Asymptomatic Dysphagia and Aspiration in Patients with Idiopathic Bronchiectasis

This is a prospective cohort study conducted between May 2019 and July 2022 at the Tel-Aviv Sourasky Medical Center, a tertiary medical center in Tel-Aviv, Israel. Consecutive adult patients (age ≥ 18 years) with bronchiectasis were recruited during their visit to the Institute of Pulmonary Medicine. Only patients diagnosed with IB after a targeted work-up were recruited [7]. Considering the study aim, we decided to include only IB patients, which do not have other confounding etiologies to explain the presence of bronchiectasis. The workup included a chest computed tomography (CT) scan to confirm the diagnosis of bronchiectasis, a detailed medical history to exclude relevant and possible causes of bronchiectasis (such as rheumatoid arthritis, Sjögren’s syndrome, chronic obstructive pulmonary disease [COPD], allergic bronchopulmonary aspergillosis (ABPA), gastroesophageal reflux disease, stroke and inflammatory bowel disease), and laboratory tests (including a full blood cell count, serum total IgE, serum immunoglobulin G subclasses, immunoglobulin A, immunoglobulin M and sputum culture). Patients with prior stroke underwent an assessment by a senior neurologist and were included in the study only if there were no detectable neurologic deficits. Patients with COPD were excluded if the bronchiectasis diagnosis was in-adjacent or after the diagnosis of COPD. Patients who had a history of dysphagia or any reason for a prior dysphagia workup, were excluded. To verify these exclusion criteria, patients were interviewed prior to study inclusion, and their past medical records from prior hospitalizations and clinic visits were then reviewed. Patients with cognitive decline were excluded from study entry. Only the findings of patients who met the inclusion criteria, consented to participate in the study and completed the study protocol are included in this report.

The study was reviewed and approved by the Institutional Review Board, (TLV-0582-18). The study is reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.

After providing written informed consent, each participant filled in the Eating Assessment Tool (EAT-10) questionnaire, which is a 10-item self-assessment screening scale aimed at identifying individuals at high risk for swallowing disorders [8]. Each parameter is graded on a scale from 0 (no problem) to 4 (severe problem). The total EAT-10 score was the sum of the scores of all 10 parameters. A cutoff score of 3 was chosen in our study for the detection of dysphagia, in accordance with a meta-analysis that utilized the diagnostic accuracy of this cutoff [9]. Each participant’s bronchiectasis severity index, based upon the information derived from their medical interview and medical records, was determined on a scale of 0–7 (0–2 = mild; 3–4 = moderate: 5–7 = severe) using the FACED score [10]. The FACED score is based on five parameters to determine bronchiectasis severity: (1) forced expiratory volume in the first second (FEV1) value (≥ 50% = 0 points, < 50% = 2 points), (2) age (< 70 years = 0 points, ≥ 70 years = 2 points), (3) chronic lung colonization by Pseudomonas (1 point), (4) extent of lung involvement (1–2 lobes affected = 0 points, > 2 lobes affected = 1 point), and (5) dyspnea severity based on the modified Medical Research Council scale (mMRC) (0–2 = 0 points, 3–4 = 1 point). We extracted the spirometry results of included patients, with the nearest results to their inclusion chosen for analysis.

All included participants underwent a fiberoptic endoscopic evaluation of swallowing (FEES) test at the Voice and Swallowing Clinic. The evaluation was performed according to the accepted protocol by an experienced otolaryngologist and a speech pathologist specializing in dysphagia [11]. The FEES test was performed with the patient seated. A flexible fiberoptic laryngoscope connected to a camera and a digital recording video system was inserted through the nose to the hypopharynx to allow adequate visualization of the larynx and pharynx. The patients were instructed to swallow three food consistencies (pureed in the form of two spoons of apple sauce containing food coloring, solid in the form of two bites of a biscuit and liquid in the form of two spoons of food colored milk and, if there was no suspicion of aspiration, further two separate sips from a cup and one uninterrupted sip under visualization. The FEES test function parameters included: saliva stasis, abnormal sensation, spillage before swallowing, abnormal bolus location before swallowing, abnormal time to swallow reflex initiation, penetration (entrance of the bolus to the larynx), aspiration (entrance of the bolus to trachea), bolus residues, spillage after swallowing and ineffective cough reflux, hypopharyngeal reflux and oral motor function. We used the “touch method” to assess laryngeal sensation [12]. If no laryngeal adductor reflex (LAR), swallowing, or coughing was induced by touching the epiglottis or arytenoids with the tip of the fiberoptic laryngoscope, laryngeal sensation impairment was assumed.

We used two methods to quantify the FEES abnormalities. Each one of the 12 FEES test function parameters received a score of 0 (no abnormality) or 1 (abnormality). The sum of all abnormalities was calculated to yield a FEES test score between 0 and 12 for each patient. We also adopted the penetration aspiration score (PAS) [13] (used to grade dysphagia on videofluoroscopic swallow studies) to quantify the FEES test scores. The score is an 8-point ordinal scale, with 1 representing the least and 8 representing the highest or most severe score. A PAS score was calculated for each patient at the completion of each FEES test. The PAS score we chose for each patient for statistical analysis was the highest score of the entire FEES, independent of the viscosity of the food bolus or the phase of swallowing which it scored [14]. Due to the limitation of FEES, we did not score the “white-out” phase. PAS scores of 6 to 8 indicated the presence of aspiration (entrance of the bolus to the trachea), and PAS scores of 2 to 5 indicated penetration to the larynx.

A receiver operating characteristic curve was generated for the estimation of The EAT-10 score’s predictive ability for aspiration (a PAS score of ≥ 6). This allowed for the calculation of area under the curve (AUC), and for determining the optimal cutoff point for the EAT-10 score. Sensitivity, specificity, positive predictive value, negative predictive value as well as confidence interval (CI) were calculated.

The median (IQR) EAT-10 score was 2 (0–8). Twenty-one patients (44.6%) had an EAT-10 score ≥ 3. A weak but statistically significant correlation was found between a higher EAT-10 score and a higher FACED score (p = 0.028, r = 0.32). In addition, patients with an EAT-10 score of above 3 show a trend towards higher FACED scores (p = 0.05).

The FEES test results are summarized in Tables 2 and 3. At least one swallowing abnormality was found in 42 patients (89.3%). Nineteen patients (40.4%) exhibited penetration-related abnormalities and 6 patients (12.7%) exhibited aspiration-related abnormalities. No patient displayed either an ineffective cough or abnormal bolus location before swallowing. Nineteen of the 42 patients with at least one swallowing abnormality had an EAT-10 score > 3. No correlations were found between the FEES test score of individual patients and their FACED or EAT-10 scores (p = 0.615 and p = 0.386, respectively) or any of the baseline or clinical characteristics.

The PAS scores are presented in Table 4. Six patients (12.7%) had a PAS score of ≥ 6, signifying aspiration. A comparison of the patients’ PAS and FACED scores revealed no correlation between them (p = 0.329).

The diagnostic accuracy of EAT-10 in detecting aspiration was calculated, and it is represented by the receiver operating characteristic curve in Fig. 1. We found that the EAT-10 score had a good predictive ability for a PAS score of above 6, indicating aspiration (AUC = 0.78, SD = 0.077, confidence interval 0.629–0.932, p = 0.028). The optimal cutoff value for the EAT-10 score to predict aspiration was 3, with sensitivity, specificity, positive predictive value and negative predictive value of 83, 60, 23, and 96%, respectively.