Subclinical posterior semicircular canal hypofunction without functional or postural impact in healthy older adults: an exploratory cross-sectional study of a structure–function dissociation

<title>Abstract</title> <p> <bold>Background</bold> . An age-related decrease in the vestibulo-ocular reflex (VOR) has been identified in the posterior semicircular canals of healthy older adults in recent studies. However, the potential functional and postural repercussions have not been examined in those articles. Pathological gain cut-off values exist only for the horizontal canals, and no validated cut-off is available for the vertical canals to date. This study sought to determine whether a reduction in posterior canal VOR gain is associated with functional and postural performance in healthy older adults. <bold>Methods</bold> . In this cross-sectional study, 40 healthy adults aged 65 years and over were assessed. VOR gain was measured in the three canal planes using the video head impulse test (vHIT). Functional capacity was assessed using the Timed Up and Go (TUG), the 10-metre walk test (10MWT), the Short Physical Performance Battery (SPPB) and the Functional Reach Test (FRT). Postural control was quantified by stabilometry in the five sensory conditions of the modified Clinical Test of Sensory Interaction on Balance (m-CTSIB). Gains were compared across planes using a Friedman test. Associations with functional and postural outcomes were examined by multivariate linear regression and by exploratory factor analysis, with canal gains analysed as continuous variables throughout. <bold>Results</bold> . Mean VOR gain was 1.05 (standard deviation 0.12) in the anterior plane, 1.01 (0.05) in the lateral plane and 0.81 (0.19) in the posterior plane, with a significant difference across planes (Friedman test, p &lt; 0.001) and the lowest gain observed in the posterior plane. The posterior plane gain was associated with none of the functional or postural outcomes in multivariate regression (p between 0.073 and 0.820), and no overall model reached the significance threshold (model p &gt;= 0.134). In exploratory factor analysis, the posterior plane gain was associated with no factor (uniqueness = 0.91), and the anterior and lateral plane gains formed a factor distinct from the functional and postural dimensions. Only the lateral plane gain was associated with walking speed (standardised beta = +0.43; p = 0.027), within a model that did not itself reach the significance threshold. <bold>Conclusions</bold> . In a subset of the healthy older population of this study, VOR gain was selectively reduced in the posterior semicircular canals, without any functional or postural repercussion being identified. This structure-function dissociation calls for caution when a low, isolated posterior gain is unexpectedly discovered in an otherwise asymptomatic older adult. <bold>Clinical trial number</bold> : not applicable. </p>