ORIGINAL RESEARCH

The Phase Angle and Mass BodyCell as Markers of Nutritional Statusin Hemodialysis PatientsClaudia M. C. Oliveira, Ms,*† Marcos Kubrusly, PhD,* Rosa S. Mota, Ms,*

Carlos A. B. Silva, PhD,†§ Gabriel Choukroun, PhD,‡ and Valzimeire N. Oliveira, PhD*

Objective: Bioelectrical impedance analysis (BIA) is a fast, noninvasive method for assessing body composition,

and its role in the evaluation of nutritional status in haemodialysis (HD) has been studied. This study aimed to com-

pare BIA parameters to clinical, biochemical, and anthropometric markers of nutrition in HD patients, such as sub-

jective global assessment modified for renal disease (SGA-1), serum albumin, body mass index (BMI), percent of

standard body weight (%SBW), deviation of triceps skinfold thickness (TSF), mid-arm circumference (MAC), mid-

arm muscle circumference (MAMC) from the standard value (50th percentile), anthropometry-derived fat (FM-A),

and fat-free mass (FFM-A).

Methods: BIA was performed 30 minutes after a HD session and the reactance (Xc), resistance (R), phase angle

(PA), body cell mass (BCM), fat mass (FM-BIA), and fat-free mass (FFM-BIA) values were recorded. The prevalence

of malnutrition was estimated according to PA and percent of BCM. The correlation between methods was assessed

through Pearson’s correlation coefficient and Bland and Altman analysis.

Results: A total of 58 patients were studied (30 women and 28 men; mean age of 49.2 6 14.8 years). The mean PA

was 6.19 6 1.33 degrees and the mean percent of BCM was 33.75 6 5.91%. The prevalence of malnutrition was

17.5% and 43.9% according to the PA and percent of BCM, respectively. PA had a negative correlation with age

and SGA-1 score and a positive correlation with percent SBW, MAC, MAMC, FFM-A, and albumin. Percent of

BCM had a negative correlation with age, MAC, MAMC and FM-A and a positive correlation with FFM-A and albumin.

A significant correlation between FFM-A and FFM-BIA was observed, as well as between FM-A and FM-BIA.

Conclusions: BIA indexes reflected nutritional state. PA and BCM seem to be less influenced by changes in vol-

ume and can be used for nutritional assessments of dialysis patients.

� 2010 by the National Kidney Foundation, Inc. All rights reserved.

MALNUTRITION is common in chronicdialysis patients; it affects approximately

one third as many hemodialysis (HD) patients asperitoneal dialysis (PD) patients. Several studiesshow evidence of malnutrition in 23% to 76% of

*Ceara State University, Fortaleza, Brazil.

†Federal University of Rio Grande do Norte, Natal, Brazil.

‡Center Hospitalier Universitaire, Amiens, France.

§University of Fortaleza, Fortaleza, Brazil.

Address reprint requests to Carlos Antonio Bruno da Silva, Uni-

versity of Fortaleza, Public Health Post-graduate Program, 1321,

Washington Soares Av, Block S-S01, Fortaleza, Ceara, Brazil

60.811-905. E-mail: carlosbruno@unifor.br and carlosbruno@secrel.com.br (C.A.B.S.)

� 2010 by the National Kidney Foundation, Inc. All rights

reserved.

1051-2276/$36.00

doi:10.1053/j.jrn.2010.01.008

314 Journal o

HD patients and in 18% to 50% of patients onPD.1–3 According to Kopple,4 one third of dialysispatients have mild-to-moderate malnutrition and6% to 8% are severely affected. The wide variationin the prevalence of malnutrition in patients onHD can be attributed to different evaluationmethods as well as to a range of factors contribut-ing to the development of the condition.

Malnutrition in cases of chronic kidney failure(CKF) may be secondary to poor nutritional in-take, severe dietary restrictions, hormonal and gas-trointestinal disorders, metabolic acidosis, drugsthat interfere with absorption of food, concurrentdisease, loss of nutrients during dialysis treatment,and increase in catabolism.5

Dialysis patients should be given regular nutri-tional evaluations to prevent, diagnose, and treat

f Renal Nutrition, Vol 20, No 5 (September), 2010: pp 314–320

MARKERS OF NUTRITIONAL STATUS IN HEMODIALYSIS PATIENTS 315

protein-caloric malnutrition. Early identificationand treatment of nutritional deficits can reducethe risk of infection and other complications andthus help prevent mortality.

There is no single best method or gold standardfor the nutritional evaluation of uremic patients,and multiple methods are usually employed. Inthe clinical setting, the methods of nutritionalevaluation used may be subjective (clinical historyand nutritional physical examination) or objective(food diary, anthropometry, biochemical studies,and bioelectrical impedance analysis [BIA]).

BIA is a fast, noninvasive, painless, relatively in-expensive, and reproducible method of evaluatingbody compartments. The BIA is based on theprinciple that different body components offer dif-ferent resistance to an electrical current. Lean tis-sues are good conductors of electric current dueto large amounts of water and electrolytes, whichprovide a low resistance to the passage of electricalcurrent. The fat, bone, and skin have low conduc-tivity and high resistance. It is also possible to cal-culate the phase angle (PA) and the body cell mass(BCM), both of which have been used as nutri-tional markers.6

The PA is derived from the arctangent of the re-actance and resistance and indicates changes in theintegrity of cell membranes and intercellularspaces.7 Smaller PAs appear to be consistent withlow reactance, cell death, or disruption of thecell membrane’s selective permeability. Larger PAappears to be more compatible with high reac-tance and larger quantities of intact cellmembranes.8

The BCM is a marker for the combination ofvisceral and somatic protein deposits. The bodymass can be divided into two components: thefat mass (FM) and fat-free mass (FFM). TheFFM can be divided into a multicompartmentalmodel consisting of: skeleton, skin, skeletal mus-cle, visceral organs, and total intracellular and ex-tracellular body water. The BCM, which isdefined as the FFM without the bone mineralmass and extracellular water, is the most metabol-ically active compartment of the body.

Chertow et al.9 validated the method of BIA byassessing the body composition of HD patients andcomparing the results of total body water andBCM obtained through BIA to those obtained us-ing methods of dilution with deuterium oxide andsodium bromide and DEXA. The authors alsofound that the coefficient of variation of repeated

measurements with BIA was below 4%, demon-strating the reliability of the test.10

The effect of variation in the state of hydrationof the patient is the main limitation of the methodbecause if the patient is hyperhydrated, the FFM isoverestimated.11 Although the BIA method ishighly accepted by the scientific community forthe evaluation of body composition, resultsmay be affected by several factors, including diet,exercise, dehydration or water retention, use of di-uretics, and menstrual cycle.

Subjects and Methods

This is a cross-sectional study in which the au-thors aimed to evaluate the PA and BCM assessedby BIA as nutritional markers and to compare BIAparameters to clinical, biochemical, and anthropo-metric markers of nutrition in HD patients. Theexclusion criteria were age less than 18 years, dial-ysis vintage less than 3 months, limb amputation,and pregnancy.

Demographic characteristics of the patients,such as gender, age, body weight, height, underly-ing kidney disease, time on dialysis, duration andfrequency of HD therapy, and interdialytic gainweight, were determined.

Subjective Global Assessment Modifiedfor Renal Disease (SGA-1)

Information was obtained regarding changes inthe usual body weight, functional activity, feedinghabits, and presence of digestive symptoms in HDpatients. A brief physical examination was per-formed to assess muscle mass, fat storage, andedema. A score was attributed to each patient.12

Body Composition Assessment

The anthropometric indexes used in this studywere dry weight (kg), height (cm), percent of stan-dard body weight (% SBW), body mass index(BMI), mid-arm circumference (MAC), mid-arm muscle circumference (MAMC), and skinfoldmeasurements. All measurements were performedafter a mid-week HD session. Patients wereweighed while wearing light clothes and no shoes.

The SBW was determined for each patient ac-cording to his or her actual weight, gender andheight using the Metropolitan Life InsuranceCompany table (1983).13 The percentage of theactual weight in relation to the standard body

OLIVEIRA ET AL316

weight (% SBW) was calculated with the formula:% SBW 5 actual weight 3 100/SBW.

Skinfold measurements were performed at foursites (biceps, triceps, subscapular, and suprailiac)on the opposite side of the vascular access usinga Lange skinfold caliper. Three measurementswere averaged for each site. Body density was cal-culated using the formula of Durnin and Womers-ley,14 and the percentage of body fat was thencalculated by Siri’s equation.15 Values for the FMand FFM were then derived.

BIA was performed with a single 50-kHz fre-quency system with tetrapolar electrodes (BIA101-A; RJL Systems, San Diego, CA) by thesame operator 30 min after an HD session withthe patients in estimated dry weight. The patientswere placed in a supine position with their armslying parallel and away from their trunk and theirlegs separated so that they were not touching eachother. Two pairs of disposable electrodes wereused: the first pair was placed on the dorsum ofthe hand over the third metacarpophalangeal jointand the wrist, and the second pair was placed overthe ipsilateral third metatarsophalangeal and anklejoints. Electrodes were placed on the arm oppositeto the vascular access site. The average of threereadings was obtained for all patients.

Resistance (R) and reactance (Xc) were mea-sured, and the PA was calculated directly fromXc and R (PA 5 arc tangent Xc/R 3 180degrees/p). PA values less than 5 were consideredabnormal.16 Estimates of the total body water(TBW), BCM, FM, and FFM were obtainedwith the Fluids and Nutrition Analyzer Program(software provided by RLJ Systems, San Diego,CA). Men and women with BCM percentageslower than 35% and 30%, respectively, of bodyweight were considered malnourished.16

Biochemical Parameters

Blood drawn for routine analysis before the HDsession was used to calculate hematocrit, hemo-globin, urea (colorimetric method), and serumalbumin (bromocresol green)

Statistical Analysis

The data are expressed as mean 6 SD. Unpairedt-tests were used to compare mean differences.The correlation between methods was assessedthrough Pearson’s correlation coefficient and

Bland and Altman analysis. A P-value , .05 wasconsidered statistically significant.

Ethical Aspects

The study was approved by the hospital ethicalcommittee before patient enrollment. Fifty-eightpatients participated in the study after giving theirinformed consent to the protocol.

Results

A total of 58 adult patients (age 49.2 6 14.8years, 51.7% females) were studied. The patientshad been on HD for 4.27 6 2.50 years, and the in-terdialytic weight gain was 1.93 6 1.04 kg. Thecauses of end-stage renal disease were hypertensivenephrosclerosis (22.4%), diabetic nephropathy(8.6%), hereditary nephritis (5.2%), polycystic kid-ney disease (5.2%), ‘‘unknown’’ (48.3%), and‘‘others’’ (10.3%). HD was performed three timesa week in 4-hour sessions using high-flux polysul-fone dialyzers and a bicarbonate bath. The demo-graphic, clinical, anthropometric, and biochemicalcharacteristics of the patients enrolled for the studyare shown in Table 1.

Resistance (R), reactance (Xc), phase angle(PA), and derived body compartment values areshown in Table 2. Women presented a significantlyhigher resistance than did men. Men had a signifi-cantly higher BCM and PA than did women.

The prevalence of malnutrition was 17.5% and43.9% according to the PA and percent of BCM,respectively. PA had a negative correlation withage (r 5 –0.525, P , .0001) and SGA-1 score (r 5

–0.533, P , .0001) and a positive correlation with% SBW (r 5 0.277, P 5 .039), MAC (r 5 0.369,P 5 .005), MAMC (r 5 0.284, P 5 .033), FFM-A(r 5 0.42, P 5 .001), and albumin (r 5 0.588,P , .0001).

BCM had a negative correlation with age(r 5 20.455, P , .0001), MAC (r 5 –0.287,P 5 .031), MAMC (r 5 –0.378, P 5 .004), andFM-A (r 5 –0.655, P ,.0001) and a positive corre-lation with FFM-A (r 5 0.529, P , .0001) andalbumin (r 5 0.328, P 5 .013) (Table 3).

A significant correlation between FFM-A andFFM-BIA (r 5 0.939, P ,.001) as well as betweenFM-A and FM-BIA (r 5 0.824, P , .001) wasfound. However, FFM-BIA was significantlydifferent from FFM-A and resulted in an overesti-mation of the FFM (mean difference 5 2.2 kg,

Table 1. Demographic, Clinical, Anthropometric, and Biochemical Characteristics of the Patients

Parameter All Patients Men Women P

Patient (n) 58 28 30

Age (yr) 49.2 6 14.8 49.1 6 15.6 49.3 6 14.3 .954

Dialysis age (yr) 4.27 6 2.50 4.20 6 2.35 4.33 6 2.66 .847

Body weight (kg) 56.5 6 12.3 64.3 6 11.2 49.3 6 8.4 .0001Height (cm) 156.0 6 10.0 164.0 6 6.0 149.0 6 7.0 .0001

BMI (kg/m2) 22.8 6 3.6 23.9 6 3.7 21.9 6 3.3 .038

Interdialytic weight gain (kg) 1.93 6 1.04 2.23 6 1.02 1.66 6 1.01 .039

SGA-1 11.66 6 2.76 11.32 6 1.98 11.97 6 3.33 .370MAC (cm) 26.28 6 3.78 27.54 6 3.52 25.09 6 3.66 .012

MAMC (cm) 22.86 6 3.52 24.73 6 3.28 21.12 6 2.81 .0001

FFM, anthropometry (kg) 43.61 6 10.33 52.13 6 7.12 35.39 6 4.68 .0001

FM, anthropometry (kg) 13.26 6 5.74 11.98 6 6.31 14.49 6 4.93 .099Hematocrit (%) 34.2 6 6.07 35.05 6 6.07 33.45 6 6.07 .319

Hemoglobin (g/dL) 11.57 6 2.07 11.96 6 2.08 11.21 6 2.02 .167

Urea (mg/dL) 164.33 6 37.02 166.5 6 38.1 162.3 6 36.51 .670Albumin (g/dL) 3.45 6 0.55 3.65 6 0.41 3.25 6 0.59 .004

BMI, body mass index; SGA-1, Subjective Global Assessment modified for renal disease; MAC, mid-arm circumference;

MAMC, mid-arm muscle circumference; FFM, free-fat mass; FM, fat mass.

MARKERS OF NUTRITIONAL STATUS IN HEMODIALYSIS PATIENTS 317

P ,.0001, CI 5 1.23–3.17 kg) and of LBM (meandifference 5 1.9 kg, P , .0001).

Figures 1 and 2 show the comparison graphicsbetween the two methods though the Bland andAltman analysis of FFM-A and FFM-BIA andLBM-A and LBM-BIA.

Discussion

The diagnosis of protein-calorie malnutrition isoften delayed in dialysis patients due to a lack ofa definitive criterion for identification. Someauthors suggest that the diagnosis should be basedon multiple nutritional indicators, including anassessment of visceral protein deposits (expressedin biochemical parameters) and somatic depositsdetermined by body composition analysis (weight,anthropometry, BIA, total body nitrogen, andDEXA scanning).2 In addition, although the im-portance of the nutritional status of dialysis patientsis widely recognized in research settings, it has beengiven less emphasis in everyday nephrology

Table 2. BIA Parameters of the Study Population

All Patients

R (ohm) 643.0 6 110.0

Xc (ohm) 68.9 6 15.1PA (o) 6.2 6 1.3

Body cell mass (kg) 19.1 6 5.7

Percentage of BCM (%) 33.75 6 5.91FFM BIA (kg) 40.97 6 8.80

FM BIA (kg) 15.05 6 6.16

R, resistance; Xc, reactance (Xc); PA, phase angle; BMC, bod

practice, especially with regard to the monitoringof dialysis adequacy, pressure levels, vascular access,and cardiovascular comorbidities.

Anthropometry is a common method forevaluating nutritional status, but it may not be suit-able for patients with CKF due to oscillations intissue hydration. In addition, not only is anthro-pometry relatively inefficient at identifying mal-nutrition in HD patients (especially during theearly stages due to a lack of reliable standards ofcomparison),17 but it is also examiner dependent.In fact, some authors believe that anthropometrymarkedly underestimates protein loss in CKFcases.18 Nevertheless, Nelson et al.17 demon-strated that anthropometry is reproducible andattains a sensitivity of up to 90%.

A studypublishedbyMadore,Wuest, andEthier19

showed a close correlation between anthropometry-evaluated lean body mass and bioelectrical imped-ance findings (r 5 0.71). Some authors have foundskinfold measurements to be more accurate thanBIA for the evaluation of the body fat compartment

Men Women P

572.0 6 70.0 706.0 6 102.0 .0001

66.9 6 12.9 70.7 6 16.8 .3516.7 6 1.2 5.7 6 1.3 .005

23.9 6 3.8 14.7 6 2.9 .0001

30.27 6 4.79 37.61 6 4.49 .000147.91 6 6.94 34.73 6 4.61 .0001

15.86 6 6.38 14.31 6 5.96 .349

y cell mass; FFM, free-fat mass; FM, fat mass.

Table 3. Correlation Between Phase Angle (PA)/Percent of BCM (% BCM) and Nutritional Markers

PA % BCM

Nutritional

Marker r P r P

Age –0.525 .0001 –0.455 .0001BMI 0.260 .051 –0.244 .068

% SBW 0.277 .039 –0.138 .311

TSF % 0.113 .402 –0.083 .538

MAC % 0.369 .005 –0.287 .031MAMC % 0.284 .033 –0.378 .004

FFM-A 0.441 .001 0.529 .0001

FM-A –0.116 .394 –0.655 .0001

FFM-BIA 0.420 .001 0.491 .0001FM-BIA 0.075 .579 –0.471 .0001

Albumin 0.588 .0001 0.328 .013

SGA-1 score –0.533 .0001 –0.162 .229Reactance 0.681 .0001 0.266 .046

Resistance –0.370 .005 –0.442 .001

40 60 80

-15

15

Mean + 2SD

Mean

Mean - 2SD

1,9

8,3

-4,5

(LBM-A + LBM-BIA)/2

(L

BM

-A

- L

BM

-B

IA

)

Figure 2. Comparison between LBM-A and LBM-BIA of the study population through use of the Blandand Altman method.

OLIVEIRA ET AL318

because of the smaller gender-related variability.20

Other authors prefer BIA to skinfold measurementsbecause of its smaller margin of examiner-dependenterror and its ability to evaluate the patients’ hydrationstatus.21 However, Woodrow et al.22 concluded that,in CKF patients, skinfold measurements and BIAgenerate results similar to the results obtained byDEXA scanning. The latter method has been widelyused as a reference in comparative studies of bodycomposition and is still recommended by theNational Kidney Foundation Kidney Disease Out-comes Quality Initiative (NKF KDOQI).23

Other BIA parameters, such as BCM, reac-tance, and PA, seem to be less influenced by vari-ations in blood volume and are therefore useful forthe evaluation of the nutritional status of dialysis

10 20 30

-15

15

Mean + 2SD

Mean - 2SD

Mean

(FFM-A + FFM-BIA)/2

2,2

2,2

8

-3,6

(F

FM

-A

- F

FM

-B

IA

)

Figure 1. Comparison between FFM-A and FFM-BIA of the study population through use of the Blandand Altman method.

patients. In our patients, the BIA-measured mal-nutrition prevalence was 17.5% (PA) and 43.9%(BCM). Barbosa-Silva et al.16 found their patients’postoperative malnutrition prevalence to be 18.5%(PA) and 59.3% (BCM).

The PA has been shown to be correlated witha number of nutritional indices, such as theSGA, anthropometric measures, and albumin,prealbumin, and creatinine levels.24–26

Many clinical studies have associated PAwith morbidity and mortality in HD patients.25

However, it remains unclear whether the relationbetween PA and survival is dependent on nutri-tional status. Low PA values have also been associ-ated with mortality in patients with HIV andcancer of the colon and pancreas.27,28

The advantage of using BCM in the evaluationof nutritional status is that, unlike FFM, it does notinclude extracellular water (which is often in-creased in CKF patients and is a frequent causeof the overestimation of the nutritional status).29

Reductions in the visceral or somatic proteinmass may be masked by a concurrent increase inextracellular water.

BCM estimation may soon become the mostimportant element in BIA. So far, nephrologistshave based their evaluations mostly on physical ex-amination and serum protein levels, despite therisk of underestimation or delayed detection ofmalnutrition.

NKF KDOQI guidelines23 do not considerBIA parameters (reactance and resistance) to bevalid or reliable enough to estimate body compo-sition and evaluate the protein-calorie nutritionalstatus of dialysis patients. Likewise, the EuropeanConsensus on the nutritional status of dialysis pa-tients30 does not recommend BIA for routine

MARKERS OF NUTRITIONAL STATUS IN HEMODIALYSIS PATIENTS 319

nutritional evaluations because the method doesnot provide adequate measurements of lean andFM in patients with a body composition differentfrom healthy adults with normal BMI values.DEXA scanning is a valid and clinically usefulevaluation method as it is less influenced by thevariations in hydration status commonly observedin dialysis patients. The method evaluates the threemain components of body composition: FM, leanmass, and bone mineral mass. The major limita-tions are the high cost of the equipment and exam-ination, the spatial requirements of the equipment,and the unavailability of the examination at a largenumber of facilities. In addition, the method doesnot distinguish between intracellular and extracel-lular water.

BIA does, however, appear to be a valid methodfor evaluating and detecting early changes in thenutritional status of dialysis patients. Further stud-ies with serial BIA measurements are required toevaluate the test’s sensitivity to nutritional changesover time and its association with morbidity andmortality (e.g., survival, hospitalization, and func-tional status).

The importance of a study may be measured byits applicability to clinical practice and the benefitit represents to patients during treatment and fol-low-up, thus potentially increasing survival rates.Although there may still not be a single bestmarker for routine clinical use, our findings sug-gest that, in the near future, bioelectric impedancewill play a substantial role in the evaluation of di-alysis patients. BIA is not observer-dependentand is easy to perform, and BIA parameters (suchas phase angle and BCM percentage) are not af-fected by variations in blood volume.

Longitudinal, prospective studies investigatingthe association between these nutritional markersand adverse events, such as hospitalization andmortality, should be carried out to clarify remain-ing issues.

References

1. Blumenkrantz MJ, Kopple JD, Gutman RA, et al: Methods

for assessing nutritional status of patients with renal failure. Am J

Clin Nutr 33(7):1567-1585, 1980

2. Ikizler TA, Hakim RM: Nutrition in end-stage renal dis-

ease. Kidney Int 50(2):343-357, 1996

3. Pecoits-Filho R, Stenvinkel P, Lindholm B, et al: Revisao:

Desnutricao, inflamacao e aterosclerose (sındrome MIA) em pa-

cientes portadores de insuficiencia renal cronica hemodialise. J

Bras Nefrol 24(3):136-146, 2002

4. Kopple JD: Protein-energy malnutrition in maintenance di-

alysis patients. Am J Clin Nutr 65(5):1544-1557, 1997

5. Guarnieri G, Antonione R, Biolo G: Mechanisms of malnu-

trition in uremia. J Renal Nutr 13(2):153-157, 2003

6. Chumlea WC, Guo SS: Bioelectrical impedance and body

composition: present status and future direction-reply. Nutr Rev

52(4):123-131, 1994

7. Heymsfield SB, Mcmanus C, Smith J, et al: Anthropometric

measuring of muscle mass: revised equations for calculating bone-

free arm muscle area. Am J Clin Nutr 36(4):680-690, 1982

8. Pupim LCB, Ribeiro CB, Kent P, et al: Atualizacao em di-

alise: Uso da bioimpedancia eletrica em pacientes em dialise. J Bras

Nefrol 22(4):249-256, 2000

9. Chertow GM, Lowrie EG, Wilmore DW, et al: Nutritional

assessment with bioelectrical impedance analysis in maintenance

hemodialysis patients. J Am Soc Nephrol 6(1):75-81, 1995

10. Dumler F, Kilates C: Body composition analysis by bioelec-

trical impedance in chronic maintenance dialysis patients: com-

parisons to the National Health and Nutrition Examination

Survey III. J Ren Nutr 13(2):166-172, 2003

11. De Lorenzo A, Barra PFA, Sasso GF, et al: Body impedance

measurements during dialysis. Eur J Clin Nutr 45(6):321-325,

1991

12. Kalantar-Zadeh K, Kleiner M, Dunne E, et al: A modified

quantitative subjective global assessment of nutrition for dialysis

patients. Nephrol Dial Transplant 14(7):1732-1738, 1999

13. 1983 Metropolitan Height and Weight Tables. Stat Bull

64(1):1-9, 1983

14. Durnin JVGA, Womersley J: Body fat assessed from total

body density and its estimation from skinfold thickness: measure-

ment on 481 men and women aged from 16 to 72 years. Br J Nutr

32(1):77-97, 1974

15. Siri WE: The gross composition of the body. Adv Biol Med

Phys 4:239-280, 1956

16. Barbosa–Silva MCG, Barros AJD, Post CLA, et al: Can

bioelectrical impedance analysis identify malnutrition in preoper-

ative nutrition assessment. Nutrition 19(5):422-426, 2003

17. Nelson EE, Hong C, Pesce AL, et al: Anthropometric

norms for the dialysis population. Am J Kidney Dis 16(1):

32-37, 1990

18. Rayner HC, Stroud DB, Salamon KM, et al: Anthropom-

etry underestimates body protein depletion in hemodialysis pa-

tients. Nephron 59(1):33-40, 1991

19. Madore F: Wuest M, Ethier JH: Nutritional evaluation of

hemodialysis patients using an impedance index. Clin Nephrol

41(6):377-382, 1994

20. Kamimura MA, Avesani CM, Cendoroglo M, et al:

L. Comparison of skinfold thickness and bioelectrical impedance

analysis with dual-energy x-ray absorptiometry for the assessment

of body fat in patients on long-term hemodialysis therapy.

Nephrol Dial Transplant 18(1):101-105, 2003

21. Oe B, de Fijter CWH, Oe PL, et al: Four-sites skinfold an-

thropometry (FSA) versus bioimpedance analysis (BIA) in assess-

ing nutritional status of patients on maintenance hemodialysis:

which method is to be preferred in routine patient care? Clin

Nephrol 49(3):180-185, 1998

22. Woodrow G, Oldroyd B, Smith MA, et al: Measurement of

body composition in chronic renal failure: comparison of skinfold

anthropometry and bioelectrical impedance with dual energy

x-ray absortiometry. Eur J Clin Nutr 50(5):295-301, 1996

OLIVEIRA ET AL320

23. National Kidney Foundation: K/DOQI: Clinical Practice

Guidelines for Nutrition in Chronic Renal Failure. Am J Kidney

Dis 35(suppl 2):S1-S140, 2000

24. Chertow GM, Lazarus JM, Lew NL, et al: Bioimpedance

norms for the hemodialysis population. Kidney Int 52(6):

1617-1621, 1997

25. Maggiore Q, Nigrelli S, Ciccarelli C, et al: Nutritional and

prognostic correlates of bioimpedance indexes in hemodialysis pa-

tients. Kidney Int 50(6):2103-2108, 1996

26. Enia G, Sicuso C, Alati G, et al: Subjective global assess-

ment of nutrition in dialysis patients. Nephrol Dial Transplant

8(10):1094-1098, 1993

27. Gupta D, Lammersfeld CA, Burrows JL, et al: Bioelectrical

impedance phase angle in clinical practice: implications for prog-

nosis in advanced colorectal cancer. Am J Clin Nutr 80(6):

1634-1638, 2004

28. Schwenk A, Beisenherz A, Romer K, et al: Phase angle

from bioelectrical impedance analysis remains an independent

predictive marker in HIV-infected patients in the era of highly ac-

tive antiretroviral treatment. Am J Clin Nutr 72(2):496-501, 2000

29. Dumler F, Kilates C: Use of bioelectrical impedance tech-

niques for monitoring nutritional status in patients on mainte-

nance dialysis patients. J Ren Nutr 10(3):116-124, 2000

30. Locatelli F, Fouque D, Heimburger O, et al: Nutritional

status in dialysis patients: a European consensus. Nephrol Dial

Transplant 17(4):563-572, 2002