OBJECTIVE: Electrical high-frequency stimulation (HFS) of deep brain structures has been successfully used as a treatment for patients with movement disorders. The mechanisms of HFS allowing therapeutic clinical effects remain unclear, which justifies experimental studies to address these questions. These experiments require an external stimulator, which may offer the possibility to deliver a current with monophasic or biphasic pulses. The aim of the present study was to quantify the evolution of a potentially deleterious effect of HFS according to the duration and/or intensity in monophasic and biphasic conditions. METHODS: In all rats, HFS was performed with monophasic pulses in deep brain structures of 1 hemisphere and with biphasic pulses symmetrically in the other hemisphere. The effect of HFS was tested, first for various durations of HFS at a constant intensity (100 microA) and, second, for measuring the effect of various current intensities of HFS at constant duration (10 minutes). At the end of each stimulation test, the volume of lesion was determined and analyzed. RESULTS: In all hemispheres in which stimulation using biphasic pulses was delivered, we never found any relevant lesions. Conversely, monophasic electrical stimulation always created a lesion: at 100 microA, a minimal duration of HFS of 5 minutes induced a tissue damage volume of 0.0055 +/- 0.0015 mm(3). For 10 minutes of HFS, a minimal intensity of 100 microA induced a tissue damage volume of 0.0062 +/- 0.0017 mm(3). Regression analysis showed that the extent of lesion increased linearly with the intensity and duration. CONCLUSION: In conclusion, this study proved that HFS using monophasic pulses systematically created tissue damage after 5 minutes of stimulation at 100 microA. HFS is safe when biphasic pulses are used for intensities as high as 2 mA and durations as long as 120 minutes. Monophasic pulses can be safely used only during short stimulation and at low intensities.