Participants
A total of sixty-one young adults (mean age= 23.49 years, SD age=2.9) took part in this study. Participants were students recruited from the University GimbernatCantabria (Spain) with no prior or ongoing vocal pathology. None of them referred to having a history of elite voice use (e.g., singing, acting, teaching, etc.). The study was carried out in the CFL clinic (Centro de Foniatría y Logopedia in Santander) with an evaluator, a phoniatrician with 40 years of experience (phoniatrics is a medical specialty focusing on voice, speech, language, hearing and swallowing disorders), who blindly evaluated the participants. Five speech therapists experts in voice disorders had the task of accompanying, supervising and guiding the participants throughout the whole experimental setting.
The experiment was carried out prospectively and double-blinded, because neither the participants nor the evaluator knew the treatment group assigned to each participant. There were three different treatment groups: Exercise group (20 participants), Gauze group (21 participants) and a Control group (20 participants). The groups were set up to have the same number of females so as to control gender differences in voice parameters (16 females in each group).
Participants were appropriately informed, and signed consent forms before the experiment. The protocol was carried out according to the guidelines approved by the
CFL (Centro de Foniatria y Logopedia) Ethics Committee in line with the Professional College of doctors and of Speech Therapists of Cantabria. Students were rewarded with a present after the experiment.
Procedures
The entire study lasted for about thirty minutes, during which participants followed the same procedure: initial self-assessment, pre-test, treatment task and posttest. Participants were individually tested before (pre-test) and after (post-test) a double-blinded assigned treatment task (Gauze, Exercise and Control) with three tests (laryngostroboscopic examinations, acoustic analysis and perceptual evaluations). Each participant was accompanied all the time by one speech therapist for guiding and supervision. A program, presented on a 21-inch PC monitor running Experiment Builder® software, assigned the participant to a treatment group, blindly and randomly by gender. The participants did not know which treatment group they had been placed in.
Firstly, participants carried out an initial self-assessment before they started the study. They filled out a personal survey, in order to evaluate the inflammatory factors that could affect the degree of vocal well-being at the level of the glottis and of the upper respiratory tract, as previous research showed that they could affect the voice quality (32,33). Those factors should be taken into account in order to control that all participants have the same homogeneity in terms of their voice quality. They had to subjectively rate different parameters according to their own perception. They were asked to rate on an ordinal scale going from 0 to 10 their perception of the degree of dysphonia (i.e., the question in Spanish was “tu voz es ronca” which is the informal term, and everybody has a subjective perception, to refer that the voice sounds breathy raspy, strained or will be soften in volume or lower in pitch). Also, they had to rate on ordinal scales from 0 to 4 (where 0 is none and 4 is a lot) the degree of vocal effort while speaking, how much they liked their voice, the amount of tobacco consumed per week, the alcohol consumed per week, the degree of nuisance perceived while speaking, and the propensity of colds suffered per year. Lastly, participants had to fill in the Spanish version of the Voice Handicap Index (VHI-30) (34) with its three subscales: emotional, physical, and functional. A total score (possible range, 0–120 points) and subscale scores (possible range, 0–40) were calculated; the higher the score, the greater the voice handicap.
Before and after the treatment tasks, participants were evaluated through laryngostroboscopies, acoustic analysis and perceptual evaluations.
Laryngostroboscopic examinations were carried out by the experienced phoniatrician with a 70° Karl Storz telelaryngoscope and a Karl Storz Pulsar stroboscopic light source
(35) and were recorded on a Windows PC Inter Core i5-6400 2.70 GHz using Adobe Premiere Pro 2.0 (36). Laryngostroboscopy creates a low-motion illusion of vocal fold vibration by compiling images at different points (30 frames per second) of each vibratory cycle, helping clinicians to interpret the physiology of the vocal folds (e.g. the symmetry and periodicity of glottic vibration, glottic closure, profile of vocal fold, edge, amplitude of vocal folds vibration and mucosal wave) (37–39). Although other techniques can be employed to measure the mucosal wave (e.g., high speed digital imaging, Videokymography), we chose this tool because it is the most commonly used by clinicians and has the advantages of a low-cost, quick and easy technique (40). Furthermore, there are no significant differences found when comparing mucosal wave parameters extracted from laryngostroboscopic images and other techniques in patients with no voice problems (13).
To evaluate the glottic function, three measures of interest were deducted from the laryngostroboscopic examination: two subjective parameters (the Amplitude of the mucosal wave and the Glottic Closure) and one objective parameter (the Maximum Opening of the Glottic Space). The Glottic Closure parameter measures the degree of glottic competence and can be complete, inconstant, or incomplete. The experienced phoniatrician, blind to the treatment group of the participants, evaluated the degree of the glottic closure rating from 0 (complete closure) to 4 (incomplete closure). The Amplitude of the Mucosal Wave parameter evaluates the progression of the wave generated by the flowing of the vocal fold epithelium on the vocal ligament due to the lax tissue contained in the superficial layer of the Lamina Propria, driven by subglottic pressure, myoelastic and aerodynamic forces. Also, the experienced phoniatrician rated this parameter on a scale from 0 (absence of mucosal wave) to 4 (normal mucosal wave). In order to have reliability in those subjective measures, an inter-rater reliability analysis was conducted. Three experienced judges outside the experimental process were asked to evaluate the Amplitude of the mucosal wave and the Glottic Closure. The external judges were three external speech therapists who have an experience of more than 35 years evaluating and treating voice disorders. They were instructed to follow the same evaluation procedure as the phoniatrician, rating from 0 to 4 the laryngostroboscopic examinations recordings.
The objective laryngostroboscopic measure, the Maximum Opening of the Glottic Space, was taken into account basing measurements on the width of the vocal folds defined by Hirano (38), which subjectively measures the distance of the vocal fold’s edge from the middle area of the glottis to the side. The width is an indicator of how much the mucosal wave moves during phonation. Trying to avoid this subjective evaluation, the maximum opening of the glottic space in relation to the width of the vocal folds was measured objectively. To do so, each participant’s recorded laryngostroboscopic video was frozen at the frame of maximum glottic opening during the vibratory cycle using Adobe Premiere Pro CS5.5. A trained speech therapist identified the line of the maximum glottic space opening and drew a horizontal line going from the Morgagni’s ventricle of the right vocal fold to the contralateral one, allowing measurement of the total pixel width of the vocal folds and glottic space. The ratio between the glottic space and the total width of the vocal folds was then calculated and expressed as a percent measurement (glottic space width X 100/ VFs total width).
This measurement is constant regardless of the distance of the glottis from the endoscopy lens and the variation in the size of the larynx between the participants.