Journal of Voice最新文献

Purpose: Long-term retention of treatment effects following discharge from voice therapy remains unclear. This study examined relapse rates 4-5 years after completion of voice therapy.

Method: A prospective observational study was conducted with 43 patients who completed voice therapy between 2019 (in person) and 2020 (telehealth). Primary outcomes included the voice handicap index-10 (VHI-10), the patient perception of voice therapy questionnaire, and additional questions regarding current vocal function.

Results: Current VHI-10 scores (M = 15.3) were significantly lower than baseline scores (M = 24.2; P < 0.001). The mean amount of voice therapy completed per individual was 2.8 sessions. Most participants (77%) reported that voice therapy was helpful, with 63% describing improved voice after discharge, 30% reporting stability, and 7% reporting relapse. Nearly half (49%) indicated that transferring skills to connected speech was the most challenging aspect of therapy, while 70% identified specific exercises as the most useful. At follow-up, 28% continued to use therapy exercises, and 77% reported being able to meet their daily vocal demands.

Conclusions: Four to five years after voice therapy, 23% of patients reported current vocal limitations, and 7% reported relapse. These rates are lower than those previously reported, suggesting favorable long-term outcomes following voice therapy.

Objectives: As the global population ages, age-related vocal fold atrophy (ARVA) is increasingly recognized. However, no nationwide epidemiological studies have been conducted. We aimed to investigate the clinical characteristics and treatments of ARVA in Japan, the world's most aged society.

Methods: This multicenter retrospective study, part of the Tokyo Voice Center Initiative, analyzed medical records of 1365 patients diagnosed with ARVA between 2018 and 2022. The primary survey examined the involvement of speech language pathologist (SLP) and the routine use of stroboscopic examinations at 34 certified institutions. The secondary survey assessed hoarseness severity (via maximum phonation time and voice handicap index-10), age of onset, treatment methods, and treatment outcomes at 14 certified institutions.

Results: All diagnoses and assessments were conducted by departments of otolaryngology-head and neck surgery. Of the 1365 patients (637 females, 728 males; mean age: 64.6 ± 14.7 years), 1097 received treatment and 268 did not. Treatments included voice therapy (n = 1031), injection augmentation (n = 162), and surgery (n = 18), with some receiving multiple therapies. Voice therapy was more effective when provided by voice-specialized SLPs. Invasive treatments yielded greater voice improvement overall. Significant differences in age and sex ratios were found between institutions (P < 0.01).

Conclusion: These findings underscore the need for cross-specialty education, broader awareness of ARVA, and establishment of a national database and severity classification system.

Objective: To explore differences in vocal fundamental frequency (F0) between premenopausal and postmenopausal women through associations with age and hormone use, including estrogen-progesterone therapy with testosterone replacement (EPT-HRT) and birth control (BC).

Methods: This cross-sectional, multi-institutional study recruited four groups: postmenopausal women taking EPT-HRT, postmenopausal women not taking HRT (HRT-naïve), premenopausal women taking BC, and premenopausal women not taking BC (hormone-naïve). The primary outcome was F0 in Hertz (Hz), compared between groups as a difference (Δ) with 95% confidence interval (CI) and correlated (r) with demographic variables.

Results: Ninety-six participants were recruited, with 24 per group: postmenopausal women on EPT-HRT (mean age:51; range:32-65), HRT-naïve postmenopausal women (60 [49-74]), premenopausal women on BC (26 [23-30]), and hormone-naïve premenopausal women (29 [22-45]). There was no significant difference in F0 between postmenopausal women with (174 ± 14 Hz) and without EPT-HRT (173 ± 20Hz) (P = 0.91) or premenopausal women with (198 ± 24 Hz) and without BC (191 ± 20 Hz) (P = 0.19). Mean F0 of postmenopausal women on EPT-HRT was lower than in both premenopausal women on BC (Δ24.3 [CI: 13.6-35.7]; P < 0.0001) and not on BC (Δ16.7[CI: 5.3-28.1]; P = 0.005). A moderate negative correlation was observed between age and F0 (r = -0.39; P < 0.001).

Conclusions: This study found that F0 decreases with age in women, with no significant effect from testosterone-containing HRT.

Objective: To adapt the original singing voice handicap index-10 (SVHI-10) into Hindi and validate it among singers without voice problems and those with dysphonia.

Study design: Cross-sectional analytic study design.

Method: The Hindi translation of the original SVHI-10 was conducted following standardized guidelines. The study included 75 singers, comprising two groups: singers without voice problems (NS) (n = 50) and dysphonic singers (DS) (n = 25). Psychometric evaluation of the Hindi SVHI-10 was performed, including assessments of internal consistency, test-retest reliability, criterion validity, and concurrent validity. Sensitivity and specificity of the SVHI-10/Hindi version were determined using receiver operating characteristic (ROC) curve analysis. Additionally, the statistical comparisons were made between both the groups.

Result: The Hindi SVHI-10 has excellent internal consistency and test-retest reliability. The Pearson product-moment correlation test indicated that the Hindi SVHI-10 is a reliable tool (r = 0.99, n = 75, P < 0.001). The ROC curve revealed 96.7% sensitivity and 85.1% specificity with a cutoff score of 10. Dysphonic singers scored significantly higher than singers without voice problems on the Hindi version of the singing voice handicap index-10.

Conclusion: The Hindi SVHI-10 is a valid, reliable, and sensitive self-report tool that effectively identifies self-perceived singing voice handicaps among Hindi-speaking singers.

Objective: This study aimed to translate, culturally adapt, and validate the Korean version of the screening index for voice disorder (SIVD-K).

Methods: Case group patients with voice problems (n = 90) and control group patients without voice problems (n = 25) participated in the study. They completed three questionnaires (the Korean version of the voice handicap index [VHI-K], the Korean version of Voice-Related Quality of Life [VRQOL-K], and the SIVD-K). Case group patients re-completed the SIVD-K to assess test-retest reliability. Finally, receiver operating characteristic analysis was performed to determine the cut-off value of SIVD-K for differentiating the case and control groups.

Results: The SIVD-K scores for the case group were significantly higher than those for the control group. The scores of SIVD-K showed significant correlations with the VHI-K and VRQOL-K. Test-retest reliability was high (Spearman's rho = 0.854), and internal consistency was acceptable. The cut-off value for differentiating case and control groups was 2.5, with scores of 3 or higher suggesting a risk for dysphonia.

Conclusion: The SIVD-K demonstrated good reliability and validity as a brief screening tool for dysphonia in Korea. While it shows promise for early detection in clinical practice, these findings should be considered preliminary because the study was conducted at a single site and the participant groups were uneven in size. Larger multicenter studies are needed to confirm its clinical applicability.

Laryngeal guarding is a term used informally by clinicians and teachers, which has yet to be clearly defined until now. This paper examines its potential role as an unconscious block within the body, hindering progress in voice rehabilitation and limiting the development of efficient singing and speaking behaviors. Through a narrative review, it traces a continuum of guarding from reflexive responses to learned behaviors, drawing on findings from pain science, fear-avoidance processes, dispositional and psychoanalytic formulations, and research on functional voice disorders and secondary gain. The review reframes persistent functional dysphonia as potentially underpinned by unconscious protective strategies, showing how anticipatory anxiety, trauma, and hypervigilance may contribute to maladaptive laryngeal co-contraction. In cases of hyperfunctional patterns or "guarding," the alteration is not only a loss of parasympathetic modulation but, in some instances, a reversal of its expected function. The paper proposes a theoretical account of why the system may shift from a restorative, protective parasympathetic state toward a heightened sympathetic or anticipatory mode. In this upregulated state, the larynx adopts a physiologic posture of defense rather than rest: an adaptive contraction that becomes maladaptive when sustained over time. This paper also offers practical insights into how therapists and pedagogues can identify the role of laryngeal guarding in voice disorders and support clients and students to move beyond this limitation. It outlines key features clinicians and teachers might observe in order to recognize and conceptualize laryngeal guarding within clinical or studio practice.

Parkinson's disease (PD) presents significant diagnostic challenges, particularly in its early stages when timely intervention is most effective. Voice analysis offers a noninvasive approach for early detection but faces challenges of class imbalance and high-dimensional data that limit machine learning (ML) classification performance. This study investigates the combined effect of the synthetic minority over-sampling technique (SMOTE) for data balancing and feature selection (FS) methods, analysis of variance, Chi-squared (χ²), and mutual information to optimize PD detection from voice recordings. These preprocessing techniques were integrated with multiple classifiers, including extreme gradient boosting (XGBoost), random forest, logistic regression, and support vector machine, using the Parkinson's Disease Classification Dataset containing 754 features extracted from voice recordings. Results demonstrate that combining SMOTE with FS significantly improves model performance compared to using either technique independently. The highest performance was achieved by XGBoost with Chi-squared (χ²) FS (96.4% accuracy, 96.9% F1-score) using approximately 600 optimal features. These findings confirm that appropriate preprocessing techniques improve the effectiveness of voice-based PD detection and support the development of accurate, noninvasive diagnostic tools powered by ML.

Objective: To verify whether speech rate and the presence of dysphonia impact the cognitive performance of university students.

Methods: The study included health sciences undergraduates aged 18 to 40 years, of any gender. Participants were randomly assigned to tests of attention, working memory, and language, recorded by women with rough dysphonic or healthy voices, and presented at different audio playback speeds (1.0×, 1.25×, 1.5×, and 2.0×). All tests were part of the Brief Neuropsychological Assessment Battery (Neupsilin). Results were analyzed using the Kruskal-Wallis test to compare cognitive performance across playback speeds, and the Mann-Whitney test to compare performance according to vocal quality (healthy or dysphonic voice).

Results: The rough dysphonic voice negatively affected the cognitive performance of university students in tasks involving attention, working memory with semantic cues (word span in sentences), and repetition of words and pseudowords. Dysphonic voice quality was favored in working memory tasks without semantic cues (ascending digit span). Playback speed interfered only with repetition tasks, with poorer performance observed above 1.75×.

Conclusion: Vocal quality negatively impacted students' cognitive performance, particularly in attention and working memory, regardless of audio playback speed.