Abstract
Objective:
Quantitative measures of swallowing function were extracted from modified barium swallowing studies (MBS) to characterize swallowing pathophysiology in patients with unilateral vocal fold immobility (UVFI).
Methods:
All individuals with UVFI completing a MBS during the prior five years were included. Demographic information, penetration-aspiration score, timing of aspiration and quantitative measures from the MBS were extracted from electronic medical records and compared across 1, 3, and 20 cc liquid bolus swallows. UVFI patient measures were compared to normal age-matched controls to identify swallowing pathophysiology associated with aspiration. The incidence of aspiration by UVFI etiology groups (i.e., central nervous system, idiopathic, iatrogenic, skull base tumor, or peripheral tumor) was also compared.
Results:
Of the 61 patients who met inclusion criteria, aspiration was observed in 23%. Maximum pharyngeal constriction was abnormal in 79% of aspirators compared to 34% of non-aspirators (p=.003). Delay in airway closure was the most common swallowing abnormality identified in the study population (62%) but was not associated with aspiration. Among the 14 individuals who aspirated, the iatrogenic and skull base tumor etiology groups comprised the majority (i.e. 36% each). However, the incidence of aspiration for the iatrogenic group was 19% compared to 50% of the skull base tumor group.
Conclusions:
Aspiration in patients with UVFI was associated with abnormally reduced pharyngeal constriction. Delayed airway closure was common in both aspirators and non-aspirators.
Keywords: dysphagia, quantitative measures, videofluoroscopic swallow studies, pathophysiology, unilateral vocal fold immobility
Introduction
The diagnosis of unilateral vocal fold immobility (UVFI) is hypothesized to increase risk of aspiration due to impaired laryngeal adduction during swallowing [1-5]. However, our previous review of outpatients with UVFI (N=415) revealed a low incidence of dysphagia signs and symptoms severe enough to warrant further instrumental assessment using modified barium swallow (MBS) studies (15%), a low incidence of pneumonia (6%), and an overall aspiration rate of only 3% [6]. These findings indicate that most patients with UVFI are able to swallow safely despite incomplete vocal fold closure. Nonetheless, identification of those patients with UVFI at risk for aspiration is critical to avoid potential pulmonary compromise. An understanding of patient characteristics and swallowing pathophysiology that predict aspiration risk in the UVFI population is needed. This preliminary study examines swallowing pathophysiology in the subset of patients with UVFI who underwent an MBS to characterize patient factors and swallowing function most predictive of aspiration.
Quantitative MBS study measures applied to the evaluation of swallowing physiology are associated with improved rater reliability and elucidation of dysphagia pathophysiology [7, 8]. Previously reported inter-rater reliability for the quantitative measures used to evaluate these MBS studies ranged from 82-90% [7]. Furthermore, patient measures can be compared to those from normal age-matched controls, allowing identification of abnormality. The use of quantitative measures has the potential to identify subtle functional abnormalities that may be missed during visual perceptual analysis alone [9]. This is illustrated by the finding that up to 60% of patients with a normal Penetration-Aspiration score are found to have a delay in airway closure by quantitative measures [7]. As airway protection is hypothesized to be the predominant pathophysiology in UVFI swallowing dysfunction, it is particularly important to apply these measurement techniques to the evaluation of the UVFI population. This study uses five quantitative measures of swallowing function, in addition to standard visual perceptual assessments of aspiration and bolus residue to describe swallowing physiology in a dysphagic UVFI population.
Methods
This retrospective study was approved by the University of Utah Institutional Review Board (Protocol # 00114169).
Participants
The University of Utah Medical Center’s electronic records were searched to identify individuals ≥18 years-old diagnosed with UVFI seen from January 1, 2013 through December 31, 2018 in the Utah Voice Disorders Center (VDC), a regional outpatient university medical center. A total of 476 individuals were initially identified using these search criteria. Sixty-one of these patients did not meet inclusion criteria upon review of their medical record. Individuals were excluded for various reasons including lack of evidence indicating impaired vocal fold mobility following SLN lesion, age limitations to exclude minors, bilateral vocal fold impairment, and diagnosis of UVFI not assigned by an otolaryngologist. Following application of exclusion criteria, 415 individuals were included in the previous study [6]. Of these 415, a total of 62 individuals completed an MBS assessment from which quantitative measures of swallowing function were performed and recorded. This subset of patients demonstrated significant signs or symptoms of dysphagia during their clinical evaluation warranting further instrumental assessment. Upon further review, one of these individuals was excluded from this study because their UVFI was observed isolated to stimulation of the vagal nerve by their implanted device. Therefore, a total of 61 individuals were included in this study.
Videofluoroscopic Study Procedures
All MBS procedures and recordings took place in the radiology suite at the University of Utah main hospital. All patients were positioned in the lateral plane of view and a disk 1.8 cm in diameter was secured on the midline of the patient’s chin prior to initiation of the MBS protocol to be used in later calibration of distance and spatial measures [7, 8]. All liquid boluses were prepared by the speech-language pathologist (SLP) for administration to the patient. Boluses were measured using a syringe and delivered to the patient via a cup or spoon. Varibar thin liquid sulfate was utilized for 1-cc, 3-cc, and 20-cc thin liquid swallows in the lateral plane of view. The VDC clinic protocol also includes a 5-cc paste bolus, a cookie bolus, and Barium tablet in the lateral plane of view as well as a 20-cc liquid bolus viewed in the anterior-posterior plane. These latter bolus consistencies were not included in the study due to a lack of normative data for comparison. Videos were recorded at 25 frames per second (fps) before January 2017 and at 30 fps after January 2017.
Frame-by-frame analysis of the MBS recordings was performed following completion of the study by the SLP who conducted the study. All SLPs in the Utah Health Voice Disorders Center are required to meet proficiency standards prior to independently performing and analyzing MBS studies. Clinicians extracted quantitative measures from the studies (see the MBS measures section below) and also rated the swallows using the Penetration-Aspiration Scale (PAS). The PAS is an 8-point rating scale used to assess patient risk of aspiration by rating penetration, aspiration and the ability of the patient to eject the aspirated contents from the airway [10]. The PAS ranges from a score of 1 (none of the bolus entered the airway at any time during the swallow) to 8 (the bolus entered the airway and the patient did not respond voluntarily to expel the bolus). A PAS score greater than 5 was considered to indicate the occurrence of aspiration.
MBS Measures
The extraction of MBS measures was conducted either manually or using a software approach. Quantitative spatial analysis of MBS studies conducted before 2016 required two software programs, IMAGE-J (http://rsb.info.nih.gov/nih-image) and Universal Desktop Ruler (http://avpsoft.com/products/udruler/). In 2016, the software utilized for these measures was Swallowtail version 1.0.8, updated to version 2.2.17 in 2017. Both methods of analysis and software programs implemented the same methodology initially described by Leonard & Kendall [11]. For the purpose of this study, quantitative measures from 1-cc, 3-cc, and 20-cc liquid boluses, as well as the PAS ratings, were extracted from clinical reports in patient medical records. Extracted measures included: 1. Total Pharyngeal Transit time (TPT), 2. Maximum Opening of the Pharyngoesophageal Segment (PESmax), 3. Pharyngeal Area during Maximal Constriction during Swallowing (PAM), 4. Maximum Hyolaryngeal Excursion (Hmax), and 5. Timing of Airway Closure (AC) relative to bolus arrival in the pharyngoesophageal segment. A detailed description of these measures was previously reported and is briefly described as follows [7, 11].
TPT is defined as the time between when the head of the bolus passes the posterior nasal spine and when the tail of the same bolus clears the upper esophageal sphincter (UES). Airway closure (AC) is measured as the time associated with approximation of the arytenoid cartilages to the laryngeal surface of the epiglottis relative to the arrival of the bolus at the UES. Hmax was measured as the distance between the hyoid position during a reference bolus hold (i.e., the position of the hyoid when a 1-cc liquid bolus is held in the oral cavity) and the position of the hyoid at maximal superior and anterior excursion during the swallow. PESmax was defined as the greatest anterior-posterior UES opening during the swallow. PAM was defined as the area measured within the outline of the pharyngeal space containing the bolus during maximum pharyngeal constriction during the swallow.
Individual quantitative measures were defined as abnormal if they fell outside of the range measured from a group of historical normal controls (n=147) for any of the three liquid bolus volumes (i.e., 1-cc, 3-cc, or 20-cc) evaluated during swallowing. Data from patients younger than 65 years old were compared to data collected from 60 control subjects ranging from 18 to 65 years in age. Data collected from patients older than 65 years were compared to data from 87 control subjects ranging in age from 66 to 87 years old. Given that reference normative measures can differ by age and sex, each patient measure was assigned a binary term of 0 for “normal” or 1 for “abnormal,” to compare normal and abnormal outcomes. This was done to assign appropriate cutoff scores by gender and age to assure accurate interpretation of normal and abnormal continuous values.
The timing of aspiration relative to the onset of laryngeal vestibule closure (LVC) was determined through visual-perceptual observation of each recording by the first and senior author for the 14 individuals who aspirated (defined by PAS>5) during the MBS study. The ratings included three options: before (0), during (1) or after (2) LVC. Interrater reliability of 90% was achieved for these measures.
Etiology Classifications
All participants completing an MBS were grouped for descriptive comparison by etiology of UVFI. These groups included participants with UVFI due to iatrogenic, idiopathic, central nervous system (CNS), skull base tumor (SBT), or peripheral tumor (PT) etiologies. These etiologic classifications were operationally defined to elucidate sites of vagal lesion associated with UVFI. Individuals classified in this study as CNS were those diagnosed with UVFI associated with cerebrovascular accident (CVA), traumatic brain injury (TBI), or a neurodegenerative disease originating in the brain (e.g., Multiple Sclerosis). Those classified as skull base tumor (SBT) were individuals whose onset of UVFI was due to a tumor at the base of skull including those pre- and post-surgical treatment. Participants in the peripheral tumor (PT) group were diagnosed with a tumor involving the recurrent laryngeal nerve in the thoracic cavity. Those in the iatrogenic group were all remaining individuals with onset of UVFI after undergoing surgery associated with injury to the RLN, and those in the idiopathic group were those without a known cause. Thus, individuals were classified into one of the following five etiologies: 1. Iatrogenic (n=26), 2. Idiopathic (n=14), 3. Central Nervous System (CNS) (n=9), 4. Skull Base Tumor (SBT) (n=10), and 5. Peripheral Tumor (PT) (n=2).
Statistical Analysis
Participant demographic and clinical measurement data were summarized using mean and standard deviation (SD) or median and interquartile range (IQR) if the variable was continuous. Categorical variables were summarized using counts and percentages. Descriptive statistics were presented for the full cohort and for the subset of subjects identified with aspiration during the MBS study. MBS measures and etiologies were compared between those identified as aspirators and non-aspirators using chi-squared or Fisher’s exact test as appropriate. Due to limited sample sizes in etiology categories (ranging from N=2 to 26), demographics and clinical outcomes were not compared statistically across etiology groups. The small number of aspiration events observed in our study cohort (N = 14) limited us from performing a multivariable analysis with aspiration as the outcome. Statistical significance was assessed at the .05 level and statistical analyses were implemented using R (v 3.6.0, R Core Team, 2019) [12].
Results
A total of 61 individuals were included in this study. Table 1 describes the demographic information of the patient population. The average age of individuals in this study was 61.8 years (ranging from 23-90+) with slightly more females than males (52% female, 48% male). Any patient over 90 years of age was considered 90+ to comply with Institutional HIPPA policies but scaled to age 90 for analysis purposes. The majority of individuals had a left sided UVFI (59%). All individuals who completed the Eating Assessment Tool-10 (EAT-10) or Dysphagia Handicap Index (DHI) had total scores indicative of dysphagia.
Table 1.
Summary of demographic and pathophysiology measures stratified by etiology.
| Variable | All N=61 N(%) |
CNS N=9 N (%) |
Iatrogenic N=26 N (%) |
Idiopathic N=14 N (%) |
Peripheral Tumor N=2 N (%) |
Skull Base Tumor N=10 N (%) |
|---|---|---|---|---|---|---|
| Gender: Female | 32 (52%) | 3(33%) | 13(50%) | 8(57%) | 2(100%) | 6(60%) |
| Male | 29 (48%) | 6 (67%) | 13 (50%) | 6 (43%) | 0 (0%) | 4 (40%) |
| Age: Mean (SD) | 61.8 (14.7) | 69.7 (11.1) | 58.7 (16.8) | 63.1 (15.1) | 66.5 (7.8) | 59.9 (10.7) |
| Median (IQR) | 61.0 (53.0, 71.0) | 68.0 (67.0, 76.0) | 59.0 (49.0, 69.8) | 61.5 (52.2, 76.8) | 66.5 (63.8, 69.2) | 62.5 (55.0, 67.0) |
| Range | (23.0, 90.0) | (53.0, 90.0) | (23.0, 88.0) | (40.0, 84.0) | (61.0, 72.0) | (36.0, 72.0) |
| Years since onset of UVP: Mean (SD) | 5.0 (7.4) | 1.6 (1.1) | 8.5 (9.8) | 2.8 (2.9) | 0.3 (0.1) | 3.0 (4.4) |
| Median (IQR) | 2.0 (0.4, 5.2) | 1.6 (0.8, 2.0) | 3.5 (1.8, 12.9) | 1.9 (0.5, 4.5) | 0.3 (0.3, 0.3) | 1.0 (0.3, 2.8) |
| Range | (−0.5, 35.0) | (0.2, 3.7) | (0.2, 35.0) | (0.1, 10.0) | (0.2, 0.4) | (−0.5, 11.3) |
| UVP side: Left | 36 (59%) | 5(56%) | 16(62%) | 9(64%) | 2(100%) | 4(40%) |
| Right | 25(41%) | 4 (44%) | 10(39%) | 5(36%) | 0(0%) | 6(60%) |
| DHI: Mean (SD) † | 46.4 (18.0) | 58.5 (15.8) | 37.1 (9.2) | 45.2 (9.1) | - | 67.0 (46.7) |
| Median (IQR) | 42.0 (34.0, 53.0) | 56.0 (52.5, 62.0) | 34.0 (32.0, 42.0) | 44.0 (40.0, 46.0) | - | 67.0 (50.5, 83.5) |
| Range | (24.0, 100.0) | (42.0, 80.0) | (24.0, 52.0) | (36.0, 60.0) | - | (34.0, 100.0) |
| EAT-10: Mean (SD) ‡ | 18.9 (10.6) | 15.0 (5.7) | 18.0 (8.8) | 11.0 (8.4) | 14.5 (10.6) | 30.8 (8.8) |
| Median (IQR) | 17.0 (10.0, 24.5) | 17.0 (10.0, 19.0) | 15.0 (12.2, 24.5) | 10.0 (5.0, 16.5) | 14.5 (10.8, 18.2) | 34.0 (24.0, 37.2) |
| Range | (0.0, 40.0) | (8.0, 21.0) | (6.0, 37.0) | (0.0, 24.0) | (7.0, 22.0) | (15.0, 40.0) |
| DHI>5 or EAT-10>2 | 55 (90%) | 9(100%) | 23(88%) | 11(79%) | 2(100%) | 10(100%) |
| PAS Score: 6 | 3 (5%) | 0(0%) | 1(4%) | 0(0%) | 1(50%) | 1(10%) |
| 7 | 3 (5%) | 0(0%) | 2(8%) | 0(0%) | 0(0%) | 1(10%) |
| 8 | 8 (13%) | 2(22%) | 2(8%) | 1(7%) | 0(0%) | 3(30%) |
| PAS score: PAS≤5 | 47 (77%) | 7(78%) | 21(81%) | 13(93%) | 1(50%) | 5(50%) |
| PAS>5 | 14 (23%) | 2(22%) | 5(19%) | 1(7%) | 1(50%) | 5(50%) |
| Abnormal Measure for any Bolus Size |
All (N=61) N (%) |
CNS (N=9) N (%) |
Iatrogenic (N=26) N (%) |
Idiopathic (N=14) N (%) |
Peripheral Tumor (N=2) N (%) |
Skull Base Tumor (N=10) N (%) |
| TPT | 5 (8%) | 1(11%) | 0(0%) | 1(7%) | 1(50%) | 2(20%) |
| PESmax | 12 (20%) | 2(22%) | 2(8%) | 1(7%) | 2(100%) | 5(50%) |
| PAM | 27 (44%) | 2(22%) | 10(39%) | 5(36%) | 1(50%) | 9(90%) |
| Hmax | 13 (21%) | 2(22%) | 5(19%) | 1(7%) | 2(100%) | 3(30%) |
| AC | 38 (62%) | 5(56%) | 17(65%) | 9(64%) | 1(50%) | 6(60%) |
Missing values: DHI or EAT-10 missing values for 5 individuals who completed neither questionnaire as their clinic visit was for the purpose of voice assessment rather than dysphagia. Patients completed either the EAT-10 or the DHI depending on the year they were evaluated at the VDC.
Based on a subset of N= 20 out of 61 completing the DHI.
Based on a subset of N= 36 out of 61 completing the EAT-10.
TPT = total pharyngeal transit time
PESmax = maximum pharyngoesophageal opening
PAM = pharyngeal area at maximum constriction
Hmax = hyolaryngeal maximum excursion
AC = timing of airway closure
Of the total population studied, 14/61 were identified to aspirate (PAS > 5) giving an overall incidence of aspiration of 23% (Table 1). The proportion of those documented with aspiration within each etiologic category was as follows: skull base tumor 5/10 (50%), iatrogenic 5/26 (19%), CNS 2/9 (22%), peripheral tumor 1/2 (50%), and idiopathic 1/14 (7%) (Table 1). A delay in airway closure (AC) was the most common abnormality of swallowing pathophysiology, found in 62% (38/61) of individuals regardless of etiology (Table 1). However, this was not associated with aspiration as it occurred at similar frequencies in both aspirators and non-aspirators (79% and 57% respectively, p=.15, Table 2). PAM was abnormal in 79% of aspirators compared to 34% of non-aspirators (p=.003) (Table 2).
Table 2.
Summary of pathophysiology and etiology stratified by aspirator/non-aspirator.
| Variables | All N=61 N (%) |
Non-Aspirator N=47 N (%) |
Aspirator N=14 N (%) |
p-value |
|---|---|---|---|---|
| Abnormal measure for any bolus size: | ||||
| TPT | 5 (8%) | 3(6%) | 2(14%) | - |
| PESmax | 12 (20%) | 6(13%) | 6(43%) | - |
| PAM | 27 (44%) | 16(34%) | 11(79%) | 0.003 c |
| Hmax | 13 (21%) | 8(17%) | 5(36%) | - |
| AC | 38 (62%) | 27(57%) | 11(79%) | 0.15c |
| Etiology: | 0.11f | |||
| Central CNS | 9 (15%) | 7(15%) | 2(14%) | |
| Iatrogenic | 26 (43%) | 21(45%) | 5(36%) | - |
| Idiopathic | 14 (23%) | 13(28%) | 1(7%) | - |
| Peripheral Tumor | 2 (3%) | 1(2%) | 1(7%) | - |
| Skull Base Tumor | 10 (16%) | 5(11%) | 5(36%) | - |
Fisher's exact test,
Chi-squared test.
Figure 1 illustrates aspirators categorized by etiology. Table 3 describes the timing of aspiration relative to LVC by etiology. Of those identified to aspirate, 21% (3/14) aspirated before LVC, 21% (3/14) aspirated during LVC, and 64% (9/14) aspirated after LVC. One individual from the SBT group was observed to aspirate during and after LVC. Of the 14 aspirators, Table 1 shows that 3 individuals received a score of 6 on the PAS, meaning they aspirated but spontaneously expelled the aspirant from the airway. Three individuals had a PAS of 7, meaning they sensed the bolus enter the airway and were observed to try to expel the aspirant without success. Additionally, 57% of aspirators (8/14) received a PAS of 8, meaning they aspirated silently at least once during their MBS. Table 4 displays the frequency of individuals who aspirated (N=14) by etiology and pathophysiology. Among the aspirators, the most common abnormal measures, regardless of etiology, were impaired pharyngeal constriction (i.e., PAM) (79%) and delayed airway closure (79%).
Fig. 1. Etiology of UVFI for aspirators (N=14).
*One participant in the Skull Base Tumor group aspirated during and after laryngeal vestibule closure
Table 3.
Aspirators within each etiology group are stratified by the timing of aspiration relative to laryngeal vestibule closure (LVC). The proportion of aspirators is indicated for each etiology group as well as their representation within categories of aspiration timing.
| Etiology | Before LVC N=3 N (%) |
During LVC N=3 N (%) |
After LVC N=9 N (%) |
|---|---|---|---|
| CNS | 0(0%) | 1(33%) | 1(11%) |
| Iatrogenic | 2(67%) | (0%) | 3(33%) |
| Idiopathic | 1(33%) | (0%) | 0(0%) |
| Peripheral Tumor | 0(0%) | 1(33%) | 0(0%) |
| * Skull Base Tumor | 0(0%) | 1(33%) | 5(56%) |
One participant in the Skull Base Tumor group aspirated during and after laryngeal vestibule closure.
LVC=laryngeal vestibule closure
Table 4.
Summary of pathophysiology stratified by etiology, among individuals who aspirated during MBS.
| Abnormal Measure for any Bolus Size |
All N=14 N(%) |
CNS N=2 N(%) |
Iatrogenic N=5 N(%) |
Idiopathic N=1 N(%) |
Peripheral Tumor N=1 N(%) |
Skull Base Tumor N=5 N(%) |
|---|---|---|---|---|---|---|
| TPT | 2 (14%) | 1(50%) | 0(0%) | 0(0%) | 1(100%) | 0(0%) |
| PESmax | 6 (43%) | 2(100%) | 0(0%) | 0(0%) | 1(100%) | 3(60%) |
| PAM | 11 (79%) | 1(50%) | 3(60%) | 1(100%) | 1(100%) | 5(100%) |
| Hmax | 5 (36%) | 1(50%) | 1(20%) | 0(0%) | 1(100%) | 2(40%) |
| AC | 11 (79%) | 2(100%) | 5(100%) | 1(100%) | 0(0%) | 3(60%) |
TPT = total pharyngeal transit time
PESmax = maximum pharyngoesophageal opening
PAM = pharyngeal area at maximum constriction
Hmax = hyolaryngeal maximum excursion
AC = timing of airway closure
Discussion
This study is the first to quantitatively characterize the pathophysiology of dysphagia and aspiration in those with UVFI. Quantitative measures were used to objectively compare pharyngeal events of a swallow to better explain differences between individuals who have UVFI and are at a higher risk of aspiration compared to those who are not. Both temporal and spatial quantitative measures were used to compare individuals who aspirated during any of the 1, 3, or 20 cc liquid bolus swallows.
Out of 61 individuals with UVFI undergoing MBS for dysphagia, only 14 (23%) were observed to aspirate. Our findings show that reduced pharyngeal constriction (as measured by PAM) significantly distinguishes between those who do and do not aspirate in this population. Most individuals who aspirated (9/14 or 64%) did so after completion of laryngeal vestibule closure due to pharyngeal residue resulting from poor pharyngeal constriction. Three individuals (21%) aspirated prior to laryngeal vestibule closure associated with delayed onset of airway closure, or due to reduced lingual-palatal closure resulting in loss of the bolus from the oral cavity prior to initiation of swallowing. Only 3 individuals among aspirators in this study (21%) were observed to aspirate during laryngeal vestibule closure indicative of impaired airway closure during the swallow. Overall, these findings do not support concerns that individuals with UVFI are at greater risk for aspiration due to impaired laryngeal closure during the swallow. Rather, our findings show that the majority of those with UVFI who aspirated exhibited impaired pharyngeal constriction and associated aspiration of pharyngeal residue. This supports a previous finding among individuals with UVFI following medialization showing that the majority of aspiration cases occurred after the swallow [13]. That is, those with UVFI at greatest risk for aspiration were those with impaired pharyngeal function in addition to impaired vocal fold mobility.
Safe and efficient swallowing requires both sensory and motor coordination and is not solely reliant on glottic closure [14]. The pharyngeal plexus as well as superior laryngeal nerve (SLN) contribute to pharyngeal constriction. The internal branch of the superior laryngeal nerve (iSLN) conducts sensory information from the laryngopharynx and larynx whereas the external branch of the superior laryngeal nerve (eSLN) provides motor innervation to the inferior pharyngeal constrictor (fast outer layer) and upper esophageal sphincter [15]. Damage to the eSLN or pharyngeal plexus may reduce pharyngeal constriction and result in increased pharyngeal residue placing individuals at a higher risk for aspiration. Thus, involvement of the pharyngeal plexus and eSLN would place those with UVFI at increased risk for aspiration due to incomplete clearance of material from the pharynx. Further evidence that additional neurologic involvement is likely present in aspirators with UVFI is the observation that the majority of individuals who aspirated in this study were silent aspirators (57%) (see Table 1). These findings indicate impaired iSLN and sensory RLN impairment. The other six individuals who aspirated scored a 6 or 7 on the penetration aspiration score, indicating that the sensory branches of the RLN and SLN were intact to sense when the bolus was aspirated, but expectoration of the aspirant was not successful.
Considering the small sample size, the two etiologic groups demonstrating the highest number of aspirators among this UVFI population were the iatrogenic (5/26; 19%) and skull base tumor (5/10; 50%) groups. Table 5 offers a detailed comparison of those within the iatrogenic population grouped by those identified with and without aspiration. Unfortunately, there were no clear patterns among iatrogenic patients to distinguish surgery sites likely to lead to aspiration. Although all of the iatrogenic group exhibited delayed airway closure, five of 26 in this group exhibited aspiration. Of those who aspirated in the iatrogenic group, 60% also exhibited abnormally reduced pharyngeal constriction. This finding demonstrates that the majority of aspirators in the iatrogenic group had impaired function of the pharyngeal constrictor musculature, either due to injury to the RLN, external SLN, or pharyngeal plexus. These findings are consistent with a previously published study of 25 individuals with UVFI reported to exhibit significantly reduced pharyngeal constriction compared to age- and sex-matched controls [16]. However, the association between pharyngeal weakness and risk of aspiration in those with UVFI was not previously assessed. Together, this prior and current study highlights the importance of studying the contribution of pharyngeal weakness toward risk of aspiration in those with UVFI in future work.
Table 5.
Surgery Type for Patients with Iatrogenic Etiology
| Surgery Type | Total number in Iatrogenic group: | Aspirators: | Non-Aspirators: |
|---|---|---|---|
| Thyroid Surgery | 8 | 1 | 7 |
| Anterior Cervical Discectomy & Fusion | 8 | 1 | 7 |
| Cardiac Surgery | 4 | 1 | 3 |
| Left carotid body tumor resection | 1 | 0 | 1 |
| Multiple hernia surgeries | 1 | 0 | 1 |
| Thoracic lymphadenectomy | 1 | 1 | 0 |
| Left neck paraganglionic tumor resection damaging vagus nerve | 1 | 1 | 0 |
| Tummy tuck | 1 | 0 | 1 |
| Mediastinoscopy | 1 | 0 | 1 |
All 61 individuals included in this study underwent a VFSS. Videofluoroscopic measures allowed quantification and meaningful interpretation of the physiology of deglutition for comparison of normal and abnormal patterns in those with UVFI. The findings of our previous study (Schiedermayer et al., 2019) showed that only 3% of the entire UVFI population (i.e. 14 of 415) aspirated [6]. This study evaluated a subset of the prior population recommended for additional assessment using MBS to elucidate the pathophysiology of dysphagia. In the 61 out of 415 (15%) of those meeting inclusion criteria for this study, only 14 of 61 (23%) were observed to aspirate. Our findings show that aspiration predominantly occurred when pharyngeal residue resulting from impaired pharyngeal constriction during the swallow was inhaled after the laryngeal vestibule re-opened upon completion of the swallow. Only 3 of 14 (21%) exhibited aspiration during the swallow. That is, aspiration in those with UVFI did not commonly occur due to impaired laryngeal closure during the swallow as was anticipated. These findings appear contrary to prior studies reporting improved patient self-report of dysphagia symptoms after surgical medialization of the immobile vocal fold [17, 18]. However, two studies addressing the same treatment compared swallow function imaging outcomes pre- and post-medialization with findings similar to this study. One of these studies conducted a retrospective evaluation of pre- and post-treatment MBS outcomes in a pediatric population showing that vocal fold mobility was not essential for safe swallow function [19]. Moreover, many of the studied children exhibited improved swallow function even though vocal fold immobility remained [19]. A more recent study by Kammer et al evaluated the contribution of vocal fold medialization to swallow function using simultaneous flexible endoscopic evaluation of swallowing (FEES) and manometry pre- to post-treatment. This study in adults demonstrated that successful vocal fold medialization did not lead to significantly improved swallow function post-treatment [20]. That is, prior work reliant on post-treatment patient self-report ratings offered positive findings in contrast to studies utilizing swallow function imaging. One possible explanation for differences between patient-based outcomes and swallow function outcomes is the belief by patients that they felt better after being surgically treated. However, measurable differences were not supported by studies utilizing imaging and physiologic characteristics of swallow function. Overall, the outcomes of swallow function studies in those with unilateral vocal fold immobility indicate that evaluation of the entire oropharyngeal apparatus and sensorimotor function is necessary to elucidate the underlying pathophysiology of dysphagia to inform treatment planning. Vocal fold medialization would be recommended for those individuals experiencing impaired laryngeal valving for adequate voice production and cough integrity necessary for pulmonary clearance. For those with pharyngeal clearance problems, additional therapeutic interventions should be considered including postural, swallow maneuvers, or dietary modifications to improve swallowing safety and effectiveness.
Overall, this study showed that aspiration in those with UVFI most commonly occurred on residue resulting from impaired pharyngeal constriction, regardless of etiology. Although etiology was not statistically compared due to the small group population sizes, individuals at greatest risk for vagal nerve injury involving the pharyngeal plexus and SLN represented the largest proportion of those who aspirated (i.e., iatrogenic and skull base tumor).
Limitations
This study is not without limitations. The UVFI MBS sample size was small, and individuals were unevenly distributed across etiology and aspiration groups, limiting the extent of statistical analysis. Additionally, the study population included only those referred for MBS, a group with more severe symptoms, and may not be representative of the larger cohort of UVFI diagnoses. That is, individuals with mild dysphagia signs or symptoms were not represented. In addition, the duration of the disease was not controlled for and participants were limited to those seen in one outpatient clinic setting in the western region of the United States. Individuals with UVFI within acute care or inpatient hospital settings were not studied. The manifestation of dysphagia in those who are hospitalized during acute phases of onset may exhibit different swallowing pathophysiology from a mobile outpatient population. It is possible that swallowing physiology reorganized over time as a compensation for acute onset of UVFI resulting in functional clearance and efficiency in the outpatient group [21]. Future work should expand to include acute stages of UVFI as well as MBS testing of those with mild dysphagia complaints to evaluate the generalizability of this study’s findings to the larger patient population.
Conclusion
Of 61 individuals with UVFI undergoing MBS assessment, 14 (23%) were observed to aspirate. Approximately 71% of aspirators in the study group were classified as having either an iatrogenic or a skull base tumor etiology of UVFI. The most common abnormalities of swallowing function in those with UVFI undergoing MBS was delayed airway closure (62%) and impaired pharyngeal constriction (44%). Statistical comparison of aspirators and non-aspirators with UVFI showed an association of impaired pharyngeal constriction with aspiration, most commonly on pharyngeal residue after laryngeal vestibule closure. These findings indicate that individuals with UVFI at greatest risk for aspiration are those with impaired pharyngeal constriction function during swallowing resulting in reduced bolus clearance.
Acknowledgments
This investigation was supported by the University of Utah Study Design and Biostatistics Center, with funding in part from the National Center for Research Resources and the National Center for Advancing Translational Sciences, National Institutes of Health, through Grant UL1TR002538 (formerly 5UL1TR001067-05, 8UL1TR000105 and UL1RR025764). Partial support was also provided by the National Institutes on Deafness and Other Communication Disorders through Grant R01DC011311.
Footnotes
All authors have approved the manuscript and have agreed to its contents. All authors certify that they have no financial disclosures related to this work. All authors certify that they have no conflict of interest to report in relation to this work.
References
- 1.Tsai MS, et al. , Unilateral Vocal Fold Paralysis and Risk of Pneumonia: A Nationwide Population-Based Cohort Study. Otolaryngol Head Neck Surg, 2018. 158(5): p. 896–903. [DOI] [PubMed] [Google Scholar]
- 2.Damrose EJ, Percutaneous injection laryngoplasty in the management of acute vocal fold paralysis. Laryngoscope, 2010. 120(8): p. 1582–90. [DOI] [PubMed] [Google Scholar]
- 3.Zuniga SA, Ebersole B, and Jamal N, Utility of Eating Assessment Tool-10 in Predicting Aspiration in Patients with Unilateral Vocal Fold Paralysis. Otolaryngol Head Neck Surg, 2018. 159(1): p. 92–96. [DOI] [PubMed] [Google Scholar]
- 4.Zhou D, Jafri M, and Husain I, Identifying the Prevalence of Dysphagia among Patients Diagnosed with Unilateral Vocal Fold Immobility. Otolaryngol Head Neck Surg, 2019. 160(6): p. 955–964. [DOI] [PubMed] [Google Scholar]
- 5.Gould FD, et al. , The Physiologic Impact of Unilateral Recurrent Laryngeal Nerve (RLN) Lesion on Infant Oropharyngeal and Esophageal Performance. Dysphagia, 2015. 30(6): p. 714–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Schiedermayer B, et al. , Prevalence, incidence, and characteristics of dysphagia in those with unilateral vocal fold paralysis. Laryngoscope, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Ellerston JK, et al. , Quantitative Measures of Swallowing Deficits in Patients With Parkinson's Disease. Ann Otol Rhinol Laryngol, 2016. 125(5): p. 385–92. [DOI] [PubMed] [Google Scholar]
- 8.Kendall KA, et al. , Objective Measures of Swallowing Function Applied to the Dysphagia Population: A One Year Experience. Dysphagia, 2016. 31(4): p. 538–46. [DOI] [PubMed] [Google Scholar]
- 9.Kendall KA, Evaluation of airway protection: Quantitative timing measures versus penetration/aspiration score. Laryngoscope, 2017. 127(10): p. 2314–2318. [DOI] [PubMed] [Google Scholar]
- 10.Rosenbek JC, et al. , A penetration-aspiration scale. Dysphagia, 1996. 11(2): p. 93–8. [DOI] [PubMed] [Google Scholar]
- 11.Kendall KA, et al. , Timing of events in normal swallowing: a videofluoroscopic study. Dysphagia, 2000. 15(2): p. 74–83. [DOI] [PubMed] [Google Scholar]
- 12.Matson JL, Fodstad JC, and Boisjoli JA, Cutoff scores, norms and patterns of feeding problems for the Screening Tool of fEeding Problems (STEP) for adults with intellectual disabilities. Res Dev Disabil, 2008. 29(4): p. 363–72. [DOI] [PubMed] [Google Scholar]
- 13.Nayak VK, et al. , Patterns of swallowing failure following medialization in unilateral vocal fold immobility. Laryngoscope, 2002. 112(10): p. 1840–4. [DOI] [PubMed] [Google Scholar]
- 14.Ha JF, Unilateral vocal fold palsy & dysphagia: A review. Auris Nasus Larynx, 2020. [DOI] [PubMed] [Google Scholar]
- 15.Mu L and Sanders I, Neuromuscular specializations within human pharyngeal constrictor muscles. Ann Otol Rhinol Laryngol, 2007. 116(8): p. 604–17. [DOI] [PubMed] [Google Scholar]
- 16.Domer AS, Leonard R, and Belafsky PC, Pharyngeal weakness and upper esophageal sphincter opening in patients with unilateral vocal fold immobility. Laryngoscope, 2014. 124(10): p. 2371–4. [DOI] [PubMed] [Google Scholar]
- 17.Cates DJ, et al. , Effect of Vocal Fold Medialization on Dysphagia in Patients with Unilateral Vocal Fold Immobility. Otolaryngol Head Neck Surg, 2016. 155(3): p. 454–7. [DOI] [PubMed] [Google Scholar]
- 18.Anis MM and Memon Z, Injection medialization laryngoplasty improves dysphagia in patients with unilateral vocal fold immobility. World J Otorhinolaryngol Head Neck Surg, 2018. 4(2): p. 126–129. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Tibbetts KM, et al. , Etiology and long-term functional swallow outcomes in pediatric unilateral vocal fold immobility. Int J Pediatr Otorhinolaryngol, 2016. 88: p. 179–83. [DOI] [PubMed] [Google Scholar]
- 20.Kammer RE, et al. , High-resolution manometry and swallow outcomes after vocal fold injection medialization for unilateral vocal fold paralysis/paresis. Head Neck, 2019. 41(7): p. 2389–2397. [DOI] [PubMed] [Google Scholar]
- 21.DeLozier KR, et al. , Impact of recurrent laryngeal nerve lesion on oropharyngeal muscle activity and sensorimotor integration in an infant pig model. J Appl Physiol (1985), 2018. 125(1): p. 159–166. [DOI] [PMC free article] [PubMed] [Google Scholar]