Ultrasound: An Emerging Modality for the Dysphagia Assessment Toolkit?

Abstract

Purpose of review

Videofluoroscopy (VFSS) and fibreoptic endoscopic evaluation of swallowing (FEES) are established instrumental techniques to support differential diagnosis and treatment of oropharyngeal dysphagia. Whilst their value is undisputed, each tool is not without limitations. The COVID-19 pandemic has restricted access to VFSS and FEES leading clinicians to explore alternative or augmentative tools to support swallowing assessment.

Ultrasound (US) is an established tool for visualisation of head and neck anatomy, including structures implicated in swallowing. While US has been utilised in swallowing research for many years, its application has not translated into common clinical practice. This review presents and debates the evidence for and against use of ultrasound for clinical swallowing assessment.

Recent findings

Evaluation of swallowing muscle morphometry and measurement of isolated swallowing kinematics are two primary uses of US in swallowing assessment that have been identified in the literature. Use of US to detect bolus flow, aspiration, and residues is in its early stages and needs further research.

Summary

US shows promise as an adjunctive modality to support assessment of swallowing. With standardisation, these measurements may have potential for transition into clinical care. Reliability and validity testing and development of normative data are imperative to ensure its use as an evidence-based instrumentation.

Introduction

Instrumental assessment plays a key role in differential diagnosis of oropharyngeal dysphagia and treatment planning. Videofluoroscopy and fibreoptic endoscopy are commonly used instrumentation within Speech and Language Therapy (SLT) (1, 2) that provide information about kinematics, bolus flow, and symptomatology (3, 4). Despite benefits, both imaging modalities have limitations. Videofluoroscopic swallowing study (VFSS) exposes patients to radiation while fibreoptic endoscopic evaluation of swallowing (FEES) is an intrusive procedure that does not allow visualisation of each stage of swallowing. External software is required for videofluoroscopic temporal and spatial measurements, and VFSS does not offer the opportunity to assess mucosal tissues or vocal fold movements. While the value of VFSS and FEES remains undisputed, the role of radiation-free and non-invasive ultrasound imaging (US) as a viable addition to the clinical dysphagia assessment toolkit warrants further exploration.

US allows visualisation of soft tissue structures, such as the tongue and floor of the mouth. Visualisation can be static, to assess morphometry (muscle structure and composition), or dynamic, to assess biomechanics (5). Another critical point of difference compared to VFSS and FEES, is the capacity for quantitative measurements of swallowing derived from the US image during an examination (6), highlighting its potential efficiency and capacity for rapid ‘point of care’ decisions. The wide variety of US technologies includes highly portable systems requiring only a probe and tablet device. These offer potential to be used in difficult to access populations, such as patients living in the community with disability. The potential for use of US in the current schema of swallowing assessments will be presented and debated.

Application of ultrasound

Muscle morphometry

US has been used to assess morphometry of muscle groups related to swallowing, such as the tongue, masseter, and submental muscles (7–9). Morphometric assessment includes measurements of muscle thickness or cross-sectional area and gradings of structural muscle degradation (10). US has been found to more precisely define and measure submental muscles when compared to MRI in healthy subjects (7). Associations have been identified between muscle morphometry, swallowing biomechanics and symptoms of dysphagia in specific patient populations (11, 12). These findings have applicability in neuromuscular disease groups such as motor neuron disease (MND), where unique sonographic profiles of swallowing muscles have been described (13).

Use of US to analyse geniohyoid muscle bulk at rest demonstrated significant differences between patients’ post-radiotherapy and young, healthy controls (14). The cross-sectional area of patients’ geniohyoid muscle was larger than that of the controls. These findings have facilitated hypothesis-formulation to further understand and treat biomechanical changes as a result of radiotherapy, as well as guiding research to understand sub-clinical muscle changes and thresholds for functional impact. In patients with nasopharyngeal carcinoma post-radiotherapy, a significant positive correlation was identified between percentage increase in US measures of cross-sectional area of the geniohyoid muscle and videofluoroscopic measures of anterior hyoid movement during liquid swallowing (15). Increased echogenicity of the masseter has been shown to significantly predict change in mastication, as measured by the timed test of mastication(16) in young adults living with Duchenne muscular dystrophy (DMD) (17). These examples demonstrate the potential applicability of US as a non-invasive biomarker for dysphagia (18).

There are promising data to suggest that US may be a suitable tool to longitudinally assess morphometric non-swallowing muscle changes. In a study of 18 boys with DMD, a significant increase in echo intensity and reduction in muscle thickness was identified in specified limb muscles with age. These muscle changes were significantly correlated with clinical parameters such as strength and functional status (19). Replication of this study in the bulbar muscle groups would support use of US for long-term, non-invasive monitoring of swallowing. Increased understanding of age-related changes in muscle morphometry, as well as the impact of confounders such as nutrition (20) and sarcopenia (21) on these measures are required before transition into clinical practice.

Whilst isolated studies have demonstrated good reproducibility of US assessed morphometric measures such as tongue thickness (22) and cross-sectional area of the geniohyoid (15, 23), more research is needed in morphometric analysis to evaluate reliability within, across raters, and over time. The opportunity that US provides to assess underlying muscle morphometry is, however, an exciting leap forward in SLT practice. Unlike FEES and VFSS, it offers an opportunity to understand changes at the muscular level and demonstrates potential as a responsive outcome measure to a therapeutic intervention.

Imaging studies have identified an association between muscle volume loss, for example in the base of tongue or geniohyoid, and swallowing impairment (24). This highlights the prospective ability of US to support targeted swallowing treatment. Its utility has been demonstrated in non-swallowing muscles and found to have validity as a tool to assess muscle volume pre- and post-exercise training (25). Whilst there is evidence supporting the relationship between isolated muscle strength and swallowing function (26), the impact of swallowing muscle training on bulbar muscle structure and volume remains unknown. Research to evaluate the potential of US to be used in the assessment of muscle change following strength training is required. The role of US as a biofeedback modality is a further area for exploration (27, 28).

Swallowing kinematics

US has been used to assess swallowing kinematics of the tongue, hyoid, larynx, and lateral pharyngeal wall (29). Assessment of hyolaryngeal movement has received particular attention (11, 15, 30–33); this may be explained by its ease of identification and potential association with airway protection (34). Percentage increase of the coronal cross-sectional area of floor of mouth muscles has been utilised as a surrogate for geniohyoid contraction and hyoid movement (15). Other studies have focused on temporal measurements of hyoid movement (11).

US measures of laryngotracheal movement have been used to assess pre- and post-surgical measures of swallowing ability in a population of 40 patients post thyroidectomy (33). The distance between the sternal notch and the first tracheal ring was compared at rest and at the highest point of laryngotracheal elevation. These measurements showed significant differences pre- and six months post-surgery (p = 0.021). Questions on swallow symptoms were taken from the thyroidectomy-related voice questionnaire (TCQ) (12) to assess swallowing function. A higher score on these questions was significantly correlated with reduced laryngotracheal elevation. Differences in temporal US measures of hyoid movement including shorter duration of hyoid elevation for thickened liquid swallowing have been identified between post-surgical thyroid populations and healthy controls (11). These data highlight the potential application of US as a tool to identify and measure longitudinal changes in swallowing, though evaluation of reliability is required in future studies.

Studies have reported relationships between US and videofluoroscopic measurement of hyoid excursion (30, 35), as well as surrogate measures for dysphagia, such as aspiration (31, 36), oral intake (32), and tube feeding (37); however, variation in procedural methods limits comparison across studies (38). Nevertheless, validity data of US in swallowing assessment are emerging. A comparison of US and VFSS measurement of hyoid displacement in subjects with dysphagia of multiple aetiologies showed a strong positive correlation (> 0.8, p < 0.01) between the two instruments (30). A cut-off value of 13.5 mm for hyoid displacement in patients with dysphagia was reported to predict presence or absence of aspiration (31). Sensitivity and specificity were reported as 83.9 and 81.0, respectively, with excellent intra- and inter-rater reliability (> 0.96, p < 0.01 and 0.89, p < 0.05). Differences in US measures of hyoid-laryngeal approximation were found in patients post-stroke with and without dysphagia as determined by the functional oral intake scale (FOIS) (32). However, no reliability data were reported.

Online US measurements of swallowing kinematics show potential as a valuable addition to the toolkit. The relative ease of imaging the shadow cast by the hyoid has led to a focus on hyoid displacement rather than less echogenic structures, such as the tongue base and epiglottis. Whilst the relevance of suprahyoid (specifically geniohyoid) muscle contraction to swallowing is supported in research (39, 40), the overall relevance of altered hyoid bone movement in disordered swallowing is disputed (41). Further high-quality studies to validate US assessed hyoid movement as a technique to identify dysphagia are required.

Symptoms of dysphagia

Evaluation of aspiration or residue are important variables in swallowing assessment and management but challenging to visualise using US. The low density of a bolus compared to soft tissue structures means US will travel with less reflection through the material, creating challenges with visualisation (42). Recent research has explored the potential application of US in visualising aspirates and residues using image-processing methods that compared movement of food and fluid to surrounding tissue (43–45). This US technique has been used to detect aspiration and residue in a randomised control trial which compared the effectiveness of US-guided swallowing care with standard swallowing care in elderly nursing home residents (44). The group receiving US-guided care were assessed with US every two weeks over an eight-week period. Any residue or aspiration detected on US were managed via a pre-determined algorithm. This group were found to have less frequent episodes of aspiration and residue at the end of the eight-week period compared to those receiving standard care. Findings were not statistically significant and there was no reduction in rate or time to aspiration pneumonia. The study is compromised by the small sample and absence of reliability assessment.

A stand-alone assessment of residue and aspiration has limited clinical application for dysphagia diagnostics and treatment. However, there is potential applicability for ‘in the moment’ symptom management (46). A critical limitation of current literature is lack of reported reliability. Thus, use of US for this purpose remains contentious. Advancement in accurate assessment of aspiration and residue is likely to require reliable identification of key anatomical landmarks, specifically the valleculae, pyriforms, and the subglottis. The ability to detect these structures using US remains disputed in the literature (37). Further validity and reliability data are required before US can be used for this purpose in the dysphagia toolkit.

Advantages and limitations of ultrasound as a swallowing assessment modality

US offers promise as a less invasive or disruptive option than VFSS or FEES. This extends its utility to patients with restricted alertness and vulnerable patient groups, such as children. Easily transportable equipment also provides opportunity for US swallowing assessment outside of the hospital setting, improving access to geographically hard to reach clinical groups.

There are, however, fundamental constraints to integration of US in clinical application at present. The most pertinent being lack of established reliability associated with data acquisition, image selection for measurement, and interpretation (38). The acquisition and measurement of dynamic images presents a challenge for intra- and inter-rater reliability beyond that of static imaging (47). Whilst small scale studies have shown promising rater-agreement on defined swallowing parameters with standard-sized equipment (22, 23, 40), further work is required prior to translation into clinical practice. The validity and reliability of easily transportable hand-held devices also warrant further exploration (48).

One of the major limitations of US-assessed kinematics is the inability to achieve a gestalt view of swallowing. The accessibility of more superficial kinematic swallowing movements may put the practising clinician at risk of overlooking critical changes in swallowing that are less accessible by US. The challenges associated with capturing less echogenic structures and movements in addition to the inability to capture multiple dynamic movement supports the use of US as an adjunctive tool only and positions VFSS as the assessment of choice for clinicians requiring a comprehensive view of swallowing biomechanics.

Whilst evidence supporting the utility of US as an adjunctive imaging modality in the dysphagia toolkit continues to increase, we are yet to determine which patient populations are mostly to benefit. There is early evidence to suggest that US swallowing assessment does not interrupt mealtimes (44), yet more research focused on the patient experience is required.

A pathway for the clinical translation of US to swallowing diagnosis

Education and training

Accurate and effective US swallowing assessment requires highly skilled operators, associated with challenging and time-consuming competency acquisition to achieve independent practitioner status (49). For SLTs to use US technologies, an understanding of the technical and practical aspects, including probe selection, probe stabilisation and placement, and high-quality image acquisition is critical (50). Collaboration across professional groups, particularly radiographers and manufacturers, will be essential to achieve high-quality imaging and training. Upskilling of SLT in practical aspects of US may help to expedite clinical integration of US to build practice-based evidence and its application across a range of patient groups.

Standard protocols and normative data

The major step to establishing US as a clinical tool is to develop clear methodological protocols and standardisation of measurements for swallowing kinematics and morphometry. Preliminary normative data are available for hyoid movement (51, 52) and tongue morphometry (8); however, larger normative data set for all measures should be developed after universal protocols are in place.

Clinical guidelines

The transition of US from research to clinical settings should account for the requirements unique to this environment. Application guidelines for clinical integration and scope of US as a swallowing assessment tool will require clear indications for use as an instrumental screening tool, a method to quantify kinematic or structural characteristics, or a tool to document change from disease progression or rehabilitation.

Future directions

Implementation of US swallowing assessment in clinical practice is premature but methods with more imminent potential for clinical translation and integration are in development. US measures of isolated swallowing kinematics and muscle morphometry have potential for transition into clinical care with standardisation. This presents opportunities to refine our understanding of the use of US via expansion of practise-based evidence. Future research should focus on the potential of US to assess aspiration, bolus flow and residue. Research should also consider the potential use of US as a biofeedback modality or tool to assess impact of therapeutic interventions (28).

Key points.

• Ultrasound has potential to be used in the assessment of muscle morphometry and swallowing kinematics.

• More validity and reliability data are required, particularly for the use of ultrasound to detect aspiration and residue.

• Work towards refinement of protocols including scanning and measurement techniques is needed.