Does the Pisa syndrome affect postural control, balance, and gait in patients with Parkinson's disease? An observational cross-sectional study

lable at ScienceDirect

Parkinsonism and Related Disorders xxx (2015) 1e6

Contents lists avai

Parkinsonism and Related Disorders

journal homepage: www.elsevier .com/locate/parkreldis

Does the Pisa syndrome affect postural control, balance, and gait inpatients with Parkinson's disease? An observational cross-sectionalstudy

Christian Geroin a, Nicola Smania a, b, Federico Schena c, Eleonora Dimitrova a,Elisabetta Verzini d, Federica Bombieri c, Francesca Nardello c, Michele Tinazzi e,Marialuisa Gandolfi a, *

a Neuromotor and Cognitive Rehabilitation Research Center (CRRNC), Department of Neurological and Movement Sciences, University of Verona, P.le L.A.Scuro 10, 37134 Verona, Italyb Neurological Rehabilitation Unit, Azienda Ospedaliera Universitaria Integrata, P.le Scuro 10, 37134 Verona, Italyc School of Sport and Exercise Sciences, Department of Neurological and Movement Sciences, University of Verona, Via Casorati 43, 37137 Verona, Italyd School of Specialization in Physical Medicine and Rehabilitation, Department of Neurological and Movement Sciences, University of Verona, Italye Neurology Unit, Movement Disorders Division, Department of Neurological and Movement Sciences, University of Verona, P.le Scuro 10, 37134 Verona,Italy

a r t i c l e i n f o

Article history:Received 23 December 2014Received in revised form16 March 2015Accepted 19 April 2015

Keywords:FallsEquilibriumProprioceptionPostural orientation

* Corresponding author. Tel.: þ39 045 8124573.E-mail addresses: [email protected] (C. Ge

(N. Smania), [email protected] (F. Schena),com (E. Dimitrova), [email protected] (E.univr.it (F. Bombieri), [email protected] (Funivr.it (M. Tinazzi), [email protected] (M.

http://dx.doi.org/10.1016/j.parkreldis.2015.04.0201353-8020/© 2015 Elsevier Ltd. All rights reserved.

Please cite this article in press as: C. Geroin,disease? An observational cross-sectional stu

a b s t r a c t

Introduction: An altered sense of verticality, associated with impaired proprioception and somatosensoryintegration deficits, has been reported in patients with Parkinson's disease (PD) but it has not beencharacterized in patients with Pisa syndrome (PS). Therefore, we investigated postural control, balance,and gait disturbances in patients with PD and PS, patients with PD but without PS, and aged-matchednormal controls.Methods: This observational cross-sectional study involved patients with PD and PS (n ¼ 10, Hoehn &Yahr score <4), patients with PD but without PS (n ¼ 10), and age-matched healthy controls (n ¼ 10). Theprimary outcome measure was the velocity of CoP displacement (VEL_MED_AP/ML) assessed by staticstabilometry in eyes open (EO) and eyes closed (EC) conditions. The secondary outcomes were otherstabilometric parameters, the Sensory Organization Balance Test (SOT), and gait analysis (GA).Results: There were no significant differences in demographic and clinical data and Berg Balance Scalescores between the groups. There was a significant main effect in the VEL_MED_AP/ML between thegroups and eye conditions (p ¼ .016). A significant main effect was found in the EO (p ¼ .01) and EC(p ¼ .04) conditions. Post-hoc comparisons showed a significant increase in VEL_CoP in both the EO andEC conditions only in the patients with PD and PS. No significant main effects on SOT and GA were found.Conclusion: Patients with PD and PS had more difficulty achieving good postural alignment with gravityand greater velocity of body sway than the other groups. Rehabilitation programs for patients with PDand PS should include spine alignment and dynamic postural training.

© 2015 Elsevier Ltd. All rights reserved.

roin), [email protected]@gmail.Verzini), federica.bombieri@. Nardello), michele.tinazzi@Gandolfi).

et al., Does the Pisa syndromedy, Parkinsonism and Relate

1. Introduction

The Pisa syndrome (PS) is a postural deformity affecting patientswith Parkinson's disease (PD) with a prevalence of 1.9% [1]. It refersto a lateral trunk flexion of more than 10� which resolves by passivemobilization or lying in the supine position [2]. The pathologicalunderpinnings of PS in PD have not yet been fully characterized, buteither central or peripheral mechanismsmay be responsible for thedeformity [2,3]. Central mechanisms, such as trunk and lower limbmuscle dystonia, may play a crucial role. Electromyography studies

affect postural control, balance, and gait in patients with Parkinson'sd Disorders (2015), http://dx.doi.org/10.1016/j.parkreldis.2015.04.020

esMLCoP

displacemen

tside(cm)

APCoP

displacemen

tside(cm)

OnsetPS

(wee

ks)

Duration

PS(yea

rs)Th

erap

yat

onset(daily

mg)

Anterior

trunk

flex

ion(�)

Backpain

(0e10

VAS)

Metronom

e(Yes/no)

Awaren

ess

(yes/no)

L(15)

A(5)

128

Levo

dop

a(700

),Pram

ipex

ole(1.31)

200

No

Yes

L(7)

A(15)

73

Levo

dop

a(400

),Pram

ipex

ole(3.15),

Rasag

iline(1)

104

No

Yes

L(10)

P(2)

126

Levo

dop

a(600

),Pram

ipex

ole(1.31)

00

No

Yes

R(5)

A(9)

32

Levo

dop

a(400

),Pram

ipex

ole(1.04)

07

No

Yes

L(19)

P(7)

31

Levo

dop

a(800

),Pram

ipex

ole(2.1),

Rasag

iline(1)

83

Yes

Yes

R(2)

P(14)

34

Pram

ipex

olo(3.15),R

asag

iline(1)

466

No

Yes

L(2)

A(13)

14*

Levo

dop

a(600

),Pram

ipex

ole(2.1)

129

No

Yes

L(10)

A(1)

31

Levo

dop

a(190

0),E

ntacapon

e(200

),Pram

ipex

ole(0.52)

200

No

Yes

L(1)

P(4)

127*

Pram

ipex

ole(1.57),R

asag

iline(1)

223

No

No

L(12)

P(21)

35*

Levo

dop

a(130

0)8

2No

Yes

drome;

R,R

ight;L,

Left;Ty

peDom

inan

tPh

enotyp

e:YO,y

ounge

ron

set;TD

,tremor

dom

inan

t;NT,

non

-tremor

dom

inan

t;an

dRPD

,rap

iddisea

seprogression

;CoP

,an

teroposterior

CoP

displacemen

t;*(m

onths);VAS,

visu

alan

alog

uescale.

C. Geroin et al. / Parkinsonism and Related Disorders xxx (2015) 1e62

[4e6] have shown greater activation of the paraspinal and/ornonparaspinal muscles bilaterally (co-contraction), ipsilateral orcontrolateral to the trunk leaning side than in healthy subjects.Peripheral mechanisms, such as myopathy and degenerative spinaland soft tissue changes, may all lead to muscle imbalance, weak-ness, and compensatory posture [2].

Postural and balance control are two sides of the same coinwhich, once impaired, may lead to reduced mobility, greaterdisability, and lower quality of life [7]. Postural control serves toalign the body with respect to gravity, a support surface, and thesurrounding environment. In balance control, the center of mass ofthe body is shifted in relation to a base of support through staticand dynamic (gait) tasks performed during daily activities [8]. PDpatients often have difficulty with postural and balance control [9],but the presumed effect of PS on these disturbances has not beenfully clarified. Unpublished data indicate that abnormal posturalalignment is not necessarily related to abnormal postural and/orbalance responses [7]. Previous studies have demonstrated, how-ever, that PD patients suffer from an altered sense of verticality andthat an impaired proprioceptive system and somatosensory inte-gration may play a causative role [9,10]. Such impairment may alsocreate an inaccurate internal representation of the body [8], pre-disposing PD patients to an increased risk of falling [10,11]. Afurther potential complication of PD, observed particularly in pa-tients with PD and PS in the early stages of the disease, is the lack oftrunk misalignment awareness [2].

To our knowledge, no studies to date have investigated posturalcontrol, balance and gait disturbances in patients with PD and PS.The aim of this study was to investigate postural control, balanceand gait disturbances in patients with PD and PS, patients with PDbut without PS, and aged-matched healthy controls. We hypothe-sized that, because of postural misalignment, the PD patients withPS would have greater difficulty with postural control and balancethan either the PD patients or the healthy controls.

2. Methods

2.1. Study population

This observational cross-sectional study involved 10 patients with PD and PS(age range 42e80 years, mean 64.4 ± 11.36 SD), 10 patients with PD but without PS(age range 64e82 years, mean 72.1 ± 6.06), and 10 age-matched healthy controls(HS) (age range 56e83 years, mean 68.7 ± 8.38). The controls were recruited fromamong the patients' familymembers who showed no signs of neurological disease atassessment during enrollment into the study. All patients were attending theoutpatient clinic of the Movement Disorders Division, Neurological RehabilitationUnit, Department of Neurological and Movement Sciences, University Hospital.Demographic and clinical characteristics are reported in Tables 1 and 2.

All patients underwent neurological evaluation before enrollment. Inclusioncriteria were: a medical diagnosis of PD confirmed according to the United KingdomParkinson's Disease Society Brain Bank criteria [12] and/or PS defined as at least 10

Table 1Demographic and clinical characteristics of the patients and the controls.

Total PS PD HS P Value

Patients, no. 30 10 10 10 e

Gender, M/F 6/4 6/4 3/7 e

Age, mean (SD), yrs 64.4 (11.36) 72.1 (6.06) 68.7(8.38)

.58b

UPDRS III score 28.7 (16.10) 29.5 (19.21) e .92a

Pull Test score 0.80 (1.03) 1.30 (1.06) e .30a

H&Y stage 2.20 (0.63) 2.30 (0.67) e .74a

PDQ8 33 (20.15) 18.3 (12.64) e .66b

Falls 3.5 (5.06) 6.10 (6.14) e .31a

Abbreviations: PS, denotes patients with Parkinson's disease and Pisa Syndrome;PD, patients with Parkinson's Disease (without PS); HS, aged-matched healthycontrols; SD, standard deviation; M, Male; F, Female; yrs, years; H&Y, stage Hoehnand Yahr stage; UPDRS, Unified Parkinson's Disease Rating Scale; Pull Test score, subitem of UPDRS scale part III; PDQ8, Parkinson' s Disease Questionnaire; Falls,number of falls in the previous month; P value a, one-way ANOVA; and P value b,KruskalleWallis Test, p significant if < .05. Ta

ble

2Dem

ographic

andclinical

featuresof

PD/PSpatients.

Case

PDfeatures

PSfeatures

PDduration

(yea

rs)

Type

PDsymptoms

onset

Direction

(left/righ

t)Deg

re(�)

111

NT

RR

202

6NT

LR

50

316

NT

LR

104

8T

LL

165

7T

RL

18

64

NT

LR

427

5NT

LL

288

3NT

LR

30

95

NT

LL

1010

7NT

RL

30

Abb

reviations:

PD,d

enotes

Parkinson'sDisea

se;PS

,PisaSy

ncenterof

pressure;ML,

med

iolateralC

oPdisplacemen

t;AP,

Please cite this article in press as: C. Geroin, et al., Does the Pisa syndrome affect postural control, balance, and gait in patients with Parkinson'sdisease? An observational cross-sectional study, Parkinsonism and Related Disorders (2015), http://dx.doi.org/10.1016/j.parkreldis.2015.04.020

C. Geroin et al. / Parkinsonism and Related Disorders xxx (2015) 1e6 3

degrees of lateral trunk flexion that can be reduced by passive mobilization or su-pine positioning (PS � 10) [2], Hoehn & Yahr (H&Y) stage <4 in “ON” medicationphase. Exclusion criteria were: severe dyskinesia or “on-off” fluctuations; PDmedication modification in the 3 months preceding enrollment into the study; needfor assistive devices to rise from a chair or bed; somatic sensation deficits involvingthe legs; vestibular disorders or paroxysmal vertigo; other neurological, orthopedicor cardiovascular co-morbidities. All patients gave their informed consent toparticipate in the study. The study was carried out according to the Declaration ofHelsinki and was approved by the Local Ethics Committee (Project number:CE2399).

3. Testing procedures

All diagnostic and neurological evaluations were performed by aspecialist in movement disorders. Clinical and demographic vari-ables were recorded: gender, age, Unified Parkinson's DiseaseRating Scale part III (UPDRS III), Pull Test score (UPDRS subitem),H&Y stage, quality of life ([PDQ8] Parkinson's DiseaseQuestionnaire-8), and a history of falls defined as a sudden event inwhich the patient came to rest on the ground or another lower levelin the previous month [13].

Further informationwas gathered for the PD and PS patients: PDduration, dominant PD phenotype according to a data-drivenapproach (younger onset, tremor dominant, non-tremor domi-nant, and rapid disease progression) [14], the more affected side,direction and degree of leaning at the thoracolumbar level asmeasured with a pocket compass needle goniometer (IncliMed®,University of Padua) [15], center of pressure (CoP) displacement,pattern of PS onset, PS duration, pharmacological therapy, degree ofanterior trunk flexion, low back pain ([VAS] visual analog scale1e10), the presence of metronome, and awareness of PS. A raterexperienced in “movement disorder” rehabilitation and unaware ofthe study objectives performed the balance and gait evaluations.

The primary outcome measure was the mean velocity of CoPdisplacement in the mediolateral and the anteroposterior (VEL_-MED_ML/AP) (mm/s) direction assessed by stabilometry (SA) toquantify postural instability [16]. The secondary outcomemeasureswere other stabilometric parameters, Berg Balance Scale (BBS),Sensory Organization Balance Test (SOT), and gait analysis.

SA was performed in the standing position on an electronicmonoaxial platform (Technobody©). The feet position on the plat-formwas standardized using a V-shaped frame for all patients. Thedistance between the two malleoli was 3 cm and the medial bor-ders of the feet were extrarotated 12� with respect to the ante-roposterior axis. The patients were evaluated while standingupright without the use of upper limb support in the eyes open(EO) and the eyes closed (EC) condition, each lasting 30 s [17,18]. Atrunk sensor placed at T12 was used for qualitativemeasurement oftrunk position in relation to CoP displacement. The secondaryoutcomes were: CoP displacement in the mediolateral and theanteroposterior direction and related standard deviations (mm)(DEV_ST_ML/AP); length of CoP trajectory (mm); and sway area(mm2).

The BBS is a 14-item validated and reliable test (test-retestICC ¼ 0.80,intrarater ICC ¼ 0,80; accuracy ¼ 0.79). It is the mostcommonly used scale that tests balance disorders during sitting,standing and positional changes in PD [19]. The SOT is a validatedbalance test that evaluates central integration deficit of sensoryinputs in patients with neurologic impairment [20]. The patientstands barefoot with arms alongside the body and feet in a heel-to-heel position and maintains standing balance under six differentsensory conditions [20]. Two sensory conditions were tested: firmand compliant surface with EO, EC and dome for each condition. Astopwatch was used to record the amount of time a patient main-tained upright posture without activating postural reactions. Five30-s trials were carried out for each condition [20].

Please cite this article in press as: C. Geroin, et al., Does the Pisa syndromedisease? An observational cross-sectional study, Parkinsonism and Relate

Spatiotemporal gait parameters were evaluated using the GAI-TRite walkway system (CIR Systems Inc, Havertown, PA). The pa-tients were not allowed to use walking aids. The followingparameters were measured: gait speed (cm/s); cadence (step/min);step length (cm); stride length (cm); width of base of support (cm)(BoS); swing phase (% of cycle); stance phase (% of cycle); singlesupport time (% of cycle); and double support time (% of cycle). Thedata from three trials were collected and their average was calcu-lated [21].

4. Statistical analysis

Descriptive statistics included frequency tables and calculationof means and standard deviation. Normal distribution waschecked using the ShapiroeWilk test. Parametric or non-parametric test were applied accordingly. Primary outcome mea-sure and DEV_ST_AP and ML were analyzed with two-way analysisof variance (ANOVA) with between factor “Groups” (HS, PD, PS) asindependent variables and within factor “velocity” (AP and MLdirection) and “visual conditions” (EO/EC). The remaining outcomemeasures were analyzed using one-way ANOVA, except for age,PDQ8, X and Y CoP in both conditions (EO and EC), the length ofCoP in the EC condition, gait speed, cadence, swing and stancephase, single and double support time which were analyzed withnon-parametric tests (KruskaleWallis and ManneWhitney test forpost-hoc analysis). Linear regression was performed to determinewhether the X CoP mean displacement (dependent variable) canbe predicted from lateral trunk bending (independent variable).P < .05 was set as a significant value for the first level of analysis.Post-hoc comparisons were performed using Tukey's multiplecomparison test to evaluate whether there was any differenceamong the three groups after adjusting for multiple testing. Sta-tistical analyses were carried out using the IBM® SPSS® Statisticsversion 16.0 for Macintosh.

5. Results

All patients were receiving chronic therapywith a dopaminergicdrug and showed good motor compensation in appendicularfunction. None had psychiatric disturbances. There were no sig-nificant differences among the groups in age, UPDRS III motorsubscore, Pull Test, H&Y stage, PDQ8, or the number of falls in theprevious month (Table 1).

In the PD and PS patients, the average duration of the defor-mity was 2.63 ± 2.6 years (range 4 monthse8 years) and wascharacterized by lateral flexion of the trunk towards the right(30.4�±16.5 on average) in five patients and towards the left side(20.40�±8.4 on average) in the other five patients. Anterior trunkflexion of various severity was present in eight out of 10 PD andPS patients (14.6�±13.5, range 8e46). Camptocormia was presentin only one patient (case no. 6). The direction of lateral trunkflexion was contralateral to the side of clinical signs of PD onset insix of the ten PS patients with identifiable motor asymmetry atonset.

5.1. Primary outcome measure

A significant main effect in the VEL_MED (F ¼ 4.873; df ¼ 2,p ¼ .016) was found. Post-hoc comparison revealed a significantdifference between the patients with PS and the healthy controls(HS) (p ¼ .018) where worse performance was noted in the PSpatients. There was a significant interaction among the groupsbetween the eyes condition and the velocity of CoP (p ¼ .007). Inthe EO condition, there was a significant main effect in the velocityof CoP displacement (F ¼ 5.437; df ¼ 2, p ¼ .01). Post-hoc

affect postural control, balance, and gait in patients with Parkinson'sd Disorders (2015), http://dx.doi.org/10.1016/j.parkreldis.2015.04.020

C. Geroin et al. / Parkinsonism and Related Disorders xxx (2015) 1e64

comparison revealed a significant difference between the PS pa-tients and the HS (p ¼ .012) and between the PS and the PD pa-tients (P ¼ .048) where worse performance was observed in the PSpatients. No significant “velocity x group” interaction was found.

In the EC condition, there was a significant main effect in theVEL_MED (F ¼ 3.650; df ¼ 2, p ¼ .04). Post-hoc comparisonrevealed a significant difference between the PS patients and theHS (p ¼ .044) where worse performance was noted in the PS pa-tients. No significant “velocity x group” interaction was found(Table 3).

5.2. Secondary outcome measures

There was no significant main effect in the DEV_ST_AP/ML.There was a significant “eyes condition x DEV_ST” interaction(p ¼ .008). No significant main effects were found in either the ECor the EO condition. No significant differences were found for theX CoP, Y CoP, and sway area in either the EO or the EC conditionand length of CoP in the EC condition. A significant main effectwas found only for length CoP in the EO condition (F ¼ 9.108;df ¼ 1, p ¼ .005). There were no statistically significant effects inBBS, the SOT scores or spatiotemporal gait parameters. Lateraltrunk bending did not predict the X CoP mean displacement(R ¼ 0.033; p ¼ .92) (Table 3).

Table 3Multiple pair-wise comparisons between the three groups for each outcome measure.

PS N ¼ 10

Stabilometric assessmentPrimary outcome measuresEo VEL_MED_AP (mm/sec) 7.3 (3.80

VEL_MED_ML (mm/sec) 5.50 (2.55Ec VEL_MED_AP (mm/sec) 11 (5.52

VEL_MED_ML (mm/sec) 8.20 (4.39Secondary outcome measuresEo X CoP mean (mm) �6.90 (7.72

Y CoP mean (mm) �0.5 (11.5DEV_ST_AP (mm) 2.6 (1.51DEV_ST_ML (mm) 2.6 (1.58Length CoP (mm) 254.4 (127Sway area (mm2) 139.8 (163

Ec X CoP mean (mm) �6.70 (8.56Y CoP mean (mm) .50 (15.2DEV_ST_AP (mm) 3.00 (1.41DEV_ST_ML (mm) 3.60 (2.17Length CoP (mm) 378.20 (188Sway area (mm2) 221.10 (247

BBS 47.90 (6.95Sensory organization balance test (0e150) (s)Firm surface eo 148.20 (5.69

ec 142.20 (19.1dome 145.90 (12.9

Compliant surface eo 135.30 (39.3ec 146.80 (10.1dome 135.10 (39.4

Spatio-temporal gait parametersGait speed (cm/s) 95.30 (32.9Cadence (step/min) 112.70 (16.4Step length (cm) 50.60 (15.1Stride length (cm) 101.50 (30.2BoS (cm) 10.70 (6.29Swing (% of gait cycle) 37.10 (3.07Stance (% of gait cycle) 62.90 (3.07Single support time (% of gait cycle) 37.10 (3.07Double support time (% of gait cycle) 26.10 (5.99

Abbreviations: PS, denotes patients with Parkinson's disease and Pisa Syndrome; PD, patstandard deviation; Eo, eyes open; Ec, eyes closed; Dome, dome condition; CoP, center of pvelocity of mediolateral CoP displacement; X CoP Mean, CoP displacement in mediolaterastandard deviation of CoP anteroposterior direction; DEV_ST_ML, standard deviation oftwo-way analysis of variance ANOVA (3 � 2 � 2); P valueb, one-way ANOVA; and P valu

Please cite this article in press as: C. Geroin, et al., Does the Pisa syndromedisease? An observational cross-sectional study, Parkinsonism and Relate

6. Discussion

The main finding is that the patients with PD and PS showed asignificantly greater velocity of CoP displacement in the ante-roposterior and mediolateral directions than either the PD patientswithout PS or the age-matched controls. As hypothesized, lateralbending of the trunk, irrespective of side and severity, increasedpostural instability in the patients with PD and PS during the staticbalance tasks. In contrast, it did not seem to affect gait performance,as measured on the dynamic balance task. Our data confirm pre-vious findings that, as compared with normal subjects, posturalsway velocity and frequency are increased in PD patients, with aworse performance in the EC condition [7]. Moreover, our resultsfurther suggest that postural stability, as assessed by stabilometry,may be more impaired in PD patients with lateral trunk flexion.

Posture control involves many underlying physiological systemsthat can be affected by pathology or subclinical constraints [22].Damage to any of these systems will result in different, context-specific instabilities [22]. It is well established that the velocity ofCoP displacement reflects the postural strategies implemented tomaintain uprightness [7,23]. In normal conditions, three mainpostural strategies are distinguished [7,24]. The ankle strategyconsists of movements at the ankle that involve the distal muscles(e.g., the tibialis anterior in response to backward tilt and thegastrocnemius in response to forward tilt). This strategy is usually

PD N ¼ 10 HS N ¼ 10 P value main effect

) 4.4 (1.51) 3.8 (1.81) .01*a

) 3.5 (1.51) 2.9 (1.37)) 7.20 (4.10) 6.50 (3.06) .04*a

) 5.30 (2.21) 4.5 (2.27)

) 1.00 (4.50) .80 (4.80) .29c

7) 5.6 (12.44) 6.3 (7.07) .39c

) 2.7 (1.16) 2.3 (0.82) .31a

) 1.9 (1.10) 1.4 (0.52).52) 159.6 (56.66) 136.4 (58.81) .005*b

.26) 98 (95.21) 47.6 (22.40) .07b

) 3 (5.94) 1.3 (3.74) .06c

8) 5.80 (12.66) 7.80 (7.80) .67c

) 3.10 (1.29) 2.50 (1.18) .29a

) 2.8 (1.03) 2.20 (1.03).33) 248.70 (118.77) 218.30 (96.28) .06c

.49) 169.60 (132.24) 100.9 (80.16) .12b

) 46.5 (9.51) e 0.71a

) 150 (0) 150 (0) .23b

9) 150 (0) 150 (0) .13b

7) 150 (0) 150 (0) .23b

7) 150 (0) 150 (0) .16b

2) 150 (0) 150 (0) .23b

1) 147.60 (7.59) 150 (0) .16b

1) 96.60 (32.36) 102.50 (14.10) .67c

3) 116.50 (12.95) 106.50 (10.19) .06c

6) 49.30 (15.65) 57.70 (5.14) .23b

9) 99.10 (31.17) 115.70 (10.37) .23b

) 8.70 (3.27) 9.20 (2.57) .45b

) 34.90 (3.81) 37.60 (1.65) .39c

) 65.10 (3.81) 62.40 (1.65) .39c

) 34.90 (3.81) 37.60 (1.65) .39c

) 30.30 (7.47) 25.20 (3.26) .67c

ients with Parkinson's Disease (without PS); HS, aged-matched healthy controls; SD,ressure; VEL_MED_AP, velocity of anteroposterior CoP displacement; VEL_MED_ML,l direction; Y CoP Mean, CoP displacement in anteroposterior direction; DEV_ST_AP,CoP mediolateral direction; BBS, Berg Balance Scale; BoS, base of support; P valuea,ec, Kruskall Wallis Test; p significant if < .05. * and in bold if statistically significant.

affect postural control, balance, and gait in patients with Parkinson'sd Disorders (2015), http://dx.doi.org/10.1016/j.parkreldis.2015.04.020

C. Geroin et al. / Parkinsonism and Related Disorders xxx (2015) 1e6 5

employed to counteract small perturbations. The hip strategyconsists of quick movements at the hip involving also the upperlimb to maintain uprightness in more precarious conditions.Finally, the stepping strategy keeps the CoP within the base ofsupport by taking a step during larger perturbations [25]. It hasbeen shown that in the quiet stance the increase in mediolateralsway may reflect decreasing postural control at the hip, whereascontrol in the anteroposterior direction is primarily related toactivation of the ankle strategy [16]. Our data showed that the CoPdisplacement in both directions was increased in the patients withPS, denoting an altered use of ankle and hip strategy to maintainthe CoP within the base of support. It has been reported that, whenankle and hip joint movement occur simultaneously, the use ofmultijoint postural strategies [24] is already impaired in PD pa-tients without PS (particularly during switching between strate-gies) [26] and that it may be further compromised by lateral trunkflexion in patients with PD and PS.

There was no statistically significant correlation betweenthe CoP displacement and the side and severity of lateral

Fig. 1. Stabilometric assessment in patients with Parkinson's disease and the Pisa Syndromsigrams of CoP displacement in the patients with PD and PS (A), in those with PD but withotrunk flexion. The position of the trunk with respect to the base of support is shown at thewithin the base of support and prevent falls.

Please cite this article in press as: C. Geroin, et al., Does the Pisa syndromedisease? An observational cross-sectional study, Parkinsonism and Relate

trunk bending. Shifting of CoP was contralateral to thedirection of trunk bending in five patients with PD and PS andipsilateral in the others (Table 2). This may have been due to thedifferent postural strategies the PS patients deployed (Fig. 1,Table 2).

It is unlikely that the severity of PS may be more associated withone postural strategy than with another. Controlateral and ipsilat-eral displacement of CoPwas observed in two patients with 50� and42� of PS (Nos. 2 and 6 respectively) (Table 2). From a biome-chanical point of view, this suggests that postural misalignmentinvolves the trunk as well as the hips and ankles in patients with PDand PS. As illustrated in Fig. 1, different strategies can be observed.The patient with PD without PS (Fig. 1B) showed shifting of CoP tothe heel of the left foot, which means that postural control deficitsare already related to the disease itself, particularly in the medio-lateral direction [7]. A patient with PD and PS (Fig. 1A) showedcontralateral displacement of CoP to the left on the toes and ante-rior to the left foot, suggesting postural instability in both themediolateral and the anteroposterior direction.

e, without the Pisa Syndrome, and age-matched controls. Legend: Pictures and kine-ut PS (B), and the controls (C). A PD patient with PD and PS with 50 degrees of lateralbottom. Different postural strategies are employed by the patients to maintain the CoP

affect postural control, balance, and gait in patients with Parkinson'sd Disorders (2015), http://dx.doi.org/10.1016/j.parkreldis.2015.04.020

C. Geroin et al. / Parkinsonism and Related Disorders xxx (2015) 1e66

Somatosensory integration deficits have been reported tocontribute to the development of balance disorders in patientswith PD but without PS [9]. The SOT scores did not show consistentdifferences, as reported in the stabilometric assessment. A possibleexplanation is that the SOT was not sufficiently sensitive to mea-sure differences in our study sample. That said, the SOT is the besttest to evaluate sensorimotor integration deficits [20], however,the stabilometric assessment performed in the EC condition sim-ulates one of the SOT tasks (stance EC condition). Our data suggestthat the patients with PD and PS relied on visual information morethan the healthy subjects. Future studies will need to include SOTcombined with stabilometric assessment as outcome measures toproperly evaluate somatosensory integration deficits in patientswith PD and PS.

As regards the contribution of proprioceptive cues from sensorysources, such asmuscle, skin and joints, the patients with PD and PShad more difficulty in properly controlling their postural orienta-tion (Fig. 1A, trunk sensor) than the PD patients without PS and thecontrols.

It is widely recognized that biomechanical constraint can alsoaffect balance [22]. Biomechanical constraint usually refers toproblems at the base of support (feet); however, we cannot rule outthat biomechanical constraints at another level such as the trunkmay contribute to postural instability. One of the hypothesizedetiologies of PS is that it is caused by a primary alteration of themusculoskeletal system (peripheral mechanisms) [3]. The existingliterature on PD and PS patients is based on magnetic resonancefindings of muscle atrophy and fatty degeneration not only ipsi-lateral but also contralateral to the trunk leaning side, which maylead to muscle imbalance and weakness and result in compensa-tory posture [2,4]. Although gait disturbances are frequentlyobserved in PD patients and are usually consistent with imbalance[9,27], we found no significant difference in gait parameters. Twopossible explanations may be put forward: the small sample sizeand a different physiological mechanism between maintainingbalance during static and dynamic tasks [27].

This study has several limitations. The small sample size lacksstatistical power. Because the SOT was performed without astabilometric force platform, it was potentially less sensitive todetecting differences consistent with stabilometric assessment.Muscle strength was not evaluated and x-ray of the spine wasnot performed. Finally, the BBS, although a dynamic integrationtest, did not turn out to be sensitive enough to determinedifferences.

In summary, the results of this pilot observational cross-sectional study suggest that, as compared with the healthy, age-matched controls, trunk side bending in patients with PD and PSis associated with significantly greater body sway velocity in theanteroposterior and medial lateral directions to maintain posturaluprightness. As compared with the age-matched controls and thePD patients, the patients with PD and PS appeared to be fairly ableto maintain their ability to walk. Rehabilitation programs shouldinclude spine alignment and postural righting strategies to main-tain balance, preserve independence, and minimize the risk offalling.

Financial disclosure/conflict of interest

The authors received no financial support for the research orauthorship of this article.

No commercial party having a direct financial interest in theresults of the research supporting this manuscript has or will confera benefit on the authors or on any organization with which theauthors are associated.

Please cite this article in press as: C. Geroin, et al., Does the Pisa syndromedisease? An observational cross-sectional study, Parkinsonism and Relate

Acknowledgments

The authors thank M. Veronese for assistance in figure editing.

References

[1] L. Bonanni, A. Thomas, S. Varanese, V. Scorrano, M. Onofrj, Botulinum toxintreatment of lateral axial dystonia in Parkinsonism, Mov. Disord. 22 (2007)2097e2103.

[2] K.M. Doherty, B.P. van de Warrenburg, M.C. Peralta, L. Silveira-Moriyama,J.P. Azulay, O.S. Gershanik, et al., Postural deformities in Parkinson's disease,Lancet Neurol. 10 (2011) 538e549.

[3] A. Castrioto, C. Piscicelli, D. P�erennou, P. Krack, B. Debû, The pathogenesis ofPisa syndrome in Parkinson's disease, Mov. Disord. 29 (2014) 1100e1107.

[4] M. Tinazzi, I. Juergenson, G. Squintani, G. Vattemi, S. Montemezzi, D. Censi, etal., Pisa syndrome in Parkinson's disease: an electrophysiological and imagingstudy, J. Neurol. 260 (2013) 2138e2148.

[5] A. Di Matteo, A. Fasano, G. Squintani, L. Ricciardi, T. Bovi, A. Fiaschi, et al.,Lateral trunk flexion in Parkinson's disease: EMG features disclose twodifferent underlying pathophysiological mechanisms, J. Neurol. 258 (2011)740e745.

[6] C. Tassorelli, A. Furnari, S. Buscone, E. Alfonsi, C. Pacchetti, R. Zangaglia, et al.,Pisa syndrome in Parkinson's disease: clinical, electromyographic, andradiological characterization, Mov. Disord. 27 (2012) 227e235.

[7] B. Schoneburg, M. Mancini, F. Horak, J.G. Nutt, Framework for understandingbalance dysfunction in Parkinson's disease, Mov. Disord. 28 (2013) 1474e1482.

[8] F.B. Horak, J.M. Macpherson, Postural orientation and equilibrium: interactionand coordination, in: Handbook of Physiology, Oxford University Press, NewYork, 1996, pp. 255e292.

[9] N. Smania, E. Corato, M. Tinazzi, C. Stanzani, A. Fiaschi, P. Girardi, et al., Effectof balance training on postural instability in patients with idiopathic Parkin-son's disease, Neurorehabil. Neural. Repair. 24 (2010) 826e834.

[10] M. Vaugoyeau, S. Viel, C. Assaiante, et al., Impaired vertical postural controland proprioceptive integration deficits in Parkinson's disease, Neuroscience146 (2007) 852e863.

[11] M.G. Carpenter, B.R. Bloem, Postural control in Parkinson patients: a propri-oceptive problem? Exp. Neurol. 227 (2011) 26e30.

[12] D.J. Gelb, E. Oliver, S. Gilman, Diagnostic criteria for Parkinson disease, Arch.Neurol. 56 (1999) 33e39.

[13] Kellogg, The prevention of falls in later life. A report of the Kellogg Interna-tional Work Group on the prevention of falls by the elderly, Dan. Med. Bull. 4(1987) 1e24.

[14] S.J. Lewis, T. Foltynie, A.D. Blackwell, T.W. Robbins, A.M. Owen, R.A. Barker,Heterogeneity of Parkinson's disease in the early clinical stages using a datadriven approach, J. Neurol. Neurosurg. Psychiatr. 76 (2005) 343e348.

[15] A.R. Gravina, C. Ferraro, A. Frizziero, M. Ferraro, S. Masiero, Goniometerevaluation of thoracic kyphosis and lumbar lordosis in subjects during growthage: a validity study, Stud. Health Technol. Inform. 176 (2012) 247e251.

[16] A.P. Stylianou, M.A. McVey, K.E. Lyons, R. Pahwa, C.W. Luchies, Postural swayinpatients with mild to moderate Parkinson's disease, Int. J. Neurosci. 121(2011) 614e621.

[17] D. Cattaneo, J. Jonsdottir, Sensory impairments in quiet standing in subjectswith multiple sclerosis, Mult. Scler. 15 (2009) 59e67.

[18] G.W. Ickenstein, H. Ambach, A. Kl€oditz, H. Koch, S. Isenmann, H. Reichmann, etal., Static posturography in aging and Parkinson's disease, Front Aging Neu-rosci. 4 (2012) 20.

[19] A.L. Leddy, B.E. Crowner, G.M. Earhart, Functional gait assessment and balanceevaluation system test: reliability, validity, sensitivity, and specificity foridentifying individuals with Parkinson disease who fall, Phys. Ther. 91 (2011)102e113.

[20] A. Shumway-Cook, F.B. Horak, Assessing the influence of sensory interactionof balance. Suggestion from the field, Phys. Ther. (1986) 1548e1550.

[21] H.B. Menz, M.D. Latt, A. Tiedemann, M. Mun San Kwan, S.R. Lord, Reliability ofthe GAITRite walkway system for the quantification of temporo-spatial pa-rameters of gait in young and older people, Gait Posture (2004) 20e25.

[22] F.B. Horak, Postural orientation and equilibrium: what do we need to knowabout neural control of balance to prevent falls? Age Ageing (Suppl. 2) (2006)ii7eii11.

[23] A. Frenklach, S. Louie, M.M. Koop, H. Bronte-Stewart, Excessive postural swayand the risk of falls at different stages of Parkinson's disease, Mov. Disord. 24(2009) 377e385.

[24] F.B. Horak, L.M. Nashner, Central programming of postural movements:adaptation to altered support-surface configurations, J. Neurophysiol. 55(1986) 1369e1381.

[25] B.E. Maki, W.E. McIlroy, The role of limb movements in maintaining uprightstance: the “change-in-support” strategy, Phys. Ther. 77 (1997) 488e507.

[26] F.B. Horak, J. Frank, J. Nutt, Effects of dopamine on postural control in parkin-sonian subjects: scaling, set, and tone, J. Neurophysiol. 75 (1996) 2380e2396.

[27] M. Gandolfi, C. Geroin, A. Picelli, D. Munari, A. Waldner, S. Tamburin, et al.,Robot-assisted vs. sensory integration training in treating gait and balancedysfunctions in patients with multiple sclerosis: a randomized controlledtrial, Front Hum. Neurosci. 8 (2014) 318.

affect postural control, balance, and gait in patients with Parkinson'sd Disorders (2015), http://dx.doi.org/10.1016/j.parkreldis.2015.04.020