Upper esophageal sphincter opening and modulation during swallowing

GASTROENTEROLOGY 1989;97:1469-78

Upper Esophageal Sphincter Opening and Modulation During Swallowing P. JACOB, P. J. KAHRILAS, J. A. LOGEMANN, V. SHAH, and T. HA Departments of Medicine and Communication Sciences and Disorders,Northwestern University and Veterans Administration Lakeside Medical Center, Chicago,Illinois

Studies were done on 8 normal subjects with synchronized videofluoroscopy and manometry to facilitate a biomechanical analysis of upper esophageal sphincter opening and volume-dependent modulation during swallowing. Movements of the hyoid and larynx, dimensions of sphincter opening, and intraluminal sphincter pressure were deterintervals during swallows of 1,5, mined at 1/3Oth-s 10, and 20 ml of liquid barium. Our analysis subdivided upper esophageal sphincter activity during swallowing into five phases: (a) relaxation, (b) opening, (c) distention, (d) collapse, and (e) closure. Sphincter relaxation occurred during laryngeal elevation and preceded opening by a mean period of 0.1s.Opening occurred as the sphincter was pulled apart via muscular attachments to the hyoid such that the hyoid coordinates at which sphincter opening and closing occurred were constant among bolus volumes. Sphincter distention after opening was modulated by intrabolus pressures rather than graded hyoid movement. The generation of intraboIus pressure coincided with the posterior thrust of the tongue that culminated in pharyngeal wall contact and the initiation of pharyngeal peristalsis. Larger volume swallows were associated with greater intrabolus pressure and increased bolus head velocity. The duration of sphincter opening increased in conjunction with a prolongation of the anterior-superior excursion of the hyoid and a delay in the onset of pharyngeal peristalsis (the event that determined the timing of sphincter closure). We conclude that transsphincteric transport of increasing swallow bolus volumes is accomplished by modulating sphincter diameter, opening interval, and flow rate (reflected by bolus head velocity). Furthermore, upper esophageal sphincter opening is an active mechanical event rather than simply a consequence of cricopharyngeal relaxation.

he upper esophageal sphincter (UES) is a complex musculoskeletal structure that constitutes the pharyngoesophageal junction. Viewed in cross section, the anterior wall of the sphincter consists of the lamina of the cricoid cartilage with the major muscular component of the sphincter, the cricopharyngeus muscle, attached to the lateral aspects of the arch of the cricoid cartilage. Thus, when closed, the sphincter has a slitlike configuration and, because of the attachment to the cricoid cartilage, the cricoid and the cricopharyngeus muscle are obliged to move in unison. Manometric and electromyographic studies further suggest that the inferior portion of the pharyngeal constrictor and the superior cuff of the esophagus functionally contribute to sphincteric function (1). A recent study analyzing UES activity during swallowing in normal subjects suggested that sphincter relaxation and sphincter opening were independent, although coordinated events (2). Relaxation preceded opening, which appeared to be an active event occurring in the course of laryngeal elevation. Furthermore, both the duration of UES opening and the anterior-posterior dimension of the UES during opening were bolus volume-dependent such that larger boluses resulted in longer opening periods and increased sphincter diameter. The mechanism of UES opening and the observed volume-related modulation of dimensions during swallowing are incompletely understood. Potential mechanisms include variations in hyoid-mediated traction or in bolus-mediated pulsion. Studies in the opossum have shown that although the cricopharyngeus is the major contributor to UES pressure, abolition of cricopharyngeal contractile activity by

T

Abbreviation

sphincter. 0 1989 by

used in this paper:

UES,

the American Gastroenterological 09X6-5085/091$3.50

upper esophageal Association

1470 JACOBETAL.

nerve sectioning or d-tubocurarine did not completely eliminate UES pressure (1). Contraction of the geniohyoid muscle during the period that the cricopharyngeus muscle was relaxed abolished the remaining UES pressure. Furthermore, electromyographic activity was recorded from the geniohyoid muscle during UES opening, suggesting that sphincter opening was an active process resulting from anterior distracting forces transmitted through the hyoid. Preliminary radiographic observations in humans suggest that bolus volume-dependent alterations in hyoid movement pattern may be partly responsible for the volume-dependent modulation of UES opening (3). The present study was undertaken to investigate these relationships and to detail the biomechanical events leading to UES opening and volume-dependent modulation during swallowing.

Materials and Methods Concurrent videofluoroscopic and manometric studies of swallowing were obtained in 8 healthy male volunteers, 22-28 yr of age, without past or current swallowing problems. Studies were done in the late afternoon, at least 4 h after a meal. The study protocol was approved by the Northwestern University Institutional Review Board. During recording sessions, subjects were seated in a chair fitted with a headrest similar to that of a dental chair that minimized head movement throughout the study. Lateral videofluoroscopic studies were done such that the radiographic image included the tip of the tongue anteriorly, the hard palate superiorly, and the subglottic air column inferiorly (4). A lead letter “x” was taped to the skin over the vertebral column and included in the fluoroscopic field, to serve as the stationary origin of the image-based x-y coordinate system used in the analysis of swallow-related movements. The fluoroscopic image was both displayed on a monitor and recorded with a videocasette recorder (Sony U-matic, model VO-5600).Two swallows each of l-,5-,lo-,and ZO-ml of liquid barium were obtained in each subject. Liquid barium was placed in the mouth with a syringe and subjects were instructed to hold the barium over the tongue until the swallow command at which time they should swallow it as a single bolus. We have found 20 ml to be the largest volume consistently swallowed by subjects as a single bolus. In instances that the subject double swallowed or that technical problems occurred, a repeat swallow of the same volume was obtained. Intraluminal manometry was done with a miniature strain gauge assembly that was ovoid in cross section (3 x 5 mm) and incorporated three pressure sensors spaced 3 cm apart with identical radial orientation such that they faced one of the flat sides (Medical Measurements Inc., Hackensack, N.J.). The strain gauge sensors were housed in asymmetric metal capsules that enabled radiographic identification of both the axial position and the radial orientation. The manometric assembly was passed nasally

GASTROENTEROLOGYVol. 97,No.6

after application of 2% xylocaine jelly to the nostril and positioned with the sensors facing posteriorly such that the middle sensor was 5 mm proximal to the subglottic air column. This position enabled recording of UES contractile activity during swallowing by the middle sensor (2) while placing the proximal sensor in the hypopharynx and the distal sensor in the cervical esophagus. Sensor positioning was checked fluoroscopically throughout the study period. Pressure tracings were displayed on a polygraph (model R-611; Beckman Instruments, Oxnard, Calif.) with the sensitivity set at 10 mmHg/cm and the chart speed at 50 mm/s during swallows. Manometric and videofluoroscopic records were synchronized using a modified videotimer (model VC 436; Thalner Electronics Laboratories, Ann Arbor, Mich.). The timer both encoded an analogue time signal on the videotape in hours, minutes, seconds, and hundredths of a second and sent a 5-ms pulse to the polygraph at whole second intervals. The timing pulse was inscribed on the manometric record using an instrumentation channel of the recorder. A numeric notation from the video display was written on one of the second marks during each swallow sequence, permitting subsequent temporal correlation between fluoroscopic and manometric records. Pressure values corresponding to fractional seconds were derived by interpolation between the second marks. Data analysis encompassed both temporal analysis of the videofluoroscopic and manometric records and a spatial analysis of the videofluoroscopic images. For all analysis, the time base for each swallow was arbitrarily defined such that the first video frame showing UES opening (defined as the first frame showing either barium or air within the UES) was time zero. This adjustment of the time base facilitated comparison of events among swallows with specific attention to the relationship of those events to UES opening. Without such a time base adjustment, similar events appeared staggered in time among swallows as they occurred at variable intervals from an earlier time reference such as the onset of bolus movement in the mouth or the initiation of hyoid elevation. The videofluoroscopic images of each swallow were initially analyzed during slow motion playback to obtain the following timing information: (a) the time of arrival of the head (leading edge) and the tail (trailing edge) of the barium bolus at each pressure sensor and (b) the timing of swallow-related tongue movements (onset and offset of ramp configuration seen early in the pharyngeal swallow as well as the onset of the subsequent posterior movement of the tongue base that terminated with contact of the tongue base to the pharyngeal wall) (5). Manometric tracings were analyzed by visual inspection and UES pressure was determined at 1/3Oth second intervals, corresponding to the times of the individual videofluoroscopic frames. The onset of UES relaxation was defined as the time at which UES pressure fell to a value of 0 mmHg. All pressure measurements were referenced to atmospheric pressure, as determined by the most proximal sensor in the air-filled hypopharynx between swallows. The nadir UES pressure was the lowest pressure value measured during the relaxation interval. The onset of pharyngeal peristalsis was defined as the beginning of

UPPER ESOPHAGEAL SPHINCTER OPENING 1471

December 1989

rapid pressure increase recorded from the sensor located in the hypopharynx. The velocity of pharyngeal peristalsis was calculated by dividing the 3-cm distance between the proximal and middle manometric sensors by the time interval between the onset of peristalsis at these two sites. Similarly, bolus head velocity was determined by dividing the 3-cm distance between the proximal and middle manometric sensors by the time of arrival of the bolus head at these two sites as seen fluoroscopically. Spatial analysis of the videofluoroscopic swallowing sequences was accomplished using an interactive computer program written to enable x-y coordinate determination of selected structures on each video frame (6). For each swallow 30-60 sequential frames (at 1/3Oth-s intervals), selected so as to encompass critical events pertaining to UES opening and closing during the swallow, were analyzed. Each of these images was digitized using an IBM PC-AT computer equipped with an image digitization board (Data Translation Frame Grabber, model DT 2851; IBM, Marlboro, Mass.). The digitized image had a configuration of 512 x 512 pixels with 8 bits per pixel, thereby allowing for 256 increments on a gray scale. The program allowed the user to position a cross-hair cursor on anatomic points of interest on each video frame and then mark these points for subsequent calculations. For this analysis, the following points were marked: (a) the anchor point (center of the lead letter “x”), (b) the posterior-superior corner of the proximal and middle manometric sensor casings, (c) the anterior-superior corner of hyoid bone, (d) the posterior-superior corner of the subglottic air column, and (e) the anterior and posterior walls of the UES (these points were the same when the sphincter was closed). The axial position of the UES was uniformly marked 1 cm distal to the subglottic air column corresponding to the typical location of the center of the LIES high-pressure zone (2). After the marking of data points on the digitized images, coordinate calculations were computed. The coordinate system was referenced to the anchor point (point 0, 0) and used the axis defined by the manometric sensor as the y-axis. Fluoroscopic magnification was corrected for by using the known distance of 3 cm between the first and second manometric sensors as a yardstick positioned within the sagittal plane. Thus, the result of this analysis yielded the x and y coordinates of each data point (in millimeters) on each digitized frame of the swallow sequence corrected for magnification, head tilt, and head movement. Numerical data among swallows were averaged and expressed as the mean ? SEM unless otherwise specified. Data among volumes were compared by paired t-testing, analysis of variance, linear regression analysis, or multiple regression analysis (77, depending on the data configuration. Regardless of the statistical tests employed, p values of ~0.05 were regarded as significant.

Results Upper

opening

Esophageal

Sphincter

Opening

Profile

Figure 1 depicts the timing and extent of UES during swallows of graded bolus volume.

1. , , , , , , , , . , . , , . , -O.‘l

0

0:1

0.i

, ,, ,, ,. ,

0.3 0.4

0:5

016

0:7

Tlmeh&condm

Figure 1. Upper esophageal sphincter opening as a function of bolus volume. Time zero represents the first video frame showing opening and distance zero represents the initial horizontal position of the sphincter relative to the anchor point before sphincter opening. Anterior opening occurred as the horizontal distance between the anterior sphincter wall and the anchor point increased, and posterior opening occurred as the horizontal distance between the posterior sphincter wall and the anchor point diminished. Values shown are mean f SEM of two swallows for each of 8 subjects. Both the maximal anterior and posterior components of opening as well as the duration of opening increased significantly proportionate to the bolus volume (analysis of variance, p < 0.01). The time-courses of anterior and posterior opening were different in that anterior opening occurred rapidly, peaking within 0.13 s of sphincter opening, whereas posterior opening peaked later, suggesting that the two components of opening were attributable to distinct mechanisms.

As summarized in Table 1, both the maximal extent of sphincter opening and the mean duration of sphincter opening varied significantly with bolus volume. These differences were further accentuated

Table

1. Volume Effect on Parameters of Upper Esophageal Sphincter Opening Bolus volume

Opening duration (sY Maximal diameter (mm)” Integral of opening diameter Immsecl”

1 ml

5 ml

10 ml

20 ml

0.34 ? 0.02

0.45 2 0.01

0.50 ‘- 0.02

0.54 f 0.03

5.6 + 0.4

8.5 2 0.5

10.5 " 0.5

11.8 k 0.4

1.67 f 0.14

3.13 + 0.23

4.29 f 0.34

5.33 * 0.53

Values are mean variance.

f

SEM. n p < 0.001, one-way

analysis

of

1472

JACOB ET AL.

GASTROENTEROLOGY Vol. 97, No. 6

by deriving the integral of the opening diameter, expressed in mmsec of opening, for each volume. As indicated in Table 1, the integral of the opening diameter varied more than threefold over the range of bolus volumes used. Figure 1 subdivides sphincter opening into a component achieved as the horizontal coordinate of the anterior sphincter wall moved away from the horizontal coordinate of the anchor point (anterior opening), and a component achieved as the horizontal coordinate of the posterior sphincter wall moved closer to that of the anchor point (posterior opening). The greater component of sphincter opening occurred as the anterior wall of the sphincter moved forward and this achieved a maximum value within the first 0.13s of opening. A lesser component of sphincter opening was attributable to posterior opening. The volume dependency of posterior opening was less obvious than that of anterior opening, with differences most apparent late in the opening period. The profile of the posterior opening component was also different than that of the anterior component in that it peaked later and persisted longer into the opening period, suggesting that the mechanism responsible for posterior opening was different from that accounting for the anterior component. Both the magnitude and the duration of the individual components increased significantly with increasing bolus volumes (p < 0.01, one-way analysis of variance). Figure 2 is an example of the most commonly observed pattern of intraluminal UES pressure recording during the course of a barium swallow. Analysis of the tracing, in conjunction with the data depicted in Figure 1, suggested that sphincter activity during swallowing could be subdivided into five phases: (a) manometric relaxation, (b) sphincter opening, (c) sphincter distention during which time the bolus head traversed the sphincter, (d) sphincter collapse, and (e) sphincter contraction occurring with the arrival of pharyngeal peristalsis. Each of these phases of sphincter activity will be analyzed in conjunction with relevant corresponding data on movements of the hyoid, larynx, and tongue. Swallow-Related

Hyoid Movement

There was pronounced superior and anterior excursion of the hyoid during swallows. Figure 3

illustrates the x-y coordinates of the hyoid at 1/30th-s intervals during the course of l- and lo-ml barium swallows. Although Figure 3 accurately depicts the pattern of swallow-related hyoid movement, the time-course of movement is obscured by overlapping points. This limitation is clarified in Figures 4 and 5, which depict the timing and extent of the horizontal and vertical vectors of the hyoid movement for

rmn HQ.

I3

c

sobs Heed 2767

SQkM Tal 28.30

fl

10 o L ___-___&L_.J---__ -10 UES’ /j ReQxauorl -20 t 2780

27.00 8

--

\

-1oL

PheMmeal Per*& 2Sao t

-

‘UES %z2

a 28.60

28.00 0

Figure 2. Representative manometric tracings from the hypopharynx and UES during a Z&ml barium swallow. The bottom tracing shows the second markings from the video timer that facilitated synchronization with the videofluoroscopic record. Time notations regarding the bolus position and the occurrence of UES opening were derived from examination of the videofluoroscopic record. Note that UES relaxation preceded opening by 0.1 s and that UES intraluminal pressure increased to about 12 mmHg after opening, during sphincter distention.

graded barium boluses. The initial hyoid movement was predominantly superior beginning about 0.4 s before sphincter opening, whereas the anterior movement began 0.2-0.3 s later. Although sphincter

relaxation began early in the course of anterior hyoid movement, sphincter opening occurred only after substantial anterior and superior excursion. In fact, as evident from Figures 3-5, sphincter opening and closing occurred at very nearly the same x-y coordinates of the hyoid, regardless of bolus volume. Sphincter opening occurred after a mean anterior hyoid excursion of 11.3 mm (range 10.9-12.5 mm) and a mean vertical excursion of 19 mm (range 17-20 mm). Larger volume swallows were characterized by persistence of the hyoid above and anterior to the coordinates at which opening occurred for proportionately longer periods. Table 2 compares the duration of anterior and vertical hyoid movements with the UES opening interval. The duration of anterior hyoid excursion was defined as the inter-

UPPER ESOPHAGEAL SPHINCTER OPENING

December 1989

1 MI Barium Swallows

1473

10 MI Barium Swallows

Anterior Movement (mm) Figure 3. Movement patterns of the hyoid during l- and lo-ml swallows (mean values of 8 subjects, two swallows each). Each circle represents the hyoid position during a single video frame (1/3Oth-s intervals] and the arrows indicate the direction of movement. Open circles denote frames during which the sphincter was closed, filled circles denote frames during which the sphincter was open, and gray circles denote frames during which the sphincter was variably open depending on the subject. Note that sphincter opening and closing occurred at nearly identical hyoid positions among subjects and among volumes. The larger volume swallows were associated with persistence of the hyoid superior to and anterior to the opening coordinates.

val during which the hyoid remained anterior to its opening coordinates and the duration of superior excursion as the interval during which it remained superior to its opening coordinates. The duration of both the anterior and superior excursions correlated with the duration of UES opening (r = 0.78, p < 0.01 for anterior and r = 0.64, p < 0.01for vertical hyoid movement).

Although the duration of hyoid movement was significantly influenced by bolus volume, the extent of movement was minimally affected. Maximal anterior hyoid excursion was 15 mm for all bolus volumes and, as evident in Figure 4, the excursion patterns for the different bolus volumes completely overlapped each other. On the other hand, maximal vertical hyoid excursion did tend to increase with mm

I

251 25

ol.........-.:..-.-...-..!......-.. I

oi-..-.-. -0.4

-0.4

...:-----..---.!.....,---0.2

0.0

0.2

0.4

0.6

Time (seconds)

Figure

4. The anterior vector of hyoid movement during I-, 5, lo-, and ZO-ml swallows (mean t SEM values of 8 subjects, two swallows each]. Hyoid position at the time of sphincter opening is shown by the vertical dotted line at time zero. Position zero on the vertical axis for each subject was the most posterior hyoid position observed in that subject during the entire study. The open circles on each tracing show the hyoid position at the time of sphincter closing. As evidenced from the extensive overlap among these tracings, the anterior vector of hyoid movement was very similar among volumes and achieved an identical maximal excursion (15 mm) for all volumes. Error bars shown are representative.

-0.2

0.0

0.2

0.4

0.6

‘Nme (seconds)

Figure 5. The superior vector of hyoid movement plotted as in Figure 4 during l-, 5-, and lo-ml swallows. The 20-ml curve was omitted because it nearly completely overlapped the lo-ml curve. Sphincter opening occurred at the 20-mm vertical position regardless of bolus volume. The maximal extent of superior hyoid movement was significantly greater for lo- and 20-ml swallows compared with l-ml swallows (p < 0.05, paired t-test). The open circles indicate the timing of the onset of pharyngeal peristalsis at the hypopharyngeal strain gauge sensor for each volume. Note that pharyngeal peristalsis was initiated just before hyoid descent, suggesting that prolonged hyoid excursion was associated with a delay in the initiation of pharyngeal peristalsis and, consequently, UES closure. Error bars shown are representative.

1474

GASTROENTEROLOGY Vol. 97, No. 6

JACOBET AL.

Table 2. Correlation Between the Duration of Hyoid Excursion and the Interval of Upper Esophageal Sphincter

Opening Bolus volume

Opening duration (s) Anterior excursion (s) Superior excursion (s)

5 ml

10 ml

0.34 k 0.02

0.45 k 0.01

0.50 + 0.02

0.54 f 0.03

0.41 r 0.05

0.52 f 0.03 0.31 !z0.05

0.62 2 0.03 0.49 2 0.07

0.65 2 0.05 0.54 * 0.04

0.20 + 0.04

Values are mean 2 SEM. ’ Linear regression

analysis, correlating

greater volumes: 20 mm for l-ml, 20.5 mm for 5-m], 22.5 mm for lo-ml, and 23 mm for 20-ml swallows [these values were not significant when analyzed by an analysis of variance but a paired t-test showed a significant difference between l-ml and both lo- and 20-ml values (p < 0.05)]. As shown in Figure 4, rapid anterior hyoid movement began 0.10-0.16s before UES opening. Manometrically, this preopening interval corresponded to the 0.1-sinterval following UES relaxation characterized by a period of increasingly negative intraluminal pressure. The nadir UES pressure ranged from a mean of -11 mmHg for l-ml swallows to -17 mmHg for 20-ml swallows. Sphincter opening was associated with a sudden increase in the intraluminal UES pressure to a value of -0 (ranging from a mean of -1 mmHg for l-ml swallows to +5 mmHg for 20-ml swallows). Figure 6 correlates traction on the anterior wall of the UES (expressed as the distance between the anterior wall of the UES and the -Hg 40

20 ml

1 ml

. Hyold-UE8 Distance

mm

ouEsPressure

22

r (PI” 0.76 (0.001) 0.64 (0.001)

with opening duration.

hyoid) with the intraluminal UES pressure during the preopening interval for IO-ml barium swallows. Hyoid traction on the sphincter and intraluminal pressure showed an inverse relationship that correlated well both in timing and in magnitude (r = 0.84, p < 0.01, multiple regression analysis for all four bolus volumes), suggesting that the negative pressure was attributable to suction created by the traction on the anterior wall of the sphincter. lntrabolus

Pressure

As described above, nadir UES pressure increased abruptly at the time of sphincter opening, reaching levels close to 0 mmHg at the time of the first video frame showing sphincter opening. As

mmHg -I

a 20 '4 B k

P fE20

16

8

B 16

14 -20 -0.2

-0.1 Timc(mccon~)

Figure

I e

12 0.0

-10

ogd

6. The relationship between intraluminal UES pressure during the 0.2-s interval before sphincter opening and traction applied to the anterior wall of the sphincter during lo-ml swallows (mean k SEM 8 subjects, two swallows each]. The distance between the hyoid and the anterior wall of the sphincter increased progressively during this interval until the anterior wall of the sphincter snapped forward. The hyoid-to-sphincter distance correlated inversely with UES pressure (r = 0.84, p < 0.01, multiple regression analysis for all four bolus volumes), suggesting that hyoid-mediated traction pulled the sphincter open.

Figure

7. Intrabolus pressure within the UES during the period of sphincter distention immediately after opening during l-, 5-, lo-, and ZO-ml swallows (mean + SEM, 8 subjects, two swallows each). The negative intraluminal pressure recorded before UES opening increased abruptly with opening. Both the magnitude and the duration of the positive intrabolus pressure varied directly with bolus volume (p < 0.01, analysis of variance).

UPPER ESOPHAGEAL SPHINCTER OPENING

December 1989

-2

10

6

2

xdrabol~PreuPn(mmagl

Figure

8. Correlation between intrabolus pressure and sphincter diameter after opening (mean -C SEM, 8 subjects, two swallows of each volume). The parameter of sphincter diameter depicted is that achieved by forward excursion of the anterior wall. The strong correlation (r = 0.93, p < 0.001, linear regression analysis) suggests that observed differences in sphincter diameter were largely determined by distensive intrabolus force. The low range of intrabolus pressures observed at all volumes implies that the sphincter is extremely compliant when relaxed within this range of swallow volume.

illustrated in Figure 2, this was followed by an interval of positive pressure corresponding to the period of sphincter distention after opening. Figure 7 shows the mean intrabolus pressures during the distention phase of sphincter opening for each bolus volume. Both the magnitude and duration of intrabolus pressure increased with increasing bolus volume, ranging from a maximal value of 1.5 mmHg for l-ml boluses to 9.9 mmHg for 20-ml boluses (p < 0.01, analysis of variance). Figure 8 correlates the mean intrabolus pressure during the distention phase of opening with the anterior dimension of UES opening for the four bolus volumes. As shown in Figure 8, the correlation between the two was highly significant (r = 0.93, p < O.OOl), suggesting that the diameter achieved during the distention phase was related to intrabolus forces acting from within the sphincter segment. Posterior sphincter opening, on the other hand, did not show a consistent relationship with intrabolus pressure and peaked consider-

1475

ably later in the opening period compared with intrabolus pressure. Maximal sphincter distention occurred within 0.13 s of opening, followed by a period of lesser sphincter diameter (collapse) and finally contraction. Sphincter contraction was evident radiographically as the tail of the barium bolus traversed the sphincter segment and manometrically as the arrival of the pharyngeal peristaltic contraction. The velocity of pharyngeal peristalsis did not vary with the 2 0.7,12.7 bolusvolume(12.7 + 0.1,13.3 + 0.6,12.2 ‘-+ 0.9 cm/s for l-, 5-, lo-, and 20-ml swallows, respectively). However, the velocity of the head of the bolus did vary significantly with increasing bolus volume [28 2 6,45 -C 7, 66 + 9, 80 + 12 cm/s for l-, 5-, lo-, and 20-ml swallows, respectively (p < 0.01, analysis of variance)]. Analysis of the videotaped swallow sequences suggested that the propulsive forces on the bolus (associated with increased intrabolus pressure as shown in Figure 7) seen early in the period of UES opening were derived from the piston action of the tongue in the pharynx. The pattern of tongue movement as viewed in the lateral images was first to assume a ramp configuration early in the swallow (5). This occurred during the period of laryngeal and hyoid elevation and was associated with the head of the bolus falling into the valleculae. Almost coincident with the bolus arriving at the valleculae, the tongue base moved posteriorly like a piston, eventually contacting the pharyngeal wall, or in some cases the inferior end of the soft palate. Bolus acceleration and anterior UES distention closely correlated with the posterior movement of the tongue base (Table 3). The onset of pharyngeal peristalsis, as recorded in the proximal sensor, coincided with the contact of tongue base with the pharyngeal wall and was sometimes seen radiographically as a bulging of the pharyngeal constrictor against the base of the tongue. In other instances, there was no radiographic correlate of pharyngeal peristalsis in this region because the tail of the bolus was distal to the area and the contour of the posterior pharyngeal wall did not reveal any anterior bulging to meet the tongue base.

Table 3. Timing of Posterior Tongue Base Movement and the Initiation of Pharyngeal Peristalsis Relative to Maximal Sphincter Distention During Swallowing Bolus volume

Onset of posterior movement Pharyngeal wall contact (s) Onset of peristalsis (s)~

(s)’

1 ml

5 ml

10 ml

20 ml

-0.19 Lk 0.03 -0.06 k 0.03 -0.04 + 0.04

-0.17 c 0.02 fO.07 2 0.03 +0.08 f 0.03

-0.09 * 0.02 +0.12 f 0.04 +0.14 + 0.02

-0.05 + 0.02 +0.15 + 0.04 +0.19 k 0.02

Values are mean f SEM. a Interval relative to maximal anterior sphincter opening: negative values indicate occurrence distention and positive values indicate occurrence after maximal distention.

before maximal

1476

GASTROENTEROLOGY Vol. 97, No. 6

JACOB ET AL.

?

r

$4

$ 3

$2 f 2

0

20

25 Luyn@d

Figure

30

35

40

Elevation (mm)

9. Correlation between superior sphincter excursion during swallowing and the magnitude of the posterior component of UES opening (8 subjects, all bolus volumes). The magnitude of sphincter elevation was determined by the vertical excursion of the subglottic air column. Larger volume swallows were characterized by a greater extent and duration of laryngeal elevation, thereby allowing for the persistence of the posterior component of opening observed in Figure 1. The strong correlation observed (r’ = 0.76, p < 0.001, polynomial regression) is consistent with the hypothesis that greater potential space for opening existed during sphincter elevation.

Laryngeal

Elevation

Superior laryngeal movement during swallowing paralleled the superior hyoid movement and showed a similar relationship to bolus volume. Sphincter opening and closure occurred during nearly maximal laryngeal elevation. Furthermore, as shown in Figure 9, the magnitude of the posterior dimension of sphincter opening correlated well with the extent of laryngeal elevation (polynomial regression r2 = 0.76,p < O.OOl), suggesting that the more superior position of the larynx provided a greater potential posterior space to accommodate the posterior opening. The corresponding relationship between laryngeal elevation and anterior sphincter opening was much weaker (r2 = 0.38,polynomial regression). The magnitude of the superior laryngeal excursion for all bolus volumes was greater than that of the hyoid excursion, resulting from a significant shortening of the infrahyoid compartment. Figure 10 shows the vertical dimension (expressed as the distance between the hyoid and the trachea) of the infrahyoid compartment during swallows of different bolus volumes. As shown in the figure, the shortening of the infrahyoid compartment started 0.2 s before UES opening and persisted throughout the period of sphincter opening.

Discussion The aim of this study was to investigate the mechanisms of UES opening and volume-dependent modulation that occur during swallowing. This was

accomplished by applying a biomechanical analysis to data obtained during synchronized concurrent manometric and fluoroscopic swallowing studies in young normal volunteer subjects. Our analysis suggested that UES function during swallowing could be subdivided into five phases, each influenced by separate factors. The five sequential phases of swallow-related UES activity (relaxation, opening, distention, collapse, and closure] are discussed in detail below. Sphincter. relaxation, defined manometrically as the fall of UES pressure to 0 mmHg, was achieved 0.1 ? 0.03 s before fluoroscopically demonstrated sphincter opening, irrespective of the bolus volume. Relaxation occurred during laryngeal elevation and previous studies have shown that this is associated with a cessation of the tonic activity of the striated muscle structures of the sphincter (1,8,9). However, UES pressure continued to decrease during the ensuing 0.1 s, reaching a nadir of - 11 to - 17 mmHg immediately before sphincter opening (Figure 2). Previous investigators have acknowledged this negative intraluminal pressure (sometimes referred to as the “Schluckatmung,” or swallow inspiration), but it has not been adequately explained (8,lO). Recordings from single “inspiratory” neuron studies in the medulla that showed short bursts of activity during this period were proposed as the central correlate of

mm

-0.4

-0.2

0.0

0.2

0.4

0.6

Time (seconds)

Figure

10. Shortening of the hyoid-sphincter compartment during swallowing (mean values 8 subjects, two swallows of each volume). The vertical position of the sphincter was derived from the vertical position of the subglottic air column. Active shortening of the infrahyoid compartment began 0.2 s before sphincter opening and persisted until sphincter closure. The magnitude of the shortening was similar for all swallow volumes but the persistence of the shortening varied with bolus volume much the same as did the period of sphincter opening. Thus, almost half of sphincter elevation occurring during swallowing occurred as a result of shortening of the infrahyoid compartment, whereas the remainder was attributable to hyoid elevation.

December1989

the Schluckatmung (11,12), however, without a demonstrable correlate of the neuronal burst in the inspiratory musculature, the significance of the observation is open to question. Our findings were that the magnitude and timing of this briefly negative preopening UES pressure corresponded to the distance separating the hyoid and the anterior wall of the LIES, a measurement we propose reflects the anterior traction being applied to the sphincter via the hyoid (Figure 6). As the UES snapped open, the hyoid-to-sphincter distance abruptly diminished and the intraluminal UES pressure increased to a value of -0 (range -1 to +5 mmHg). Description of the biomechanics of sphincter opening and the relevance of swallow-related hyoid and laryngeal movements requires some anatomic considerations (13). The cricopharyngeus muscle is attached to the lateral aspects of the cricoid arch and forms a sling around the hypopharyngeal-esophageal junction posteriorly. Human cadaveric studies by Nilsson et al. (14) have shown that the prevertebral space at the level of the larynx and hypopharynx is composed of a thin layer of fatty tissue. Furthermore, a distinct slit is present through this tissue, extending along the prevertebral fascia, that allows for the larynx, trachea, and attached UES to move up and down in unison during swallowing without oblique forces acting on the sphincter. The hyoid is suspended above the larynx by the suprahyoid and infrahyoid musculature. The main attachment of the hyoid to the larynx is the thyrohyoid muscle, a component of the infrahyoid group of cervical strap muscles. Superiorly, the hyoid attaches to the mandible by the anterior group of suprahyoid muscles (geniohyoid, anterior belly of digastric, and mylohyoid), and to the basicranium and pterygoid by the posterior group of suprahyoid muscles (posterior belly of digastric and stylohyoid). This unique arrangement allows the hyoid and larynx along with the attached UES to function as a mobile unit during swallowing. Contraction of the anterior and posterior groups of suprahyoid muscles allows for anterior and superior excursion of the hyoid. The muscular connection of the larynx to the hyoid results in the transmission of forces acting on the hyoid to the larynx via the hyoid fulcrum. Findings from this study suggest that UES opening results from anterior traction corresponding to anterior movement of the hyoid being transmitted to the anterior wall of the UES via the infrahyoid muscles. Evidence for this mechanism is twofold. First, the anterior excursion of the hyoid bone showed close temporal and spatial association with UES opening; rapid anterior hyoid movement invariably began 0.1-0.16 s before sphincter opening (Figure 4) and the x-y coordinates of the hyoid at the time of

UPPERESOPHAGEALSPHINCTEROPENING

1477

opening and closing are remarkably similar regardless of bolus volume (Figures 3-5). Second, the magnitude of the hyoid tug on the UES correlated inversely with the negative intraluminal UES pressure during the period of UES relaxation, as discussed earlier (Figure 6). This anterior tug culminated in sphincter opening, evident manometrically as an abrupt increase in intraluminal pressure to near atmospheric pressure, and fluoroscopically as the anterior wall of the sphincter appearing to jump forward. Previous studies on anesthetized opossum demonstrated electromyographic activity in the geniohyoid muscle during cricopharyngeal and inferior constrictor relaxation, an observation that supports the hypothesis that the relaxed sphincter is elevation of actively pulled open (1).Furthermore, the UES as a result of hyoid elevation and shortening of the infrahyoid compartment (Figure 10) results in a more horizontal alignment of mandible, hyoid, and UES, thereby improving the mechanical advantage of the anterior tug in facilitating sphincter opening. That bolus volume did not substantially influence the extent of the anterior or superior hyoid movement suggests that the extent of hyoid excursion is not a major mechanism for the modulation of the UES opening diameter during swallowing. Maximal UES distention during swallows of all volumes was achieved by movements of both the anterior and posterior walls of the sphincter away from the preopening coordinate (Figure 1) and occurred early in the course of the opening interval. Manometrically, the distensive phase of sphincter opening was associated with a positive intrabolus pressure early in the relaxation interval (Figure 2). Instantaneous intrabolus pressure correlated closely with anterior UES opening (Figure 8),suggesting that modulation of anterior sphincter wall excursion resulted from the increased intrabolus pressure associated with larger volume swallows. In contrast to the anterior component of sphincter opening, the extent of the posterior component of UES opening correlated best with the extent of sphincter elevation, suggesting that, perhaps because of the anatomy of the cervical vertebrae, more potential space existed posteriorly as the sphincter elevated with the larynx (Figure 9). The timing of the intrabolus pressure leading to sphincter distention corresponded to the posterior thrust of the tongue that acted as a piston to force the barium bolus contained in the oropharynx down the hypopharynx and into the esophagus with a bolus head velocity ranging from 20-80 cm/s depending on bolus volume (Table 3), values similar to those previously reported (15). The onset of pharyngeal peristalsis coincided with contact between the base of the tongue and the posterior pharyngeal wall at

1478

JACOB ET AL.

the termination of the tongue thrust. In contrast to the velocity of bolus head, the velocity of pharyngeal peristalsis, and hence the tail of the barium bolus did not vary with bolus volume. These observations lead us to concur with the conclusion of Cerenko et al. (16) that the initial pharyngeal bolus driving force for liquid swallows is generated by the tongue base acting like a plunger in the oropharynx, rather than pharyngeal peristalsis per se. After the period of sphincter distention, the diameter of the sphincter diminished and intrabolus pressure returned to a value near zero, a period that we refer to as sphincter collapse. This condition persisted until sphincter closure occurred, an event that coincided with the arrival of the pharyngeal peristaltic contraction. Thus, the overall interval of UES opening was determined by the timing of the initiation of pharyngeal peristalsis. The initiation of pharyngeal peristalsis, in turn, was related to the pattern of hyoid movement such that prolonged anterior and superior hyoid excursion was associated with a later initiation of peristalsis. This suggests that the hyoid coordinates indirectly represent the tongue contour and that pharyngeal peristalsis occurs in a continuum beginning with posterior tongue movement and tongue contact with the pharyngeal wall. We conclude from these studies that the UES is subject to complex regulation during swallowing that allows for the accommodation of a wide range of bolus volumes. Transport of increased bolus volumes across the sphincter is accomplished by increasing the diameter, and hence the cross-sectional area of the sphincter during swallowing, increasing the duration of sphincter opening, and increasing the velocity of the head of the bolus. Manipulating these three variables should result in an increased flow rate as well as period of flow across the sphincter. Furthermore, UES opening seems to be an active mechanical event dependent on muscular traction to the anterior sphincter wall rather than simply a consequence of cricopharyngeal relaxation.

References 1.

2.

Asoh R, Goyal RK. Manometry and electromyography of the upper esophageal sphincter in the opossum. Gastroenterology 1978;74:514-20. Kahrilas PJ, Dodds WJ, Dent J, Logemann JA, Shaker R. Upper

GASTROENTEROLOGY Vol. 97, No. 6

esophageal sphincter function during deglutition. Gastroenterology 1988;95:52-62. 3. Dodds WJ, Man KM, Cook IJ, Kahrilas PJ, Stewart ET, Kern MK. Quantification of swallow-induced hyoid movement. AJR 1988;150:1307-9. 4. Logemann JA. Manual for the videofluorographic study of swallowing. San Diego: College-Hill Press, 1986. 5. Shedd DP, Scatliff JH, Kirschner JA. The buccopharyngeal propulsive mechanism in human deglutition. Surgery 1960; 48:846-53. 6. Logemann JA, Kahrilas

PJ, Begelman J, Dodds WJ, Pauloski BR. Interactive computer program for biomechanical analysis of videoradiographic studies of swallowing. AJR 1989;153:

277-80. 7. Slinker BK, Glantz SA. Multiple

linear regression is a useful alternative to traditional analysis of variance. Am J Physiol 1988;255:R353-67. 8. Doty RW, Bosma JF. An electromyographic analysis of reflex deglutition. J Neurophysiol 1956;19:44-60. 9. Van Overbeek JJM, Wit HP, Paping RI-IL, Segenhout HM. Simultaneous manometry and electromyography in the pharyngoesophageal segment. Laryngoscope 1985;95:582-4. 10. Fyke FE, Code CF. Resting and deglutition pressures in the pharyngo-esophageal region. Gastroenterology 1955;29:2434. 11. Hukuhara

12.

13. 14.

15.

T, Okada H. Effects of deglutition upon the spike discharges of neurons in the respiratory center. Jpn J Physiol 1956;6:162-6. Sumi T. The activity of brain-stem respiratory neurons and spinal respiratory motoneurons during swallowing. J Neurophysiol 1963;26:466-77. Warwick R, Williams PL, eds. Grays anatomy. 35th ed. Philadelphia: WB Saunders, 1980. Nilsson ME, Isacsson G, Isberg A, Schiratatzki H. The mobility of the upper esophageal sphincter in relation to the cervical spine-a morphologic study. Doctoral thesis. Karolinska Institute, Huddinge, Sweden, 1988. Fisher MA, Hendrix TR, Hunt JN, Murrills AJ. Relation between volume swallowed and velocity of the bolus ejected from the pharynx into the esophagus. Gastroenterology 1978;

74:1238+0. 16. Cerenko D, McConnel

FMS, Jackson RT. Quantitative assessment of pharyngeal bolus driving forces. Otolaryngol Head Neck Surg 1989;100:57-63.

Received March 7, 1989. Accepted June 7, 1989. Address requests for reprints to: P.J. Kahrilas, M.D., Northwestern University Medical School, GI Section, Department of Medicine, 1526 Wesley Towers, 250 East Superior Street, Chicago, Illinois 60611. This project was supported by National Institutes of Health National Research Service Award fellowship #0821201 (P.J.), a fellowship from the Schweppe Foundation (P.J.K.), a grant from the Veterans Administration Medical Research Service (P.J.K.), and United States Public Health Service grant #POl-CA-40007 (J.A.L.).