Abstract
We read with great interest the article ‘Olfactory hallucinations in primary headache disorders: case series and literature review’ by Coleman et al. (1). The authors suggested that olfactory hallucinations, also known as phantosmia, are an uncommon but distinct manifestation of migraine aura. We would like to report an additional case in which phantosmia and migraine are tightly intermingled in a very original manner.
A 33-year-old male described olfactory hallucinations whose onset was reported after a facial trauma in a road traffic accident, when he was 22 years old. On initial admission, CT scan showed fractures of the left frontal and bilateral ethmoidal sinuses. Since then, the patient has complained of anosmia and recurrent olfactory hallucinations, which were described as neutral or more rarely as pleasant or unpleasant. Moreover, this patient, with no familial or personal history of migraine, reported severe bifrontal pulsatile headache associated with nausea. Attacks lasted 6–24 hours and occurred 10 days per month. Interestingly, headaches were systematically associated with intense and unpleasant phantosmias and never occurred during pleasant or neutral phantosmias. There was neither any ensuing visual, somatosensory, motor or phasic deficit nor any impairment of consciousness. Conversely, in the absence of unpleasant phantosmias, the patient did not have any headache. The duration of hallucination and the normality of electroencephalography recordings argued against the diagnosis of epilepsy. Brain MRI was normal. During attacks, ibuprofen and triptans (almotriptan or eletriptan) provided effective pain relief. Treatment with gabapentin was effective to prevent both migraine-like headache and olfactory hallucinations.
Olfactory hallucinations are thought to have several causes, including rhinosinusitis, epilepsy, psychiatric conditions and, as more recently suggested, migraine aura (1). In our patient with recurrent phantosmias, the diagnosis of aura is not appropriate, unless we suppose that migraine crisis could be associated systematically with isolated olfactory auras. If so, it would also need to be supposed that only unpleasant auras (but not pleasant auras) can trigger a migraine crisis. Because of the past history of acquired anosmia, it can be hypothesized that hallucinations are associated with olfactory deafferentation due to loss of sensory input, a mechanism similar to that described as visual hallucinations in Charles Bonnet syndrome.
The close temporal relationship between migraine attacks and phantosmias raises the question of the interaction between the trigeminal and the olfactory systems. Interactions between the olfactory and the trigeminal nociceptive system have been demonstrated in a recent fMRI study investigating migraine attacks (2). The authors showed that pure olfactory stimuli increased activity in the olfactory system but also in the rostral part of the pons. Several sites of interaction between the two systems have also been shown at central levels, such as the piriform cortex, the rostral insula and the orbitofrontal cortex (3). These areas are also involved in the olfactory hedonic judgment. In our patient, the unpleasant hallucinations could be associated with activation of such areas and, through an interaction between olfactory and trigeminal systems, they may trigger migraine headache.
In conclusion, the reported cases of olfactory migraine aura (1) and olfactory deafferentation associated with migraine-like headache in this case report emphasize the role of the olfactory system in migraine pathophysiology.