Fatigue is highly reported in people with MS (pwMS), with a prevalence up to 80%, showing a high impact on quality of life (Kluger et al., 2013; Filippi et al., 2018). Recent taxonomies have indicated two types of fatigue: trait fatigue and state fatigue. Trait fatigue is referred to as an overall perception of fatigue over a longer time which is frequently assessed by questionnaires such as the modified fatigue impact scale (MFIS). State fatigue is activity dependent. State fatigue can be either assessed by asking the perceived fatigue by a visual analog scale (VAS), or assessed objectively by performance decline during, for example, maximal strength contractions or walking (Kluger et al., 2013; Van Geel et al., 2019). So far, research on motor fatigability has often been based on evaluation of the capacity to maintain maximal isometric muscle force measured on the body function level of International Classification of Functioning, disability and health (ICF); or the susceptibility to muscle fatigue (Severijns et al., 2017). With the use of the twitch interpolation technique, it has shown that motor fatigability in MS is mainly influenced by central factors such as reduced voluntary central drive. These theories assume a larger contribution of central than peripheral neural activation to motor fatigability (Steens et al., 2012; Bachinger et al., 2019; Severijns et al., 2019).

In recent years the focus of the assessment of motor fatigability have been changed from body function level of ICF to walking at ICF activity level. Walking-related fatigability has been quantified in pwMS by comparing walking velocity during different paradigms (Phan-Ba et al., 2012), and more recently, by comparing the last vs. the first minute of the 6 min walking test (6MWT), quantified by a “distance walked index” (Leone et al., 2016; Van Geel et al., 2020). Walking-related fatigability limits the mobility of pwMS in daily life, and is prevalent in up to half of pwMS with moderate to severe impairments (Leone et al., 2016). During the 6MWT, pwMS perceive more balance problems and gait impairments, and report overall higher state fatigue levels across minutes (Van Geel et al., 2021).

Walking requires accurate and consistent antiphase interlimb coordination pattern, to ensure safe ambulatory functions. Interlimb coordination involves a distributed network at both supraspinal and spinal level (Swinnen, 2002). It has been suggested that defaults in the central nervous system (Debaere et al., 2001; Swinnen, 2002; Swinnen and Carson, 2002), such as muscle firing timing, slowed motor nerve conduction time and reduced voluntary drive by the motor cortex (Swinnen and Carson, 2002; Wurdeman et al., 2011; Swanson and Fling, 2018) may contribute to the impaired coordination. When walking at self-selected and fast speed, pwMS with mild to moderate disability shows a worse interlimb coordination pattern than healthy controls (HC) (Richmond et al., 2020). Moreover, when performing a 6-min walking task, interlimb coordination pattern deteriorates each minute (Plotnik et al., 2020). Interlimb coordination in pwMS may be impacted by the presence of heterogenous symptoms, ranging from impaired dynamical balance, decreased muscle strength, asymmetric gait, compensatory movements, or perception of fatigue or performance fatigability. Therefore, a seated 6-min coordination task (6MCT) was designed to mimic the 6-min walking test (6MWT) but with minimalized balance requirements or major muscular effort, which are required during a regular walking task. Our main hypothesis is that a change in coordination pattern would be observed during the seated 6MCT, by the end of the task as a result of fatigability. This change is hypothesized to be more present in a group with walking-related performance fatigability, as both walking fatigability and coordination are assumed to be mostly (but not exclusively) influenced by central neural activation mechanisms.

The aim of this study was consequently to examine more fundamental antiphase interlimb coordination pattern while seated in pwMS, with and without walking fatigability in comparison to healthy controls.

This is the first study that investigates an antiphase coordination pattern of the lower limbs during a seated 6 min coordination task (6MCT) in pwMS and changes over time. The seated 6MCT was designed to examine more fundamental interlimb coordination of the lower limbs, assuming minimized balance and strength demands, which are required during walking. Our main hypothesis was that a change in coordination pattern over time, due to performance fatigability, would be observed during seated 6MCT in pwMS with WF, by means of change in antiphase coordination movements (ABS, CV, PCI), and decrease in spatial and temporal movement parameters (movement amplitude, movement frequency). We hypothesized this change, as a similar trend of deterioration in coordination pattern over each minute have been observed in pwMS while performing a 6-min walking task (Plotnik et al., 2020).

The exploratory results of this study showed that interlimb coordination changes over time, observed by an increase in their variability in generating antiphase left-right movement CV(φ), toward the end of the task. As expected, this change in variability was greatest in pwMS with WF. Besides the observed change over time, the mean values of CV(φ) and PCI differed significantly between HC and pwMS with WF, with worse values in pwMS with WF. Noteworthy is that ABS(φ) in antiphase generation did not differ between group and did not changed significantly over time. These results suggest that pwMS with WF have more difficulties in the control of seated antiphase coordination of the lower limbs compared to HC.

Regarding the temporal and spatial movement parameters during the 6MCT, we expected to observe a similar trend of decline in performance in movement frequency or amplitude in pwMS with WF, as they show a decrease in walking distance during the 6MWT. As the 6MWT requires not only a consistent and accurate antiphase coordination pattern, but also balance and strength, the 6MCT was designed to minimalize balance requirements or major muscular effort, and therefore in theory stresses more central neural mechanism. This 6MCT was based on a similar test in previous research to measure phase coordination deficits, in stroke, elderly and Parkinson (Swinnen et al., 1997; Serrien, 1998; Serrien and Stelmach, 2000). We found that only in pwMS with WF, the movement amplitude significantly decreased over time, and not movement frequency. One of the first studies that used this interlimb seated coordination task in Parkinson disease found significant differences in movement amplitude compared to HC. Temporal measure (cycle duration) and absolute error (ABS) did not significantly differ between patients and HC. Together, these results suggest that during a seated coordination task, the temporal movement parameter is less influenced by neurological diseases, like pwMS and Parkinson disease, than spatial movement parameters. However, direct comparison is not possible due to difference in population, but also in length of the task (24 trials of 15 s vs. sustained 6 min) and performing the task while listening to a metronome (1 Hz) (Swinnen et al., 1997).

In the WF group, the change in perceived state fatigue during the 6MCT was lower than during the 6MWT, indicating that the seated task was perceived less fatiguing than walking. However, despite low levels of perceived state fatigue, the test detected differences in antiphase interlimb coordination pattern between groups. These results suggest that the 6MCT can be used as an alternative for patients who are not able to perform the 6MWT due to extreme fatigue or having severe motor disabilities. This assumption should, however, be further investigated, as the 6MCT cannot be directly compared to walking, as explained further. We note that, although pwMS without walking fatigability did show walking impairments assessed by the T25FWT and 6MWT, no significant differences were found in seated interlimb coordination performance, nor in change in perceived state fatigue, between pwMS in the NWF group and HC. Besides, we note that abnormal trait fatigue was present in many pwMS with WF (see Table 1, mean MFIS = 45), assessed by the MFIS. However, no significant difference in trait fatigue was found in pwMS with or without walking-related performance fatigability, and thus trait fatigue seems to be rather a general MS characteristic. This result may suggest that trait fatigue do not directly relates to walking-related performance fatigability. These results highlight the importance of interlimb coordination in walking-related performance fatigability.

In contrast to 6MCT, walking is for example characterized by different sensory-motor interactions, different and higher muscle activity, higher dynamic balance needs, metabolic cost, higher automaticity, etc., which all seems to be different in the 6MCT. Moreover, Meesen et al. (2005) found that movement with the head is a contributing factor for disturbance of coordination. Serrien and Stelmach (2000) confirms that movements integrating different limbs show more coordination effort in elderly. Future research could investigate these perturbations and multi-limb coordination patterns during the seated coordination test, to increase the ecologically validity of the 6MCT toward walking. The 6MCT should therefore be considered as a model that allows (in theory) isolation of the central neural drive mechanisms, taking lower limbs into account.

Future studies, should take the following limitation of this study into account. This research instructed participants to perform as many movements as possible, so move as fast as possible during 6 min, similar to the instructions of the 6MWT, where patients are instructed to walk as fast as possible, and cover as much distance as they can. Therefore, movement frequency could have been prioritized and the main focus of the participants over the 6 min instead of movement amplitude. The decrease in amplitude might have been a compensation strategy to keep the frequency equal, and perform as many movements as possible. Future studies could focus on different instructions and look at the possible influence of prioritization of amplitude or frequency. However, movement frequency is often prioritized in other tasks of the locomotor system, wherefore we do not think that the instruction has exclusively influenced our results (Malone et al., 2012; Motl et al., 2012; Comber et al., 2017).

Methodological considerations apply. The small sample size, especially in subgroups, and no a priori power analysis, should be considered when interpreting the results. A power analysis could not be made given it was the first time that the task was used in persons with MS. Besides, no one-to-one matching was performed between HC and MS groups. Next, we cannot exclude the impact of the relative weight of the levers, muscle strength asymmetries, accuracy of axis alignment of the device during the task, the liberty of comfortable range of motion of the legs and optional use of back support on differences in task performance between participants. Because of individualization purpose, participants were not restricted in their movement amplitude, nor externally paced in their movement frequency, similar to 6MWT were the step length and cadence are not standardized. One may argue that muscle strength asymmetries may have impacted the results. In this research, maximal lower limb strength was unfortunately not recorded given strength requirements in the seated coordination task were limited and likely not importantly affecting interlimb coordination control. Still, future research is advised to include maximal lower leg muscle strength as strength asymmetry might be present in pwMS (Farrell et al., 2021) and can then be included as factor in statistical analyses. Further research should also elaborate on more standardized task constraints regarding movement amplitude or correct for variability in movement amplitude to avoid that variability expressed in absolute values could influence group comparisons.

Our exploratory results suggest that the antiphase coordination of the lower limbs change over time while performing a seated 6-min coordination task. These changes were apparent by an average higher coordination variability and a decrease in movement amplitude. The observed changes where more prominent in pwMS with walking-related performance fatigability. Results suggest indirectly the influence of central mechanism contributing to coordination patterns in the walking-related performance fatigability group. Further studies in a powered sample size with new methodological considerations are warranted to determine the relative contribution of centrally driven interlimb coordination and walking fatigability.

The raw data supporting the conclusions of this article will be made available by the authors after request.

The studies involving human participants were reviewed and approved by the Ethical Committee of Hasselt University, MS Centre Overpelt and Hospitalo-Facultaire Universitaire de Liège approved the protocols (B707201835771 and B9115201836892). The patients/participants provided their written informed consent to participate in this study.

FV was responsible for study conceptualization, data collection, data processing, and writing of the manuscript. MGo was responsible for data analyzing, data interpretation, and writing of the manuscript. MGe was responsible for data processing. RM was responsible for equipment conceptualization, data processing, and data interpretation. LT and LM were responsible for data interpretation and writing of the manuscript. PF was responsible for study conceptualization, data interpretation, and writing of the manuscript. All authors contributed to the article and approved the submitted version.

PF was steering committee member of Neurocompass, participated to advisory board meetings of BIOGEN IDEC, and received teaching honoraria for EXCEMED and PARADIGMS. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.