Hemifacial spasm secondary to vascular compression of the facial nerve
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Hemifacial spasm secondary to
vascular compression of the
facial nerve
Idiopathic hemifacial spasm is an uncommon,
benign, but distressing condition characterized by
the insidious onset of paroxysmal, repetitive,
Daniel R. Neagoy, M.D.* clonic contractions of the facial musculature. T h e
Donald F. Dohn, M.D. condition usually begins with contractions of the
_ . , , , , . ,. orbicularis oculi and spreads first through con-
Department of Neurological Surgery r °
tiguous muscles to involve all the muscles in-
nervated by the seventh nerve on one side of the
face. 1
In a review of 106 cases of cryptogenic hemi-
facial spasm in 1945, Ehni and Woltman 1 pointed
out several characteristics of the condition. They
found that the disorder occurred only in adults
and that the mean age of onset was 45 years, al-
though recently hemifacial spasm has been re-
ported in an 8-year-old child. 2 Ehni and Woltman 1
observed a sex distribution of six women to four
men. Either side of the face was affected in either
sex with equal frequency. In six of their patients
the spasms were bilateral, but not synchronous
nor symmetric. T h e spasms consisted of intermit-
tent, irregular series of single muscle twitches oc-
* Fellow, Department of Neurologi- curring in rapid sequence. In 12 patients the
cal Surgery. spasms were observed in sleep.
205
Downloaded from www.ccjm.org on February 10, 2022. For personal use only. All other
uses require permission.206 Cleveland Clinic Quarterly Vol. 41, No. 3
Psychic upsets, fatigue, and volun- cause of idiopathic hemifacial spasm.
tary movement of the face were the Other authors have pointed to an
most common factors that either pre- association between posterior fossa
cipitated or increased the severity of neoplasms and hemifacial spasm. Re-
the spasms. None of the patients could villa 9 ' 1 0 observed spasms in patients
voluntarily stop the spasms, although with cerebellopontine angle neuri-
nine patients experienced periods of nomas, meningiomas, and cholestea-
spontaneous remission lasting from a tomas. Gardner and Sava 11 reported
few weeks to 3 years. In three patients the case of a patient with hemifacial
the spasms were accompanied by tri- spasm in whom the seventh nerve was
geminal neuralgia on the ipsilateral squeezed between petrous bone and
side, a condition referred to previously pons. T h e brainstem had been dis-
by Cushing 3 as "tic convulsif." placed, presumably by an expanding
T h e failure of the spasms to spread lesion on the opposite side. In 1966
beyond the distribution of the facial Gardner and Dohn 12 reported a case
nerve and to be affected by a capsular of bilateral hemifacial spasm in a pa-
infarct led Ehni and Woltman 1 to con- tient with basilar impression due to
clude that the lesion causing the dis- Paget's disease.
order must be located distal to the Some authors have emphasized that
corticobulbar pathway. Regeneration hemifacial spasm can be the result of
following division of the facial nerve compression of the seventh nerve by
at the stylomastoid foramen is usually vascular structures in the posterior
accompanied by recurrence of spasm. fossa. Campbell and Keedy 13 observed
Division of the nerve at this point fol- a cirsoid aneurysm of the basilar artery
lowed by anastomosis with another in two patients in whom hemifacial
cranial nerve is accompanied by rein- spasm was accompanied by trigeminal
nervation of the face, but not by re- neuralgia. Laine and Nayrac 14 de-
currence of spasm. 4 ' 5 These observa- scribed a similar lesion in a patient
tions led Ehni and Woltman to with hemifacial spasm. Of 19 patients
conclude that the lesion responsible in whom posterior fossa exploration
for the spasm must lie within the seg- was performed for hemifacial spasm,
ment of the nerve between the stylo- Gardner and Sava 11 found three in
mastoid foramen and the facial nu- whom the seventh nerve was com-
cleus. pressed by a cirsoid aneurysm of the
Although most agree that compres- basilar artery. In three other cases, the
sion of the nerve by some pathologic nerve was compressed by an arterio-
process causes the spasms, the nature venous malformation and in seven
of the pathologic process is variable. others by a redundant anterior inferior
Woltman et al, 6 Proud, 7 and Pulec, 8 cerebellar or auditory artery. Jan-
on decompressing the facial nerve in netta 15 reported eight cases of hemi-
its canal through the temporal bone, facial spasm in which he performed
observed edema and thickening of the microsurgical exploration and found
sheath surrounding this portion of the the facial nerve compressed by vascular
nerve. They concluded that infratem- structures in the cerebellopontine
poral pathology is the most common angle. Decompression was followed by
Downloaded from www.ccjm.org on February 10, 2022. For personal use only. All other
uses require permission.Winter 1974 Hemifacial spasm 207
relief of the spasms in each of his pa- hemifacial spasm which, in most, was
tients. reported as having begun in the or-
Gardner and Dohn 12 presented a bicularis oculi. It had spread and in-
vertebral angiogram which demon- volved all the facial muscles on one
strated a redundant loop of basilar side at the time of surgery. T h e left
artery projecting into the left cere- side was affected in eight patients and
bellopontine angle in a patient with the right side in six. T h e duration of
both left hemifacial spasm and left tic symptoms varied between 16 months
douloureux. T h e cerebellopontine and 20 years. Ipsilateral facial weak-
angle was not explored in this case, ness was reported in five of the pa-
however. More recently Eckman et al 16 tients. T h e spasms were accompanied
presented angiographic evidence of by tic douloureux on the ipsilateral
elongated and tortuous vertebral and side in one patient. In another, con-
basilar arteries invading the cerebello- stant dull pain was associated with the
pontine angle on the appropriate side spasms. Audiometry disclosed some de-
in four patients with hemifacial spasm. terioration in auditory acuity compati-
Similar aberrant vessels believed to be ble with neurologic disease in four pa-
the cause of hemifacial spasm have tients. T h e deficit was bilateral, al-
been demonstrated angiographically though most severe on the side of the
by others. 17-19 lesion in three and unilateral on the
We agree that idiopathic hemifacial same side as the hemifacial spasm in
spasm is most often caused by anoma- one. Additional abnormal neurologi-
lous vascular structures compressing cal findings in one patient included a
the seventh nerve in the cerebellopon- spastic, ataxic gait, positive Romberg
tine angle. In recent years 45 patients sign, and fasciculations in both thighs.
have undergone exploration of the
cerebellopontine angle for hemifacial Angiographic findings
spasm at the Cleveland Clinic. In 14 of Of the four vertebrobasilar angio-
these patients, the facial nerve was grams which demonstrated the aber-
found to be compressed by an elon- rant blood vessels preoperatively we
gated and tortuous vertebral or basilar have selected two (Figs. 1 and 2). T h e
artery. Preoperative vertebral angiog- changes observed in these are repre-
raphy in four of these 14 patients sentative of those seen also in the other
clearly demonstrated the abnormality two. They consist of elongation and an
in each. T h e purpose of this paper is anomalous curvature of the distal seg-
to analyze briefly these 14 cases and to ment of the vertebral and proximal
demonstrate the aberrant blood vessels basilar artery with a convexity directed
observed in two of the four abnormal toward the right cerebellopontine
angiograms. angle in case 1 (Fig. 1 A,B). In case 2
the distal segment of vertebral artery
Analysis of cases (see Table) is elongated and doubled over on it-
At the time of operation, the 14 pa- self, forming a loop which extends into
tients ranged in age from 44 to 71 the left cerebellopontine angle (Fig. 2
years. There were five men and nine A,B). T h e terminal segment of the
women. All had typical unilateral artery pursues a transverse course
Downloaded from www.ccjm.org on February 10, 2022. For personal use only. All other
uses require permission.Downloaded from www.ccjm.org on February 10, 2022. For personal use only. All other
uses require permission.u .5
0 >
"v> itV
S
»
'S a.
0
o
*4u> 'Sco Q.
o
^ 0 a
m0m
r>-
0 e
6? m I c
3
O
.2. E
V -=¡
-O 3 I-X O T! 1 O- S
O OV _~oV « ' o o
« 2 « oV r tbo1 ' a X
Û. o rt ni
s t S a o _ S
M o" S = w-> Oo
S Si
S S-
so g =3 .S
Downloaded from www.ccjm.org on February 10, 2022. For personal use only. All other
uses require permission.210 Cleveland Clinic Q u a r t e r l y V o l . 41, No. 3
disclosed an e l o n g a t e d a n d tortuous
vertebral o r basilar artery distorting
and compressing the seventh nerve in
all 14 patients. T h e o p e r a t i n g micro-
scope was utilized in four of the dissec-
tions. T h e o p e r a t i v e findings of the
Fig. 1. A, B, Case 1. Right vertebral angio-
gram, arterial phase, half-axial and lateral
projections. Elongation resulting in an arch
of vertebrobasilar artery projecting into the
right cerebellopontine angle (arrow).
toward the point of origin of the basi-
lar artery.
Fig. 2. A, B, Case 2. Left vertebral angio-
Operative findings gram, arterial phase, anteroposterior and lat-
eral projections. Redundant loop of vertebral
C e r e b e l l o p o n t i n e angle e x p l o r a t i o n artery extends into left cerebellopontine angle
through a suboccipital craniectomy (arrow).
Downloaded from www.ccjm.org on February 10, 2022. For personal use only. All other
uses require permission.W i n t e r 1974 H e m i f a c i a l spasm 211
four cases in which a n g i o g r a p h y was serted to separate the two structures
p e r f o r m e d will be discussed in detail. after a neurolysis was p e r f o r m e d on
In case 1 the seventh nerve was the nerve. T h e nervus intermedins was
found to be compressed on its antero- identified and divided.
medial aspect by an a n o m a l o u s verte- In the fourth case an elongated and
bral artery. B e n e a t h t h e nerve was seen tortuous internal auditory artery
an e x t r a o r d i n a r y a m o u n t of c h o r o i d coursed laterally between the facial
plexus. It was believed that by remov- a n d auditory nerves and gave rise to a
ing the choroid plexus, pressure on the small arterial plexus which overlay the
nerve f r o m the artery could be elimi- facial nerve as it entered the internal
nated. T h e choroid plexus was there- auditory meatus. Anteriorly, the sev-
fore partially removed by coagulation. e n t h nerve was contiguous with a dis-
T h e seventh nerve was believed also tended and tortuous segment of verte-
to be compressed by the anterior in- bral artery. T h e seventh nerve was dis-
ferior cerebellar artery. These two sected away from the eighth nerve and
structures were separated and a piece a neurolysis was p e r f o r m e d on the por-
of absorbable gelatin sponge (Gel- tion of seventh nerve p r o x i m a l to the
foam) was interposed between them. arterial plexus.
A segment of the superior aspect of the I n case 9 the offending lesion, an
seventh nerve t h o u g h t to be nervus atherosclerotic basilar aneurysm, was
intermedins was divided. A neurolysis m o r e extensive than those seen in most
was p e r f o r m e d on the intact portion of o u r o t h e r cases. It was directly com-
of the seventh nerve a c c o r d i n g to the pressing three c r a n i a l nerves and dis-
m e t h o d of G a r d n e r . 2 0 torting the brainstem as well. T h e seg-
In case 2 a similar ectatic loop of m e n t of a b n o r m a l vessel compressing
vertebral artery was f o u n d impinging the seventh nerve is shown in Figure 3.
on the seventh a n d eighth nerve com-
plex near the b r a i n s t e m in the left R e s u l t s of surgery
cerebellopontine angle. T h e seventh
Surgery was considered successful in
a n d eighth nerves together were sepa-
13 of the 14 cases. I n one it was consid-
rated from the artery by interposing a
ered a failure because the p a t i e n t ini-
piece of Gelfoam. T h e nervus inter-
medins was isolated a n d sectioned. A
neurolysis was p e r f o r m e d on the intact
p o r t i o n of the seventh nerve.
In case 3 there was a n unusual bony
thickening of the petrous ridge, par-
ticularly in the region of the internal
auditory meatus. An a b e r r a n t ectatic
loop of vertebral a r t e r y projected lat-
erally to m e e t the seventh nerve at its
point of e m e r g e n c e from the brain-
stem. T h e artery could easily be dis-
Fig. 3. Case 9. Operative photograph show-
placed with a probe, but immediately
ing loop of tortuous aberrant basilar artery
resumed its original position when re- (arrow) adjacent to seventh and eighth nerve
leased. A pledget of G e l f o a m was in- complex.
Downloaded from www.ccjm.org on February 10, 2022. For personal use only. All other
uses require permission.212 Cleveland Clinic Quarterly Vol. 41, No. 3
dally had only 50% to 75% relief and and vibration over the contralateral
the spasms progressed to their original side of the body and extremities.
level of intensity by the 5th month fol-
lowing surgery. Discussion
Of the IS successful cases, surgery in We have presented 14 patients with
12 was followed by complete relief hemifacial spasm in whom the facial
and, in one, by partial relief of the nerve was distorted by an elongated
facial spasms. Patient 3 is categorized and ectatic vertebral or basilar artery.
as having only partial relief because We are convinced that this was the
chewing was occasionally accompanied cause of the spasms in these patients.
by involuntary closure of the left eye. In all cases compression of the facial
Occasional mild facial twitching began nerve was actually observed during
recurring about 9 months postopera- cerebellopontine angle exploration.
tively in this patient. Three others Neurolysis of the facial nerve was per-
were symptom free before signs of re- formed in all cases, sectioning of the
currence at intervals ranging from 8 to nervus intermedius in nine, and sepa-
18 months. Eight patients showed no ration of the aberrant vessel from the
signs of recurrence when last seen 4 nerve in seven. These procedures were
weeks to 2 years postoperatively. followed by complete relief in 12 cases
Five patients had ipsilateral facial and partial relief in one.
weakness for the first time after sur- We want to emphasize that the
gery. Four others in whom facial weak- point of compression is most often lo-
ness was noted before surgery experi- cated proximally along the nerve just
enced worsening of the weakness after- after it emerges from the brainstem.
ward. One of the four patients with If this segment is not explored care-
preoperative facial weakness had com- fully, preferably with the operating
plete paralysis of facial musculature microscope, the lesion may easily be
following surgery. In most cases the overlooked.
deficit was transient and the strength T h e 14 patients reported in this
returned to preoperative levels. T h e study constitute 0.30% of all the pa-
patient in whom complete paralysis de- tients who have undergone cerebello-
veloped experienced only partial re- pontine angle exploration for hemi-
covery of the deficit. facial spasm at the Cleveland Clinic.
Seven of the patients also experi- In another large percentage of our pa-
enced some deterioration of their hear- tients, not included in this study, the
ing. T h e deficit was bilateral but worse facial nerve was compressed by a less
on the ipsilateral side in two. Most pa- well defined vessel, such as the anterior
tients recovered at least partially. inferior cerebellar or internal auditory
In addition to dizziness, of which artery. These findings corroborate our
two patients complained, other com- contention that compression of the fa-
plications were observed in only one cial nerve by an aberrant vascular struc-
patient. These included nystagmus on ture in the cerebellopontine angle is
lateral gaze, diplopia due to a contra- the most common cause of hemifacial
lateral fourth nerve palsy, a dimin- spasm.
ished ipsilateral corneal reflex, and a Others have postulated that disturb-
decrease in sensitivity to light touch ances of nerve conduction could arise
Downloaded from www.ccjm.org on February 10, 2022. For personal use only. All other
uses require permission.Winter 1974 Hemifacial spasm 213
as a consequence of compression by an contractions has been the subject of
abnormal vascular structure. Sunder- considerable discussion. Woltman et
land, 21 in his study of the neurovascu- al° were among the first to postulate
lar relationships of the internal audi- that local irritation of the nerve by
tory meatus, was one of the first, fn a compression and ischemia might cause
subsequent study of 210 autopsy speci- spontaneous discharge of impulses or
mens he found that tortuosity and de- facilitate the activation of inactive
viation of the vertebral and basilar fibers by impulses traveling through
arteries from their normal location oc- adjacent active fibers in patients with
curred frequently. 22 It was of a degree hemifacial spasm.
sufficient to deform the pontomedul- Hering 24 was the first to show in his
lary junction and compress the seventh laboratory experiments of 1882 that
and eighth nerves in 27 instances. Un- such cross stimulation does indeed oc-
fortunately, the clinical records of the cur between adjacent nerve fibers. He
cases examined were not available so created an "artificial synapse" in frogs
that he could not correlate the ana- by dividing the sciatic plexus in the
tomical changes with the symptoma- pelvis. He was able to demonstrate that
tology that had existed during life. impulses traveling in ascending sen-
In a recent study of the pathogenesis sory afferent fibers would cross over to
of elongation and distention of intra- stimulate efferent motor fibers at the
cranial arteries at the base of the brain level of the interrupted plexus.
in 34 autopsy specimens, Sacks and In a somewhat more sophisticated
Lindenburg 2:i observed atrophy caused experiment, Granit et al 25 created an
by pressure exerted on adjacent cranial artificial synapse by placing a ligature
nerves and brain in 28 cases. They around the sciatic nerve in frogs so
emphasized that arterial elongation loosely that it did not block conduc-
and distention must be considered in tion of the original impulse. By record-
the differential diagnosis of any pa- ing with electrodes from the roots of
tient with neurologic signs and symp- the sciatic nerve which they had di-
toms suggestive of a space occupying vided intraspinally, they demonstrated
lesion. that impulses descending in the motor
An important point should be fibers would cross stimulate afferent
learned from the present study: in all sensory fibers at the level of the liga-
patients with hemifacial spasm, ture. A response could also be pro-
whether or not accompanied by other duced in motor fibers by stimulating
neurologic signs, angiography of the the cut ends of the afferent sensory
vertebrobasilar system should be per- fibers. Moreover, they were able to
formed preoperatively. Of the four pa- abolish the response by removing the
tients evaluated in our series, this tech- ligature and irrigating the compressed
nique revealed the nature of the pa- segment of nerve with Ringer's solu-
thology by demonstrating an anoma- tion much as is done today during an
lous vascular structure projecting into intracranial neurolysis.
the region of the cerebellopontine In recent years Gardner 26 ' 27 has been
angle in each. one of the strongest proponents of the
Just how pressure on a nerve pro- theory that hemifacial spasm is caused
duces paroxysmal repetitive muscle by cross stimulation at a point of nerve
Downloaded from www.ccjm.org on February 10, 2022. For personal use only. All other
uses require permission.214 Cleveland Clinic Quarterly Vol. 41, No. 3
compression. He, in fact, suggests that 13. Campbell E, Keedy C: Hemifacial spasm:
a note on the etiology in two cases. J
the direction of cross stimulation is
Neurosurg 4: 342-347, 1947.
probably afferent to efferent and con-
14. Laine E, Nayrac P: Hemispasm facial
tends that this explains why, when gueri par intervention sur la fosse pos-
other intracranial procedures fail, sec- térieure. Rev Neurol 80: 38-40, 1948.
tioning of the nervus intermedius is 15. Jannetta PJ: Microsurgical exploration
usually curative. The observations and decompression of the facial nerve in
hemifacial spasm. Curr Top Surg Res 2:
made in our series of patients add 217-220, 1970.
further support to this theory. 16. Eckman PB, Kramer RA, Altrocchi PH:
Hemifacial spasm. Arch Neurol 25: 81-87,
References 1971.
1. Ehni G, Woltman HW: Hemifacial spasm. 17. Kerber CW, Morgalis MT, Newton TH:
Review of one hundred and six cases. Arch Tortuous vertebro basilar system; a cause
Neurol Psych 53: 205-211, 1945. of cranial nerve signs. Neuroradiology 4:
2. Shaywitz BA: Hemifacial spasm in child- 74-77, 1972.
hood treated with carbamazepine. Arch 18. Kramer RA, Eckman PB: Hemifacial
Neurol 31: 63, 1974. spasm associated with redundancy of the
3. Cushing H: The major trigeminal neu- vertebral artery. Am J Roentgenol Ra-
ralgias and their surgical treatment based dium Ther Nucl Med 115: 133-136, 1972.
on experiences with 332 gasserian opera- 19. Carella A, Caruso G, Lamberti P: Hemi-
tions. Am J Med Sci 160: 157-184, 1920. facial spasm due to elongation and ectasia
4. Coleman CC: Surgical treatment of facial of the distal segment of the vertebral
spasm. Ann Surg 105: 647-657, 1937. artery. Neuroradiology 6: 233-236, 1973.
5. Harris W, Wright AD: Treatment of 20. Gardner WJ: Concerning the mechanism
clonic facial spasm; (a) by alcohol injec- of trigeminal neuralgia and hemifacial
tion, (b) by nerve anastomosis. Lancet 1: spasm. J Neurosurg 19: 947-958, 1962.
657-662, 1932. 21. Sunderland S: The arterial relations of the
6. Woltman HW, Williams HL, Lambert internal auditory meatus. Brain 68: 23-27,
EH: An attempt to relieve hemifacial 1945.
spasm by neurolysis of the facial nerve. 22. Sunderland S.: Neurovascular relations
Proc Staff Meet Mayo Clin 26: 236-240, and anomalies at the base of the brain. J
1951. Neurol Neurosurg Psychiatry 11: 243-257,
7. Proud GO: Surgical treatment of hemi- 1948.
facial spasm. South Med J 46: 66-67, 1953. 23. Sacks JG, Lindenburg R: Dolicho-ectatic
8. Pulec J L : Idiopathic hemifacial spasm; intracranial arteries. Symptomatology and
pathogenesis and surgical treatment. Ann pathogens of arterial elongation and dis-
Otol Rhinol Laryngol 81: 664-676, 1972. tension. Johns Hopkins Med J 125: 95-106,
9. Revilla AG: Differential diagnosis of tu- 1969.
mors at the cerebellopontine recess. Johns 24. Hering E: Beitrage zur allgemeinen
Hopkins Hosp Bull 83: 187-212, 1948. nerven-und muskelphysiologie. Neunte
10. Revilla AG: Neurinomas of the cerebello- Mittheilbung. Ueber Nervenreizung durch
pontine recess. A clinical study of one den Nervenstrom. 39 pp. Wien, 1882. Re-
hundred and sixty cases including opera- printed from Sitzungb Akad Wiss Math
tive mortality and end results. Johns Hop- Naturwiss CI (Wien) 85: 237, 1882.
kins Hosp Bull 80: 254-296, 1947. 25. Granit R, Leksell L, Skoglund CR: Fibre
11. Gardner WJ, Sava GA: Hemifacial spasm interaction in injured or compressed re-
—a reversible pathophysiologic state. J gion of nerve. Brain 67: 125-140, 1944.
Neurosurg 19: 240-247, 1962. 26. Gardner WJ: Cross talk—the paradoxical
12. Gardner WJ, Dohn DF: Trigeminal neu- transmission of a nerve impulse. Arch
ralgia—hemifacial spasm—Paget's disease. Neurol 14: 149-156, 1966.
Significance of this association. Brain 89: 27. Gardner WJ: Trigeminal neuralgia. Clin
555-562, 1966. Neurosurg 15: 1-56, 1968.
Downloaded from www.ccjm.org on February 10, 2022. For personal use only. All other
uses require permission.You can also read
