Diagnostic Accuracy of Aerodigestive Ultrasound for Predicting Swallowing Disorders

Swallowing disorders management requires an accurate recognition of implicated anatomical structures and pathways. The usual clinical assessment of swallowing disorders lack of reliability and accuracy. The gold standard remains the videofluoroscopy. However, this imaging technique lacks reliability and standardisation. Moreover, videofluoroscopy not easily available is time and material consuming and implies ionisation. Ultrasound imaging, which can be performed at the patient's bedside, is a non-invasive tool. It allows the evaluation of the main structures involved in all the swallowing pathways and may be a promising tool for swallowing disorders assessment.

The investigators aim to develop an ultrasound predictive model to diagnose swallowing disorders, and assess its reliability and accuracy.

The investigators hope a high reliability and accuracy of the ultrasound predictive model in the swallowing disorders diagnostic. As ultrasonography is easy-to-perform, rapidly available, non-invasive and inexpensive, it may be a valuable alternative to videofluoroscopy in swallowing disorders diagnostic.

1. Hypothesis The investigators hypothesise that a new ultrasound predictive model is valid and reliable to diagnose swallowing disorder.
2. Objectives Primary objective:

• To create a new ultrasound predictive model based on the most relevant ultrasound measurements of the swallowing structures. Secondary objectives:
• To assess the accuracy of the ultrasound predictive model in the swallowing disorders diagnostic;
• To assess the inter- and intra-reliability of the ultrasound measurements;
• To assess the accuracy of each ultrasound measurements.

1. Study design: The prospective observational single-centre study will be performed per the ethical standards of the Declaration of Helsinki and will be reported according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement.
2. Population: One hundred outpatients at risk of swallowing disorders (neck cancer, neurological diseases, previous ICU stay) will be enrolled in the Dysphagia Diagnostic Unit at the Forcilles' Hospital, during a 2-year period. All consecutive patients will be screened at hospital admission, and will be included if they satisfy the eligibility criteria. All patients will provide oral consent to participate, in accordance with the clinical research French laws. Duration of the patient's participation will take 3 hours.
3. Data collection Data will be recorded using a local case report form. Demographic data, comorbidities, medical and surgical history, reason for swallowing assessment will be recorded at inclusion. Clinical, videofluoroscopy and ultrasound measurements will also be recorded. On-site monitoring is planned every month. Participants will first undergo an ultrasound examination, then the clinical examination and videofluoroscopy imaging.
4. Ultrasound examination: Ultrasonography will be performed using a Sonosite E2 Expert ultrasound device (Digital Color Doppler Ultrasound System; Sonoscape medical corp. China) with a 7-10 MHz linear and a 3-5 MHz probes. Detailed information for ultrasound procedure is described in the research protocol. Ultrasonography will be performed with the patients in near-to-supine position (30°). The hyoid bone movement, tongue kinetics and geniohyoid muscle measurement will be performed in a sagittal plane with the curvilinear probe. The probe will be placed between the posterior border of the symphysis and the anterior margin of the hyoid bone. The anterior belly of digastric and geniohyoid muscles evaluation will be performed in a frontal plane with the linear probe. The probe will be placed in the submandibular area, behind the symphysis. Ultrasonographer will be blinded from all patients' status and previous examination. Reliability will be assessed in the first twenty participants. All ultrasound measurements will be repeated twice by the first ultrasonographer to assess the intra-operator reliability. A second ultrasound operator will perform all the ultrasound measurement, allowing the inter-operator reliability evaluation.
5. Videofluoroscopy (VFSS): VFSS is a radiographic procedure that provides a direct, dynamic view of oral, pharyngeal, and upper oesophageal function during swallowing. The radiologist and SLP work as a team to follow a standard protocol for conducting this examination. During this procedure, the speech-language therapist presents food and fluid mixed with barium. The barium is necessary to view structures via videofluoroscopy during the swallow. The following procedure will be followed: 5 ml of thin fluid; 5 ml of ultra-thick liquid; 5 ml of solid aliment, then 30 ml of thin fluid. The VFSS allow the assessment of the characteristics of the swallow and the patterns of bolus movement, including initiation of the swallow, nasopharyngeal reflux, pharyngeal clearance, and laryngeal penetration and aspiration. At any time, the radiologist may not initiate or stop the examination if it may be harmful to the patient. The reason to stop the test will be recorded. In case of penetration or aspiration, the speech language therapist will use the 'Penetration-Aspiration Scale'.
6. Statistical plan: Continuous variables will be expressed as the mean (± standard deviation) and compared using the Student's t test if the null hypothesis is not rejected by the Shapiro-Wilk test. Continuous variables will be expressed as the median (interquartile range) and compared with the Mann-Whitney U or Kruskal-Wallis test if the null hypothesis is rejected by the Shapiro-Wilk test. For categorical variables, the proportions of patients in each category will be calculated. Then the groups will be compared using Chi squared test. Continuous ultrasound variables will be dichotomised or grouped by class, depending on the linear relation between the logit of the swallowing disorder diagnostic and the ultrasound variable. The threshold for continuous ultrasound variables will be calculated with the ROC curve analysis. A multivariate logistic regression model will be carried out to assess the relation between swallowing disorders and ultrasound measures associated in univariate analyses. All statistically significant ultrasound variables will be included in the model. Variable selection will be stepwise, based on Akaike Information Criterion. To check multicollinearity between independent variables, the variance inflation factor will be calculated before performing multivariate logistic regression. Multicollinearity will be regarded as present when the variance inflation factor is > 5. Goodness of fit will be assessed by Hosmer-Lemeshow method. The outcome of each ultrasound variable from the multivariable logistic regression will be used to weight it in the global ultrasound predictive model. Intra- and inter-operator reliability will be assessed using the kappa coefficient for categorical variables or intra-class coefficients for continuous variables. For all tests, a p-value ≤ 0.05 will be considered statistically significant. All statistical analysis will be made using R software (version 3.6.1, www.R-project.org).
7. Sample size calculation:

The investigators based the sample size calculation on the number of independent variables which will be included in the logistic multivariable regression model. The investigators use the criterion of one variable per 10 events for binary logistic regression analysis. The investigators plan to include the 5 most relevant ultrasound measures in the model. As prevalence of swallowing disorders in our unit is estimated at 0.5, the investigators plan a sample size of 100 patients.

Reliability will be assessed on the first 20 patients. With an 80% predicted kappa, a minimal value of the kappa of 50%, 2 observations, a 95% certainty and a 80% power, 20 patients will be required.