Prevalence, Incidence, and Characteristics of Dysphagia in Those With Unilateral Vocal Fold Paralysis

Abstract

Objectives/Hypothesis:

The purpose of this study was to investigate the impact of dysphagia definition on the incidence and overall prevalence of dysphagia in patients with unilateral vocal fold paralysis (UVP) stratified by etiology.

Study Design:

Retrospective medical chart review.

Methods:

Data was collected from the records of individuals diagnosed with UVP from 2013 to 2018, including patient demographics, dysphagia questionnaire total scores, clinical evaluation dysphagia symptoms, and instrumental swallow assessment outcomes. The annual incidence and overall prevalence of dysphagia were calculated by etiology as counts and percentages across five operational definitions of dysphagia.

Results:

A total of 415 individuals met inclusion criteria for the study. Annual prevalence estimates ranged from 19% to 55%, depending on the definition of dysphagia used. The highest prevalence of dysphagia occurred when defined by symptoms or signs identified by the clinician (55%). The lowest prevalence of dysphagia occurred using a definition of abnormal swallowing function documented during instrumental assessment (19%). Dysphagia questionnaire scores were more frequently abnormal in those with iatrogenic than idiopathic etiology of UVP (adjusted P = 0.014). Rate of instrumental assessment and documentation of aspiration was highest for central UVP etiology (33%). On average, pneumonia was rare (6%) irrespective of UVP etiology.

Conclusion:

Up to 55% of patients diagnosed with UVP complained of dysphagia, but only 21% had dysphagia symptoms severe enough to prompt instrumental assessment. Incidence and severity of dysphagia varied depending on UVP etiologic category as well as dysphagia definition. The etiology of UVP may impact dysphagia risk and severity in this population and warrants further investigation.

INTRODUCTION

Individuals with unilateral vocal fold paralysis (UVP) are believed to be at increased risk for dysphagia based on the assumption that impaired laryngeal closure during swallowing will lead to aspiration.^1–4 Epidemiologic research into the prevalence of dysphagia symptoms in patients with UVP has substantiated these concerns. However, these prior studies provided limited identification of risk factors and patient population characteristics most predictive of dysphagia to direct clinical assessment and treatment decisions.

Previous studies reporting on dysphagia prevalence in patients with UVP report a wide range of results (20%–72%).^1–15 This wide range of estimated dysphagia prevalence reflects varied study methodologies, particularly with regard to the definition of dysphagia used to characterize those with UVP and dysphagia. Because dysphagia is a symptom (patient-reported difficulty swallowing) as well as a sign (observed abnormality of swallowing function), confusion regarding the meaning of “dysphagia” assessed in these studies is likely. Although patients with UVP may experience a sensation of swallowing changes with a UVP, accurate determination of swallow function and aspiration risk requires direct observation of swallowing function in this patient population to link symptoms with the underlying pathophysiology. Additional confounding variables in previous studies included heterogeneity of patient populations and inconsistent stratification of study results by UVP etiology. Consequently, clinicians may currently anticipate more serious dysphagia than likely occurs with the majority of UVP patients.

To clarify the risk of swallowing impairment in patients with UVP, this study evaluated the prevalence of dysphagia within the same population across several operational definitions of dysphagia employed in previous studies on the UVP population. In addition, the impact of UVP etiology on dysphagia diagnosis was studied as a potential risk factor for clinically significant dysphagia in patients with UVP.

MATERIALS AND METHODS

Participants

This retrospective study was approved by the University of Utah Institutional Review Board (IRB) (protocol #00114169). The University of Utah Health electronic medical record warehouse was searched to identify individuals ≥18 years-old diagnosed with UVP seen from January 1, 2013, through December 31, 2018, in the University of 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 additional inclusion criteria upon review of their medical record. Reasons for exclusion included: superior laryngeal nerve (SLN) impairment without evidence of impaired vocal fold (VF) motion (N = 17), bilateral impairment of VF mobility (N = 14), endoscopy showing normal bilateral VF mobility (N = 12), patient under 18 years of age (N = 12), and UVP diagnosis that did not involve an otolaryngologist (N = 6). Thus, 415 individuals met inclusion criteria for this study (see Fig. 1).

Fig. 1.

Cohort flowchart. This flowchart displays the process by which the final inclusion/exclusion criteria were applied to the cohort studied.

Demographic Data

The medical records of all patients meeting inclusion criteria were reviewed by one of two clinically certified speech-language pathologists (SLP) (M.S. AND B.S.) trained in extracting medical record information relevant to this study by the senior SLP author (J.B-K.). In addition, ~10% of the 415 medical records (N = 42) were reviewed by both raters to assess interrater reliability. Medical record raters acquired demographic information including age, sex, duration of the disease, etiology, laterality of UVP (i.e. left or right vocal fold immobility), and the presence/absence of a diagnosis of pneumonia subsequent to onset of UVP. The patient’s first team-based clinic visit within the period of study was used for data collection. When the etiology of UVP was unclear from this visit, additional visits were reviewed to determine etiology. Etiology was extracted verbatim from the medical record before classification as one of six etiology categories (i.e., Iatrogenic, Idiopathic, cerebrovascular accident/traumatic brain injury (CVA/TBI), neurodegenerative disease, intubation injury, and other).

Patient Self-Report Data

The chief complaint for the clinic visit was recorded (e.g., dysphonia, dyspnea, dysphagia). When multiple reasons were listed, the clinic visit documentation was inspected to determine the order of priority in which the patient listed complaints. During both the scheduling of appointments in the clinic and clinical evaluation, patients are asked to indicate their highest priority complaint(s). They are then recorded in order of priority as indicated by the patient. The chief complaints were categorized as the primary (i.e., first symptom listed) or secondary reason (i.e., second symptom listed) for the visit, as indicated by the order of each listed within the medical record.

Additional information recorded from the medical record included patient self-assessment questionnaires administered based on the stated reason for the patient’s clinic visit. Two patient self-assessment dysphagia questionnaires have historically been used in the VDC. The Eating Assessment Tool-10 (EAT-10)¹⁶ was administered until 2016, after which time the Dysphagia Handicap Index (DHI)¹⁷ was adopted (see Appendix 1). A total of 83 patients completed the DHI, and 99 completed the EAT-10. The total score for each instrument was recorded. DHI total scores greater than 5,¹⁷ and EAT-10 total scores greater than 2¹⁶ were used to define the presence of dysphagia.

Dysphagia Symptom Description

The type, consistency (i.e., consistent versus intermittent), and severity of dysphagia symptoms discussed with the patient during the clinic visit were noted from medical record documentation and recorded as yes/no outcomes. These included patient report of the following: 1) intermittent coughing and choking during eating, 2) consistent coughing and choking every time the patient eats, 3) sensation of food stuck in the throat, and 4) food coming out of the nose. In addition, complaints of dysphonia were recorded.

Instrumental Assessment of Dysphagia

Documentation of completion of an instrumental assessment (i.e., flexible endoscopic evaluation of swallowing (FEES) or modified barium swallow (MBS)) based on the findings of the clinic visit was recorded. In addition, inspection of the FEES and MBS reports was completed to determine whether abnormal swallow function was identified within narrative impression summaries or indicated by deviation of quantitative measures from normative values of swallowing function (yes/no).^18,19 Abnormal swallow function was indicated by documentation stating the observation or measurement of impaired timing, pathway, or clearance of bolus materials from the upper aerodigestive pathway during oral preparation and swallowing. For those completing the MBS study, the Penetration-Aspiration Scale (PAS)²⁰ score was recorded, and ratings greater than 5 were considered positive for aspiration. During FEES assessment, identification of directly observed or inferred airway penetration or aspiration was also noted.

Operational Definitions of Dysphagia

Five different operational definitions of dysphagia were used to assess the annual incidence and overall prevalence of dysphagia in UVP. The current study’s operational definitions were developed to reflect variants of dysphagia studied in prior literature. Review of previous research on this topic identified a range of methods used to define the presence of dysphagia, including patient self-report of symptoms, medical record documentation of signs and symptoms during clinical examination, administration of dysphagia questionnaires (e.g., DHI or EAT-10), instrumental assessment (MBS/FEES), and PAS rates.⁴ Thus, this study’s definitions included:

1. Dysphagia symptoms were listed as the chief complaint for the clinic appointment. This definition only included patient-reported symptoms.^4,8

2. An abnormal total score on a dysphagia self-report questionnaire. For this study, we recorded the presence of dysphagia when the DHI total score > 5, or EAT-10 total score > 2.^3,5

3. The presence of signs or symptoms of dysphagia was documented by the clinician during the evaluation. This definition included both patient-reported symptoms and clinician-observed signs.^{2,4,5,8,11,13,14}

4. Instrumental assessment of swallowing (i.e., FEES or MBS) was recommended and completed based on findings from the clinical evaluation.^2–4,7,8

5. Instrumental assessment (FEES or MBS) was documented as showing abnormal swallow function.^{2–4,6–12,15}

In addition, the entire UVP population was assessed to determine the incidence and prevalence of dysphagia when any one of the operational definitions above was endorsed for the presence of dysphagia.

Statistical Analysis

All statistical analyses were conducted using R (version 3.4.4).²¹ Age, duration of disease, and dysphagia questionnaire total scores from the EAT-10 and DHI were summarized using mean, standard deviation (SD), or median and interquartile range (IQR). Categorical variables (i.e., presence/absence of pneumonia, side of UVP, male/female, etiology, dysphagia definition membership) were summarized using counts and percentages. Dysphagia definitions were also summarized by calendar year (2013–2018), and comparisons by etiology were conducted using Fisher exact tests. Post hoc pair-wise comparisons were conducted on dysphagia definitions found to be statistically significant across etiology groups. Multiple post hoc comparisons were adjusted by the Bonferroni method (multiplying the P values by the number of pair-wise tests). Statistical significance was assessed at the 0.05 level, and all tests were two-tailed.

Interrater reliability on medical record data extraction was determined on ~10% of the patients. Both raters independently completed data collection on 42 patients, blind to each other’s data extraction. The Cohen’s Kappa statistic (k)²² and 95% confidence interval (CI) were calculated to assess levels of agreement between the two raters for each of the five definitions of dysphagia. The two raters achieved “strong” levels of agreement²³ with estimated k values (95% CI), as follows for each operational dysphagia definition as numbered above: 1) 0.83 (95% CI 0.61–1.00), 2) 0.86 (95% CI 0.66–1.00) for DHI > 5 or EAT-10 > 2), 3) 0.95 (95% CI 0.84–1.00), 4) 0.85 (95% CI 0.63–1.00), and 5) 0.92 (95% CI 0.76–1.00).

RESULTS

The average age (SD) of the entire population was 58.6 (16.3), ranging from 18 to 93 years with 41% (170 of 415) males and 59% (245 of 415) females (see Table I). The average duration of UVP was 3.6 (± 7) years ranging from less than 1 year to 54 years. The majority of patients exhibited a left-sided (60%, 249 of 415) compared to a right-sided (40%, 166 of 415) paralysis (see Table I).

TABLE I.

Summary of Patient Demographic Characteristics Stratified by Etiology.

Demographic Variable	Total N = 415	Iatrogenic N = 206	Idiopathic N = 107	Tumor N = 31	Intubation Injury N = 20	CVA/TBI N = 19	Neurodegenerative Disease N = 4	Other N = 28
Males	170 (41%)	84 (41%)	36 (34%)	15 (48%)	6 (30%)	8 (42%)	3 (75%)	18 (64%)
Females	245 (59%)	122 (59%)	71 (66%)	16 (52%)	14 (70%)	11 (68%)	1 (25%)	10 (36%)
Age-mean (SD)	58.6 (16.3)	57 (15.8)	61.3 (16.3)	59.4 (15.4)	53.1 (17.6)	67.4 (16.7)	72.8 (11.4)	54.1 (16.5)
Median (IQR)	60 (47, 70.5)	59 (46, 68)	63 (52, 73)	62 (53.5, 69.5)	54.5 (40.5, 65.2)	69 (55, 79)	75 (69, 78.8)	58.5 (45, 67.2)
Range	(18, 93)	(18, 88)	(24, 93)	(23, 81)	(20, 84)	(29, 90)	(57, 84)	(21, 75)
Duration of disease (yrs) mean (SD)	3.6 (7)	4.4 (7.6)	2 (2.6)	2.9 (9.4)	2.8 (6.3)	2.3 (2.2)	0.4 (0.4)	6.1 (11.8)
Median (IQR)	0.8 (0.2, 3)	0.8 (0.2, 5)	0.8 (0.2, 2.8)	0.5 (0.2, 1)	0.5 (0.2, 2.2)	2 (0.5, 3.5)	0.2 (0.2, 0.5)	2 (0.5, 3.2)
Range	(0, 54)	(0, 41)	(0, 13)	(0.1, 52)	(0, 28)	(0.2, 9)	(0.2, 1)	(0.1, 54)
Left-sided UVP	249 (60%)	116 (56%)	65 (61%)	23 (74%)	14 (70%)	11 (58%)	1 (25%)	19 (68%)
Right-sided UVP	166 (40%)	90 (44%)	42 (39%)	8 (26%)	6 (30%)	8 (42%)	3 (75%)	9 (32%)
Pneumonia post-onset of UVP	23 (6%)	11 (5%)	6 (6%)	2 (6%)	1 (5%)	0 (0%)	0 (0%)	3 (11%)
DHI completed	83 (20%)	46 (22%)	22 (21%)	2 (6%)	1 (5%)	4 (21%)	1 (25%)	7 (25%)
DHI total score-mean (SD)†	43.3 (21.2)	43.5 (19.1)	37 (20.9)	96 (2.8)	38 (NA)*	58.5 (15.8)	54 (NA)*	37.1 (21.2)
Median (IQR)	40 (28, 56)	40 (30, 55)	37 (22.5, 49.5)	96 (95, 97)	38 (38, 38)	56 (52.5, 62)	54 (54, 54)	38 (25, 47)
Range	(6, 100)	(16, 100)	(6, 90)	(94, 98)	(38, 38)	(42, 80)	(54, 54)	(6, 72)
EAT-10 completed	99 (24%)	53 (26%)	13 (12%)	15 (48%)	5 (25%)	4 (21%)	1 (25%)	8 (29%)
EAT 10-mean (SD)‡	17.6 (11.2)	20 (11.2)	8.5 (9.4)	20.8 (11.1)	12.2 (6.5)	16.8 (10.8)	10 (NA)*	15.5 (9.7)
Median (IQR)	16 (9.5, 25)	19 (10, 29)	6 (2, 12)	22 (12, 29)	15 (6, 15)	14.5 (8, 23.2)	10 (10, 10)	14.5 (9.8, 24.5)
Range	(0, 40)	(0, 40)	(0, 32)	(4, 40)	(5, 20)	(8, 30)	(10, 10)	(3, 27)
FEES or MBS completed based on clinical findings	89 (21%)	48 (23%)	15 (14%)	7 (23%)	3 (15%)	7 (37%)	1 (25%)	8 (29%)
MBS completed	62 (15%)	32 (16%)	13 (12%)	4 (13%)	2 (10%)	6 (32%)	1 (25%)	4 (14%)
PAS on MBS > 5§	14 (4%)	9 (28%)	1 (8%)	1 (25%)	0 (0%)	2 (33%)	0 (0%)	1 (25%)
PAS on MBS = 7§	3 (1%)	2 (1%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	1 (4%)
PAS on MBS = 8§	8 (2%)	5 (2%)	1 (1%)	0 (0%)	0 (0%)	2 (11%)	0 (0%)	0 (0%)
PAS mean (SD), median and range	3.1 (2.6) 2(1 – 8)	3.3 (2.7) 2(1–8)	2.2 (2) 2(1–8)	3.2 (2.6) 3(1–6)	3 (1.4) 3(2–4)	4.2 (3.3) 3.5(1–8)	1 (NA) 1(1–1)	3.2 (2.5) 2(2–7)

^*Mean and SD are not available from a single observation (N = 1).

^†Based on a subset of N = 83 out of 415 who had completed DHI.

^‡Based on a subset of N = 99 out of 415 who had completed EAT-10.

^§Based on a subset of N = 62 out of 415 who had completed MBS.

DHI = Dysphagia Handicap Index; EAT = Eating Assessment Tool; PAS = Penetration-Aspiration Scale; SD = standard deviation; UVP = unilateral vocal fold paralysis; FEES = Flexible Endoscopic Evaluation of Swallowing; MBS = Modified Barium Swallow; CVA = Cerebrovascular Accident; TBI = Traumatic Brain Injury.

Incidence and Prevalence of Dysphagia

Table II shows the annual incidence and overall prevalence of dysphagia in the population of patients diagnosed with UVP (N = 415) for each operational definition studied. The average number of new UVP cases diagnosed with dysphagia annually ranged from 13 to 43 individuals, depending on the definition of dysphagia applied. Overall prevalence estimates across dysphagia definitions ranged from 19% to 55%. Forty-eight percent of patients (198 of 415) reported dysphagia as one of the reasons for their clinic visit. Of those, 92% (182 of 198) completed a dysphagia questionnaire (see Table I), and 96% (174 of 182) of these patients met criteria for dysphagia based on the results of the DHI or EAT-10. Thus, 42% (174 of 415) of individuals in this study had DHI or EAT-10 total scores indicative of dysphagia (see Table II). However, dysphagia was most likely to be identified when defined as “the presence of dysphagia signs or symptoms were noted during the clinic visit.” That is, the clinician documented the patient’s report regarding swallowing problems during the case intake or documented signs of dysphagia (55%, 230 of 415). Dysphagia had the lowest prevalence when defined as completion of an instrumental assessment (MBS or FEES) based on the findings of the clinic visit (21%, 89 of 415) and/or documentation of abnormal swallow function on the instrumental assessment (19%, 77 of 415). Overall, 62% (257 of 415) of the population was documented with one or more of the dysphagia operational definitions.

TABLE II.

Incidence and Prevalence of Dysphagia by Operational Definition.

Operational Definition of Dysphagia	2013 N = 89	2014 N = 61	2015 N = 67	2016 N = 48	2017 N = 79	2018 N = 71	Avg No. New Cases per Year	Overall Prevalence N = 415
Dysphagia symptoms were listed as chief complaint for the clinic appointment	50 (56%)	31 (51%)	37 (55%)	21 (44%)	32 (41%)	27 (38%)	33	198 (48%)
Dysphagia indicated by DHI >5, or EAT-10 > 2*	28 (31%)	26 (43%)	30 (45%)	20 (42%)	33 (42%)	37 (52%)	29	174 (42%)
Presence of dysphagia signs or symptoms noted during the clinic visit	58 (65%)	39 (64%)	44 (66%)	24 (50%)	35 (44%)	30 (42%)	38	230 (55%)
FEES/MBS assessment was completed based on clinical evaluation	19 (21%)	16 (26%)	18 (27%)	9 (19%)	15 (19%)	12 (17%)	15	89 (21%)
Completed instrumental assessment documentation identified abnormal swallow function†	18 (20%)	12 (20%)	16 (24%)	9 (19%)	12 (15%)	10 (14%)	13	77 (19%)
Total population identified by the presence of 1 or more of the above definitions	59 (66%)	41 (67%)	46 (69%)	28 (58%)	41 (52%)	42 (59%)	43	257 (62%)

^*Based on a subset of N = 182 out of 415 who had completed either DHI or EAT-10.

^†Based on a subset of N = 89 out of 415 who had completed MBS or FEES.

DHI = Dysphagia Handicap Index; EAT = Eating Assessment Tool; FEES = Flexible Endoscopic Evaluation of Swallowing; MBS = Modified Barium Swallow.

Patient Self-Report Outcomes

Table III shows the overall frequency of symptoms reported as the chief complaint for the clinic appointment and their ranking as primary versus secondary. Dysphonia was the most frequently reported reason for the clinic visit among patients (92%, 381 of 415) and was the highest-ranking as the primary reason (83%, 345/415). Dysphagia (48%, 198 of 415) was the second most reported reason for the clinic visit but was the primary reason in only 8% (34 of 415) of patients. Instead, 33% (138 of 415) of patients reported dysphagia as a secondary reason for their clinic visit. Other less frequently reported symptoms included dyspnea (21%), cough (14%), throat clearing (4%), and globus sensation (3%).

TABLE III.

Most Common Symptoms Documented as the Chief Complaint for the Clinic Appointment.

Overall Symptom	Total Frequency (N = 415)	Primary Symptom (N = 415)	Secondary Symptom (N = 415)
Dysphonia	381 (92%)	345 (83%)	24 (6%)
Dysphagia	198 (48%)	34 (8%)	138 (33%)
Dyspnea	89 (21%)	14 (3%)	42 (10%)
Cough	57 (14%)	13 (3%)	27 (7%)
Throat clearing	18 (4%)	1 (.24%)	10 (2%)
Globus sensation	11 (3%)	2 (.5%)	5 (1%)

UVP Etiology

Table I displays the demographic characteristics of the patient population stratified by the etiology of UVP. DHI, EAT-10, and MBS values were calculated on a subset of 83, 89, and 62 subjects out of a total of 415, respectively. Iatrogenic injury was determined to be the most common etiology of UVP (50%, 206 of 415). The second most common etiology of UVP was idiopathic (26%, 107 of 415). Other etiologic categories included tumor, intubation injury, CVA/TBI, and neurodegenerative disease. The “other” category (7%, 28 of 415) was assigned when the etiology of UVP was unclear or represented only one individual for that diagnosis. Twice as many females compared to males were diagnosed with UVP due to an idiopathic etiology or intubation injury. There was nearly equal representation of males and females in the iatrogenic and tumor etiology categories.

Pneumonia was documented post-onset of UVP in 6% (23 of 415) of patients. Pneumonia rate was similar for the idiopathic, iatrogenic, tumor, and intubation injury categories (about 6%), whereas no pneumonia was reported for the TBI/CVA or neurodegenerative categories. The highest prevalence of pneumonia (11%, 3 of 28) occurred for the “other” category.

The presence of dysphagia as defined by DHI total score > 5 or EAT-10 total score > 2 was distributed differently across etiologic groups (P = 0.007) and showed a statistically significant difference between idiopathic and iatrogenic etiologies (adjusted P = 0.014). Based on this definition, dysphagia was less frequent among patients with UVP due to CVA/TBI, iatrogenic injury, idiopathic etiology, and intubation (30%–47%). Dysphagia was more frequently present when UVP was due to tumor or “other” etiology (55% and 54%, respectively). Among the four subjects with a neurodegenerative disease, the frequency of dysphagia was found to be 50%.

Dysphagia Symptoms

Table IV shows the frequency of clinical symptoms of dysphagia stratified by etiology, excluding the “other” category, leaving 378 patients. Dysphonia was the most frequently identified symptom across all etiologies of UVP (92%, 354 of 387). Intermittent coughing/choking during eating was identified in 29% to 35% of those with iatrogenic, tumor, and intubation injury with smaller frequency in the idiopathic (16%, 17 of 109) and CVA/TBI (21%, 4 of 18) etiologies. The sensation of food being stuck in the throat was identified in less than one-third of patients irrespective of etiologic category. Consistent coughing/choking during eating and food coming out of the nose occurred in less than 10% of any etiologic category. Interestingly, none of the clinical signs and symptoms were represented in the four patients classified with a neurodegenerative disease with the exception of dysphonia.

TABLE IV.

Frequency of Dysphagia Symptoms Reported During the Clinic Visit.

Dysphagia Symptoms During Clinical Evaluation	Iatrogenic (N = 205)	Idiopathic (N = 109)	Tumor (N = 31)	Intubation Injury (N = 20)	CVA/TBI (N = 18)	Neurodegenerative Disease (N = 4)
Intermittent coughing/choking during eating	72 (35%)	17 (16%)	9 (29%)	7 (35%)	4 (21%)	0 (0%)
Consistent coughing/choking during eating	16 (8%)	6 (6%)	3 (10%)	0 (0%)	2 (11%)	0 (0%)
Sensation of food stuck in throat	56 (27%)	20 (19%)	9 (29%)	1 (5%)	5 (26%)	0 (0%)
Food coming out of the nose	4 (2%)	1 (1%)	1 (3%)	0 (0%)	0 (0%)	0 (0%)
Dysphonia	190 (92%)	97 (91%)	29 (94%)	18 (90%)	17 (89%)	3 (75%)

Instrumental Assessment

A total of 62 individuals (15%, 62 of 415) completed an MBS evaluation. As shown in Table I, a PAS score > 5, indicating aspiration, was documented in 23% (14 of 62) of those completing an MBS study and 3% (14 of 415) of the total UVP population. This result is consistent with prior literature reporting rates between 23.4%¹¹ and 29%⁹ in patients undergoing instrumental evaluation. Three patients were rated with a PAS of 7, and eight patients were rated as 8. Thus, the majority of those with UVP rated with aspiration exhibited silent aspiration indicative of both SLN and recurrent laryngeal nerve (RLN) involvement. Etiologies associated with silent aspiration occurred across the CVA/TBI, iatrogenic, and idiopathic categories.

DISCUSSION

The results of this study demonstrated that the annual number of new cases of dysphagia among the UVP population differed by definition of dysphagia (i.e., 13–43), as did overall prevalence estimates (19%–55%). The highest estimates of dysphagia prevalence resulted from applying the operational definition “the presence of signs or symptoms of dysphagia noted during the clinic visit” (55%), followed by “dysphagia symptoms were listed as the chief complaint for the clinic appointment” (48%), and “a DHI total score > 5, or EAT-10 total score > 2” (42%). Lower prevalence estimates resulted from operational definitions based on “FEES/MBS assessment was recommended and completed based on clinical evaluation findings” (21%) and “instrumental assessment was documented as showing abnormal swallow function” (19%). The majority of patients who underwent instrumental assessment were found to have abnormal swallow function. The range of prevalence estimates in this study were comparable to prior studies investigating the prevalence of dysphagia in UVP using a similar range of dysphagia operational definitions (20%–72%).^4,5,10,13,24 By applying these various definitions to a single patient population, this study demonstrated the impact of dysphagia operational definition on the results of dysphagia prevalence estimates in those with UVP.

Overall, only 21% of the UVP population in this study exhibited dysphagia signs or symptoms during their clinical evaluation severe enough to warrant instrumental assessment. Of these, abnormal findings were reported in 87%; however, aspiration was observed in 23% of those undergoing MBS. This falls in the lower estimates reported in prior studies that defined dysphagia by the observation of laryngeal penetration and aspiration (PAS ≥3) during instrumental assessment (20%–50%).^1,3,6,11 The current study defined aspiration rates using PAS scores greater than 5 (i.e., observation of aspiration) rather than inferring aspiration in those with laryngeal penetration that was not ejected from the laryngeal vestibule during the swallow (PAS = 3–5). Had the current study used the same criteria as prior studies applied, aspiration rates would be elevated given that the average PAS score was 3.1. However, the median PAS score (PAS = 2) supports that the majority of those with UVP did not exhibit laryngeal penetration (PAS = 3–5) or aspiration (PAS > 5).

Interestingly, of the 14 individuals identified to aspirate in the current study, three were rated with a PAS score of 7 and eight were rated with a score of 8 indicative of silent aspiration. This finding suggests that the majority of aspirators exhibited silent aspiration indicative of impaired SLN and RLN function. The etiology representing the largest number of individuals identified with aspiration were those with an iatrogenic etiology of UVP (N = 9) or CVA/TBI (N = 2). These findings partially overlap with one prior study that evaluated aspiration by etiology using categories similar to the present study.¹¹ Bhattacharyya et al. found higher median PAS scores in patients diagnosed with “intracranial” etiology (i.e., stroke) (PAS = 4) than in other etiology groups (PAS = 2.0–2.5).¹¹ However, their iatrogenic etiologies were split across a “chest cause” (N = 45) and “neck cause” (N = 5), both showing a median PAS score of 2.0. The current study showed similar median PAS scores across etiologies; however, the majority of those observed to aspirate were from both iatrogenic and CVA/TBI categories.

In the present study, the etiologic categories most often associated with further instrumental assessment included CVA/TBI, neurodegenerative disease, iatrogenic, tumor, and “other.” Patients with idiopathic and intubation injury etiologies were referred for instrumental assessment less frequently. Theoretically, individuals with isolated injury to the RLN are capable of achieving airway safety due to the compensatory contributions of the supraglottic larynx, the unimpaired side of the larynx, normal hyolaryngeal excursion, and pharyngeal constriction and shortening.^10,25 Support for this compensation theory was found in a piglet model in which dynamic reorganization of the pharyngeal swallow and adequate airway protection were demonstrated subsequent to isolated RLN damage.²⁶ In addition, Dworkin and Treadway found that 70% of a retrospective cohort of adults diagnosed with idiopathic UVP (N = 30) did not report any form of dysphagia.²⁵ Only one individual (3%) complained of severe dysphagia characterized by consistent chronic coughing and gagging on liquids and solid food items.²⁵ Similarly, in this study, 33 (31%) of patients with an idiopathic UVP complained of dysphagia; only 15 (14%) underwent instrumental assessment for dysphagia; and one patient was found to aspirate. These findings suggest that, from a clinical perspective, idiopathic etiologies may present a lower risk of dysphagia and aspiration than injuries affecting multiple vagal branches, multiple peripheral nerves, or the central nervous system. To this point, the sudden change in swallowing function that occurs with an iatrogenic injury involving multiple peripheral nerves and aerodigestive structures may preclude swallowing adaptation resulting in more dysphagia signs and symptoms during early postoperative intervals.

Although idiopathic and iatrogenic etiologies for UVP might be expected to impact swallowing function similarly, this study found differences between these categories. That is, patients with an iatrogenic etiology more frequently complained of dysphagia, had higher DHI or EAT-10 scores, exhibited increased referral for instrumental assessment of dysphagia, and demonstrated higher rates of aspiration during the MBS study than in those with an idiopathic etiology. However, additional comparison of these two patient populations is needed using quantitative analysis of videofluoroscopic evaluation to better define swallowing physiology similarities and differences.

Prior studies using temporospatial measures of swallowing function to assess swallowing physiology in those with UVP found a greater degree of swallowing dysfunction in patients suffering from high vagal and additional cranial nerve involvement beyond the RLN vagal branch.^{7,8,10–12,14} These patients showed prolonged total pharyngeal transit times, reduced laryngeal elevation, and pharyngeal residue in the oropharynx.^7,9–12,15 CNS and brainstem level injuries are more likely to result in bilateral pharyngeal and laryngeal impairment in addition to vocal fold paralysis.^6,7 The findings of this study were consistent with prior findings that the CVA/TBI etiology was associated with the highest dysphagia questionnaire total scores, more likely to be referred for instrumental assessment, and had the highest rate of aspiration during MBS assessment.

Future investigation of swallowing physiology in patients with UVP is needed to confirm the results of this study. Instrumental assessment using MBS recordings would enable observation and measurement of oropharyngeal swallowing physiology. Future studies that systematically evaluate the physiology of individuals with impaired swallow function associated with UVP across varied etiologies would elucidate specific risk factors in this patient population. In addition, longitudinal investigation is needed to define the impact of aspiration seen during the FEES or MBS on the long-term risk of pneumonia in those with UVP across various etiologies.

CONCLUSION

This study found that the operational definition of dysphagia impacts the results of prevalence studies of dysphagia in the UVP population. Whereas the majority of patients diagnosed with UVP in a university outpatient setting complained of dysphagia, only a small proportion had dysphagia symptoms severe enough to prompt instrumental assessment. Although most of those undergoing instrumental assessment exhibited abnormal swallowing function, aspiration was rare. Further, rates of pneumonia were low in the overall UVP population. Variability in the incidence of dysphagia by UVP etiologic category warrants further evaluation because this study found the highest rates of dysphagia in patients with neurodegenerative and central etiologies, intermediate rates in patients with tumor and iatrogenic etiology, and the lowest rates in those with idiopathic vocal fold paralysis. Thus, etiology may be a factor for consideration when determining dysphagia severity and aspiration risk in those with UVP.

APPENDIX

Eating Assessment Tool (EAT-10)¹⁶

The patient rates each item from 0 (No problem) to 4 (Severe problem). Total scores range from 0-40, with 0 being least severe and 40 being most severe.

1. My swallowing problem has caused me to lose weight.

2. My swallowing problem interferes with my ability to go out for meals.

3. Swallowing liquids takes extra effort.

4. Swallowing solids takes extra effort.

5. Swallowing pills takes extra effort.

6. Swallowing is painful.

7. The pleasure of eating is affected by my swallowing.

8. When I swallow food sticks in my throat.

9. I cough when I eat.

10. Swallowing is stressful.

Dysphagia Handicap Index (DHI)¹⁷

The patient rates each item as Never (0), Sometimes (2), or Always (4). Total scores range from 0-100, and subscale scores range from 0-34 (Physical), 0-34 (Functional), and 0-28 (Emotional), with 0 being least severe, and higher scores indicating greater severity. An additional question asks the patient to rate their swallowing difficulty on a 1-7 scale.

1. 1P. I cough when I drink liquids.

2. 2P. I cough when I eat solid food.

3. 3P. My mouth is dry.

4. 4P. I need to drink fluids to wash food down.

5. 5P. I’ve lost weight because of my swallowing problem.

6. 1F. I avoid some foods because of my swallowing problem.

7. 2F. I have changed the way I swallow to make it easier to eat.

8. 1E. I’m embarrassed to eat in public.

9. 3F. It takes me longer to eat a meal than it used to.

10. 4F. I eat smaller meals more often due to my swallowing problem.

11. 6P. I have to swallow again before food will go down.

12. 2E. I feel depressed because I can’t eat what I want.

13. 3E. I don’t enjoy eating as much as I used to.

14. 5F. I don’t socialize as much due to my swallowing problem.

15. 6F. I avoid eating because of my swallowing problem.

16. 7F. I eat less because of my swallowing problem.

17. 4E. I am nervous because of my swallowing problem.

18. 5E. I feel handicapped because of my swallowing problem.

19. 6E. I get angry at myself because of my swallowing problem.

20. 7P. I choke when I take my medication.

21. 7E. I’m afraid that I’ll choke and stop breathing because of my swallowing problem.

22. 8F. I must eat another way (e.g., feeding tube) because of my swallowing problem.

23. 9F. I’ve changed my diet due to my swallowing problem.

24. 8P. I feel a strangling sensation when I swallow.

25. 9P. I cough up food after I swallow.

Please circle the number that matches the severity of your swallowing difficulty (1 = no difficulty at all; 4 = somewhat of a problem; 7 = the worst problem you could have)