Walking Speed as an Alternative Measure of Functional Status in Patients with Lumbar Spinal Stenosis

OBJECTIVE: The main objective was to compare estimated walking perimeter (WP) and actual WP during a free walking test (6-minute walk test [6MWT]) in patients with lumbar spinal stenosis (LSS). The second objective was to describe the correlation between measured gait parameters and functional parameters. METHODS: This was a single-center prospective study including 38 patients with symptomatic LSS (23 men, 15 women; mean age, 69.3 +/- 7.9 years). The intervention was the 6MWT. For the main outcome measure, patients were first asked to verbally estimate their WP. Then, WP was evaluated using a 6MWT at a self-selected speed. In the absence of need to stop, and if a distance of 500 m was not reached, the test was extended to this distance. Specific functional scores (Oswestry Disability Index and Quebec Back Pain Disability Scale) were recorded, and a quality of life questionnaire was completed. RESULTS: WP was estimated to be less than 500 m in 21 of 38 patients, but only 7 patients actually stopped walking before reaching 500 m. The median estimate error in these patients was 200 m (interquartile range, 65-250). The shorter the estimated distance was, the greater the estimation error (r = -0.63, P = 0.002). The average walking speed was slow. Functional parameters (Oswestry Disability Index and Quebec Back Pain Disability Scale) were weakly and inversely correlated with real WP (r = -0.44 and r = -0.31, respectively) and moderately inversely correlated with measured walking speed (r = -0.51, P <= 0.001 for both). CONCLUSIONS: Direct measurement of free walking speed should be considered as a valid functional assessment in current practice for patients with LSS instead of estimated WP. To assist therapeutic decision-making, the most relevant type of walking test (duration, distance, velocity) needs to be determined.