Wearable sensors detect impaired gait and coordination in LBSL during remote assessments

Abstract

Background Leukoencephalopathy with brainstem and spinal cord involvement and lactate elevation (LBSL) is a rare leukodystrophy with motor impairment due to biallelic mutations in DARS2, which encodes mitochondrial aspartyl tRNA synthetase. Progressive ataxia is the primary feature. Objective The study objective is to determine the feasibility of remotely collecting quantitative gait and balance measures in LBSL. Methods The study design uses wearable accelerometers and the scale for the assessment and rating of ataxia (SARA) scale to assess gait and postural sway in LBSL and control participants' homes through video conferencing. Results Lateral step variability (LSV), which indicates stride variability, and elevation of the step at mid‐swing are increased for LBSL patients during brief walking tests. During stance with the eyes closed, LBSL participants show rapid accelerations and decelerations of body movement covering a large sway area and path. Both the LSV and sway area during stance with the feet together and eyes closed correlate strongly with the SARA. Conclusions Wearable accelerometers are valid and sensitive for detecting ataxia in LBSL patients during remote assessments. The finding of large increases in the sway area during stance with the eyes closed is intriguing since dorsal column dysfunction is universally seen in LBSL. This approach can be applied to related rare diseases that feature ataxia.

Digital Measures of Postural Sway Quantify Balance Deficits in Spinocerebellar Ataxia

“…Previous studies support our finding that greater differences in sway measures between SCA and controls can be found in standing tasks with EC or narrow stance conditions. 19,[25][26][27][28]42 These findings highlight the importance of selecting (and standardizing) conditions for postural sway analysis that are both feasible for most participants yet challenging enough to exacerbate sway in ataxia patients. Quantitative, continuous-scale measures of extent of postural sway in the least challenging condition that all our SCA cohorts could accomplish (NSEO) related better to severity of disease (SARA score) than the number of stance conditions that subjects could accomplish (see Fig.…”

“…This improved concurrent validity with increased difficulty in the stance condition is in agreement with other studies correlating sensor-derived sway measures with SARA scores. [27][28][29] Furthermore, modest correlations were observed between sway measures and patientreported outcomes (PROM Phys2, ABC, and FARS-ADL), particularly in the EO and EC natural stance conditions. Importantly, the discriminative sway measures were not correlated with age or different for men versus women (not shown).…”

“…In particular, accelerometer-derived sway path length, area, and RMS on the lower trunk have previously demonstrated sensitivity in discriminating even premanifest SCA2 from controls, with path length and sway area also showing sensitivity in early-manifest ataxia. 27,28 Additionally, the postural sway path length from the estimated body center of gravity, measured using digital cameras, has shown to be sensitive in distinguishing premanifest ataxia in genetically confirmed SCA from controls. 25 Earlier studies, prior to genetic confirmation of SCA, used force plates to characterize postural sway in patients with ataxia with similar sway measures as sensor-based sway measures and demonstrated similar, excessive sway due to cerebellar involvement.…”

“…9 Several studies have demonstrated that accelerometer-based sensors worn on the lower trunk are effective in discriminating sway patterns of groups with ataxia from healthy control (HC) groups. [25][26][27][28][29] Digital sway metrics derived from inertial sensors also show a correlation with genetically estimated time to ataxia onset, 25,[27][28][29][30] but these studies have been limited to small groups of individuals with ataxia. Although sway measures are extremely sensitive to specific test conditions, it is unclear which sway test conditions and digital sway measures best serve as sensitive and reliable outcomes that are related to disease severity and patient-reported outcomes.…”

“…Body‐worn accelerometers and/or gyroscopes are inexpensive, compact, and easily donned, and can automatically provide quantitative postural sway measures that are accurate, valid, reliable, and sensitive to neurological disorders 9 . Several studies have demonstrated that accelerometer‐based sensors worn on the lower trunk are effective in discriminating sway patterns of groups with ataxia from healthy control (HC) groups 25‐29 . Digital sway metrics derived from inertial sensors also show a correlation with genetically estimated time to ataxia onset, 25,27‐30 but these studies have been limited to small groups of individuals with ataxia.…”