigated. It is possible to observe that the mean age of the malnourished individuals was 76.4 years and the length of hospital stay was 6.4±7.9 days. Table 2 shows the nutritional indicators AC, TST and MAMC according to their percentiles and BMI according to the cut-off points for elderly individuals. One third (33.3%) of the TST of individuals who were classified as malnourished by the MNA was under the 10th percentile (<P10), 55.5% were between the 10th and the 90th percentiles and 11% were above the 90th percentile (P>90). Regarding the MAMC of those who were classified as malnourished by the MNA, 55.5% were under the 10th percentile (<P10) and 44.4% were between the 10th and 90th percentiles. Classification of the CC cutoffs into percentiles was not done because there was no available reference for this indicator, classifying the cutoffs into percentiles. Table 2 also lists other data. Table 1 Characteristics of the population according to the studied variables and nutritional status determined with the MNA Nutritional status by the MNA Variables Malnourished At risk Well of malnourishment nourished Age (years) 76.4± ± ±7.2 Length of hospital stay (days) 6.4± ± ±6.2 Current weight (kg) 47.9± ± ±13.9 Usual weight (kg) 55.4± ± ±15.9 Ideal weight (kg) 63.2± ± ±7.5 BMI (kg/m 2 ) 19.2± ± ±5.9 CC (cm) 27.1± ± ±3.9 AC (cm) 23.2± ± ±4.5 TST (mm) 13.0± ± ±10.4 MAMC (mm) 185.8± ± ±31.2 Lymphocytes (cel/mm 3 ) ± ± ±706.6 Hemoglobin (mg/dl) 10.4± ± ±2.4 HEI (kcal) ± ± ±537.3 TER (kcal) ± ± ±319.8 %HEI/TER 53.9± ± ±26.8 BMI: body mass index, CC: calf circumference, AC: arm circumference, TST: triceps skinfold thickness, MAMC: mid-arm muscle circumference, HEI: habitual energy intake, TER: total energy requirement, %HEI/TER: % of the habitual energy intake in relation to the total energy requirement. Table 4 182

3 08 MERHI/c/pppp_04 LORD_c 21/10/11 10:08 Page183 JNHA: NUTRITION Table 2 Classification of the assessed indicators according to the nutritional status determined with the MNA Nutritional status by the MNA Variables Malnourishment Risk of malnourishment Well nourished Total N(%) N(%) N(%) N(%) Arm circumference* Triceps skinfold thickness * Mid-arm muscle circumference* Body mass index ** <P10 5 (55.5) 7(21.2) 5(7.5) 17(15.6) P10-P90 4(44.4) 20(60.6) 40(59.7) 64(58.7) >P90-6(18.1) 22(32.9) 28(27.7) <P10 3(33.3) 7(21.2) 6(8.9) 16(14.7) P10-P90 5(55.5) 20(60.6) 38(56.7) 63(57.8) >P90 1(11.1) 6(18.1) 23(34.3) 30(27.5) <P10 5(55.5) 6(18.1) 12(17.9) 23(21.1) P10-P90 4(44.4) 22(66.6) 37(55.2) 63(57.8) >P90-5(15.1) 18(26.8) 23(21.1) 22 7(77.8) 18(54.5) 11(16.4) 36(33.0) >22<27 2(22.2) 10(30.3) 32(47.4) 44(40.4) 27-5(15.5) 24(35.8) 29(26.7) * Cutoffs for the classification into percentiles, Burr and Phillips :- Arm circumference (AC) - <P10: <223 mm for women and <219 mm for men; P10-P90: from 176 to 305 mm for women and from 198 to 302 mm for men; >P90: >266 mm for women and >262 for men. as a function of age. Triceps skinfold thickness (TST) - <P10: <11.3 mm for women and <4.3 mm for men; P10-P90: from 7.0 to 28.5 mm for women and from 3.9 to 15.2 mm for men; >P90: >19.0 mm for women and >10.6 for men as a function of age. Mid-arm muscle circumference (MAMC) - <P10: <172 mm for women and <196 mm for men; P10-P90: from 150 to 236 mm for women and from 180 to 266 mm for men; >P90: >214 mm for women and >236 for men as a function of age.** Body mass index (BMI) - Lipschitz Table 3 Comparison of the assessed indicators according to the nutritional status determined with the MNA Nutritional status by the MNA Variables Malnourishment + Risk of Well nourished P-value malnourishment Females / Males 40.5% / 59.5% 35.8% / 64.2% * Age Group % 61.2% * % 29.8% % 8.9% Arm circumference <P % 7.4% * P10-P % 59.7% >P % 32.8% Triceps skinfold thickness <P % 8.9% * P10-P % 56.7% >P % 34.3% Mid-arm muscle circumference <P % 17.9% * P10-P % 55.2% >P % 26.8% Age (X±DP) 72.9± ± ** Length of hospital stay (X±DP) 7.6± ± ** Body mass index (X±DP) 21.6± ±5.8 <0.0001** Calf circumference (X±DP) 31.1± ± ** Arm circumference (X±DP) 25.7± ±4.5 <0.0001** Triceps skinfold thickness (X±DP) 12.3± ± ** Mid-arm muscle circumference (X±DP) 217.4± ±31.2 <0.0001** Lymphocyte count (X±DP) ± ± ** Hemoglobin (X±DP) 11.8± ± ** Habitual energy intake (X±DP) ± ± ** %HEI /TER (X±DP) 71.7± ± ** %CEH / TER: habitual energy intake in relation to the total energy requirement; *Chi-square test **Mann-Whitney test 183

4 08 MERHI/c/pppp_04 LORD_c 21/10/11 10:08 Page184 Predictive factors of MNA classification (malnourishment or risk of malnourishment versus well nourished) analyzed by the univariate logistic regression model Variables P-value Odds ratio Confidence interval (95%) Age ; Gender F x M ; BMI ; Calf circumference ; Arm circumference < ; Triceps skinfold thickness ; Mid-arm muscle circumference ; Lymphocytes ; Hemoglobin ; Habitual energy intake ; Total energy requirement ; %HEI/TER ; AC <P10 x >P ; AC P10-P90 x >P ; TST <P10 x >P ; TST P10-P90 x >P ; MAMC <P10 x >P ; MAMC P10-P90 x >P ; BMI normal x excess weight ; BMI underweight x excess weight < ; In order to compare the assessed indicators to the nutritional status determined by the MNA, the group of malnourished individuals (M) was grouped with the group of individuals at risk of malnourishment (RM), since the number of malnourished individuals was small (N=9), and for the statistical analyses (Table 3). There was a statistically significant difference between the groups M + RM and N regarding age (p=0.0146), BMI (p<0.0001), CC (P=0.0004), AC (p<0.0001), TST (p=0.0005) and MAMC (p<0.0001), but no statistical difference in the length of hospital stay (p=0.3769). Low HEI was seen in the three groups of patients classified by the MNA (Table 1) yet there was no statistical difference among them (Table 3). Univariate logistic regression analysis was done to determine if these indicators could predict MNA classification (Table 4). Table 4 shows that if the AC of an individual is below the 10th percentile (<P10), he or she is 8.8 times more likely to be M or RM according to the MNA than a patient whose AC is above the 90th percentile (>P90). Regarding TST, the risk was lower than 1 (OR=0.920) and when TST<P10, the individual was 5.4 times more likely to be malnourished (OR=5.476). Regarding underweight versus overweight according to the BMI, OR=10.909, p< Other results can be seen in Table 4. Finally, multiple logistic regression analysis was used to identify the best model to predict MNA classification (Table 5) starting with the entire model which included all the variables listed in Table 4. This first model identified AC as the best variable to predict MNA classification, with an accuracy of 81%. Since AC is already included in the MNA, a new model excluding AC was proposed, also excluding the laboratory tests lymphocyte count and hemoglobin, since there was not enough information for this type of analysis (within the study period). This second model found that BMI is the best variable to 184 predict MNA classification, with an accuracy of 74%. Finally, since BMI is already a nutritional indicator extensively used in clinical practice, a new model was proposed leaving out this variable (BMI). The new model identified both TST and MAMC as capable of predicting MNA classification, with an accuracy of 78% (TST: p=0.0040; OR=0.920 and MAMC: p=0.0010; OR=0.976). Table 5 Predictive factors of MNA classification (malnourishment or risk fo malnourishment versus well nourished), analyzed by the multiple logistic regression model Variables P-value Odds ratio Confidence interval (95%) 1st model * RM+M x WN Arm circumference (cm) ; Statistics c=0.814 (accuracy) 2nd model ** RM+M x WN BMI normal x excess weight ; BMI underweight x excess weight < ; Statistics c=0.738 (accuracy) 3rd model *** RM+M x WN Triceps skinfold thickness (mm) ; Mid-arm muscle circumference (cm) ; statistics c=0.778 (accuracy) RM: risk of malnourishment, M: malnourishment, WN: well nourished. * Using all the variables in Table 4 and the stepwise variable selection process. **Using most variables in Table 4 except AC (which is already included in the MNA), lymphocytes and hemoglobin (because the numbers were not enough), and the stepwise variable selection process. *** Using most variables in Table 4 except AC (which is already included in the MNA), lymphocytes and hemoglobin (because the numbers are not enough) and BMI (because it is already extensively used in clinical practice) and the stepwise variable selection process. Discussion This study investigated the nutritional status of hospitalized elderly patients by the MNA and other assessment indicators that are routinely used in hospitals. However, the strength of this study was to investigate if these routine nutritional indicators could predict MNA classification, thus conferring credibility to the study. Such knowledge would allow other indicators (anthropometry or laboratory tests) to be used to determine the nutritional status of hospitalized elderly patients when the MNA cannot be used. It is already known that the MNA is a good tool to predict the functional status of institutionalized elderly patients at risk of malnutrition (23). It is also recommended as a routine geriatric assessment tool by ESPEN (European Society for Clinical Nutrition and Metabolism) (24). The global MNA nature allows the inclusion of important factors which do not only classify the nutritional status but also indicate when intervention is necessary to guarantee proper care. Inadequate food intake is the cause of malnutrition while physical and cognitive limitations can prevent adequate food intake (25). Cereda et al., 2008 (23), showed that poorer

5 08 MERHI/c/pppp_04 LORD_c 21/10/11 10:08 Page185 functional status was associated with low BMI, sarcopenia and reduced oral intake and the MNA reliably identifies at-risk institutionalized elderly needing higher standards of care, particularly related to eating. Routine documentation of oral intakes and feeding assistance might be useful to prevent weight loss, sarcopenia and functional status deterioration. In hospital settings, a low MNA score is associated with increased mortality, prolonged length of stay and greater likelihood of discharge to nursing homes. Malnutrition is associated with functional and cognitive impairment and eating difficulties. The MNA detects risk of malnutrition before severe changes in weight or serum proteins occur (26). However, MNA administration takes time, which is not always available to health professionals in hospitals for the nutritional care of elderly patients. In a recent study, Kaiser et al., 2009 (27), proposed the validation of a shorter MNA (MNA -SF) version and claimed this to be a valid instrument for geriatric care, with the option of using CC when BMI calculation is not possible. This short MNA (MNA -SF) version would demand less time for the routine nutritional assessment of the geriatric population. This study reinforces the importance of the MNA as an instrument to assess the nutritional status of the elderly since it represents a global assessment instrument. It also shows other nutritional assessment parameters that can be used safely in this population, especially when the MNA is not available or possible. There was no statistical difference among the groups regarding laboratory data (lymphocyte counts and hemoglobin values) yet the mean values were lower in malnourished individuals and in those at risk of malnourishment (Tables 1 and 3), which has also been observed in other recent studies (28), and the use of univariate analysis showed that the parameters BMI, CC, TST and MAMC can predict MNA classification. This study also found that as TST increases, nutritional risk decreases (OR=0.9). When multiple logistic regression is used to analyze the third model, one observes that as TST and MAMC increases, the risk of malnutrition decreases, with an accuracy of (Table 5). Finally, the data in this study show that the factors that can predict MNA classification are, first, AC, then BMI and finally TST and MAMC equally, which could be taken once or twice weekly in routine clinical practice. Thus, if it is not possible to use the MNA, it is possible to use one of these indicators in its place. Analysis and comparison of nutritional assessment methods need to be investigated very attentively, taking into account the differences in interpretation and the type of population before the instrument can be used in clinical practice. A study that analyzed 44 different nutritional assessment instruments found that the methodology was not appropriate in most of them (9). Another study which compared the accuracy of traditional nutritional assessment parameters verified that the methods used were considered weak predictors of clinical outcomes, death, infection and length of hospital stay (29). Other studies indicate that Malnutrition Universal Screening Tool (MUST) JNHA: NUTRITION 185 and serum albumin are sensitive in the identification of nutritional risk (30). In this study, 38.5% of the patients were malnourished or at risk of malnourishment and this diagnosis and the investigation of the other nutritional indicators were done in the first 48 hours of their hospital stay, which may suggest that these patients were already malnourished or at risk of malnourishment at admission. The fact that malnourished individuals and those at risk of malnourishment had a HEI similar to that of well nourished individuals (1331.1±412.9 and ±537.3) was eye-catching, but there was no statistical significance (p=0.9479), (Table 3); this may have been influenced by the kind of disease. This fact reinforces the importance of nutritional assessment at the beginning of patients stay so that intervention strategies can be used soon enough to improve their nutritional status and clinical course of the disease and reduce their length of stay (11, 12, 31). It is also important to emphasize the difficulty of assessing the nutritional status of hospitalized elderly patients since many of them are bedridden, some are unconscious, and thus cannot be weighed. In these cases, the CC has been considered a good indicator (32, 33). In this study, univariate regression analysis investigated CC as a predictive factor of the MNA, with p= and OR=0.08; malnourished patients or those at risk of malnourishment had a mean CC of 27.1±2.8 and 32.3±4.5, respectively, which confirms an association between CC and the risk of becoming malnourished in the elderly, as seen in other recent studies (31). Conclusion In the conditions of this study, first AC, then BMI and lastly TST and MAMC together were considered factors that predict MNA classification, that is, if it is not possible to use the MNA, the other nutritional indicators can be used for the diagnosis of the nutritional status of hospitalized elderly patients. Acknowledgments: Sponsored by: Research Support Fund from the Pontifical Catholic University of Campinas, Brazil. Competing Interest: The authors declare that they have no competing interests. Authors contributions: - Vania Aparecida Leandro Merhi was involved in the protocol and study design, analysis, carried out the statistical analysis, writing of the article and critically reviewed the article.- José Luiz Braga de Aquino was involved in the study design and critically reviewed the article. The authors read and approved the final manuscript. References 1. Guigoz Y, Vellas B, Garry PJ. Mini nutritional assessment: A practical assessment tool for grading the nutritional state of elderly patients. Facts and Research in Gerontology 1994; Supplement (2): Bauer JM, Sieber CC. Significance and diagnosis of malnutrition in the elderly. Z Arztl Fortbild Qualitatssich 2007; 101(9): Cuyac Lantigua M, Santana Porbén S. The Mini Nutritional Assessment of the elderly in the practice of a hospital geriatrics service: inception, validation and operational characteristics. Arch Latinoam Nutr 2007; 54(3): Ferreira LS, Nascimento LF, Marucci MF. Use of the mini nutritional assessment

6 08 MERHI/c/pppp_04 LORD_c 21/10/11 10:08 Page186 tool in elderly people from long-term institutions of southeast of Brazil. J Nutr Health & Aging 2008; 12(3): González Hernández A, Cuyá Lantigua M, González Escudero H, Sánchez Gutiérrez R, Cortina Martínez R, Barreto Penié J, Santana Porbén S, Rojas Pérez A. Nutritional status of Cuban elders in three different geriatric scenarios: community, geriatrics service, nursery home. Arch Latinoam Nutr 2007; 57(3): Tsai AC, Ho CS, Chang MC. Assessing the Prevalence of Malnutrition with the Mini Nutritional Assessment (MNA) in a Nationally Representative Sample of Elderly Taiwanese. J Nutr Health & Aging 2008; 12(4): Lei Z, Qingyi D, Feng G, Chen W, Shoshana Hock R, Changli W. Clinical study of mini-nutritional assessment for older chinese inpatients. J Nutr Health & Aging 2009; 13(10): Izawa S, Kuzuya M, Okada K, Enoki H, Koike T, Kanda S, Iguchi A. The nutritional status of frail elderly with care needs according to the mini-nutritional assessment. Clinical Nutrition 2006; 25(6): Jones JM: The methodology of nutritional screening and assessment tools. J Hum Nutr Diet 2002; 15(1): Martins CP, Correia JR, do Amaral TF: Undernutrition risk screening and length of stay of hospitalized elderly. J Nutr Elder 2005, 25(2): Rauen MS, Moreira EAM, Calvo MCM, Lobo AS. Avaliação do estado nutricional de idosos institucionalizados. Rev Nutr 2008; 21(3): Gaino NM, Leandro-Merhi VA, Oliveira MRM. Idosos hospitalizados: estado nutricional, dieta, doença e tempo de internação. Rev Bras Nutr Clin 2007; 22(4): BRASIL: Lei nº , de 4 de janeiro de Dispõe sobre a política nacional do idoso, cria o Conselho Nacional do Idoso e dá outras providências. [ 14. World Health Organization. Physical status: the use and interpretation of anthropometry. Geneva: World Health Organization; WHO technical report series Frisancho AR. New norms of upper limb fat and muscle areas for assessment of nutritional status. Am J Clin Nutr 1981; 34: Burr ML, Phillips MK. Anthropometric norms in the elderly. Br J Nutr 1984; 51: Lipschitz DA. Screening for nutritional status in the elderly. Prim Care 1994; 22(1): Universidade Federal de São Paulo. Escola Paulista de Medicina. Programa de Apoio a Nutrição (NUTWIN) programa de computador, versão 1.5. São Paulo: UNIFESP/ EPM; Harris J, Benedict F. A biometric study of basal metabolism in man. Washington D.C. Carnegie Institute of Washington Conover WJ. (1971). Practical Nonparametric Statistics. John Wiley & Sons Inc. Nova Iorque. 21. Tabachnick BG, Fidell LS. (2001). Using Multivariate Statistics. 4ª ed. Allyn&Bacon. Needham Heights. MA. USA. 22. SAS System for Windows (Statistical Analysis System), versão Service Pack 3. SAS Institute Inc, , Cary, NC, USA. 23. Cereda E, Valzolgher L, Pedrolli C: Mini nutritional assessment is a good predictor of functional status in institutionalised elderly at risk of malnutrition. Clin Nutr 2008; 27(5): Kondrup J, Allison SP, Elia M, Vellas B, Plauth M; Educational and Clinical Practice Committee, European Society of Parenteral and Enteral Nutrition (ESPEN). ESPEN guidelines for nutrition screening Clin Nutr 2003; 22(4): Bo M, Massaia M, Raspo S; Bosco F, Cena P; Molaschi M, Fabris F: Preventive factors of in-hospital mortality in older patients admitted to a medical intensive care unit. J Am Geriatr Soc 2003; 51: Guigoz Y: The Mini Nutritional Assessment (MNA) review of the literature-what does it tell us? J Nutr Health & Aging 2006; 10(6): Kaiser MJ, Bauer JM, Ramsch C, Uter W, Guigoz Y, Cederholm T, Thomas DR, Anthony P, Charlton KE, Maggio M, Tsai AC, Grathwohl D, Vellas B, Sieber CC. Validation of the mini nutritional assessment short-form (MNA -SF): a practical tool for identification of nutritional status. J Nutr Health & Aging 2009; 13(9): Venzin RM, Kamber N, Keller WCF, Suter PM, Reinhart WH. How important is malnutrition? A prospective study in internal medicine. Eur J Clin Nutr 2009; 63 (3): Beghetto MG, Luft VC, Mello ED, Polanczyk CA. Accuracy of nutritional assessment tools for predicting adverse hospital outcomes. Nutr Hosp 2009; 24(1): Pereira Borges N, Alegria Silva BD, Cohen C, Portari Filho PE, Medeiros FJ. Comparison of the nutritional diagnosis, obtained through different methods and indicators, in patients with cancer. Nutr Hosp 2009; 24(1): Correia IMTD, Waitzberg DL. The impact of malnutrition on morbidity, mortality, length of hospital stay and costs evaluated through a multivariate model analysis. Clin Nutr 2003; 22(3): Cuervo M, Ansorena D, Garcia A, González Martínez MA, Astiasarán I, Martínez JA. Valoración de la circunferencia de la pantorrilla como indicador de riesgo de desnutrición en personas mayores. Nutr Hosp 2009; 24(1): Mclellan KCP, Staudt C, Silva FRF, Bernardi JLD, Frenhani PB, Leandro-Merhi VA. The use of calf circumference measurement as an anthropometric tool to monitor nutritional status in elderly inpatients. J Nutr Health & Aging 2010; 14(4): DOI /s