Introduction

The primary headache disorders, migraine and the trigeminal autonomic cephalalgias (TACs), are severely disabling neurologic conditions and constitute a therapeutic challenge. Each of these disorders now has strict and validated criteria for diagnosis, promulgated in the International Classification of Headache Disorders, third edition (ICHD-3), beta version (Headache Classification Committee of the International Headache Society (IHS), 2013).

The sphenopalatine ganglion (SPG), also called the pterygopalatine ganglion, is the switching station for the final common pathway for cluster headache (CH), may serve as the final synapse in the other TACs, and may also be pivotal in the efferent pathway to the dura and meninges in migraine. Inhibiting SPG outflow offers a specific and targeted approach for primary headache termination or prevention, and this chapter reviews the current state of this approach.

Diagnostic Criteria of the Primary Headache Disorders Targeted With SPG Inhibition

Episodic migraine (EM) is defined by the ICHD-3 as a headache lasting 4–72 h, with at least 2 h of moderate to severe intensity, throbbing quality, unilateral location, and aggravation by routine physical activity. There must be at least 1 h of nausea or photophonophobia. Chronic migraine (CM) is transformed from EM and requires headache at least 15 days/month for at least 4 h/day for at least 3 months and at least 8 days/month reaching a migraine level, responding to a migraine-specific drug such as an ergot or triptan, or recognized by the patient as migrainous.

The TACs now constitute CH, paroxysmal hemicrania, short-lasting unilateral headache attacks, and hemicrania continua. Episodic CH (ECH) is defined as unilateral attacks of 15 min to 3 h/day, occurring 0–8 times daily with at least one of various symptoms: agitation, sense of fullness in the ipsilateral ear, ipsilateral facial or forehead flushing or edema, ipsilateral ptosis, ipsilateral miosis, ipsilateral conjunctival injection, ipsilateral lacrimation, ipsilateral rhinorrhea, or nasal stuffiness. The flushing, edema, red eye, tearing, and nasal symptoms and signs are manifestations of parasympathetic activation. The miosis and ptosis are a partial Horner’s and are the clinical indicators of sympathetic paresis.

Cluster attacks in ECH occur for cycles or bouts of daily or near-daily headaches generally lasting 6–12 weeks, with long periods of remission between cycles. Chronic CH (CCH) has been described as not manifesting clinically significant remissions. In the ICHD-2 and ICHD-3 beta version, remissions of at least 1 month/year defined ECH. There is currently discussion as to whether to increase the time required for remission for ECH to at least 3 months per year ( ).

SPG Anatomy and Headache Pathophysiology

CH and migraine are likely initiated by central generators: in the case of CH, the ipsilateral posterior hypothalamus, and in the case of migraine the upper brainstem in the region of the dorsal raphe and periaqueductal gray. These generators may in turn trigger a central cascade of events that result in activation of the superior salivatory nucleus (SSN), a pontine visceral seventh cranial nerve nucleus. A synapse in the SSN leads to parasympathetic outflow from the SSN to the SPG via the greater superficial petrosal nerve.

Sympathetic postganglionic fibers, most of which synapsed in the superior cervical ganglion, join with these parasympathetics to form the vidian nerve. The parasympathetics of the vidian nerve synapse in the SPG, while the postganglionic sympathetics pass through the SPG without synapsing. Both the parasympathetic fibers and the sympathetic fibers synapse on target organs activated in the TACs, and both likely also proceed to the meninges, where the peripheral pain mechanisms of migraine are initiated. The target organs of SPG autonomic output include lacrimal nasal, pharyngeal, and palatine glands, the orbit, and, as noted, cerebral and meningeal blood vessels. The outflows for TACs and migraine probably both go through the SPG.

In addition, peripheral afferents destined to join ophthalmic (V1) and maxillary (V2) trigeminal divisions course through the SPG on their way. Thus the SPG acts as both an efferent and an afferent nexus, increasing its potential value as a therapeutic mark ( ).

One addition to the picture of this anatomy was provided by Akerman et al., who traced how oxygen works to terminate CH. Oxygen inhibits outflow from the SSN to the SPG, thus confirming the efferent pathway ( ).

Brief History of the Role of SPG Modulation in Headache Treatment

The concept of neuromodulation of the SPG for treatment of headaches derives from procedures that appear to have benefit in blocking the SPG. The concept is simple. If the SPG is the exit point for TACs and migraine, hindering outflow should stop or reduce headaches.

The history of these blocks goes back to 1908, when Sluder described SPG cocaine or alcohol blocks for what he called “nasal headaches” ( ). SPG blocks have also been described as effective in terminating headaches with caine anesthetics and steroids ( ).

The positive effects of nerve/ganglion blocks led to trials with destructive lesioning or surgery. These included cryosurgery ( ), ganglionectomy ( ), stereotactic radiosurgery ( ), and radiofrequency ablation ( ). There were also trials of injecting alcohol into the SPG, with damaging effects. The effects of these lesions can be helpful in difficult-to-treat headaches such as CCH, but may be transient.

Because of the temporary effects of blocks and the potentially harmful effects of lesioning, neuromodulation was proposed for inhibiting outflow from the SPG. As noted throughout this textbook, neuromodulation is attractive for its reversibility and potential for altering parameters of stimulation to optimize outcomes. Clinicians skilled in neuromodulation immediately saw the potential of this approach for the SPG in terms of being able to change stimulation site, length, bandwidth, intensity, frequency, and electrode activity. With a multielectrode stimulator or lead, the anode and cathode can be changed to influence both the location and the of stimulation.

Several proof-of-concept case reports were published on SPG stimulation for primary headaches. Ibarra described successful subzygomatic stimulation of the SPG for a man with CCH ( ). Ansarinia and colleagues, including the authors and one of the editors (AC, AR, SJT), described SPG stimulation provided in the operating room for terminating CCH attacks. Success was linked to anatomic location with physiologic paresthesia correlation ( ). Tepper et al. (including AC and AR) described similar results for CM ( ).

Historically, the next problem was how to place a more permanent stimulator in the relatively difficult to reach pterygopalatine fossa, in which the SPG is located. A multidisciplinary panel including the authors and one of the editors (AC, AR, SJT) was convened to consider options. The subzygomatic approach used in the above three reports was felt to be difficult, in that a tunneled lead would exit the cheek and be subject to movement and cosmetic problems. A lateral nasal approach would be an open channel for infection. Dr. Frank Papay suggested a transoral subperiosteal approach, used for reconstructive plastic surgery but requiring a redesign of the neurostimulator.

An engineering team including one of the authors (AC) devised a small six-electrode stimulator with no external leads that could be fixed to the skull above the teeth with the lead in the pterygopalatine fossa next to the SPG (see Fig. 63.1 ). By this fixation, lead migration was eliminated. Both the power and the instructions for stimulation come from a rechargeable and programmable remote controller (see Fig. 63.2 ).

(A) Anatomy of the anterior midface and (B) pterygopalatine fossa and sphenopalatine ganglion (SPG).

Autonomic Technologies Inc., Pulsante microstimulator and remote controller.

The device is placed in a minimally invasive procedure that can be done in less than 60 min. It has maximal targeted flexibility of programming, and can be inserted or removed easily at patient request without permanent damage.

Randomized Controlled Trial