Evaluation of Nutritional Status in Patients with Kidney Disease: Usefulness of Dietary Recall

Philippe Chauveau, MD,* Emmanuelle Grigaut, RD,* Anne Kolko, MD, Patricia Wolff, RD,* Christian Combe, MD,* and Michel Aparicio, MD*
Background: Three-day food recall and normalized protein nitrogen appearance calculation from pre- and postdialysis plasma urea are the most commonly used techniques to assess nutritional intake, but a 7-day dietary recall is probably more accurate to approach dietary intake in clinical practice. Methods: A total of 99 hemodialyzed patients from two units were analyzed in a 7-day dietary record with a large range of age and without having any signs of malnutrition. Dietary protein intake was estimated from the recall and calculated (normalized protein catabolic rate) from urea kinetic modeling. Calorie intake and quality and repartition of nutrients were estimated from diaries. Results: Repartition of nutrients was close to that of a reference population except for a lower glucidic contribution (glucide 47%, lipid 36%, protein 16%). Normalized protein catabolic rate and dietary protein intake were well correlated (R2 0.4), but a large variability existed from day to day, according to age (older patients are less variable) and day of dialysis (long or short interval). Conclusion: A large variation in alimentary intake exists from patient to patient and day to day. A 7-day evaluation of nutrient intake, dialysis adequacy, and nutritional parameters seems to be a good solution to guide dietetic counseling. 2007 by the National Kidney Foundation, Inc.

ALNUTRITION, which is highly prevalent in patients with kidney disease, is associated with poor clinical outcomes.1 Malnutrition is primarily caused by an inadequate balance between lower dietary intake and higher needs. The rst cause of lower nutrients intake is progressive anorexia, which is associated with the progression of renal failure.2 Other causes worsen the decrease in nutritional intake, such as a large number of medications, gastropathy, psychosocial factors, infection, and inammation. In patients on dialysis, despite the correction of numerous metabolic disorders related to chronic renal failure, the negative balance tends to worsen. Food intake decreases in relation to inadequate dialysis, postdialysis

*From the Dpartement de Nphrologie et de Transplantation Rnale, Centre Hospitalier Universitaire, Bordeaux, France. Service de Nphrologie, Hpital Foch, Suresnes, France. Address reprint requests to Philippe Chauveau, MD, Dpartement de Nphrologie, Hpital Pellegrin, 33076 Bordeaux, France. E-mail: ph.chauveau@wanadoo.fr 2007 by the National Kidney Foundation, Inc. 1051-2276/07/1701-0017$32.00/0 doi:10.1053/j.jrn.2006.10.015

fatigue, hypotension, dry mouth, and ageusia. The reduction in appetite is in relation to inammation markers.3 Poor appetite is also associated with a higher morbidity and mortality rate.3 On the other hand, protein requirements are higher because of the loss in nutrients and amino acids, loss of protein in peritoneal dialysis, inammation in relation to uid or dialysate or bioincompatible membrane, infection of vascular access, and peritonitis.4 As in the general population, the evaluation of nutritional status requires more than one marker. It is recommended to associate body composition assessment with biochemical markers and evaluation of food intake.5 Associated with weight, height, and body mass index calculation, the assessment of body composition could be easily performed by anthropometry and more accurately by bioelectrical impedance analysis, dual x-ray absorptiometry, or near-infrared reactance. Serum levels of albumin and prealbumin are the biochemical markers most often used to assess visceral protein stores, whereas changes in serum creatinine over time may indicate a change in
Journal of Renal Nutrition, Vol 17, No 1 ( January), 2007: pp 88-92

88

DIETARY RECALL IN CKD

89

muscle mass, which is a good indicator of somatic protein reserves. The assessment of energy and protein intake is the third part of a standard evaluation. Food record and 24-hour food recall are the most commonly used methods, but several concerns have been raised about the validity and usefulness of these short-duration dietary reports. This is a time-consuming method, nutrient and uid intakes are underestimated, and energy intake is underestimated when compared with energy expenditure from 10% to 20%.6 Underreporting of protein intakes is generally observed in the general population. Last, food intake varies widely with time, and the coefcient of variation is close to 20% in a 3-day recall and much higher if a 1-day technique or a food frequency questionnaire is used.7 The evaluation of protein intake through nitrogen balance is accurate and easier to perform. Total nitrogen appearance is highly correlated with urea nitrogen appearance, and formulas based on urea kinetic modeling have been developed in patients with CKD, as in patients on hemodialysis or peritoneal dialysis. These formulas have been validated in clinically stable patients with neutral nitrogen balance. The aim of this article is to demonstrate the usefulness and validity of a 7-day dietary report in clinical practice in a large population of patients on hemodialysis.

Table 1. Nutrient Intake of 99 Patients on Hemodialysis Compared With the French Population
Patients on Hemodialysis Calories: Kcal/d Male Female Protein (%) Lipid (%) Glucide (%) 1789 1979 1572 16 36 47 408 370 342 3 6 7 French Population

2400 1780 14%18% 38%40% 39% 41% RDA 50%

RDA, recommended dietary allowance.

each dialysis session of the week, the report was reviewed with a dietitian. Quantities were estimated using the SU-VI-MAX8 manual, which contains photos of portion size, or standard units. Data were computed (Bilnut 4.0 SCDA Nutrisoft, Le Hallier, Cerelles, France) using the French database by the same dietitian.

Biology Pre- and postdialysis blood samples were drawn using National Kidney Foundation-Kidney Disease Outcomes Quality Initiative procedure guidelines at each session of the week. Dialysis adequacy was estimated by urea Kt/V. Normalized protein nitrogen appearance (nPNA) was calculated from the urea generation rate after measurements of pre- and postdialysis plasma urea. Statistical Analysis All statistical analysis was performed with Stat View version 5 for Windows (Abacus Concept, Berkeley, CA). Data are expressed as means standard deviation. Unpaired data were analyzed with the Mann-Whitney U test. Simple correlations were studied using the Spearman rank test. A P value of .05 or less was considered signicant.

Methods
Patients Ninety-nine patients from two hemodialysis centers and one self-dialysis unit were included in this study. The mean age was 45 19 years (range 14-84 years), 54 were male, and the dialysis vintage was 97 88 months. The mean weight was 59 13 kg, and body mass index was 21 3 kg/m2. For most of the patients, dialysis duration was 4 hours three times per week using highly biocompatible membranes. The mean albumin level was 42 4 g/L, and plasma bicarbonate was 22 3 mmol/L. All patients were able to cook and report all their meals. Dietary Recall After careful instructions in each site, patients recorded a diary every day of the week. During

Results
The mean daily global nutrients intakes are reported in Table 1 and are compared with estimated values for the general French population reported in a national governmental program (The Unit of Nutritional Surveillance and Epidemiology, 1997-2003, www.invs.sante.fr). Calorie intake in patients on hemodialysis was

90

CHAUVEAU ET AL
1.3

Calories Kcal/kg/d

Dpi gram/kg/d

33 31 29 27 25 <40 >40-65

1.2 1.1 1 .9 <40 >40-65

65

65

Figure 1. Mean daily calorie (Kcal/kg/d) and protein (g/kg/d) intakes according to age. Dpi, dietary protein intake.

lower than in the French population. Repartition of nutrients was close to what is reported in the general population, but carbohydrate intake was lower than generally recommended. The mean calorie intake was 29.8 7.5 Kcal/ kg/d, and the mean protein intake was 1.18 0.28 g/kg/d. When patients were stratied according to age ( 40 years, 41-65 years, and 65 years), calorie and protein intake were signicantly lower in the older group (27 5 Kcal/kg/d and 1.05 0.2 g/kg/d, respectively) and below the usual recommendations (Fig. 1). The mean Kt/V was 1.56 0.32 units, and no correlation existed between dialysis dose and nutrient intake. Variation from day to day was wide and higher than in the general population. The mean daily intake and coefcient of variation were reported in Table 2 for nutrients and some oligo-elements. Daily variations in food intake are illustrated in Figure 2, which shows, in a 75-year-old man, a signicant decrease in nutrient intake on every dialysis day compared with nondialysis days. As previously reported, there is a good correlation between nPNA calculation and dietary protein intake (Dpi) (Fig. 3). A better correlation exists between Dpi and PNA, the third
Table 2. Mean Daily Nutrient Intake of Seven-Day Dietary Recall and Mean Coefcient of Variation per Patient
Mean Daily Intake Sodium mg/d Potassium, mg/d Phosphate, mg/d Protein, g/d Lipid, g/d Glucide, g/d 1900 1926 860 68 68 200 990 663 300 23 27 64 Mean Coefcient (min-max) 52% (7%80%) 34% (6%55%) 35% (5%58%) 34% (6%58%) 40% (4%100%) 32% (7%77%)

session of the week compared with midweek session, or the rst session after a 3-day interval. The most valuable explanation is that the patient gives a better estimation of the global intake.

Discussion
The current study conrms that food intake is reduced in patients on maintenance hemodialysis treatment; this reduction is more pronounced in older patients. The mean calorie and protein intakes reported in our patients is signicantly higher than those reported in others series, notably from the United States. In the Hemodialysis (HEMO) study the mean calorie intake at baseline was 22.7 Kcal/kg versus 29.8 Kcal/kg. Our patients were younger (45 vs. 56 years) and thinner (59 vs. 69 Kg).9 Moreover, large differences in both the type and amount of food ingested were observed from day to day, particularly from dialysis to nondialysis days. Again, the same observation was made in the analysis of the rst 1900 patients included in the HEMO study; the mean protein intake was 5 g less (10%) on the nondialysis day.10 The results justify that the recommendation of a food record exceeding

Figure 2. Individual example of energy daily intake in a 75-year-old man weighing 75 kg. There is a signicant decrease of calories intake on the day of dialysis. SD, standard deviation.

DIETARY RECALL IN CKD

91

Figure 3. Correlation between diet recalls (2 or 3 days) and protein nitrogen appearance calculation. A better correlation exists between the last session urea kinetic modeling calculation and 2-day recall. Dpi, dietary protein intake; PNA, protein nitrogen appearance.

the minimal time of 3 days is needed for data collection. In hemodialyzed patients, protein and energy requirements are not different from normal, but food intake is most often below the usual recommendations even in stable patients receiving an appropriate dialysis prescription and not presenting obvious signs of malnutrition. Although the emphasis has been put on protein deciency, we have observed in our patients, as already reported in some other series, that low caloric energy intake seemed to be more common and severe than decreased protein intake.11 This is an important point because an adequate energy intake is necessary to ensure the maintenance of a neutral or positive nitrogen balance. Food intake was more reduced in the elderly patients than in the younger patients, as reported in other series of patients on hemodialysis11,12 and in other hospitalized populations. The reduction is favored by a higher incidence of concomitant illnesses, the socioeconomic and psychologic factors, and the physiologic deciencies related to aging. Moreover, intradialytic complications (especially hypotension) may contribute to the reduced alimentary intake on dialysis days. Conversely, some elderly, especially those living alone, increase their food intake on dialysis days when they meet regularly with the staff and other patients. A good correlation exists between nPNA calculation and Dpi. In a general population, a good agreement is generally found between the estimation of nitrogen excretion and dietary report, even in a long period of 1 week to 1 month. An underestimation of calorie intake, but not protein intake, was reported Saturday and Sunday.7 We found a better correlation

between nPNA and Dpi the last 2 days of the week than the 3-day interval or midweek session. This is probably because of the conjunction of many factors: a less accurate estimation of nitrogen balance with the formula used for 3 days, an underestimation of food intake on the weekend, and a better understanding of compliance to the food recall after 3 to 5 days. All of these points and the wide variations of alimentary intake from day to day (weekends and weekdays, dialysis and nondialysis days) justify the recommendation of a 7-day food record to reliably evaluate qualitative and quantitative aspects of food intake.

References
1. Heimburger O, Qureshi AR, Blaner WS, et al: Handgrip muscle strength, lean body mass, and plasma proteins as markers of nutritional status in patients with chronic renal failure close to start of dialysis therapy. Am J Kidney Dis 36:1213-1225, 2000 2. Kopple JD, Greene T, Chumlea WC, et al: Relationship between nutritional status and the glomerular ltration rate: results from the MDRD study. Kidney Int 57:1688-1703, 2000 3. Kalantar-Zadeh K, Block G, McAllister CJ, et al: Appetite and inammation, nutrition, anemia, and clinical outcome in hemodialysis patients. Am J Clin Nutr 80:299-307, 2004 4. Bossola M, Muscaritoli M, Tazza L, et al: Malnutrition in hemodialysis patients: what therapy? Am J Kidney Dis 46:371386, 2005 5. Clinical practice guidelines for nutrition in chronic renal failure. K/DOQI, National Kidney Foundation. Am J Kidney Dis 35:S1-140, 2000 6. Livingstone MB, Black AE: Markers of the validity of reported energy intake. J Nutr 133(Suppl 3):895S-920S, 2003 7. van Staveren WA, de Boer JO, Burema J: Validity and reproducibility of a dietary history method estimating the usual food intake during one month. Am J Clin Nutr 42:554-559, 1985 8. Hercberg S, Deheeger M, Preziosi P: SU.VI.MAX study coordinators, Portions Alimentaires, Paris, France, Polytechnica

92

CHAUVEAU ET AL
11. Lorenzo V, de Bonis E, Runo M, et al: Caloric rather than protein deciency predominates in stable chronic haemodialysis patients. Nephrol Dial Transplant 10:1885-1889, 1995 12. Movilli E, Mombelloni S, Gaggiotti M, et al: Effect of age on protein catabolic rate, morbidity, and mortality in uraemic patients with adequate dialysis. Nephrol Dial Transplant 8:735739, 1993

9. Burrowes JD, Larive B, Chertow GM, et al: Self-reported appetite, hospitalization and death in haemodialysis patients: ndings from the Hemodialysis (HEMO) Study. Nephrol Dial Transplant 20:2765-2774, 2005 10. Burrowes JD, Larive B, Cockram DB, et al: Effects of dietary intake, appetite, and eating habits on dialysis and nondialysis treatment days in hemodialysis patients: cross-sectional results from the HEMO study. J Ren Nutr 13:191-198, 2003