Longitudinal Neuroimaging over 30 Days: Temporal Characteristics of Migraine

• Published in: Annals of neurology Vol. 87; no. 4; pp. 646 - 651
• Main Authors: Schulte, Laura H., Mehnert, Jan, May, Arne
• Format: Journal Article
• Language: English
• Published: Hoboken, USA John Wiley & Sons, Inc 01.04.2020
• ISSN: 0364-5134; 1531-8249
• DOI: 10.1002/ana.25697

Objective Although migraine is defined by the headache and headache‐associated symptoms, the true beginning of a migraine attack lies in the premonitory phase. To understand the generation of attacks, one needs to investigate the phase before headache starts. The premonitory phase of migraine is characterized by a well‐described complex of symptoms. Its duration, however, is not clearly defined, and there are no biomarkers to help define when this phase starts. Methods Here, we used functional magnetic resonance imaging (MRI) to elucidate the duration of the premonitory phase in spontaneous human migraine attacks. Because migraine attacks are hardly predictable and thereby the premonitory phase is difficult to catch, we scanned 9 patients daily over a minimum period of 30 days using a well‐established paradigm for functional MRI of trigeminal nociception. Results Seven patients were included in the analysis, thus providing cumulative data of 27 spontaneous human migraine attacks including scans before, during, and after migraine pain as well as interictal scans. As a response to painful trigeminal stimulation, activation of the hypothalamus was present within the last 48 hours before headache onset but not earlier. Interpretation Using hypothalamic activation as a potential marker for the premonitory phase of migraine in this unique dataset, our data corroborated a duration of 48 hours for the premonitory phase of migraine. We suggest applying this time criterion in future studies when focusing on this phase of the migraine cycle. ANN NEUROL 2020;87:646–651