Beyond Broca and Wernicke: Epilepsy surgery in the language areas

Epilepsy surgery in language areas is challenged by the intricacies of presurgical workup and surgical planning. In recent decades, the view of language-related circuitry has shifted from being localized in a few cortical centers to a distributed, dynamically interconnected system, increasing complexity. In this framework, neuropsychology, functional neuroimaging and neurophysiological assessments play an essential role in minimizing the risk of language deficits. A comprehensive preoperative neuropsychological assessment is essential for providing a baseline against which to compare postoperative performance, as well as for helping to define functional deficit and epileptogenic zones. Functional magnetic resonance imaging (fMRI) represents a fundamental noninvasive alternative to the traditional Wada test for determining hemispheric language lateralization and localization. Resting state fMRI (rs-fMRI) may offer a compelling alternative, particularly in populations where task performance is difficult or unreliable. The role of Stereo-electroencephalography (Stereo-EEG) and the intracerebral electrical stimulations (ES) is pivotal for both the identification of the epileptogenic zone and the mapping of eloquent cortices. The use of all available cortical stimulation methods, combined with test batteries targeting different language subdomains and designed to avoid the induction of post-discharges, appears to be the optimal approach for maximizing patient safety. The preservation of language function also relies on highly specific intraoperative mapping techniques, primarily employed in cooperative patients during awake craniotomy. Likewise, cortico-cortical evoked potentials could represent a valuable "task-free" neurophysiological alternative. Surgical strategies involving language areas vary from focal resections to larger lobar, multilobar or hemispheric procedures. Minimally invasive options such as Stereo-EEG-guided radiofrequency thermocoagulation (RF-TC) and laser interstitial thermal therapy (LITT) have broadened therapeutic possibilities. Selection depends on EZ accessibility and extent. Modern epilepsy surgery within language-eloquent areas increasingly relies on integrated multimodal approaches. Future directions include the use of machine learning to analyze large postoperative datasets and predict long-term functional and seizure outcomes.