Which factors influence the accuracy of knee ROM data of the BPMpathway e-rehabilitation system? A concurrent validity study with healthy subjects

AIMS AND OBJECTIVES: As the population in different countries ages, there is an expected increase in the demand for TKA procedures in the coming decades. This increasing demand will face the healthcare systems and health care providers with new challenges. After TKA physiotherapy and exercise lead to short term improvements in physical function. Physiotherapy exercise provided at home is an appealing approach with the possibility of wider acceptability and uptake in the near future. E-rehabilitation or telerehabilitation are the terms used to define medical rehabilitation programs that are implemented at home with the use of information and communication technologies (ICT). The BPMpathway (270 Vision Ltd., Andover, UK) is an internet home-based biofeedback and remote patient monitoring system. The system comprises a single high-performance wearable wireless sensor, a patient application that can be downloaded for free and a software application for the clinic. The aim of the present study was to test the concurrent validity of the ROM data displayed by the BPMpathway system for knee exercises. MATERIALS AND METHODS: This is a concurrent validity study conducted in a laboratory setting. The ROM data obtained with the BPMpathway system was compared with the ROM data acquired simultaneously with a twin-axis digital goniometer (BIOPAC Systems, Inc., California USA). The effects of sensor positioning (position of the sensor on the limb) and ergonomic factors (chair height) on ROM were investigated. The sample consisted of healthy male subjects with different body compositions. The data acquisition was carried out in both systems at a sample rate of 100Hz. Data was analyzed with MATLAB. The Root Mean Squared Error (RMSE) was used to calculate the difference between the values measured by the two devices. Additionally the mean ± SD values for each system were plotted and the absolute difference of the mean curves was plotted as measurement error. A ROM mean difference of 5° between both systems was considered as clinically irrelevant. RESULTS: A 20% increase or reduction in sitting height led to a RMSE increase. A ventral shift of the BPMpathway sensor by 45° and 90° caused significant measurement errors. A vertical shift was associated with a diminution of the measurement errors. The lowest RMSE (2.4°) was achieved when the sensor was placed directly below the knee. The effects of the manipulations were similar for all subjects. RMSE values were higher in subject A, which had a high body mass index (BMI). CONCLUSION: The knee ROM data measured by the BPMpathway system seems reliable and comparable to the data of the concurrent system used in this investigation, provided the instructions of the manufacturer are respected concerning the sitting position of the subject for knee exercises, and disregarding the same instructions for sensor positioning, by placing the sensor directly below the knee. Changes in seating position and rotation shifts of the sensor induced measurement errors.