Article Text

Refractory primary and secondary headache disorders that dramatically responded to combined treatment of ultrasound-guided percutaneous suprazygomatic pterygopalatine ganglion blocks and non-invasive vagus nerve stimulation: a case series
  1. Andre P Boezaart1,2,
  2. Cameron R Smith1,
  3. Yury Zasimovich1,
  4. Rene Przkora1,
  5. Sanjeev Kumar1,
  6. Olga C Nin1,
  7. Louis C Boezaart3,
  8. Daniel AJ Botha4,
  9. André Leonard5,
  10. Miguel A Reina1,6 and
  11. Juan A Pareja7
  1. 1Anesthesiology, University of Florida College of Medicine, Gainesville, Florida, USA
  2. 2Lumina Health, Surrey, UK
  3. 3Private Ophthalmology Practice, Cape Town, South Africa
  4. 4Private Family Practice, Still Bay, South Africa
  5. 5Private Neurology Practice, Mossel Bay, South Africa
  6. 6Anesthesiology, CEU San Pablo University Faculty of Medicine, Madrid, Spain
  7. 7Neurology, Hospital Universitario Quirón Madrid, Madrid, Spain
  1. Correspondence to Andre P Boezaart, Anesthesiology, University of Florida College of Medicine, Gainesville, FL 32610, USA; ABoezaart{at}


In 1981, Devoghel achieved an 85.6% success rate in treating patients with treatment-refractory cluster headaches with alcoholization of the pterygopalatine ganglion (PPG) via the percutaneous suprazygomatic approach. Devoghel’s study led to the theory that interrupting the parasympathetic pathway by blocking its transduction at the PPG could prevent or treat symptoms related to primary headache disorders (PHDs). Furthermore, non-invasive vagus nerve stimulation (nVNS) has proven to treat PHDs and has been approved by national regulatory bodies to treat, among others, cluster headaches and migraines.

In this case series, nine desperate patients who presented with 11 longstanding treatment-refractory primary headache disorders and epidural blood patch–resistant postdural puncture headache (PDPH) received ultrasound-guided percutaneous suprazygomatic pterygopalatine ganglion blocks (PPGB), and seven also received nVNS. The patients were randomly selected and were not part of a research study. They experienced dramatic, immediate, satisfactory, and apparently lasting symptom resolution (at the time of the writing of this report). The report provides the case descriptions, briefly reviews the trigeminovascular and neurogenic inflammatory theories of the pathophysiology, outlines aspects of these PPGB and nVNS interventions, and argues for adopting this treatment regime as a first-line or second-line treatment rather than desperate last-line treatment of PDPH and PHDs.

  • postdural puncture headache
  • nerve block
  • post-dural puncture headache

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, an indication of whether changes were made, and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


  • Primary headache disorders (PHDs) are challenging to treat effectively because the precise mechanisms that trigger attacks or episodes are unknown, and they are particularly refractory. Furthermore, postdural puncture headache (PDPH) treatment is mostly effective with an epidural blood patch (EBP), but it requires another epidural injection.


  • Percutaneous suprazygomatic pterygopalatine ganglion blocks (PPGB) and non-invasive vagus nerve stimulation (nVNS) can provide immediate and long-lasting relief of the pain and other symptoms associated with primary and some secondary headache disorders and prevent further attacks. Repeated EBP failed to provide relief to our patients with PDPH.


  • Research into the impulses that trigger the mechanisms that cause the pain associated with PHDs will help further demystify these disorders’ elusive etiopathogenesis and pathophysiology. This case series argues for adopting PPGB and nVNS as first-line and second-line treatments rather than desperate last-line treatment modalities for PHDs and PDPHs and to compare standardized versions of them to modern drug therapies prospectively.


Devoghel1 developed a treatment regime based on results published in 19332 to treat patients with longstanding treatment-refractory cluster headaches.1 He unsuccessfully treated the patients over at least 4 years with antihistamines, phenothiazines, H1-antagonists, methysergide, indomethacin, prednisone, propranolol, and lithium carbonate before the pterygopalatine ganglion (PPG) alcoholizations. Similar to more recent studies,3 4 Devoghel found that the percutaneous suprazygomatic approach to the PPG is the simplest and safest technique, and he achieved a remarkable success rate of 85.8% of patients who went into apparent permanent remission.1

While the etiopathogeneses of most secondary headaches are generally well understood,5 that of primary headache disorders (PHDs) still eludes us.6–9 However, regarding the pathophysiology of the pain associated with PHDs, the trigeminovascular theory (TVT), although disputed, has stood the test of time to simplify and comfortably explain information and experience obtained in research and clinical practice.10–14 According to the TVT, pain associated with PHD involves a vicious cycle of parasympathetically induced vasodilation and inflammation caused by inflammatory mediators in the cerebral extraparenchymal blood vessels and the meninges.15 16 It is reasonable to argue that interrupting this pathway with a conduction block could prevent vasodilatation and inflammation (whichever phenomenon causes the other) and, thus, nociception of the PHDs and, theoretically, that of postdural puncture headaches. Blocking the PPG is also theorized to be effective in treating ipsilateral autonomic symptoms classically associated with trigeminal autonomic cephalalgias, which follow similar anatomical pathways.17

Transcutaneous cervical vagus nerve stimulation (GammaCore, ElectroCore, Rockway, New Jersey, USA) has been approved for episodic cluster headaches since 2017.18 A comprehensive update on the mechanism of action of non-invasive vagus nerve stimulation (nVNS) for PHD has recently been published by Silberstein and colleagues, in which they provide an in-depth review of the relationship between the vagus nerve and the trigeminal autonomic reflex.19 These and the potent anti-inflammatory actions via the cholinergic anti-inflammatory process of nVNS are briefly discussed in the Discussion section.20 21

This communication presents nine case studies of desperate patients with severe treatment-refractory and long-standing suffering with PHDs who responded dramatically to percutaneous suprazygomatic pterygopalatine ganglion blocks (PPGB) and transcutaneous auricular vagus nerve stimulation (taVNS) (Parasym, London, UK) and briefly discusses the pathophysiology as it is currently understood as well as aspects and rationale of these treatment regimes.

Case descriptions

The patients in this case series included five patients with PHD (epicrania fugax, chronic paroxysmal hemicrania, new daily persistent headache, episodic cluster headache, and menstrual migraine without aura), one patient with two types of PHDs (menstrual followed by chronic retinal migraine), one with primary and secondary headache disorders (tension-type headache and medication overuse headache), and two patients with secondary headache disorders (postdural puncture headache) (table 1). The patients in this case series were randomly selected as they presented to individual practitioners and did not form part of a research project. They were all treated with PPGB (the video ‘Ultrasound Guided Sphenopalatine Ganglion Block’ on the Regional Anesthesiology and Acute Pain Medicine YouTube channel details this approach),1 3 4 22–25and seven of the nine also received taVNS.26–28 These were offered as desperate attempts to manage their conditions. All cases (except for patients #8 and #9) suffered for many years from debilitating, severe refractory headaches and presented to individual practitioners who had vast experience with PPGB procedures and taVNS. These practitioners had abandoned epidural blood patch (EBP) to treat PDPH and replaced it with PPGB with excellent results for the previous 7 years. They had also gained extensive experience with PPGB for various persistent PHD disorders and postoperative pain management for conditions such as adult and pediatric tonsillectomy, pediatric cleft palate repair, functional endoscopic sinus surgery, and others. As with Devoghel’s cases,1 all of the patients (except #2 and #6, who presented to seek alternative non-pharmacological treatments because of unwanted adverse effects) failed to respond to all known and locally available therapies (table 1) and were desperate for relief.

Table 1

Case series of primary and secondary headache disorders

Figure 2

(A) Before and (B) immediately after images of the patient with epicrania fugax (#1) undergoing a percutaneous suprazygomatic pterygopalatine ganglion block (photographs published with patient’s kind permission).

All the patients were Caucasian and provided written consent to publish their case studies. Patients were managed in different private practice out-of-hospital settings, thus precluding institutional review board approval for publication of the case reports.

The PPGBs were all performed with 4 mL of 0.5% ropivacaine and 3 mg (1 mL) of betamethasone per side via a 22-gage B-bevel needle after thorough numbing of the skin and subcutaneous tissue with 1%–2% lidocaine via a 27-gage needle. All the patients also received mild conscious sedation with low-dose propofol and midazolam as required. taVNS, applied to the tragii or conchae of the patients’ ears at a frequency of 30 Hz and pulse width of 250 µs for at least 60 min per day (but typically for longer and up to 180 min), was added to try to minimize any further exacerbations.


This case series presents nine patients with 11 treatment-resistant and longstanding primary (eight) and secondary (three) headache disorders. The patients responded dramatically and potentially (as of the time of writing this report) permanently to PPGB and taVNS.

Despite the variety of disorders, we propose that their pathophysiologies were comparable due to a common anatomical and biochemical pathway and process involved in generating pain, thus yielding similar results.

The etiopathogenesis of PHDs eludes us and that of PDPH is perhaps better understood, while the pathophysiology of these conditions is still being investigated and debated. It is complex, and several theories attempt to explain what causes the pain of PHD.6–10 15 29 Although the neurogenic inflammatory theory29 has to a large extent replaced the trigeminovascular theory (TVT) as the current favorite, the TVT still adequately, although perhaps in an oversimplified manner, and comfortably explains why PPGB, by blocking the parasympathetic pathway, is successful in treating PHD and PDPH, both characterized by meningeal cerebral extraparenchymal blood vessel dilatation and meningeal inflammation, whichever is first and of whatever thus far unknown other mechanism or cause(s)10–14 (please see figure 1 for an abbreviated explanation of the TVT).

Figure 1

Pathophysiology of PHD according to TVT. Impulses generated from different brain areas (orange arrows) travel to the salivatory nucleus (1) from where parasympathetic impulses travel via the facial nerve (2), geniculate ganglion (3), greater petrosal nerve (4), and Vidian nerve (5) to synapse in pterygopalatine ganglion (6). From here, postsynaptic parasympathetic impulses travel via the maxillary nerve (7) and meningeal nerves (8) to reach the meninges (10, 11) and extraparenchymal meningeal blood vessels (12), which are dilated and inflammatory mediators are released. Either the vasodilatation or the inflammation or both cause nociception, and the nociception impulses travel via the trigeminal ganglion (16) and trigeminal nerve (17) to the principal (18), mesencephalic (19), and spinal (20) nuclei of the trigeminal nerve causing the release of further inflammatory mediators to spread to other areas of the brain and be interpreted as pain (reproduced with the kind permission of Mary K. Bryson and Lumina Health). PHD, primary headache disorder; TVT, trigeminovascular theory.

On the other hand, according to the neurogenic inflammatory theory,29 inflammatory mediators—among others, substance-P, acetylcholine, and calcitonin gene-related peptides—are released by the trigeminal autonomic reflex from the trigeminal system and cause the extraparenchymal cerebral vasodilatation, an epiphenomenon that, according to this theory, does not per se cause the headache.

Nociceptive impulses originate from the inflamed meningeal areas and pass through the trigeminal ganglion, which comprises trigeminal pseudounipolar neurons, to the trigeminal nerve’s principal sensory, mesencephalic, and spinal nuclei. Additional inflammation and nociceptive impulses are generated in and around these nuclei and join those from the meninges to spread to other brain areas where the impulses are interpreted as pain. These other brain areas may include, but are not limited to, the superior salivatory nucleus and the trigeminal nerve and its branches, the spinal nucleus of the trigeminal nerve and the cervical nerves and muscles, the parabrachial nucleus, the periaqueductal gray matter, the hypothalamus, and the thalamus.

The etiopathogenesis of PDPH is, likewise, still not completely understood, but better than that of PHDs. In a 2003 porcine study, Boezaart30 demonstrated a loss of approximately 0.3 mL of CSF per kg body weight of these animals (around 23–25 kg) aspirated from a cisterna magna needle puncture caused an immediate and significant increase in cerebral cortical blood flow. This, in turn, was immediately reversed by distant lumbar EBP—iatrogenic peridural hematoma31 —which is known as an early and potent stimulus for cerebral vasoconstriction.32 While vasodilation (in the case of PDPH, based on the Monro-Kellie hypothesis33), and inflammation, most likely due to the cortical vasodilatation or some trigeminal reflex, are probably shared features in some PHDs and PDPH, we recognize that these conditions’ specific etiology, triggers, and pathophysiological mechanisms differ.34

Despite the theoretical mechanistic differences and poorly understood pathophysiologies, treating both PHD and PDPH with PPGB may share similar reasons to be effective.35 These may include the disruption of parasympathetic-activated vasodilatation or inflammation at the PPG and the modulation of inflammatory neurotransmitters. Finally, it may have the potential to interrupt trigeminal activation and thus the transmission of pain signals during PHDs and PDPHs. By blocking conduction through the PPG, the transmission of parasympathetic vasodilatory or inflammatory signals from the superior salivatory ganglion (figure 1) and pain signals from the trigeminal nerve to the central nervous system are disrupted, relieving headache symptoms.

There are several approaches to blocking the PPG.36 However, suprazygomatic seems the safest percutaneous approach.3 4 We have, over a 7 year period, not encountered any serious side effects in using PPGB on hundreds of patients for various indications (see Background section) other than transient numbness of the upper jaw and teeth and rarely lower jaw and tongue for as long as the local anesthetic agent is active, and postprocedure short-lived cheek tenderness and swelling. These could easily be treated by patient reassurance and with non-steroidal anti-inflammatory agents. Other extremely rare complications may be encountered, such as puncturing of the maxillary artery, hematoma formation, and intravascular injection. However, to date, we have not met these side effects, nor to the best of our knowledge have they been reported in literature. Patient discomfort during the procedure is limited to an absolute minimum if a fine needle is used for skin and subcutaneous local anesthetic lignocaine infiltration (or for the PPGB) combined with mild levels of conscious sedation as indicated.

In the experience of the authors4 36 and as argued by Narouze,37 among others, the transnasal sphenopalatine ganglion block (SGB) yields inconsistent results. It may block the autonomic nerve fibers in the nasal cavity and interrupt some of the autonomic symptoms associated with PHD. Still, due to its unreliable results, we do not support its use. It may be argued that it is the safest approach to blocking the PPG (SGB). We concur with Narouze37 that the transnasal approach most likely does not reach the PPG. If it does, it does so unreliably and with ineffective, minimal dosages of local anesthetic agents.

We cannot explain why short-acting local anesthetic agents combined with relatively short-acting (~3 months) steroids could have such lasting effects. Breaking the vicious cycles of vasodilatation and inflammation may partially explain it. However, the long-term effects may also be explained by our patients’ compliant use of taVNS postprocedure. This long-lasting effect was especially apparent in the patient described as case #6 (CH, who experienced years of remission after single-injection superior cervical (stellate) ganglion blocks). Since remission recurred on three occasions after autonomic blocks in this patient, it is unlikely to be purely coincidental or a placebo effect. However, the patients with PPDH (cases #8 and #9), in whom the PBGB could not possibly have stopped the CSF leaks and whose leaks could arguably still have been active in the form of fistulae at the time of the blocks, were not treated with nVNS. This finding should be further debated and ultimately clarified by ongoing research. The answer to this question may pave the way for radiofrequency ablation38 or other destructive techniques such as alcoholization1 or PPG electrical stimulation.39 The immediate effects on the patient with epicrania fugax (case #1, figure 2) can be partially explained by the blocking of the autonomic nerve fibers that run with the ophthalmic (V1) and maxillary (V2) nerves.

The precise mechanisms underlying the therapeutic effects of nVNS in PHD are still being investigated, and additional research is needed for a comprehensive understanding.8 Evidence suggests that nVNS interacts with multiple aspects of headache pathophysiology.19 These include short-term and long-term neurotransmitter modulation and thus a lower frequency of attacks, autonomic regulation leading to a decreased firing of trigeminal neurons, and hence acute relief of parasympathetic nervous system symptoms.19 nVNS furthermore causes nociceptive modulation, mitigating increased nociception and acute pain relief, decreased susceptibility to cortical spreading depression initiation and frequency, and thus acute attack (aura) resolution and reduced attack risk.19 Finally, nVNS has potent anti-inflammatory effects through its effect on the cholinergic anti-inflammatory process.20 40 This all ultimately leads to symptom relief and improved headache management with nVNS.


The conclusive clinical benefit of PPGB and nVNS, separately or combined for treating PHDs and, to a lesser extent, that of PDPH remains to be determined. Further research is required to compare this treatment strategy prospectively to other newer pharmacologic therapies. Because of the simplicity, relative safety, and cost-effectiveness of PPGB and nVNS, these treatment modalities should be offered early on to patients with therapy-refractive or therapy-resistant PHDs and early on to patients with PDPH not responding to simple conservative non-invasive measures to eliminate yet another epidural injection (EBP). For the same reasons, we strongly advocate that it should be offered to patients as first-line or second-line treatment after failed simple pharmacological treatment with OTCs, triptans, etc., rather than desperate last-resort treatments—especially in middle-to-low-income developing countries where newer, expensive therapy modalities and super-specialized services and treatments are not readily available.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants. The cases described in this case series were all managed in the offices of various private practitioners who are not attached or affiliated to nor associated with any institution. Institutional review boards or ethical review boards are therefore not applicable as the practitioners, who own their practices and who are the coauthors, have complete authority over their private practices. To comply with this, they would in fact have to give themselves approval, which would not make any sense. Participants gave informed consent to participate in the study before taking part.


The authors express their deepest gratitude to the patients for permitting them to publish their case reports. They also sincerely thank Dr Natasha Beukes (PhD), Waldi du Toit, Gareth Coetzee, and Kirsten Holman of Lumina Health, and Bryan Penberthy (MFA) of the University of Florida for their valuable contributions to editing and preparing the manuscript.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Presented at Some of the cases (1-5) have been presented as abstracts and poster presentations at the American Academy of Pain Medicine Annual Meeting on March 23, 2023, in Florida, USA.

  • Contributors All authors contribute to project development, data collection and management, and manuscript writing.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.