(1)
School of Behavioral and Brain Sciences, The University of Texas, Dallas, TX, USA
Abstract
Hemifacial spasm can be cured by microvascular decompression (MVD) operations of the root exit zone of the facial nerve. This fact was the basis for the (“ephaptic”) hypothesis stating that the anatomical location of the pathology that generates the signs of HFS, spasm in the mimic muscle on one side of the face and synkinesis, was the root exit zone of the facial nerve. However, later intracranial recording from the facial nerve provided strong experimental support of a different hypothesis about the pathology of HFS, namely, that the anatomical location of the pathology is the facial motonucleus.
Intraoperative measurements of neural conduction times provided evidence against the “ephaptic” hypothesis and showed evidence that hyperactivity of the facial motonucleus could explain the symptoms of HFS. Studies of the blink reflex supported the hypothesis that the facial motonucleus is hyperactive in people with HFS. The results of animal experiments showed that signs of HFS could be caused by facial motonucleus hyperactivity.
It was hypothesized that the abnormalities in the facial motonucleus in HFS were caused by activation of maladaptive neuroplasticity that was activated by the irritation of the root of the facial nerve by a blood vessel. These findings were supported by the results of animal studies.
Since a similar close contact with a blood vessel is present in at least 50 % of individuals who do not have any symptoms of spasm, it was concluded that a second factor in addition to vascular contact with the facial nerve root must be present in order to create the signs of HFS.
MVD operations have a success rate of over 85 %, and when combined with monitoring of the abnormal muscle contraction, success rates of 97 % have been reported. No other treatment has been shown to have noticeable success in relieving the signs of HFS.
Keywords
Hemifacial spasmVascular compressionNeuroplasticityFacial nucleusHyperactivity3.1 Introduction
Hemifacial spasm (HFS) is a rare disease (incidence: 0.74 per 100,000 in white men and 0.81 per 100,000 in white women; (Auger 1979)) that has its onset relatively late in life. HFS is characterized by episodes of spasm separated by periods of essentially normal function of the mimic musculature (Ehni and Woltman 1945). The disorder can be cured by moving a blood vessel (artery or vein) off of the intracranial portion of the facial nerve using a surgical technique known as microvascular decompression (MVD) (Jannetta 1970).
Microvascular decompression (MVD) operations have been shown to be effective in treating hemifacial spasm (HFS), trigeminal neuralgia (tic douloureux) (TGN), and glossopharyngeal neuralgia (GPN). Symptoms of some disorders associated the eighth cranial nerve, such as some forms of tinnitus and disabling positional vertigo (DPV), a balance disorder, can also be alleviated by MVD. In the following Chap. 8 in this book, we will discuss the forms of tinnitus and DPV that are treatable by MVD. Vascular contact with the hypoglossal nerve seems to be associated with a rare disorder, hemilingual spasm (Osburn et al. 2010). Some forms of spasmodic torticollis also seem to be associated with close contact between a blood vessel and a cranial nerve (CNXI) (Freckmann et al. 1981, 931; Nakai et al. 1989).
In this chapter, we will discuss HFS and its treatment with MVD of the intracranial portion of the facial nerve. For a review of history of the MVD operation, see Møller (1998).
Vascular contact with a cranial nerve is common, occurring in at least 50 % of people without causing any symptoms (Sunderland 1948). This means that symptoms of disease only occur when another factor (pathology) is present. This other factor is unknown (Møller and Jannetta 1984), but since it does not seem to cause symptoms on its own and symptoms only occur when vascular contact is also present, the disease can be cured by eliminating the vascular contact. This explains why MVD operations are effective in eliminating the symptoms of diseases such HFS, TGN, and GPN.
3.2 Results of MVD Operations for HFS
Large studies have shown that the cure rate for MVD operations for HFS (Barker et al. 1995) is commonly 85 %. This is similar to MVD operations for TGN (Barker et al. 1996) and GPN (Laha and Jannetta 1977) and probably some forms of spasmodic torticollis (Nagata et al. 1989). Intermediate nerve neuralgia (geniculate neuralgia) can also be cured by MVD operations (Tang et al. 2014). (Treatment of vascular compression disorders of the eighth cranial nerves is discussed in Chaper 8 in this book.)
A study of 1174 operations with 1 year or more follow-up (Hyun et al. 2010) showed that at the 1-year follow-up of 1105 patients, 5.9 % had residual spasm, thus a cure rate of 94.1 %. This study also reported that transient hearing loss occurred in 31 patients (2.6 % of all the patients operated upon), permanent hearing loss in 13 (1.1 %), transient facial weakness in 86 (7.3 %), permanent facial weakness in nine (0.7 %), cerebrospinal fluid leak in three (0.25 %), and cerebellar infarction or hemorrhage in two (0.17 %).
The precise location of the vascular contact along the intracranial portion of the facial nerve occurs varies among people with HFS. De Ridder et al. (2002) showed evidence that vascular compression syndromes arise from vascular contact along the CNS segment of the cranial nerves. Campos-Benitez and Kaufmann (2008) studied 115 patients who were operated upon for HFS (Campos-Benitez and Kaufmann 2008). Thirty-eight percent of these patients had multiple vessels in contact with the facial nerve. When the intracranial portion was examined, it was found that the prevalence of vascular contact along the cranial portion of the nerve was not distributed evenly. The root exit point was the primary offending area in 10 %, the attached segment in 64 %, the Obersteiner–Redlich zone (transition between central and peripheral axonal myelination) in 22 %, and the distal cisternal portion in 3 %. As judged during the operation, the severity of compression was mild in 27 %, moderate in 61 %, and severe in 12 % of patients. Failure to alleviate the HFS occurred in nine cases, but this failure was not related to compression location, severity, or vessel type. The vessels that were in close contact with the facial nerve were the anterior inferior cerebellar artery (in 43 %), posterior inferior cerebellar artery (in 31 %), vertebral artery (in 23 %), or a large vein (in 3 %).
3.3 Symptoms and Signs of Hemifacial Spasm
HFS is characterized by periods of spasm engaging only the mimic muscles of one side of the face. In some people with HFS, more or less pronounced degrees of synkinesis are present. The signs typically develop gradually, beginning with brief contractions of the muscles around the eye. The condition progresses over several years. The intensity of the spasm around the eye increases and spreads to other mimic muscles lower down on the face. Typically, after many years (10–15), most mimic muscles have been recruited, including the platysma, though the muscles of the forehead are excluded as they have a different innervation pattern (bilateral facial nerve innervation).
In addition to the episodes of spasm and synkinesis, HFS is characterized by an abnormal muscle response (AMR) which can be elicited from mimic muscles by electrical stimulation of a branch of the facial nerve (Esslen 1957; Nielsen 1984a; Møller and Jannetta 1984, 1985; Møller 1993). This AMR can be demonstrated by EMG recordings from mimic muscles that are innervated by a different branch of the facial nerve. The initial component of the AMR has a latency of approximately 10 ms when, for example, elicited from the temporal or zygomatic branch of the facial nerve and recorded from the mentalis muscle (Fig. 3.1).
Fig. 3.1
(a) Electrode montage for recording the abnormal muscle response (Artwork by Zahra Akhavi). (b) Placement of needle electrodes for recording the AMR in a patient to undergo MVD for HFS
It is not known how specific the AMR is for HFS, but it can normally not be elicited in people who do not have HFS, and in HFS, it disappears after successful treatment with MVD. The AMR is also known as the “lateral spread response.”
The AMR can normally be recorded under surgical anesthesia although the amplitude of the AMR will be slightly lower than when the patient was awake. The AMR has some similarities with the F-response. (The F-response can be elicited by electrical stimulation of a peripheral motor nerve (Møller and Jannetta 1987). The response is caused by firing of alpha motoneurons by antidromic activity in their motor nerves. Like the F-response, the AMR is assumed to be a measure of excitability of the facial motor nucleus.
The initial component of the AMR that can be recorded in response to electrical stimulation of a peripheral branch of the facial nerve is followed by EMG potentials that may last 100 ms or more (Fig. 3.2). The prolonged response reflects repetitive firings of motoneurons that produce repetitive muscle contractions (spasm).
Fig. 3.2
Typical recordings of the abnormal muscle response in a patient with HFS who is to undergo an MVD operation. The response was recorded from the mentalis muscle and elicited by stimulation of the zygomatic branch of the facial nerve (Reprint with permission from Møller (1997); originally published in Møller and Jannetta (1985)
The AMR usually persisted until the blood vessel that was in close contact with the facial nerve lifted off the facial nerve. In some patients, especially patients who had had their spasm for a short time only, the AMR decreased and even disappeared before the facial nerve was exposed (Møller and Jannetta 1985). When that happens, the AMR can almost always be activated again by stimulation of the facial nerve branch at a high rate (50 pps) for a short time (Fig. 3.3). This procedure sometimes has to be repeated in some MVD operations to maintain the AMR before the facial nerve is exposed.
Fig. 3.3
Recordings of the AMR in a patient undergoing an MVD operation for HFS. Consecutive recordings (beginning at top) from the mentalis muscle in response to electrical stimulation of the temporal branch of the facial nerve are shown. The effect on the abnormal muscle response from increasing the stimulus rate from 5 to 50 pps for a short period of stimulation is shown in a patient in whom the abnormal muscle response was very small when recorded in the beginning of the operation (Reprint with permission from Møller (2011) Fig. 15.8; originally from Møller and Jannetta (1985))
In the example shown in Fig. 3.3, the amplitude of the AMR decreased before the facial nerve was exposed. Increasing the stimulus rate from 5 pp to 50 pp for a brief period brought the AMR back to the amplitude it had at the beginning of the operation.
Many patients have multiple vessels in contact with the facial nerve, but often times, only one vessel needs to be moved off the facial nerve in order to achieve complete relief from the spasm. A visual inspection is not sufficient; the only way to discern which vessels need to be moved is through intraoperative monitoring of the AMR. The AMR disappears when the blood vessel that is associated with the symptoms is moved off the intracranial portion of the facial nerve (Møller and Jannetta 1985) (Figs. 3.4 and 3.5). This principle has led to the development of a monitoring method that can help identify which vessel should be moved off the facial nerve to eliminate the symptoms (Møller and Jannetta 1987).
Fig. 3.4
AMR recorded from the mentalis muscle in response to consecutive stimulations of the zygomatic branch of the facial nerve in a patient undergoing an MVD operation for HFS. Then recordings started at the top of the left-hand column. A blood vessel was lifted off the root of the facial nerve at the time the top recording of the second column was done (Reprint with permission from Møller (2011), Fig. 15.9; originally from Møller and Jannetta (1985))
Fig. 3.5
Recordings of the AMR in a patient undergoing an MVD operation for HFS. The responses were recorded from the mentalis muscle in response to stimulation of the zygomatic branch of the facial nerve. The response to consecutive stimulation beginning at the top of the left-hand column, continuing from the topof the middle column. The single arrow near the bottom of the left-hand column indicates when a vessel was lifted off the root of the facial nerve. The double arrows at the top of the middle column indicate when the vessel was allowed to fall back on the nerve root (Reprint with permission from Møller (2011), Fig. 15.7; originally from Møller and Jannetta (1985))
The blink reflex (also known as the supraorbital nerve reflex (Nielsen 1984b)) elicited by stimulation of the supraorbital nerve is abnormal in people with HFS in that it engages several muscle groups including the mentalis muscle, while it normally only involves the orbicularis oculi muscles.
When a blink reflex response is elicited in people with HFS through electrical stimulation of the supraorbital nerve, the reflex response not only includes the muscles around the eye but also other muscles of the face (Kim and Fukushima 1984; Nielsen 1984b; Auger 1979). This spread of blink reflex activity to other parts of the face is a sign of the synkinesis of mimic muscles often observed in people with HFS. This abnormality has been named the “lateral spread response” (Nielsen 1984b).
Valls-Sole and Tolosa (1989) found that the R1 component of the blink reflex response recorded from the mentalis or the orbicularis oris muscle has a slightly longer latency than the response from the orbicularis oculi in the normal blink reflex response (12.6 ± 1.3 ms). This prolongation of the R1 component of the response recorded from muscles of the lower face could be explained by the fact that facial nerves travel a greater distance to the muscles of the lower face than to the muscles of the upper face. Nielsen reported similar results (Nielsen 1984b).
Despite having seemingly distinct and clear symptoms, hemifacial spasm is frequently reported to be misdiagnosed in the primary care setting (Martinez et al. 2014).
Spasm of mimic muscles that resembles hemifacial spasm occasionally occurs together with trigeminal neuralgia, known as tic convulsive (Yeh and Tew 1984).
An atypical form of HFS where the spasm progresses upward over the face has been described (Jannetta 1990).
3.4 Pathophysiology of HFS
It was probably Dandy (1934) who first recognized and described how vascular conflict of cranial nerves could cause a specific disorder, in his case, trigeminal neuralgia (TGN). When Dandy (1929) sectioned the trigeminal nerve in the cerebellopontine angle (CPA) to treat patients with TGN, he observed vascular compression of the trigeminal nerve, and he later reported that he believed this compression to be the cause of tic douloureux (Dandy 1934). Cushing (1920) had earlier hypothesized that TGN could be caused by pressure from a tumor on the trigeminal root, but it was probably Taarnhøj (1952, 1956) who first described the beneficial effect of decompressing the trigeminal nerve root to treat trigeminal neuralgia. Much later, Gardner and Sava (1962) reported the presence of vascular compression of the seventh cranial nerve root in patients with HFS (Møller 1998).
Several theories have been presented explaining the pathology of those disorders that can be cured by moving a blood vessel off the intracranial portion of the fifth, seventh, and eighth cranial nerves (for a review, see Møller (1991). It was earlier believed that it was the pounding of an artery into a nerve root that injured the nerve and thereby caused the symptoms. It has also been assumed that arteries in close contact with a nerve root were important, but it is now known that veins (Wang et al. 2013) and very small arteries are important to move off the facial nerve to effectively treat HFS in some patients (Jannetta 1984).
It is also known that symptoms of HFS may be caused by close contact between a vessel and other parts of the intracranial portion of the facial nerve than the Obersteiner–Redlich zone (Kondo et al. 1980; Møller and Jannetta 1987).
Investigators in different fields of medicine have been fascinated by the pathology of diseases that are so effectively cured by MVD. One of the unanswered questions has been the anatomical location of the site of the pathology; another question has regarded an explanation for the fact that both HFS and TGN are very rare disorders, while vascular contact (compression) is commonly found in asymptomatic people.
Two hypotheses explaining the role of the close vascular contact have prevailed. One suggests that hyperactivity of the facial motonucleus is the cause of the spasm and synkinesis (Ferguson 1978). The other hypothesis states that cross talk (ephaptic transmission) between individual nerve axons of the facial nerve occurs (ephaptic communication) where it is in contact with a blood vessel causing ectopic excitation that could explain the signs of HFS (Esslen 1957; Williams et al. 1952; Gardner 1966; Gardner and Sava 1962; Nielsen 1984a).
That close contact between injured nerve fibers could cause direct communication (“cross talk”) between axons of nerve was first suggested by Granit et al. (1944).
Ephaptic transmission seems a logical explanation of many different pathologies including the synkinesis in HFS. Ephaptic transmission, however, is a rare phenomenon (Granit et al. 1944) and has only been verified in very few instances. The ephaptic transmission hypothesis was nonetheless favored for many years as an explanation for HFS (Nielsen 1984a, b).
Ferguson (1978), however, was one of the first investigators to point out that ephaptic transmission between denuded facial nerve axons is not sufficient to explain the symptoms and signs of HFS. More recently, Esteban and Molina-Negro (1986) arrived at similar conclusions on the basis of preoperative studies of people with HFS. These investigators questioned whether the ephaptic transmission hypothesis could explain HFS because it seemed unlikely that the nerve fibers could be in contact with each other on a scale sufficient to cause the characteristically massive contraction of nearly all the facial muscles (Møller 1999).
Results of studies using electrophysiological recording during MVD operations of patients with HFS reported by Møller and Jannetta (1984) provided physiological evidence against the ephaptic hypothesis and in favor of the hyperactivity hypothesis.
That stimulation of one branch of the facial nerve produced EMG responses from not only the muscles that are innervated by the stimulated nerve branch but also from muscles innervated by other branches of the facial nerve indicates that there is an abnormal cross-transmission somewhere in the path of the antidromic facial motor nerve. The branch that is stimulated must come in contact with the branch that innervates the muscles from which the EMG potentials are recorded.
Studies using the recording of the AMR during MVD operations for HFS supported the hypothesis that the anatomical location of the physiological abnormalities in HFS that causes spasm and synkinesis is the facial motonucleus. Measurements of the conduction time in the individual segments of the assumed path of the AMR in these intraoperative studies revealed that the latency of the AMR was approximately 2 ms longer that it would have been if the cross talk occurred in the facial nerve at the location of the vascular contact (Møller and Jannetta 1984) (Fig. 3.6).
Fig. 3.6
Comparison between the AMR elicited by stimulation of the temporal branch of the facial nerve recorded from the mentalis muscle (middle tracing) and the response from the intracranial portion of the facial nerve at the location of the vascular contract in response to stimulation of the temporal branch of the facial nerve (bottom tracing). The top tracing is the response from the mentalis muscle to electrical stimulation of the intracranial portion of the facial nerve at the location of the vascular contact. If the cross talk that causes the AMR occurred in the facial nerve at the location of the vascular contact, the tracing of the response from the mentalis muscle to stimulation of the facial nerve would have been shifted to the left coinciding with the occurrence of the negative peak of the recording from the facial nerve
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
