KeKK y words: Biompedance, disability, muscle strength, physical performance, resistance.Correspondence: Mikko P. BjB örkman, MD, PhD, Researcher, Clinics of Internal Medicine and Geriatrics, Helsinki University Central Hos-pital, POB 340, FI-00029 HUS, Helsinki, Finland.E-mail: mikko.bjb orkman@helsinki.fiReceived Received January 13, 2012; accepted in revised form July 10, 2012.First published ahead of print November 12, 2012 as DOI: 10.3275/8719

Bioimpedance spectcc roscopy as a measure of phyhh sicalfuff nctcc ioning in nursing home residentnn s

Aging Clinical and Expx erimental Research

Mikko P. BjB örkman1, Harriet Finne-Soveri2, Taru K. Pilvi3 and Reiji o S. Tilvis1,4

1Clinics of Internal Medicine and Geriatrics, Helsinki University Central Hospital, 2Division of Healthand Social Services, The National Institute for Health and Welfare, 3Valio Ltd, R&D, 4Institute of ClinicalMedicine, Department of Internal Medicine, Geriatric Unit, University of Helsinki, Helsinki, Finland

ABSTRTT ARR CT. Backgk rorr und and aims: Intracellular re-sistance (R(( i), a raw measure of bioimpedance spec-troscopy (B(( ISII )S , has been sugggg ested foff r assessment ofomuscle health. The associations ofo repe eated BISII mea-surerr ments with fuff nctioning and nutritional status wererrinvestigated in nursrr ing home residents suffff eff ring frompoor health and disabilities. MeMM thods: A total ofo 106nursrr ing home residents (a(( ge 83±8 yrsrr , 75% women)were recruited. Whole body and calfl BISII measures(lean body mass, resistance at 50 kHzHH , and Ri), heightand calfl electrorr de distance (D(( )D wererr used to calculate sixmuscle indices. HaHH nd grip and knee exee tension strerr ngthswere measured and data on Activities ofo Daily Living(A(( DL), mobility scorerr , and Mini Nutritional Assessment(M(( NMM ANN ) collected. Repeated measurements were per-foff rmed at 3 (B(( ISII )S and 6 months (B(( ISII , muscle strength,ADL, mobility, and MNMM ANN ). Results: All bioimpedancemuscle indices were lower in women than men and as-sociated with MNMM ANN . HoHH wever,r the calfl skeletal muscle in-dex (S(( MIMM =D2/2 R// i) associated with muscle strength mea-surerr ments at baseline and consistently with mobility andADL also at 6-month re-examination. WhWW en comparedto the highest tertile ofo SMIMM percent change (c(( ut point+0.7%), the patients in the lowest tertile (c(( ut point -11.6%) had a 5.3-f- off ld risk (p(( =0.00400 )4 foff r mobility declinewithin the 6-month foff llow-up. ThTT is risk association alsoremained significant aftff er controlling foff r age, gender,baseline mobility, and percent change in body weight.Conclusions: Calfl intracellular resistance related toelectrorr de distance is associated with the activities ofo dai-ly living rerr fe lecting mobility in typyy ical nursrr ing home rerr s-idents and a decrease in this index indicates a marked-ly increased risk foff r mobility decline.(A(( ging Clin ExEE px Res 2012; 24: 612-618)8©2012, Editrice Kurtis

INTRODUCTIONDisabilities related to muscle health are common

threats to the autonomy of the oldest old. Thus, the im-portance of objb ective assessment and measurement ofmuscle mass, strength, and performance has been em-phasized (1). These measurements oftff en include magneticresonance imaging (MRI), computerized tomography,dual x-ray absorptiomery, hand grip strength, quadri-ceps strength and usual gait speed in addition to otherlower limb tests. Howevevv r, advavv nced cognitivevv impairmentand severe physical frailty hamper or are obstacles forthese measurements. Thus, the assessment of physicalfunctioning in these patients has been based on obser-vational scales on activities of daily living.

During recent decades evidence on bioimpedance as ameasure of muscle health and nutritional status has beenaccumulating (2, 3). The devices today are oftff en portableand the co-operation needed for the measurement isminor compared to the imaging techniques. Howevevv r, theaccuracy of bioimpedance in determining body compo-sition for example in obese and oldest old patients hasbeen questioned, and population-specific algorithmsbased on age, gender, height, body weight have been de-veloped in order to increase accuracy (4). However, pre-cise measurement of height and body weight for examplecan be also problematic (knkk ee height, liftff to weight) in frailolder patients with severe cognitive impairment andphysical disability.

In addition to the algorithm-based bioimpedance anal-ysis, single-frequency (50 kHz) bioimpedance vector anal-ysis of raw resistance and reactance data has been con-sistently associated with muscle strength in middle-agedand older patients (5). Furthermore, new bioimpedancespectroscopy devices scan a wide range of frequencies al-lowing the calculation of intracellular resistance (Ri),

612 Aging Clin Exp Res, Vol. 24, No. 6

Bioimpedance and physical functioning

Aging Clin Exp Res, Vol. 24, No. 6 613

which is a surrogate for intracellular water (ICW) (6).Since fat and bone cells have very low content of water,the measurement of segmental Ri of calf could be a use-ful method to measure lower limb skeletal muscle.

In order to test whether and to what extent measuresof bioimpedance spectroscopy (BIS) could be used as asurrogataa e indicataa or of phyhh sical fuff nctioning, a series of mea-surements were performed repeatedly in an older sampleof typical Finnish nursing home residents.

METHODSThe data are based on randomized controlled trial of

whey protein supplementation conducted in a municipalnursing home. The study procedures have been pre-sented in detail elsewhere (BjB örkman et al. submitted).Briefly, a total of 106 out of 134 residents living in thetwtt o units were willing to partrr icipate in this study. Both thetest and control groups received a glass of regular fruitjuice three times a day for 6 months, but in the testgroup mixtxx utt re of whww ey proteins (75% β-lactoglobulin, 25%α-αα lactalbumin) weww re added to the juice totaling in dailyll sup-plementation of approximately 20 g of protein. Theroutine care of residents on these wards included physicalrehabilitation with MOTOmed Movement Therapy (7).The study protocol was approved by the Ethics Com-mittee of the Helsinki University Central Hospital. In-formed consent was obtained from each resident and/ortheir next of kin before any study procedures.

Complete data of the Minimum Data Set 2.0 (MDS2.0) of interRAI were collected. The MDS 2.0 derivedlong form of the Activivv ties of Daily Livivv ng score (A(( DL)LL (8),Cognitive Performance Scale (CPS) (9), and Case Mix In-dex (CMI) (10) were used to assess physical performance,cognitive function, and the use of resources, respectively.ADL scores range from 0 to 28, where 0 stands for in-dependent and 28 for totally dependent. CPS scoresrange from 0 to 6, whww ere 0 stands for intact cognition and6 for very severe dementia. CMI score of a typical nurs-ing home resident is 1.000 and higher values indicate in-creased use of resources. FuFF rtrr hermore, a mobilitytt score us-ing the MDS 2.0 was calculated. The means of 7-dayaa self-performance (0=independent, 1=supervision, 2=limitedassistance, 3=extensive assistance, 4=total dependence,8=activity did not occur) and provided physical support(0=no setup or physical help from staff, 1=setup help on-ly, 2=one person physical assist, 3=two+ persons phys-ical assist, 8=activity did not occur) scores in transferbetween surfaces, walking in room, and toilet use weresummed together. Nutritional status was evaluated withMini Nutritional Assessment (MNA) (11). A mean ofthree measurements was recorded from both limbs ofhand grip strength in kilograms (JAMAR dynamometer,Saehan Corp., Masan, Korea) (12) and knee extensionstrength in Newtons (MicroFet2, Hoggan Health, WestJordan, UT) (13). Body composition was measured with

a single channel, tetra polar BIS devivv ce (SFB7, ImpediMedLtLL d., Eight Miles Plains, Queensland, Australia) that scans256 frequencies between 4 and 1000 kHz. Raw datawere analyzed with the supporting software (version5.3.1.1, SFB7, ImpediMed Ltd., Eight Miles Plains,Queensland, Australia) supplied by the manufacturer in or-der to obtain values for Ri (Ω), resistance at 50 kHz (Ω),and fataa free mass (kgkk ). The whww ole bodydd vavv lues weww re relataa edto height and calf measurements to electrode distance asfollows: whole body muscle index (WMI) = height2/w// holebody Ri (cm2/Ω), skeletal muscle index (SMI) = elec-trode distance2/c// alf Ri (cm2/ Ω), whww ole bodyd 50 kHkk z mus-cle index (W(( 50MI) = height2/w// hole body resistance at 50kHz (cm2/Ω), and skeletal 50 kHz muscle index (S50MI)= electrode distance2/calf resistance at 50 kHz (cm2/Ω)(14, 15). The fat free mass calculataa ed by BIS softff wtt are wasused to determine the BIS muscle mass index (BIS MMI)= fat free mass/height2 (kg/m2). Finally, the calculatedmuscle mass according to Jansen et al. (16) was used todetermine Jansen muscle mass index (Jansen MMI) =muscle mass according to Jansen et al./height2 (kg/m2).The impedance measurements were repeated after 3and 6 months. MDS 2.0 derived variables and MNAwere collected again after 6 months.

The continuous variables descriptive values were ex-pressed by means with standard deviations (SD) and me-dians with range. For the vavv riables with normal (Gaussian)distribution, statistical comparisons between the groupswere made by student’s t-test or analysis of variance.Group differences of variables with non-normal (skewed)distributuu ion weww re compared bybb either Mann-W- hitney U-testor Kruskal-Wallis test. Adjusted bivariate correlationswere calculated by partr ial correlations. Logistic regressionmodels were used to calculate the odds ratios and toadjd ust for confounding factors.

RESULTLL SThe maja ority of the participants (age 83±8 yrs, 75%

women) required assistance in their activities of daily liv-ing and/o// r suffered from cognitivevv impairment. Nearlyll allpatients (91.8%) were at least at risk of malnutrition(MNA<24), but only 20.4% filled the MNA criteria (<17)of malnutrition. The mean baseline CMI±SD was1.071±0.322, indicating that the patients were typicalnursing home residents in Finland.

All muscle indices were lower in women (age 84.7±8.1yrs) than in men (age 79.6±7.6 yrs, p=0.005) also aftff ercontnn rtt olling foff r thtt e age diffff eff rerr nce betee wtt eww en gendnn ersrr (T(( aTT ba le 1).AfAA tff ett r contnn rtt orr lling foff r botoo htt agaa e andnn gendnn er, thtt e bioimpedancemuscle indices associated moderately with MNA both atbaseline and 6-montnn h re-examinataa ion (T(( aTT ba le 2). AtAA baselinethe calf SMI and BIS MMI associated with both musclestrength measurements, mobilitytt score and ADL. The as-sociations of SMI with mobility and ADL remained sig-nificant at 6 months as well, whww ereas the significance with

M.P. BjB örkman, H. Finne-Soveri, T.K. Pilvi et al.

614 Aging Clin Exp Res, Vol. 24, No. 6

the muscle strength measurements was lost. Howevevv r, theassociations of other tested bioimpedance muscle in-dices (WMI, S50MI, W50MI, BIS MMI, and Jansen MMI)with physical functioning and muscle strength appearedweaker and were less consistent (Table 2). SMI correlat-ed modestly with WMI (r=0.494, p<0.001) S50MI(r=0.572, p<0.001), W50MI (r=0.599, p<0.001), BISMMI (r=0.566, p<0.001) and Jansen MMI (r=0.598,

p<0.001). Furthermore, baseline measurement of handgrip strength was impossible in 36% of the patients, therespective figure being 44% for knee extension strength.In contrast to bioimpedance indices, calf circumference didnot correlate significantlyl with any of the muscle functions(mobility score, ADL, hand grip and knee extension) orMNA either at baseline or at 6-month re-examination af-ter controlling for age and gender (data not shown).

Men Women p-value

Number 26 80Age, yrs 79.6±7.6 84.7±8.1 0.005Case mix indexa 0.98 (0.45-1.31) 1.15 (0.42-1.87) 0.076Cognitive performance scale, 0-6a 3 (0-6) 3 (0-6) 0.563Mini Nutritional Assessment score 20.8±3.4 19.1±3.4 0.042Activities of Daily Living, 0-28a 14.4±6.8 17.4±8.0 0.100Mobility scoreb, 0-24a 7.7±4.8 8.9±5.7 0.361Hand grip strength, kg (n=21/52) 18.1±6.9 9.7±5.1 <0.001Knee extension strength, n (n=19/44) 170±60 131±48 0.009Body mass index, kg/m2 24.9±4.5 24.2±5.1 0.549Muscle mass indices adjd usted for age

Whole body muscle indexc, cm2/Ω 14.3±0.9 9.6±0.5 <0.001Skeletal muscle indexd, cm2/Ω 1.28±0.08 0.9±0.05 <0.001Whole body 50 kHz muscle indexe, cm2/Ω 53.8±1.7 39.0±0.9 <0.001Skeletal 50 kHz muscle indexf, cm2/Ω 8.2±0.4 5.8±0.2 <0.001BIS muscle mass indexg, kg/m2 16.8±0.5 14.7±0.3 0.002Jansen muscle mass indexh, kg/m2 8.6±0.2 6.0±0.1 <0.001

Values are mean±standard deviation for unadjd usted and mean±standard error for adjd usted variables.aHigher scores indicate worse performance; bphysical performance and provided physical support in transfer, walking in the room, and toileting;cheight2/w// hole body intracellular resistance; delectrode distance2/c// alf intracellular resistance; eheight2/w// hole body resistance at 50 kHz; felectrode distance2/c// alfresistance at 50 kHz; gfat free mass according to bioimpedance spectroscopy (BIS)/height2; hmuscle mass according to Jansen et al. 2004/height2.

Table 1 - Characteristics ofo residents by gender.

WMIb, SMIc, W50MIe, S50MIf, BIS MMIg, Jansen MMMM Ih,cm2/Ω cm2/Ω cm2/Ω cm2/Ω kg/m2 kg/m2

Baseline evaluationMini Nutritional Assessment 0.200(*) 0.562** 0.390** 0.356** 0.443** 0.325**Activities of Daily Livinga -0.030 -0.461** -0.172 -0.184* -0.224* -0.273*Mobility scorea,b -0.103 -0.461** -0.159 -0.214* -0.251* -0.238*Hand grip strength (n=68/106) 0.434** 0.502** 0.305* 0.171 0.373** 0.243*Knee extension strength (n=59/106) 0.255* 0.284* 0.237* 0.176 0.301* 0.192

Six-month re-examinationMini Nutritional Assessment 0.326** 0.507** 0.401** 0.440** 0.308** 0.355**Activities of Daily Livinga -0.287** -0.512** -0.203* -0.211* -0.088 -0.292**Mobility scorea,b -0.338* -0.549** -0.172 -0.201* -0.153 -0.284*Hand grip strength (n=59/97) 0.289* 0.197 0.128 -0.095 0.259* 0.216Knee extension strength (n=50/97) 0.086 0.144 0.062 0.052 0.193 0.200

(*)p<0.100, *p<0.050, **p<0.010. aHigher scores indicate worse performance; bphysical performance and provided physical support in transfer, walking inthe room, and toileting; cheight2/w// hole body intracellular resistance; dheight2/w// hole body resistance at 50 kHz; eelectrode distance2/c// alf intracellular resistance;felectrode distance2/calf resistance at 50 kHz; gfat free mass according to bioimpedance spectroscopy/height2; hmuscle mass according to Jansen et al.2004/height2.

Taba le 2 - PaPP rtial agegg and gegg nder adjd usted correrr lations ofo bioimpedance muscle indices with nutritional status, physical fuff nctioning,gg and mus-cle strentgth.

Bioimpedance and physical functioning

Aging Clin Exp Res, Vol. 24, No. 6 615

The SMI values were veryrr stable in the repeated mea-surements. In fact, SMI measured at 3 and 6 months cor-relata ed closelyll wiww th the baseline vavv lues in both intervrr evv ntiongroups (Fig. 1). However, the mean SMI decreased from0.98±0.42 to 0.93±0.42 (p(( =0.009) within the 6-monthfollow-up. Furthermore, the 6-month changes in WMI,W50MI, BIS MMI and Jansen MMI were statistically in-significant (data not shown). The changes in SMI associ-ated with the changes in mobility scores (Fig. 2) and lessso with those in ADL scores (3-month p=0.073, 6-month p=0.190). When compared to the highest tertileof SMI percent change (cut point +0.7%), the patients inthe lowest tertile (cut point -11.6%) had a 5.3-fold risk

(p(( =0.004) for mobilitytt decline within the 6-month follow-up (T(( aba le 3). This risk associataa ion also remained significantafter controlling for age, gender, and baseline mobility.FuFF rtrr her controlling for percent change in body weight andintervention group did not abolish this association.

DISCUSSIONThe measures of bioimpedance associated well with

physical functioning in these typical nursing home res-idents. However, in contrast to the other bioimpedancemuscle indices, a decrease in calf Ri indicated significantdecline in mobility within the 6-month follow-up. Of thehuge database this simple measurement of calf Ri in re-lation to electrode distance emerged as an indepen-dent associate of activities of daily living reflecting mo-bility.

Fig. 1 - Correlation ofo baseline calfl intracellular resistance skele-tal muscle index (S(( MIMM )I with the three- (A(( ) and six-month (B(( )measurements by intervention group (test group = solid lines anddots; control group = dashed line and open circles). Three-monthcorrelations: test-group r=0.883, p=0.001; Control groupr=0.898, p<0.001. Six-month correlations: test-group r=0.917,p=0.001; Control group r=0.890, p<0.001.

Fig. 2 - The mean three- (A(( ) and six-month (B(( )B changes in calfl in-tracellular resistance skeletal muscle index (S(( MIMM )I by six-monthchanges in mobility. Standard error indicated by T-TT bars.

Baseline calf SMI, cm2/Ω

Thre

e-m

onth

calf

SM

I,cm

2 /Ω

0.00 0.50 1.00 1.50 2.00 2.50

2.50

2.00

1.50

1.00

0.50

0.00

A

Baseline calf SMI, cm2/Ω

Six

-mon

thca

lfS

MI,

cm2 /Ω

0.00 0.50 1.00 1.50 2.00 2.50

2.50

2.00

1.50

1.00

0.50

0.00

B

Six-month change in mobility score

Mea

nth

ree-

mon

thch

ange

inca

lfS

MI,

%

Improvement No change Decline

p=0.07630.00

20.00

10.00

0.00

-10.00

Six-month change in mobility score

Mea

nsi

x-m

onth

chan

gein

calf

SM

I,%

Improvement No change Decline

p=0.07315.00

10.00

5.00

0.00

-5.00

-10.00

-15.00

A

B

M.P. BjB örkman, H. Finne-Soveri, T.K. Pilvi et al.

616 Aging Clin Exp Res, Vol. 24, No. 6

bility decline in the present study was also independentof changes in body weight. The number of obese olderpatients with or without co-existing sarcopenia is in-creasing (26, 27) and evidence on benefits of controlledweight loss programs of older people aiming at mainte-nance of skeletal muscle mass is also emerging (28-30).Thus, there will be an increasing need for methods al-lowing frequent follow-up of body composition in a clin-ical setting including the nursing homes. Even thoughhigh correlation between algorithm-based bioimpedanceanalysis and dual x-ray absorptiometryrr in relatively youngwell functioning older adults has been shown (3), the re-sults of this study suggest that the use of raw BIS datasuch as calf Ri may be more appropriate in frail and dis-abled older patients.

It can be argued that the an even simpler measure-ment of calf circumference could be used as a surrogatefor muscle mass. However, in contrast to bioimpedanceindices tested in the present study, calf circumference didnot correlate significantly with any of the muscle func-tions (mobility score, ADL, hand grip and knee exten-sion) or MNA either at baseline or at 6-month re-ex-amination after controlling for age and gender. BISmeasures several frequencies to measure bioelectricalimpedance (Z). Biological cell membranes behave ascapacitors, and Z is frequency-dependent. With directcurrent or at zero frequency, there is theoretically noconduction through biological cells, and Z is purely re-sistive and a function of extracellular water (ECW).With alternating current, capacitors charge and dis-charge the current at the frequency rate. Therefore, asfrequency increases, the amount of ICW measured in-creases. At some infinitely high frequency (>10 MHz),the charge and discharge of current through the cells be-

Previvv ous literature has mostly addressed the use of BISin evaluation of hydration (17, 18). However, calf watervolume measured by algorithm-based BIS has also beenshown to correlate highly (r=0.93) with segmental mus-cle volume measured by MRI in healthy adults (19). Fur-thermore, the changes in muscle volume after 72-hourbed rest measured by BIS also correlated (r=0.70) withthose measured by MRI. Whole body and calf Ri mea-sured by BIS have also been associated with maximaloxygen uptake independently of age, gender and self-re-ported physical activity in healthy men and women (20).Finally, ICW volume of the lower leg calculated from calfBIS Ri measurements has been associated with lowerlimb muscle strength and performance in healthy youngeras well as older men (age 20-88 years) (21). Thus, the re-sults of the present study are consistent with these pre-vious observations, but to the best of our knowledgethis is the first study to show the association between thechanges in raw calf Ri data and mobilitytt decline. The factthat the present study used raw data without population-specific algorithms, that most BIS devices are portableand that the measurements require low co-operation(comparable to electrocardiography) of the patients em-phasizes the importance of this observation.

The importrr ance of frequent body weight measurementin nursing home residents has been emphasized. Weightloss of both nursing home residents and communitydwelling older people has been associated with a numberof adverse health outcomes including death (22, 23).However, the accurate measurement of body weightby nursing home staff has proven to be challenging(24). Furthermore, at least in older men it is particularlythe loss of skeletal muscle mass that independently pre-dicts all-cause mortalityt (25). The risk association for mo-

Tertile I Tertile II Tertile III

Three-month change in SMIbNumber at risk 27 28 28Decline in mobilitya 14 14 8Unadjd usted 2.69 (0.88-8.21) 2.50 (0.83-7.55) 1.00 (reference)Model 1c 3.50 (1.05-11.63) 3.16 (0.98-10.17) 1.00 (reference)Model 2d 3.39 (1.00-11.5) 3.14 (0.97-10.15) 1.00 (reference)Model 3e 3.28 (0.97-11.17) 3.40 (1.02-11.31) 1.00 (reference)

Six-month change in SMIbNumber at risk 28 27 28Decline in mobilitya 19 9 8Unadjd usted 5.28 (1.69-16.51) 1.25 (0.40-4.93) 1.00 (reference)Model 1c 7.34 (2.12-25.41) 1.58 (0.47-5.29) 1.00 (reference)Model 2d 7.78 (2.09-28.93) 1.61 (0.48-5.41) 1.00 (reference)Model 3e 7.74 (2.08-28.85) 1.63 (0.48-5.52) 1.00 (reference)

aPhysical performance and provided physical support in transfer, walking in the room, and toileting; belectrode distance2/calf intracellular resistance; cadjd ustedfor age, sex, and baseline mobilitytt ; dadjd usted for height, and percent change in body weight in addition to model 1; eadjd usted for intervrr ention group (test vs con-trol) in addition to model 2.

Table 3 - Odds ratios with 95% confidence intervals foff r decline in mobilitya during six months according to tertiles ofo percentchange in skeletal muscle index (S(( MIMM )I b

ff.

Bioimpedance and physical functioning

Aging Clin Exp Res, Vol. 24, No. 6 617

come so fast that the effects of the capacitors become in-significant. Z becomes purely resistive, and both ECWand ICW are fully measured. Once the resistance of ze-ro and infinite frequency are determined by BIS mea-surement, ICW resistance can be computed by simplesubtraction. Skeletal muscle tissue holds a large amountof water that is partitioned into ICW and ECW (thesum of interstitial fluid and blood plasma) fractions.Therefore, in addition to muscle cell mass, skeletalmuscle also contains ECW that may not be related tomuscle strength and function. Thus, the poorer perfor-mance of other bioimpedance measures in comparisonto SMI in the present study may be related to theirlacking ability to differentiate between intramuscularwater compartments (intra- vs extracellular).

The lack of a standard measurement of muscle mass(MRI, computerized tomography or dual x-ray absorp-tiometryrr ) is a maja or limitation to this study. However, thetransportation of residents to a radiology unit was not ac-ceptable. Secondly, partly due to the relatively smallsample size we are not able to rule out residual con-founding. Thirdly, the number of men was small. Finally,the mobility score used in this study was not a validatedtool for assessment of physical functioning. However, theseven items (bed mobility, transfer between surfaces, lo-comotion, dressing, eating, toilet use, and personal hy-giene) included in the MDS 2.0 ADL score are not all re-lated to muscle health. Transfer between surfaces, walk-ing in room, and toilet use are the very basic activities ofdailyll livivv ng and they oftff en present significant challenges tothe muscle healtll h of nursing home residents. FuFF rtrr hermore,the frequent and noticeable nature of these items was likekk -lyll to increase the accuracycc of the nurse reportrr s on self-ff per-formance and provided physical support. The mobilityscore correlated also strongly with ADL (r=0.865,p<0.001) and fairly with CPS (r=0.339, p<0.001). Therepeated measurements of bioimpedance measures andphysical functioning in truly old and disabled typical nurs-ing home population are the most important strengths ofthis study.

CONCLUSIONSSMI of BIS is based on raw measurement of calf Ri re-

lated to electrode distance and does not require popula-tion specific algorithms. It is associated with the activitiesof daily living reflecting mobility in typical nursing homeresidents and a decrease in this index indicates a marked-ly increased risk for mobility decline.

ACKNOWLEDGEMENTSThis work was supported in co-operation by Finnish Funding

Agency for Technology and Innovation and Valio LtLL d. The dedicationand commitment of Kustaankartano nursing home staff, residentsand their next of kin during this trial are greatly appreciated. The au-thors also kindly acknowledge InterRAI for the use of instrumentsand data.

REFERENCES1. Cruz-Jentoft AJ, Baeyens JP, Bauer JM et al. Sarcopenia:

European consensus on definition and diagnosis: Report of theEuropean Working Group on Sarcopenia in Older People. AgeAgeing 2010; 39: 412-23.

2. Diaz EO, Villar J, Immink M, Gonzales T. Bioimpedance or an-thropometry? Eur J Clin Nutr 1989; 43: 129-37.

3. Ling CH, de Craen AJ, Slagboom PE et al. Accuracy of directsegmental multi-frequency bioimpedance analysis in the assess-ment of total body and segmental body composition in middle-aged adult population. Clin Nutr 2011; 30: 610-5.

4. De Lorenzo A, Andreoli A, Matthie J, Withers P. Predictingbody cell mass with bioimpedance by using theoretical methods:a technological review. J Appl Physiol 1997; 82: 1542-58.

5. Norman K, Pirlich M, Sorensen J et al. Bioimpedance vevv ctor anal-ysis as a measure of muscle function. Clin Nutuu r 2009; 28: 78-82.

6. Jaffrin MY, Morel H. Body fluid volumes measurements byimpedance: A review of bioimpedance spectroscopy (BIS) andbioimpedance analysis (BIA) methods. Med Eng Phys 2008;30: 1257-69.

7. Diehl W, Schüle K, Kaiser T. Use of an assistive movementtraining apparatus in the rehabilitation of geriatric patients.NeuroGeriatrie 2008; 5: 3-12.

8. Carpenter GI, Hastie CL, Morris JN, Fries BE, Ankri J.Measuring change in activities of daily living in nursing home res-idents with moderate to severe cognitive impairment. BMCGeriatr 2006; 6: 7.

9. Morris JN, FrFF ies BE, Mehr DR et al. MDS Cognitivevv PerformanceScale. J Gerontol 1994; 49: M174-82.

10. FrFF ies BE, Schneider DP, FoFF ley WJWW , Gavaa avv zzi M, Burkekk R, CorneliusE. Refining a case-mix measure for nursing homes: ResourceUtilization Groups (RUG-III). Med Care 1994; 32: 668-85.

11. Guigoz Y, Lauque S, Vellas BJ. Identifying the elderly at risk formalnutrition. The Mini Nutritional Assessment. Clin Geriatr Med2002; 18: 737-57.

12. Bohannon RW. Hand-grip dynamometry predicts future out-comes in aging adults. J Geriatr Phys Ther 2008; 31: 3-10.

13. Suzuki M, Yamada S, Inamura A et al. Reliability and validity ofmeasurements of knee extension strength obtained from nursinghome residentnn s wiww th dementia. Am J Phyhh s Med Rehaba il 2009; 88:924-33.

14. Yamada Y, Masuo Y, Yokoyama K et al. Proximal electrodeplacement improves the estimation of body composition in obeseand lean elderlyll during segmental bioelectrical impedance analyll sis.Eur J Appl Physiol 2009; 107: 135-44.

15. Yamada Y, Schoeller DA, Nakamura E, Morimoto T, Kimura M,Oda S. Extracellular water may mask actual muscle atrophy dur-ing aging. J Gerontol A Biol Sci Med Sci 2010; 65: 510-6.

16. Janssen I, Baumgartrr ner R, Ross R et al. Skeletal muscle cutuu pointsassociated with elevated physical disability risk in older men andwomen. Am J Epidemiol 2004; 159: 413-21.

17. Raimann J, Liu L, Ulloa D, Kotanko P, Levivv n NW. Consequencesof ovevv rhydration and the need for dryrr weight assessment. ContribNephrol 2008; 161: 99-107.

18. Zhu F, Kotanko P, Handelman GJ et al. Estimation of normal hy-drataa ion in dialyll sis pataa ients using whww ole bodydd and calf bioimpedanceanalysis. Physiol Meas 2011; 32: 887-902.

19. Bartok C, Schoeller DA. Estimation of segmental muscle volumeby bioelectrical impedance spectroscopy. J Appl Physiol 2004;96: 161-6.

M.P. BjB örkman, H. Finne-Soveri, T.K. Pilvi et al.

618 Aging Clin Exp Res, Vol. 24, No. 6

20. Stahn A, Strobel G, Terbr lanche E. VO(2maxaa )x prediction from mul-ti-frequency bioelectrical impedance analysis. Physiol Meas 2008;29: 193-203.

21. Wallace JI, Schwartz RS, LaCroix AZ, Uhlmann RF, PearlmanRA. Involuntary weight loss in older outpatients: incidence andclinical significance. J Am Geriatr Soc 1995; 43: 329-37.

22. Ritchie CS, Locher JL, Roth DL, McVie T, Sawyww er P, Allman R.Unintentional weight loss predicts decline in activities of daily liv-ing function and life-space mobility over 4 years among com-munity-dwelling older adults. J Gerontol A Biol Sci Med Sci2008; 63: 67-75.

23. Salva A, Coll-Planas L, Bruce S et al. Nutritional assessment ofresidents in long-term care facilities (LTLL CFs): recommendations ofthe task force on nutrition and ageing of the IAGG European re-gion and the IANA. J Nutr Health Aging 2009; 13: 475-83.

24. Simmons SF, Peterson EN, You C. The accuracy of monthlyweight assessments in nursing homes: implications for the iden-tification of weight loss. J Nutr Health Aging 2009; 13: 284-8.

25. Szulc P, Munoz F, Marchand F, Chapurlat R, Delmas PD. Rapid

loss of appendicular skeletal muscle mass is associated withhigher all-cause mortality in older men: the prospective MINOSstudy. Am J Clin Nutr 2010; 91: 1227-36.

26. Stenholm S, Harris TB, Rantanen T, Visser M, Kritchevskykk SB,Ferrucci L. Sarcopenic obesity: definition, cause and conse-quences. Curr Opin Clin Nutr Metab Care 2008; 11: 693-700.

27. Flegal KM, Carroll MD, Ogden CL, Curtin LR. Prevalence andtrends in obesity among US adults,1999-2008. JAMA 2010;303: 235-41.

28. Villareal DT, Banks M, Sinacore DR, Siener C, Klein S. Effect ofweight loss and exercise on frailty in obese older adults. ArchIntern Med 2006; 166: 860–6.

29. Villareal DT, Chode S, Parimi N et al. Weight loss, exercise, orboth and physical function in obese older adults. N Engl J Med2011; 364: 1218-29.

30. Shea MK, Nicklkk as BJ, Houston DK et al. The effect of intentionalweight loss on all-cause mortality in older adults: results of a ran-domized controlled weight-loss trial. Am J Clin Nutr 2011; 94:839-46.

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具