Poster Presentation Sydney Spinal Symposium 2023

Characterising the pathological gait signatures of degenerative lumbar spine diseases using inertial wearable sensors: an observational study (#47)

Pragadesh Natarajan 1 2 3 , R Dineth Fonseka 2 3 , Monish M Maharaj 1 2 4 , Lianne Koinis 2 , Ralph J Mobbs 1 2 3 4
  1. NeuroSpine Surgery Research Group (NSURG), Sydney, Australia
  2. Wearables and Gait Assessment Research (WAGAR), Randwick, NSW, Australia
  3. Faculty of Medicine, University of New South Wales, Sydney, NSW, Australia
  4. Neurospine Clinic, Prince of Wales Private Hospital, Randwick, NSW, Australia

Introduction
Degenerative diseases of the lumbar spine are associated with quantitatively altered gait patterns. Recent advances in wearable accelerometry provide an inexpensive and convenient means of objectivelyassessing gait in the clinical setting. Pathological ‘gait signatures’ are yet to be established for lumbar spine pathologies by wearable sensor-based quantitative gait analysis.


Purpose
To examine the quantitative gait patterns associated with lumbar disc herniation (LDH), lumbar spinal stenosis
(LSS) and mechanical low back pain (MLBP) using a chest-based inertial wearable sensor. ‘Gait signatures’ compared to an age-matched control population, and reported as statistically significant mean difference (%) from ‘normative’ gait parameters. 

Methodology

Procedure: Participants fitted at the sternal angle with inertial measurement unit (MetaMotionC, Mbientlab Inc.) and walked unobserved at a self-selected pace for 120m along an obstacle-free, carpeted hospital corridor.
Gait metrics: Spatial, temporal, asymmetry and variability parameters of gait were compared with age-matched (+/- 2 years) control participants recruited from the community. Sensor accuracy: Validated in LSS and healthy controls.

 

Results
No significant differences in age, body mass index, smoking and
diabetes between lumbar spine and control groups.
All lumbar spine groups had spatiotemporal increases to step time,
stance time, swing-time, double-support time and single-support
time with decreases in gait velocity and step length. Pathological
gait signatures were unique between groups (Figure 2).
LDH group involved marked asymmetry, with step length asymmetry
(+39.1%, p=0.018), step time asymmetry (+23.0%, p=0.026), singlesupport asymmetry (+35.1%, p=0.016). LDH group also involved
variation in step length (+29.0%, p=0.029).
CMLBP group involved no asymmetry but marked variability in
particular metric: single support time (+49.0%, p=0.031).
LSS group involved both asymmetry (+24.9%, p=0.039) and
variability (+36.3%, p=0.043) in step length.

 

Conclusions
Wearable-based gait analysis is capable of detecting gait abnormalities in lumbar spine pathologies such as LDH,
LSS and CMLBP. Subtypes have unique 'pathological signatures' of gait impairment.