Clinical and video-assisted examination of the vestibulo-ocular reflex: a comparative study

The assessment of the vestibulo-ocular reflex (VOR) is one of the main steps in clinically evaluating patients with dizziness. It can be performed at the bedside with common head-impulse test in which eye position is analysed at the end of the head-thrust. It is an important test due to its high specificity but low sensitivity.

We studied 179 patients with different types of balance- affecting disorders. The results were analysed in contingency tables. The clinical test was classified as normal or abnormal according to the absence or existence, respectively, of fixation saccades once the head-thrust was ended. The video head-impulse test (vHIT) was classified according to vestibulo-ocular reflex (VOR) gain and presence of fixation saccades. The speed of the slow phase of spontaneous nystagmus was also quantified, as well as the caloric test results.

There were significant differences (Chi-square test, P=0.00) for the findings in the clinical evaluation and with the vHIT: 32.1% of the tests performed yielded different findings in both tests. In the vHIT, the differences were due to the finding of normal gain with saccades; in these patients, the mean canal paresis was significantly abnormal: 39% ± 10%.

The distribution of findings for the VOR bedside examination and for that with the help of a video system are significantly different; as such, the video head-impulse examination is not simply an added VOR detection and registration system. The difference relies mainly on a vHIT response characterised as of normal gain but with fixation saccades. These have been considered as the cause for the low sensitivity of the bedside VOR examination and sometimes regarded as normal responses; we have demonstrated that these findings are abnormal according to the findings of higher canal paresis in the caloric test.