- Top of page
- Case Report
- Discussion and Conclusion
A 28-year-old man visited our outpatient clinic because he had suffered 3 stereotyped headache attacks: 2 years, 1 year, and 6 weeks before his visit. Each attack immediately followed repeated and violent sneezing, and consisted of a severe holocranial, pulsating headache with nausea, vomiting, photophobia, and phonophobia, increasing after exercise and lasting 1 day on all 3 occasions. All symptoms completely resolved after 1 night of sleep. Only during the first attack, the patient noticed some blurring of vision for several hours during the headache, but he never experienced diplopia, speech problems, confusion, paraesthesias, or loss of muscle strength. There were no symptoms of allodynia. At the time of the visit to the hospital, the patient was symptom-free. He admitted to have become afraid of sneezing because of these 3 experiences. The patient had, however, sneezed several times since his first headache attack, mostly in series of 10-20 powerful sneezes, but these events did not all induce a headache. He also had to cough regularly, but this had never resulted in a headache attack. Except for the day following each of 3 attacks, he never had any headaches. His medical history was further unremarkable; he did not take any medication, did not smoke, nor drank more than 2 glasses of alcohol and 1 cup of coffee per day. He refrained from cannabis, serotonergic drugs, sympathomimetics, or other (neuroactive) compounds. The patient's mother suffered from moderate headache attacks without accompanying symptoms once or twice per year not provoked by sneezing. At examination, patient's blood pressure was 150/80 mmHg and at repeated measurement 120/75 mmHg. The general and neurological examinations were also without any remarkable findings. A magnetic resonance imaging (MRI) examination, including a magnetic resonance angiography (MRA) of the intracranial and extracranial vessels 7 weeks after the last attack, was unremarkable. In specific, there were no structural abnormalities such as an aneurysm that remains of old hemorrhages, intracranial or extracranial arterial dissection, pituitary apoplexy, signs of low cerebrospinal fluid pressure, or signs of vasoconstriction or tonsilar descent in the posterior fossa. At follow up, 6 weeks and 6 months later, the patient was still symptom-free, although he had sneezed and coughed several times since his first visit.
Discussion and Conclusion
- Top of page
- Case Report
- Discussion and Conclusion
The patient we describe here suffered from 3 acute headache attacks after sneezing. Although he was not investigated directly after the occurrences, making a definite diagnosis difficult, his clinical history and normal MRI and MRA examinations nevertheless make several possible causes for these 3 “thunderclap headaches” very unlikely, as the MRI examination in our patient did not show any signs of structural abnormalities known to cause thunderclap headache.[1, 2, 4, 6] Besides, headaches after a subarachnoid hemorrhage do not usually resolve in a night's sleep, and a “delayed” MRI will often show remnants of old hemorrhages or an aneurysm indicating a “warning leak.”[9, 10] Although a reversible benign central nervous system (CNS) angiopathy could not be excluded completely, we feel that in our patient, the overnight resolution of the symptoms, the absence of specific precipitants, and the time course (3 separate attacks in 2 years) made such an angiopathy very unlikely.[2, 4, 6] Sneezing has, however, been described as trigger factor for reversible benign CNS angiopathy.
For thunderclap headache, not because of a structural lesion, several other possible diagnoses exist of “primary,” “idiopathic,” or even “cryptogenic” (with a suspected cause that can not be found) origin: “primary thunderclap headache” (PTH; International Headache Society [IHS] no. 4.6), “PCH” (IHS no. 4.2), and “primary exertional headache”(PEH; IHS no. 4.3).
In our patient, PTH is not likely for the occurrence on 3 separate occasions and resolution by sleep.[6, 11] In many cases of PTH, a lingering less severe headache persists for weeks after headache onset. Furthermore, there have been no reports on repeated PTH exclusively after sneezing. PCH can be precipitated by sneezing and other non-cough triggers, such as straining at stool, bending down, heavy lifting, Valsalva maneuver, laughing, crying, and sexual activity.[3, 5, 7] The pathophysiology of PCH is unclear and is thought to be related to a transient increase of intracranial pressure because of venous or arterial distension in combination with other factors such as hypersensitivity of baroreceptors or release of vasoactive peptides.[3, 7, 12] PCH was, however, not likely to be the correct diagnosis in our patient, as in this entity, the headache lasts for 30 minutes at maximum. Associated features such as nausea, vomiting, photophobia, and phonophobia are also uncommon.[3, 5, 7] A diagnosis of PEH[3, 7, 11] could still be considered, although the patient did not perform prolonged physical exercise before the attacks. In this entity, headache can last from 5 minutes up to 48 hours, and migraine features are not excluded by the criteria. In fact, in PEH, the headache is often indistinguishable from migraine, and comorbidity of PEH and migraine is quite common.[7, 13, 14] Sneezing has been described as a trigger-factor but specifically in association with a Chiari malformation.
Although in our patient all 3 headache attacks fulfilled IHS criteria for migraine without aura, strictly, we were not allowed to make this diagnosis, as this requires the occurrence of at least 5 typical attacks. Nevertheless, based on these diagnostic considerations and on the remarkably stereotypic attacks in our patient on 3 occasions resolving after one night of sleep, we made a tentative diagnosis of “migraine without aura after sneezing,” which we here call “sneeze-migraine.”
There are many precipitating factors for migraine, including physical activity, straining, stress, cessation of stress, skipping meals, fatigue, changes in sleep pattern, menstruation, and weather changes.[8, 14-19] Additionally, rare syndromes are described, such as “crying migraine,” “laughing migraine,” “Chinook wind migraine,” and “hair wash or head bath migraine.”[17, 19] Sneezing has not often been mentioned as a trigger factor. In fact, the IHS criteria mention the precipitant “sneezing” only once in: 7.4.1 – “Headache attributed to increased intracranial pressure or hydrocephalus caused by neoplasm,” and not in association with migraine. The only study that included sneezing in an analysis of trigger factors for migraine associated it with coughing and found that “coughing/sneezing” was reported by 7/38 (18.4%). Coughing is much more often associated with headache/migraine than sneezing. A PubMed search (July 2012) resulted in 74 hits for “migraine and cough,” 85 for “migraine and coughing,” but only 8 for “migraine and sneeze” and 7 for “migraine and sneezing.” Obviously, there was much overlap between the retrieved articles. A combined search of “migraine and sneeze and coughing” revealed only 4 hits, of which only 2 were relevant.[8, 12] Although there seems to be much more attention for the association between migraine and coughing, a pathophysiological relation between sneezing and migraine is more likely, as discussed hereafter.
The pathophysiological mechanisms of coughing are largely unknown. Anatomically, vagal afferent nerve responses from the airways are transmitted to synapses in the brainstem nucleus tractus solitarius, the so-called cough centre. Central cough receptors use glutamate, substance P, calcitonin gene-related peptide, and neurokinin A as neurotransmitters. Physiologically, “protective coughs” (to clear the airway) and “sensory-driven coughs” (in response to an unpleasant sensation) can be distinguished. Little is also known of the sneeze reflex that, similar to the cough reflex, is predominantly protective but with a completely different anatomy with the irritation coming from the nasal mucosa. The reflex has 2 phases: a nasal or sensitive phase, when distal branches of the trigeminal nerve are triggered and stimuli are transmitted to the sneezing center in the lateral medulla.[22-24] In the second phase, the efferent or respiratory phase, not only trigeminal but also parasympathetic branches of the oculomotor nerve are activated. During sneezing, the expelled air and its particles reach estimated speeds of 150-1045 km/hour. Complications of sneezing, although rare, therefore include acute aortic dissection, cerebral venous thrombosis, fractures of the stapes footplate, abortion, emphysema, pneumocephalus, wide-angle glaucoma, retinal hemorrhage, and costal fractures. Besides, in response to nasal irritation, sneezing has been described as a consequence of lateral medullary brainstem syndrome, temporal lobe and grand mal epilepsy, and psychogenic factors. Interestingly, recently, a patient was described with paroxysmal sneezing after stimulation of the posterior hypothalamus for refractory cluster headache, illustrating that parasympathetic stimulation can lead to sneezing possibly through hypothalamic influence on trigeminal nerve function.
It seems that sneezing and coughing involve different CNS mechanisms with different brainstem nuclei and pathways/tracts.[20, 21, 24] In sneezing, the trigeminal nerve complex plays a role, which is possibly affected by input from the hypothalamus. These pathways are also involved in migraine, where trigeminal neurons become hypersensitive to changing pressure in the head, explaining worsening of headache after bending over and sneezing.[15, 26-28] Trigeminovascular projections play an important role in the triggering of a migraine attack.
The patient described here had never had a migraine attack after coughing and also had many series of sneezing not followed by headache. This is in line with the theory that triggers of migraine are not absolutely determinant but often only cause attacks in susceptible individuals in combination with other exogenous or endogenous exposures leading to increased vulnerability.[14, 15] The presence of a precipitant is called “a necessary but not a sufficient factor to initiate an attack.” More than one trigger may be needed simultaneously for attack initiation. Another possible explanation of the migraine attacks that occurred after sneezing in our patient is that sneezing might not have been the trigger for the attacks but merely a premonitory epiphenomenon. According to this view, what a patient perceives as a trigger factor is actually the result of heightened sensitivity of the brain in the prodromal phase. The trigger therefore should not be considered a trigger but an accompaniment of the attack itself.[15, 30] Premonitory symptoms are very heterogeneous and include fatigue, phonophobia, yawning, craving, a stiff neck, and concentration problems.[30-32] Sneezing, however, has never been identified as a premonitory symptom, which could be a sign of low prevalence, or more likely of underreporting. Finally, it is also possible that the combination of sneezing and migraine in this patient was purely due to coincidence. We cannot rule out completely that 3 separate migraine attacks coincidentally occurred after sneezing in an otherwise headache-free patient, but we estimate that this is very unlikely.