Journal of the Neurological Sciences 316 (2012) 79–85

Contents lists available at SciVerse ScienceDirect

Journal of the Neurological Sciences

j ourna l homepage: www.e lsev ie r .com/ locate / jns

Uses of the postural stability test for differential diagnosis of hereditary ataxias

J. Schwabova a,⁎, F. Zahalka c, V. Komarek b, T. Maly c, P. Hrasky c, T. Gryc c, O. Cakrt d, A. Zumrova b

a Department of Neurology, Charles University in Prague, 2nd Faculty of Medicine and the Motol University Hospital, V Úvalu 84, Prague 5, 150 06, Czech republicb Department of Paediatric Neurology, Charles University in Prague, 2nd Faculty of Medicine and the Motol University Hospital, V Úvalu 84, Prague 5, 150 06, Czech republicc Research Sport Center, Charles University in Prague, Faculty of Physical Education and Sport, Jose Martího 31, Prague 6, 162 52, Czech republicd Department of Rehabilitation and Sport Medicine, Charles University in Prague, 2nd Faculty of Medicine and the Motol University Hospital, V Úvalu 84, Prague 5, 150 06, Czech republic

⁎ Corresponding author. Tel.: +420 224 436 801; faxE-mail address: schwabova@gmail.com (J. Schwabov

0022-510X/$ – see front matter © 2012 Elsevier B.V. Aldoi:10.1016/j.jns.2012.01.022

a b s t r a c t

a r t i c l e i n f o

Article history:Received 19 June 2011Received in revised form 16 January 2012Accepted 24 January 2012Available online 13 February 2012

Keywords:AtaxiaSensory ataxiaCerebellar ataxiaCoordination impairmentSpinocerebellar ataxia type 2Friedreich ataxiaDifferential diagnosisCentre of pressure

Friedreich's ataxia (FRDA) and spinocerebellar ataxia type 2 (SCA 2) are among the most commonly diagnosedhereditary ataxias in Czech Republic. Although criteria differentiate the ataxias, disorder onset symptoms maybe similar.Our goal was to determine whether and to what degree of validity posturographic examination may be utilized,with the aim of differential diagnosis; which specific posturographic parametres are suitable for differential diag-nosis; and which differences in FRDA and SCA 2 patient posturographic findings may be established.17 SCA 2 and 12 FRDA patients were examined with ten healthy controls. A multi-sensor tenzometric platformwas used for posturographic examination. Toe standing position was added to basic tests, including standing po-sition with and without visual control.There was no difference between patients in standing position with visual control but there were distinct dif-ferences between FRDA and SCA 2 patients, based on upright stance without visual control and medio-lateraldeviation.There were no differences between patients in toe standing position, suggesting not only the cerebellum, butalso deep sensation, helps to create the so-called adaptive controller.Posturography is attested to as a useful method for differential diagnosis of hereditary ataxias and provides neu-rophysiological findings in cerebellar and sensoric ataxias.

© 2012 Elsevier B.V. All rights reserved.

1. Introduction

From a clinical and neurological perspective, ataxia is a syndromeappearing as a result of cerebellar affliction and/or its centripetal orcentrifugal paths of all varieties of etiology. One of the causes of themanifestation of ataxia may be a group of hereditary ataxias with alltypes of inheritance. Friedreich ataxia (FRDA) and autosomal domi-nant spinocerebellar ataxia type 2 (SCA 2) are among the most com-monly diagnosed hereditary ataxias in the Czech Republic [1,2]. FRDAwas long thought to be a disorder particular to childhood age butgreater data specification, resulting from the development of DNAanalysis, now shows that FRDA belongs even among differential diag-nosis of ataxias with onsets in adulthood [1].

In both FRDA and SCA 2 there is a progressive atrophy of the cer-ebellum as well as spinocerebellar and dorsal column paths [3]. In thebegining, the clinical portrait, may be very similar. Genealogical datadoes not necessarily assist in the aims of DNA diagnosis.

One of the first symptoms in patients tends to be the loss of stabil-ity during regular day-to-day activities, both in static situations such

: +420 224 436 820.a).

l rights reserved.

as standing and, especially, in dynamic activities such as stair-climbing [4,5]. Neurological findings, as with neuroimaging methodfindings, may be weak at this stage. It is possible to differentiate be-tween nosological units by means of typical clinical features such asRomberg sign, limb ataxia, truncal ataxia, spasticity and plantar re-flex, typical among FRDA patients [6] and SCA 2 patients with cerebel-lar ataxia and slow eye movements [5]. MRI is also helpful indifferential diagnosis: ponto-cerebellar atrophy may be found onthe MRI among SCA 2 patients, while spinal atrophy without cerebel-lar atrophy is typical among FRDA patients.

For quantification and the possibility of comparing clinical find-ings, a range of clinical testing scales were developed. Such scalesare unfortunately to some extent subjectively influenced. The mostcommonly used scales are the International Cooperative Ataxia Rat-ing Scale (ICARS) [7] and the newer SARA scale (Scale for the assess-ment and rating of ataxia) [8] Test results are in the later stages of thedisorder mostly in accordance with morphological findings. The clin-ical portrait of SCA 2 patients is framed namely by cerebellar ataxia.Such is not the case with FRDA as it is mostly a sensoric ataxia. Resultsin differentiation are not overly positive or beneficial in the earlystages of the disorder [3].

Among electrophysical methods, posturography has been usedmore and more in recent years. Posturography helps to eliminate

80 J. Schwabova et al. / Journal of the Neurological Sciences 316 (2012) 79–85

the subjectivity of examiner and examined. On the basis of posturo-graphic examination of ataxia patients, greater COP (centre of pres-sure) deviation with a typical cycle frequency and an inability toadequately react to deviation in position, with a tendency to over-correct movement, was demonstrated [9,10]. The importance of theRomberg's test was confirmed [11].

On the basis of basic posturographic examination, sensoric ataxiais differentiated on the basis of the Romberg´s test [12]: Uprightstance without visual control. No provocative maneuvres are typicallyapplied for cerebellar symptomology. After having analyzed the situ-ation, we focused on the possibility of expanding the testing pro-gramme by including toe standing position on tiptoes.

From a neurological perspective, upright stance shares three mainsystems: visual appartus, somatosensor and vestibular system. Thecerebellum is thought to be critical for postural coordination andprobably plays several different roles in control of posture involvingsensorimotor integration [13]. Lesions in different regions of the cer-ebellum produce very different effects on postural control [14]. Forexamaple lesions of the lateral hemispheres can produce profounddisorders of timing for arm and hand coordination without significanteffects on posture and gait [15]. Toe standing position on tiptoes is ap-parently a common motor activity, during which relatively greatchanges occur in the centre of balance [16,17]. It is a basic motorskill whose consequence determines the quality of the motion carriedout [18]. An important role in the coordination of balanced reaction isplayed by the cerebellum.

The lower limbs – namely the feet in the frontal foot and heelareas – are used in upright stance. The centre of pressure workingunder each foot is found closer to the heel area and in ideal cases, asymmetrical burden is to be expected.

When in toe standing position, pressure on the heel area is less-ened and the burden is gradually transferred to the frontal footarea. In that moment, the heel is no longer in contact with the surfaceunderfoot.

Medio-lateral deviation of the centre of pressure on a flat surfacetypically changes position towards the frontal foot. As in every freemovement, the combination of opposing forces – agonist and antago-nist – is important as regards muscle engagement in the toe standingposition. The basic consideration is the assumption that the combina-tion of agonist and antagonist is manifest in the change of pressure onthe flat surface, from the heel area of the foot to the frontal foot. Inco-ordination is typical of patients with cerebellar lesion [13] and isidentified by means of irregular changes in the centre of pressure[19]. Since the action has its own clear beginning and end, it is calleda closed-loop motor skill. From the perspective of motor learning,such a skill is expected to be stable and automatic in healthy individ-uals [20]. It is, however, a difficult action for patients with stabilityand coordination problems [21].

There exists currently a wide range of work comparing clinicaltests with postural stability parameters [22–25]. It has been demon-strated that, according to selected parameters, it is possible to clearlydifferentiate healthy and unhealthy individuals, although inter-comparison does not lead to clear conclusions [24,25]. One causemay be that provocative maneuvers in changes of position and thebodily movement of patients are a component part of clinical scalesand in some works not a part of posturographic examination.

Measuring dynamic body responses during simple motor maneu-vers creates objective evidence about changes in time and space andbodily reactions to internal and external subjects. Initiated subjectsare in most cases expected to arise from bodily movements carriedout during regular day-to-day activities, since they are expected to in-crease the risk of loss of stability and, alongside loss of stability, in-crease the related risk of falling [4]. Several provocative maneuverslower standing position stability. Among these is loss of eye controlduring measurement. It is possible in the same way to increase testdifficulty – even when performing motor maneuvers such as toe

standing position on tiptoes while standing, raising arms, or a combi-nation of both movements [16,17,26].

The goal of our study was to test a) whether and to what extentposturographic examination is a helpful and useful procedure in thedifferential diagnosis of hereditary ataxias prior to DNA analysis;aa) and to test the value of toe standing position as a suitableand useful maneuver in the testing of ataxia patients; and aaa) todetermine differences in postural stability parameters in patientswith a preponderance of cerebellar rather than sensoric elementsof ataxia.

2. Methods

2.1. Study sample

17 SCA 2 patients and 12 FRDA patients (Table 1) were tested withten health controls. The disorder was verified on a molecular level.Patients chosen were able to stand, without the need of support,with or without visual control. Clinical findings were evaluated bymeans of the ICARS and SARA scales. No further criteria was broughtto bear in order to best simulate the regular approach of a physician inpractice. The average age of the SCA 2 patients was 43.8 (18–58) andthe group was comprised of eleven men and six women with an aver-age period of clinical symptoms stretching back 10.8 years (1–29).The average age of the FRDA patients was 31.9 (19–59) and thegroup was comprised of seven men and five women with an averageperiod of clinical symptoms stretching back 10.91 years (5–27). Pa-tients were divided into two groups – ICARSb30=GM (Group ofMild Ataxia) and ICARS>30=GS (Group of Severe Ataxia) – to com-pare postural stability results in relation to disability.

Healthy controls had neither orthopedic nor neurological case-history. Neurological findings were normal. ICARS was 0.

2.2. Assessment of postural stability parameters

The pressure method was used for posturographic examination,measuring pressure on sensors in the FOOTSCAN platform (RSscan,Belgium). The FOOTSCAN is 0.5 m by 0.4 m and has approximately4100 sensors sensitive from one-tenth of N/cm2 and a 500 Hz sensingfrequency. Pressure on individual sensors is measured and the centreof pressure calculated on the contact area, or COP. Standard standingposition with wide base was the selected standard stance. Hip widthdelimited stance, but was measured out afterwards by antropometertransferred to the base, determining instep distance. Transparentsheeting for tracing foot position was placed between feet and theFOOTSCAN platform to ensure individual conformity during repeatedexamination. Stance was measured according to standard practice ata length of 30 s [27]. The following parameters were used to evaluateCOP centre of pressure: medio-lateral directional deviation (Delta X);anterio-posterior directional deviation (Delta Y); the whole course oftotal travelled way (TTW); and standard deviation of COP velocity(VelocitySD).

Parameters were viewed after a one-second time period after ini-tial movement during toe-standing position, without the help ofarms, in order to evaluate pressure changes in toe-standing position.

TTWwas chosen among evaluated parameters, as well as trajecto-ry change of centre of pressure in medio-lateral directional deviation:Delta X from initialization to stabilization in toe-standing position.Size of confident elipse was a further evaluated parameter and wascalculated at intervals from the initialization of movement until thestabilization of toe-standing position (set at 1 s). The confident elipseis determined by the surface, which will in all probability locate theactual position of designated central pressure points. The final evalu-ated parameter was pressure under the right and left lower limb dur-ing initialization of toe-standing position, expressed by co-relation,the Right to Left ratio, or R:L.

Table 1Clinical characteristics of patients.

Subject Sex Disorder Age(years)

Diseaseduration(years)

ICARS SARA ABC scale(%)a

Cerebellarataxia

Sensoryloss

Rombergsign

Pyramidaltract signs

Tendonreflexes

Slow eyemovements

Neuroimaging, atrophy of

1 F SCA 2 47 7 35 16 47.5 Yes No Yes/no No Normal Yes Brain stem and cerebellum2 M SCA 2 51 6 28 10 89.4 Yes No Yes/no No Reduced Yes Brain stem and cerebellum3 F SCA 2 34 9 34 13 75.6 Yes No Yes No Reduced Yes N.A.4 F SCA 2 55 11 30 12.5 40.6 Yes No Yes No Normal Yes N.A.5 M SCA 2 59 14 17 7 64.4 Yes No No No Normal No N.A.6 M SCA 2 57 13 20 7 60 Yes No No No Normal Yes Cerebellum7 M SCA 2 50 11 11 2 89.7 Yes No No No Normal No N.A.8 F SCA 2 48 13 49 21 9.1 Yes Yes Yes No Reduced Yes Cerebellum9 M SCA 2 28 1 3 2 100 Mild No No No Normal No N.A.10 F SCA 2 52 19 8 4 30.3 Mild No No No Normal Yes N.A.11 M SCA 2 58 11 55 26 37.5 Yes No Yes No Normal Yes N.A.12 M SCA 2 40 29 36 19 78.1 Yes No No No Normal No Normal cerebellum13 M SCA 2 38 9 26 10 61.9 Yes No No No Normal No Brain stem and cerebellum14 M SCA 2 18 14 49 18 18.8 Yes No Yes No Normal Yes Brain stem and cerebellum15 F SCA 2 29 1 1 0 99.8 Mild No No No Normal No N.A.16 F SCA 2 46 6 33 16 46.3 Yes No Yes/no No Normal Yes Brain stem and cerebellum17 M SCA 2 35 9 18 5 91.9 Yes No Yes No Normal Yes N.A.18 M FRDA 29 11 66 29 0 Yes Yes Yes Yes Reduced No Cerebellum and spine19 F FRDA 40 13 20 10.5 62.5 Yes Yes Yes Yes Reduced No Normal cerebellum/spine20 M FRDA 31 8 16 13 90.6 Yes No Yes Yes Reduced No Normal cerebellum/spine21 F FRDA 23 9 53 32 43.1 Yes Yes Yes Yes Reduced No Normal cerebellum/spine22 M FRDA 19 5 37 17 49.4 Yes Yes Yes Yes Absent No Normal cerebellum/spine23 F FRDA 41 14 28 12 30.6 Yes No Yes Yes Reduced No Normal cerebellum/spine24 M FRDA 32 9 19 13.5 83.1 Yes No Yes Yes Reduced No Normal cerebellum/spine25 M FRDA 20 4 39 17 48.8 Yes Yes Yes Yes Absent No Normal cerebellum/spine26 F FRDA 22 12 38 17 39.4 Yes Yes Yes Yes Reduced No Normal cerebellum/spine27 M FRDA 37 27 27 13.5 25 Yes Yes Yes Yes Reduced No Periventricular/spine normal28 F FRDA 59 12 21 9 50.6 Yes Yes Yes Yes Reduced No Normal cerebellum/spine29 M FRDA 30 7 23 10 78.1 Yes Yes Yes Yes Reduced No Normal cerebellum/spine

SCA 2: autosomal dominant spinocerebellar ataxia type 2.FRDA: Friedreich ataxia.ICARS: International Cooperative Ataxia Rating Scale.SARA: Scale for the assessment and rating of ataxia.ABC Scale: Activities-specific balance confidence scale.N.A.: no data available.Neuroimaging was not performed for patients with a genetically definitive diagnosis who came from a known family.

a No confidence=0%, optimal confidence =100%.

81J. Schwabova et al. / Journal of the Neurological Sciences 316 (2012) 79–85

2.3. Statistical analysis

Significant differences between observed changes were evaluatedwith the assistance of multiple analysis of variability with fixed effectswith the design of the double multivariate model. Fixed effects in SCA2, FRDA and control groups as well as visual control with open andclosed eyes were independently variable. Such indicators of posturalstability as TTW, DeltaX, DeltaY and VelocitySD were dependentlyvariable. The effects were evaluated as the main factors (effects) ofgroup and visual control as well as the effect of their inter-relatedactivity — interaction between group versus visual control.

A range of comparison was undertaken when there was statisticalsignificance in themain factor bymultiple comparison ofmeans (Bonfer-onni's post-hoc test). Rejection of the null hypothesiswas assessed at thelevel of pb0.05. Effect sizewas assessedusing the “Eta square” coefficient(η2), which explains the proportion of variance of the monitored factor.

Observation of differences of average group postural stability pa-rameters between GM and GS was compared by means of the para-metric t-test (in case of preservation of distribution data on theGauss curve). In case of disruption of assumptions about the normal-ity of dissociation of data the Man–Whitney U test was used. Statisti-cal analysis was performed using IBM SPSS 19.0.

3. Results

3.1. Standing position

A significant effect was determined in the main effect (group) whencomparison was made between groups (F8,66=3.38, pb0.01,

η2=0.290). The univariate ANOVA did not establish significant differ-ence in TTW parameters between observed groups (F2,35=2.92,p>0.05, η2=0.14). A simple analysis of variability demonstrated a sig-nificant effect in Delta X deviation parameters: F2,35=5.92, pb0.01,η2=0.25; DeltaY: F2,35=7.68, pb0.01, η2=0.31 and VelocitySD:F2,35=3.31, pb0.05, η2=0.16), pb0.01. On the basis of a post hoctest we determined a significant difference in Delta X parametres be-tween the FRDA vs. CONTROL group (pb0.05) and SCA 2 vs CONTROL(pb0.05). Similar results were determined in Delta Y parametres be-tween FRDA vs CONTROL (pb0.01) and SCA 2 vs CONTROL (pb0.01).

The between-group effect was more notable in both deviationsfrom the perspective of non-static relevance. In medio-lateral (X) de-viations, the factor accounts for 25,3% of overall evidence variability.In antero-posterior (Y), the factor accounts for 30.5% of overall evi-dence variability.

3.2. Standing position with or without visual control

Mulitiple analyses of variability demonstrated a significant effectof the main effect (visual control) among observed parametres(F4,32=4,55, pb0.01, η2=0.363). A noteworthy visual control effectwas evident among all observed parameters. The univariate ANOVAestablished the significant effect of visual control among all observedpostural stability parameters (TTW: F1,35=12,20, pb0.01, η2=0.26;DeltaX: F1,35=8,82, pb0.01, η2=0.20; DeltaY: F1,35=18,21,pb0.01, η2=0.34; VelocitySD: F1,35=12,97, pb0.01, η2=0.27).

The greatest substantive relevance of the visual control effect wasdetermined in the DeltaY parameter, accounting for 34,2% of overallevidence variability.

Fig. 2. Effect of interaction among observed factors in observed parametres (visual vs.non visual control).

82 J. Schwabova et al. / Journal of the Neurological Sciences 316 (2012) 79–85

Interaction between visual control and individual groups was sig-nificant (F8,66=2,94, pb0.01, η2=0.262). In all observed parame-ters, a significant effect was determined in the interaction ofobserved factors (Group vs. Visual control: Table 1, Figs. 1–4), con-firming diverse changes of observed parameters among chosengroups dependent upon visual control. Deviation in both directionsshowed the greatest value from the perspective of substantive rele-vance. A significantly different interaction effect was determined bypairwairs comparison in the DeltaX parameters of the FRDA groupthrough comparing visual controls with results without visual control(pb0.01). Medio-lateral directional deviation worsened by more thanfive times the original value during loss of visual control, in observa-tion of the FRDA group. The same parameters worsened by 12,6% inthe SCA 2 group, but such a difference is not significant (Table 2).On the basis of further observed parameters it was not possible to dif-ferentiate between the observed group of neurological patients onthe basis of observed postural stability values because they embodiedcomparable changes of value during loss of visual control.

3.3. Toe-standing position

On the basis of multiple analyses of variability, a significant effectwas determined among the main factor (group) (F8,66=2.17, pb0.05,η2=0.209). A significant effect factor on the CI elipsis parameter; theOffset elipsis (pb0.05); and DeltaX (pb0.01) was determined on thebasis of a simple (univariate) analysis of variability. Post-hoc testsdemonstrated significant differences between FRDA vs. CONTROLand SCA 2 vs. CONTROL in DeltaX and Offset E parametres. In casesof confident elipses, a significant difference between SCA 2 andCONTROL groups was determined (Table 3). The inter-group effectwas most distinctive in the DeltaX parametre – accounting for 32.7%of overall evidence variability – from the perspective of substantiverelevance.

3.4. Comparison of postural stability parameters for FRDA and SCA 2groups in relation to degree of disability

During division of the basic group of neurological patients accord-ing to set criteria (ICARSb30=Group of Mild Ataxia-GM andICARS>30=Group of Severe Ataxia-GS), we determined a significantdifference only in medio-lateral deviation; The GM group averageGM=82.9±39.3 mm was significantly lower in comparison withthe GS group average GS=135.0±78.3 mm, t24=−2.28, pb0.05,

Fig. 1. Effect of interaction among observed factors in observed parametres (visual vs.non visual control).

Fig. 5). In further parameters we noted insignificant differences ingroup averages (TTW GM=246.9±95.0 mm vs. TTW GS=274.1±104.7 mm, t24=−0.67, p>0.05, ELIPSA GM=1191.9±1042.6 mm2

vs. ELIPSA GS=2439.9±1794.8 mm2, U=104, p>0.05, OffsetElipsaGM=0.68±0.23 vs. OffsetElipsa GS=0.59±0.24, t24=0.95, p>0.05).

4. Discussion

The study originated in the need to find the most reliable electro-physiological instrument for distinguishing types of ataxia and, indoing so, to facilitate the aims of DNA analysis among patients as-sumed to have hereditary ataxias; and to determine differences inposturographic findings among FRDA and SCA 2 patients with notablefactors of sensoric versus cerebellar ataxia. Since previously existingliterature was less than satisfactory, namely as regards partial correla-tion between clinical testing scales and posturographic examination,we decided to enrich and expand upon routine posturographic

Fig. 3. Effect of interaction among observed factors in observed parametres (visual vs.non visual control).

Fig. 4. Effect of interaction among observed factors in observed parametres (visual vs.non visual control).

83J. Schwabova et al. / Journal of the Neurological Sciences 316 (2012) 79–85

testing by adding toe standing position as a means of testing cerebel-lum coordination abilities.

Results demonstrated that posturographic testing in erect stand-ing position with sight control clearly differentiates the group of pa-tients from healthy controls but does not allow for differentiation ofgroups of patients among themselves.

Romberg maneuvre examination confirmed on a significant statis-tical level a worse result among patients with FRDA.

Upright stance results in anterior-posterior plane were in accor-dance with earlier published works [9,28,29]. In our sampling, signif-icant differences in latero-lateral deviation were determined. One of

Table 2Effect of interaction among observed factors in observed parametres (visual vs. non visual

Visual control

M(SE)

TTW (mm) FRDA 288.92(117.87)‡

SCA 2 623.74(93.19)†

CONTROL 289.68(107.60)DeltaX (mm) FRDA 9.76(4.99)‡

SCA 2 26.06(3.95)CONTROL 5.59(4.56)

DeltaY(mm) FRDA 21.00(4.56)‡

SCA 2 32.78(3.96)†

CONTROL 9.15(4.57)VelocitySD (mm.s-1) FRDA 7.28(2.90)‡

SCA 2 16.14(2.30)†

CONTROL 7.21(2.65)

FRDA: Friedreich's ataxia.SCA 2: spinocerebellar ataxia type 2.TTW: Total travel way of centre of pressure.DeltaX: Centre of pressure range in left-right way (medio-lateral axis).DeltaY: Centre of pressure range in forward-backward way (anterior-posterior axis).VelocitySD: Standard deviation of velocity centre of pressure.M: Mean.SE: Standard error of mean.F: F statistic.Df: Degree of freedom.Eta: Effect size coefficient.⁎ Significant interaction effect (group⁎visual control) at pb0.05.⁎⁎ Significant interaction effect (group⁎visual control) pb0.01.† Mean difference is significant at pb0.05; based on post-hoc comparison.‡ Mean difference is significant at pb0.01; based on post-hoc comparison.

the reasons our observations differed from those of the above-mentioned works may be that a more significant neurological deficitexisted among our observed groups.

It is interesting that when upright stance examination dependsprimarily upon sensoric information, the greater distinctiveness of asingle patient among FRDA patients is not as manifest as expectedamong patophysiological disorders. The most probable explanationfor this is the strong compensatory assistance of an intact vestibularand, namely, visual, analyzer unit. The stage of the disorder may ofcourse play a role, although ICARS scale results do not confirm amore significant difference in neurological deficit among one of thegroups.

The elimination of sight control, the Romberg sign, led to a distinctclarification of the situation. Our results confirmed statistically signif-icant differences both among groups of patients and health controlsas well as between SCA 2 and FRDA.

Although standing position without visual control clearly distin-guished between FRDA and SCA 2 patients at first glance (Figs. 1–4), itis clear after statistical processing that, from the perspective of statisticalsignificance, the only reliable marker for differentiating between thetwo groups is the DeltaX parameter (medio-lateral deviation).

We further anticipated that data obtained would be made moreprecise with the help of maneuvres difficult to coordinate. Such mea-sures have, according to expectation, worse results among patientswith cerebellar atrophy, such as those in the SCA 2 group. Toe-standing position, itself a simple option among such activities, waschosen as a movement maneuver. After statistical processing accord-ing to selected parameters such as CE, DeltaX and R:L ratio, we wereable to clearly differentiate patients from health controls but wereunable to differentiate groups of patients from one another.

It may be assumed that it will be necessary to change the approachto interpretation of results in order to differentiate between the twogroups of patients and seek possibilities to express qualitative ratherthan quantitative differences among the groups, as well as, perhaps,to look for differences with the aid of 3D motion analysis of toe-standing position.

control).

Non visual control Interaction effect

M(SE) F df Eta

1363.24(340.62) 3.75⁎ 2.35 0.1761116.58(269.29)297.79(310.95)51.55(9.96) 6.03⁎⁎ 2.35 0.25629.81(7.88)5.77(9.10)78.97(13.91) 6.55⁎⁎ 2.35 0.27254.20(11.00)9.27(12.70)44.65(10.61) 4.38⁎ 2.35 0.231.68(8.39)7.12(9.68)

Table 3Significant differences among observed parameters in toe standing position.

FRDA M(SE) SCA 2 M(SE) Control M(SE) F df Eta

TTW (mm) 262.40(27.52) 252.50(21.75) 221.92(25.12) 0.681 2.35 0.037CE (mm2) 1636.20(387.75) 1616.25(306.54)b,† 489.67(353.97) 3.51⁎ 2.35 0.167DeltaX (mm) 104.50(16.04)a,† 98.75(12.68)b,‡ 28.08(14.64) 8.51⁎⁎ 2.35 0.327R:L ratio 0.65(0.07)a,† 0.65(0.05)b,† 0.88(0.06) 4.95⁎ 2.35 0.22

FRDA: Friedreich's ataxia.SCA 2: spinocerebellar ataxia type 2.TTW: Total travel way of centre of pressure.DeltaX: Centre of pressure range in left-right way (medio-lateral axis).DeltaY: Centre of pressure range in forward-backward way (anterior-posterior axis).VelocitySD: Standard deviation of velocity centre of pressure.M: Mean.SE: Standard error of mean.F: F statistic.Df: Degrees of freedom.Eta: Effect size coefficient.⁎ Significant main effect (GROUP); based on univariate ANOVA comparison at pb0.05.⁎⁎ Significant main effect (GROUP); based on univariate ANOVA comparison at pb0.01.† Mean difference is significant at pb0.05; based on post-hoc comparison.‡ Mean difference is significant at pb0.01; based on post-hoc comparison.a Significant difference between FRDA vs. control; based on post-hoc comparison.b Significant difference between AD SCA 2 vs. control; based on post-hoc comparison.

84 J. Schwabova et al. / Journal of the Neurological Sciences 316 (2012) 79–85

Toe standing position is a voluntary movement. To perform a vol-untary movement, the brain appears to carry out two types of compu-tation: a) when given a desired change in the proprioceptively orvisually defined sensory state of the limb, and aa) when given aplanned motor command [30]. These sensorimotor maps are collec-tively called 'internal models' of action [30,31].

Recent studies indicate that the cerebellum is fundamentally in-volved in formulating internal reaching models [30]; Inaccurate sim-ulation of reaching attests to ataxic symptoms [32]. Basic musclemotor commands may be generated by feedback control, feedforwardcontrol, or a combination of the two [33].

During toe standing position, the cerebellum attempts to co-ordinate body response with a feedforward control, required for rela-tively fast movement.

Schweighofer et al. suggests that feedback control is important forunskilled movements and unpredictable interactions [33]. When theerror occurs, the cerebellum corrects with a feedback control.

Therefore, cerebellar degeneration would decrease adaptive con-trol of locomotion, for which the cerebellum is necessary [30]. Itseems that deep sensation is a necessary element for the construction

Fig. 5. Comparison of difference in observed group parameters: mild and severe ataxiagroups.

of the adaptive controller, as suggested by loss of deep sensation duringlarge body sway and cerebellar degeneration.

We were further interested in how the mild and severe ataxiagroups would prove different from one another in toe-standing posi-tion. Such differentiation was not possible for individual disorders,due to the small number of patients. The result was interesting as itwas possible to differentiate between groups with the aid of the DeltaXparameter, medio-lateral deviation.

Patients were differentiated from healthy control in upright stancewith visual control; Groups of patients among themselves were dif-ferentiated in upright stance without visual control as well. It appearsthat the parametre may have meaning connected with the gravity ofthe illness. It may perhaps mean that it is a parametre neither typicalnor specific for certain types of ataxia, but a non-specific parametrecommensurate with the level of effort expended to maintain balance.

The importance of using the Romberg maneuvre grows as a result,thanks to which SCA and FRDA groups from the same sampling wereconvincingly differentiated between. Currently published works[24,25] comparing clinical tests with posturographic parametreshave not brought clear-cut results. With a view to the fact that clinicaltests consist of various provocative moments such as narrowing ofbase and closed eyes, it appears that in the future it will be advanta-geous to compare results with postural stability tests, as such testsclearly set out neurological deficits in patients.

Acknowledgments

Supported by GAUK (Grant agency of Charles University in Prague)96909; IGA MZCR (Ministry of Health of the Czech Republic - InternalGrant Agency) 100005-4; GACR P407/11/P784 (Czech Science Founda-tion); MSM 002160864 and VZ FNM MZO 0064203-6505.

References

[1] Bhidayasiri R, Perlman SL, Pults SM, Geschwind DH. Late-onset Friedreich ataxia:phenotypic analysis, magnetic resonance imaging findings, and review of theliterature. Arch Neurol 2005;62(12):1865–9.

[2] Zumrova A. Problems and possibilities in the differential diagnosis of SyndromeSpinocerebellar Ataxia. Neuro Endocrinol Lett 2005;26(2):98–108.

[3] Moseley ML. Incidence of dominant spinocerebellar and Friedreich triplet repeatsamong 361 ataxia families. Neurology 1998;51(6):1666–71.

[4] Globas C, du Montcel ST, Baliko L, Boesch S, Depondt C, Di Donato S, et al. Earlysymptoms in spinocerebellar ataxia type 1, 2, 3, and 6. Mov Disord 2008;23(15):2232–8.

[5] Durr A. Autosomal dominant cerebellar ataxias: polyglutamine expansions andbeyond. Lancet Neurol 2010;9(9):885–94.

85J. Schwabova et al. / Journal of the Neurological Sciences 316 (2012) 79–85

[6] Pandolfo M. Friedreich ataxia: the clinical picture. J Neurol 2009;256(1):3–8.[7] Trouillas P, Takayanagi T, Hallett M, Currier RD, Subramony SH, Wessel K, et al.

International cooperative ataxia rating scale for pharmacological assessment ofthe cerebellar syndrome. The Ataxia Neuropharmacology Committee of the WorldFederation of Neurology. J Neurol Sci 1997;145(2):205–11.

[8] Schmitz-Hubsch T, du Montcel ST, Baliko L, Berciano J, Boesch S, Depondt C, et al.Scale for the assessment and rating of ataxia: development of a new clinical scale.Neurology 2006;66(11):1717–20.

[9] Mauritz KH, Dichgans J, Hufschmidt A. Quantitative analysis of stance in late cor-tical cerebellar atrophy of the anterior lobe and other forms of cerebellar ataxia.Brain 1979;102(3):461–82.

[10] Dichgans J, Mauritz KH, Allum JH, Brandt T. Postural sway in normals and atacticpatients: analysis of the stabilising and destabilizing effects of vision. Agressologie1976;17:15–24.

[11] Jansen EC, Larsen RE, Olesen MB. Quantitative Romberg's test. Measurement andcomputer calculation of postural stability. Acta Neurol Scand 1982;66(1):93–9.

[12] Bastian AJ. Mechanisms of ataxia. Phys Ther 1997;77(6):672–5.[13] Babinski J. De l'asynergie cerebelleuse. Rev Neurol 1899;7:806–16.[14] Horak FB, Macpherson JM. Postural orientation and equlibrium. In: Smith JL, edi-

tor. Handbook of physiology. New York: Oxford University Press; 1996. p. 255–92.[15] Diener HC, Muller A, Thron A, Poremba M, Dichgans J, Rapp H. Correlation of clinical

signs with CT findings in patients with cerebellar disease. J Neurol 1986;233(1):5–12.

[16] Guitton D, Kearney RE, Wereley N, Peterson BW. Visual, vestibular and voluntarycontributions to human head stabilization. Exp Brain Res 1986;64(1):59–69.

[17] Rogers MW, Pai YC. Dynamic transitions in stance support accompanying leg flex-ion movements in man. Exp Brain Res 1990;81(2):398–402.

[18] Choi Y, Qi F, Gordon J, Schweighofer N. Performance-based adaptive schedulesenhance motor learning. J Mot Behav 2008;40(4):273–80.

[19] Ienaga Y, Mitoma H, Kubota K, Morita S, Mizusawa H. Dynamic imbalance in gaitataxia. Characteristics of plantar pressuremeasurements. J Neurol Sci 2006;246(1-2):53–7.

[20] Adams JA. A closed-loop theory of motor learning. J Mot Behav 1971;3(2):111–49.[21] Kawato M, Gomi H. A computational model of four regions of the cerebellum

based on feedback-error learning. Biol Cybern 1992;68(2):95–103.[22] Schwabová J, Zahálka F, Komárek V, Malý T, Hráský P, Gryc T, et al. Validita mezi-

národní škály pro pacienty s ataxií— a scale for the assessment and rating of ataxia.Cesk Slov Neurol N 2010;73(6):68–72.

[23] Schwabová J, Zahálka F, Malý T, Zumrová A, Komárek V, Hráský P, et al. Activitiesof daily living scale— the tool for clinical statemonitoring of spinocerebellar ataxiaand Friedreich ataxia patients. Arch Int J Med 2009;2(4):295–8.

[24] Yabe I, Matsushima M, Soma H, Basri R, Sasaki H. Usefulness of the Scale forAssessment and Rating of Ataxia (SARA). J Neurol Sci 2008;266(1–2):164–6.

[25] Manabe Y, Honda E, Shiro Y, Sakai K, Kohira I, Kashihara K, et al. Fractal dimensionanalysis of static stabilometry in Parkinson's disease and spinocerebellar ataxia.Neurol Res 2001;23(4):397–404.

[26] Crenna P, Frigo C. A motor programme for the initiation of forward-orientedmovements in humans. J Physiol 1991;437:635–53.

[27] Kapteyn TS, Bles W, Njiokiktjien CJ, Kodde L, Massen CH, Mol JM. Standardizationin platform stabilometry being a part of posturography. Agressologie 1983;24(7):321–6.

[28] Van de Warrenburg BP, Bakker M, Kremer BP, Bloem BR, Allum JH. Trunk sway inpatients with spinocerebellar ataxia. Mov Disord 2005;20(8):1006–13.

[29] Diener HC, Dichgans J. Pathophysiology of cerebellar ataxia. Mov Disord 1992;7(2):95–109.

[30] Chen H, Hua SE, Smith MA, Lenz FA, Shadmehr R. Effects of human cerebellar thala-mus disruption on adaptive control of reaching. Cereb Cortex 2006;16(10):1462–73.

[31] Wolpert DM, Ghahramani Z. Computational principles of movement neuroscience.Nat Neurosci 2000;3:1212–7.

[32] Bastian AJ, Martin TA, Keating JG, Thach WT. Cerebellar ataxia: abnormal controlof interaction torques across multiple joints. J Neurophysiol 1996;76(1):492–509.

[33] SchweighoferN, ArbibMA, KawatoM. Role of the cerebellum in reachingmovementsin humans. I. Distributed inverse dynamics control. Eur J Neurosci 1998;10(1):86–94.