Effect of a speech prosthesis on electromyographic activity levels of the levator veli palatini muscle activity during syllable repetition

OBJECTIVES: To examine whether repeated production of a syllable effects levator veli palatini muscle activity for speakers with velopharyngeal incompetence and whether the effect can be changed by a speech prosthesis. DESIGN: Repeated-measures analysis; each subject produced the speech sample /pu/ more than 200 times in each of 2 experimental conditions. SETTING: Graduate dental school in Japan. PARTICIPANTS: Four patients with operated cleft palate with a speech prosthesis. INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: Electromyographic traces were highband-pass filtered at 30 Hz, rectified, and smoothed with a time constant of 30 ms. Electromyographic traces were made of each production in 2 conditions: (1) without the prosthesis and (2) with the prosthesis. The regression slope of the linear regressor line, when smoothed levator activity was the criterion variable and the number of productions was the explanatory variable, was calculated in each condition. RESULTS: The mean value of levator activity was significantly smaller with the prosthesis in than without it (P<.10, t test). With the prosthesis, the regression slope was significant for all 4 subjects, whereas it was insignificant without the prosthesis for 3 of 4 subjects (P<.10, t test). Absolute values of the regression slope were significantly smaller with the prosthesis than without it for all subjects (P<.10, t test). Comparison of the regression slopes for the 2 conditions identified a significant difference in slopes between the 2 conditions (P<.10, t test). CONCLUSION: In operated cleft palate patients with velopharyngeal incompetence, a speech prosthesis can stabilize both temporal changes in levator muscle activity and connected speech, such as conversation.