Introduction: Degenerative Cervical Myelopathy [DCM] is the leading cause of non-traumatic spinal cord injury and present in around 2.2% of the adult population.1 The true prevalence is thought to be much higher, but detection and diagnosis is limited by a lack of clinician awareness of the condition.1 The earliest signs of DCM are gait changes, but no diagnostic tools assessing gait have been validated in a clinical setting.
Methods: 50 people [25 with DCM and 25 healthy controls] were recruited from a neurosurgical clinic in Sydney, Australia and assessed using a new gait assessment tool. The gait assessment tool uses a smartphone video recording of a person's walking and then extracts their gait features. An Area Under the Curve was completed on each variable to determine a binary cut-off for a positive test. Diagnostic accuracy was then calculated for each gait feature, with the top 5 selected for inclusion into a cluster. DCM was diagnosed by a neurosurgeon using imaging [cord signal change confirmed on Magnetic Resonance Imaging] and a physical examination [i.e., presence of upper motor neuron lesion signs].
Results: 28 gait variables were assessed for their diagnostic accuracy. A cluster of the 5 most significant gait variables was found to have a sensitivity of 96%, specificity of 83%, positive likelihood ratio of 14.00 and negative likelihood ratio of 0.04. Post hoc analysis found that sensitivity could be increased to 100% if a 6th variable was included.
Discussion: This study presents promising results for a new diagnostic tool for DCM. A new diagnostic tool for DCM could improve detection of the condition and may help reduce unnecessary imaging in those without signs of DCM. Before the diagnostic tool is endorsed, it needs to be assessed in a fully powered diagnostic accuracy study.