User Tools

Site Tools


Microvascular decompression for hemifacial spasm

The definitive treatment for hemifacial spasm is microvascular decompression (MVD), which cures the disease in 85% to 95% of patients according to reported series. In expert hands, the MVD procedure can be done with relatively low morbidity.

A video demonstrates the surgical steps of a MVD at left facial REZ in a 41-year-old man who presented with typical hemifacial spasm on the left side due to VIIth nerve REZ compression by PICA. A classical retromastoid and infrafloccular approach was performed to avoid stretching of the VIIIth nerve and access the VIIth nerve ventro-caudally. The next step is insertion-along the brainstem, VII-VIIIth nerves REZ, and flocculus-of a plaque made of Teflon felt (Edward-type) which is semi-rigid, and by principle does not exert direct compression on the facial REZ, thus avoiding compression and/or transmission of pulsations on the VIIth nerve. The patient's postoperative period was uneventful and clinical outcome good 1).


The nerve function of Cranial Nerve VIII is at risk during microvascular decompression for hemifacial spasm. Intraoperative monitoring of brainstem auditory evoked potentials (BAEPs) can be a useful tool to decrease the danger of hearing loss.

Auditory function was studied before and after surgery in 143 consecutive patients who were operated on for hemifacial spasm by microvascular decompression of the intracranial portion of the facial nerve. The acoustic reflex was abnormal preoperatively in 41% of the patients, indicating that the vascular abnormalities that caused the hemifacial spasm also affected the auditory nerve. Three patients suffered a profound hearing loss in the ear on the operated side, and one lost hearing function totally. In addition, 24 patients had a moderate elevation in the pure-tone threshold at one or more octave frequencies. Of these, 16 patients experienced a hearing loss at only one frequency (8000 Hz), while eight had a threshold evaluation of no more than 20 dB in the speech frequency range (500, 1000, and 2000 Hz). Two patients were deaf on the side of the spasm before the operation. Three patients were not tested postoperatively, and one patient was tested only after surgery. Thus, in this series of 143 patients, only 2.8% suffered a significant hearing loss as a complication of facial nerve decompression to relieve hemifacial spasm 2).

Brainstem auditory evoked potential monitoring during microvascular decompression for hemifacial spasm

Case series


Endoscopic-assisted microvascular decompression (MVD) was performed in 42 patients with HFS with complicated local anatomy from Janurary 2008 to Janurary 2012 in the Department of Neurosurgery, Hospital Affiliated with Nanjing Medical University of Wuxi, Wuxi, China., in the event of a significant blind spot, endoscopic exploration was performed with multi-angle 360-degree observation, including exploration of the brainstem facial nerve root exit zone (REZ) and exploration of the distal end of the nerve, and the relationships between blood vessels and nerves were carefully determined. After surgery, endoscopic examination was performed again to rule out vascular omissions, avascular excessive stretch, kinking, or formation of new compressions. The relevant data of all cases were retrospectively analyzed.

All patients were followed for 18-30 months, 41 patients had complete remission without recurrence (97.6%), 3 cases recovered to grade 0 from discharge grade I, 1 case of hearing loss was fully restored in 6 months, and 1 case of grade II was not significant increased to the end of follow-up.

Neuroendoscopy is an effective supplement to traditional MVD in treating HFS. In particular, in patients with complicated or abnormal local anatomy (for example small posterior fossa volume, abnormal fullness of the cerebellar flocculus, petrous bone block, local thickening of arachnoid adhesions, and unidentified offending vessels), neuroendoscopy can greatly improve the effectiveness of surgery 3).


A retrospective analysis of 370 patients who underwent microvascular decompression for HFS was performed. The patients were divided into four groups based on the offending arteries, namely anterior inferior cerebellar artery (AICA), posterior inferior cerebellar artery (PICA), vertebral artery, and multiple offending arteries. Affected side, age at onset, presence of hypertension, and sigmoid sinus area and dominance were compared between groups. The mean age of patients with a left HFS was significantly greater than that of patients with a right HFS (P=0.009). The AICA affected primarily the right side and PICA and multiple offending arteries the left side (P<0.001). Side of sigmoid sinus dominance was significantly different among groups (P<0.001). The offending arteries in HFS may be related to these differences. AICA was associated with right-sided symptoms, younger age at onset, and presence of left dominant sigmoid sinus, while PICA was associated with left-sided symptoms, older age at onset, and smaller right sigmoid sinus area 4).

Twenty-six patients underwent late redo MVD in our institution from January 1, 2011 to December 31, 2015. The clinical features, surgical findings, outcomes, and complications of the repeat MVD were analyzed retrospectively.

Twenty-four (92.3 %) patients were cured immediately after the redo MVD. Delayed relief was found in two (7.7 %) patients; it took 6 days and 2 weeks for them to obtain complete relief. No recurrence was found during follow-up. Surgical complications including three (11.5 %) facial paralysis and one (3.8 %) hearing loss.

We suggested that repeat MVD can be performed 2 years after the first MVD if the spasm was not resolved. Repeat MVD for HFS is effective 5).

Three hundred seventy-two patients received microvascular decompression (MVD) under intraoperative electrophysiological monitoring in Nanjing Drum Tower Hospital in 2014; the characteristic AMR of HFS was observed in 359 patients during the operation. And the 359 patients were divided into two groups based on whether AMR had remained before the beginning of the decompression procedure for offending vessels. Thirty-three patients who showed a permanent disappearance of AMR before the beginning of decompression were regarded as group I. Dural opening and the succeeding CSF drainage produced a permanent disappearance of AMR in 13. During the dissection of lateral cerebellomedullary cistern, a permanent disappearance of AMR was found in 20 patients. Thirty-two patients were cured immediately; delayed resolution (7 days after surgery) was found in one patient. No complications were observed and no recurrence was found during the follow-up period in the 33 patients. In the other 326 patients (group II), AMR disappeared temporarily before the beginning of the decompression procedure for offending vessels in 42 patients. After decompression, AMR disappeared completely in 305 patients. Two hundred sixty-seven patients were cured immediately and 57 patients got a delayed resolution (2 days to 45 weeks after surgery). The two left did not get a complete abolition of spasm. Three cases of hearing loss, one hoarseness, and nine delayed facial paralysis were observed. The reason of early abnormal muscle response disappearance may be that the degree of neurovascular compression was not serious; these patients were more likely to get an immediate cure. Continuous intraoperative electrophysiological monitoring of AMR is necessary 6).

Between June 2005 and May 2014, 13 patients with facial hemispasm were operated, underwent microvascular decompression. The age, sex, duration of symptoms before surgery, and surgical finds, were all evaluated. In addition, postoperative results were also analyzed.

Seven patients were women and 6 were men. The average age of the patients was 53 years. The average time between onset of symptoms and surgery ranged from 3 to 9 years. In all cases the facial hemispasm was typical, one with concomitant trigeminal neuralgia, observed in all neurovascular compression intraoperative. In decreasing order of frequency, the cause of compression was anterior inferior cerebellar artery, posterior inferior cerebellar artery, dolicomega basilar artery and dolicomega vertebral artery. The average time of postoperative follow-up after the surgery was 24 months. Complete relief from spasm occurred in 62%; 30% disappearance after 3 weeks-2 months (8% partial) and in 8% had no improvement. Regarding postoperative complications: 3 patients had facial paresis II-III in House-Brackman scale and 1 patient presented CSF leak. None of the patients in the serie had hearing loss or deafness.

The microvascular decompression for facial hemispasm is a safe an effective procedure, which allows complete resolution of the disease in most cases 7).

42 patients included in the analysis consisted of 22 females and 20 males, with an average follow-up duration of 76 months (range 24-132 months). Intraoperative investigation revealed that an artery other than the VA was responsible for the neurovascular compression in all cases : posterior inferior cerebellar artery (PICA) in 23 patients (54.7%) and anterior inferior cerebellar artery (AICA) in 11 patients (26.2%). All patients became symptom-free after MVD. Neither recurrence nor postoperative neurological deficit was noted during the 2-year follow-up, except in one patient who developed permanent deafness. Cerebrospinal fluid (CSF) leak occurred in three patients, and one required dural repair.

Transposition of the VA using a bioglue-coated Teflon sling retraction technique is a safe and effective surgical technique for HFS involving the VA. A future prospective study to compare clinical outcomes between groups with and without use of this novel technique is required 8).

Joo et al., performed a preliminary study of 13 patients with HFS in 2010. They increased the stimulation rate from 10.1 Hz/sec to 100.1 Hz/sec by 10-Hz increments, and they elevated the average time from 100 times to 1000 times by 100-unit increments at a fixed stimulus rate of 43.9 Hz. After defining the optimal stimulation rate and the number of trials that needed to be averaged for IOM of BAEPs, they also identified the useful warning criteria for this protocol for MVD surgery. From January to December 2013, 254 patients with HFS underwent MVD surgery following the new IOM of BAEPs protocol. Pure-tone audiometry and speech discrimination scoring were performed before surgery and 1 week after surgery. To evaluate the usefulness of the new protocol, the authors compared the incidence of postoperative hearing impairment with the results from the group that underwent MVD surgery prior to the new protocol.

Through a preliminary study, the authors confirmed that it was possible to obtain a reliable wave when using a stimulation rate of 43.9 Hz/sec and averaging 400 trials. Only a Wave V amplitude loss > 50% was useful as a warning criterion when using the new protocol. A reliable BAEP could be obtained in approximately 9.1 seconds. When the new protocol was used, 2 patients (0.8%) showed no recovery of Wave V amplitude loss > 50%, and only 1 of those 2 patients (0.39%) ultimately had postoperative hearing impairment. When compared with the outcomes in the pre-protocol group, hearing impairment incidence decreased significantly among patients who underwent surgery with the new protocol (0.39% vs 4.02%, p = 0.002). There were no significant differences between the 2 surgery groups regarding other complications, including facial palsy, sixth cranial nerve palsy, and vocal cord palsy.

There was a significant decrease in postoperative hearing impairment after MVD for HFS when the new protocol for IOM of BAEPs was used. Real-time IOM of BAEPs, which can obtain a reliable BAEP in less than 10 seconds, is a successful new procedure for preventing hearing impairment during MVD surgery for HFS 9).


A retrospective study evaluated the length of cerebellar retraction and the changes of intraoperative brainstem auditory evoked potential (BAEP) during microvascular decompression (MVD), and assessed the predictive value of the hearing loss as a prognostic indicator for the treatment outcome of hemifacial spasm (HFS).

This series included 1,518 consecutive patients affected with HFS who underwent MVD, during which BAEP was monitored. Patients were divided into two groups based on whether hearing loss occurred following decompression or not. Each patient underwent a similar procedure performed by one neurosurgeon. The two patients groups were matched with regard to sex, age, and degree of spasm. RESULTS:

Among the 1,518 patients, 106 (6.98 %) displayed functional hearing changes. Hearing loss was permanent in 12 patients (0.79 %). Of the 1,412 patients with stationary hearing compared with preoperative audiometry, 96 patients were selected who were individually matched with respect to sex, age, and degree of spasm. BAEP changed immediately after cerebellar retraction in 7 of 12 hearing-loss patients, suggesting the importance of retraction on hearing outcomes. The distance from the cerebellar surface of the petrous temporal bone to the neurovascular compression point was measured. The median distance of cerebellar retraction in the hearing-loss group was 13.77 mm, which was longer than the median distance in the control group.

Preoperative measurement of the cerebellar retraction distance can be a valuable clue to predict and prevent postoperative hearing loss in MVD for HFS 10).


84 consecutive patients affected with hemifacial spasm who underwent microvascular decompression during which BAEPs were monitored. During surgery, Wave I, I to V interpeak interval, latency, and amplitude of Wave V were recorded and measured. Auditory function was studied before and after surgery and expressed as a pure tone average in all patients. Then, correlations were made between hearing impairment after surgery and intraoperative BAEP changes in an attempt to define warning values.

Seventy-four patients (88%) had no hearing loss after surgery (Group 1). Eight patients (9.5%) had hearing impairment with a decrease in pure tone average of more than 20 dB (Group 2). Two patients (2.3%) experienced a definitive and complete hearing loss on the side operated on (Group 3). Among intraoperative BAEP changes, latency of Peak V was the most frequently observed and the most significant phenomenon, especially during cerebellar retraction and the decompression step of the microvascular decompression procedure. In the group of patients without hearing loss (Group 1), the mean delay in latency of Peak V was 0.61 millisecond (standard deviation, +/-0.36 ms); in the group with hearing decrease (Group 2), the mean delay was 1.05 milliseconds (standard deviation, +/-0.64 ms); and in the group with deafness (Group 3), Wave V was abolished.

From a practical standpoint, three warning values, based on delay in latency of Peak V, were established for use during surgery: an initial one at 0.4 millisecond (“watching” signal) at the safety limit; a second one at 0.6 millisecond (risk “warning” signal), which is the mean value corresponding to the group of patients without postoperative hearing loss; and an ultimate one at 1 millisecond (“critical” warning), before irreversibility. These warnings should help the surgeon to avoid or correct maneuvers that are dangerous for hearing function, which is mandatory in functional surgery 11).

Case reports

Takeda et al. report the successful treatment of a patient with hemifacial spasm due to a tortuous vertebral artery that appeared to have developed to compensate for agenesis of the ipsilateral carotid artery. The 51-year-old man presented with a 1-year history of progressive left hemifacial spasm. His medical history was otherwise unremarkable except for untreated mild hypertension. Magnetic resonance angiography and bone window computed tomography demonstrated congenital agenesis of the left carotid artery and compression of the root exit zone of the left facial nerve by a tortuous left vertebral artery (VA). Microvascular decompression was performed via a left suboccipital craniotomy, and the offending vessel was identified using endoscopy. The vertebral artery was successfully transposed using polytetrafluoroethylene (PTFE) tape and a PTFE ball (Bard PTFE felt, Tempe, Arizona). This is the first report of a patient with hemifacial spasm caused by an ectatic VA associated with agenesis of the ipsilateral carotid artery 12).

A 61-year-old female presented with 4 years history of left-sided hemifacial spasm. Head MRI and angiography indicated left vertebral artery dissecting aneurysm which compressed ipsilateral cranial nerves Ⅶ and Ⅷ. Microvascular decompression was performed. The dissecting aneurysm was pushed apart and the distal part of the parent artery was adhered to the dura on the petrosum. The compressed nerves were totally decompressed. The symptom of facial spasm was completely resolved immediately after surgery and did not recur during 6 months of follow up 13).

Sindou M, Esqueda-Liquidano M, Brinzeu A. Microvascular Decompression for Hemifacial Spasm. Neurosurgery. 2014 Sep 24. [Epub ahead of print] PubMed PMID: 25255262.
Møller MB, Møller AR. Loss of auditory function in microvascular decompression for hemifacial spasm. Results in 143 consecutive cases. J Neurosurg. 1985 Jul;63(1):17-20. PubMed PMID: 4009269.
Zhi M, Lu XJ, Wang Q, Li B. Application of neuroendoscopy in the surgical treatment of complicated hemifacial spasm. Neurosciences (Riyadh). 2017 Jan;22(1):25-30. doi: 10.17712/nsj.2017.1.20150567. PubMed PMID: 28064327.
Chung M, Han I, Chung SS, Huh R. Side predilections of offending arteries in hemifacial spasm. J Clin Neurosci. 2016 Jul;29:106-10. doi: 10.1016/j.jocn.2015.10.041. PubMed PMID: 26898581.
Jiang C, Xu W, Dai Y, Lu T, Jin W, Liang W. Failed microvascular decompression surgery for hemifacial spasm: a retrospective clinical study of reoperations. Acta Neurochir (Wien). 2016 Nov 5. [Epub ahead of print] PubMed PMID: 27817006.
Jiang C, Xu W, Dai Y, Lu T, Jin W, Liang W. Early permanent disappearance of abnormal muscle response during microvascular decompression for hemifacial spasm: a retrospective clinical study. Neurosurg Rev. 2016 Dec 15. [Epub ahead of print] PubMed PMID: 27981401.
Campero A, Herreros IC, Barrenechea I, Andjel G, Ajler P, Rhoton A. [Microvascular decompression in hemifacial spasm: 13 cases report and review of the literature]. Surg Neurol Int. 2016 Apr 1;7(Suppl 8):S201-7. doi: 10.4103/2152-7806.179545. Spanish. PubMed PMID: 27127708; PubMed Central PMCID: PMC4828948.
Lee SH, Park JS, Ahn YH. Bioglue-Coated Teflon Sling Technique in Microvascular Decompression for Hemifacial Spasm Involving the Vertebral Artery. J Korean Neurosurg Soc. 2016 Sep;59(5):505-11. doi: 10.3340/jkns.2016.59.5.505. PubMed PMID: 27651870; PubMed Central PMCID: PMC5028612.
Joo BE, Park SK, Cho KR, Kong DS, Seo DW, Park K. Real-time intraoperative monitoring of brainstem auditory evoked potentials during microvascular decompression for hemifacial spasm. J Neurosurg. 2016 Nov;125(5):1061-1067. PubMed PMID: 26824371.
Lee MH, Lee HS, Jee TK, Jo KI, Kong DS, Lee JA, Park K. Cerebellar retraction and hearing loss after microvascular decompression for hemifacial spasm. Acta Neurochir (Wien). 2015 Feb;157(2):337-43. doi: 10.1007/s00701-014-2301-8. Epub 2014 Dec 18. PubMed PMID: 25514867.
Polo G, Fischer C, Sindou MP, Marneffe V. Brainstem auditory evoked potential monitoring during microvascular decompression for hemifacial spasm: intraoperative brainstem auditory evoked potential changes and warning values to prevent hearing loss–prospective study in a consecutive series of 84 patients. Neurosurgery. 2004 Jan;54(1):97-104; discussion 104-6. PubMed PMID: 14683545.
Takeda R, Ookawara M, Fushihara G, Kobayashi M, Fujimaki T. Successful Treatment of Hemifacial Spasm Caused by an Ectatic Vertebral Artery Accompanying Agenesis of the Carotid Artery. Surg J (N Y). 2016 Sep 22;2(3):e105-e107. doi: 10.1055/s-0036-1593447. eCollection 2016 Jul. PubMed PMID: 28825001; PubMed Central PMCID: PMC5553477.
Ou C, Wang S, Chen Y, Mo J, Zhao X. [Microvascular decompression for hemifacial spasm induced by vertebral artery dissecting aneurysm: one case report]. Zhejiang Da Xue Xue Bao Yi Xue Ban. 2016 May 25;45(5):536-539. Chinese. PubMed PMID: 28087915.
microvascular_decompression_for_hemifacial_spasm.txt · Last modified: 2018/07/18 18:02 by administrador