The role of cochlear and vestibular afferents in long-latency cervical vestibular evoked myogenic potentials.

OBJECTIVE: To examine the origin of cervical vestibular evoked myogenic potential (cVEMP) late waves (n34-p44) elicited with air-conducted click stimuli. DESIGN: Using a retrospective design, cVEMPs from normal volunteers were compared to those obtained from patients with vestibular and auditory pathologies. STUDY SAMPLE: (1) Normal volunteers (n = 56); (2) severe-to-profound sensorineural hearing loss (SNHL) with normal vestibular function (n = 21); (3) peripheral vestibular impairment with preserved hearing (n = 16); (4) total vestibulocochlear deficit (n = 23). RESULTS: All normal volunteers had ipsilateral-dominant early p13-n23 peaks. Late peaks were present bilaterally in 78%. The p13-n23 response was present in all patients with SNHL but normal vestibular function, and 43% had late waves. Statistical comparison of these patients to a subset of age-matched controls showed no significant difference in the frequencies, amplitudes or latencies of their ipsilateral early and late peaks. cVEMPs were absent in all patients with vestibular impairment. CONCLUSION: The presence of long-latency cVEMP waves was not dependent on the integrity of sensorineural hearing pathways, but instead correlated with intact vestibular function. This finding conflicts with the view that these late waves are cochlear in origin, and suggests that vestibular afferents may assume a more prominent role in their generation.

Bone-conducted vestibular and stretch reflexes in human neck muscles.

In normal humans, tapping the forehead produces a neck muscle reflex that is used clinically to test vestibular function, the cervical vestibular evoked myogenic potential (cVEMP). As stretch receptors can also be activated by skull taps, we investigated the origin of the early and late peaks of the bone-conducted cVEMP. In twelve normal participants, we differentially stimulated the vestibular and neck stretch receptors by applying vibration to the forehead (activating both vestibular and stretch receptors) and to the sternum (activating mainly stretch receptors). Patients with bilateral vestibulopathy (BVP; n = 26) and unilateral vestibular loss (uVL; n = 17) were also investigated for comparison. Comparison of peaks in normal subjects suggested that the early peaks were vestibular-dependent, while the later peaks had mixed vestibular and stretch input. The late peaks were present but small (1.1 amplitude ratio) in patients with BVP and absent VEMPs, confirming that they do not strictly depend on vestibular function, and largest in age-matched controls (1.5 amplitude ratio, p = 0.049), suggesting that there is an additional vestibular reflex at this latency (approx. 30 ms). Patients with uVL had larger late peaks on the affected than the normal side (1.4 vs 1.0 amplitude ratio, p = 0.034). The results suggest that the early responses in SCM to skull vibration in humans are vestibular-dependent, while there is a late stretch reflex bilaterally and a late vestibular reflex in the contralateral muscle.