Masseteric-to-Facial Nerve Transfer and Selective Neurectomy for Rehabilitation of the Synkinetic Smile

Key Points

Question

Is masseteric-to-facial nerve transfer combined with selective neurectomy associated with improved synkinetic smile?

Findings

In this case series, 7 patients underwent masseter-to-facial nerve transfer with selective neurectomy for synkinesis. Patients experienced a statistically significant improvement in multiple eFACE domains including smile, dynamic function, synkinesis, midface and smile function, and lower face and neck function at 1-year mean follow-up.

Meaning

Masseteric-to-facial nerve transfer with selective neurectomy may provide significant smile improvement with a long-term decrease in synkinesis for patients.

Abstract

Importance

Synkinesis is the involuntary movement of 1 area of the face accompanying volitional movement of another; it is commonly encountered in patients affected by facial palsy. Current treatments for synkinesis include biofeedback for muscular retraining and chemodenervation via the injection of botulinum toxin. Chemodenervation is effective in reducing unwanted muscle movement, but it requires a commitment to long-term maintenance injections and may lose effectiveness over time. A permanent solution for synkinesis remains elusive.

Objective

To evaluate masseteric-to-facial nerve transfer with selective neurectomy in rehabilitation of the synkinetic smile.

Design, Setting, and Participants

In this case series, 7 patients at a tertiary care teaching hospital underwent masseteric-to-facial nerve transfer with selective neurectomy for synkinesis between September 14, 2015, and April 19, 2018. The medical records of these patients were retrospectively reviewed and demographic characteristics, facial palsy causes, other interventions used, and changes in eFACE scores were identified.

Intervention

Masseteric-to-facial nerve transfer.

Main Outcomes and Measures

Changes in eFACE scores (calculated via numeric scoring of many sections of the face, including flaccidity, normal tone, and hypertonicity; higher scores indicate better function and lower scores indicate poorer function) and House-Brackmann Facial Nerve Grading System scores (range, 1-6; a score of 1 indicates normal facial function on the affected side, and a score of 6 indicates absence of any facial function [complete flaccid palsy] on the affected side).

Results

Among the 7 patients in the study (6 women and 1 man; median age, 49 years [range, 41-63 years]), there were no postoperative complications; patients were followed up for a mean of 12.8 months after surgery (range, 11.0-24.5 months). Patients experienced a significant improvement in mean (SD) eFACE scores in multiple domains, including smile (preoperative, 65.00 [8.64]; postoperative, 76.43 [7.79]; P = .01), dynamic function (preoperative, 62.57 [15.37]; and postoperative, 75.71 [8.48]; P = .03), synkinesis (preoperative, 52.70 [4.96]; and postoperative, 82.00 [6.93]; P < .001), midface and smile function (preoperative, 60.71 [13.52]; and postoperative, 78.86 [14.70]; P = .02), and lower face and neck function (preoperative, 51.14 [16.39]; and postoperative, 66.43 [20.82]; P = .046). Preoperative House-Brackmann Facial Nerve Grading System scores ranged from 3 to 4, and postoperative scores ranged from 2 to 3; this change was not significant.

Conclusion and Relevance

This study describes the application of masseteric-to-facial nerve transfer with selective neurectomy for smile rehabilitation in patients with synkinesis, with statistically significant improvement in smile symmetry and lower facial synkinesis as measured with the eFACE tool. This technique may allow for long-term improvement of synkinesis and smile. This study is only preliminary, and a larger cohort will permit more accurate assessment of this therapeutic modality.

Level of Evidence

4.

This case series evaluates masseteric-to-facial nerve transfer with selective neurectomy in rehabilitation of the synkinetic smile.

Introduction

Facial nerve dysfunction can occur as a result of Bell palsy, acoustic neuroma growth or extirpation, head and neck cancer, iatrogenic injuries, varicella zoster virus (Ramsay Hunt syndrome), trauma, autoimmune diseases, infectious diseases, and idiopathic conditions¹ and can have a profound functional, aesthetic, and psychosocial effect on an individual.² There is ongoing controversy regarding how best to objectively assess patient outcomes in facial palsy.³ Facial palsy is most commonly graded with the House-Brackmann Facial Nerve Grading System (FGS), although others, such as the Sunnybrook and Yanagihara grading systems, are also used when more detail is required in the evaluation.^4,5 Most facial paralysis assessment scales nevertheless oversimplify facial nerve function, providing a single digit to summarize the status of an entire hemiface or, at best, 1 number each for resting symmetry, dynamic function, and synkinetic movement. The eFACE instrument, in contrast, is a validated grading tool that evaluates individual domains of facial function (eg, periocular, smile and midface, lower face and neck, static symmetry, dynamic function, and synkinetic movement), permitting segmental comparisons. It allows a more specific evaluation of synkinesis and facial dysfunction after nerve palsy.^6,7,8 There are advantages and disadvantages to each grading modality, and many clinicians use multiple methods to track patient outcomes over time.

Synkinesis is the unintentional motion of an area of the face accompanying volitional movement of another area⁹; it is commonly encountered in patients with a history of flaccid facial palsy. Synkinesis is theorized to result from 3 different mechanisms: aberrant regeneration with multiply terminal axons innervating multiple muscle groups, ephatic transmission (the phenomenon in which axons near a site of injury stimulate one another), and hyperexcitability at the level of the facial nucleus.¹⁰

There is ongoing debate regarding the most appropriate treatment modality for synkinesis, and a criterion standard therapy has yet to emerge.³ The most common treatments for synkinesis are conservative and include biofeedback for muscular retraining and chemodenervation with botulinum toxin injection.^10,11 Chemodenervation is effective in reducing unwanted muscle movement, but it requires a commitment to undergo regular injections, as its effects wear off over time, and it should be used cautiously in the perioral and periocular regions to prevent iatrogenic lagophthalmos or oral incompetence. Historically, patients who have failed chemodenervation or have declined botulinum toxin injections may have been offered selective neurectomy as second-line therapy.^12,13 Periocular selective neurectomy appears to be ineffective in the long term,¹⁴ but a 2019 article by Azizzadeh et al¹⁵ demonstrated that cervicofacial selective neurectomy can be beneficial as first-line therapy for synkinesis, significantly improving patients’ smiles.

Rehabilitation of the smile in the flaccid face, in contrast, has historically been accomplished with either reinnervation of native mimetic musculature or with muscle transfer depending on the time interval between paralysis onset and intervention.¹⁶ Transfer of the masseteric nerve has found favor for this application in recent years because of its convenient location with respect to the facial nerve and the minimal morbidity associated with its transfer compared with other techniques, although its efficacy as a single procedure for synkinetic smile rehabilitation is low, reported at only 20%.^17,18,19,20 Although the masseteric nerve has a substantial axon count and therefore adequate power to drive a smile, it has a lower basal firing rate than the facial nerve and can result in facial droop when it is used as the sole rehabilitation option for the flaccid face. Fortunately, the synkinetic face has sufficient resting tone to obviate the problem of ptosis with masseteric-to-facial nerve transfer, making the intervention a theoretically good solution for restoring smile in patients with synkinesis. In addition, masseter-to-facial nerve transfer has been shown to restore a spontaneous normal, or near-normal, smile in up to 24% of patients, and many patients anecdotally describe the ability to smile without needing to clench the jaw.²¹

Herein, we describe masseteric-to-facial nerve transfer, a technique commonly used for facial reanimation, combined with selective neurectomy as a long-term treatment option for synkinesis. We hypothesize that this combination will lead to improved smile symmetry and excursion while decreasing synkinesis, as measured by the eFACE.

Methods

We performed masseteric-to-facial nerve transfer with selective neurectomy to improve the smile and symptoms of synkinesis in a series of patients at Madigan Army Medical Center, Tacoma, Washington, and San Antonio Military Medical Center, San Antonio, Texas, between September 14, 2015, and April 19, 2018. We retrospectively reviewed the medical records of these patients and identified demographic characteristics, causes of facial palsy, and other interventions used. We recorded FGS scores before and after surgery at multiple time points (range, 1-6; a score of 1 indicates normal facial function on the affected side, and a score of 6 indicates absence of any facial function [complete flaccid palsy] on the affected side). We also analyzed preoperative and postoperative videos of patients going through multiple facial exercises using eFACE software. Videos for eFACE analysis were taken preoperatively and at least 10 months postoperatively. Videos were taken at least 3 months after any botulinum toxin injections. Approval for this study was granted by the Madigan Army Medical Center Institutional Review Board. Patients provided written consent for the use of their photos. The Madigan Army Medical Center Institutional Review Board waived the need for patient consent for inclusion in the study given the study’s retrospective nature and analysis of deidentified data only.

Technique

We offer masseteric-to-facial nerve transfer and selective neurectomy to patients with persistent oro-ocular synkinesis who have either failed botulinum toxin therapy or who do not wish to continue with regular injections, instead preferring a longer-term solution. We begin with a modified Blair incision, carried down to the parotidomasseteric fascia, after which a flap is developed to the anterior edge of the parotid gland. Next, blunt dissection deep to the superficial musculoaponeurotic layer exposes the facial nerve branches. Our goal is to identify at least 5 to 8 facial nerve branches, with at least 2 superior to a buccal branch–producing smile and 2 inferior to it. Once isolated, the branches are marked with vessel loops. Typically, the branch producing the best smile excursion with the least synkinetic stimulation is located near the Zuker point.²² Other nerve branches can typically be found evenly spaced at 1- to 1.5-cm intervals.

Next, the masseteric nerve is isolated via blunt dissection through the masseter muscle at a point 3 cm anterior to the tragus, 1 cm inferior to the zygomatic arch, and 1.5 cm deep to the muscle fascia.²³ The facial nerve branches are individually stimulated from superior to inferior to produce a “map” of the facial nerve (Video 1). Figure 1A demonstrates the close anatomical relationship of the buccal branch of the facial nerve and the masseteric nerve. The facial nerve branch producing the most aesthetically pleasing smile with the least synkinesis is identified (Video 2), and branches causing synkinetic midface, oral, and ocular stimulation are transected. The neurectomies are performed by removing an approximately 1-cm segment of nerve to minimize the potential for regrowth. Care is taken to avoid neurectomy of branches to the orbicularis oculi, particularly of the lower eyelid, to prevent postoperative lagophthalmos. Branches causing stimulation of mentalis, depressor anguli oris, depressor labii inferioris, and platysma are also transected to eliminate unwanted inferior pull of the oral commissure during smile. The cervical branch to the platysma can be found roughly 1.5 to 2 cm inferior to the angle of the mandible, perpendicular to a line from the menton to the mastoid tip.²⁴ Neurorrhaphy is then performed from the masseteric nerve to the isolated facial nerve “smile” branch with 10-0 nylon stitches, a nerve sheath, and tissue sealant (Figure 1B). After the neurorrhaphy, the wound is irrigated and the incision is closed in layers over a Penrose drain.

Figure 1. Preoperative Anatomy of the Nerve to the Masseter and Facial Nerve and Postoperative Anatomy After Transfer of the Masseteric Nerve to a Buccal Branch of the Facial Nerve and Multiple Selective Neurectomies.

A, Preoperative anatomy. The descending branch of the masseteric nerve and the buccal branch of the facial nerve lie in close proximity to each other. There is redundant midface innervation from the facial nerve. The zygomaticus major and orbicularis oris are also shown. B, Postoperative anatomy. When performing neurectomy, a segment of nerve is removed to inhibit nerve regeneration and reinnervation. These illustrations are copyrighted by Dr Wind and reproduced with permission.

Video 1. Buccal Branch Stimulation.

Download video file ^{(22.8MB, mp4)}

Intraoperative identification of a buccal nerve branch with good smile excursion.

Video 2. Distal Facial Nerve Stimulation.

Download video file ^{(100.3MB, mp4)}

Intraoperative stimulation of multiple distal facial nerve branches to identify nerves for selective neurectomy and masseteric-to-facial nerve transfer.

Statistical Analysis

Scores in the eFACE domains of smile, static, dynamic, synkinesis, periocular, lower face and neck, and midface and smile (higher scores indicate better function, and lower scores indicate poorer function) were compared preoperatively and postoperatively with paired-samples t tests. Preoperative scores, postoperative scores, and the difference in preoperative to postoperative scores in patients with a history of Bell palsy were compared with scores from patients with a history of acoustic neuroma with independent-samples t tests. Similarly, independent-samples t tests were used to compare eFACE domain scores of patients who underwent a platysmectomy with scores of those who did not. Preoperative and postoperative FGS scores were compared with paired-samples t tests. We performed statistical analysis using IBM SPSS Statistics, version 24 (IBM Corp). P < .05 was considered significant.

Results

Seven patients with synkinetic facial palsy underwent masseteric-to-facial nerve transfer and selective neurectomy for smile rehabilitation between September 14, 2015, and April 19, 2018. There were no postoperative complications. Six patients were women, and median patient age was 49 years (range, 41-63 years). Bell palsy was the cause of 4 patients’ paralysis, while the other 3 experienced paralysis after removal of acoustic neuroma. One patient underwent attempted synkinetic smile reanimation with a 2-strand cross-face nerve graft that ultimately failed based on a complete lack of conduction as measured by cross-face electroneuronography at 1 year after the procedure. Three patients also underwent platysmectomy to relieve synkinetic hyperactivation of the platysma. All patients had tried multiple rounds of botulinum toxin injection for synkinesis. Three patients underwent selective neurectomy in a staged fashion, while 4 underwent selective neurectomy at the time of masseteric-to-facial nerve transfer. One patient required a revision selective neurectomy. Table 1 shows the cause of palsy for each patient as well as all procedures performed during the study period. Figure 2 shows the preoperative and postoperative smile of 1 of our study patients.

Table 1. Causes of Facial Paralysis in Study Patients and Procedures Performed to Improve Synkinetic Facial Palsy During the Study Period.

Patient No.	Cause of Facial Paralysis	Platysmectomy	Selective Neurectomy	Masseteric-to-Facial Nerve Transfer	Chemical Denervation	Additional Procedures After Development of Synkinesis
1	Acoustic neuroma	Yes	Yes	Yes	Yes	Rhytidectomy
2	Acoustic neuroma	No	Yesa	Yes	Yes	Cross-face nerve graft (to buccal branch of facial nerve)
3	Acoustic neuroma	No	Yes	Yes	Yes	None
4	Bell palsy	No	Yesa	Yes	Yes	None
5	Bell palsy	No	Yes	Yes	Yes	Brow lift
6	Bell palsy	Yes	Yes	Yes	Yes	None
7	Bell palsy	Yes	Yes	Yes	Yes	Rhytidectomy

^aPatients 2 and 4 underwent 2 selective neurectomy procedures.

Figure 2. Patient With a History of Acoustic Neuroma Extirpation and Subsequent Synkinetic Facial Paralysis on the Right Treated With Selective Neurectomy and Masseteric Nerve Transfer.

A, Preoperative significant smile asymmetry. B, Improvement of smile asymmetry after surgery.

Patients underwent masseteric-to-facial nerve transfer a mean of 7.8 years after the onset of facial palsy (range, 1.6-25 years). Patients have been followed for a mean of 12.8 months after surgery (range, 11.0-24.5 months). Preoperative FGS scores ranged from 3 to 4, and postoperative scores ranged from 2 to 3; this change was not significant. Figure 3 compares preoperative and postoperative eFACE scores (Table 2). Scores improved after masseteric-to-facial transfer in each category. Patients experienced a significant improvement in mean (SD) eFACE scores in the smile domain (preoperative, 65.00 [8.64]; postoperative, 76.43 [7.79]; P = .01), dynamic function domain (preoperative, 62.57 [15.37]; postoperative, 75.71 [8.48]; P = .03), synkinesis domain (preoperative, 52.70 [4.96]; postoperative, 82.00 [6.93]; P < .001), lower face and neck function domain (preoperative, 51.14 [16.39]; postoperative, 66.43 [20.82]; P = .046), and midface and smile function domain (preoperative, 60.71 [13.52]; postoperative, 78.86 [14.70]; P = .02). Improvement in the static and periocular domains were not significant. There was no significant difference in preoperative or postoperative eFACE scores from any domain between patients with a history of acoustic neuroma extirpation and those with a history of Bell palsy. Similarly, the amount of improvement seen in eFACE scores for each domain, from preoperative to postoperative scores, was not significantly different between patients with different causes of palsy. However, patients who underwent platysmectomy had significantly better postoperative midface and smile domain scores than those who did not undergo platysmectomy (95% CI of the difference, –17.47 to 40.14; P = .03). Platysmectomy did not have a significant association with scores in other eFACE domains or preoperative to postoperative changes in domain scores.

Figure 3. Comparison of Preoperative and Postoperative eFACE Scores.

Patients demonstrated improvement in every category after masseteric-to-facial nerve transfer with selective neurectomy. The box defines the 25th percentile (bottom line), median (center line), and the 75th percentile (top line). The whiskers extend to the upper and lower extremes of the percentiles. The dots indicate outliers. LFN indicates lower face and neck; MSF, midface and smile.

^aThe improvement met statistical significance.

Table 2. Preoperative and Postoperative eFACE Scores.

eFACE Category	Score, Mean (SD)		Difference (95% CI)	P Value
	Preoperative	Postoperative
Smile	65.00 (8.64)	76.43 (7.79)	11.43 (3.53 to 19.33)	.01
Static function	81.86 (10.56)	84.57 (7.74)	2.71 (–2.90 to 8.33)	.28
Dynamic function	62.57 (15.37)	75.71 (8.48)	13.14 (2.93 to 24.10)	.03
Synkinesis	52.70 (4.96)	82.00 (6.93)	29.30 (5.02 to 31.27)	<.001
Periocular	75.86 (9.19)	79.87 (9.43)	4.01 (–5.02 to 12.45)	.34
Lower face and neck function	51.14 (16.39)	66.43 (20.82)	15.29 (0.36 to 30.22)	.046
Midface and smile function	60.71 (13.52)	78.86 (14.70)	18.15 (3.93 to 32.36)	.02

Discussion

Herein, we present a retrospective review that demonstrates improved smile and decreased synkinesis after masseteric-to-facial nerve transfer combined with selective neurectomy. Improvements in the static facial appearance and periocular function were not statistically significant, although an approximately 10- to 12-point improvement in eFACE score was noted in these domains, which is consistent with the selective neurectomy results of Azizzadeh et al,¹⁵ reported as statistically significant owing to a larger sample size. Also, while selective neurectomy was performed to improve oro-ocular synkinesis, it was performed conservatively in our series for facial nerve branches innervating the orbicularis oculi to avoid weakening periocular muscles to the point of risking poor eye closure and exposure keratopathy. Furthermore, selective periocular neurectomy is not a permanent procedure, as classically described.¹⁴ There is sufficient midface redundancy of terminal zygomatic and buccal facial nerve branches that periocular innervation tends to regenerate over time with recurrence of oro-ocular synkinesis, although nerve caps that may improve the longer-term efficacy of neurectomy are under development. Regardless, dynamic smile can be improved with masseteric-to-facial nerve transfer. We hypothesize that the transfer improves smile by powering the terminal buccal nerve branch and smile by the more powerful masseteric nerve and by removing aberrant innervation of synkinetic facial nerve branches. In our experience, masseteric-to-facial nerve transfer is more successful when facial nerve branches innervating the lower face and platysma are cut, as this minimizes activation of muscles that would pull the oral commissure downward. It frees the zygomaticus muscles from having to work against the action of other muscles, and it improves the overall smile excursion. Furthermore, in our study, patients who underwent platysmectomy had significantly better postoperative eFACE scores in the midface and smile domain, as contraction of the platysma during attempted smile can lead to inferior pull on the oral commissure. Platysmectomy has been described previously for treatment of facial synkinesis and is associated with significant improvement in quality of life.²⁵ To our knowledge, no previous studies have found objective eFACE changes, however.

The human face performs myriad functions, and each one must be rehabilitated individually; our technique is one more weapon in the facial surgeon’s armamentarium for smile restoration. A similar study by Biglioli et al²⁶ in 2017 retrospectively reviewed 18 cases in which patients with midface synkinesis had undergone masseteric-to-facial nerve transfer for smile rehabilitation; 12 patients in their series also underwent 2-stage cross-face nerve grafting, and the other 6 underwent a selective neurectomy as well as other ancillary procedures. They reported that 72% of patients with preoperative hypertonicity experienced at least partial relaxation immediately after surgery without recurrence of symptoms. They measured improvement in facial movement using the modified FGS; before the procedure, their patients were reported to have grade III or IV palsy, whereas after the procedure they had grade II or III palsy. They did not, however, use a segmental facial function assessment instrument, such as the eFACE, making outcome evaluations somewhat less specific. Our results are consistent with those of Biglioli et al²⁶; however, ours are possibly more specific, as our cohort had preoperative FGS grades of 3 to 4, improving to 2 to 3 postoperatively. We did not find significant changes in FGS scores after surgery, first, as this tool is not designed to specifically evaluate smiling, and second, as our sample size is small. The eFACE instrument proved to be a much more useful measurement tool, identifying statistically significant changes for multiple facial zones in our patients, thereby validating our hypothesis that a long-term improvement in smile and reduction in mid- to lower-facial synkinesis could be achieved by combining selective neurectomy and masseteric-to-facial nerve transfer.

A 2019 study by Azizzadeh et al¹⁵ also investigated a modified selective neurectomy technique for synkinetic smile rehabilitation and did use the eFACE instrument to document its outcome. They found that neurectomy without masseteric-to-facial nerve transfer produced an improvement in all eFACE domains, which is consistent with our findings, although our study was insufficiently powered to demonstrate statistical significance in the static and periocular domains. As is common with facial reanimation operations, multiple procedures were performed in the study by Azizzadeh et al¹⁵; more than half of the patients in their series (56%) underwent concurrent rhytidectomy, and 27% also underwent reneurotization of the zygomatic musculature without use of the masseteric nerve, which may have contributed to improvement beyond what could be achieved with neurectomy alone. Although we are unable to comment on the statistical significance of outcome differences between the study by Azizzadeh et al¹⁵ and our own, in our series, the preoperative to postoperative differences in eFACE scores for the dynamic, synkinetic, lower face and neck, midface and smile, and smile domains are all greater (differences of 2 points for dynamic domain, 5 points for synkinetic domain, 1 point for lower face and neck domain, 9 points for midface and smile domain, and 3 points for smile domain), indicating that masseteric-to-facial nerve transfer may provide an advantage over selective neurectomy alone and warrants further study.

Limitations

Our study is limited by its small sample size and preliminary nature. A larger patient cohort will allow more robust analysis of the true efficacy of masseteric-to-facial nerve transfer with selective neurectomy. Furthermore, all patients in our study underwent surgery with the senior author (M.H.H.); his results may not be generalizable and replicable.

Conclusions

A criterion standard treatment for synkinesis remains elusive, but masseteric-to-facial nerve transfer with selective neurectomy provides significant smile improvement with long-term decrease in synkinesis for patients who fail botulinum toxin therapy or, for other reasons, do not wish to continue with regular injections. Our study is only preliminary, and a larger cohort examined with the eFACE instrument will allow more accurate assessment of this therapeutic modality.