The risk factors of trigeminocardiac reex caused by Onyx embolization during cerebrovascular intervention operation

Trigeminocardiac reex (TCR) is a brainstem reexive response of hemodynamic instability during surgery. However, few risk factor analysis and management strategies of TCR have been highlighted in the literature. The purpose of this study was to demonstrate the risk factors for Onyx embolization during cerebrovascular intervention operation so as to optimize perioperative management and reduce the incidence of TCR. We performed a retrospective study on consecutive patients with Onyx embolization over a 4-years period. According to whether TCR episode, they were divided into TCR group and control group. Cases and controls were compared regarding patient characteristics, comorbidities, hemodynamics and complications. We also constructed several multivariable regression models to analyze the risk factors.


Introduction
Trigeminocardiac re ex (TCR) is a unique brain stem re ex that manifests as the sudden onset of hemodynamic perturbation in heart rate and blood pressure as a result of stimulation of any branches of the trigeminal nerve [1]. TCR has been described during intracranial, maxillofacial, ophthalmic surgery, endovascular treatment of intracranial dural arteriovenous, and carotid-cavernous stulas as well as radiofrequency lesioning of the trigeminal ganglion [2].
In recent years, with the development of neurointerventional technology, the treatment of Onyx embolization has gradually become an independent and important measures for the cerebrovascular malformation. This technique achieves the purpose of blood ow redistribution by blocking blood supply and occluding arteriovenous stula, which has been clinically con rmed [3]. However, with the wide application of Onyx in interventional operation, its common complications and potential risks should not be ignored. Onyx embolization in cerebrovascular interventional surgery might trigger TCR, leading to severe hemodynamic uctuations and even cardiac arrest [4][5][6][7][8][9]. In addition, the occurrence of TCR is unpredictable, and there is no appropriate prophylactic approaches to prevent. Therefore, TCR is beginning to reactivation in our eld of vision, which needs to be paid enough attention.
Due to its relatively low incidence and high potential risk, there is no clear consensus on prevention or treatment of TCR, therefore, we need to recognize this phenomenon as soon as possible. In this study, we aimed to retrospectively investigated the incidence and related risk factors of TCR during the treatment of cerebrovascular embolization with Onyx to guide clinical practice.

Materials And Methods
This retrospective observational cohort study was approved by the Institutional Review Board (IRB) of The First A liated Hospital of Nanjing Medical University/Jiangsu Province Hospital (JSPH; Jiangsu, China; IRB approval number: 2020-SR-312), and registered on Chinese Clinical Trial Register (chictr.org.cn ChiCTR2000038839; October 5, 2020). Considering the retrospective design of this study, the requirement for informed consent was waived by the IRB.

Participants and Study Design
This study utilised data stored and managed in the electronic medical record system of JSPH on 215 patients who were decided to perform Onyx embolization operation under general anesthesia by neurosurgeon, from January 2016 to May 2020 at a single academic institution. Patients < 18 years old were excluded from the study. Similarly, patients with preoperative confusion, communication di culties were not eligible. Lastly, we excluded patients with severe complications unrelated to cerebrovascular embolism occurred during perioperative period. According to whether TCR episode during the operation, they were divided into TCR group (T group) and control group (C group). In all cases, the archived cases and anesthetic records were reviewed in detail, and the data were true and reliable. All the cases for the study period were screened by a group of medical record technicians in the medical informatics team who were not informed of the purpose of this study.

Monitoring and Anesthesia
All patients were performed with standardization of anesthesia induction and maintenance, intraoperative anesthetic management was performed with continuous monitoring of peripheral capillary oxygen saturation, electrocardiography (ECG), arterial blood pressure, end-tidal carbon dioxide partial pressure (PetCO2) and Bispectral index (BIS). Anesthesia was induced with midazolam 0.05 mg/kg, fentanyl 3 mg/kg, propofol 2 mg/kg, cisatracurium 0.15mg/kg. Mechanical ventilation (Drager, Fabius-Plus, Germany) was used during the operation to keep PetCO2 at 30-45 mmHg. All patients received 1%-2% sevo urane and 2-5 mg/kg·h propofol for anesthesia maintenance until the end of operation, and 0.1 mg/kg·h cisatracurium for muscle relaxation to half an hour before the end. The depth of anesthesia was guided by BIS < 60 to prevent intraoperative awareness.

Variables and Outcomes
Perioperative variables were ascertained from patient records. The following data were recorded: patient age, sex, Body Mass Index (BMI), comorbidities, protopathy, supplying vessels, PetCO2, BIS value (record per 5 minutes during embolization), hemodynamic indexes (baseline, before embolization, embolization and end of procedure) and the use of atropine, what's more the postoperative outcomes. For the purposes of this study, TCR was de ned as the sudden onset of bradycardia triggered by stimulation of the trigeminal nerve and its anatomic branches. The bradycardia is characterized by a reduction in HR of 20% or more from the baseline, and/or asystole. Change in MAP is an optional criterion for the de nition of TCR [10][11], but not included as part of the TCR de nition in this study.

Statistical Analysis
All the statistical analysis was performed with SPSS 23.0 (IBM Software Inc., USA). Categorical variables were presented using numbers with percentages and were analyzed with chi-square test or Fisher's exact test, whereas the continuous variables were expressed as mean ± standard deviation, and were compared with the Student's t-test for unpaired samples when a normal deviation was assumed. Univariate and stepwise multivariate logistic regression analysis were performed to determine the risk factors of TCR episode. All clinically sensible covariates were included in the model. For all analysis, a P value of < 0.05 was considered statistically signi cant.

Patient Characteristics
We identi ed 215 patients who required endovascular embolization therapy with Onyx under general anesthesia during the study period. Of these, 11 patients with preoperative disturbance of consciousness, and 9 children younger than 18 years were excluded. Therefore, our nal study cohort included 195 patients who ful lled our selection criteria (Fig. 1).
The demographic and clinical characteristics over the entire cohort are presented in detail in Table 1. The mean (SD) age of our cohort was 45.3 (15.1) years. Gender distribution was 121 male (62.1%) and 74 female (37.9%) patients. In terms of comorbidities, hypertension was present in 20.5% of patients, diabetes mellitus in 6.2%, bradycardia in 11.8%, and the history of cerebral hemorrhage in 36.9%. Regarding preoperative diagnosis, 93 patients (47.7%) were diagnosed with arteriovenous malformation (AVM), 55 (28.2%) with dural arteriovenous stula (DAVF), 16 (8.2%) with carotid-cavernous stula (CCF), and 31 (15.9%) with intracranial tumors.   *P < 0 .05. Table 4 shows the patients with signi cant hemodynamic uctuations. Intraoperatively, bradycardia occurs during TCR episode, however, blood pressure may appear to be either lower or higher. 13 patients presented with hypotension, while 14 patients with hypertension. Among them, whose embolized vessel is the middle meningeal artery is more likely to present with cardiac arrest and hypotension when TCR episode. In addition, this study results also showed that 16 patients just experienced a 20% increase in blood pressure and no signi cant decrease in heart rate during the process of embolization. Whereas, we did not nd patients presenting with a simple hypotension.

Prognosis
The composite adverse events occurred in 4 of 27 (14.8%) patients in the TCR group and in 39 of 168 (23.2%) of the patients in the control group (P > 0.05). The incidence of postoperative dizziness, headache, PONV, and muscle weakness were quite similar in two groups (P > 0.05) ( Table 5).

Discussion
The TCR is a critical cardiovascular event found in several surgical procedures involving a structure innervated by the trigeminal nerve. TCR occurs during treatment of endovascular embolization, which results in an intense autonomic disturbance of the heart that manifests as the sudden onset of bradycardia, hypotension, arrhythmias and even heart arrest. Sudden hemodynamic uctuations may affect the course of surgery and patient prognosis. Therefore, TCR and related risk factors have gained much interest during recent years. In this study, we have analyzed, the relation between DAVF and TCR occurrence highlight a strong association. We also found a higher pooled risk for embolization vessel supplied by the middle meningeal artery for hypotension and occurrence of asystole during TCR, suggesting a more severe re ex variant under these conditions.
The physiological mechanism of TCR is that signals triggered by either peripheral or central stimulation on the sensory nerve endings of the trigeminal nerve are relayed to the sensory nucleus of the trigeminal nerve. This afferent pathway continues along the short internuncial nerve bers in the reticular formation to connect with the efferent pathway in the motor nucleus of vagus nerve and nucleus ambiguus. The bers of vagus or sympathetic nerves end in the myocardium, leading to autonomic changes that usually manifests as a negative chronotropy [12]. However, there are two major subtypes of peripheral or central TCR. Previous studies have suggested that peripheral stimulation of the trigeminal nerve co-actives vagal and sympathetic nerves, resulting in both hypertension (peripheral vasoconstriction) and bradycardia [13]. By contrast, central stimulation may cause hypotension and bradycardia by generating profound activation of the cardiac vagal branch and speci c inhibition of the inferior cardiac sympathetic nerve [14].
In this study, we observed that all the TCR patients presented with heart rate deceleration > 20%, or even cardiac arrest, but just 13 patients experienced blood pressure reduction. On the contrary, 14 patients presented with blood pressure increase. Therefore, we suggest that central stimulant from chemical agents and distinct endovascular embolization might share the common efferent pathway for vagus activation, but the induce of inhibitory or promotive effect on efferent sympathetic pathway may coexistence in the process of central TCR, which depends on the site of stimulation. Hypertension may be associated with a greater susceptibility of sympathetic nerve stimulation, while the occurrence of hypotension is partly in a heart-rate-dependent manner, with signi cant blood pressure reduction when the heart rate is too slow. The low incidence of hypotension in this study may be related to our team's close monitoring of intraoperative hemodynamic changes, which usually prevented the continued deterioration of TCR.
In this cohort, we found that in the TCR group, the patients with protopathy of DAVF accounted for 77.8%, and the patients whose supplying vessel was the middle meningeal artery accounted for 70.4%, which was signi cantly higher than that in the control group. Therefore, we suspected that the patients with DAVF or middle meningeal artery embolization are the risk factors for TCR episode during DMSO or Onyx injection. These results are consistent with the previous case reports that TCR occurs mainly in patients with DAVF during cerebrovascular interventional embolization operation.
DAVFs are abnormal connections between dural arteries and venous sinuses. The dura mater is innervated in part by branches of the trigeminal nerve and receives vascular supply from the meningeal artery as well as meningeal branches of the occipital artery [15]. These vessels are primarily involved in the blood supply to the sensory area of the trigeminal nerve. Therefore, when the meningeal artery receives mechanical or chemical stimulation, trigeminal nerve excitability increases, leading to the episode of TCR. To some extent, these results support the anatomically speci c triggering of TCR. Anesthesiologists and neurosurgeons should be alert to the occurrence of TCR during DAVF embolization.
Studies have con rmed that Onyx has no effect on hemodynamics by itself [16]. However, mechanical stimulation of the trigeminal nerve during DMSO or Onyx injection is an essential inducement of TCR, and this effect is closely related to the dose and rate of injection [17]. Experts recommend that the injection rate of both DMSO and Onyx should not exceed 0.1 ml/min [18]. In this study, all procedures were performed by the same experienced neurosurgeon. Therefore, it seems that we can ignore the effect of injection speed on this study.
The occurrence of TCR is challenging to predict, and there is no precise detection index. Currently, preventive measures for TCR mainly include the use of anticholinergic drugs to reduce vagal excitability and increase the heart rate [19]. However, experts considered that preemptive administration of anticholinergics for the prevention of TCR may be ineffective. It has been observed that bradycardia and hypotension in TCR includes both excessive activation of vagal and inhibition of adrenergic vasoconstriction after electrical stimulation of the spinal trigeminal tract and trigeminal nuclear complex [20].
Atropine can only block the cholinergic bers, yet cannot completely prevent bradycardia or hypotension.
In this study, we also found that some patients often showed MAP increasing during embolization treatment, and atropine may increase the risk of hypertension while increasing the heart rate. In addition, atropine is at risk of refractory arrhythmia, so prophylactic use is not recommended [21]. The results of this study showed that although there was no signi cant difference in HR between the TCR group and control group, preoperatively and before embolization, however, multivariate logistic regression analysis showed that bradycardia before embolization was a risk factor for TCR. Therefore, it may be necessary to prophylactically correct bradycardia before embolization in high-risk patients.
In view of the large number of similar TCR events, the recognition and treatment of TCR has become increasingly adequate. The surgeon should rst be informed to discontinue the procedure once TCR has occurred, resulting in a deterioration in hemodynamics. Withdrawal of stimulation usually reverses this phenomenon immediately, without the need for anticholinergic drugs. For refractory bradycardia and hypotension, anticholinergic agents such as epinephrine or atropine should be initiated. Furthermore, we need to reassess and increase the depth of anesthesia. A meta-analysis [22] showed that lighter anesthesia (CSI > 60) were associated with a 1.2-fold increased incidence of bradycardia and a 4.5-fold increased risk of cardiac arrest than deeper anesthesia (CSI < 40). In this study, all patients had moderate anesthesia depth, and there was no signi cant difference in BIS value between the two groups. During operation 4 patients occurred cardiac arrest which resumed spontaneous circulation after discontinuation of the procedure and administration of atropine or a small dose of epinephrine. In addition, 6 patients with recurrent HR < 40 bpm were given atropine 0.5mg, and the TCR was successfully corrected. However, the incidence of postoperative adverse events were similar between the two groups, which may be mainly attributed to the in-depth understanding of TCR and timely treatment.
The limitation of this study is being a single-centered study with a relatively limited sample size and potential bias in patient selection. Meanwhile, this study is a retrospective study, in which only BIS values were used as the evaluation index for anesthesia depth. It is di cult to track the real-time blood concentration of various anesthetics, and it is impossible to exclude the in uence of drugs on TCR episode.
In summary, the TCR episode may due to chemical stimulus of DMSO and Onyx injection, leading to a signi cant decrease in HR under a standardized anesthetic protocol. This study con rms that patients with DAVF, CCF and middle meningeal artery embolism are the risk factors for TCR during Onyx endovascular embolization. Therefore, anesthesiologists and neurosurgeons should strengthen the identi cation of TCR risk factors, and strive to achieve early prevention and treatment to reduce the incidence of TCR-related adverse events.  Figure 1 Study Flow Chart.