A comprehensive review of deep brain stimulation for drug-resistant epilepsy.

Background: Deep brain stimulation (DBS) has emerged as a palliative neurosurgical treatment option for drug-resistant epilepsy (DRE), particularly in patients who are not candidates for resective surgery or who continue to experience seizures after surgical intervention. Multiple thalamic and extrathalamic targets have been investigated; however, optimal target selection remains challenging and requires a comprehensive understanding of epileptogenic networks. Objective: To provide a comprehensive overview of current evidence on DBS for DRE and refractory status epilepticus, with a focus on anatomical targets, mechanisms of action, clinical outcomes, and considerations for target selection. Methods: We reviewed the existing literature on DBS in epilepsy, including randomized controlled trials, observational studies, meta-analyses, and relevant experimental data. Major targets-including the anterior nucleus of the thalamus (ANT), centromedian nucleus (CM), pulvinar, mediodorsal nucleus (DM), subthalamic nucleus (STN), and other emerging targets-were examined in terms of connectivity, proposed mechanisms, and clinical efficacy. Results: Among available targets, ANT-DBS has the strongest clinical evidence, demonstrating sustained seizure reduction in randomized trials and long-term follow-up studies. CM-DBS shows particular promise in generalized epilepsy, especially Lennox-Gastaut syndrome, likely through modulation of thalamocortical and reticular networks. The pulvinar has emerged as a potential target for temporal and posterior quadrant epilepsies, reflecting its extensive cortical connectivity. Other targets, including the DM, STN, hippocampus, hypothalamus, nucleus accumbens, and cerebellum, have shown variable efficacy in smaller studies and may be relevant for specific epilepsy subtypes or network configurations. Across targets, therapeutic effects are likely mediated by modulation of distributed epileptogenic networks involving limbic, sensorimotor, and arousal systems. Conclusion: DBS represents an important therapeutic option for DRE, expanding the scope of neuromodulation beyond traditional surgical approaches. Optimal outcomes depend on individualized target selection based on seizure semiology, network characteristics, and anatomical considerations. As clinical experience and technological advances continue to evolve, further studies are required to refine patient selection, improve targeting strategies, and optimize stimulation paradigms.