Headache Drugs Logo
Home | About Dr. Robbins | Archived Articles | Headache Books | Topic Index  

Back to List



Botulinum Toxin: Efficacy in Migraine, Tension
and Cluster Headache: A Review
Trupti Gokani, M.D. and Lawrence Robbins, M.D.
Posted February 2002

Botulinum neurotoxin has been utilized for various types of headache. A number of studies, both controlled and open-label, have demonstrated efficacy (to various degrees) of botulinum toxin for migraine and chronic daily headache. In addition, several small studies on cluster headache patients indicate that botulinum toxin may be useful in selected cluster patients.

This study discusses the history and pharmacology of botulinum toxin, the pathophysiology of headache, and reviews the literature on botulinum toxin for headache.

Patients with headache often do not adequately respond to preventative medications.1  For those patients who are refractory to the usual first and second-line medications, and the non-medication therapies, the choices are somewhat limited. These include monoamine oxidase inhibitors, long-acting opioids, stimulants, daily triptans, intravenous dihydroergotamine, etc.2  Botulinum toxin injections have been utilized in the last several years as one other possibility for these difficult to treat patients.

Previous studies have indicated that there may be some role for botulinum toxin in the treatment of migraine headache. The studies have varied as to the efficacy of botulinum toxin for muscle contraction or chronic daily headache. There have been several small studies indicating possible efficacy in the treatment of cluster headache. The purpose of this study was to review previous work on botulinum toxin in relation to headache.

History of Botulinum Toxin:
Botulinum neurotoxin is produced by the anaerobic bacterium Clostridium botulinum. In 1897 in Belgium, Emile P. Van Ermengen isolated the spore-forming bacillus from raw pork meat, along with post-mortem tissue from individuals who had died of botulism. He concluded that a toxin existed leading to this disease process.3  Botulism, resulting from small amounts of this toxin entering the bloodstream, leads to a symmetric descending paralysis with prominent bulbar symptoms and involvement of the autonomic nervous system. This disease has been recognized since the early 19th century, when it was believed to occur as the result of eating spoiled sausage. The word "botulism" derives from the Latin word botulus, which means "black sausage."

Dr. Alan Scott, an ophthalmologist, was the first to use this toxin for the treatment of strabismus.4  In 1989, the toxin gained full FDA approval, with indications for strabismus, blepharospasm, and hemifacial spasm. Since that time, the use of botulinum toxin has expanded to involve treatment of dystonia, achalasia, wrinkles, and headache, among other conditions.

Different strains of Clostridium botulinum produce eight serotypes, designed A, B, C1, C2, and D through G. All are proteases with a similar structure, composed of a light chain linked by a disulfide bond to a heavy chain.5  Botulinum toxin type A (BTA) and type B (BTB), when injected into muscle, causes flaccid paralysis by inhibiting release of acetylcholine from nerve terminals. The toxin also inhibits release of acetylcholine in all parasympathetic and cholinergic postganglionic sympathetic neurons. Botulinum toxin type A prevents release of acetylcholine (ACh) presynaptically by cleaving a membrane bound protein, SNAP 25. This results in inhibition of calcium-activated release of ACh. The other serotypes of botulinum toxins act on different sites, such as VAMP and syntaxin.

Pathophysiology of Headache Disorders:
In order to hypothesize the possible mechanism of action of botulinum toxin in headache, it is helpful to review the pathophysiology of the various headache disorders: migraine, chronic daily headache, and cluster.

Migraine Headache Pathophysiology:
Migraine is a neurovascular pain syndrome. The initial stimulus is from a trigger   physical, chemical, or psychological   that leads to a chemical change in the dorsal raphe and locus ceruleus. The dorsal raphe has projections that terminate on cerebral arteries and alter cerebral blood flow. Cranial vascular changes lead to activation of trigeminal afferents. The dura mater and large cerebral vessels, in the anterior and middle cranial fossae, are primarily innervated by branches of the ophthalmic division of the trigeminal and, in the posterior fossa, by the upper cervical dorsal roots.6,7,8

Activation of the trigeminal ganglion leads to release of substance P (SP) and calcitonin gene-related peptide.9  Release of vasoactive neuropeptides (SP and CGRP) occurs at vascular terminations of the trigeminal nerve by means of antidromic stimulation. These peptides include a sterile inflammation along with vasodilation which activates the trigeminal nociceptive afferents on the vessel wall leading to propagation of pain.10  The cycle of pain continues as the trigeminal nucleus is activated, leading to release of inflammatory mediators activating trigeminal afferents on the vessel wall, leading to further pain. After passing through the trigeminal ganglion, the trigeminal afferents project into the trigeminocervical complex, which consists of the trigeminal nucleus caudalis and dorsal horns of C1 and C2. The second order neurons project to the contralateral thalamus, and then on to activate the cortex, and periaqueductal gray, anterior cingulate cortex, insular cortex and frontal cortex. Overall, there is activation of nociceptors of pain, in conjunction with a reduction of the normal functioning of the endogenous pain control pathways that normally gate the pain. This leads to the generation of migrainous symptoms.

Chronic Daily Headache (CDH) Pathophysiology:
The underlying pathogenesis of CDH is less delineated than that of migraine. Some believe that there exists a primary headache spectrum with tension-type headache on one end and migraine on the other.11  Studies show that several mechanisms contribute to the development of CDH. First, there may be an abnormal excitation of peripheral nociceptive afferent fibers. This may be due to chronic neurogenic inflammation. The accumulation of pain-producing metabolites sustains nociceptive input and thereby leads to central sensitization. Pericranial muscle hardness and tenderness are increased in CDH patients. Secondly, an enhanced responsiveness of the trigeminal nucleus caudalis neurons leads to generation of pain signals. This may be the result of supraspinal facilitation. In addition, these patients may have a decrease in pain modulation (involving serotonin and nonephinephrine as neurotransmitters). There may be a generation of spontaneous central pain. These factors may operate individually or in combination, leading to CDH.12  Recent research has focused on the role of nitrous oxide in CDH.

Cluster Headache Pathophysiology:
In order to explain the pathogenesis of cluster headache, one must take into account three crucial features of this disorder. These include the trigeminal distribution of the pain, ipsilateral autonomic features, and the tendency for the attacks to cluster with striking circadian and circannual consistency.12  As in migraine, the ophthalmic branch of the trigeminal nerve relays the pain signal, leading to release of SP and CGRP. Vasodilation of dural blood vessels and neurogenic inflammation occurs. There exists a connection between the trigeminal nucleus and the superior salivatory nucleus, where first-order parasympathetic neurons originate. These fibers travel with the seventh cranial nerve and synapse in the pterygopalatine ganglion. The postganglionic efferents innvervate the secretory glands of the nasal mucosa and the lacrimal glands. This explains the ipsilateral autonomic symptoms with cluster.13,14  The carotid sympathetic plexus is also involved, leading to Horner’s syndrome during attacks. The rhythmicity of the attacks appears to be related to hypothalmic activation or dysfunction.15

Mechanism of Action of BTA:
For the various headache disorders, different theories have arisen explaining the proposed efficacy of botulinum toxin, according to the known pathophysiology of pain transmission.

One of the theories proposed involves a reduction of muscular hyperactivity.16  Since patients with cervical dystonia have less pain after BTA treatment, a mechanism of decreased contractility of the musculature seems valid as an explanation for pain relief. The prevention of acetylcholine release from the presynaptic terminal would lead to a relaxation in the muscle. However, the actual effects of BTA are not so simply explained. This theory does not explain why pain relief often occurs earlier than the muscular relaxation. Also, this would not explain why pain relief occurs in areas that display no muscle tension.17   Lastly, some believe that the muscle overactivity is more likely a reflex central process, with pain leading to muscle activity, rather than vice-versa.18  Thus, this theory is not entirely satisfactory as an explanation for the efficacy of BTA.16

Another theory involves the "normalization of excessive muscle spindle activity".16  Studies have revealed active loci of spontaneous EMG activity corresponding to an excessive release of acetylcholine at the neuromuscular junction leading to abnormally excessive end plate activity. This results in extrafusal muscle contraction in the vicinity of the extrafusal motor endplates.19  Transmission has been shown to be inhibited at gamma motor neurons in the muscle spindle, which may lead to a decrease in muscle overactivity.18  Evidence of inhibition of infrafusal muscle fibers along with a change in muscle spindle activity has been demonstrated.20  The resulting decrease in spindle activity results in less sensory information traveling from the 1a afferents from the muscle to the CNS. This decrease in sensory input will effect a domain outside of the directly involved pathway due to multiple reflex and centrally projecting pathways that are connected. Thus, areas outside of the injection area may be affected in a similar fashion.

Another interesting mechanism of action involves its possible entry into the CNS leading to pain modulation. As discussed earlier, the pain of migraine is secondary to neurogenic inflammation of dural and meningeal vessels.10  BTA was found to inhibit release of substance P from trigeminal nerve endings, along with activating expression of substance P in the raphe nuclei.21,22  In addition, inhibition of SNAP-25 by BTA could block neurotransmitter exocystosis, therefore decreasing pain. This is due to the co-localization of vasoactive intestinal peptide and neuropeptide Y with acetylcholine in parasympathetic neurons.23  BTA was found in the dorsal root ganglia forty-eight hours after injection. This corresponds to pain relief.24  In addition, there may be a decompression of afferent nociceptive neurons, leading to a decrease in excitatory metabolites secondary to the muscle relaxation, which may contribute to the analgesic effects.

BTA may reduce parasympathetic outflow. In cluster patients, as previously mentioned, the involvement of the trigemino-cervical complex, the SSN, and the pterygopalatine ganglion are integral in the autonomic symptoms and vasomotor control.13,14  BTA may block the parasympathetic outflow loop leading to analgesia.

Review of the Literature:
Numerous studies have been published on the use of botulinum toxin for migraine and tension headache. Cluster headache and BTA / BTB has been the focus of several small studies. The following is a review of the literature.

Optimal dosing of BTA has been evaluated in several studies. The dosing for migraine has ranged widely, from 10 units to 150 units per patient. One multicenter double-blind study of 123 patients demonstrated that 25 units (low dose BTA) was adequate to significantly decrease migraines. In addition, when 25 units was compared to 75 units, there were significantly fewer adverse events. In this study, the patients reported decreased severity of migraines, less number of days utilizing migraine medications, and reduced migraine-associated vomiting.25  A further retrospective study of BTA with 77 migraineurs resulted in 46% with complete headache improvement, 30% with partial improvement, and the remainder were non-responders. The mean dose in this study was 35.5 units of BTA per patient.26  Another study, open label over 3 years, utilized 80-150 units of BTA per patient. Sixty-seven percent of migraine sufferers responded favorably in this study.27

As to where to inject, the studies have involved different protocols. Glabellar injections may lead to more complete relief.26  Several double-blind, placebo-controlled studies injected combinations of frontalis, temporalis, and glabellar sites.25  Other studies utilized suboccipital sites, although for posterior muscles larger doses are necessary to achieve an effect. Although there is some indication that injecting posteriorally adds to efficacy of botulinum toxin for migraine, further studies are necessary to delineate optimal sites for injection. If injections are done primarily in the frontal and temporal regions, it appears that low dose botulinum may be as, or more, effective than higher doses.25

Chronic Daily Headache:
Studies on botulinum toxin for tension headache have not been as favorable as those for migraine.28,29,30,31  In one study where higher doses of BTA were utilized (80-150 units), 58% of those with CDH did achieve positive outcomes.27  These same investigators then conducted a double-blind, placebo-controlled, randomized study involving 40 patients with chronic tension-type headache.27  The number of headache-free days was significantly increased in the BTA group at 3 months post treatment. Other studies have not been quite as positive for tension-type headache, however. One study that was double-blind, placebo-controlled and randomized revealed no significant differences through 12 weeks for chronic tension-type headache.32

A recent study (by Robbins) of BTA for refractory chronic daily headache evaluated 87 patients in an open-label fashion.33 The results of this study are as follows:
This was an open label, non-randomized non-blinded study. Eighty-seven participants, aged 23-67, were enrolled in the study, and 79 patients completed the study. Each patient had the diagnosis of moderate or severe chronic daily headache, refractory to the usual preventive medications. The patients recorded headache severity utilizing a visual analog scale for one month prior to treatment, and three months following treatment.

Each patient received low dose BTA injections, 12 injections of 2 units BTA each. The symmetrical injections, 6 on each side, were done frontally and temporally. See figure I for sites of injection.

The results were: Thirty-six of 79 patients (46%) did not respond to the injections (response = at least a 2 point decrease on a 10 point vas). Forty-three of 79 (54%) were considered positive responders. Among the positive responders, 44% had a mild response (2 to 3 point decrease on the vas). Forty-seven percent had a moderate response (4 to 5 point decrease on the vas), while 9% had an excellent response (more than 5 point decrease on the vas).

Among the mild responders (19 of 43 = 44% of responders), in the second month, 16% declined to no response, 63% continued to have a mild response, while 10.5% of these mild responders had a moderate response and 10.5% reported an excellent response. During the 3rd month with these mild responders, 68% now had no response, 10.5% had a mild response, 10.5% had a moderate response, and 10.5% reported an excellent response.

For the moderate responders (20 of 43 = 47% of total responders), during the 2nd month 15% now had no response, 15% had only a mild response, while 70% continued with a moderate response. During the 3rd month, 50% now declined to no response, 10% continued with a mild response, while 40% continued with a moderate response.

For the excellent responders, which were only 4 of 43 responders in total, during the 2nd month all 4 remained excellent. During the 3rd month 2 of the 4 declined to a mild response, 1 had a moderate response and 1 of the 4 patients continued with an excellent response.

In summary, this study revealed a modest effect of BTA on CDH, but only 24 of 79 patients achieved a moderate or excellent response.

Adverse events were generally mild, with 6 patients reporting mild ptosis, 2 with bilateral edema of the eyelids, and 2 reported a dramatic increase in headache.33

Cluster Headache:
A limited number of studies have been performed regarding botulinum toxin in patients with cluster headache. One study with 2 patients yielded excellent results, where both patients had no further clusters after 1 week of treatment. The effects lasted 10-12 weeks.34  The results of a recent study35 on botulinum toxin for cluster headache (by Robbins) are as follows:
Ten chronic cluster headache patients received botulinum toxin for refractory clusters.

The injections were done with low dose botulinum toxin; either 24 units of BTA per patient, or the equivalent BTB (1200 units). These were patients who had been refractory to the usual preventive and abortive medications for cluster headache. Side effects were minimal in this study. For the 7 chronic cluster patients, the injections were moderately effective in 3/7, and extremely effective for one. The botulinum toxin was not effective for 3 of the chronic cluster sufferers. Of the 3 episodic cluster sufferers, 1 obtained complete relief and 1 had moderate relief. The 3rd achieved complete relief with the first set of injections, but only moderate improvement after a further set of injections one year later. One other cluster study found that 2 of 4 patients improved, with doses ranging widely from 24 to 150 U.36  Therefore, while the results in cluster patients have not been very encouraging, certain patients do benefit from the injections.

Botulinum Type A vs. Botulinum Type B:
The authors have had extensive experience with both types of botulinum toxins for headache, having used each in approximately 130 patients. As they are different strains with somewhat differing pharmacology, exact equivalencies of units are not possible. However, for practical purposes, we have equated 100 units of botulinum toxin A to approximately 5000 units of type B. Advantages of BTA may include decreased pain upon administration, while type B is more stable, leading to somewhat greater ease of use. However, we have not observed differences in results between BTA and BTB. Both strains of botulinum appear to be quite safe when used in the usual (low) doses for headache.

Botulinum toxin (BTA, BTB) is effective for certain selected patients with migraine headache. The efficacy appears to be less well demonstrated for those with CDH. In addition, certain patients with refractory cluster headache respond to low dose botulinum toxin. The safety of low dose botulinum toxin has been well established. There are a number of unanswered questions regarding the use of botulinum toxin in headache patients. The optimal sites of injection have yet to be delineated. While studies have indicated that low dose botulinum may be as, or more, effective than higher doses, this may not be true for posterior injections. It does appear that, anecdotally, BTA (Botox) and BTB (Myobloc) may be somewhat equivalent in the treatment of headache. Further studies are needed in order to determine the role of botulinum toxin in headache patients. However, it does appear that botulinum toxin will play some role in the management of headache.

  1. Robbins L, Maides J. Efficacy of preventive medications for chronic daily headache. American Journal of Pain Management, October 1999, vol. 9, 4:130-138.
  2. Robbins L. Management of Headache and Headache Medications. 2nd ed. New York: Springer-Verlag:2000:126-132.
  3. von Ermengem E. Ueber einen neuen anaeroben Bacillus und seine Beziehungen zum Botulismus. Zeitschrift fur Hygeine und Infektionskrankheiten 1897; 26: 1-56; (translation reprinted in Reve Infect Dis 1979; 1:710-719).
  4. Scott AB. Botulinum toxin injection into extraocular muscules as an alternative to strabismus surgery. Ophthalmology 1980; 87:1044-1049.
  5. Brin MF. Botulinum toxin: chemistry, pharmacology, toxicity, and immunology. Muscle Nerve 1997; Supp. 6: S146-168.
  6. Goadsby PJ, Zagami AS, Lambert GA. Neural processing of craniovascular pain: a synthesis of the central structures involved in migraine. Headache 1991; 31: 365-371.
  7. May A, Goadsby PJ. The trigeminovascular system in humans: pathophysiologic implications for primary headache syndromes of the neural influences on the cerebral circulation. J Cereb Blood Flow Metab. 1999:19; 115-127.
  8. Weiller C, May A, Limmroth V, Juptner M, Kaube H, Schayck RV, Coenen HH, Diener HC. Brainstem activation in spontaneous migraine attacks. Nat Med. 1995; 1:658-660.
  9. Goadsby PJ, Edvinsson L, Ekman R. Release of vasoactive peptides in the extracerebral circulation of man and the cat during activation of the trigeminovascular system. Ann of Neurol. 1988: 23:193-196.
  10. Markowitz S, Saito K, Moskowitz MA. Neurogenically mediated leakage of plasma proteins occurs from blood vessels in dura mater but not in brain. J Neuroscience. 1987: 7: 4129-4136.
  11. Raskin, NH. Headache, 2nd edition, Churchill Livingston, New York: 1988 pp.215-224.
  12. Silberstein S, Lipton R, Dalessio D. Wolff’s Headache and other head pain. 2001: 260-265; 290-291.
  13. Lambert, GA, Bogduk N, Goadsby PJ, et al. Decreased carotid arterial resistance in cats in response to trigeminal stimulation. J Neurosurgery. 61:307-315.
  14. Goadsby, PJ, Edvinsson L. Human in vivo evidence for trigeminovascular activation in cluster headache. Brain. 117:427-434.
  15. May A, Bahra A, Buchel C, et al. Hypothalamic activation in cluster headache attacks. Lancet. 352:275-278.
  16. Gobel H, Heinze A, Katja Heinze-Kuhn, Austermann K. Botulinum toxin A in the treatment of headache syndromes and pericranial pain syndromes. Pain. 2001.91:195-199.
  17. Brin MF, Swope DM, O’Brain C, Aboasi S, Pogoda JM. Botox for migraine: double-blind, placebo controlled region-specific evaluation. Cephalalgia 2000: 20:421-422.
  18. Priori A, Beradrelli A, Mercuri B, Manfredi M. Physiological effects produced by botulinum toxin treatment of upper limb dystonia: changes in reciprocal inhibition between forearm muscles. Brain. 1995: 118:801-807.
  19. Simons DG. Fibrositis/Fibromyalgia: a form of myofascial trigger points? Amer J Med 1986:81 (Suppl 3A): 9-98.
  20. Rosales RL, Arimura K, Takenaga S, Osame M. Extrafusal and intrafusal muscle effects in experimental botulinum toxin-A injection. Muscle Nerve. 1996: 19 (4): 488-496.
  21. Ishikawa H, Mitsui Y, Yoshitomi, Mashimo K, Aoki S, Mukano K, Shimizu K. Presynaptic effects of botulinum toxin type A on the neuronally evoked response of albino and pigmented rabbit iris sphincter and dilator muscles. Jpn J Ophthalmol. 2000: 44(2):106-109.
  22. Van den Bergh P, De Beukelaer M, Deconinck N. Effect of muscle denervation on the expression of substance P in the ventral raphe-spinal pathway of the rat. Brain Res. 1996: 707 (2): 206-212.
  23. Hohne-Zell B, Galler A, Schepp W et al. Functional importance of synaptobrevin and SNAP-25 during exocytosis of histamine by rat gastric enterochromaffin-like cells. Endocrinology. 1997: 138:5518-5526.
  24. Weigand H, Wellhoner HH. The action of botulinum A neurotoxin on the inhibition by antidromic stimulation of the lumbar monosynaptic reflex. Naunyn Schmiedebergs Arch Pharmacol 1977: 298(3):235-238.
  25. Silberstein S, Mathew N, Saper J, Jenkins S. Botulinum toxin type A as a migraine preventive treatment. Headache 2000; 40:445-450.
  26. Binder WJ, Brin MF, Blitzer A, Schoenrock LD, Pogoda JM. Botulinum toxin type A (Botox) for treatment of migraine headaches: an open-label study. Otolaryngol Head Neck Surg 2000; 123:669-676.
  27. Smuts JA, Baker MK, Smuts HM, Rheta Stassen JM, Rossouw E, Barnard PWA. Prophylactic treatment of chronic tension-type headache using botulinum toxin type A. Eur J Neurology 1999; 6(suppl 4): S99-S-102.
  28. Porto M. Botulinum toxin type A injections for myofascial pain syndrome and tension-type headache. Eur J Neurology 1999; 6(suppl 4): S103-S109.
  29. Relja M. Botulinum toxin type A in the treatment of tension-type headache. Presented at the 9th World Congress on Pain, Vienna, Austria, August 22-27, 1999, and at the International Conference 1999: Basic and Therapeutic Aspects of Botulinum and Tetanus Toxins, Orlando, FL, November 16-18, 1999.
  30. Wheeler AH. Botulinum toxin A, Adjunctive therapy for refractory headaches associated with pericranial muscle tension. Headache 1998; 38:468-471.
  31. Schulte-Mattler WJ, Wieser T, Zierz S. Treatment of tension-type headache with botulinum toxin: a pilot study. Eur J Med Res 1999; 4:183-186. These data also presented at the International Conference 1999: Basic and Therapeutic Aspects of Botulinum and Tetanus Toxins, Orlando, FL, November 16-18, 1999.
  32. Rollnik JD, Tannenberger O, Schubert M, Schneider U, Dengler R. Treatment of tension-type headache with botulinum toxin type A: A double-blind, placebo-controlled study. Headache 2000; 40:300-305
  33. Robbins L. Botulinum Toxin A for Refractory Chronic Daily Headache. Abstract presented at the 10th Congress of the International Headache Society, June 29-July 2, 2001.
  34. Freund BJ, Schwartz M. The use of Botulinum toxin-A in the treatment of refractory cluster headache: case reports. Presented at Headache World 2000, London,England, Sept. 3-7, 2000, and published as an abstract in: Cephalalgia 2000; 4:329.
  35. Robbins L. Botulinum toxin for cluster headache, 10 patients. Presented at the 10th Congress of the International Headache Society, June 29-July 2, 2001.
  36. Smuts JA, Barnard PWA. Botulinum toxin type A in the treatment of headache syndromes: a clinical report of 79 patients. Cephalalgia 2000: 20:332.