Multimodal Neuroimaging as a SUDEP Predictor: What Is Known and What Still Needs to Be Uncovered?

Introduction: Sudden Unexpected Death in Epilepsy (SUDEP) is the most common cause of death in patients with poorly controlled epilepsy. To date, a higher risk of developing SUDEP is mainly identified by clinical factors, among which generalized tonic-clonic seizures and their frequency stand out as part of the highly debated SUDEP-7 Scoring. This review investigates the role of neuroimaging-based approaches as a tool to help predict SUDEP.MethodsWe carried out a systematic search of the literature to identify multimodal neuroimaging modifications (i.e., MRI, fMRI, PET, and SPECT) in patients with epilepsy who died from SUDEP. The following databases were used: PubMed and Google Scholar. The review was registered on the PROSPERO platform (Registration code: CRD42024558765).ResultsFifteen articles were selected, investigating 104 SUDEP cases compared with 792 non-SUDEP epileptic patients and 280 healthy controls (HC) (overall mean age 33.9 +/- 1.6). Results suggest that SUDEP and non-SUDEP cases and HC differ anatomically and functionally. In the SUDEP population, MRI data indicate relevant volume differences in the gray matter of the hippocampus and cerebellar cortices. In addition, functional imaging reveals discrepancies in network modulation within the brainstem and its relationship with several cortical structures. Although less consistent, PET and SPECT scan data point toward alterations in metabolism and perfusion in the frontal and brainstem areas.ConclusionThe reviewed data support correlations between the occurrence of SUDEP and neuroimaging multimodal alterations that could be relevant in death prediction.