Oropharyngeal Dysphagia Assessment in Behçet Disease Using EAT‐10 and Fiberoptic Endoscopic Evaluation of Swallowing Authors: Emre Demirel, Emel Tahir, Ahmet Kıvanç Cengiz Published: 2026 ABSTRACT ABSTRACTObjectiveTo provide a preliminary symptom‐based and standardized single‐consistency FEES screening assessment of swallowing in adults with Behçet disease using the Eating Assessment Tool‐10 (EAT‐10) and fiberoptic endoscopic evaluation of swallowing (FEES).MethodsThis single‐center cross‐sectional observational study with prospective enrollment was conducted between September 2022 and March 2024. Adult patients with Behçet disease completed the EAT‐10 and underwent FEES using an IDDSI level 3 moderately thick bolus (yellow pudding). FEES outcomes were scored using the Penetration‐Aspiration Scale (PAS), Yale Pharyngeal Residue Severity Rating Scale for vallecular and pyriform sinus residue (YPRSRS‐V and YPRSRS‐P), Functional Outcome Swallowing Scale (FOSS), and Functional Oral Intake Scale (FOIS).ResultsTwenty‐seven patients were included (20 women, 7 men). Mean age was 45.7 ± 8.0 years and mean disease duration was 10.4 ± 7.6 years. The mean EAT‐10 score was 4.0 ± 6.6, and 10 patients (37.0%) had an abnormal EAT‐10 score (≥ 3). With the tested IDDSI level 3 bolus, PAS was ≥ 2 in 1 patient (3.7%). Vallecular residue (YPRSRS‐V ≥ 2) was observed in 5 patients (18.5%) and pyriform sinus residue (YPRSRS‐p ≥ 2) in 2 patients (7.4%). FOSS was ≥ 2 in 1 patient (3.7%), and FOIS was ≤ 6 in 1 patient (3.7%).ConclusionIn Behçet disease, patient‐reported swallowing complaints were relatively common, whereas objective FEES evidence of penetration or aspiration was uncommon. Pharyngeal residue, particularly vallecular residue, was more frequent than airway invasion under this standardized bolus condition. EAT‐10‐based screening with selective FEES may help identify patients who may benefit from multidisciplinary swallowing evaluation.Level of Evidence4.Keywords: Behçet disease, dysphagia, EAT‐10, FEES, pharyngeal residue Introduction Behçet disease is a chronic, relapsing, multisystem inflammatory vasculitis classically characterized by recurrent oral aphthae, genital ulceration, and ocular involvement. Beyond this triad, musculoskeletal, neurologic, gastrointestinal, vascular, and dermatologic manifestations are well recognized. For otolaryngologists, recurrent inflammation of the oral cavity, pharynx, and larynx is clinically relevant because it may influence airway symptoms, voice, and swallowing [1, 2]. Dysphagia in Behçet disease is clinically plausible but insufficiently studied. Reported mechanisms include recurrent oral or pharyngeal ulceration, laryngeal ulceration, pharyngolaryngeal stenosis, pharyngeal webs, and cicatricial narrowing after repeated inflammation [3, 4, 5, 6, 7, 8, 9]. However, the published literature is dominated by case reports and small surgical series describing severe structural phenotypes rather than systematic swallowing assessment in unselected patients. Behçet disease may also involve the esophagus. Older reports and cohort studies have described ulcerative esophagitis, dysphagia associated with esophageal involvement, and manometric abnormalities, indicating that swallowing complaints may arise from both oropharyngeal and esophageal mechanisms [10, 11, 12, 13, 14, 15]. Available cohort data nevertheless suggest that esophageal involvement is uncommon overall, supporting selective rather than routine esophageal investigation in patients without suggestive symptoms [15]. Rare neurologic mechanisms further widen the differential diagnosis. Neuro‐Behçet may affect central swallowing pathways, and unusual dysphagic phenotypes such as palato‐pharyngo‐laryngeal rhythmic myoclonus have been reported [16]. Swallowing complaints in Behçet disease should therefore not be interpreted solely as evidence of overt structural disease. Systematic swallowing data in Behçet disease remain scarce. Symptom severity and instrumental abnormality are not equivalent: patients may report globus, throat clearing, effort, or sticking in the absence of aspiration, whereas endoscopic assessment defines swallowing safety and efficiency more directly. The Eating Assessment Tool‐10 (EAT‐10) is a validated symptom questionnaire, and Flexible Endoscopic Evaluation of Swallowing (FEES) allows direct assessment of penetration, aspiration, and post‐swallow residue using standardized outcome measures such as the Penetration‐Aspiration Scale (PAS), Yale Pharyngeal Residue Severity Rating Scale (YPRSRS), Functional Outcome Swallowing Scale (FOSS), and Functional Oral Intake Scale (FOIS) [17, 18, 19, 20, 21, 22, 23, 24]. This study therefore aimed to provide a preliminary symptom‐based and standardized single‐consistency FEES screening assessment of swallowing in adults with Behçet disease. Specifically, we examined the frequency of abnormal EAT‐10 scores, abnormal FEES outcomes under an IDDSI level 3 moderately thick bolus condition, and their associations with disease severity and disease duration. Methods This single‐center cross‐sectional observational study with prospective enrollment was conducted between September 2022 and March 2024 at a tertiary otolaryngology center. The protocol was approved by the local ethics committee (OMUKAEK‐2022/70), and all participants provided written informed consent. The study was conducted in accordance with the Declaration of Helsinki. Adult patients with Behçet disease followed in the rheumatology clinic were consecutively recruited. Exclusion criteria were pregnancy, prior head and neck surgery, prior head and neck radiotherapy, head and neck malignancy, active upper aerodigestive infection, and neurologic disorders known to affect swallowing. Demographic and clinical variables included age, sex, disease duration, treatment‐based disease severity, and major clinical manifestations of Behçet disease. These clinical manifestations included oral aphthous ulceration, genital ulceration, uveitis, and arthritis/arthralgia. Disease severity was categorized according to treatment requirements. Patients controlled with colchicine and/or corticosteroids were categorized as having mild disease. The moderate–severe disease category included patients requiring additional immunosuppressive or biologic treatment such as azathioprine, methotrexate, or tumor necrosis factor inhibitors. All participants completed the EAT‐10 before endoscopic assessment. This 10‐item questionnaire rates each item from 0 (no problem) to 4 (severe problem), yielding a total score from 0 to 40. An EAT‐10 score of 3 or greater was considered abnormal [17, 18]. FEES was performed transnasally in the seated position without topical anesthesia using a 3.7‐mm flexible endoscope (Karl Storz, Tuttlingen, Germany), in accordance with standard descriptions and training recommendations. Because this study was designed as a preliminary standardized outpatient screening assessment rather than a comprehensive multi‐consistency physiologic FEES protocol, all participants received the same bolus condition: three 5‐mL trials of an IDDSI level 3 moderately thick bolus (yellow pudding). This protocol was selected to reduce interpatient protocol variability and to provide a uniform assessment of airway invasion and pharyngeal residue in a Behçet disease cohort with limited prior instrumental swallowing data. Although comprehensive FEES protocols commonly use multiple consistencies and volumes, standardized single‐consistency bolus conditions can be useful in focused research settings when the aim is to evaluate swallowing under a controlled bolus condition [19, 20, 25]. Swallowing safety was rated with PAS. Post‐swallow residue was rated separately for the valleculae and pyriform sinuses using YPRSRS‐V and YPRSRS‐P. Functional swallowing severity and oral intake were documented with FOSS and FOIS, respectively. For descriptive analyses, PAS ≥ 2, YPRSRS ≥ 2, FOSS ≥ 2, and FOIS ≤ 6 were considered abnormal [21, 22, 23, 24]. Continuous variables were summarized as mean ± standard deviation or median (range), and categorical variables were summarized as counts and percentages. Comparisons between two independent groups were performed using the Mann–Whitney U test for continuous variables and the chi‐square or Fisher exact test for categorical variables where appropriate. Associations between continuous variables were evaluated using Spearman rank correlation. Exploratory analyses were also performed to examine the associations between major clinical manifestations of Behçet disease and swallowing measures. For these analyses, comparisons between symptom‐positive and symptom‐negative groups were performed using the Mann–Whitney U test. Because oral aphthous ulceration was present in nearly all patients and several swallowing measures showed limited variability, these analyses were interpreted cautiously as exploratory. Statistical significance was defined as p < 0.05. Results Twenty‐seven patients were included, comprising 20 women (74.1%) and 7 men (25.9%). Mean age was 45.70 ± 8.01 years, and mean disease duration was 10.44 ± 7.63 years. Twelve patients (44.4%) were classified as having mild disease and 15 (55.6%) as having moderate–severe disease. Baseline characteristics and overall swallowing outcomes are summarized in Table 1. Major clinical manifestations of Behçet disease included oral aphthous ulceration in 26 patients (96.3%), genital ulceration in 17 (63.0%), uveitis in 6 (22.2%), and arthritis/arthralgia in 13 (48.1%). TABLE 1Patient characteristics and swallowing outcomes.CharacteristicsValueSample size, n 27Gender, n (%)Female 20 (74.1%), Male 7 (25.9%)Age, years (mean ± SD; range)45.70 ± 8.01; 22–59Disease duration, years (mean ± SD; range)10.44 ± 7.63; 1–30Disease severity, n (%) Mild; 12 (44.4%) Moderate–severe; 15 (55.6%) EAT‐10 (mean ± SD; range)4.04 ± 6.58; 0–25EAT‐10 ≥ 3, n (%)10 (37.0%)PAS ≥ 2, n (%)1 (3.7%)YPRSRS‐V ≥ 2, n (%)5 (18.5%)YPRSRS‐P ≥ 2, n (%)2 (7.4%)FOSS ≥ 2, n (%)1 (3.7%)FOIS ≤ 6, n (%)1 (3.7%)Abbreviations: EAT‐10, Eating Assessment Tool – 10; FOIS, Functional Oral Intake Scale; FOSS, Functional Outcome Swallowing Scale; PAS, Penetration Aspiration Scale; YPRSRS‐V/P, Yale Pharyngeal Residue Severity Rating Scale Pyriform Sinus/Vallecula. Patient characteristics and swallowing outcomes. Mild; 12 (44.4%) Moderate–severe; 15 (55.6%) Abbreviations: EAT‐10, Eating Assessment Tool – 10; FOIS, Functional Oral Intake Scale; FOSS, Functional Outcome Swallowing Scale; PAS, Penetration Aspiration Scale; YPRSRS‐V/P, Yale Pharyngeal Residue Severity Rating Scale Pyriform Sinus/Vallecula. The mean EAT‐10 score was 4.04 ± 6.58 (range 0–25). Ten patients (37.0%) had an abnormal EAT‐10 score (Table 1). Among symptom items, 6 patients (22.2%) reported the need to clear the throat and 6 patients (22.2%) reported a lump‐in‐the‐throat sensation, whereas no patient reported liquid dysphagia or choking. Overall, 7 patients (25.9%) reported at least one swallowing‐related symptom (Table 2). TABLE 2Patient‐reported swallowing‐related symptoms.Symptom n %Dysphagia with liquid boluses00.0%Need to clear throat622.2%Feeling of a lump in the throat (globus)622.2%Choking00.0%At least one symptom725.9% Patient‐reported swallowing‐related symptoms. FEES findings obtained with three 5‐mL trials of the IDDSI level 3 moderately thick bolus showed that overt airway invasion was uncommon under the tested bolus condition. Only 1 patient (3.7%) had a PAS score of 2 or greater. Vallecular residue of 2 or greater was observed in 5 patients (18.5%), and pyriform sinus residue of 2 or greater was observed in 2 patients (7.4%). FOSS was 2 or greater in 1 patient (3.7%), and FOIS was 6 or less in 1 patient (3.7%) (Table 1). In the provided statistical analysis, EAT‐10 scores were higher in the moderate–severe disease group than in the mild disease group (5.40 ± 7.62 vs. 2.33 ± 4.77, p = 0.01). Patients with an abnormal EAT‐10 score had shorter disease duration than those with a normal EAT‐10 score (7.00 ± 5.93 vs. 12.47 ± 7.94 years, p = 0.03). Patients with vallecular residue had shorter disease duration than those without vallecular residue (7.40 ± 7.86 vs. 11.14 ± 7.59 years, p = 0.04) and higher EAT‐10 scores (6.40 ± 8.02 vs. 3.50 ± 6.31, p = 0.03). Disease duration correlated negatively with EAT‐10 score (Spearman r = −0.42, p = 0.02). Detailed associations are shown in Table 3. TABLE 3Associations between clinical variables and swallowing measures.ComparisonResult p‐valueMild vs. Moderate–severe (EAT‐10, mean ± SD)2.33 ± 4.77 vs. 5.40 ± 7.620.01* EAT‐10 < 3 vs. ≥ 3 (disease duration, years mean ± SD)12.47 ± 7.94 vs. 7.00 ± 5.930.03* YPRSRS‐V < 2 vs. ≥ 2 (disease duration, years mean ± SD)11.14 ± 7.59 vs. 7.40 ± 7.860.04* YPRSRS‐V < 2 vs. ≥ 2 (EAT‐10, mean ± SD)3.50 ± 6.31 vs. 6.40 ± 8.020.03* Disease duration vs. EAT‐10 r = −0.42 a 0.02 a a Spearman's rank correlation.*Mann–Whitney U‐test. Associations between clinical variables and swallowing measures. Spearman's rank correlation. Mann–Whitney U‐test. Exploratory analyses were also performed to evaluate the associations between major clinical manifestations of Behçet disease and swallowing measures. Oral aphthous ulceration was present in nearly all patients and was therefore not suitable for meaningful comparative analysis. No statistically significant associations were observed between genital ulceration or arthritis/arthralgia and EAT‐10, PAS, YPRSRS‐V, YPRSRS‐P, FOSS, or FOIS. Overall, these exploratory analyses did not demonstrate a robust association between major clinical manifestations of Behçet disease and subjective or objective swallowing outcomes (Table 4). TABLE 4Associations between major clinical manifestations of Behçet disease and swallowing measures.Genital ulceration absent vs presentUveitis absent vs presentArthritis/arthralgia absent vs presentEAT‐10, mean ± SD7.40 ± 9.59 vs. 2.06 ± 2.73 (p = 0.280)4.95 ± 7.17 vs. 0.83 ± 2.04 (p = 0.058)3.64 ± 6.65 vs. 4.46 ± 6.75 (p = 0.879)PAS, mean ± SD1.00 ± 0.00 vs. 1.06 ± 0.24 (p = 0.490)1.05 ± 0.22 vs. 1.00 ± 0.00 (p = 0.656)1.07 ± 0.27 vs. 1.00 ± 0.00 (p = 0.374)YPRSRS‐V, mean ± SD1.20 ± 0.42 vs. 1.18 ± 0.39 (p = 0.911)1.24 ± 0.44 vs. 1.00 ± 0.00 (p = 0.209)1.21 ± 0.43 vs. 1.15 ± 0.38 (p = 0.719)YPRSRS‐P, mean ± SD1.00 ± 0.00 vs. 1.12 ± 0.33 (p = 0.293)1.05 ± 0.22 vs. 1.17 ± 0.41 (p = 0.368)1.14 ± 0.36 vs. 1.00 ± 0.00 (p = 0.181)FOSS, mean ± SD1.10 ± 0.32 vs. 1.00 ± 0.00 (p = 0.220)1.05 ± 0.22 vs. 1.00 ± 0.00 (p = 0.656)1.00 ± 0.00 vs. 1.08 ± 0.28 (p = 0.335)FOIS, mean ± SD6.90 ± 0.32 vs. 7.00 ± 0.00 (p = 0.220)6.95 ± 0.22 vs. 7.00 ± 0.00 (p = 0.656)7.00 ± 0.00 vs. 6.92 ± 0.28 (p = 0.335) Note: Data are presented as mean ± SD. p‐values were calculated using the Mann–Whitney U‐test. Oral aphthous ulceration was not included in comparative analysis because it was present in nearly all patients. These analyses were considered exploratory because of the small sample size and the limited variability of several swallowing measures. Associations between major clinical manifestations of Behçet disease and swallowing measures. Note: Data are presented as mean ± SD. p‐values were calculated using the Mann–Whitney U‐test. Oral aphthous ulceration was not included in comparative analysis because it was present in nearly all patients. These analyses were considered exploratory because of the small sample size and the limited variability of several swallowing measures. Discussion This study combined symptom screening and a standardized single‐consistency FEES screening protocol in adults with Behçet disease. Three findings stand out. First, patient‐reported swallowing complaints were not rare. Second, penetration or aspiration was uncommon with the tested IDDSI level 3 moderately thick bolus. Third, under this standardized bolus condition, pharyngeal residue, especially vallecular residue, was more frequent than airway invasion. Together, these findings suggest that swallowing complaints in Behçet disease may more often reflect impaired efficiency, throat discomfort, or sensory burden than overt aspiration risk during this moderately thick bolus condition [17, 18, 19, 20, 21, 22, 23, 24, 25, 26]. This interpretation is consistent with the existing Behçet literature. ENT reviews and case reports show that recurrent mucosal inflammation can involve the oral cavity, pharynx, and larynx and, in selected patients, progress to laryngeal ulceration, pharyngolaryngeal stenosis, pharyngeal webs, or severe cicatricial narrowing requiring intervention [1, 2, 3, 4, 5, 6, 7, 8, 9]. These reports confirm that dysphagia in Behçet disease can be clinically important, but they largely represent advanced structural phenotypes. Our cohort extends the literature by showing that swallowing‐related symptoms may also occur in patients without overt destructive upper aerodigestive lesions and without frequent airway invasion during the tested IDDSI level 3 bolus condition. The symptom profile is also informative. Throat clearing and globus were more frequent than choking or liquid dysphagia. This pattern argues against aspiration as the dominant mechanism in most patients and instead supports a role for throat irritation, altered pharyngeal clearance, mucosal sensitivity, or cautious swallowing behavior. The association between vallecular residue and higher EAT‐10 scores supports this interpretation because even limited residue may cause repeated clearing, persistent throat awareness, or a sense of sticking [22, 27]. Esophageal involvement remains an important complementary explanation. Behçet‐related oesophagitis, dysphagia associated with esophageal ulceration, and manometric abnormalities have all been reported [10, 11, 12, 13, 14, 15]. Accordingly, a normal or near‐normal FEES does not exclude disease‐related swallowing dysfunction. In practice, these findings support a stepwise approach: FEES can define oropharyngeal residue and airway invasion, whereas persistent symptoms unexplained by FEES should prompt consideration of gastroenterologic evaluation when the clinical history suggests esophageal disease [15]. Rare neurologic mechanisms, including neuro‐Behçet‐related rhythmic myoclonus, should also be considered when symptoms appear disproportionate to local endoscopic findings [16]. Exploratory analyses based on major clinical manifestations of Behçet disease did not demonstrate clear associations with swallowing outcomes. This may indicate that swallowing complaints in Behçet disease are not determined simply by the dominant systemic manifestation profile, but may instead reflect a more complex interaction between local mucosal involvement, symptom perception, disease activity, and compensatory swallowing behavior. The absence of significant findings should be interpreted cautiously because of the small sample size, the low variability of several swallowing measures, and the near‐universal presence of oral aphthous ulceration in the cohort. The associations with disease severity and disease duration should also be interpreted cautiously. Higher EAT‐10 scores in the moderate–severe treatment group suggest that greater systemic disease burden may be accompanied by greater swallowing‐related symptom burden. By contrast, shorter disease duration was associated with abnormal EAT‐10 scores and with vallecular residue. These findings may reflect higher inflammatory activity earlier in the disease course, reduced adaptation in recently affected patients, treatment effects over time, or differences in symptom reporting. Because disease activity indices and longitudinal follow‐up were not included, these associations should be considered hypothesis‐generating rather than causal. The single‐consistency protocol is central to interpreting the present findings. Comprehensive clinical FEES protocols commonly include multiple consistencies and volumes; therefore, the present protocol cannot determine consistency‐dependent swallowing breakdown. IDDSI level 3 moderately thick boluses are generally less challenging for airway protection than thin liquids, while increased viscosity may increase pharyngeal residue. Thus, the low penetration/aspiration rate in this study should not be interpreted as evidence that airway invasion is uncommon across all consistencies in Behçet disease. Rather, our findings show that, under three standardized 5‐mL IDDSI level 3 trials, airway invasion was uncommon and residue was the more frequent endoscopic abnormality [19, 20, 25, 26]. The clinical implication is practical. EAT‐10 can be used as a first‐line screen during Behçet follow‐up, and abnormal scores or persistent throat symptoms can trigger otolaryngologic evaluation. FEES then helps distinguish major airway safety problems from residue‐dominant or symptom‐predominant presentations and may guide counseling, swallowing therapy referral, dietary adjustment, or further etiologic work‐up. This combined approach is useful in Behçet disease because dysphagia may be inflammatory, structural, sensory, esophageal, neurologic, or mixed in origin [17, 18, 19, 20, 27]. This study has limitations. The principal limitation is the single‐consistency FEES protocol. Only three 5‐mL trials of an IDDSI level 3 moderately thick bolus were administered; therefore, the study cannot evaluate consistency‐dependent swallowing physiology or determine whether thin liquids, larger volumes, sequential drinking, puree, or solid consistencies would have produced different PAS or residue findings. Because moderately thick boluses may reduce airway invasion compared with thin liquids while increasing residue, the low penetration/aspiration rate and the predominance of vallecular residue should be interpreted only in relation to the tested IDDSI level 3 condition [25, 26]. The sample was small and derived from a single tertiary center. The investigation focused on the oropharyngeal phase and did not systematically evaluate the esophagus. Other potential contributors to dysphagia, such as reflux, xerostomia, medication effects, deconditioning, or sarcopenia, were not systematically assessed. Future studies should include larger cohorts, standardized disease activity measures, multi‐consistency instrumental assessment, repeated follow‐up, complementary esophageal evaluation when clinically indicated, and standardized capture of anonymized FEES images or video segments. Conclusion Among adults with Behçet disease, dysphagia symptoms were relatively common, whereas FEES evidence of penetration or aspiration with three 5‐mL trials of an IDDSI level 3 moderately thick bolus was rare. Pharyngeal residue, especially vallecular residue, was the most frequent objective abnormality under the tested bolus condition. These findings suggest that swallowing complaints in Behçet disease may reflect reduced swallowing efficiency or throat symptom burden rather than major airway compromise during IDDSI level 3 moderately thick bolus trials. EAT‐10 screening with selective FEES may therefore be a practical clinical strategy in this population, but comprehensive multi‐consistency protocols are needed before drawing conclusions about swallowing physiology across bolus types. Funding The authors have nothing to report. Ethics Statement This study was approved by the local institutional ethics committee of Ondokuz Mayıs University (No: 2022/70). Consent Written informed consent was obtained from all participants. Conflicts of Interest The authors declare no conflicts of interest. Acknowledgments The authors have no acknowledgments to declare. Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request. References Associated Data Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request.