Facial reanimation by cross-facial nerve grafting: Report of five cases. (Original Article).

Abstract

Facial nerve repair is a dynamic reanimation technique. Direct nerve repair by suturing or grafting can provide good results within a specific time frame. Immediate nerve repair has been successful in cases of laceration injuries, but nerve grafting techniques are typically delayed when it is clear that direct suturing to the nerve trunk cannot be achieved without tension. Delayed nerve grafting is also employed following ablative procedures and in cases of trauma that cause segmental nerve deficits. Cross-facial nerve grafting is particularly useful when the peripheral branches are intact and the main trunk of the facial nerve is inaccessible. This method is also typically performed in a delayed fashion. Rehabilitation of the facial nerve and subsequent reinnervation of the mimetic motor endplates are achieved through axonal growth. In this article, we describe a consecutive series of five patients who developed facial paralysis following cranial surgery for acoustic neuroma. Each underwent successful cross- facial nerve grafting during the first week following their initial surgery. Each received a sural nerve graft to at least two main divisions of the VIIth cranial nerve. We discuss our operative technique and the degree of restored nerve function.

Introduction

A unilateral loss of facial movements causes not only disfigurement but dysfunction. At rest there is an unnatural relaxation of the eyelid, the cheek, and the corner of the mouth, which can lead to dry eye, ectropion, alar collapse, nasal obstruction, muffled speech, and drooling. Cosmetically, the face is asymmetric, as the eyebrow and eyelids droop, the frontal creases are lost, the cheeks and lips sag, and the nasolabial fold flattens.

The muscles of facial expression are under the control of the VIIth cranial nerve. Their movement not only reflects a person's emotion, they maintain ocular, nasal, and masticatory function. The loss of these functions can be debilitating.

Techniques to address these deficits include both dynamic (e.g., nerve grafting and muscle transposition) and static (e.g., fascial and alloplastic slings) procedures. The goal of surgical intervention is to achieve an aesthetic result at rest while enhancing sphincter control (ocular, nasal, and oral). Each technique is limited, yet each can provide some restoration of both form and function. A combination of techniques can optimize results. The choice of treatment plan is dependent on the type of facial nerve injury, the resultant deficit, the prognosis for recovery, and the wishes of the patient.

Nerve repair is a dynamic reanimation technique that effects functional recovery by reinnervating the muscles of facial expression. (1) Direct suturing of the nerve endings is best accomplished immediately following the injury. In cases of laceration (as a result of either trauma or iatrogenic injury), suturing should take place prior to the development of fibrosis and scarring. (2) When direct suturing to the nerve trunk cannot be achieved without tension or because an ablative procedure or trauma has resulted in segmental nerve deficits, delayed nerve grafting techniques should be employed.

Cross-facial nerve grafting was first described by Scaramella in 1971. (3) His basic idea was to stimulate the facial nerve branches on the paralyzed side with nerve energies from the nonparalyzed side of the face. He accomplished this with a cable nerve graft of the sural nerve. In this article, we discuss our operative technique and the degree of restored nerve function in a series of five patients.

Patients and methods

We performed cross-facial nerve grafting on two men and three women, aged 32 to 53 years, who had sustained an iatrogenic nerve trauma during neurosurgery (table). Each of these patients had undergone surgery to remove an acoustic neuroma in a single-stage procedure. In each case, surgery caused damage to the facial nerve, which resulted in a complete paralysis of the operated side of the face (figures 1, 2, and 3). No patient experienced a disruption of the facial nerve at its connection to the brainstem, and none experienced any decrease in function on the contralateral side.

We performed the grafting on all five patients within 1 week of their neurosurgery. The distal branches of the facial nerve were isolated by a bilateral parotidectomy incision. With a nerve stimulator, the branches of the facial nerve on the nonparalyzed side were dissected; specific attention was directed at the zygomaticofrontal, buccal, and even marginal mandibular branches. The dominant branches and strong secondary branches were identified. The strong secondary branches were chosen for grafting in order to prevent a postoperative deficit on the donor side (figure 4). Next, the branches of the paralyzed side were dissected, and the nerve stimulator was used to map the corresponding branches. When there was minimal stimulation, anatomic matching was performed. Branches on both sides were tagged. The dominant branch from the functional eye was not grafted in any patient.

A sural nerve graft of sufficient length was obtained in the standard manner. Both legs were prepped in case one sural nerve was not sufficient. Most patients required two or three grafts. In order to place the graft, a subcutaneous tunnel was created above the upper lip, into which a 4.5 pediatric endotracheal tube was placed. The sural nerve grafts were reversed from their anatomic orientation and passed through the tube. The tube was removed, and the sural nerve grafts remained in the subcutaneous tunnel. Using an operating microscope, we dissected the fascicles and performed neurorrhaphy to the size-matched branches. Epineural suturing was performed with a 9-0 nylon microsuture. A gold weight was placed in the paralyzed eyelid at this time.

Results

Follow-up ranged from 3 months to 3 years. Postgraft functional status was graded on a scale of 1 (no improvement) to 5 (restoration of normal function). Although no patient achieved a return to completely normal function, most did attain acceptable results in several branches (table). Following graft surgery, all patients experienced some discomfort at the donor site, especially those who had grafts taken from both legs.

Discussion

Scaramella first performed his procedure--anastomosing a branch of the normal pes anserinus to the trunk of the paralyzed facial nerve--in 1968 on a patient who had undergone resection of an acoustic neuroma. (3) At that time, the nerve graft was tunneled submentally to link the platysmal branch of the normal side to the trunk of the paralyzed side. In his second case, Scaramella used the buccal branch for an anastomosis with a sural nerve graft to the paralyzed facial nerve trunk. (4)

In 1996, Scaramella published a report of his results on 11 patients. (5) Those results were variable with respect to tone and function. Five of the 11 patients had good tone, two fair, one poor, two were deemed failures, and one patient experienced increasing weakness over time. Some of these patients did experience some restoration of some function. In most of these cases, a nerve was grafted from the cervicofacial branch to the trunk of the paralyzed side. These patients had undergone repair between 5 weeks and 1 year after they had sustained their initial nerve injury. Those patients who had been operated on at 5 and 6 weeks postinjury experienced some restoration of some tone and some function. Scaramella later modified his technique to include an anastomosis between the ansa cervicalis and the ipsilateral paralyzed cervicofacial branch of the VIIth cranial nerve.

In 1971, Smith described his technique for performing cross-facial grafting by placing a supralabially tunneled sural nerve graft between a buccal-zygomatic branch and a peripheral branch on the paralyzed side. (6) He reported improvements in symmetry in all three patients he treated. Two of these patients had been treated 6 months after they had sustained a facial nerve injury in automobile accidents. The other patient, who had developed a facial palsy' during infancy, received a VIIth-to-XIIth nerve crossover graft at the age of 8 years. Even this patient experienced some improvement in symmetry.

Anderl reported his results with four sural grafts in 23 patients with facial paralysis. (7,8) He placed separate anastomoses between the branches in different stages. A graft was tunneled subcutaneously in the frontal region to link fascicles from the zygomatic donor site to the paralyzed side. After 4 to 6 months, the neuroma on the transplanted end (considered a sign of regeneration) was amputated, and the sural graft was anastomosed to the paralyzed zygomatic segment. Likewise, two sural grafts were tunneled subcutaneously above the lip to link different groups of fascicles to the paralyzed side, with the final anastomosis to the buccal segments placed 4 to 6 months later. The marginal segment was treated in a similar manner. These patients had been treated at various intervals following their initial injury, ranging from 3 months to 19 years. Poor results (no or little symmetry, minimal muscle action, and/or poor lid closure) were observed in only four of the 23 patients; these four patients had been tre ated 8 months, 10 months, 17 years, and 19 years following their initial injury.

Baker and Conley reported their results on 10 patients in 1979. (9) Their technique required that a sural nerve graft be sutured to the proximal and distal ends of the cervical division of the donor side. The graft was tunneled supralabially and then anastomosed to the main trunk of the paralyzed nerve. Only six of the 10 patients experienced improvement, and that improvement was rated as only fair. The exact details as to the timing and nature of the initial nerve injuries were not noted.

In 1993, Inigo et al reported on the treatment of patients with hemifacial microsomia with a cross-facial technique in which the sural nerve was anastomosed to the donor side and tunneled supralabially; then its fascicles were sutured directly to perioral muscles. (10) These patients had been diagnosed with facial paralysis between 5 weeks and 10 months of age, and all were treated before they reached the age of 1 year. All patients experienced excellent results with regard to recovery of movement.

With our technique, the sural graft is tunneled supralabially, and sural fascicles are anastomosed to size-equivalent branches on the donor and the paralyzed sides. Each of the five main branches of the facial nerve is variable in its orientation and with respect to the number of secondary branches that stem from each. A theoretical risk of cross-facial nerve grafting is the disruption of the innervation on the donor side. For this reason, we performed intraoperative nerve mapping, which allowed us to identify the "dominant" branch so that only secondary branches would be used for anastomosis. Additionally, there is a component of shared innervation between the main branches such that the buccal branch might actually contribute to the zygomaticofrontal region. Therefore, during the cross-facial anastomosis, a component of shared reinnervation might contribute to the overall result. With our technique, the sural nerve graft is placed in the reverse orientation. This facilitates the group fascicular repair to a llow nerve regrowth.

In the earlier reported cases of cross-facial nerve grafting, surgery was often delayed for months--and in some cases years--following the initial nerve injury. In our series, the nerves were grafted within I week of the initial damage. In each case, the patient had undergone cranial surgery that had caused irreversible damage to the VIIth nerve, which led to a complete facial paralysis. Our early intervention allowed for nerve regrowth prior to damage to the motor endplate and subsequent muscle atrophy. Additionally, early intervention allowed for maximal preservation of nerve integrity on the injured side. Although our patients were allowed some time to recover from their neurosurgery, the residual edema of the damaged nerve was still present, and this allowed for ease in suturing the nerve to the graft.

Traditional teaching holds that the repair of lacerations should be performed immediately, an opinion that stems from observations made during World War. (11) Although primary repair is the treatment of choice, much controversy exists as to the exact timing of nerve repair with respect to axonal growth and the eventual electrophysiologic results. (12) Metabolic activity in injured nerves is greatest at 21 days postinjury, and therefore some authors recommend that attention to injured nerves be delayed until then. (13) To the contrary, Barrs found that axon counts in delayed grafts following nerve transection were actually lower and that there was no evidence of increased nerve growth in repairs that had been performed on day 21. (14) Ferreira et al performed Anderl's two-stage procedure on 38 patients. (15,16) Although they noted individual variations in surgical results, they did conclude that patients who were treated within 6 months of sustaining their initial injury subsequently experienced better control of their facial movements. Pulec found the same good results with grafts that were placed up to 2 years after the initial trauma. (17,18)

The House-Brackmann scale (Ito VI) is the standard for grading facial paralysis following acoustic neuroma resection. (19) However, although this scale is adequate for assessing postinjury function, it has its limits with respect to assessments of reinnervation. For this reason, we evaluated our results on an anatomic basis with a brief description of function according to each segment that was addressed surgically (table).

The disadvantages of cross-facial nerve grafting have been summarized by Baker and Conley. (9) Certainly, the surgical intrusion into the normal side of the face is a significant disadvantage, but protection of the normal side might reduce the degree of axonal input that is necessary for an optimal result. Obviously, no nerve grafting technique is perfect. Patients might experience mass movement and synkinesis, and the reinnervated side might never appear to be completely normal again. Even so, a patient stands only to gain from increased tone, improved symmetry, and perhaps even purposeful movement. Cross-facial nerve grafting remains a powerful tool in addressing facial paralysis.

From the Department of Otolaryngology, New York University, New York City (Dr. Galli); and the Department of Plastic Surgery (Dr. Valauri) and the Department of Otolaryngology (Dr. Komisar), Lenox Hill Hospital, New York City.

References

(1.) Hoffman WY. Reanimation of the paralyzed face. Otolaryngol Clin North Am 1992;25:649-67.

(2.) Coker NJ. Management of traumatic injuries to the facial nerve. Otolaryngol Clin North Am 1991;24:215-27.

(3.) Scaramella LF. L' anastomosi tra i due nervi facciali. Archivio Italiano Otologia, Rinologia, Laringologia e Patologia Cervicofacciale 1971 ;82:209- 15.

(4.) Scaramella LF. On the repair of the injured facial nerve. Ear Nose Throat J 1979;58:127-33.

(5.) Scaramella LF. Cross-face facial nerve anastomosis: Historical notes. Ear Nose Throat J 1996;75:343, 347-52, 354.

(6.) Smith JW. A new technique of facial animation. Transactions of the Vth International Congress of Plastic and Reconstructive Surgery. Melbourne: Butterworth, 1971:83-4.

(7.) Anderl H. Cross-face nerve transplant. Clin Plast Surg 1979;6:433-49.

(8.) Anderl H. Reconstruction of the face through cross-face nerve transplantation in facial paralysis. Chir Plast 1973;2:17-46.

(9.) Baker DC, Conley J. Facial nerve grafting: A thirty year retrospective review. Clin Plast Surg 1979;6:343-60.

(10.) Inigo F, Ysunza A, Ortiz-Monasterio F, Drucker-Colin R. Early postnatal treatment of congenital facial palsy in patients with hemifacial microsomia. Int J Pediatr Otorhinolaryngol 1993;26:57-66.

(11.) Gidley PW, Gantz BJ, Rubinstein JT. Facial nerve grafts: From cerebellopontine angle and beyond. Am J Otol 1999;20:781-8.

(12.) Terris DJ, Fee WE, Jr. Current issues in nerve repair. Arch Otolaryngol Head Neck Surg 1993;l19:725-31.

(13.) McCabe BF. Injuries to the facial nerve. Laryngoscope 1972;82:1891-6.

(14.) Barrs DM. Facial nerve trauma: Optimal timing for repair. Laryngoscope 1991;l0l:835-48.

(15.) Ferreira MC, Marchese AT, Spina V. Injertos transfaciales de nervio en el tratamiento de la paralysis facial observaciones iniciales. Chir Plast Ibero Lat Am 1977;3:301-6.

(16.) Ferreira MC. Cross-facial nerve grafting. Clin Plast Surg 1984;11:211-4.

(17.) Pulec IL. Facial nerve grafting. Laryngoscope 1969;79:1562-83.

(18.) Pulec JL. Facial nerve neuroma. Ear Nose Throat J 1994;73:721 2.725-39,743-52.

(19.) House JW, Brackmann DE. Facial nerve grading system. Otolaryngol Head Neck Surg 1985;93:146-7.

Table

Characteristics of the five patients with grade 6/6 facial nerve palsy
following resection of an acoustic neuroma who underwent cross-facial
nerve grafting

          Age(Yr)/       Type of        Duration of       Outcome (+)
Pt.         sex      branch graft (*)    follow-up   B          ZF

1          50/F     B to ZF, B; M to M    18 mo      4           4
2          32/F      B to B(upper and      3 yr      4          N/A
                      lower); M to M
3          45/M     ZF to ZF; B to B;      2 yr      2           3
                          M to M
4          53/F       B to B; M to M       3 mo      2          N/A
5          45/M       B to B; M to M      18 mo      3          N/A

           Outcome
             (+)
Pt.       M

1         4
2         3

3         2

4         2
5         2

(*)B = buccal; ZF = zygomaticofrontal; M = marginal.

(+)Postgrafting functional status by anatomic division, rated on a scale
of 1 to 5; 1 = no improvement; 2 = improvement in resting tone; 3 =
normal tone, but no movement; 4 = normal tone, movement with effort; 5 =
normal function.