espen_guidelines_geriatrics nutrition 200625330-60.pdf

7/27/2019 Espen_guidelines_Geriatrics Nutrition 200625330-60.pdf

1/31

Clinical Nutrition (2006) 25, 330360

ESPEN GUIDELINES

ESPEN Guidelines on Enteral Nutrition: Geriatrics$

D. Volkerta,,1, Y.N. Bernerb, E. Berryc, T. Cederholmd, P. Coti Bertrande,A. Milnef, J. Palmbladg, St. Schneiderh, L. Sobotkai, Z. Stangaj,DGEM:$$ R. Lenzen-Grossimlinghaus, U. Krys, M. Pirlich, B. Herbst,T. Schutz, W. Schroer, W. Weinrebe, J. Ockenga, H. Lochs

aHead Medical Science Division, Pfrimmer-Nutricia, Erlangen, GermanybHead Geriatric Department, Meir Hospital, Kfar Saba, IsraelcDepartment of Human Nutrition & Metabolism, Hebrew University, Hadassah Med School,

Jerusalem, IsraeldDepartment of Public Health and Caring Science, Uppsala University, Uppsala, SwedeneUnite de Nutrition Clinique, CHUV, Lausanne, SwitzerlandfHealth Services Research Unit, University of Aberdeen, Aberdeen, UKgDepartment of Medicine, Karolinska Institute, Huddinge University Hospital, Huddinge, SwedenhGastroenterologie et Nutrition Clinique, Hopital de lArchet, Nice, FranceiMetabolic Care Unit, Department of Gerontology and Metabolic Care, Charles University,Faculty of Medicine, Hradec Kralove, Czech RepublicjInternal Medicine and Clinical Nutrition, Inselspital/University Hospital, Bern, Switzerland

Received 18 January 2006; accepted 19 January 2006

KEYWORDSGuideline;Clinical practice;Evidence-based;Recommendations;

Summary Nutritional intake is often compromised in elderly, multimorbidpatients. Enteral nutrition (EN) by means of oral nutritional supplements (ONS)and tube feeding (TF) offers the possibility to increase or to insure nutrient intake incase of insufficient oral food intake.

The present guideline is intended to give evidence-based recommendations forthe use of ONS and TF in geriatric patients. It was developed by an interdisciplinaryexpert group in accordance with officially accepted standards and is based on all

ARTICLE IN PRESS

http://intl.elsevierhealth.com/journals/clnu

0261-5614/$ - see front matter & 2006 European Society for Clinical Nutrition and Metabolism. All rights reserved.doi:10.1016/j.clnu.2006.01.012

Abbreviations: ADL, activities of daily living; BCM, body cell mass; BMI, body-mass index; CI, confidence interval; EN, enteralnutrition; FFM, fat-free mass; IADL, instrumental activities of daily living; MAC, mid-arm circumference; MAMC, mid-arm musclecircumference; NGT, nasogastric tube; ONS, oral nutritional supplement; OR, odds ratio; PEG, percutaneous endoscopic gastrostomy;RR, relative risk; SD, standard deviation; TF, tube feeding; TSF, triceps skin fold$For further information on methodology see Schutz et al.173 For further information on definition of terms see Lochs et al.174Corresponding author. Tel.: +499131 7782 31; fax: +499131 7782 86.

E-mail address: [email protected] (D. Volkert).1Dorothee Volkert had been employed at the Department of Nutrition Science, University of Bonn, until May 31, 2005; she was not

industry employed during the development of the guidelines.$$The authors of the DGEM (German Society for Nutritional Medicine) guidelines on enteral nutrition in geriatrics are

acknowledged for their contribution to this article.
http://intl.elsevierhealth.com/journals/clnuhttp://localhost/var/www/apps/conversion/tmp/scratch_10/dx.doi.org/10.1016/j.clnu.2006.01.012mailto:[email protected]:[email protected]://localhost/var/www/apps/conversion/tmp/scratch_10/dx.doi.org/10.1016/j.clnu.2006.01.012http://intl.elsevierhealth.com/journals/clnuhttp://intl.elsevierhealth.com/journals/clnu


2/31

Enteral nutrition;Oral nutritionalsupplements;Tube feeding;Geriatric patients;Undernutrition;Malnutrition;Elderly;Aged-80-and-over

relevant publications since 1985. The guideline was discussed and accepted in aconsensus conference.

EN by means of ONS is recommended for geriatric patients at nutritional risk, incase of multimorbidity and frailty, and following orthopaedic-surgical procedures. Inelderly people at risk of undernutrition ONS improve nutritional status and reducemortality. After orthopaedic-surgery ONS reduce unfavourable outcome. TF is clearlyindicated in patients with neurologic dysphagia. In contrast, TF is not indicated infinal disease states, including final dementia, and in order to facilitate patient care.

Altogether, it is strongly recommended not to wait until severe undernutrition hasdeveloped, but to start EN therapy early, as soon as a nutritional risk becomesapparent.

The full version of this article is available at www.espen.org.& 2006 European Society for Clinical Nutrition and Metabolism. All rights reserved.

ARTICLE IN PRESS

Summary of statements: Geriatrics

Subject Recommendations Grade173 Number

Indications In patients who are undernourished or at risk of

undernutrition use oral nutritional supplementation toincrease energy, protein and micronutrient intake,maintain or improve nutritional status, and improvesurvival.

A 2.1

In frail elderly use oral nutritional supplements (ONS) toimprove or maintain nutritional status.

A 2.2

Frail elderlymay benefit from TF as long as their generalcondition is stable (not in terminal phases of illness).

B 2.2

In geriatric patients with severe neurological dysphagiause enteral nutrition (EN) to ensure energy and nutrientsupply and, thus, to maintain or improve nutritional

status.

A 2.3

In geriatric patients after hip fracture and orthopaedicsurgery use ONS to reduce complications.

A 2.4

In depression use EN to overcome the phase of severeanorexia and loss of motivation.

C 2.6

In demented patients ONS or tube feeding (TF) may leadto an improvement of nutritional status.

2.7

In early and moderate dementia consider ONSandoccasionally TFto ensure adequate energy and nutrientsupply and to prevent undernutrition.

C 2.7

In patients with terminal dementia, tube feeding is notrecommended.

C 2.7

In patients with dysphagia the prevention of aspirationpneumonia with TF is not proven.

2.9

ONS, particularly with high protein content, can reducethe risk of developing pressure ulcers.

A 2.10

Based on positive clinical experience, EN is alsorecommended in order to improve healing of pressureulcers.

C 2.10

ESPEN Guidelines on Enteral Nutrition 331
http://www.espen.org/http://www.espen.org/


3/31

ARTICLE IN PRESS

Application In case of nutritional risk (e.g. insufficient nutritionalintake, unintended weight loss 45% in 3 months or 410%in 6 months, body-mass index (BMI) o20 kg/m2) initiateoral nutritional supplementation and/or TF early.

B 2.1

In geriatric patients with severe neurological dysphagiaEN has to be initiated as soon as possible.

C 2.3

In geriatric patients with neurological dysphagiaaccompany EN by intensive swallowing therapy until safeand sufficient oral intake is possible.

C 2.3

Initiate enteral nutrition 3hours after PEG placement. A 3.2

Route In geriatric patients with neurological dysphagia preferpercutaneous endoscopic gastrostomy (PEG) tonasogastric tubes (NGT) for long-term nutritional support,since it is associated with less treatment failures andbetter nutritional status.

A 2.3

Use a PEG tube if EN is anticipated for longer than 4weeks.

A 3.1

Type of

formulaDietary fibre can contribute to the normalisation of bowelfunctions in tube-fed elderly subjects.

A 3.4

Grade: Grade of recommendation; Number: refers to statement number within the text.

Terminology

Geriatric patienta biologically elderly patient who is at acute risk of loss of independence due to acuteand/or chronic diseases (multiple pathology) with related limitations in physical, psychological, mental

and/or social functions. The abilities to perform the basic activities of independent daily living arejeopardised, diminished or lost. The person is in increased need of rehabilitative, physical, psychologicaland/or social care to avoid partial or complete loss of independence.

Elderlya term used to describe a particular age group, i.e. over 65 years.Very old or very elderlya term to describe those over 85 years of age.Frail elderlyFrail elderly are limited in their activities of daily living due to physical, mental,

psychological and/or social impairments as well as recurrent disease. They suffer from multiple pathologieswhich seriously impair their independence. They are therefore in particular need of help and/or care and arevulnerable to complications.

Reduced capacity for rehabilitationThis means that the older the patient, the more difficult it is torehabilitate that patient back to normal or to his/her previous state. Specifically, the restoration ofmuscle mass after illness requires much greater effort in terms of exercise and nutrition in the elderly

compared with the younger patient. It is also implicit that other functions, including mental, are similarlymore resistant to rehabilitation.Functional statusThis term is being used in a general sense to describe global function, e.g. the ability

to perform activities of daily living (ADL), or specific function, e.g. muscle strength or immune function.

Introduction

The risk of undernutrition is increased in elderlypatients due to their decreased lean body mass andto many other factors that may compromisenutrient and fluid intake. Consequently, an ade-quate intake of energy, protein and micronutrients

has to be ensured in each patient independently ofhis/her previous nutritional status. Since restorationof body cell mass (BCM) is more difficult than inyounger persons, preventive nutritional support hasto be considered.

Nutritional care should be integrated appropri-ately into the overall care plan, which takes into

D. Volkert et al.332


4/31

account all aspects of the patient, personal, social,physical and psychological. A complete assessment ofthe patient should include that of nutritional status orrisk, followed by a nutritional programme reflectingethical as well as clinical considerations. In designingthe programme, it should be remembered that themajority of sick elderly patients require at least 1 g

protein/kg/day and around 30 kcal/kg/day of energy,depending on their activity. Many elderly people alsosuffer from specific micronutrient deficiencies, whichshould be corrected by supplementation.

Oral nutritional therapy via assisted feeding anddietary supplements is often difficult, time-con-suming and demanding in elderly patients (due tomultimorbidity and slow responses). However,assisted oral feeding and supplements are able tosupport the physical and psychological rehabilita-tion of most elderly patients. Therefore, even intimes of declining financial and human resources, it

is unacceptable to initiate tube feeding (TF) merelyin order to facilitate care or save time.Decision making concerning TF in the elderly is

often difficult, and in many cases ethical questionsarise (see Guidelines Ethical and legal aspectsin enteral nutrition). In each case, the followingquestions should be asked:

Does the patient suffer from a condition that islikely to benefit from enteral nutrition (EN)?

Will nutritional support improve outcome and/oraccelerate recovery?

Does the patient suffer from an incurable

disease, but one in which quality of life andwellbeing can be maintained or improved by EN?

Does the anticipated benefit outweigh thepotential risks?

Does EN accord with the expressed or presumedwill of the patient, or in the case of incompetentpatients, of his/her legal representative?

Are there sufficient resources available to manageEN properly? If long-term EN implies a differentliving situation (e.g. institution vs. home), will thechange benefit the patient overall?

Sedation of the patient for acceptance of thenutritional treatment is not justified.

The present guidelines are based on studies inelderly subjects or in those in whom the averageage of the study participants is 65 years or more.

1. What are the aims of EN therapy ingeriatrics?

Provision of sufficient amounts of energy,protein and micronutrients.

Maintenance or improvement of nutritionalstatus.

Maintenance or improvement of function,activity and capacity for rehabilitation.

Maintenance or improvement of quality oflife.

Reduction in morbidity and mortality.

Therapeutic aims for geriatric patients do notgenerally differ from those in younger patientsexcept in emphasis. While reducing morbidity andmortality is a priority in younger patients, ingeriatric patients maintenance of function andquality of life is often the most important aim.Considering the reduced adaptive and regenerativecapacity of the elderly, EN may be indicated earlierand for longer periods than in younger patients.

1.1. Can EN improve energy and nutrient intake

in geriatric patients?

EN (oral nutritional supplement (ONS) and/or TF)increases energy and nutrient intake in geriatric

patients (Ia). Percutaneous endoscopic gastro-stomy (PEG) feeding is superior to nasogastric

feeding in this respect (Ia).

Comment: In a recent Cochrane analysis, ONS ledto an increase in energy and nutrient intake in 29out of the 33 analysed trials which had reportedintake. In three studies no difference in total intakewas found, since patients reduced their voluntaryfood consumption1 (Ia). The success of ONS issometimes limited by poor compliance due to lowpalatability, side effects such as nausea anddiarrhoea, and by cost.210 Variety and alterationin taste (different flavours, temperature andconsistency), encouragement and support by staff,as well as administration between the meals (andnot at meal times) are all important in order toachieve increased energy and nutrient intake.

Randomised controlled trials of TF in patientswith neurological dysphagia that compared naso-gastric (NG) with PEG feeding have shown that93100% of the prescription was administered via

the PEG, versus 55

70% via a NG tube.11,12 In threestudies with supplemental overnight NG TF, be-tween 1000 and 1500 kcal were administered pernight in addition to daily food intake. Total energyand nutrient intake was, therefore, markedlyimproved.1315

1.2. Can EN maintain or improve the nutritional

status of elderly patients?

ONS can maintain or improve nutritional status(Ia). Several studies have shown that TF also

ARTICLE IN PRESS



5/31

maintains or improves nutritional paramentersirrespective of the underlying diagnosis. Themetabolic consequences of ageing which can leadto sarcopenia and a severely reduced nutritionalstatus at the time of tube placement can impair oreven prevent successful nutritional therapy (III).

Comment: The administration of ONS has beenreported to have positive effects on nutritionalstatus irrespective of the main diagnosis. Weightloss, during acute illness and hospitalisation, can beprevented by the provision of food of high energyand protein density, combined with between mealsnacks, and by the use of ONS, when normal intakeis insufficient. Sometimes weight gain can even beachieved. Milne et al.1 analysed the percentageweight change in 34 randomised controlled trialswith 2484 elderly patients and showed a meanweight increase of 2.3% (pooled weighted meandifference; 95% confidence interval (CI) 1.92.7%)1

(Ia). Changes to anthropometric parameters areless consistent, but may reflect improvement ofnutritional status in general1 (Ia). Effects on bodycomposition have only occasionally been investi-gated. Increases in fat-free mass (FFM) (Ib)16,17

(IIa)18 and BCM (Ib)19 in supplemented patientshave been reported by some investigators whereasothers could not detect any change (Ib)2022 (IIa)23.

Several observational studies exploring the ef-fect of TF in multimorbid geriatric patients haveshown improvements in nutritional status, e.g.maintenance of body weight2427 (III) and either

maintenance25,27 (III) or increase in albuminlevels24,26,28 (III). It should be emphasised, how-ever, that changes in albumin more usually reflectchanges in disease rather than nutritional sta-tus.29,30 In two studies of frail, mainly dementednursing home residents, weight gain has beenreported.31,32 Improvements in nutritional statushave also been described in patients with neurolo-gical dysphagia, in whom PEG feeding provedsuperior to nasogastric feeding (NGT)11,12 (Ib).The effects of nocturnal TF supplementary to dailyfood intake in elderly patients with hip fracture or

fractured neck of femur, are inconsistent.

1315

Bastow et al.13 have reported the greatest benefitin undernourished patients (Compare 2.4).

The effectiveness of TF on nutritional status maybe limited by compliance with the tubes, and byside effects. The nutritional status of the frailelderly is often very reduced at the time of tubeplacement,2426,3338 and is accompanied by sarco-penia which is more difficult to reverse in the oldcompared with the young.3941 Resistance training,if tolerated, may add to the effectiveness ofnutritional support.9,42 Many tube fed patients are

bedridden, and consequent immobility furtherenhances muscle wasting and prevents gain in leanmass. Weighing is also problematic in thesepatients.

1.3. Does EN maintain or improve functional

status or rehabilitative capacity?

Adequate nutrition is a prerequisite for anyfunctional improvement, although studies aretoo few and diverse to allow a general state-ment. Some studies have been positive and somenegative in this respect.

Comment: Available data concerning the effect ofONS on the functional capacity of elderly patientsare inconsistent, although several studies reportfunctional improvements. Thus, Gray-Donald et al.7

(Ib), observed a significantly lower frequency offalls in supplemented free-living frail elderly

compared with non-supplemented and Unossonet al.43 (Ib) describe a higher activity level inlong-term care residents after 8 weeks of ONS.Improvements in the ability to perform basicactivities of daily living (ADL) are reported in agroup of female patients after hip fracture byTidermark et al.44 (Ib), in a subgroup of severelyundernourished geriatric patients by Potter45 (Ib)and in a subgroup of patients with good acceptanceof a 6 months supplementation by Volkert et al.2

(Ib). Woo et al.46 (Ib) describe a significantlyimproved ADL status in patients during recoveryfrom chest infection after 3-months intervention

compared with the control group. Several studies,however, detected no difference between inter-vention and control groups with respect to inde-pendence in ADL (Ib)19,20,4749 (IIa)6,50. Mobilitywasalso unchanged in several studies (Ib)3,43,47 (IIa)6.Similarly, hand grip strength was unaltered in moststudies (Ib)3,6,7,17,21,5153 (IIa)18 but this may be oflimited relevance as it only tests muscle function ofthe upper body. One randomised trial54 (IIa) as wellas two non-randomised23,55 and one uncontrolledtrial56 (IIb) report an improved hand grip strength insupplemented patients. In four trials, the effects

on mental capacity were assessed and again nochanges were observed (Ib)20,43,52 (IIa)50.At the time of tube placement, geriatric patients

are often in a significantly compromised generalcondition as well as severely functionally im-paired.24,27,36,5759 Trials in nursing homes alsodescribe a high degree of frailty and dependencein PEG-fed residents32,36,6063 (III).

Apart from the fractured femur studies withsupplementary overnight TF (Compare 2.4) onlya few, uncontrolled trials have reported theeffects of TF on either functional status or

ARTICLE IN PRESS



6/31

rehabilitative capacity in other groups of elderlypatients.24,33,36,64,65 Callahan et al.24 evaluated 72PEG-fed patients with severe physical and mentalimpairments before and after PEG placement usingseveral ADL scales. Improvements in functionalstatus were only rarely observed (improvement ofinstrumental activities of daily living (IADL) in 6%,

ADL 10%, upper body functions 18%, lower bodyfunctions 29%) (IIb). Kaw and Sekas,36 using theFunctional Independence Measure Scale (FIM), alsofailed to show significant improvements after 18months in functional status in tube-fed nursinghome residents who were in reduced generalcondition (52% demented, 48% completely ADLdependent) (III). Weaver et al.65 used a Quality ofLife Scale adapted from Spitzer, in which orienta-tion, communicative capacity, ability to self-care,and continence were assessed. In a mixed popula-tion of PEG-fed patients (median age 76 years), no

significant change was detected after long-termEN. Relatives of the patients with the lowest valueon the scale tended to answer no to the questionwhether they would wish TF in a similar situationfor themselves (IIb). Nair et al.33 observed nochanges in function measured by the KarnovskyPerformance Scale after 6 months of PEG feeding in31 surviving patients aged 8478 years (IIa). OnlySanders et al.64 describe an improvement in ADL in25 stroke patients (mean age 80 years) with EN viaPEG. At the time of PEG placement 84% of thepatients had a Barthel index (0100 points) of 0points (completely dependent; mean 0.5 points).

After 6 months of EN a mean increase of 4.8 pointswas observed. Six patients (24%) showed a clearimprovement (Barthel index increase from 0.5 to 9points), in 10 patients (40%), however, no or only aminimal improvement was observed (IIa).

1.4. Does EN reduce length of hospital stay?

In geriatric patients, length of hospital stay isdetermined not only by nutritional status butalso by other factors. Available results concern-ing the effect of EN on length of hospital stay areconflicting.

Comment: Undernutrition increases the risk ofcomplications thereby increasing the length ofhospital stay in geriatric patients.6669 Consequently,improvement in nutritional status using EN shouldresult in a reduced length of hospital stay. In geriatricpatients, however, length of hospital stay is not onlydetermined by nutritional status but also by otherfactors, e.g. the assurance of adequate care afterdischarge. In addition, in times of declining financialresources, length of hospital stay is only a poorreflection of the effects of EN.

Available study results about the impact of EN onlength of stay are conflicting. In 2002 Milne et al.70

analysed seven studies with 658 participants andreported a statistically significant benefit of ONSwith respect to hospital stay. Mean length of staywas 3.4 days shorter in the supplemented com-pared with the unsupplemented group (95% CI

6.1

0.7 days) (Ia). The addition of three new trialsto the meta-analysis, however, shifted the resultsto non-significant effects.1 If patients with hip orfemoral neck fracture are regarded separately,several studies report significantly shorter length ofstay in supplemented patients7174; this could nothowever be confirmed by others75 (Compare 2.4)

The effects of TF on length of hospital stay haveonly occasionally been measured11,13,15 and requirefurther study.

1.5. Does EN improve quality of life?

The effect of ONS and TF on quality of life isuncertain.

Comment: Although quality of life is crucial in theevaluation of therapeutic benefit in geriatrics, onlya few studies have examined the effect of EN uponit. Studies investigating the effect of ONS haveemployed different parameters, e.g: general well-being, subjective health, SF 36, EQ-5D, HospitalAnxiety and Depression Scale (HADS). Some reportimprovements (IIa)3,54,76, whereas others observeno changes7,22,51 (IIa). These few available data donot allow any firm conclusion about the effects of

ONS on quality of life.In patients requiring TF, impairments of cogni-

tion, vigilance and speech can make assessingquality of life difficult. About 60% of the patientsin the trial of Callahan et al.24 were unable tocommunicate at the time of PEG placement, andthe majority of patients with preserved ability tocommunicate were cognitively impaired (IIb). Inthe cohort of 215 patients investigated by Banner-man et al.77 data on quality of life could only begathered in 30 patients (IIb). Verhoef and vanRosendaal78 used semi-structured interviews (with

either patients or their relatives), the KarnovskyPerformance Scale as well as the Quality of LifeIndex, in order to measure subjective quality of lifein patients after PEG placement (mean age 66718years). About 85% of the patients who were stillalive after one year and still fed via PEG (n 23)were not able to run a household, 67% weredependent in personal care and 19% were feelingvery ill. However, the majority of patients andcaregivers felt that it had been the right decision toagree to the PEG. All 10 patients who were aliveafter one year and could be asked, stated that they

ARTICLE IN PRESS



7/31

would decide in favour of PEG again. The Karnovskyindex deteriorated in three of these 10 survivingpatients and improved in six (IIb). According to theauthors, these results do not necessarily imply aclear improvement in quality of life.78 Weaver etal.65 evaluated subjective quality of life by inter-view and observed a correlation between subjec-

tive and objective quality of life (Compare 1.3).Significant changes in subjective quality of lifewere not detected (IIb). Abitbol et al.26 used both abehaviour scale and a depression scale in order toassess quality of life in 59 institutionalised patients(mean age 85 years) who received EN via a PEG.The patients were bedridden, their health statuswas reduced, and infections were present in 25%.After 3 months of EN via a PEG, quality of lifescores were unchanged, although the depressionscale tended to improve. However, 16 of thesurviving patients (27%) resumed full oral nutrition

and six patients (10%), returned to their own homewith a functioning PEG tube (IIb). In a cohort of 38long-term home EN patients, quality of life waspoorer in elderly than in younger patients.79

All in all, these studies do not allow for anygeneral conclusions about effects of EN on qualityof life. TF may also have side effects that mayadversely affect quality of life, e.g. gastrointest-inal symptoms, aspiration, the discomfort of thetube, or the need to use restraints.

1.6. Does EN improve survival in geriatric

patients?

ONS improve average survival (Ia). In patientswho need TF due to the severity of disease, anincrease in survival is not proven.

Comment: Meta-analysis of the data from 32randomised controlled trials with 3017 partici-pants revealed a lower mortality risk in supple-mented elderly subjects than in controls (relativerisk (RR) 0.74; 95% CI 0.590.92)1 (Ia). Participantswere supplemented for at least 1 week andobserved for at least 2 weeks. A further meta-analysis from 12 randomised controlled trials

(n

1146) and five non-randomised studies on theeffect of ONS in hospitalised geriatric patients withmixed diagnoses reached similar conclusions (RR0.58; 95% CI 0.40.83)80 (Ia). In contrast, a meta-analysis from five studies on the effect of proteinand energy supplementation, mainly in hip fracturepatients, showed no effect on mortality risk.75

Studies on supplementary overnight TF in hipfracture patients have produced similar results(Compare 2.4).

The effect of TF on the survival of elderlypatients without a hip fracture was investigated

in nine non-randomised controlled studies (non-randomised for ethical reasons) (Table 1) and sev-eral uncontrolled observational studies (Table 2).

Four ofthe controlled studieswere carried out inhospitals,33,81,82,84 five in nursing homes.6063,83

Two of the studies were prospective,33,81 and theothers were retrospective comparisons of EN vs. no

EN. In five studies, participants with advanceddementia were investigated.33,61,62,81,84 The mostrecent of these studies was retrospective anddescribes a mean survival of 59 and 60 days in 23severely demented dysphagic patients with PEGand in 18 patients without PEG.84 A databaseanalysis from Mitchell et al.62 in 1386 nursing homeresidents with severe cognitive impairmentwhere 135 were enterally fedshowed no increasein survival (III). Mortality rate after one year wassurprisingly low (15%). Meier et al.81 prospectivelystudied 99 acutely ill patients with advanced

dementia, seventeen of whom were already beingfed by PEG at the time of hospital admission, 51had a PEG inserted in hospital, and the remaining31 consumed regular food orally. Half of all patientsdied during the following 6 months irrespective ofthe nutritional regimen. Nair et al.33 observed ahigher mortality rate in 55 severely dementedpatients with PEG after 6 months compared with acontrol group without a PEG (44% vs. 26%).According to the authors, the groups were com-parable regarding age, gender and comorbidity.PEG patients, however, suffered more often fromsevere hypoalbuminaemia (mean albumin con-

centration 28.675 vs. 33.274g/l in the controlgroup) suggesting more severe underlying inflam-matory disease. The only trial that detecteda significantly reduced mortality in nursing homeresidents with severe cognitive impairment isthe data base analysis from Rudberg et al.61

After 30 days, 15% had died in the group ofenterally fed patients compared with 30% in thecontrol group. After 1 year, the difference wasless distinct, but still statistically significant (50%vs. 61%). The control group was comparableregarding dementia, comorbidity, functional status

and BMI (III).Two further non-randomised controlled studies innursing home patients with various diagnoses and alow percentage of demented patients also failed toshow prolonged survival in the enterally fedpatients.60,63 In the databank analysis from Mitchellet al.63 mortality in 551 tube-fed nursing homeresidents with chewing and swallowing difficultieswas even higher than in 4715 residents withoutnutritional therapy (III). Approximately half of theparticipants showed severe cognitive impairments(66% of tube-fed patients vs. 46% of the control

ARTICLE IN PRESS



8/31

ARTICLE IN PRESS

Table1

Mortalityintube-fe

delderlysubjects(controlled,

non-ra

ndomisedstudies).

Article

Study

Typeof

EN

Patients

Diagnosis

Mortality(%)

Firstauthor

Type

Place

n

Age(years)

Dementia

(%)

CVE

(%)

CA

(%)

Dysphagia

(%

)

Othercharacteristics

30

6

1

M7SD

Rang

e

day

mon

year

Mitchell62

R(database)

NH

TF

135

87(Md)

(65107)

100severe

47

6

63%instablecondition,

30%decubitus,

33%severe

ADL-dependent,

84%chewingor

swallowingproblems

o5

ca.

15

No

1251

87(Md)

(65107)

100severe

27

7


15%decubitus,

45%severeADL-

dependent,

61%chewingor

swallowingproblems

o5

ca.

15

Meier8

1

P

H

68PEG,

31no

99

84.8

(63100)

100advanced

0

0

Allacutelyill,

56%decubitus,

62%infections

ca.

20

50

65

Nair33

P

H

PEG

55

83710

100advanced

0

0

NoCA,

CVE,

severe

disease,

ENduetoloworal

intake

44

No

33

8078

100advanced

0

0

NoCA,

CVE,

severedisease

26

Rudberg

61

R(database)

NH

NG

353

8577

X65

93cog.

imp.

(63severe)

10

0

100%dysphagia&eating

dependence,

96%dependentin6ADL

15

50

No

1192

8677

X65

93cog.

imp.

(64severe)

10

0

100%dysphagia&eating

dependence,

96%dependentin6ADL

30

61



9/31

ARTICLE IN PRESS

Table1(continued)

Article

Study

Typeof

EN

Patients

Diagnosis

Mortality(%)

Firstauthor

Type

Place

n

Age(years)

Dementia

(%)

CVE

(%)

CA

(%)

Dysphagia

(%

)

Othercharacteristics

30

6

1

M7SD

Rang

e

day

mon

year

Mitchell63

R(database)

NH

TF

551

87(Md)

X65

31(66severe

cog.

imp.)

59

7

100%chewingor

swallowingdifficulties,


12%decubitus,

83%severely

ADL-dependent

22

No

4715

87(Md)

X65

50(46severe

cog.

imp.)

30

6

100%chewingor

swallowingdifficulties,


9%decubitus,

46%severely

ADL-dependent

12

Bourdel-

Marchasson

60

R

NH

PEG

58

7479

n.a.

(NH55%)

n.a.

(NH

19%)

n.a.

53

36%anorexia,

10%unconscious,

allseverelydependent,

66%decubitus

14

No

50

8278

n.a.

(NH55%)

n.a.

(NH

19%)

n.a.

44

56%anorexia,

0%unconscious,

allseverelydependent,

14%decubitus

10

Cowen82

R

H

All

149

76712

20

56

0

100

Seriouscomorbidity,

42%hemiplegia,

32%CHF,20%decubitus,

70%alert,

85%urine-incontinent

27

62

PEG

80

60

Spontaneousimprovement

10

No

18

78

No/NG

51

Croghan

83

R

NH

All

40

69

(3196)

25

90

5

83

55%aspiration,

20%mobile

Tube

15

53

No

7

43

ADL

Activitiesofdailyliving,

CA

cancer,CHF

congestiveheartfailure

,cog.i

mp.

cognitiveimpairment,CVE

cerebrovascularevent,EN

enteralnutrition,

H

Hospital,

ONS:oralnutritionalsupplements,

TF

tubefeeding,

EN

enteralnutritio

n(

ONS&TF)Md

Median,

M7SDM

ean7standarddeviation,

mon

months,

n.a.

notavailable,

NG

nasogastrictube,

NH

nursinghome,

PEG

percutaneousendoscopicgastrostomy,P

prospective,

R

retrospective.



10/31

ARTICLE IN PRESS

Table2

Mortalityintube-fe

delderlysubjects(observationalstud

ieswithoutcontrolgroup).

Article

Study

type

Typeof

EN

Patients

Diagnosis

Mortality

Firstauthor

n

Age(yr)

(Range)

Dementia

CVE

CA

Dysphagia

Otherchara

cteristics

30day

3mon

6mon

1year

M7SD

(%)

(%)

(%)

(%)

(%)

(%)

(%)

(%)

Nursinghomeresidents

Golden32

R

PEG

102

8976

(71104)

89severe

20

0

100

Persistentd

ysphagia,

low

intake,

75%compl.

ADL-depend

ent,

stablecond

ition,

noterminalstage,

LEatleast

1mon

12

24

38

Abuksis57

R

PEG

47

84711

(44100)

87

49

0

94%desorie

nted,

96%bedridden

4

Kaw36

R

PEG

46

74

(1996)

52

24

7

48%comple

telyADL-

dependent,

only4%could

decideinfavourofPEG

themselves,

poorgeneral

condition

20

50

70%470

Geriatricpatients(all465yrormea

nage465yr)

Lindemann85

P

PEG

36

83

(X65)

100

0

0

11

84%lowint

ake(53%chron,

31%acute),

6%behaviouraldisorder

25

42

Sanders59

R

PEG

103

77

100

0

0

100

allseverely

ADL-dependent

(BI0-5

P)

54

78

81

90

Dwolatzky86

P

PEG

32

8576

(X65)

84

53

3

28

72%refusal

toeat

5

45

NG

90

8279

(X65)

68

43

2

37

63%refusal

toeat

20

80

Abuksis57

R

PEG

67

80716

(26103)

52

30

10

31

79%bedridden,

11%unconscious

29

Paillaud35

R

PEG

73

8379

(X65)

45

4

45

49%anorexia,

30%infection

44%reducedmobility,

44%decubitus

32

52

63

Fay27

R

PEG

80

70.2

32

52

23

79

31%decubitus,

91%inneed

ofassistancein

ADL,

76%fa

ecal-,


17

55

70

NG

29

69.8

13

41

28

41

21%decubitus,

86%inneed

ofassistancein

ADL,

66%fa

ecal-,


28

45

70

Callahan24

P

PEG

99

7979

(6098)

35

41

13

35%neuro-degenerative

disorder,se

verephysicaland

mentalimp

airment

22

50

Ciocon25

P

NG

70

82

(6595)

34

47

50%refusal

toeat,

3%oesopha

gus-obstruction,

multiple&

advanceddisease

5

41

Quill87

R

NG/G

55

470

(X70)

31

49

27

69%incomp

etent

Abitbol26

P

PEG

59

8377

50%485

30

2

42

31%MNwit

houtdys,

25%refusal

toeat

54%decubitus,49%pulmonary

infection

25

Bussone88

R

PEG

155

84

(7098)

24

3

35%neurol,

38%depression

16

Bussone89

P

PEG

101

83.6

(7098)

22

36

4

38%depression

14

Markgraf90

R/P

PEG

54

87

(6594)

24

72%neurol,

multimorbid

33

Raha91

?

PEG

161

79

(5399)

81

88

12%MN

20

39

Finucane92

P

PEG

28

82

(6899)

93

100

7%Parkinso

n;

NG-intolera

nt

8



11/31


12/31

group) and 83% and 46%, respectively, wereseverely dependent in basic ADLs. The mortalityrate after one year was comparably low in bothgroups (22% and 12%, respectively). Bourdel-March-asson et al.60 (III) reported in a mixed population of108 severely dependent nursing home residents amortality rate of 14% in the PEG group vs. 10% in

the group without nutritional support. Gastroin-testinal and pulmonary complications were also notsignificantly different. The prevalence of dementiain the nursing home was reported to be 55% and ofstroke 19%. Specific prevalence data for the studygroup, however, are not given.

Two trials in dysphagic patients reach differentconclusions. Croghan et al.83 report no differencein mortality between 15 tube-fed and seven orallyfed nursing home residents suffering from aspira-tion, who underwent videofluoroscopic swallowingevaluation mainly because of stroke. Cowen et al.82

(III) recruited 149 severely ill hospital patients withdysphagia and compared the mortality of threesubgroups after one year: Death had occurred in60% of 80 patients who had received a PEG, in 10%of 18 patients who did not receive a PEG becausetheir clinical situation had improved in hospital,and in 78% of 51 patients who did not receive a PEGfor other reasons (28 had refused EN, 12 had diedbefore PEG placement, one patient was transferredto another hospital and 10 patients were fed via aNGT).

The study by Cowen et al.82 is an example of thedifficulty of all non-randomised controlled studies,

i.e. there is a lack of comparability between theintervention and control group. The enterally fedpatients from almost all studies described aboveare probably not comparable with the patients inthe control group. The only exception is the studyfrom Rudberg et al.61 In the studies from Meieret al.81 and Murphy and Lipman84 the groups are notproperly described. In the non-randomised studies,the enterally fed patients obviously differed fromthose patients who did not receive ENfor avariety of reasons. The decision not to use EN isprobably linked to the status of the patients in

some respects. Moreover, the heterogeneity ofgeriatric patient populations provides a multitudeof factors which may influence outcome, e.g. maindiagnosis, comorbidity, nutritional status and gen-eral condition, mood, various functional para-meters including cognition, vigilance, self-careability, mobility and continence which are presentat the same time in different combinations and to avarying extent.

Observational studies reporting mortality ofenterally fed elderly subjects focus on mortalityafter 30 days or after 1 year (Table 2). However,

comparisons between studies are generally difficultdue to the heterogeneous populations involved thatare often not properly characterised. In most of thestudies, between 10% and 30% of the participantsdied after 30 days. Lower mortality rates arereported by Abuksis et al.57 and Dwolatzkyet al.86 mainly in the demented elderly, by

Finucane et al.92 and Horton et al.98 in geriatricpatients with predominantly cerebrovascularevents, and by Ciocon et al.25 in a mixed populationof elderly patients. Extremely high 30 day mortalityrates of 46% and 54% are described by Schneideret al.115 and Sanders et al.59 in the dementedelderly. One year after initiation of EN, mortalityrates between 15% and 90% are reported (Table 2).The highest as well as the lowest mortality rate isreported in demented patients59,62 (Compare 2.7).

Mitchell et al. who performed a meta-analysis ofseven controlled studies on mortality with or

without PEG, draw the conclusion that the impactof TF on survival is not known because the level ofevidence is limited.116 Further studies are neededin groups in whom nutrition may further reasonablybe expected to influence mortality.

2. EN in specific diagnostic groups

2.1. Is EN indicated in patients with under-

nutrition?

Undernutrition and risk of undernutrition repre-

sent essential and independent indications forEN in geriatric patients. ONS is recommended inorder to increase energy, protein and micronu-trient intake, maintain or improve nutritionalstatus, and improve survival in patients who areundernourished or at risk of undernutrition (A).ONS and/or TF are recommended early in

patients at nutritional risk (e.g. insufficientnutritional intake, unintended weight loss 45%in 3 months or410% in 6 months, BMIo20 kg/m2) (B).

Comment: Undernutrition in geriatric patients is

associated with poor outcome. Essential signs ofundernutrition in the elderly are unintended weightloss 45% in 3 months or 410% in 6 months as wellas a BMI below 20 kg/m2. Risk of undernutrition isindicated by loss of appetite, reduced oral intakeand stress (physical as well as psychological).

In a Cochrane analysis of 49 studies including4790 randomised elderly patients with manifestundernutrition or risk of undernutrition, positiveeffects of ONS have been shown: there is increasein energy and nutrient intake, maintenance orimprovement of nutritional status and reduction of

ARTICLE IN PRESS



13/31

mortality risk1 (Ia) (Compare 1.1, 1.2 and 1.6).ONS are, therefore, clearly recommended (A).Effects on functionality and quality of life are,however, uncertain (Compare 1.3 and 1.5).

The effects of TF in undernourished elderlypatients are unclear due to limited data. Veryoften TF is not initiated until advanced under-

nutrition has developed, which is a clear impedi-ment to the success of nutritional therapy(Compare 1.2). Results from several studies how-ever, indicate maintenance or improvement ofnutritional parameters in undernourished elderlypatients after TF2426 (III). Effects on functionalstatus and quality of life are uncertain (Compare1.3 and 1.5).

It is highly recommended to initiate nutritionalsupport, not only in manifest undernutrition, but assoon as there are indications of nutritional risk, andas long as physical activity is possible, EN

together with rehabilitative exercise

can help tomaintain muscle mass (C). Early routine nutritionalscreening is mandatory. Several tools (e.g. ESPENguidelines,117 MNA118) are available for this pur-pose.

2.2. Is EN indicated in frail elderly?

In frail elderly, ONS are recommended in order toimprove or maintain nutritional status (A).

Frail elderly may benefit from TF as long as theirgeneral condition is stable (not in terminal

phases of illness). TF is therefore recommended

early in case of nutritional risk (B), wherenormal food intake is insufficient.

Comment: Frail elderly are limited in their ADL dueto physical, mental, psychological and/or socialimpairments a well as recurrent disease. Theysuffer from multiple pathology which seriouslyimpairs their independence. Therefore they are inparticular need of help and care and are vulnerableto complications. An inadequate intake of fluidsand nutrients is a common problem in thesesubjects. Frail elderly therefore are at high risk ofundernutrition and its serious consequences. Ex-

perience has shown that the ability to eat sufficientamounts orally is inversely associated with theextent of frailty. Decreasing oral intake maytherefore be an indication of the progress orseverity of disease or frailty.

ONS lead to a significant increase in energy andnutrient intake as well as to a stabilisation orimprovement of nutritional status in mixed samplesof multimorbid elderly with acute and/or chronicdiseases, at home as well as in nursing homes andhospitals (Table 3). Effects on functional status andquality of life are uncertain due to limited data.

Effects on length of hospital stay and mortalityhave been investigated only occasionally. Potter etal.127 found a reduced length of hospital stay onlyin a subgroup of patients with adequate initialnutritional status. Data on mortality are contro-versial in frail elderly.8,127

Clinical experience shows that frail elderly, at

nutritional risk, may benefit from TF as long as theirgeneral condition is stable. Observational studiesindicate a relatively good prognosis in tube-fed frailelderly nursing home residents with good healthstatus32,57 (III) (Table 2). Although data are scarce,it is recommended that nutritional support beinitiated early, as soon as there are indications ofnutritional risk and as long as physical activity ispossible since ENtogether with rehabilitativeexercisecan help to maintain muscle mass (C).Nutritional screening has to be implemented as amatter of routine for early detection of risk of

undernutrition. Several tools (e.g. ESPEN guide-lines,117 MNA118) are available for this purpose.TF is not recommended in frail elderly who have

progressed to an irreversible final stage, e.g. withextreme frailty and advanced disease (irreversiblydependent in ADL, immobile, unable to commu-nicate, as well as high risk of death) (IV).

2.3. Is EN indicated in geriatric patients with

neurological dysphagia?

In geriatric patients with severe neurologicaldysphagia, EN is recommended in order to

ensure energy and nutrient supply and, thus, tomaintain or improve nutritional status (A). Forlong-term nutritional support PEG should be

preferred to NGT, since it is associated with lesstreatment failures, better nutritional status (A),and it may also be more convenient for the

patient. In patients with severe neurologicaldysphagia TF has to be initiated as soon as

possible (C). EN should accompany intensiveswallowing therapy until safe and sufficient oralintake from a normal diet is possible (C).

Comment: In neurological dysphagia, nutritional

therapy depends on the type and extent of theswallowing disorder. Nutritional therapy may rangefrom normal food, to mushy meals (modifiedconsistency), thickened liquids of different consis-tencies or total EN delivered via NGT or PEG.Nutritional therapy and swallowing therapy have tobe closely coordinated. Typical complications ofneurological dysphagia are aspiration with bronch-opulmonary infections136139 and undernutrition,causing extended length of hospital stay andrecurrent hospitalisations.139141 Mortality due todysphagia is significantly enhanced.139 Patients

ARTICLE IN PRESS



14/31

ARTICLE IN PRESS

Table3

Oralsupplementatio

ninmixedcohortsoffrailelderly.

Article

Study

type

Patients

Supplements

Results

Firstauthor

n

Age(yr)

Nutritional

Place

Energy

Protein

Duration

Intake

N

utritionalstatus

Functional

status

Quality

oflife

M7SD

(range)

status

(kcal/d)

(g/d)

E

Prot

Chandra

119

RCT

30

(7084)

MN

Athome

Individ.

n.a.

4wks

n.a.

n.a.

W

eight+

T

SF+

A

lb,

PA+

immuneresponse+

n.a.

n.a.

Gray-Donald7

RCT

50

78

BMI1973

Athome

500700

1726

12wks

(+)

n.a.

W

eight+

s

kinfolds

A

MC,

CC

Handgrip

falls+

Well-

being

subjective

health

(460)

Payette

3

RCT

83

8077

BMI2073

Athome

500700

1726

16wks

+

(+)

W

eight+

s

kinfolds

A

MC,

CC

Handgrip

mobility

daysinbed+

Emotional

role

functioning+

(465)

Volkert

2

RCT

46

85

MN

Athome

250

15.0

6mon

n.a.

n.a.

W

eight

ADL+(incompliant

subgroup)

n.a.

(7598)

BMI1972

Woo

46

RCT

81

73

BMI2075

Athome

500

17.0

1mon

+

+

W

eight+(m)

f

atmass+

F

FM+(m)

ADL+

activity+

mentalfunction

appetite

sleep+

n.a.

(465)

Wouters22

RCT

68

82

BMI2472

Nursing

home

250

8.8

6mon

+

+

w

eight+

F

FM,

FM,

CC

A

lb,

PA

Handgrip

ADL

mobility

sleep+

(X65)

Wouters120

RCT

55

83

BMI2472

Nursing

home

250

8.8

6mon

n.a.

n.a.

V

it.

C,

E,

Cysteine+

A

ntiox.

capacity+

n.a.

n.a.

(X65)

Banerjee

121,

122

RCT

63

81

n.a.

Nursing

home

265

18.6

14wks

+

T

SF+

A

lb,

Trf,

PA

%

T-Lymphocytes

C

omplementC3

Skinproblems+

n.a.

(6098)

Beck123

RCT

16

85

BMI20(M)

Nursing

home

380

5.0

2mon

n.a.

w

eight

n.a.

n.a.

(6596)

MNA1723,5

Ek124

RCT

482

80

28.5

%MN

Nursing

home

400

16.0

26wks

n.a.

n.a.

S

kintest+

n.a.

n.a.

Fiatarone

20

RCT

50

8871

BMI

25.5

(M)

Nursing

home

360

15.0

10wks

n.a.

W

eight+

F

FM

FM(+)

A

lb,

Fe,

HDL

V

it.

D,

E,

Folate

ADL

depression

mentalfunction

n.a.

(470)

Hankey

125

RCT

14

8172

weight

45kg,

Alb

33g/L

Nursing

home

680

n.a.

8wks

+

n.a.

w

eight(+)

T

SF,AMC

+

Albumin

n.a.

n.a.

(475)



15/31

ARTICLE IN PRESS

Table3(continued)

Article

Study

type

Patients

Supplements

Results

Firstauthor

n

Age(yr)

Nutritional

Place

Energy

P

rotein

Duration

Intake

N

utritionalstatus

Functional

status

Quality

oflife

M7SD

(range)

status

(kcal/d)

(g/d)

E

Prot

Larsson

8

RCT

435

80

29%MN

Nursing

home

400

1

6.0

26wks

n.a.

n.a.

n.a.

n.a.

Lauque

53

RCT

35

85

BMI2271

Nursing

home

300500

2

030

60days

+

+

W

eight+

Handgrip

MNA+

n.a.

(465)

MNA

17-2

3.5

Unosson

43

RCT

430

80

26%MN

Nursing

home

400

1

6.0

26wks

n.a.

n.a.

n.a.

Activity+,

mobility

mentalfunction

generalwell-being

n.a.

Hubsch19

RCT

72

86

MN

Hospital

500

3

0.0

3wks

+

+

W

eight

FFM

BCM+

A

lb,

Trf,

RBP

V

it.

B1,

C+

ADL(+)

n.a.

(7599)

McEvoy126

RCT

51

n.a.

MN

Hospital

644

3

6.4

4wks

n.a.

n.a.

W

eight+

T

SF+,

A

MC

A

lb

n.a.

n.a.

Potter127

RCT

381

83(Md)

Non-o

bese

hospital

540

2

2.5

Hospital

(Md17days)

+

n.a.

W

eight+

A

MC(+)

ADL+(MN)

n.a.

(6199)

Bunker1

28

NRT

58

80

BMI

24.4

(M);

Athome

200(in

under-

weight

patients

300)

2

0.0

12wks

n.a.

n.a.

A

lb,

PA

,RBP+

Fe,

Zn,

Se+

ly

mphocyte-populations

skintest(+)

n.a.

n.a.

(7085)

19%o20

Cederholm55

NRT

23

7471

MN

Athome

400

4

0.0

3mon

n.a.

n.a.

W

eight+

T

SF,

A

MC+

A

lb,

Orosomucoid

skintest+

Handgrip+

peakflow

n.a.

BMI17(M)

Bos1

8

NRT

23

79

MN

Hospital

400

3

0.0

10days

+

+

W

eight+

FFM+

A

lb,

Trf,

PA

C

RP,IGF-I

Im

munglobulin

C

omplementC3

Handgrip(+)

n.a.

(6990)

BMI

2173

Bourdel-M.1

29

NRT

672

83

Alb3275

Hospital

400

3

0.0

15days

+

+

n.a.

Decubitus(+)

n.a.

(465)



16/31

ARTICLE IN PRESS

Chandra

130

UCT

21

460

MN

Athome

500

17.5

8wks

n.a.

n.a.

A

lb,

PA,

Trf,

RBP+

Z

n+,

F

erritin

s

kintest+

lymphocyte

p

opulations+

n.a.

n.a.

Gray-Donald131

UCT

14

7976

MN

Athome

500

KA

12wks

+

+

A

lb(+),RBP,

H

b

lymphocyte

c

ount+

Handgrip

Well-

being+

(460)

Lipschitz132

UCT

12

75

Highrisk

Athome

1050

39.0

16wks

+

+

W

eight+

A

lb,

TIBC,

Vit.

+

H

b,

metals

lymphocytecount

s

kintest

n.a.

n.a.

Harrill133

UCT

18

89(Md)

n.a.

Nursing

home

355

13,0

30days

(+)

(+)

V

it.

A,

C,

B1,

B2+

A

lb,

Hb,

Ht,Fe

n.a.

n.a.

Welch134

UCT

15

81

Alb32g/L

Nursing

home

n.a.

n.a.

6mon

+

+

W

eight+

A

lb,

Hb,

Ht+

F

e,

TIBC,

Trf

Decubitus+

n.a.

Bourdel-M.2

3

UCT

11

87

MN

Hospital

400

30.0

4wks

W

eight+

m

usclemass

A

lb+

Handgrip+

n.a.

BMI1873

Joosten135

UCT

50

8376

BMI24.5

74

Hospital

600

19.0

1376days

+

n.a.

n

.a.

n.a.

n.a.

Alb3676g/

L

Katakity

56

UCT

12

(7184)

n.a.

n.a.

204

9.0

12wks

n.a.

n.a.

H

b

V

it.

C,

D,

B1

Handgrip+

mentalfunction

darkadaption

n.a.

ADL

activitiesofdailyliving,A

lb

albumin,

AMC

arm

muscle

circum

ference,

Antiox.

antioxidative,

BCM

bodycellmass,

BMI

body-massindex

[Kg/m

2],CC

calf

circumference,

CRP

C-reactivep

rotein,

E

energy,Fe

iron,

FFM

fatfreemass,

FM

fatmass,

Hb

hemoglobin

,Ht

hematocrit,n.a.

notavailable,M

Mean,

(m)

male

participants,

Md

median,

MN

malnutrition,

MNA

MiniNutritionalAsses

sment,mon

months,NRT

non-random

isedtrial,PA

prealbumin,

Prot

protei

n,

RCT

randomised

controlledtrial,RBP

Retinolbin

dingprotein,

Ref.

reference,

SD

standarddeviation,

Se

Selen,

TIBC

totali

ronbindingcapacity,TSF

Tricepsskinfo

ld,

Trf

Transferrin,

UCT

uncontrolledtrial,Vit.vi

tamin,

wks

weeks,Zn

Zinc.

+improvementinsupplementedgr

oup(SG)comparedtocontrolgroup(CG).

(+)trendtowardsimprovement,n

otsignificant;

nodifferenceSGCG.



17/31

with acute stroke and dysphagia often alreadyexhibit a poor nutritional status on hospital admis-sion, which negatively impacts on outcome andcosts: length of hospital stay is extended, rehabi-litation is delayed and survival is reduced.141143

These results are confirmed by the current inter-national FOOD study.144

Controlled trials studying the effects of EN afterdysphagic stroke are not available, since controlgroups without nutritional support would be un-ethical. It is common sense, however, that energyand nutrient supply has to be ensured in thesepatients in order to maintain nutritional status andto avoid the development of undernutrition. Due tothe strong physiological plausibility based on thefact that patients with severe neurological dysphy-gia are not able to sustain their life withoutnutritional support, this recommendation wasrated at the highest level.

Nutritional status: In a Cochrane analysis ofinterventions for dysphagia in acute stroke ENdelivered via PEG was associated with a greaterimprovement of nutritional status when comparedto EN delivered via NGT.145 These results are basedon a randomised controlled trial conducted byNorton et al.11 (Ib) in 30 patients and on unpub-lished data from the authors of the Cochraneanalysis from 19 further patients. In anotherrandomised controlled trial in 40 patients withneurological dysphagia (mean age 60 years), thegroup receiving PEG also exhibited weight gain aswell as an increase in mean serum albumin and

transferrin. Due to a high drop out rate noevaluation was undertaken in the NGT group12 (Ib).

Functional status: Sanders et al.64 reported animprovement in ADL in 25 stroke patients (meanage 80 years) with EN via PEG (PEG placement onaverage 14 days after stroke). At the time of PEGplacement Barthel index was 0 points (completelydependent) in 84% of patients (mean 0.5 points).After 6 months of EN a mean increase of 4.8 pointswas observed. Six patients (24%) showed a clearimprovement (Barthel index increase from 0.5 to 9points). In 10 patients (40%), however, no or only a

minimal improvement was observed (IIa).Resuming oral nutrition: Dysphagia may bereversible in stroke patients.146 In various studiesbetween 4% and 29% of patients resumed full oralnutrition after 431 months11,92,93,95,112,115 (III)(Table 4). In the British Artificial Nutrition Survey(BANS) no difference between 65- and 75-year oldelderly people and younger adults (1664 years)was found, although resumption of oral nutritionwas slightly reduced in the elderly above the age of75 years112 (Table 4). Schneider et al.115 report therate of resuming oral nutrition in different diag-

nostic groups of tube-fed patients at home. Among148 neurological patients with dysphagia (mean age75 years), 24% regained the ability to eat sufficientamounts orally within the study period of 2427494days.

Mortality: Clear statements about the effect ofEN on overall mortality after dysphagic stroke are

not possible since the investigated groups are tooheterogeneous, and control groups without nutri-tional support would be unethical (Compare 1.6).In the study of Norton et al.11 mortality after 6weeks was significantly lower in the PEG group thanin the group fed by NGT (12% vs. 57%), due probablyto the lower percentage of the prescribed intakereached in the latter. In the recent multicentreFOOD trial147 no difference in 6-month mortalitywas found between 162 dysphagic stroke patientswith PEG and 159 patients with NGT. However,these results are of limited value since only those

patients were enrolled in whom the responsibleclinician was uncertain of the best feeding practice.Furthermore the duration of the intervention isunclear and there was a greater delay to first TF inthe PEG group than in the nasogastic group. Becauseof these methodological problems, results of theFOOD trial have to be interpreted with caution.

Timing of tube placement: In patients withsevere neurological dysphagia, TF has to beensured immediately unless there are compellingreasons against it. Studies investigating the role ofearly TF after acute cerebrovascular events in age-mixed samples have shown that early TF is feasible

also in elderly patients148,149 and has a positiveimpact on survival148 and length of hospital stay144

(III). In a retrospective analysis of stroke patients(19% of patients 465 years) by Nyswonger andHelmchen,149 the group receiving TF within 72 hafter the cerebrovascular event had a reducedhospital stay compared to patients that received TFafter 72 h (III). Taylor148 found that patients, whohad spent less than 5 days without nutrient supply,had a lower mortality than patients who had morethan 5 days without nutrition. Interestingly, thisdifference was statistically significant only in

patients aged 465 years and was less distinct inyounger patients. The authors conclude that olderpatients react more sensitively to food deprivationthan younger patients and that TF should beinitiated as early as possible in this group (III).

In the recent multicentre FOOD trial147 nodifference in outcome was found between dyspha-gic stroke patients who received EN via a PEGwithin 7 days of hospital admission and anothergroup in whom TF was avoided for at least 7 days.Again, these results are of limited value because ofmethodological problems (see above).

ARTICLE IN PRESS



18/31

ARTICLE IN PRESS

Table4

Resumingoralnutritionafterenteralnutritioninelderly

patients.

Article

Study

Patients

TypeofEN

Proportion

resumingfull

oralnutrition(%)

Timeperiod

Firstauthor

Type

Place

n

Ag

e(yr)

Proportionof

elderly

M7SD

(Range)

Neurologicdysphagia

Finucane92

P

Hospital

28

82

(6899)

PEG

4%

6months

Elia112

P

Athome

2970

(X75)

EN

10%

12months

Elia112

P

Athome

1230

(6575)

EN

15%

12months

Norton11

P

Hospital

16

76

PEG

19%

6months

Schneider115

P

Athome

148

75

(Md)

(197)

EN

24%

4months(M)

Wijdicks95

R

Hospital

63

74

(Md)

(4198)

PEG

28%

236months

(Md4months)

James93

R

Hospital

126

80

(Md)

(5394)

PEG

29%

471months

Mixedcohorts

(Md31months)

Quill87

R

Hospital

55

470

(X70)

51%480y

r

NG

4%

Clarkston96

R

Hospital

42

71

.4

(3399)

PEG

7%

2months

Dwolatzky86

P

Hospital

122

(X65)

PEG/NG

8%

3months

Markgraf103

R

Hospital

84

69

714

(3598)

65%X65y

r

PEG

12%

14229days

(M108days)

Markgraf90

R/P

Hospital

54

87

(6594)

PEG

13%

14229days

(M133days)

Bussone88

R

Hospital

155

84

(7098)

PEG

14%

Larson100

R

Hospital

314

(392)

66%460y

r

PEG

14%

Skelly58

P

Hospital

74

69

(Md)

(2890)

PEG

15%

6months

Tan109

R

Hospital

44

65

(1494)

PEG

16%

144months

Howard113

R

Athome

887

79

78

(X65)

EN

17%

12months

Nicholson104

R

Hospital

168

70

(Md)

(1696)

PEG

21%

4months(Md)

Wolfsen111

R/P

Hospital

201

66

716

PEG/PEJ

21%

2757353days

(Md144days)

Sali107

R

Hospital

32

75

(3888)

PEG

22%

28months

Mitchell63

R

Athome

551

87

(Md)

(65107)

TF

25%

12months

Taylor110

P

Hospital

97

76

.5

(o197)

PEG

25%

1day7yr

(Md327days)

Abitbol26

P

Hospital

59

83

77

50%485y

r

PEG

27%

12months

Verhoef78

P

Hospital

71

66

718

(1789)

PEG

28%

12months

EN

enteralnutrition,

M

mean,

Md

Median,

NG

nasogastrictube,

P

prospective,

PEG

percutaneousendoscopicgastrostomy,PEJ

percutaneousend

oscopicjejunostomy,

R

retrospective,

SD

standarddeviation,

TF

tubefeeding,

yr

years.



19/31

ARTICLE IN PRESS

Table5

Supplementaryover

nighttubefeedinginelderlyfracture

patients.

Article

Patients

Supplement

Results

Firstauthor

n

Age

(yr)

Diagnosis

Energyand

protein/day

Duration

Intake

Nutritionalstatus

Clinicalcourse

Bastow13

58CG

80

Femurneck

+1000kcal

1639days

Tota

lintakem

Anthropometrym

ADL

64SG

81

fracture&

malnutrition

+28gprot

Md26days

Foodintake

Proteinsm

LORk

LOSk

Mo

rtality(k)

Hartgrink14

67CG

8378

Hipfracture&riskof

pressuresores

+1500kcal

7and14days,

resp.

mDespitelow

tole

rance

Intendedtofeed:

Alb,

TP

Pre

ssuresores

only40%tolerated

tub

e41wk

62SG

8477

+60gprot

Actuallyfed:

Albm

,TPm

Sullivan15

10CG

7776

Hipfracture&good

nutritionalstatus

+1383kcal

1676days

m

Alb,

transferrin

Complications

8SG

7572

+86gprot

ADL

LOS

In-hospital

mo

rtality

6-m

onthsmortalityk

ADL

activitiesofdailyliving,

Alb

albumin,

CG

controlgroup,

LORlengthofrehabilitation,

LOS

lengthofstay,Md

median,

prot

protein,

SGS

upplementedgroup,

TP

totalprotein,

yr

years.

m

increase,

k

decrease(orimprov

ementinthesupplmentedgroupcomparedtothecontrolgroup);

nodifference

betweenthegroups.

Meanormean7standarddevia

tion.



20/31

In earlier studies, long periods of 4463 daysbetween the acute event and PEG placement arenoticeable.91,93,107 Three studies on the naturalcourse of dysphagia after stroke show that sponta-neous remission of the swallowing difficulty occurs714 days after the acute event in 7386%.150152

Based on clinical experience, prognosis of dyspha-

gia seems to be better in medial cerebral infarctthan in brain stem infarct (IV). If severe dysphagiapersists longer than 14 days after the acute event,a PEG should be placed immediately. Controlledtrials on the ideal timing and length of TF inneurological dysphagia, that also consider thevarying kinds and extents of swallowing disorders,are still not available.

2.4. Is EN indicated after orthopaedic surgery in

geriatric patients?

ONS are recommended in geriatric patients after

hip fracture and orthopaedic surgery in order toreduce complications (A).

Comment: Voluntary oral intake is often insuffi-cient to meet the enhanced requirements ofenergy, protein and micronutrients after orthopae-dic surgery. Rapid deterioration in nutritionalstatus, and impaired recovery and rehabilitationare common.

The results of several randomised studies of ENafter hip fracture are summarised in a Cochraneanalysis75 that includes eight trials testing supple-mentary overnight TF, five trials with ONS and three

studies regarding the effects of supplementary oralprotein. The quality of most of the studies and theavailability of outcome data were considered poorby the authors of the Cochrane analysis.75 Inaddition, a recent randomised controlled study153

and two non-randomised trials with ONS areavailable.4,6,154

Energy and nutrient intake: Administration ofONS leads to a significant increase in energy andnutrient intake.75 However, several trials71,74,155

have shown that the daily requirements for energyand protein are still not met. This may be due to

poor compliance of less than 20%,7

to intolerance ofsupplements by some patients,155 and to require-ments being markedly increased.

Supplementary overnight TF enables the admin-istration of larger amounts of enteral formu-lae,1315 but is of limited tolerance in practice. Inthe trial of Hartgrink et al.14 only 40% tolerated thisintervention longer than 1 week and only one-quarter for the whole study period of 2 weeks.

Nutritional status: Information about the effectsof ONS on nutritional status is sparse and incon-sistent. Delmi et al.71 observed a larger increase in

albumin and transferrin levels in supplementedpatients than in the unsupplemented control group(Ib), whereas Lawson et al.154 and Williams et al.6

detected no difference with respect to serumalbumin (IIa). In the study of Lawson et al.154 BMIand mid-arm muscle circumference (MAMC) werealso unaffected, however transferrin and haemo-

globin decreased less than in the unsupplementedgroup. Williams et al.6 reported a positive effect ontriceps skinfold thickness (TSF) and MAMC in thesupplemented group. In contrast Tidermark et al.44

registered weight loss, and Brown and Seabrock74

observed decreases in body weight, mid-armcircumference (MAC) and TSF in the supplementedas well as in the control group.

Positive effects of protein supplementation onbone density and parameters of bone metabolismwere described by Tkatch et al.72 and Schurchet al.73 (Ib). A 6-month administration of protein-

enriched supplements led to a significant attenua-tion of loss of bone mineral density when comparedto the control group. Even short-term supplemen-tation (o40 days) was accompanied by a smallerdecrease in proximal femur bone mineral densitythan in the unsupplemented group. However, otherskeletal sites were unaffected. Moreover, proteinrepletion was shown to be associated with anincrease in serum osteocalcin72 and insulin-likegrowth factor-I,73 both of which are importantmediators of bone metabolism.

The effect of supplementary overnight TF onnutritional status of elderly patients with either

hip or femoral neck fracture was investigated inthree randomised controlled studies1315 (Ia)(Table 5). Initial nutritional status as well as resultswere inconsistent. Clear improvements were re-ported by Bastow et al.13 who divided theirpatients into thin and very thin according toanthropometric measurements. In both interven-tion groups (thin and very thin), anthropo-metric parameters (body weight, TSF, MAC)and postoperative prealbumin increased during1639 days. Very thin patients had the greatestbenefit from the nutritional therapy. No change in

serum albumin was observed in the study ofHartgrink et al.14 in 62 patients intended to receivesupplementary TF. An evaluation of the actuallytube-fed patients however (n 25 after 1 week,n 16 after 2 weeks), revealed increased serumconcentrations of albumin and total protein. Noeffects on plasma proteins were reported in thestudy of Sullivan et al.15 who examined patientswith a relatively good nutritional status (BMI24.1kg/m2, albumin 32 and 35 g/l, respectively),with respect to albumin, transferrin and cholester-ol values.

ARTICLE IN PRESS



21/31

Length of hospital stay: Data concerning thelength of hospital stay are inconsistent. Delmi etal.71 found a significantly shorter length of hospitalstay (including rehabilitation) in patients receivingONS (median 24 days) compared to control patients(median 40 days) (Ib). Protein administration in thetrials of Tkatch et al.72 and Schurch et al.73 was

also associated with a significantly reduced lengthof stay (30 and 21 days, respectively). In five otherstudies, however, the observed differences werenot significant.44,75

A positive impact ofsupplementary overnight TFon the length of hospital stay of geriatric patientsafter hip or femur neck fracture cannot be firmlyconcluded from the data available.1315

Functional status: Data regarding functionalstatus are heterogeneous and unsatisfactory. TheCochrane analysis of Avenell and Handoll75 refers tofour studies investigating this aspect. Only one of

them showed positive effects of ONS on ADL-functions after 6 months.44 The non-randomisedtrial of Williams et al.6 showed a trend towardsimproved mobility and greater independence athospital discharge in supplemented patients. Oralsupplementation of calcium, protein and vitaminsin the study of Espaulella et al.47 showed nosignificant changes in mobilisation, ADL status anduse of walking aids when compared to the controlgroup receiving an isocaloric placebo as well.

Bastow et al.13 assessed the time between thepatient0s operation and the achievement of phy-siotherapy goals (e.g. recovering independent mobi-

lity). Thin patients (according to anthropometricmeasurements; see above) receiving supplementaryovernight TF, achieved independent mobility in 10days, while thin control patients did so in 12 days.Very thin patients from the intervention groupreached this goal after 16 days whereas very thincontrol patients needed 23 days to regain indepen-dent mobility (Po0:05) (IIa). ADL status at discharge,however, was not affected by the intervention.13

Postoperative complications and mortality: ONShave a positive impact on the rate of postoperativecomplications. Thus, Lawson et al.154 in their

recent non-randomised study found a significantlylower rate of complications in post-operativelysupplemented orthopaedic patients than in thoseunsupplemented (IIa). In the study of Tkatch etal.72 the complication rate in protein supplementedpatients was significantly lower during hospitalstay, as well as 7 months later, compared to thecontrol group with isocaloric placebo. The pooledanalysis of five randomised studies in the meta-analysis of Avenell and Handoll75 revealed aborderline reduction of the risk of complicationsin supplemented patients (RR 0.61, 95% CI

0.361.03). When risks for mortality and complica-tions were combined in these five studies, thechances of an unfavourable outcome were reducedin supplemented patients (RR 0.52, 95% CI0.320.84)72 (Ia).

If mortality was considered separately in themeta-analysis of five studies with ONS, no reduction

in mortality risk was found.75 The same was true inthe study of Espaulella et al.47 Combining mortalityoutcome of all the studies with supplementaryovernight TF did not produce a significant riskreduction either (RR 0.99; 95% CI 0.51.97).75 Thepooled analysis of studies using ONS or overnight TFin geriatric patients with either hip or femoral neckfracture also did not show a significant reduction ofmortality risk in the enterally fed patients whencompared to controls (RR 0.94; 95% CI 0.591.50).75

2.5. Is EN indicated in the perioperative phase of

major surgery in geriatric patients?There is no evidence that nutritional therapy inelderly patients undergoing major surgery (e.g.

pancreatic surgery, head and neck surgery)should be different from that in younger pa-tients. We therefore refer to the Guidelines.Surgery and transplantation.

It is generally recognised, however, that elderlyare at higher risk of being undernourished thanyounger patients and restoration of BCM is moredifficult. Therefore, preventive nutritional supporthas to be considered.

2.6. Is EN indicated in elderly patients with

depression?

EN is recommended in depression in order toovercome the phase of severe anorexia and lossof motivation (C).

Comment: Depression is common in elderly pa-tients, but often not recognised due to thedifficulty of discriminating it from other symptomsof old age. Anorexia and refusal to eat are integralsymptoms of this disease, and depression is there-

fore regarded as a major cause of undernutrition inthe elderly.156 Undernutrition may itself contributeto the depressive states often seen in theelderly.157 Depression can be treated by severalmethods, especially by drugs, although this maytake some time to be effective. Based on positiveclinical experience and expert opinion, EN isrecommended in the elderly suffering from depres-sion in order to support the patient during the earlyphase of severe anorexia and loss of motivation,thereby preventing the development of under-nutrition with its serious consequences (C).

ARTICLE IN PRESS



22/31

ARTICLE IN PRESS

Table6

Prevalenceofaspira

tionpneumoniaintube-fedelderlyp

atients.

Firstauthor

Stud

ytype

Patients

Aspi

rationpneumonia(AP)

n

Age

(years)

Diagnos

es

Befo

re

After

Timeperiod

Patel160

P

24

72

CVE,

CA,

dementia

58%

14/24(58%)(all)

UntilAPordeath

12/14(86%)(withAP)

Paillaud35

R

73

8379

Mixed

15%

53%

2,

6,

12months

Sali107

P

32

75

Mixed

9%

3/5(60%)deathsduetoAP

2488days

Abitbol26

P

59

8377

Nursing

homeresidents

49%

51%

30days

Baeten161

P

90

72

CA,

neu

rologicaldisease

6%

12months

Wijdicks95

P

63

74

Apoplex

16%

Hospitalstay

Peschl151

P

33

76

Cerebraldysfunctions

18%

236months

Kaw36

R

46

74

Neurolo

gicaldisease,

dementia

22%

6months

Stuart108

R

125

70

CA,

dem

entia,

cachexia

28%

12,

18months

Bourdel-Marchasson60

R

46

8179

Mixed

39%

30days

Fay27

R

80

70

Apoplex,

dementia,

PEG

6%/32%

14/192days

29

70

Parkinson

NG

24%/46%

14/141days

Golden32

R

102

8976

dementia

51%

6months

AP

aspirationpneumonia,

CA

carcinoma,

CVE

cerebrovascularevent,

NG

nasogastrictube,

P

prospective,R

retrospective.

Meanormean7standarddevia

tion.



23/31

2.7. Is EN indicated in dementia?

ONS or TF may lead to an improvement innutritional status in demented patients. In earlyand moderate dementia ONSand occasionallyTFmay contribute to ensuring an adequateenergy and nutrient supply and to preventingundernutrition from developing; they are there-

fore recommended (C). In those with terminaldementia, TF is not recommended (C). Thedecision in each case must be made on anindividual basis.

Comment: An indequate intake of energy andnutrients is a common problem in dementedpatients. Undernutrition may be caused by severalfactors including anorexia (common cause: poly-pharmaco-therapy), insufficient oral intake (for-getting to eat), depression, apraxia of eating or,less often, enhanced energy requirement due tohyperactivity (constant pacing).158 In advancedstages of dementia, dysphagia may develop andmight be an indication for EN in a few cases.

Some studies with ONS have shown improve-ments in body weight (Ib)49,159 (IIa)50. In tube-feddemented elderly patients, two studies reportedweight gain31,32 (III), but two others reported nochange (III)24 (IIb)86. Available trials regarding theeffects of ONS (Ib)49 (IIa)50 or TF24,33,36 onfunctional status, report no improvement (Com-

pare 1.5). In terms of survival most studies show nobenefit.33,81,84,94 On the other hand, Rudberg etal.61 described lower mortality, compared to con-

trols, at 30 days and 1 year in enterally fed patientswith severe cognitive impairment (IIb). Very lowmortality rates have been reported in PEG-feddemented nursing home residents.32,57,62 On theother hand, in one retrospective study comparingmortality rates in different diagnostic groups,outcome was worst among the demented59 (III).

In conclusion, tube-fed demented patients varyconsiderably with respect to their prognosis. Out-come and also the success of nutritional therapy indemented patients are strongly influenced by theseverity of disease, the kind and extent of

comorbidities and by their general condition. It istherefore recommended that adequate and highquality nutrition is ensured, especially in the earlyand middle stages of dementia, in order to preventundernutrition developing and to help maintain astable general condition (C).

TF may be useful in some demented patients.The following aspects have to be considered indecision-making:

presumed or previously expressed wishes of thepatient with respect to TF;

severity of the disease; the individual prognosis and life expectancy of

the demented patient; the anticipated quality of life of the patient with

or without TF; the anticipated complications and impairments

due to TF;

the mobility of the patient.

The decision for or against TF has always to bemade individually and together with relativesand care givers, legal custodian, family doctorand therapists, and in case of doubt, with legaladvice.

For patients with terminal dementia (irreversi-ble, immobile, unable to communicate, completelydependent, lack of physical resources) TF is notrecommended (C).

2.8. Is EN indicated in geriatric patients withcancer?

In principal, nutritional therapy in geriatricpatients with cancer does not differ from young-er cancer patients (see Guidelines on Non-surgical oncology).

Comment: It is generally recognised, however, thatelderly are at higher risk of being undernourishedthan younger patients and restoration of BCM ismore difficult. Therefore, preventive nutritionalsupport has to be considered.

2.9. In patients with dysphagia does TF preventaspiration pneumonia by improving functional

status?

Due to the heterogeneity of the studies, and lackof data on prevalence before the TF, firmconclusions can not be drawn.

Comment: Dysphagia may enhance aspiration frompharyngeal contents, but, on the other hand, TFmay enhance reflux and aspiration of gastriccontents. Several studies have reported the pre-valence of aspiration pneumonia in tube-fed

elderly patients (Table 6). Due to the heterogeneityof patient groups and lack of data on theprevalence of aspirat

espen_guidelines_geriatrics nutrition 200625330-60.pdf

Documents