Effects of Increasing Vocal Tract Insertion Depth in Semi-Occluded Vocal Tract Exercises on an Excised Canine Model

Abstract

Introduction:

Straw phonation therapy, a form of semi-occluded vocal tract (SOVT) exercise, is commonly used to help treat various voice disorders. Although straw phonation therapy has been studied extensively for decades, the impact of straw depth on vocal function remains unexplored. This study aims to quantify the effects ofvarious straw vocal tract insertion depths (VTID) into the vocal tract on common aerodynamic parameters such as phonation threshold pressure (PTP), phonation threshold flow (PTF), and phonation threshold power (PTW) in an ex vivo canine model. It was hypothesized that increasing the VTID of the straw would reduce the PTP, PTF, and PTW.

Methods:

Five excised canine larynges were mounted on a pseudolung apparatus and attached to a simulated vocal tract in an acoustic sound booth. Two straw depths (20.0 mm and 60.0 mm) were tested to determine the effect of VTID on ease of phonation as indicated by the aerodynamic parameters of PTP, PTF, and PTW. The control had no straw and a VTID of 0.0 mm. The straw diameter and length above the simulated vocal tract were consistent between the straws. Sustained phonationwas achieved and aerodynamic data was collected and analyzed.

Results:

Both straw treatment groups exhibited significant reductions in PTP and PTW compared to the control. However, there were no significant differences between the 20.0 mm or 60.0 mm straw depths in PTP, PTF, or PTW.

Conclusion:

The presence of a straw significantly reduced PTP and PTW, but VTID did not appear to influence these outcomes. This supports previous straw phonation therapy research suggesting that straw phonation therapy is beneficial primarily due to the presence of the straw rather than the depth of insertion. Future studies should explore the combined effects of varying straw diameters, lengths, and depths to optimize SOVT therapy for clinical use.

INTRODUCTION

Various voice disorders affect millions of people around the United States, as Fujiki & Thibeault et al report the lifetime prevalence of a disordered voice in adolescents is 24.3% ¹. The prevalence and negative effects of vocal disorders in the elderly population could be even greater Pernambuco et al report that the prevalence of vocal disorders in the general population of people 60 years old or older could be up to 29.1% ². Voice disorders can profoundly impact job performance, especially for professions that require extensive voice use, such as teachers, musicians, salespeople, physicians, lawyers, and customer service representatives ³. Voice disorders reduce one’s ability to communicate, leading to a reduction in quality of life. Semi-occluded vocal tract (SOVT) exercises are one form of voice therapy utilizing a straw to increase supraglottal pressure. This induces an efficient glottal closure pattern that supports sustained phonation ⁴. Straw phonation is effective in treating symptoms of dysphonia ⁵, vocal fatigue ⁶, hyperfunctional voice disorders (HFVD) ⁷, and vocal lesions ⁸. This form of therapy is commonly known in literature as straw phonation therapy. It has been increasingly used in clinical settings since it was first published in 1904 ⁹. Straw phonation therapy has also been shown to reduce phonation threshold pressure (PTP) making phonation easier to achieve ¹⁰. This is achieved by introducing an effectively lengthened vocal tract and minimizing the cross-sectional area, thus changing the impedance and increasing inertance ¹¹. Beyond voice disorders, straw phonation therapy has also been shown to have neurological benefits in treating neurological diseases such as Parkinson’s Disease in conjunction with therapeutic group singing ¹². SOVT exercises help patients identify the feeling of proper, strong phonation in the form of a resonant voice, which allows them to implement this into their speech in their everyday lives ¹⁰. In human subjects, controlling supraglottic pressure using a continuous positive airway pressure (CPAP) device combined with a straw-shaped mouthpiece significantly lowered PTP during a long-duration task at 2 cm H₂O supraglottic pressure. PTF and acoustic parameters did not change significantly ¹³. InJiang & Maytag et al, researchers evaluated tube phonation in water to investigate the interaction between airflow and vocal folds that produced phonation ¹⁴. Changing the depth of the tube in the water created changes to the oral pressure, resulting in a change in the shape of the vocal tract. For this therapeutic approach to be effective, it was determined that the depth of the tube must be less than the oral pressure to facilitate a bubble being released. However, the study did not specify the exact depth or specific configuration of the vocal tract required to achieve this effect. It was found that this change in shape can be beneficial to patients with hyperfunctional voices. The results of this study presented the subjects reporting easier and more comfortable phonation following straw phonation ¹⁴. In this study, we investigate how the vocal tract insertion depth (VTID) of a SOVT affects the ease of phonation as indicated by aerodynamic measures. VTID refers to the distance that the SOVT is inserted into the vocal tract or mouth of the user. In Tangney et al., researchers tested four different straw lengths and inner tube diameters to determine which length and diameter produced the greatest phonation ease by analyzing parameters such as PTP, PTF, contact quotient, and supraglottal pressure ⁹. This study found that the straws with lengths of 5 and 25 cm, with a depth of 0 cm, and inner tube diameters of 9 and 15 mm significantly decreased PTP ⁹. Optimization of SOVT straw dimensions of length and diameter was further investigated by Titze et al. Flow resistances and oral pressure were effectively displayed in nonograms that can be used to select an optimal straw depending on the situation and desired resistance ¹⁵. Regner et al. analyzed and quantified the minimum flow rate required for phonation which falls within the onset-offset PTF ratios bound by [0.707, 1.0] ¹⁶. Our study aims to expand the research to determine the most advantageous depth of the straw, leading to further optimizations of SOVT straw phonation therapy.

Phonation threshold power (PTW) may be a useful measure encapsulating the aerodynamic and biomechanical vocal fold properties and is derived from the interaction between PTP and PTF. PTW represents the lowest level of aerodynamic power necessary to start phonation which helps quantify its energetic relationships ¹⁷. The equations provided show that PTW can be calculated as the product of PTP and PTF, and further detailed expressions involve various physical parameters: the transglottal pressure coefficient $\left(k_t\right)$, damping constant ($B$), mucosal wave velocity ($c$), prephonatory glottal half-width $\left(x_o\right)$, vocal fold thickness ($T$), vocal fold length ($L$), and glottal airflow density ($\rho$) ¹⁷.

$$\mathrm{P}\mathrm{T}\mathrm{P}=\left(k_{t}Bc{x}_o\right)/T$$

$$\mathrm{P}\mathrm{T}\mathrm{F}=4L\sqrt{\left(2Bc{x}_o^3\right)/\rho T}$$

$$\mathrm{P}\mathrm{T}\mathrm{W}=\mathrm{P}\mathrm{T}\mathrm{P}\times \mathrm{P}\mathrm{T}\mathrm{F}=k_{t}L\sqrt{\left(8B^3{c}^3{x}_o^3\right)/T^3\rho }$$

SOVT can increase supraglottal pressure, significantly reduce PTP, and increase the ease of phonation ¹⁸. This paper aims to understand how altering the depth of the SOVT straw into the mouth maximizes the benefits of phonation therapy through PTP, PTF, and PTW in an ex vivo canine model. PTW can represent airflow-tissue energy transfer more completely than PTP or PTF alone, which can be a more useful parameter for assessing laryngeal health. PTW can potentially be utilized as a screening parameter that is sensitive to biomechanical laryngeal change. SOVT therapy normally utilizes straws without intuitively inserting them deeper into the vocal tract and closer to the glottis. While lengths and widths of straws have been researched, the impact of the depth of the straw into the vocal tract has yet to be understood. We aim to quantify this relationship using aerodynamic measurements to determine if VTID can improve the benefits of SOVT straw phonation therapy. According to a study, straw phonation decreased PTP due to the increase in inertive reactance in the vocal tract ⁹. Reactance, in contrast to resistance, facilitates phonation by supporting the efficient transfer of energy within the vocal system. As the straw is inserted closer to the vocal tract, it is hypothesized that the inertive reactance will be greater, resulting in a lower PTP. This ease in PTP directly correlates to the belief that PTF will also decrease at a larger insertion depth. A lower PTP, implying the vocal folds can vibrate more easily, suggests a reduction in the airflow required to sustain phonation. Increasing the VTID of the straw is hypothesized to increase inertance by decreasing the cross-sectional area of the vocal tract ¹⁹. Titze et al used computational methods to create inertance curves for various SOVT configurations. These inertagams graphically demonstrated how adjustments in length, diameter, and shape of the vocal tract impact inertive reactance, allowing for a precise way to understand how various straw configurations affect the dynamics of phonation. This computational approach showed that altering effective vocal tract configurations changed inertegrams depending on distance of the constriction from the glottis ²⁰. Therefore, changing VTID may be a useful way to further optimize SOVT therapy.

Excised canine larynges were used in this study to simulate consistent phonation in a controlled setting. Two straws with identical dimensions other than VTID were used to elucidate the effect of depth in SOVT straw phonation as measured by PTP, PTF, and PTW. It was hypothesized that a straw treatment would decrease PTP, PTF, and PTW, with the largest reduction using the largest straw VTID of 60.0 mm.

METHODS

Five canine larynges of unknown breeds and sexes were used for data collection, all from canines that were ethically euthanized for purposes unrelated to this study. All guidelines provided by the University of Wisconsin-Madison Institutional Animal Care and Use Committee were followed. Canine larynges are optimal for laryngeal research, as they are more readily available than human larynges. They are biologically comparable due to their size and vocal fold histology ⁹. The average vocal fold lengths of the layrnges used was 13.1 mm ± 1.95 mm. The larynges are frozen at −16°C in a 0.9% saline solution to prevent damage, dehydration, and tissue decay. After preparing and excising the larynges for data collection in the procedure outlined by Jiang & Titze et al, they were mounted on the pseudolung apparatus and connected to the simulated vocal tract as seen in Figure 1 ²¹. This pseudolung apparatus simulates the human respiratory system and allows the excised larynx to be phonated similarly to in vivo function. The pseudolung was in a sound-attenuated room isolated from outside interference of sound, temperature, and humidity. The larynx was mounted to the pseudolung, and airflow to induce phonation was controlled manually by a needle valve and measured by an airflow meter (model FMA-1601 A; Omega Engineering Inc., Stamford, CT) ⁹. The epiglottis of the excised canine larynges were glued to the laryngeal insert of the simulated vocal tract to ensure a proper seal was created to prevent air leakage, as seen in Figure 3. Subglottal pressure was measured using an Omega Pressure Transducer (part number MMG005V10W2MA0T1A4, Omega Engineering, Inc, Stamford, CT) and was recorded through the combination of all the raw data inputs to a data acquisition board (model USB-6229; National Instruments Corporation, Austin, TX). The inputs were recorded and visualized by a custom LabVIEW (Nation Instruments Corporation, Austin, TX) program ⁹.

FIGURE 1:

pseudolung and excised bench setup.

FIGURE 3:

Larynx mounted on the psuedolung apparatus with a straw insert in the simulated vocal tract.

This study used two different VTIDs; the short straw treatment experimental group had a diameter of 8.0 mm and was inserted 20.0 mm into the straw insert of the simulated vocal tract as seen in Figure 2, meaning that the VTID was 20.0 mm. The long straw experimental group had a diameter of 8.0 mm and was inserted 60.0 mm into the straw insert of the simulated vocal tract, meaning the VTID was 60.0 mm. For both straw treatments, the length of the straw above the simulated vocal tract was 50.0 mm. The control for the experiment was no straw being inserted into the straw insert of the simulated vocal tract, therefore its VTID was 0.0 mm. Each of the five larynges was tested for the control and VTID treatment groups, and five seconds of continuous and consistent phonation was recorded for three separate trials for each condition. Proper phonation was defined as when the phonation was audibly stable as perceived by the fundamental frequency and harmonics which can be quantified as when PTP is reached ⁹. PTP and PTF were measured by analyzing the subglottal pressure and subglottal flow ascertained from the custom LabVIEW program at phonation onset. Phonatory conditions were consistent and achieved from manual abduction with prongs, however tension was not directly measured. Ideal adduction was determined by altering the position of the metal prongs until the larynx could produce ideal phonation as indicated by auditory perception and defined harmonics. Once this was achieved, the metal prongs were held constant for each larynx throughout the study. The larynges were hydrated with 0.9% saline solution and given a 5-minute rest period following each condition to minimize dehydration and fatigue. The order of the larynges and treatment types were randomized to negate confounding variables. The custom LabVIEW program was recalibrated before each set of data collection to ensure accurate quantification.

FIGURE 2:

schematic for the straw insert in the simulated vocal tract setup.

DATA ANALYSIS

Repeated measures ANOVA tests were used to determine the significance between averages of the control and the straw insertion treatment groups for the PTP, PTF, and PTW. All statistical analyses were conducted with a significance level of ɑ = 0.05. Statistical analyses were conducted in RStudio using unpaired t-tests with a Bonferroni correction to identify significant differences between conditions which resulted in a new alpha value of .0166.

RESULTS

PTP:

Repeated measures ANOVA analysis of PTP (F[28] = 26.044, P-value = 4.08 × 10⁻⁷) showed a significant difference between the averages of the control and the experimental groups of 20.0 mm and 60.0 mm. Unpaired t-test analyses determined that the PTP of the control (18.46 ± 3.49 cm H₂O) was significantly larger than 20.0 mm VTID (13.10 ± 1.40 cm H₂O) (p = 1.44 × 10⁻⁵) and 60.0 mm VTID 13.54 ± 1.98 cm H₂O) (p = 4.77 × 10⁻⁵); however, there was no significant difference between 20.0 mm VTID and 60.0 mm VTID (p = 0.247) (Figure 4). Overall, PTP values ranged from 10.59 to 24.56 cm H₂O.

FIGURE 4:

Bar graph of phonation threshold pressure (PTP) with different vocal tract insertion depths. P-values were obtained from repeated ANOVA measures versus control with a significance level of ɑ = 0.016. ns = p > 0.016, * = p < 0.016, *** = p < 0.001, **** = p < 0.0001. Error bars show one standard deviation.

PTF:

Repeated measures ANOVA analysis of PTF (F[28] = 0.941, P-value = 0.402) did not show a significant difference between the PTF averages of the control and experimental groups of 20.0 mm VTID and 60.0 mm VTID (p > 0.05) (Figure 5). Unpaired t-tests showed no significant differences between the means of the control (4.48 ± 0.56 L/min) and experimental groups of 20.0 mm VTID (4.31 ± 0.37 L/min) and 60.0 mm VTID (4.52 ± 0.44 L/min), as expected due to the non-significant results of the repeated measures ANOVA analysis. Overall, PTF values ranged from 3.63 to 5.44 L/min.

FIGURE 5:

Bar graph of phonation threshold flow (PTF) with different vocal tract insertion depths. P-values were obtained from repeated ANOVA measures versus control with a significance level of ɑ = 0.016. ns = p > 0.016, * = p < 0.016, *** = p < 0.001, **** = p < 0.0001. Error bars show one standard deviation.

PTW:

Repeated measures ANOVA analysis of PTW (F[28] = 28.297, P-value = 1.89 × 10⁻⁷) showed a significant difference between the averages of the control and the experimental groups of 20.0 mm VTID and 60.0 mm VTID. Unpaired t-test analyses determined that the PTW of the control (82.19 ± 15.78 cm H₂O × L/min) was significantly larger than 20.0 mm VTID (56.69 ± 9.64 cm H₂O × L/min) (p = 9.82 × 10⁻⁶) and 60.0 mm VTID (60.59 ± 5.86 cm H₂O × L/min) (p = 5.14 × 10⁻⁵); however, there was no significant difference between 20.0 mm VTID and 60.0 mm VTID (p = 0.0967) (Figure 6). Overall, PTW values ranged from 41.57 to 108.90 cm H₂O × L/min.

FIGURE 6:

Bar graph of phonation threshold power (PTW) with different vocal tract insertion depths. P-values were obtained from repeated ANOVA measures versus control with a significance level of ɑ = 0.016. ns = p > 0.016, * = p < 0.016, *** = p < 0.001, **** = p < 0.0001. Error bars show one standard deviation.

DISCUSSION

The purpose of this study was to determine if using an optimized straw at various VTIDs could improve the effectiveness of straw phonation therapy, measured by phonation threshold pressure (PTP), phonation threshold flow (PTF), and phonation threshold power (PTW). Our study tested the hypothesis that a straw treatment would decrease PTP, PTF, and PTW, with a more significant reduction occurring from a larger VTID than a smaller one. No significant differences in PTP, PTF, or PTW were found between the short (20.0 mm VTID) and long (60.0 mm VTID) straw treatments. This suggests that, while the presence of a straw affects PTP and PTW, the increased VTID does not significantly enhance the therapy’s effectiveness in an ex vivo model.

Our findings confirmed that using a straw can lower the pressure required to start phonation, which aligns with our hypothesis and previous research ^{6, 10, 22}. However, the anticipated PTP, PTF, and PTW reduction with the longer depth was not observed. This indicates that, while using a straw insertion can influence PTP and PTW, the length and increased VTID do not provide additional significant benefits.

In this study, PTP and PTW results showed significant differences between the control and both VTID treatment groups, but not between VTID treatments. This indicated that straw use effectively reduces the energy transfer from airflow to tissues required for phonation initiation. The absence of significant changes between short and long VTIDs suggests that the combined effect of PTP and PTF did not alter sufficiently to produce pronounced differences in PTW. Given the relatively novel use of PTW as a metric, these findings contribute to a deeper understanding of its behavior in vocal fold mechanics. The equations highlight that while reducing PTP tends to lower PTW, any lack of change in PTF can limit the overall reduction in PTW, suggesting that both pressure and flow dynamics need to be considered in optimizing therapy.

The physics of acoustic inertance can explain the changes in airflow due to straw insertion. While straw phonation works by effectively lengthening the vocal tract, a larger glottal gap introduces an effectively smaller cross-sectional area of the vocal tract. Inertive reactance, $I$, which facilitates self-sustained oscillations is defined by $I=pL/A$. $p$ is the density of the air column, $L$ is the length of the vocal tract, and $A$ is the crosssectional area ¹⁹. Therefore, lengthening VTID may be an alternative way to increase inertance via a decreased cross-sectional area rather than solely lengthening a straw. However, our results indicate that in the controlled ex vivo setting, the expected increase in inertance from deeper straw insertion did not translate into measurable improvements in phonation metrics. This could be because the airflow resistance of the tube, which did not change across the experimental tube conditions, controlled the aerodynamic outcome measures more than any potential changes in inertive reactance that varying vocal tract-tube insertion depth might yield. A maximum depth is achieved when the straw is longer than the vocal tract, contacting the glottis. This would likely interrupt phonation and is not practical for therapeutic use due to limitations imposed by the pharyngeal reflex. The pharyngeal reflex occurs when an object interacts with the uvula, the base of the tongue, or the tonsil area ²³.

Our research on VTID investigated whether this alteration in resistance would significantly reduce the effort required for phonation. Although the larger VTID did not show a significant additional benefit, understanding how straw depth affects airflow can guide future adjustments in therapy techniques and potentially guide future adjustments in therapy protocols. Clinically, this suggests that therapy protocols can prioritize straw use for reducing phonation effort, without emphasizing depth variations, thereby simplifying patient instructions and potentially improving adherence.

Increasing the sample size of larynges could further the reliability of the results. Furthermore, testing more VTIDs between 0.0 mm and 60.0 mm would be beneficial to quantifying the effects of the straws at a wider range of VTIDs on PTP, PTF, and PTW, which would help supplement the data collected in this project. However, because there were no significant differences in PTP, PTF, or PTW between straw depths, it is hypothesized that different VTIDs would not demonstrate significant differences.

CONCLUSION

This study focused on quantifying the PTP, PTF, and PTW in ex vivo canine larynges using optimized straws at different VTIDs to understand how straw phonation therapy with a semi-occluded vocal tract can treat voice disorders effectively. Two treatment groups of varying straw VTIDs were compared to a control group without straw usage. Our results showed a significant decrease in PTP and PTW in both treatment groups compared to the control, confirming the therapeutic potential of straw phonation. However, we found no significant differences between the 20.0 mm and 60.0 mm VTID groups for PTP, PTF, or PTW. Future studies could benefit from directly measuring tension and testing variations in straw depth, oral diameter, and length to understand further insights into optimizing straw phonation therapy. Applying this research to human subjects could allow for more clinically relevant applications, such as measuring patient-reported outcomes or vocal fold function under different straw conditions, providing valuable data to improve clinical recommendations.