By Peter R. Sabatini M.D.

Faculty Sponsor: Yadranko Ducic M.D., F.A.C.S.

Cranial nerve VII, the facial nerve, primarily provides motor input to the muscles of facial expression. The input that the facial nerve provides can be interrupted in a variety of ways and the end result is predictable.  Facial nerve paralysis, depending on the etiology of the injury can manifest in an array of clinical presentations. The most recognized and accepted way to characterize facial paresis is the House-Brackman scale (Fig 1).  The focus of this discussion will be on the treatment of complete facial paralysis or HB VI/VI.

Sunderland classified peripheral nerve injury in terms of severity on a scale of I to V (Fig 2).  Class I injury termed neuropraxia is equivalent to a temporary conduction block that is secondary to external pressure on the axon. Recovery is swift after the insult is removed and there is return to normal function (ie. arm going to sleep).  Class V injury is complete transection of the nerve, with the others falling in between.  This paper will be focused on Class IV and Class V injuries in which return to normal function is unlikely or impossible without intervention. 

Facial paresis can result from blunt or sharp trauma, intentional and unintentional surgical transection, viral infection, congenital conditions, Lyme disease, acute otitis media and sarcoidosis. An overview of how the facial nerve can be tested is helpful in understanding the algorithm for reanimation procedures. The nerve itself can be stimulated electrically and the end result, muscle contraction, observed. This is the basis for the most common tests, the Nerve Excitability test (NET), Maximum Stimulation test (MST), and Electroneurography (ENOG). It is important to note that all these tests may record false positive results (i.e. muscle contraction) if they are performed within 72 hours of paresis. This is the approximate amount of time for Wallerian degeneration to occur. Wallerian degeneration is the retrograde and anterograde degeneration of the nerve's axonal sheath. This phenomenon takes place with all injuries of Sunderland Class II or greater.

Briefly, the NET is performed by placing a stimulating electrode over the stylomastoid foramen, a return electrode on the forearm and giving 0.3 msec pulses of electricity in increasing amounts until a muscle twitch is seen.  This is performed on the non-paralyzed side first and compared with the paralyzed side. A significantly abnormal result is greater than 3.5mA difference between sides. This predicts, in Bell's palsy with complete clinical paralysis, whether or not the patient can expect a complete recovery (80% accuracy).

The MST is set up similarly to the NET except in this test the current at which the maximum twitch is seen is recorded and compared between sides. It is hypothesized that in this way one could estimate the percent of nerve fascicles that were damaged. This test is not commonly used because it is very uncomfortable for patients and does not give superior prognostic information.

The ENOG is different in that it uses a bipolar electrode system. The stimulating probe is inserted transcutaneously at the stylomastoid foramen and the nasolabial groove. Maximum electrical stimulation is used and the amplitude of muscle contraction is recorded and compared between sides. A difference of greater than 90% correlates with a poor prognosis for meaningful recovery without surgical intervention.  This test, and the NET, is most commonly used in conjunction with another important confirmatory test, electromyography.

Electromyography (EMG) is a test that records muscle potentials. A recording needle is placed into the muscle bed and the patient is asked to move their face. If the needle records voluntary muscle potentials then prognosis for recovery is excellent for that muscle group. Other possible results of EMG in the setting of facial paralysis include fibrillation or denervation potentials, which confirm that the nerve connection to that muscle group is no longer viable. Polyphasic innervation potentials are seen after enough time has passed for reinnervation to occur and electrical silence indicates non-functional muscle that can no longer be recovered by reinnervation. 

Facial reanimation is the practice of restoring symmetry and function to a patient with a debilitating and disfiguring paralysis. The art lies in deciding upon the right treatment plan for the patient.  Factors such as patient age, prognosis, time since injury, type of injury, nutritional status, and history of radiation therapy all play a role in the decision making process. 

Obviously, the most desirable outcome in facial reanimation surgery would be complete return of pre-injury function and aesthetics. Unfortunately, this can rarely be achieved and the algorithm for facial nerve repair begins with the optimal procedure, direct reanastomosis of the injured nerve and proceeds to static procedures that offer no voluntary function but restore symmetry.  Where each patient fits in is the key to preoperative planning.  It is important to note that in cases where the etiology of facial paralysis is unclear, appropriate radiological images should be obtained to rule out a malignancy. This includes a CT scan of the temporal bone, an MRI of the Parotid gland and an MRI with gadolinium of the cerebellopontine angle and internal auditory canal.

The simplest solution to an acute transection of the facial nerve is reanastomosis. This is possible if both ends are readily identifiable and a tension-free anastomosis can be performed. In the case of a poly-trauma patient where immediate anastomosis is not possible it is important that the distal facial nerve branches be located with a nerve stimulator and tagged before 72 hours if possible. After 72 hours, Wallerian generation will have taken place and the nerves will no longer stimulate a muscle twitch, making them difficult to locate and repair.  Direct neurorraphy is the most desirable repair for an acute facial nerve transection.

In cases where it is not possible to perform direct anastomosis secondary to inability to achieve a tension-free anastomosis, the next step in the algorithm would be to perform cable grafting from the functioning proximal nerve stump to functional distal facial nerve branches. Three peripheral nerves are most commonly used to achieve this purpose: the greater auricular nerve, the sural nerve, and the medial antebrachial cutaneous nerve.

The greater auricular nerve offers a similar diameter to the facial nerve and a favorable distal branching pattern. The nerve contralateral to the paralyzed side is taken. The greater auricular nerve offers up to 10 cm of length. Donor site morbidity includes numbness of the skin overlying the angle of the mandible to the lobule of the ear.

The sural nerve is by far the nerve that offers the most length, up to 70cm. It is also advantageous in that it lies distant to the site of injury and facilities a two team approach, saving the patient from a more prolonged general anesthetic time. Donor site morbidity manifests as numbness over the lateral aspect of the leg. This is especially important to remember in patients with long-standing diabetes or peripheral vascular disease, as they will need to pay extra attention to this area.

The medial antebrachial cutaneous nerve, like the greater auricular nerve, has a similar diameter and favorable branching pattern. Up to 20 cm can be harvested.

When faced with limited branching of the donor nerve there is a certain priority that is given to the distal facial nerve branches. The buccal and zygomatic branches take equal first priority, as they will affect the all important orbitozygomatic and orobuccal complexes. Second in importance is the marginal mandibular branch followed by the frontal and then cervical branches.

If cable grafting is not feasible due to lack of functional proximal facial nerve then a nerve transfer procedure would be in order. This involves an anastomosis between all or some of the ipsilateral hypoglossal nerve.  The ipsilateral nerve can be exposed through a standard parotid-type incision and transected and anastomosed directly to the facial nerve stump. An alternative involves harvesting a cable graft and anastomosing it to about 30% of the hypoglossal nerve leaving the other 70% intact. Similar results and less morbidity are theoretically possible by this method. Nerve transfer procedures hope to achieve static tone and some mass facial movement.

Electromyography guides much of the decision making process when planning facial rehabilitation procedures. The jump from nerve transposition grafting to the next preferable option, muscle transfer, is made when there is no evidence of viable distal facial musculature.  This corresponds with electrical silence on EMG. Just as polyphasic potentials preclude a reinnervation procedure, fibrillation potentials preclude a muscle transfer procedure. Muscle transfer procedures most commonly involve the masseter and temporalis muscles.

Masseter transfer is generally useful for rehabilitation of the buccal-smile complex and for a sagging oral commisure.  It may be approached intraorally or through a facelift incision.

Temporalis transfer may be used for the orbitozygomatic and buccal-smile areas because of its wide base of insertion. 

Microneurovascular transfer is another option that involves transfer of a free muscle graft along with its neurovascular pedicle.  This may be utilized in the small subset of patients with absent distal facial nerve fibers. The graft is harvested (e.g. Lattisimus dorsi, gracilis, inferior rectus…) and sutured to the contralateral facial nerve.

In patients who are poor surgical candidates or have a poor prognosis, static facial procedures offer a good immediate option for return of symmetry and some function. Static procedures are also used in conjunction with other dynamic procedures to provide immediate symmetry while reinnervation occurs. Examples of static procedures include static facial slings with alloderm, gore-tex, fascia lata, or suture suspension. The above are typically used for a sagging oral commisure or external nasal valve collapse.  The sling is anchored at the zygomatic arch with a suture or with a screw.

Finally, other adjunctive procedures can be included in the care of the patient with facial paralysis. Endoscopic brow lifts may relieve lid ptosis, canthoplasty can relieve epiphora that is associated with ectropion.  Insertion of gold weight is essential in the initial care of a patient with complete facial paralysis to protect against exposure keratitis.  Alar batten grafting can provide support for external nasal valve collapse. A mid-face lift can help regain some lost facial symmetry.

References:

Lore` JM, Medina JE (2005). An Atlas of Head & Neck Surgery. Fourth edition. Philadelphia: Elsevier. 367, 381-391

Cummings CW, Flint PW, Haughey BH, Robbins KT, Thomas JR, Harker LA, Richardson MA, Schuller DE (2005). Cummings Otolaryngology Head & Neck Surgery. Fourth edition. Philadelphia: Elsevier. 822-852