Nerve graft from the neck restores facial movement after cancer surgery, 12-month study shows

When surgeons remove a malignant tumor of the parotid gland, the facial nerve that animates the face often has to come away with it. The result can be devastating: a face frozen on one side, eyelids that no longer close, a mouth corner that droops, and expression muscles that begin to waste away within months. A new retrospective study from Beijing now offers one of the clearest long-term pictures yet of what happens when those lost nerve segments are immediately rebuilt with a graft taken from the patient’s own neck, and the findings suggest the technique deserves wider attention.

The research, published in Frontiers in Oral Health, followed 36 patients for an average of 18.5 months, with a minimum follow-up of 12 months, after surgical repair of facial nerve defects using an autologous great auricular nerve transplantation. The team, led by Huawei Liu and Min Hu of the First Medical Center of PLA General Hospital, together with colleagues from Peking University School and Hospital of Stomatology and other Beijing institutions, retrospectively analyzed cases performed between January 2011 and December 2020. Thirty of the patients had lost facial nerve segments to parotid gland malignancies, while six had suffered traumatic injuries to the maxillofacial region.

The technical logic behind the choice of donor nerve is central to the study. The great auricular nerve, or GAN, is the largest cutaneous branch of the cervical plexus, arising from the second and third cervical nerves and dividing into anterior and posterior branches near the parotid gland. It supplies sensation to the skin over the parotid gland, the mastoid process, and parts of the outer ear. Crucially, its median diameter closely matches that of the facial nerve trunk, and anatomical studies have shown a positive correlation between the two. That size compatibility, combined with its location within the same surgical field, makes the GAN an almost tailor-made donor for facial nerve reconstruction.

In the operating room, the surgeons harvested a GAN segment measuring one centimeter longer than the facial nerve defect, an intentional overshoot to ensure tension-free coverage. The proximal and distal ends of the facial nerve and the graft were then anastomosed under a double binocular surgical microscope using 9-0 nylon microsurgery sutures, an extremely fine thread finer than a human hair. When the harvested GAN branches were insufficient to cover all defected facial nerve branches, the team adopted a selective repair strategy based on functional priority. The zygomatic branch, which controls eyelid closure and protects the cornea from exposure keratitis, was repaired first, followed by the marginal mandibular branch governing lower lip movement and oral corner symmetry, then the buccal branch controlling nasolabial groove morphology, and finally the remaining branches. If the initial graft segment fell short in length or branch count, the dissection was extended proximally toward the cervical plexus to obtain longer or more branching segments.

Functional recovery was assessed with the House-Brackmann grading system, the standard six-point scale for facial nerve function ranging from Grade I, meaning normal function, to Grade VI, meaning total paralysis. Two experienced senior physicians who had not participated in the surgeries evaluated each patient independently, blinded to clinical information and surgical regimens, after pretraining that achieved a consistency Kappa above 0.8. Their assessments relied on standardized preoperative and postoperative photographs and videos captured in both static and dynamic states. Among the 12 patients with House-Brackmann results available at both 6 and 12 months, 8, or 66.7 percent, showed improvement between the two time points, and 4 remained stable. No patient deteriorated during that interval, indicating that facial nerve function continues to improve well beyond the half-year mark after grafting.

The defects themselves averaged 2.6 centimeters, ranging from 1.0 to 6.5 centimeters, and the tumor cohort included 8 mucoepidermoid carcinomas, 12 adenoid cystic carcinomas, 3 malignant pleomorphic adenomas, 4 acinar cell carcinomas, and 3 parotid duct carcinomas. Twelve patients had single-branch defects and 18 had multibranch defects. Intraoperative frozen sections confirmed no residual tumor in the facial nerve stump before reconstruction. For patients whose malignancies were tightly adherent to the facial nerve and could not be separated during surgery, resection of the affected nerve segments and immediate GAN reconstruction was performed in the same operation to guarantee complete tumor removal. One case of tumor recurrence occurred during follow-up and was controlled with 60 Gy radiotherapy.

The statistical analysis revealed two factors that significantly retarded recovery. Facial nerve defects longer than 3 centimeters were associated with poorer outcomes: only 1 of 11 patients with such long defects reached an acceptable Grade I-III recovery, compared with 11 of 25 patients whose defects measured 3 centimeters or less, a difference that reached statistical significance. Preoperative facial paralysis combined with parotid gland malignancy was an even stronger predictor, with only 2 of 14 patients who entered surgery already paralyzed achieving Grade I-III function, versus 10 of 16 who were not paralyzed beforehand. The authors explain the biology behind this: parotid malignancies invade the facial nerve axons, causing denervation of the target muscles and depriving facial neurons of nutritional supply. Muscle atrophy and weakness can become irreversible after roughly six months of facial nerve damage, as the intramuscular nerve sheath deteriorates and supportive satellite cells are lost. Notably, whether a defect involved a single branch or multiple branches did not influence the outcome, which the researchers attribute to the ability of autologous nerve transplantation to guide regenerating axons in the correct direction toward their target muscles, preventing the directionless axonal sprouting at facial nerve bifurcations that can produce nerve balls and graft failure.

Sacrificing the GAN carries an obvious cost: the ear and periauricular region lose part of their sensory innervation. Using a validated seven-item questionnaire covering the anterior and posterior superior helix, lobule, concha, preauricular area, infraauricular area, and postauricular area, the team recorded the sensations patients experienced after the graft was taken. Complete numbness, or lack of feeling, was the dominant abnormal sensation at 78.1 percent, followed by diminished but not absent sensation, or lack of sensitivity, at 53.1 percent, and abnormal sensation at 40.6 percent. Yet when the frequency, duration, anxiety, extension, interference with daily routine, and worry about these abnormal feelings were quantified, the median scores for interference and worry were both zero. The sensory deficits, in other words, did not significantly affect quality of life, and patients were far more concerned about tumor recurrence and recovery from facial paralysis than about the numbness behind their ears. Consistent with this, a five-year follow-up study cited by the authors found that postoperative sensations and one-year quality of life were comparable in parotid tumor patients whether or not the GAN was preserved.

The study’s conclusion is that autologous GAN transplantation is an effective approach for repairing facial nerve defects and can enable acceptable recovery of facial nerve function during the 12-month follow-up period. The findings align with earlier reports, including a series of 213 cable graft interposition cases showing acceptable recovery for defects up to 4 centimeters, and evidence that postoperative radiotherapy does not undermine the functional results of immediate facial nerve repair. In the present cohort, all 12 patients with adenoid cystic carcinoma treated with intraoperative radioactive seed implantation achieved postoperative recovery of facial nerve function.

The authors are candid about the limitations. The study was single-center and retrospective, with a small sample of 36 patients and no control group treated with other nerve grafts or direct anastomosis, so only univariate analysis was possible and the results should be regarded as exploratory. The selective repair strategy may have objectively improved the overall House-Brackmann grading, and a refined regional facial nerve function evaluation system could more accurately capture the recovery of individual branches in future work. Donor-site sensations were mainly assessed within 12 months, leaving long-term sensory monitoring data scant. Whether GAN grafting is superior to alternative techniques, such as hypoglossal-facial nerve transfer, which carries its own risk of donor nerve dysfunction, must be tested in prospective randomized controlled trials. Still, for patients facing the loss of the nerve that animates their face, the message from a decade of Beijing cases is quietly hopeful: a nerve borrowed from the neck, spliced in under a microscope in a single operation, can bring movement back to a paralyzed face, and the price paid behind the ear is one most patients barely notice.

Subject of Research: Surgical repair of facial nerve defects by an autologous great auricular nerve transplantation: a retrospective, 12-month, follow-up study

Article Title: Surgical repair of facial nerve defects by an autologous great auricular nerve transplantation: a retrospective, 12-month, follow-up study

Article References: Liu, H., Ning, Q., Mu, X., Bi, W., Sun, Q., Cai, Z., Li, S., Shan, X., & Hu, M. (2026). Surgical repair of facial nerve defects by an autologous great auricular nerve transplantation: a retrospective, 12-month, follow-up study. Frontiers in Oral Health, 7, Article 1920749. https://doi.org/10.3389/froh.2026.1920749

Image Credits: AI Generated

DOI: 10.3389/froh.2026.1920749

Keywords: facial nerve, great auricular nerve, nerve grafting, parotid gland cancer, House-Brackmann grading, microsurgery, Surgical, repair, facial, nerve, defects, autologous