Abstract
Traumatic facial paralysis may have devastating effects on patients both functionally, emotionally, and aesthetically. In this chapter, the word trauma is used in its broad sense as any injury to the facial nerve caused by a noninfectious, nonmetabolic etiology. These include not only injuries secondary to penetrating and nonpenetrating accidents but also secondary to surgery, whether advertent or inadvertent. The pathophysiology and end result are the same: disruption of axonal flow with subsequent incomplete or complete mimetic muscle paralysis.
Perhaps the most important variable in treatment decision making is the duration of paralysis followed by the anatomic location of injury. The evidence on intracranial facial nerve decompression in the acute or subacute period after trauma is controversial and the literature is lacking but regardless of intra- or extracranial nerve injury, early intervention is key to provide the best outcomes. The type of intervention, whether direct coaptation, nerve grafting, or nerve transfers, relies on the physiologic salvageability of the mimetic muscles, and if feasible, will provide the best results. If this golden window period is missed, then most reconstructive efforts will be directed towards providing a new neuromuscular unit via either a muscle transfer or more commonly a functional muscle transplant.
This chapter will discuss the preoperative variables affecting the surgical choices, details of dynamic techniques, and as importantly, secondary nondynamic interventions, which are paramount to achieve optimal outcomes in this challenging yet enjoyable group of patients.
Keywords
facial paralysis, facial palsy, facial nerve trauma, nerve grafts, nerve transfers, functional muscle transplants, dynamic reanimation, paralytic lagophthalmos
Introduction
The field of facial paralysis provides the reconstructive surgeon an incredible opportunity to help patients with what is undoubtedly a devastating injury. The field intricately involves restoration of function, form, and even emotion, demanding that the surgeon use a wide spectrum of surgical tools and principles, including microsurgery, peripheral nerve surgery, and aesthetic facial surgery with the goal of restoring the patient’s appearance, as feasibly possible, to normalcy. Although traumatic facial palsy may differ in etiology and mechanism from other facial palsies, the reconstructive principles, philosophy, approach, and techniques remain quite similar. Hopefully this chapter will provide the reader a thinking process and technical tools to achieve optimal outcomes.
Epidemiology and Etiology
Each year approximately 20 of 100,000 people are diagnosed with facial palsy. Among more than 40 different etiologies, the most common include facial palsy of unknown etiology (Bell’s palsy), infections, trauma, and iatrogenic facial nerve injury during extirpative surgery. The exact incidence of traumatic facial palsy is not as clear, but is approximated at 16%. Although intraoperative facial nerve injury, whether inadvertent or during cosmetic or extirpative surgery, may not be considered as trauma in the classic sense, this chapter discusses these types of injuries due to similarities in clinical presentation, diagnoses, and management principles.
Surgical Anatomy
The facial nerve is a mixed nerve that carries fibers from and into three distinct nuclei in the midbrain. The motor nucleus originates fibers that innervate the mimetic musculature, the superior salivatory nucleus sends preganglionic parasympathetic fibers to the lacrimal, sublingual, and submandibular glands, and the solitary tract nucleus receives sensory fibers from the external auditory canal and taste fibers from the anterior two-thirds of the tongue. Upper motor neurons from the cerebrum converge via the corticobulbar tract into the facial motor nucleus in the pons. The forehead is innervated by fibers from both cortices while the lower two-thirds of the face is innervated from the contralateral side. As the fibers converge in the brain stem to synapse in the facial nerve motor nucleus, the remainder of the nerve, the lower motor neuron, innervates the entire ipsilateral hemiface. From the brain stem, the facial nerve enters the petrous portion of the temporal bone at the internal auditory meatus, coursing through the bone until it exits through the styloid–mastoid foramen. This is the longest and narrowest course of a nerve within a bone in the body, making it susceptible to compression ( Fig. 1.5.1 ). Along its intratemporal portion, the facial nerve gives off several branches, including parasympathetic fibers innervating the lacrimal gland via the greater petrosal nerve and more distally the sublingual and submandibular glands via the chorda tympani, motor fibers innervating the stapedius muscle, and taste fibers innervating the anterior two-thirds of the tongue.
Although variations exist in the extracranial facial nerve anatomy, once the facial nerve exits the stylomastoid foramen, it sends a small sensory branch to the posterior wall of the external auditory canal and tympanic membrane, followed by the postauricular nerve that innervates the posterior auricular and occipitalis muscles, and two small branches innervating the stylohyoid and posterior belly of the digastric muscles. It courses 1–2 cm and at the pes anserinus divides into two major trunks: the superior temporofacial trunk and the inferior cervicofacial trunk. The first divides into the temporal (frontal) branch with minimal arborizations and the zygomatic and buccal branches that have significantly more arborizations creating an intertwining network of nerves in the midface. The cervical trunk divides into the mandibular branch and the cervical branch with minimal arborizations between them. The frontal branch innervates the frontalis muscles and is responsible for brow elevation, anterior and superior auricular muscles, corrugator and procerus muscles, and contributions to the orbicularis oculi muscle. The zygomatic and buccal branches innervate the mimetic muscles of the midface, including the orbicularis oculi, levators of the lip, buccinator and rizorius muscles, and the orbicularis oris muscle, and depressors of the lip. The cervical trunk innervates the mentalis muscle via the marginal mandibular nerve and the platysma via the cervical division. A full review of the facial nerve anatomy and its branches may be found in any anatomy atlas ( Table 1.5.1 ).
| Branch of CN VII | Muscle | Action |
|---|---|---|
| Posterior auricular | Posterior auricular | Pulls ear up and backwards |
| Occipitofrontalis, occipital belly | Moves scalp backward | |
| Temporal | Anterior auricular | Pulls ear up and forwards |
| Superior auricular | Raises ear | |
| Occipitofrontalis, frontal belly | Wrinkles forehead and raises eyebrow | |
| Corrugator supercilii | Pulls eyebrow medially and down | |
| Procerus | Pulls medial angle of eyebrow down producing transverse wrinkles over bridge of nose | |
| Temporal and zygomatic | Orbicularis oculi | Closes eyelids gently (palpebral part) or forcefully (orbital part) and contracts skin around eye |
| Zygomatic and buccal | Zygomaticus major | Pulls corners of mouth up and lateral |
| Buccal | Zygomaticus minor | Elevates upper lip |
| Levator labii superioris | Elevates upper lip (helps form nasolabial furrow) | |
| Levator labii superioris alaeque nasi | Elevates medial nasolabial fold and nasal ala (opens nostril) | |
| Risorius | Pulls corner of mouth lateral (aids smile) | |
| Buccinator | Pulls corner of mouth backward and compresses cheek against teeth | |
| Levator anguli oris | Raises angles of mouth upward (helps form nasolabial furrow) | |
| Orbicularis oris | Closes and protrudes lips | |
| Nasalis, alar part | Flares nostrils by pulling cartilage down and laterally | |
| Nasalis, transverse part | Compresses nostrils | |
| Buccal and marginal mandibular | Depressor anguli oris | Pulls corner of mouth down and lateral |
| Depressor labii inferioris | Pulls lower lip down and lateral | |
| Marginal mandibular | Mentalis | Pulls skin of chin upward |
| Cervical | Platysma | Pulls down corners of mouth |
Based on the anatomic level of facial nerve injury, different deficits will occur. Unilateral upper motor neuron injuries will cause injuries to the contralateral lower two-thirds of the face sparing the forehead, while injuries distal to the facial nerve nucleus will affect both the upper and lower parts of the ipsilateral face. More proximal injuries may involve the lacrimal gland via the greater petrosal nerve, the auditory apparatus via the stapedial nerve, and taste via the chorda tympani and lingual nerve, in addition to motor deficits. More distal injuries may spare the former and create purer motor deficits and small sensory deficits. Knowledge of this anatomy is consequential in both the diagnosis and localization of the facial nerve injury as well as the treatment plan.
History and Physical Exam
A thorough history and physical examination is critical in the evaluation and treatment plan. Establishing the duration, mechanism, and pretraumatic facial nerve function, especially in older patients, is pertinent. Age, comorbidities, and previous surgical history will also partially dictate the type of future intervention.
In multitrauma cases, initial assessment must include a full trauma evaluation for other injuries. Low Glasgow Coma Scores suggest traumatic brain injuries and possible skull base fractures affecting the facial nerve. External auditory canal otoscopic examination can reveal a step deformity or laceration. If hemotympanum or tympanic membrane perforation with blood or CSF is seen, longitudinal fracture of the temporal bone should be suspected. Conductive hearing loss is also generally associated to longitudinal temporal fractures while sensorineural loss is most commonly associated to transverse fractures. Regional facial examination for resting tone and voluntary activity is necessary. Thorough mimetic muscle assessment from forehead to neck is crucial. Whether an intracranial or extracranial nerve injury is suspected, any facial motion suggests a contusion rather than laceration and implies continuity of nerve and excellent prognosis. In midface penetrating injuries, parotid duct injuries must be excluded. Additionally, remaining cranial nerves must be evaluated for both injury and their potential as donor nerves for reanimation.
Clinical presentation may suggest the level of facial nerve injury. Injuries proximal to the midbrain nuclei, termed upper motor neuron injuries, will spare the forehead due to the bilateral contributions of the cerebral hemispheres yet produce contralateral lower two-third paralysis. In the trauma setting these injuries usually result in death or preclude rehabilitation, but patients must be assessed individually. Injuries inclusive and distal to the midbrain nuclei will result in complete ipsilateral facial paralysis. The more proximal the injury, the more fiber types are included, and symptoms will include decreased lacrimation, hyperacousis, changes in taste, sensory, and motor deficits. The more distal the injury, the resultant injury is more localized. Hence complete knowledge of facial nerve anatomy is paramount in the diagnosis of facial nerve trauma. Midface lacerations involving proximal buccal and zygomatic nerve injuries will mostly cause midface paralysis unless they are isolated to single branches, in which cases the injury may be unnoticeable due to arborization between the branches. In this setting, concomitant parotid duct injury should be ruled out.
Special attention needs to be given to patients suffering from high-energy injuries involving blast or high velocity projectiles or patients with multilevel injuries as seen in animal bites. These patients usually present with facial nerve injuries at several levels, coinciding with soft tissue trauma including muscle, skin, mucosa and bone.
Ancillary Testing
CT Scans
Temporal bone fractures are a product of high-energy blunt trauma commonly resulting in fracture, hemorrhage, nerve trauma, vascular damage, with disruption of the middle or inner ear structures, classically classified into longitudinal, transverse, or oblique fractures. Longitudinal fractures often result from lateral to medial forces extending through the facial nerve canal, possibly causing intraneural hemorrhage, transection, or bone compression. They can disrupt the ossicular chain, resulting in conductive hearing loss. Transverse fractures often result from anterior posterior forces with a fracture line often traversing the vestibulocochlear apparatus causing sensorineural hearing loss and equilibrium disorders. Transverse fractures more commonly injure the facial nerve due to proximity to the nerve’s labyrinthine segment. Oblique, also termed mixed, fractures include both longitudinal and transverse components. Additional classifications are based on degree of involvement of the petrous portion of the temporal bone, or the otic capsule. Temporal bone computed tomography (CT) scans should be performed in thin-section 1 mm cuts to avoid interpreting normal suture lines as fractures.
MRI
Magnetic resonance imaging (MRI) has lower sensitivity and specificity in depicting temporal bone fractures compared to CT scans. It may demonstrate fluid in the middle ear and mastoid air cells as a high signal on T2-weighted images or suggest hemorrhage, revealed by a bright signal in the labyrinth or middle ear. Recent MR neurography has some promise in depicting intratemporal facial nerve injury.
ENoG
Electroneurography is an objective test that measures evoked compound muscle action potentials using skin electrodes. Nerve injury is expressed as percentage of function relative to the healthy contralateral side. It is preferable to perform ENoG only after 3 days, when Wallerian degeneration is completed otherwise there is a risk for a false negative since conduction will still occur. Conversely, ENoG is of little value if performed after 21 days since interpretation is unreliable. Optimally it should be performed at 3 days followed every 3–5 days to demonstrate a trend. Some surgeons use a decrease of 90% on the injured side as an indication for temporal bone decompression, although this remains controversial.
EMG
Electromyography measures voluntary mimetic muscle response with needle electrodes detecting action potentials during muscle contraction. Denervated muscles display fibrillation potentials while muscle that are reinnervating demonstrate polyphasic potentials. Electrical silence indicates severe muscle atrophy and degradation of motor end plates. EMG is best performed at least 10–14 days after injury to allow Wallerian degeneration to set in. The value of the EMG in the acute facial nerve injury period is unclear but may have some value in the subacute period, up to 12–18 months, possibly helping decide between nerve transfers relying on salvageability of the patient’s native mimetic muscles versus using a new muscle for reanimation. A more detailed discussion will follow.
Location, Mechanism, and Duration of Injury – Effects on Management Strategies
The evaluation of a new patient with facial palsy must include the three crucial factors in future treatment decisions: location, mechanism, and perhaps most important, the duration of paralysis. Location can be grouped into intracranial or extracranial, mechanism into blunt or penetrating, and duration into acute, subacute, and chronic.
Location and Mechanism of Injury
Intracranial Facial Nerve Injuries
Most intracranial nerve injuries occur secondary to blunt trauma. Penetrating injuries are usually associated to gunshot wounds and are frequently lethal. The mechanism is usually either complete transection of the nerve or irreversible compression injury due to collapse of the osseous canal within the petrous portion of the temporal bone. The zone of injury is usually beyond the area of transection and difficult to assess ( Fig. 1.5.2 ).
Blunt injuries resulting in temporal bone fractures are often encountered in motor vehicle accidents, altercations, or falls from heights. Approximately 7%–10% of temporal bone fractures result in facial nerve injury. Temporal bone fractures have several classifications. Fracture line orientation relative to the petrous bone defines fractures as longitudinal (70%–80%), transverse (10%–20%), and oblique (10%). Facial paralysis occurs most commonly in transverse fractures (50%) but may also occur in longitudinal fractures (25%). More modern CT-based classifications assess whether fractures are otic capsule sparing or violating, the latter being twice as likely to cause facial paralysis. Four types of facial nerve trauma have been found in temporal bone fractures. In 76% of longitudinal fractures either bony impingement or intraneural hematoma was found, and in 15% the nerve was transected. In the remainder of patients, no visible pathology was found other than neural edema. In transverse fractures, 92% were transected and 8% had impingement ( Fig. 1.5.3 ). Similar to penetrating intracranial facial nerve injuries, surviving patients are initially often in critical condition and the facial paralysis is often unnoticed.
Extracranial Facial Nerve Injuries
Extracranial penetrating trauma may include either low-energy penetrating objects, higher-energy projectiles or blasts, or a combination of sharp, crush, and avulsion injuries.
Low-energy penetrating trauma, commonly secondary to sharp objects as knives or broken glass, usually results in localized cuts through the facial nerve branches resulting in paralysis of one or more of the facial nerve branches. These injuries are mostly clean, associated with minimal nerve gaps, and frequently result in well-aligned proximal and distal nerve endings. These injuries may span proximally from the styloid foramen to the small distal branches just proximal to the mimetic muscles. Proximal injuries at the level of the pes anserinus result in complete unilateral facial paralysis while more distal injuries may result in more regional palsies or none. Injuries at nearly any level of the frontal or marginal mandibular branches will result in complete paralysis of the forehead or depressors of the lip due to the lack of arborization with synergistic nerve branches, while injury to the zygomatic or buccal branches is more forgiving due to increased branch arborization.
Combined sharp, crush, or avulsion injuries, as often seen in animal bites, high-energy projectiles, or blasts, are usually characterized by contaminated, multilevel traumatic injury that may involve multilevel facial nerve injury, often creating long nerve gaps with or without muscle injury or overlying tissue deficiencies. Blunt injury to the extracranial portion of the face is a rare cause of facial paralysis, and will usually result in temporary neuropraxia and complete recovery.
Muscle Injury
Although not a common cause of facial paralysis, muscle injury may have similar effects and clinical presentation as nerve injuries, especially in the setting of acute trauma, in which both muscle and nerve may be contused. Over time, when the nerve recovers the muscle injury usually becomes more apparent since motion is seen in the proximal innervated muscle but animation is inhibited because the distal muscle is detached and will not produce the intended motion at the muscle insertion. These injuries usually result in more focal akinesia compared with proximal nerve injuries ( Fig. 1.5.4 ).
Duration of Paralysis, Pre-Injury Function, and Choice of Surgical Techniques
The most important variable when evaluating facial palsy patients is the duration of paralysis and degree of function prior to injury. As opposed to patients whose facial palsy is due to neoplasms, infections, or Bell’s palsy, in whom palsy duration is not always clear, most facial nerve trauma patients have no history of prior paralysis, are frequently younger, and palsy duration is directly traced to the time of injury. Duration may be divided into three quite indistinct periods: acute (0–72 hours), subacute (up to 6–12 months), and longstanding facial paralysis (beyond 12–18 months). Differentiation should be more physiological than chronological. Distinction between acute and subacute facial paralysis is somewhat artificial since management may be similar and depends more on the type of injury rather than duration of paralysis, but common to both is the potential to salvage the patient’s mimetic muscles, as opposed to longstanding facial paralysis, when a new neuromuscular unit must be introduced as part of the treatment strategy due to mimetic muscle denervation atrophy.
Acute Facial Paralysis
Historically, acute facial palsy may be defined as a period between the time of trauma up to 72 hours, after which Wallerian degeneration begins and stimulation of the distal nerve branches is not feasible. This is likely the period in which treating the paralysis with either direct nerve repair, nerve grafting, or nerve transfer will yield the best results although results may vary with each of these techniques depending on the mechanism and location of facial nerve injury.
Acute Intracranial Facial Nerve Injuries
Decompression.
The use of steroids in these cases has not been demonstrated to confer the benefit it has in the setting of Bell’s palsy, therefore the benefits versus risks of infection should be weighed. There is scant high-level evidence as to whether acute intracranial facial nerve decompression helps prevent facial paresis, and only few reports support this approach and are mainly retrospective, lack controls, and suffer from small patient numbers. The main difficulty in achieving high-level evidence is the difficulty in comparing a control group whose natural history is often favorable, to an intervention group, proving the latter fare better. Additionally, these injuries are frequently in the setting of traumatic brain injuries with most patients intubated and either intentionally sedated or in comatose conditions, rendering early clinical assessment of facial nerve function difficult. If examination is possible, sensorineural hearing loss bears a poorer prognosis than conductive hearing loss and partial facial palsy is highly predictive of full recovery versus poorer prognosis with complete palsy. Overall, many of these patients recover facial nerve function without intervention.
ENoG, if performed within 14 days of injury and demonstrating more than 95% degeneration compared to the normal side, has been suggested by some authors as a criterion for surgery, but no strong supportive data exists. In most cases, especially when several weeks have passed since the trauma, treatment is directed more by a combination of high-resolution CT scans and EMGs demonstrating fibrillation potentials, and is often based on the surgeon’s personal opinion and experience with intracranial decompression. Technical details of intracranial nerve decompression are beyond the scope of this text and are more relevant for the neuro-otologist than the plastic and reconstructive surgeon, but close collaboration is paramount.
Primary nerve repair.
In the setting of trauma, primary neuroraphy would be a rare scenario, since most injuries would necessitate nerve grafting due to some degree of zone of injury to the nerve and the inability to mobilize the nerve sufficiently and perform a tension-free coaptation.
Nerve grafting.
In cases of intracranial trauma with laceration and discontinuity of the facial nerve, acute nerve grafting may be considered. The literature is scant and the situation is further complicated by the difficulty in assessing the zone of nerve injury. If the injury is close to the brain stem, a proximal nerve may not be available for coaptation, rendering grafting impossible. If a nerve graft is optional, suturing may be difficult and using fibrin glue may be preferable, although there is no evidence of superiority of one technique over another. Additionally, current literature on immediate intracranial nerve grafting in the setting of tumor resection demonstrates poor functional results, albeit restoration of some facial tone, especially in the perioral, periorbital, and buccinator muscle areas, but the trauma setting provides even less optimal conditions for immediate nerve grafting.
Acute Extracranial Facial Nerve Injuries
Primary repair.
Primary neuroraphy is always the treatment of choice if feasible and provides the best results. The advantages of repair at this stage are twofold. Distal branches are still stimulable prior to onset of Wallerian degeneration and identification of branches is easier in the unscarred environment. Two conditions must be met for primary neuroraphy to result in successful outcomes. The first is that both proximal and distal nerve endings must exist. The second is that direct nerve coaptation must be absent of tension. Any tension on repair will uniformly result in scarring and inhibit axonal regeneration through the coaptation. Prior to induction, communicating with the anesthesiologist to avoid the use of paralytics is important and local injection should include only epinephrine for hemostasis. Depending on the location of injury, exploration may be performed via the wound edges or a preauricular incision. A nerve stimulator with several options of amperage and frequency enabling repetitive stimulation and tetany is necessary for accurate facial nerve branch mapping. Ragged nerve edges may be trimmed with a fresh blade while cutting on a sterile wooden tongue blade. Repair of midface branches, including the zygomatic and buccal branches, results in better function than repairs of the marginal mandibular and temporal branches, although repair of the latter is recommended in the acute setting.
Nerve grafting.
Where a nerve gap exists despite mobilization of the nerve endings, the second-best option is nerve grafting the gap. The options for nerve grafting are abundant. Where only one branch needs repair, the great auricular nerve may be used since it is usually within the same operative field, but more commonly several branches may need repair, in which case a sural nerve would provide more graft material. The advantage of the latter is that one nerve may be either cut into several longitudinal nerve segments or if size mismatch exists, i.e. the sural nerve is too large, the sural nerve can usually be safely neurolyzed along its fascicles for relatively long distances, providing several fascicles of decreased diameter for better size match. Additional potential donor nerves are the medial or lateral antebrachial cutaneous nerves.
Whether direct nerve repair or nerve grafting is performed, coaptation is usually performed with either 9-0 or 10-0 nylon under microscopy. Prior to coaptation, the nerve endings may be cleared from any surrounding connective tissue to alleviate identification of the epineurium and prevent any redundant tissue disrupting the coaptation. The key is to approximate the nerve ending via epineural sutures, minimizing fascicular injury and avoiding tension. If a fascicle has “escaped” between the nerve endings, it may either be gently pushed back or shortened and reinserted into the coaptation. Some authors use fibrin sealants to minimize the need for sutures in coaptation, but the potential exists that glue may infiltrate the interface and inhibit axonal growth.
In cases where a proximal nerve is not found or is too close to the styloid foramen, unroofing the mastoid portion of the nerve is warranted to enable nerve grafting. If a proximal nerve is not found, then nerve transfers are justified, enabling the use of native mimetic muscles. In younger patients (under 60), concomitant cross facial nerve grafting can be performed, with future planning to perform selective end-to-end coaptation once axons have reached the contralateral side or performing end-to-side coaptations in the primary surgery. In rare cases when a distal branch is not found, but a proximal branch is available, the latter may be used to innervate a functional muscle transplant.
Nerve transfers.
Although acute nerve transfers in the trauma setting are uncommon, they may be used concomitantly in cases where planned extirpation of the facial nerve is performed and nerve grafting is unlikely to yield optimal results or is not feasible. Nerve transfers are used much more frequently in the subacute phase and are detailed below ( Fig. 1.5.5 ).
