ARTICLE IN PRESS Pressure and Flow Comparisons Across Vocal Pathologies *Linda Carroll, †Ann Rooney, *,‡Thomas J. Ow, and *,‡Melin Tan, *†Bronx and ‡New York, New York Summary: Objective. The aim of this study was to aid in the distinction among hyperadductive dysphonias by evaluating peak glottal pressure, release burst, and mid and final airflow values across repeated /pa/ syllable trains. Methods. Sixty subjects were assessed for aerodynamic patterns during onset-offset for the /papapapapa/ task in modal voice. Subject groups included adductory spasmodic dysphonia (AdSD), benign vocal fold lesion, primary muscle tension dysphonia (MTD-1), secondary muscle tension dysphonia with an identifiable primary benign vocal fold lesion (MTD2), vocal fold paresis or paralysis, and normal controls. Results. Increased peak pressure (PP) was found for AdSD and MTD-2 subjects compared with controls. Release burst and mid airflow were not significantly different among groups. Final airflow was significantly higher for AdSD compared with the other groups. Final airflow was significantly lower for MTD-1. Conclusions. Significant differences in aerodynamics are seen in subjects with AdSD compared to MTD. AdSD was characterized by higher PP and higher final airflow. MTD-1 was characterized by lower final airflow, whereas MTD-2 was characterized by higher PP. Aerodynamic evaluation may aid in differential diagnosis for those patients in whom distinction among hyperadductive disorders is challenging. Key Words: Spasmodic dysphonia–Muscle tension dysphonia–Laryngeal aerodynamics–Voice pressure and flow–Larynx. BACKGROUND In normal voice and motor control, aerodynamic measures are characterized by a fairly steady and reliable onset-offset of the glottal flow in the phonemic coordination from consonant to vowel, as well as a fairly consistent strength of intraoral pressure during [p] production. During /pa/ syllable trains, there is a buildup of intraoral pressure as the lips close for [p], leading to a peak pressure (PP), and then a drop-off of pressure as the lips open to begin the final portion of [p] production. As the lips open, the airflow previously held behind the closed lips is suddenly released into the “release burst” (RB), which signals the end portion of the [p] production. Glottal airflow then stabilizes during the [a] portion of the /pa/ syllable train. As the syllable train is repeated, glottal airflow shuts down at the end of the [a] as the lips close for the next [p] production. Dysphonic speech may be marked by voicing discoordination because of mass effect, neurologic disturbance, muscular tension, or mucus loading on the vocal folds. Although voicing disturbances may be examined through acoustic signals, aerodynamic measures permit a detailed examination of flow and pressure changes. Prior studies have examined differences in aerodynamic measures among different pathologic etiologies of dysphonia.1,2 However, little research has evaluated the pressure-flow coordination at syllable transitions in dysphonic subjects. Although peak intraoral pressures may be consistent in the dysphonic population, control of pressure-flow discoordination, particularly during the coordination of one [pa] syllable to the next [pa] production Accepted for publication April 6, 2017. Presented at The Voice Foundation Annual Symposium Philadelphia, PA, June 2017. From the *Department of Otorhinolaryngology—Head and Neck Surgery, Montefiore Medical Center, Bronx, New York; †Department of Rehabilitation Medicine, Montefiore Medical Center, Bronx, New York; and the ‡Department of Otorhinolaryngology—Head and Neck Surgery, Albert Einstein College of Medicine, New York, New York. Address correspondence and reprint requests to Melin Tan, Department of Otorhinolaryngology—Head and Neck Surgery, Montefiore Medical Center, 3400 Bainbridge Avenue, 3rd Floor, Bronx, NY 10536. E-mail: mtangel@montefiore.org Journal of Voice, Vol. ■■, No. ■■, pp. ■■-■■ 0892-1997 © 2017 The Voice Foundation. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jvoice.2017.04.004 in a repeated /pa/ syllable train, may yield important information on glottic function skills. This finding may be particularly true for patients with adductory spasmodic dysphonia (AdSD) who have reduced neuromuscular control at the glottis, when compared to other non-neurologic causes of dysphonia. Higgins et al reported on aerodynamic differences among AdSD, muscle tension dysphonia (MTD), and normal subjects.3 Higgins et al found no significant difference in the mean phonatory airflow across the different groups, but did find a very large intersubject variation in the mean phonatory airflow for patients with both AdSD and MTD. The hypothesis of the present study is that a closer examination of aerodynamic management of onset-offset during the /pa/ syllable train may yield important information in the diagnostic assessment of AdSD, and may provide important information in the differentiation of AdSD from other dysphonias. STATEMENT OF PURPOSE The present study sought to determine whether aerodynamic characteristics could be used to differentiate AdSD from other dysphonias. To determine differences across disorders, an institutional review board-approved study examined the aerodynamics (glottal pressure peak, airflow, and RB) across /pa/ syllable trains for the following groups of subjects: spasmodic dysphonia (SD), benign vocal fold lesions, primary muscle tension dysphonia (MTD-1), secondary muscle tension dysphonia with an identifiable primary benign vocal fold lesion (MTD-2), unilateral vocal fold paresis or paralysis (VFP), and normal nondysphonic patients (control). METHODS Sixty adult subjects were enrolled on an ongoing basis at a single institution’s voice clinic and categorized into one of six equal groups based on a diagnosis by a single fellowshiptrained laryngologist. Subjects were enrolled until 10 subjects were collected for each of the following groups: group 1 (AdSD, ARTICLE IN PRESS 2 Journal of Voice, Vol. ■■, No. ■■, 2017 TABLE 1. Patient Demographics Group No. of Subjects Sex (n) Mean Age (y) (Range) 1 10 Female (10) 73.1 (64–89) 2 10 Female (9) Male (1) 40.2 (21–59) 3 10 60.8 (43–75) 4 10 5 10 6 10 Female (9) Male (1) Female (9) Male (1) Female (9) Male (1) Female (9) Male (1) 42.4 (17–55) Specific Diagnosis (n) AdSD (9) AdSD with tremor (1) Nodules (4) Polyp (4) Cyst (2) MTD-1 (10) 68.4 (51–89) MTD-2 with nodules (6) MTD-2 with Polyp (4) Unilateral paralysis (10) 36.4 (31–55) Normal (10) Abbreviations: AdSD, adductory spasmodic dysphonia; MTD-1, primary muscle tension dysphonia; MTD-2, secondary muscle tension dysphonia with an identifiable primary benign vocal fold lesion. N = 10), group 2 (benign vocal fold lesion, N = 10), group 3 (MTD-1, N = 10), group 4 (MTD-2, N = 10), and group 5 (VFP, N = 10). Subjects were compared to group 6 (control subjects), who were considered to be normal and denied current dysphonia (N = 10). Subject demographics are available in Table 1. Aerodynamic measures obtained were part of the normal voice evaluation protocol during the voice clinic. Aerodynamic measures were analyzed for repeated /pa/ syllable trains among 50 dysphonic subjects and 10 control subjects. Control subjects denied any current dysphonia and denied knowledge of laryngeal pathology. Aerodynamic signals were captured and analyzed using Pentax Medical Phonatory Aerodynamic System, model 2 (Pentax Medical, Montvale, NJ). A tight-fitting pneumotachograph mask was held against the subject’s face, covering the oral and nasal cavities. Apressure tube was positioned into the oral cavity, avoiding contact with the tongue, and in accordance with manufacturer specifications. All subjects were assessed for aerodynamics during onsetoffset for the /papapapapa/ task in a modal voice. Subjects repeated /pa/ syllable trains at ~2 /pa/ segments per second for a total of five to seven syllables on a single breath flow using a steady, comfortable pitch and comfortable volume, consistent with their conversational speaking voice. A minimum of three /pa/ syllable train tokens were obtained for analysis. Recordings were made before any treatment. Although measures of pressure, airflow, intensity, and fundamental frequency were obtained, only data related to pressure and airflow were presented in this study. FIGURE 1. Graphic representation of the PP of the first syllable [p], the highest airflow during the RB for that [p] production, the airflow at the MA, and the FA measured immediately before pressure increase for the subsequent [p] in a control subject. FA, final airflow for the [a] segment; MA, middle of the [a] segment; PP, pressure peak; RB, release burst. ARTICLE IN PRESS Linda Carroll, et al 3 Pressure and Flow Across Vocal Pathologies Pressure peak and RB were determined across the /pa/ syllable task, between subjects and between groups. In addition, values for airflow following the completion of the RB (for [p]) and airflow shutoff at the end of the vowel segment leading into the next /pa/ syllable were examined. Data were hand-measured for the PP of the first syllable [p] production and the highest airflow during the RB for that [p] production, as well as the airflow at the middle of the [a] segment (MA) measured following the RB and at the midportion of the [a] segment, and the final airflow for the [a] segment (FA) measured immediately before the pressure increase for the next [p]. Figure 1 shows the airflow measures with the graphical locations of the PP, RB, MA, and FA in a control subject. Data were averaged for each subject across a minimum of three /pa/ syllable trains. The primary objective was to determine if aerodynamic measures could differentiate between any of the five pathologic groups and controls. Because each group contained 10 subjects with four trials for each measure, repeated measures analysis of variance (ANOVA) was carried out for each outcome variables (PP, RB, MA, and FA) to determine if there was a significant difference between the groups. Data were analyzed for normality and outliers. When the assumption of sphericity was violated, the Box correction factor was applied and a P value of <0.05 was considered significant. To test the differences in outcomes between specific groups, one-way ANOVA was carried out, and Bonferroni correction was applied to adjust for multiple testing. RESULTS Whereas the analysis of RB and MA showed no significant difference among different groups of subjects, the analysis of PP and FA revealed distinctions between SD, MTD-1, and MTD-2 subjects. The mean PP and FA values for each group are presented in Table 2, with ANOVA Bonferroni correct P value results. Repeated measures ANOVA demonstrated that there was a significant difference between groups for measured PP. Similarly, a difference was noted between groups for FA. Differences between groups were not observed for RB or MA. For PP and FA, post hoc analysis demonstrated significant differences between individual groups. PP was significantly different between the control group and those with SD or MTD-2. PP among AdSD subjects was significantly higher than MTD-1, benign lesion, VFP, and controls. The FA for the AdSD group was significantly higher than that for any other group. The FA for the MTD-1 group was significantly lower than those for all the other groups (Figure 2). Subjects with benign lesions had significantly higher FA than those with MTD-1. Subjects with MTD-1 had significantly lower FA than those with MTD-2, VFP, or controls. DISCUSSION Aerodynamic measures may be useful in the assessment of laryngeal function for patients with dysphonia and is known to aid in the differentiation between vocal pathology and normal voice. Prior investigation suggests a disorder-specific pattern of glottal airflow patterns seen in the aerodynamic control across voice disorders including benign mucosal lesions, unilateral vocal fold paralysis, MTD-1, and vocal fold atrophy.1 Hyperadductive voice disorders may be particularly amenable to a detailed evaluation of aerodynamics because of the impact on glottal airflow by muscular forces (functional or organic). MTD involves the hyperadduction of the vocal folds with characteristic changes TABLE 2. Mean ± Standard Deviation for Each Group and P Value From One-Way Analysis of Variance with Bonferroni Correction for (A) Peak Pressure and (B) Final Airflow (A) Peak pressure (cmH2O) Benign lesion MTD-1 MTD-2 Vocal fold paralysis Controls 10.099 ± 3.155 9.830 ± 4.009 11.835 ± 3.942 10.211 ± 3.477 7.623 ± 1.971 Spasmodic Dysphonia Benign Lesion MTD-1 MTD-2 Vocal Fold Paralysis 13.069 ± 6.315 P = 0.017 P = 0.004 P = 1.000 P = 0.005 P = 0.000 10.099 ± 3.155 9.830 ± 4.009 11.835 ± 3.942 10.211 ± 3.477 P = 1.000 P = 1.000 P = 1.000 P = 1.000 P = 0.908 P = 1.000 P = 1.000 P = 0.977 P = 0.024 P = 1.000 Spasmodic Dysphonia Benign Lesion MTD-1 MTD-2 Vocal Fold Paralysis 0.394 ± 0.223 P = 0.023 P = 0.000 P = 0.001 P = 0.000 P = 0.002 0.285 ± 0.102 0.150 ± 0.091 0.233 ± 0.136 0.215 ± 0.157 P = 0.001 P = 1.000 P = 1.000 P = 1.000 P = 0.030 P = 0.075 P = 0.020 P = 1.000 P = 1.000 P = 1.000 (B) Final airflow (L/s) Benign lesion MTD-1 MTD-2 Vocal fold paralysis Controls 0.285 ± 0.102 0.150 ± 0.091 0.233 ± 0.136 0.215 ± 0.157 0.232 ± 0.105 Note: Significant values are in bold. Abbreviation: MTD-1, primary muscle tension dysphonia; MTD-2, secondary muscle tension dysphonia with an identifiable primary benign vocal fold lesion. ARTICLE IN PRESS 4 Journal of Voice, Vol. ■■, No. ■■, 2017 FIGURE 2. Individual aerodynamic sample for subjects with (A) AdSD, (B) MTD-1, and (C) MTD-2. Peak pressure is significantly elevated in AdSD subjects. The final airflow is significantly higher for the AdSD subjects, whereas the final airflow is significantly lower for the MTD-1 subjects. AdSD, adductory spasmodic dysphonia; MTD-1, primary muscle tension dysphonia; MTD-2, secondary muscle tension dysphonia with an identifiable primary benign vocal fold lesion. in subglottal pressure, which, when coupled with maximum phonation time, may suggest the diagnosis of MTD.2 A detailed examination of aerodynamic parameters might aid in the differentiation between hyperadductive disorders including AdSD, MTD-1, and MTD-2. With a close examination of the aerodynamics during /pa/ syllable train, there was a difference for PP and FA aerodynamics compared to normal controls, and a significantly higher FA for AdSD subjects as they tried to control the offset of voicing. There were notable differences among groups with AdSD and with MTD-1 and MTD-2. The examination of PP revealed that subjects with AdSD and MTD-2 demonstrated a significantly higher PP than control subjects. Among AdSD patients, this reflects greater glottic closure and pressure buildup, consistent with the underlying disease. Those with MTD-2 may be using greater compensatory effort, consistent with typical laryngeal findings. The final airflow for AdSD and MTD-1 subjects was significantly different from controls. Representative aerodynamic tracings for four subjects with AdSD and four subjects with MTD-1 are shown in Figures 3 and 4, respectively (cursors are marked at PP and RB moments for each subject, and the output was rescaled to permit optimum comparison across subjects). Subjects with AdSD had an increased FA, whereas subjects with MTD-1 had a decreased FA. Subjects with MTD-2 performed similarly to controls for FA and were significantly different from subjects who had MTD-1. Subjects with MTD-2 had similar FA values compared to those with a benign lesion, and were similar to controls. FA in subjects with benign lesion may be the primary influence rather than compensatory muscle tension. Based on these findings, the present study suggests that a detailed aerodynamic evaluation can aid in discerning the differential diagnosis of AdSD vs. MTD. Close examination of PP and FA may be a diagnostic measure to differentiate AdSD, MTD-1, and MTD-2. Table 3 reveals a summary of the aerodynamic TABLE 3. Summary Table of Differentiating Aerodynamic Characteristics for AdSD, MTD-1, and MTD-2 When Compared to Controls Vocal Pathology AdSD MTD-1 MTD-2 Peak Pressure (cmH2O) Final Airflow (L/min) Increased Increased Decreased Increased Abbreviations: AdSD, adductory spasmodic dysphonia; MTD, muscle tension dysphonia; MTD-1, primary muscle tension dysphonia. ARTICLE IN PRESS Linda Carroll, et al Pressure and Flow Across Vocal Pathologies AdSD Subject 6 AdSD Subject 7 Frequency Frequency Intensity Intensity Airflow Airflow Pressure Pressure AdSD Subject 8 AdSD Subject 10 Frequency Frequency Intensity Intensity Airflow Airflow Pressure Pressure FIGURE 3. Sample aerodynamic patterns for four subjects with AdSD. AdSD, adductory spasmodic dysphonia. MTD-1 Subject 1 MTD-1 Subject 2 Frequency Frequency Intensity Intensity Airflow Airflow Pressure Pressure MTD-1 Subject 8 MTD-1 Subject 9 Frequency Frequency Intensity Intensity Airflow Airflow Pressure Pressure FIGURE 4. Sample aerodynamic patterns for four subjects with MTD-1. MTD-1, primary muscle tension dysphonia. 5 ARTICLE IN PRESS 6 characteristics of AdSD, MTD-1, and MTD-2. AdSD is characterized by a higher PP and a higher final airflow. MTD-1 is characterized by lower values of final airflow. MTD-2 is characterized by higher values of PP. Overall, RB values and MA values were not valuable for differentiating vocal pathologies. In the present study, AdSD subjects’ control of airflow during the interphonemic environment contrasted with all other groups. Rather than reducing airflow at the end of the [a] in preparation for a smooth transition to pressure buildup for [p], AdSD subjects kick glottal airflow up. Additionally, as a group, they may demonstrate difficulties with balance of flow and pressure during voicing changes. CONCLUSIONS Aerodynamic coordination is a key component to ease of voice and speech. The present study found a significant difference between diagnostic groups for PP and final airflow during /pa/ Journal of Voice, Vol. ■■, No. ■■, 2017 syllable trains. In particular, there was a significant difference between AdSD, MTD-1, and MTD-2 groups. AdSD was characterized by a higher PP and a higher final airflow. MTD-1 was characterized by a lower final airflow, whereas MTD-2 was characterized by a higher PP. Examination of aerodynamics may aid in the diagnosis for patients in whom distinction among hyperadductive disorders is challenging. REFERENCES 1. Dastolfo C, Gartner-Schmidt J, Yu L, et al. Aerodynamic outcomes of four common voice disorders: moving toward disorder-specific assessment. J Voice. 2016;30:301–307. 2. Zheng YQ, Zhang BR, Su WY, et al. Laryngeal aerodynamic analysis in assisting with the diagnosis of muscle tension dysphonia. J Voice. 2012; 26:177–181. 3. Higgins MB, Chait DH, Schulte L. Phonatory air flow characteristics of adductor spasmodic dysphonia and muscle tension dysphonia. J Speech Lang Hear Res. 1999;42:101–111.