■ OBJECTIVE: Vestibular evoked myogenic potential (VEMP) testing is a new method of assessing the vestibular system but its clinical usefulness remains unestablished. Our objectives were to assess the clinical usefulness of this test in identifying vestibular disease, and to evaluate whether the degree of asymmetry used to classify an abnormal response (30% or 40%) would affect its accuracy.

■ METHODS: We clinically evaluated 88 consecutive patients complaining of dizziness with a neurotologic physical examination including vestibular evalutation (spontaneous nystagmus, head-shake testing, gaze nystagmus, and Hallpike maneuver). ENG testing and MRI scanning were done when indicated. All patients underwent VEMP testing using a head lift or head turn protocol to activate neck muscles. We considered 30% and 40% asymmetry criteria as abnormal and correlated the VEMP findings with the clinical evaluation.

■ RESULTS: Twenty patients had no VEMP response and were excluded from the study. Analysis of the remaining 68 patients showed that 41% (20 of 49) of patients with a peripheral vestibular disorder had an abnormal VEMP on the ipsilateral side using the 30% amplitude asymmetry criterion (p<0.01) (t-test), and 27% (13 of 49) had an abnormal VEMP on the ipsilateral side using the 40% criterion (p<0.01). The likelihood of a false positive (abnormal test in a normal patient) was very low (3%, p<0.01). Sixty-two percent (5 of 8) of patients suspected of central vestibular dysfunction had delayed latency responses.

■ CONCLUSIONS: The very low false positive rate implies that an abnormal VEMP test is a good indicator of the presence of vestibular dysfunction, but a normal VEMP test does not exclude vestibular dysfunction. More patients with vestibular disease are identified (41% vs. 27%) just as accurately using a 30% than 40% amplitude asymmetry criterion. VEMP testing appears to be a valuable addition to the vestibular test battery and should be considered in the evaluation of dizzy patients.