1

2

3

4

5

6 Q1

7

891011

1213

141516 Q2

1718

19

20

21

22

23

Mechanisms of Ageing and Development xxx (2013) xxx–xxx

G Model

MAD 10701 1–9

Effect of resistance-type exercise training with or without proteinsupplementation on cognitive functioning in frail and pre-frail elderly:Secondary analysis of a randomized, double-blind, placebo-controlledtrial

Ondine van de Rest a,*, Nikita L. van der Zwaluw a, Michael Tieland a,b, Jos J. Adam c,Gert Jan Hiddink d,e, Luc J.C. van Loon b,f, Lisette C.P.G.M. de Groot a,b

a Division of Human Nutrition, Wageningen University, P.O. Box 8129, 6700 EV Wageningen, The Netherlandsb Top Institute Food and Nutrition, P.O. Box 557, 6700 AN Wageningen, The Netherlandsc Department of Human Movement Sciences, School for Mental Health and Neuroscience (MHeNS), Maastricht University Medical Centre, P.O. Box 616, 6200

MD Maastricht, The Netherlandsd Communication Strategies; Communication, Technology and Philosophy – Centre for Integrative Development (CTP–CID), Social Sciences, Wageningen

University, P.O. Box 8130, 6700 EW Wageningen, The Netherlandse Manager Research Nutrition and Health, Dutch Dairy Association (NZO), P.O. Box 165, 2700 AD Zoetermeer, The Netherlandsf Department of Human Movement Sciences, NUTRIM School for Nutrition, Toxicology and Metabolism Maastricht University Medical Centre, P.O. Box 616,

6200 MD Maastricht, The Netherlands

A R T I C L E I N F O

Article history:

Received 8 May 2013

Received in revised form 5 December 2013

Accepted 16 December 2013

Available online xxx

Keywords:

Frailty

Aging

Cognitive functioning

Exercise training

Protein supplementation

A B S T R A C T

Physical activity has been proposed as one of the most effective strategies to prevent cognitive decline.

Protein supplementation may exert an additive effect. The effect of resistance-type exercise training

with or without protein supplementation on cognitive functioning in frail and pre-frail elderly people

was assessed in a secondary analysis. Two 24-week, double-blind, randomized, placebo-controlled

intervention studies were carried out in parallel. Subjects performed a resistance-type exercise program

of two sessions per week (n = 62) or no exercise program (n = 65). In both studies, subjects were

randomly allocated to either a protein (2 � 15 g daily) or a placebo drink. Cognitive functioning was

assessed with a neuropsychological test battery focusing on the cognitive domains episodic memory,

attention and working memory, information processing speed, and executive functioning. In frail and

pre-frail elderly, resistance-type exercise training in combination with protein supplementation

improved information processing speed (changes in domain score 0.08 � 0.51 versus �0.23 � 0.19 in the

non-exercise group, p = 0.04). Exercise training without protein supplementation was beneficial for attention

and working memory (changes in domain scores 0.35 � 0.70 versus �0.12 � 0.69 in the non-exercise group,

p = 0.02). There were no significant differences among the intervention groups on the other cognitive tests or

domain scores.

� 2013 Published by Elsevier Ireland Ltd.

Contents lists available at ScienceDirect

Mechanisms of Ageing and Development

jo ur n al ho mep ag e: www .e lsev ier . c om / lo cate /m ec hag ed ev

24252627282930

1. Introduction

Current estimates indicate that 35.6 million people worldwideare living with dementia, and because of the aging population thisnumber is predicted to double by 2030 and more than triple by2050 (World Health Organization and Alzheimer’s DiseaseInternational, 2012). Interventions targeted at risk factors for

313233343536

* Corresponding author. Division of Human Nutrition, Wageningen University,

P.O. Box 8129, 6700 EV Wageningen, The Netherlands. Tel.: +31 317 485867;

fax: +31 317 482782.

E-mail address: Ondine.vandeRest@wur.nl (O. van de Rest).

Please cite this article in press as: van de Rest, O., et al., Effectsupplementation on cognitive functioning in frail and pre-frail eldecontrolled trial. Mech. Ageing Dev. (2013), http://dx.doi.org/10.1016

0047-6374/$ – see front matter � 2013 Published by Elsevier Ireland Ltd.

http://dx.doi.org/10.1016/j.mad.2013.12.005

cognitive decline and Alzheimer’s Disease (AD) may offeropportunities for the development of an optimal preventivestrategy. Dementia is preceded by a decline in cognitive function-ing; however, this decline in cognitive functioning is not uniformacross all older individuals or across all cognitive domains. One ofthe factors that have been associated with cognitive decline isphysical activity. Several meta-analyses and systematic reviewshave summarized the evidence for the impact of physical activityon cognitive function in older individuals that are provided byeither observational studies (Arab and Sabbagh, 2010; Sofi et al.,2010), randomized controlled trials (RCTs) (Angevaren et al., 2008;Colcombe and Kramer, 2003; Heyn et al., 2004; Hindin andZelinski, 2012; O’Connor et al., 2010; Smith et al., 2010; van Uffelen

of resistance-type exercise training with or without proteinrly: Secondary analysis of a randomized, double-blind, placebo-/j.mad.2013.12.005

37 et38 Ge39 co40 ob41 in42

43 va44 al45 Be46 ex47 fle48 ex49 fo50 tr51 (252 w53 co54 (C55 re56 re57 ra58 co59 an60 2061 st62 m63 sig64 (L65 re66 us67 an68 Vo69

70 fu71 on72 co73 th74 Ho75 be76 a

77 su78 id79 2080 ac81 (T82

83 be84 su85 ex86 re87 m

88 2.

89 2.

90

91 2092 th93 m94 an95 un96 tio97 w98 w99 di

100101102103104105106107

108

109110111112113114115116117118119120121122123124125126127128129130131132

133

134135136137138139140141142143144145146147

148

149150151152153154155156157158159

O. van de Rest et al. / Mechanisms of Ageing and Development xxx (2013) xxx–xxx2

G Model

MAD 10701 1–9

al., 2008), or both (Bherer et al., 2013; Rolland et al., 2008).nerally, it has been suggested that physical activity improvesgnitive function and prevents cognitive decline, but results ofservational as well as experimental studies are limited andconsistent to draw firm conclusions.

The heterogeneous results may be explained not only byrying study populations and differences in study duration, butso by the different intensities and types of exercise employed.neficial effects, though not all significant, of various kinds ofercise programs were observed: aerobic, strength, balance,xibility, or a combination of these. It is unclear which type ofercise program is most effective, for what aspect of cognition andr which specific population. The importance of resistanceaining is stressed in the meta-analysis by Colcombe et al.003) who showed that aerobic-based training exercises togetherith resistance training exercises had a greater beneficial effect ongnition than programs of aerobic-based exercise training aloneolcombe et al., 2003), and is also postulated by a recent literatureview of Bherer et al. (2013). In addition, the few RCTs that usedsistance training alone were largely positive. Study durationsnged from 8 to 52 weeks and effects were mainly observed on thegnitive domains memory and executive functioning (Colcombed Kramer, 2003; Liu-Ambrose et al., 2010; Nagamatsu et al.,13; O’Connor et al., 2010; van Uffelen et al., 2008). A very recent

udy that was performed in frail older adults showed that 3onths of a combination of aerobic and strength exercisenificantly improved several domains of cognitive performance

anglois et al., 2013). Studies on different types of exercise inlation to brain function, structure, and connectivity assesseding magnet resonance imaging (MRI) have also started to emerged the results are promising as has been summarized byelcker-Rehage and Niemann (2013).In addition to exercise, dietary protein might improve cognitive

nctioning. van de Rest et al. (2013) performed a literature review the role of protein intake in relation to cognitive functioning andncluded that this has barely been studied. Five observational andree case-control studies were found, showing mixed results.wever, all three RCTs that were performed showed someneficial effects. A population of physically frail elderly would besuitable target population for a combined exercise and proteinpplementation intervention, since physical frailty has beenentified as a risk factor for cognitive decline and AD (Boyle et al.,10), frail elderly generally have a limited habitual physicaltivity level and moreover, they have a suboptimal protein intakeieland et al., 2012a).

We hypothesized that resistance-type exercise training wouldneficially affect cognitive functioning, in particular in subjectspplemented with dietary protein, which would augment theercise effect. Therefore, we examined the effect of 24-weeksistance-type exercise training with or without protein supple-entation on cognitive function in frail and pre-frail elderly.

Subjects and methods

1. Subjects

Individuals �65 years of age were recruited between December09 and October 2010, by using an existing database of volunteers,rough distribution of information flyers, and by local informationeetings. Potentially eligible elderly were screened for pre-frailtyd frailty using the five criteria from Fried et al. (2001): (1)intentional weight loss, (2) weakness, (3) self-reported exhaus-n,(4) slow walking speed, and(5) low physical activity. Pre-frailty

as classified when one or two criteria were present and frailtyhen three or more criteria were present. Individuals who wereagnosed with cancer, chronic obstructive pulmonary disease,

Please cite this article in press as: van de Rest, O., et al., Effecsupplementation on cognitive functioning in frail and pre-frail eldcontrolled trial. Mech. Ageing Dev. (2013), http://dx.doi.org/10.101

muscle disease, type 2 diabetes (plasma glucose concentration�7.0 mmol/L) (Alberti and Zimmet, 1998), or renal insufficiency(estimated Glomerular Filtration Rate <60 mL/min/1.73 m2 (Man-dayam and Mitch, 2006) were excluded. For participants in theresistance-type exercise training program, a resting electrocardio-gram was performed to exclude silent ischemia. The WageningenUniversity Medical Ethical Committee approved the study and allsubjects gave their written informed consent.

2.2. Study design

For the current study, we used data of two randomized,placebo-controlled trials that were carried out in parallel. Bothtrials covered a 24-week intervention period and investigated theeffect of protein versus placebo supplementation; in the absence orpresence of additional resistance-type exercise training (Tielandet al., 2012b, c). Primary outcome in both studies was muscle mass,but one of the secondary outcome measures was cognitivefunctioning.

Sample size was calculated based on an expected difference inlean body mass of 1.1 kg between groups (Borsheim et al., 2008;Esmarck et al., 2001) with an SD of 1.4 kg, a minimum of 24subjects per treatment group would be required to detect adifference (power = 80%, a = 0.05). With an expected dropout rateof 25% (Bonnefoy et al., 2003; Chin et al., 2002), a sample size of 30subjects per treatment group was considered adequate.

After inclusion, 62 of the 127 subjects participated in a 24-weekresistance-type exercise training program and 65 did not performany exercise program. All subjects were randomly allocated toeither protein or placebo supplementation for 24 weeks. Allocationwas carried out independently for both trials by an independentperson using a computer-generated random numbers in stratifiedpermuted blocks of size 4 and was stratified by gender. All outcomemeasures were collected at baseline and after 12 (except mostcognitive function tests) and 24 weeks of intervention.

2.3. Resistance-type exercise program

The supervised resistance-type exercise training was per-formed twice per week under personal supervision for a 24-weekperiod. The sessions were performed in the morning and afternoonwith at least 72 h between sessions. The training consisted of a 5-min warm-up on a cycle ergometer, followed by four sets on theleg-press and leg-extension machines and three sets on chestpress, lat pulldown, pec-dec, and vertical row machines (Techno-gym, Rotterdam, the Netherlands). The workload started at 50% of1-Repitition Maximum (1-RM) (10–15 repetitions per set) to 75%of 1-RM (8–10 repetitions) to stimulate muscle hypertrophy.Resting periods of 1 min were allowed between sets and 2 minbetween exercises. To evaluate changes in muscle strength, 1-RMwas repeated after 4, 8, 12, 16, and 20 weeks of training. Workloadintensity was adjusted based on the 1-RM outcomes.

2.4. Protein supplementation

Twice daily, the subjects received either a 250-mL protein-supplemented beverage containing 15 g protein (MPC80; milkprotein concentrate), 7.1 g lactose, 0.5 g fat, and 0.4 g calcium, or amatching placebo supplement containing no protein, 7.1 glactose, and 0.4 g calcium (FrieslandCampina Consumer ProductsEurope, Wageningen, the Netherlands). All beverages were vanillaflavored to mask the contents of the drinks , and the packageswere non-transparent. The subjects consumed one beverage afterbreakfast and one beverage after lunch. Adherence to the proteinand placebo drinks was judged by returned ticked calendars andnon-consumed beverages. Staff members and subjects were

t of resistance-type exercise training with or without proteinerly: Secondary analysis of a randomized, double-blind, placebo-6/j.mad.2013.12.005

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212213214215216217218219220221222

223

224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257

258

2592602612622632642653

O. van de Rest et al. / Mechanisms of Ageing and Development xxx (2013) xxx–xxx 3

G Model

MAD 10701 1–9

blinded toward treatment allocation until completion of dataanalysis.

2.5. Cognitive performance

At baseline, the Mini-Mental State Examination (MMSE) wasperformed to determine general cognitive status (Folstein et al.,1975). At baseline, after 12 weeks (only reaction time) and after 24weeks of intervention a standard battery of neuropsychological testswas performed by well-trained research assistants to assess thecognitive domains episodic memory, working memory and atten-tion, information processing speed, and executive functioning. TheWord Learning Test measures immediate and delayed memory andretrieval of newly acquired verbal material (Brand and Jolles, 1985).The forward test of the Wechsler Digit Span Task measures attentionand the backward test measures working memory (Wechsler, 1981).The Trail Making Test version A measures sensorimotoric speed andversion B measures concept shifting interference (executivefunction) (Reitan, 1958). The Stroop Color-Word Test measuresselective attention and susceptibility to behavioral interference(Stroop, 1935). The Verbal FluencyTest measures semantic memoryand language (Lezak et al., 2004). The finger precuing task is a four-choice reaction time task (Miller, 1982) requiring participants torespond to spatial-location stimuli on a computer screen withdiscrete button-press responses by index and middle fingers of bothhands. The screen always showed a reference row consisting of fourplus signs indicating the four possible stimulus locations. A trial wasstarted by showing a cue signal, which could be four plus signs in allfour positions (neutral uncued condition) or only two plus signs,thatis, the hand-cued condition in which the cue specifies two fingers onthe same hand; the finger-cued condition in which the cue specifiesthe same finger on two hands; or the neither-cued condition inwhich the cue specifies different fingers on two hands. Those cuesreduce the number of possible stimulus locations from four to two.After a preparation interval of 2000 ms, the target signal (one plussign) appeared in one of the cued positions. Participants had to reactas quickly as possible to the target signal by pressing thecorresponding response key. Outliers were defined as reaction timeless than 150 ms or in excess of 2000 ms and were excluded fromdata analysis (mean percentage of outliers was 1.5% based on 160trials per test session). Reaction time was based only on correcttrials.

2.6. Blood sampling

Fasted blood samples were collected at baseline, after 12 weeks,and directly after the intervention to determine glucose homeo-stasis and renal function (only baseline data shown). Plasmaglucose concentrations were analyzed with a COBAS FARA analyzer(Uni Kit III; Roche, Basel, Switzerland). Insulin was analyzed byradioimmunoassay (Insulin RIA Kit; LINCO Research Inc, St Charles,MO). Serum creatinine was measured by using Roche ModularSystem P (Roche Diagnostics GmbH, Mannheim, Germany).

2.7. Dietary intake

Participants recorded their food intake for 3 days at baseline,after 12 weeks, and after 24 weeks to assess potential changes in

Episodic memory ¼ ðZWLT total 1�5 immediate recall þ ZWLT

Attention and working memory ¼ ðZDigit span forward þ ZD

Information processing speed ¼ ð�ZTrail Making Test Part

Executive functioning ¼ ðZVerbal Fluency Animal þ ZVerbal Flu

Please cite this article in press as: van de Rest, O., et al., Effectsupplementation on cognitive functioning in frail and pre-frail eldecontrolled trial. Mech. Ageing Dev. (2013), http://dx.doi.org/10.1016

daily food intake during the intervention period (only baselinedata shown). Trained dieticians gave oral and written instruc-tions about recording the type of foods and estimating theportion sizes in household measures, and they checked the foodrecords for completeness, obtained additional information aboutunclear items or amounts , and used household measures toimprove the estimation of portion size. Dietary intake data werecoded (type of food, time of intake, and amount) and energy andmacronutrient intakes were calculated using a nutrient compu-tation program (Komeet) and the Dutch food compositiondatabase (NEVO, 2006).

2.8. Other measurements

Physical performance was assessed by the short physicalperformance battery (SPPB), which consists of three components:balance, gait speed, and chair-rise ability (Guralnik et al., 1994).Scores of 0-4 were based on the categories of performance in thebalance tests, on the time necessary to complete the walk, and onthe time needed to perform the chair-rise test. A summaryperformance score of 0–12 was calculated by summing the scoresof the tests. Habitual physical activity data were quantified using atri-axial accelerometer (ActiGraph GTX3, 2009, Pensacola, FL, USA)worn on the hip for 1 week. Change of acceleration per second andepochs of 60 s were used. After 7 days, data were uploaded andanalyzed using the MAH/UFFE analyzer, version 1.9.0.3 (MRCEpidemiology Unit, Cambridge, UK). Data files that did not meet10 h of monitoring per day on at least 5 days as well as files thatincluded periods of >100 min without activity were excluded fromthe analysis. Mental health was assessed using the Center forEpidemiologic Studies Depression Scale (CES-D), a 20-item, self-report scale developed to measure depressive symptoms experi-enced in the past week (Radloff, 1977). Overall health status andquality of life of the subjects were assessed using the 12-Item ShortForm Health Survey (SF-12) (Ware et al., 1996) and the EuroQol-5D(EQ-5D) (Brooks, 1996). The SF-12 is a short form of the widelyused SF-36 that generates a physical (PCS12) and a mental (MCS12)composite score. The EQ-D5 defines health in five dimensions and avisual analogue scale indicates health status from 0 to 100. Heightwas measured at baseline with a wall-mounted stadiometer to thenearest 0.1 cm. Body weight was measured in the fasted state tothe nearest 0.1 kg with a calibrated digital scale (ED-6-T; Berkel,Rotterdam, The Netherlands). Blood pressure was measured in themorning after a 10-min rest with an Omron HEM-907 (Lake Forest,IL, USA) device. At each visit four measurements were done, withan interval of 2 min, of which the first measurement was discarded.The mean value of the three subsequent measurements for systolicand diastolic blood pressure was taken.

2.9. Statistics

Differences in baseline characteristics among the four inter-vention groups were compared using analysis of variance (ANOVA)for continuous variables or Chi square for categorical variables. Z-scores were created for cognitive tests at baseline and 24 weeksusing the baseline data as norm data to calculate the grand meanand SD per test. Individual neuropsychological tests were clusteredinto compound scores for four neuropsychological domains: Q

decay þ ZWLT recognitionÞ=3

igit span backwardÞ=2

A þ ZStroop mean I and II � ZReaction time uncued � ZReaction time cuedÞ=4

ency Letter P þ ZStroop interference � ZTrail making interference=4:

of resistance-type exercise training with or without proteinrly: Secondary analysis of a randomized, double-blind, placebo-/j.mad.2013.12.005

266

267 ac268 rig269 ne270 20271 ca272 St273 20274 (r275 an276 cu277 va278 of279 pr280 re281 su282 tim283 w284 ou285 tio286 co287 in288 po289 fa290 an291 de292 pe

293 3.

294 3.

295

296 ha297 fra298 cr299 in300 pa301 ex302 pl303 fro304 dr305 co306 ad307 ca308 ex309

310 w311 am312 di313 in314 <2315 th316 co

317 3.

318

319 in320 sc321 ch322 in323 gr324 w325 gr326 to

327328329330331332333334335336337338339340341342343344345346

347

348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390

O. van de Rest et al. / Mechanisms of Ageing and Development xxx (2013) xxx–xxx4

G Model

MAD 10701 1–9

To take into account errors made during the Stroop test,curacy (maximum right answers–amount of errors/maximumht answers) and speed-accuracy trade-off (accuracy/timeeded to complete the task) were calculated (Hausdorff et al.,05). To control for the effect of motor speed on performance, welculated interference measures for TMT (TMT B/TMT A) androop part 3 (Stroop part 3-(part 1 + part 2)/2) (Valentijn et al.,05). For the finger precuing task trials, with errors or outliers

eaction time <150 ms or >2000 ms) were excluded from dataalysis. The three cued conditions of the reaction time test (hand-ed, finger-cued, and neither-cued) were averaged to obtain oneriable cued reaction time. Analyses were performed for the effect

resistance-type exercise versus non-exercise within bothotein-supplemented groups and separately for the effect ofsistance-type exercise versus non-exercise within both placebo-pplemented groups. Differences between interventions over-

e were analyzed using analyses of covariance (ANCOVA), inhich the baseline data were included as the covariate, thetcome measures as the dependent factor and exercise interven-n as the fixed factor. Linear mixed models with unstructuredvariance structure were used to assess the effect of thetervention on reaction time, which was measured at three timeints. Time, treatment, and their interaction were defined as fixed

ctors, whereas subject was defined as random factor. Data werealyzed by intention-to-treat and an a-level of 0.05 was used totermine statistical significance. All statistical analyses wererformed using SPSS Statistics v19.

Results

1. Subjects

Details of the participant flow of recruitment and screeningve been published before (Tieland et al., 2012b, c). In total, 127il and pre-frail elderly men and women fulfilled to the inclusion

iteria and were willing to participate: 62 in the study with and 65 the study without resistance-type exercise training. Nineteenrticipants withdrew during the intervention: 11 from theercise intervention (five from the protein and six from theacebo group) and eight from the non-exercise intervention (fourm the protein and four from the placebo group). Of these

opouts, eight were willing to have final measurements, so theyuld be included in the intention-to-treat analysis. The averageherence to the protein and placebo drinks based on tickedlendars and non-consumed returned beverages was 98% in theercise study and 93% in the non-exercise study.Mean age of the participants at baseline was 79 � 8 years, 39%

ere male and these distributions were not significantly differentong the intervention groups. There were also no significant

fferences in any of the other baseline characteristics among thetervention groups (Table 1). In total, 15 subjects had an MMSE score5, suggesting a cognitive impairment (Mungas, 1991). However,

e overall median MMSE score of 28 showed that most subjects weregnitively healthy.

2. Cognitive functioning

At baseline mean cognitive function scores for each of thedividual tests and for the four cognitive domains, compositeores did not differ among the groups (Table 2). Table 3 shows theanges in cognitive tests and domains after 24 weeks oftervention. When comparing the exercise and non-exerciseoup who received additional protein supplementation for 24eeks, the non-exercise group improved compared to the exerciseoup on Verbal Fluency Animals (2.4 � 4.1 more animals compared

�0.6 � 4.1 animals, respectively, p < 0.01). On the domain of

Please cite this article in press as: van de Rest, O., et al., Effecsupplementation on cognitive functioning in frail and pre-frail eldcontrolled trial. Mech. Ageing Dev. (2013), http://dx.doi.org/10.101

information processing speed, the exercise group (change in domainscore 0.08 � 0.51) improved significantly compared to the non-exercise group (change in domain score �0.23 � 0.19) (p = 0.04)

After 24 weeks of placebo supplementation, the exercise groupsignificantly improved on the Digit Span forward test (0.7 � 1.7versus �0.3 � 1.7 for the exercise and non-exercise group, respec-tively, p = 0.04) (Table 3). Accordingly, this significant improvementwas also observed in the corresponding cognitive domain of attentionand working memory (change in domain score 0.35 � 0.70 in theexercise group versus �0.12 � 0.69 in the non-exercise group,p = 0.02). On the other cognitive tests or domain scores, no significantdifferences were found among the intervention groups.

Both uncued and mean cued reaction time (and also all threeindividual cued conditions) significantly improved overtime in allintervention groups (p < 0.05) (Table 4). However, neither exercisewith protein supplementation nor exercise without proteinsupplementation showed a significant benefit compared to non-exercise Q4. Overall error rate was very low (1.7%), with small, non-significant reductions in error rate for all groups at the end of thetreatment compared to baseline.

4. Discussion

In this study performed in frail and pre-frail elderly, 24 weeks ofresistance-type exercise training in combination with proteinsupplementation improved the cognitive domain informationprocessing speed and the same training without protein supple-mentation was beneficial for attention and working memory.

To the best of our knowledge, this is the first study thatinvestigated the effect of resistance-type exercise in combinationwith protein supplementation on cognitive functioning. Ascognitive function was not the primary outcome of the study,the total number of subjects included was not based on cognitionfunction as outcome measure, but on the primary outcome musclemass. However, compared to the other intervention studies thathave been performed on resistance-type exercise and cognitivefunction, this was still one of the largest studies (Busse et al., 2008;Cassilhas et al., 2007; Lachman et al., 2006; Langlois et al., 2013;Liu-Ambrose et al., 2010; Moul et al., 1995; Nagamatsu et al., 2013;Panton et al., 1990; Perrig-Chiello et al., 1998; Tsutsumi et al.,1998) and the only one that combined exercise with proteinsupplementation. Furthermore, cognitive function was measuredin a standardized way with an extensive neurological test battery,which enabled us to thoroughly investigate the effects of theintervention on multiple cognitive domains that are sensitive toage-related cognitive decline.

As frailty has been associated with cognitive impairment (Boyleet al., 2010), we expected this study population to be relativelycognitively impaired and therefore subjective to change. However,except 15 individuals who scored <25 points on the MMSE, thegroup in general was still cognitively healthy. Still, we foundbeneficial changes in cognitive functioning on the cognitivedomain information processing speed after the exercise plusprotein intervention and on attention and working memory afterthe exercise intervention without additional protein supplemen-tation. In more cognitively impaired elderly, exercise might benefitother individual neuropsychological tests or cognitive domains aswell.

The current study of 24 weeks might also have been too short topick up an effect of the interventions on the other cognitivedomains, because the process of cognitive decline is a rather slowprocess. According to the IANA taskforce, trials studying effect oncognitive functioning should have a study duration of at least 18months (Vellas et al., 2007). Though, compared to the otherresistance-type exercise intervention studies that did find effectson cognitive function, our intervention was of average length. Only

t of resistance-type exercise training with or without proteinerly: Secondary analysis of a randomized, double-blind, placebo-6/j.mad.2013.12.005

Table 2Baseline cognitive test scores of Dutch frail and pre-frail elderly, by intervention group.

Protein (n = 65) Placebo (n = 62)

Cognitive tests, maximum score Exercise (n = 31) Non-exercise (n = 34) Exercise (n = 31) Non-exercise (n = 31)

Episodic memory

Word learning test – Immediate recall 1–5, 75 words 31.8 � 10.9a 34.6 � 11.8 33.6 � 10.0 33.5 � 10.5

Word learning test – Delayed recall, 15 words 6.1 � 2.7 6.2 � 3.1 6.5 � 3.1 5.9 � 2.9

Word learning test – Decay (delayed recall–WLT5) �2.0 � 1.8 �2.8 � 1.5 �2.3 � 1.8 �2.6 � 2.0

Word learning test – Recognition, 30 words 27.5 � 2.5 28.2 � 1.9 28.1 � 1.8 27.1 � 2.7

Attention and working memory

Digit Span forward, 16 points 8.0 � 1.8 7.9 � 1.4 7.9 � 1.7 8.3 �2.1

Digit Span backward, 14 points 5.2 � 1.9 5.2 � 1.3 5.5 � 1.8 6.0 � 2.3

Information processing speed

Stroop 1b 1.85 � 0.33c 1.91 � 0.36 1.91 � 0.37d 1.80 � 0.33

Stroop 2b 1.47 � 0.28c 1.5 � 0.4 1.51 � 0.32d 1.44 � 0.28

Trail making part A (s)e 63.7 � 29.0 61.6 � 27.2 54.4 � 24.3d 62.6 � 23.0

Reaction time uncued (ms)e 833 � 290c 764 � 301 787 � 221d 830 � 243

Reaction time, mean cued conditions (ms)e 832 � 309 753 � 330 779 � 240 821 � 263

Executive functioning

Stroop interference (Part 3–(Part 1 + Part 2/2))b �0.89 � 0.29c �0.90 � 0.24 �0.89 � 0.25d �0.79 � 0.21

Trail making test (Part B/Part A) 1.93 � 0.63 1.77 � 0.61 1.68 � 0.49d 1.83 � 0.62

Word Fluency – Animals 20.6 � 6.3 21.4 � 6.4 20.0 � 6.8 20.6 � 5.8

Word Fluency – Letter P 13.0 � 6.2 15.2 � 5.8 14.7 � 5.5 14.1 � 3.8

Domain specific Z-scores

Episodic memory �0.01 � 0.63 0.04 � 0.60 0.09 � 0.62 �0.13 � 0.77

Attention and working memory �0.07 � 0.90 �0.11 � 0.64 �0.04 � 0.87 0.22 � 1.13

Information processing speed �0.06 � 0.74d 0.10 � 1.00 0.11 � 0.81d �0.09 � 0.75

Executive functioning �0.14 � 0.64d 0.06 � 0.60 0.05 � 0.58d 0.08 � 0.55

a Mean � SD (all such values).b Corrected for errors (time needed/accuracy).c n = 30.d n = 29.e Higher scores indicate more time was needed to complete the task, so poorer performance.

Table 1Baseline characteristics of 127 Dutch frail and pre-frail elderly by intervention group.

Protein (n = 65) Placebo (n = 62)

Characteristic Exercise (n = 31) Non-exercise (n = 34) Exercise (n = 31) Non-exercise (n = 31)

Age (years) 77.7 � 8.8a 77.9 � 8.1 79.2 � 6.3 81.2 � 7.4

Male, n (%) 11 (36) 14 (41) 10 (32) 15 (48)

Education Low/middle/high (%) 10/55/36 9/59/32 3/71/26 0/55/45

BMI (kg/m2) 28.7 � 4.5 27.0 � 4.6 28.2 � 4.6 26.2 � 3.2

Smokers, n (%) 2 (7) 5 (15) 1 (3) 1 (3)

Number of cigarettes/cigars per day 8 12 5 10

Alcohol consumers (%) 74 68 77 71

Alcohol consumption (g/day)b 11 (6–22)c 15 (5–30) 9 (3–17) 11 (3–22)

MMSE (Mini-Mental State Examination) 28 (27–29) 29 (26–30) 28 (27–29) 28 (26–30)

Range 23–30 21–30 23–30 22–30

CES-D 8.2 � 6.9 6.6 � 5.7 6.1 � 6.0 7.0 � 4.1

PCS12 (0–100 points) 42.6 � 6.4 43.9 � 10.0 42.7 � 9.8 42.7 � 9.8

MCS12 (0–100 points) 56.0 � 8.2 54.0 � 7.9 56.6 � 7.2 54.9 � 8.0

EQ-5D (0–1 points) 0.7 � 0.2 0.7 � 0.2 0.8 � 0.1 0.7 � 0.2

Frailty score (1–5)d 1.5 � 0.7 1.7 (1–2) 1.6 � 0.8 2.0 (1–3)

Frail/pre-frail, n 4/27 7/27 7/24 11/20

Physical activity score (counts/minute)e 159 � 111 153 � 101 122 � 69 115 � 88

Baseline glucose (mmol/L) 5.4 � 0.5 5.2 � 0.4 5.2 � 0.5 5.3 � 0.4

Baseline insulin (mU/L) 19.6 � 6.9 18.0 � 7.4 18.1 � 6.7 18.2 � 7.1

eGFR (mL/min/1.73 m2) 80.6 � 14.1 84.4 � 20.2 79.3 � 13.9 77.6 � 12.5

Blood pressure, systolic (mmHg) 142 � 19.2 152 � 25 143 � 20.3 150 � 23

Blood pressure, diastolic (mmHg) 74 � 8.3 77 � 10 73 � 9.8 75 � 9

SPPB (0–12) 8.0 � 2.4 8.9 � 2.8 7.9 � 2.4 7.8 � 3.6

Energy intake, kcal 1980 (1687–2226) 1981 (1513–2295) 1698 (1543–2128) 1932 (1593–2261)

Protein intake, g/day (En%) 72 (61–85) (16) 72 (57–92) (16) 75 (62–85) (17) 72 (64–78) (16)

Fat intake, g/day (En%) 76 (49–95) (33) 83 (55–107) (36) 71 (58–92) (35) 73 (57–94) (33)

Carbohydrate intake, g/day (En%) 234 (190–270) (47) 198 (166–256) (44) 184 (162–254) (45) 216 (173–273) (48)

Abbreviations: BMI, body mass index; MMSE, Mini-Mental State Examination; CES-D, Centre for Epidemiologic Studies Depression Scale; pCS12, physical composite score;

MCS12, mental composite score; EQ-5D, EuroQol-5D; eEGR, estimated Glomerular Filtration Rate; SPPB, short physical performance battery.a Mean � SD (all such values).b Mean use in consumers only.c Median (Q1–Q3) (all such values).d Higher scores indicate being more frail according to Fried et al. (2001).e Significant difference between the protein and the placebo group (p < 0.05).

O. van de Rest et al. / Mechanisms of Ageing and Development xxx (2013) xxx–xxx 5

G Model

MAD 10701 1–9

Please cite this article in press as: van de Rest, O., et al., Effect of resistance-type exercise training with or without proteinsupplementation on cognitive functioning in frail and pre-frail elderly: Secondary analysis of a randomized, double-blind, placebo-controlled trial. Mech. Ageing Dev. (2013), http://dx.doi.org/10.1016/j.mad.2013.12.005

391 tw392 in393 ou394 be395 se396 an397 in398 an399 nu400 w

401402403404405406407408409410

Table 3Changes on cognitive test scores of Dutch frail and pre-frail elderly, by intervention group (only participants with baseline and end data).

Cognitive tests/domains Exercise (n = 28) Non-exercise (n = 30)

Protein Baseline 24 weeks Change Baseline 24 weeks Change p

Episodic memory

Word learning test – Immediate recall, 75 words 31.9 � 10.9a 35.1 � 10.9 3.3 � 7.4 35.9 � 11.1 38.8 � 12.0 2.9 � 7.6 0.83

Word learning test – Delayed recall, 15 words 6.2 � 2.6 6.8 � 3.2 0.6 � 1.9 6.6 � 2.9 6.8 � 3.6 0.2 � 2.3 0.53

Word learning test – Decay �2.0 � 1.8 �2.1 � 1.5 �0.2 � 2.6 �2.7 � 1.6 �2.6 � 2.1 0.1 � 2.4 0.37

Word learning test – Recognition, 30 words 27.4 � 2.5 27.4 � 2.8 0.0 � 2.3 28.4 � 1.9 27.7 � 2.2 �0.7 � 1.9 0.54

Attention and working memory

Digit Span forward, 16 points 7.7 � 1.6 7.9 � 1.9 0.1 � 1.4 8.0 � 1.4 8.1 � 1.7 0.1 � 1.4 0.92

Digit Span backward, 14 points 5.2 � 1.9 5.8 � 1.9 0.6 � 1.5 5.3 � 1.3 5.4 � 1.5 0.0 � 1.2 0.16

Information processing speedb

Trail making Part A (s)c 59.4 � 21.1 58.2 � 25.8 �1.1 � 24.2 57.2 � 21.8 60.8 � 36.9 3.7 � 20.1 0.85

Stroop 1 (s)d 1.9 � 0.3 1.9 � 0.3 0.0 � 0.2 2.0 � 0.3 2.0 � 0.3 �0.0 � 0.2 0.95

Stroop 2 (s)d 1.5 � 0.3 1.5 � 0.3 0.0 � 0.2 1.6 � 0.3 1.6 � 0.3 0.1 � 0.1 0.60

Reaction time, uncued (ms)c 834 � 295 744 � 181 �90 � 162 686 � 136 644 � 154 �42 � 122 0.20

Mean reaction time cued (ms)c 830 � 312 726 � 226 �104 � 149 667 � 147 604 � 172 �63 � 107 0.19

Executive functioningb

Stroop interference d �0.9 � 0.3 �0.9 � 0.2 0.0 � 0.2 �0.9 � 0.2 �0.8 � 0.2 0.1 � 0.2 0.42

Word Fluency – Animals 21.2 � 6.1 20.6 � 5.8 �0.6 � 4.1 21.8 � 6.2 24.2 � 6.5 2.4 � 4.1 <0.01

Word Fluency – Letter P 13.0 � 6.5 13.6 � 6.1 0.6 � 4.3 15.7 � 5.6 15.7 � 6.7 0.0 � 4.9 0.99

trail making test (Part B/Part A)c 1.86 � 0.54 1.86 � 0.57 0.01 � 0.76 1.75 � 0.63 1.77 � 0.57 0.01 � 0.60 0.75

Domain specific Z-scores

Episodic memory �0.02 � 0.64 0.05 � 0.77 0.07 � 0.62 0.11 � 0.57 0.13 � 0.83 0.01 � 0.57 0.79

Attention and working memory �0.16 � 0.85 0.03 � 0.97 0.19 � 0.63 �0.03 � 0.63 0.01 � 0.79 0.04 � 0.57 0.38

Information processing speedb �0.05 � 0.71 0.03 � 0.41 0.08 � 0.51 0.34 � 0.60 0.10 � 0.54 �0.23 � 0.19 0.04

Executive functioningb �0.09 � 0.64 �0.05 � 0.52 0.04 � 0.44 0.07 � 0.60 0.25 � 0.64 0.17 � 0.43 0.09

Exercise (n = 27) Non-exercise (n = 28)

Placebo Baseline 24 weeks Change Baseline 24 weeks Change p

Episodic memory

Word learning test – Immediate recall, 75 words 33.3 � 10.5 37.0 � 10.3 3.7 � 6.2 34.4 � 9.8 36.5 � 10.4 2.1 � 5.5 0.36

Word learning test – Delayed recall, 15 words 6.5 � 3.3 6.9 � 3.4 0.4 � 2.2 6.1 � 2.8 6.8 � 3.5 0.7 � 2.4 0.72

Word learning test – Decay �2.1 � 1.7 �2.4 � 1.8 �0.3 � 2.3 �2.7 � 1.8 �2.2 � 2.0 0.5 � 2.6 0.58

Word learning test – Recognition, 30 words 28.2 � 1.8 27.9 � 2.1 �0.3 � 1.6 27.4 � 2.4 27.7 � 2.6 0.3 � 2.8 0.78

Attention and working memory

Digit Span forward, 16 points 8.0 � 1.7 8.8 � 2.7 0.7 � 1.7 8.3 � 2.2 8.0 � 2.1 �0.3 � 1.7 0.04

Digit Span backward, 14 points 5.6 � 1.8 6.2 � 2.1 0.5 � 1.6 6.2 � 2.3 6.0 � 2.4 �0.1 � 1.7 0.22

Information processing speede,f

Trail making Part A (s)c 55.9 � 25.9 51.2 � 17.6 �4.7 � 15.5 63.9 � 24.3 60.1 � 25.1 �3.8 � 14.8 0.50

Stroop 1 (s)d 1.9 � 0.4 2.0 � 0.4 0.0 � 0.2 1.8 � 0.3 1.8 � 0.4 0.0 � 0.2 0.85

Stroop 2 (s)d 1.5 � 0.3 1.6 � 0.4 0.1 � 0.2 1.4 � 0.3 1.5 � 0.3 0.0 � 0.2 0.47

Reaction time, uncued (ms)c 792 � 239 722 � 189 �69 � 138 832 � 236 812 � 242 �21 � 144 0.10

Mean reaction time cued (ms)c 794 � 261 696 � 206 �98 � 148 813 � 260 772 � 271 �41 � 135 0.10

Executive functioninge,g

Stroop interference d �0.9 � 0.3 �0.9 � 0.3 0.0 � 0.2 �0.8 � 0.2 �0.8 � 0.2 �0.0 � 0.2 0.84

Word Fluency – Animals 20.6 � 7.3 20.2 � 7.7 �0.4 � 4.4 20.9 � 5.6 22.0 � 6.3 1.1 � 3.7 0.21

Word Fluency – Letter P 15.7 � 5.5 15.9 � 6.1 0.2 � 4.8 14.6 � 3.8 14.9 � 5.7 0.3 � 4.1 0.99

Trail making test (Part B/Part A)c 1.62 � 0.47 1.59 � 0.43 �0.04 � 0.59 1.83 � 0.65 1.79 � 0.69 �0.04 � 0.81 0.37

Domain specific Z-scores

Episodic memory 0.11 � 0.65 0.13 � 0.71 0.02 � 0.57 �0.08 � 0.68 0.12 � 0.87 0.20 � 0.79 0.51

Attention and working memory 0.05 � 0.85 0.40 � 1.27 0.35 � 0.70 0.28 � 1.16 0.16 � 1.13 �0.12 � 0.69 0.02

Information processing speede,f 0.11 � 0.87 0.18 � 0.47 0.07 � 0.48 �0.14 � 0.78 �0.01 � 0.46 0.13 � 0.50 0.32

Executive functioninge,g 0.09 � 0.62 0.12 � 0.55 0.02 � 0.42 0.10 � 0.58 0.17 � 0.61 0.07 � 0.42 0.78

a Values are mean � SD (all such values). Differences between the two groups were measured using ANCOVA (baseline cognitive function as covariate).b n = 26 for the exercise–protein group.c Higher scores indicate more time was needed to complete the task, so poorer performance.d Corrected for errors (time needed/accuracy).e n = 24.f n = 27.g n = 26.

O. van de Rest et al. / Mechanisms of Ageing and Development xxx (2013) xxx–xxx6

G Model

MAD 10701 1–9

o studies were longer with, respectively, 9 and 12 months oftervention (Busse et al., 2008; Liu-Ambrose et al., 2010) and sincer study population of frail and pre-frail elderly was expected to

more sensitive, effects could have occurred faster. Additionallyen the extensive exercise program that was used in our studyd the fact that this was performed by frail and pre-fraildividuals, 24 weeks was a substantial period for this target groupd a longer study duration could have resulted in a relatively largember of dropouts and age-related co-morbidities interfering

ith the effects of the intervention.

Please cite this article in press as: van de Rest, O., et al., Effecsupplementation on cognitive functioning in frail and pre-frail eldcontrolled trial. Mech. Ageing Dev. (2013), http://dx.doi.org/10.101

The majority of the other intervention studies that focused onthe effects of resistance training on cognitive function alsoobserved an effect, mostly on the cognitive domains of memoryand executive functioning (Colcombe and Kramer, 2003; Langloiset al., 2013; Liu-Ambrose et al., 2010; Nagamatsu et al., 2013;O’Connor et al., 2010; van Uffelen et al., 2008). We did observe aneffect of the resistance training on attention and workingmemory, but not on episodic memory. Additionally, we foundan effect of resistance-type exercise plus protein supplementationon information processing speed. It may be that for improvement

t of resistance-type exercise training with or without proteinerly: Secondary analysis of a randomized, double-blind, placebo-6/j.mad.2013.12.005

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447

448

449

450

451

452

453

454

455

456

457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497

498

499500

Table 4Reaction time of Dutch frail and pre-frail elderly measured at baseline, and after 12 and 24 weeks, by intervention group.

Exercise Non-exercise

Protein Baseline

(n = 30)

12 weeks

(n = 30)

24 weeks

(n = 27)

Baseline

(n = 34)

12 weeks

(n = 28)

24 weeks

(n = 32)

Treatment*time

interaction

Treatment

effect

Time

effect

Reaction time (RT) in millisecond

RT uncued 833 � 290a,b 832 � 268 769 � 220 764 � 301 663 � 151 671 � 183 0.69 0.17 <0.01

RT hand-cued 798 � 321 756 � 267 725 � 249 716 � 318 597 � 170 613 � 210 0.92 0.86 <0.01

RT finger-cued 847 � 304 844 � 303 765 � 262 772 � 333 640 � 179 643 � 208 0.22 0.13 <0.01

RT neither-cued 850 � 310 820 � 266 762 � 259 771 � 342 648 � 184 650 � 213 0.90 0.47 <0.01

RT mean cued 832 � 309 807 � 276 751 � 255 753 � 330 628 � 176 635 � 208 0.79 1.00 <0.01

Placebo Baseline

(n = 29)

12 weeks

(n = 27)

24 weeks

(n = 24)

Baseline

(n = 31)

12 weeks

(n = 29)

24 weeks

(n = 28)

RT uncued 787 � 221 741 � 182 722 � 189 830 � 243 778 � 196 808 � 238 0.43 0.27 0.01

RT hand-cued 748 � 234 689 � 201 670 � 196 767 � 263 726 � 208 740 � 260 0.37 0.41 0.02

RT finger-cued 790 � 239 752 � 211 706 � 217 840 � 261 773 � 215 799 � 291 0.25 0.32 <0.01

RT neither-cued 800 � 256 752 � 218 711 � 210 857 � 275 781 � 221 785 � 251 0.52 1.00 <0.01

RT mean cued 779 � 240 731 � 208 696 � 206 821 � 263 760 � 212 775 � 266 0.37 1.00 <0.01

a Mean � SD (all such values). Differences between the two groups were measured using linear mixed models.b Higher scores indicate more time was needed to complete the task, so poorer performance.

O. van de Rest et al. / Mechanisms of Ageing and Development xxx (2013) xxx–xxx 7

G Model

MAD 10701 1–9

in speed -related tasks, where the neuromuscular system isinvolved, additional protein is needed to reinforce the effect ofexercise. Also, in our cognitive domain of information processingspeed, the finger precuing task was a major component. This is acomplex reaction time task, which involves more centralprocessing and may be more likely to show improvements thansimple reaction time tasks. More trials of sufficient duration andsize, performed in susceptible target groups, and with extensiveand sensitive neuropsychological test batteries like ours, areneeded to disentangle these effects.

Although the suggestion that physical exercise training isbeneficial for cognitive functioning is becoming more and moreestablished, the underlying biological mechanisms are not yetelucidated. In the review of Bherer et al. (2013), the current ideashave been summarized and include direct effects at the supramo-lecular and molecular level. In animal studies, physical activity hasbeen shown to induce angiogenesis, neurogenesis, neural cellproliferation, or synaptogenesis. The evidence for molecularmechanisms also comes from animal studies and includes changesin molecular growth factors such as brain-derived neurotrophicfactor (BDNF) and insulin-like growth factor (IGF-1) (Bherer et al.,2013). The findings for an exercise-related increase in BDNF havealso been shown for the first time in humans and were associatedwith an increased hippocampal volume and improved memory(Erickson et al., 2011). Recently, also differences in underlyingmolecular mechanisms for different types of exercise werepostulated; aerobic training increased BDNF in the hippocampusand resistance training increased IGF-1 levels (Cassilhas et al., 2012).Something to take into consideration in the future studies is thatthere are also some factors that might moderate the effects ofphysical activity on brain health, such as apolipoprotein E (APOE),BDNF, catechol-O-methyltransferase (COMT), and the dietaryomega-3 fatty acids (Erickson et al., 2013; Leckie et al., 2012).Furthermore, studies on physical activity and brain function,structure, and connectivity assessed using MRI are still in theirearly stages, but with these studies could also provide more insightin underlying mechanisms (Voelcker-Rehage and Niemann, 2013).

In addition to the resistance-type exercise program, subjectswere also supplemented with either protein or placebo. One of thepossible links between protein and cognitive functioning may bethe fact that dietary protein contains some indispensable aminoacids, tryptophan, and tyrosine, in particular, which act asprecursors for several brain neurotransmitters (Fernstrom,2012). However, the number of studies on supplementation withprotein or some individual amino acids is very small and moreresearch is needed to elucidate the effects on cognitive functioning.

Please cite this article in press as: van de Rest, O., et al., Effectsupplementation on cognitive functioning in frail and pre-frail eldecontrolled trial. Mech. Ageing Dev. (2013), http://dx.doi.org/10.1016

Our main interest was in the effects of resistance-type exercisetraining, with or without protein supplementation, on cognitivefunctioning. This is the first time that the effects of a resistance-type exercise intervention could be examined in a frail elderlypopulation. However, to examine this interesting effect, analyseshad to be performed across the two trials, which may be amethodological limitation. Although not designed as such, the twotrials together also approach a 2 � 2 factorial design. To examinepossible interaction effects between exercise and protein, weexplored the results of adding an interaction term to our models,despite the fact that the trials were each powered for main effectsand not for interaction effects. There was no significant interactionbetween exercise and protein supplementation on any of thecognitive domains, but on the individual cognitive tests asignificant interaction was observed for the uncued and cuedreaction time tests (p = 0.05 for both tests). However, taking intoaccount the limited power, these p-values have to be interpretedwith caution. Moreover, this interaction was not observed in thecorresponding cognitive domain of information processing speed.Still, a possible interaction between exercise and proteinsupplementation has to be taken into account in the futurestudies using similar interventions in parallel.

In summary, in this population of frail and pre-frail elderly, 24weeks of resistance-type exercise training in combination withprotein supplementation was beneficial for the cognitive domaininformation processing speed and the same training withoutprotein supplementation was beneficial for attention and workingmemory.

The aim of the NU-AGE project is to obtain more knowledge onhow a healthful diet can impact on and counteract age-relateddisease and functional decline. In this article, we focus on one ofthe most important functional declines in elderly: cognitivedecline, on which both diet and physical activity play an importantrole. Within the current studies, we were able to investigate theeffect of exercise with or without protein supplementation, as anexample for a study on interaction or synergy between diet andphysical activity, an area which is relatively unexplored so far. TheNU-AGE project will open new perspectives to test the relationshipbetween physical status (measured by ad hoc test) and activity(measured by ActiGraph device) and health and cognitive statusafter a Mediterranean diet-based nutritional trial.

5. Author contributions

The authors’ contributions were as follows: Study design: OvdR,MT, GJH, LJCvL, and LCPGMdG. Data collection: NvdZ, MT. Data

of resistance-type exercise training with or without proteinrly: Secondary analysis of a randomized, double-blind, placebo-/j.mad.2013.12.005

501 an502 pr503 Du504 da505 re

506 6.

507

508 (T509 th510 Ag511 sin512 in513 co514

515 an

516 Ac

517

518 st519 w

520 Ap

521

522 th

523 ReQ5

524 Al525

526

527

528 An529

530

531 Q6532 Ar533

534 Bh535

536

537 Bo538

539

540

541 Bo542

543

544 Bo545

546

547 Br548

549 Br550 Bu551

552

553

554 Ca555

556

557 Ca558

559

560 Ch561

562

563 Co564

565 Co566

567

568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627628629630631632633634635636637638639640641642643644645646647648649650651652653

O. van de Rest et al. / Mechanisms of Ageing and Development xxx (2013) xxx–xxx8

G Model

MAD 10701 1–9

alyses and writing article: OvdR and NvdZ. Expertise and dataocessing of reaction time test: JA. GJH is an employee at thetch Dairy Association (NZO), but he had no interference withta collection and data analysis. All authors provided criticalvision of the manuscript.

Funding

This study was financed by Top Institute Food and NutritionIFN) and co-financed by the Dutch Dairy Association (NZO) ande European Union’s Seventh Framework Program under Grantreement No. 266486 (‘NU-AGE: New dietary strategies addres-g the specific needs of the elderly population for healthy aging

Europe’). None of the authors had any personal or financialnflicts of interest.These trials were registered at clinicaltrials.gov as NCT01110369

d NCT01109628.

knowledgements

We gratefully acknowledge all participants of the ProMuscleudy. Furthermore, we thank all colleagues and students whoorked in this study for their help and dedication.

pendix A. Supplementary data

Supplementary data associated with this article can be found, ine online version, at http://dx.doi.org/10.1016/j.mad.2013.12.005.

ferences

berti, K.G., Zimmet, P.Z., 1998. Definition, diagnosis and classification of diabetesmellitus and its complications. Part 1: diagnosis and classification of diabetesmellitus provisional report of a WHO consultation. Diabetic Med.: J. Br. DiabeticAssoc. 15, 539–553.

gevaren, M., Aufdemkampe, G., Verhaar, H.J., Aleman, A., Vanhees, L., 2008.Physical activity and enhanced fitness to improve cognitive function in olderpeople without known cognitive impairment. Cochrane Database of SystematicReviews, CD005381.

ab, L., Sabbagh, M.N., 2010. Are certain lifestyle habits associated with lowerAlzheimer’s disease risk? J. Alzheimers Dis.: JAD 20, 785–794.

erer, L., Erickson, K.I., Liu-Ambrose, T., 2013. A review of the effects of physicalactivity and exercise on cognitive and brain functions in older adults. J. AgingRes. 2013, 657508.

nnefoy, M., Cornu, C., Normand, S., Boutitie, F., Bugnard, F., Rahmani, A., Lacour,J.R., Laville, M., 2003. The effects of exercise and protein-energy supplements onbody composition and muscle function in frail elderly individuals: a long-termcontrolled randomised study. Br. J. Nutr. 89, 731–739.

rsheim, E., Bui, Q.U., Tissier, S., Kobayashi, H., Ferrando, A.A., Wolfe, R.R., 2008.Effect of amino acid supplementation on muscle mass, strength and physicalfunction in elderly. Clin. Nutr. 27, 189–195.

yle, P.A., Buchman, A.S., Wilson, R.S., Leurgans, S.E., Bennett, D.A., 2010. Physicalfrailty is associated with incident mild cognitive impairment in community-based older persons. J. Am. Geriatr. Soc. 58, 248–255.

and, N., Jolles, J., 1985. Learning and retrieval rate of words presented auditorilyand visually. J. Gen. Psychol. 112, 201–210.

ooks, R., 1996. EuroQol: the current state of play. Health Policy 37, 53–72.sse, A.L., Filho, W.J., Magaldi, R.M., Coelho, V.A., Melo, A.C., Betoni, R.A., Santarem,

J.M., 2008. Effects of resistance training exercise on cognitive performance inelderly individuals with memory impairment: results of a controlled trial.Einstein 6, 402–407.

ssilhas, R.C., Viana, V.A., Grassmann, V., Santos, R.T., Santos, R.F., Tufik, S., Mello,M.T., 2007. The impact of resistance exercise on the cognitive function of theelderly. Med. Sci. Sport Exer. 39, 1401–1407.

ssilhas, R.C., Lee, K.S., Fernandes, J., Oliveira, M.G., Tufik, S., Meeusen, R., de Mello,M.T., 2012. Spatial memory is improved by aerobic and resistance exercisethrough divergent molecular mechanisms. Neuroscience 202, 309–317.

in, A.P.M.J., de Jong, N., Schouten, E.G., van Staveren, W.A., Kok, F.J., 2002. Physicalexercise or micronutrient supplementation for the wellbeing of the frail elder-ly? A randomised controlled trial. Br. J. Sport. Med. 36, 126–131.

lcombe, S., Kramer, A.F., 2003. Fitness effects on the cognitive function of olderadults: a meta-analytic study. Psychol. Sci. 14, 125–130.

lcombe, S.J., Erickson, K.I., Raz, N., Webb, A.G., Cohen, N.J., McAuley, E., Kramer,A.F., 2003. Aerobic fitness reduces brain tissue loss in aging humans. J. Gerontol.A-Biol. 58, 176–180.

Please cite this article in press as: van de Rest, O., et al., Effecsupplementation on cognitive functioning in frail and pre-frail eldcontrolled trial. Mech. Ageing Dev. (2013), http://dx.doi.org/10.101

Erickson, K.I., Voss, M.W., Prakash, R.S., Basak, C., Szabo, A., Chaddock, L., Kim, J.S.,Heo, S., Alves, H., White, S.M., Wojcicki, T.R., Mailey, E., Vieira, V.J., Martin, S.A.,Pence, B.D., Woods, J.A., McAuley, E., Kramer, A.F., 2011. Exercise trainingincreases size of hippocampus and improves memory. Proceedings of theNational Academy of Sciences of the United States of America 108, 3017–3022.

Erickson, K.I., Banducci, S.E., Weinstein, A.M., Macdonald, A.W., 3rd, Ferrell, R.E.,Halder, I., Flory, J.D., Manuck, S.B., 2013. The brain-derived neurotrophic factorVal66Met polymorphism moderates an effect of physical activity on workingmemory performance. Psychol. Sci. 24, 1770–1779.

Esmarck, B., Andersen, J.L., Olsen, S., Richter, E.A., Mizuno, M., Kjaer, M., 2001.Timing of postexercise protein intake is important for muscle hypertrophy withresistance training in elderly humans. J. Physiol. 535, 301–311.

Fernstrom, J.D., 2012. Large neutral amino acids: dietary effects on brain neuro-chemistry and function. Amino Acids.

Folstein, M.F., Folstein, S.E., McHugh, P.R., 1975. ‘‘Mini-mental state’’. A practicalmethod for grading the cognitive state of patients for the clinician. J. Psychiatr.Res. 12, 189–198.

Fried, L.P., Tangen, C.M., Walston, J., Newman, A.B., Hirsch, C., Gottdiener, J., Seeman,T., Tracy, R., Kop, W.J., Burke, G., McBurnie, M.A., 2001. Frailty in older adults:evidence for a phenotype. J. Gerontol. A-Biol. 56, M146–M156.

Guralnik, J.M., Simonsick, E.M., Ferrucci, L., Glynn, R.J., Berkman, L.F., Blazer, D.G.,Scherr, P.A., Wallace, R.B., 1994. A short physical performance battery assessinglower extremity function: association with self-reported disability and predic-tion of mortality and nursing home admission. J. Gerontol. 49, M85–M94.

Hausdorff, J.M., Yogev, G., Springer, S., Simon, E.S., Giladi,N., 2005. Walking is more likecatching than tapping: gait in the elderly as a complex cognitive task. Exp. BrainRes. Experimentelle Hirnforschung. Experimentation cerebrale 164, 541–548.

Heyn, P., Abreu, B.C., Ottenbacher, K.J., 2004. The effects of exercise training onelderly persons with cognitive impairment and dementia: a meta-analysis.Arch. Phys. Med. Rehab. 85, 1694–1704.

Hindin, S.B., Zelinski, E.M., 2012. Extended practice and aerobic exercise interven-tions benefit untrained cognitive outcomes in older adults: a meta-analysis. J.Am. Geriat. Soc. 60, 136–141.

Lachman, M.E., Neupert, S.D., Bertrand, R., Jette, A.M., 2006. The effects of strengthtraining on memory in older adults. J. Aging Phys. Act. 14, 59–73.

Langlois, F., Vu, T.T., Chasse, K., Dupuis, G., Kergoat, M.J., Bherer, L., 2013. Benefits ofphysical exercise training on cognition and quality of life in frail older adults. J.Gerontol. B-Psychol. 68, 400–404.

Leckie, R.L., Weinstein, A.M., Hodzic, J.C., Erickson, K.I., 2012. Potential moderatorsof physical activity on brain health. J. Aging Res. 2012, 948981.

Lezak, M.D., Howieson, D.B., Loring, D.W., 2004. Neuropsychological assessment.Oxford University Press, New York.

Liu-Ambrose, T., Nagamatsu, L.S., Graf, P., Beattie, B.L., Ashe, M.C., Handy, T.C., 2010.Resistance training and executive functions: a 12-month randomized con-trolled trial. Arch. Intern. Med. 170, 170–178.

Mandayam, S., Mitch, W.E., 2006. Dietary protein restriction benefits patients withchronic kidney disease. Nephrology 11, 53–57.

Miller, J., 1982. Discrete versus continuous stage models of human informationprocessing: in search of partial output. J. Exp. Psychol. Human 8, 273–296.

Moul, J.L., Goldman, B., Warren, B.J., 1995. Physical Activity and Cognitive Perfor-mance in the Older Population. J. Aging Phys. Act. 3, 135–145.

Mungas, D., 1991. In-office mental status testing: a practical guide. Geriatrics 46 ,54–58, 63, 66.

Nagamatsu, L.S., Chan, A., Davis, J.C., Beattie, B.L., Graf, P., Voss, M.W., Sharma, D.,Liu-Ambrose, T., 2013. Physical activity improves verbal and spatial memory inolder adults with probable mild cognitive impairment: a 6-month randomizedcontrolled trial. Q7J. Aging Res. 861893.

NEVO, 2006. NEVO-tabel. Nederlands Voedingsstoffenbestand (Netherlands foodcomposition database 2006). The Netherlands Nutrition Centre, The Hague.

O’Connor, P.J., Herring, M.P., Caravalho, A., 2010. Mental health benefits of strengthtraining in adults. Am. J. Lifestyle Med. 4, 377–396.

Panton, L.B., Graves, J.E., Pollock, M.L., Hagberg, J.M., Chen, W., 1990. Effect of aerobicand resistance training on fractionated reaction time and speed of movement. J.Gerontol. 45, M26–M31.

Perrig-Chiello, P., Perrig, W.J., Ehrsam, R., Staehelin, H.B., Krings, F., 1998. The effectsof resistance training on well-being and memory in elderly volunteers. AgeAgeing 27, 469–475.

Radloff, L.S., 1977. The CES-D scale: a self-report depression scale for research in thegeneral population. Appl. Psych. Meas. 1, 385–401.

Reitan, R., 1958. Validity of the Trail Making Test as an indicator of organic braindamage. Percept. Mot. Skills 8, 271–276.

Rolland, Y., Abellan van Kan, G., Vellas, B., 2008. Physical activity and Alzheimer’sdisease: from prevention to therapeutic perspectives. J. Am. Med. Dir. Assoc. 9,390–405.

Smith, P.J., Blumenthal, J.A., Hoffman, B.M., Cooper, H., Strauman, T.A., Welsh-Bohmer, K., Browndyke, J.N., Sherwood, A., 2010. Aerobic exercise and neuro-cognitive performance: a meta-analytic review of randomized controlled trials.Psychosom. Med. 72, 239–252.

Sofi, F., Valecchi, D., Bacci, D., Abbate, R., Gensini, G.F., Casini, A., Macchi, C., 2010.Physical activity and risk of cognitive decline: a meta-analysis of prospectivestudies. J. Intern. Med. 269, 107–117.

Stroop, J., 1935. Studies of interference in serial verbal reactions. J. Exp. Psychol. 18,643–662.

Tieland, M., Borgonjen-Van den Berg, K.J., van Loon, L.J., de Groot, L.C., 2012a.Dietary protein intake in community-dwelling, frail, and institutionalizedelderly people: scope for improvement. Eur. J. Nutr. 51, 173–179.

t of resistance-type exercise training with or without proteinerly: Secondary analysis of a randomized, double-blind, placebo-6/j.mad.2013.12.005

654

655

656

657

658

659

660

661

662

663

664

665

666

667

668

669

670

671

6726736746756766776788679680681682683684685686687688

O. van de Rest et al. / Mechanisms of Ageing and Development xxx (2013) xxx–xxx 9

G Model

MAD 10701 1–9

Tieland, M., Dirks, M.L., van der Zwaluw, N., Verdijk, L.B., van de Rest, O., de Groot,L.C.P.G.M., van Loon, L.J., 2012b. Protein Supplementation Increases Muscle MassGain During Prolonged Resistance-Type Exercise Training in Frail Elderly People:A Randomized, Double-Blind, Placebo-Controlled Trial. J. Am. Med. Dir. Assoc..

Tieland, M., van de Rest, O., Dirks, M.L., van der Zwaluw, N., Mensink, M., van Loon,L.J., de Groot, L.C., 2012c. Protein supplementation improves physical perfor-mance in frail elderly people: a randomized, double-blind, placebo-controlledtrial. J. Am. Med. Dir. Assoc. 13, 720–726.

Tsutsumi, T., Don, B.M., Zaichkowsky, L.D., Takenaka, K., Oka, K., Ohno, T., 1998.Comparison of high and moderate intensity of strength training on mood andanxiety in older adults. Percept. Mot. Skills 87, 1003–1011.

Valentijn, S.A., van Boxtel, M.P., van Hooren, S.A., Bosma, H., Beckers, H.J., Ponds,R.W., Jolles, J., 2005. Change in sensory functioning predicts change in cognitivefunctioning: results from a 6-year follow-up in the Maastricht aging study. J.Am. Geriatr. Soc. 53, 374–380.

van de Rest, O., van der Zwaluw, N.L., de Groot, L.C., 2013. Literature review on therole of dietary protein and amino acids in cognitive functioning and cognitivedecline. Amino Acids 45, 1035–1045.

Please cite this article in press as: van de Rest, O., et al., Effectsupplementation on cognitive functioning in frail and pre-frail eldecontrolled trial. Mech. Ageing Dev. (2013), http://dx.doi.org/10.1016

van Uffelen, J.G., Chin, A.P.M.J., Hopman-Rock, M., van Mechelen, W., 2008. Theeffects of exercise on cognition in older adults with and without cognitivedecline: a systematic review. Clin. J. Sport Med.: Official Journal of the CanadianAcademy of Sport Medicine 18, 486–500.

Vellas, B., Andrieu, S., Sampaio, C., Wilcock, G., European Task Force, g, 2007.Disease-modifying trials in Alzheimer’s disease: a European task force consen-sus. Lancet Neurol. Q6, 56–62.

Voelcker-Rehage, C., Niemann, C., 2013. Structural and functional brain changesrelated to different types of physical activity across the life span. Neurosci.Biobehav. R.

Ware Jr., J., Kosinski, M., Keller, S.D., 1996. A 12-Item Short-Form Health Survey:construction of scales and preliminary tests of reliability and validity. Med. Care34, 220–233.

Wechsler, D., 1981. Manual for the Adult Intelligence Scale-Revised. PsychologicalCorporation, New York.

World Health Organization and Alzheimer’s Disease International, 2012. Dementia:a public health priority. World Health Organization.

of resistance-type exercise training with or without proteinrly: Secondary analysis of a randomized, double-blind, placebo-/j.mad.2013.12.005