Predictors of laryngeal complications in patients implanted with the Cyberonics vagal nerve stimulator

Since its approval by the US Food and Drug Administration in 1997 for management of medically refractory seizures, more than 35,000 patients have been implanted with the Cyberonics vagal nerve stimulator. Preliminary reports described transient vocal changes in the majority of subjects, which were thought to be short-term. However, these reports were for the most part based upon perceptual evaluations by the subjects themselves. Later reports described possibly more permanent recurrent laryngeal nerve injury and recommended measuring the nerve diameter to use the safest spiral cuff electrode. To date, no study has systematically evaluated vocal fold mobility in subjects before and after implantation. The objectives of this study were to determine the true incidence of both short- and long-term recurrent laryngeal nerve injuries and determine whether there are any potential indicators to predict in which patients long-term nerve deficits may develop.

Thirteen subjects underwent preimplantation laryngeal electromyography, videolaryngoscopy, measurement of the maximum phonation time, Voice Handicap Index determination, and Consensus Auditory-Perceptual Evaluation of Voice. Two weeks after implantation, all subjects underwent videolaryngoscopy. Three months after implantation and activation of the device, all subjects were reevaluated.

Six of the 13 subjects had significant vocal fold mobility abnormalities at 2 weeks. Significant electromyographic abnormalities were detected before implantation in 5 subjects. All 5 of these subjects, at 3 months after implantation, had prolonged left vocal fold paresis.

The authors conclude that perioperative vocal fold paresis occurs in approximately 50% of subjects. Further, laryngeal electromyography performed before implantation of the vagal nerve stimulator is a statistically significant predictor (p < .05) of which patients may be at risk for extended vocal fold abnormalities. Possible explanations for this phenomenon are offered. Surgical modifications to limit vagal nerve injury are offered.