Aerodynamic measures of glottal function: what extra can they tell us and how do they guide management?

Abstract

Purpose of review

This article will define the major advances in laryngeal aerodynamics research from recent evidence-based literature.

Recent findings

Recently published research focuses on new applications of aerodynamic parameters to improve patient diagnosis and outcomes, as well as further elucidating the mechanisms of phonation using computational modeling and excised larynges.

Summary

Although there is an extensive amount of research on improving the diagnosis and treatment of voice disorders using aerodynamics, the majority of recent literature lacks any conclusive evidence on new methods for use in the clinic; further research in these is needed. The best practices for resonance tube phonation in water and semi-occluded voice therapy are being investigated, as is the exact mechanism by which glottal airflow interacts with vocal folds to produce phonation. It is recommended that clinicians evaluate patients with Parkinson’s disease on the basis of airflow declination and lung volume expended per syllable to avoid dependence on an acoustic signal. In addition, advances in modeling laryngeal disorders and structures will contribute to future research into treatments and diagnosis. Now that the groundwork has been laid, it is crucial to begin evaluating such techniques in patient populations.

INTRODUCTION

Voice production is the result of a complex interaction between the aerodynamic input to the larynx from the lungs, the reactionary vibration of the vocal folds, and the resulting acoustic output. A disordered voice may be a result of problems throughout the respiratory and vocal tracts, as well as the vocal folds themselves. For example, subglottic stenosis or asthma diminishes the air support for phonation and results in a breathy or strained voice. Epiglottic malformations similarly restrict airflow above the larynx, significantly changing the voice quality [1]. However, most patients experiencing disordered voice have an abnormality interfering with their glottal function. In this review article, we discuss the aerodynamic measurements of glottal function and their value in the clinic.

During phonation, airflow from the lungs builds up beneath the closed glottis and pushes the vocal folds apart, allowing the air to flow through the glottis. As the glottis begins to close, because the air column continues to move toward the mouth without replacement from the intraglottal flow, a negative pressure area is created above the folds [2–4].

Changes to glottal aerodynamics often manifest in perception of voice quality and changes in acoustic parameters. Therefore, in addition to measuring the aerodynamic parameters detailed in the following article, acoustic parameters, as well as perceptual voice analysis, are often evaluated.

APPLICATIONS OF PARAMETERS

It is well established that phonation threshold pressure (PTP) is affected by changes to vocal fold shapes and behaviors [5–7]. Increases in PTP can indicate the presence of a disorder, such as a vocal fold lesion or vocal fold paralysis, and often manifest as decreased voice quality [8,9]. However, PTP is also significantly affected by supraglottal pressure, and manipulating supraglottic pressure has been a special subject of research over the past year [10,11]. Resonance tube phonation was introduced in the 1960s as a voice therapy method [12]. Later modifications included straw phonation therapy, or semi-occluded voice therapy, that functions by increasing supraglottal resistance to flow and reducing the pressure difference across the glottis. Such a tube phonation therapy also causes changes in vocal tract shape; thus it can be helpful for patients with pressed or hyperfunctional voices [13].

Enflo et al. [14▪] have evaluated resonance tube phonation in water (RTPW). In this method, one end of a resonance tube is held tightly in the mouth, whereas the other is immersed in water to a certain depth; then the patient phonates such that bubbles are produced. The oral pressure must exceed the depth of the tube immersion to release a bubble from the end of the tube, and the oral pressure drops once the bubble is ejected. This creates a pulsatile oral pressure that may be beneficial to vocal tract shapes. Further, the authors point out that patients are also made aware of their airflow with this method. In this study, the effects of RTPW on voice quality, PTP, and collision threshold pressure (the minimum pressure required to produce sustained vocal fold collision) were evaluated. RTPW produced audible beneficial effects on voice quality, although they were smaller for formally trained or practiced singers, as well as a significant increase in collision threshold pressure and a nonsignificant increase in PTP. The authors hypothesized that the increase in collision threshold pressure may be due to vocal loading during the therapy and increased blood flow to the vocal folds, although more work is needed to determine this exact mechanism. In addition, evaluating RTPW using objective acoustic measures could provide greater support for the practice and mechanistic insight.

Similarly to RTPW, semi-occluded voice therapy, or straw phonation therapy, increases the supraglottal pressure and reduces PTP [10,15]. Different combinations of constrictions and elongations of the vocal tract, as well as the potential addition of oppositional airflow, were evaluated using an excised canine larynx model [16▪]. A significant decrease was observed for phonation threshold flow (PTF) with the combination of 200 ml/s oppositional airflow and 6.5mm constriction, and significant decreases in PTP were observed for the 114 and 200ml/s oppositional airflows, 30-cmextension, and 17.5 and 6.5mm constrictions. The authors observe that more drastic changes might be noted with the addition of a humidifier to the supraglottal airflow. Future studies should determine the optimal combination of constriction, extension, and oppositional airflow, as well as confirm the findings in humans.

ADVANCES IN MODELING

The inaccessibility of the larynx in humans and a small patient population are two major factors in the amount of research devoted to developing new excised and computational models. These models, building on previous studies, are becoming more elaborate and robust. In order to advance research on the effects of supraglottic structures and deformities on aerodynamic and acoustic measures of voice, Smith et al. [17▪] were able to add a simple, full-size vocal tract to excised canine larynges so that in-vivo phonation could be more accurately approximated. They found, as expected by previous studies [18], that the addition of a vocal tract caused a significant decrease in phonation threshold pressure and flow by increasing the nonlinear interactions between the source (the larynx) and the filter (the vocal tract). This addition of the vocal tract to excised larynx modeling will make research even more applicable to patients.

A big trend in computational modeling over the past years has been new ways of modeling disorders [19–21]. Xue et al. [22▪] modeled the tension imbalance produced by unilateral vocal fold paralysis. In the normal state, the interaction of the vocal folds with the airflow through the glottis represents a nonlinear self-oscillator; the addition of a tension imbalance, as in vocal paralysis, produces nonlinear dynamic behaviors, such as bifurcations and chaotic vibration, that often manifest with vocal fatigue and a breathy and hoarse voice. Modeling of this disorder offers insight into its physical mechanism. The simulation predicted more glottal leakage in the imbalance model, leading to the breathy quality of the voice, and showed a difference in phase and amplitude between the normal and soft fold. The model also predicted that less energy was transferred to the soft vocal fold than to the normal fold. Because the authors only evaluated a slight tension imbalance of 20%, further research into the glottal leakage and energy transfer with different levels of tension imbalance are warranted.

PHONATION THRESHOLD POWER

Regner and Jiang [23] proposed a novel parameter, phonation threshold power (PTW). PTW is mathematically defined as the product of phonation threshold pressure and phonation threshold flow and conceptually defined as the minimum power needed to produce sustained vocal fold oscillation. Because it is highly dependent on both glottal configuration and vocal fold biomechanics, it was proposed as a potentially useful clinical and research parameter. For the first time, Zhuang et al. [24▪] evaluated the feasibility of its use in humans, its utility in detecting disease, and its sensitivity to treatment. The study found that PTW significantly increased in individuals with either vocal fold mass lesions or a vocal fold mobility disorder compared with normal individuals, but it was not able to distinguish between the two pathological groups. A significant change in PTW was also observed after polyp excision. The authors suggest that PTW, because it gives a more complete representation of vocal fold physiology than either PTP or PTF alone, has application for a wider range of disorders. In addition, because PTW showed higher area under the curve values than PTP or PTF alone, it is better able to determine laryngeal pathophysiology from normal. However, as for any newly established parameter, the range of normal PTW should be established to create a baseline for physicians in the clinic. The authors also proposed to standardize the current method of collecting PTW, as two separate tasks were used to collect PTP and PTF.

NEW EVALUATION OF PARKINSON’S DISEASE

Patients with Parkinson’s disease often suffer a lack of control of their laryngeal muscles, causing voicing errors and impairing patients’ abilities to communicate [25]. Voice onset time has been used to evaluate voice symptoms of neurological disease, although this depends on a clear acoustic signal, which some patients with Parkinson’s disease may be unable to produce, and does not yield insight into laryngeal physiology. Hammer [26▪▪] examined voice onset time (VOT) and airflow declination during speech in patients with Parkinson’s disease to elucidate the mechanisms of laryngeal and respiratory action in Parkinson’s disease. Patients with Parkinson’s disease, compared with healthy controls, had a shorter VOT, exhaled less of their lung volume per syllable during speech, and had a larger airflow declination (indicating faster phonation onset). There was a large correlation between both VOT and lung air volume expended per syllable with voice severity; VOT showed a medium correlation with airflow declination and lung air volume expended per syllable. Therefore, airflow declination and lung air volume expended per syllable may be clinically useful in patients with Parkinson’s disease, as they can function independently of a clear acoustic signal; the authors propose to use these parameters in tandem with VOT, as well as imaging techniques. This study also provides some illumination into the laryngeal and respiratory contributions to speech in patients with Parkinson’s disease, and will lay the groundwork for further research into these mechanisms. More research is necessary to validate these aerodynamic parameters with more complex speech tasks than monosyllable production.

GLOTTAL JET DYNAMICS

A significant portion of recent research into laryngeal aerodynamics has been looking to determine how intraglottal flow contributes to vocal fold vibration.

In one study [27▪▪], the velocity fields and intraglottal geometry at different subglottal pressures were measured with particle imaging velocimetry (PIV) during vibration in an excised canine larynx. Although Bernoulli’s equation can predict the movement of the central glottal jet, certain aspects of the flow dynamics cannot be predicted in the same way. With increasing subglottal pressure, glottal flow separates from the vocal folds, causing flow vortices at the superior aspect of the vocal folds. The flow was not observed to separate at the glottal entrance, contradicting the findings of a previous computational model; more research is needed to reconcile this and other potential discrepancies between computer models of the glottal flow and excised larynx measurements. Furthermore, it remains to be seen what effects the flow separation vortices created at higher subglottal pressures have on the vibration of the vocal folds, as well as what contribution they might have to sound production; however, studies have been done to determine the direct effects on pressure of the flow separation vortices.

Khosla et al. [28▪▪] looked to illuminate the mechanism by which airflow is manipulated by the vocal folds into sound, and how airflow supports sustained vocal fold oscillation. Much of the research into glottal jet dynamics has been accomplished by using theoretical models, which requires a number of contradicting assumptions to be made as follows: that flow separation occurs at the glottal exit, not within the glottis; that flow vortices produced by flow through the glottis do not affect the pressure; or that flow separation vortices create significant negative pressures and cause intraglottal flow separation. Using excised larynges and PIV, the authors evaluated the glottal flow fields at low, medium, and high subglottal pressures. At low pressures, the first assumption held as very little flow separation was seen; however, at higher pressures unilateral vortices appeared, and bilateral vortices appeared at the highest pressures. The data also indicated that when flow separation vortices appeared, the negative pressures were significantly lower than those predicted by theoretical models. There was no evidence to support the second assumption. Future studies will be done to determine the relative contributions of such flow separation vortices and the inherent tissue mechanics of the vocal folds. These mechanisms can help contribute to therapeutic methods that focus on increasing vocal loudness without increasing subglottal pressure to the point where it causes trauma.

Similarly, Oren et al. [29▪▪] aimed to show the ability of intraglottal flow separation vortices to produce negative pressure at the superior vocal fold edge. Using excised larynges, the flow fields were evaluated with PIV; the intraglottal pressure was computed from these fields. At increasing subglottal pressures, the magnitude of the negative pressure produced by the flow separation vortices increased as well. The authors point out that further work is necessary to determine the role of flow separation vortices in vocal fold closing.

This line of research has potential clinical applicability and should be pursued further. A better understanding of the glottal closure mechanism may lead to therapy modifications that prevent trauma from vocal misuse.

CONCLUSION

Aerodynamic parameters, because they reflect both vocal fold biomechanics and glottal geometry, are particularly useful in clinical evaluation of voice. The medical and research communities continue to develop and refine methods of voice measurement, shifting toward new applications of parameters, as well as continuing to examine the mechanisms of voice production.

This review article summarizes the recent advances made in laryngeal aerodynamics and how results from evidence-based studies may inform clinical diagnoses and treatments. Future research will be directed toward increasing sensitivity and improving patient outcomes, as well as into understanding the interactions between the respiratory airflow and the vocal folds in voice production.

KEY POINTS.

• Commonly used tube phonation techniques need to be standardized and systematically investigated to identify therapeutic benefits and best practices.

• Phonation threshold power, the product of phonation threshold pressure and phonation threshold flow, may provide a fuller representation of laryngeal health, as PTP and PTF alone are sensitive to different laryngeal changes.

• New research into more robust computational and excised larynx models shows promise for future studies of voice disorders and elucidating the mechanism of phonation.

• Research into glottal jet dynamics reflects growing interest into the interaction between glottal airflow and vocal fold movement.

• Future research directed toward patient diagnosis and outcomes is crucial.