Washington University School of MedicineDigital Commons@Becker

Open Access Publications

2016

Long-term effects of robotic hippotherapy ondynamic postural stability in cerebral palsyYoung Joo ChaYonsei University

Megan StanleyWashington University School of Medicine in St. Louis

Tim ShurtleffWashington University School of Medicine in St. Louis

Joshua S.H. YouYonsei University

Follow this and additional works at: https://digitalcommons.wustl.edu/open_access_pubs

This Open Access Publication is brought to you for free and open access by Digital Commons@Becker. It has been accepted for inclusion in OpenAccess Publications by an authorized administrator of Digital Commons@Becker. For more information, please contact engeszer@wustl.edu.

Recommended CitationCha, Young Joo; Stanley, Megan; Shurtleff, Tim; and You, Joshua S.H., ,"Long-term effects of robotic hippotherapy on dynamicpostural stability in cerebral palsy." Computer Assisted Surgery.21,. . (2016).https://digitalcommons.wustl.edu/open_access_pubs/5719

Full Terms & Conditions of access and use can be found athttp://www.tandfonline.com/action/journalInformation?journalCode=icsu21

Download by: [Washington University in St Louis] Date: 10 April 2017, At: 15:21

Computer Assisted Surgery

ISSN: (Print) 2469-9322 (Online) Journal homepage: http://www.tandfonline.com/loi/icsu21

Long-term effects of robotic hippotherapy ondynamic postural stability in cerebral palsy

Young Joo Cha, Megan Stanley, Tim Shurtleff & Joshua (Sung) H. You

To cite this article: Young Joo Cha, Megan Stanley, Tim Shurtleff & Joshua (Sung) H. You (2016)Long-term effects of robotic hippotherapy on dynamic postural stability in cerebral palsy, ComputerAssisted Surgery, 21:sup1, 111-115, DOI: 10.1080/24699322.2016.1240297

To link to this article: http://dx.doi.org/10.1080/24699322.2016.1240297

© 2016 The Author(s). Published by InformaUK Limited, trading as Taylor & FrancisGroup.

Published online: 25 Oct 2016.

Submit your article to this journal

Article views: 169

View related articles

View Crossmark data

RESEARCH ARTICLE

Long-term effects of robotic hippotherapy on dynamic postural stability incerebral palsy

Young Joo Chaa, Megan Stanleyb, Tim Shurtleffb and Joshua (Sung) H. Youa

aSports�Movement Institute & Technology (S�MIT), Department of Physical Therapy, Yonsei University, Wonju, South Korea;bOccupational Therapy Program, School of Medicine, Washington University, St. Louis, MO, USA

ABSTRACTBackground: Dynamic postural instability is a common neuromuscular impairment in cerebralpalsy (CP), which often includes balance dysfunction and an associated risk of serious falls.Robotic hippotherapy has recently become a widespread clinical application to facilitate posturalcore stabilization, strength, and endurance through repetitive vestibular and proprioceptivestimulation to the spine via the sensorimotor system pathways. However, the long-term effectsof robotic hippotherapy on dynamic postural instability in CP remain unclear.Objective: To examine the long-term effects of robotic hippotherapy on dynamic postural stabil-ity in CP.Methods: An advanced three-dimensional biomechanical eight-camera video motion capture(VMC) system was used to compute the center of mass (COM) pathway, which represents inter-vention-related spinal core instability. The robotic hippotherapy system was used to improvedynamic postural stability and associated balance performance. Robotic hippotherapy exercisewas provided for 45minutes/session, 2–3 times a week over the 12-week period.Results: Abnormal mean COM pathway length, standard deviation, and range substantiallydecreased after 12 weeks of robotic hippotherapy. The initial x-axis COM was greater than thatof the y-axis. However, the amount of abnormal anterior–posterior and medio-lateral posturalsway substantially decreased after robotic hippotherapy.Conclusions: This study provides the first compelling evidence that the robotic hippotherapy issafe and effective for postural instability control and sitting balance dysfunction that mitigatesthe risk of falls in CP.

KEYWORDSCenter of mass; cerebralpalsy; core stability; robotichippotherapy

1. Introduction

There is an accumulating body of evidence that cor-roborates the therapeutic efficacy of horseback ridingfor facilitating gross motor function in children withcerebral palsy (CP).[1–4] Nevertheless, robotic hippo-therapy has recently appeared as an alternative in clin-ical environments where real horses are not readilyaccessible or unaffordable.[5] Limited accessibility tohorses, weather-dependence, and cost may all contrib-ute to the increased use of robotic horses.[6] Althoughrobotic hippotherapy does not allow patients to inter-act with a live horse, this therapy has the advantageof allowing regular therapy with no apparent spatio-temporal or weather constraints.[7] Our robotic hippo-therapy system is designed to facilitate stretching,rhythmic trunk rotation, core stabilization, strength,endurance, and cardiopulmonary function via the

sensorimotor system (vestibular, propriocep-tive).[2,6,8–11] The system simulates live horse move-ments, including walking (6 km/h), trotting (15 km/h),cantering (25 km/h), and galloping (60 km/h). Thesemovements are featured as exercise modes composedof 100 different two-dimensional movement patternswith 100 different exercise modes. Conceptually,robotic hippotherapy was developed based on the col-lective integration of the best clinical evidence [2,8,9]and the fact that pediatric neurorehabilitation must befun, motivating, intensive, task-specific, rhythmic,repetitive, and integrated, and should be implementedover a long period of time to produce measurablechanges in muscle size,[2,6] motor behavior,[10] andneuroplasticity.[11] Intensive repetitive, rhythmicmovement is a hallmark of horse movement, likehuman locomotion. An average horse cadence is

CONTACT Joshua (Sung) H. You, PT, PhD, Professor neurorehab@yonsei.ac.kr Department of Physical Therapy, Yonsei University, Director ofSports�Movement Institute & Technology (S�MIT), 1 Yonseidae Kil, Wonju City, Kangwon-do 220-710, South Korea� 2016 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. This is an Open Access article distributed under the terms of the CreativeCommons Attribution-NonCommercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits unrestricted non-commercial use, distribution, and reproductionin any medium, provided the original work is properly cited.

COMPUTER ASSISTED SURGERY, 2016VOL. 21, NO. S1, 112–116http://dx.doi.org/10.1080/24699322.2016.1240297

reported to be 100 steps per minute at a preferred ormedium walking speed,[12] which can serve as atherapeutic perturbation and a challenge to posturalstability, since the client needs to adaptively re-stabil-ize his or her perturbed equilibrium repeatedly.However, the long-term effects of robotic hippother-apy on dynamic postural instability in CP, which isconceptually defined as an inability to maintain thecenter of mass (COM) within the base of support (BOS)in response to external perturbation or movement,remain unclear.[13,14] Therefore, the purpose of thisstudy was to examine the long-term effects of robotichippotherapy on dynamic postural stability. Wehypothesized that a long-term robotic hippotherapyintervention would improve dynamic postural stabilityand balance performance.

2. Materials and methods

The study included a pre-test, a 12-week hippotherapyintervention, and a post-test. An 11-year-old childdiagnosed with CP and associated core instability wasthe subject in this study. At the pre- and post-tests,intervention-related postural instability was estimatedusing the trajectories of the COM. The COM was meas-ured using an eight-camera video motion capture(VMC) system using Cortex software, Version 1.0.0.198(Motion Analysis Corporation, Santa Rosa, CA) and a50-cm� 50-cm force plate (Kistler; Winterthur,Switzerland). To identify intervention-related changesin dynamic postural stability, 31 reflective markers(9mm) were attached to anatomical landmarks on thechild's head, trunk, and lower extremities to track kine-matic motion. As illustrated in Figure 1, a three-dimen-sional figure was created in the visual motion capture

(VMC) software using reflective surface markers placedon anatomical landmarks. This VMC system was set torecord at a sampling rate of 60Hz with an accuracy of0.5mm. A force plate was also used to record groundreaction forces at 300Hz and was later synchronizedto compute the COM. The child was positioned on acustomized, motorized barrel in the motion capturevolume. The motorized barrel moved reciprocally inan anterioposterior direction at three speeds: 0.5, 0.6,and 0.7 Hz.

Data analysis for the COM measurement wasimplemented by synchronizing the kinematic dataand kinetic data. Kinematic data obtained from sur-face marker tracking were edited using Cortex soft-ware to configure three-dimensional coordinates foreach trial. Data recorded for the second half (7.5 s) ofeach 15-s trial were processed for further analysis,which represented relatively consistent reciprocatingmovement of the customized, motorized barrel. Thekinematic variables measured from these tracked dataincluded head and trunk segment kinematics anglesduring motorized barrel perturbation. The head seg-ment kinematic angle was determined by computingthe intersection angle between the line segments ofthe vertex marker and the C7 marker, and the hori-zontal line segment between the front and the backbarrel markers. The trunk segment kinematic anglewas determined by the intersection angle betweenthe line segment of the C7 marker and the L5marker, and the horizontal line of the barrel markers.More detailed information has been describedpreviously.[8]

The robotic hippotherapy system (FORTIS-102,Daewon Fortis, Ha Nam, Kyungi, Korea) was used toimprove dynamic postural stability. Robotic hippother-apy exercise was administered by experienced thera-pists 2–3 times a week over the 12-week period. Eachinterventional session lasted 45minutes and includedsitting in various positions on a moving robotic horsethat walked and/or trotted (e.g., forward astride, sidesit, tall kneel, reverse astride, and quadruped). Thetherapy involved several transitions between positions,some of which occurred while the horse was mov-ing.[9] As described in the manufacturer's guidelines(Figure 2), therapy involving the device includes 12visits to the lab. The robotic horse movement levelwas set at course #100 with a slow speed or in a com-fortable walking mode (levels 1–50), which imitatesrhythmic horse movement designed for core stabiliza-tion; trunk rotation in forward astride position, side-sit-ting, backward astride, and tall kneel; and improvedpostural and locomotor movement complexity in alldirections.[6]

Figure 1. Three-dimensional figure created in visual motioncapture (VMC) software using reflective surface markers onanatomical landmarks.

COMPUTER ASSISTED SURGERY 113

3. Results

The mean COM pathway length, standard deviation,and range decreased from pre-robotic hippotherapy topost-robotic hippotherapy (Table 1). Specifically, theinitial X-axis COM was greater than that of the Y-axis.

However, the amount of abnormal anterior–posteriorand medio-lateral postural sway decreased as a resultof the intervention.

4. Discussion

The present investigation highlighted the long-termeffects of robotic hippotherapy on dynamic posturalstability in a child with CP. As anticipated, after 12weeks of robotic hippotherapy, our biomechanicalanalysis results demonstrated that both X- and Y-axisCOM pathway lengths decreased in anterior–posteriorand medio-lateral spinal core instability during amotorized barrel perturbation. This finding supportshippotherapy's effectiveness for improving posturalstability during dynamic movement using objectivequantification of spinal core stability in a child withCP.[6,8,9,15] These objective improvements in dynamicpostural stability were correlated to functional move-ment performance, specifically dynamic sitting andstanding balance, after the intervention. As a result ofthe therapy, the child was more stable and uprightwhile sitting and standing. Prior to the intervention,the child demonstrated greater postural sway, as evi-denced by the COM pathway length measurement.This leads to postural instability, poor balance, and ahigher risk of falling.[16] The child's parent reportedthat the child occasionally fell prior to the interven-tion. These falling episodes were associated with poorpostural stability. The therapy allowed the child to bemore confident while walking, jumping, and running.In addition, there was a significant reduction in appar-ent abnormal postural sway and staggering locomotorbehavior (Figure 3).

Figure 2. Experimental setup for robotic hippotherapy.

Table 1. Static stability variables.COM(x-axis)

COM(y-axis)

Pre Post Pre Post

Sway pathlength (cm)

36.78 ± 16.97 33.56 ± 0.75 21.28 ± 0.80 19.20 ± 0.52

Figure 3. Dynamic COM trajectory changes between the pre-/post-robotic hippotherapy intervention.

114 Y. J. CHA ET AL.

The underlying neurophysiological rationale forsuch improvements in postural core stability and bal-ance performance may result from the integrative andrepetitive vestibular and proprioceptive sensorimotorstimuli provided during robotic hippotherapy. Robotichippotherapy may have facilitated upright posture,anticipatory postural adjustment control, equilibriumreaction, stretching of shortened hip abductors andstrengthening of the lumbopelvic musculature. Suchneurophysiological outcomes were reflected inimproved postural alignment, symmetry, and musclesize and strength in trunk muscles as a result of therobotic hippotherapy.[6] These findings are also repre-sented by our data regarding COM postural stabilitymeasurements. In the present study, the typicalrobotic hippotherapy session lasted 45minutes andthe total sensorimotor stimulations included as manyas 3000–5000 repetitions of postural control challenge.This exceeded the required number of repetitions thatare typically offered in conventional neurorehabilita-tion or may be equivalent to the hippotherapy.[5,6]Lang et al. [17] demonstrated that the number of rep-etitions in customary neurorehabilitation (physical andoccupational therapy) care sessions included 6.0 and291.5 repetitions per session for balance and gait(steps), particularly in stroke rehabilitation. However,previous experimental studies showed that as many as400–600 repetitions are needed to produce neuroplas-ticity and associated motor recovery after ischemiclesions.[11,18] In this context, innovative robotic hip-potherapy and virtual reality systems should be con-sidered as an alternative movement therapy toprovide interactive and motivating virtual reality exer-cises with a greater number of repetitions (Lang). Infact, Lee et al. [6] reported that robotic hippotherapycan provide rhythmic two-dimensional postural controland lumbopelvic core stabilization exercise with largenumbers of repetitions (3000–5000 per each session),which surpasses conventional neurorehabilitation andpotential requirements for neuroplasticity and mayallow functional motor recovery to occur. Recentrobotic hippotherapy studies [5] demonstrated thatrobotic hippotherapy was capable of therapeuticstimulation with safe, repetitive, and variable modes ofmovement perturbation. Park et al. [5] showed thatthe robotic hippotherapy acceleration was approxi-mately five times less than that of real horse move-ment (0.67m/s2 vs. 3.22m/s2, respectively). Thissuggests the issue of patient safety when performinghippotherapy exercises with a real horse, especially forpatients who exhibit inherently unstable postural con-trol if not carefully screened.

Undoubtedly, the four-legged robotic horse systemprovides 6 degrees of freedom (DOF), which mimicsreal horse movement; this can be useful for trainingnovice ‘healthy’ riders, but can be dangerous for peoplewith postural control dysfunction, because of a higherrisk of falls.[12,13] Moreover, as documented previ-ously,[5] greater acceleration movement was observedwith real horse hippotherapy because it involves mul-tiple DOFs just like the four-legged robotic horse, whichrender a great challenge for children with moderate orsevere CP. Hence, in a real hippotherapy situation, two-side walkers including a therapist usually work togetherto ensure patient safety. In robotic hippotherapy, thereis a greater liberty to progress and adjust the robotichorse movement from slow amplitudes to large, fastmovement amplitudes. This enables a therapist toaccommodate patients with moderate and severe pos-tural control dysfunction, and then build up the chal-lenge in fine increments as they progress in theirpostural core stability, endurance, strength, and coord-ination. Such postural core improvements will not onlyreduce the risk of falls in children with CP, but also helpto address the common neuromuscular impairments inCP such as scoliosis, lordosis, and kyphosis; this isbecause previous evidence has demonstrateddecreased deep core muscle size such as transversusabdominis muscle in adolescent scoliosis.[6,19,20]Taken together, these results provide therapeutic evi-dence that robotic hippotherapy can be used as analternative therapy in the management of CP patientswith core instability. Nevertheless, further clinical trialsare needed to generalize our findings. The objectivequantification of these outcome measures usingadvanced motion analysis systems can provide import-ant clinical evidence regarding the results of robotichippotherapy. Such data will further inform cliniciansand third-party supporters about the benefits of robotichippotherapy.

5. Conclusion

Robotic hippotherapy is an effective method of enhanc-ing dynamic postural stability in CP patients with pos-tural instability, poor balance, and a high fall risk. In thisstudy, a long-term robotic hippotherapy interventiondecreased COM anterior–posterior and medio-lateralsway. These data suggest that robotic hippotherapyimproves spinal core stability and balance.

Disclosure statement

The authors report no conflicts of interest. The authors aloneare responsible for the content and writing of this article.

COMPUTER ASSISTED SURGERY 115

Funding

This study was funded by Daewon Fortis, Ha Nam, Kyungi,Korea, and in part by “Brain Korea 21 PLUS Project (GrantNo. 2016-51-0009)” sponsored by the Korean ResearchFoundation for Department of Physical Therapy in GraduateSchool, Yonsei University.

References

[1] Casady RL, Nichols-Larsen DS. The effect of hippother-apy on ten children with cerebral palsy. Pediatr PhysTher. 2004;16:165–172.

[2] You SH, Jang SH, Kin YH, et al. Virtual reality-inducedcortical reorganization and associated locomotorrecovery in chronic stroke: an experimenter-blindrandomized study. Stroke. 2005;36:1166–1171.

[3] Frank A, McCloskey S, Dole RL. Effect of hippotherapyon perceived self-competence and participation in achild with cerebral palsy. Pediatr Phys Ther.2011;23:301–308.

[4] Zadnikar M, Kastrin A. Effects of hippotherapy andtherapeutic horseback riding on postural control orbalance in children with cerebral palsy: a meta-ana-lysis. Dev Med Child Neurol. 2011;53:684–691.

[5] Park JH, Shurtleff T, Engsberg J, et al. Comparisonbetween the robo-horse and real horse movementsfor hippotherapy. Biomed Mater Eng. 2014;24:2603–2610.

[6] Lee DR, Lee NG, Cha HJ, et al. The effect of robo-horseback riding therapy on spinal alignment andassociated muscle size in MRI for a child with neuro-muscular scoliosis: an experimenter-blind study.NeuroRehabilitation. 2011;29:23–27.

[7] Zurek G, Dudek K, Pirogowicz I, et al. Influence ofmechanical hippotherapy on skin temperatureresponses in lower limbs in children with cerebralpalsy. J Physiol Pharmacol. 2008;59:819–824.

[8] Shurtleff TL, Engsberg JR. Changes in trunk and headstability in children with cerebral palsy after hippo-therapy: a pilot study. Phys Occup Ther Pediatr.2010;30:150–163.

[9] Shurtleff TL, Standeven JW, Engsberg JR. Changes indynamic trunk/head stability and functional reachafter hippotherapy. Arch Phys Med Rehabil. 2009;90:1185–1195.

[10] Han JY, Kim JM, Kim SK, et al. Therapeutic effects ofmechanical horseback riding on gait and balance abil-ity in stroke patients. Ann Rehabil Med. 2012;36:762–769.

[11] Nudo RJ, Wise BM, SiFuentes F, et al. Neural sub-strates for the effects of rehabilitative training onmotor recovery after ischemic infarct. Science.1996;272:1791–1794.

[12] Clayton HM, Walk this way. USDF Connection; 2002.p. 39–42.

[13] Goldie PA, Bach TM, Evans OM. Force platformmeasures for evaluating postural control: reliabilityand validity. Arch Phys Med Rehabil. 1989;70:510–517.

[14] Pai YC, Patton J. Center of mass velocity-position pre-dictions for balance control. J Biomech. 1997;30:347–354.

[15] Shurtleff T, Engsberg J. Long-term effects of hippo-therapy on one child with cerebral palsy: a researchcase study. Br J Occup Ther. 2012;75:359–366.

[16] Hasson CJ, Van Emmerik RE, Caldwell GE, et al.Predicting dynamic postural instability using center ofmass time-to-contact information. J Biomech.2008;41:2121–2129.

[17] Lang CE, MacDonald JR, Gnip C. Counting repetitions:an observational study of outpatient therapy for peo-ple with hemiparesis post-stroke. J Neurol Phys Ther.2007;31:3–10.

[18] Kleim JA, Barbay S, Nudo RJ. Functional reorganizationof the rat motor cortex following motor skill learning.J Neurophysiol. 1998;80:3321–3325.

[19] Linek P, Saulicz E, Kuszewski M, et al. Ultrasoundassessment of the abdominal muscles at rest and dur-ing the ASLR test among adolescents with scoliosis.Clin Spine Surg. 2016. [Epub ahead of print]. doi:10.1097/BSD.0000000000000055.

[20] Linek P, Saulicz E, Wolny T, et al. Ultrasound evalu-ation of the symmetry of abdominal muscles in mildadolescent idiopathic scoliosis. J Phys Ther Sci.2015;27:465.

116 Y. J. CHA ET AL.