Polyneuropathies (PNP) cover a group of diseases that primarily cause damage to peripheral nerve fibers in symmetric, distal, length-dependent “glove and stocking” distribution (Findling et al., 2018). Affected patients suffer from sensitivity impairments that can manifest in paresthesia, dysesthesia, numbness, and pain in the hands and feet (Hanewinckel et al., 2016; Alam et al., 2017). PNP is also associated with reduced muscular strength (Rotta et al., 2000; Ferreira et al., 2017). This may directly result from damage to motor nerve fibers in specific types of PNP or advanced disease settings or indirectly from a lack of movement due to the aforementioned sensitivity symptoms (Andreassen et al., 2006; Martinelli et al., 2013; Hafsteinsdottir and Olafsson, 2016; Broers et al., 2019).

Functionally speaking, PNP significantly raises the risk of postural instability, meaning balance problems and gait disturbances (Mustapa et al., 2016; Kneis et al., 2020b). For example, in older populations, PNP accounts for around 18% of gait disorders (Pirker and Katzenschlager, 2017) and raises the probability of falling (Cavanagh et al., 1992; Richardson and Hurvitz, 1995; Stolze et al., 2004; Hanewinckel et al., 2017; Findling et al., 2018). PNP can therefore reduce both our everyday mobility and quality of life considerably (Hoffman et al., 2015; Win et al., 2019) and strain healthcare resources (Gordois et al., 2003; Pike et al., 2012; Mengel et al., 2018). Standardized clinical tests usually include gait and balance tasks to assess one’s mobility status and fall risk. Observation of these tests can be refined by motion capture systems enabling specific analysis, e.g., about gait abilities. It is well known that people with PNP habitually walk slower, take shorter steps, and reveal more gait variability than age-matched healthy individuals (Wuehr et al., 2014; Hanewinckel et al., 2017; Marshall et al., 2017; Kneis et al., 2020a). However, it remains unclear whether the altered gait, especially in terms of slowness, is caused by the primary deficit, i.e., damaged peripheral nerve structures, or whether it is a secondary strategy to prevent accidents such as falls (Courtemanche et al., 1996; Dingwell et al., 2000; Dingwell and Cavanagh, 2001; Menz et al., 2004; Wuehr et al., 2014). As PNP makes evident, damage to peripheral nervous structures leads to proprioceptive impairments (Hanewinckel et al., 2016; Findling et al., 2018). Reliable proprioceptive feedback is essential for precise posture control, as it contains information about alterations via tendon or muscle lengths and joint angles (Allum et al., 1998; Dietz, 2002; Masani et al., 2003; Shaffer and Harrison, 2007). Disturbances in the proprioceptive signals thus imply inaccurate motor control mechanisms (Dietz et al., 2002; Pearson, 2004). In PNP, this somatosensory deficit is believed to promote the aforementioned balance and gait disturbances. However, its exact interrelation with the slowing of gait remains unclear (Courtemanche et al., 1996; Dingwell et al., 2000; Menz et al., 2004; Wuehr et al., 2014). Besides gait and balance, the functional performance of upper limbs also suffers from PNP. There is evidence of less accuracy and slower execution speed during goal-directed arm movements (Hondzinski et al., 2010), as well as reduced functional hand performance during fine motor tasks compared to healthy individuals, while strength capacity seems unaffected (Kender et al., 2022). Until now, there has been no known whole-body gait movement analysis in PNP considering both upper and lower-body motor behavior. We assume that altered motor pattern in the lower limbs while walking is also reflected in the upper limbs (Wannier et al., 2001; Meyns et al., 2013; Pearcey and Zehr, 2019).

Our approach in the present study is to capture the whole-body movement of PNP patients in a standardized mobility test acknowledged as clinically relevant, i.e., the Timed Up and Go (TUG) test (Salarian et al., 2010; Christopher et al., 2019). The TUG test covers essential demands of daily living (standing up and sitting down, accelerating and decelerating walking, turning around) and, thus, stimulates different body systems responsible for posture stability, coordinated movements, and force development. TUG was originally done at a preferred movement speed (Podsiadlo and Richardson, 1991), operating as a reliable and approved sign of vitality (Studenski et al., 2011; Perera et al., 2016). For a more differentiated perspective of PNP-related sensorimotor impairments, we will add two further TUG conditions that challenge patients’ executive resources: executing TUG while counting backward to provoke cognitive-motor interference (Montero-Odasso et al., 2012; Montero-Odasso and Hachinski, 2014; Bayot et al., 2018; Schniepp et al., 2019), and at a fast movement speed to assess acceleration capacities associated with disability and functional reserve capacity (Wuehr et al., 2014; Artaud et al., 2015; Middleton et al., 2015). For motion behavior analysis, we will extract common gait parameters and additionally focus on joint velocities across the whole body in comparison to matched healthy control individuals. As the first step, we aim to verify that the TUG performance of PNP patients falls below that of healthy individuals. We assume that PNP-related gait alterations will reflect the poorer performance level in each TUG condition. Furthermore, we hypothesize that velocity across the whole body can characterize PNP-specific motion patterns depending on the TUG condition and joints analyzed. We believe that our approach will enable us to derive additional velocity-based parameters that will help us better understand motor behavior and movement organization in neurological diseases.

No adverse events occurred during the tests, and all participants performed all test conditions. We included data from N = 40 participants (20 PNP:20 CG) in our analysis. The comparative groups PNP and matched CG exhibited similar anthropometric parameters (Table 1). Participants’ characteristics revealed significant group differences in fall incidence, maximum walking distance, fear of falling (FES-I), mobility performance (Tinetti POMA; Table 1), medicine intake, and chronic disabilities (Supplementary Table S1). All included patients had relevant PNP symptoms (Table 2).

The main objective of the present study was to analyze the motion patterns of PNP patients along the body axis for a deeper understanding of altered motion behavior caused by PNP. We assessed PNP patients’ performance in a clinically relevant functional test, i.e., the Timed-Up-and-Go test (TUG), by comparing gait-related parameters and joint-velocity profiles across the whole body to a group of matched healthy individuals. As hypothesized, PNP patients performed TUG slower than the control group during all test conditions, i.e., at preferred and fast movement speed and while executing a cognitive task (dual-task condition). More specifically, we found a PNP-related slowed gait pattern during TUG execution determined by reduced gait speed, shorter steps, and prolonged step time, as well as lower mean velocities in all other measured body joints. Interestingly, the slowing factor of PNP patients’ joint velocities was independent of the respective joint and, therefore, individual mean joint velocity, which tends to vary across the body. Moreover, this factor applied to each TUG condition similarly and amounted, on average, to 0.87 compared to the healthy control group.

As expected, PNP patients’ overall TUG performance fell below that of the healthy control group across all conditions. Patients needed approximately 20% more time to complete TUG independently of the test condition. By extracting gait-related parameters, we detected a slowed walking pattern in PNP (preferred: 80.7 cm/s; fast: 134.7 cm/s) compared to the CG (preferred: 109.7 cm/s; fast: 148.8 cm/s) determined by shorter steps (preferred: 54 cm vs. 59 cm) and prolonged step time (preferred: 0.60 s vs. 0.57 s). These results concur with other studies investigating PNP-related gait impairments (Fernando et al., 2013; Wuehr et al., 2014; Marshall et al., 2017), identifying 139 cm/s for fast walking (Wuehr et al., 2014), and 93–110 cm/s for preferred walking in PNP patients (Menz et al., 2004; Wuehr et al., 2014; Marshall et al., 2017). In contrast to other studies (Petrofsky et al., 2005; Brach et al., 2008; Wuehr et al., 2014; Brown et al., 2015), we failed to observe a wider gait base in PNP patients than in healthy individuals. Basically, the TUG condition significantly affected TUG performance in both groups in a similar way but at a different speed level.

These findings also apply to joint velocity profiles in PNP compared to CG. As with gait-related parameters, we used the walking sequences of the TUG to extract the mean joint velocities of ankles, knees, hips, shoulders, elbows, and wrists in body coordinates, respectively. Across the whole body, PNP patients reduced, on average, their joint velocities with a factor of 0.87 compared to CG. Interestingly, the velocity reductions occurred independently of the respective joint or test condition. This means that in each of the three test conditions, the average joint velocity of each joint was reduced by 13% in PNP patients compared to healthy individuals. Furthermore, joint velocities (of PNP and CG) were in a linear relationship to each other, and gait speed (COM velocity in space) referred to the respective test condition (Figure 2B). We deduce that despite PNP-related slowing, patients maintained movement patterns in terms of velocity distributions across joints similarly to healthy individuals.

PNP affects the peripheral nerves in a distal symmetric distribution that results in impaired somatosensory information. Especially proprioceptive feedback is essential for posture stability (Dietz, 2002) as proprioceptors provide continuous feedback about joint positions and velocity, thus facilitating our orientation in space with respect to the ground and the different body segments (Lackner and DiZio, 2005; Shaffer and Harrison, 2007). The proprioceptors’ signal carries velocity information relying on changes in lengths (tendons or muscles) or the joint angles that are primarily used to control motion accurately (Allum et al., 1998; Dietz, 2002; Masani et al., 2003; Shaffer and Harrison, 2007). Because of their proprioceptive deficit, PNP patients often suffer from postural instability, which manifests in balance problems (Horlings et al., 2008; Sawacha et al., 2009; Brown et al., 2015; Kneis et al., 2016, 2020b; Mustapa et al., 2016) and gait disturbances (Allet et al., 2008; Sawacha et al., 2009; Shin et al., 2021) and may raise the risk of falls (Stolze et al., 2004; Allet et al., 2008). Patients often experience a loss of muscular strength also due to the PNP-induced impaired sensorimotor interplay (Andreassen et al., 2006; Martinelli et al., 2013; Ferreira et al., 2017). A slower gait speed, including shorter steps and longer step times, may thus become manifest from a diminished propulsion capacity, but it may also follow PNP’s safety management (Dingwell et al., 2000; Mustapa et al., 2016; Findling et al., 2018). Slower, shorter steps imply a relatively higher proportion of the double support phase associated with greater stability (Williams and Martin, 2019). Vice versa, minimizing the support surface during the single-leg stance phase requires more effort to maintain balance, thus challenging the postural system (Williams and Martin, 2019). Furthermore, reducing movement speed may give patients more time to generate an adequate response (Williams and Martin, 2019) and thus contribute to movement accuracy despite the proprioceptive deficit. We are assuming that PNP patients’ slower gait implies the slowdown of upper limbs’ joint velocities according to the hypothesis of strong linear velocity interlimb coupling (Orsal et al., 1990). As mentioned above, our PNP patients showed velocity distributions across joints resembling those of healthy individuals, which corresponds to the manifestation of PNP-induced nerve damage -that is, the symmetrical pattern across extremities.

Despite PNP patients’ reduced overall velocity, we conclude that velocity control mechanisms are largely intact in PNP (Zehr and Duysens, 2004; Meyns et al., 2013). We even propose that slowing movement velocity is potentially a secondary compensatory safety strategy rather than one triggered by the primary physiological deficit. We base this assumption on the fact that our PNP patients were able to adjust their movement speed situationally like healthy individuals but move within a lower individual speed zone. For example, patients increased their movement speed when asked to, as in the fast TUG condition. Furthermore, the concurrent execution of a cognitive task led to the slowing down of movements in PNP to the same extent as in healthy individuals, but again at a lower level. There is evidence that multi-task conditions require an allocation of attentional resources to each task that often results in slowed movement speed (Hausdorff et al., 2008; Montero-Odasso et al., 2012). In general, PNP patients may require greater effort in order to allocate more cognitive resources than healthy individuals to generate well-coordinated movements despite their PNP-related impairments (Courtemanche et al., 1996). We, therefore, suggest that PNP patients scale their postural control strategy along with the quality of the sensory signals they receive. This supposition is supported by findings of studies addressing stance control in PNP patients. We found that the postural behavior of PNP patients is modifiable by an exercise intervention; that people with PNP performed a sensory reweighting and an adjustment of velocity control towards the postural behavior of healthy individuals (Kneis et al., 2019, 2020b).

Our assumption of PNP patients’ preventive safety strategy is highlighted by the correlation between their individual slowing factor with their fear of falling (FES-I), mobility performance (Tinetti POMA), and self-reported maximum walking distance. We suggest that the individual slowing factor measures the degree of PNP mobility impairment.

The 3-meter walking sequences of the TUG test cover a relatively short distance for gait analysis, encompassing both acceleration and deceleration phases. While we obtained plausible values for the extracted gait parameters, our experimental setup did not adhere to common guidelines for gait analysis. Specifically, it deviated from the standard practice of continuous walking over a minimum distance of 10 to 20 strides, as outlined in the literature (Hollman et al., 2010, 2011). Furthermore, our step detection methodology relied on velocity thresholds derived from our own validation work, rather than direct force plate measurements, which are considered the gold standard (Bilney et al., 2003; Steinert et al., 2019). These factors, coupled with the limited sample size, emphasize the exploratory nature of our research, and underscore the need for a cautious interpretation of the results. The diagnostic utility of the TUG test lies in its multidimensional assessment of functional performance. However, in this study, we deliberately focused only on the walking sequence of the TUG test, excluding other motion tasks such as standing up, turning, and sitting down. These additional sequences are vital for a comprehensive understanding of the significance of the TUG test. In the future, we plan to incorporate these sequences into our motion analysis. Moreover, we will also extract additional biomechanical parameters such as joint torques and moments, offering a more comprehensive view of motor behavior during the TUG test.

Our results indicate that PNP patients reduce all their mean joint velocities similarly, resulting in a patient-specific slowness factor (on average 13%) compared to healthy individuals, regardless of joint or condition. This slowing of all joints in body coordinates correlates strongly with the individual gait speed in space coordinates (COM velocity). We assume that this global slowing is caused by the reduced quality of proprioceptive signals. We maintain that PNP patients’ altered gait, as assessed by mean absolute joint velocities, is not only determined by the obvious balance requirements leading to smaller and slower steps but that it results from a shift in general speed of an otherwise undisturbed whole body motion pattern.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving humans were approved by Ethics Committee of the University of Freiburg, Germany. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

IW, SW, and CM drafted the manuscript and analyzed and interpreted the data. IW and CM designed the study. IW recruited patients and collected and analyzed data. VL wrote the code for MATLAB and analyzing the TUG parameters. All authors contributed to the article and approved the submitted version.

IW, SW, and CM were partially funded by the European Union’s Horizon 2020 research and innovation program (grant agreement no. 769574) and the European Union’s Horizon 2021 research and innovation program (grant agreement no. 101057747). We acknowledge the funding of the article processing charge by the Open Access Publication Fund of the University of Freiburg.

The 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.