SA3 - Motor Nerve to the Masseter Versus Cross-Facial Nerve Grafting in Pediatric Free Gracilis Muscle Transfer: Objective Analysis of Smile Symmetry

Abstract: Statements of the Problem: Facial nerve palsy in children produces significant functional and esthetic deficits that often require surgical smile reanimation. Free gracilis muscle transfer (FGMT) is the most widely used procedure for dynamic facial reanimation in long-standing pediatric paralysis. However, the optimal donor nerve for gracilis reinnervation remains debated. Cross-facial nerve grafting (CFNG) provides spontaneous smile activation but requires staged reconstruction and prolonged reinnervation, whereas the motor nerve to the masseter (MNM) allows single-stage reconstruction with higher axonal density and potentially stronger contraction.^1,2 The extent to which these physiologic differences translate into measurable improvements in dynamic facial symmetry remains unclear. This study evaluates donor nerve selection using objective machine-learning analysis of smile symmetry following pediatric FGMT.

Materials and Methods Used in the Investigation of the Problem: A retrospective review was conducted of pediatric patients with unilateral facial palsy who underwent FGMT at a tertiary pediatric center between 2007 and 2020. Standardized preoperative and postoperative photographs at rest (repose), smile, and open smile were analyzed using an automated machine-learning facial landmarking algorithm. The software identified key perioral landmarks and generated 5 validated symmetry metrics: commissure excursion, smile angle, commissure height, upper lip height, and lower lip height deviation.

Method of Data Analysis: Sample size: 12 patients.

Duration of study: 2007-2020 with follow-up at 1, 2, and 3 years postoperatively.

Statistical methods: Linear mixed-effects modeling was used to assess the independent effects of donor nerve (MNM vs CFNG), time, and their interaction on symmetry outcomes. Statistical significance was defined as P < .05.

Results of the Investigation: Twelve patients with a mean age of 12.2 years met the inclusion criteria. Four patients underwent MNM reinnervation, and 8 underwent CFNG. No significant differences in symmetry were observed at rest across all 5 metrics over time or between techniques.

Dynamic expressions demonstrated significant differences. In smile analysis, the MNM group showed significantly improved commissure excursion symmetry compared with CFNG (P = .004), with significant interaction effects at postoperative years 1 and 2. In open-smile analysis, MNM demonstrated significantly improved commissure height symmetry (P = .026) with interaction effects at years 1 and 3. MNM also significantly improved upper lip height symmetry (P = .048), with interaction effects at all 3 timepoints. Although MNM alone did not significantly improve smile angle symmetry, significant interaction effects at years 1 and 3 indicated greater improvement over time compared with CFNG.

Outcomes Data: Objective machine-learning analysis demonstrated superior dynamic symmetry outcomes with MNM reinnervation compared with CFNG, particularly in commissure excursion, commissure height, and upper lip height during smile activation.

Conclusions Relevant to the Problem: Masseteric nerve reinnervation demonstrated superior dynamic symmetry compared with cross-facial nerve grafting following pediatric FGMT, particularly in commissure excursion, commissure height, and upper lip symmetry. These findings suggest that MNM may provide more reliable functional restoration in patients with significant preoperative asymmetry. Objective machine-learning-based symmetry analysis may help guide donor nerve selection, improve surgical planning, and support individualized treatment strategies in pediatric facial reanimation.


References

1. Chen K, Nguyen SA, Chun WB, et al. Outcomes and Innervation of Gracilis for Pediatric Facial Paralysis: A Systematic Review. Laryngoscope. Published online December 30, 2025.

2. Harrison DH. Surgical correction of unilateral and bilateral facial palsy. Postgrad Med J. 2005;81(959):562-567.