[Reliability of electrogustatory threshold--electrogustometry using CNV]

CNV (contingent negative variation), an event-related potential, is induced when an oncoming imperative stimulus (S2) is anticipated after recognition of a warning stimulus (S1). With this in mind, we assessed the usefulness of electrogustometry using CNV as an objective gustatory test with an electric gustatory stimulus as a warning stimulus, S1. A gustatory test was performed on a total of 100 sides of the tongue in 50 subjects. Excluding no response cases on 5 sides in 5 subjects, the CNV threshold was successfully determined on 95 sides in 50 subjects. When the CNV threshold values obtained and the conventionally measured subjective threshold values were compared, a very high correlation was demonstrated between the two thresholds, with a correlation coefficient of 0.961. To assess the reliability of the CNV threshold, three additional measurements on 5 sides in 5 subjects with normal subjective threshold values were performed on different days, and the CNV threshold value variation among the three measurements in all 5 subjects never exceeded 13 microA. Separately, subjective threshold and CNV threshold values were compared in 4 patients with facial palsy (hence with gustatory abnormality) who consented to a follow-up examination. The difference between the subjective threshold and CNV threshold values in all 4 patients never exceeded 10 microA. These findings demonstrate that the reliability of the CNV threshold is satisfactory. Next, a gustatory imitation experiment was performed to assess the usefulness of electrogustometry using CNV as a means of detecting gustatory abnormality malingering. The experiment was performed on a total of 6 subjects: 3 physicians with knowledge of CNV (knowledge group) and 3 others with no knowledge of CNV (no-knowledge group). The differences between the imitated threshold and the subjective threshold values in all of these subjects never exceeded 10 microA, and the difference between imitated threshold and CNV threshold values never exceeded 13 microA. Comparison between the knowledge group and the no-knowledge groups showed that the imitated threshold values were very close to the subjective threshold and CNV threshold values. Based on the above results, it was concluded that gustometry using CNV is useful and can serve well as a method of objective evaluation of gustatory sensation. The results also suggested that it will be useful in diagnosing malingering.