Facial nerve action potentials: a study to assess waveform reliability.

Abstract

Objective: To assess the reliability of the orthodromic facial nerve action potential (FNAP), recorded from the intratemporal portion of the facial nerve on stimulation within the cerebellopontine angle.

Study design: Prospective study.

Setting: Tertiary referral center.

Patients: Ten consecutive patients undergoing translabyrinthine resection of vestibular schwannoma.

Intervention: Diagnostic.

Main outcome measures: Ten consecutive FNAPs were recorded on stimulation of the facial nerve within the cerebellopontine angle. The FNAP recording probe was placed directly on the nerve surface after the fallopian canal was opened at the second genu. Ten consecutive compound muscle action potentials (CMAPs) were recorded simultaneously from surface electrodes overlying the facial musculature, by use of a standardized electrode placement technique. The stimulating and recording equipment were removed (excluding CMAP surface electrodes) and reapplied, and FNAP and CMAP data were recorded for a second time (test/retest). Peak-to-peak amplitudes of all waveforms were calculated.

Results: The average FNAP peak-to-peak amplitude for all patients was larger than the CMAP peak-to-peak amplitude (2.60 mV and 1.07 mV, respectively). Random effects analysis of variance was performed to assess the individual components of variation. This showed that CMAP was less variable than FNAP for replicate error (10 consecutive FNAPs and CMAPs) and test/retest error. However, subject variance was less for FNAP, where subject variance was by far the largest contributor to overall variation. The reliability coefficient for FNAP was 0.995 and for the CMAP was 0.982, where absolute reliability is 1.0.

Conclusion: These data confirm that the FNAP, recorded by the technique described here, is a reliable waveform when compared with the CMAP and is a valid method for assessing facial nerve function.