Clinical presentations and surgical outcomes of hemifacial spasm involving the vertebral artery

Objectives We aimed to determine whether vertebral artery (VA) affects the success rate of microvascular decompression(MVD) in hemifacial spasm(HFS) patients. We compared the clinical presentations and surgical outcomes between VA-involved and non-VA HFS patients. Methods The study comprised 313 HFS patients who underwent MVD between January 2015 and December 2018. There were 59 patients in the VA-involved group and 254 patients in the non-VA group. The clinical results mainly included the rate of HFS remission and the complications. Postoperative neurosurgery-related problems and neurological issues were among the complications.


Introduction
Hemifacial spasm (HFS) is a benign functional disease with an average incidence of about one in 100,000 people, and the number of female patients is about twice that of males [1,4]. The main manifestation of HFS is paroxysmal and involuntary twitching of the facial muscles innervated by the face nerve, which is not relieved by rest. Although Nerve Origin Hypothesis/ Peripheral Theory and Nuclear Origin Hypothesis/ Central Theory have been proposed, neurovascular compression remain the relatively recognized pathogenesis [7,8,37]. Anticonvulsants may be recommended as drug treatments in early stages of the disease, but they are often ineffective [2]. Botulinum toxin therapy is currently a recognized treatment method, although 20% of patients experience postoperative complications and symptom relief can last just a few months. Microvascular decompression (MVD) is thought to be the only treatment that can cure HFS. The overall success rate of MVD surgery has exceeded 90%, with complications occurring in less than 1%, due to the advancement of minimally invasive neurosurgery and nerve monitoring technology.

Page 3/11
The main purpose of MVD surgery is to use bio-soluble materials to adequately separate the problematic blood vessels from the facial nerve and relieve nerve compression. The anterior inferior cerebellar artery and the posterior inferior cerebellar artery are the most commonly reported blood vessels that cause HFS during surgery. The vertebral artery (VA) is relatively rare. However, because the vertebral arteries are often thick and tortuous, and part of the atherosclerosis develops, the procedure is obviously more di cult. Furthermore, injury to the small perforating arteries that nourish vital surrounding brain tissues may cause serious complications [13,23]. However, several studies have found no signi cant difference in clinical outcomes between VA-involved and non-VA HFS [15,32,35]. Therefore, this study used singlecenter data to retrospectively analyze the long-term effects and postoperative complications of patients with VA-involved HFS.

Study population and method
This study included 313 patients, with well-diagnosed primary HFS, who attended the West China

Patient management and follow-up
All patients were subjected to magnetic resonance imaging (MRI) before surgery, to ensure correct diagnosed. The MVD procedure of each patient was performed by the same surgeon under electrophysiological monitoring. The clinical results mainly included the remission degree of HFS after the operation: free spasm, partial remission, ineffective, delayed remission, and postoperative recurrence.
Complications mainly included postoperative neurosurgery-related problems and neurological issues.
Nerve complications included vertigo, facial palsy, motor or Sensory disturbances, whereas neurosurgical complications comprised cerebellar edema, CSF leak, intracranial infection, car infection, and fat liquefaction. The short-term clinical outcomes and complications were mainly obtained through clinical records, whereas the long-term outcomes and complications were mainly acquired through outpatient follow-up or telephone interviews. All 313 patients were followed-up for more than two year.

Statistical analyses
The continuous variables were reported as Mean±SD and analyzed using the Student's t-test. Patients with favorable or adverse outcomes were compared using χ2 or sher exact test for categorical variables and clinical data. All analyses were performed using the SPSS program (Version 25.0, IBM Corp., Armonk, New York, USA), which was also used to generate illustrations. Statistical signi cance was set at P < 0.05.

Discussion
It has been generally accepted that HFS is caused by a vascular compression of the facial nerve at the root exit zone (REZ) [2,12]. Vertebral artery (VA) involvement has previously been reported in 4.4−18.4 % of HFS patients [18,27,31]. In our case series, patients with VA and non-VA compressed the facial nerve in 59 (18.8%) and 254 (81.2%) cases, respectively. In the VA-involved group, indirect compression by the VA along with the small branches occurred more frequently than direct compression by the VA alone. When compared to non-VA HFS, we found female predominance and left-sided predominance in VA-associated HFS, and the difference was statistically signi cant. A similar outcome has been proven in extensive research [35]. The left-sided predominance may result from several factors, including the origin of the left VA directly from the aortic arch, a higher percentage of left-dominant VA, and higher ow velocity and volume in the left VA [32]. Compared with the mean duration of patients in the previous studies, the duration in this study is relatively longer, and there was no signi cant difference between the VA and non-VA group. This phenomenon may be attributed to the fact that most patients decline surgical treatment until their lives are signi cantly affected by the symptoms.

Results
The 313 HFS patients underwent MVD between January 2017 and December 2018. Detailed information on demographic features and baseline characteristics are shown in Table 1. There were 59 (67.4%) patients with VA-involved and 254 (32.6%) patients non-VAHFS.The mean age and duration in the VAinvolved group were 51 years and 63 months, respectively; 31 patients were female, and 41 cases were on the left side. In the non-VA group, 182 patients were female, and the mean age and duration were 50 years and 68 months, respectively, and 138 cases were on the left side. The female-male ratio was obviously higher in the VA-involved group (P=0.005). The difference in mean age and duration between the VAinvolved group and non-VA group was not statistically signi cant (P 0.05). However, the VA-involved HFS was more severe on the left side (P=0.034). The most frequently offending vessels in the VA-involved group were the VA with anterior inferior cerebellar artery (58.7%) and the anterior inferior cerebellar artery alone (64.3%) in the non-VA HFS.
All patients were followed up for at least two years. In the VA-involved group, 52 (88%) patients were spasem-free, 44 patients experienced immediate postoperative remission, and 10 patients had delayed remission which lasted from half a month to a year. Two patients relapsed during the follow-up period. In the non-VA group, 288 (92.8%) patients were spasem-free, 208 experienced immediate postoperative remission, and 34 patients had delayed remission. Two patients relapsed during the follow-up period. The rate of partial remission (8.5%) in the VA-involved group was signi cantly higher than in the non-VA group (P=0.014).
There was no discernible difference between the two groups (all P 0.05) in terms of neurosurgery-related complications and neurological issues. Furthermore, vital complications and mortality were not recorded among the patients.
Since Jenetta's rst vascularonever compression for hemifacial spasm (HFS) in 1966 [12], Microvascular decompression has been widely used in the treatment of vascular nerve syndrome, Safety and e cacy of MVD in HFS have also been veri ed, with a success rate ranging from 90.9-94.3% [17,33]. Patients with HFS are often in their 50s. Aging promotes atherosclerosis of vertebral arteries and makes their walls to become more rigid and brittle. Therefore, MVD is considered more di cult and is associated with a worse outcome when compared to MVD for HFS unrelated to the VA [34]. Several studies have investigated the effect of vertebral artery on HFS. The total spasem-free rate does not differ signi cantly between the VA involved and the non-VA groups which was in line with the results obtained in this study [10,21,22].
Meantime, we also found the delay and partial remissions were more frequent in the VA-associated group. The recurrence rate of VA involved groups was comparable to non-VA groups in our study. Some hypotheses have been proposed for the pathogeneses of delayed healing for HFS [5]. Demyelination of the nerve bers and hyperexcitability of facial motor nucleus are widely acknowdged and edema of the facial nerve may be another risk factors. According to the ndings in the operation, most of the vertebral arteries not only make facial nerve displacement but also cause obvious indentation. The facial nerve becomes demyelinated due to the long-term and continuous blood vessel compression. Even if the facial nerve is fully decompressed, it often takes a long time to repair the myelin sheath. It has been reported that delayed cure is associated with the slow reversal of the plastic changes in the facial nucleus [29]. According to the ndings in the second surgery, the causes for persistent or recurrent NVC were often associated with Te on adhesions and secondary granulomas [6]. Vasculature changes are also a frequent causative factor, especially elongated vertebral artery [3,28]. Therefore, VA-linked HFS may be a risk factor for delay and partial remissions. However, there is no signi cant difference in recurrence rate between the two groups in our study, it may be related to the insu cient follow-up period.
Although the incidence of permanent complications don't have difference, postoperative transient nerve complications in HFS with VA were more common than HFS attributed to small vascular compression [10]. The MVD procedures were mainly comlpleted through the narrow space between vessels and nerves. Due to the large VA maybe obscured surgical eld, excessive retraction of the cerebellar occulus and overstretching of the cranial nerves sometimes were need to achieve su cient decompression, which increase the complications. Thus, MVD might has a higher surgical risk for VAinvolved HFS. Adequate evaluation of preoperative imaging may identify neurovascular contacts and reveal more accurate details of the neighbouring structures, and then help surgeon choose the appropriate operative approach [25]. The appropriate craniotomy, full release of arachnoid membrane and effective use of the natural space of the brain sulcus are often important for providing us with more space [19,30]. Meantime, endoscopic microvascular decompression provide an more clear eld of vision for surgeron and is a better alternative for the treatment of hemifacial spasm [36]. Some new surgeical techniques have also been reported to mobilize the VA safely [9,11,14,16,20,24]. However, these techniques are not always effective in all cases, particularly in cases with a high number of perforator blood vessels. Minor damages to the small perforator blood vessels that supply brainstem can have catastrophic repercussions. In this situation, intraoperative electrophysiological mornitoring may be an important and usefull auxiliary tool. Because intraoperative electrophysiological monitoring can provide the real-time response and help the surgeon to judge whether decompression is adequate and when to nish surgery.