ADVERTISEMENT
Thalamus May Be Key to Classifying Migraine From Cluster Headache
A new study utilizing magnetic resonance imaging (MRI) to identify headache biomarkers suggests that the thalamus may play a central role in classifying migraine from cluster headache patients. Findings were published in the Annals of Neurology.
“Although the clinical phenotype of these 2 primary headaches can be different, they share some pathophysiological mechanisms,” wrote Roberta Messina, MD, Division of Neuroscience, Neuroimaging Research Unit, Institute of Experimental Neurology, Milan, Italy, and co-authors in the study background. “Machine learning techniques provide biomarkers for diagnosis, prognosis, and personalized treatments and shed light on disease pathophysiology.”
The study included 20 patients with migraine (10 with and 10 without aura), 20 cluster headache patients, and 15 controls from the population attending headache clinics at the King’s College Hospital as well as the school’s staff and students. Researchers ensured that those scanned did not report having a headache 48 hours before or after the MRI.
>>News: In Migraine Prevention, Tolerability Higher for Erenumab Versus Topiramate
Using a supervised machine learning method and multimodal imaging data, the study authors worked to identify MRI patterns that differentiated migraine from cluster headache patients and imaging features that the two groups had in common. In training the model to distinguish between migraine and cluster headache patients from control groups, the authors achieved an overall accuracy of 89% and 98% respectively.
Researchers found that the best way to distinguish migraine and cluster headache patients from controls was to examine a combination of “different patterns of brain activation and morphometry.” They incorporated volumetric and RS fMRI measures with DTI and pCASL data, which provided significant insight regarding hemodynamic, functional, macrostructural, and microstructural changes of the brain.
The most telling MRI patterns in classifying migraine and cluster headache patients from controls included brain resting state functional connectivity (RS FC) networks of the periaqueductal gray (PAG) and hypothalamus. Study results showed “significant functional interaction” between the hypothalamus and brain regions responsible for pain control and visual processing in both migraine and cluster headache patients. The authors also found specific involvement of the networks connecting the left and right PAG to the cerebellum, insula, and frontal, temporal, and occipital areas in the whole group of headache patients as well as when differentiating between migraine and cluster headache.
>>News: Genome-Wide Analysis Identifies 86 Previously Unknown Migraine Risk Loci
The MRI model designed to distinguish cluster headaches from migraine had the lowest accuracy rate at 78%. Researchers believe that the accuracy may have been thrown off by the fact that 9 patients with cluster headache also had a history of “definite or probable” migraine without aura. The most telling MRI features attested to the central role played by the thalamus in classifying between migraine and cluster headache. For cluster headache patients, there was a lower functional interaction between the left thalamus and parietal brain regions, including the precuneus and angular gyrus.
“Based on our results, we could speculate that an abnormal processing of the inner-generated sensory stimuli may lead to the sense of agitation and the compulsion to move described by patients with cluster headache. This hypothesis is in line with previous evidence showing an association between abnormal thalamocortical activity and the presence of agitation in patients with restless leg syndrome or psychiatric disorders,” the authors concluded.
Reference
