A comparison of the angular velocity between the vocal folds during phonation and throat clearing under high-speed digital imaging

Abstract

INTRODUCTION: Several studies have reported that loud phonation and frequent throat clearing are problematic behaviors for vocal hygiene. It is presumed that strong collision between the vocal folds during loud phonation or throat clearing injures the laryngeal tissue. However, in normal video images (30 frames/s), it is not possible to continuously measure the velocity of rapid vocal fold adduction. In the present study, we attempted to investigate whether the velocity of vocal fold adduction in the onset of loud phonation and throat clearing is faster in comparison to natural phonation, and whether humming, as a vocal training technique, affects the velocity of vocal fold adduction in the onset of phonation using high-speed digital imaging (HSDI) with high time resolution.

MATERIALS AND METHODS: Twenty normal healthy adults were enrolled in the present study. A transnasal flexible fiberscope connected to a high-speed camera was inserted, then each participant was asked to perform weak/strong throat clearing (TC) and three phonatory tasks: natural /e:/ phonation (NP), loud /e:/ phonation (LP) and humming /m:/ phonation (HP), and the high-speed laryngeal findings were recorded at a rate of 4,000 frames/s. The vocal fold angular velocities were calculated during vocal fold adduction from three points (the anterior commissure and the tips of the bilateral vocal processes) using the Dipp-Motion Pro motion analysis software program (DITECT, Japan). In addition, the average angular velocities were calculated in the ranges of 100–80%, 80–20% and 20–0% from all of the angular changes. These measurements were compared among the five tasks.

RESULTS: The pattern of the changes in the angle between the vocal folds drew sigmoid curves and polynomial curves in NP/HP and TC/LP, respectively. The angular velocity during weak/strong TC and LP continuously accelerated, whereas the angular velocity during NP and HP accelerated once, but then decelerated. The average angular velocity of each range in strong TC was significantly greater than that in NP; in particular, the 20–0% average angular velocity was approximately two-fold greater than that in NP. The average angular velocity of each range in HP was significantly smaller than that in NP, and the 20–0% average angular velocity was approximately half of that which was observed in NP.

CONCLUSION: The present study demonstrated that HSDI enables the continuous analysis of the velocity of vocal fold adduction, and that TC/LP increases the velocity of vocal fold adduction in a continuous manner, producing intense collision just before vocal fold contact. In contrast, HP was found to decrease the velocity of vocal fold adduction, perhaps leading to the weakening of the collision force at the end of vocal fold adduction.