Differential Vibratory Characteristics of Adductor Spasmodic Dysphonia and Muscle Tension Dysphonia on High-Speed Digital Imaging

Abstract

Objectives: The purpose of this study was to quantify disorder-specific signature kinematic disturbances of vibratory motion in adductor spasmodic dysphonia (AdSD) and muscle tension dysphonia (MTD), in voice disturbances of a severe nature, with the use of high-speed digital imaging (HSDI). A secondary hypothesis of the study was to investigate the sensitivity and specificity of the signature kinematic features obtained from HSDI, in differentiating between AdSD and MTD. Methods: We used vibratory features from automated extraction of vocal fold motion waveforms and glottal cycle montage analysis from HSDI for differential kinematic profiling of AdSD and MTD. Results: Novel features of motion irregularities and micromotions (as small as 27 ms) were greater in number for AdSD, whereas reduced motion irregularities, absence of oscillatory breaks, absence of micromotions, and increased hyperfunction characterized the MTD group. Oscillatory breaks (as small as 8 ms), although present only in the AdSD group, were not statistically significant because of their reduced number of occurrences compared to the other features. Further montage analysis of successive glottal cycles of oscillatory breaks in the AdSD group revealed 3 different kinematic patterns within the AdSD group, indicative of likely AdSD with: 1) possible predominant thyroarytenoid muscle involvement, 2) possible predominant cricothyroid muscle involvement, and 3) possible combined involvements of the thyroarytenoid and lateral cricoarytenoid muscles. Four consistent but unique kinematic patterns were identified within the MTD group: 1) diplophonia, 2) vocal fry, 3) breathy phonation, and 4) pressed phonation. Sensitivity and specificity analysis revealed that only motion irregularity was a significant predictor of the presence of AdSD. Conclusions: Fine kinematic analysis from HSDI can be used to aid detailed clinical profiling of the source characteristics of AdSD and MTD.