I

O

Crd

H

a

b

S

0d

ntensive and Critical Care Nursing (2011) 27, 194—201

a va i la b le at www.sc iencedi rec t .com

jo ur n al homepage: www.elsev ier .com/ iccn

RIGINAL ARTICLE

hanges in nutritional status in ICU patientseceiving enteral tube feeding: A prospectiveescriptive study

yunjung Kima, Smi Choi-Kwonb,∗

University of California San Francisco, School of Nursing, CA, United StatesSeoul National University, Research Institute of Nursing Science, College of Nursing, 28 Yeongeon-dong, Jongro-gu, Seoul,outh Korea

Accepted 6 May 2011

KEYWORDSEnteral nutrition;Intensive care units;Nutritional status

SummaryObjectives: This study aimed to assess the changes in nutritional status in Korean ICU patientsreceiving enteral feeding, and to understand the contribution of baseline nutritional status andenergy intake to nutritional changes during the ICU stay.Methods: This was a prospective study of nutritional changes in 48 ICU patients receiving enteralfeeding for 7 days. The Subjective Global Assessment scale was used upon admission. In addition,anthropometric measures (triceps skinfold thickness, mid-arm circumference, mid-arm musclecircumference, body mass index and percent ideal body weight) and biochemical measures(albumin, prealbumin, transferrin, haemoglobin and total lymphocyte count) were evaluatedtwice, upon admission and 7 days after admission.Results: Seventy-five percent of ICU patients were severely malnourished at admission. Althoughthe nutritional status worsened in both the patients with suspected malnourishment and thepatients with severe malnutrition at admission, the nutritional status worsened significantlymore in the patients with severe malnutrition than in the patients with suspected malnour-ishment. Moreover, a number of nutritional measures significantly decreased more in underfedpatients than in adequately fed patients. The most significant predicting factor for underfeedingwas under-prescription.Conclusion: The ICU patients in our study were severely malnourished at admission, and their

nutritional status worsened during their ICU stay even though enteral nutritional support wasprovided. The changes in nutritional status during the ICU stay were related to the patients’baseline nutritional status and underfeeding during their ICU stay. This study highlights an urgentneed to provide adequate nutrit© 2011 Elsevier Ltd. All rights re

∗ Corresponding author. Tel.: +82 2 740 8830.E-mail address: smi@snu.ac.kr (S. Choi-Kwon).

964-3397/$ — see front matter © 2011 Elsevier Ltd. All rights reserved.oi:10.1016/j.iccn.2011.05.002

ional support for ICU patients.

served.

al tu

Ir

S

AaIerdstwsfeuTfpow

D

AfssEh

NFSam1t(meteamascmaotaab

Changes in nutritional status in ICU patients receiving enter

Introduction

Malnutrition is a common problem in hospitalised patients.More than 35% of intensive care unit (ICU) patients arereported to be malnourished upon admission to the ICU(Nisim and Allins, 2005; Sungurtekin et al., 2008). More-over, two-thirds of 200 patients experience worsening oftheir nutritional status during their hospital stay (Barr et al.,2004). Similarly, malnutrition is highly prevalent in patientshospitalised in Korean ICU, occurring in up to 87% of patients(Lee et al., 2003).

Malnutrition can lead to a higher risk of infection(Krishnan et al., 2003), development of pressure ulcers(Thomas, 2001), and reduced wound healing (Jagoe et al.,2001), which can result in prolonged hospital stays(Rodriguez, 2004) and increased healthcare costs (Martinet al., 2004). Eventually, malnutrition can increase morbid-ity and mortality (Martin et al., 2004).

Although enteral nutrition is the preferred method fornutritional support in ICU patients, patients with enteralfeeding are at great risk of malnutrition due to insufficientnutritional intake (Parrish, 2003). This may be in part due toinsufficient prescription (Krishnan et al., 2003), incompletedelivery of prescribed enteral nutrition (Kim et al., 2010;O’Leary-Kelley et al., 2005; Reid, 2006) or gastrointesti-nal (GI) intolerance of tube feeding (Petros and Engelmann,2006). Furthermore, the patient’s baseline nutritional sta-tus assessed on admission may have a negative influence onchanges in nutritional status during the ICU stay (Barr et al.,2004).

To date, however, there have been few attempts to studythe changes in nutritional status in patients with enteralnutrition, particularly so with relation to the nutritional sta-tus at admission. Furthermore, there is little data aboutmalnutrition in Korean ICU patients. Nothing is known aboutthe factors that contribute to the nutritional changes inthis population. Therefore, evaluating the nutritional statusat admission and monitoring the nutritional changes duringthe ICU stay are essential for developing an evidence-basedenteral nutrition protocol to improve nutritional status inKorean ICU patients.

The aims of the current study, therefore, were to assessthe nutritional status of patients receiving enteral tubefeeding in the ICU at admission and to evaluate its effectson nutritional status over the 7 days after admission. Thisstudy also aimed to understand the contribution of energyintake during hospitalisation to the changes in nutritionalstatus during the ICU stay.

Methods

Study design

This was a prospective, descriptive study of nutritionalchanges in enterally fed ICU patients. Patients wererecruited from the medical ICU of Seoul National Univer-

sity Hospital (SNUH) in South Korea. Data was collectedfrom September 2003 to October 2004. The study obtainedapproval from the Institutional Review Board at theSeoul National University College of Nursing (Ref:2010-13).

taba

be feeding: A prospective cohort study 195

nformed consent was obtained from patients or legal sur-ogates of patients.

ample

dult medical patients (≥18 years of age) who had beendmitted to the medical ICU were screened for eligibility.nclusion in the study was restricted to patients who startednteral tube feeding after admission to the ICU, had noteceived preoperative or postoperative care, did not haveo-not-resuscitate orders, had received nothing by mouthince admission, and had not received total parenteral nutri-ion. A total of 80 patients meeting the eligibility criteriaere initially enrolled in this study, but 32 patients who

topped enteral feeding were excluded from the analysisor the following reasons: they were changed to oral or par-nteral nutrition (n = 14), transferred or discharged from thenit (n = 10), or expired (n = 8) within the 7 day study period.herefore, 48 patients who received enteral tube feedingor 7 days were included in this study. The data from theatients excluded was used only to compare subjective andbjective nutritional status on admission between patientsho were included and those who were excluded.

ata collection

fter screening patients, we collected demographic datarom patients’ electronic medical records (EMRs). Theeverity of disease assessed with the Patient Severity Clas-ification Tool (PSCT) (KHNA, 1994) was also obtained fromMRs. The PSCT ranges in level from one to six, and theigher the level, the more severe the disease.

utritional status dataor the subjective assessment of nutritional status, theubjective Global Assessment (SGA) scale was used upondmission (Desky et al., 1987). This scale has been recom-ended for use as a validated assessment method (Delmore,

997). Although the information was obtained primarily fromhe caregivers for patients with communication problemsn = 19), the SGA scale has been reported to be a reliableethod with the agreement (k = 0.9) between raters (Sheean

t al., 2010). For objective assessments of nutritional status,he ninth revision of the International Classification of Dis-ase, Clinical Modification (ICD-9-CM) (Swails et al., 1996),nthropometric measurements, and biochemical measure-ents were evaluated twice-upon admission and 7 days

fter admission. Anthropometric measures included tricepskinfold thickness (TSF), mid-arm circumference (MAC), per-ent ideal body weight (PIBW), body mass index (BMI) andid-arm muscle circumference (MAMC) (Bickley, 2003). TSF

nd MAC were measured three times in the arms withoutedema or wound dressing using a calliper (Caldwell, Jus-ics & Co Inc.) and an inelastic measuring tape, and then

mean value was obtained (Lee and Kwon, 2000). Therms were held vertically so as not to touch any surfaceecause wound dressings, oedema, and the patients’ posi-

ion may influence measurements (Sabol, 2004). MAMC, BMInd PIBW were calculated using equations (ADA, 2000). Foriochemical measures, we obtained the values for serumlbumin, haemoglobin, total lymphocyte count, and C-

1

rauaS

EEicdEo(eaAparrieefepmwGgdd3t(eerewtc

D

SdDaitnsis

R

C

AwstsiTs

nmfiwawpto

C

SmBda(ds

Cnh

Tnmgd(gd(o

c(tmams

96

eactive protein (CRP) from EMRs at admission and 7 daysfter admission. Prealbumin and transferrin were analysedsing blood samples drawn upon admission and 7 days afterdmission, and then sent to the clinical research centre atNUH (Sabol, 2004).

nergy intake datanergy prescription. Data on prescription of enteral feed-ng including calorie and concentration (normal, 1.5 timesoncentrated, and 2 times concentrated) of the formula,etermined by ICU physicians each day, were obtained fromMRs. Energy prescription was classified as follows: 90—110%f the energy requirement (adequate prescription), >110%over-prescription) and <90% (under-prescription) (McClavet al., 1998). Energy requirements were calculated upondmission using the Harris—Benedict equation (ADA, 2000).dequacy of energy intake. For adequacy of energy intake,atients were categorised into the adequately fed groupnd the underfed group. Adequate feeding was defined aseceiving an energy intake greater than 90% of the energyequirement, and underfeeding as receiving an energyntake less than 90% of the energy requirement (Binnekadet al., 2005; De Jonghe et al., 2001). Data regarding thenergy intake received was obtained from EMRs every dayor 7 days. It included additional energy provided from par-nteral sources, such as dextrose for mixing antibiotics andropofol for sedation (O’Leary-Kelley et al., 2005). Theean number of kilocalories received per day for 7 daysas then calculated (Krishnan et al., 2003).I intolerance. To assess digestive status, we obtained theastric residual volumes (GRVs) from EMRs every day for 7ays and re-calculated the mean volume of each aspirationuring 7 days and categorised them into less than 30 cm3,0—50 cm3 and over 50 cm3. In our hospital, nurses recordedhe GRVs aspirated before each feeding in patients’ EMRsHeyland et al., 1995). Episodes of vomiting were recordedvery day for 7 days and categorised into presence ofpisodes over the 7 days (no and yes) because vomiting wasare (Montejo, 1999). The consistency and amount of stool,valuated using a diarrhoea scale (Hart and Dobb, 1988),ere also obtained from EMRs for 7 days. We calculated

he diarrhoea score per day for 7 days from EMRs, and thenalculated the mean value for 7 days.

ata analysis

PSS version 15.0 for windows was used to analyse theata. The level of statistical significance was set at p < 0.05.escriptive analysis was used to characterise the sample atdmission. The paired t-test was used to determine changesn anthropometric measures and biochemical measures overhe 7-day period, and to analyse the influence of baseline

utritional status and adequacy of energy intake during ICUtays. The logistic regression test was used to examine thempact of enteral nutrition prescription, baseline nutritionaltatus, and GI intolerance on the adequacy of energy intake.

(iBd

H. Kim, S. Choi-Kwon

esults

haracteristics of the study sample

total of 48 patients (mean age, 57.2 years; 70.8% male)ere included in the study. No significant differences in

ubjective or objective nutritional status were found withhe SGA (Chi-square = 1.47, p = 0.479) and ICD-9-CM (Chi-quare = 3.82, p = 0.575) between the 48 patients who werencluded in and the 32 patients excluded from the study.he demographic and nutritional characteristics of the studyubjects are shown in Table 1.

Based on the SGA, 75% (n = 36) were severely mal-ourished and 25% (n = 12) were suspected being ofalnourished upon admission. Based on the ICD-9-CM, onlyve patients (10%) were not malnourished and 90% (n = 43)ere malnourished on admission. During the study period,bout two-thirds of the patients (62.5%) were underfed,hilst 37.5% were adequately fed. More than half of theatients (54.2%) had under-prescription of enteral nutri-ion, 39.6% had adequate prescription and only 6.2% hadver-prescription.

hanges in nutritional status

even days after admission, the anthropometric measure-ents of TSF (p < 0.001), MAC (p < 0.001), MAMC (p < 0.001),MI (p = 0.016) and PIBW (p = 0.012) were significantlyecreased when compared to the measurements takent admission. The biochemical measurements, prealbuminp = 0.045) and CRP (p < 0.001) were significantly decreased 7ays after admission when compared to the levels at admis-ion (Table 2).

hanges in nutritional status according to baselineutritional status and energy intake over 7 days ofospitalisation

o assess the contribution of baseline nutritional status toutritional changes, we classified patients into the severelyalnourished group (n = 12) and the suspected malnourished

roup (n = 36) based on the SGA at admission (Table 3). Sevenays after admission, TSF (p < 0.001), MAC (p < 0.001), MAMCp < 0.001) and prealbumin (p < 0.05) had decreased in bothroups, whereas BMI (p = 0.039) and PIBW (p = 0.039) wereecreased only in the severely malnourished group. CRPp < 0.001) and total lymphocyte (p = 0.012) also decreasednly in the severely malnourished group.

To assess the contribution of energy intake to nutritionalhanges, patients were divided into the underfed groupn = 30) and the adequately fed group (n = 18). The nutri-ional status of the underfed group worsened significantlyore than that of the adequately fed group (Table 4). Whilst

dequately fed patients had significant decreases in foureasures, underfed patients had decreases in seven mea-

ures during the 7 days after admission. TSF (p < 0.05), MAC

p < 0.001), MAMC (p < 0.01) and CRP (p < 0.01) had decreasedn both the underfed and adequately fed group, whereasMI (p = 0.041), PIBW (p = 0.049) and prealbumin (p = 0.027)ecreased only in the underfed group.

Changes in nutritional status in ICU patients receiving enteral tube feeding: A prospective cohort study 197

Table 1 General and nutritional characteristics of the subjects.

Characteristic n (%)

Age 57.2 (13.8)a

Gender Male 34 (70.8)Female 14 (29.2)

Diagnosis Lung disease 9 (18.8)GI disease 1 (2.1)Heart disease 8 (16.7)Cancer 8 (16.7)Leukaemia 5 (10.4)Septic shock 2 (4.2)Neurologic disorder 10 (20.8)Nephrologic disorder 5 (10.4)

Severity of disease (PSCTb) Fifth level 20 (41.7)Sixth level 28 (58.3)

SGAb Well nourished 0 (0)Suspected being of malnourished 12 (25.0)Severely malnourished 36 (75.0)

ICD-9-CMb No malnutrition 5 (10.4)Malnutrition of mild degree 12 (25.0)Malnutrition of moderate degree 7 (14.6)Energy malnutrition 1 (2.1)Protein malnutrition 22 (45.8)Severe protein-energy malnutrition 1 (2.1)

Feeding Under-prescription 26 (54.2)Prescription Adequate prescription 19 (39.6)

Over-prescription 3 (6.2)Energy intake Underfed 30 (62.5)

Adequately fed 18 (37.5)Feeding Regular 35 (72.9)Concentration 1.5 times concentrated 4 (8.3)

2 times concentrated 9 (18.8)Digestive status Residual of less than 30 cm3 42 (87.5)

Residual of 30—50 cm3 2 (4.2)Residual of over 50 cm3 4 (8.3)

Vomiting No 47 (97.9)Yes 1 (2.1)

Diarrhoea scale 3.9 (2.6)a

a Mean (SD); All values are presented as n (%) except footnote ‘‘a’’ where mean (SD) is presented.b al As

iin

D

TnwtWimalnourished, and one-fourth were suspected of being mal-

PSCT, Patient Severity Classification Tool; SGA, Subjective Globrevision, Clinical Modification.

Factors contributing to adequate feeding

To assess the impact factors had on the likelihood patientswould receive underfeeding, we performed a direct logis-tic regression. The full model containing all five predictors(prescription of enteral nutrition, baseline nutritional sta-tus, digestive status, vomiting and diarrhoea) was significant(Chi-square(df) = 11.01(5), p < 0.05). The strongest predic-tor of underfeeding was under-prescription, with an oddsratio of 5.33 (CI: 1.34, 21.15). This indicates that patientswho were under-prescribed were over five times more likelyto be underfed than those who had adequate prescriptionor over-prescription of enteral feeding, controlling for allother factors in the model (Wald(df) = 5.65(1), p < 0.05).

Patients who were severely malnourished at admission were2.32 (CI: 1.47, 3.72) times more likely to be underfedcompared to those who were suspected to be malnour-

nft

sessment; ICD-9-CM, International Classification of Disease, ninth

shed (Wald(df) = 3.31(1), p < 0.05). However, GI intolerancencluding digestive status, vomiting or diarrhoea did not sig-ificantly predict underfeeding.

iscussion

o our knowledge, this is the first study to demonstrate theutritional changes over time during ICU stays in patientsith enteral feeding as well as the factors contributing to

he changes in nutritional status in Korean ICU patients.e found that all patients in our study were malnour-

shed at admission, three-fourths of patients were severely

ourished. The nutritional status of study subjects wasurther aggravated during the ICU stay, which was relatedo inadequate nutritional intake.

198 H. Kim, S. Choi-Kwon

Table 2 Changes in anthropometric and biochemical measures over 7 days.

Variables Normal range At admission 7 days afteradmission

T p Value

Mean (SD) Mean (SD)

TSF (mm) M ≥12.5, F ≥16.5 19.47 (6.44) 17.28 (5.90) −6.247 <0.001MAC (cm) M ≥29.3, F ≥28.5 25.62 (4.72) 23.86 (4.11) −9.000 <0.001MAMC (cm) M ≥25.3, F ≥23.2 19.50 (3.46) 18.51 (3.23) −6.118 <0.001BMI 18.5—24.9 21.25 (3.19) 20.79 (2.95) −2.491 0.016PIBW (%) 90—110 100.68 (16.8) 97.19 (14.7) −2.620 0.012Albumin (g/dl) 3.5—5.5 2.95 (0.43) 2.91 (0.39) −0.666 0.509Prealbumin (mg/dl) 15—32 11.96 (6.91) 9.80 (5.95) −2.732 0.045Transferrin (mg/dl) 250—300 101.8 (34.0) 103.6 (32.3) 0.284 0.778Haemoglobin (g/dl) M:12—16, F:13.5—17.5 10.98 (2.08) 10.54 (1.72) −1.693 0.097Total lymphocytes (n/mm3) ≥1600 723.1 (567.1) 873.1 (683.6) 1.972 0.054CRP (mg/dl) ≤1.0 8.67 (6.79) 5.36 (5.67) −4.348 <0.001

, mid

hd(e2de

rIdi

TSF, triceps skinfold thickness; MAC, mid-arm circumference; MAMCideal body weight; CRP, C-reactive protein.

The prevalence rate for malnutrition in our study wasigher than that found in previous studies (37—40%) con-ucted by Sungurtekin et al. (2008) (n = 124) and Barr et al.2004) (n = 200). Several factors may account for the differ-

nces in the prevalence of malnutrition (Schwebel et al.,000). First, the higher prevalence rate may have beenue to the use of different methods for assessment. How-ver, this is not likely because we obtained almost the same

tdhC

Table 3 Nutritional changes according to baseline nutritional sta

Suspected maMean (SD)

TSF Pre 20.31(8.29)**

Post 18.66 (8.10)

MAC Pre 25.68 (5.50)**

Post 24.13 (5.14)

MAMC Pre 19.30 (4.45)**

Post 18.61 (4.54)

BMI Pre 23.00 (3.69)

Post 22.49 (3.49)

PIBW Pre 108.65 (16.34Post 105.80 (16.17

Albumin Pre 3.29 (0.40)*

Post 3.00 (0.40)

Prealbumin Pre 14.08 (8.33)*

Post 11.16 (6.26)

Transferrin Pre 107.08 (36.38Post 120.68 (39.84

Haemoglobin Pre 12.01 (2.17)

Post 11.30 (1.97)Total lymphocyte Pre 905.0 (410.2)

Post 836.0 (594.5)CRP Pre 5.54 (4.73)

Post 3.92 (4.90)

TSF, triceps skinfold thickness; MAC, mid-arm circumference; MAMC, midideal body weight; CRP, C-reactive protein.

* p < 0.05.** p < 0.01.

-arm muscle circumference; BMI, body mass index; PIBW, percent

esults when we classified nutritional status according to theCD-9-CM. Second, it may have been due to the severity ofiseases in our population when compared to previous stud-es (Atalay et al., 2008; Giner et al., 1996). Our hospital is a

ertiary care university hospital and all subjects had a severeegree of critical illness. Third, the high prevalence mayave been due to severe inflammation, supported by highRP levels in our subjects (Schols et al., 1996). Finally, it

tus.

lnutrition (n = 12) Severe malnutrition (n = 36)Mean (SD)

19.20 (5.82)**

16.82 (5.03)25.60 (4.51)**

23.76 (3.78)19.56 (3.14)**

18.48 (2.74)20.67 (2.82)*

20.22 (2.56)) 98.03 (16.38)*

) 94.31 (13.31)2.83 (0.38)2.88 (0.37)11.25 (6.35)*

9.35 (5.86)) 100.13 (33.52)) 98.18 (27.92)

10.63 (1.96)10.28 (1.59)

662.5 (603.2)*

885.2 (718.3)9.71 (7.60)**

5.85 (5.90)

-arm muscle circumference; BMI, body mass index; PIBW, percent

Changes in nutritional status in ICU patients receiving enteral tube feeding: A prospective cohort study 199

Table 4 Nutritional changes according to energy intake.

Adequately fed (n = 18) Underfed (n = 30)Mean (SD) Mean (SD)

TSF Pre 21.39 (5.96)* 16.28 (5.09)**

Post 19.24 (5.73) 14.02 (4.71)MAC Pre 26.93 (3.89)** 23.45 (5.27)**

Post 25.14 (3.61) 21.73 (4.10)MAMC Pre 20.21 (3.14)** 18.32 (3.74)**

Post 19.22 (3.05) 17.34 (3.27)BMI Pre 22.32 (2.80) 19.48 (3.07)*

Post 21.78 (2.69) 18.15 (2.70)PIBW Pre 104.33 (20.57) 94.62 (13.25)*

Post 101.01 (14.39) 90.82 (13.87)Albumin Pre 3.01 (0.42) 2.85 (0.46)

Post 2.95 (0.43) 2.84 (0.33)Prealbumin Pre 11.63 (7.80) 11.17 (3.90)*

Post 10.25 (5.64) 9.06 (6.53)Transferrin Pre 108.62 (36.92) 104.54 (37.21)

Post 100.27 (32.48) 95.39 (21.12)Haemoglobin Pre 10.95 (2.28) 11.03 (1.77)

Post 10.58 (2.02) 10.47 (1.13)Total lymphocyte Pre 718.00 (624.79)* 731.67 (472.39)

Post 918.67 (803.67) 797.22 (425.13)CRP Pre 8.04 (6.21)** 9.72 (7.73)**

Post 5.43 (6.24) 5.25 (4.74)

TSF, triceps skinfold thickness; MAC, mid-arm circumference; MAMC, mid-arm muscle circumference; BMI, body mass index; PIBW, percentideal body weight; CRP, C-reactive protein.

* p < 0.05.**

(dddsdt

nbcwpaeifipsrtsn

p < 0.01.

may have been due to the difference in populations (differ-ent countries). Consistent with our result, a study conductedin a Korean ICU demonstrated that 140 out of 161 patients(87%) were malnourished when admitted to the ICU (Leeet al., 2003).

Despite enteral nutritional support, nutritional statusmeasured with anthropometric and biochemical parametersworsened during the ICU stay. We found that prealbuminlevels decreased significantly overall during the ICU stay(p < 0.05), although no significant changes were found inserum albumin. This may be due to the short half-life (1.9days) of serum prealbumin when compared to that of albu-min (20 days) (Sabol, 2004). However, prealbumin can beinfluenced by an acute phase response to critical illness, sothe determination of CRP in conjunction with prealbuminmay help distinguish an acute-phase response from malnu-trition (Cresci, 2005). Because low levels of both prealbuminand CRP indicate malnutrition (Brinkmann, 2010), our datashowed that the nutritional status of our patients wors-ened.

For the anthropometric measurements, BMI and PIBWshowed significant decreases in our patients and particu-larly so in severely malnourished patients. The decreasein BMI and PIBW might indicate dehydration, because total

body weight in ICU patients may fluctuate due to ongoingfluid losses (Campbell et al., 2005). This is unlikely, how-ever, since MAC, a useful index representing protein status(Sabol, 2004) also decreased significantly during the ICU stay

atam

p < 0.01). The loss of muscle mass may also have been in partue to the decrease in the synthesis of contractile proteinsue to immobilisation (Jork and Konstadt, 1983). However,ecreased MAC, along with decreased TSF-which representsubcutaneous fat loss (Sarhill et al., 2004)-strongly suggestsecreases in BMI and PIBW were due mainly to a deteriora-ion in nutritional status.

In order to investigate the reasons for the worsening ofutritional status, we analysed the relationship betweenaseline nutritional status, energy intake, and nutritionalhanges. Nutritional status worsened more in patients whoere severely malnourished than in those who were sus-ected of being malnourished at admission. We do not have

clear explanation as to why the nutritional status wors-ned even though the patients received enteral feeding. Its likely that they may have received inadequate enteraleeding (Barr et al., 2004). In further analysis, worsen-ng of nutritional status was more extensive in underfedatients compared to adequately fed patients during the ICUtay. Despite the fact that malnourished patients urgentlyequire ample nutritional support for the repair of nutri-ional deficits, they may not receive adequate nutritionalupport during their ICU stay. It is interesting to note thatutritional status was aggravated in adequately fed patients

s well as in underfed patients during the ICU stay. Nutri-ional deterioration presented even in patients who weredequately fed during the ICU stay and were not severelyalnourished at admission. These results indicate that it is

2

nc

tstd2

mPntd(jhdros

ai(dutptti2he

L

2ootatsoetbpmoseuiu

C

Ta

tpsisetoeotc(

sespnt

C

T

F

Ttf(

A

WStc

R

A

A

B

B

B

B

00

ecessary to provide rigorous nutritional support for criti-ally ill patients to improve their nutritional status.

We found that about two-thirds of patients received lesshan their nutritional requirements, particularly so in theeverely malnourished group. Our results are consistent withhe findings of previous studies in which ICU patients oftenid not receive adequate enteral feeding (De Jonghe et al.,001; O’Leary-Kelley et al., 2005; Rice et al., 2005).

GI problems such as large GRVs, vomiting and diarrhoeaay be reasons for inadequate feeding (Mentec et al., 2001;

etros and Engelmann, 2006). However, in this study this wasot the case, since we excluded the patients who stoppedhe enteral feeding due to GI problems. Diagnostic proce-ures and surgery might have had an effect on underfeedingO’Leary-Kelley et al., 2005; O’Meara et al., 2008). Our sub-ects, however, were patients with chronic illness havingad no prior surgery. They also had few diagnostic proce-ures that interfered with the feeding schedule. In addition,outine nursing procedures did not interrupt the deliveryf enteral feeding in our study because they were usuallycheduled to avoid feeding time in our ICU.

We found that under-prescription of enteral nutrition wasn independent contributing factor to inadequate energyntake. Similarly, McClave et al. (1999) and De Jonghe et al.2001) reported that physicians prescribed 66% and 78% ofaily energy requirements. Bourgault et al. (2007) describednder-prescription as one of the barriers to meeting nutri-ional goals. In clinical settings, enteral nutrition is usuallyrescribed not by a specific algorithm or protocol, but byhe physician’s opinion. Therefore, it is important to raisehe awareness of physicians regarding nutritional supportn order to provide adequate nutrition (De Jonghe et al.,001). The necessity of a standardised nutritional protocolas been suggested (Bourgault et al., 2007; O’Leary-Kelleyt al., 2005).

imitations

There are limitations to the findings of our study, andur results should therefore be interpreted cautiously. Outf 80 patients, 48 patients were included in this study, buthere were no significant differences in nutritional statust admission between the patients who were included andhe patients who were not. Although this study has a smallample size and a single ICU setting limiting generalisationf results, we included patients with less GI trouble, whichnabled us to interpret the data more easily. Another poten-ial limitation is that the TSF assessment technique maye difficult to implement in ICU patients, resulting in theossibility of measurement error. However, our measure-ents were obtained in patients whose arms did not have

edema or wound dressings and the arm was held vertically,o as not to rest on any surface, minimising measurementrror. Finally, energy requirements were calculated oncepon admission. However, since the BMI did not increasen a single patient over 7 days, it is unlikely that we havenderestimated energy requirements.

onclusions

he ICU patients in our study were severely malnourishedt admission and their nutritional status worsened during

B

H. Kim, S. Choi-Kwon

he ICU stay, even though enteral nutritional support wasrovided. The patients’ baseline nutritional status at admis-ion and underfeeding, due mainly to under-prescription,nfluenced the changes in nutritional status during their ICUtay. Therefore, it is important that energy requirements arestimated accurately and are then matched to prescriptionso provide adequate enteral nutrition. Accurate assessmentf energy needs is the foundation for the prescription ofnteral nutrition for ICU patients. Mathematical estimatesf energy requirements such as the Harris—Benedict equa-ion or measuring energy expenditure, such as by indirectalorimetry, can help ensure that underfeeding is avoidedO’Leary-Kelley et al., 2005).

A standardised enteral nutrition protocol including pre-cription and delivery of enteral nutrition may help meet thenergy requirements of patients in the ICU. Future researchhould devote attention to the development of a nutritionalrotocol based on the assessment of nutritional status andeeds, and on factors related to delivery of enteral nutri-ion.

onflict of interest

here are no conflicts of interest concerning this study.

unding

his work was supported by Basic Science Research Programhrough the National Research Foundation of Korea (NRF)unded by the Ministry of Education, Science and TechnologyMEST) (810-20090018).

cknowledgements

e gratefully acknowledge the dedicated support of Hyunjuhin, a manager of the nursing department. We also thankhe nurses from the medical ICU for their support in dataollection.

eferences

DA. Manual of clinical dietetics. American Dietetic Association;2000.

talay BG, Yagmur C, Nursal TZ, Atalay H, Noyan T. Use of subjectiveglobal assessment and clinical outcomes in critically ill geriatricpatients receiving nutrition support. JPEN — J Parenter EnterNutr 2008;32:454—9.

arr J, Hecht M, Flavin KE, Khorana A, Gould MK. Outcomesin critically ill patients before and after the implementationof an evidence-based nutritional management protocol. Chest2004;125:1446—57.

ickley LS. Bates’ guide to physical examination and history taking.8th ed. Philadelphia: Lippincott and Wilkins; 2003.

innekade JM, Tepaske R, Bruynzeel P, Mathus-Vliegen EMH, de HaanRJ. Daily enteral feeding practice on the ICU: attainment of goalsand interfering factors. Crit Care 2005;9:R218—25.

ourgault AM, Ipe L, Weaver J, Swarta S, O’Dea PJ. Development ofevidence-based guidelines and critical care nurses’ knowledge

of enteral feeding. Crit Care Nurse 2007;27(4):17—29.

rinkmann T. The role of prealbumin as a marker of mal-nutrition. Clin Lab Int. http://www.cli-online.com/index.php?id=2260.[Accessed August 16, 2010].

al tu

M

M

N

O

O

P

P

R

R

R

S

S

S

S

S

S

S

Changes in nutritional status in ICU patients receiving enter

Campbell CG, Zander E, Thorland W. Predicted vs measured energyexpenditure in critically ill, underweight patients. Nutr ClinPract 2005;20:276—80.

Cresci G. Nutrition support for the critically ill patient: a guide topractice. Florida: Taylor & Francis; 2005.

De Jonghe B, Appere-De-Vechi C, Fournier M, Tran B, Merrer J,Melchior JC, et al. A prospective survey of nutritional supportpractices in intensive care unit patients: what is prescribed?What is delivered? Crit Care Med 2001;29(1):8—12.

Delmore G. Assessment of nutritional status in cancer patients:widely neglected? Support Care Cancer 1997;5:376—80.

Desky AS, McLaughlin JR, Baker JP. What is the subjective globalassessment of nutritional status? JPEN — J Parenter Enter Nutr1987;11:8—13.

Giner M, Laviano A, Meguid MM, Gleason JR. In 1995 a correlationbetween malnutrition and poor outcome in critically ill patientsstill exists. Nutrition 1996;12(1):23—9.

Hart GK, Dobb GJ. Effect of a fecal bulking agent on diarrhea duringenteral feeding in the critically ill. JPEN — J Parenter Enter Nutr1988;12:465—8.

Heyland D, Cook DJ, Winder B, Brylowski L, Van deMark H, GuyattG. Enteral nutrition in the critically ill patients: a prospectivesurvey. Crit Care Med 1995;23(6):1055—60.

Jagoe RT, Goodship TH, Gibson GJ. The influence of nutritional sta-tus on complications after surgery for lung cancer. Ann ThoracSurg 2001;71:936—43.

Jork P, Konstadt S. Effect of hindlimb immobilization on mus-cle function and protein composition. J Clin Orthoped1983;174:222—8.

KHNA. Cost analysis of nursing care. Life Nursing; 1994.Kim H, Shin JA, Shin JY, Cho OM. Adequacy of nutritional support

and reasons for underfeeding in neurosurgical intensive care unitpatients. Asian Nurs Res 2010;4(2):102—10.

Krishnan JA, Parce PB, Martinez A, Diette GB, Brower RG.Caloric intake in medical ICU patients: consistency of carewith guidelines and relationship to clinical outcomes. Chest2003;124(1):297—305.

Lee GN, Kwon OH. Clinical pathology file. Medical Culture Co; 2000.Lee S, Choi M, Kim Y, Lee J, Shin C. Nosocomial infection of

malnourished patients in an intensive care unit. Yonsei Med J2003;44(2):203—9.

Martin CM, Doig GS, Heyland DK, Morrison T, Sibbald WJ. Mul-ticentre, cluster-randomized clinical trial of algorithms forcritical-care enteral and parenteral therapy (ACCEPT). Can MedAssoc J 2004;170(2):197—204.

McClave SA, Lowen CC, Kleber MJ, Nicholson JF, Jimmerson SC,McComell JW, et al. Are patients appropriately fed accordingto their caloric requirements? JPEN — J Parenter Enter Nutr1998;22:375—81.

McClave SA, Sexton LK, Spain DA, Adams JL, Owens NA, SullinsMB, et al. Enteral tube feeding in the intensive care unit:factors impeding adequate delivery. Crit Care Med 1999;27(7):1252—6.

T

be feeding: A prospective cohort study 201

entec H, Dupont H, Bocchetti M, Cani P, Ponche F, Bleichner G.Upper digestive intolerance during enteral nutrition in criticallyill patients: frequency, risk factors, and complications. Crit CareMed 2001;29(10):1955—61.

ontejo JC. Enteral nutrition-related gastrointestinal complicationsin critically ill patients: a multicenter study. Crit Care Med1999;27(8):1447—53.

isim AA, Allins AD. Enteral nutritional support. Nutrition2005;21:109—12.

’Leary-Kelley CM, Puntillo KA, Barr J, Stotts N, Douglas MK. Nutri-tional adequacy in patients receiving mechanical ventilation whoare fed enterally. Am J Crit Care 2005;14(3):222—31.

’Meara D, Mireles-Cabodevila E, Frame F, Hummell AC, HammelJ, Dweik RA, et al. Evaluation of delivery of enteral nutrition incritically ill patients receiving mechanical ventilation. Am J CritCare 2008;17(1):53—61.

arrish CR. Enteral feeding: the art and the science. Nutr Clin Pract2003;18:76—85.

etros S, Engelmann N. Enteral nutrition delivery and energyexpenditure in medical intensive care patients. Clin Nutr2006;25:51—9.

eid CL. Frequency of under and overfeeding in mechanically venti-lated ICU patients: causes and possible consequences. J HumanNutr Diet 2006;19:13—22.

ice TW, Swope T, Bozeman S, Wheeler AP. Variation in enteralnutrition delivery in mechanically ventilated patients. Nutrition2005;21(7—8):786—92.

odriguez L. Nutritional status: assessing and understanding itsvalue in the critical care setting. Crit Care Nurs Clin North Am2004;15(4):509—14.

abol VK. Nutrition assessment of the critically ill adult. AACN ClinIssues 2004;15(4):595—606.

arhill N, Mahmoud F, Walsh D, Nelson KA, Komurcu S, Davis M, et al.Evaluation of nutritional status in advanced metastatic cancer.Support Care Cancer 2004;11(10):652—9.

chols A, Slangen J, Volovics A, Wouters E. Body weight and survivalin COPD. Am J Respir Crit Care Med 1996:A452.

chwebel C, Pin I, Barnoud D, Devouassoux G, Brichon PY,Chaffanjon P, et al. Prevalence and consequences of nutri-tional depletion in lung transplant candidates. Eur Respir J2000;16(6):1050—5.

heean PM, Peterson SJ, Gurka DP, Braunschweig CA. Nutritionassessment: the reproducibility of subjective global assessmentin patients requiring mechanical ventilation. Eur J Clin Nutr2010;64(11):1358—64.

ungurtekin K, Sungurtekin U, Oner O, Okke D. Nutrition assessmentin critically ill patients. Nutr Clin Pract 2008;23(6):635—41.

wails WS, Samour PQ, Babineau TJ, Bistrian BR. A proposed revi-sion of current ICD-9-CM malnutrition code definitions. J Am Diet

Assoc 1996;96:370—3.

homas DR. Improving outcome of pressure ulcers with nutri-tional interventions: a review of the evidence. Nutrition2001;17(2):121—5.