R

SKADA

NGgnbuceoec©

Sttsiacf

C

T

D

l

DB

J

EVIEW

ubjective Global Assessment in Chronicidney Disease: A Review

lison L. Steiber, PhD, RD, LD,* Kamyar Kalantar-Zadeh, MD, PhD, MPH,†onna Secker, MS, RD,‡ Maureen McCarthy, MPH, RD, CSR, LD,§shwini Sehgal, MD,¶ and Linda McCann, RD, LD�

utritional assessment of patients with chronic kidney disease is a vital function of health care providers. Subjectivelobal Assessment (SGA) is a tool that uses 5 components of a medical history (weight change, dietary intake,astrointestinal symptoms, functional capacity, disease and its relation to nutritional requirements) and 3 compo-ents of a brief physical examination (signs of fat and muscle wasting, nutrition-associated alternations in fluidalance) to assess nutritional status. SGA was originally used to predict outcomes in surgical patients; however, itsse has gone beyond this function and population. In chronic kidney disease patients, SGA is incorporated into theomplete nutritional assessment. Validation of SGA as a screening tool for surgical patients was done by Detskyt al in 1984. Since that time, SGA has been altered by different researchers and clinicians to better meet the needsf the patients they served. Validation of the altered SGA formats has not been thoroughly done. Further work instablishing validity and reliability of each version of SGA in different patient populations should be done to enablelinicians and researchers to properly use this nutritional assessment tool.2004 by the National Kidney Foundation, Inc.

tfBm(Qt

bmsrpsipmtct(tw

UBJECTIVE GLOBAL ASSESSMENT(SGA) is a tool used by health care providers

o assess nutritional status and aid in the predic-ion of nutrition-associated clinical outcomes,uch as postoperative infections1 and/or mortal-ty.2 The tool has many strengths in the clinicalnd research setting: it is inexpensive; is rapid toonduct; can be used effectively by providersrom different disciplines, such as nursing, dieti-

*Department of Nutrition, Case Western Reserve University,leveland, OH.†Harbor-UCLA, UCLA David Geffen School of Medicine,

orrance, CA.‡Hospital for Sick Children, Department of Clinical Dietetics,ivision of Nephrology, Toronto, Ontario, Canada.§McMinnville Kidney Center, McMinnville, OR.¶Division of Nephrology, MetroHealth Medical Center, Cleve-

and, OH.�Satellite Healthcare, Redwood City, CA.Address reprint requests to Alison Steiber, PhD, RD, LD,epartment of Nutrition, Case Western Reserve University, Dentaluilding, Room 201, 10900 Euclid Ave, Cleveland, OH 44106.© 2004 by the National Kidney Foundation, Inc.1051-2276/04/1404-0002$30.00/0

sdoi:10.1053/j.jrn.2004.08.004

ournal of Renal Nutrition, Vol 14, No 4 (October), 2004: pp 191-200

ians, and physicians; and in some studies has beenound to be reproducible, valid, and reliable.3,4

ecause of its strengths, SGA has been recom-ended by the National Kidney Foundation

NKF) Kidney Disease/Dialysis Outcomes anduality Initiative (K/DOQI) for use in nutri-

ional assessment in the adult dialysis population.5

However, for all its potential, SGA has yet toe thoroughly validated in the maintenance he-odialysis and peritoneal dialysis population. A

tudy recently published disputed the validity andeliability of SGA in hemodialysis patients. Coo-er et al6 examined SGA ratings between 2 ob-ervers and against total body nitrogen. Thesenvestigators concluded that SGA can detect theresence of malnutrition but not the degree ofalnutrition.6 An additional complication in de-

ermining the usefulness of SGA in both thelinical and research arenas is the modification ofhe original tool. In the chronic kidney diseaseCKD) literature, a minimum of 5 different SGAools have been reported,1,3,7-9 almost none ofhich have been tested in a large validation

tudy.

191

AbMal(bSiAooDa.nicddcSSTp

fTabtoSaaw

nuiptstcuoemgdmenTnocwtmtv1bvwa

opppbF

s

STEIBER ET AL192

To address these issues, a Subjective Globalssessment Consensus Conference was organizedy the Department of Nutrition of the School ofedicine of Case Western Reserve University

nd held on November 7 and 8, 2003, in Cleve-and, OH. The objectives of the conference were1) to review the methods, techniques, and toolseing used for SGA; (2) to examine the validity ofGA; and (3) to identify how and by whom SGAs being used in clinical practice and research.ttendance at this conference was by invitationnly; announcements were placed in the Journalf Renal Nutrition, the American Journal of Kidneyisease, the Journal of the American Dietetic Associ-

tion, and on an SGA website: http://www-nephrology.rei.edu/sgahome.htm. The an-ouncements requested applications from peoplenterested in attending and/or presenting at theonference. Thirty individuals (physicians andietitians) were invited to attend. During theay-and-a-half conference, presentations in-luded original research results, experiences withGA in clinical practice, and experiences withGA in education programs for dietetics students.hroughout the conference, attendees partici-ated in roundtable discussions to generate ideas

igure 1. Plan for scholarly work by the SGA Con-

sensus Conference Group.

or validating SGA within the renal population.he consensus of the group of professionals who

ttended this conference is that further study muste conducted to standardize and validate SGA forhe CKD population. Figure 1 outlines the rec-mmended plan for further scholarly work withGA. This article is one component of that plan,nd is intended to review current literature avail-ble on SGA and to make recommendations onork to be done.

History of SGADetsky et al1,10 published the first reports of a

utritional assessment tool, entitled SGA, thatsed clinical judgment to assess nutritional statusn preoperative surgical patients and to predictostoperative infections; SGA had the best sensi-ivity and specificity for predicting infection afterurgery. SGA was quickly used in other popula-ions such as elderly patients,11-13 patients withancer14 or liver transplants,15 and adult patientsndergoing maintenance dialysis.2,3,6,16 Theriginal SGA form (Fig 2) as reported by Detskyt al1 had clinicians score 5 components of aedical history (ie, weight change, dietary intake,

astrointestinal symptoms, functional capacity,isease and its relation to nutritional require-ents) and 3 components of a brief physical

xamination (ie, signs of fat and muscle wasting,utrition-associated alternations in fluid balance).he patient is then assigned a rating of wellourished (A), moderately undernourished (B),r severely undernourished (C) by subjectiveonsideration of the data collected in the 8 areas,ithout adhering to a rigid scoring system. From

his original form, the tool has been modified byany others in an attempt to increase its predic-

ive value and reproducibility.2,7,8 Hirsch et alalidated SGA in 175 gastroenterology patients in990. That study found significant differencesetween well-nourished and moderately or se-erely undernourished patients in serum albumin,eight, midarm muscle circumference (MAMC),

nd triceps skinfold measurements.17

The first validation study in CKD patientsccurred in 199316 with continuous ambulatoryeritoneal dialysis (CAPD) patients. SGA waserformed on 23 CAPD and 36 hemodialysisatients, and significant correlations were seenetween the subjects’ SGA ratings and values for

erum albumin, bioelectrical impedance,

SUBJECTIVE GLOBAL ASSESSMENT IN CHRONIC KIDNEY DISEASE 193

Figure 2. A, B, and C original SGA.

Mtoapci6ctseosmwi7ebntpp(wSsMwgc

Pv7(dt

orstvnTafiEw

mtt0cs

dd7aocmcipMfnoppiaiTeiotawwarts

vcsfumkram

STEIBER ET AL194

AMC, percent body fat, and normalized pro-ein catabolic rate. This study’s SGA methodol-gy was used in the next major study in Canadand the United States (CANUSA) in the CKDopulation. CANUSA was a multicenter studyonducted in Canada and the United States thatnvestigated mortality and nutritional status in80 patients on peritoneal dialysis.12 This studyhanged Detsky’s A, B, C method of rating SGAo a 7-point scale (Fig 3). The components as-essed remained the same, but the rating scale wasxpanded. Using survival analysis, the relative riskf death was increased with worsening nutritionaltatus as defined by SGA and loss of lean bodyass.2 A major outcome of the CANUSA studyas that a 1-unit decrease in SGA equaled a 25%

ncrease in mortality for CAPD patients. The-point rating scale has been pilot tested by Vissert al3 and Jones et al.4 The cross-sectional studyy Visser et al3 on 13 hemodialysis and 9 perito-eal dialysis patients showed that SGA was posi-ively correlated with body mass index (BMI),ercent body fat, and MAMC. In a recentlyublished article by Jones et al,4 both the A, B, C3-point) scale and the 7-point scale SGA formsere conducted with 72 hemodialysis patients.tatistical differences were found between SGAcores (both A, B, C and 7-point scales) for

AMC and serum creatinine.4 The A, B, C scaleas also statistically different between A and Broups with the serum C-reactive protein con-entration.3

Kalantar-Zadeh et al,7 Stenvinkel et al,9 andifer et al8 have each studied different modifiedersions of SGA in samples ranging from 41 to,719 patients. Modifications in the rating scaleie, from 7 points to 49 or 57 points) and theirection of data collection (ie, from prospectiveo retrospective8) have been made.

In 1999, Kalanter-Zadeh et al7 presented an-ther version of the SGA that was originallyeferred to as modified quantitative SGA and inubsequent publications as the Dialysis Malnutri-ion Score (DMS).18,19 This fully quantitativeersion of SGA used the 7 original SGA compo-ents and created a quantitative scoring system.he scoring was a 5-point scale with 1 as normal

nd 5 as very severe malnutrition (Fig 4). Thenal score was the total sum of all 7 components.ach component was rated on a scale of 1 to 5

ith a possible total range from 7 to 35. This a

ethod of SGA scoring produced high correla-ions with objective nutritional indicators such asotal iron-binding capacity (TIBC) (r � 0 to.77) and MAMC (r � 0 to 0.66) and moderateorrelations with serum albumin, BMI, bicepkinfold, age, and years on dialysis.7

The Malnutrition-Inflammation Score (MIS),eveloped by Kalantar-Zadeh, is a recently intro-uced, fully quantitative tool that is based on theoriginal SGA components and also includes 3

dditional items (BMI and serum concentrationsf albumin and serum TIBC).18,20 Each MISomponent has 4 levels of severity from 0 (nor-al) to 3 (very severe). The sum of all 10 MIS

omponents ranges from 0 to 30, denoting thencreasing degree of severity (Fig 5). In a 2001rospective study on 83 hemodialysis patients,IS was compared with conventional SGA, its

ully quantitative version (DMS), anthropometry,ear-infrared measured body fat percentage, lab-ratory measures including serum C-reactiverotein (CRP), and 12-month prospective hos-italization and mortality rates.18 MIS had signif-cant correlations with prospective hospitalizationnd mortality as well as measures of nutrition,nflammation, and anemia in dialysis patients.he correlations were higher for MIS thanither the conventional SGA or DMS withndividual laboratory values as a predictor ofutcome. In a 2004 recent multicenter study byhe same group of investigators, the mortalitynd hospitalization predictability of the MISas assessed in 378 hemodialysis patients; MISas found to be comparable with serum CRP

nd serum interleukin-6.20 The MIS is cur-ently being used in the multicenter Nutri-ional and Inflammatory Evaluation in Dialysistudy (www.NIEDstudy.org).21,22

In 1999, Stenvinkel et al9 published anotherersion of the SGA. Although these researchersited Detsky et al and Baker et al in their methodsections, Stenvinkel et al changed the scoringrom the original A, B, C scale to a 4-point scalesing 1 as normal nutritional status and 4 as severealnutrition.9 Data on 109 adults with chronic

idney failure were analyzed by creating a biva-iate variable with SGA scores 2 to 4 as one groupnd an SGA score of 1 as another group. In thisanner they found those with scores between 2

nd 4 were older, more frequently had a history

osla

Sg

SUBJECTIVE GLOBAL ASSESSMENT IN CHRONIC KIDNEY DISEASE 195

f tobacco use, and had significantly lower BMI,erum creatinine, serum albumin, urine urea, andean body mass (measured by dual-energy x-ray

Figure 3. The 7-p

bsorptiometry).9 T

The Dialysis Outcomes and Practice Patternstudy (DOPPS) study created m-SGA that wasraded retrospectively using a patient interview.

cale SGA form.

oint s

he score was based on the caregiver’s ratings

rad

ms

Fpm

STEIBER ET AL196

elative to weight loss, visual somatic store loss,ppetite, nausea and vomiting, energy level, and

igure 4. The fully quantitative version of the SGarameters are used, and the values are summealnutrition.

isease burden. The rating for m-SGA is normal, l

oderate (any 3 areas rated as a moderate orevere level), or severe (at least 3 areas at severe

so known as modified SGA or DMS. Five scalealue of 7 is normal, and 35 is the most severe

A, ald. A v

evel). Those patients who rated a severe m-SGA

lcr

lpso

tsrofcr

Fsss to 17

SUBJECTIVE GLOBAL ASSESSMENT IN CHRONIC KIDNEY DISEASE 197

evel had a relative risk of 1.33 for mortalityompared with those with a moderate or normalating,8 which was statistically significant.

Although each of the versions has strengths, theirack of uniformity makes it difficult both to com-are research results on nutritional status from onetudy to the next and to provide consistent meth-

igure 5. MIS. *Major comorbid conditions includeevere coronary artery disease, moderate to severeequelae, and metastatic malignancies or s/p recenerum transferrin are �200 (0), 170 to 200 (1), 140

dology guidance for clinicians wishing to use this e

ool. Currently the NKF regularly offers trainingessions at its Clinical Nephrology meetings to trainenal dietitians in the use of the 7-point SGA. Nother formal training forum currently exists. There-ore, it is assumed that the majority of renal dietitiansurrently conducting SGA are using the versionecommended by K/DOQI and studied by Visser

estive heart failure class III or IV, full-blown AIDS,ic obstructive pulmonary disease, major neurologic

otherapy. �Suggested equivalent increments for0 (2), and �140 mg/dL.

congchront chem

t al3 and Jones et al.4

mf(SserrmrtgcCsills

trbgacec

Ctamd

teepdm7pec3w

SiHStampsOstaesmcsacitortl

saArma7tlpsfbciloica

STEIBER ET AL198

Current Literature With SGA as aNutritional Assessment Tool

Table 1 includes studies that used SGA as aethod of nutritional status determination for

urther comparisons against a dependent variableeg, mortality). From this table it is clear thatGA, using either the A, B, C or the 7-pointcale, detects the presence of malnutrition; how-ver, the controversy appears when SGA is cor-elated with serum albumin. In some studies se-um albumin was significantly lower in the SGAalnourished group,9,23-25 whereas in others, se-

um albumin was not significantly different be-ween the normal and the malnourishedroups.4,26 Serum albumin is one of the mostommonly used indicators for malnutrition in theKD population, and although it is affected by

everal other factors including inflammation, thisnconsistency has raised questions about the va-idity of SGA. To that end, incorporating serumaboratory markers for malnutrition may be aolution, as done in the MIS.

Studies have shown significant differences be-ween SGA categories for many other nutrition-elated variables, ie, BMI, MAMC, serum preal-umin, TIBC or transferrin, ferritin, insulin-likerowth factor 1, phase angle (bioelectrical imped-nce analysis), percent body fat, lean body mass,omorbidity state (diabetes, cardiovascular dis-ase, etc), c-reactive protein and cytokines, andreatinine clearance.4,7,9,24-31

The risk of mortality has been assessed byANUSA,2 Lawson et al,27 Davies et al,28 Kalan-

ar-Zadeh,18,20 and Pifer et al,8 with all showingstatistically significant increase in risk or rate ofortality with the presence of malnutrition as

etermined by SGA.Interventional trials using kilocalorie and pro-

ein supplements, such as in the studies by Caglart al32 and Steiber et al,33 have shown varyingffects on changes in an individual’s pre-SGA andost-SGA rating, depending on the interventionuration. The trial by Caglar et al32 was 6onths, included 85 patients, and used the

-point scale. They were able to show an im-rovement in the 7-point SGA over time; how-ver, Steiber et al33 did not see a significanthange in pre-SGA and post-SGA scores over a-month period when the A, B, C rating system

as used in 22 patients. b

RecommendationsA review of the literature indicates that use of

GA as a nutrition assessment tool for CKD patientss growing, in both the clinical and research settings.owever, given the variability of published results,

GA cannot be considered a gold standard in nu-rition assessment for CKD patients. The validitynd reliability of SGA must be proven in a large,ulticenter trial with sufficient power to be able to

revent type I and II errors. Additionally, the study’sample must represent the current CKD population.ne of the difficulties associated with conducting a

tudy such as this is choosing which version of SGAo test. It may be that different SGA versions areppropriate for different patient disease states, differ-nt age stages, or different clinical purposes (eg,creening preoperatively versus full assessment ofaintenance hemodialysis patients). Another diffi-

ulty is data collection. To get a representativeample, data would need to be collected from allreas of the country in a random manner. Thisould be done in a way similar to that of Beto et al34

n a nationally collaborative research projecthrough the National Kidney Foundation’s Counciln Renal Nutrition (CRN). Using this model,egistered dietitians from local CRN groupshroughout the United States could randomly col-ect data on patients in their dialysis centers.

Many of the studies reviewed collapsed the SGAcores into 2 groups (normal and malnourished) fornalyses. For instance, Julien et al,30 Lawson et al,27

bdullah et al,35 and Jones et al,26 used the A, B, Cating system and all dichotomized the final results byerging the B and C groups together for comparison

gainst the A-rated group. Davies et al28 used the-point scale and collapsed it into 6 to 7, 3 to 5, and 1o 2 for analysis, and then grouped those with a 5 oress into a “malnourished group” and compared thoseatients with the 6 to 7 group. This method of analysisubstantiates the conclusion of Cooper et al,6 whoound that SGA detects the presence of malnutritionut not the degree. It is possible that the need for theollapsed groups in such studies has more to do withnadequately powered studies or analytical tools (eg,ogistic regression) than the lack of detectable precisionf SGA. When presenting results of SGA in aggregate,t may be useful to show them in both a full and aollapsed or aggregated format. This would highlightny linear relationships as well as show differences

etween those with and without malnutrition.

boroacrC

scowmsr

T

cH tion Sco

SUBJECTIVE GLOBAL ASSESSMENT IN CHRONIC KIDNEY DISEASE 199

In a large study with sufficient power, SGA maye able to detect differences between all 7/5 pointsr A, B, and C. Similarly, a continuous score mayesolve the issue independent of sample size. The-retically, with careful methodology and statisticalnalysis, a large, nationally representative studyould be designed to determine the validity andeliability of SGA within the diverse United States

able 1. Studies Using SGA as a Tool in Their Me

First Author Journal*

Year;Volume (No.):Page Range

RatingMethod

Maiorca Nephrol DialTransplant

1995;10 ABC

Cianciaruso AJKD 1995;26(3) ABC

Maggiore A Kidney Int 1996;50(6) ABC

Jones CH Nephrol DialTransplant

1997 ABC

Abdullah Miner ElectrolyteMetabolism

1997;23(3-6) ABC

Noh H Perit Dial Int 1998;18(4)Kalantar-Zadeh K AJKD 1998;31(2) ABC

Kalantar-Zadeh K Nephrol DialTransplant

1999;14(7):1732-1738

5-point

Biesenbach G Nephrol DialTransplant

1999;14(3) ABC

Passadakis P Adv Perit Dial 1999;15

Visser R Adv Perit Dial 1999;15:222-225

7-point

Davies SJ Kidney Int 2000;57(4) 7-pointKalantar-Zadeh K AJKD 2001;38(6):

1251-12634-point, plus

3 new itemLawson J JREN 2001;11(1) ABC

Sezer S Adv Perit Dial 2001;17

Julien J EDTNA 2001;27(4) ABC

Cooper BA AJKD 2002;40(1):126-132

ABC

Caglar K Kidney Int 2002;62 7-point

Bakewell A Q J Med 2002;95(12) 7-point

Steiber A JREN 2003;13(3) ABC

Kalantar-Zadeh K Nephrol DialTransplant

2004;19(6):1507-1519

4-point, plus3 new item

Abbreviations: SGA, subjective global assessment; LBM, lean body mircumference; IGF-1, insulin growth factor-1; TNF�, tumor necrosis factor alD-PNI, Hemodialysis Prognostic Nutritin Index; MIS, Malnutrition-Inflamma*Medline abbreviations used.

KD population. a

Until the issue of which form of SGA is bestuited to the hemodialysis population is determined,linicians who are currently using one of the formsf SGA should continue to perform SGA. SGA isithout a doubt a useful tool for nutritional assess-ent. However, as with all of the available tools, it

hould be used in conjunction with anthropomet-ic, laboratory, and dietary intake measures to form

ogy

Main ComparisonVariable(s) n Results

rvival 578 No difference in survivalbetween SGA groups

e 487 Older patient 2 SGAscore

ioelectricalimpedanceanalysis

131 SGA 1 as phase angle 1,not predictive in patientswith worst SGA rating

utrition parameters 76 LBM, CrCl, BMI, MAMC,handgrip, weight 2 in Band C groups

F-1, TNF� 20 B and C groups 2 IGF-1and 1 TNF�

ortality 106boratoryparameters

59 C group has 2 TIBC

b, TIBC,anthropometry

41 hemodialysis Fully quantitative SGA hadgood correlation withlaboratory andanthropometricnutritional markers

iabetic versusNondiabetic

30 No difference betweenSGA groups

ioelectricalimpedanceanalysis

47 Correlation between phaseangle and SGA

MI, anthropometry,albumin

13 hemodialysis9 peritoneal

dialysis

7-point SGA scale is avalid and reliable tool forassessing nutritionalstatus among end-stagerenal disease patients

141RP, mortality,hospitalization

83 hemodialysis MIS predicted clinicaloutcome

ortality 87 1 mortality in B and Cgroups

b 100 Alb 2 in malnourishedpatients

b, prealb 32 Prealb 1 in A versus Band C groups

tal body nitrogen 76 SGA differentiated severelymalnourished patientsfrom those with normalnutrition, but was not areliable predictor ofdegree of malnutrition

me-dependentchange

85 SGA 1 over 6 mo

cidence ofmalnutrition

70 SGA 2 over time (NS)

D-PNI 22 HD-PNI 2 in B and Cgroups (NS)

RP, cytokines,mortality,hospitalization

378 hemodialysis MIS was superior toalbumin and was similarto CRP and IL-6 inpredicting clinicaloutcome

l, creatinine clearance; BMI, body mass index; MAMC, midarm muscle, total iron-binding capacity; alb, serum albumin; prealb, serum prealbumin;

re; CRP, C-reactive protein NS, not significant.

thodol

Su

Ag

B

N

IG

MLa

Al

D

B

B

sC

M

Al

Al

To

Ti

In

H

sC

ass; CrCpha; TIBC

comprehensive nutritional assessment.

sE

GnN

tt

aJ

g3

sr

qp

hD

ar

atE

pa

tt

asc

Gt5

Mgt

m8

gt

t

t3

AtT

CiN

lo1

ah

at

at1

inA

nu

tR

ep

sr

mE

pt

oI

niN

nhCl

m

STEIBER ET AL200

References1. Detsky AS, McLaughlin JR, Baker JP, et al: What is

ubjective global assessment of nutritional status? JPEN J Parenternteral Nutr 11:8-13, 19872. Canada-USA (CANUSA) Peritoneal Dialysis Studyroup. Adequacy of dialysis and nutrition in continuous perito-eal dialysis: Association with clinical outcomes. J Am Socephrol 7:198-207, 19963. Visser R, Dekker FW, Boeschoten EW, et al: Reliability of

he 7-point subjective global assessment scale in assessing nutri-ional status of dialysis patients. Adv Perit Dial 15:222-225, 1999

4. Jones CH, Wolfenden RC, Wells LM: Is subjective globalssessment a reliable measure of nutritional status in hemodialysis?Ren Nutr 14:26-30, 20045. K/DOQI, National Kidney Foundation: Clinical practice

uidelines for nutrition in chronic renal failure. Am J Kidney Dis5:S1-140, 2000 (suppl 2)6. Cooper BA, Bartlett LH, Aslani A, et al: Validity of

ubjective global assessment as a nutritional marker in end-stageenal disease. Am J Kidney Dis 40:126-132, 2002

7. Kalantar-Zadeh K, Kleiner M, Dunne E, et al: A modifieduantitative subjective global assessment of nutrition for dialysisatients. Nephrol Dial Transplant 14:1732-1738, 19998. Pifer TB, McCullough KP, Port FK, et al: Mortality risk in

emodialysis patients and changes in nutritional indicators:OPPS. Kidney Int 62:2238-2245, 20029. Stenvinkel P, Heimburger O, Paultre F, et al: Strong

ssociation between malnutrition, inflammation, and atheroscle-osis in chronic renal failure. Kidney Int 55:1899-1911, 1999

10. Detsky AS, Baker JP, Mendelson RA, et al: Evaluating theccuracy of nutritional assessment techniques applied to hospi-alized patients: Methodology and comparisons. JPEN J Parenternteral Nutr 8:153-159, 198411. Christensson L, Unossons M, Ek AC: Measurement of

erceived health problems as a means of detecting elderly peoplet risk of malnutrition. J Nutr Health Aging 7:257-262, 2003

12. Christensson L, Unosson M, Ek AC: Evaluation of nu-ritional assessment techniques in elderly people newly admittedo municipal care. Eur J Clin Nutr 56:810-818, 2002

13. Covinsky KE, Covinsky MH, Palmer RM, et al: Serumlbumin concentration and clinical assessments of nutritionaltatus in hospitalized older people: Different sides of differentoins? J Am Geriatr Soc 50:631-637, 2002

14. Bauer J, Capra S, Ferguson M: Use of the scored Patient-enerated Subjective Global Assessment (PG-SGA) as a nutri-

ion assessment tool in patients with cancer. Eur J Clin Nutr6:779-785, 200215. Stephenson GR, Moretti EW, El Moalem H, et al:alnutrition in liver transplant patients: Preoperative subjective

lobal assessment is predictive of outcome after liver transplan-ation. Transplantation 72:666-670, 2001

16. Enia G, Sicuso C, Alati G, et al: Subjective global assess-ent of nutrition in dialysis patients. Nephrol Dial Transplant

:1094-1098, 199317. Hirsch S, de Obaldia N, Petermann M, et al: Subjective

lobal assessment of nutritional status: Further validation. Nutri-ion 7:35-37, 1991

18. Kalantar-Zadeh K, Kopple JD, Block G, et al: A malnu-

rition-inflammation score is correlated with morbidity and mor- 2

ality in maintenance hemodialysis patients. Am J Kidney Dis8:1251-1263, 200119. Kalantar-Zadeh K, Rodriguez RA, Humphreys MH:

ssociation between serum ferritin and measures of inflamma-ion, nutrition and iron in haemodialysis patients. Nephrol Dialransplant 19:141-149, 200420. Kalantar-Zadeh K, Kopple JD, Humphreys MH, et al:omparing outcome predictability of markers of malnutrition-

nflammation complex syndrome in hemodialysis patients.ephrol Dial Transplant 2004 (in press)21. Kalantar-Zadeh K, Block G, Humphreys MH, et al: A

ow, rather than a high, total plasma homocysteine is an indicatorf poor outcome in hemodialysis patients. J Am Soc Nephrol5:442-453, 200422. Kalantar-Zadeh K, Block G, McAllister CJ, et al: Appetite

nd inflammation, nutrition, anemia and clinical outcome inemodialysis patients. Am J Clin Nutr 2004 (in press)23. Sezer S, Ozdemir FN, Akman B, et al: Predictors of serum

lbumin level in patients receiving continuous ambulatory peri-oneal dialysis. Adv Perit Dial 17:210-214, 2001

24. Passadakis P, Sud K, Dutta A, et al: Bioelectrical imped-nce analysis in the evaluation of the nutritional status of con-inuous ambulatory peritoneal dialysis patients. Adv Perit Dial5:147-152, 199925. Kalantar-Zadeh K, Kleiner M, Dunne E, et al: Total

ron-binding capacity—estimated transferrin correlates with theutritional subjective global assessment in hemodialysis patients.m J Kidney Dis 31:263-272, 199826. Jones CH, Newstead CG, Will EJ, et al: Assessment of

utritional status in CAPD patients: Serum albumin is not aseful measure. Nephrol Dial Transplant 12:1406-1413, 199727. Lawson JA, Lazarus R, Kelly JJ: Prevalence and prognos-

ic significance of malnutrition in chronic renal insufficiency. Jen Nutr 11:16-22, 200128. Davies SJ, Phillips L, Griffiths AM, et al: Analysis of the

ffects of increasing delivered dialysis treatment to malnourishederitoneal dialysis patients. Kidney Int 57:1743-1754, 200029. Biesenbach G, Debska-Slizien A, Zazgornik J: Nutritional

tatus in type 2 diabetic patients requiring haemodialysis. Neph-ol Dial Transplant 14:655-658, 1999

30. Julien JP, Combe C, Lasseur C: Subjective global assess-ent of nutrition a useful diagnostic tool for nurses? EDTNARCA J 27:193-196, 200131. Qureshi AR, Alvestrand A, Danielsson A, et al: Factors

redicting malnutrition in hemodialysis patients: A cross-sec-ional study. Kidney Int 53:773-782, 1998

32. Caglar K, Fedje L, Dimmitt R, et al: Therapeutic effectsf oral nutritional supplementation during hemodialysis. Kidneynt 62:1054-1059, 2002

33. Steiber AL, Handu DJ, Cataline DR, et al: The impact ofutrition intervention on a reliable morbidity and mortalityndicator: The hemodialysis-prognostic nutrition index. J Renutr 13:186-190, 200334. Beto JA, Bansal VK, Hart J, et al: Hemodialysis prognostic

utrition index as a predictor for morbidity and mortality inemodialysis patients and its correlation to adequacy of dialysis.ouncil on Renal Nutrition National Research Question Col-

aborative Study Group. J Ren Nutr 9:2-8, 199935. Abdullah MS, Wild G, Jacob V, et al: Cytokines and thealnutrition of chronic renal failure. Miner Electrolyte Metab

3:237-242, 1997