Navigated rTMS for the treatment of tinnitus: a pilot study with assessment by fMRI and AEPs

Objective: Repeated transcranial magnetic stimulation (rTMS) of auditory cortex has been proposed to treat refractory chronic tinnitus, but the involved mechanisms of action remain largely unknown. The purpose of this pilot study was to evaluate the impact of rTMS on auditory cortex activity in a series of tinnitus patients, using for the first time both functional magnetic resonance imaging (fMRI) of the brain and auditory evoked potentials (AEPs).

Method: In six patients with chronic, lateralized refractory tinnitus, we performed five sessions of neuronavigated rTMS delivered at 1Hz over the secondary auditory cortex (defined on morphological MRI), contralateral to tinnitus side. The effects of rTMS were assessed on clinical scales, fMRI, and AEPs (N1 and P2 components).

Results: The clinical impact of rTMS on tinnitus was good for three patients (25-50% improvement of tinnitus severity compared to baseline), moderate for two patients (15% improvement), and null for one patient who had the most severe tinnitus at baseline. The changes induced by rTMS on fMRI data varied with the baseline level of auditory cortex activation before rTMS. This baseline level of activation was itself related to the severity of tinnitus. Thus, cortical stimulation increased auditory cortex activation in patients who had less severe tinnitus and low level of activation before rTMS, whereas it decreased auditory cortex activation in patients who had more severe tinnitus and higher level of activation before rTMS. Regarding AEPs, rTMS decreased N1 amplitude in all patients, except in the patient who had the most severe tinnitus at baseline and showed no improvement after rTMS. Conversely, P2 amplitude decreased after rTMS only in patients with severe tinnitus, at least for auditory stimulation contralateral to tinnitus, but increased in patients with less severe tinnitus.

Conclusions: The changes produced by rTMS in auditory cortex activity, as assessed by fMRI and AEPs, appeared to depend on a process of disease-related homeostatic cortical plasticity, regardless of the therapeutic impact of rTMS on tinnitus.