- Email: [email protected]
Reliability of Mini Nutritional Assessment in Hemodialysis Compared With Subjective Global Assessment
ORIGINAL RESEARCH
Reliability of Mini Nutritional Assessment in Hemodialysis Compared With Subjective Global Assessment Baris Afsar, MD,* Siren Sezer,* Zubeyde Arat, MD,* Emre Tutal,* Fatma Nurhan Ozdemir,* and Mehmet Haberal† Protein-energy malnutrition (PEM) is common in hemodialysis patients. Subjective Global Assesment (SGA) and Mini Nutritional Assessment (MNA) are two tools for monitoring PEM. Our aim was to determine reliability of MNA in detecting malnutrition in hemodialysis patients in comparison with SGA. The study population consisted of 137 patients with pure PEM with no signs of chronic inflammation. Nutritional statuses of patients were assessed concomitantly by SGA and MNA. Ninety-two patients were in SGA-A, 40 patients were in SGA-B, and 5 patients were in SGA-C. Forty-seven patients were in MNA-1, 77 patients were in MNA-2, and 13 patients were in MNA-3. Albumin (P ⫽ .0001), prealbumin (P ⫽ .0001), body mass index (P ⫽ .01), creatinine (P ⫽ .0001), and nPNA (P ⫽ .04) were statistically different between SGA groups. Creatinine (P ⫽ .001), blood urea nitrogen (P ⫽ .017), albumin (P ⫽ .001), prealbumin (P ⫽ .005), body mass index (P ⫽ .0001), and nPNA (P ⫽ .005) were statistically different between MNA groups. Fifty-two patients who had no evidence of malnutrition according to SGA were defined as having moderate malnutrition according to MNA. Seven patients who were in a state of moderate malnutrition determined by SGA were in good nutritional status according to MNA. SGA identified 8 patients as moderately malnourished; the same patients were defined as having severe malnutrition in MNA. Our results suggest that MNA might underestimate the nutritional status of hemodialysis patients who are not in an inflammatory state and may not be a reliable method for detecting moderate malnutrition when compared with SGA. © 2006 by the National Kidney Foundation, Inc.
P
ROTEIN-ENERGY malnutrition (PEM) is a commonly seen health problem in end stage renal disease (ESRD) patients. Various studies have shown signs of malnutrition in 23% to 76% of hemodialysis (HD) and 18% to 50% of peritoneal dialysis (PD) patients.1-4 There is strong association between nutritional status and clinical outcome in HD patients, providing support to the hypothesis that PEM may cause or contribute to morbidity and mortality.5 *Department of Nephrology, Baskent University Hospital, Ankara, Turkey. †Department of General Surgery, Baskent University Hospital, Ankara, Turkey. Address reprint requests to Baris Afsar, MD, 3. Cadde 50, Sokak 9/8 06500, Bahçelievler, Ankara, Turkey. E-mail: afsarbrs@yahoo. com © 2006 by the National Kidney Foundation, Inc. 1051-2276/06/1603-0021$32.00/0 doi:10.1053/j.jrn.2006.01.012
Journal of Renal Nutrition, Vol 16, No 3 ( July), 2006: pp 277-282
Assessment and monitoring of protein and energy statuses are essential to prevent, to diagnose, and to treat uremic malnutrition. Subjective Global Assessment (SGA) and Mini Nutritional Assessment (MNA) are two tools for monitoring PEM. The basis of SGA is to determine whether nutrient assimilation has been restricted because of decreased food intake, maldigestion, or malabsorption; whether any effects of malnutrition on organ function have occurred; and whether patients’ disease processes influence nutrient requirements.6 SGA has been used for nutritional assessment of chronic HD patients and is independently associated with mortality in this patient population.7 MNA is a simple, noninvasive, inexpensive, and easily preformed method used especially in the elderly population.8 The data about the use of MNA for determining PEM in ESRD patients is scarce. In this study, we attempted to determine the reliability of MNA in 277
278
AFSAR ET AL
detecting malnutrition in HD patients in comparison with SGA.
Methods This cross-sectional study was undertaken in the HD unit of Baskent University Ankara Hospital. The study received local ethical approval, and verbal informed consent was obtained from all patients before enrolment. All patients were receiving 4-hour HD 3 times weekly with standard bicarbonate dialysis (sodium 138 mmol/L, bicarbonate 35 mmol/L, potassium 1.5 mmol/L, calcium 1.25 mmol/L, and magnesium 1.8 mmol/L) by either high-flux HD and 1.1 to 1.7 m2 hollow fiber or flat plate dialyzer at the time the research was conducted. The group was composed of patients with pure PEM with no signs of chronic inflammation, defined as C-reactive protein concentrations ⬍10 mg/dL. Medical problems including any clinical events that could cause inflammation (such as infections, hepatitis, operations, trauma, nonspecific fever, and new cardiovascular events including acute coronary syndromes, cerebrovascular disease, and peripheric arterial disease) were noted, and patients with these conditions were excluded. Initially, 154 patients were included in the study. Five patients with chronic liver disease (4 hepatitis C, 1 hepatitis B), 2 with diabetic foot ulcers, 1 with bladder carcinoma, 1 with nonspecific fever, 1 with recent myocardial infarction, and 1 with recent history of trauma (traffic accident) were excluded from the study. Two patients receiving nutritional support were not included. Four patients refused to participate in the study. The final study population consisted of 56 (41%) female and 81 male (59%) patients (mean age, 41.4 ⫾ 12.1 years; mean HD duration, 72.4 ⫾ 38.1 months). The nutritional statuses of HD patients were assessed concomitantly using SGA and MNA tests. History used in SGA focused on 5 areas. Percentage of body weight loss in the previous 6 months was characterized as mild (⬍5%), moderate (5% to 10%), or severe (⬎10%). Dietary intake was classified as normal or abnormal as judged by a change in intake and whether the current diet was nutritionally adequate. Presence of gastrointestinal symptoms such as nausea, vomiting, diarrhea, and abdominal pain were recorded. A patient’s functional capacity was defined as bedridden, suboptimally active, or full
capacity. The last feature of history concerned metabolic demands of patients’ underlying disease states. In physical examination, loss of subcutaneous fat was measured in the triceps region and mid-axillary line at the level of the lower ribs. The second feature was muscle wasting in temporal areas and in deltoids and quadriceps, as determined by loss of bulk and tone detectible by palpation. Presence of edema in the ankle and sacral regions and the presence of ascites were noted. The findings of history and physical examination were used to categorize patients as well-nourished (no malnutrition), as having moderate malnutrition, or as having severe malnutrition.6 MNA was carried out using a MNA scale consisting of 18 point-weighted questions in 4 categories. The first part consisted of 4 anthropometric measurements including body mass index, mid-upper arm and calf circumferences, and weight loss during the past 3 months, followed by 6 global questions regarding accommodation type, pharmaceutical consumption, acute diseases (including psychological stress), mobility, neuropsychologic problems, and pressure sores/skin ulcers. The third part consisted of questions to assess dietary intake, including how many whole meals were eaten, food choice of the patient, fluid intake per day, and how much help was required during meals. The final part of the scale consisted of 2 self-assessments of whether food intake was sufficient and of their own health status. The answer point scale gave a maximum of 30 points. Less than 17 points indicated malnutrition, 17 to 23.5 points indicated a risk for malnutrition, and ⱖ24 points indicated that the person was wellnourished.9 All patients were adequately dialyzed, expressed as equilibrated spKt/V. Urea kinetic modeling was performed to assess the delivered dose of dialysis. HD dose was evaluated using the following formula: spKt ⁄ V⫽ ⫺Ln 共R ⫺ 0.008 ⫻ t兲 ⫹ 共4 ⫺ 关3.5 ⫻ R兴 ⫻ UF ⁄ W兲 where spKt/V is a single-pool Kt/V, R is the ratio of postdialysis to predialysis serum urea nitrogen, t is time on dialysis in hours, UF is the amount of ultrafiltration in liters, and W is postdialysis body weight in kilograms.
279
RELIABILITY OF MNA AND SGA
The normalized protein equivalent of total nitrogen appearance (nPNA), also known as normalized protein catabolic rate (nPCR), was calculated to estimate the daily protein intake by the following formula: nPNA 共g/kg/d兲 ⫽ C/关36.3 ⫹ 5.48 共spKt/V兲 ⫹ 53.5/共spKt/V兲兴 ⫹ 0.168 where C is the predialysis concentration of serum urea nitrogen in milligrams per deciliter. Body mass index was calculated as the ratio of dry weight in kilograms (end-dialysis weight) to height squared (in square meters). The mean laboratory parameters of the last 6 months, including serum hemoglobin, albumin, prealbumin, C-reactive protein, predialysis calcium and phosphorus, predialysis blood urea nitrogen and creatinine, total cholesterol, high-density lipoprotein, low-density lipoprotein, and triglycerides, were measured before beginning the dialysis session. Postdialysis serum urea nitrogen concentrations used to calculate the urea reduction ratio were also measured. According to the results of SGA, the patients are classified into 3 SGA groups, defined as SGA-A, SGA-B, and SGA-C. The groups were composed of patients who were well-nourished (no malnutrition), moderately malnourished, and severely malnourished, respectively. Based on MNA results, the patients were also divided into 3 groups defined as patients with no malnutrition, MNA-1; patients with malnutrition risk or moderate malnutrition, MNA-2; and patients with severe malnutrition, MNA-3.
Table 1. Demographic Characteristics of the Patients
*Mean ⫾ standard deviation.
SGA groups SGA-A (n ⫽ 92, 67.2%) SGA-B (n ⫽ 40, 29.2%) SGA-C (n ⫽ 5, 3.6%) MNA groups MNA-1 (n ⫽ 47, 34.3%) MNA-2 (n ⫽ 77, 56.2%) MNA-3 (n ⫽ 13, 9.5%)
Age (y)
HD Duration (mo)
P
41.5 ⫾ 12.4
74.0 ⫾ 37.7
NS*
42.2 ⫾ 11.8
66.1 ⫾ 36.2
NS
33.6 ⫾ 2.0
80.2 ⫾ 12.3
NS
43.0 ⫾ 11.7
73.6 ⫾ 39.7
NS
40.7 ⫾ 13.0
70.2 ⫾ 37.7
NS
39.5 ⫾ 6.6
75.3 ⫾ 13.5
NS
*NS, not significant (P ⬎ .05).
Statistical Analysis All values are expressed as mean ⫾ SD and percent. The data were analyzed using the program SPSS 11.0 for Windows (SPSS Inc, Chicago, IL). In statistical analysis, SGA and MNA were considered as categorical variables. The within-group parameters were analyzed by Kolmogorov-Smirnov with a Lilliefors significance level and Shapiro-Wilk statistics for normality. Normally distributed continuous parameters between 3 groups were analyzed by one-way analysis of variance, and between 2 groups were analyzed by the Student t test. Parameters that were not normally distributed between the 3 groups were compared by the Kruskal-Wallis test and between 2 groups by the Mann-Whitney U test. A P value ⬍ 0.05 was accepted as statistically significant.
Results
Parameters Age (mean, y) Male/female (n) Dialysis duration (mean, mo) Etiologies (n) Nephrolithiasis Diabetes mellitus Glomerulonephritis Amyloidosis Hypertension Unknown
Table 2. Classification of Patients According to SGA and MNA
41.4 ⫾ 12.1* 81/56 72.4 ⫾ 38.1* 137 63 20 13 11 10 20
The patient characteristics, including etiology of renal failure, are summarized in Table 1. To evaluate the degree of reproducibility, the same physician repeated the SGA and MNA after 10 days on a subset of 15 patients without reference to the first SGA and MNA evaluations. The reliability coefficients (alpha) for between 2 SGA and MNA assessments were 0.91 and 0.93, respectively, representing a good degree of reproducibility.
280
AFSAR ET AL
Table 3. Comparison of 3 SGA Groups SGA Groups Hemoglobin (g/dL) Creatinine (mg/dL) BUN (mg/dL) Albumin (g/dL) Prealbumin (mg/dL) Cholesterol (mg/dL) Triglyceride (mg/dL) Phosphorus (mg/dL) BMI spKt/V nPNA
SGA-A (n ⫽ 92)
SGA-B (n ⫽ 40)
SGA-C (n ⫽ 5)
P
11.3 ⫾ 0.9 10.8 ⫾ 2.1 79.1 ⫾ 13.0 4.2 ⫾ 0.3 36.2 ⫾ 4.7 162.9 ⫾ 33.3 152.2 ⫾ 57.6 5.4 ⫾ 1.3 22.8 ⫾ 3.4 1.1 ⫾ 0.2 1.0 ⫾ 0.1
11.3 ⫾ 1.1 9.6 ⫾ 2.7 80.3 ⫾ 14.5 3.7 ⫾ 0.3 32.2 ⫾ 4.1 155.8 ⫾ 27.1 133.3 ⫾ 43.9 5.3 ⫾ 1.3 21.2 ⫾ 3.6 1.1 ⫾ 0.1 1.0 ⫾ 0.1
10.7 ⫾ 0.5 7.2 ⫾ 0.3 68.6 ⫾ 3.9 3.4 ⫾ 0.1 29.8 ⫾ 4.9 148.0 ⫾ 10.6 132.2 ⫾ 4.4 4.6 ⫾ 0.6 19.4 ⫾ 1.1 1.1 ⫾ 0.1 0.8 ⫾ 0.2
.3 .0001* .2 .0001* .0001* .3 .1 .4 .01* .6 .04*
*Kruskal-Wallis and one-way ANOVA.
According to SGA results, 92 patients (67.2%) were in SGA-A, 40 patients (29.2%) were in SGA-B, and 5 patients (3.6%) were in SGA-C. There were no statistically significant differences between ages and HD durations of SGA groups (Table 2). Forty-seven patients (34.3%) were in MNA-1, 77 patients (56.2%) were in MNA-2, and 13 patients (9.5%) were in MNA-3; the patients were also similar in terms of mean ages and HD durations (Table 2). Comparison of 3 SGA groups showed that the albumin (P ⫽ .0001), prealbumin (P ⫽ .0001), body mass index (P ⫽ .01), creatinine (P ⫽ .0001), and nPNA (P ⫽ .04) of the patients were statistically different between SGA groups. Other parameters, including spKt/V, did not reach statistical significance (P ⬎ .05) (Table 3). We combined the small number of severely malnourished patients in SGA-C (5 patients) with the moderately malnourished SGA-B (n ⫽ 40) patients and reanalyzed the parameters between SGA-A and SGA-B and C. The two groups still differed in terms of albumin (P ⫽ .0001), prealbumin (P ⫽ .0001), body mass index (P ⫽ .01),
and creatinine (P ⫽ .006), but the difference for nPNA disappeared (P ⫽ .4) (data not shown). When we compared 3 MNA groups, we found that creatinine (P ⫽ .001), blood urea nitrogen (P ⫽ .017), albumin (P ⫽ .001), prealbumin (P ⫽ .005), body mass index (P ⫽ .0001), and nPNA (P ⫽ .005) were statistically different (Table 4). The remaining parameters were not statistically different (P ⬎ .05). We combined the small number of severely malnourished patients in MNA-3 (13 patients) and moderately malnourished patients in MNA-2 (n ⫽ 77) and reanalyzed the parameters between MNA-1 and MNA-2 and 3; body mass index (P ⫽ .0001) and creatinine (P ⫽ .04) differed statistically between MNA-1 and MNA-2 and 3, and the differences between albumin (P ⫽ .17), prealbumin (P ⫽ .42), blood urea nitrogen (P ⫽ .6), and nPNA (P ⫽ .2) had disappeared (data not shown). Our data showed that the MNA and SGA results regarding patients’ nutritional statuses were inconsistent with each other. Fifty-two patients who had no evidence of malnutrition according to SGA were defined as
Table 4. Comparison of 3 MNA Groups MNA Groups Hemoglobin (g/dL) Creatinine (mg/dL) BUN (mg/dL) Albumin (g/dL) Prealbumin (mg/dL) Cholesterol (mg/dL) Triglyceride (mg/dL) Phosphorus (mg/dL) BMI spKt/V nPNA
MNA-1
MNA-2
MNA-3
P
11.4 ⫾ 0.9 11.0 ⫾ 2.0 79.8 ⫾ 11.3 4.1 ⫾ 0.4 35.4 ⫾ 5.0 158.9 ⫾ 32.0 159.7 ⫾ 60.2 5.2 ⫾ 1.1 24.5 ⫾ 3.2 1.0 ⫾ 0.2 1.0 ⫾ 0.1
11.2 ⫾ 1.0 10.5 ⫾ 2.3 80.2 ⫾ 14.7 4.0 ⫾ 0.3 35.4 ⫾ 4.5 161.3 ⫾ 32.7 140.1 ⫾ 51.5 5.5 ⫾ 1.4 21.4 ⫾ 3.2 1.1 ⫾ 0.1 1.0 ⫾ 0.2
11.2 ⫾ 0.9 7.1 ⫾ 1.2 69.0 ⫾ 6.2 3.6 ⫾ 0.2 30.6 ⫾ 5.5 158.9 ⫾ 16.9 131.0 ⫾ 21.4 4.7 ⫾ 0.8 18.9 ⫾ 1.5 1.1 ⫾ 0.1 0.9 ⫾ 0.09
.57 .001* .017* .001* .005* .9 .1 .08 .0001* .2 .005*
*Kruskal-Wallis and one-way ANOVA.
RELIABILITY OF MNA AND SGA Table 5. Cross Table of Nutritional Statuses of the Patients According to SGA and MNA SGA Groups (n)
MNA groups (n) MNA-1 MNA-2 MNA-3 Total
SGA-A
SGA-B
SGA-C
Total
40 52 0 92
7 25 8 40
0 0 5 5
47 77 13 137
having moderate malnutrition according to MNA. Seven patients who were in a state of moderate malnutrition determined by SGA were in good nutritional status according to MNA. SGA identified 8 patients as moderately malnourished, whereas the same patients were defined as having severe malnutrition in MNA. Only 5 patients were found to have severe malnutrition by both SGA and MNA (Table 5).
Discussion In this study, we found that MNA may not be a reliable method for detecting moderate malnutrition in a subset of HD patients who are not in an inflammatory state when compared with SGA. In our opinion, MNA might underestimate the nutritional status of HD patients by identifying them as having moderate malnutrition, although they are in good nutritional status according to SGA. PEM is one of the major predictors of poor clinical outcome in the HD population.10 A variety of factors may cause or contribute to the development of malnutrition in renal failure, including inadequate nutrient intake, inadequate dialysis, loss of proteins and amino acids during dialysis, increased catabolism, metabolic acidosis, and hormonal and metabolic disturbances. So far, we do not have a single definite test by which to assess nutritional status in ESRD patients. SGA was originally developed to identify poor nutritional status in subjects undergoing gastrointestinal surgery, but because the technique of SGA allows rapid, easy, and equipment-free scoring of nutritional status, it has been used widely to quantify the prevalence of malnutrition in HD and PD patients.11-14 SGA has a high predictive value for determining mortality in ESRD patients.15,16 SGA is accepted as a reliable method for detecting malnutrition in ESRD patients and is recommended in screening for malnutrition
281
according to the United States, United Kingdom, and European guidelines.17-19 MNA is another method of detecting malnutrition. The MNA scores correlate with nutritional intake and anthropometric and biological nutritional parameters, and may predict mortality in geriatric patients.20,21 To our knowledge, there exists no study in the literature regarding the use of MNA for the assessment of the nutritional status of ESRD patients. In our study, we attempted to analyze the reliability of MNA in detecting PEM when compared with SGA. The reliability of SGA in showing the nutritional status of HD patients in our study was reinforced by its compatibility with the anthropometric (body mass index) and biochemical nutritional parameters (albumin, prealbumin, creatinine) as well as nPNA in well-nourished and moderately and severely malnourished patients. When the small number of severely malnourished patients in SGA-C was combined with moderately malnourished patients in SGA-B (SGA-B and C) and was compared with wellnourished patients (SGA-A), except for nPNA, the difference between both anthropometric and biochemical nutritional parameters remained significant, showing that SGA readily distinguishes malnourished patients (SGA-B and C) from wellnourished patients (SGA-A). When the small number of severely malnourished patients in MNA-3 was combined with moderately malnourished patients in MNA-2 (MNA-2 and 3) and was compared with well-nourished patients (MNA-1), we found that although the correlation of MNA scores with anthropometric parameters was sustained, the correlation with biochemical nutritional parameters and nPNA were lost. In other words, although MNA distinguished severely malnourished patients from others, unlike SGA, the method failed to identify patients with malnutrition (MNA-2 and 3) before severe malnutrition developed. In our study, we found that 52 patients who had no malnutrition according to SGA scores were identified as having malnutrition according to MNA (Table 5). This misinterpretation might result in unnecessary nutritional intervention in this group of patients, who were assessed as having malnutrition according to MNA, but in reality were not malnourished. Conversely, 7 patients who were in a state of malnutrition according to SGA were regarded as having PEM by MNA. This might
282
AFSAR ET AL
lead to underdiagnosis of truly malnourished patients, causing a delay of necessary interventions and resulting in deterioration of the health status. The compliance of MNA scores with SGA scores in detecting the severely malnourished patients is not considered to be a great concern because this group of patients can be diagnosed easily by medical stuff even without any method. There may be reasons for MNA to not to detect nutritional status correctly in dialysis patients. The questions in MNA are mostly subjective in nature. The patients may not correctly remember the answers of the questions regarding the last 3 months, may be reluctant to answer the questions, or the answers may be blunted because of neuropsychiatric and intellectual problems. Because MNA is a method normally used in patients without any dietary restrictions, ESRD patients may not be correctly assessed because of their dietary modifications. There are not enough data in the literature regarding the use and reliability of MNA in determining nutritional status in HD patients. As far as we know, this study is the first in the literature in which SGA and MNA were concomitantly performed and the reliability of MNA was tested in the assessment of nutritional status in the HD population. In conclusion, we suggest that MNA may not be as reliable as SGA in detecting PEM in the HD population.
References 1. Cianciaruso B, Brunori G, Kopple JD, et al: Cross-sectional comparison of malnutrition in continuous ambulatory peritoneal dialysis and hemodialysis patients. Am J Kidney Dis 26:475-486,1995 2. Marckmann P: Nutritional status of patients on hemodialysis and peritoneal dialysis. Clin Nephrol 29:75-78, 1988 3. Bergstrom J, Lindholm B: Nutrition and adequacy of dialysis. How do hemodialysis and CAPD patients compare? Kidney Int Suppl 40:S39-50, 1993 4. Qureshi AR, Alvestrand A, Danielsson A, et al: Factors predicting malnutrition in hemodialysis patients. A cross-sectional study. Kidney Int 53:773-782, 1998 5. Hakim RM, Levin N: Malnutrition in hemodialysis patients. Am J Kidney Dis 21:125-137, 1993
6. Pupim LB, Ikizler TA: Assessment and monitoring of uremic malnutrition. J Ren Nutr 14:6-19, 2004 7. Pifer TB, McCullough KP, Port FK, et al: Mortality risk in hemodialysis patients and changes in nutritional indicators: DOPPS. Kidney Int 62:2238-2245, 2002 8. Guigoz Y, Lauque S, Vellas BJ: Identifying the elderly at risk for malnutrition. The Mini Nutritional Assessment. Clin Geriatr Med 18:737-757, 2002 9. Saletti A, Lindgren EY, Johansson L, et al: Nutritional status according to Mini Nutritional Assessment in an institutionalized elderly population in Sweden. Gerontology 46:139-145, 2000 10. Lowrie EG, Huang WH, Lew NL, et al: The relative contribution of measured variables to death risk among hemodialysis patients, in Friedman EADeath on Hemodialysis. Amsterdam, Kluwer Academic, 1994, pp 121–141 11. Detsky AS, Baker JP, O’Rourke K, et al: Predicting nutrition-associated complications for patients undergoing gastrointestinal surgery. J Parenter Enteral Nutr 11:440-446, 1987 12. Enia G, Sicuso C, Alati G, et al: Subjective global assessment of nutrition in dialysis patients. Nephrol Dial Transplant 8:1094-1098, 1993 13. Kawaguchi Y, Sugino N, Arai J, et al: Nutritional assessment of patients on continuous ambulatory peritoneal dialysis. Nippon Jinzo Gakkai Shi 35:843-851, 1993 14. Young GA, Kopple JD, Lindholm B, et al: Nutritional assessment of continuous ambulatory peritoneal dialysis patients: An international study. Am J Kidney Dis 17:462-471, 1991 15. McCusker FX, Teehan BP, Thorpe KE, et al: How much peritoneal dialysis is required for the maintenance of a good nutritional state? Canada-USA (CANUSA) Peritoneal Dialysis Study Group. Kidney Int Suppl 56:S56-61, 1996 16. Lawson JA, Lazarus R, Kelly JJ: Prevalence and prognostic significance of malnutrition in chronic renal insufficiency. J Ren Nutr 11:16-22, 2000 17. K/DOQI, National Kidney Foundation: Clinical practice guidelines for nutrition in chronic renal failure. Am J Kidney Dis 35:S1-140, 2000 (suppl 2) 18. Treatment of Adults and Children With Renal Failure– Standards Document (ed 3). London, England, Royal Collage of Physicians of London/Renal Association, 2002 19. Locatelli F, Fouque D, Heimburger O, et al: Nutritional status in dialysis patients: A European consensus. Nephrol Dial Transplant 17:563-572, 2002 20. Persson MD, Brismar KE, Katzarski KS, et al: Nutritional status using Mini Nutritional Assessment and Subjective Global Assessment predict mortality in geriatric patients. J Am Geriatr Soc 50:1996-2002, 2002 21. Vellas B, Guigoz Y, Baumgartner M, et al: Relationships between nutritional markers and the mini-nutritional assessment in 155 older persons. J Am Geriatr Soc 48:13001309, 2000
Reliability of Mini Nutritional Assessment in Hemodialysis Compared With Subjective Global Assessment Baris Afsar, MD,* Siren Sezer,* Zubeyde Arat, MD,* Emre Tutal,* Fatma Nurhan Ozdemir,* and Mehmet Haberal† Protein-energy malnutrition (PEM) is common in hemodialysis patients. Subjective Global Assesment (SGA) and Mini Nutritional Assessment (MNA) are two tools for monitoring PEM. Our aim was to determine reliability of MNA in detecting malnutrition in hemodialysis patients in comparison with SGA. The study population consisted of 137 patients with pure PEM with no signs of chronic inflammation. Nutritional statuses of patients were assessed concomitantly by SGA and MNA. Ninety-two patients were in SGA-A, 40 patients were in SGA-B, and 5 patients were in SGA-C. Forty-seven patients were in MNA-1, 77 patients were in MNA-2, and 13 patients were in MNA-3. Albumin (P ⫽ .0001), prealbumin (P ⫽ .0001), body mass index (P ⫽ .01), creatinine (P ⫽ .0001), and nPNA (P ⫽ .04) were statistically different between SGA groups. Creatinine (P ⫽ .001), blood urea nitrogen (P ⫽ .017), albumin (P ⫽ .001), prealbumin (P ⫽ .005), body mass index (P ⫽ .0001), and nPNA (P ⫽ .005) were statistically different between MNA groups. Fifty-two patients who had no evidence of malnutrition according to SGA were defined as having moderate malnutrition according to MNA. Seven patients who were in a state of moderate malnutrition determined by SGA were in good nutritional status according to MNA. SGA identified 8 patients as moderately malnourished; the same patients were defined as having severe malnutrition in MNA. Our results suggest that MNA might underestimate the nutritional status of hemodialysis patients who are not in an inflammatory state and may not be a reliable method for detecting moderate malnutrition when compared with SGA. © 2006 by the National Kidney Foundation, Inc.
P
ROTEIN-ENERGY malnutrition (PEM) is a commonly seen health problem in end stage renal disease (ESRD) patients. Various studies have shown signs of malnutrition in 23% to 76% of hemodialysis (HD) and 18% to 50% of peritoneal dialysis (PD) patients.1-4 There is strong association between nutritional status and clinical outcome in HD patients, providing support to the hypothesis that PEM may cause or contribute to morbidity and mortality.5 *Department of Nephrology, Baskent University Hospital, Ankara, Turkey. †Department of General Surgery, Baskent University Hospital, Ankara, Turkey. Address reprint requests to Baris Afsar, MD, 3. Cadde 50, Sokak 9/8 06500, Bahçelievler, Ankara, Turkey. E-mail: afsarbrs@yahoo. com © 2006 by the National Kidney Foundation, Inc. 1051-2276/06/1603-0021$32.00/0 doi:10.1053/j.jrn.2006.01.012
Journal of Renal Nutrition, Vol 16, No 3 ( July), 2006: pp 277-282
Assessment and monitoring of protein and energy statuses are essential to prevent, to diagnose, and to treat uremic malnutrition. Subjective Global Assessment (SGA) and Mini Nutritional Assessment (MNA) are two tools for monitoring PEM. The basis of SGA is to determine whether nutrient assimilation has been restricted because of decreased food intake, maldigestion, or malabsorption; whether any effects of malnutrition on organ function have occurred; and whether patients’ disease processes influence nutrient requirements.6 SGA has been used for nutritional assessment of chronic HD patients and is independently associated with mortality in this patient population.7 MNA is a simple, noninvasive, inexpensive, and easily preformed method used especially in the elderly population.8 The data about the use of MNA for determining PEM in ESRD patients is scarce. In this study, we attempted to determine the reliability of MNA in 277
278
AFSAR ET AL
detecting malnutrition in HD patients in comparison with SGA.
Methods This cross-sectional study was undertaken in the HD unit of Baskent University Ankara Hospital. The study received local ethical approval, and verbal informed consent was obtained from all patients before enrolment. All patients were receiving 4-hour HD 3 times weekly with standard bicarbonate dialysis (sodium 138 mmol/L, bicarbonate 35 mmol/L, potassium 1.5 mmol/L, calcium 1.25 mmol/L, and magnesium 1.8 mmol/L) by either high-flux HD and 1.1 to 1.7 m2 hollow fiber or flat plate dialyzer at the time the research was conducted. The group was composed of patients with pure PEM with no signs of chronic inflammation, defined as C-reactive protein concentrations ⬍10 mg/dL. Medical problems including any clinical events that could cause inflammation (such as infections, hepatitis, operations, trauma, nonspecific fever, and new cardiovascular events including acute coronary syndromes, cerebrovascular disease, and peripheric arterial disease) were noted, and patients with these conditions were excluded. Initially, 154 patients were included in the study. Five patients with chronic liver disease (4 hepatitis C, 1 hepatitis B), 2 with diabetic foot ulcers, 1 with bladder carcinoma, 1 with nonspecific fever, 1 with recent myocardial infarction, and 1 with recent history of trauma (traffic accident) were excluded from the study. Two patients receiving nutritional support were not included. Four patients refused to participate in the study. The final study population consisted of 56 (41%) female and 81 male (59%) patients (mean age, 41.4 ⫾ 12.1 years; mean HD duration, 72.4 ⫾ 38.1 months). The nutritional statuses of HD patients were assessed concomitantly using SGA and MNA tests. History used in SGA focused on 5 areas. Percentage of body weight loss in the previous 6 months was characterized as mild (⬍5%), moderate (5% to 10%), or severe (⬎10%). Dietary intake was classified as normal or abnormal as judged by a change in intake and whether the current diet was nutritionally adequate. Presence of gastrointestinal symptoms such as nausea, vomiting, diarrhea, and abdominal pain were recorded. A patient’s functional capacity was defined as bedridden, suboptimally active, or full
capacity. The last feature of history concerned metabolic demands of patients’ underlying disease states. In physical examination, loss of subcutaneous fat was measured in the triceps region and mid-axillary line at the level of the lower ribs. The second feature was muscle wasting in temporal areas and in deltoids and quadriceps, as determined by loss of bulk and tone detectible by palpation. Presence of edema in the ankle and sacral regions and the presence of ascites were noted. The findings of history and physical examination were used to categorize patients as well-nourished (no malnutrition), as having moderate malnutrition, or as having severe malnutrition.6 MNA was carried out using a MNA scale consisting of 18 point-weighted questions in 4 categories. The first part consisted of 4 anthropometric measurements including body mass index, mid-upper arm and calf circumferences, and weight loss during the past 3 months, followed by 6 global questions regarding accommodation type, pharmaceutical consumption, acute diseases (including psychological stress), mobility, neuropsychologic problems, and pressure sores/skin ulcers. The third part consisted of questions to assess dietary intake, including how many whole meals were eaten, food choice of the patient, fluid intake per day, and how much help was required during meals. The final part of the scale consisted of 2 self-assessments of whether food intake was sufficient and of their own health status. The answer point scale gave a maximum of 30 points. Less than 17 points indicated malnutrition, 17 to 23.5 points indicated a risk for malnutrition, and ⱖ24 points indicated that the person was wellnourished.9 All patients were adequately dialyzed, expressed as equilibrated spKt/V. Urea kinetic modeling was performed to assess the delivered dose of dialysis. HD dose was evaluated using the following formula: spKt ⁄ V⫽ ⫺Ln 共R ⫺ 0.008 ⫻ t兲 ⫹ 共4 ⫺ 关3.5 ⫻ R兴 ⫻ UF ⁄ W兲 where spKt/V is a single-pool Kt/V, R is the ratio of postdialysis to predialysis serum urea nitrogen, t is time on dialysis in hours, UF is the amount of ultrafiltration in liters, and W is postdialysis body weight in kilograms.
279
RELIABILITY OF MNA AND SGA
The normalized protein equivalent of total nitrogen appearance (nPNA), also known as normalized protein catabolic rate (nPCR), was calculated to estimate the daily protein intake by the following formula: nPNA 共g/kg/d兲 ⫽ C/关36.3 ⫹ 5.48 共spKt/V兲 ⫹ 53.5/共spKt/V兲兴 ⫹ 0.168 where C is the predialysis concentration of serum urea nitrogen in milligrams per deciliter. Body mass index was calculated as the ratio of dry weight in kilograms (end-dialysis weight) to height squared (in square meters). The mean laboratory parameters of the last 6 months, including serum hemoglobin, albumin, prealbumin, C-reactive protein, predialysis calcium and phosphorus, predialysis blood urea nitrogen and creatinine, total cholesterol, high-density lipoprotein, low-density lipoprotein, and triglycerides, were measured before beginning the dialysis session. Postdialysis serum urea nitrogen concentrations used to calculate the urea reduction ratio were also measured. According to the results of SGA, the patients are classified into 3 SGA groups, defined as SGA-A, SGA-B, and SGA-C. The groups were composed of patients who were well-nourished (no malnutrition), moderately malnourished, and severely malnourished, respectively. Based on MNA results, the patients were also divided into 3 groups defined as patients with no malnutrition, MNA-1; patients with malnutrition risk or moderate malnutrition, MNA-2; and patients with severe malnutrition, MNA-3.
Table 1. Demographic Characteristics of the Patients
*Mean ⫾ standard deviation.
SGA groups SGA-A (n ⫽ 92, 67.2%) SGA-B (n ⫽ 40, 29.2%) SGA-C (n ⫽ 5, 3.6%) MNA groups MNA-1 (n ⫽ 47, 34.3%) MNA-2 (n ⫽ 77, 56.2%) MNA-3 (n ⫽ 13, 9.5%)
Age (y)
HD Duration (mo)
P
41.5 ⫾ 12.4
74.0 ⫾ 37.7
NS*
42.2 ⫾ 11.8
66.1 ⫾ 36.2
NS
33.6 ⫾ 2.0
80.2 ⫾ 12.3
NS
43.0 ⫾ 11.7
73.6 ⫾ 39.7
NS
40.7 ⫾ 13.0
70.2 ⫾ 37.7
NS
39.5 ⫾ 6.6
75.3 ⫾ 13.5
NS
*NS, not significant (P ⬎ .05).
Statistical Analysis All values are expressed as mean ⫾ SD and percent. The data were analyzed using the program SPSS 11.0 for Windows (SPSS Inc, Chicago, IL). In statistical analysis, SGA and MNA were considered as categorical variables. The within-group parameters were analyzed by Kolmogorov-Smirnov with a Lilliefors significance level and Shapiro-Wilk statistics for normality. Normally distributed continuous parameters between 3 groups were analyzed by one-way analysis of variance, and between 2 groups were analyzed by the Student t test. Parameters that were not normally distributed between the 3 groups were compared by the Kruskal-Wallis test and between 2 groups by the Mann-Whitney U test. A P value ⬍ 0.05 was accepted as statistically significant.
Results
Parameters Age (mean, y) Male/female (n) Dialysis duration (mean, mo) Etiologies (n) Nephrolithiasis Diabetes mellitus Glomerulonephritis Amyloidosis Hypertension Unknown
Table 2. Classification of Patients According to SGA and MNA
41.4 ⫾ 12.1* 81/56 72.4 ⫾ 38.1* 137 63 20 13 11 10 20
The patient characteristics, including etiology of renal failure, are summarized in Table 1. To evaluate the degree of reproducibility, the same physician repeated the SGA and MNA after 10 days on a subset of 15 patients without reference to the first SGA and MNA evaluations. The reliability coefficients (alpha) for between 2 SGA and MNA assessments were 0.91 and 0.93, respectively, representing a good degree of reproducibility.
280
AFSAR ET AL
Table 3. Comparison of 3 SGA Groups SGA Groups Hemoglobin (g/dL) Creatinine (mg/dL) BUN (mg/dL) Albumin (g/dL) Prealbumin (mg/dL) Cholesterol (mg/dL) Triglyceride (mg/dL) Phosphorus (mg/dL) BMI spKt/V nPNA
SGA-A (n ⫽ 92)
SGA-B (n ⫽ 40)
SGA-C (n ⫽ 5)
P
11.3 ⫾ 0.9 10.8 ⫾ 2.1 79.1 ⫾ 13.0 4.2 ⫾ 0.3 36.2 ⫾ 4.7 162.9 ⫾ 33.3 152.2 ⫾ 57.6 5.4 ⫾ 1.3 22.8 ⫾ 3.4 1.1 ⫾ 0.2 1.0 ⫾ 0.1
11.3 ⫾ 1.1 9.6 ⫾ 2.7 80.3 ⫾ 14.5 3.7 ⫾ 0.3 32.2 ⫾ 4.1 155.8 ⫾ 27.1 133.3 ⫾ 43.9 5.3 ⫾ 1.3 21.2 ⫾ 3.6 1.1 ⫾ 0.1 1.0 ⫾ 0.1
10.7 ⫾ 0.5 7.2 ⫾ 0.3 68.6 ⫾ 3.9 3.4 ⫾ 0.1 29.8 ⫾ 4.9 148.0 ⫾ 10.6 132.2 ⫾ 4.4 4.6 ⫾ 0.6 19.4 ⫾ 1.1 1.1 ⫾ 0.1 0.8 ⫾ 0.2
.3 .0001* .2 .0001* .0001* .3 .1 .4 .01* .6 .04*
*Kruskal-Wallis and one-way ANOVA.
According to SGA results, 92 patients (67.2%) were in SGA-A, 40 patients (29.2%) were in SGA-B, and 5 patients (3.6%) were in SGA-C. There were no statistically significant differences between ages and HD durations of SGA groups (Table 2). Forty-seven patients (34.3%) were in MNA-1, 77 patients (56.2%) were in MNA-2, and 13 patients (9.5%) were in MNA-3; the patients were also similar in terms of mean ages and HD durations (Table 2). Comparison of 3 SGA groups showed that the albumin (P ⫽ .0001), prealbumin (P ⫽ .0001), body mass index (P ⫽ .01), creatinine (P ⫽ .0001), and nPNA (P ⫽ .04) of the patients were statistically different between SGA groups. Other parameters, including spKt/V, did not reach statistical significance (P ⬎ .05) (Table 3). We combined the small number of severely malnourished patients in SGA-C (5 patients) with the moderately malnourished SGA-B (n ⫽ 40) patients and reanalyzed the parameters between SGA-A and SGA-B and C. The two groups still differed in terms of albumin (P ⫽ .0001), prealbumin (P ⫽ .0001), body mass index (P ⫽ .01),
and creatinine (P ⫽ .006), but the difference for nPNA disappeared (P ⫽ .4) (data not shown). When we compared 3 MNA groups, we found that creatinine (P ⫽ .001), blood urea nitrogen (P ⫽ .017), albumin (P ⫽ .001), prealbumin (P ⫽ .005), body mass index (P ⫽ .0001), and nPNA (P ⫽ .005) were statistically different (Table 4). The remaining parameters were not statistically different (P ⬎ .05). We combined the small number of severely malnourished patients in MNA-3 (13 patients) and moderately malnourished patients in MNA-2 (n ⫽ 77) and reanalyzed the parameters between MNA-1 and MNA-2 and 3; body mass index (P ⫽ .0001) and creatinine (P ⫽ .04) differed statistically between MNA-1 and MNA-2 and 3, and the differences between albumin (P ⫽ .17), prealbumin (P ⫽ .42), blood urea nitrogen (P ⫽ .6), and nPNA (P ⫽ .2) had disappeared (data not shown). Our data showed that the MNA and SGA results regarding patients’ nutritional statuses were inconsistent with each other. Fifty-two patients who had no evidence of malnutrition according to SGA were defined as
Table 4. Comparison of 3 MNA Groups MNA Groups Hemoglobin (g/dL) Creatinine (mg/dL) BUN (mg/dL) Albumin (g/dL) Prealbumin (mg/dL) Cholesterol (mg/dL) Triglyceride (mg/dL) Phosphorus (mg/dL) BMI spKt/V nPNA
MNA-1
MNA-2
MNA-3
P
11.4 ⫾ 0.9 11.0 ⫾ 2.0 79.8 ⫾ 11.3 4.1 ⫾ 0.4 35.4 ⫾ 5.0 158.9 ⫾ 32.0 159.7 ⫾ 60.2 5.2 ⫾ 1.1 24.5 ⫾ 3.2 1.0 ⫾ 0.2 1.0 ⫾ 0.1
11.2 ⫾ 1.0 10.5 ⫾ 2.3 80.2 ⫾ 14.7 4.0 ⫾ 0.3 35.4 ⫾ 4.5 161.3 ⫾ 32.7 140.1 ⫾ 51.5 5.5 ⫾ 1.4 21.4 ⫾ 3.2 1.1 ⫾ 0.1 1.0 ⫾ 0.2
11.2 ⫾ 0.9 7.1 ⫾ 1.2 69.0 ⫾ 6.2 3.6 ⫾ 0.2 30.6 ⫾ 5.5 158.9 ⫾ 16.9 131.0 ⫾ 21.4 4.7 ⫾ 0.8 18.9 ⫾ 1.5 1.1 ⫾ 0.1 0.9 ⫾ 0.09
.57 .001* .017* .001* .005* .9 .1 .08 .0001* .2 .005*
*Kruskal-Wallis and one-way ANOVA.
RELIABILITY OF MNA AND SGA Table 5. Cross Table of Nutritional Statuses of the Patients According to SGA and MNA SGA Groups (n)
MNA groups (n) MNA-1 MNA-2 MNA-3 Total
SGA-A
SGA-B
SGA-C
Total
40 52 0 92
7 25 8 40
0 0 5 5
47 77 13 137
having moderate malnutrition according to MNA. Seven patients who were in a state of moderate malnutrition determined by SGA were in good nutritional status according to MNA. SGA identified 8 patients as moderately malnourished, whereas the same patients were defined as having severe malnutrition in MNA. Only 5 patients were found to have severe malnutrition by both SGA and MNA (Table 5).
Discussion In this study, we found that MNA may not be a reliable method for detecting moderate malnutrition in a subset of HD patients who are not in an inflammatory state when compared with SGA. In our opinion, MNA might underestimate the nutritional status of HD patients by identifying them as having moderate malnutrition, although they are in good nutritional status according to SGA. PEM is one of the major predictors of poor clinical outcome in the HD population.10 A variety of factors may cause or contribute to the development of malnutrition in renal failure, including inadequate nutrient intake, inadequate dialysis, loss of proteins and amino acids during dialysis, increased catabolism, metabolic acidosis, and hormonal and metabolic disturbances. So far, we do not have a single definite test by which to assess nutritional status in ESRD patients. SGA was originally developed to identify poor nutritional status in subjects undergoing gastrointestinal surgery, but because the technique of SGA allows rapid, easy, and equipment-free scoring of nutritional status, it has been used widely to quantify the prevalence of malnutrition in HD and PD patients.11-14 SGA has a high predictive value for determining mortality in ESRD patients.15,16 SGA is accepted as a reliable method for detecting malnutrition in ESRD patients and is recommended in screening for malnutrition
281
according to the United States, United Kingdom, and European guidelines.17-19 MNA is another method of detecting malnutrition. The MNA scores correlate with nutritional intake and anthropometric and biological nutritional parameters, and may predict mortality in geriatric patients.20,21 To our knowledge, there exists no study in the literature regarding the use of MNA for the assessment of the nutritional status of ESRD patients. In our study, we attempted to analyze the reliability of MNA in detecting PEM when compared with SGA. The reliability of SGA in showing the nutritional status of HD patients in our study was reinforced by its compatibility with the anthropometric (body mass index) and biochemical nutritional parameters (albumin, prealbumin, creatinine) as well as nPNA in well-nourished and moderately and severely malnourished patients. When the small number of severely malnourished patients in SGA-C was combined with moderately malnourished patients in SGA-B (SGA-B and C) and was compared with wellnourished patients (SGA-A), except for nPNA, the difference between both anthropometric and biochemical nutritional parameters remained significant, showing that SGA readily distinguishes malnourished patients (SGA-B and C) from wellnourished patients (SGA-A). When the small number of severely malnourished patients in MNA-3 was combined with moderately malnourished patients in MNA-2 (MNA-2 and 3) and was compared with well-nourished patients (MNA-1), we found that although the correlation of MNA scores with anthropometric parameters was sustained, the correlation with biochemical nutritional parameters and nPNA were lost. In other words, although MNA distinguished severely malnourished patients from others, unlike SGA, the method failed to identify patients with malnutrition (MNA-2 and 3) before severe malnutrition developed. In our study, we found that 52 patients who had no malnutrition according to SGA scores were identified as having malnutrition according to MNA (Table 5). This misinterpretation might result in unnecessary nutritional intervention in this group of patients, who were assessed as having malnutrition according to MNA, but in reality were not malnourished. Conversely, 7 patients who were in a state of malnutrition according to SGA were regarded as having PEM by MNA. This might
282
AFSAR ET AL
lead to underdiagnosis of truly malnourished patients, causing a delay of necessary interventions and resulting in deterioration of the health status. The compliance of MNA scores with SGA scores in detecting the severely malnourished patients is not considered to be a great concern because this group of patients can be diagnosed easily by medical stuff even without any method. There may be reasons for MNA to not to detect nutritional status correctly in dialysis patients. The questions in MNA are mostly subjective in nature. The patients may not correctly remember the answers of the questions regarding the last 3 months, may be reluctant to answer the questions, or the answers may be blunted because of neuropsychiatric and intellectual problems. Because MNA is a method normally used in patients without any dietary restrictions, ESRD patients may not be correctly assessed because of their dietary modifications. There are not enough data in the literature regarding the use and reliability of MNA in determining nutritional status in HD patients. As far as we know, this study is the first in the literature in which SGA and MNA were concomitantly performed and the reliability of MNA was tested in the assessment of nutritional status in the HD population. In conclusion, we suggest that MNA may not be as reliable as SGA in detecting PEM in the HD population.
References 1. Cianciaruso B, Brunori G, Kopple JD, et al: Cross-sectional comparison of malnutrition in continuous ambulatory peritoneal dialysis and hemodialysis patients. Am J Kidney Dis 26:475-486,1995 2. Marckmann P: Nutritional status of patients on hemodialysis and peritoneal dialysis. Clin Nephrol 29:75-78, 1988 3. Bergstrom J, Lindholm B: Nutrition and adequacy of dialysis. How do hemodialysis and CAPD patients compare? Kidney Int Suppl 40:S39-50, 1993 4. Qureshi AR, Alvestrand A, Danielsson A, et al: Factors predicting malnutrition in hemodialysis patients. A cross-sectional study. Kidney Int 53:773-782, 1998 5. Hakim RM, Levin N: Malnutrition in hemodialysis patients. Am J Kidney Dis 21:125-137, 1993
6. Pupim LB, Ikizler TA: Assessment and monitoring of uremic malnutrition. J Ren Nutr 14:6-19, 2004 7. Pifer TB, McCullough KP, Port FK, et al: Mortality risk in hemodialysis patients and changes in nutritional indicators: DOPPS. Kidney Int 62:2238-2245, 2002 8. Guigoz Y, Lauque S, Vellas BJ: Identifying the elderly at risk for malnutrition. The Mini Nutritional Assessment. Clin Geriatr Med 18:737-757, 2002 9. Saletti A, Lindgren EY, Johansson L, et al: Nutritional status according to Mini Nutritional Assessment in an institutionalized elderly population in Sweden. Gerontology 46:139-145, 2000 10. Lowrie EG, Huang WH, Lew NL, et al: The relative contribution of measured variables to death risk among hemodialysis patients, in Friedman EADeath on Hemodialysis. Amsterdam, Kluwer Academic, 1994, pp 121–141 11. Detsky AS, Baker JP, O’Rourke K, et al: Predicting nutrition-associated complications for patients undergoing gastrointestinal surgery. J Parenter Enteral Nutr 11:440-446, 1987 12. Enia G, Sicuso C, Alati G, et al: Subjective global assessment of nutrition in dialysis patients. Nephrol Dial Transplant 8:1094-1098, 1993 13. Kawaguchi Y, Sugino N, Arai J, et al: Nutritional assessment of patients on continuous ambulatory peritoneal dialysis. Nippon Jinzo Gakkai Shi 35:843-851, 1993 14. Young GA, Kopple JD, Lindholm B, et al: Nutritional assessment of continuous ambulatory peritoneal dialysis patients: An international study. Am J Kidney Dis 17:462-471, 1991 15. McCusker FX, Teehan BP, Thorpe KE, et al: How much peritoneal dialysis is required for the maintenance of a good nutritional state? Canada-USA (CANUSA) Peritoneal Dialysis Study Group. Kidney Int Suppl 56:S56-61, 1996 16. Lawson JA, Lazarus R, Kelly JJ: Prevalence and prognostic significance of malnutrition in chronic renal insufficiency. J Ren Nutr 11:16-22, 2000 17. K/DOQI, National Kidney Foundation: Clinical practice guidelines for nutrition in chronic renal failure. Am J Kidney Dis 35:S1-140, 2000 (suppl 2) 18. Treatment of Adults and Children With Renal Failure– Standards Document (ed 3). London, England, Royal Collage of Physicians of London/Renal Association, 2002 19. Locatelli F, Fouque D, Heimburger O, et al: Nutritional status in dialysis patients: A European consensus. Nephrol Dial Transplant 17:563-572, 2002 20. Persson MD, Brismar KE, Katzarski KS, et al: Nutritional status using Mini Nutritional Assessment and Subjective Global Assessment predict mortality in geriatric patients. J Am Geriatr Soc 50:1996-2002, 2002 21. Vellas B, Guigoz Y, Baumgartner M, et al: Relationships between nutritional markers and the mini-nutritional assessment in 155 older persons. J Am Geriatr Soc 48:13001309, 2000