The assessment of facial palsy
by Glen Croxson and Tim Eviston
The management of facial nerve disorders represents a special, complex and diverse area of medicine. The challenge of managing facial nerve disorders was summarized by Kamerer (1986), who said "seldom can the diagnostic acumen of the physician be more severely tested than in the evaluation and treatment of patients with facial nerve paralysis''. Meeting this challenge requires the application of existing skills as well as the acquisition of new ones. The following is a primer on what information we look for when we assess a patient with facial palsy and how some of the signs may be interpreted.
A careful history explores the onset of facial weakness, its progress, prodromal symptoms, predispositions, associated symptoms and, if relevant, the nature of recovery.
The rate of onset of facial paralysis may give valuable diagnostic information. Slowly progressing weakness, rapid painless complete paralysis, and fluctuant weakness should suggest tumour (May, 1986). Bilateral synchronous involvement can signal Guillain-Barre syndrome, Melkersson Rosenthal syndrome, or bilateral idiopathic paralysis (IP). Recurrent paralysis on the ipsilateral side raises the spectre of tumour, whereas contra lateral involvement is usually recurrent IP (May, 1986).The completeness of facial weakness has important diagnostic and prognostic implications. Complete paralysis is associated with more severe nerve damage than incomplete lesions and is therefore related to more serious pathologies and worse prospects of recovery (May, 1986). Rapid progress to complete paralysis may occur in IP, herpes zoster, trauma or tumours of benign or malignant variety (Fisch, 1980). Slowly progressive paralysis for greater than three weeks with or without fluctuation is also likely to be neoplastic.The diagnostic and prognostic value of prodromal symptoms have been examined by Adour et al (1978), and are discussed under prognostication. The relevance of these prodromal symptoms against their incidence in healthy cohorts is questionable, and their presence has no prognostic value.Idiopathic paralysis has a familial pattern with 14% of patients having a family member with the condition (May, 1986; Willbrand, 1974). Similarly, other congenital and acquired conditions (eg: Melkersson Rosenthal syndrome, Mobius syndrome) require exclusion by family history (May 1987). Hypertension, diabetes, pregnancy, neoplasia, systemic illness and previous ear disease or head injury may be relevant (Adour, 1982). A history of previous facial paralysis, especially in relation to trauma, surgery or systemic illness may be diagnostic (May, 1987).
The presence of specific associated symptoms should be noted. Mastoid pain, earache, neck pain and unilateral occipital headache occur in 50% (May, 1986) to 60% (Adour, 1982) of patients with Idiopathic Paralysis. Other cranial nerve symptoms including subjective facial sensory disturbance, tinnitus, dizziness, dysacusis, vertigo, dysgeusia and tearing abnormalities should be sought (Adour, 1978). Skin eruptions, vesicles, facial swelling and other neurological disturbances are of relevance.
The nature and pattern of facial nerve recovery is related to the degree of injury of the nerve, which in turn relates to the disease. Of primary interest is the degree of paralysis, the duration of paralysis, the rapidity of recovery, the completeness of recovery, and the presence of sequelae. Above all else, the recovery pattern must be consistent with the diagnosed condition.
Incomplete paralysis is an excellent prognostic sign associated with transient (days to weeks) facial dysfunction and complete resolution of weakness without sequelae (Fisch, 1984). The persistence of weakness, incomplete resolution, recurrence or evolution of recovery sequelae following incomplete paralysis indicates a neoplastic process (May, 1986).
Complete paralysis is related to more severe nerve injury. The injury however may fall anywhere along the spectrum described by Sunderland from first degree to fifth degree (Sunderland, 1978). Recovery will depend upon the degree of injury. First degree injury is a chemical neurological blockade, unassociated with structural nerve damage. Neural deblocking occurs early (hours to days) and is related to rapid complete recovery. Second degree injury involves axonal disruption, Wallerian degeneration and axon regeneration. Recovery requires axon regrowth occurring at I mm per day from the site of injury to the motor end plates. Consequently, intratemporal injuries recover between 3-6 months after injury depending upon the site of the lesion. Regrowth down existing endoneural tubes results in appropriate and often plentiful axonal reinnervation, giving complete or near complete recovery with little or no sequelae. Once recovery of function begins, progress is rapid with improvement achieved within several months.
Higher degrees of injury involving the endoneural tube (third degree injury), perineurium (fourth degree injury), and epineurium (fifth degree injury), have more profound effects upon the completeness of recovery than the duration of paralysis. Progressive loss of soft tissue components of the nerve superstructure with third, fourth and fifth degree injury result in worsening recovery patterns (Sunderland, 1978). Loss of the endoneural tube influences the number and direction of regenerating neurones. As the number of successfully regenerating neurones decreases, the residual facial weakness increases. Also, the percentage of aberrant regeneration is related to the sequelae of synkinesis, hypenonicity, facial spasm and tics. Ironically, the most energetic and successful regeneration may give the most deforming sequelae and consequently a poor recovery. The more severe the injury, the less rapid the recovery once it begins. Most recovery is achieved in 18 months but continues up to five years following severe nerve injury (May, 1986).
Finally, the recovery pattern must match the diagnosed condition. Incomplete recovery with sequelae resulting from a partial or incomplete paralysis cannot be IP. Similarly absent recovery after six months again cannot be IP (May, 1987).
Clinical symptom s and patterns of weakness that do not fit the expected diagnosed disease profile should be viewed with great suspicion. At least 10% of all facial nerve paralyses are incorrectly diagnosed as idiopathic (May, 1986).
A systematic approach to examination should be followed, assessing weakness, functional and cosmetic deficits, neurologic and general ENT abnormalities.
Upper motor neuron weakness should be distinguished from lower motor neuron by eliciting forehead sparing, preservation of spontaneous expression, abnormalities of speech and long tract signs in upper motor neurone disease (May, 1986).
Generalised lower motor neuron weakness should be separated from isolated facial nerve division or branch weakness which may indicate peripheral or intraparotid pathology. Unilateral weakness should be confirmed although the presence of bilateral paralysis is difficult to assess in the absence of a reference side.
The tone and muscle power of each group of muscles innervated by the five branches of the facial nerve should next be estimated. Whilst splinting the normal side with the fingers, the frontalis, orbicularis oculi, nasalis, zygomaticus, orbicularis oris and platysma are tested. The tone is noted at rest.
The major functional deficits of facial paralysis are incomplete eye closure, oral incompetence and nasal obstruction. Of these, the eye is the most significant with the largest potential for debilitating sequelae.
Eye assessment should include measurement of the degree of exposure in soft and hard closure, assessment of corneal sensation, completeness of Bell's phenomena, and fluorescein staining for evidence of exposure keratitis or ulceration. Schirmer tear testing should be performed and is discussed under topognostic testing. Acuity should be documented by an ophthalmologist prior to eyelid surgery or in medicolegal cases. The finding of a poor Bell's phenomena, anaesthetic cornea, and dry eye requires urgent ophthalmological referral (Levine, 1986).
Oral incompetence may be associated with cheek biting, inability to wear dentures, food build up in the gingivobuccal fold and dental caries, especially in bilateral paralysis (Downs, 1986). Nasal obstruction due to nasal alar collapse is least troublesome but should be examined for.
The assessment of abnormal regeneration of motor function is important. Synkinesis is best unmasked by asking the patient to whistle and observing ipsilateral eye closure. Similarly, closing the eyes may result in elevation of the ipsilateral corner of the mouth. Following complete degeneration, axonal regrowth occurs preferentially to the middle third of the face, with axons avoiding angular pathways to the upper and lower face. Hypertonia of the middle third of the face with deepening of the nasolabial groove and "hung-up" corner of the mouth is caused by relative over-reinner vation. Likewise, the over-supply of regenerated nerves is also responsible for observed middle third hyperkinesia, tics and spasm. Relative hypotonia, weakness and even lack of recovery are seen in the upper and lower thirds of the face, typical of regeneration after severe degeneration (Croxson, 1989).
The appraisal of cosmetic defects involves integrating the patient's perception of his own appearance with the physician's assessment of weakness and recovery pattern. Full discussion is outside the scope of this review, but provides a new and exciting field of expansion for the otolaryngologist (Croxson, 1989; Conley, 1986).
Full examination focusing on eye movement (nystagmus) the trigeminal nerve, and all lower cranial nerves is important. Where relevant, tests of balance, co-ordination and gait should be included.
Topognostic/Site of Lesion Testing
Topognostic testing involves evaluating greater superficial petrosal nerve function through lacrmination, stapedius branch function through the acoustic reflex, and chorda tympani function using taste or submandibular gland salivation . The purpose of topognostic testing is to identify the site of injury of the facial nerve by examining the function of the intratemporal branches. The constant anatomical branching of the facial nerve suggested that such a system should work when Tschiassny proposed the first test protocol in 1953. Unfortunately, the experience of many workers has shown the limited value of site of lesion testing.
There are four reasons why topognostic testing may give unreliable results -
1. Acute versus chronic or longstanding lesions.
Acute injuries of the facial nerve, such as traumatic transection, affect all axons of all size and function simultaneously. Slowly progressive injuries, such as tumour compression, cause differential injury to axons, depending upon their cable diameter, the presence of myelination and their function, motor or autonomic. Longstanding injuries are associated with axons in various states of degeneration and regeneration according to their function and regeneration capacity. Hence the relevance of topognostic testing is limited to acute nerve injuries (Gantz, 1984).
2. Discrete versus diffuse injuries.
The reliability of site of lesion testing presumes discrete isolated facial nerve injury. Unfortunately the majority of facial nerve lesions do not provide such an injury. Adour (1975) believes IP results in a patchy demyelination process involving the intracranial and intratemporal portions of the facial nerve. Herpes zoster, like IP, probably causes diffuse nerve injury as well (Adour, 1975). Jahrsdoerfer (1987) has shown that temporal bone fractures are commonly complex rather than simply longitudinal or transverse with the capacity for multiple facial nerve trunk injuries or branch injury with sparing of the main trunk. May (1986) has confirmed isolated greater superficial petrosal nerve transection in temporal bone fractures with intact facial nerves. Diffuse or multiple facial nerve injury invalidates topognostic testing.
3. Difficulty with measurement techniques.
The techniques of topognostic testing will be described below. Acoustic reflex and submandibular gland secretion testing are reasonably objective, but taste testing and tests of lacrimat ion present problems with measurement tech niques which further compromise the reliability of topognostic testing.
4. Anatomical and physiological variability.
May (1986) attributes unexpected topognostic findings to individual anatomical diversity of the facial nerve pathway. Fisch (1977) reported that 61% of patients with unilateral geniculate ganglionectomy had bilateral diminished lacrimation suggesting a bilateral or central influence on lacrimation.