Objective

Gait disorders is the most common and disabling dysfunction in patient with multiple sclerosis (PwMS). Several studies have been using gait parameters to determine the interventional effects in PwMS. However, few studies interested initially to determine reproducibility of these parameters which could overestimate effects. The aim of the study was to examine the reproducibility in different spatio-temporal parameters of gait measured during short and long distances.

Material/patients and methods

Fifty-eight PwMS (EDSS 4–7) and 19 healthy peers were included in this study. The gait evaluation was carried out using the GaitRite system (CIR System Inc., USA). For short distance, patients walk on a 25 foots walkway in a fixed order in the follow conditions: self-selected comfortable speed, maximum speed and dual task (arithmetic calculation). Then, they performed a Timed Up and Go test (TUG) and a 6 Minutes Walk Test (6MWT). Two identical measures were conducted at one-week interval without therapeutic intervention. For the reproducibility evaluations a Student t -test (for systematic error), Intra-class Correlation Coefficients (ICC) and Minimum Detectable Change (MDC95) were calculated.

Results

The absence of systematic error was found only for the gait in fast speed condition. The reliability of the gait parameters had moderate to high ICC values (all conditions ICC > 0.7). The MDC95 was systematically higher in PwMS compared to healthy persons. MDC95 was high for gait speed in all conditions (> 34% – PwMS), TUG (51.5% – PwMS) and 6MWT (31.68% – PwMS). The smallest MDC95 (3.2 – healthy and 6.07 – PwMS in comfortable speed) was found for stance time parameter for both populations and all conditions.

Discussion–conclusion

Despite systematic bias were found for several conditions, the outcome measures were high reproducible across 1 week in the most parameters. MDC95 values are high in PwMS probably due to a higher variability associated with the disease. Stance time which was the best parameter in terms of MDC could be choose as a pertinent outcome of gait improvement. This study could improve the knowledge of in PwMS’ gait evaluation as it shows good reproducibility for different conditions in a large sample size.