There is, as yet, no known cause for cluster headache. However, cluster headache has two major clinical features: the trigeminal (nerve) distribution of pain and the associated ipsilateral (same-sided) autonomic symptoms. Firstly, the pain producing innervation (stimulation) of the cranium projects through branches of the trigeminal and upper cervical nerves to the trigeminocervical complex from where nociceptive pathways project to higher centres. Secondly, the accompanying ipsilateral (same-sided) autonomic symptoms suggest cranial parasympathetic activation (that's the teary eye, runny and blocked nose) and sympathetic hypofunction (the droopy eye and constriction of the pupil).
However, the third and possibly most important clinical feature is the uncanny timing of both attacks and bouts themselves, which originally suggested an involvement in the brain's master-clock: the hypothalamus or more particularly the suprachiasmatic nucleus (SCN). It is now thought that an abnormality within the hypothalamus is the root cause of the pain, which, when in cycle, releases hormones and chemicals that innervate the trigeminal ganglion, in turn causing the domino effect of pain and cranial autonomic symptoms through the trigeminal nerve down one side of the face and head (and sometimes the neck). This theory is backed up by the regularity of attacks (circannual [time of year] and circadian [time of day]), much lower levels of plasma testosterone (in males) during attacks and bouts, and alterations in the natural production of a variety of hormones/chemicals that affect the biologic clock.
Furthermore, PET (positron emission tomography) studies conducted on the brains' of CH sufferers in the late 1990s demonstrated that there is also ipsilateral (same-sided) hypothalamic activation within the brain: there are direct hypothalamic-trigeminal connections and the hypothalamus is known to have a modulatory role on the nociceptive and autonomic pathways. In summary these studies showed an increase in functional activity of the hypothalamus amongst CH sufferers which is not seen in migraine, and is the prime reason why CH is thought to be caused by an abnormality within the hypothalamus. These abnormalities were seen both when sufferers were undergoing an attack and also whilst pain free and were interpreted as an excessive growth of grey cells within the hypothalamus, or more possibly, within the SCN.
There is also evidence that genetic factors may play an important role in CH, and although the type and number of genes involved is unclear, one recent study focussed upon orexin (or hypocretin), a neuropeptide used by the hypothalamus for signalling.