Surface Matrix Electrode Array-Guided Recurrent Laryngeal Nerve Mapping Without Nerve Exposure: A Rabbit Proof-of-Concept Pilot Study

Jiandi Wu, Wenqing Hu, Jie Hao, Tao Li, Fengwu Tan, Xiaohong Chen

Background: Precise recurrent laryngeal nerve (RLN) localization remains challenging in thyroid surgery, particularly in re-operative cases where peri-neural fibrosis distorts anatomy and limits conventional single-point neural monitoring. We aimed to design, optimize stimulation parameters for, and preliminarily evaluate the feasibility of a surface matrix electrode array guide for structured RLN mapping under a skin-incision-only, nerve-unexposed condition.Methods: Nine New Zealand White rabbits were assigned to first-, second-, and third-operation groups to model progressive peri-neural fibrosis. A 10 × 10 mm flexible array guide with a 5 × 5 matrix of 25 stimulation sites (2.0 mm spacing) was applied to paratracheal soft tissue under a skin-incision-only condition (18 sides; 90 accuracy assessments). Localization accuracy was defined as electrophysiological reference point-to-nerve distance ≤2 mm, using the surgically exposed RLN as the reference standard.Results: Effective stimulation distance was 1–2 mm at 0.5 mA and 4–5 mm at 1.0 mA, justifying the 2.0 mm inter-site spacing and 0.5 mA working current. Overall descriptive site-level localization accuracy was 92.2% (83/90 sites). The array guide consistently reconstructed the RLN trajectory across all 18 evaluated sides, with similar descriptive accuracy between the left and right sides. Seven initially false-negative sites in re-operative animals were recovered by increasing the stimulation current to 0.8–1.0 mA. One case of an anomalous RLN course with lateral displacement toward the carotid sheath was successfully identified without prior anatomical assumptions.Conclusions: These preliminary findings provide the electrophysiological rationale and initial accuracy data necessary to justify progression to large-animal validation studies, representing a required step prior to clinical translation. Given the small sample size, rabbit model anatomy, and mild-to-moderate fibrosis burden, these findings should be interpreted as hypothesis-generating and are not directly extrapolable to clinical practice without further large-animal validation.