Clinical reports show that bilateral subthalamic nucleus (STN) stimulation is effective in improving parkinsonian gait. Quantitative analysis of the efficacy of STN stimulation on gait is of interest and can be carried out using a commercially available stride analyser. Ten parkinsonian patients (5 men, 5 women) with a mean age of 55.8, SD 9.6 years were included in our study. They had a mean duration of Parkinson's disease (PD) of 13.3, SD 4.5 years and a motor examination score (part III of the Unified Parkinson's Disease Rating Scale) (UPDRS) of 43, SD 13 in off-stimulation off-drug condition. All the patients had bilateral chronic STN stimulation which had started from 3 to 36 months before the study. Patients were evaluated in off-drug and on-drug conditions both with and without stimulation. We analysed the principal gait measures: velocity, cadence, stride length, gait cycle, duration of single and double limb support. The clinical parkinsonian signs were evaluated with the part III of the UPDRS. In the off-drug condition, STN stimulation significantly (p < 0.05) improved velocity and stride length. The effect was similar to that of levodopa. When STN stimulation was switched on at the best of the levodopa induced effect, no further improvement was observed. The UPDRS motor score was significantly (p < 0.001) decreased after both stimulation and levodopa. In conclusion, STN stimulation is effective on parkinsonian gait.