Background/Purpose:   Somatosensory function has been shown to be altered in knee osteoarthritis (OA); for example, we previously observed that in knee OA there are generalized deficits of vibratory perception in the upper and the lower extremity compared to age-matched controls, though it remains unclear whether these deficits worsen during the course of OA. Biomechanical interventions that reduce dynamic loading during gait are promising strategies that may improve pain and retard disease progression, though their relationship to somatosensory function is unknown. Here, we present data from a double-blind randomized controlled study of a flexible shoe (“mobility” shoe), which has previously been shown to reduce medial compartment knee loading, on the progression of vibratory deficits in participants with knee OA. 

Methods:   Participants were enrolled in a randomized controlled study of a flexible-soled shoe (“mobility” shoe) vs an identical appearing “control” shoe with stiffer sole.  This was a 6-month primary study and participants had the opportunity to continue for an additional 18 months.  Here we evaluate the interim 48 week data.  All participants had radiographic (KL grades ≥ 2) and symptomatic (at least 30/100 mm on VAS) medial compartment knee OA.  In those with bilateral OA, the most symptomatic side was considered the “affected” side.  Investigators and participants were blinded to shoe assignment.  Vibratory perception threshold (VPT) was measured using a biothesiometer that provided vibratory stimulation at multiple predetermined anatomic sites, including an upper extremity site.  VPT was recorded as the first sensation of vibration (volts).

Results: 50 participants were randomized, 22 to the active “mobility” shoe and 28 to the control group.  24 participants continued with the study extension, 11 in the “mobility” group (7 female, mean age 59±4 years) and 13 in the control group (10 female, mean age 52±9 years).   Changes in VPT at the various sites are summarized in Table 1.  VPT worsened significantly over 48 weeks at each site in the control group except the medial ankle, but there were no significant changes in the mobility group.   Changes in VPT were not correlated with baseline age or with pain improvement.  

Conclusion:   Lower extremity OA is known to be associated with altered somatosensory function, including vibratory perception deficits.  Here we demonstrate that in knee OA, vibratory perception worsens significantly over time. In addition, alteration in footwear with sustained mechanical unloading during gait appears to arrest that progression of vibratory loss both in the affected lower extremity as well as in the upper extremity in such patients. The clinical significance and relationship of these findings to OA disease progression need better elucidation in future studies.

 Table