Abstract
Summary
Movement analysis (IMA) and activity monitoring (AM) using a body-fixed inertia-sensor can discriminate patients with ankle injuries from controls and between patients of different pathology or post-injury time. Weak correlations with PROMs show its added value in objectifying outcome assessment.
Introduction
Ankle injuries often result in residual complaints calling for objective methods to score outcome alongside subjective patient-reported outcome measures (PROMs). Inertial motion analysis (IMA) and activity monitoring (AM) using a body-fixed sensor have shown clinical validity in patients suffering knee, hip and spine complaints. This study investigates the feasibility of IMA and AM 1) to differentiate patients suffering ankle injuries from healthy controls, 2) to compare different ankle injuries, 3) to monitor ankle patients during recovery.
Methods
32 patients suffering ankle problems (ankle sprain, n=17, 42 ±26yrs; operatively managed ankle fracture, n=15, 44 ±17yrs) during short-term recovery (6weeks to 3months post-trauma) were compared to 22 matched healthy controls (41 ±13yrs). Function was measured using IMA: Gait, Sit-Stand (STS) and 10s one-leg Balance test (OLB) were performed while wearing a 3D inertia-sensor at the lower back. Physical activity was measured during 4 successive days using a 3D accelerometer (AM), attached at the upper leg. Validated algorithms were used to derive motion parameters (e.g. speed for gait, bending angle for STS, sway area for OLB, # steps for AM). Moreover three ankle specific questionnaires were completed: Manchester-Oxford Foot and Ankle Questionnaire (MOXFQ), Foot and Ankle Ability Measure (FAAM), AOFAS ankle-hindfoot score. Data was analysed using paired t-tests, one-way ANOVA and Pearson's r.
Results
Both patient groups performed worse in function (IMA), activity (AM) and scales (PROM). Even at 3 months mean walking speed (−26%) and cadence (−9%) were slower while pelvic obliquity during ST (+18%) and sway during OLB (+230%) was higher than in controls (p<0.05). Activity was less regarding step/day (−31%) and sit-stand transfers (−11%, p<0.05). Also PROMs were significantly worse in patients (range 17–37%). Comparing in-between patients, significant differences were found in PROMs and function (IMA) with fracture patients walking slower (−30%) or showing more sway during one-legged balance (+250%). Mean activity scores (steps, bouts, cadence) were also less for fractures (p>0.05). During recovery, PROMs improved significantly, while function (IMA) and activity (AM) only showed small or selective improvements (e.g. 8° reduction in bending angle during ST, 20% increase in cadence). Only weak to moderate correlations were found between PROM, IMA (r-range: 0.32–0.79) and AM parameters (r-range: 0.31–0.71).
Discussion & Conclusion
Function, physical activity and PROMs were impaired in patients with ankle injuries compared to healthy subjects, with the fracture group performing worse on IMA and PROMs than the sprain group. In this first study ever to measure activity in ankle patients, it was shown that e.g. fracture patients walk ca. half as many steps and bouts as age-matched controls even though at 6weeks they were unable to walk 20m without walking aids (floor effect of IMA function test). The lack of improvements in IMA and AM during recovery while PROMs improved significantly, together with the few and only weak-to-moderate correlations between PROMs, IMA and AM shows that subjective and objective recovery differ and highlight the added value of IMA and AM for objectifying routine clinical outcome assessment.
