This study evaluated the effectiveness of gabapentin in treating tinnitus in two populations: Tinnitus with associated acoustic trauma and tinnitus without associated acoustic trauma. The hypothesis was that gabapentin would decrease both subjective and objective features of tinnitus in the trauma group, but would be less effective in the non-trauma group.'

Methods. A prospective, placebo-controlled, single-blind study of the effect of gabapentin on tinnitus was employed. Audiograms and personal histories were used to categorize tinnitus etiology as either secondary to acoustic trauma, or not associated with acoustic trauma. Participants were restricted to those with moderate-to-severe tinnitus for at least one year. All participants received gabapentin in a graduated ascending-descending dose series over 20 weeks (peak dose of 2400 mg/day).

Results. There was a significant improvement in tinnitus annoyance for the trauma group (p = 0.05). Other subjective aspects of tinnitus were not significantly affected in either group. Between-subject variability of therapeutic response was considerable. Nevertheless, considering subjective loudness ratings, 4/19 non-trauma participants, and 6/20 trauma participants showed an improvement of 20 percent or better. Considering psychoacoustic loudness estimates, 4/19 non-trauma and 6/20 trauma participants showed a 15 dB (HL) improvement. Evenly dividing each group into high and low responders revealed significant improvement in loudness at 1800 and 2400 mg/day for the trauma high-response subgroup (p = 0.007). No significant improvement was obtained for other subgroups.

Inclusion Criteria:

• non-pulsatile tinnitus present > 1 year
• Tinnitus Handicap Questionnaire score > 30
• ability to perform psychophysical matching procedure

Exclusion Criteria:

• evidence of depression
• renal insufficiency
• conductive hearing loss