Extracranial vertebral artery rupture likely secondary to “cupping therapy” superimposed on spontaneous dissection

Abstract

The extracranial vertebral artery (VA) is vulnerable to dissection and the V3 segment is the most common location for dissection. Dissection accounts for about 2% of all ischemic strokes and can occur after trauma or chiropractic neck maneuvers. We report an extremely rare case of spontaneous extracranial VA dissection presenting with posterior neck hematoma aggravated after cupping therapy, a treatment in traditional Oriental medicine. We treated the patient successfully by endovascular treatment without any complication.

Introduction

Unruptured dissection of the vertebral arteries (VAs) is rare and more likely to involve the intradural rather than the extracranial portion of the artery.¹ Neck pain on the side of the injury is the most common symptom of VA dissection (VAD),² and may include symptoms of posterior fossa ischemia or infarction, such as dizziness, vertigo, ataxia and syncope.^3,4 In addition to these common symptoms, unusual presentations include toothache,⁵ cervical radiculopathy,⁶ Brown-Séquard syndrome⁷ and extracranial vertebral arteriovenous fistula.⁸

We report an extremely rare case of spontaneous extracranial VAD presenting with posterior neck hematoma aggravated after cupping therapy, a treatment in traditional Oriental medicine. We treated the patient successfully by endovascular treatment of the VA without any complication. To the best of our knowledge, a clinical presentation of posterior neck hematoma with a VAD has been reported only once before.

Case presentation

A 54-year-old male shipyard worker was referred to our institution with severe left posterior neck pain and swelling. On admission day, while using a hammer, he felt a sudden onset of severe neck pain. The degree of pain at the first symptom attack was visual analog scale for pain (VAS) 5. He went to a Korean traditional clinic for cupping therapy. The neck pain became much more severe, VAS 8. His posterior neck began swelling, albeit without neurological symptoms. Non-enhanced brain computed tomography (CT) demonstrated no remarkable finding. However, an inter- and intramuscular hematoma was revealed at the left suboccipital area. There was a hematoma in the epidural space at the C1 level and slight compression of the spinal cord (Figure 1). Brain CT angiography showed contrast media extravasation at an extra-spinal segment of the left VA at the V3 segment (Figure 2). Transfemoral cerebral angiography (TFCA) was carried out, and confirmed a posterior projection pseudoaneurysm of about 17 mm at the V3 segment. The left posterior inferior cerebellar artery (PICA) was of extradural origin but the neck of the pseudoaneurysm was more proximal than the origin of the left PICA (Figure 3). There was no arteriovenous communication at the ruptured point. We decided to occlude the parent artery because the dominancy of VA was contralateral.

Figure 1.

Non-contrast-enhanced head and neck computed tomography (CT) demonstrates inter- and intramuscular neck hematoma at the C1 level with slight cord compression.

Figure 2.

Brain computed tomography (CT) angiography (a) axial, (b) coronal, and (c) sagittal maximum intensity projection images show contrast media extravasation (black arrows) at extra-spinal segment of the left vertebral artery (V3 segment).

Figure 3.

Left vertebral artery oblique angiography revealed (a) a 17 mm sized posterior projection pseudoaneurysm (white arrow indicates rupture point) arising from the V3 segment and proximal luminal narrowing of the vertebral artery. (b) The origin of the left posterior inferior cerebellar artery (PICA) (black arrow) arises from the extradural segment of the vertebral artery and the neck of the pseudoaneurysm (white arrow) is more proximal to the origin of the PICA.

A technique using two microcatheters was chosen for the left VA sacrifice. To confirm retrograde flow from the right VA to left PICA, a 5Fr angiocatheter was placed in the right proximal VA for right VA angiography. To secure the patency of the PICA, a large sized detachable coil was inserted temporarily at the origin of the PICA, and the proximal VA was occluded with another microcatheter. After removing the distal coil, we occluded the patent VA while maintaining patency of the PICA.

Final angiography revealed complete occlusion of the left VA, with patent left PICA flow through the right VA (Figure 4). The patient was discharged from the hospital one week after embolization procedure without neurologic deficit. Seven months later, brain CT angiography showed completely occlusion of the pseudoaneurysm with patent PICA flow (Figure 5).

Figure 4.

Post-procedural right vertebral artery (VA) angiography confirms complete obliteration of the left vertebral artery (VA) disease segment with patent left posterior inferior cerebellar artery (PICA) (white arrow) flow through the right VA.

Figure 5.

Follow-up brain computed tomography (CT) angiography three-dimensional (3D) image after seven months shows complete occlusion of the pseudoaneurysm with patent posterior inferior cerebellar artery (PICA) flow.

Discussion

VAD is an important cause of posterior circulation ischemic stroke in young adults.⁹ Vertebral and basilar arteries are important in the posterior circulation of the brain; therefore VAD typically presents with neck or occipital pain preceding a posterior circulation ischemic stroke.¹⁰ The etiology of VAD may be either traumatic or spontaneous.² In most spontaneous cases, the cause of VAD is obscure but there is often a history of a minor precipitating event associated with hyperextension or rotation of the neck, including practicing yoga, chiropractic therapy, painting a ceiling, coughing, sneezing, vomiting and the act of resuscitation.¹¹ Genetic connective-tissue diseases such as Marfan syndrome, Ehlers-Danlos syndrome, and fibromuscular dysplasia entail a high risk of VAD.¹² Our patient has no genetic predisposing factors for VAD. He has no history of using tobacco or medication such as oral contraceptives.¹² In this case, the only notable predisposing factor was that the patient is a shipyard worker and therefore liable to hyperextension or rotation of the neck. His neck pain began while hammering, and posterior neck swelling developed after cupping therapy in a Korean traditional clinic. He has no acupuncture therapy history at the Korean traditional clinic, so there is no possibility that an acupuncture needle penetrated the VA and caused a suboccipital lesion hematoma. We presume that when spontaneous extracranial VAD occurred, his neck pain began and the cupping therapy caused a VA rupture and posterior neck hematoma with a pseudoaneurysm.

Cupping is traditionally performed with a small round cup made of thick glass with a rolled rim to maintain air-tight contact with the skin to preserve the vacuum.¹³ Nowadays, cupping is increasingly carried out with plastic cups and a manual hand-pump to create a vacuum, for chronic pain such as neck, back and shoulder.¹³ Some authors explain from a biomechanical perspective of cupping, the vacuum pressure inside the cup bulges the skin and the suction force exerts an effect on the soft tissue composed of skin, fat, and muscle layer.¹³ Blunt and Lee describe a patient who received cupping to the neck and developed intracranial hemorrhagic stroke 14 hours later, with no other cause identified.¹⁴ Cupping of the cervical area can influence the carotid arteries and VAs supplying the cerebral circulation. Cupping to the neck might be a possible mechanism leading to stroke.¹⁴ Cupping pressure has a pressure range of 150.01 mmHg to 750 mmHg. Ciszek et al. studied rupture pressure and patterns with 10 human cadaver cerebral arteries such as the middle cerebral artery, anterior communicating artery, internal carotid artery and vertebral and basilar arteries.¹⁵ The vessels were pressurized with saline and observed until they ruptured. The average rupture pressure was 2.21 ± 0.59 atm (1 atm = 760 mmHg).¹⁵ Therefore normal blood vessels are well able to withstand high pressure, so it is difficult to assume that cupping alone causes direct vessel rupture and neck hematoma. However, the tensile radial stresses generated by cupping potentially facilitate posterior neck hematoma formation with VA rupture in the presence of VAD.

Although there have been some reports on the natural course of acute VAD, it is still not well understood. The treatment strategy for acute VAD is also controversial. But, most agree that prevention of rebleeding of ruptured VAD is an essential goal for treatment. However, extracranial VAD rupture is extremely rare, and we don’t know the natural history of this disease and an exact treatment strategy. The treatment options of extracranial VAD rupture with hemorrhage could be deconstructive treatments such as endovascular internal trapping, surgical trapping or reconstructive treatments using coils and stents, flow diverter, or direct surgical repair of ruptured point. However, deconstructive treatment is not possible in some situation such as dominant VAD rupture with poor collateral flow or PICA-involved VAD.

Concerned about expansion of dissection into intracranial VA, we decided to occlude the disease segment. Non-dominant type VA and a PICA-proximal ruptured point could facilitate sacrificing the VA.

Conclusion

The clinical diagnosis of VAD is not easy with acute-onset neck pain, especially young patients with no evidence of cervical trauma and disease. With undiagnosed VAD, traditional remedies such as cupping and chiropractic therapy to reduce pain can aggravate spontaneous dissection and worsen symptoms. We report an extremely rare case of extradural VA rupture caused by neck hematoma after neck cupping therapy. Physicians should be aware of the possible clinical manifestations and findings on radiographic imaging. Early diagnosis of VAD could prevent symptom aggravation and permanent neurologic deficit.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.