am j clin nutr 2007 secker 1083 9

Subjective Global Nutritional Assessment for children13

Donna J Secker and Khursheed N Jeejeebhoy

ABSTRACTBackground: Subjective Global Assessment (SGA), a method ofnutritional assessment based on clinical judgment, has been widelyused to assess the nutritional status of adults for both clinical andresearch purposes.Objective: Foreseeing benefits of its use in children, we chose toadapt SGA and test its validity and reproducibility in the pediatricpopulation.Design: We prospectively evaluated the preoperative nutritionalstatus of 175 children (aged 31 d to 17.9 y) having major thoracic orabdominal surgery with the use of Subjective Global NutritionalAssessment (SGNA) and commonly used objective measurements.Each child underwent nutritional assessment by 2 independent as-sessors, one performing measurements of anthropometrics andhandgrip strength and one performing SGNA. To test interraterreproducibility, 78 children had SGNA performed by a third asses-sor. Occurrence of nutrition-associated complications was docu-mented for 30 d postoperatively.Results: SGNA successfully divided children into 3 groups (wellnourished, moderately malnourished, severely malnourished) withdifferent mean values for various anthropometric and biochemicalmeasures (P 0.05). Malnourished children had higher rates ofinfectious complications than did well-nourished children (P 0.042). Postoperative length of stay was longer for malnourishedchildren (8.2 10 d) than for well-nourished children (5.3 5.4 d)(P 0.002). No objective nutritional measures showed associationwith outcomes, with the exception of serum albumin, which was notclinically predictive because mean concentrations were in the nor-mal range irrespective of the presence or absence of complications.Conclusion: SGNA is a valid tool for assessing nutritional status inchildren and identifying those at higher risk of nutrition-associatedcomplications and prolonged hospitalizations. Am J Clin Nutr2007;85:10839.

KEY WORDS Malnutrition, children, surgery, nutritional as-sessment, Subjective Global Assessment, SGA, outcomes

INTRODUCTIONCurrent methods of assessing nutritional status in children rely

on a combination of objective anthropometric, dietary, biochem-ical, and immunologic measures. Although epidemiologicallyuseful, these measures have several shortcomings that hampertheir effectiveness in clinical practice, and no single objectivemarker has the sensitivity and specificity to be a reliable index ofprotein-energy malnutrition or predictive of nutrition-relatedcomplications (15).

Almost 25 y ago, Baker et al (6) published a study showing thatnutritional assessment using clinical judgment was a reproduc-ible and valid technique for evaluating the nutritional status ofadults. With the use of a Bayesian analysis of pretest and posttestprobabilities they showed that a structured clinical approach,called Subjective Global Assessment (SGA), was able to predictnutrition-associated complications (NACs; ie, infections, use ofantibiotics, length of stay) better than serum albumin, transferrin,delayed cutaneous hypersensitivity, the Prognostic Nutrition In-dex, creatinine-height index, triceps skinfold thickness, and var-ious combinations of these measures (710). The method of SGAwas since refined (11, 12) and was applied and validated inter-nationally for clinical, epidemiologic, and research purposes(1320).

We hypothesized that SGA could be adapted to identify pre-surgical malnutrition and to predict postsurgical nutrition-associated morbidities that lead to prolonged hospital stay inpediatric patients. The adapted version was called SubjectiveGlobal Nutritional Assessment (SGNA). We then performed aprospective cohort study to determine whether SGNA could bet-ter predict NACs and length of hospital stay than current objec-tive methods of pediatric nutritional assessment.

SUBJECTS AND METHODS

SubjectsConsecutive children scheduled for surgery by surgeons in the

Division of General Surgery at The Hospital for Sick Children, alarge pediatric academic health science center located in To-ronto, Canada, were considered for the study. Children weredeemed eligible if they were aged 31 d to 17.9 y, required majorabdominal or noncardiac thoracic surgery on a nonemergency

1 From the Department of Clinical Dietetics and Division of Nephrology,The Hospital for Sick Children, Toronto, Canada (DJS), and the Division ofGastroenterology, St Michaels Hospital and the Institute of Medical Sci-ences, University of Toronto, Toronto, Canada (KNJ).

2 Supported by the Canadian Foundation for Dietetic Research, the Cana-dian Institutes of Health Research (CIHR) Doctoral Research Award (DJS),the CIHR Clinician Scientist Training Program in Clinical Nutrition (DJS),and the Hospital for Sick Children Research Institute Research TrainingCentre (Restracomp) (DJS). CDC growth charts were provided by AbbottLaboratories, Ltd.

3 Reprints not available. Address correspondence to DJ Secker, Depart-ment of Clinical Dietetics, The Hospital for Sick Children, 555 UniversityAvenue, Toronto, ON, Canada M5G 1X8. E-mail: [email protected].

Received August 17, 2006.Accepted for publication November 27, 2006.

1083Am J Clin Nutr 2007;85:10839. Printed in USA. 2007 American Society for Nutrition

by guest on December 6, 2013

ajcn.nutrition.orgD

ownloaded from

basis, and had not undergone surgery in the 30 d before screening.Surgery was deemed major if it was performed in the operatingroom, involved general anesthesia, and required at least onenights stay in the hospital after the surgical procedure. Subjectswere precluded from study entry if they were preterm infants whohad not reached 31 d corrected age, were clinically unstable, or,along with their caregivers, were non-English speaking (becauseSGNA depended on clear verbal communication with the chil-dren and caregivers). Written consent was obtained from theresponsible caregivers of children aged 16 y or directly fromadolescents16 y. Informed assent was obtained from childrenaged 7-15.9 y. The study was approved by the Research EthicsBoard of The Hospital for Sick Children.

Study designFive staff dietitians served as assessors for the study after

training for standardized performance of objective measure-ments and the SGNA. Study results are not based on a singleassessor but are a composite of the data from the 5 differentassessors, who performed the assessments in random order. Eachchild underwent nutritional assessment by 2 independent asses-sors blinded to the results of the other, one performing measure-ments of anthropometrics and handgrip strength, and one per-forming the SGNA. To test interrater reproducibility of theSGNA, a subset of 78 children were seen by a third assessor whoalso performed the SGNA. Nutritional assessments were per-formed the day before surgery for prior-day admissions or in-patients and the morning of surgery for same-day admissions;assessments took place in the Clinical Investigation Unit, thepatients individual hospital room, or the General Surgery Same-Day Admission Unit. A blood sample for measurement of serumalbumin and transferrin was obtained coincidental with routinepreoperative blood work (ie, complete blood count) or by theanesthetist at the time of intravenous induction. Children werefollowed in a purely observational fashion, from the day of sur-gery until 30 d after surgery, for the development of nutrition-associated infectious and noninfectious morbidities related totheir surgical procedure. Medical and surgical risk factors thatmight influence the occurrence of complications and outweighnutritional risk were documented. Data were accrued prospec-tively and entered into a database (MICROSOFT ACCESS2000, version 9.0; Microsoft Corp, Redmond, WA) and thenpartially exported to NUTSTAT (Epi Info, version 3.3.2; Centersfor Disease Control and Prevention, Atlanta, GA) for calculationof age, body mass index (in kg/m2), and anthropometric z scores.Analyses compared the ability of SGNA and anthropometric andbiochemical measures to identify malnourished children andthose at increased risk of developing complications, as well as therate of agreement between the pairs of dietitians who performedSGNA on the same child.

Subjective Global Nutrition AssessmentAssessors used a SGNA questionnaire to guide their interview

and to gather information on the childs recent and current heightand weight history as well as parental heights, dietary intake(type, volume, and frequency of liquid and solid feedings forinfants and frequency of eating and a brief description of a typicaldays intake for children; rating of appetite and recent changes;feeding or eating problems; diet restrictions), frequency and du-ration of gastrointestinal symptoms (loss of appetite, vomiting,

diarrhea, and constipation for all ages plus stomach pain andnausea for children), and current functional capacity and recentchanges (alertness, amount of energy or activity for infants andschool attendance, ability to run and play games or sports withfriends, and time sleeping for children). They then performed thenutrition-related physical examination, looking at specific siteson the body for signs of fat and muscle wasting (combined asgeneral wasting for infants and young toddlers), as well asedema. Finally, they considered the presence or absence of thespecified historical features, metabolic demands of the underly-ing condition, and physical signs associated with malnutrition tocome to a global assessment of the patients nutritional status,assigning a global rating of well-nourished, moderately malnour-ished, or severely malnourished. A rigid scoring system based onspecific criteria was not used.

Objective nutritional assessmentWe examined the following objective markers of nutritional

status: length or height; weight; percentage of ideal body weightfor height; body mass index-for-age; midarm circumference;triceps skinfold thickness; midarm muscle area; handgripstrength; concentrations of serum albumin, transferrin, and he-moglobin; and total lymphocyte count. Prealbumin measure-ment was unavailable at our center and therefore was not in-cluded. Measurements were performed with the use of calibratedequipment and standardized techniques (2123). Anthropomet-ric z scores were calculated with the use of NUTSTAT anthro-pometric software package (Epi Info, version 3.2.2; Centers forDisease Control and Prevention) (24) or common reference data(25, 26). Preterm infants (n 22) were assessed with the use ofcorrected age up to age 2 y.

Biochemical analysis was performed by the hospitals De-partment of Pediatric Laboratory Medicine with the use of thebromocresol green dye binding method for serum albuminand immunonephelometry for serum transferrin. The z scoresfor hemoglobin and transferrin were calculated from datafrom the American National Health and Nutrition Examina-tion Survey (27).

Outcome measuresPatients were followed for 30 d after surgery for the develop-

ment of NACs related to their surgical procedure. Complicationswere subdivided into infectious and noninfectious complica-tions, as well as major and minor complications, according topublished objective criteria (2831). Secondary outcomes werelength of postoperative hospitalization, use of nonprophylacticantibiotics, and unplanned rates of reoperation and readmission.Monitoring after discharge continued until 30 d after surgery andoccurred by either a routine follow-up clinic visit or telephoneinterview shortly thereafter.

Statistical analysesAccording to previous observations (6, 32), we planned to

enroll 175 children for the study to have the capacity to detect a3-fold difference in incidence of postoperative complicationsbetween the malnourished and well-nourished groups (two-sided level: 0.05; level: 0.8). To examine the relation amongSGNA and individual objective measurements, means of theobjective measures for the 3 groups defined by SGNA were

1084 SECKER AND JEEJEEBHOY


ajcn.nutrition.orgD

ownloaded from

tested with analysis of variance or the Kruskal-Wallis nonpara-metric test and Tukeys honest significant difference or theMann-Whitney nonparametric test for post hoc analysis. Asso-ciations among numerical equivalents of the SGNA groups withobjective measures were tested by correlation analysis(Kendalls ). Correlation analysis (Kendalls ) and multino-mial logistic regression were used to explore the relation ofindividual components of SGNA with the final SGNA rating.The association between SGNA and clinical morbidity wastested with the use of the chi-square test or Kendalls test forcategorical variables and the Kruskal-Wallis and Mann-WhitneyU nonparametric tests for length of stay. Relations between ob-jective measures of nutritional status and morbidities were ex-amined by the Students t test, Mann-Whitney U nonparametrictest, or correlation analysis (Kendalls ).

Analysis was performed with the use of SPSS 14.0 forWINDOWS GRADPACK (SPSS Inc, Chicago, IL). Data col-lection, entry, and analysis were performed by the principal au-thor (DJS). Results are reported as means SDs, unless statedotherwise. Results were considered significant at P 0.05.

RESULTSBetween February 2003 and August 2004, 180 children were

enrolled and completed the study. One hundred seventy-fivechildren (99 males, 76 females) ranging in age from 31 d to 17.9 y(x SD: 8.1 6.1 y) made up the final study group. Althoughheterogeneous in terms of age, distribution by age group wasfairly even (Table 1). Data from 5 children were excluded fromfinal analysis for either discharge within 24 h of surgery (n 4)or preemptive surgery before the assessments were completed(n 1). Because of an outbreak of severe acute respiratorysyndrome, study activity was held for an 8-wk period betweenMarch and May 2003.

Almost half of the children (n 80, 46%) had one or moredisease processes or anomalies [eg, inflammatory bowel disease;Hirschsprung disease; vertebral, anal, cardiac, tracheal, esopha-geal, renal, and limb anomalies (VACTERL association; a ran-dom association of birth defects); cerebral palsy; Down syn-drome; cystic fibrosis; sickle cell disease]. Sixteen children (9%)had cancer. On a 6-point scale of medical complexity [AmericanSociety of Anesthesiologists (ASA) classification of physicalstatus score (33)], with 1 being the lowest complexity and 6 beingthe highest, the majority of children were rated preoperatively bythe anesthetist as either ASA 2 (47%) or ASA 3 (32%). All but 8children (5%) received prophylactic antibiotics before surgery.The most commonly performed surgical procedures are listed inTable 1. Distribution of surgical procedures classified by risk ofmicrobial infection (clean, clean-contaminated, contaminated,dirty) did not differ across the 3 SGNA groups (P 0.88) nor didthe proportion of children with cancer (P 1.0).

Concurrent validity of SGNANutritional status, as determined by SGNA, was compared

with objective measurements. With the use of SGNA, 85 children(49%) were classified as well-nourished, 64 children (36%) wereclassified as moderately malnourished, and 26 children (15%) asseverely malnourished. The 3 groups of children had mean agesthat did not differ, but they had significantly different mean

values for 7 of the 8 anthropometric measures, including hand-grip strength and serum albumin, but not transferrin, hemoglo-bin, or total lymphocyte count (Table 2). It should be noted that,although the mean serum albumin concentrations of the 3 groupsof children classified by SGNA were statistically different, theywere not clinically different in that even the lowest mean serumalbumin (ie, the severely malnourished group, 38 7 g/L) fellwell within the hospitals normal reference range. SGNA, withthe use of numerical equivalents of the clinical groups, showedmoderate-to-fair correlation with all of the objective anthropo-metric and biochemical measurements of nutritional status, ex-cept transferrin z score and total lymphocyte count (Table 2).Handgrip strength, as a surrogate marker of muscle function andimpaired functional status, decreased with worsening nutritionalstatus (P 0.001).

Predictive validity of SGNASeventy children had at least one complication, experiencing

97 complications in total; no deaths occurred. The majority ofcomplications were infectious in origin (80%), with an equalpercentage being major (49%) or minor (51%) in nature. Occur-rence of complications did not differ by sex or by age, with the

TABLE 1Characteristics of the study population (n 175)

Value

Sex [n (%)]Male 99 (57)Female 76 (43)

Age (y) 8.1 6.1 (0.0817.9)131 d2 y [n (%)] 51 (29)25 y [n (%)] 22 (13)512 y [n (%)] 45 (26)1218 y [n (%)] 57 (32)

ASA score [n (%)]21) Normally healthy patient 25 (16)2) Patient with mild systemic disease 73 (47)3) Patient with severe systemic

disease that is not incapacitating51 (32)

4) Patient with incapacitatingsystemic disease that is a constantthreat to life

8 (5)

5) Moribund patient not expected tosurvive for 24 h with or withoutoperation

0

6) Patient declared brain dead whoseorgans are being removed fordonation

0

Primary surgical procedures [n (%)]Bowel resection, abdominal pull-

through, ostomy closure43 (25)

Excision of tumors and cysts 18 (10)Cholecystectomy 13 (7)Fundoplication 11 (6)Interval appendectomy 10 (6)Pectus repair 9 (5)Lung lobectomy 5 (3)Other 66 (38)

1 x SD; range in parentheses.2 American Society of Anesthesiologists (ASA) classification of phys-

ical status score (32). Higher score reflects more severe illness. Eighteen of175 children did not have an ASA score recorded by the anesthetist.

PEDIATRIC SUBJECTIVE GLOBAL NUTRITIONAL ASSESSMENT 1085


ajcn.nutrition.orgD

ownloaded from

exception of minor complications, which occurred more often inyounger children. A breakdown of complications is provided inTable 3. Children identified by SGNA as malnourished (mod-

erately or severely) experienced a higher rate of infectious com-plications (P 0.04) and minor (infectious and noninfectious)complications (P 0.03) compared with children classified aswell-nourished (Table 4).

Postoperative length of stay was more than twice as long forseverely malnourished children (19.0 58.8 d) compared withwell-nourished (5.3 5.4 d) and moderately malnourished(8.4 11.1 d) children (P 0.002, Kruskal-Wallis nonparamet-ric test). Mean postoperative hospitalization for severely mal-nourished children was skewed by 1 child whose postoperativestay was 306 d. Although unusual, an extreme length of stay suchas this does occur in malnourished patients and was not seen in

TABLE 2Relation between Subjective Global Nutritional Assessment (SGNA) and anthropometric measures of nutritional status1

No. ofsubjects

SGNA classification Significance

Well nourishedModerately

malnourishedSeverely

malnourished P2 r3P for

association

n

Age (y) 175 8.3 6.2 (85)4 8.2 6.2 (64) 7.2 6.0 (26) 0.621 0.033 0.577Weight-for-age (z score) 175 0.12 1.20a (85) 0.68 1.23b (64) 1.88 1.18c (26) 0.001 0.387 0.001Height-for-age (z score) 175 0.06 1.13a (85) 0.51 1.22a (64) 1.47 1.57b (26) 0.001 0.258 0.001Ideal body weight (%) 175 103.1 16.0a (85) 98.3 17.2b (64) 92.0 14.8b (26) 0.002 0.213 0.001BMI-for-age (z score) 124 0.19 1.15a (59) 0.30 1.25a (47) 1.21 1.15b (18) 0.001 0.280 0.001Midarm circumference (z score) 154 0.25 1.38a (76) 0.52 1.07b (58) 1.28 0.81c (20) 0.001 0.342 0.001Tricep skinfold thickness (z score) 135 0.35 1.43 (67) 0.10 0.84 (51) 0.38 0.73 (19) 0.065 0.167 0.014Midarm muscle area (z score) 137 0.11 1.17a (67) 0.55 1.09b (51) 1.27 0.82c (19) 0.001 0.333 0.001Handgrip (z score) 91 0.69 1.38a (49) 1.22 0.96a (32) 2.12 0.99b (10) 0.001 0.287 0.001Albumin (g/L) 171 41 4a (83) 40 6ab (62) 38 7b (26) 0.027 0.156 0.011Transferrin (z score) 109 0.24 1.47 (55) 0.46 2.05 (39) 0.22 1.75 (15) 0.925 0.014 0.841Hemoblobin (z score) 154 0.46 1.60 (76) 1.03 1.87 (56) 1.19 2.08 (22) 0.100 0.136 0.032Total lymphocyte count (109/L) 110 4.3 3.1 (49) 3.9 3.3 (42) 4.2 3.4 (19) 0.681 0.051 0.499

1 Values in the same row with different superscript letters are significantly different, P 0.05 (Tukeys honestly significant difference or Mann-Whitneynonparametric test, as appropriate).

2 P values were determined with the use of ANOVA or the Kruskal-Wallis nonparametric test.3 Kendalls correlation coefficient was used for ordinal data.4 x SD; n in parentheses (all such values).

TABLE 3Breakdown of postoperative complications

No. ofepisodes Children

n (% of sample)Major infectious complication 39 36 (21)

Bacteremia or fungemia 4Clinical sepsis 2Severe surgical site infection (deep

incisional, organ, or space)5

Pneumonia 8Significant surgical site infection (superficial

incisional)18

Major symptomatic urinary tract infection 1Major febrile atelectasis 1

Major noninfectious complication 9 9 (5)Anastomotic stricture 1Gastrointestinal bleeding 1Gastrointestinal obstruction 6Persistent air leak 1

Minor infectious complication 39 37 (21)Gastroenteritis 12Secondary surgical site infection (superficial

incisional)9

Urinary tract infection 4Eye, ear, nose, throat, mouth infection 4Lower respiratory tract infection 2Skin infection (ie, candida) 2Minor early sepsis 6

Minor noninfectious complication 10 8 (5)Atelectasis 5Pleural effusion 5

TABLE 4Relation between malnutrition and outcomes

Outcome

Groups1

P2Well-nourished

(n 85)Malnourished

(n 90)n (%)

Infectious complications2 24 (28) 39 (43) 0.042Noninfectious complications2 7 (8) 10 (11) 0.614Major complications2 16 (19) 27 (30) 0.114Minor complications2 14 (17) 28 (31) 0.033Nonprophylactic antibiotic use2 27 (32) 34 (38) 0.404Unplanned reoperation2 5 (6) 6 (7) 0.831Unplanned readmission2 5 (6) 12 (13) 0.096Postoperative length of stay (d)3 5.3 5.4 8.2 10 0.001

1 Data presented as the number of children with that outcome; % withinnutritional classification in parentheses.

2 P values were determined with the use of Pearson chi-square test forall outcomes except postoperative length of stay, which was tested with theMann-Whitney nonparametric test for independent means.

3 x SD; 1 child with severe malnutrition and a postoperative stay of306 d was removed from this analysis as an extreme outlier.



ajcn.nutrition.orgD

ownloaded from

our well-nourished population. When this child was excluded asan extreme outlier, the overall difference in mean stay across the3 groups remained significant (P 0.02, Kruskal-Wallis). Whenthe moderately and severely malnourished children weregrouped together, mean length of postoperative hospitalizationwas 2.9 d (55%) longer than for the well-nourished children (P0.001) (Table 4). The percentages of children receiving nonpro-phylactic antibiotics or requiring unplanned reoperation or read-mission were not significantly different between the well-nourished and malnourished children (Table 4).

No association was observed among any of the objective mea-sures of nutritional status and outcomes, with the exception ofheight-for-age and postoperative length of stay (r0.18, P0.001), and serum albumin, which was associated with infectious(no complication, 41.4 4.5 g/L; complication, 38.2 5.9 g/L;P 0.001) and minor (no complication, 40.8 5.2 g/L; com-plication, 38.65.3 g/L; P0.01) complications, postoperativelength of stay (r 0.20, P 0.001), and use of nonprophy-lactic antibiotics (no antibiotic, 41.44.5 g/L; antibiotic, 38.25.9 g/L; P 0.001). However, it must be emphasized that meanserum albumin concentrations were within the normal referencerange for all 3 nutritional groups; therefore, serum albumin wasclinically not useful in identifying malnutrition or predictingoutcomes.

Interobserver reproducibility of SGNAInterrater agreement occurred in 44 (56%) of 78 children hav-

ing duplicate SGNA assessments (unweighted : 0.28). Thisrepresents only fair agreement between the assessors (34, 35).More discrepancies occurred between classifications of normal-to-moderate malnutrition (22 of 34; 65%) than moderate-to-severe malnutrition (11 of 34; 32%).Individual components of SGNA and their contributionto the overall SGNA rating

Each of the individual historical and physical characteristicswas univariately correlated with the overall SGNA ratings. Forinfants and toddlers, the individual SGNA components that mostinfluenced the overall SGNA rating were physical signs of wast-ing, gastrointestinal symptoms, and the metabolic stress of theunderlying disease. For the older age groups, controlling for age,the most influential components were physical evidence of fatwasting, serial weight loss, gastrointestinal symptoms, and stunt-ing of height.

DISCUSSIONThis study shows that SGNA is capable of identifying malnu-

trition, NACs, and increased length of stay in hospitalized chil-dren. As such, it is a useful clinical tool that links nutritionalstatus to outcome and distinguishes children at risk of an adverseoutcome and prolonged hospitalization. Although serum albu-min showed a significant relation to the occurrence of compli-cations, it was abnormal in only 11 children and was clinicallynondiscriminatory because the mean concentrations for the 3nutritional groups were in the normal range.

The relation of SGNA with commonly used objective mea-surements of nutritional status was as good as that reported instudies of SGA in adult surgical patients (6, 10, 36). A commonlyoverlooked marker of nutritional status is functional capacity andmuscle power (37). With the use of maximal handgrip strength as

a surrogate measure of muscle function, we showed that mal-nourished children had physical evidence of lower functionalcapacity than did children who were well nourished, a relationnot previously shown. Hand dynamometry was easy to performand the children enjoyed testing their strength.

Children classified by SGNA as malnourished experiencedmore frequent infectious complications than well-nourishedchildren as well as more minor infectious and noninfectiouscomplications combined. None of the objective anthropometricor biochemical markers of nutritional status, aside from albumin,showed association with complications or other morbidity. Oth-ers have also found a lack of association with biochemical mark-ers of visceral proteins or immune function that have been said tobe useful techniques of assessing nutritional status (6, 36) but arenow thought to be indicators of inflammation or infection ratherthan malnutrition (1, 38, 39).

The finding of longer postoperative stays for children identi-fied by SGNA as malnourished is particularly important becauseextended hospitalization translates into increased hospital costswhere, at our center, the average cost of an inpatient day on thegeneral surgery ward is $587. On the basis of the lengths ofpostoperative stay observed in this study, the average malnour-ished child would stay 2.9 d (55%) longer than a well-nourishedchild and generate added costs of $1702. Adult studies have alsoreported significantly longer postoperative hospitalization formalnourished persons (6, 4042) and associated increases inhospital and home-care costs (42, 43).

In contrast with the other objective measurements, serum al-bumin was associated with infectious and minor complicationsand 2 of the 4 markers of morbidity. Although correlated withoutcomes, serum albumin concentrations were well within thenormal reference range for even the severely malnourished chil-dren and could not have been used clinically to identify malnu-trition. Only 11 (6%) of 171 children were hypoalbuminemic. All11 children were identified by SGNA as malnourished; hence,the addition of albumin to SGNA would not have identified anyadditional children as malnourished.

Concordance between assessors performing duplicate SGNAin this study was lower than desired and lower than reported inoriginal studies of SGA in adults (agreement range: 7991%; :0.660.78) (6, 10, 36). Considerable variability was observedamong the 5 dietitian assessors in terms of their years of work andpediatric experience. Unlike the original SGA studies, in whichassessors were members of the gastroenterology team who fre-quently performed rounds and discussed patients and therapytogether, none of the dietitians had previous experience with thegeneral surgery population. Training has perhaps the most sig-nificant impact on concordance or reproducibility. After review-ing all factors, we believe that more hands-on refresher trainingseveral times throughout the study would have resulted in im-proved rates of interrater agreement. Despite reduced concor-dance based on formal testing, SGNA performed by 5 differentassessors nevertheless identified children likely to have compli-cations and prolonged hospital stay and therefore is likely to beclinically useful for this purpose. In contrast, currently usedtechniques for identifying malnutrition were not able to detect therisk of complications or longer length of stay.

Whereas objective measures of nutritional status cannot takeinto account all of the variables that a clinician should considerto identify malnutrition, SGNA represents an explicit model ofthe thought process to be used in assessing nutritional status. It



ajcn.nutrition.orgD

ownloaded from

directs the clinician to piece together the nutritional componentsto make a complete picture; that is, Is this child short or thin? Isit by nature or because of inadequate intake? If intake is inade-quate, what is the cause? Is there physical evidence of wasting tosupport the historical findings? Are losses of body mass affectingthe childs ability to perform? Is the childs nutritional status andfunctioning worsening or beginning to improve? Is the childlikely to continue to have problems? What needs to be treated? Asa result, after completing the SGNA and identifying the malnour-ished child, the clinician has determined whether the childsnutritional status is likely to improve or worsen, has establishedthe potential cause or causes of malnutrition, and identifiedwhere to target intervention.

A possible source of bias in our study is our exclusion ofnon-Englishspeaking children and caregivers; however, it isunlikely that this would have altered our conclusions about theability of SGNA to predict the risk of morbidity, because the biaswould potentially reduce the numbers of severely malnourishedchildren that we saw. Because such a reduction would make itmore difficult to show statistical differences in outcomes, theinclusion of such patients would only reinforce our findings.

In conclusion, SGNA and its use of clinical judgment is a validmethod for assessing nutritional status in children. SGNAshowed good correlation with currently used objective nutri-tional measures of nutritional status and was the only true nutri-tional measure that identified children with higher incidence ofinfectious and minor complications and longer postoperativehospital stays, undesirable consequences for children and fami-lies that also increase hospital costs. Is SGNA valid in all chil-dren? We suggest that SGNA will be a useful tool in a widevariety of children with chronic and systemic conditions, giventhat our study included children with mild-to-moderate medicalcomplexity and a wide assortment of comorbidities and under-lying illnesses, most of which involved the gastrointestinal tract.Further study is needed to determine whether the fair degree ofreproducibility that we observed in this study was a function ofthe heterogeneity of the patients, the assessors in our study, orboth and whether ongoing training throughout the study periodcan effectively improve interrater agreement.

We thank the children and their families for their participation in thisresearch. In addition, we thank Joann Herridge, Laura Stefanizzi, SarahFarmer, and Veronika Langos for participating as assessors for the study andthe staff of the Divisions of General Surgery and Anaesthesiology of TheHospital for Sick Children for their cooperation throughout.

DJS conceived the hypothesis and study design, conducted the statisticalanalyses, and drafted the manuscript. KNJ contributed to the concept anddesign of the study and interpretation and discussion of the results. Bothauthors critically revised the drafted manuscript. Neither author had anyconflict of interest.

REFERENCES1. Seres D. Surrogate nutrition markers, malnutrition, and adequacy of

nutrition support. Nutr Clin Pract 2005;20:30813.2. Zemel BS, Riley EM, Stallings VA. Evaluation of methodology for

nutritional assessment in children: anthropometry, body composition,and energy expenditure. Annu Rev Nutr 1997;17:21135.

3. Figueroa-Colon R. Clinical and laboratory assessment of the malnour-ished child. In: Suskind RM, Lewinter-Suskind L, eds. Textbook ofpediatric nutrition. 2nd ed. New York, NY: Raven Press Ltd, 1993:191205.

4. Mascarenhas MR, Zemel B, Stallings VA. Nutritional assessment inpediatrics. Nutrition 1998;14:10515.

5. Taylor Baer M, Bradford Harris A. Pediatric nutrition assessment: iden-tifying children at risk. J Amer Diet Assoc 1996;16:714.

6. Baker JP, Detsky AS, Wesson DE, et al. Nutritional assessment: a com-parison of clinical judgment and objective measurements. N Engl J Med1982;306:96972.

7. Baker JP, Detsky AS, Whitwell J, Langer B, Jeejeebhoy KN. A com-parison of the predictive value of nutritional assessment techniques.Hum Nutr Clin Nutr 1982;36:23341.

8. Jeejeebhoy KN, Baker JP, Wolman SL, et al. Critical evaluation of therole of clinical assessment and body composition studies in patients withmalnutrition and after total parenteral nutrition. Am J Clin Nutr 1982;35(suppl):111727.

9. Detsky AS, Baker JP, Mendelson RA, Wolman SL, Wesson DE,Jeejeebhoy KN. Evaluating the accuracy of nutritional assessment tech-niques applied to hospitalized patients: methodology and comparisons.JPEN J Parenter Enteral Nutr 1984;8:1539.

10. Detsky AS, Baker JP, ORourke K, et al. Predicting nutrition-associatedcomplications for patients undergoing gastrointestinal surgery. JPEN JParenter Enteral Nutr 1987;11:4406.

11. Detsky AS, McLaughlin JR, Baker JP, et al. What is subjective globalassessment of nutritional status? JPEN J Parenter Enteral Nutr 1987;11:813.

12. Detsky AS, Smalley PS, Chang J. Is this patient malnourished? JAMA1994;271:548.

13. Ulander K, Grahn G, Jeppsson B. Subjective assessment of nutritionalstatus-validity and reliability of a modified Detsky index in a Swedishsetting. Clin Nutr 1993;12:139.

14. Niyongabo T, Melchior JC, Henzel D, Bouchaud O, Larouze B. Com-parison of methods for assessing nutritional status in HIV-infectedadults. Nutrition 1999;15:7403.

15. Duerksen DR, Yeo TA, Siemens JL, OConnor MP. The validity andreproducibility of clinical assessment of nutritional status in the elderly.Nutrition 2000;16:7404.

16. Stephenson GR, Moretti EW, El-Moalem H, Clavien PA, Tuttle-Newhall JE. Malnutrition in liver transplant patients: preoperative sub-jective global assessment is predictive of outcome after liver transplan-tation. Transplantation 2001;72:66670.

17. Wyszynski DF, Perman M, Crivelli A. Prevalence of hospital malnutri-tion in Argentina: preliminary results of a population-based study. Nu-trition 2003;19:1159.

18. Waitzberg D, Caiaffa W, Correia I. Hospital malnutrition: the BrazilianNational Survey (IBRANUTRI): a study of 4000 patients. Nutrition2001;17:57380.

19. Steiber A, Kalantar-Zadeh K, Secker D, McCarthy M, Sehgal A,McCann M. Subjective Global Assessment in chronic kidney disease: areview. J Ren Nutr 2004;14:191200.

20. Gupta D, Lammersfeld C, Vashi P, Burrows J, Lis C, Grutsch J. Prog-nostic significance of Subjective Global Assessment (SGA) in advancedcolorectal cancer. Eur J Clin Nutr 2005;59:3540.

21. US Department of Health and Human Services, Maternal and ChildHealth Bureau. Using the CDC Growth Charts: accurately weighing andmeasuring: technique [electronic training module]. 2001. Internet:http://www.cdc.gov/growthcharts/ (accessed 15 August 2006).

22. US Department of Health and Human Services, Maternal and ChildHealth Bureau. Using the CDC Growth Charts: accurately weighing andmeasuring: equipment [electronic training module]. 2001. Internet:http://www.cdc.gov/growthcharts/ (accessed 15 August 2006).

23. Mathiowetz V, Weber K, Volland G, Kashman N. Reliability and va-lidity of grip and pinch strength evaluations. J Hand Surg 1984;9A:2226.

24. US Department of Health and Human Services, Centers for DiseaseControl and Prevention, Epidemiology Program Office, Division of Pub-lic Health Surveillance and Informatics. NutStat: a nutritional anthro-pometry program. Epi Info Version 3.2.2. 2005. Internet: http:www.cdc.gov/epiinfo/MANUAL/NutStat.htm (accessed 15 August 2006).

25. Mathiowetz V, Wiemer DM, Federman SM. Grip and pinch strength:norms for 6- to 19-year olds. Am J Occup Ther 1986;40:70511.

26. Frisancho A. Anthropometric standards for the assessment of growth andnutritional status. Ann Arbor, MI: The University of Michigan Press,1990.

27. Hematological and nutritional reference data for persons 6 months to 74years of age: United States, 197680. Vital Health Stat 11 1982;(232):ivi, 1173.



ajcn.nutrition.orgD

ownloaded from

28. Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. CDC defini-tions for nosocomial infections. In: Olmsted RN, ed. APIC infectioncontrol and applied epidemiology: principles and practice. St Louis,MO: Mosby, 1996:A-1A-20. Internet: http://www.apic.org/pdf/cdcdefs.pdf (accessed 15 August 2006).

29. Mangram AJ, Horan TC, Pearson MI, Silver LC, Jarvis WR. Guidelinefor prevention of surgical site infection, 1999. Hospital Infection ControlPractices Advisory Committee. Infect Control Hosp Epidemiol 1999;20:25078.

30. Buzby GP, Williford WO, Peterson OL, et al. A randomized clinical trialof total parenteral nutrition in malnourished surgical patients: the ratio-nale and impact of previous clinical trials and pilot study on protocoldesign. Am J Clin Nutr 1988;47:35765.

31. The Veterans Affairs Total Parenteral Nutrition Cooperative StudyGroup. Perioperative total parenteral nutrition in surgical patients.N Engl J Med 1991;325:52532.

32. Jevsevar DS, Karlin LI. The relationship between preoperative nutri-tional status and complications after an operation for scoliosis in patientswho have cerebral palsy. J Bone Joint Surg Am 1993;75:8804.

33. American Society of Anesthesiologists. New classification of physicalstatus. Anesthesiology 1963;24:111.

34. Portney L, Watkins M. Statistical measures of reliability. In: Foundationof clinical research: applications to practice. 2nd ed. Norwich, CT:Appleton and Lange, 2000:55786.

35. Donner A, Eliasziw M. Sample size requirements for reliability studies.Stat Med 1987;6:4418.

36. Hirsch S, de Obaldia N, Petermann M, et al. Subjective global assess-ment of nutritional status: further validation. Nutrition 1991;7:358.

37. Klein S, Kinney J, Jeejeebhoy K, et al. Nutrition support in clinicalpractice: review of published data and recommendations for future re-search directions. National Institutes of Health, American Society forParenteral and Enteral Nutrition, and American Society for ClinicalNutrition. JPEN J Parenter Enteral Nutr 1997;21:13356.

38. Fuhrman M, Charney P, Mueller C. Hepatic proteins and nutrition as-sessment. J Am Diet Assoc 2004;104:125864.

39. Lark R, Williams C, Stadler D, et al. Serum prealbumin and albuminconcentrations do not reflect nutritional state in children with cerebralpalsy. J Pediatr 2005;147:6957.

40. Kyle UG, Pirlich M, Schuetz T, Lochs H, Pichard C. Is nutritionaldepletion by nutritional risk index associated with increased length ofhospital stay? A population-based study. JPEN J Parenter Enteral Nutr2004;28:99104.

41. Naber TH, Schermer T, de Bree A, et al. Prevalence of malnutrition innonsurgical hospitalized patients and its association with disease com-plications. Am J Clin Nutr 1997;66:12329.

42. Reilly J, Hull S, Albert N, Waller A, Bringardener S. Economic impactof malnutrition: a model system for hospitalized patients. JPEN JParenter Enteral Nutr 1988;12:3716.

43. Chima C, Barco K, Dewitt M, Maeda M, Teran J, Mullen K. Relationshipof nutritional status to length of stay, hospital costs, and discharge statusof patients hospitalized in the medicine service. J Am Diet Assoc 1997;97:9758.



ajcn.nutrition.orgD

ownloaded from

am j clin nutr 2007 secker 1083 9

Documents