Current Pain and Headache Reports

, 18:409

Supraorbital and Supratrochlear Stimulation for Trigeminal Autonomic Cephalalgias


    • Department of Physical Medicine and RehabilitationHarvard Medical School
    • Pain and Wellness Center
  • Edrick Lopez
    • Department of Physical Medicine and RehabilitationHarvard Medical School
    • Spaulding Rehabilitation Hospital
  • Nicholas K. Muraoka
    • Department of Physical Medicine and RehabilitationHarvard Medical School
    • Spaulding Rehabilitation Hospital
Trigeminal Autonomic Cephalalgias (M Matharu, Section Editor)

DOI: 10.1007/s11916-014-0409-4

Cite this article as:
Vaisman, J., Lopez, E. & Muraoka, N.K. Curr Pain Headache Rep (2014) 18: 409. doi:10.1007/s11916-014-0409-4
Part of the following topical collections:
  1. Topical Collection on Trigeminal Autonomic Cephalalgias


Trigeminal autonomic cephalalgias (TAC) is a rare primary headache disorder with challenging and limited treatment options for those unfortunate patients with severe and refractory pain. This article will review the conventional pharmacologic treatments as well as the new neuromodulation techniques designed to offer alternative and less invasive treatments. These techniques have evolved from the treatment of migraine headache, a much more common headache syndrome, and expanded towards application in patients with TAC. Specifically, the article will discuss the targeting of the supratrochlear and supraorbital nerves, both terminal branches of the trigeminal nerve.


Cluster headacheChronic cluster headacheMigraine headacheNeuromodulationOccipital nervePainPeripheral nerveSphenopalatine ganglionSupraorbital nerveSupratrochlear nerveStimulationTranscutaneous nerve stimulationTrigeminal autonomic cephalalgia


Trigeminal autonomic cephalalgias (TAC) is a group of primary headache conditions of relatively short duration associated with cranial autonomic symptoms. The term TAC encompasses several disabling neurovascular syndromes, the most common being cluster headaches (CH), short-lasting unilateral neuralgiform headache with conjunctival injection and tearing (SUNCT), and short-lasting unilateral neuralgiform headache with cranial autonomic features syndrome (SUNA) [1].

Although TAC is a relatively rare disorder, the human toll for the headaches sufferers remains very high, many patients displaying a high suicidal ideation rate along with decreased productivity because of job loss or disabilities [2]. As a result, there is an increased need for early detection, assuring an effective, safe, and evidence-based treatment.

Over the past 2 decades, the neuromodulation field has witnessed a robust growth and has expanded from its original application for the treatment of the failed back syndrome to other various intractable painful neuropathic conditions such as chronic regional pain syndrome, postherpetic neuralgia, and headaches.

For the treatment of severe, intractable headache conditions, a number of peripheral nerves including the occipital, supraorbital, and supratrochlear nerves became the target of stimulation at their location in the epifascial and subcutaneous tissue. The modality has gained popularity since 1999, when Weiner and Reed described their technique for occipital stimulation in a number of patients with occipital neuralgia [3]. Thus, the greater and the lesser occipital nerves became a common target for the treatment of various drug-resistant, refractory headaches such as occipital neuralgia, chronic migraine, hemicrania continua, chronic CH, and hypnic headache [4, 5].

As an alternative to the stimulation of the occipital nerves, the 2 terminal branches of the trigeminal nerve, the supraorbital and supratrochlear nerves, became a target for the treatment of intractable TAC, giving the presumed involvement of the trigemino-vascular axis and the frontal localization of the pain.

The present article will discuss the application of neuromodulation techniques for various headache conditions, focusing on the supraorbital and supratrochlear stimulation, various treatment options, and the technical challenges related to the actual implantation process.

Classification and Treatment of the Trigeminal Autonomic Cephalalgias

The trigeminal autonomic cephalalgias are a group of primary headache conditions accompanied by cranial autonomic symptoms such as lacrimation, conjunctival injection, nasal congestion, and rhinorrhea. Interestingly, these are the same symptoms elicited by stimulating the ophthalmic/V1 branch of the trigeminal nerve in healthy patients [6].

Cluster Headaches

Cluster headaches are the most common of the trigeminal autonomic cephalalgia subtypes; they classically present periorbital or in the eye and are strictly unilateral during each episode, but attacks can affect the contralateral side about 20 % of the time [7]. Men are affected more frequently than females at a ratio of 5:1. Attacks typically last 30–60 min and occur at a frequency from once every other day to 8 times a day. During active periods, the attacks may be provoked by alcohol, histamine, or nitroglycerin [1]. The quality of the pain is described as severe, throbbing, and lancinating causing the individual to be restless, in contrast to the person suffering a migraine, who prefers to be still. Prodrome symptoms are reported almost universally and vary from twinges to pressure to a pulsating feeling in the area that becomes painful during the attack [8]. Chronic CH (CCH) occurs over the interval of more than 1 year without remission or with remissions lasting less than 1 month [1]. The chronic form accounts for up to 10 % of patients with CH.

CH are best treated acutely with supplemental oxygen [9] and subcutaneous [10] or intranasal [11] Sumatriptan. Oral triptans have not been shown to be effective as acute treatment of cluster headache and are associated with medication overuse headaches [12]. Preventative treatments include prednisolone and methysergide for episodic cluster headaches; and verapamil, topiramate, gabapentin, lithium, melatonin, and greater occipital nerve blocks for CCH [13].

Paroxysmal Hemicrania

Paroxysmal hemicrania typically affects unilaterally the temporal, peri-auricular, peri-orbital, and maxillary areas. Pain referral to the neck, shoulder, and arm can be common. Attacks last an average of 13–29 min [14]. Compared with cluster headaches, paroxysmal hemicrania occur more frequently, from 8 to 30 attacks per day. Paroxysmal hemicrania does not appear to affect either gender more frequently [15].

One of the diagnostic criteria of paroxysmal hemicrania is that it is responsive to indomethacin, so this NSAID is the first-line treatment of paroxysmal hemicrania symptoms [16]. In individuals with contraindications to indomethacin intake, there are case reports of topiramate being used as successful prophylactic treatment for paroxysmal hemicrania [17].


In SUNCT, both conjunctival injection and tearing are present. SUNCT can be viewed as a subtype of SUNA, where SUNA can have any of the cranial autonomic symptoms in addition to conjunctival injection and tearing, such as nasal stuffiness, rhinorrhea, ptosis, and sweating. SUNCT and SUNA are both primary headache syndromes characterized by a high frequency of symptoms of short duration. Average duration of SUNCT attacks are 1 min, with maximum length of 2 min; in comparison, SUNA attacks can last as long as 10 min. SUNCT attacks can come as frequently as 200 per day [1]. Prevention of these TAC varieties is with oral Lamotrigine. Intravenous lidocaine infusions can be used as a transition to oral Lamotrigine [18]. Subcutaneous infusion of lidocaine proved to be effective as abortive treatment on 78 % occasions in patients with SUNCT and SUNA [19].

Despite optimal medical management of headache symptoms, up to 20 % of patients with chronic cluster headaches are refractory to these treatments [20]. Headaches are considered intractable if there are persistent symptoms or intolerable side-effects after an adequate trial (appropriate dose and length of treatment) of medications [21].

A variety of invasive procedures have emerged to address these intractable headaches. There are case reports of surgical interventions targeting various structures including: the trigeminal nerve and its branches, the Gasserian ganglion (sensory ganglion of the trigeminal nerve located in Meckel’s cave) and the sphenopalatine ganglion. These procedures carry their associated risk of morbidity and experience is that there is recurrence of symptoms in as little as 6 months, necessitating repeat procedures [22].

Partial sectioning of the main trigeminal sensory root and/or the nervus intermedius has been a treatment for refractory cluster headaches, however, in a case series of 13 patients, all but 1 had a return of symptoms by 2 years postprocedure [23]. Microvascular decompression of the trigeminal nerve has been reported in medically intractable SUNCT and SUNA patients. Although 12 of the 19 patients (63 %) were pain-free at median follow-up of 14 months, the remaining patients did not experience any benefit and 2 patients had persistent complications such as ataxia and hearing impairment [18].

Neurolysis of the Gasserian ganglion, which is the sensory ganglion of the trigeminal nerve, is done with the idea that symptoms of TACs affect the sensory distribution of the trigeminal nerve. There has been chemical neurolysis with glycerol, however, 63 % of the treated patients had persistent headaches at 1-year follow-up [22]. Radiofrequency ablation of the Gasserian ganglion has been attempted as well, with potentially debilitating side effects including abducens nerve palsy [24], facial anesthesia dolorosa and corneal anesthesia, which can lead to blindness [25].

Percutaneous radiofrequency ablation of the sphenopalatine ganglion, which is located in the pterygopalatine fossa, has shown some benefits in episodic CH, but the procedure seems to be less effective in patients with chronic CH [26]. Gamma knife surgery of the sphenopalatine ganglion with or without targeting the ipsilateral trigeminal nerve has been done for chronic cluster headaches in patients with symptoms refractory to medications and even other surgical procedures. Satisfactory pain relief was only seen in 59 % of the patients at 34-month follow-up with half of the group developing facial sensory impairments after the procedure [27]. Microvascular decompression of the sphenopalatine ganglion has been attempted for refractory cluster headache, however, a small case series failed to show a significant difference in presurgical vs postsurgical pain scores, attack frequency, or quality of life [28].

Neuromodulation for pain is a growing field that is receiving much interest and has shown promise. For this reason, it is felt that destructive surgical options should not be included in modern treatment options for the trigeminal autonomic cephalalgias [13].

Neuromodulation – Evolving Therapy (other than for HA)

The first concept for the application of neurostimulation in modern medicine was 1965, when Melzak and Wall published their gate control theory of pain in Science: selective stimulation of large diameter myelinated fibers causes inhibition of the small fiber pathway that transmit pain and temperature sensation [29]. Norman Shealy is credited with the first clinical application of neuromodulation when he placed an electric lead attached to an external cardiac generator over the dorsal column of the spinal cord in the intrathecal space to treat intractable pain in a terminally ill cancer patient [30].

Hosobuchi and colleagues developed an invasive approach of deep brain stimulation in 1973 by implanting leads into the thalamus of individuals with facial anesthesia dolorosa [31]. Tsubokawa et al used motor cortex stimulation to alleviate central pain originating in the thalamus [32].

Peripheral nerve stimulators are used to manage pain by targeting a specific peripheral nerve. Peripheral nerve field stimulation (PNfS) is another technique to induce paresthesia over an area of pain that overlaps several cutaneous nerve distributions and not confined to a single peripheral nerve distribution [33]. The mechanism of PNfS is also suggested to work via the gate control theory of pain. The subcutaneous layer of the skin has a high density of A beta terminal sensory fibers and electrodes are placed under the skin in the region of pain [34]. A recent randomized control study looking at PNfS for chronic axial low back pain refractory to medications, interventions, and surgery found 24 of 32 patients (80 %) who had a trial system placed responded to the trial and underwent implantation of permanent device. At 1-year follow-up, 70 % had good to excellent pain relief with an additional 17 % reporting fair relief in symptoms [35].

Specifically related to the trigeminal autonomic cephalalgias, there have been attempts at targeting various structures for neurostimulation. The most invasive is deep brain hypothalamic stimulation. The rationale for this is based on PET scan results of nitrous-oxide induced symptoms in individuals with chronic cluster headaches suggesting increased blood flow in the diencephalon, specifically the hypothalamus [36]. Experience with this in 58 patients has shown success in 60 % of the treated group, however, because of the invasive nature of the procedure, there is a small but real risk of brain hemorrhage [37]. Another target has been the occipital nerve, which has been effective in about two-thirds of the treated group [38].

Stimulation as Treatment for Headaches

The supratrochlear and supraorbital nerves are branches of the frontal nerve, which is the largest branch of the ophthalmic division of the trigeminal nerve. The supratrochlear nerve ascends medially under the corrugator and frontal belly of the occipitofrontalis muscles to supply the innervation of the skin above the orbital area. The supraorbital nerve passes via the supraorbital foramen and then ascends just lateral to the supratrochlear nerve to finally divide into a smaller medial and a larger lateral branch. Both branches are situated in the frontal belly of the occipitofrontalis muscle and then pierce the muscle and the epicranial aponeurosis. Therefore, they are amenable to stimulation as they emerge from the orbit and further divide into their terminal branches, which supply the skin of the forehead. Hence, it makes sense to attempt stimulating both nerves in the supraorbital area for the treatment of TAC.

Building on this theory Vaisman et al performed an observational, open-label, retrospective case series of 5 patients with intractable trigeminal autonomic cephalalgia who were implanted with a supraorbital/supratrochlear neuromodulation system [39••]. Trial stimulation led to implantation of all 5 patients. All patients reported improvement in their activities of daily living. The follow-up of the patients ranged between 18 months and 36 months, with a mean of 25.2 months. Following the implant, the Visual Analog Scale score was reduced to a mean of 1.6 from an initial mean score of 8.9. Three patients were completely weaned off opioid medications, while 2 patients continued to take opioid at a lower dosage. All patients experienced a decrease of the adjuvant neuropathic drugs. The authors found that supraorbital/supratrochlear peripheral nerve stimulation appears to be a fairly safe and potentially effective therapeutic modality for patients with intractable TAC. It was found that similar to occipital nerve stimulation, the supraorbital nerve blocks do not accurately predict who will respond to a trial of supraorbital/supratrochlear nerve stimulation [40].

A few cases of supraorbital nerve stimulation for other types of headaches have been reported. In 1 retrospective case series, 10 patients had permanent implantation of a supraorbital nerve stimulator for the treatment of chronic, intractable frontal, and frontotemporal neuritis unresponsive to medication [41]. Permanent implantation resulted in significant reductions in pain and use of opioids. Adverse events were minor, limited to 3 patients who had lead migrations and 1 with a minor scalp infection. Reed et al reported on 7 patients who were successfully treated for intractable chronic migraine with dual stimulation of both the occipital and supraorbital nerves [42]. These patients were followed for a period ranging from 1 to 35 months. The authors concluded that this combination led to a substantial better outcome than occipital nerve stimulation alone. Slavin et al reported on 5 patients undergoing supraorbital nerve stimulation for craniofacial pain [40]. In addition, a single case report of supraorbital nerve stimulation for supraorbital neuralgia was also published [43]. Prior to the aforementioned Vaisman et al study [39••] only 1 report using strictly supraorbital stimulation for the treatment of chronic CH had been published [44]. Goadsby reported 50 % success in treating 2 CH patients with occipital nerve stimulation [45] In a review of neurostimulation for primary headache disorders in 2009, Schwedt recommended that “further studies are required to determine the safety and efficacy of supraorbital nerve stimulation for treating headache disorders” [46].

Not surprisingly, other anatomic targets, such as cervical posterior epidural space, cervical medial branch nerves, and the sphenopalatine ganglion were found to be valuable targets for neuromodulation. For example, high epidural spinal cord stimulation was reported successful in 7 patients suffering from intractable CH [47]. One case of intractable episodic CH treated successfully with percutaneous cervical zygapophyseal radiofrequency ablation has been reported [48].

The sphenopalatine ganglion can be amenable to electrical stimulation using the classical infrazygomatic approach during an acute episode of CH and the relief of pain occurs after several minutes of stimulation [49, 50]. Recently, Schoenen et al showed that stimulation of the SPG could be an effective novel therapy for chronic cluster headache sufferers, with dual beneficial effects, acute pain relief and observed attack prevention, and has an acceptable safety profile compared with similar surgical procedures [51•]. Theirs was a multicenter, multiple CH attack study in patients suffering from refractory CCH. Each CH attack was randomly treated with full, sub-perception, or sham stimulation. Pain relief at 15 min following SPG stimulation and device- or procedure-related serious adverse events were evaluated. Nineteen of 28 (68 %) patients experienced a clinically significant improvement. Pain relief was achieved in 67.1 % of full stimulation-treated attacks compared with 7.4 % of sham-treated and 7.3 % of sub-perception-treated attacks. Five serious adverse events occurred (3 lead revisions and 2 explantations) and most patients (81 %) experienced transient, mild/moderate loss of sensation within distinct maxillary nerve regions but most resolved within 3 months. Mueller et al followed 10 CCH patients implanted with occipital nerve stimulators over a 12 month period. The authors found that the frequency, duration, and severity of the cluster attacks were reduced in 90 % of patients [52].

Burns et al published a large study [53] that included 14 patients with intractable CCH treated with occipital nerve stimulation. A total of 6 patients reported meaningful improvement of the pain. Improvement evolved over weeks or months, although attacks returned in a few days when the device malfunctioned due to battery depletion. Magis et al [54] followed a total of 8 patients with intractable CCH after occipital nerve implantation. Again the results appeared to be favorable. Interestingly, the authors found a delay of at least 2 months between implantation and significant clinical improvement, suggesting that the procedure acts via a slow neuromodulation process at the level of the upper brain stem or diencephalic centers. The general impression from the few studies using occipital stimulation for patients with intractable CH is that relief occurred slowly over time while autonomic phenomena appeared to persist [55].

Other studies found supraorbital nerve stimulation effective in relieving neuropathic facial pain including ophthalmic postherpetic neuralgia [5658]. The mechanism of pain relief for patients undergoing peripheral nerve stimulation remains elusive, despite some prevailing theories, including increased blood flow, increased serotonin and dopamine at the spinal cord level, inhibition of nociceptive responses of wide-dynamic range neurons, and a possible alteration of pain inhibitory circuits [5961].

There is a building body of solid literature indicating that peripheral nerve stimulation is a potentially promising therapy for patients suffering from chronic migraine (CM). Serra et al [62] recently published a prospective, randomized cross-over trial looking at occipital nerve stimulation for the treatment of chronic migraine and medication overuse headache. They found that occipital nerve stimulation appeared to be safe and effective in reducing headache intensity and frequency, improving quality of life, and reducing medication use in patients suffering from chronic migraine or medication overuse headache. Adding to the evidence, Silberstein et al [63] published a randomized, controlled multicenter study, in which patients diagnosed with CM were implanted with a neurostimulation device near the occipital nerves and randomized 2:1 to active or sham stimulation. The primary endpoint was a difference in the percentage of responders (defined as patients that achieved a 50 % reduction in mean daily visual analog scale scores) in each group at 12 weeks. They did not find significant difference in the percentage of responders in the active group compared with the control group. Nonetheless, although this study failed to meet its primary endpoint, this is the first large-scale study of peripheral nerve stimulation of the occipital nerves in chronic migraine patients that showed significant reductions in pain, headache days, and migraine-related disability.

In addition, Schoenen et al [64] published in 2013 a double-blinded, randomized, sham-controlled trial looking at supraorbital transcutaneous nerve stimulation (TNS) as a treatment for migraine. They concluded that treatment with a supraorbital TNS is effective and safe as a preventive therapy for migraine.

Intractable chronic CH has been shown to be responsive to deep brain stimulation [6567], but it is more invasive and can also lead to much more severe complications such as intracerebral hemorrhage and/or death [68]. For those patients with intractable TAC who do not respond to suboccipital peripheral nerve stimulation, or who might not be candidates for DBS, another option is stimulation of the trigeminal nerve branches. Clearly, direct stimulation of a peripheral nerve is less invasive and a safer procedure than deep brain stimulation [69].

Considering that CH arises unilaterally in the first and second trigeminal nerve distribution [70], it makes sense to stimulate peripheral nerves which, in fact, are branches of the trigeminal nerve. Also, because central sensitization mechanisms appear to play a significant role in the pathophysiology of CH [71], it makes more sense to seek a neuromodulation technique for the treatment of this condition.

Device Implantation and Technical Considerations

As the intracranial trigeminal location is not easily amenable to stimulation, the peripheral branches of the trigeminal nerve become an appealing target for neuromodulation. Similar to the model of spinal cord stimulation, a trial of stimulation is initially performed in order to identify the best candidates with the general goal that the stimulation will be felt over the skin above the corresponding orbit.

The trial 8-contact lead is placed in the operating room under sterile conditions. Light sedation is provided so the patient can verify adequate coverage of the painful region. A curved Tuohy needle is navigated just above the supraorbital margin of the frontal bone and directed medially towards the glabella. As a landmark, the supraorbital notch is usually easily palpated along the supraorbital prominence, on the same line with the pupil. The lead is placed through the needle following which the needle is removed and the lead is anchored to the skin. Fluoroscopy is used to confirm the final lead placement (Fig. 1). Once the intraoperative test is deemed satisfactory, the patient is discharged home with appropriate instructions on how to use the device. Typically the trial is extended for a period of 10–14 days. During the trial the patients carry on with their normal daily activities. The trial lead is exiting the temporal area of the skin where it is covered by a sterile dressing and is connected to a handheld programmer, usually similar in size to a cellular phone. The patients receive a few programs and are welcome to use the trial stimulation as much as they deem necessary during the trial in order to hopefully achieve satisfactory pain relief.
Fig. 1

An octrode lead was implanted above the superior orbital margin (arrow)

A positive response to the stimulator trial is a 50 % reduction in pain from the baseline visual analog score and a permanent implant would be considered in these individuals. Those patients who pass the trial stimulation are brought back to the operating room where the actual implant takes place. General anesthesia is induced and typically the procedure takes up to 2.5 h. A small incision is made above the zygomatic process of the frontal bone and the lead is placed in a similar manner and anchored to the underlying superficial fascia with a nonabsorbable suture. The next step is the tunneling of the lead and its extension to the internal pulse generator, which is placed in a superficial pocket in the subclavicular area just above the pectoralis muscle.

In the retrospective case series of 5 patients the main complication was skin erosion and secondary skin infection, which occurred in 2 patients [39••]. Interestingly, the skin infection occurred 1 year after the implant and required a lead revision; however, the stimulation was successfully regained.

Extrapolating the supraorbital and supratrochlear implantation to the occipital nerve implantation, where many more implants have been performed, it is conceivable that other complications could arise if indeed this procedure gains more popularity. These complications may include: lead migration, device malfunction with disconnection and lead fracture, pain at the site of the lead or internal pulse generator, infections and allergic reactions to surgical material. (63).


Neuromodulation appears to become a frontline therapy for those rare patients with TAC and intractable pain. Supraorbital and supratrochlear stimulation should be considered alone or in combination with other extra cranial targets such as the occipital nerve or the sphenopalatine ganglion.

Compliance with Ethics Guidelines

Conflict of Interest

Julien Vaisman declares that he has no conflict of interest. Edrick Lopez declares that he has no conflict of interest. Nicholas K. Muraoka declares that he has no conflict of interest.

Human and Animal Rights and Informed Consent

This article does not contain any studies with human or animal subjects performed by any of the authors.

Copyright information

© Springer Science+Business Media New York 2014