Michael A. Crary
`Giselle D. Carnaby (Mann)
`University of Florida Health Science Center,
`Gainesville
`
`Michael E. Groher
`University of Redlands, Redlands, CA
`
`Biomechanical Correlates of Surface
`Electromyography Signals Obtained
`During Swallowing by Healthy Adults
`
`Purpose: The purpose of this study was to describe biomechanical correlates of the
`surface electromyographic signal obtained during swallowing by healthy adult
`volunteers.
`Method: Seventeen healthy adults were evaluated with simultaneous videofluoroscopy
`and surface electromyography (sEMG) while swallowing 5 mL of liquid barium sulfate.
`Three biomechanical swallowing events were analyzed: hyoid elevation, pharyngeal
`constriction, and opening–closing of the pharyngoesophageal segment. For each
`biomechanical event and from the sEMG signal, the authors identified onset, peak,
`and offset time points. From these points, duration measures were calculated. Means
`and 95% confidence intervals were calculated for each measure. Subsequently,
`correlations were evaluated between timing aspects of the sEMG traces and each
`biomechanical event.
`Results: Swallow onset in the sEMG signal preceded the onset of all biomechanical
`events. All biomechanical events demonstrated a strong correspondence to the sEMG
`signal. The strongest relationship was between hyoid elevation–anterior displacement
`and the sEMG signal.
`Conclusions: These results suggest that the sEMG signal is a useful indicator of major
`biomechanical events in the swallow. Future studies should address the impact of
`age and disease processes, as well as bolus characteristics, on the biomechanical
`correlates of sEMG signals obtained during swallowing.
`
`KEY WORDS: electromyography, videofluoroscopy, swallowing assessment
`
`S wallowing is a complex sensorimotor function that incorporates
`
`activity from multiple muscle groups in the upper aerodigestive
`tract. Muscle activity associated with swallowing movements may
`be evaluated with intramuscular or surface electromyographic tech-
`niques. The intramuscular approach uses various forms of needle
`electrodes placed directly into specific muscles. Intramuscular electro-
`myography is used primarily to evaluate activation patterns of specific
`muscles and timing coordination among various muscles. Conversely, the
`surface electromyographic signal is obtained from electrodes adhered to
`the skin over a group of muscles to be studied. Unlike intramuscular
`electromyography, the surface technique lacks muscle specificity. Electro-
`myographic signals obtained through surface electrodes represent simul-
`taneous muscle activity from multiple muscles in the region of interest.
`Thus, although both techniques provide data that can be used to estimate
`the muscular activity associated with swallowing, each technique provides
`different information.
`
`Using the intramuscular technique with a variety of animal models,
`Doty and Bosma (1956) demonstrated a systematic and consistent timing
`
`186
`
`Journal of Speech, Language, and Hearing Research  Vol. 49  186–193  February 2006  AAmerican Speech-Language-Hearing Association
`1092-4388/06/4901-0186
`
`Petitioner - Avation Medical, Inc.
`Ex. 1035, p. 186
`
`

`

`pattern of muscle activation during swallowing. In jux-
`taposition, recent studies involving human participants
`have demonstrated considerable intersubject variability
`across activation patterns of specific muscles during
`swallowing (Gay, Rendell, & Spiro, 1994; Spiro, Rendell,
`& Gay, 1994). Furthermore, intramuscular electromyo-
`graphic results have demonstrated variability in the
`activation of specific muscles across various swallowing
`tasks (McCulloch, Perlman, Palmer, & VanDaele, 1996;
`Ertekin et al., 1997). The primary contribution of the
`intramuscular approach to electromyography (EMG) is
`the ability to study individual muscles. The timing and
`pattern of muscle activation are common foci of this ap-
`proach in the study of swallowing.
`
`Surface EMG (sEMG) has been used primarily as
`follows: (a) to identify the presence of swallowing activity,
`(b) to analyze swallowing functions (timing and ampli-
`tude), and (c) as a biofeedback strategy in the treatment
`of swallowing disorders (Bryant, 1991; Crary, 1995; Crary,
`Carnaby (Mann), Groher, & Helseth, 2004; Haynes, 1976;
`Huckabee & Cannito, 1995; McKeown, Torpey, & Ghem,
`2002; Vaiman, Segal, & Eviatar, 2004). Despite extensive
`research and clinical applications of this technique, lim-
`ited information exists regarding specific swallowing bio-
`mechanical activities that may be associated with sEMG
`correlates of swallowing. sEMG does not provide infor-
`mation about the activity of specific muscles or move-
`ments; however, studies have evaluated specific muscles
`that are associated with sEMG signals obtained during
`swallowing. For example, Palmer, Luschei, Jaffe, and
`McCulloch (1999) evaluated specific muscle activity con-
`tributing to sEMG signals obtained from the submental
`site. These investigators concluded that the mylohyoid,
`geniohyoid, and anterior belly of the digastric muscles
`contributed the most information to the submental
`sEMG signal obtained during swallowing. In related
`work, Perlman, Palmer, McCulloch, and VanDaele (1999)
`used both intramuscular and surface electromyography
`to study activation patterns and timing relationships
`among laryngeal, pharyngeal, and submental muscles
`involved in swallowing. Specific muscles revealed a
`volume-dependent response for duration and pattern
`of activation during swallowing. Furthermore, although
`the pattern of muscle activation demonstrated high in-
`trasubject agreement, considerable variability was ob-
`served across participants.
`
`Few studies have evaluated movement correlates of
`electromyographic signals during swallowing. Ertekin
`et al. (1995, 1997) studied relationships among laryngeal
`elevation, as measured by a movement transducer on the
`anterior neck over the larynx; submental sEMG signals;
`and intramuscular EMG signals obtained from the
`cricopharyngeus muscle during swallowing. These stud-
`ies revealed that laryngeal elevation was significantly
`related to both submental and cricopharyngeal muscle
`
`activity. Specifically, laryngeal elevation was significantly
`related to increased submental sEMG activity and to de-
`creased cricopharyngeal muscle activity. Ding, Larson,
`Logemann, and Rademaker (2002) also identified a
`strong temporal relationship between laryngeal eleva-
`tion and the submental sEMG signal. Using electro-
`glottography as a measure of laryngeal elevation, these
`investigators reported a close temporal relationship be-
`tween submental sEMG signals and laryngeal elevation
`in both normal swallows and in swallows using the
`Mendelsohn maneuver (sustained laryngeal elevation
`during the swallow).
`
`Although these studies have addressed laryngeal
`elevation during swallowing in reference to sEMG ac-
`tivity, few studies have evaluated other biomechanical
`events in reference to the sEMG signal. For example,
`researchers believe that the submental muscle complex
`has an indirect influence on the opening of the phar-
`yngoesophageal segment (PES). The PES typically re-
`mains closed at rest and opens during a swallow. Goyal
`(1984) observed that submental muscle activation con-
`tributes to superior and anterior displacement of the
`hyo–laryngeal complex. This movement provides the
`traction necessary to distend the PES, contributing to
`lowered pressure in this sphincter that facilitates open-
`ing during swallowing. This relationship among sub-
`mental muscle activation, hyo–laryngeal elevation, and
`opening of the PES is so well recognized that clinical
`techniques to improve PES opening have focused on
`strengthening the submental muscles (Shaker et al.,
`1997, 2002). Despite this clinical application, no study
`has described potential direct relationships between sub-
`mental sEMG activity and opening of the PES.
`
`Comparison of specific biomechanical events with
`sEMG signals requires a method to simultaneously ob-
`serve the swallowing events while recording the EMG
`activity. Using simultaneous fluoroscopic examination
`and sEMG information from four muscles sites (orbicu-
`laris oris, masseter, submental group, and laryngeal strap
`muscles) in 5 healthy participants, Vaiman, Eviatar, and
`Segal (2004) evaluated sEMG signals in reference to the
`sequence or stages of swallowing events. Although these
`investigators reported ‘‘synchronism’’ between the fluo-
`roscopic images and the sEMG signals, no details were
`provided regarding the simultaneous technique or the
`interpretation of their results in reference to specific bio-
`mechanical events occurring during swallowing.
`
`Prior reports have demonstrated that laryngeal
`elevation, as measured by indirect methods, is related
`to muscle activity in the submental region and in the PES
`(cricopharyngeus muscle). Beyond these reports, little
`information is available documenting specific biome-
`chanical correlates of sEMG signals obtained during
`swallowing. Information within the sEMG signal may be
`important in understanding clinical applications for this
`
`Crary et al.: Swallow Correlates of sEMG Signals 187
`
`Petitioner - Avation Medical, Inc.
`Ex. 1035, p. 187
`
`

`

`technique. For example, knowledge of specific biome-
`chanical events contributing to the sEMG signal ob-
`tained from a specified location during swallowing may
`help clinicians develop and apply effective swallowing
`therapies.
`
`In the present study, we evaluated specific biome-
`chanical correlates of sEMG activity obtained from the
`submental region of the anterior neck during swallowing
`by healthy adults. Our purpose was to identify and de-
`scribe timing relationships between three biomechanical
`aspects of swallowing and the corresponding sEMG
`signal. Specifically, we evaluated timing aspects of hyoid
`elevation and anterior movement, pharyngeal constric-
`tion, and PES opening compared with the timing of
`sEMG events recorded simultaneously with each swal-
`low. These biomechanical swallow aspects were selected
`as representative of cardinal events during swallowing:
`elevation, constriction, and bolus outlet. We anticipated
`that hyoid movement would be strongly related to sEMG
`activity during swallowing. Hyoid elevation is strongly
`tied to laryngeal elevation during swallowing, and both
`are the result of activity in muscles close to the surface
`in the submental region. Because hyoid and laryngeal
`elevation and anterior movement during swallowing are
`thought to facilitate opening of the PES, we also anti-
`cipated a strong relationship between this swallowing
`event and the sEMG signal. Because muscles responsible
`for pharyngeal constriction are deeper than submental
`muscles, we anticipated that pharyngeal constriction
`during swallowing would demonstrate the weakest rela-
`tionship with the sEMG signal.
`
`Method
`Participants
`Seventeen healthy, adult volunteers participated in
`this study. The study group included 8 women and 9 men
`with an average age of 28.17 years (range = 21–39 years).
`All participants completed a health survey questionnaire
`to ascertain whether there was a history of swallowing
`difficulties or medical problems, or current medical prob-
`lems or medications that might influence swallowing func-
`tion. All participants signed an informed consent form.
`This study was approved by the Institutional Review
`Board and the Human Radiology Review Committee at
`the University of Florida Health Science Center.
`
`Procedures and Materials
`Participants stood during the study and were exam-
`ined within a fluoroscopic unit. Simultaneous video-
`fluoroscopic and sEMG data were collected using the Kay
`Swallowing Workstation. Videofluoroscopic images ob-
`tained at a rate of 30 frames per s from the lateral view
`
`were used to verify the presence of a swallow event and to
`calculate the timing parameters of each swallow. sEMG
`signals were obtained from a single three-point, circular,
`dry, disposable electrode with a 2.25-in. diameter. Each
`patch contained three electrodes in a triangle configu-
`ration. Two electrodes were recording electrodes and the
`third served as the ground. Interelectrode distance was
`0.25 in. edge-to-edge and 0.75 in. center-to-center. To
`facilitate consistent electrode placement across partic-
`ipants, the center point of each electrode patch was
`placed inferior to the hyoid bone in the midline, anterior
`neck. Recording electrodes were oriented toward the chin
`whereas the ground electrode was oriented toward the
`larynx (Figure 1). Obtained sEMG signals were processed
`by the Kay Swallowing Signals Lab, which was inter-
`faced with the Kay Elemetrics Swallowing Workstation.
`Sampling rate was 500 Hz and the raw signal was band-
`pass filtered (50–250 Hz), integrated (time constant =
`50 ms), and rectified. Simultaneous videofluoroscopic
`and sEMG examinations were recorded on formatted
`0.50-in. videotape using a video recorder. The video re-
`corder and the Swallowing Signals Lab are computer
`integrated within the swallowing workstation. The video
`recorder contains a vertical interval time-code generator
`and reader with video field resolution (.0167 s). The sys-
`tem allows physiologic and video data to be recorded and
`analyzed (postacquisition) concurrently. Software is able
`to read and link time-code data from the video signal and
`digital data from the Swallowing Signals Lab.
`
`Three swallows were analyzed from each partici-
`pant swallowing 5 mL of standard barium liquid con-
`trast for a total of 51 swallow events. Liquid barium
`contrast was provided to each participant from a med-
`icine cup. Participants were instructed to hold the liquid
`in their mouth and remain still until asked to swallow.
`The request to swallow was provided only when the
`G
`sEMG baseline activity returned to resting levels (
`4 mV
`root-mean-square) after the participant placed the liquid
`bolus in his or her mouth.
`
`Data Analysis
`Videofluoroscopic images were analyzed indepen-
`dent of the sEMG signal by an experienced judge (the
`third author) who was blinded to the sEMG signal. Three
`biomechanical aspects of each swallow were evaluated
`from the fluoroscopic image: hyoid elevation and return
`to baseline, pharyngeal constriction and return to base-
`line, and PES opening and closing. The baseline position
`prior to swallowing was established during quiet breath-
`ing (rest position) with the bolus held in the mouth. The
`video was viewed in slow motion until the posterior move-
`ment of the bolus in the mouth indicated the initiation
`of a swallow. Biomechanical events were measured in-
`dividually beginning with hyoid movement and followed
`
`188
`
`Journal of Speech, Language, and Hearing Research  Vol. 49  186–193  February 2006
`
`Petitioner - Avation Medical, Inc.
`Ex. 1035, p. 188
`
`

`

`Figure 1. Anterior and lateral views of electrode placement on anterior
`neck. Center point on the electrode patch was placed in the midline neck
`slightly inferior to the hyoid bone. Recording electrodes were oriented
`toward the chin (white leads). The ground electrode (black lead) was
`oriented toward the larynx.
`
`in order by pharyngeal constriction and PES opening.
`For each biomechanical component, the onset, peak, and
`offset of movement were identified from the fluoroscopic
`video. Onset was identified as the first video frame de-
`picting movement of that structure that continued into
`the swallow event. Peak movement was identified as the
`point of maximum excursion of each structure. For PES
`opening, peak excursion was determined visually as that
`time point in which the PES demonstrated its widest
`point of opening during bolus transit. The offset point
`was the video frame in which the structure returned to
`the resting, preswallow position.
`
`sEMG traces were also evaluated for onset, peak,
`and offset of activity during swallowing events. Each
`sEMG trace was evaluated independent of the video-
`
`fluoroscopic images by an experienced judge (the third
`author). Onset was identified as the point of upward
`excursion of the sEMG trace from resting baseline that
`led into the swallowing event. Peak was the highest
`amplitude point of the sEMG swallow trace. Offset was
`the point at which sEMG activity returned to baseline.
`
`Because subjective judgment was used for both
`videofluoroscopic and sEMG measures, interjudge reli-
`ability was estimated. To establish interjudge reliability
`of these measures, a second judge (the first author), who
`was blinded to the original results, measured the same
`parameters of 10 randomly selected samples from the
`51 total swallow events. As with the original measures,
`videofluoroscopic images and sEMG traces were eval-
`uated independently. Intraclass correlation coefficients
`were high for all comparisons (range of lower 95% con-
`fidence interval [CI] = .9989–1.000), suggesting strong
`consistency between raters.
`
`To compare specific time points (onset, peak, offset)
`across biomechanical events and with the sEMG signal,
`the respective time points for hyoid movement were set to
`Time 0. Differences between hyoid time points and other
`events were calculated from this zero point.
`
`Three additional temporal parameters were calcu-
`lated from the direct measures for each biomechanical
`event and from the sEMG trace. Onslope represented the
`time from the onset to the peak amplitude. Offslope rep-
`resented the time from the peak amplitude to the offset
`point. Total duration represented the time between the
`onset and offset points for each measure. The respective
`measures are depicted graphically in Figure 2.
`
`Correlation analyses (Pearson product–moment r
`and R2) were used to evaluate the relationship between
`
`Figure 2. Graphic depiction of timing measures obtained from surface
`electromyography signals and the fluoroscopic video of three biome-
`chanical swallowing events.
`
`Peak
`
`Onslope
`
`Offslope
`
`♦
`
`•
`
`fi Onset
`
`Offset
`
`•
`
`•
`
`Total Duration
`
`Crary et al.: Swallow Correlates of sEMG Signals 189
`
`Petitioner - Avation Medical, Inc.
`Ex. 1035, p. 189
`
`

`

`Table 1. Average onset, peak, and offset times relative to hyoid movement time points.
`
`Event
`
`Onset
`
`Peak
`
`Offset
`
`Onslope
`
`Offslope
`
`Total duration
`
`sEMG activity
`
`Hyoid movement
`
`–0.062
`(–0.132, 0.007)
`0
`
`–0.095
`(–0.144, –0.047)
`0
`
`–0.028
`(–0.107, 0.052)
`0
`
`Pharynx constriction
`
`PES Open/close
`
`0.141
`(0.077, 0.205)
`0.457
`(0.384, 0.530)
`
`0.126
`(0.095, 0.156)
`0.001
`(–0.027, 0.028)
`
`0.108
`(0.046, 0.171)
`–0.418
`(–0.484, –0.351)
`
`0.520
`(0.438, 0.572)
`0.572
`(0.500, 0.644)
`0.573
`(0.519, 0.627)
`0.123
`(0.111, 0.134)
`
`0.826
`(0.722, 00.931)
`0.740
`(0.667, 0.814)
`0.736
`(0.679, 0.793)
`0.345
`(0.332, 0.376)
`
`10.35
`(10.24, 10.45)
`10.31
`(10.22, 10.41)
`10.31
`(10.24, 10.38)
`00.477
`(0.455, –0.499)
`
`sEMG = surface electromyography; PES = pharyngoesophageal segment. Average onslope, offslope, and total durations are calculated
`Note.
`measures. Means and 95% confidence intervals (in parentheses) are presented. Data are presented in seconds.
`
`each of the three biomechanical events and the correspond-
`ing sEMG information. For these analyses, hyoid move-
`ment values were not set to Time 0. All statistical analyses
`were completed using SPSS software (Version 11.0).
`
`Results
`Descriptive results are presented as mean duration
`of each measure and the corresponding 95% CI (Table 1).
`On the basis of the mean data (Figure 3), sEMG activity
`preceded all biomechanical events for each time point
`(onset, peak, and offset). Hyoid movement appeared most
`closely related to the sEMG signal. On average, onset of
`pharyngeal constriction followed onset of hyoid eleva-
`tion, with PES opening occurring last. PES peak opening
`is nearly simultaneous with peak hyoid elevation and
`precedes the peak point for pharyngeal constriction. Off-
`set points are similar for hyoid elevation and the sEMG
`signal, but pharyngeal constriction follows these points
`and PES closure precedes each of the events.
`
`Calculated measures (onslope, offslope, and total du-
`ration) indicated that durations for sEMG activity, hyoid
`movement, and pharyngeal constriction were compara-
`
`Figure 3. Graphic depiction of timing results (onset, peak, offset) for the
`surface electromyography (sEMG) signal and three biomechanical events
`during the swallow. Data are mean values for each time point.
`
`Onset
`
`Peak
`
`Offset
`
`PES
`
`Pharynx
`
`Hyoid
`
`sEMG
`
`•
`
`A
`
`I.
`
`■IU
`
`•
`
`A
`
`I.
`
`•
`
`A
`
`.
`
`•
`
`-0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
`
`1
`
`1.1 1.2 1.3 1.4
`
`Time (sec)
`
`ble with offslope being longer than onslope in each case.
`Average total durations for these three measures were
`nearly identical. In contrast, PES calculated durations
`were shorter than all other measures. PES onslope was
`approximately 75% shorter than the other measures.
`PES offslope and total durations were approximately
`50%–60% shorter than other measures.
`
`Correlations and corresponding variance estimates
`between each biomechanical event and the sEMG signal
`were high for onset, offset, and peak measures (Table 2).
`Calculated duration measures of onslope, offslope, and
`total duration did not result in the same strength of re-
`lationship. Overall, hyoid movement was most strongly
`related to the sEMG signal.
`
`Discussion
`The empirical and clinical uses of sEMG analyses
`of swallowing activity are dependent on a basic under-
`standing of the relationship between the sEMG signal
`and swallow-related movements. Results from this study
`identified hyoid bone movement as the biomechanical
`swallowing event most closely linked to the sEMG sig-
`nal. This link is logical because muscles responsible for
`hyoid movement are superficial and close to the region of
`measurement from the surface electrodes. In addition,
`because hyoid movement is among the earliest of bio-
`mechanical events during the swallow, the sEMG signal
`may have been dominated by this activity, overshadow-
`ing other swallow movements. In the normal swallow,
`hyoid movement is closely tied to laryngeal movement.
`Thus, the primary result of the present study extends
`previous findings of Ertekin and colleagues (1995, 1997)
`and Ding et al. (2002): Both hyoid and laryngeal move-
`ments are closely related to sEMG signals obtained from
`the upper anterior neck midline.
`
`The sequence of biomechanical movements com-
`pared with the sEMG signal is in general agreement
`with earlier EMG findings of Perlman and colleagues
`
`190
`
`Journal of Speech, Language, and Hearing Research  Vol. 49  186–193  February 2006
`
`Petitioner - Avation Medical, Inc.
`Ex. 1035, p. 190
`
`

`

`Table 2. Results of correlation analyses between sEMG and biomechanical data points.
`
`sEMG measure
`
`Biomechanical event
`
`Onset
`
`Peak
`
`Offset
`
`Onslope
`
`Offslope
`
`Total duration
`
`Hyoid
`
`Pharynx
`
`PES
`
`1.00
`.9998
`1.00
`.9997
`1.00
`.9997
`
`1.00
`.9999
`1.00
`.9999
`1.00
`.9999
`
`1.00
`.9997
`1.00
`.9996
`1.00
`.9997
`
`.502
`.2516
`.299
`.0894
`–.358
`.1280
`
`.524
`.2742
`ns
`
`ns
`
`.539
`.2908
`ns
`
`–.345
`.1188
`
`sEMG = surface electromyography. The top number represents the correlation value (r) and the bottom number
`Note.
`represents the accountable variance (R2). ns = not significant; all other values are significant at p G .05.
`
`(1999), who reported that the submental sEMG signal
`preceded muscle activation in the superior pharyngeal
`constrictor, thyroarytenoid, interarytenoid, and crico-
`pharyngeus muscles. Present findings on sequence of
`activities also concur with sEMG sequences reported by
`McKeown et al. (2002). Specifically, submental muscle
`activation precedes laryngeal elevation, whereas crico-
`pharyngeus activation follows laryngeal elevation. In
`the present study, sEMG activation preceded initiation
`of hyoid bone elevation, followed in order by onset of
`pharyngeal constriction and then PES opening. Kendall,
`Leonard, and McKenzie (2003) noted a similar sequence.
`They reported that peak hyoid elevation typically pre-
`cedes peak PES opening (mean differences between
`0.04 s to 0.10 s depending on bolus volume). Our results
`support their findings but suggest a closer temporal re-
`lationship between peak hyoid elevation and PES open-
`ing (mean difference of 0.001 s). This discrepancy may
`have resulted from different bolus volumes used between
`the two studies (5 mL in the present study vs. 1, 3, and
`20 mL in Kendall’s study). In fact, the smallest difference
`between peak hyoid elevation and maximum PES open-
`ing in Kendall et al.’s study was identified for the 3-mL
`bolus that was closest to the 5-mL bolus volume used in
`the present study.
`
`The calculated measures of onslope, offslope, and
`total duration did not result in the same strength of as-
`sociation as the specific point measures (onset, peak, off-
`set). This result seems to reflect increased intersubject
`variance in the calculated measures that are perhaps
`inherent to timing pattern differences among the respec-
`tive biomechanical components of the swallow. In prior
`studies, researchers have commented on extensive inter-
`subject variability in muscle activation patterns (Gay
`et al., 1994; Spiro et al., 1994; Palmer et al., 1999; Perlman
`et al., 1999) and in biomechanical events (Kendall, 2002).
`In the present study, substantial intersubject variability
`was reflected in the 95% CIs obtained for both the sEMG
`and biomechanical results. Collectively, these findings
`
`point to variation in muscle activation patterns and the
`resulting biomechanical movements in normal swallow-
`ing, even when the swallow is limited to a single material
`and volume. Thus, although the individual time-point
`measures of onset, peak, and offset varied systematically
`with corresponding measures on the sEMG graphic
`trace, the relative timing between these point measures
`appears to have varied substantially, thus decreasing the
`strength of the obtained correlations.
`
`One clinical application of sEMG techniques has
`been to provide patients with biofeedback regarding
`muscle activation during swallowing attempts (Bryant,
`1991; Crary, 1995; Crary et al., 2004; Huckabee &
`Cannito, 1995). Results from the present study suggest
`that such approaches would be helpful in treating
`patients with dysphagia, whose problems reside in the
`pharyngeal aspects of swallowing, specifically hyoid ele-
`vation, and by extension, laryngeal elevation. However,
`in our study, specific time-point measures associated
`with both pharyngeal constriction and PES opening–
`closing were strongly associated with concurrent mea-
`sures obtained from the sEMG signal. This observation
`suggests that some characteristics of these biomechan-
`ical events may be incorporated within sEMG biofeed-
`back provided to patients with pharyngeal dysphagia. In
`addition, hyoid elevation has been linked to PES opening
`(Goyal, 1984; McConnel, 1988). Therefore, biofeedback
`that facilitates increased hyoid elevation during swallow-
`ing attempts may facilitate PES opening as well.
`
`The present study was limited to evaluation of a
`single bolus size and consistency swallowed by a cohort of
`young, healthy adults. Accommodation is a well-known
`aspect of swallowing activity in which swallow physiol-
`ogy adapts to materials that are swallowed. Leonard,
`Kendall, McKenzie, and Gonc¸alves (2000) identified volume-
`specific adjustments in hyoid elevation, pharyngeal con-
`striction, and PES opening but not in hyoid-to-larynx
`approximation in healthy adults. Thus, volume is
`one variable that may differentially impact swallowing
`
`Crary et al.: Swallow Correlates of sEMG Signals 191
`
`Petitioner - Avation Medical, Inc.
`Ex. 1035, p. 191
`
`

`

`biomechanics. Given the impact of what may be selective
`accommodation within the normal swallowing mecha-
`nism, the current biomechanical evaluation should be re-
`assessed in consideration of characteristics that may alter
`swallow physiology. Furthermore, future studies should
`incorporate additional biomechanical events closely tied
`to hyoid movement (e.g., laryngeal movement), to chal-
`lenge and/or expand the current results. Last, both
`age and disease impact swallow physiology. Thus, it
`is likely that the interpretation of sEMG correlates
`of swallowing activity also will be impacted by these
`factors. It will be important to replicate these findings
`in adults across the age span and in patients with dys-
`phagia resulting from different disease processes. Com-
`parison of results addressing these factors with the
`present results obtained from healthy, young adults
`will help to expand researchers’ understanding of the
`relationship between swallow biomechanics and sEMG
`signals obtained during swallowing.
`
`Results of the present study indicate a strong as-
`sociation between the sEMG signal and certain biome-
`chanical events occurring during swallowing, specifically
`hyoid movement. These results are encouraging but
`should be viewed as an initial step toward enhanced
`use of sEMG technology in the areas of swallowing phys-
`iology and the treatment of swallowing disorders. Im-
`proved understanding of the contribution of sEMG
`technology to the study and rehabilitation of swallowing
`activity will advance research and clinical applications
`of this technology.
`
`References
`
`Bryant, M. (1991). Biofeedback in the treatment of a selected
`dysphagic patient. Dysphagia, 6, 140–144.
`
`Crary, M. A. (1995). A direct intervention program for
`chronic neurogenic dysphagia secondary to brainstem
`stroke. Dysphagia, 10, 6–18.
`
`Crary, M. A., Carnaby (Mann), G. D., Groher, M. E., &
`Helseth, E. (2004). Functional benefits of dysphagia
`therapy using adjunctive sEMG biofeedback. Dysphagia,
`19, 160–164.
`
`Ertekin, C., Pehlivan, M., Aydogdu, I., Ertas, M.,
`Uludag, B., & Celebi, G. (1995). An electrophysiological
`investigation of deglutition in man. Muscle & Nerve,
`18, 1177–1186.
`
`Gay, T., Rendell, J. K., & Spiro, J. (1994). Oral and
`laryngeal coordination during swallowing. Laryngoscope,
`104, 341–349.
`
`Goyal, R. K. (1984). Disorders of the cricopharyngeus
`muscle. Otolaryngology Clinics of North America,
`17, 115–130.
`
`Haynes, S. N. (1976). Electromyographic biofeedback treat-
`ment of a woman with chronic dysphagia. Biofeedback
`and Self Regulation, 1, 121–126.
`
`Huckabee, M. L., & Cannito, M. P. (1995). A direct
`intervention program for chronic neurogenic dysphagia
`secondary to brainstem stroke. Dysphagia, 10, 6–18.
`
`Kendall, K. A. (2002). Oropharyngeal swallowing variabil-
`ity. Laryngoscope, 112, 547–551.
`
`Kendall, K. A., Leonard, R. J., & McKenzie, S. W. (2003).
`Sequence variability during hypopharyngeal bolus transit.
`Dysphagia, 18, 85–91.
`
`Leonard, R. J., Kendall, K. A., McKenzie, S. W., &
`Gonc¸ alves, M. I. (2000). Structural displacements in
`normal swallowing: A videofluoroscopic study. Dysphagia,
`15, 146–152.
`
`McConnel, F. (1988). Analysis of pressure generation and
`bolus transit during pharyngeal swallowing. Laryngoscope,
`98, 71–78.
`
`McCulloch, T. M., Perlman, A. L., Palmer, P. M., &
`VanDaele, D. J. (1996). Laryngeal activity during
`swallow, phonation, and the Valsalva maneuver: An
`electromyographic analysis. Laryngoscope, 106,
`1351–1358.
`
`McKeown, M. J., Torpey, D. C., & Ghem, W. C. (2002).
`Non-invasive monitoring of functional distinct muscle
`activations during swallowing. Clinical Neurophysiology,
`113, 354–366.
`
`Palmer, P. M., Luschei, E. S., Jaffe, D., & McCulloch,
`T. M. (1999). Contributions of individual muscles to the
`submental surface electromyogram during swallowing.
`Journal of Speech, Language, and Hearing Research, 42,
`1378–1391.
`
`Perlman, A. L., Palmer, P. M., McCulloch, T. M., &
`VanDaele, D. J. (1999). Electromyographic activity from
`human laryngeal, pharyngeal, and submental muscles
`during swallowing. Journal of Applied Physiology, 86,
`1663–1669.
`
`Ding, R., Larson, C. R., Logemann, J. A., & Rademaker,
`A. W. (2002). Surface electromyographic and electroglotto-
`graphic studies in normal studies under two swallow
`conditions: normal and during the Mendelsohn maneuver.
`Dysphagia, 17, 1–12.
`
`Shaker, R., Easterling, C., Kern, M., Nitschke, T.,
`Massey, B., Daniels, S., et al. (2002). Rehabilitation of
`swallowing by exercise in tube-fed patients with pharyn-
`geal dysphagia secondary to abnormal to UES opening.
`Gastroenterology, 122, 1314–1321.
`
`Doty, R. W., & Bosma, J. F. (1956). An elecromyographic
`analysis of reflex deglutition. Journal of Neurophysiology,
`19, 44–60.
`
`Ertekin, C., Aydogdu, I., Yuceyar, N., Pehlivan, M.,
`Ertas, M., Uludag, B., & Celebi, G. (1997). Effects of
`bolus volume on oropharyngeal swallowing: An electro-
`physiologic study in man. American Journal of Gastro-
`enterology, 92, 2049–2053.
`
`Shaker, R., Kern, M., Bardan, E., Taylor, A., Stewart,
`E. T., Hoffman, R. G., et al. (1997). Augmentation of
`deglutitive upper esophageal sphincter opening in the
`elderly by exercise. American Journal of Physiology, 272,
`G1518–G1522.
`
`Spiro, J., Rendell, J. K., & Gay, T. (1994). Activation and
`coordination patterns of the suprahyoid muscles during
`swallowing. Laryngoscope, 104, 1376–1382.
`
`192
`
`Journal of S