CAN SENSOR-BASED BALANCE ASSESSMENT REPLACE CONVENTIONAL TESTS IN STROKE REHABILITATION?

M. Geerars^1,2,3, R.A. Felius^1,4, N.C. Wouda^1,2,5, A.M. Visser-Meily^2,6, M.F. Pisters^2,7,8

¹Utrecht University of Applied Sciences, Research Group Lifestyle and Health, Utrecht, Netherlands, ²UMC Utrecht Brain Center, Utrecht University, Department of Rehabilitation, Physiotherapy Science & Sport, Utrecht, Netherlands, ³Axioncontinu, Rehabilitation Center de Parkgraaf, Physiotherapy Department Neurology, Utrecht, Netherlands, ⁴Vrije Universiteit Amsterdam, Faculty of Human Movement Sciences, Amsterdam, Netherlands, ⁵De Hoogstraat Rehabilitation, Physiotherapy Department Neurology, Utrecht, Netherlands, ⁶Centre of Excellence for Rehabilitation Medicine, UMC Utrecht Brain Centre, University Medical Centre Utrecht, and De Hoogstraat Rehabilitation, Utrecht, Netherlands, ⁷Center for Physical Therapy Research and Innovation in Primary Care, Center for Physical Therapy Research and Innovation in Primary Care, Julius Health Care Centers, Utrecht, Netherlands, ⁸Fontys University of Applied Sciences, Research Group Empowering Healthy Behaviour, Department of Health Innovations and Technology, Eindhoven, Netherlands

Background: Balance problems in sub-acute stroke survivors are often assessed with the conventional balance tests Trunk Control Test (TCT), Berg Balance Scale (BBS) and Mini Balance Evaluation System Test (MiniBESTest). These tests have some disadvantages: they are subjective, have floor- and ceiling effects or are time consuming. Objective balance assessment with a movement sensor, a so called ‘inertial measurement unit” (IMU), might reduce the time investment and burden of measuring balance, and reduce the problem of floor and ceiling effects as seen in conventional tests.

Purpose: In this study we explored IMU balance assessment as compared to conventional balance tests. We investigated the relationship between the conventional and IMU balance tests and examined the floor-and ceiling effects of all tests.

Methods: An observational study was conducted with sub-acute stroke survivors in five inpatient rehabilitation centers in the Netherlands. Participants were able to sit unsupported for one minute. Both the conventional tests TCT, BBS and miniBESTest (if applicable) and IMU balance test were carried out. The IMU balance test measured postural sway in five tasks: sitting (un)stable, stance stable with eyes open (EO) and eyes closed (EC), and stance unstable. All analysis were carried out at admission and at discharge, to cover floor and ceiling effects. The relationship between the conventional tests and the IMU was determined by correlation testing. The floor -and ceiling effects were assessed by investigating the score distributions, skewness, and kurtosis of all balance tests.

Results: At admission 105 participants were measured and 90 at discharge. A strong correlation (r= -.70, p<.001) was found between the BBS and IMU stance EO at discharge. There were moderate correlations between the BBS and IMU stance EC and between the MiniBESTest and IMU stance unstable at both measurement moments. No correlation was found between the TCT and IMU sitting tasks. The IMU had similar ceiling effects compared to conventional tests.

Conclusions: Sensor-based balance assessment derives partly the same and partly unique information about balance in sub-acute stroke survivors, as compared to conventional measures. Our findings demonstrate ceiling effects in the conventional balance tests, and in the IMU. The clinical relevance of IMUs is yet to be shown in future research.

Implications: An IMU cannot replace conventional balance tests but might be a candidate for quick and repeated balance assessment in individual stroke survivors.

Funding acknowledgements: Study: funded by: SIA-RAAK (RAAK.PRO.03.006). SMB: funded by a VIDI grant (016.Vidi.178.014), Dutch Organization for Scientific Research (NWO).

Keywords:
Stroke
Balance
Inertial measurement unit

Topics:
Neurology: stroke
Disability & rehabilitation
Innovative technology: robotics

Did this work require ethics approval? Yes

Institution: University Medical Center Utrecht

Committee: The medical ethical review committee of Utrecht, the Netherlands

Ethics number: METC number: 20-462/C