OUP user menu

The spectrum of Möbius syndrome: an electrophysiological study

Harriëtte T. F. M. Verzijl, George W. Padberg, Machiel J. Zwarts
DOI: http://dx.doi.org/10.1093/brain/awh502 1728-1736 First published online: 13 April 2005

Summary

We studied the nature and extent of facial muscle innervation and the involvement of the motor and sensory long tracts in Möbius syndrome, in order to shed light on the pathophysiological mechanism of the syndrome. Standardized blink reflexes, direct responses of the facial nerves to the orbicularis oculi muscles and concentric needle electrode electromyography in orbicularis oculi and/or oris muscles were measured in 11 patients with Möbius syndrome, of whom six participated in MRI studies, all showing absent facial nerves. We performed motor- and somatosensory-evoked potentials in seven Möbius patients. We demonstrated three distinct patterns of abnormalities suggesting different sites of the primary lesion in different patients. (i) Presence of normal blink reflexes and facial compound motor action potentials, normal habituation tests, a reduced recruitment in the facial muscles and an aberrant ‘blink reflex-like’ response of the orbicularis oculi muscle upon stimulation of the facial nerve region, which suggests a supranuclear origin of the defect. (ii) Absent blink reflexes, absent direct responses of the facial nerves and absent motor activity on needle electromyography, indicating a defect at the facial nuclear level. However, the nuclear defect might mask an additional supranuclear defect, which cannot, therefore, be excluded in these patients. (iii) A disperse pattern of facial compound action potentials combined with long latencies that were recorded with concentric needle electrodes, indicating involvement of motor axons in the facial nerve, possibly secondary to nuclear involvement. An additional supranuclear defect cannot be excluded in these cases. All evoked potentials studied were normal. The electrophysiological findings of the facial muscles show a spectrum of disturbances varying in degree of severity and diverse in the extent of structures involved, in 11 Möbius patients. At one end of the spectrum are patients with completely immobile faces in whom electrophysiological testing shows no signs of involvement of the facial nuclei, nerves or muscles, suggestive of a dysfunction at the supranuclear level. At the other extreme of the spectrum are patients with complete absence of responses upon facial nerve stimulation and absence of motor unit activity. This is at least indicative of a defect at the facial nuclear level. While a supranuclear defect is compatible with the concept that Möbius syndrome is a developmental disorder of the lower brainstem, intact facial nuclei as part of the syndrome has not been suggested before. The findings corroborate the concept of the Möbius syndrome being a complex regional developmental disorder of the brainstem.

  • blink reflex
  • electromyography
  • evoked potentials
  • facial nerve
  • Möbius syndrome
  • pathogenesis
  • BAEP = brainstem auditory-evoked potential
  • cR2 = contralateral late blink response
  • EMG = electromyography
  • (f)CMAP = (facial) compound motor action potential
  • iR1 = ipsilateral early blink response
  • iR2 = ipsilateral late blink response
  • MEP = motor-evoked potential
  • MUAP = motor unit action potential
  • SSEP = somatosensory-evoked potential

Introduction

In 1888, Möbius reported patients with congenital, non-progressive, bilateral facial and abducens nerve palsy (Möbius, 1888). Varying definitions have hampered a clear delineation of the Möbius syndrome. Isolated congenital facial palsy and the extended phenotype of congenital facial palsy with ocular muscle weakness, with or without craniofacial dysmorphisms and congenital abnormalities of the extremities, were criteria for the diagnosis of Möbius syndrome (Kumar, 1990). Recently, we suggested facial palsy with impairment of ocular abduction as the primary criterion for Möbius syndrome (Verzijl et al., 2003). Dysfunction of other cranial nerves, orofacial malformations, limb malformations and musculoskeletal system defects are commonly associated features, but they are not necessary for the diagnosis making the syndrome extremely variable in its clinical manifestation.

Möbius syndrome is most frequently described as a sporadic condition. However, rare instances of autosomal dominant, autosomal recessive and even X-linked recessive inheritance have been described (Verzijl et al., 2003). It is distinct from congenital facial palsy, in the sense that in Möbius syndrome abduction impairment is present.

Although neither the aetiology nor the pathogenesis of the syndrome have yet been elucidated, there are two theories: a developmental rhombomeric defect including the facial cranial nerve nuclei due to a genetic cause (Legum et al., 1981; Stabile et al., 1984), or an interruption in the vascular supply of the brainstem resulting in ischaemia in the region of the facial cranial nerve nuclei owing to an environmental, mechanical or a genetic cause (Bavinck and Weaver, 1986; Lipson et al., 1989; Miller et al., 1989; Miller and Stromland, 1991; St Charles et al., 1993; Leong and Ashwell, 1997; Gonzalez et al., 1998; Smets et al., 2004). The postulated pathogenetic mechanisms are based on limited post mortem observations (Towfighi et al., 1979), and the site of the primary lesion has been localized in the nucleus, the nerve or the muscle. In a recent pathological study of our group, we presented evidence to exclude a primary developmental disorder of the facial muscle and a primary defect in the periphery of the facial nerve as possible aetiological mechanisms in autosomal dominant congenital facial palsy (Verzijl et al., 2005a).

In our clinical study, we argued that sporadic Möbius syndrome is more than a cranial nerve or cranial nuclear developmental disorder, and could be viewed as a developmental disorder of the rhombencephalon, including motor nuclei and axons, as well as traversing long tracts (Verzijl et al., 2003). Our pathological study offered support for this hypothesis, which was corroborated recently by pathological data on two patients with Möbius syndrome (Verloes et al., 2004). Also, radiologically, we provided evidence for Möbius syndrome being part of a more complex congenital anomaly of the posterior fossa with hypoplasia of the entire brainstem and with other congenital brain abnormalities (Verzijl et al., 2005b). Electrophysiologically, this hypothesis has been confirmed by the results of brainstem auditory-evoked potentials (BAEPs) in several case reports of Möbius patients who had, in the vast majority, normal hearing (Stabile et al., 1984; Sudarshan and Goldie, 1985; Erro et al., 1989; Govaert et al., 1989; Harbord et al., 1989; Kawai et al., 1990; Fujita et al., 1991; Gilmore et al., 1991; Noro et al., 1991; Singh et al., 1992; D'Cruz et al., 1993; Gonzalez et al., 1993; Hamaguchi et al., 1993; Hatanaka et al., 1993; Kankirawatana et al., 1993; Jaradeh et al., 1996; Bonanni and Guerrini, 1999; Larrandaburu et al., 1999; Nunes et al., 1999; Pedraza et al., 2000) and by the results of motor-evoked potentials (MEP) in one patient (Bonanni and Guerrini, 1999) and somatosensory-evoked potentials (SSEPs) in five of seven patients studied (Sudarshan and Goldie, 1985; Govaert et al., 1989; Gilmore et al., 1991; Hatanaka et al., 1993; Bonanni and Guerrini, 1999). These results are consistent with a lesion of the central pathways in the brainstem, suggesting a widespread involvement of the brainstem, possibly including the long tracts, as part of the syndrome.

Furthermore, in our radiological paper we demonstrated the absence of the facial nerve in six patients with Möbius syndrome (Verzijl et al., 2005b). This finding was puzzling in light of the often observed residual motor function in lower facial muscles, which is the characteristic pattern of facial weakness in Möbius patients. Our radiological findings argue that other cranial nerves aberrantly innervate some facial muscles, especially in the lower part of the face. This innervation is probably similar to the mechanism of reinnervation described in Duane syndrome (Blodi, 1970; Huber, 1974).

No systemic study has been carried out on the electrophysiological characteristics of the facial muscles and nerves in Möbius syndrome. Several case reports mention electrophysiological examination without discussing its significance. To the best of our knowledge, one electrophysiological study of a series of seven Möbius patients has been reported, suggesting a brainstem process predominantly affecting the facial nuclei and their internuclear connections (Jaradeh et al., 1996).

In order to study the nature and extent of facial muscle innervation in Möbius syndrome and to clarify the pathophysiological mechanism of the syndrome, we assessed standardized blink reflexes and direct responses of facial nerves, and we performed electromyography of facial muscles in 11 patients with Möbius syndrome, six of whom lacked facial nerves. Additionally, to find electrophysiologically further evidence for a lesion of the motor and sensory long tracts at brainstem level, we performed MEPs and SSEPs in seven Möbius patients.

Subjects and methods

Subjects

In a previous study, we examined 37 sporadic Dutch Möbius patients in a nationwide survey (Verzijl et al., 2003). All patients met the inclusion criteria of non-progressive congenital facial weakness and impairment of ocular abduction. For the present study, we randomly selected out of this group 11 cooperative patients in whom we performed blink reflexes, direct responses of facial nerves and electromyography (EMG) of facial muscles. Six of these patients participated in brain MRI studies; facial nerves were absent in all of them. In seven cooperative patients we assessed MEPs and SSEPs; of these, six patients had MRI, of which five showed mild brainstem hypoplasia.

Blink reflex

The blink reflex was elicited according to the standard technique of Kimura (2001). The electrodes were placed over the orbicularis oculi muscles and the supraorbital nerves were stimulated by a single stimulus of 0.1 ms duration. The ipsilateral early component of the response (iR1) is considered to be delayed if its latency exceeds 13.0 ms, and the late ipsi- and contralateral components (iR2 and cR2) delayed if their latencies exceed 40 and 41 ms, respectively. In addition, the latency difference between the two sides is also considered abnormal if it exceeds 1.2 ms for R1 and 5.0 ms for R2. The normal amplitude (Kimura, 2001) and duration (Berardelli et al., 1985) of the R1 and R2 components are shown in Table 1. If the blink reflex was present, we performed a habituation test. The recovery cycle of the R2 component of the blink reflex was studied by applying two pulses, with an interval of 1 s, to the supraorbital nerve at an intensity of three times the threshold (Kimura, 2001; Aramideh and Ongerboer, 2002). In normal subjects the amplitude of the R2 decreases to ∼30% of unconditioned values (Berardelli et al., 1985).

View this table:
Table 1

Electrophysiological characteristics of blink reflexes and direct responses of the facial nerves to the orbicular oculi muscle in 11 patients with Möbius syndrome

CaseSideBlink reflexDirect response facial nerve
Latency (ms)*Amplitude (mV)*Duration (ms)Latency (ms)*Amplitude (mV)*
NR/LR1<13.000.3821.0 ± 0.8<4.11.21
iR2<40.000.5347.0 ± 1.8
cR2<41.000.4947.6 ± 1.7
1R/L
2R
10.25 (n)0.6
L
28.00 (n)0.2
3RR1
iR2(L:41.2§)(0.8§)
cR2
LR18.801.7182.252.6
iR238.201.42538.11.1
cR237.001.330
4RR1
iR2
cR2
LR16.951.00183.102.6
iR231.700.3826
cR231.000.4038
5R/L
6R
10.35 (n)1.5
L
21.50 (n)0.2
7RR19.000.24213.351.4
iR229.650.3231
cR229.950.3038
LR18.950.30201.901.7
iR232.000.4037
cR230.900.3236
8R/L
9R/L
10R3.201.0
L
11R/L
  • * Upper limits of normal latency and amplitude of R1 and R2 components and fCMAP (mean ± 3 SD) (Kimura, 2001)

  • Normal duration of R1 and R2 components (mean ± 1 SD) (Berardelli et al., 1985)

  • Patients in whom the facial nerve could not be visualized in a recently performed radiological study

  • § On stimulation of the right facial nerve, a motor response was obtained in the left orbicularis oculi muscle

  • On stimulation of the left facial nerve late motor responses were present in the left orbicularis oculi muscle

  • N = normal subject; n = recorded with a concentric needle; − = no response obtained; R = right; L = left.

Direct response of the facial nerve and electromyography of the facial muscles

The nerves were stimulated with a bipolar block electrode, and all the EMG recordings were made using bipolar surface or concentric needle electrodes placed on or in various facial muscles. The EMG signals were recorded and stored using a two-channel EMG System (Medelec Synergy®, Oxford Instruments, Witney, UK). Sweep duration was 20–200 ms. Bandpass filters were set at 20 Hz and 3 kHz. The method for eliciting facial nerve conduction was adapted from Kimura (2001). The electrodes were placed over the orbicularis oculi muscles. Direct electrical stimulation of the facial nerve below the ear using constant square wave current of 0.1 ms duration elicited a facial compound motor action potential (fCMAP) of the muscle. The stimulus strength was increased to a supramaximal level. The upper limit of normal latency of the fCMAP is set at 4.1 ms. In addition, the latency difference between the two sides should not exceed 0.6 ms. Table 1 shows the normal amplitude of the direct response (Kimura, 2001).

Concentric needle EMG studies were performed in the orbicularis oculi and/or oris muscles (Table 2). At rest, attention was focused on spontaneous activity. On planned voluntary muscle contraction, shape and duration of motor unit action potentials (MUAPs) were noted, and at maximal effort, the pattern of recruitment and the amplitude of the MUAPs were reported.

View this table:
Table 2

Electromyographic findings of facial muscles in 11 patients with Möbius syndrome

CaseSideElectromyography
2ROrbicularis oculi and oris muscles: impaired recruitment, polyphasic motor units of slightly increased duration (10–15 ms) and amplitude (500–1000 μV), no spontaneous activity
LOrbicularis oris muscle: polyphasic single motor unit of normal duration (8–10 ms) and slightly increased amplitude (500–1000 μV), fibrillation potentials
3ROrbicularis oculi muscles: no voluntary contraction, insertion activity, or motor units
LOrbicularis oculi and oris muscles: impaired recruitment, motor units of increased duration (10–20 ms) and amplitude (1000–3000 μV), no spontaneous activity
4ROrbicularis oculi muscles: no voluntary contraction, insertion activity, or motor units
LOrbicularis oculi and oris muscles: impaired recruitment, polyphasic motor units of slightly increased duration (10–15 ms) and amplitude (500–1000 μV), no spontaneous activity
6RNot examined
LOrbicularis oris muscle: polyphasic single motor unit of normal duration (8–10 ms) and low amplitude (150–300 μV), no spontaneous activity; orbicularis oculi muscle: no voluntary contraction, insertion activity, or motor units
8RNot examined
LOrbicularis oculi and oris muscles: no voluntary contraction, some insertion activity, no motor units
10RNot examined
LOrbicularis oculi muscle: no voluntary contraction, insertion activity, no motor units
  • Cases 1, 5, 7, 9 and 11 were not examined. L = left; R = right.

Evoked potentials

MEPs were evoked by a Magstim 200 (2.0 tesla) magnetic stimulator and recorded from both tibialis anterior and vastus medialis muscles. SSEPs were evoked by applying electrical stimuli to both median and tibial nerves. Both evoked potentials were elicited according to standard techniques (Dumitru et al., 2002).

Results

Subjects

The clinical characteristics of eight male and three female patients with Möbius syndrome, the youngest being 12 years and the oldest 49 years of age, are summarized in Table 3. Clinical examination disclosed, in seven patients, a paralysis of the upper facial muscles and a relative sparing of the lower facial muscles including the perioral muscles, and in six patients, a complete facial palsy with in three of them a bilateral paralysis. The extent of the residual facial muscle function is illustrated in Fig. 1. Duane syndrome type I, characterized by marked limitation of abduction with minimally defective or normal adduction and retraction of the adducting eye, was present in one case. Six patients had Duane syndrome type II, characterized by marked limitation of both abduction and adduction of the eyes, and retraction of the eye on attempted adduction, and two patients had horizontal gaze palsy. Trigeminal sensory nerve functions were normal in all subjects. All patients except one showed global hypoglossia, three patients had an asymmetrical dysplastic tongue and one patient a paretic tongue. In all cases, oral dysfunction, consisting of impaired swallowing, palatal weakness, regurgitation or dysarthria was present. These and other clinical characteristics of the patients have been reported in detail (Verzijl et al., 2003). The abnormalities of the seventh cranial nerve and other malformations of the fossa posterior in six of the 11 patients are listed in Table 4.

Fig. 1

The characteristic distribution of the facial palsy for Möbius syndrome, with more severe involvement of the upper facial muscles and a relative sparing of the lower facial muscles. Raising eyebrows (A), smiling (B).

View this table:
Table 3

Clinical characteristics of 11 patients with Möbius syndrome

Patient*SexAge (years) at examinationFacial palsyHorizontal gaze palsyDuane syndrome typeHearing lossHypoglossiaAsymm. dysplastic tongueHypoglossal paresisOral dysfunctionCraniofacial malformations
Right upper/lowerLeft upper/lower
12M12++/++++/+III+++
27F22++/+++/+III+++
39M21++/+++/+III++++
413F20++/+++/−+++
517M23++/++++/++III+++
622M49++/+++/+++++
723M21++/++++/++III+++++
824F26++/++++/++III++++
926M12++/+++/++++++
1028M34++/+++/++++++
1134M12++/++/+I+++
  • * Superscript numbers correspond to the case numbers in a previous clinical study of Möbius syndrome (Verzijl et al., 2003)

  • Conductive hearing loss of 20 dB on both sides

  • M = male; F = female; ++ = paralysed; + = paretic or affected.

View this table:
Table 4

Abnormalities of the seventh cranial nerve and other malformations of the fossa posterior in six patients with Möbius syndrome

CaseRight 7th nerveLeft 7th nerveHypoplasia of the ponsOther malformations
2NPNPArnold–Chiari, pineal cyst
3NP*NP*+
5NPNP+Asymmetrical posterior fossa
7NP?+Mastoid enlargement left, hypoplastic hemi-cerebellum, unilateral ventricle enlargement
9NPNP*+
11NPNP+Arnold–Chiari
  • * No clear separation of structures in the internal acoustic meatus, no structure is recognizable on the locus of the facial nerve

  • NP = not present; ? = inconclusive perpendicular reconstruction, suggestive of absence of the seventh cranial nerve.

Blink reflex

Table 1 shows the blink reflex responses. In eight cases (cases 1, 2, 5, 6, 8, 9, 10 and 11), electrically elicited blink reflexes showed absent R1 and R2 responses. In cases 3 and 4, iR1 and iR2 responses on the right side could not be elicited; on the left side they produced a motor response with prolonged latency and normal amplitude in the left orbicularis oculi muscle (Fig. 2A). On stimulation of the left facial nerve, both normal and high voltage, and late and normal voltage motor responses were present in the left orbicularis oculi muscle; no motor responses were obtained in the left zygomatic muscle (Fig. 2B). Since an electrical stimulus to peripheral nerves, specifically the median nerve at the wrist, may induce a startle response in the orbicularis oculi muscles (Aramideh and Ongerboer, 2002), the right median nerve was stimulated in case 3, but no motor response was obtained in the left orbicularis oculi muscle. In case 7, a normal blink reflex was obtained. In cases 3, 4 and 7, the habituation test was normal. MRI in cases 3 and 7 revealed absence of both facial nerves.

Fig. 2

(A) In case 3, on stimulation of the right facial nerve, a motor response with prolonged latency and normal amplitude was obtained in the left orbicularis oculi muscle [channels 1.1 (latency 48.60 ms, amplitude 0.6 mV), 2.1 (latency 42.00 ms, amplitude 0.7 mV), 3.1 (latency 33.00 ms, amplitude 1.0 mV)]. No motor response was observed in the right orbicularis oculi (channels 1.2, 2.1, 3.2). (B) On stimulation of the left facial nerve, a high voltage motor response with normal latency [channels 1.1 (latency 2.4 ms, amplitude 2.9 mV), 2.1 (latency 2.00 ms, amplitude 2.3 mV), 3.1 (latency 2.35 ms, amplitude 2.6 mV)], and a second, late normal voltaged motor response was found in the left orbicularis oculi muscle, in case 3 [channels 1.1 (latency 40.00 ms, amplitude 1.1 mV), 2.1 (latency 36.75 ms, amplitude 0.9 mV), 3.1 (latency 37.60 ms, amplitude 1.2 mV)]. No motor response was seen in the right orbicularis oculi (channels 1.2, 2.1, 3.2).

Direct response of the facial nerve and electromyography of the facial muscles

Electrophysiological characteristics of the direct response of the facial nerves of 11 patients with Möbius syndrome are summarized in Table 1. Direct responses of the facial nerves were absent in seven cases bilaterally (cases 1, 2, 5, 6, 8, 9 and 11) and in three cases unilaterally (cases 3, 4 and 10). On MRI, facial nerves were absent in cases 2, 3, 5, 7, 9 and 11. The left fCMAPs in cases 3 and 4 had normal distal latency and duration and high amplitude. In case 10, the right fCMAP showed a normal response. In case 7, who presented with a symmetrical facial diplegia and in whom neuroimaging revealed absence of both facial nerves, the facial motor distal latencies were asymmetric, the right being prolonged compared with the left, but both were within the limits of normal; the amplitudes were slightly increased. In all cases, a high stimulation threshold was required to elicit a response. Using a needle electrode, in cases 2 and 6, the responses were found to be severely delayed and disperse (Fig. 3). The electromyographic findings are listed in Table 2. In five cases needle EMG was not performed.

Fig. 3

Using a needle electrode in case 6, on direct stimulation of the left facial nerve a severely delayed (latency 21.50 ms, amplitude 0.2 mV) and disperse motor response was obtained.

Evoked potentials

MEPs were performed in four of the initially seven cooperative patients. In three patients, the studies could not be finished because of patients' discomfort. MEPs of the tibialis anterior or vastus medialis muscles, obtained in cases 2, 7, 8 and 11, were normal. In all seven patients, normal SSEPs of the median and peroneal nerves were recorded at both arms and legs.

Discussion

To date, only one electrophysiological study on the facial nerves and muscles of a series of Möbius patients has been reported. Jaradeh et al. (1996) suggested a brainstem process predominantly affecting the facial nuclei and their internuclear connections. Other studies on electrophysiological findings in Möbius syndrome mainly deal with case reports without discussing the significance of electrophysiology. In the literature, blink reflex studies have been carried out in 12 patients. Blink reflexes were absent in five cases (Govaert et al., 1989; Kankirawatana et al., 1993; Jaradeh et al., 1996; Bonanni and Guerrini, 1999; Larrandaburu et al., 1999) and long latencies were found in seven (Hatanaka et al., 1993; Jaradeh et al., 1996). Facial nerve conduction studies have been performed in 22 cases, 11 of whom revealed no response to the orbicularis oculi muscle (Hellstrom, 1949; Richards, 1953; van Allen and Blodi, 1960; Hanson and Rowland, 1971; Rubinstein et al., 1975; Govaert et al., 1989; Jaradeh et al., 1996; Tran et al., 1998; Criado and Aytes, 1999); seven of these patients did reveal a response to the orbicularis oris muscle (Hellstrom, 1949; Richards, 1953; Hanson and Rowland, 1971; Rubinstein et al., 1975; Tran et al., 1998; Criado and Aytes, 1999). The latency to the orbicularis oculi muscles was prolonged (Olson et al., 1970; Nardelli et al., 1983; Jaradeh et al., 1996; Kiratli and Erdener, 2000) and normal (Harris et al., 1983; Hatanaka et al., 1993; Jaradeh et al., 1996) in six cases despite facial palsy, and a reduced amplitude of the fCMAP was demonstrated in nine cases (Harris et al., 1983; Hatanaka et al., 1993; Jaradeh et al., 1996). In none of the above-mentioned cases has imaging of the facial nerves been performed. Needle EMG studies have been carried out in some 19 cases, revealing absent action potentials during planned activity in 10 cases (Breinin, 1957; Merz and Wojtowicz, 1967; Abid et al., 1978; Nardelli et al., 1983; Sudarshan and Goldie, 1985; Govaert et al., 1989; Kankirawatana et al., 1993; Jaradeh et al., 1996; Kiratli and Erdener, 2000). Impaired recruitment was seen in four instances (Harrison and Parker, 1960; van Allen and Blodi, 1960; Harris et al., 1983; Kankirawatana et al., 1993), and a single motor unit pattern in one case (Olson et al., 1970). Although not always mentioned, MUAPs were found to be normal in three cases (Jaradeh et al., 1996), high voltage in one (Olson et al., 1970), and polyphasic, low voltage and disperse in four cases (Jaradeh et al., 1996). EMG revealed spontaneous activity in the form of fibrillations in five cases (Hellstrom, 1949; van Allen and Blodi, 1960; Jaradeh et al., 1996). From all these reports one can draw the overall conclusion that electrophysiological study performed in an individual patient is too incomplete for any deductions to be made about the site of the primary lesion.

The electrophysiological findings in our study support our hypothesis that Möbius syndrome is more than a cranial nuclear developmental disorder, and that it should be viewed as a diffuse developmental disorder of the rhombencephalon. By electrophysiological means, we demonstrated three sites of primary lesion in our patients.

(i) In two patients with the characteristic pattern of facial weakness in Möbius syndrome, i.e. severe involvement of the upper facial muscles with a relative sparing of the lower facial muscles (cases 3 and 4), and in two patients with complete facial diplegia (cases 4 and 7), within the two cases studied with MRI absence of facial nerves on MRI (cases 3 and 7), normal blink reflexes and fCMAPs and normal habituation tests were found, suggesting a supranuclear dysfunction. The reduced recruitment of the facial muscles supports a supranuclear aetiological component (cases 3 and 4). The aberrant ‘blink reflex-like’ response of the orbicularis oculi muscle upon stimulation of the facial region (case 3) suggests the presence of abnormal projections to the facial nucleus that, most probably, extend via interneurons located in the supranuclear region, since the habituation of the blink reflex responses is normal. The inability to record blink reflexes despite the presence of fCMAPs (cases 2, 6 and 10), points to greater conduction abnormalities in the proximal portion of the facial nerve or in the brainstem, or to a possible trigeminal nerve involvement. We postulate extensive reorganization of motor and sensory connections in a way similar to that found in the Duane syndrome (Blodi, 1970; Huber, 1974). Our findings indicating a defect at the supranuclear level are in agreement with reported abnormalities of BAEPs, suggesting a pontine anomaly at a supranuclear site to be part of the syndrome (Harris et al., 1983; Stabile et al., 1984; Sudarshan and Goldie, 1985; Erro et al., 1989; Hamaguchi et al., 1993; Hatanaka et al., 1993).

(ii) Absent blink reflexes, absent direct responses of the facial nerves (cases 1, 3, 4, 5, 8, 9 and 11), and in addition absent motor activity on needle EMG (cases 3, 4, 8 and 10), indicate a defect at the facial nuclear level. In cases 3, 5, 9 and 11, MRI studies were performed and revealed both facial nerves to be absent. In addition, paucity of potentials in the pattern of recruitment suggests an incomplete development of the facial nucleus (cases 2 and 6). A supranuclear defect, however, might be masked by the defect at the nuclear level and cannot, therefore, be excluded in these patients.

(iii) Disperse patterns of facial compound action potentials combined with long latencies, recorded with concentric needle electrodes, indicating a motor facial nerve disorder that might be secondary to a nuclear involvement, were found in cases 2 and 6. These findings could be either on account of a primary motor nerve cell disorder resulting in small, poorly myelinated axons or the result of secondary sprouting with small fibres conducting slowly. Alternatively, aberrant innervating fibres from the trigeminal nerve may be poorly myelinized. The absence of synkinesia, owing to reinnervation as seen in Bell's palsy, is remarkable and argues against reinnervation in Möbius syndrome. In addition, the presence of fibrillation potentials suggests insufficient or absent reinnervation. An additional supranuclear defect cannot, however, be excluded in these cases.

All three electrophysiological types of defect, representing three different sites of lesion, might be combined in a single patient, which results in a spectrum of electrophysiological disturbances and could explain the variability in the clinical manifestation of the syndrome. However, of our two patients with distinct levels of electrophysiological defects on each side (cases 3 and 4), one showed symmetrical and one showed asymmetrical facial weakness; patients with asymmetrical facial weakness (cases 1, 10 and 11) appeared to have the same defect at both sides.

In a recent radiological study, the proximal portion of the facial nerve appeared to be absent in all six patients studied; two of them had facial diplegia and four demonstrated the characteristic pattern of facial weakness in Möbius syndrome. If any residual motor function was present, it was only found in some lower facial muscles (Table 3). These findings suggest that another cranial nerve, probably the trigeminal, less likely the glossopharyngeal or hypoglossal nerve, aberrantly innervates facial muscles, especially in the lower part of the face. This is probably due to a similar mechanism of reinnervation as that found in the Duane syndrome (Blodi, 1970; Huber, 1974). The fact that we could stimulate electrophysiologically the facial nerve in three patients, in whom MRI had revealed facial nerves to be absent (cases 2, 3 and 7) probably supports our hypothesis of aberrant innervation. One cannot exclude the possibility that the facial nerve, which could not be visualized in our radiological study, was severely hypoplastic or atrophic and escaped visualization despite the high spatial resolution (0.6 × 0.5 × 1.0 mm in plane resolution) of the three-dimensional CISS MR method (Verzijl et al., 2005b). However, then, we would have not obtained the wide spread of electrophysiological responses.

We started out on this study with the assumption that published BAEP studies in Möbius syndrome had demonstrated a pontine anomaly at a supranuclear site, which was the reason why we did not record BAEPs in our patients. It is amazing that, based on these studies, it was not concluded previously that Möbius syndrome is more than a cranial nerve nuclear developmental disorder and should be viewed as a more regional developmental disorder of the brainstem, possibly including the central pathways. In a recent radiological and pathological study, we obtained support for this hypothesis by demonstrating brainstem hypoplasia (Verzijl et al., 2005a, b). The results of the evoked potentials in this study, however, do not offer electrophysiological evidence for a lesion of the long tracts traversing through the brainstem in Möbius syndrome. Normal MEPs and SSEPs can be explained by the fact that only the conduction time of the fastest fibres is measured; lesion of only part of the long tracts would leave enough normal fibres to measure. Therefore, based on normal evoked potentials in our Möbius patients, a reduction of the traversing long tracts at pontine level can not be excluded.

Conclusions

The most prominent finding of this study is the fact that the electrophysiological results in patients with Möbius syndrome show a spectrum of disturbances varying in degree of severity and extent of structures involved, suggesting defects at three levels i.e. supranuclear, nuclear and peripheral, which might be combined in one patient. At one end of the spectrum are patients with completely immobile faces, some of whom lack facial nerves on MRI, who show no abnormalities of the facial nerve or muscles on electrophysiological testing. This suggests a dysfunction at the supranuclear level. This situation is corroborated by abnormal late responses on stimulation of the facial nerve. At the other end of the spectrum are patients with totally absent responses on facial nerve stimulation, absence of motor unit activity and absence of facial nerves on neuroimaging. These findings indicate at least a defect at the facial nuclear level; a supranuclear defect cannot be excluded.

Our findings support the hypothesis that the primary site of the lesion in Möbius syndrome is not necessarily restricted to the cranial nerve nuclei, but is localized more regionally in the brainstem.

Acknowledgments

We wish to thank the patients for their participation in this research. This work was funded by NWO grant 920-03-021.

References

View Abstract