Serial Electromyography of the Thyroarytenoid Muscles Using the NIM-Response System in a Canine Model of Vocal Fold Paralysis

Objectives: We sought to determine whether serial intraoperative laryngeal electromyography (L-EMG) or evoked L-EMG predicts vocal fold (VF) recovery following iatrogenic injury. Methods: Six beagles were sedated, and bipolar needle electrodes were inserted into each thyroarytenoid (TA) muscle. Endotracheal tube surface electrodes were also placed. As the sedation lightened, L-EMG activity was recorded from all electrodes with an intraoperative nerve monitoring system. The neck was opened, and direct recurrent laryngeal nerve (RLN) stimulation was performed. Subsequently, 4 animals underwent crush injury of the left RLN, and 2 animals underwent nerve transection. The L-EMG procedures were repeated every I to 2 weeks until left VF motion was observed in the dogs that suffered RLN crush injury. At each time point, the neck was opened and both RLNs were stimulated. Results: Fibrillation potentials were detected in all animals after RLN injury. A change to electrical silence was seen in the animals in the crush injury group that were evaluated during the week preceding VF recovery. Fibrillation potentials and VF immobility persisted in the transection group throughout the complete time course of these experiments. The first appearance of an evoked response coincided with or occurred after the return of left VF motion in the crush injury group. The threshold, latency, and amplitude differed from those of the controls and approached normal values over time. No response was detected in the transected nerves. Conclusions: The disappearance of fibrillations on intraoperative L-EMG was noted in the animals tested the week before the return of VF motion, and the return of motor unit action potentials was seen along with return of VF function. Evoked L-EMG was not helpful in predicting the return of VF mobility, but it may help quantify degrees of RLN injury and predict the speed of recovery.