Abnormal vestibular-evoked myogenic potentials as a risk factor for unpredicted falls in spinocerebellar ataxia: a preliminary study

Two authors retrospectively reviewed the medical records of 37 patients with genetically confirmed autosomal dominant SCA at the dizziness clinic of Pusan National University Hospital from January 2019 to March 2022. We finally recruited 26 (9 men, median age: 52, age range: 21–67) for analyses after excluding 11 patients with incomplete evaluation due to a bed-ridden state (n = 4), a follow-up loss (n = 4), refusal of the testing (n = 2), and a previous history of vestibular neuritis (n = 1). The subtypes of SCA included SCA1 in eight from four families, SCA2 in 14 from eight families, and SCA6 in four from three families. There were no differences in age, sex, and disease duration among the subtypes (Table 2).

This study followed the tenets of the Declaration of Helsinki and was performed according to the guidelines of the Institutional Review Board of Pusan National University Hospital (2108–001-105).

The assessments for cerebellar ataxia were based on the Scale for the Rating and Assessment of Ataxia (SARA) that was scored by the sum of the gait, stance, sitting, speech disturbance, finger-chase, nose-finger test, fast alternating hand and heel-shin movements. [5] Self-administered questionnaires for evaluation of dizziness and health-related QoL included the dizziness handicap inventory (DHI) [16, 17] and the five-level EuroQoL five-Dimension (EQ-5D-5L). [18] To assess the symptoms during walking in daily life, the following yes or no questions were also added; "Q1: Have you had unpredictable falls?" "Q2: Do you feel more dizzy in moving conditions, such as in a moving elevator or escalator?" "Q3: Do you feel a pulling sensation to one side while walking (lateropulsion)?".

Eye movements were recorded at a sampling rate of 60 Hz using 3D-video-oculography (VOG, SLMED, Seoul, Korea). Spontaneous nystagmus (SN) with or without visual fixation, gaze-evoked nystagmus (GEN), and horizontal saccades and smooth pursuit were pursued following the methods described previously in detail. [19] Head and eye movements were also recorded during head impulses using a video-based equipment (vHIT, SLMED, Seoul, South Korea). The detailed methods and normative data on gain of the vestibulo-ocular reflex (VOR) for six semicircular canals (SCCs) were described previously. [20]

The detailed methods for recording cervical and ocular vestibular evoked myogenic potentials (VEMPs) and the subjective visual vertical (SVV) were described previously. [20, 21] VEMPs were recorded using a Nicolet Viking Select unit (Nicolet Biomedical, Madison, WI, USA) on the same day. Cervical VEMP (cVEMP) was recorded from the ipsilateral sternocleidomastoid muscle (SCM) that was activated by elevating and turning the head contralaterally in the supine position. Electromyographic activities of the SCM were simultaneously recorded using a surface electrode from the upper half of the SCM with a ground electrode on the forehead and a reference electrode over the upper sternum. A short alternating tone burst (110 dB nHL; 121.5 dB SPL; 500 Hz; ramp = 2 ms; plateau = 3 ms) was given at 2.1 Hz monoaurally using a headphone. During the recording, the EMG activities of the SCM were also monitored and digitized at 1 kHz using an analog-to-digital converter (NI PCI-4461, National Instruments, Austin, TX, USA). The LabVIEW program (National Instruments) was used to analyse the peak-to-peak amplitudes and calculate the mean tonic activation during the recording. The absolute cVEMP amplitude was then normalized against the mean tonic activation of the SCM during the recording. [21] We confirmed that none of the patients had the conditions leading to conductive hearing loss, such as middle ear infections, and no patients reported subjective hearing impairments. To record ocular VEMP (oVEMP), patients sat with their head upright while fixating on a target that was displaced on the wall, more than 2 m away from the eyes, with an angle of gaze more than 20° upward. Electromyographic activities were recorded using a surface electrode placed on the infra-orbital ridge 1 cm below the center of the lower eyelid. The reference electrode was attached 2 cm below the active electrode, and the ground electrode on the forehead. Bilateral oVEMPs were induced tapping the hairline at AFz up to 60 times using an electric reflex hammer (VIASYS Healthcare, CA, USA). The latency of cVEMP was measured as the time when the positive wave (p13) at around 13 ms and the negative wave (n23) at around 23 ms appeared, and the amplitude between these two waves was normalized to the SCM contraction. For oVEMP, the latency of the negative wave at around 10 ms (n10) and the amplitude between n10 and the subsequent positive wave were measured. We compared the VEMPs of the patient group with those of 26 age-matched control (Median age = 53, IQR 40–58, p = 0.096). Abnormal VEMPs were defined as no responses to stimulation or asymmetry of the amplitudes > 20%. [20, 21]

The SVV tilt was measured while seating the patient in a darkroom. The patients were instructed to align a randomly tilted rod (80 cm long and 0.3 cm wide) vertically using a joystick at least five times per each session. The normal value of SVV is ± 3° during binocular viewing (a negative value indicates a counterclockwise rotation). [20, 21]

Detailed description of the ocular motor findings is provided in Table 1. No patients exhibited SN with or without fixation. However, all patients showed cerebellar ocular motor dysfunction that included GEN, impaired smooth pursuit, or hypermetric saccades. GEN was observed in all patients with SCA6 (p = 0.040). Slowing of horizontal saccades was observed in 15 (58%) patients and was more prevalent in patients with SCA2 (13/14, 93%, p < 0.0001, Table 2).

The head impulse gain of the VOR was decreased for the horizontal semicircular canal (HC, 0.89 ± 0.10 vs. 0.94 ± 0.05, p = 0.004, t-test), but was normal for the vertical semicircular canals (0.95 ± 0.11 vs. 0.96 ± 0.04, p = 0.427, t-test). Overall, 14 patients (54%) showed a subnormal gain for at least one of the semicircular canals, 38% (10/26) for the HC and 27% (7/26) for the vertical canals. The reduced gain for the HC was less commonly observed in patients with SCA2 (2/12, p = 0.022).

Nine patients (35%) exhibited abnormal oVEMP (n = 6, 23%) or cVEMP (n = 7, 27%) (Table 1). Of them, bilateral absence of responses was observed in seven, oVEMP in five and cVEMP in five. Two other patients with abnormal cVEMP were marked by a unilateral absence of wave formation, one (SCA1) of them with an asymmetry oVEMP at 36%.

Most patients with SCA1 (7/8, 88%) demonstrated various patterns of abnormal VEMPs [abnormal cVEMP in 6 (75%, median age 54, range 43–65), abnormal oVEMP in 5 (63%, median age 56, range 43–65)] while only two patients (14%) with SCA2 and none of the patients with SCA6 showed abnormal VEMP (Table 2).

In the cross-analyses of the variables (Fisher's Exact Test), the patients with abnormal cVEMP or oVEMP showed a higher SARA score (p = 0.014) and a history of unpredicted falls (p = 0.046). Especially, the patients with abnormal oVEMP showed elevated SARA scores (p = 0.017) (Fig. 1B).