Objective.The study explored the hypothesis that spontaneous nystagmus (Ny) in migraine patients can be triggered or modulated by painful trigeminal stimulation, providing evidence of a functional connection between vestibular and trigeminal systems. Background.-Vertigo attacks are reported by subjects with migraine or a familiar history of migraine, also independently of headache episodes. Idiopathic vertigo is three times more frequent in migraine patients than in controls. Vestibular investigations in migraine patients have consistently demonstrated spontaneous Ny both of central and peripheral origin. Design.-In the first phase of the study 10 outpatients experiencing migraine without aura (MO) and 10 healthy volunteers were submitted to the registration of spontaneous primary-position Ny in the dark by Ulmer’s video-ocular-nystag-mographic equipment. Two electrodes for electrical stimulation were applied on the supraorbital point of one side of the head and the intensity of stimulation corresponding to pain threshold was calculated. Spontaneous ocular movements were recorded for 5 minutes at baseline and after a sequence of five electric pulses (square waves of .5 Hz frequency and 50 μs duration, at pain threshold intensity). Nystagmographic responses were expressed as latency after stimulation, direction of the quick phase, and duration. The second phase of the study explored, with the same procedure, the effects on Ny of supraorbital versus median nerve stimulation in other 10 MO patients. Responses to stimulation were considered the appearance of de novo Ny after stimulation in subjects without baseline Ny, or the change of the frequency (at last a 50%variation) or of the direction of Ny after stimulation in subjects with baseline Ny. The latency and the duration of responses to stimulation were also calculated. Results.-In the first series supraorbital painful electric stimulation was able to modify or to evoke Ny in 8 of 10 migraineurs and in none of 10 volunteers (Fisher’s exact test, P< .01). Both the baseline and the induced Ny were second degree, stationary persistent, with a linear slow phase and were suppressed by visual fixation. In the second series, supraorbital nerve stimulation was able to induce or modify Ny in all of 10 patients but only in 1 patient Ny was induced by median nerve stimulation. Characters of Ny were the same as previously described. Statistical comparison of the responses at the two sites of stimulation was significant (Fisher’s exact test, P < .01). In those 7 patients who presented de novo Ny after stimulation it was possible to calculate Ny latency and duration. The mean latency was 25 s (SD: 16, range: 14 to 60). The mean duration was 120 s (SD: 94, range: 20 to 290). Conclusion.-The main result of our study is that in migraine patients painful trigeminal stimulation elicits de novo, or modifies pre-existing spontaneous Ny, generally increasing it. The finding was obtained after trigeminal stimulation, but not after median nerve stimulation. We suggest that painful trigeminal stimulation can induce an imbalance of the vestibular system in migraine patients and possibly explain their predisposition to vertigo. Our data require confirmation by other studies. Objective. The study explored the hypothesis that spontaneous nystagmus (Ny) in migraine patients can be triggered or modulated by painful trigeminal stimulation, providing evidence of a functional connection between vestibular and trigeminal systems. Background.-Vertigo attacks are reported by subjects with migraine or a familiar history of migraine, also independently of headache episodes. Idiopathic vertigo is three times more frequent in migraine patients than in controls. of the study 10 outpatients experiencing migraine without aura (MO) and 10 healthy volunteers were submitted to the registration of spontaneous primary-position Ny in the dark by Ulmer’s video-ocular-nystag-mographic equipment. Two electrodes for electrical stimulation were applied on the supraorbital point of one side of the head and the intensity of stimul Spontaneous ocular movements were recorded for 5 minutes at baseline and after a sequence of five electric pulses (square waves of .5 Hz frequency and 50 μs duration, at pain threshold intensity). Nystagmographic responses were expressed as latency after stimulation, direction of the quick phase, and duration. The second phase of the study explored, with the same procedure, the effects on Ny of supraorbital versus median nerve stimulation in other 10 MO patients. de novo Ny after stimulation in subjects without baseline Ny, or the change of the frequency (at last a 50% variation) or of the direction of Ny after stimulation in subjects with baseline Ny. The latency and the duration of responses to stimulation were also Results. -In the first series supraorbital painful electric stimulation able to modify or to evoke Ny in 8 of 10 migraineurs and in none of 10 v oBoth the baseline and the induced Ny were second degree, stationary persistent, with a linear slow phase and were suppressed by visual fixation. In the second series, supraorbital nerve stimulation was able to induce or modify Ny in all of 10 patients but only in 1 patient Ny was induced by median nerve stimulation. Characters of Ny were the same as previously described. Statistical comparison of the responses at the two sites of stimulation was significant (Fisher’s exact test, P <.01). In those 7 patients who presented de novo Ny after stimulation it was possible to calculate Ny latency and duration. The mean latency was 25 s (SD: 16, range: 14 to 60). The mean duration was 120 s Conclusion: -The main result of our study is that in migraine patients painful trigeminal stimulation elicits de novo, or modifies pre-existing spontaneous Ny, generally increasing it. The finding was obtained after trigeminal stimulation, b ut not after median nerve stimulation. We suggest that painful trigeminal stimulation can induce an imbalance of the vestibular system in migraine patients and possibly explain their predisposition to vertigo. Our data require confirmation by other studies.