Analysis of vowel-consonant-vowel sequences in patients with partial glossectomies using 2D ultrasound imaging

Background: The effect of partial glossectomy surgery on speech is still not completely understood. In previous research, we found that the loss of tongue tissue due to a lateral partial glossectomy was compensated by an increase of both the average tongue height and the velocity of midsagittal tongue movements (Rastadmehr, Bressmann, Smyth & Irish, 2008). Since Rastadmehr et al. had used a connected speech task, the aim of the present study was to evaluate the effect of the glossectomy surgery on tongue height in individual vowel-consonant-vowel (VCV) sequences. Methods: 19 patients (12 men and 7 women) with tongue cancer took part in the present study. The patients’ tongue movement was recorded using midsagittal Bmode ultrasound a few days before the glossectomy and two months postoperatively. During each examination the patients repeated the following VCV sequences five times: aka, ala, ara, asa, asha and ata. The midsagittal tongue movements were quantitatively analyzed using the Ultra-CATS software. The type of defect reconstruction (local closure vs. free flap) comprised the between-subject factor in the analysis. Results: Following the surgery, all patients showed an increase of the tongue height. The statistical analysis of the tongue height (ANOVA) revealed a significant main effect of time (pre- vs. post-surgery) for the VCV sequences “ara” (df= 1, F= 8.11, p= 0.012), “asa” (df= 1, F= 12.86, p= 0.002) and “asha” (df=1, F= 7.02, p= 0.018). For these sounds, the tongue height was significantly higher after the surgery. The analysis across the two groups of patients (local closure vs. free flap) did not show a significant main effect of reconstruction type for any of the VCVs. Discussion: As in the previous research, we found a significant increase in the postsurgical tongue height during the production of the target consonants. This effect was noted for all sounds but was statistically significant for “r”, “s” and “sh”, which are among the most frequently distorted speech sounds in this patient population. We interpret the results as evidence for active and often hypermetric articulatory compensation in partial glossectomy patients. Reference: Rastadmehr, O., Bressmann, T., Smyth, B., Irish, J.C. Increased midsagittal tongue velocity as indication of articulatory compensation in patients with lateral partial glossectomies (2008) Head and Neck, 30 (6), pp. 718-726.