Asystole during onyx embolisation of a dural AV fistula: The trigeminocardiac reflex

Abstract

There are fewer than 20 published case reports of bradycardia or asystole during intracranial embolisation procedures. These are well described in open neurosurgical procedures, particularly involving the skull base. We present a case of a 59-year-old male patient who presented for elective embolisation of a dural arteriovenous fistula. During the injection of Onyx, the patient experience sudden asystole, which recurred after a second Onyx injection. Following successful treatment, a third injection proceeded without incident.

Introduction

Haemodynamic instability during endovascular procedures is relatively rare, but is described. There are fewer than 20 published case reports of bradycardia or asystole during embolisation procedures. These are well described in open neurosurgical procedures, particularly involving the skull base. We recently experienced such a case that we feel carries important teaching points for all members of the neurovascular team involved in the care of these patients.

Clinical case

A 59-year-old male presented with ongoing tinnitus and headaches, and underwent MRI imaging showing a Borden and Cognard Class III dural arteriovenous fistula (dAVF) involving the transverse sinus and torcula, with significant venous congestion over the entire cerebellum. He was otherwise well, with no significant cardiac history. Following discussion of treatment options in the clinic, the patient elected to proceed with endovascular treatment of this fistula and was admitted for this. He had been assessed and cleared by the pre-operative anaesthetic clinic prior to the procedure.

The procedure was performed under general anaesthesia and systematic heparinisation. The patient underwent routine general anaesthesia induction without incident, and there were no haemodynamic abnormalities during induction. After initial diagnostic angiography (Figure 1), a 90 cm Neuron Max guide catheter (Penumbra, Inc, USA) was placed in the proximal left vertebral artery (Figure 2) and, with the aid of an X-Pedion 0.010” hydrophilic microwire (Covidien, EV3, USA), an Apollo 2.7 French (Covidien, EV3, USA) microcatheter was advanced to the posterior meningeal artery using a Fargo Mini 040 intermediate catheter (Balt, France) for support. A microcatheter injection was performed to confirm the fistulous point arising from the posterior meningeal artery (Figure 3). Once all catheters were in a stable position, embolisation of the fistula proceeded.

Figure 1.

Lateral view of a left vertebral artery angiogram showing the Borden type III dural arteriovenous fistula being fed mainly from branches of the posterior meningeal artery with signs of cerebellar venous congestion.

Figure 2.

Lateral view of a left vertebral artery angiogram following the procedure, showing interval resolution of the dural arteriovenous fistula.

Figure 3.

Superselective microcatheter angiogram in the same projection, showing the left posterior meningeal artery injection. The fistulous point can be seen (arrow) along with faint early venous reflux (arrowheads).

The embolisation procedure was carried out using Onyx (EV3; Neurovascular, Irvine, California, USA). Dimethyl sulfoxide (DMSO) was injected without consequence, but following injection of the first few 0.1 cc of Onyx the patient experienced asystole. The injection was immediately paused, and return of spontaneous circulation achieved after both chest compressions and the use of intravenous adrenaline and atropine. After the patient was stabilised, and following discussion with the anaesthetist, a second Onyx injection was performed, with an identical haemodynamic response. This too was successfully treated in a similar fashion. Again the patient was stabilised, and following discussion, a third Onyx injection proceeded without any more cardiac episodes.

We subsequently achieved good angiographic occlusion of the fistula, with opacification of the fistulous point extending to the foot of the draining vein (compare Figure 1 with Figure 2). Final angiography showed no thromboembolic complications as a result of the procedure.

The patient awoke without any new neurological symptoms related to the procedure. He remained in a monitored bed for 48 hours following the procedure, during which time he experienced no more haemodynamic instability. There were no reported respiratory symptoms, subsequent cardiac workup was negative and he was discharged 2 days later. He reported some improvement in his presenting neurological symptoms over the following few days and was asymptomatic on his later follow-up clinic visits.

Discussion

This case demonstrates an important consideration for intracranial embolisation procedures. The trigeminocardiac reflex (TCR) is the presumed cause of this patient’s haemodynamic instability. The TCR has been described as a physiological reflex that occurs in response to stimulation of the trigeminal nerve. The fistula in our case was adjacent to the transverse sinus and the torcula. The dura mater of the posterior fossa is innervated in part by branches of the trigeminal nerve and receives vascular supply from branches of the posterior meningeal artery (arising from the vertebral artery) as well as meningeal branches of the occipital artery.^1,2 Hence, embolisation of these vessels can stimulate the trigeminal nerve and cause a TCR.

Like all brainstem reflexes, the phylogenetic origin of the TCR is unclear but many authors believe that like the other vagal reflexes it is an older ‘protective’ reflex. The TCR has been well described in various open neurosurgical procedures, but there are relatively few published cases related to its occurrence in neuroendovascular cases.³ Of note, the TCR has only been reported with the use of Onyx and not with other embolic materials. It has been reported in the injection of both the DMSO and the Onyx. It is therefore likely it is the DMSO itself that has this effect, through an as-yet unknown mechanical or biochemical effect on the dural nerves.⁴ Although temporary, this haemodynamic effect needs to be quickly addressed or it can lead to serious morbidity and/or mortality.

There is no way of predicting which patients will present with this response. Therefore, our recommendation is that interventionalists need to be aware of the possibility of this response during all intracranial embolisation procedures, but particularly with the use of Onyx for dural fistulas. A brief ‘pause’ prior to injection to inform anaesthesia colleagues of any potential instability is therefore our standard practice.

It is also reassuring that this is a transient phenomenon; in all cases in the literature^5–8 the TCR resolved following treatment with chest compressions and/or atropine without permanent sequelae.

Conclusion

Practitioners using Onyx for embolisation of dAVF should be aware of the potential for haemodynamic instability during the case. The anaesthetist and other members of the treating team should be pre-notified prior to injection of the embolic material. These cases usually respond well to treatment.

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.