C

EeA

WVN

a

Rb

c

d

A

V

h0

omplementary Therapies in Medicine (2014) 22, 614—620

Available online at www.sciencedirect.com

ScienceDirect

j ourna l ho me pa g e: www.elsev ierhea l th .com/ journa ls /c t im

xercise training and music therapy inlderly with depressive syndrome:

pilot study

. Verrusioa,∗, P. Andreozzia, B. Mariglianob, A. Renzic,. Gianturcoa, M.T. Peccid, E. Ettorrea, M. Cacciafestaa,. Gueli a

Department of Cardiovascular, Respiratory, Nephrological, and Geriatric Sciences, Sapienza University ofome, Rome, ItalyInternal Medicine, Campus Bio-medico of Rome, Rome, ItalyDepartment of Dynamic and Clinical Psychology, ‘‘Sapienza’’ — University of Rome, Italy‘‘Science of Aging’’ Interdepartmental Research Center — Sapienza University of Rome, Italyvailable online 6 June 2014

KEYWORDSAnxiety;Depression;Exercise;Music therapy;Elderly;Rehabilitation

SummaryObjective: Recent studies have thrown doubt on the true effectiveness of anti-depressants inlight and moderate depression. The aim of this study is to evaluate the impact of physical train-ing and music therapy on a sample group of subjects affected by light to moderate depressionversus subjects treated with pharmacological therapy only.Design and setting: Randomized controlled study. Patients were randomized into two groups.Subjects in the pharmacotherapy group received a therapy with antidepressant drugs; the exer-cise/music therapy group was assigned to receive physical exercise training combined withlistening to music. The effects of interventions were assessed by differences in changes inmood state between the two groups.Main outcome measures: Medically eligible patients were screened with the Hamilton AnxietyScale and with the Geriatric Depression Scale. We used plasmatic cytokine dosage as a stress

marker.Results: We recruited 24 subjects (mean age: 75.5 ± 7.4, 11 M/13 F). In the pharmacotherapygroup there was a significant improvement in anxiety only (p < 0.05) at 6-months. In the exer-cise/music therapy was a reduction in anxiety and in depression at 3-months and at 6-months(p < 0.05). We noted an average reduction of the level of TNF-a from 57.67 (±39.37) pg/ml to35.80 (±26.18) pg/ml.

∗ Corresponding author at: Department of Cardiovascular, Respiratory, Nephrological, and Geriatric Sciences, Sapienza University of Rome,iale del Policlinico 155, 00161 Roma, Italy. Tel.: +39 3490745274.

E-mail address: walter.verrusio@uniroma1.it (W. Verrusio).

ttp://dx.doi.org/10.1016/j.ctim.2014.05.012965-2299/© 2014 Elsevier Ltd. All rights reserved.

Exercise training and music therapy with depressive syndrome 615

Conclusions: Our training may potentially play a role in the treatment of subjects with mild tomoderate depression. Further research should be carried out to obtain more evidence on effectsof physical training and music therapy in depressed subjects.© 2014 Elsevier Ltd. All rights reserved.

ont

aaraomIFttat

eetm

td

M

Ta1mparpaaoIotbgbfdDepression Scale (GDS) score between 5 and 12; absence ofpharmacological treatment. Patients suffering from severedepression, or for whom physical exercise was not rec-

Introduction

Depression and anxiety are frequent disorders in the elderly.Beyond the age of 65, the risk of falling ill with depressionincreases threefold compared to the general population.1

The Food and Drug Administration (FDA) recently releasedthe data underlying the approval of six of the most widelysold anti-depressants: in 47 controlled clinical studies, only10—20% showed benefits effectively attributable to thepharmacological action of the molecule employed, whilethe placebo effect was responsible for the improved moodof 80—90% of the subjects.2 In confirmation of this find-ing, a recent study showed that drugs against depressionproved partially effective only in subjects suffering fromsevere depression.3 In recent years, therefore, extensivescientific evidence has thrown doubt on the true effective-ness of anti-depressants, especially for patients affected bylight or moderate depression. And we need, especially in thefield of geriatrics, to look for new tools for the treatment ofmood disturbances.

Experience has been gained, over the last few years,with cognitive therapy (CT). A study carried out on 240patients suffering from moderate to severe depressionshowed that, although the percentage of response totreatment with CT overlapped, to a certain extent, withthe response to pharmaceuticals, the long-term incidenceof new episodes of depression in the group treated withSelective serotonin reuptake inhibitors (SSRIs) was 54%,compared to 17% among patients treated with CT.4 Thissuggests not only that the effect of CT lasts longer than thatof pharmaceuticals, but that CT can produce changes whichdiffer from those brought about by pharmaceutical therapy,making it a valid alternative to the use of anti-depressants.In the field of geriatrics, other works have highlighted thepositive effect of music on treating cognitive disturbances5

and behavioral and psychological symptoms of dementia(BPSD)6 and on a whole series of unfavorable stress-inducedcircumstances, including anxiety and depression.7 Recentresearch efforts examining the effects of listening to musicon the brain have discovered that music is able to increasecerebral synaptic plasticity.8 In other words, though theyact through different mechanisms and have differenteffects, both depression and music work in a shared sub-strate of certain areas of the brain, giving rise to a seriesof changes.9 The fact is that exposure to sound leads toincreased neurogenesis at the hippocampus, where neuronimpoverishment due to loss and/or scarce regeneration isthought to be an underlying cause of a variety of mooddisturbances, including depression.10 Therefore there is a

very close link between depression and music.

Another therapeutic strategy recently proposed for thetreatment of depressed subjects is physical activity. Physicaltraining helps control mood disturbances, especially those

odsu

f light to moderate intensity, with a level of effective-ess comparable to that of psychological or pharmacologicalherapies, especially over the long term.11

In the present study, we used plasmatic cytokine dosages a stress marker. Recent research has pointed to anctive role on the part of pro-inflammatory cytokines in theegulation of synaptic plasticity, demonstrating that hyper-ctivation of the immune system can be identified as partf the pathogenesis of depression.12 Patients suffering fromood disturbances have been found to present increases in

nterleukin 1 (IL1), Interleukin 6 (IL6) and Tumor Necrosisactor alpha (TNF�), tied to a greater risk of cognitive dis-urbances and mood depression, plus a reduced response toherapy.13 The cytokine dosage, therefore, can be used as

parameter for evaluating the effectiveness of a proposedherapy.

The primary aim of this study is to determine whetherxposure to music and physical training can have a positiveffect on the mood of elderly subjects suffering from lighto moderate depression, and whether the improvement inood persists over time.A secondary aim of the study is to evaluate the effec-

ive influence of aerobic physical training on the concurrentisturbances of the subjects examined.

aterials and methods

his was a randomized controlled trial. This study included series of elderly subjects (n = 24; mean age: 75.5 ± 7.4;3 females) diagnosed according to the DSM-IV criteria withajor depression with mild to moderate severity.14,15 Eacharticipant received information about this study in writingnd an individual verbal explanation of this study from theesearcher. Participation was entirely voluntary. Individualarticipants in this study gave written informed consent and

signed declaration of consent. This study was conductedccording to the principles expressed in the Declarationf Helsinki and was approved by the ‘‘Science of Aging’’nterdepartmental Research Center of Sapienza Universityf Rome. We evaluated the concurrent conditions throughhe Cumulative Illness Rating Score (CIRS) and its Comor-idity Index (CInd).16 The main diseases found in the testroup were: hypertension, obesity, diabetes mellitus, car-ohydrate intolerance, dyslipidemia. Consistency with theollowing criteria of inclusion was assessed for each patient:iagnosis of light to moderate depression, with a Geriatric

mmended (blood pressure — SBP — >200 mmHg and/oriastolic blood pressure — DBP — >110 mmHg; diabeticubjects with fasting blood glucose — FBG — >250 mg/dl;nstable angina pectoris; arrhythmias; severe heart valves

6

dw

cccs

r((o

-

-

r9n3n

awslwnpembbrecfddisoscd

•••

vdrdwofi

soavtacodwc

R

2iCt2

bd

twppliW6

ga(wa

mmwl

tr

16

iseases; aneurysms; any kind of severe systemic diseases),ere excluded.

Participants were randomly assigned to the pharma-otherapy group or the exercise/music therapy group usingomputerized random numbers. Group allocation was notommunicated to the patients until the first exercise/musicession or the first prescription of the drug.

An electronic chart was filled out for each subject,ecording the following parameters at the start of the studyW0), at the three-month point (W12), after six monthsW24) and, finally, in the exercise/music therapy group atne month after the end of the study (W28):

pharmacotherapy group: GDS for assessing mood; Hamil-ton Anxiety Scale (HAS) score for evaluating anxiety;

exercise/music therapy group: GDS and HAS; standardlaboratory techniques were used to determine, after anovernight fast, plasma total cholesterol (TC), high-densitylipoprotein cholesterol (HDLc), triglycerides (TG), FBGand plasmatic cytokine dosage (TNF�, IL1, IL6). Waist cir-cumference (WC) was measured at the level of iliac crestwith the patients standing. The body mass index (BMI) wasdetermined by dividing the weight (kg) by the square ofheight (m).

The pharmacotherapy group subjects were assigned toeceive a antidepressant medication involving an SSRI in

patients (paroxetine 20 mg/die) and a specific seroto-ergic antidepressant (NaSSA) in 2 patients (mirtazapine0 mg/die), associated with benzodiazepine (Alprazolam) aseeded.

The exercise/music therapy group subjects weressigned to receive physical exercise training combinedith listening to music. Each patient engaged in two

essions of physical exercise a week, with each sessionasting approximately an hour. The routine consisted in:arm-up of muscles, general gymnastics or postural gym-astics, aerobic training on stationary bicycles or treadmills,ost-workout decompression. The physical activity was mod-rately intense so as not to exceed the target pulse rate,eaning 75% of the maximum pulse rate for the patienteing treated (based on the theoretical maximum pulse ratey age, or on the Borg scale).17 Blood pressure, resting heartate, pulse oximetry and FBG were assessed before and afterxercise. While the patients exercised, a collection of musi-al pieces was played. Three different play-lists were usedor the study, with the style of music selected in accor-ance with the tastes of the patients, who were intervieweduring the recruitment phase. Three different genres weredentified: Jazz (Mt.1), Classical (Mt.2), Modern (Mt.3). Theample group being examined was divided into 3 subgroupsf four patients each, one for each play-list. The differentongs on the various play-lists each corresponded to a spe-ific phase of the physical training session; each session wasivided as follows:

warm up (light exertion, slow rhythm), main part (moderate-intense exertion, fast rhythm), cool down (decompression phase, slow rhythm).

saoa

W. Verrusio et al.

For the portion of the study involving depression, theariations in mood were measured with the GDS, a briefepression screening inventory composed of 15 items thatequire yes or no answers. A score of 5 or more indicatesepressed individuals.18 The state of anxiety was evaluatedith the HAS. It consists of 14 items, each defined by a seriesf symptoms. Each item is rated on a 5-point scale, rangingrom 0 (not present) to 5 (severe). A score of 18 or morendicates anxiety.19

An overall description was performed for each of the con-idered variables. For qualitative variables, the frequenciesf each modality were determined. For quantitative vari-bles mean and standard deviation were calculated. Meanalues were compared using Student’s t-test. The change inhe parameters between groups was compared by one-waynalysis of variance and HAS score at baseline was used asovariate (ANCOVA), to control the effect of this variable onutcome at 12 and 24 months (since it appeared marginallyifferent in the two groups). A 95% confidence intervals (CI)as calculated. A p value below 0.05 was deemed signifi-ant.

esults

4 patients were randomized into two groups: 12 subjectsnto the pharmacotherapy group (mean age: 76.1 ± 7.1;Ind: 2.4 ± 2.3; males/females: 4/8) and 12 subjects intohe exercise/music therapy group (mean age: 74.8 ± 8; CInd:.1 ± 0.7; males/females: 7/5) (Fig. 1).

All patients completed the study protocol.The characteristics of the patients including age, Comor-

idity Index, GDS and HAS scores were not significantlyifferent between the two groups at baseline.

Table 1 shows test scores in the two groups.In the pharmacotherapy group there were minimal varia-

ions in GDS and HAS scores, however there was a significantithin group change in anxiety observed after 24 weeks com-ared to baseline data (p < 0.05) (Fig. 2). At 3-months, werescribed in association therapy benzodiazepine (Alprazo-am) to 6 patients and we increased antidepressant dosagen 4 patients (paroxetine, from 20 mg/die to 40 mg/die).e also changed antidepressant therapy of 9 patients at

-months because of side effects.In the exercise/music group there were significant within

roup reductions both in anxiety and in depression observedfter 12 and 24 weeks compared to baseline data (p < 0.05)Fig. 2). In general, the physical training in our study wasell tolerated and the exercise/music group did not takeny antidepressant medication during study period.

There were significant differences between the phar-acotherapy group and exercise/music group for HASeasurements at week 12 and at week 24 and for GDS ateek 24 (p < 0.05), although the covariate for HAS at base-

ine was included (Table 1).The cytokine dosages in the serum of the exercise/music

herapy group subjects examined pointed to a linear cor-elation between high levels of cytokine and a high GDS

core. In fact, of the 12 patients examined, 9 presented average GDS score of 7.44 (±1.23) and undetectable dosesf cytokines in the blood plasma, while 3 subjects showed

average GDS score of 11.66 (±0.57) combined with high

Exercise training and music therapy with depressive syndrome 617

Assessed for eligibili ty (n=24)

Excluded (n=0) Αll par ticipan ts assessed for e lig ibili ty met the inclusion cri teria

Analysed (n=12)

Lost to foll ow-up (n=0)

Continued fo llow up (n=1 2)

All ocate d to Pharmacoth erapy gro up (n=12) Received allocated inte rvention ( n=12)

Lost to foll ow-up (n=0)

Continued fol low up (n=1 2)

Allocated to Ex erci se/music ther apy group (n=12) Received allocated inte rvention (n=12)

Analysed (n=12)

Rand omized (n=24 )

stu

ppw1oW

Figure 1 The

levels of TNF-�. The following dosage reading, taken at 24weeks, showed that the average level of TNF-� in the threemost depressed subjects had fallen from 57.67 (±39.37)pg/ml to 35.80 (±26.18) pg/ml.

A new assessment of mood carried out 4 weeks (W28)after the suspension of the rehabilitation program resultedin only minimal differences in the GDS and HAS scores. Forthese parameters, the difference between W24 and W28 was

minimal (GDS: −0.17 ± 0.58; HAS: 0.17 ± 1.03).

As for the secondary aim of the study, after 6 monthswe observed the following in the exercise/music therapygroup: a reduction of 7 mmHg in the average systolic blood

D

Ts

Table 1 Anxiety (HAS) and depression (GDS) in the two groups. V(W24) and differences between groups not-adjusted and adjusted

PharmacotherapygroupMean (SD)

Exercise/musictherapy groupMean (SD)

Differencesbetween gro[95% CI]

HASW0 24.2 (±5.3) 22.2 (±5.07) −2.08 [−6.5W12 23.5 (±3.5) 19.6 (±4.8) −3.92 [−7.5W24 22 (±4.7) 16.5 (±2.7) −5.50 [−8.8

GDSW0 8.4 (±1.8) 8.5 (±2.2) 0.09 [−1.9W12 8.6 (±2.9) 7.6 (±1.6) −1.2 [−2.84W24 8 (±2.5) 5.5 (±1) −2.92 [−4.3

p < 0.05 = significant.

dy flow chart.

ressure (p < 0.05) and 3 mmHg in the average diastolic bloodressure of the group examined; a variation in the lipid level,ith average reductions of 27 mg/dl in TC (p < 0.05) and of4.75 mg/dl in the TG; a reduction of 16.7 mg/dl in the levelsf basic FBG (p < 0.05); an average reduction of 6.2 cm in theC of the sample group examined (Table 2).

iscussion

he primary aim of this study is to determine whether expo-ure to music and physical training can have a positive effect

alues at baseline (W0), 12 weeks later (W12), 24 weeks laterfor HAS score at baseline.

upsF; pNot-adjusted

F; pAdjusted

F p F p

1 to 2.34] 0.95 0.34 / /0 to −0.33] 5.14 0.03 4.55 0.040 to −2.20] 11.92 0.00 13.33 0.00

7 to 2.14] 0.01 0.92 / / to 0.84] 1.08 0.31 1.74.209 to −0.61] 10.44 0.01 8.72 0.01

618 W. Verrusio et al.

F d staH he tw

oepsgelicma6

lcsbetoda

aat

F

aodmrpgatatvfirTdcabocpat

igure 2 Change in depression and anxiety scores. Means anAS = Hamilton Anxiety Scale. *Significant difference between t

n the mood of elderly subjects suffering from light to mod-rate depression and whether the improvement in moodersists over time. The results at the third and sixth monthhow a twofold positive effect in the exercise/music therapyroup, reducing both depression and the symptoms of anxi-ty in the group. We also observed progressive improvementinked to an increase in the number of sessions, point-ng to the possibility of a dose-dependent effect. On theontrary, in the pharmacotherapy group we observed onlyinimal variations in GDS and HAS scores at the third month

nd we observed a significant reduction in anxiety only at-months.

We assessed whether the positive results of our trainingasted over time, subjecting the subjects of the exer-ise/music therapy group to a new control 4 weeks afteruspension of the rehabilitation program. The results foroth depression and anxiety showed only minimal differ-nces in the GDS and HAS scores. With regard to physicalraining, our study shows that it has both a beneficial effectn mood and a positive influence on an entire series of con-itions that present a high rate of morbidity in the geriatricge bracket.

In light of these results, we hold that physical trainingssociated with exposure to music is capable of setting off

series of positive effects leading to modifications in bothhe depressive and anxious components of the mood.20

Nevertheless the current study faces certain limitations.irstly, the small size of the sample. Other studies have

eaai

Table 2 Side effects of antidepressant medications and seconda

Antidepressant drugs

MAOIs Hypertensive crisis, flushing, nauseaTCAs Constipation, weight gain, sexual dysfunction,

dizziness, memory disturbances, sedationSSRIs Dyspepsia, nausea, sexual dysfunctionNASSAs Weight gain, drowsiness, dizziness, headacheNARI Hypotension and tachycardia, insomnia,

urinary retentionNSRIs Dyspepsia, nausea, constipation

MAOI, monoamine oxidase inhibitors; TCA, tricyclic antidepressants; SSRand specific serotonergic antidepressants; NARI, selective noradrenaliinhibitors.

ndard deviation (error bars). GDS = Geriatric Depression Scale;o groups.

nalyzed the effect of music exposure or exercise trainingn depression,21—23 but our study provides preliminary evi-ence of the combined benefit of exercise and listening tousic for mood disturbances in elderly patients. For this

eason we have chosen to study two groups of patients, aharmacotherapy group versus an exercise/music therapyroup, exploring directly synergistic therapy. However, whennalysing our results it is not possible to disentangle the con-ribution of each component to the overall effect. Secondly,lthough the very small number of test subjects means thathe result cannot be considered statistically significant, theariations in cytokines levels of our study are similar tondings reported in other studies that have pointed to a cor-elation between high levels of cytokines and depression.24

hese preliminary results seem to confirm that the cytokineosages in the serum may potentially play a role as indi-ator of effectiveness for evaluating the effectiveness of

antidepressive therapy but further studies are needed toetter characterize these correlations. Thirdly, the resultsf our training in the exercise/music therapy group, espe-ially over the short term, can be attributed in part to thelacebo effect, which, nevertheless, can be used to gooddvantage when treating depressed subjects, especially inhe early phases of a rehabilitation program. For the placebo

ffect can encourage the depressed subject to take a morective role in the healing process while helping to create

good physician—patient relationship, in this way creatingdeal conditions for pursuing the therapeutic strategy over

ry effects of our rehabilitation program.

Training (value W24)

• Reduction of 7 mmHg in the average systolic pressureand 3.2 mmHg in the average diastolic pressure;• Average reductions of 15% in total cholesterol and of10% in the triglycerides;• Reduction of 15% in the levels of basic glycaemia;• Average reduction of 6.2 cm in the waistmeasurements

Is, selective serotonin reuptake inhibitors; NASSAs, noradrenergicne reuptake inhibitor; NSRIs, norepinephrine serotonin reuptake

e

gia

C

Ti

R

1

1

1

1

1

Exercise training and music therapy with depressive syndrom

the long term, at which point the results will be influencedto a lesser extent by the placebo effect.

We feel that, despite the limitations of this pilot study,the results can have interesting implications for cliniciansin the management of light and moderate depression.It is known that modifications in the neuro-transmissionof elderly subjects render them more vulnerable tothe extra-pyramidal and anticholinergic effects of anti-depressant pharmaceuticals, resulting in the frequent onsetof side effects.1 This also delays the construction of aphysician—patient relationship that is important for themanagement of depressed patients and requires more timein order to find an effective therapy that is well toleratedby the patient. On the contrary, a variety of different mech-anisms can be identified to explain the positive effect ofphysical exercise on depression: the fact that the depressedsubjects are taking care of themselves heightens their self-reliance and esteem; the exercise leads to patients to havemore positive thoughts and plans; the alliance establishedbetween the patient and the care-giver in ‘‘taking action’’to cure the depression and anxiety. Moreover, the resultsof the present study demonstrate the positive effect of ourtraining on all the risk factors linked to the metabolic syn-drome. It is safe to say, that, even in terms of its ‘‘sideeffects’’, our rehabilitation program proved to be safe andeffective (Table 2). Unlike other studies,25,26 the distinguish-ing feature of our study is the use of the style of musicselected in accordance with the tastes of the patients andthe results show that our training had a beneficial effect,albeit at varying levels, in each of the groups of patientsthat listened to the different styles of music. Other originalelement of our study is the way it has combined two suchdifferent techniques, namely physical exercise and musictherapy. We have divided the training session in 3 phases(warm up, main part, cool down) in order to create closeassociation between musical rhythm and intensity of theactivity being performed, in order to increase the synergisticeffect of physical exercise associated with music therapy.

Future studies should be randomized controlled trials ofexercise training and music therapy in depressed patients,including a systematic investigation of possible biologi-cal action mechanism (e.g., reduction in inflammatorybiomarkers).

Conclusions

The preliminary results allow us to state that:

• Our training may potentially play a role in the treatmentof subjects with mild to moderate depression.

• It appears that the improvement in mood may persist overshort term: at a distance of 4 weeks from the suspensionof our program, we observed that the benefits obtainedwere essentially stable. In other words, our training mayserve as a tool in the rehabilitation of mood disturbances.

• Thanks to the multidimensional approach of our program,we can treat both mood disturbances and the variouscomorbid conditions that can aggravate the psycho-physical health of elderly patients.

1

619

Further research should be carried out using a largerroup of patients to obtain more evidence on effects of phys-cal training and music therapy in rehabilitation of subjectsffected by light to moderate depression.

onflict of interest

he authors declare that they have no competing financialnterests.

eferences

1. Marigliano V. Manuale breve di geriatria. Società Editrice Uni-verso; 2007. p. 62—75.

2. Kirsch I, Deacon BJ, Huedo-Medina TB, Scoboria A, Moore TJ,Johnson BT. Initial severity and antidepressant benefits: a meta-analysis of data submitted to the Food and Drug Administration.PLoS Med 2008;5(February (2)):45.

3. Fournier JC, DeRubeis RJ, Hollon SD, Dimidjian S, AmsterdamJD, Shelton RC, et al. Antidepressant drug effects and depres-sion severity. A patient-level meta-analysis. J Am Med Assoc2010;303(1):47—53.

4. DeRubeis RJ, Sigle GJ, Hollon SD. Cognitive therapy verusmedication for depression: treatment outcomes and neuralmachanisms. Nat Rev Neurosci 2008;9(October (10)):788—96.

5. Cacciafesta M, Ettorre E, Amici A, Cicconetti P, Martinelli V,Linguanti A, et al. New frontiers of cognitive rehabilitation ingeriatric age: the Mozart Effect (ME). Arch Gerontol Geriatr2010;51(November—December (3)):e79—82.

6. Raglio A, Bellelli G, Traficante D, Gianotti M, Ubezio MC, Vil-lani D, et al. Efficacy of music therapy in the treatment ofbehavioral and psychiatric symptoms of dementia. AlzheimerDis Assoc Disord 2008;22(April—June (2)):158—62.

7. Guétin S, Soua B, Voiriot G, Picot MC, Hérisson C. The effectof music therapy on mood and anxiety-depression: an observa-tional study in institutionalised patients with traumatic braininjury. Ann Phys Rehabil Med 2009;52:30—40.

8. Kim H, Lee MH, Chang HK, Lee TH, Lee HH, Shin MC, et al.Influence of prenatal noise and music on the spatial memoryand neurogenesis in the hippocampus of developing rats. BrainDev 2006;28(March (2)):109—14.

9. Rauschecker Josef P. Cortical plasticity and music. Ann N Y AcadSci 2001;930(June):330—6.

0. den Heijer T, Tiemeier H, Luijendijk HJ, van der Lijn F, KoudstaalPJ, Hofman A, et al. A study of the bidirectional associationbetween hippocampal volume on magnetic resonance imagingand depression in the elderly. Biol Psychiatry 2011;(June).

1. Carek PJ, Laibstain SE, Carek SM. Exercise for the treatment ofdepression and anxiety. Int J Psychiatry Med 2011;41(1):15—28.

2. Khairova RA, Machado-Vieira R, Du J, Manji HK. A potential rolefor pro-inflammatory cytokines in regulating synaptic plastic-ity in major depressive disorder. Int J Neuropsychopharmacol2009;12(May (4)):561—78.

3. Okada K, Kurita A, Takase B, Otsuka T, Kodani E, Kusama Y,et al. Effects of music therapy on autonomic nervous sys-tem activity, incidence of heart failure events, and plasmacytokine and catecholamine levels in elderly patients with cere-brovascular disease and dementia. Int Heart J 2009;50(January(1)):95—110.

4. American Psychiatric Association. Diagnostic and statisticalmanual of mental disorders. 4th ed. Washington: American Psy-

chiatric Association; 1994.

5. Lux V, Aggen SH, Kendler KS. The DSM-IV definition of severity ofmajor depression: inter-relationship and validity. Psychol Med2010;40(October (10)):1691—701.

6

1

1

1

1

2

2

2

2

2

2

2

20

6. Conwell Y, Forbes NT, Cox C, Caine ED. Validation of a mea-sure of physical illness burden at autopsy: the Cumulative IllnessRating Scale. J Am Geriatr Soc 1993;41:38—41.

7. Borg G. Borg’s perceived exertion and pain scales. Hum Kinet1998.

8. Ji S, Ja Y. Geriatric Depression Scale (GDS): recent evidence anddevelopment of a shorter version. In: Clinical gerontology: aguide to assessment and intervention. The Haworth Press Ltd.;1986. p. 165—73.

9. Hamilton M. The assessment of anxiety states by rating. Br JMed Psychol 1959;32(50).

0. Koelsch S, Fritz T, Cramon V, Müller DY, Friederici KAD. Inves-tigating emotion with music: an fMRI study. Hum Brain Mapp2006;27:239—50.

1. Ströhle A. Physical activity, exercise, depression and anx-

iety disorders. J Neural Transm 2009;116(June (6)):777—84.

2. Zhang JM, Wang P, Yao JX, Zhao L, Davis MP, Walsh D, et al.Music interventions for psychological and physical outcomes in

W. Verrusio et al.

cancer: a systematic review and meta-analysis. Support CareCancer 2012;20(December (12)):3043—53.

3. Erkkilä J, Punkanen M, Fachner J, Ala-Ruona E, Pöntiö I, Ter-vaniemi M, et al. Individual music therapy for depression:randomised controlled trial. Br J Psychiatry 2011;199(August(2)):132—9.

4. Rushaniya A, Machado-Vieira R, Du J, Manji HK. A potential roleroe pro-inflammatory cytokines in regulating synaptic plastic-ity in major depressive disorder. Int J Neuropsychopharmacol2009;12(May (4)):561—78.

5. Leea EJ, Bhattacharyab J, Sohnc C, Verresa R. Monochordsounds and progressive muscle relaxation reduce anxiety andimprove relaxation during chemotherapy: a pilot EEG study.Complement Ther Med 2012;20:409—16.

6. Cross K, Flores R, Butterfield J, Blackman M, Lee S. The effect of

passive listening versus active observation of music and danceperformances on memory recognition and mild to moderatedepression in cognitively impaired older adults. Psychol Rep2012;111(October (2)):413—23.

Reproduced with permission of the copyright owner. Further reproduction prohibited withoutpermission.