Clinical Phenotypes, Neuroimaging, and Outcomes in Prolonged and Persistent Non-Motor Migraine Aura: A Systematic Review and Episode-Level Synthesis

Background/Objectives: Prolonged and persistent migraine aura may mimic cerebral ischemia, but its clinical course and neuroimaging features are poorly characterized. We synthesized episode-level evidence on non-motor aura lasting more than 60 min without reported established infarction. Methods: This systematic review (International Prospective Register of Systematic Reviews [PROSPERO] CRD420261461238) followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines. PubMed, Scopus, and Google Scholar were searched from inception to August 2026 and the searches were rerun with an extended strategy on 5 October 2026, supplemented by backward and forward citation searching. Eligible reports provided individual-episode clinical data and temporally related neuroimaging. Owing to publication selection, heterogeneous reporting, and clustering, the synthesis was descriptive. Results: Forty-eight studies described 80 episodes in 79 individuals (49 prolonged; 31 persistent). Visual aura occurred in 63/80 episodes. Although magnetic resonance imaging was performed in 77/80 episodes, patient-level informative diffusion imaging was available in only 28/80 and imaging-based perfusion data in 28/80. Five of the 28 episodes with informative diffusion imaging were diffusion-positive (17.9%); the acute apparent diffusion coefficient (ADC) was reduced in 3, non-restricted in 1, and unreported in 1, and all diffusion-weighted imaging (DWI) abnormalities resolved on follow-up. Unusual imaging findings are more likely to be published; this proportion is therefore hypothesis-generating rather than a frequency estimate. In the longer-follow-up stratum, aura improvement was reported in 36/36 prolonged and 18/22 persistent episodes, with complete resolution in 35/35 and 15/22, respectively; a nerve-block series with a fixed 24-h assessment (14 episodes) was analyzed separately. Conclusions: Published cases show heterogeneous phenotypes and cannot support estimates of prognosis or treatment effectiveness. Standardized prospective data collection should incorporate symptom-level duration, imaging timing, DWI/ADC reporting, treatment timing, and protocol-defined follow-up.