Prognosis of Dysphagia in Pediatric Patients With Vocal Fold Immobility

Abstract

Objective

This study aims to determine the incidence and prognosis of dysphagia in pediatric patients with true vocal fold (TVF) immobility or hypomobility.

Study Design

A single‐center retrospective chart review.

Setting

A single‐institution tertiary‐care center.

Methods

A total of 89 pediatric patients diagnosed with vocal fold hypo/immobility with a modified barium swallow (MBS) performed were examined. Patient demographic information and etiology of vocal fold immobility as well as laterality were reviewed. Changes in MBS findings over time were assessed.

Results

A total of 89 pediatric patients were identified with a mean follow‐up of 35.4 months. The most common etiology of TVF hypo/immobility was cardiothoracic surgery (58.4%).Immobility was observed in 80.6% of patients. Patients with unilateral disease were more likely to present with dysphonia than bilateral disease (40.3% vs 9.1%, odds ratio [OR] 6.75, 95% confidence interval [CI] 1.77‐44.5, P = .01). MBS results did not vary statistically with respect to laterality, hypomobility versus immobility, or etiology. Of the 33 children who demonstrated aspiration on their initial MBS, 48.5% demonstrated complete resolution over median of 10.5 weeks. Rates of recovery differed only with respect to hypomobility versus immobility (87.5% vs 33.3%, OR 14.0, 95% CI 2.01‐286, P = .0133).

Conclusion

Based on the study results, about half of pediatric patients with vocal fold hypo/immobility will have resolution of dysphagia at about 10.5 weeks. This may helpful when deciding on when to obtain follow up imaging/exam on pediatric patients with dysphagia. The only factor that confers improved prognosis is hypomobility when compared to complete immobility.

After laryngomalacia, true vocal fold (TVF) immobility or paresis is the second most common pediatric laryngeal pathology. ¹ Etiologies of TVF immobility and hypomobility include cranial neuropathies (eg, recurrent laryngeal nerve), intrinsic structural disease (eg, glottic scarring), central neurologic disease (eg, tumor or stroke), or systemic disease. Often vocal fold hypomobility is acquired through interventions such as prolonged, repeat, or traumatic intubations and cardiothoracic surgery. Given the myriad of pathologies that can result in this laryngeal disorder, it is difficult to elucidate overall prognosis and return of vocal fold mobility. Rate of complete recovery varies anywhere from 8% to 82%, with the largest study reporting only 28% resolution.

TVF immobility or hypomobilty can lead to significant morbidity in the pediatric population. The vocal cords serve 3 primary functions: (1) phonation, (2) airway protection through glottic competence, and (3) airway patency. Permanent disruption of these vital functions can result in the need for surgical intervention and eventual tracheotomy or gastrostomy tube dependence. Oropharyngeal dysphagia is defined by the presence of laryngeal penetration or aspiration of food contents secondary to dysfunction of the oral and pharyngeal phase of swallowing. ³ Leder et al demonstrated that adults with TVF immobility have 40% rate of aspiration and approximately 2.5 greater odds of aspirating compared to patients who undergo laryngoscopy, which does not reveal vocal cord immobility. ⁴ Chronic dysphagia can lead to malnutrition, failure to thrive, reactive airway disease, chronic cough, bronchiectasis, and aspiration pneumonia. ³

Although numerous studies have investigated the return of TVF function in pediatric patients, there is a paucity of research investigating functional swallow outcomes. When described, swallow outcomes have been primarily assessed either subjectively through patient‐reported symptomatic improvement (eg, tolerated diet) or through association with the rate of gastrostomy tube dependence. The modified barium swallow (MBS) is a video fluoroscopic study that remains one of the best available tools to objectively assess oropharyngeal and upper esophageal swallow function. ³

Previous studies shown have revealed that clinical recovery of swallow in pediatric patients does not correlate with return of vocal cord function. ⁵ , ⁶ , ⁷ When MBS was utilized to evaluate swallow function, it was implemented only for the initial assessment, which is not predictive of return of swallow function. ⁵ The purpose of our study is to elucidate the incidence and prognosis of oropharyngeal dysphagia utilizing MBS in pediatric patients with vocal fold hypomobility.

Methods

Cleveland Clinic institutional review board approval and exemption was obtained before initiation of this investigation. A retrospective chart review was performed of all pediatric patients, in the last 10 years, who were diagnosed with unilateral or bilateral TVF hypomobility or immobility via flexible fiberoptic laryngoscopy performed by fellowship‐trained pediatric otolaryngologists at a single tertiary care academic center. Only patients who had an MBS study conducted were included (89 of 177 patients). Patients were excluded if they did not have an MBS or a flexible laryngoscopy confirming vocal fold immobility/hypomobility. Electronic medical records were used for retrospective data extraction. Vocal fold immobility was defined as no evidence of movement of the vocal fold or arytenoid. Hypomobility was defined as some evidence of movement without complete closure of the vocal folds.

Variables collected include patient characteristics (age at the time of diagnosis, sex, and ethnicity), presenting symptoms, comorbidities, follow‐up time, severity, laterality of vocal fold hypomobility, suspected etiologies, initial MBS results, and time elapsed between subsequent MBS studies. Time to recovery was defined as time elapsed between initial failed MBS and first MBS to demonstrate resolution of dysphagia. Documented status of TVF mobility was recorded as follows: hypomobility or immobility and unilateral or bilateral. Data regarding positioning of immobile cords were not documented consistently and therefore not collected.

MBS swallow results were assessed for presence or absence of penetrance and/or aspiration. If aspiration was silent, this was also recorded. Failure of a swallow study was defined as a patient with evidence of penetrance or aspiration.

Statistical Analysis

For comparison of categorical and continuous variables, χ ² and t‐test were used, respectively, with threshold for significance set for P < .05. Odds ratios (OR) with 95% confidence intervals (95% CI) were reported to reflect effect size for group comparisons. Comparisons of groups with respect to mutually exclusive result categories with more than 2 possible values were performed using Fisher's exact test. Statistical analysis was performed in JMP 9.0 (SAS Institute, Inc., Cary, NC, USA).

Results

Eighty‐nine pediatric patients were identified (52.8% male, mean age 4.6 years, standard deviation [SD] 5.5 years) with mean follow‐up was 35.4 months (SD 37.1 months). Ethnicities of the patients included: Caucasian (64.0%), black (15.7%), multiracial (7.9%), and other (12.4%). The laterality and severity of vocal fold hypomobility are summarized in Table 1. Sixty‐seven of the 89 patients (75.3%) had unilateral disease. Complete immobility and hypomobility were observed in 77.5% and 16.9% of our patients, respectively.

Table 1.

Distribution of Vocal Fold Hypomobility

	Unilateral (75.3%)	Bilateral (24.7%)
	(n = 67)	(n = 22)
Laterality	Left 83.6%	N/A
Complete immobility	54 (80.6%)	15 (68.2%)
Hypomobility	9 (13.4%)	6 (27.3%)
Unspecified severity	4 (5.9%)	1 (4.5%)

There was no statistically significant difference between males and females with respect to prevalence of vocal fold hypomobility (P = .67) or unilateral versus bilateral disease (P = .85). The most common presenting symptoms were dysphagia (46.1%, n = 41), dysphonia (43.8%, n = 39), and stridor or respiratory distress (20.2%, n = 18). As displayed in Table 2, when comparing unilateral disorder to bilateral disorder, the only presenting symptom that differed with statistical significance between the 2 groups was dysphonia (55.2% vs 9.1%, respectively, OR 12.3, 95% CI 3.25‐81.2, P = .0002). As presented in Figure 1, etiologies included cardiothoracic surgery (58.4%), intubated‐related (10.1%), central lesion or neurological surgery (9.0%), trauma (7.9%), and other (14.6%). Comorbidities included congenital structural heart defect, (51.7%), upper airway including laryngomalacia, subglottic stenosis and pulmonary disease (37.1%), central nervous system (CNS) pathology (32.6%), prematurity (20.2%), gastrointestinal disease (16.9%), other chromosomal abnormality or syndrome (13.5%), tracheoesophageal fistula (TEF) and/or laryngeal cleft (5.6%), Arnold‐Chiari malformation (3.4%), and other (20.2%). Of note, trauma included patients with traumatic brain injuries and direct trauma to the neck.

Table 2.

Comparing Presenting Symptoms in Unilateral Versus Bilateral Vocal Fold Hypomobility

	Unilateral	Bilateral	P value
Dysphagia	29 (43.3%)	12 (54.5%)	.36
Dysphonia	37 (55.2%)	2 (9.1%)	.0002, OR 12.3, 95% CI 3.25‐81.2
Respiratory distress/stridor	13 (19‐4%)	5 (22.7%)	.74

Figure 1.

Summary of etiologies of true vocal fold paralysis or paresis in our patient population.

Initial Swallow Outcomes

A total of 178 MBS studies were obtained across 89 patients (mean 1.35, SD 1.35, range 1‐9). Initial MBS outcomes at the time of vocal fold hypomobility are outlined in Table 3. Rates of observed penetrance or aspiration (“failed“ study) did not differ with statistical significance in (1) unilateral versus bilateral disorder (56.7% vs 50.0%, respectively, OR 1.31, 95% CI 0.50‐3.47, P = .63), (2) hypomobility versus immobility (46.7% vs 59.4%, respectively, OR 0.60, 95% CI 0.19‐1.85, P = .58) or (3) dysphonia versus normophonia (46.9% vs 65.0%, OR 0.48, 95% CI 0.20‐1.11, P = .13). When comparative analysis was performed including only aspiration in the “failed” study group (penetrance was considered as a “pass”), again, there was no observed difference. Similarly, the overall distribution of swallow outcomes did not differ with statistical significance across the 3 categories (P = .19, P = .80, and P = .24, respectively). Failure rates of initial MBS studies did not differ with statistical significance when comparing the presence versus absence of the following etiologies: cardiothoracic surgery (51.9% versus 59.5%, P = .52, OR 0.74, 95% CI 0.31‐1.72), CNS pathology (75.0% vs 53.1%, P = .29, OR 2.65, 95% CI 0.57‐18.8), intubation‐related (66.7% vs 53.8%, OR 1.72, 95% CI 0.42‐8.60, P = .51), and trauma (71.4% vs 53.7%, OR 2.16, 95% CI 0.44‐15.7, P = .45).

Table 3.

Distribution of Initial Modified Barium Swallow Outcomes

	Negative	Penetrance	Aspiration	Silent aspiration	P value (Fisher's exact test)
Unilateral (n = 67)	29 (43.3%)	5 (7.5%)	14 (20.9%)	19 (28.4%)	.19
Bilateral (n = 22)	11 (50.0%)	1 (4.5%)	8 (36.4%)	2 (9.1%)
Hypomobility (n = 15)	8 (53.3%)	o (0%)	4 (26.7%)	3 (20.0%)	.8
Immobility (n = 69)	28 (40.6%)	6 (8.7%)	17 (24.6%)	18 (26.1%)
Without dysphonia (n = 40)	14 (35.0%)	2 (5.0%)	13 (32.5%)	11 (27.5%)	.24
With dysphonia (n = 49)	26 (53.1%)	4 (8.2%)	9 (18.4%)	10 (20.4%)

Follow‐Up Swallow Outcomes

Figures 2 and 3 illustrate the rates of aspiration resolution and time to recovery in our patient population. Of the 43 children who demonstrated aspiration on their initial MBS, 33 had a follow‐up MBS, 8 of whom (24.2%) showed complete resolution by their second study over median 8.0 weeks (Interquartile range [IQR] 14.3) and an additional 8 patients (24.2%) on their third study over median 11.5 weeks (IQR 22.5) from time of first study. Overall, 48.5% demonstrated complete resolution of dysphagia over median 10.5 weeks (IQR 14.8). Both rate of recovery (44.4% vs 66.7%, OR 0.4, 95% CI 0.05‐2.42, P = .40) and time to recovery (P = .42) did not vary with statistical significance when comparing unilateral and bilateral disorder. Rate of aspiration resolution varied with statistical significance in hypomobility versus immobility group (87.5% vs 33.3%, OR 14.0, 95% CI 2.01‐286, P = .01); however, the time to recovery did not (mean 13.8 weeks vs 38.0 weeks, P = .37). Of the patients with persistent dysphagia direct laryngoscopy bronchoscopy was performed in all but 3 patients and laryngeal cleft and TEF were ruled out. One of the patients with persistent dysphagia did have a repaired TEF. Seventeen patients in the study did have tracheostomies, and 29.4% (5 of the 17) of the patients with tracheostomies had persistent dysphagia.

Figure 2.

Evolution of overall swallow outcomes in our patient population.

Figure 3.

Density histogram plot of overall cohort time to recovery of dysphagia.

Discussion

Through retrospective chart review, our authors aimed to evaluate swallow function objectively using MBS in children with impaired vocal fold mobility and to determine prognosis overtime. This study found that the mean time of dysphagia resolution in pediatric patients with either vocal fold immobility or hypomobility was 13.8 weeks. Nearly half of patients showed recovery at 2.5 months from initial diagnosis.

Much of our current understanding of swallow function in patients with impaired vocal fold mobility stems from adult literature focused on unilateral disease. However, even in this population, results have varied significantly. Schiedermayer et al elegantly described the challenges in elucidating swallow outcomes in patients with impaired vocal fold mobility. ¹¹ These include inadequate delineation between subjective and objective dysphagia, and an innately heterogeneous patient population with respect to disease etiology and existing comorbidities. Additionally, there is a paucity of studies with consistent methodology that account for these confounding variables leading to heterogeneous results. In turn, this has rendered it difficult to counsel patients on prognosis and expected outcomes.

The pathophysiology of dysphagia secondary to impaired vocal fold mobility has not been clearly elucidated; however, several plausible hypotheses exist: (1) reduced airway protection due to impaired adduction, (2) insensate laryngopharynx, (3) weak cough, (4) impaired clearance of oropharyngeal secretions, and (5) pharyngeal and/or upper esophageal dysmotility or weakness. ⁶ , ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² , ¹³ Most likely, several of these mechanisms are contributing concomitantly. For example, through endoscopic evaluation of swallowing in adults, Tabaee et al were able to demonstrate increased aspiration rates of pureed foods in subgroups with concurrent impaired sensation, adductor reflex, and/or pharyngeal constriction when compared to patients with either finding in isolation. ¹⁰

The American Academy of Otolaryngology–Head and Neck Surgery (AAO‐HNS) has published Clinical Practice Guidelines (CPG) for the management of dysphonia. ¹³ , ¹⁴ The expert panel devoted careful attention to neonatal hoarseness, calling for laryngoscopic evaluation to identify laryngeal pathologies including vocal fold hypomobility that “might affect [the] ability to swallow.” Similarly, the CPG discusses that dysphonia in children may also be the harbinger of a serious etiology, particularly if it is (1) persistent or (2) associated with other symptoms including dysphagia. Although the CPG's key action statements address indications for a diagnostic laryngoscopy and the management of impaired vocal fold mobility, it does not comment on the necessity for ancillary swallow testing. Moreover, the AAO‐HNS does not currently have any published practice guidelines on dysphagia. Due to the lack of formal testing criteria across several multidisciplinary organizations, the decision to pursue objective swallow evaluation in patients with confirmed vocal fold mobility disorders is often variable and ultimately provider or institution‐dependent.

The pediatric literature is scarce of objective swallow assessment in patients with impaired vocal fold mobility. When described, swallow outcomes are often secondarily reported and seldom contain subvariant analyses. Nevertheless, our study adds to the current literature including that of Turong et al who report a 45% aspiration and/or penetrance rate in 55 pediatric patients with unilateral vocal fold hypomobility caused by cardiac surgery. ¹⁵ Tibetts et al had a cohort of 36 pediatric patients with unilateral immobility with documented MBS studies, 69.4% of whom aspirated on initial assessment. ⁵ In a larger cardiac series of 2255 children by Sachdeva et al, laryngoscopic confirmation of vocal fold hypomobility was noted in 1.7% of children (n = 38). An MBS was performed in 29 of 38 children. They report a high aspiration rate of 82.7% (24/29) and penetrance of 13.8% (4/29). Moreover, 18 of 38 patients with impaired vocal fold mobility required gastrostomy tube placement before discharge from hospital. An additional 7 patients required nasogastric tubes. ¹⁶

Our group reports 89 pediatric patients with known vocal fold hypomobility of various etiologies, both unilateral versus bilateral, and hypomobility versus immobile. MBS revealed nearly 60% of patients either had silent aspiration, aspiration, or penetrance on their first MBS study. Subvariant analysis of our initial MBS results was performed, and there was no statistically significant difference with respect to unilateral versus bilateral disease, hypomobility versus immobility or etiology. To our knowledge, Jabbour et al report the largest single series (n = 404) of pediatric patients with impaired vocal fold mobility (immobility only). ² Using G‐tube dependence as a surrogate for swallow outcomes, they reported 40.8% (n = 165) dysphagia in their cohort. Overall, our reported dysphagia rate of 56% was within range of the previously cited adult and pediatric literature.

Given previous evidence that impaired vocal fold movement does not correlate with swallow return, ² , ⁵ monitoring children through serial laryngoscopic examinations may not be reliable in predicting resolution of dysphagia and appropriateness for repeat MBS testing. Overall, reported rates of swallow recovery trend as greater than return of mobility, suggesting compensation. In 1 study, 73% of patients with unilateral impaired vocal fold movement were feeding orally compared to only 45% at time of presentation, despite only a 3% vocal fold mobility recovery rate. ¹⁷ Moreover, Jabbour et al found that despite only a 28% recovery rate of vocal fold immobility in their cohort, 45.8% of patients had resolution in their presenting dysphagia symptoms. ² Overall, it seems the natural course of dysphagia is favorable, either through recovery of function and/or compensation. Tibbets et al found that 79.5% of pediatric patients were feeding orally at follow‐up compared to 31.5% at time of diagnosis. Longitudinal follow‐up MBS studies (mean 2.3 total studies) were pursued in 25 of their 36 patients; however, the results of those studies were not reported. ⁵ To our knowledge, Sachdeva et al is the only group to report the results of follow‐up swallow testing in pediatric patients. ¹⁶ A total of 28 cardiovascular surgery patients (median follow‐up 12 months) underwent MBS testing, of which 17 required a follow‐up MBS (6.2 ± 4.1 months later). Eight patients demonstrated resolution, while 4 improved, 3 remained stable, and 2 patients progressed.

Management of impaired vocal fold movement‐related dysphagia in children can be challenging. In part, this is due to our poor understanding of its overall prognosis. This fundamental gap in knowledge renders it difficult to know when to observe with close surveillance or intervene surgically. As depicted in Figure 3, our cohort represents a wide range of time to recovery. Nevertheless, nearly half of our patients recovered within 3 months. Of the larger studies, Jabbour et al report a mean 4.3 months to resolution of vocal fold hypomobility by laryngoscopy. As suspected, this is longer than our reported time to resolution of dysphagia, presumably due to compensation. Moreover, our reported prevalence of dysphagia and the median time to recovery of 10.5 weeks is likely an overestimate, as it represents a subset of patients with more severe disease as patients included had a diagnosis of vocal fold immobility, paresis, and an MBS.

Overall, our results suggest that nearly half of pediatric patients with vocal fold hypomobility and dysphagia spontaneously resolve within approximately 2.5 months. Moreover, we found that patients with hypomobility TVF have improved prognosis when compared to complete immobility. These findings may be invaluable to the clinician surgeon who must decide when to offer repeat swallow testing and surgery. Temporary injection laryngoplasty (IL) may be offered up front as one awaits recovery. Cohen et al performed 11 injections in 6 pediatric patients with dysphagia or aspiration and noted 85% clinical improvement. ¹⁹ A recent systematic review by Aires et al comprehensively analyzed 22 studies in pediatric unilateral vocal fold hypomobility and differentiated swallow outcomes by subjective and objective measures. ²⁰ The review included several studies evaluating IL and reinnervation. The largest study assessing IL for dysphagia included 33 patients, 21 of whom (63.6%) advanced their diet following objective improvement on swallow study within 5 to 15 days post‐injection. ²¹ Notably, injection within 6 months of diagnosis of vocal fold hypomobility resulted in improved success. Caloway et al conducted the largest reinnervation study (n = 17) for aspiration. ²² They report 94.1% success, defined by diet enhancement by at least one‐half consistency. When assessing only the objective studies included in the 2021 systematic review, our authors estimate that 44 of 63 subjects improved with IL, while 29 of 32 subjects improved with reinnervation. These rates are comparable to the reported meta‐analysis, which included both subjective and objective outcomes: 84.5% (CI 82.6%‐88.4%) improvement for IL and 91.6% (CI 88.2‐94.9%) for reinnervation.

Limitations

There are limitations inherent to our study question and design that merit consideration. First, there has been longstanding controversy with respect to definition of terms in vocal fold hypomobility. As a result, terms have varied and evolved over time in the literature. Inter‐examiner variability in flexible fiberoptic laryngoscopy and MBS testing have also made it difficult to investigate vocal fold hypomobility on a large scale. The authors feel that the single‐institution nature of our design minimizes examiner heterogeneity. Flexible fiberoptic laryngoscopy was performed only by 3 practicing fellowship‐trained pediatric otolaryngologists who share consensus with respect to definitions and documentation as detailed in our methods section.

Although laryngeal EMG is the gold standard to diagnose and differentiate between hypomobility and immobility, the technology is not routinely used in pediatric cases as many will not tolerate it without anesthesia and may not be accessible in all institutions. One survey study found less than 30% of providers use EMG to diagnose vocal fold function disorders. ²⁴ , ²⁵ In addition, studies have shown that laryngoscopy is the predominant tool used by most laryngologists in the diagnosis of hypomobility and there is fair inter‐rater reliability in diagnosis. ²⁶ , ²⁷ In this study, 3 fellowship‐trained pediatric otolaryngologists diagnosed the hypomobility and immobility, thereby relying on subspecialty expertise. There is an absolute need for further studies utilizing laryngeal electromyography in evaluating children with dysphagia

Second, given the retrospective nature of this study, the reasons for not obtaining additional MBS testing in some patients were unclear. This could have potentially increased our overall rate of recovery as presumably, most patients who did not have a repeat MBS had clinical resolution of their dysphagia symptoms. In addition, the exact time of vocal fold insult in some cases is unknown and could impact time to recovery. While other comorbid medical conditions such as genetic abnormalities, laryngeal cleft, TEF, tracheostomy and CNS lesions were captured numbers were too small for bivariable analysis but could impact prognosis. Despite these acknowledged limitations, to our knowledge, this is the largest pediatric series to date that objectively evaluates swallowing function over time in patients with either unilateral or bilateral vocal fold hypomobility without intervention. The need for high‐quality randomized controlled trials is evident. The clinical consequences of withholding possible therapeutic options from pediatric patients with active dysphagia can be grave, rendering it difficult to ethically acquire a substantial control group. ²³ In addition, the necessity for a robust control group is further underlined by the relatively high rates of spontaneous resolution observed in this disease entity. We echo previous authors. ¹⁸ in imploring others who surgically manage vocal fold hypomobility to document and present their swallow outcomes to advance our collective understanding.

Conclusions

Our study suggests that objective recovery in dysphagia can be appreciated in nearly half of patients with vocal fold hypomobility within 10.5 weeks without any intervention. An important prognostic indicator may be the status of vocal fold mobility as hypomobility versus immobility. One may offer temporary injection in the interval, after which a repeat swallow study may guide further definitive surgical management. This information can prove crucial in counseling families.

Author Contributions

Khashayar Arianpour, design, conduct, analysis, presentation, editing; Samantha Anne, analysis, editing; Rachel Georgopoulos, design, conduct, analysis, editing.

Disclosures

Competing interests

None.

Funding source

None.