Acoustic voice analysis does not identify presence of penetration/aspiration as confirmed by videofluoroscopic swallowing study.

Abstract

To investigate whether acoustic voice analysis can identify the presence of penetration/aspiration (P/A) as confirmed by videofluoroscopic swallowing study (VFSS). Repeated measures within subjects. Rehabilitation department in a tertiary teaching hospital. Patients (N=44) with swallowing disorders referred for VFSS. Patients were asked to sustain phonations /a/ for at least 3 seconds before and after swallowing 5mL of liquid barium during a standardized VFSS. The acoustic voice analysis program was used to analyze vocal quality change. Five acoustic parameters including average fundamental frequency, relative average perturbation, shimmer percentage, noise-to-harmonic ratio, and voice turbulence index were analyzed for each participant before and after swallowing during VFSS. Differences in the pre- and postmeasures were compared between those participants who demonstrated VFSS-confirmed P/A (n=17) and those who did not (n=27). No significant changes were noted in the 5 acoustic parameters in or between P/A and nonpenetration/aspiration groups (P>.05). As used in this study, acoustic voice analysis does not identify the presence of P/A confirmed by VFSS.