Multidimensional Assessment of Voice After Vertical Partial Laryngectomy: A Comparison With Normal and Total Laryngectomy Voice

Abstract

The aim of this cross-sectional cohort study was to analyze the acoustic, stroboscopic, and perceptual parameters in patients who had undergone vertical partial laryngectomy (VPL) and to compare them with normal subjects (N) and total laryngectomy (TL) patients. This study was carried out in a tertiary referral Cancer Services Centre. We analyzed data from a total of 51 individuals; six VPL patients, 27 TL patients, and 18 N subjects. Acoustic analysis and videostroboscopy were performed using a computerized speech studio. Perceptual analysis was carried out using the GRBAS scoring system. The mean ages in the three groups were 40.9+/-13.5 (N), 54.3+/-9.5 (VPL), and 63.9+/-10.5 years (TL). Acoustic analysis yielded the following results for N, VPL, and TL, respectively: median fundamental frequency (155.2, 224.9, 106.3 Hz), shimmer (0.6, 1.9, 1.3 dB), and jitter (0.3%, 3.3%, 4.2%). The median maximum phonation time was 21.8 (N), 9.4 (VPL), and 10.3 seconds (TL). The median words per minute (WPM) were 168.5 (N), 126 (VPL), and 127 WPM (TL). Acoustic analysis of VPL voice was significantly different from normal voice (Mann-Whitney, P<0.05) and approximated better to TL parameters. The results of TL and VPL groups show poorer values and larger variability for nearly all the quantitative measures as compared to N subjects. Videostroboscopy demonstrated variable level and amount of closure of the vibrating segment in the reconstructed larynx of the VPL group. VPL GRBAS scores were similar to TL scores with good interrater reliability. The multidimensional assessment of voice in VPL patients is significantly worse than in N subjects and more closely resembles that of patients who have undergone TL with surgical voice restoration.