Association between perceived fatigue and gait parameters measured by an instrumented treadmill in people with multiple sclerosis: a cross-sectional study

BACKGROUND: Multiple sclerosis (MS) is a multi-focal progressive disorder of the central nervous system often resulting in diverse clinical manifestations. Symptomatic fatigue is quite common in people with MS (PwMS), with prevalence as high as 85%. Nevertheless, it remains poorly understood and its association with walking capabilities unclear. Therefore, the objective of this investigation was to examine the relationship between symptomatic fatigue and spatio-temporal parameters of gait in PwMS based on an instrumented treadmill. METHODS: One hundred and twenty-four relapsing-remitting patients diagnosed with MS, 84 women and 40 men aged 42.6 (S.D = 11.9), participated in this investigation. A convenience sample of 25 apparently healthy subjects, 15 women and 10 men aged 40.3 (S.D = 11.1), served as controls. Gait spatiotemporal parameters were obtained using the Zebris FDM-T Treadmill (Zebris1 Medical GmbH, Germany). The Modified Fatigue Impact Scale (MFIS), a self-reported questionnaire, was used to determine the level of symptomatic fatigue in the MS study group. PwMS were divided into two groups: fatigued and non-fatigued. RESULTS: Forty-four PwMS were classified as suffering from fatigue (mean MFIS = 52.0, S.D = 13.7); 80 were classified as non-fatigued (mean MFIS = 14.5, S.D = 14.5). Individuals in the fatigued group walked slower than those in the non-fatigued group; 1.7 (S.D = 2.4) vs. 2.4 (S.D = 1.0); P < 0.001, respectively. Moreover, fatigued patients took smaller steps, had a shorter stride length, prolonged stance, double support phase and a shorter single support phase compared to the non-fatigued group. In the total group, fatigue was significantly correlated with 10 (out of 14) spatiotemporal parameters of gait, however, correlation scores <0.40 were considered as weak correlations. According to step one of the linear logistic regression analysis, the temporal gait component was found to explain 5.1% of the variance related to symptomatic fatigue, R2 = 0.051, χ2 (1) = 6.511, P = 0.011. Step two of the model added the gait spatial component, thus increasing the explaining variance to 9.3%; R2 = 0.093, χ2 (2) = 12.12, P = 0.002. The asymmetry gait parameter did not contribute to the equation. CONCLUSIONS: Perceived fatigue is related to walking speed in PwMS, nevertheless its contribution to level of fatigue is limited.