Introduction

Swallowing involves oral, pharyngeal, laryngeal, and esophageal structures working in tandem to efficiently propel food and liquid completely from the mouth into the stomach, whilst maintaining airway protection for the prevention of penetration and aspiration of boluses. Swallowing kinematics can be quantified spatially (how far structures move) and temporally (when structures move, and how fast they move, relative to each other and relative to bolus flow). Hyoid bone displacement can be measured by its overall displacement, or separated into its superior and anterior vectors [1, 2]. It is thought that superior hyoid displacement is an important contributing factor to epiglottic inversion and laryngeal vestibule closure for the prevention of penetration and aspiration of boluses before and during the swallow, while anterior hyoid displacement assists in mechanical opening of the pharyngoesophageal segment to allow for bolus clearance into the esophagus [3,4,5,6,7,8,9]. Research demonstrates that impairments in superior and anterior displacement of the hyoid bone place individuals at increased risk of penetration, aspiration, and post-swallow residue; however, it is important to note that some studies demonstrate impaired hyoid displacement that does not necessarily predict compromised laryngeal closure or bolus clearance [9,10,11,12,13,14,15].

Videofluoroscopic swallow studies (VFSS) are used to analyze spatial and temporal swallowing kinematics in addition to observing adverse airway protective events such as penetration and aspiration of foods and liquids. While many swallowing kinematic measurement methodologies exist, one standardized and frequently used approach to kinematic assessment is the Dynamic Swallow Study (DSS) protocol, originally developed by Rebecca Leonard and Katherine Kendall [11, 16,17,18,19,20,21]. The DSS protocol involves having an examinee establish a “pseudo-rest” position by way of holding a 1 cc liquid bolus in their mouth. After establishing the pseudo-rest position, the examinee performs a single cued swallow of 1, 3, and/or 20 cc nectar-thick liquids. Upon completion of the exam, the examiner can perform a frame-by-frame fluorographic analysis of a variety of spatial and temporal swallowing kinematics [11, 16, 22,23,24,25,26,27,28,29,30,31,32].

The “Hmax” is a DSS hyoid displacement measure describing the total distance traveled by the hyoid bone from the pseudo-rest position to the point of maximal anterior–superior hyoid displacement at or near the height of 1, 3, or 20 cc nectar-thick liquid swallows. While healthy volunteer DSS Hmax norms have been previously established, normative DSS data separately describing superior and anterior hyoid displacement, and their relationships to one another, have not been previously investigated [16]. Given the unique physiologic contributions of superior hyoid movement and anterior hyoid movement, establishing separate normative data within the context of a DSS protocol, and describing their kinematic relationship to one another, would provide meaningful information for clinical and research purposes.

The aims of this study were to (1) establish normative data for maximal superior hyoid displacement (Hsup), maximal anterior hyoid displacement (Hant), and the ratio of superior to anterior hyoid displacement (SAratio) in non-dysphagic patients within the context of the DSS maximal hyoid displacement (Hmax) measure and (2) assess the effects of age, sex, and bolus size on these measures.

Methods

Record Review

Records were reviewed for consecutive non-elderly (< 65) and elderly (≥ 65 years) male and female non-dysphagic patients who presented for VFSS at an outpatient, tertiary swallowing center. Non-dysphagic patients were included if VFSSs revealed a Functional Oral Intake Scale (FOIS) score of 1; a Penetration Aspiration Scale score ≤ 2; complete horizontal and vertical epiglottic inversion for all swallows during the VFSS; and normal (i.e., < 2 SD below mean) pharyngeal constriction ratio (PCR), hyolaryngeal approximation (HL), pharyngoesophageal segment opening (PESmax), and maximal hyoid displacement (Hmax) DSS displacement measures for 1, 3, and 20 cc nectar-thick liquid swallows [16, 33, 34]. Patients having the following criteria were excluded: a history of oral cavity, laryngeal, or pharyngeal surgical intervention (with the exception of routine dental work); a history of radiation therapy to the head and/or neck; diagnosis of a neurologic and/or neuromuscular disease; and/or a formal diagnosis of a structural- or motility-based esophageal abnormalities.

Videofluoroscopic Swallow Studies

Videofluoroscopic examinations were performed at the University of California San Francisco (UCSF) Medical Center in accordance with established and routine VFSS+DSS protocols. Two different fluoroscopic machines were used: the Axiom Luminos TF (Siemens Healthcare, USA) and the Luminos Agile Max (Siemens Healthcare, USA). Patients were presented with 1, 3, and 20 cc nectar-thick liquid barium boluses (40% w/v Varibar Nectar Barium Sulfate Suspension) and were instructed to (1) hold the liquid bolus in their mouth (i.e., the “pseudo-rest” position) and then (2) attempt to swallow the entire bolus in a single swallow when cued by the clinician. All video segments were recorded in a lateral viewing plane with an image-capturing rate of 25–30 images per second (depending on the capabilities of the fluoroscopy machine) and a magnification level of 1–2 ×. All VFSS videos were saved directly into a picture archiving and communication system (PACS) for later review and analysis.

Hyoid Displacement Measurement Methodology

Both superior hyoid displacement (Hsup) and anterior hyoid displacement (Hant) were measured at the same points in time within the same video frames—during the pseudo-rest position (i.e., 1 cc bolus hold) and during maximal overall hyoid displacement (i.e., Hmax) at or near height of swallow for 1, 3, and 20 cc liquid swallows. The Hsup was defined as the change in vertical hyoid position from the pseudo-rest position (“Hsup-rest”) to the Hmax position (“Hsup-max”). The Hant was defined as the change in horizontal hyoid position from the pseudo-rest position (“Hant-rest”) to the Hmax position (“Hant-max”). The relationship between superior and anterior hyoid displacement was calculated by dividing Hsup by Hant, in order to get the superior-to-anterior hyoid displacement ratio (SAratio). See Figs. 1a–d for more detailed instructions on measurement methodology.

Fig. 1
figure 1

a Measurement methodology: Establish horizontal and vertical planes relative to the larynx during a pseudo-rest position (1 cc bolus hold) by connecting the anterior-inferior hyoid to anterior–superior tracheal air column. b Move the line established in Fig. 1a from the larynx to the spine with rotation or distortion. Ensure that the line crosses over the anterior-inferior point of cervical spine 2 (P1). Make note of a second identifiable point (P2) that is inferior to both P1 point and the hyoid bone but that also lays along the same line—extend the line inferiorly as necessary. In this example, P2 is along the superior border of C4. c Next, trace a line from the anterior-inferior hyoid bone to the P1–P2 line drawn in Fig. 1b until a 90º intersection is made. Extend the P1–P2 line superiorly as needed. Measure the distance from P2 to the 90º intersection (“Hsup-rest”) and the distance from the hyoid to the 90º intersection (“Hant-rest”). d Advance to the point of maximal overall hyoid displacement (Hmax). Repeat the instructions outlined in Fig. 1c by retracing a line using the original P1 and P2 points. Trace a line from the anterior-inferior hyoid until a 90º intersection is made. Measure from P2 to the 90º intersection (“Hsup-max”), and from the hyoid to the 90º intersection (“Hant-max”)

Full size image

Data Abstraction and Reliability Testing

All measures were performed by one primary rater (JC). Ten percent of the video clips were selected at random and were repeated for analysis 1 week later by the primary rater to facilitate intra-rater reliability estimation. A second rater (JL) analyzed the same randomly selected video segments in order to assess inter-rater reliability. Both raters had ≥ 2 years of experience performing DSS kinematic analysis and were blinded to patient’s history and identity.

Statistical Analysis

Descriptive statistics were performed for demographic information, Hmax, Hsup, Hant, and SAratio. A mixed-design repeated-measures analysis of variance (ANOVA) with post hoc analysis and Bonferroni corrections, using bolus size (1, 3, 20 cc) as a within-subject factor, and age (elderly and non-elderly) and gender (male and female) as between subject-factors, was used to assess differences with Hsup, Hant, and SAratio as a function of age, sex, and bolus size. A p < 0.05 was set as the significance level for all statistical tests. Two-way random effects, absolute agreement, intraclass correlation coefficients (ICC) were used to calculate intra- and inter-rater reliabilities. Interpretation of ICC was judged to be ‘excellent’ if ≥ 0.90, ‘good’ if between 0.75 and 0.90, ‘moderate’ if between 0.50 and 0.75, and ‘poor’ if < 0.50 [35]. Statistical analyses were performed using SPSS statistical package version 24 (SPSS Inc., Chicago IL).

Results

Patient Demographics

One hundred sixty-five patients met the above inclusion–exclusion criteria and were included for initial study analysis. Four of the exams were extreme outliers (> 3 SD away from the Hmax mean) and were therefore excluded. A total of 161 exams were included in the final analysis. Non-elderly men (n = 39) had an average age of 50.3 years (± 12.1), with an age range from 18 to 64 years. Elderly men (n = 42) had an average age of 76.7 years (± 7.7), with an age range from 65 to 94 years. Non-elderly women (n = 46) had an average age of 46.5 years (± 13.8), with an age range from 18 to 64 years. Elderly women (n = 34) had an average age of 77.3 years (± 8.9), with an age range from 65 to 96 years.

Intra- and Inter-rater Reliabilities

Sixteen exams were selected at random and yielded a total of 102 repeated anterior and superior hyoid displacement measures. Intraclass correlation coefficient estimates demonstrated excellent inter-and intra-rater reliabilities for superior hyoid displacement and good intra-and inter-rater reliabilities for anterior hyoid displacement. For Hsup, inter-rater ICC was 0.929 (95% CI 0.866–0.962) and intra-rater ICC was 0.973 (95% CI 0.951–0.986). For Hant, inter-rater ICC was 0.821 (95% CI 0.639–0.908) and intra-rater ICC was 0.887 (95% CI 0.801–0.938).

Superior Hyoid Displacement (H sup)

Age, sex, and bolus size norms [mean ± standard deviation (range)] are outlined for Hsup (Table 1). Hsup was significantly effected by bolus size, F (2, 312) = 94.158, p < 0.005, and sex, F (1, 156) = 16.869, p < 0.0005, but not age (p > 0.05). Pairwise comparisons were performed for statistically significant differences between bolus size with Bonferroni corrections being made, revealing significant differences (p < 0.0005) between all three bolus sizes (i.e., 1 and 3 cc, 1 and 20 cc, and 3 and 20 cc). Three-way mixed ANOVA revealed a statistically significant three-way interaction between age, sex, and bolus size, F (2, 312) = 3.712, p = 0.026, partial η2 = 0.023 (Fig. 2). Statistical significance of a simple two-way interaction was accepted at a Bonferroni-adjusted alpha level of 0.025. There was a statistically significant simple two-way interaction of sex and age for 20 cc bolus size, F (1, 156) = 6.254, p = 0.013, but not for 1 cc or 3 cc bolus sizes (p > 0.05). Statistical significance of a simple simple main effect was accepted at a Bonferroni-adjusted alpha level of 0.025. There was a statistically significant simple simple main effect of age for males for 20 cc bolus size, F (1, 156) = 7.758, p = 0.006, but not for females (p > 0.05).

Table 1 Maximal superior hyoid displacement (Hsup) in millimeters (mm)
Full size table
Fig. 2
figure 2

Three-way interaction between age, sex, and 20 cc bolus size for superior hyoid displacement, with simple simple main effects between young and elderly males

Full size image

Anterior Hyoid Displacement (H ant)

Age, sex, and bolus size norms [mean ± standard deviation (range)] are outlined for Hant (Table 2). Hant was statistically effected by bolus size, F (2, 312) = 6.845, p = 0.001, and age, F (1, 156) = 10.449, p = 0.001, but not sex (p > 0.05). Pairwise comparisons were performed for statistically significant differences between bolus size with Bonferroni corrections being made, revealing significant differences (p < 0.01) between 1 and 3 cc and 1 and 20 cc, but not between 3 and 20 cc (p = 0.497). Three-way mixed ANOVA was run to understand the effects of age, sex, and bolus size on Hant. There were no statistically significant three- or two-way interactions between age, sex, and bolus size (p > 0.05). Statistical significance of a simple simple main effect was accepted at a Bonferroni-adjusted alpha level of 0.025.

Table 2 Maximal anterior hyoid displacement (Hant) in millimeters (mm)
Full size table

Superior-to-Anterior Ratio of Hyoid Displacement (SAratio)

Age, sex, and bolus size norms [mean ± standard deviation (range)] are outlined for SAratio (Table 3). SAratio was not significantly effected by bolus size, age, or sex (p > 0.05). A three-way mixed ANOVA was run to understand the effects of age, sex, and bolus size on SAratio. There were no statistically significant three- or two-way interactions between bolus size, age, and sex for SAratio (p > 0.05).

Table 3 Ratio of superior to anterior hyoid displacement (SAratio)
Full size table

Discussion

This study establishes normative data for superior and anterior hyoid displacement, and the ratio between these two measures, for elderly and non-elderly males and females for 1, 3, and 20 cc liquid boluses within the context of the Hmax DSS measurement. Superior and anterior hyoid displacement varied significantly by bolus size. Sex was noted to significantly impact superior hyoid displacement but not anterior hyoid displacement. Age was seen to significantly impact anterior hyoid displacement, but not superior hyoid displacement. Neither bolus size, age, nor sex significantly impacted the ratio of superior to anterior hyoid displacement. Generally speaking, superior and anterior displacements became larger with increasing bolus sizes, and was larger in males compared to females, and non-elderly when compared to elderly. These findings are similar with previous studies, which demonstrate that hyoid displacement increases with increasing bolus sizes, and may be impacted by age and sex [11, 16,17,18, 29].

The hyoid bone was found to frequently move an average of two to three times more superiorly than as it does anteriorly. In some instances, anterior displacement was noted to be negative, in other words, moving slightly posterior at time of maximal overall hyoid displacement. It is unclear if these patterns for increasingly larger superior-to-anterior hyoid displacement ratios and large variability of anterior hyoid displacement reflect outcomes inherent to this measurement methodology (both superior and anterior measures taken at the same point in time), or if they truly demonstrate a highly variable nature in anterior hyoid movement in normal swallow function. This result is significant, as findings of marginal-to-nil (or even negative) anterior hyoid displacements relative to large superior hyoid displacements may not reflect a disrupted swallow pattern, but rather be a variant of normal healthy swallow, as was observed with multiple exams in the present study. This finding would also call into question the extent to which anterior hyoid displacement plays in functional airway protection and pharyngoesophageal segment opening in non-dysphagic and dysphagic patients.

Two key points should be noted when extrapolating the data for swallow function interpretation. Firstly, Hsup and Hant displacement measures are made by establishing vertical and horizontal planes relative to the orientation of the larynx. This methodology varies from many studies evaluating superior and anterior hyoid displacement norms, which use vertical and horizontal planes relative to the lateral fluoroscopic viewing plane or anterior aspects of the spine, and as a result, superior movement appears generally longer and anterior movement appears generally shorter than previously reported [6, 18, 21, 36,37,38]. Secondly, it should be recognized that while maximal anterior and superior hyoid displacements tend to occur within close temporal proximity to each other, these displacement measures often occur at different time points during the swallow. Hsup and Hant may therefore not reflect true maximal superior and anterior hyoid displacements outside of the context of DSS’s Hmax. However, Hmax was chosen as the single point of analysis for Hsup and Hant measurements in hopes of capitalizing on the excellent reliability that was previously reported for Hmax. This method proved successful, as the reliability of the present two measures also appear to have good-to-excellent intra- and inter-rater reliabilities. Further, the Hmax was chosen as the single point of analysis for the two measures in this study in an effort to minimize clinical burden by collapsing three separate hyoid displacement time frames into one, therefore reducing the time required to collect and calculate the measures. All that is needed to obtain Hmax is a Pythagoreans theorem calculation, rather than an additional measurement, where Hmax = (H 2sup  + H 2ant )0.5.

The main limitation of this study is that analyses were performed on “non-dysphagic” individuals previously seen in an outpatient otolaryngology clinic rather than on healthy volunteers. Strict inclusion–exclusion criteria were established to rule out abnormal physiology and people with compromised bolus clearance and airway protection; however abnormal physiology not representative of healthy volunteers cannot be ruled out. However, one could argue that even if the findings of the present study were not representative of “normal healthy” physiology, the present information is still valuable in understanding the deviant non-dysphagic physiology that still results in complete bolus clearance and airway protection.

Conclusion

This study presents age, sex, and bolus size norms for Hsup, Hant, and SAratio, taken at time at maximal hyoid displacement (Hmax), within the context of a Dynamic Swallow Study protocol for videofluoroscopic swallow studies. Establishing normative data for anterior and superior hyoid displacement should allow clinicians using DSS to more thoroughly identify areas of swallowing impairment, determine targets for therapeutic intervention, and track individual physiologic changes over time.