Ten out of 100,000 people between the ages of 50 and 60 are usually reported to have partial facial spasms (HFS) [1-4]. The symptoms usually progress in frequency and severity, starting with intermittent twitches in the orbicularis oculi muscle, and spreading downward to the ipsilateral facial muscles. The most common cause of HFS is neurovascular compression along the root exit zone (REZ) of the facial nerve and the most common offending vessel is the anterior inferior cerebellar artery or posterior inferior cerebellar artery [5]. HFS is also caused by secondary etiologies such as tumors or vascular malformations [6,7]. Medical treatment of HFS includes administration of anticonvulsants (carbamazepine, clonazepam, gabapentin, etc.), anticholinergic drugs, antipsychotics, or botulinum toxin injections [8-13]. However, medical treatment is increasingly less effective and side effects occur as the dose escalation. Microvascular decompression (MVD) is indicated after unsatisfactory or decreasing effect of medical treatment or becomes intolerable to side effects. In most series, the percentage of patients with total relief ranged between 85% and 90%. In Samii et al.’s reports [14], the rate of symptom resolution following MVD has been reported to be over 92% at 6 months with 90% of patients remaining spasm free at 1 year. Common complications include hearing loss (7-26%), facial nerve injury (2.8-8.3%), and cerebrospinal fluid (CSF) leak (2-3%) [15-20].

We report the surgical outcomes and complications of MVD for HFS in our institute’s last 23 years.

Two hundred forty-four patients who had got MVD between June 1998 and July 2021 were reviewed retrospectively. All patients followed up for more than 24 months. Preoperative brain magnetic resonance imaging (MRI) and magnetic resonance angiography (MRA) were performed in all patients, and constructive interference in steady state (CISS) images were taken after 2009. Institutional Review Board approval and informed consent of the patients were exempted because this study was conducted retrospectively using existing data.

Of the 244 patients, female patients were 160, and male patients were 84. The average age was 53.8 years old (19-78 years old). The average duration of symptoms was 4 years (1 month-20 years). The affected side was left 141 (female:male=82:59) and right 103 (female:male=78:25). Offending vasculatures were anterior inferior cerebellar artery (AICA) 150 (61.5%), posterior inferior cerebellar artery (PICA) 37 (15.2%), vertebral artery (VA) 16 (6.6%), basilar artery (BA) two (0.8%), vein eight (3.3%), AICA+VA eight (3.3%), AICA+BA four (1.6%), AICA+vein two (0.8%), PICA+VA 15 (6.1%), PICA+BA one (0.4%) and VA+vein one (0.4%) (Table 1). Two hundred twenty-six patients (92.6%) completely recovered from HFS, two patients (0.8%) got reoperation, and complications occurred in 16 patients (6.6%) (Table 2).

Total symptom free rate was 92.6%. Before the IOM period (1998.06-2012.06), 96.7% of 150 patients were freed from HFS, and after the IOM period (2012.07-2021.07), 86.2% of 94 patients were freed from HFS (Table 2).

In 61 patients with preoperative facial palsy (female:male=40:21, average age: 55.1 years old, average duration of HFS: 5 years, left:right=36:25) on the affected side, facial palsy was improved to normal in 56 patients (91.8%) and five patients (8.2%) showed slight improvement from House-Brackman Grade (HBG) 3 to 2. There were no patients with worsening paralysis symptoms. Almost patients showed HBG 3. The spasm-free rate of patients with preoperative facial palsy were similar those without preoperative facial palsy (95% versus 92.8%). A peculiarity of the surgical field compared to patients without facial palsy is that the arachnoid thickened more often in patients with facial palsy (12 cases [19.7%] in 61 patients versus four cases [2.2%] in 183 patients, chi-square test: odds ratio 22.16) (Fig. 1). The HFS duration of patients with facial palsy was slightly longer than that of patients without facial palsy (5 years compared to 4.2 years).

Out of 16 complications, wound dehiscence was in three (1.2%), postoperative facial paralysis in five (2.0%), permanent hearing loss four (1.6%), transient hearing impairment in one (0.4%), tinnitus in one (0.4%), hoarseness plus tinnitus plus hearing impairment in one (0.4%), and one patient (0.4%) died from subarachnoid hemorrhage (SAH) that occurred immediately after surgery. There was no CSF leak. Complications in pre-IOM period were wound dehiscence in one patient), facial palsy in two patients, and death in one patient. In post-IOM period wound infection, hearing impairment, facial palsy, tinnitus, and lower CN deficit (hoarseness+tinnitus+hearing impairment) was two, five, three, one, and one, respectively (Table 2). More complications occurred in the post-IOM period compared to the pre-IOM period (chi-square test: odds ratio 6.39).

MVD has been established as a first-line surgical treatment for patients with HFS and has been reported to provide relief from spasms in over 90% of cases [5,21,22]. However, although MVD is a reliable treatment method, general anesthesia and the difficulty of handling the area around the brainstem are still reasons for reluctance to perform surgery. We aimed to investigate the results of MVD for HFS, and to evaluate the outcome and morbidity of this treatment. We retrospectively studied 244 patients who underwent retrosigmoid MVD last 23 years.

In our series, the female patients were 160 and male patients were 84. The proportion of female patients was about twice as high, and the average age of the patients was also in their mid-fifties as already known. The left and right incidence rates were the same in the entire patient group, but when the analysis was performed separately for female and male, the left side was more affected in male.

MRI is the gold standard for neurovascular conflicts diagnosis MRI, focusing on the facial nerve trajectory, was performed to diagnose and localize the offenders, and to exclude tumors or other vascular malformations. Improvement in imaging modalities since 1990 would result in overestimation of discordance. Recently, for better information, neurovascular conflicts are documented by MRI with three-dimensional constructive CISS, alone or together with three-dimensional time-of-flight MRA [23,24].

In about offending vessels the most common conflict vessels was AICA 150 (62.3%), followed by PICA 37 (15.2%), VA 16 (6.5%), BA two (0.8%), vein eight (3.3%), and AICA+VA eight (3.3%), AICA+BA 4 (1.6%), PICA+VA 15 (6.1%), PICA+BA 1 (0.4%) and VA+vein 1 (0.4%). Most of the vessels colliding with facial REZ were AICA and PICA.

We were interested in the changes in the surgical outcome and postoperative paralysis symptoms of 61 patients with facial paralysis before surgery. Almost patients showed HBD 3. A peculiarity of the surgical field compared to patients without facial palsy is that the arachnoid thickened more often in patients with facial palsy (12 cases [19.7%] in 61 patients versus four cases [2.2%] in 183 patients). We believe that the reason for the thick arachnoid membrane observed in patients with facial paralysis before surgery suggests the possibility of some previously unknown inflammation.

The overall rate of symptom-free without complications after MVD was 92.6%. This result is similar to previous reports. Out of 16 complications, wound dehiscence was in three (1.2%), postoperative facial paralysis in five (2.0%), permanent hearing loss four (1.6%), transient hearing impairment in one (0.4%), tinnitus in one (0.4%), hoarseness plus tinnitus plus hearing impairment in one (0.4%), and one patient (0.4%) died from SAH that occurred immediately after surgery. Our results also showed a complication rate similar to that previously known. The risk of permanent CN deficit was 1-2% for facial palsy, 2-3% for non-functional hearing loss, 0.5-1% for lower CN dysfunction. Risk of stroke was at 0.1% and mortality at 0.1%.

We analyzed the surgical results and complications before and after IOM. Before IOM, one serious complication, death, occurred, but the complication rate was 2.7%. After IOM, there were no serious complications, but the complication rate was 12.7%. It is thought that the introduction of the IOM has led to an increase in the incidence of complications even though the operator or surgical method has not changed. It is thought that the cause of the problem is that the vascular and nerve structures are further manipulated until the loss of the LSR is confirmed. It is particularly valuable when LSR induced before craniotomy disappears immediately after the culprit vessel is moved off the facial nerve; this is important clinically to help surgeons to confirm whether adequate decompression has been achieved [25,26]. Since Møller and Jannetta [27] documented that spasms are more likely to persist if LSR is still present at the end of the operation, LSR has been studied as an independent prognostic predictor of surgical outcome. Its value, however, it is still a matter of debate [28-34]. In Wei et al.’s report [35] in 2018 patients who underwent MVD with intraoperative LSR monitoring did not exhibit better clinical outcomes than those who underwent MVD without intraoperative LSR monitoring at the 1-week or 1-year follow-up examination. In most of their cases (all but 18), LSR monitoring did not play a guiding role during surgery. A meta-analysis conducted by Sekula et al. [36] indicated that the chance of a cure if the LSR disappeared during Teflon implantation was 4.2 times greater than if the LSR persisted. Hatem et al. [21] demonstrated that an excellent result is still likely to be obtained in patients with LSR persistence after successful decompression, and delayed spasm relief strongly supports the hypothesis that HFS is not only due to the neurovascular compression but also to severe nerve demyelination and/or hyperactivity of the facial nucleus. Although LSR exists in almost all patients with typical HFS, the exact mechanism of this phenomenon has not yet been well elucidated. It was presumed to be related to cross-transmission between axons at the lesion site because LSR disappeared immediately in most cases after the offending vessel was separated from the facial nerve [27,37,38]. However, LSR could also be due to hyperactivity of the facial nucleus [39]. As such, there are various causes of LSR, so manipulating blood vessels and nerves and inserting more Teflon until the LSR disappears in the surgical field can only increase the possibility of complications.

MVD has a durable effect on the improvement of HFS and also improve HFS and simultaneous palsy if preoperative facial palsy is present. When using an IOM, it is better to use it at a level that prevents or predicts complications rather than relying solely on LSR loss.