Disclosures: Anne Kuwabara: Nothing to disclose

Objective: There are a number of clinical tests used to evaluate gait, including 40 meter Fast-Paced Walk Test (40mFPWT), 6-minute Walk Test (6MWT) and Self-Paced Walk Test (SPWT). The goal of this study was to 1) determine which gait features are most sensitive to distinguish between disease groups and 2) determine which functional test resulted in the identification of the greatest number of significant gait features that distinguish between disease groups.

Design: We collected multi-site inertial measurement (IMU) data during three walking tests for subjects with lumbar spinal stenosis, knee osteoarthritis and matched controls.

Setting: Outpatient

Participants: Thirty patients with lumbar stenosis and knee osteoarthritis compared to healthy controls with equal distribution of patients per cohort.

Interventions: N/A

Main Outcome Measures: We collected multi-site inertial measurement (IMU) data during three walking tests including 40mFPWT, 6MWT, and SPWT. Analyses performed include: Feature selection, test with Loess regression and non-parametric ANOVA.

Results: There were a number of gait parameters that clearly differentiated between disease populations and controls. The strongest feature for differentiating between groups was median stance time. Other features that were able to differentiate between conditions across all three walking tests included: mean cycle duration, cadence, double support %, gait speed, and toe clearance (min/max). In terms of the walking test, the greatest number of differentiating parameters were uncovered from the SPWT.

Conclusions: This data clearly demonstrates that gait parameters, measured using a simple IMU (equivalent to the hardware in a smartphone) can clearly differentiate between knee osteoarthritis, lumbar spinal stenosis and control subjects. The SPWT produced the greatest number of differentiating features, indicating that a longer walking test (beyond 40 meters or 6 minutes) may be necessary to discern the true functional barriers in diseased populations. Smartphone applications that collect gait parameters could be utilized to collect real-time functional information about patients that improves diagnostic and treatment precision.

Level of Evidence: Level III

« Back to AAPM&R Annual Assembly 2019

PM&R Meeting Abstracts - https://pmrjabstracts.org/abstract/gait-features-for-discriminating-between-mobility-limiting-musculoskeletal-disorders-lumbar-spinal-stenosis-and-knee-osteoarthritis/