To correlate the reduction in migraine frequency with change in phosphene threshold of transcranial magnetic stimulation during levetiracetam treatment. Several case series have suggested levetiracetam efficacy may be effective in the management of migraine. Phosphene threshold is reduced in patients with migraine with aura, migraine without aura, and menstrual migraine. Preventive treatment may raise phosphene threshold while reducing headache frequency. Subjects experiencing 4-10 migraine attacks per month and not currently receiving preventive treatment for the indication of migraine were recruited into an open-label trial using levetiracetam, and asked to record headache symptoms, severity, duration, and acute medication use in a daily diary. Following a 28-day qualifying baseline period, subjects were titrated over 6 weeks to either a total daily dose of 3000 mg or their maximum tolerated dose (minimum tolerated daily dose of 1000 mg required). Transcranial magnetic stimulation was performed at day 28 and days 26, 28, 84, and 154. The visual cortex of each subject was stimulated 2 times at 20% power. Power was increased by 10% increments until at least one of the 2 stimulations produced a positive phosphene response. Once a positive response was achieved, a random order of 5 stimulation intensities surrounding the initial positive threshold was generated and given 3 times per session. Stimulation intensities were -10%, -5%, 0%, +5%, and +10% in relation to the positive threshold achieved. To eliminate a learning curve distortion, only observations at days 28, 84, and 154 were used for analysis. The mean phosphene threshold was defined as the average of the lowest positive threshold of the 3 stimulation sequences per visit. Ordinary least squares regression was used to evaluate the association between the change in mean daily headache rate from visit 3 to visit 7 and the change in mean transcranial magnetic stimulation threshold during the same period. Sixty-one subjects were enrolled. Twenty-one subjects were discontinued (because of poor study compliance or attack frequency) during the baseline phase prior to study drug initiation, and an additional subject whose data were not analyzed because of suspect quality. During the first 6 weeks on study drug (titration phase), 8 subjects dropped out (20.5%). Full analysis of the remaining 31 subjects, who reached a maintenance dose after 6 weeks on study medication, was performed. Subjects were largely white, female, and had a mean age of 41 +/- 13 years. Increasing age (beta = 1.27, P = .09), nonwhite race (beta = 6.90, P = .03), and diagnosis of tension-type headache (beta = 6.12, P = .095) were found to be associated with a higher mean transcranial magnetic stimulation threshold. Conversely, increasing body mass index was found to be associated with a lower mean transcranial magnetic stimulation threshold (beta = -1.19, P = .005). The number of migraine attacks decreased from 4.24 during the baseline interval to 2.53 during the interval preceding visit 7 (P = .001). There was a small but significant increase in transcranial magnetic stimulation threshold from visit 3 to visit 5 (P = .03) and visit 3 to visit 7 (P = .03 omnibus test). However,the difference between visit 5 and visit 7 was not statistically significant (P = .88). The mean transcranial magnetic stimulation threshold did not change from visit 5 to visit 7. Phosphene threshold increased during treatment with levetiracetam. At the 10% significance level, headache frequency and phosphene threshold were negatively correlated.