INTRODUCTION: Parkinson's disease (PD) significantly impairs gait, particularly in advanced stages. Deep brain stimulation (DBS) has emerged as a therapeutic option for refractory motor symptoms, yet its effects on locomotion remain heterogeneous. The objective was to synthesize the available biomechanical evidence on changes in spatiotemporal gait parameters in PD patients undergoing DBS. METHODS: We conducted a systematic review and meta-analysis in accordance with PRISMA 2020. We included original studies with pre- and/or post-DBS gait analyses using instrumented technologies. Methodological quality was assessed with the Cochrane Handbook. RESULTS: Nineteen DBS studies were included. The studies exhibited heterogeneous assessment conditions, encompassing different time points before and after surgery, and both on and off medication and stimulation. The targets analyzed included the subthalamic nucleus (STN), the internal segment of the globus pallidus (GPi), and the pedunculopontine nucleus (PPN). The most frequently reported spatiotemporal parameters were gait speed, step and stride length, cadence, and double support time. Data acquisition employed optoelectronic systems, pressure platforms, and inertial sensors. CONCLUSION: DBS improves spatiotemporal gait parameters in PD, particularly under active stimulation. However, methodological heterogeneity and the limited representation of GPi and PPN constrain generalizability. Instrumented systems enable objective and reproducible assessment, which is key to optimizing functional follow-up and therapeutic decision-making.