Title : A data driven approach to prehabilitation and rehabilitation for hip and knee replacement patients
Abstract:
A patient’s ability to walk has a profound effect on their health, functional ability and independence. Patients with late stage hip or knee osteoarthritis (OA) undergo joint replacement surgery so that they can walk normally again without pain. However, patients will have adapted their gait kinematics to minimize pain, which results in certain muscles being weakened. GaitSmart, a sensor based technology that accurately measures gait kinematics in a clinic was used to collect data in a number of studies on hip and knee replacement patients. For hip OA the range of motion at the hip joint is affected, whilst for knee OA the range of motion in swing and stance are affected. Data also shows that one year post op, only 50% of patients return to a normal gait, suggesting that rehabilitation is not optimized.
Prehabilitation prior to surgery is designed to help patients strengthen their weakened muscles in preparation for surgery, whilst rehabilitation is to strengthen weakened muscles once the joint pain has been replaced. Identifying gait abnormalities and providing corrective exercises is conventionally performed by physiotherapists and outcomes are subjective questionnaires, such as Oxford hip score (OHS) or Oxford knee score (OKS) and EQ5D.
This paper describes an alternative, fully automated, data driven approach using GaitSmart. The range of motion for both hips and knees are presented in an easy to understand report, using traffic light coding and scoring for the clinician and patient. This data is used to generate a musculoskeletal model from which weakened muscles are identified and an optimal set of exercises determined. Patients receive their report with the exercises and ask to perform the exercises daily at home. In rehabilitation, this process is repeated 4 times, each with a 4 week interval to give time for the muscles to strengthen.
A pilot randomized control trial on post op hip and knee replacement patients was carried out at NNUH, comparing physiotherapy led rehabilitation with this novel data driven approach. The results improvements in gait kinematics for hip and knee patients combined were greater in the GaitSmart intervention group than the physiotherapy group. This resulted in a clinically meaningful increase in gait speed, which was statistically significant. For the patient reported outcomes, the change in OKS was statically significant for the intervention group whilst for the OHS it was greater for the control group. For the EQ5D-5L there was no clear distinction between the two groups. Thus, the intervention programme was more effective than physiotherapy in meeting the primary goal of helping patients to walk better.
A new study is currently underway offering patients a prehabilitation programme, so that their muscles are stronger before surgery. Preliminary findings on this will also be presented.
