Abstract
Importance
Glottal competence and cough effectiveness are associated with aspiration. In unilateral vocal fold paralysis (UVFP), injection laryngoplasty (IL) is the mainstay treatment for those with dysphonia and dysphagia. However, few data exist explaining how IL exerts its effects on aspiration prevention.
Objective
To evaluate the effect of IL on voluntary cough strength, glottal closure, and swallowing function. Furthermore, the effects were compared between patients with active treatment/disease and those without.
Study Design
Prospective cohort study.
Setting
Laryngology clinic of single tertiary hospital.
Participant
Adult patients underwent IL for UVFP between January 2021 and April 2023.
Main Outcome Measures
Maximum volitional cough pressure (MCoughP) was quantified before and after IL, as well as clinical voice outcomes and normalized glottal gap area. For those with a complaint of aspiration, the Eating Assessment Tool (EAT-10) questionnaire and Penetration-Aspiration Scale (PAS) were also evaluated.
Results
Forty-one patients were included (26M:15F; age range 32 to 80 years old, mean age 58.7). Clinical voice outcomes and glottal closure were significantly improved in all patients, as well as EAT-10 score (16.22 ± 11.83 at pre-IL and 10.83 ± 11.17 at post-IL, P = .008) and PAS [2 (IQR 1.25, 2) at pre-IL and 1 (IQR 1, 2) at post-IL with 20 mL bolus, P = .02; 2 (IQR 2, 4) at pre-IL and 2 (IQR 1, 2) at post-IL with cup-sipping, P = .007]. Twelve patients had ongoing treatment/systemic diseases, who had significant improvement in voice and glottal closure but not in PAS. MCoughP was significantly increased in ongoing treatment/systemic diseases (−) group [42.90 ± 20.17 cmH2O at pre-IL, 51.33 ± 21.15 at post-IL, 95% CI (−16.63, −0.23), P = .04] but decreased in ongoing treatment/systemic diseases (+) group [67.98 ± 40.06 cmH2O at pre-IL, 55.32 ± 31.68 at post-IL, 95% CI (1.37, 23.97), P = .03].
Conclusions and Relevance
IL significantly improved glottal competence and voice outcomes. However, increased volitional cough pressure and improved swallowing safety were only demonstrated in patients without ongoing treatment or systemic diseases. Patient’s general condition may contribute to the observed effect of IL on dysphagia. Rehabilitation to optimize cough strength and prevent aspiration is beneficial for patients with deteriorating general conditions.
Key Message
Injection laryngoplasty significantly improved glottal closure and voice outcomes in all UVFP patients. However, improvements in cough pressure and swallowing safety were only observed in patients without systemic disease or ongoing treatment.
This highlights that systemic health conditions may attenuate injection laryngoplasty’s benefits on airway protection, emphasizing the need for swallowing safety improvement strategies and rehabilitation in patients with deteriorating general conditions.
Introduction
In unilateral vocal fold paralysis (UVFP), a paramedian or laterally displaced vocal fold (VF) leads to glottal insufficiency, 1 and may subsequently cause dysphonia, weak cough strength, and aspiration. The prevalence of dysphagia has been reported as high as 20% to 72%.2,3 Injection laryngoplasty (IL) medializes the paralyzed VF by refilling materials and corrects the glottal gap. 4 Many studies have described the utility of IL in improving voice quality in patients with UVFP, but only a few studies assessing its efficacy in improving swallow outcomes emerged in recent decade, as in a majority of patients showing improvement in diet intake, subjective symptoms of dysphagia, or Penetration-Aspiration Scale (PAS) scores after IL.5 -8
Cough is a reflexive mechanism to protect the airway from aspiration. Glottal closure helps create the high intrapleural/intrathoracic pressures due to isometric expiratory muscle contraction in the compression phase occurring in cough. With VF paralysis, the compression phase is affected, resulting in reduced cough strength, which can be indicative of an increased risk of respiratory complications. 9 Indeed, a direct correlation between increased cough strength and decreased pneumonia risk has been described in literature. 10 Remarkably, little research has focused on enhancing glottic closure in improving cough strength for aspiration prevention. Early works demonstrated variable gains in peak cough flow or inspiratory pressure, yet were limited by small sample sizes, heterogeneous populations, and lack of swallowing function assessments.8,11 -13 Collectively, these findings suggest potential benefits of IL for cough enhancement, but robust, comprehensive studies are needed to confirm its efficacy and clinical relevance for swallowing safety.
Despite the known positive effect of IL on aspiration as well as the link between cough strength and aspiration, how VF augmentation exerts its effects on the prevention of aspiration, possibly via enhanced glottic closure or increased cough strength is unknown. Based on our previous study, 14 the feasibility of a predetermined flexible endoscopic evaluation of swallowing (FEES) protocol for evaluation of the safety of swallowing in patients with UVFP and positive effects of IL had been demonstrated. In the present study, in addition to glottal closure, cough pressure measurements were also incorporated, as a surrogate of the ability to clear aspiration. The objectives of our study were (1) to prospectively evaluate the effect of IL on glottal competence and voluntary cough strength and (2) to determine whether improved glottic closure and/or cough strength after IL plays a role in improving swallowing function and swallowing safety.
Materials and Methods
Study Design and Study Population
After approved by Institutional Review Board of National Taiwan University Hospital (NTUH), prospective recruitment began for patients with a diagnosis of glottic insufficiency secondary to UVFP and scheduled for IL, at otolaryngology department of NTUH, a tertiary medical center. All participants provided written informed consent. UVFP with associated glottic insufficiency was confirmed by videostrobolaryngoscopy by a laryngologist (W.H.T.). Those with pre-existing oropharyngeal dysphagia, previous laryngeal surgery, cerebral vascular accident, and history of pulmonary disease (chronic obstructive pulmonary disease, asthma, emphysema, and other restrictive or reactive lung diseases) were excluded. Demographic characteristics, including age, sex, etiology of the paralysis, and laterality, ongoing systemic disease at the period of IL and history of cardiovascular and renal diseases with impaired function, were documented. Voice and swallowing evaluations, as well as cough strength, were performed after inclusion before IL and 2 weeks after IL, which was performed in an office setting. Huvederm (Allergan) was injected through the cricothyroid membrane lateral to the involved vocal process and vocal ligament at the midportion of the true vocal fold with a disposable 25-G needle under the guidance of transnasal flexible video laryngoscopy. The injection was considered adequate when a minimum glottal gap on phonation was observed.
Voice Outcome Assessment
Self-evaluation with the Chinese version of the Voice Handicap Index-10 (VHI-10), 15 perceptual grading with the GRBAS scale, 16 S/Z ratio, and maximum phonation time (MPT) were recorded. The MPT task was obtained by having the patient sit straight and sustaining the vowel sound /a/ at normal speaking volume and loudness for as long as possible after maximum inhalation. A mean of 3 consistent trials was computed. The normalized glottal gap area (NGGA) was analyzed using the formula (minimum glottal − gap area/membranous vocal fold length2) × 100 with ImageJ software (ImageJ; National Institutes of Health), on videostrobolaryngoscopy-captured images at modal pitch. 17
Swallowing Function assessment
All patients completed the Eating Assessment Tool (EAT-10) as a standardized self-evaluation of swallowing function.18,19 Patients who reported symptoms suggestive of dysphagia—either through a positive EAT-10 score or as a chief complaint—underwent further instrumental assessment. FEES was performed by a laryngologist (W.H.T.) and an speech-language pathologist (H.L.C.) who had more than 5 years of experience in swallowing therapy using a distal chip flexible fiberoptic rhinolaryngoscope (ENF-VT3; Olympus,). Standard FEES protocol was followed with slight modifications. 20 Different bolus volumes were administered in the following order: 10, 20, and 90 mL thin liquid. The test liquids were dyed with 5% methylene blue. The patients were instructed to swallow 10 and 20 mL in 1 swallow, while the 90 mL volume was administered in a cup for serial free sipping. All volumes were administered twice, except the 90 mL volume, which was administered once. The presence and severity of aspiration were evaluated using PAS, 21 which was calculated as follows: a score of 1 = normal, 2 to 5 = penetration, and 6 to 8 = aspiration, as assessed from the most impaired swallowing across the study. Results for each test bolus were recorded as PAS 10, PAS 20, and PAS Cup.
Cough Strength Evaluation Protocol
Participants were carefully instructed on the standardized experimental protocol to ensure reproducible cough-strength measurements. Participants were seated upright in a comfortable position without restraining clothing. A commercialized respiratory muscle trainer (Bravo™ Trainer; GaleMed Corporation) connected with a digital manometer (GiO; GaleMed Corporation) was employed for measuring maximum inspiratory pressure (MIP), maximum expiratory pressure (MEP), and maximum cough pressure (MCoughP).22 -24 The respiratory muscle trainer is an exercise tool consisting of a detachable mouthpiece and Plexiglas tube with an internal spring-loaded valve calibrated in cmH2O, which had shown in literatures its wide accessibility and growing use is respiratory muscle training for patients with voice disorder, 25 and patients at risk for penetration/aspiration.23,24,26,27 Measurements were recorded based on a study by Park et al 26 Briefly, participants were asked to hold a disposable mouthpiece in their mouths with a nose clip blocking airflow through the nasal cavity. They were then asked to breathe in to total lung capacity and blow hard into the tubing for measurement of MIP and MEP, respectively. For voluntary cough measure (MCoughP), the instruction given to each participant was as the following (1) relaxing and breathing into the tube; (2) following 3 tidal volume breaths, breathing deeply, and then coughing hard, like something went into the wrong way; and (3) completing 3 trials of the voluntary cough.
Three measures of were made until 3 values were obtained within ±5% of one another. The average of the 3 values was used as the average measure of each parameter.
Statistical Analysis
Patient characteristics, NGGA, VHI-10, EAT-10 scores, and pressure data were compared before and after-IL using a paired t-test. Change in PAS scores was compared using non-parametric methods of analysis (Wilcoxon-signed rank test). Wilcoxon rank-sum test was used to evaluate differences among groups based on whether there was ongoing treatment/systemic disease. We performed logistic regression to investigate the factors influencing the odds of improved cough pressure. Statistical significance was set at P < .05. All statistical analyses were performed using SPSS 26 software (SPSS).
Results
Participants
A total of 50 patients were enrolled in the study between January 2021 and April 2023. Nine were excluded for incomplete data due to lack of follow-up related to the COVID-19 pandemic or unspecified reasons. Finally, 41 patients were included in the final analysis (26 male15 female, mean age 58.7 years). Twelve patients had ongoing systemic treatment at the period of IL, including 10 with immunotherapy/radiation for esophageal cancer and chemotherapy for lung cancer; 1 with end-stage renal diseases under hemodialysis; 1 with ventricular hypertrophy with decreased cardiac function. Eighteen patients reported choking/aspiration and completed swallowing function evaluation. Demographic characteristics are summarized in Table 1.
Demographics of the Participants.
Voice, Swallowing Outcomes, and Cough Pressures Changes Before and After IL
Significant differences in voice and swallowing function were demonstrated in subjective, perceptual grading and instrumental evaluations (Table 2). Following IL, patients had a significant decreased VHI-10 [26.83 ± 6.63 at pre-IL, 13.70 ± 6.91 at post-IL, 95% CI (10.18, 16.07), P < .001] score. Glottal closure was improved as shown on NGGA [10.30 ± 9.28 at pre-IL, 1.01 ± 2.30 at post-IL, 95% CI (6.19, 12.39), P < .001]. Of the 18 patients with a complaint of choking, 15 reported improvements after IL. EAT-10 score was 16.22 ± 11.83 at pre-IL and was 10.83 ± 11.17 at post-IL [95% CI (1.61, 9.17), P = .008]. There was 11.8%, 17.6%, and 58.8% of the patients having a PAS score >2 with 10 mL bolus, 20 mL bolus, and cup sipping, respectively, at pre-IL. The percentage largely decreased to 5.9%, 0%, and 11.8% at post-IL. After IL, all of the patients who had penetration/aspiration (PAS >2) with 10, 20 mL bolus before IL achieved a safe swallowing (PAS 1 and PAS 2). Four out of 5 patients who had penetration/aspiration (PAS >2) with cup sipping before IL achieved a safe swallowing postoperatively. The remaining one presented with a PAS score of 5 before IL and with PAS 3 after IL. Cough pressure measurement did not differ significantly before and after IL.
Comparison of Voice, Swallowing Outcomes and Cough Pressures of All Patients Before and After IL.
Abbreviations: EAT-10, Eating Assessment Tool-10; IL, injection laryngoplasty; IQR: interquartile range; MCoughP, maximum cough pressure; MEP, maximum expiration pressure; MIP, maximum inspiration pressure; MPT, maximum phonation time; NGGA, normalized glottal gap area; PAS, Penetration-Aspiration Scale; SD, standard deviation; VHI-10, Voice Handicap Index-10. GRBAS, grade roughness breathiness asthenia and strain in GRBAS scale.
Statistically significant improvement at P ≤ .05.
Comparison Between Groups With and Without Ongoing Treatment or Systemic Diseases
A chi-square test revealed a significant association between improvement in MCoughP and ongoing treatment/systemic diseases, χ2(1) = 13.81, P < .001. We further compared between patients with and without ongoing treatment or systemic diseases. Age and baseline swallowing functions, cough parameters, and glottal closure were similar. It showed that change in MCoughP significantly differed between groups [mean change 8.43 cmH2O, SD 20.30 in treatment/disease−; mean change −12.67 cmH2O, SD 17.78 in treatment/disease+ group, 95% CI (7.25, 34.94), P = .004]. In subgroup analysis, for treatment/disease− group, clinical voice and swallowing outcomes including VHI-10, EAT-10, and PAS 20, PAS Cup were significantly improved after IL; while for treatment/disease+ group, only clinical voice outcomes were significantly improved (Figure 1). MCoughP was significantly increased in those without ongoing treatment or systemic diseases [42.90 ± 20.17 cmH2O at pre-IL, 51.33 ± 21.15 at post-IL, 95% CI (−16.63, −0.23), P = .04] but significantly decreased in those with ongoing treatment or systemic diseases [67.98 ± 40.06 cmH2O at pre-IL, 55.32 ± 31.68 at post-IL, 95% CI (1.37, 23.97), P = .03]. EAT-10 score was significantly improved in patients without ongoing treatment or systemic diseases [15.13 ± 10.64 at pre-IL, 9.60 ± 10.13 at post-IL, 95% CI (0.92, 10.15), P < .05], as well as PAS Cup [median 2, IQR (2, 5.25) pre-operatively, median 2, IQR (1, 2) post-operatively, P = .02]. NGGA was significantly improved in both groups [9.82 ± 9.37 at pre-IL, 0.97 ± 2.59 at post-IL, 95% CI (5.24, 12.47), P < .001 in treatment/disease− group; 11.58 ± 9.38 at pre-IL, 1.11 ± 1.31 at post-IL, 95% CI (3.24, 17.7), P = .01 in treatment/disease+ group].

Comparison of EAT-10, PAS con, MCoughP, and glottal closure before and after IL in patients with ongoing disease or treatment group and in patients without ongoing disease or treatment group. In patients without ongoing treatment and diseases, EAT-10, PAS con, MCoughP, and glottal gap were significantly improved after IL. Contrarily, in patients with ongoing treatment and diseases, MCoughP was significantly decreased, while glottal gap was significantly improved. EAT-10 and PAS con were stationary.
Factors Associated with the Improvement in Cough Pressure
A logistic regression was performed to ascertain the effects of age, ongoing treatment/systemic diseases and post-IL NGGA on the likelihood of an improved cough pressure. The logistic regression model was statistically significant, χ2(3, N = 34) = 14.97, P = .002. Participants without ongoing treatment/systemic diseases were 35 time as likely to have an improved MCoughP [OR = 35.04, 95% CI (3.20, 384.16)].
Discussion
In the present study, we investigated the effect of IL on clinical voice, swallowing outcomes, and volitional cough pressure. VHI-10 score, MPT, S/Z ratio, and GRBAS scale were significantly improved in all participants, as well as EAT-10 score and PAS with larger boluses. Improved glottal closure was also successfully achieved. When compared to patients with ongoing treatment or systemic disease, those without had a significantly greater increase in cough pressure and more prominent improvement in swallowing function.
The beneficial effects of IL on voice and swallowing in UVFP patients are consistent with existing literature.2,5 -7,14,28 -30 However, the etiology of the VF dysfunction varied and clinical courses in the mixed population may be dissimilar. For instance, UVFP following esophagectomy in esophageal cancer patients may yield differing responses depending on whether concurrent chemoradiotherapy was administered—even when glottal closure is achieved. In this study, patients without ongoing disease or systemic treatment showed significant improvements in both voice and swallowing function, including swallowing safety, whereas in those with active disease or treatment, significant improvements were observed only in voice outcomes (Figure 1). These findings suggest that the effect of IL was more pronounced on voice than swallowing. Given that swallowing safety involves a complex interplay of structural integrity, neuromuscular coordination, 31 motor and sensory coordination, and muscle synergy, 32 a single therapeutic intervention such as IL may be insufficient to fully restore function in medically compromised individuals.
It is generally accepted that glottic competence is important for effective cough (in airflow and pressure) and airway protection mechanism.8,11,12 IL aims to medialize the paralyzed VF, which can help close the glottic gap. In UVFP, the procedure improves airway function by enhancing VF closure during adduction and typically does not make the glottic airway narrower.33,34 Rameau et al demonstrated that glottic competence improved cough efficiency in terms of an increased cough volume acceleration and a decreased peak expiratory flow rise time, but not peak expiratory pressure. 13 Our findings align partially with these results: patients without ongoing treatment or systemic disease exhibited significant improvements in MCoughP, whereas those with ongoing disease showed a decrease. This decline may be attributable to generalized weakness, reducing their ability to engage accessory respiratory muscles during cough. Similarly, literature had showed that impaired airway protection—such as reduced cough strength—has been observed in several conditions and disease states including aging, respiratory sarcopenia (related to physical inactivity, malnutrition, endocrine changes, etc.), and neuromuscular diseases.35,36 Nevertheless, improved NGGA was observed in all participants, suggesting that enhanced glottic closure may still contribute positively to swallowing safety. Supporting this, a study comparing post-laryngectomy patients to age-matched controls highlighted the importance of a functional larynx in ensuring cough effectiveness. 37 Thus, the corrected glottic competence may contribute to the overall improved swallowing function and safety despite MCoughP was not significantly increased.
Our findings underscore a gap in the current literature regarding the interplay between laryngeal competence, voluntary cough generation, and swallowing function. Much of the published work focused on reflexive cough and risk of silent aspiration.38,39 Others investigated the relationship of VF atrophy to swallowing safety and cough function in patients with neurological disorders or Parkinson’s disease.40,41 While several publications have examined the impact of IL on cough airflow parameters,8,11 -13 limited data exist on its effectiveness on swallowing. A recent systematic review on IL’s impact on aspiration outcomes in acute UVFP highlighted this issue: of the 14 case series reviewed (primarily retrospective), none could be quantitatively synthesized due to outcome heterogeneity, lack of stratification, small sample sizes, and inadequate baseline equivalence. 7 Our prospective study addressed these gaps by comprehensively evaluating dysphagia symptoms, swallowing function, glottal competence, and cough strength. Additionally, subgroup analysis based on the presence of systemic disease or ongoing treatment delineates the empirical observations in clinical practice. Dysphagia is a complex and multifactorial problem. 42 In addition to anatomical structures, physiological factors are also vital for normal swallowing function. 43 Systemic diseases and chemotherapy-related deconditioning, sarcopenia, cachexia, and mucosal side effects can change the physiology of swallowing and lead to dysphagia.43 -46 A cross-sectional study from a specialist cancer center reported that across all patients with various cancer types, 54% reported dysphagia and 20% reported dysphagia for liquids. 47 These rates are concerning when we consider the vulnerability of cancer patients and the negative effects of dysphagia on treatment tolerance, quality of life, and survival.47,48 The present study demonstrated that patient’s general condition may contribute to the observed effect of IL on swallowing in spite of complete glottal closure. Our findings reinforce the need to identify dysphagia and provide rehabilitation on improving swallowing safety for this patient populations, aligning with existing literature. 49
There are some limitations to the study. First, volitional cough, not reflexive cough, was tested. Although volitional cough serves as a final safeguard to airway, involuntary cough may be more impactful in aspiration prevention. Future studies should include both to better assess IL’s impact on cough airflow dynamics and swallowing safety. Moreover, although it was demonstrated that significant improved MCoughP was associated with improved swallowing function in terms of EAT-10 and PAS, which also agreed with literatures,8,50 future studies with larger sample size and case–control design are warranted to further delineate its clinical significance. Second, while Rameau et al showed that IL improves cough flow dynamics but not peak expiratory pressure, 13 our digital manometer cannot measure time-related variables. Despite this limitation, the device remains useful for monitoring pressure changes and could be complemented in future studies by respiratory muscle training to assess potential additive effects. 51 Third, FEES and cough strength evaluation were not blinded, although raters were unaware of patient-reported outcomes. Future studies incorporating single-blind design would minimize observer bias. Fourth, assessments were performed at a 2-week follow-up—after expected post-IL swelling subsided. Although improvements have been reported as early as 2 hours and up to 6 months post-IL,52 -54 future studies with serial follow-up timepoints or a long-term follow-up protocol are needed to delineate the long-term effect. Lastly, although we included patients with various etiologies (cancer patients, idiopathic, etc.), those with pre-existing dysphagia (eg, cerebral vascular accident, previous laryngeal surgery, etc.) were excluded during screening to ensure that the core etiology is recurrent laryngeal nerve injury. The small sample size limited further subgroup analysis by etiology. Further studies with age-matched or palsy duration-matched controls may provide more robust insights, although age was not a significant factor in our regression model.
Conclusion
In general, patients with UVFP had significant improvements in glottal competence, voice outcomes as well as dysphagia symptom severity and swallowing safety after IL. Volitional cough pressure was significantly increased in patients without ongoing treatment or systemic diseases, while it was significantly decreased in those with ongoing treatment or systemic diseases. Patient’s general condition may contribute to the observed effect of IL, which seemed to be more profound in voice than in swallowing functions in patients with ongoing treatment or systemic diseases. Additional assessment of factors that affect cough airflow dynamics to optimize cough strength is needed, and rehabilitation and prevention of aspiration is suggested, especially for patients with deteriorating or fluctuating general conditions.
Footnotes
Acknowledgements
The authors thank the Ministry of Science and Technology, National Taiwan University Hospital, and National Taiwan University for their supports. The authors would also like to thank the physicians and nursing staff at their institution for the support.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Authorship
Tsung-Lin Yang: Guarantor of the article, who was also in charge of study design, data interpretation, and manuscript revision.
Wen-Hsuan Tseng: Drafted the manuscript, performed the study, acquired data, interpreted, and analyzed the data.
Hsiang-Ling Chiu: Acquisition of data and analysis of data.
Tzu-Yu Hsiao: Study design, data interpretation.
Tsung-Lin Yang: Study design, data interpretation, manuscript revision.
All authors have approved the final version of the manuscript.
