Abstract
Background
There are cases in the headache literature described as “cluster-migraine,” but none of stereotyped cases of cluster headache evolving into status migrainosus. We believe this is the first documented case of “cluster-status migrainosus”.
Case
A 54 year-old female hospital administrator presented with headaches with a unique periodicity and semiology that were acutely responsive to high-flow oxygen. She experienced cluster headache attacks every Thursday morning at 3–4 am, which would evolve into status migrainosus lasting through Sunday evening. These attacks were preceded by prodromal depressed mood changes and fluid retention, and later followed by postdromal euphoria and auto-diuresis. These attacks initially occurred every other week and progressed to weekly attacks for 1.5 years. These headaches did not respond to trials of propanolol, sodium valproate, topiramate, amitriptyline, gabapentin, and carbamazepine for preventive treatment or to oral sumatriptan and butalbital-acetaminophen-caffeine for acute treatment. We started her on high-flow 100% oxygen for cluster headache, which successfully aborted greater than 80% of her weekly cluster headache attacks and prevented them from evolving into status migrainosus.
Conclusion
We believe this is the first case of “cluster-status migranosus” described in the medical literature. High-flow oxygen both aborted the cluster headaches and prevented the ensuing status migrainosus.
Introduction
There are case reports and series in the medical literature of headache presentations with features of both CH and migraine that have been called “cluster-migraine” (1). These cases have been different from one another and do not seem to represent a distinct entity. Some physicians argue that overlapping symptoms between CH and migraine signify clinical variability with the spectrum of these disorders (1). There has also been speculation regarding the “transformation” of migraine into CH (2). In a study of over 1600 patients, 14 were found to have migraine occurring first, then either replaced by CH or continued concomitantly during their CH years (3). However, in a case series of 10 patients with coexisting migraine and CH, nine had both forms of headache occurring synchronously (4). We believe that this is the first reported case of stereotyped episodes of cluster headache progressing into status migrainosus, with an atypical biweekly and then weekly periodicity.
Case report
A 54-year-old female with prior history of migraine with aura was evaluated for a new semiology of headache. The patient described the headaches as a severe, stabbing, left side-locked headache, orbital and/or retroauricular in location and associated with ipsilateral eye lacrimation, ptosis, and nasolabial flattening. These attacks would begin at 3–4 am on Thursday mornings and cause the patient to rock in bed and pace around the room. This type of headache would persist for several hours before evolving into a throbbing/pulsating left side-locked headache associated with photophobia, phonophobia, and severe nausea without vomiting, which would persist until Sunday night. While euthymic at baseline, one to two days before each headache attack, she would experience sudden, significant prodromal depressed mood. She also reported prodromal fluid retention. At the end of each weekly headache cycle, she would experience a euphoric mood elevation and auto-diuresis. These headaches initially occurred every other week, but then increased in frequency to once a week and had been present for 1.5 years at the time of our initial evaluation. While her previous episodic migraines were responsive to butalbital-acetaminophen-caffeine and oral sumatriptan, her new headaches were not responsive to these oral medications.
Her typical bedtime was 10pm; she tried going to bed at different times to see if this would prevent her headache attacks, but regardless of bedtime she would be awakened by these headache attacks at 3–4 am.
Timeline of patient’s headache history and treatments.
About 8 years prior to her presentation, she experienced migraine chronification, which was treated with botulinum toxin injections every 12 weeks for eight cycles. She reverted to episodic migraine, and her last botulinum toxin injection was six years prior to this presentation. About 10 years prior to presentation, she was diagnosed with Ramsay-Hunt syndrome, affecting the right V2 more than V1 and V3 distributions with residual post-herpetic facial neuralgia. However, she stated her residual right-sided facial pain was much less severe than this left-sided headache.
The patient endorsed a maternal family history of migraine and denied family history of cluster headache. At this time, she was working as a registered nurse and hospital administrator but denied significant work stressors. She denied a change to her weekly schedule that would have explained the onset of new headaches on Thursday mornings. She also denied any tobacco, alcohol or recreational drug use.
The neurological exam and general physical exam were generally unremarkable, except for allodynia in the right V1-3 distribution. The patient had several normal brain MRIs, with the most recent MRI within one year of presentation.
Upon onset of this new headache semiology, the patient was treated by a neurologist for migraine with multiple preventive medications. Propranolol was discontinued for dizziness and hypotension, sodium valproate was discontinued for inefficacy, topiramate caused clouded mentation, amitriptyline was discontinued for inefficacy, gabapentin was discontinued for suicidal ideation, and carbamazepine was discontinued for inefficacy.
After referral to us for another opinion, we discussed that her headache had features suggestive of cluster headache at onset before evolving into a presentation consistent with status migrainosus. During the first clinic visit, we performed a left suboccipital steroid injection with left greater and lesser occipital nerve blocks, which only provided one headache-free week. We also provided her with prescriptions for high-flow 100% oxygen as well as subcutaneous sumatriptan. We considered various cluster headache prevention therapies; the patient was not a candidate for verapamil due to baseline hypotension, she had previously tried sodium valproate and topiramate without benefit, and she did not wish to begin lithium. As such, we started her on melatonin. She soon after discontinued the melatonin due to daytime somnolence, after which we performed a single cycle of PREEMPT protocol botulinumtoxin injections.
While awaiting the oxygen tank and supplies, she did not try the subcutaneous sumatriptan due to her concern for vasospasm.
She started on high-flow 100% oxygen at 12 L/min with a non-rebreather mask for 15 minutes at the onset of her early-morning cluster headaches, which she reported aborted the cluster headaches greater than 80% of the time and prevented them from evolving into status migrainosus. Oxygen remained effective for her for five months, until her attacks spontaneously remitted.
Discussion
This patient presented with a 1.5-year history of side-locked, nocturnal-onset cluster headaches evolving into status migrainosus (5). These curiously occurred in a weekly pattern, initially every other week, and then weekly. These attacks were preceded by a prodromal phase and followed by a postdromal phase most typical of migraine. They were not responsive to her previously effective acute migraine treatments or to numerous trials of migraine preventive medications, but were acutely responsive to high-flow oxygen, which both aborted the cluster headache and prevented the ensuing status migrainosus. Although CH and migraine are usually distinct clinical disorders, our patient vividly described a combination of the two. To the best of our knowledge, this is the first such case of cluster-status migrainosus to be described in the medical literature.
Migrainous features may delay the diagnosis of cluster headache in women, and despite having signs and symptoms suggestive of CH, our patient’s diagnosis was likely delayed by her associated migrainous features. In a survey of over 1000 patients with CH aimed at determining gender differences in clinical presentation of CH, women reported higher rates of nausea, vomiting, and aura associated with their CH. Women also tended to have a second peak of CH after 50 years of age, and with nighttime attacks with a higher pain intensity than men (6).
While high-flow oxygen is very effective for acute treatment of CH, the diagnosis of CH is clinical and is not based on responsiveness to high-flow oxygen. In a double-blind, randomized, controlled trial for acute CH treatment, pain freedom at 15 minutes was 88% in the group receiving 100% oxygen as compared to 20% in the control group (100% room air, with both groups receiving treatment at 12 L/min for 15 minutes via a non-rebreather face mask) (6). The need for rescue medication at 15 minutes was 30% in the 100% oxygen group versus 76% in the control group. This is a much more robust response than was seen in a small randomized, crossover, placebo-control trial studying high-flow oxygen for acute migraine treatment (7). In this study, 22 migraine patients were treated with 100% oxygen versus medical air (21% oxygen) at 10–15 L/min for 30 minutes via a non-rebreather face mask. There was no significant difference between the groups in their pre-specified primary endpoint of mean decrease in visual analog score (VAS) at 30 minutes. Another study of 204 mostly non-CH headache patients in an emergency department found a significant but modest reduction in VAS score at all time points from 15 to 60 minutes among those receiving 100% oxygen versus room air via non-rebreather mask for 15 minutes (8). High-flow oxygen prevented few patients from requiring pharmacotherapy to treat their headache attack; the percentage of patients requiring analgesic medications differed between the two groups but was still relatively high (72.5% in the oxygen group versus 86.3% in the placebo group; p = 0.005). Only 1% of the subjects in this study were diagnosed with CH; most were diagnosed by the treating emergency medicine physician as having tension-type headache, migraine, or undifferentiated headache. The aforementioned studies suggest that high-flow oxygen has a more meaningful effect in CH than in non-CH primary headache disorders.
Our differential diagnosis considered several named conditions. Unlike patients with migraine with autonomic features, our patient had unilateral autonomic features and experienced autonomic features prior to, but not simultaneously with, her migrainous features. Unlike patients with hemicrania continua, our patient had several headache-free days every week. And unlike patients with long-lasting autonomic symptoms with associated hemicranias (LASH), her autonomic symptoms did not precede or follow her headache and were less prominent than her pain. Indomethacin has not yet been tried as the patient was satisfied with her response to high-flow oxygen.
There are questions about her case that remain. While it is common for cluster headaches to occur on a circadian schedule at the same hour each day (9), it is atypical for CH to occur with weekly or every-other-week periodicity. We do not have a physiological explanation for this weekly periodicity of her headache attacks.
In summary, in this uncommon presentation best described as “cluster-status migrainosus”, treating the cluster headache at onset with high-flow oxygen aborted both the cluster headache and prevented the patient’s ensuing status migrainosus. High-flow oxygen is a safe, well-tolerated, and effective acute treatment for CH that can serve an important role in clinicians’ armamentarium for the treatment of appropriately selected headache patients.
Footnotes
Clinical implications
We believe this is the first description of cluster-status migrainosus in the medical literature. High-flow oxygen aborted the acute cluster attacks and prevented the typical ensuing status migrainosus.
Cluster headache in women may go undiagnosed due to higher associations with nausea, vomiting, and aura.
High-flow oxygen can be a uniquely effective treatment in appropriately selected headache patients.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.