Fundamental to the concept of idiopathic or primary headache, including mig
raine, tension-type headache and cluster headache, is the currently accepte
d view that these conditions are due to abnormal brain function with comple
tely normal brain structure(1). Cluster headache is one such idiopathic hea
dache with many similarities to migraine, including normal brain structure
on magnetic resonance imaging and abnormal function in the hypothalamic gre
y matter by positron emission tomography(2). Given the consistency of the p
ositron emission tomography findings with the clinical presentation, we sou
ght to assess whether the brains of such patients were structurally normal.
We used voxel-based morphometry, an objective and automated method of anal
yzing changes in brain structure, to study the structure of the brains of p
atients with cluster headache. We found a co-localization of structural cha
nges and changes in local brain activity with positron emission tomography
in the same area of the brain in the same patients. The results indicate th
at the current view of the neurobiology of cluster headache requires comple
te revision and that this periodic headache is associated with a hitherto u
nrecognized brain abnormality in the hypothalamic region. We believe that v
oxel-based morphometry has the potential to change in the most fundamental
way our concept of primary headache disorders, requiring a radical reapprai
sal of the tenet of structural normality.