Migraine Surgery: Does it make sense?
Cerebral TorqueThis post is for educational purposes only and may contain errors. Please talk to your neurologist.
Let’s start with what migraine surgery is.
Migraine surgery involves decompressing various nerves in the face and head that are proposed to trigger migraine attacks. Proponents argue that decompressing these trigger sites can prevent migraine attacks or reduce their frequency and/or severity. According to their current research, the compression of said nerves may be caused by muscle, fascia, fat, or even bone.
The underlying theory is that compression or irritation of sensory trigeminal or cervical nerve branches in certain anatomical regions can generate or transmit migraine pain signals. Specific trigger points have been mapped across the head, face, and neck including areas in the glabellar and brow region, temples, external ear, mastoid area, occiput, upper neck, and nasal cavity.
With all these trigger sites, the next obvious question is how plastic surgeons choose which sites to address in a particular patient. Well, they ask the patient where the pain starts and then they instruct them to point to the pain with a finger. Then, depending on the site, they will inject botulinum toxin or administer a nerve block to determine if they are a candidate for the surgery or they will use a doppler ultrasound to identify possible culprit blood vessels.
Various surgical approaches have been developed to access and decompress different nerves depending on the anatomically distinct trigger site region. These include open procedures, endoscopic techniques, or a combination of open and endoscopic approaches.
Some of the most frequent target sensory nerves include the supratrochlear, supraorbital, zygomaticotemporal, auriculotemporal, greater occipital, lesser occipital, and third occipital nerves. Dissection and neurolysis along the nerve's path are performed to release any structural compressions (muscle, fascia, fat, and bone) which may be contributing to headache pain.
Now that we have a cursory understanding of migraine surgery, let’s discuss why the biological basis of it is not exactly coherent.
Migraine is understood to be a neurological disorder ORIGINATING in the CENTRAL nervous system, not arising from compression of PERIPHERAL craniofacial nerves or muscles. Decades of research point to migraine resulting from abnormal brain activity, neurotransmitter imbalances, and neuropeptide involvement. The headache phase likely stems from activation of the trigeminal nerve system and pain sensitization in the brainstem. NO facial nerves have been substantively linked to directly precipitating migraine attacks. Decompressing these nerves would not be expected to impact the underlying brain dysfunctions causing migraine. The rationale behind trigger site surgery contrasts sharply with modern scientific understanding of migraine pathophysiology.
If you care for a more in-depth discussion of migraine pathophysiology, see my comment on Reddit where I debunk a “biohacker”:
Proponents of migraine surgery will ALWAYS cite a 2009 randomized placebo-controlled trial showing benefit, but this study had major shortcomings. This is, in fact, their holy grail migraine surgery study.
First, the study:
This was a randomized, double-blind, placebo-controlled trial investigating the efficacy of surgical deactivation of migraine headache trigger sites.
75 patients with moderate to severe migraine headaches meeting ICHD-II criteria were enrolled. Patients had their predominant migraine trigger site identified (frontal, temporal, occipital) and were randomly assigned to receive either actual surgery or sham surgery at that site.
49 patients received actual surgery (19 frontal, 19 temporal, 11 occipital) and 26 received sham surgery (10 frontal, 9 temporal, 7 occipital).
Outcomes were assessed at baseline and 1 year after surgery.
At 1 year, 28/49 (57.1%) in the actual surgery group reported complete elimination of migraine compared to only 1/26 (3.8%) in the sham group (p<0.001). 41/49 (83.7%) in the actual surgery group had ≥50% reduction in migraine frequency, intensity, duration or index compared to 15/26 (57.7%) in sham group (p=0.014). All migraine measures (frequency, intensity, duration, MIDAS, etc.) were significantly improved from baseline in the actual surgery group but not in the sham group.
The study concludes surgical deactivation of peripheral migraine trigger sites is an effective alternative treatment for patients with frequent, moderate-severe migraine refractory to standard treatments.
Now, the very critical flaws of this study:
The study relied on the false assumption that botulinum toxin is an effective prophylactic treatment for EPISODIC migraine. However, multiple high-quality randomized controlled trials have conclusively shown that botulinum toxin provides no statistically or clinically significant benefit over placebo injections for reducing episodic migraine frequency, intensity, or duration. Response rates to botulinum toxin and placebo are remarkably high in episodic migraine trials, often over 50%, highlighting the powerful placebo influence.
Sources:
https://headachejournal.onlinelibrary.wiley.com/doi/
https://journals.sagepub.com/doi/10.1111/j.1468-2982.2007.01315.x
https://pubmed.ncbi.nlm.nih.gov/17716321/
https://journals.sagepub.com/doi/10.1111/j.1468-2982.2004.00754.x
Therefore, the criteria for selecting surgical candidates actually selected heavily for patients that are placebo responders. Of 130 patients receiving botulinum toxin injections, 76 "responders" showing at least 50% improvement were selected for the trial. However, given the lack of efficacy of botulinum toxin for episodic migraine, these responders were likely just exhibiting placebo responses, not true drug effects. Selecting these subjects biased the trial towards showing surgical benefits.
Furthermore, the innervation of migraine is complex and involves multiple branches of nerves, making anatomical localization unreliable. Even if pain is localized in one location during one migraine attack, migraine pain is often bilateral and variable in location from one attack to the other and/or during an attack.
The use of questionable selection criteria in this migraine surgery study seems to stem from an inappropriate analogy between migraine and carpal tunnel syndrome that some plastic surgeons continue to purport.
It has become common for plastic surgeons to compare migraine surgery to carpal tunnel release procedures. However, this analogy rests on a flawed understanding of migraine neurophysiology.
Carpal tunnel syndrome is a focal compression neuropathy caused by entrapment of the median nerve at the wrist. The pathology is localized, and surgical release of the transverse carpal ligament decompressing the nerve often successfully relieves symptoms.
Migraine, on the other hand, is not a peripheral nerve disorder. It is a complex neurological disease involving the trigeminal vascular system, central pain pathways, and multiple regions of the brain. Migraine attacks originate from dysfunctions in these central nervous system processes.
Equating migraine surgery to carpal tunnel release therefore vastly oversimplifies the underlying pathology. It treats migraine as if it were a minor peripheral nerve issue rather than the serious and multifaceted central disorder that it is.
Moreover, relying on patient-reported predominant pain sites to identify "trigger points" incorrectly reduces migraine to a localized compression neuropathy model. Effective migraine treatment requires an understanding of the diverse neurobiological factors involved.
This leads us to the next issue. Is a sham surgery even possible in this case? Would you not be able to tell if you had a fake surgery that just consisted of an incision?
Specifically, patients undergoing frontal migraines surgery involving corrugator resection would have expected smoothing of frown lines and reduced wrinkling in the glabellar area. These visible cosmetic changes would allow patients to guess they received active surgery rather than sham procedures.
Supporting this unblinding hypothesis is the finding that frontal migraine surgery demonstrated greater benefits than occipital procedures. With occipital surgery on the back of the head, patients could not see the results and would have minimal cues as to whether they had active or sham surgery. The significantly larger impact of frontal surgery, visible to patients, strongly suggests an unblinding influence whereby visible cosmetic changes enhanced placebo responses.
Similar unblinding issues due to visible cosmetic changes have been documented in studies of botulinum toxin for migraine treatment. Patients receiving botulinum toxin could discern whether they had active treatment based on observed smoothing of wrinkles, and those perceiving cosmetic improvement had higher expectations of headache relief.
Altogether, the sham-controlled design of this surgical trial was vulnerable to unblinding and enhancement of placebo effects due to visible cosmetic changes from the frontal procedures. Effective blinding is critical in surgical trials but was likely suboptimal in this study, confounding the results.
The idea that the placebo effect cannot last up to 1 year like in this study or even up to 5 years is purely anecdotal and not founded in science.
In fact, several sham-controlled trials of placebo surgeries for various conditions demonstrate that sham procedures can induce remarkably durable placebo effects extending well beyond 1 year.
In a trial of sham surgery for angina pectoris, patients receiving a ligation of their internal mammary artery showed substantial symptom improvements at 6 months compared to medication therapy. However, those receiving a sham incision without artery ligation showed identical improvements, indicating a 6-month placebo effect from the sham procedure.
Source: https://linkinghub.elsevier.com/retrieve/pii/S0033062012001715
In a study of endolymphatic sac surgery for Meniere’s disease, patients undergoing the real surgery showed control of vertigo attacks at 1 year. But those receiving sham surgery showed similar vertigo control through THREE YEARS and more sustained than oral placebos.
Finally, the success rates according to this study are so absurd that it only makes sense that it was placebo.
There was obvious backlash against the study due to aforementioned placebo issues, so we now have a 5-year follow up study by the same author.
The study: https://pubmed.ncbi.nlm.nih.gov/20966820/
And now to the problems…again.
There was no control group. The original study randomized patients to treatment or a control group, but the control group only completed 1 year of follow-up. Without a control group maintained over the 5-year follow-up, there is no way to definitively attribute the improvements seen in the treatment group to the surgical intervention rather than confounding factors like placebo effects, natural course of migraine, regression toward the mean, or spontaneous improvement/remission.
A sham surgery control arm is essential for the entire duration of long-term follow-up in surgical trials. Comparing 5-year results just to baseline is inadequate because, as we have previously seen, placebo responses can extend for years- especially after a surgical procedure.
Moreover, this study enrolled patients during active migraine flares when they presented for treatment. By chance alone, their migraine frequency, intensity, and duration are expected to decrease over time after enrollment even without any intervention. The reductions seen over 5 years could simply reflect this expected improvement as extreme symptoms at enrollment regress to each patient's average migraine baseline.
Without a control group, there is no way to isolate true treatment effects from this expected regression toward individual baselines. The long-term improvements may have occurred equally with or without surgery. Analyzing changes only within the treatment group versus baseline is prone to regression artifacts, since the baseline represents a symptom spike rather than patients' usual migraine status.
A randomized control group provides a way to factor out regression toward the mean by comparing surgery outcomes to the natural improvements in the control arm over the 5-year timeline. But, in this study, that comparison can only be made at 1 year, not long-term.
Furthermore, the average age of patients in the study was about 44 years old. This also represents when the rate of remission from migraine attacks is relatively high. Again, without a control group, assessing for the 5-year outcome seems like a large oversight.
Source: https://rdcu.be/dojLv
Faced with a lack of biological rationale for the procedure, plastic surgeons use the following studies to explain how peripheral nerve surgery impacts the centrally mediated pathology of migraine disease.
https://pubmed.ncbi.nlm.nih.gov/24673461/
https://pubmed.ncbi.nlm.nih.gov/19425099/
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4058402
https://pubmed.ncbi.nlm.nih.gov/23707274/
It’s important to note that these are not clinical studies, but anatomical studies.
While interesting, it still does not benefit migraine surgery. Imaging studies show that migraine attacks begin with neural excitation originating near the brainstem, followed by cortical spreading depression. This points to central trigeminovascular neurons as the primary migraine trigger.
Many migraine therapies like triptans that modulate central serotonin receptors are effective for acute attacks. This suggests central pathways are key for migraine pain.
Trigeminal nerve axons project centrally to the trigeminal nucleus and trigeminocervical complex. So peripheral nerves provide input to central pathways but don't drive migraine on their own. While trigeminal afferents do release inflammatory neuropeptides like CGRP during migraine, this is thought to be an effect rather than a cause of attacks. Moreover, migraine can persist even after disruption of peripheral trigeminal nerve pathways, indicating central mechanisms are sufficient for headache generation.
Therefore, while peripheral nerves connecting to the CNS may modulate central pain pathways, the bulk of evidence indicates migraine fundamentally arises from dysfunctional central neurophysiology. Since peripheral nerves do not drive migraine on their own, surgically manipulating them may be ineffective for prevention or treatment. Migraine surgery aims to treat a peripheral consequence rather than the underlying central cause of headaches.
Other types of surgeries that ACTUALLY made sense have shown to be ineffective.
This is why a proper, well-designed study is so important.
Take for example a procedure known as arthroscopic partial meniscectomy (APM). This surgery is one of the most popular surgeries in the United States and sham surgery in this scenario is possible as there are no cosmetic differences.
The reality of this surgery is that orthopods (😉) know this surgery doesn’t work for knee pain (this isn’t controversial), but they perform it anyway.
How do they know this? A properly done study.
Take this study: https://www.nejm.org/doi/full/10.1056/NEJMoa1305189#t=article
This randomized controlled trial examined whether arthroscopic partial meniscectomy improves outcomes compared to sham surgery in patients with knee pain and a torn meniscus, but no arthritis.
146 patients aged 35-65 were randomly assigned to receive either real partial meniscectomy or a sham procedure where the surgery was simulated but no tissue removed. After 12 months, there was no significant difference between the two groups in improvement in knee pain after exercise, Lysholm knee score, or Western Ontario Meniscal Evaluation Tool score.
The authors concluded that in patients without knee osteoarthritis, undergoing arthroscopic partial meniscectomy did not provide additional benefit over sham surgery. This suggests that for patients with a torn meniscus but no arthritis, arthroscopic surgery may not be an effective treatment for knee pain.
It makes perfect sense for this surgery to work, but it just does not. Other studies found the surgery to be just as effective as physical therapy alone.
Source: https://www.nejm.org/doi/full/10.1056/nejmoa1301408#t=article
Now, when confronted with migraine surgery that is heavily lacking in a biological rationale, it should be significantly critiqued and properly studied.
The next question some may ask is why some procedures, like botulinum toxin injections, work on migraine if nerve decompression is lacking in biological evidence.
These procedures also modulate the central nervous system.
Botox and peripheral nerve blocks can "calm things down" both at peripheral sites and in the central trigeminal pathways to reduce widespread sensitization and reactivity to pain.
Botox inhibits the release of key neurotransmitters involved in migraine pathogenesis, including calcitonin gene-related peptide (CGRP), substance P, and glutamate. The inhibition of these neuropeptides blocks peripheral sensitization of trigeminal nociceptors, which is thought to be an important initial step in migraine attacks. Additionally, Botox prevents the translocation of transient receptor potential (TRP) channels TRPA1 and TRPV1 to the neuronal cell membrane, where they normally amplify nociceptive signaling. By inhibiting neurotransmitter release and TRP channel translocation, Botox effectively "switches off" peripheral sensitization. This then prevents the cascade of central sensitization in the trigeminal nucleus caudalis and higher brain centers that generates the pain, photophobia, phonophobia, and other symptoms experienced during a migraine attack.
Therefore, the multimodal mechanisms of action of Botox in blocking key steps in migraine pathogenesis make it an optimal procedure for chronic migraine patients as opposed to surgery.
If surgical candidates are chosen by their response to Botox or nerve blocks, why not just continue using Botox?
The cost to the patient is lower since insurance will cover Botox for migraine, but not migraine surgery.
The complication risk for the patient is also lower.
Patients with migraine also have a chance at spontaneous remission or improvement often occurring with increasing age, after menopause, or during pregnancy.
If nerve decompression is necessary, then how is remission explained with increasing age? How does pregnancy in the second and third trimesters result in migraine relief for many? I wrote an article on this here
Why do we know the genetics involved in hemiplegic migraine if relief is only a few snips away? Read more about this here
Finally, if migraine is purely a peripheral neuropathy that somehow impacts the CNS, the comorbidities associated with migraine disease would be difficult to explain. Carpal tunnel syndrome patients don’t have a high incidence of gastroparesis cases, for instance.
With a success rate of over 90% in some migraine surgery studies, one may assume the problem is purely anatomical, but we know this is not the case.
Why not just give it a try? Nothing to lose? Let’s talk about some of the potential risks and complications.
All surgeries carry risks.
Because this surgery is done under anesthesia, this is the first risk, and it is not a minimal one.
Other risks include, but are not limited to, infection, hemorrhage during surgery or post-operatively if a vessel is accidentally nicked, nerve damage (there will always be, at least, temporary neuropathy and gabapentin will be given. This is also another confounding variable in migraine surgery studies as gabapentin is also used for migraine prevention along with other possible medications administered), vascular injury, muscle or soft tissue damage (this may lead to cosmetic deformities, asymmetry, or loss of function), New headache triggers or worsening migraine, facial fat grafting risks if this procedure is done to cushion the nerve, chronic pain due to damage to nerves, scarring, or inflammation, etc.
Anesthesia dolorosa aka painful numbness may also be a consequence of migraine surgery. It is a debilitating chronic pain condition characterized by persistent severe pain in areas of the body that lack sensation. Migraine surgery can damage sensory nerves and lead to anesthesia dolorosa if the nerves start misfiring or malfunctioning, sending incorrect pain signals to the brain. Patients with anesthesia dolorosa following migraine surgery describe feeling intense burning, throbbing, or electric pain in areas of their head, neck, or face that have lost normal sensation. This chronic neuropathic pain can be very difficult to treat, often requiring a combination of medications, nerve blocks, electrical stimulation, or further surgeries to try to mitigate the pain signals.
Expert Opinion Advises Caution
Considering the lack of biological rationale and adequate supporting evidence, the American Headache Society and other experts advise against routine migraine trigger site surgery outside of clinical trials. They contend claims of benefits are premature and likely explained by placebo factors. More research is first needed to definitively determine whether this surgery provides true effectiveness for migraine patients. Less invasive options should be tried first, like Botox injections which have demonstrated efficacy in trials and FDA approval. For those wishing to try migraine surgery, carefully weighing the risks versus unproven benefits is advised.
