PM R 8 (2016) 962-970www.pmrjournal.org

Original Research

Feasibility of Using Tetrax Biofeedback Video Games forBalance Training in Patients With Chronic Hemiplegic Stroke

Jen-Wen Hung, MD, Min-Yuan Yu, OTR, Ku-Chou Chang, MD, Hsuei-Chen Lee, PhD,Yen-Wei Hsieh, OTR, Po-Chih Chen, OTR

Abstract

Background: Decreased weight bearing on the affected lower limb and poor weight shifting are common after a stroke occurs.The Tetrax biofeedback system is a center-of-pressure controlled video game system designed for patients with balance deficits.Although it is a commercial product, information about its clinical use for patients affected by stroke is limited.Objective: To investigate the feasibility and potential efficacy of the Tetrax biofeedback system for balance training in patientswith chronic stroke.Design: Feasibility study.Setting: Rehabilitation department of a medical center.Participants: Participants who had sustained a hemiplegic stroke at least 6 months prior to enrollment but were still able to standindependently for more than 5 minutes.Methods: Participants were randomly assigned to an intervention group (IG) or control group (CG). All participants receivedconventional rehabilitation training. The IG also received 20 minutes of exposure to Tetrax biofeedback games controlled bychange in center of pressure 3 times a week for 6 weeks.Main Outcome Measurements: The primary outcome was feasibility, addressed by adherence, safety, and satisfaction. Thesecondary outcome was efficacy, which was evaluated by the subtests of physiological profile assessment, posturography, TimedUp and Go, and Forward Reach tests. We used percentage change (post-training score e pretraining score/pretraining score) toquantify the intervention effects. Mann-Whitney U tests were used to analyze differences in percentage of change betweengroups.Results: A total of 14 participants were assigned to the IG, and 13 were assigned to the CG; 12 participants in the IG and 11 in theCG completed the study. In the IG group, those who completed the 6-week intervention attended 89.5% of planned sessions. Nomajor adverse events or falls occurred within the intervention sessions. With use of 5-point Likert scales, participants rated theirenjoyment of Tetrax games as 4.33 � 0.78, their motivation as 4.17 � 1.03, and perceived helpfulness as 4.25 � 0.97. The IGdemonstrated a significantly greater improvement in reaction time (P ¼ .002), proprioception (P < .001), symmetric weightbearing (P ¼ .027), Timed Up and Go (P < .001), and Forward Reach (P < .001) compared with the CG.Conclusions: Using Tetrax biofeedback video games for balance training is a feasible adjunctive program that may augmentconventional therapy in persons affected by chronic hemiplegic stroke.Level of Evidence: II

Introduction Decreased weight bearing on the affected lower limb

Balance deficits are frequently reported in strokesurvivors, with 83% having impaired balance after anacute stroke [1]. The reduced ability to control balancehas been associated with ambulatory dysfunction and anincreased risk of falls [2]. Therefore, one of the majorcomponents of rehabilitative intervention after a strokeoccurs is to improve balance control.

1934-1482 ª 2016 by the American Academy of Physical Medicine and Reha(http://creativecommons.org/licenses/by-nc-nd/4.0/).http://dx.doi.org/10.1016/j.pmrj.2016.02.009

and poor weight shifting are common balance problemsafter a stroke [3-5]. Visual biofeedback using force platesystems has been adopted as a treatment modality forbalance dysfunction [6,7]. The visual biofeedbacktraining augments extrinsic information about tasksuccess to the performer via a computer screen. Suchsystems have been applied to patients with hemiplegia,and some positive effects related to symmetric weight

bilitation. This is an open access article under the CC BY-NC-ND license

963J.-W. Hung et al. / PM R 8 (2016) 962-970

bearing were noted [8]. However, patients might easilylose interest in performing stereotyped, repetitive,weight-shifting tasks. Lack of interest impairs theeffectiveness of the therapeutic exercise. Enrichedvirtual environments could optimize motor learning bymanipulating practice conditions that engage motiva-tional, cognitive, motor control, and sensory feed-backebased learning mechanisms [9]. Some videogames, which are classified as nonimmersive virtualreality systems, are designed with the relevant conceptsof neuroplasticity [10]. Coupling of biofeedback exer-cise directly to video games has been developed and hasreceived a high degree of interest [11]. The use ofcenter of pressure (COP)-controlled, game-based exer-cise regimes has been reported to improve subjects’dynamic balance control [11-16]. The majority of suchstudies rely on off-the-shelf systems with commercialsoftware designed for entertainment, not specificallyfor rehabilitation purposes. Such programs may be toochallenging or inappropriate for persons with balancedeficits. We previously demonstrated that the com-mercial exergaming system, Nintendo Wii Fit, is bene-ficial for balance function in persons with chronic stroke[16]. However, we also found that patients with mod-erate spasticity or slow movement usually struggledwith playing commercial games and felt frustrated. Inaddition, the difficulty levels of commercial games areusually not adjustable.

Commercial systems developed specifically for ther-apeutic training are rare. The Tetrax biofeedback sys-tem [17] is a COP-controlled, video gameebasedexercise system designed for patients with eitherorthopedic or neurologic problems. Although it is acommercial product, limited information about thefeasibility of clinical use for patients with balance def-icits, such as stroke, is available. Safety is the primaryconcern because there is high risk of falling or muscu-loskeletal injury for hemiplegic patients performingweight-shifting movements. It is also not known if pa-tients would accept and be satisfied using such exer-gaming balance training. In addition, the potentialbeneficial effects of Tetrax games training are currentlyunknown. Therefore, the objectives of this study wereto examine the feasibility and potential efficacy of theTetrax biofeedback system for balance training in pa-tients with chronic hemiplegic stroke. It was hypothe-sized that using Tetrax biofeedback games for balancetraining would be feasible and clinically beneficial forpatients with chronic hemiplegic stroke.

Methods

Subjects

The Ethics Research Committee of the study hospitalapproved the study. All participants provided theirwritten informed consent prior to participation in the

study. This study was conducted in the department ofrehabilitation of a tertiary care hospital.

Patients were recruited for the study if they hadsustained a hemiplegic stroke at least 6 months prior toenrollment, were older than 18 years, were able tounderstand verbal instructions and learn, had adequatevisual acuity (appropriate correction was allowed), andhad impaired balance (Berg balance test score <56) buthad the ability to stand independently (without assis-tance from any device) for more than 5 minutes. Pa-tients with bilateral hemispheric or cerebellar lesions,receptive aphasia, severe visual field deficits or hemi-neglect, or any condition that prevented them fromcompleting the program, such as transportation issuesor medical problems, were excluded from the study.

Participants were randomly allocated to either theintervention group (IG) or control group (CG) at a 1:1ratio by using a computer-generated randomizationschedule.

Procedures

All participants received conventional outpatientrehabilitation therapy (which accounted for an averageof 50 minutes of physiotherapy and an additional 50minutes of occupational therapy per day), 3 days aweek, in an outpatient setting.

The IG received additional Tetrax biofeedback bal-ance training 3 days per week (20 minutes per day) for 6weeks in the same outpatient setting. The Tetrax sys-tem was equipped with 4 independent force platesunder the toes and heels (Figure 1). The foot pressurewas acquired via each force plate. An interface boxcaptured the data from the force plates for display on apersonal computer, which contained the Tetraxbiofeedback games software. The video games werecontrolled via the change in the player’s COP.

The Tetrax biofeedback exercise system included 11games. These games were designed to help patientsimprove their balance by using various software chal-lenges, and each was created to focus on a differentaspect of balance control. Eight games were chosenbased on common balance deficits after a stroke. Thetherapeutic goals included the achievement of evenweight distribution and improvement in right-left andfront-back weight transfer. The following 8 games werechosen to improve balance (Figure 2):

1. Catch: Catch one ball by moving the other ball. Themovable ball is moved with changing pressure of thepatient’s feet.

2. Skyball: Move the baseball glove by using right-leftmovement of the patient’s feet to catch thebaseballs.

3. Tag: Move the soccer player by using front-backmovement of the patient’s feet to avoid the soccerballs.

Figure 1. Tetrax biofeedback video game system (Tetrax Ltd).

964 Tetrax Balance Training in Hemiplegics

4. Gotcha: Move the bowling pins by using right-leftmovement of the patient’s feet to avoid thebowling balls.

5. Speedball: Move the basketball hoop by using front-back movement of the patient’s feet to catch thebasketballs.

6. Immobilizer: Keep the top of each of the columnswithin the set section. The columns change withdiffering pressure of the 4 foot parts on the plates.

7. Target: Move a red ball by changing pressure of thepatient’s feet to catch targets in different directions.

8. Freeze: Keep the ball inside the circle. The ball wasmoved by differing pressures of the patient’s feet andkept in position when the patient remained steady,even when the ball was invisible.

The Skyball and Gotcha games emphasized left-rightweight shifting; Tag and Speedball, anterior-posteriorweight shifting; Target and Gotcha, multidirection weightshifting; Immobilizer, static symmetric weight bearing;and Freeze, full weight bearing of a limb. The therapeuticchallenge level was adjusted by target size, number oftargets appearing on the screen at the same time, speedof target movement, and range of percentage of theparticipant’s weight that must be placed to move thevirtual objects. At the end of each game, the finalscore and percentage of corrected hits on the left/

right side are shown on the screen to give objectivefeedback about the participant’s performance. Thesupervising therapist chose 3-5 games for each inter-vention session and set the challenge level. The gameselection and challenge progression were individual-ized according to the ability and needs of eachparticipant. For example, the Skyball and Gotchagames would be selected for patients with a left-rightweight-shifting problem, whereas Tag and Speedballgames would be selected for patients with an anterior-posterior weight-shifting problem.

Outcome Measures

The feasibility of using the Tetrax biofeedback sys-tem for balance training in patients with chronic strokewas the primary outcome of interest, specificallyfocusing on adherence, safety, and satisfaction.Adherence was represented by session attendance.Safety was assessed by recording adverse events,including falls both within the treatment sessions andduring the study period. Satisfaction was evaluated atthe end of the 6-week intervention. Participants used 5-point Likert scales rating enjoyment, motivation, andperceived helpfulness of the Tetrax games from“strongly disagree” (1) to “strongly agree” (5). We alsoasked participants to compare the enjoyment andhelpfulness of the Tetrax biofeedback games and con-ventional rehabilitation training. Participants used“yes/no” responses regarding continued use of thegame and whether they would recommend that otherpatients receive the games training.

The possible efficacy of Tetrax games was the sec-ondary outcome of interest and was assessed by testsinvolving both body function and activity domains. Thebody functions were assessed using subtests of physio-logic profile assessments and posturography. Balanceactivities were assessed by Timed Up and Go (TUAG) andForward Reach (FR) tests. Participants were assessedbefore and after the 6-week intervention.

Physiologic Profile Assessment Subtests

The physiologic profile assessment is a validatedbattery of sensorimotor measurements used to identifysubjects at risk of falling [18]. Four subtests were usedin this study, including (1) proprioception (in degrees) ofknee joints using a lower limb matching task with aninscribed vertical protractor placed between the seatedparticipant’s legs; (2) maximal isometric quadricepsstrength (in kilograms) measured while sitting using aspring gauge; (3) simple reaction time (in milliseconds)using a light stimulus and a finger press on a switch asthe response; and (4) postural sway area (in centimeterssquared) using a sway meter to measure bodydisplacement at the waist level with the participantstanding on a foam rubber mat for 30 seconds under 2

Figure 2. The 8 Tetrax system games used in the study. (1) Speedtrack. (2) Skyball. (3) Tag. (4) Gotcha. (5) Speedball. (6) Immobilizer. (7) Target.(8) Freeze.

965J.-W. Hung et al. / PM R 8 (2016) 962-970

conditionsdeyes open and closed. Maximal medialelateral and anterioreposterior sway (in centimeters)were summed to calculate the final score for eachcondition. Better performance was indicated by largerscores of quadriceps strength and lower scores of pro-prioception, reaction time, and sway area.

Tetrax Posturography System for AssessingDegree of Symmetric Weight Bearing

The Tetrax posturography system is based on theassessment of the vertical pressure fluctuations on 4independent force plates, each placed beneath the

966 Tetrax Balance Training in Hemiplegics

heels and toes of the subject while standing in an up-right position. The weight percentages on the heel andtoe force plates by the affected foot were summed asthe total amount of weight loading on the affected leg.The closer the values were to 50%, the more symmetricwas the weight bearing.

Forward Reach and Timed Up and Go Tests

For the FR test [19], the participants stood next to awall with their unaffected arm raised to 90� and theirfingers extended. The distance (in centimeters) that asubject could reach forward without moving or liftingthe feet was measured. The inter-rater and intra-raterreliabilities were usually high, in the range of 0.93-0.99 [20].

In the TUAG test [21], participants were asked to risefrom a standard chair, walk a 3-m distance, turn, walkback to the chair, and sit down. The time required tocomplete the test (in seconds), beginning at the word“go” and ending when the participant’s back touchedthe backrest of the chair, was recorded. The TUAG testexamines several movements necessary during activitiesof daily living, such as standing up, sitting down,turning, and walking. It showed excellent reliability(interclass correlation coefficient >0.95) in people withchronic stroke [21].

Statistical Analysis

The Statistical Package for the Social Sciences(version 12.0 for Windows, IBM Corp, Armonk, NY) wasused for statistical analyses. Descriptive statistics were

Figure 3. Subject recruitmen

used to summarize baseline characteristics and feasi-bility data. Based on the small sample size, we usedMann-Whitney U tests and c2 tests to compare the de-mographics and clinical data between groups. To helpquantify the magnitude of intervention effects of eachintervention, we used percentage of change for furtheranalyses. The percentage of change was determined bypost-training score e pretraining score/pretrainingscore. Mann-Whitney U tests were used to analyze dif-ferences in percentage of change between groups. Allsignificant tests were 2-tailed, and differences wereconsidered to be statistically significant at P < .05.

Results

Adherence

Thirty-one patients with hemiplegic stroke werescreened for eligibility. Three patients were excludedbecause they did not meet the inclusion criteria, andone did not give informed consent; thus 27 patientswere included in the study. After baseline assessment,14 and 13 patients were randomly assigned to the IG andthe CG, respectively. During the intervention period, 2patients in the IG withdrew from the study, includingone patient with back pain caused by an incidental fallin the bathroom after 6 training sessions and anotherwho refused to continue after 12 training sessions. Inthe CG, one patient withdrew because of worsening ofpre-existing hydrocephalus requiring shunt insertion,and another patient withdrew because of a trans-portation problem. Thus 23 patients (12 in the IG and 11in the CG) completed the study (Figure 3). In the IG

t and attrition flowchart.

967J.-W. Hung et al. / PM R 8 (2016) 962-970

group, those who completed the 6-week interventionattended 89.5% of planned sessions. The main reasonsfor missed sessions were personal issues not related tothe intervention.

Safety

No major adverse events or falls occurred within theintervention sessions. Three patients reported mild painin their lower limbs at an early stage of the interven-tion. This pain did not disrupt the study. After severalsessions of the intervention, the reports of pain sub-sided spontaneously.

Satisfaction

Twelve participants who completed the interventionprograms provided comments. Ten participants (83%)agreed or strongly agreed that the Tetrax games wereenjoyable; the mean enjoyment score was 4.33 � 0.78.Four participants found the Tetrax games more enjoy-able than their routine rehabilitation training. However,3 participants reported that their enjoyment of playingthe games declined by the end of training because theybecame familiar with the games. Nine participants (75%)agreed or strongly agreed that the games increasedtheir motivation for balance training; the mean moti-vation score was 4.17 � 1.03. Ten participants (83%)agreed or strongly agreed that the Tetrax games werehelpful for balance function; the mean helpfulnessscore was 4.25 � 0.97. Three participants reported thatsuch game intervention was as useful as conventionaltherapy, 4 reported that it was complementary to con-ventional therapy, and 3 reported that video gameswere better than conventional therapy. Eight partici-pants (67%) wanted to continue the training as part oftheir routine rehabilitation training. Eleven participants(92%) recommended that other patients receive the

Table 1Characteristics of the participants

Characteristic Intervention Group (n ¼ 12)

Age, y 52.75 (46.70; 62.90)Gender

Male 8 (66.7)Female 4 (33.3)

Type of strokeInfarction 9 (75)Hemorrhage 3 (25)

Affected sideRight 8 (66.7)Left 4 (33.3)

Onset duration, mo 17.50 (10.67; 21.93)Assistive device

None 9Cane 1Quadricane 2

The values are presented as median [first quartile; third quartile] or N (%)

Tetrax biofeedback training. The participants preferredthe Catch game, followed by the Gotcha game; only oneparticipant liked the Immobilizer game.

Training Effects

No differences in demographics or baseline balance-related variables were observed between groups(Tables 1 and 2).

Intervention effects of both groups are summarized inTable 2. Compared with the CG, the IG demonstrated asignificantly greater improvement both in body functionand activity domains, including reaction time (P ¼ .002),proprioception (P < .001), symmetric weight bearing(P ¼ .027), the TUAG test (P < .001), and the FR test(P ¼ .01) (Table 2).

Discussion

Our findings suggest that Tetrax balance gametraining is a feasible and effective intervention forpatients with chronic stroke.

Research has been published in favor of intensive andrepetitive task-specific training for functional recoveryafter a stroke [22,23]. However, conventional repetitiveexercises usually cause patients to lose interest. Lack ofinterest can decrease the potential effectiveness of thetherapy. The engaging nature of a game-based approachmay serve to increase motivation and repetitive prac-tice [24,25]. Similar to other exergaming studies[16,26], the adherence rate (89.5%) of the IG was high.Participants’ enjoyment and perceived benefit for theirbalance function were also high. However, we notedthat the types of games in the Tetrax biofeedback sys-tem were limited compared with the games in com-mercial entertainment systems; therefore, someparticipants lost interest in playing the games at thelate stage of the intervention. Only two thirds of

Control Group (n ¼ 11) P Value

55.20 (45.30; 65.50) .951.99

8 (72.7)3 (27.3)

.998 (72.7)3 (27.3)

.2204 (36.4)7 (63.6)

18.60 (10.47; 32.53) .559.667

713

.

Table

2Therapeuticeffectsin

balance

perform

ance

betw

eeninterventionandco

ntrolgroups

Variable

InterventionGroup(n

¼12

)ControlGroup(n

¼11

)Percentage

ofChange

Baseline

AfterTreatment

Baseline

AfterTreatm

ent

Intervention

Control

PValue

Muscle

strength

Kneeextension,kg

Affectedside

28.83(12.50

;31

.00)

26.83(12.50

;37

.17)

21.83(9.00;

24.33)

21.67(9.33;

28.17)

0.07

(0.05;

0.20

)0.04

(0.00;0.14

).291

Nonaffectedside

31.00(21.83

;38

.50)

35.17(25.00

;38

.33)

33.50(20.83

;38

.50)

33.83(25.33

;40

.33)

0.09

(0.00;

0.21

)0.01

(�0.03

;0.09)

.159

Reactiontime,ms

249.30

(235

.80;

270.40

)23

6.60

(208

.40;

254.90

)23

6.40

(221

.40;

268.80

)23

0.40

(222

.50;

268.90

)�0

.08(�

0.11

;�0

.03)

0.00

(�0.03

;0.03

).002

*Posturalsw

ayarea

Witheye

sopen,cm

23.44

(2.11;

12.01)

4.98

(2.02;

7.78

)8.63

(4.81;

13.45)

5.66

(2.18;

14.18)

0.27

(�0.46

;0.78

)0.24

(0.06;

0.62

).944

Witheye

sclosed,cm

228

.66(15.21

;44

.51)

19.11(6.97;

35.97)

24.53(15.96

;61

.35)

31.23(9.55;

61.15)

�0.27(�

0.38

;0.06

)�0

.19(�

0.45

;0.21

).944

Proprioce

ption,

�2.80

(1.20;

5.80

)1.40

(0.80;

0.20

)1.80

(1.00;

2.20

)1.80

(1.40;

2.60

)�0

.42(�

0.60

;�0

.25)

0.20

(�0.09

;0.33

)<.001

*Weightbearing,

affectedleg,

%46

.17(36.68

;49

.60)

47.47(38.88

;54

.09)

46.15(40.96

;50

.71)

41.28(39.28

;50

.86)

0.10

(�0.00

;0.19

)�0

.03(�

0.11

;0.03

).027

*Tim

edUpandGo,s

14.45(9.16;

18.19)

11.62(8.58;

17.67)

16.81(16.05

;23

.85)

17.05(16.00

;24

.06)

�0.12(�

0.20

;�0

.06)

0.00

(�0.02

;0.01

)<.001

*Forw

ard

reach

distance

,cm

28.50(25.00

;31

.00)

29.50(26.50

;35

.00)

29.00(23.50

;32

.50)

28.50(24.50

;34

.50)

0.08

(0.04;

0.13

)0.00

(�0.02

;0.04

).010

*

Theva

luesare

presentedasmedians[firstquartiles;

thirdquartiles].

*Statisticallysign

ifica

ntatP<

.05.

968 Tetrax Balance Training in Hemiplegics

participants wanted to continue the training. In addi-tion, participants preferred games with more variationsand tolerable level of challenge; only one participantliked the immobilizer game, which is similar to the forceplate biofeedback training. Bower et al [27] pointed outthat acceptability of the game-based intervention mayhave been enhanced through the provision of a largerrange of activities, more engaging and varied in-terfaces, aligning the tasks more closely to participantgoals, and enabling individuals to work at their level ofchallenge. Our findings support their recommendation.

Importantly, no major adverse events occurred withinthe treatment sessions in this study. However, 3 par-ticipants reported having mild pain in their lower limbsat an early stage of intervention, possibly as a result ofrepeated weight shifting movements. Mild and short-lasting pain incidents have also been reported in otherexergaming studies [26,28]. It is suggested that post-exercise pain should be carefully monitored and thatactivities be adapted where necessary. Stretching ex-ercises were recommended before each Tetraxbiofeedback balance training session, and training in-tensity should be slowly increased to avoid such adverseeffects.

This study demonstrated that, in addition to theconventional rehabilitation training for chronic stroke, 6weeks of balance training with the Tetrax biofeedbacksystem might improve participants’ balance within bothbody function and activity levels. First, in the bodyfunction domain, it improved symmetric weight bearing.This beneficial effect is synergistic with the benefit frombiofeedback force plate systems [8]. We chose gameswith the intention of either making weight bearing moresymmetrical, such as the Immobilizer game, orincreasing the weight-bearing stance of the affectedleg, such as the Gotcha and Speedball games. Accordingto the task-specific learning theory, the improvement insymmetric weight bearing was not surprising. In addi-tion, we also noted an improvement in proprioception ofthe affected leg after Tetrax training. This effect waspossibly due to repetitive weight shifting during thetraining course that increased joint compression,thereby improving proprioceptive feedback. Finally,faster reaction times were noted after the Tetrax gamesintervention. In the video games, the targets wererandom and varied in direction and speed, and thus theparticipants were required to make timely, goal-directed shifts in the COP to produce the appropriateresponse within seconds. Such training may enhance theimmediate response of participants, and therefore theirreaction time decreased.

Second, in the activity domain, several studies havereported no clinical benefit from use of force platebiofeedback training alone (without interactive com-puter games) in functional balance activity, asmeasured by either Berg balance test or TUAG scores[8,29]. Conversely, the effects of the interactive

969J.-W. Hung et al. / PM R 8 (2016) 962-970

computer game exercises were beneficial in functionalbalance activities [15,16]. We had a similar finding, withthe IG showing greater improvement in the TUAG testcompared with the CG. By combining force plate andinteractive computer games as biofeedback training,subjects were required to perform episodic bodymovements in order to interact with the video game.These random goal-oriented movements might be morehelpful in improving functional performance than theconstant feedback from a force plate [15,30]. We alsofound significantly more improvement of FR distance inthe IG compared with the CG, probably because of therepeated front-back movements required during certainTetrax games training.

Limitations

This study had several limitations. Tetrax games wereplayed on a rigid surface with no progression to acompliant surface, thus potentially limiting the balancedemands and the balance cost of the exercises. How-ever, even at the late stage of intervention, some par-ticipants still had difficulty playing games on the rigidsurface, and thus playing on a compliant surface maynot be suitable for them.

Investigation of efficacy was limited by the smallsample size, use of a control group that did not receivean equivalent amount of additional therapy, and par-ticipants’ engagement in concurrent therapy. Cautionalso must be exercised when attempting to generalizeour results, because our findings were based on rela-tively young stroke survivors (average age: 54 years)who did not have severe cognitive deficits and were ableto stand independently to some extent. The duration ofthe intervention was short (20 minutes per session for 3sessions per week for 6 weeks), which may haveunderestimated the effects of the intervention. Lack offollow-up assessment also limited the documentation ofcarryover effects of the Tetrax biofeedback system. Theassessor was not blinded to the group allocation, whichwas another drawback of our study. In addition, becausewe excluded patients who could not complete the6-week training program, our participants might have ahigher than usual adherence rate.

Future Study Design

The results in this feasibility study were not poweredto make any definitive conclusion about the efficacy ofTetrax biofeedback training. A single blinded, random-ized, controlled study recruiting an adequate samplesize of participants with varying ages and testing againsta dose-matched control group is needed to explore theeffects of this intervention. In addition, assessing thelong-term effects of Tetrax biofeedback training is alsosuggested.

Conclusion

Using Tetrax biofeedback video games for balancetraining is a feasible adjunctive program that mayaugment conventional therapy in persons with chronichemiplegic stroke. Further study to provide more robustdata about the efficacy of this intervention iswarranted.

References

1. Tyson SF, Hanley M, Chillala J, Selley A, Tallis RC. Balancedisability after stroke. Phys Ther 2006;86:30-38.

2. Laufer Y, Sivan D, Schwarzmann R, Sprecher E. Standing balanceand functional recovery of patients with right and left hemiparesisin the early stages of rehabilitation. Neurorehabil Neural Repair2003;17:207-213.

3. Dickstein R, Nissan M, Pillar T, Scheer D. Foot-ground pressurepattern of standing hemiplegic patients. Major characteristics andpatterns of improvement. Phys Ther 1984;64:19-23.

4. Horak FB, Esselman P, Anderson ME, Lynch MK. The effects ofmovement velocity, mass displaced, and task certainty onassociated postural adjustments made by normal and hemi-plegic individuals. J Neurol Neurosurg Psychiatry 1984;47:1020-1028.

5. Winstein CJ. Knowledge of results and motor lear-ningdimplications for physical therapy. Phys Ther 1991;71:140-149.

6. Bourbonnais D, Bilodeau S, Lepage Y, Beaudoin N, Gravel D,Forget R. Effect of force-feedback treatments in patients withchronic motor deficits after a stroke. Am J Phys Med Rehabil 2002;81:890-897.

7. Geiger RA, Allen JB, O’Keefe J, Hicks RR. Balance and mobilityfollowing stroke: Effects of physical therapy interventions withand without biofeedback/forceplate training. Phys Ther 2001;81:995-1005.

8. Barclay-Goddard R, Stevenson T, Poluha W, Moffatt ME, Taback SP.Force platform feedback for standing balance training afterstroke. Cochrane Database Syst Rev 2004;4:CD004129.

9. Levin MF, Weiss PL, Keshner EA. Emergence of virtual reality as atool for upper limb rehabilitation: Incorporation of motor controland motor learning principles. Phys Ther 2015;95:415-425.

10. Saposnik G, Levin M. Virtual reality in stroke rehabilitation:A meta-analysis and implications for clinicians. Stroke 2011;42:1380-1386.

11. Betker AL, Szturm T, Moussavi ZK, Nett C. Video game-based ex-ercises for balance rehabilitation: A single-subject design. ArchPhys Med Rehabil 2006;87:1141-1149.

12. Anderson F, Annett M, Bischof WF. Lean on Wii: Physical rehabili-tation with virtual reality Wii peripherals. Stud Health TechnolInform 2010;154:229-234.

13. Cho KH, Lee KJ, Song CH. Virtual-reality balance training with avideo-game system improves dynamic balance in chronic strokepatients. Tohoku J Exp Med 2012;228:69-74.

14. Mhatre PV, Vilares I, Stibb SM, et al. Wii Fit balance board playingimproves balance and gait in Parkinson disease. PM R 2013;5:769-777.

15. Szturm T, Betker AL, Moussavi Z, Desai A, Goodman V. Effects of aninteractive computer game exercise regimen on balance impair-ment in frail community-dwelling older adults: A randomizedcontrolled trial. Phys Ther 2011;91:1449-1462.

16. Hung JW, Chou CX, Hsieh YW, et al. Randomized comparison trialof balance training by using exergaming and conventional weight-shift therapy in patients with chronic stroke. Arch Phys MedRehabil 2014;95:1629-1637.

970 Tetrax Balance Training in Hemiplegics

17. BeamMed Ltd (Sunlight). The Tetrax biofeedback system. Availableat http://www.sunlightnet.com/International/html/TetraxBiofeedback.html. Accessed March 7, 2016.

18. Lord SR, Menz HB, Tiedemann A. A physiological profile approach tofalls risk assessment and prevention. Phys Ther 2003;83:237-252.

19. Tyson SF, DeSouza LH. Reliability and validity of functional balancetests post stroke. Clin Rehabil 2004;18:916-923.

20. Lin MR, Hwang HF, Hu MH, Wu HD, Wang YW, Huang FC. Psycho-metric comparisons of the timed up and go, one-leg stand, func-tional reach, and Tinetti balance measures in community-dwellingolder people. J Am Geriatr Soc 2004;52:1343-1348.

21. Ng SS, Hui-Chan CW. The timed up & go test: Its reliability andassociation with lower-limb impairments and locomotor capacitiesin people with chronic stroke. Arch Phys Med Rehabil 2005;86:1641-1647.

22. Veerbeek JM, van Wegen E, van Peppen R, et al. What is the evi-dence for physical therapy poststroke? A systematic review andmeta-analysis. PLoS One 2014;9:e87987.

23. Langhorne P, Bernhardt J, Kwakkel G. Stroke rehabilitation.Lancet 2011;377:1693-1702.

24. Thomson K, Pollock A, Bugge C, Brady M. Commercial gaming de-vices for stroke upper limb rehabilitation: A systematic review. IntJ Stroke 2014;9:479-488.

25. Peters DM, McPherson AK, Fletcher B, McClenaghan BA, Fritz SL.Counting repetitions: An observational study of video game play inpeople with chronic poststroke hemiparesis. J Neurol Phys Ther2013;37:105-111.

26. Bower KJ, Clark RA, McGinley JL, Martin CL, Miller KJ. Clinicalfeasibility of the Nintendo Wii for balance training post-stroke: Aphase II randomized controlled trial in an inpatient setting. ClinRehabil 2014;28:912-923.

27. Bower KJ, Louie J, Landesrocha Y, Seedy P, Gorelik A,Bernhardt J. Clinical feasibility of interactive motion-controlledgames for stroke rehabilitation. J Neuroeng Rehabil 2015;12:63.

28. Yong Joo L, Soon Yin T, Xu D, et al. A feasibility study usinginteractive commercial off-the-shelf computer gaming in upperlimb rehabilitation in patients after stroke. J Rehabil Med 2010;42:437-441.

29. Walker C, Brouwer BJ, Culham EG. Use of visual feedback inretraining balance following acute stroke. Phys Ther 2000;80:886-895.

30. Brauer SG, Woollacott M, Shumway-Cook A. The interacting effectsof cognitive demand and recovery of postural stability in balance-impaired elderly persons. J Gerontol A Biol Sci Med Sci 2001;56:M489-M496.

Disclosure

J.-W.H. Department of Rehabilitation, Chang Gung Memorial HospitaleKaohsiung Medical Center, Chang Gung University College of Medicine, 123,Ta-Pei Road, Niao-Sung District, Kaohsiung City, Taiwan, Republic of China.Address correspondence to: J.-W.H.; e-mail: hung0702@adm.cgmh.org.twDisclosure: nothing to disclose

M.-Y.Y. Department of Rehabilitation, Chang Gung Memorial HospitaleKaohsiung Medical Center, Kaohsiung City, Taiwan, Republic of ChinaDisclosure: nothing to disclose

K.-C.C. Division of Cerebrovascular Diseases, Department of Neurology, ChangGung Memorial Hospital, Kaohsiung City, Taiwan, Republic of China; Collegeof Medicine, Chang Gung University, Kaohsiung City, Taiwan, Republic of ChinaDisclosure: nothing to disclose

H.-C.L. Department of Physical Therapy and Assistive Technology, Exercise andHealth Science Research Center, National Yang-Ming University, Taipei, Taiwan,Republic of ChinaDisclosure: nothing to disclose

Y.-W.H. Department of Rehabilitation, Chang Gung Memorial HospitaleKaohsiung Medical Center, Kaohsiung City, Taiwan, Republic of ChinaDisclosure: nothing to disclose

P.-C.C. Department of Rehabilitation, Chang Gung Memorial HospitaleKaohsiung Medical Center, Kaohsiung City, Taiwan, Republic of ChinaDisclosure: nothing to disclose

Submitted for publication July 20, 2015; accepted February 25, 2016.