Available online at www.sciencedirect.

com

Surgical Neurology 69 (2008) 153 – 157

www.surgicalneurology-online.com
Hemifacial Spasm
Microvascular decompression for hemifacial spasm:
analyses of operative complications in 1582 consecutive patients
Ryoong Huh, MD a,1 , In Bo Han, MD a,1 , Ji Young Moon, MD a ,
Jin Woo Chang, MD b , Sang Sup Chung, MD a,⁎
a
Department of Neurosurgery, College of Medicine, Pochun CHA University, Bundang-gu, Sungnam 463-712, South Korea
b
Department of Neurosurgery, Yonsei University College of Medicine, Seoul, South Korea
Received 28 January 2007; accepted 9 July 2007

Abstract Background: Microvascular decompression is the most reliable treatment for HFS, but it may cause
complications. The aim was to identify factors affecting the prognosis after MVD and to establish
appropriate means to reduce complications.
Method: We retrospectively reviewed 1524 patients with HFS who underwent MVD and were
followed for more than 6 months since January 1987. The mean follow-up duration was 30.9 months
(6-197 months).
Results: The effect of MVD was satisfying (excellent or good) in 94.6% (n = 1442). The failure and
recurrence rates were 2.1% (n = 32) and 0.4% (n = 6), respectively. Postoperative complications were
noted in 545 (35.8%) patients. Among them, facial palsy, hearing deficit, and low cranial nerve
palsies were found in 18.6% (n = 283), 7.2% (n = 109), and 2.8% (n = 43), respectively. However,
permanent facial weakness, hearing deficit, and lower cranial nerve palsies such as hoarseness and
dysphagia were encountered in 1.2% (n = 18), 2.1% (n = 32), and 0.1% (n = 2), respectively. The
more immediate and severe the facial palsy was, the more permanent it remained, with statistical
significance (P b .05). There was a trend that the more immediate and severe the hearing deficit was,
the more permanent the deficit remained, without statistical significance (P = .673).
Conclusion: Early (occurrence within 24 hours after operation) and severe cranial nerve
deficits, including facial, hearing, and lower cranial nerve deficits after MVD, entail the risk to
stay permanent.
© 2008 Elsevier Inc. All rights reserved.
Keywords: Facial palsy; Hearing deficit; Hemifacial spasm; Microvascular decompression

1. Introduction decompression is a highly accepted and effective method for

Hemifacial spasm is a disorder characterized by twitching
and spasm of the muscles innervated by the facial nerve. It is
generally believed that HFS is caused by vascular compres-
sion of the root exit zone of the facial nerve. Microvascular
Abbreviations: BAEP, brainstem auditory evoked potential; CSF,
cerebrospinal fluid; EMG, electromyography; HFS, hemifacial spasm;
MVD, microvascular decompression.
⁎ Corresponding author. Tel.: +82 31 780 5260; fax: +82 31 780 5269.
E-mail address: nshanib@naver.com (S.S. Chung).
1
These authors contributed equally to this study.
0090-3019/$ – see front matter © 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.surneu.2007.07.027
treatment of patients with HFS. It has been reported that
there is an 85% to 90% cure rate with MVD for HFS and
84% success at 10 years from surgery, with an operative
morbidity of less than 10% [4,9,17,26]. However, the results
of MVD surgery were not always satisfactory because of
incomplete cure, recurrence of symptoms, or surgical
complications. The risks of the operation are generally
related to the surgical approach and particularly to surgery on
the seventh nerve [15]. The reported complications after
MVD included facial palsy, hearing impairment, vocal cord
paralysis, intracranial hemorrhage, wound infection, menin-
gitis, vertebral artery injury, and CSF leakage. Among them,
154 R. Huh et al. / Surgical Neurology 69 (2008) 153–157

facial palsy and hearing deficit are not uncommon 1 week. In case of facial palsy, physical therapy was
complications resulting from disturbance adjacent to the performed to facial muscles starting from 5 to 7 days
facial nerve or auditory nerve [2,10,25,26]. Although most after operation.
cases of postoperative facial weakness and hearing impair- Descriptive statistics were performed with SPSS 10.0
ment are transient, permanent facial weakness and hearing software (SPSS, Chicago, Ill). Significance was assigned
impairment may develop. In this study, we evaluated the when probability was less than .05.
long-term follow-up results and postoperative complications
and tried to determine the prognostic factors related to cranial
3. Results
nerve deficits after MVD.
The mean age of patients was 49.2 years (range, 15-78
years), and the ratio of male to female patients was 1:3.6. The
2. Methods mean duration of symptoms was 7.9 years (range, 7 months-
40 years). There was no statistically significant difference in
A total of 1582 patients with HFS underwent MVD. A the laterality. The mean follow-up duration was 30.9 months
total of 1609 operations was performed by the senior (range, 2.4-196.8 months).
author (SS Chung) at the Severance Hospital, Yonsei Excellent surgical outcome was obtained in 90.3% and
University, Seoul, Korea (September 1978-February 2003), good in 4.3%. The overall success rate was 94.6%. The
and at the Bundang CHA Hospital, Pochon CHA failure rate of MVD was 2.1% (n = 32), and spasm
University, Sungnam, Korea (March 2003-September recurrence was noted in 0.4% (n = 6). Of the patients with
2005). Modern microsurgical technique has been per- persistent or recurrent HFS, 13 (0.9%) patients who had
formed since 1987 in our institutes [1,2]. We retro- compressing vessels identified on postoperative 3-dimen-
spectively reviewed 1524 patients who were followed for sional magnetic resonance angiography underwent repeated
more than 6 months since January 1987. Patients with MVD. Eleven had excellent results with the second MVD, 1
symptomatic HFS secondary to tumors or vascular had a good result, and 1 had a fair outcome.
malformation were excluded from this study. Preopera- The operative complications are summarized in Table 1.
tively, 3-dimensional magnetic resonance angiography Postoperative complications were noted in 35.8% (n = 545),
(3D = TOF MRA) has been used since 1992, and pure but permanent deficits were encountered in 3.6% (n = 55).
tone audiometry (PTA) and facial EMG were performed There was no operation-related mortality in this study.
before operation. Brainstem auditory evoked potential has Postoperative permanent hearing deficit was more
been monitored during operation since 1983, but facial common than facial weakness (2.1% vs 1.2%). The onset
EMG was not performed intraoperatively. The clinical of permanent facial weakness occurred within 24 hours in
results for each patient were assessed in a personal 88.9% (n = 16). Delayed facial weakness was noted in 94
interview or by telephone. All patients received a (6.2%) patients. The onset of delayed facial weakness
questionnaire addressing the persistence of any operative occurred between postoperative days 8 and 28 (mean, 13.1
complications. days). Among them, 92 (97.9%) patients improved to
The patients were classified into the following 5 grades, complete recovery, and the mean time to recovery was
on the basis of the degree of HFS present: (1) “excellent” if 8.6 weeks (range, 0.7-43.7 weeks). However, 2 patients
there was no residual spasm; (2) “good” if the HFS was more remained with permanent severe facial weakness. Immediate
than 90% resolved; (3) “fair” for spasm relief of 50% to 90%; facial weakness was encountered in 66.8% (n = 189). Of the
(4) “poor” if the HFS was less than 50% resolved; and (5) patients with immediate facial weakness, permanent facial
“failure” for all remaining results. weakness was found in 8.5% (n = 16). When analyzed on the
Among the postoperative complications, the “immedi-
ate” cranial nerve deficit was assigned to the occurrence of
deficit within 24 hours. The occurrence 24 hours after
Table 1
operation was defined as “delayed.” Facial palsies were The complications of MVD and their frequencies
classified into the following 3 grades: (1) “mild” for facial
Complications No. of patients (%)
palsy of House-Brackmann grade I or II; (2) “moderate”
for grade III; and (3) “severe” for grade IV or V. Hearing Transient (%) Permanent (%) Total
deficits were also classified into 4 grades, on the basis Facial palsy 265 (17.4) 18 (1.2) 283
of PTA: (1) “mild” for 26 to 40 dB hearing loss (HL); Hearing impairment 77 (5.1) 32 (2.1) 109
Hemorrhage 35 (2.3) 0 (0) 35
(2) “moderate” for 41 to 70 dB HL; (3) “severe” for
Lower CN palsy 41 (2.7) 2 (0.1) 43
71 to 90 dB HL; and 4) “deafness” for more than 91 dB Infection 38 (2.5) 3 (0.2) 41
HL. When cranial nerve dysfunction developed post- CSF leakage 17 (1.1) 0 (0) 17
operatively, steroid medication (dexamethasone, 20 mg/d) Others 17 (1.1) 0 (0) 17
was initiated, and this was gradually discontinued over Total 490 (32.2)* 55 (3.6)* 545*
2 weeks. Naloxone (6 mg/d), IV, was also administered for n = 1524. CN indicates cranial nerve.
 R. Huh et al. / Surgical Neurology 69 (2008) 153–157 155

basis of grade of facial palsy, of the patients with severe Table 3

facial weakness, immediate facial weakness was noted in Classification by the severity of hearing impairment
81.2% (n = 13), and permanent facial weakness was found in Postoperative day Before 24 h After 24 h Total (n)
61.5% (n = 8) of patients with immediate facial weakness. Mild Transient 43 6 49
In the moderate facial weakness group, permanent facial Permanent 6 0 6
weakness was noted in only 5 (9.6%) patients with Moderate Transient 11 3 14
Permanent 4 2 6
immediate moderate facial weakness. In the mild group,
Severe Transient 12 0 12
permanent facial weakness was found in only 3 (2.4%) Permanent 11 1 12
patients with immediate mild facial weakness. These Deaf Transient 0 0 0
indicated that permanent facial weakness frequently occurred Permanent 9 1 10
within 24 hours after operation, and the more severe the Total (n) 96 13 109
facial palsy was, the more permanent the facial palsy For total transient disorders, n = 75; total permanent disorders, n = 34; and
remained, with statistical significance (P b .05) (Table 2). total, n = 109. P = .673.
The mean time to recovery from transient facial palsy was
6 weeks (mean, 4 weeks; SD, 7.1 weeks). Most of the
patients recovered within 2 months. Among 4 patients who cranial nerve palsies was 26.6 weeks and was longer than
recovered 8 months after operation, 2 patients had mild that of facial palsy or hearing impairment.
facial palsy, and 2 patients had moderate facial palsy.
Of the 109 (7.2%) patients with hearing deficit,
4. Discussion
immediate hearing deficit was noted in 88.1% (n = 96),
but delayed occurrence was encountered in 13 patients. To The cranial nerve injuries during MVD may lead to
exclude temporary conduction hearing impairment related to postoperative facial weakness, hearing impairment, or lower
middle ear fluid rather than neurosensory loss, postoperative cranial nerve palsy such as dysphagia and hoarseness, and
PTA was performed. As shown in Table 3, permanent these complications can decrease the patient's subjective
hearing deficit occurred immediately in 88.2% (n = 30) of satisfaction despite the complete disappearance of facial
patients with permanent hearing deficit. Permanent hearing spasm [3-5,11,16-18,21,23].
deficit was encountered in 62.5% (n = 20) of patients in the The efficacy of MVD in treating HFS in our patients
immediate severe or deaf group. These suggested that the was similar to that reported in other published series,
onset of permanent hearing deficit usually occurred within especially concerning the high rate of long-term success,
24 hours after operation, and the more severe the hearing together with the low rates of recurrence and operative
deficit was, the more permanent the deficit remained, but complication [1,8,12,14,22,26]. There was no operation-
this trend did not reach statistical significance (P = .673) related death. Postoperative facial weakness was more
(Table 3). The mean time to recovery from transient common than hearing deficit. Of 109 patients with hearing
hearing impairment was 10 weeks (mean, 4 weeks; SD, deficit, 22 (20.2%) suffered deafness or severe permanent
13.9 weeks). This indicated that the recovery from damage loss of hearing, whereas only 3.5% (n = 10) of 283
to the auditory nerve after operation was slower than patients with facial weakness remained with severe
recovery from damage to the facial nerve. permanent facial weakness. The mean time to recovery
The lower cranial nerve (IX-X) deficits, such as from transient hearing impairment was longer than that of
hoarseness and dyspahgia, were encountered in 2.8% (n = facial palsy. This indicated that the time to recovery from
43). Among them, complete recovery was achieved in 95.3% auditory nerve dysfunction was slower than that of facial
(n = 41). Two patients remained with mild permanent palsy nerve dysfunction. We believe that this can be attributed to
(Table 1). The mean time to recovery from transient lower the fact that the motor nerves (facial nerves) are less
susceptible than the cochlear nerve to traction and/or
ischemic traumas.
The transient facial palsy improved within 8 months, and
Table 2 hearing impairment improved within 12 months. Therefore,
Relationship between the severities and the time of occurrence of facial palsy
1-year or longer follow-up monitoring is required before
Postoperative day Before 24 h After 24 h Total (n) determination of “permanence” with respect to cranial nerve
Mild Transient 121 60 181 dysfunction. When facial palsy developed within 24 hours
Permanent 3 0 3 after operation, and the initial grade of facial palsy was
Moderate Transient 47 31 78
severe, there was a high possibility of permanent facial palsy
Permanent 5 0 5
Severe Transient 5 1 6 (P b .05). Although there was no statistical significance (P =
Permanent 8 2 10 .673), permanent hearing impairment developed more
Total (n) 189 94 283 frequently in patients who had immediate and severe hearing
For total transient disorders, n = 265; total permanent disorders, n = 18; and impairment. Consequently, when cranial nerve palsies
total, n = 283. P b .05. immediately occurred, and the initial grade of palsies was
156 R. Huh et al. / Surgical Neurology 69 (2008) 153–157
severe, the prognosis was poor, and there was more frequent
occurrence of permanent palsy.
The pathophysiologic mechanism of postoperative facial
weakness is still unclear, particularly because it can occur
some days after operation. It has been suggested that a viral
cause could be involved. The manipulation of the nerve
could stimulate a dormant virus, possibly localized in the
geniculate ganglion. In addition, it has been reported that
possible causes include facial nerve exit zone injury via the
Teflon felt, delayed facial nerve edema, or a microcirculation
disturbance due to vasospasm. Despite the uncertain cause,
the facial weakness can resolve spontaneously, with
excellent outcomes [5,7,13,15,20,23,24].
The cranial nerve injuries during MVD might result from
direct nerve damage or traction injuries. In our series, to
avoid direct nerve damage, the best exposure for HFS has
been achieved from below upward, that is, “approaching the
ninth nerve to expose the area of compression at the seventh
exit zone,” since 1987, and the incidence of facial palsy and
hearing impairment significantly decreased below the level
of 5%, compared with the early period (between 1978 and
1986). Many neurosurgeons prefer to carry out MVD
without any fixed retractor, primarily using withdrawal of
CSF and gentle exposure of pertinent anatomy with their
handheld suction. In our patients, however, the retraction of
the cerebellum was performed. The eighth nerve has the
longest extracerebral central part of all the nerves in the
cerebellopontine angle and is highly vulnerable to traction.
Therefore, the retraction was done in a direction perpendi-
cular rather than longitudinal to the axis of the eighth nerve
to minimize risk of postoperative hearing impairment.
It has been reported that particular change in “lateral
spread reflex” and BAEP monitoring can be used to predict
postoperative function. However, the role of intraoperative
BAEP and facial EMG in preventing cranial nerve deficit is
uncertain [6,17,19,20,24,26]. We did not perform facial
EMG during operation, and BAEP has been monitored since
1983. The rate of hearing impairment after operation began
to decrease from 1987, at the end of learning period, and the
improvement in the late period could be due to the
improvement on the neurosurgeon's learning curve.
4.1. Conclusion
When cranial nerve palsies occur within 24 hours after
operation and when the initial grade of the cranial palsies is
severe, the prognosis is poor, and there is a high possibility
of permanent cranial nerve dysfunction. To determine the
permanence, a 2-year wait minimum would be best,
especially considering that lower cranial nerve palsy may
improve during this time span.
References
[1] Auger RG, Piepgras DG, Laws Jr ER. Hemifacial spasm: results of
microvascular decompression of the facial nerve in 54 patients. Mayo
Clin Proc 1986;61:640-4.
[2] Barker FG, Jannette PJ, Bissonette DJ. Microvascular decompression
for hemifacial spasm. J Neurosurg 1995;82:201-10.
[3] Chung SS, Chang JH, Park YG. Microvascular decompression for
hemifacial spasm: a long-term follow-up of 1,169 consecutive cases.
Stereotact Funct Neurosurg 2001;77:190-3.
[4] Elston JS. Botulinum toxin treatment of hemifacial spasm. J Neurol
Neurosurg Psychiatry 1986;49:827-9.
[5] Hanakita J, Kondo A. Serious complications of microvascular
decompression operations for trigeminal neuralgia and hemifacial
spasm. Neurosurgery 1998;22:348-52.
[6] Hatem J, Sindou M, Vial C. Intraoperative monitoring of facial EMG
responses during microvascular decompression for hemifacial spasm.
Prognostic value for long-term outcome: a study in a 33-patient series.
Br J Neurosurg 2001;15:496-9.
[7] Hengstman GJ, Gons RA, Menovsky T. Delayed cranial neuropathy
after neurosurgery caused by herpes simplex virus reactivation: report
of three cases. Surg Neurol 2005;64:67-70.
[8] Illingworth RD, Porter DG, Jakubowski J. Hemifacial spasm: a
prospective long term follow up of 83 cases treated by microvascular
decompression at two neurosurgical centres in the United Kingdom.
J Neurol Neurosurg Psychiatry 1992;60:72-7.
[9] Jannetta PJ, Kassam A. Hemifacial spasm. J Neurol Neurosurg
Psychiatry 1999;66:255-6.
[10] Jannette PJ. Typical or atypical hemifacial spasm. J Neurosurg 1998;
89:346-7.
[11] Kalkanis SN, Eskandar EN, Carter BS. Microvascular decompression
surgery in the United States, 1996 to 2000: mortality rates, morbidity
rates, and the effects of hospital and surgeon volumes. Neurosurgery
2003;52:1251-61.
[12] Kondo A. Follow-up results of microvascular decompression in trige-
minal neuralgia and hemifacial spasm. Neurosurgery 1997;40:46-52.
[13] Kuroki A, Itagaki S, Nagai O. Delayed facial palsy after microvascular
decompression for hemifacial spasm. Facial Nerve Res 1991;11:
147-50.
[14] Lovely TJ. Efficacy and complications of microvascular decompres-
sion: a review. Neurosurg Q 1998 [In press].
[15] Lovely TJ, Getch CC, Jannetta PJ. Delayed facial weakness after
microvascular decompression of cranial nerve VII. Surg Neurol 1998;
50:449-52.
[16] Lovely TJ, Lowry DW, Jannetta PJ. Functional outcome and the effect
of cranioplasty after retromastoid craniectomy for microvascular
decompression. Surg Neurol 1999;51:191-7.
[17] McLaughlin MR, Jannetta PJ, Clyde BL. Microvascular decompres-
sion of cranial nerves: lessons learned after 4400 operations.
J Neurosurg 1999;90:1-8.
[18] Moller AR. Vascular compression of cranial nerves II. Pathophysiol-
ogy. Neurol Res 1999;21:439-43.
[19] Moller MB, Moller AR. Loss of auditory function in microvascular
decompression for hemifacial spasm. Results in 143 consecutive cases.
J Neurosurg 1985;63:17-20.
[20] Polo G, Fischer C, Sindou MP. 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;54:97–106.
[21] Rhee DJ, Kong DS, Park K. Frequency and prognosis of delayed facial
palsy after microvascular decompression for hemifacial spasm. Acta
Neurochir (Wien) 2006;148:839-43.
[22] Roh SW, Chung SS, Park YG, Rhim SC, Lee KC. Significance of
brainstem auditory evoked potentials in posterior fossa surgery.
J Korean Neurosurg Soc 1987;16:1061-72.
[23] Samii M, Gunther T, Iaconetta. Microvascular decompression to treat
hemifacial spasm: long-term results for a consecutive series of 143
patients. Neurosurgery 2002;50:712-8.
[24] Sindou MP. Microvascular decompression for primary hemifacial
spasm. Importance of intraoperative neurophysiological monitoring.
Acta Neurochir (Wien) 2005;147:1019-26.
 R. Huh et al. / Surgical Neurology 69 (2008) 153–157
[25] Wang A, Jankovic J. Hemifacial spasm: clinical findings and treatment.
Muscle Nerve 1998;21:1740-8.
[26] Wilkins RH. Hemifacial spasm: a review. Surg Neurol 1991;36:
251-77.
Commentary
The authors analyzed surgical complications of MVD
surgery for HFS. The data plural is based on their 1582
patients and 1609 surgeries during the period from 1978 to
157
2003. Dr P J Jannetta introduced the MVD for HFS in the
Journal of Neurosurgery in 1977. In this current article,
the surgery was performed only by Dr S S Chung, and his
long experience is expressed in this article. His data will
serve as preoperative explanation for a patient. I would like
to congratulate him for his great effort and good results.
Masashi Fukui, MD
Department of Neurosurgery
Sasebo Kyosai Hospital
Sasebo 857-8575, Japan