Cluster headache vs. migraine — why the distinction matters
Cluster headache is sometimes called “suicide headache.” The name is grim, but it captures something true. The pain is consistently rated as more severe than childbirth, kidney stones, or gunshot wounds in published surveys of patients who have experienced more than one of those. Attacks last fifteen minutes to three hours, strike on one side of the head behind or around the eye, and arrive in clusters of one to eight per day during a cycle that can run weeks or months.
People who have not had cluster headache often assume it is just a bad migraine. It is not. The mechanisms differ. The treatments differ. And the urgency differs: a cluster patient mid-cycle is in a category of suffering that very few other conditions reach.
Migraine involves cortical spreading depression and a complex neurovascular cascade. Cluster headache appears to be driven by hypothalamic dysfunction interacting with the trigeminal autonomic reflex. The autonomic features — tearing, nasal congestion, drooping eyelid, restlessness during the attack — are diagnostic. Patients pace, rock, and sometimes bang their heads against walls. They do not lie still in dark rooms the way many migraine patients do.
This distinction matters clinically because the treatment ladder is different, the response patterns are different, and the role of an emerging therapy like ketamine has to be understood in the context of cluster-specific evidence rather than extrapolated from migraine data.
Why standard preventatives often fail
The American Headache Society’s 2023 treatment guidance is clear about first-line care. For acute attacks, high-flow oxygen at 12–15 liters per minute via non-rebreather mask and subcutaneous sumatriptan are first-line abortives. For prevention during a cycle, verapamil is the cornerstone, often at doses higher than those used in cardiology, with EKG monitoring as the dose climbs. A short course of corticosteroids or a greater occipital nerve block can bridge patients while verapamil is being titrated. Galcanezumab is FDA-approved for episodic cluster prevention.
For many patients, this works. For a stubborn minority — particularly those with chronic cluster headache, defined as attacks without a remission of at least three months — first-line and second-line treatments fail. Verapamil is not tolerated, the cycle breaks through galcanezumab, oxygen is not enough during the worst attacks, and the sumatriptan injection limit has been reached. These are the patients who end up searching for anything that might work.
Ketamine is FDA-approved as an anesthetic; its use for cluster headache is off-label. Headache specialists have nonetheless been studying it because the underlying pharmacology fits the disease.
The mechanism: NMDA, glutamate, and the trigeminovascular system
Ketamine is a non-competitive NMDA receptor antagonist. NMDA receptors govern the way the nervous system amplifies pain, a process called central sensitization. In chronic pain conditions generally, sustained glutamate signaling at NMDA receptors keeps the spinal cord and brainstem in a hyper-excitable state. Ketamine’s effect on these receptors can interrupt that loop.
Cluster headache is not a classic chronic pain syndrome, but the trigeminovascular system that generates the pain shares the same NMDA-mediated machinery. Animal models of trigeminal pain show that NMDA blockade reduces nociceptive transmission. Clinically, this raises a hypothesis: a sufficiently dosed ketamine exposure might quiet the trigeminal pathway long enough to abort an active cluster cycle, the way a hard reset can sometimes break a feedback loop in a stuck system.
Ketamine also has effects on opioid receptors, monoamine systems, and inflammatory mediators. Researchers are not entirely sure which of these mechanisms matters most for cluster headache. The clinical signal showed up first; the mechanistic explanation is still being worked out.
What the systematic review found
The most rigorous summary of the existing evidence is the 2024 systematic review by Granella and colleagues, published in Pain and Therapy. The review pulled together four reports covering 68 patients with cluster headache who received intravenous ketamine. Outcomes were tracked at two weeks and beyond.
The headline numbers were striking. At the two-week mark, 100% of episodic cluster patients in the pooled data were pain-free, and 54% of chronic cluster patients were pain-free. Among the chronic patients who responded, many remained pain-free for weeks to months. Adverse events during the infusions were typically mild and self-limited — the dissociative and cardiovascular effects ketamine is known for, managed with standard monitoring.
Those numbers come with caveats the authors are explicit about. The studies pooled were small. None were placebo-controlled. The quality of evidence is graded as low. Open-label and retrospective designs cannot rule out placebo effect, regression to the mean, or natural cycle resolution. A signal this large in such a refractory population is hard to dismiss, but it is a signal, not proof.
Intranasal vs. IV ketamine for acute attacks
The other named study worth knowing is Petersen and colleagues, published in Headache in 2022. This was an open-label proof-of-concept trial of intranasal ketamine for the abortive treatment of individual cluster attacks rather than for breaking the cycle as a whole.
In Petersen 2022, patients self-administered intranasal ketamine at the onset of an attack and rated pain over the following minutes. A majority of treated attacks showed meaningful pain relief within roughly fifteen minutes. The trial was small and uncontrolled, so it cannot establish efficacy on its own, but it pointed at a different use case: rapid-onset abortive therapy that a patient could carry rather than an in-clinic infusion course meant to break a cycle.
The two routes serve different roles. Intravenous ketamine, given over a multi-hour or multi-day protocol, is being studied as a cycle-breaker. Intranasal ketamine is being studied as an abortive for individual attacks. Music City Ketamine works in the IV space for the cycle-breaking question; intranasal ketamine for cluster abortive use is generally a question for your headache specialist and a compounding pharmacy.
Evidence quality: small studies, large effect sizes
It is worth being direct about what the evidence is and is not.
- The studies are small. Sixty-eight patients across four reports is a slim foundation for any conclusion.
- The studies are mostly uncontrolled. Without a placebo arm, we cannot fully separate the drug effect from natural cycle resolution or expectation effects.
- The effect sizes are large. Pain-free rates of 100% and 54% in a population that has, by definition, failed standard care are not the kind of numbers seen in modest treatments.
- The safety signal is acceptable. Across the published reports, infusions were tolerated with the expected ketamine side effect profile and standard monitoring.
This is a pattern familiar from early ketamine research in chronic pain and depression: small open-label data with striking effect sizes, eventually followed by larger and more rigorous trials. The next step in cluster headache is the larger trial.
Where the upcoming RCT may shift things
A multicenter randomized placebo-controlled trial of ketamine plus magnesium for chronic cluster headache, registered on ClinicalTrials.gov as NCT04814381, is currently recruiting at French sites. It is the first prospectively designed RCT for this indication. Magnesium is included because it has its own NMDA-modulating properties and is sometimes used adjunctively in headache medicine.
If that trial reports a positive result with a credible effect size, ketamine moves from “promising open-label data” to a more standard place in the refractory-cluster ladder. If the trial is null or shows a small effect against placebo, the picture shifts toward caution and patient selection. Either result is useful. The current honest answer is that we are operating on the best evidence we have, which is real but incomplete.
Until that trial reads out, ketamine for cluster headache is a reasonable option for refractory patients under specialist care. It is not a first-line treatment. It is not appropriate for someone who has not yet completed an adequate trial of high-flow oxygen and verapamil.
How we approach refractory cases in Nashville
If you have chronic or refractory cluster headache and your headache specialist has discussed ketamine as a potential option, here is what working with us looks like.
We start with a consultation. We will review your diagnosis, your treatment history, your current medications, and ideally a note or referral from your neurologist. Cluster headache is not a self-diagnosis call. We need confidence that the diagnosis is correct and that first-line treatments have been adequately tried.
If ketamine is appropriate, we plan a protocol that fits the cycle-breaking literature. Marla Peterson, CRNA, oversees every infusion, with anesthesia-level monitoring throughout each session — continuous pulse oximetry, blood pressure, and heart rate. A CRNA is available throughout. The setting is private, calm, and built for the kind of rest that matters when you are mid-cycle.
We coordinate with your headache specialist on dosing and follow-up. We do not replace your neurologist. Cluster headache management is long-term, and the people who know your history, your prior responses, and your overall plan are the people best positioned to quarterback your care.
For practical expectations on the first visit, our walkthrough on what a first ketamine infusion is like applies, with cluster-specific dosing adjustments. Safety considerations follow the framework in our overview of ketamine safety. You can also see how the broader process works for any indication.
Honest expectations
We want to be straightforward about what ketamine can and cannot do for cluster headache.
- It is not a cure. Cluster headache is a long-term neurological condition. No current treatment, including ketamine, eliminates the underlying disease. The realistic question is whether ketamine can break a cycle, reduce attack frequency, or shorten a cycle’s duration.
- Not everyone responds. The systematic review showed weaker results in chronic patients than in episodic patients, and a meaningful share of patients did not become pain-free. We cannot guarantee a response.
- Duration of relief varies. Some patients in the published data remained pain-free for weeks to months; others relapsed sooner. We cannot promise a specific duration.
- It is not first-line. If you have not yet had an adequate trial of high-flow oxygen and verapamil under a headache specialist, those come first. Ketamine is a refractory-care option.
- Cost is a real factor. Insurance typically does not cover ketamine for cluster headache because the indication is off-label. At Music City Ketamine, sessions are $475 each, and we are transparent about cost from the start.
- This decision belongs to you and your specialist. Do not start, stop, or change any of your headache medications without talking to your prescribing provider.
If you have lived through a cluster cycle, you already know how high the stakes are. The fairest summary of the evidence is this: ketamine may help, especially for refractory and episodic patients, the safety signal in monitored infusion settings is acceptable, and the first large RCT is on its way. That is enough to make ketamine worth a serious conversation. It is not enough to make it a first move.