Authors: Henry Dunne¹, Thomas Stone¹, Robin Borchert¹, George Choa¹, Daniel Nicolae², Terry Fawden², Iwan Roberts², Manohar Bance², Andrew Gee²
¹Cambridge University Hospital / Cambridge University NHS Foundation Trust
²University of Cambridge
Background: Benign Positional Paroxysmal Vertigo (BPPV), the commonest cause of vertigo, is caused by displacement of otoliths into the semicircular canals. A particle repositioning manoeuvre/Epley Manoeuvre (PRM) is used to treat this – by taking a patient’s head through positions that sequentially align segments of the posterior canal with the vertical plane, canaloliths are directed out the canal back into the utricle. It is rarely performed outside of specialist ENT/Vestibular therapy clinics, leading to treatment delays. The manoeuvre assumes inter-subject homogeneity of the canal’s orientation. However, the degree of this consistency has not yet been quantified. Here performed an anatomical study to assess variation in the semicircular canal orientation. We then used computational modelling to examine the alignment of canal segments with the vertical plane during the particle repositioning manoeuvre. We developed an earable device that records head movements in space, and in conjunction with the anatomical data, we developed an app that uses head-in-space data from the earable to facilitate visualisation of the canal orientations during PRMs.
Methods:
• 68 CT Temporal bones were segmented and measured.
• A principal component analysis was performed to create a plane of best fit for each canal and their orientations were evaluated.
• Several modes of anatomical variation were identified.
• A computational model facilitated examination of the orientation of the canal during PRM.
• Earable device with IMUs was developed, and head-in-space algorithms validated against VICON motion capture.
Results:
– Mean canal height: 8.7mm (sd=0.55) circumference 7.3mm (sd= 0.53)
– The mean angle of deviation of a canal’s plane relative to the mean orientation was 9.5 degrees (range 7-16 degrees).
– Several modes of variation of canal orientation were identified.
– The apical canal segment aligns poorly with the vertical plane during the PRM.
– Earable head-in-space tracking correlated strongly with VICON recordings.
Discussion: Here we used CT images to identify that there is a degree of inter-subject variability in posterior semicircular canal anatomy. This, and the finding that PRM fails to align the apical canal segment with the vertical plane may explain why the PRM is unsuccessful in a proportion of patients. We propose an amendment to the PRM and developed an app and earable that allows visualisation of the semicircular canals during the PRM supporting BPPV treatment. We hope this will increase efficacy of the PRM as well increase BPPV patients access to treatment by supporting non ENT specialists to perform the PRM.
