| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Mechanical Engineering, Pennsylvania State University, University Park 16802; 2 Department of Radiology, Thomas Jefferson University Hospital, and 3 Division of Gastroenterology, Temple University Hospital, Philadelphia, Pennsylvania 19140
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
We analyzed local longitudinal shortening by combining concurrent ultrasonography and manometry with basic principles of mechanics. We applied the law of mass conservation to quantify local axial shortening of the esophageal wall from ultrasonically measured cross-sectional area concurrently with measured intraluminal pressure, from which correlations between local contraction of longitudinal and circular muscle are inferred. Two clear phases of local longitudinal shortening were observed during bolus transport. During luminal filling by bolus fluid, the muscle layer distends and the muscle thickness decreases in the absence of circular or longitudinal muscle contraction. This is followed by local contraction, first in longitudinal muscle, then in circular muscle. Maximal longitudinal shortening occurs nearly coincidently with peak intraluminal pressure. Longitudinal muscle contraction begins before and ends after circular muscle contraction. Larger longitudinal shortening is correlated with higher pressure amplitude, suggesting that circumferential contractile forces are enhanced by longitudinal muscle shortening. We conclude that a peristaltic wave of longitudinal muscle contraction envelops the wave of circular muscle contraction as it passes through the middle esophagus, with peak longitudinal contraction aligned with peak circular muscular contraction. Our results suggest that the coordination of the two waves may be a physiological response to the mechanical influence of longitudinal shortening, which increases contractile force while reducing average muscle fiber tension by increasing circular muscle fiber density locally near the bolus tail.
muscle contraction; ultrasound; mechanics
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
WHEREAS CONTRACTION OF ESOPHAGEAL circular muscle during a swallow can be quantified with manometric measurement of intraluminal pressure, contraction of longitudinal muscle is difficult to measure precisely in humans. Consequently, the role of longitudinal muscle in bolus transport remains poorly understood. In a classic study by Dodds et al. (5), four metal markers were inserted into the muscle wall of a feline esophagus to measure longitudinal motions during normal bolus transport. Their data suggested the existence of a wave of localized longitudinal shortening that appeared to traverse the distal esophagus in concert with the bolus tail. Qualitatively similar characteristics have been observed in the human esophagus with widely spaced metal clips endoscopically attached to the mucosa (6, 16). In this study, we analyze the relationship between circular and longitudinal muscle contraction near the bolus tail by directly measuring local shortening of the esophagus concurrently with intraluminal pressure during normal bolus transport. This is done by coupling concurrent endoscopic ultrasonography and intraluminal manometry with a basic conservation law from mechanics.
A consequence of contraction of longitudinal muscle is shortening in
the axial direction. For this reason, longitudinal muscle contraction
of the human esophagus has been inferred from longitudinal shortening
measured by using widely spaced metal clips attached to the esophageal
mucosa (6, 9, 13, 16). The change in spacing between
adjacent clips gives a relatively crude measure of longitudinal
shortening over esophageal segments 3-10 cm in length. A
characteristic of these studies, however, is that the motion of widely
spaced mucosal clips measures global rather than local shortening. To
illustrate this point, imagine a segment of the esophagus of initial
length L* = 1 cm shortening to length L = 0.5 cm
in the absence of any other shortening of the esophagus. Local
longitudinal shortening is given by L/L* = 0.5. However, if
the same isolated local contraction were measured by using clips spaced
3 cm apart, for example, global longitudinal shortening would be
calculated as L/L* = (3 
0.5)/3 = 0.83. Similarly, 5-cm spaced clips measure L/L* = 0.9. Thus
mucosal clips spaced outside a localized contracting segment
underestimate longitudinal shortening (larger L/L*
implies less shortening), and the estimate varies with clip spacing.
Furthermore, the relative motion between a clip on the mucosal surface
and the underlying muscle layer (from shear distortion of the mucosa)
introduces additional unknown error in the estimate of local
longitudinal shortening.
To avoid the uncertainty associated with mucosal clips, we developed a new procedure for accurately measuring local longitudinal shortening of the esophageal wall using a high-frequency ultrasound transducer at fixed locations within the lumen. As discussed in METHODS, application of the law of mass conservation (a fundamental law of physics) to the incompressible matter within esophageal muscle implies that a local increase in cross-sectional area of the muscle wall is inversely proportional to longitudinal shortening of the muscle layer locally at the same cross-section. Thus measurement of cross-sectional area of the muscle layer(s) can be related quantitatively to local shortening (or lengthening) of the same muscle layer(s). Furthermore, we shall show that measuring only changes in esophageal muscle wall thickness (1, 14, 24), as during luminal filling, for example, is generally insufficient to deduce local longitudinal shortening.
In this study, cross-sectional area was measured during bolus transport at a point in the middle esophagus from ultrasound images generated from a high-frequency transducer. Placement of the ultrasound probe within an endoscope adjacent to a single manometric sidehole allowed concurrent measurement of intraluminal pressure with local longitudinal shortening at the same luminal location. From these data we infer a correlation between longitudinal and circular muscle contraction from concurrent measurement of muscle cross-sectional area and intraluminal pressure.
| |
METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
The principle of mass conservation applied to muscle. Mass conservation is a basic law of physics that underlies the deformation of all matter. In context with muscle, the law can be understood by imagining a volume within the muscle where each material particle along the surface of the volume is tagged by a marker (a material particle may be thought of as extremely large numbers of molecules within a submicroscopic volume much smaller than any volume of interest). As the muscle deforms, so does the surface defined by these same mass particles. Mass conservation states that at all times, before, during, and after its change in shape, the mass within the volume defined by the same tagged material particles cannot change (23).
Whereas the mass within the material volume cannot change, the density of the material (mass divided by volume) can change if the size of the material volume changes during deformation. This is the case, for example, when air within a closed cylinder is compressed by a piston. Indeed, a fundamental characteristic of a gas is that when pressure is applied to the enclosing surface of a material gas volume, the relative change in volume is as large as the relative change in pressure. By contrast, when pressure is applied to liquids or solids, the change in volume is negligible. Liquids and solids are described as "incompressible," meaning that the density of the substance and material volumes do not change with changes in pressure (to a high degree of accuracy). At the microscopic level, muscle is a matrix of liquid (e.g., intracellular and interstitial water) and deformable solid (e.g., cellular material). Therefore, muscle is incompressible, and a predefined macroscopic segment of the esophageal wall can only change its volume during deformation if either vapor pockets of significant size exist within the muscle or significant liquid content were to leave the muscle volume during contraction and be taken up again after contraction. There is no evidence to suggest that either of these scenarios is relevant during esophageal muscle contraction, and the incompressibility of muscle (including esophageal muscle) has been verified experimentally by several investigators (7, 8, 22). Furthermore, any fluid that might be speculated to leave the muscle layer during contraction will lead to an underestimate of muscle cross-section and, as we show next, local longitudinal shortening. Thus any observations made from measurements of local longitudinal shortening can be enhanced only if fluid exchange existed.Measuring local longitudinal shortening. Incompressibility is tacitly assumed in muscle clip studies when global longitudinal muscle contraction is inferred from relative displacement of two clips. Here we explicitly apply the principle to quantify local shortening from measurement of change in cross-sectional area from ultrasound images.
Consider the schematic in Fig. 1, a short material segment of esophageal wall muscle at two instants in time, in the resting state and after longitudinal contraction. Because this axial muscle segment contains the same incompressible material components within and between cells before and after contraction (i.e., the surfaces of the segment are always attached to the same material particles), the volume of this muscle segment must also be the same before and after contraction. It follows that a contraction of the muscle segment in the longitudinal direction must be accompanied by an increase in cross-sectional area according to the equation LA=L*A*, where L*A* and LA are the volumes of the muscle segments before and after contraction. Thus
|
(1) |
|
Concurrent ultrasonography with manometry. Concurrent ultrasound and water-perfused manometry data from Miller et al. (14) were analyzed from four healthy subjects with no swallowing disorders (ages 23-25 years), lying supine (at 30° inclination) and for 24 swallows. The subjects were asked to swallow 10-ml boluses of water or gelatin mix. The gelatin bolus was designed to reduce the level of air bubbles within the bolus, which disrupt the ultrasound waves and render many images unsuitable for analysis. The resulting liquid was roughly the thickness of honey at room temperature.
Images were digitized from VHS videotape at 5 frames/s, corresponding to 50-60 images/swallow. Because of the time-consuming nature of the image analysis (see Image processing) and the inability to fully analyze some swallows due to significant image loss from air swallowed with the bolus (which is quite common and interferes with ultrasound), time-resolved image analysis and curve fitting for complete swallows were feasible in eight swallows from two subjects from which the changes in muscle cross-sectional area and local longitudinal shortening were deduced during entire bolus transport sequences. The time-resolved data analyzed in detail were chosen by the quality of the ultrasound images and the extent of air-induced image degradation. However, 16 additional swallows were analyzed over fewer images to determine the resting state geometry, maximum local longitudinal shortening (i.e., maximum cross-sectional area ratio Amus/A*mus), and maximum intraluminal pressure (Pamp). The details of swallows from each subject are shown in Table 1. Further data collection details are given in Ref. 14.
|
Image processing.
Figure 2, a, b, and c, shows
typical ultrasound images of the esophageal cross-section obtained at
three different times during a swallow: the relaxed state
(a; i.e., before a swallow), the esophagus distended by the
bolus (b), and the maximally contracted state
(c). The inner wall of the circular muscle and the outer wall of the longitudinal muscle are shown by dashed lines on the same
images in Fig. 2, a', b', and c',
respectively. The white annulus in the middle of each image is the
catheter, inside of which the ultrasound transducer is housed. Because
the transducer spins to obtain the image, the concentric circular bands
that appear in the image are artifacts and do not correspond to any anatomic structure.
|
|
Analysis.
Effective radii to the inner circular muscle
(Rin), to the intermuscular
connective tissue (Rmid), and to the outer
longitudinal muscle (Rout) were defined as shown
in Fig. 3, where the measured areas Alumen,
Acirc, and Along were
placed in concentric circles. Thus the effective radii were defined by
|
|
(2) |
|
) by
|
|
(3) |
|
|
(4) |
| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
The wall muscle layer.
The resting state was measured before the initiation of a swallow.
Averaged over eight swallows, the resting state in the middle esophagus
was characterized by wall muscle layer area and thickness
A*mus = 0.40 ± 0.03 cm2 and *mus = 0.138 ± 0.007 cm; longitudinal muscle area and thickness A*long = 0.218 ± 0.013 cm2 and *long =
0.069 ± 0.005 cm; radius to circular muscle
R*in = 0.40 ± 0.05 cm; and
intraluminal pressure P* = 0.12 ± 0.80 mmHg
(means ± SD).
mus and
long) and intraluminal pressure and inner
muscle radius (Rin) at a fixed point in the middle esophagus averaged over eight swallows and centered on the time
of maximum total muscle cross-sectional area
(Amus/A*mus). These figures
contain a great deal of information that shall be interpreted in terms
of muscle contractile physiology in the DISCUSSION. We
focus first on the characteristics of the total muscle layer (Fig. 4).
|
|
2 s. During most of this period, until about 3 s, the thickness of
muscle layer decreased in approximate inverse proportion to
Rin, but with
Amus/A*mus hovering close to
1 (the slight increase in
Amus/A*mus during this
period is only marginally significant). Intraluminal pressure rose to a
plateau of 4.4 mmHg during this period, which reflects intrabolus
pressure preceding the arrival of the circular muscle contraction wave
(3). During this distension phase (roughly 7 to 3 s),
the bolus fills the lumen and the muscle layer thickness decreases in
the absence of significant longitudinal shortening or circular muscle
contractile activity.
A local longitudinal shortening phase followed as both the thickness
and cross-sectional area of the muscle increased above their resting
state values, beginning their rise at about the same time (3 s) and
reaching coincident maxima at t = 0. The increase in
Amus/A*mus > 1 necessarily
implies increasing local longitudinal shortening, with maximum relative
reduction in longitudinal length of L/L* = 0.36 having
occurred ~0.25 s before maximum luminal pressure. The maximum muscle
area was 1.11 ± 0.10 cm2, significantly greater than
the resting area 0.40 ± 0.03 cm2 (P < 0.001).
Whereas average Amus/A*mus > 1 implies local longitudinal shortening, the rapid increase in
intraluminal pressure above the slightly elevated intrabolus pressure
implies circular muscle contraction, luminal narrowing, and subsequent
luminal closure (3). We estimate from Fig. 4A
that this rise in average intrabolus pressure began at about 2.0 s.
The upstroke in local longitudinal shortening, by contrast, was
estimated subjectively from Fig. 4A to begin at 3 s,
roughly 1 s before the upstroke in pressure. Indeed, Fig.
4A shows that a wave of local longitudinal shortening enveloped the wave of circular muscle contraction:
Amus/A*mus began a rapid
increase before the upstroke in intraluminal pressure and returned to
its resting state level well after intraluminal pressure had returned
approximately to its resting state (t > 4.6 s).
In Fig. 4A, the difference in duration between the waves of
circular muscle contraction and local longitudinal shortening waves is
indicated qualitatively by defining subjectively the downstrokes in
pressure and longitudinal shortening as the times when pressure and
Amus/A*mus reached plateaus
after an initially rapid drop. Defined in this way, the duration of the
longitudinal shortening wave was 1.45 times longer than the duration of
the circular muscle contraction wave. A more precise ratio of
durations, defined by the periods at which pressure and
Amus/A*mus are above their
half-maxima, gives the same ratio of durations to within the accuracy
of the data.
The distension phase is heralded by an increase in radius
Rin concurrent with a decrease in muscle
thickness mus in the absence of significant
change in Amus/A*mus.
Circular muscle contraction is heralded by a sudden rise in
intraluminal pressure coincident with a decrease in
Rin at about 2 s. In contrast, longitudinal
shortening begins with a rapid increase in
Amus/A*mus, which, in Fig.
4, is coincident with an increase in mus at
t 
3 s, reaching its peak at the time of peak
Amus/A*mus (i.e., peak local
longitudinal shortening). Maximum local longitudinal shortening
preceded the peak in intraluminal pressure by only 0.25 s; they
were very nearly coincident. At the times of maximal longitudinal
shortening and intraluminal pressure, the average radius
Rin had reached a plateau of ~0.46 cm, above the resting state value R*in = 0.40 cm.
The plateau in average Rin, which began at peak
longitudinal shortening and circular muscle contraction, remained in
the range 0.44-0.46 cm until longitudinal shortening and
intraluminal pressure had reduced to low levels and reached plateaus at
~3.7 s, after which Rin decreased to its
resting state value together with intraluminal pressure. During this
reduction, however,
Amus/A*mus and
mus remained above their resting state values
(Amus/A*mus 1.1-1.2), suggesting a persistence of lower-level longitudinal shortening beyond the period of circular muscle contraction.
The longitudinal muscle layer.
Figure 5 shows the variation in longitudinal shortening of the isolated
longitudinal muscle layer
(Along/A*mus) together with
intraluminal pressure, inner muscle radius Rin,
and longitudinal muscle thickness long, with
t = 0 defined at peak total muscle cross-sectional area
(Amus/A*mus). Comparison of
Fig. 5 with Fig. 4 shows that the temporal variations in longitudinal
shortening of the longitudinal muscle layer were the same as
longitudinal shortening of the entire muscle layer, to within the
precision of the data. In particular, a wave of local longitudinal
shortening enveloped the pressure wave, with upstroke, downstroke, and
duration essentially the same as longitudinal shortening of the entire
muscle layer. Note that it is more difficult to identify and measure
accurately the cross-sectional area and thickness of the longitudinal
muscle layer from ultrasound images than to measure the total muscle
layer, so the precisions of Along and
long are less than that of
Amus/A*mus and
mus. The difference between peak
Along/A*mus (Fig.
5A) and peak
Amus/A*mus (Fig.
4A), for example, is not statistically significant
(P = 0.25).
Cross-sectional shape of the lumen.
To illustrate the change in cross-sectional shape of the lumen during
the passage of the bolus and contraction wave, Fig. 6 shows three-dimensional reconstructions
(two space and one time coordinates) of the cross-sectional shape of
the inner wall of the esophagus over time during a representative
swallow. To evaluate the correlation between the change in shape of the
luminal cross-section and the waves of local longitudinal shortening
and circular muscle contraction, we have shaded the surfaces of the
figures so that the gray level is proportional to the magnitude of
Amus (i.e., local longitudinal shortening) in
Fig. 6A and to the magnitude of intraluminal pressure in
Fig. 6B.
|
Correlations.
As discussed in METHODS, the process of carefully done
image analysis required for Figs. 4 and 5 was highly time intensive, limiting the full-wave analysis to eight swallows. However, to establish statistically the correlation between levels of circular muscle contraction and local longitudinal shortening, peak
cross-sectional areas were quantified for 24 swallows. Figure
7 shows that larger peak longitudinal
shortening
(Amus/A*mus)max
was correlated with larger maximum intraluminal pressure (i.e.,
pressure amplitude Pamp), on average (r
= 0.61, P < 0.001, N = 24). That is, stronger closure force near the bolus tail tended
to be accompanied by greater local longitudinal shortening in the same
local area. Furthermore, from Fig. 8, a
shorter delay between the onset of circular muscle contraction and
onset of longitudinal shortening (
t) was negatively
correlated with larger Pamp (r =
0.85, P < 0.01, N = 8). Thus the
strength of the luminal closure force was altered both by the magnitude
of longitudinal shortening and by the closeness of temporal
coordination between longitudinal and circular muscle contraction.
|
|
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Measurement of longitudinal shortening. In this study, we have quantified shortening of longitudinal muscle locally in the middle human esophagus during bolus transport. The method combines explicit measurement of circular, longitudinal, and total muscle layer cross-sectional geometry with the incompressibility of the muscle layer to unambiguously deduce shortening (or lengthening) in the longitudinal direction within an axially thin segment of the esophageal wall segment at the location of the transducer. Whereas a single-point measurement using the method introduced here does not provide the motion of material points on the esophageal wall, the approach yields a purely local measurement of longitudinal shortening without inaccuracies inherent in the use of widely spaced clips placed on the mucosa.
Incompressibility of muscle layers has been confirmed experimentally (7, 8, 22), and the accuracy of the method is limited primarily by the accuracy with which the cross-sectional area of the muscle layer under consideration can be measured from endoscopic ultrasound images (or other imaging techniques). The method is applicable throughout the gastrointestinal tract wherever it is possible to image the cross-section with sufficient resolution to quantify muscle cross-sectional area. However, practical application of the method is limited by the time-intensive nature of interactive edge detection and analysis of the hundreds of images which underlie the data, for example, in Figs. 4-8.The two phases of muscle activity in the middle esophagus.
Whereas one might anticipate that shortening of the esophageal wall
muscle implies a change in muscle layer thickness, Figs. 4 and 5 make
it clear that muscle thickness cannot be used alone as an indicator of
longitudinal shortening. This is because, in the absence of
longitudinal shortening, the muscle layer thickness mus varies approximately in inverse
proportion to the luminal radius Rin. Thus
mus is sensitive to change in luminal radius, particularly to distension of the esophageal lumen by filling, as is
apparent in Figs. 4 and 5. Relative change in muscle layer cross-sectional area, on the other hand, unambiguously reflects local
relative change in the longitudinal dimension of the muscle layer,
through Eq. 1. We conclude from Figs. 4 and 5, therefore, that the changes in geometry at the middle esophagus in response to
bolus transport occur in two broad parts: 1) a "distension phase," which occurs in the absence of longitudinal shortening or
circular muscle contraction, but during which muscle wall thickness decreases while luminal radius increases as the esophagus is filled, followed by 2) a "contraction phase," whereby
longitudinal shortening and circular muscle contraction act in concert;
local longitudinal shortening increases cross-sectional area and muscle
layer thickness while circular muscle contraction decreases luminal
radius and raises intraluminal pressure.
/*,
reach plateaus 10-15% different from 1, suggesting that the
esophageal wall remains in a state of mild longitudinal shortening for
some period after the passage of waves of circular muscle contraction
and local longitudinal shortening. Whereas the lumen is noncircular in
general, the circular geometry of the lumen during contraction (Fig. 6) suggests relatively uniform circumferential active fiber tension in the
circular muscle layer during peristalsis.
Comparisons with mucosal clip studies. Dodds et al. (5) measured the axial motion of four tantalum wires inserted into the in vivo longitudinal muscle over the distal half of the feline esophagus. Although coarser in resolution than the current study, their results suggested the progression of a wave of longitudinal shortening into the lower esophagus roughly coincident with the bolus tail. Pouderoux et al. (16) placed three clips at ~4-cm intervals in the distal 8-10 cm of the human esophagus and concluded, consistent with the current study, that the "contracting longitudinal segment was advancing ahead of the contracting circular muscle segment." An earlier study with mucosal clips by Edmundowicz and Clouse (6) measured change in length of upper and lower halves of the human esophagus and observed that early in the swallow the lower half lengthened while the upper half shortened, consistent with the clip motions in the cat esophagus measured by Dodds et al. (5). Their coarse measurements are also consistent with the later clip data of Pouderoux et al. (16), who observed that two adjacent 4-cm segments of the distal esophagus initially lengthen and later shorten in peristalsis-like fashion. In the current study, the wave of longitudinal shortening is preceded by negligible change in esophageal length locally, indicating that the middle-esophageal ultrasound transducer was located above the more distal esophageal segments that initially lengthen and later shorten.
In the introduction, we pointed out that the quantification of longitudinal shortening using mucosal clips will change with the distance between the clips; the more widely spaced the clips, the more underpredicted will be the measurement of longitudinal shortening. Thus it is not surprising that we measure significantly greater local shortening than previous clip studies. Figure 4 gives maximal local longitudinal shortening of the muscle wall layers in the middle esophagus to be L/L* 0.34 ± 0.02. Kahrilas et al.
(9) and Pouderoux et al. (16) give rough
estimates of minimum L/L* between 0.6 and 0.8 over
esophageal segments 3-5 cm in length, whereas Edmundowicz and
Clouse (6) measure much lower levels of shortening,
maximum L/L* 0.92 and 0.98 over the distal and proximal
halves of the esophagus roughly 10-14 cm in length (remember that
lower values of L/L* imply greater longitudinal shortening).
The lower levels of longitudinal shortening measured in the clip
studies no doubt reflect the global nature of the measurement.
In fact, whereas the cross-sectional area-based method introduced here
arguably yields the most precise measurement of local longitudinal
shortening, clip-based measurements are complementary in that the
relative change in length of the segment between mucosal clips
(ignoring potential error in measured clip position from mucosal
deformation) may be shown to give the average of the local longitudinal
shortening over that segment. The results in Figs. 4 and 5 show that
local shortening varies strongly along the esophagus, with a peak
nearly coincident with the circular muscle contraction wave. Thus
average longitudinal shortening measured with clips will always
underestimate maximum local shortening in the segment between the
clips, and the underestimate will depend on the distance over which the
average is taken (i.e., the distance between the clips).
Coordinated waves of local longitudinal and circular muscle contraction. The data in Figs. 4 and 5 are presented as time changes in a variable at a fixed location in the middle esophagus. However, Figs. 4 and 5 may also be interpreted qualitatively as variations along the esophagus at a fixed time by imagining the peaks in intraluminal pressure and local longitudinal shortening in these figures to be at the transducer location in the middle esophagus and assuming that the spatial wave forms change only slowly as they propagate distally. The data to the left of these peaks in the figures, then, represents the esophagus distal to the transducer (note that the bolus arrives in advance of the pressure peak), whereas the data to the right of the peaks represents the esophagus proximal (3). Applied to Fig. 6, where the time axis is reversed, this interpretation places the distal esophagus on the right and the proximal esophagus on the left of the three-dimensional images. The time axes can be converted roughly to distance along the esophageal lumen by multiplying time by the peristaltic wave speed, typically ~2 cm/s in the middle esophagus. The resulting image is an approximation to the spatial structure along the esophageal lumen of coordinated waves of circular muscle contraction and local longitudinal shortening as they propagate distally through the middle esophagus. The clip studies discussed above indicate that, although the spatial structure of these peristaltic waves likely changes slowly as the waves move into the distal esophagus, the peaks in the circular muscle and local longitudinal shortening waves remain closely aligned.
Previous studies using mucosal clips have implicitly assumed that measured longitudinal shortening may be interpreted as indicating contraction of longitudinal muscle (note, for example, the reference to the "contracting longitudinal segment" in the quote above from Ref. 16). Recognizing that studies of longitudinal shortening, using either mucosal clips or the method introduced here, are ultimately directed at the contractile activity of longitudinal muscle, it is of interest to consider possible interpretations of longitudinal shortening in context with circular and longitudinal muscle contraction and the law of mass conservation. To interpret the observed longitudinal shortening in terms of longitudinal muscle contraction, consider the possibility that longitudinal shortening within a segment might occur in the absence of longitudinal muscle contraction within the same segment. In this case, the observed longitudinal shortening at peak intraluminal pressure is possible only if the longitudinal muscle in segments proximal or distal to the shortened segment were to extend. Because muscle has the property that it can only contract (or relax from a previous contraction), it is difficult to imagine a physiologically consistent scenario in which the longitudinal shortening measured in this study (and previous studies) could result from other than contraction of longitudinal muscle fibers. It has been remarked that contraction of circular muscle fibers concentrated within a narrow axial segment could, in principle, create a bulge in area, and consequently local longitudinal shortening, in adjacent esophageal segments as a result of local deformation of the muscle layer from compression by the contracting circular muscles. Although this effect is, in principle, a possible contributor to longitudinal shortening, the relative magnitude of the effect is undoubtedly minimal, given the material stiffness of muscle. Nevertheless, if the effect were measurable, it could serve only to reduce the peaks in A/A* in Figs. 4 and 5, implying a slight underprediction of maximal local longitudinal shortening, and the conclusions from this study, and related clip studies, remain unaffected. We conclude that the local longitudinal shortening measured in this study may be interpreted as indicating contraction of longitudinally aligned muscle fibers either within the esophageal wall or within the muscularis mucosa. Inferring the existence of circular muscle contraction by deviations from baseline of intrabolus intraluminal pressure, Figs. 4-6 indicate that a longitudinal muscle contraction wave (LMCW) accompanies the peristaltic circular muscle contraction wave (CMCW) as it traverses the esophagus, with peak longitudinal contraction well coordinated with peak circular muscular contraction. The LMCW envelops the CMCW, extending axially over a broader region of the esophagus. In particular, whereas longitudinal muscle contraction extends perhaps 2 cm distal to circular muscle contraction in the middle esophagus (assuming a wave speed of 2 cm/s), low-level contraction of longitudinal muscle appears to extend at least twice that distance proximally from the CMCW (especially apparent in Fig. 6). It may be that this low-level longitudinal shortening that persists in the proximal esophagus after passage of the circular muscle contraction wave reflects establishment of global shortening of the esophagus that persists until the contraction waves have passed through the lower esophageal sphincter and the esophagus has returned to its resting state.A physiological interpretation of local longitudinal shortening. The observation that longitudinal shortening envelops circular muscle contraction as the peristaltic wave traverses the esophagus was made in excised opossum esophagi by Sugarbaker et al. (21), leading them to suggest a physiological role for longitudinal muscle contraction. Sugarbaker et al. hypothesized that longitudinal shortening leads to local increases in circular muscle fiber number density within axial segments, which in turn increases the closure force generated by circular muscle contraction in that segment. Our observation that local longitudinal shortening peaks close to the peak in luminal pressure is consistent with this hypothesis.
Defining the force acting to close a segment of the esophagus of length L as Fclosure, an approximate relationship between average circular muscle fiber tension Tfiber and the closure force Fclosure can be developed from a force balance called the "Laplace equation"
|
(5) |
circ is the thickness of the
circular muscle layer and const = 1/(2
) is a constant.
This expression states that for given closure force
Fclosure, the average circular muscle fiber tension Tfiber can be reduced by increasing the
thickness of the muscle layer circ.
Alternatively, for given muscle fiber tension Tfiber, closure force
Fclosure is enhanced by increasing muscle layer
thickness circ. Figs. 4 and 5 show that
circ increases under local longitudinal
shortening, thus reducing the average circular muscle fiber tension for
given closure force, or increasing closure force for given fiber tension.
Closure force Fclosure, it may be shown, is
directly proportional to intraluminal pressure P. Thus if
the Sugarbaker et al. hypothesis is valid, one should expect to find a
correlation between peak intraluminal pressure
Pamp and maximum local longitudinal shortening
(Amus/A*mus)max.
This is, in fact, the result in Fig. 7, where the correlation
coefficient is r = 0.61. [The correlation would be
significantly higher if the two lower points at
(Amus/A*mus)max 2.6 were missing.] This conclusion is potentially significant for
the transport of solid boluses, for which clearance typically requires
multiple circular muscle contraction waves, each contraction wave
moving the bolus partially along the esophagus before passing over. The
correlation in Fig. 7 between closure force and longitudinal shortening
implies the generation of higher axial propulsive force on solid
boluses compared with no longitudinal shortening. This conjecture might
explain the in vitro result of Ren and Schulze-Delrieu
(17), who found that the transport of a solid bolus
through an excised opossum esophagus was enhanced if the esophagus was
allowed to shorten globally.
Figure 8 indicates a strong correlation between a lower delay
t between onset of circular and longitudinal muscle
contraction and higher pressure amplitude Pamp,
or closure force. The following speculative scenario is offered as a
possible explanation for this observation in terms of bolus transport.
Maximal longitudinal shortening occurs nearly coincident with peak
intraluminal pressure (Figs. 4-6). If, in addition, the rate of
increase in pressure is approximately the same from swallow to swallow,
then a shorter time delay t between the initiation of
circular and longitudinal muscle contraction would allow intraluminal
pressure to reach higher amplitude. In this way, the physiological
system could modulate the degree of contractile force required to
maintain closure for a given swallow by controlling the delay between
onset of circular and longitudinal muscle contraction.
Esophageal wall dimensions.
We have measured the thicknesses of the muscle layers in the middle
esophagus from the ultrasound images in the resting state to be
mus = 1.38 ± 0.07 mm
(long circ =
0.69 ± 0.05 mm) and mucosa 3.0 mm.
These data likely represent the most accurate measurements to date of
in vivo esophageal muscle layer thicknesses. Others' data include
dimensions from surgically removed esophagi from cadavers (2, 10,
20) in which the excised esophagus has shortened by
30-50%, and three other ultrasound studies, one of which is from
the same data set used in the current study (14), and a
related data set other from the same group (11). The one independent ultrasound study (4) quoted only an
approximate estimate of 3 mm (no precision bounds), presumably over the
middle-to-distal esophagus in achalasics and apparently including part
or all of the mucosa. Furthermore, whereas Biancani et al.
(2) did quote dimensions allowing for estimates of mucosal
thickness in the middle esophagus that match those in this study, they
did not quote surgically obtained dimensions of muscle wall thickness directly, and it was necessary to deduce these from their plots and
equations. Nevertheless, taking into account shortening of excised
esophagi, their measurements were within range of the more careful
ultrasound studies presented here and in Refs. 11 and 14.
| |
ACKNOWLEDGEMENTS |
|---|
This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant R01-DK-41436.
| |
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: J. G. Brasseur, Dept. of Mechanical Engineering, The Pennsylvania State Univ., University Park, PA 16802 (E-mail: brasseur{at}jazz.me.psu.edu).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Received 30 June 2000; accepted in final form 20 June 2001.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
1.
Balaban, DH,
Yamamoto Y,
Liu J,
Pehlivanov N,
Wisniewski R,
DeSilvey D,
and
Mittal RK.
Sustained esophageal contraction: a marker of esophageal chest pain identified by intraluminal ultrasonography.
Gastroenterology
116:
29-37,
1999[Web of Science][Medline].
2.
Biancani, P,
Zabinski MP,
and
Behar J.
Pressure, tension and force of closure of the human lower esophageal sphincter and esophagus.
J Clin Invest
56:
476-483,
1975.
3.
Brasseur, JG,
and
Dodds WJ.
Interpretation of intraluminal manometric measurements in terms of swallowing mechanics.
Dysphagia
6:
100-119,
1991[Medline].
4.
Deviére, J,
Dunham F,
Rickaert F,
Bourgeois N,
and
Cremer M.
Endoscopic ultrasonography in achalasia.
Gastroenterology
96:
1210-1213,
1989[Web of Science][Medline].
5.
Dodds, WJ,
Stewart ET,
Hodges D,
and
Zboralske FF.
Movement of the feline esophagus associated with respiration and peristalsis.
J Clin Invest
52:
1-13,
1972.
6.
Edmundowicz, SA,
and
Clouse RE.
Shortening of the esophagus in response to swallowing.
Am J Physiol Gastrointest Liver Physiol
260:
G512-G516,
1991
7.
Girerd, XJ,
Acar C,
Mourad J-J,
Boutouyrie P,
Safarm ME,
and
Laurent S.
Incompressibility of the human arterial wall: an in vitro ultrasound study.
Hypertension
10, Suppl6:
S111-S114,
1992.
8.
Goyal, RK,
Biancani P,
Phillips A,
and
Spiro HM.
Mechanical properties of the esophageal wall.
J Clin Invest
50:
1456-1465,
1971.
9.
Kahrilas, PJ,
Wu S,
Lin S,
and
Pouderoux P.
Attenuation of esophageal shortening during peristalsis with hiatus hernia.
Gastroenterology
109:
1818-1825,
1995[Web of Science][Medline].
10.
Liebermann-Meffert, D,
Allgöwer M,
Schmid P,
and
Blum AL.
Muscular equivalent of the lower esophageal sphincter.
Gastroenterology
76:
31-38,
1979[Web of Science][Medline].
11.
Liu, JB,
Miller LS,
Goldberg BB,
Feld RI,
Alexander AA,
Needleman L,
Castell DO,
Klenn PJ,
and
Millward CL.
Transnasal US of the esophagus: preliminary morphologic and function studies.
Radiology
184:
721-727,
1992
12.
Macovski, A.
Medical Imaging Systems. Englewood Cliffs, NJ: Prentice Hall, 1995.
13.
Massey, BT,
Gorny JM,
Kern MK,
Arndorfer RC,
Ryan C,
and
Hofmann C.
Is esophageal longitudinal muscle function affected by deglutitive inhibition? (Abstract).
Dysphagia
12:
111,
1997.
14.
Miller, LS,
Liu J-B,
Colizzo FP,
Marzano J,
Barbarevech C,
Hedwig K,
Leung L,
and
Goldberg BB.
Correlation of high-frequency esophageal ultrasound and manometry in the study of esophageal motility.
Gastroenterology
109:
832-837,
1995[Web of Science][Medline].
15.
Miller, LS,
Liu J-B,
Klenn PJ,
Dhuria M,
Feld RI,
and
Goldberg BB.
High-frequency endoluminal ultrasonography of the esophagus in human autopsy specimens.
J Ultrasound Med
12:
563-566,
1993[Abstract].
16.
Pouderoux, P,
Lin S,
and
Kahrilas PJ.
Timing, propagation, and effect of esophageal shortening during peristalsis.
Gastroenterology
112:
1147-1154,
1997[Web of Science][Medline].
17.
Ren, J,
and
Schulze-Delrieu K.
Movement of wax particles by contractions of the isolated opossum esophagus.
Am J Physiol Gastrointest Liver Physiol
258:
G164-G170,
1990
18.
Russ, JC.
The Image Processing Handbook. Boca Raton: CRC, 1995.
19.
Sonka, M,
Zhang X,
Siebes M,
Bissing MS,
DeJong SC,
Colling SM,
and
McKay CR.
Segmentation of intravascular ultrasound images: a knowledge-based approach.
IEEE Trans Med Imaging
14:
719-732,
1995[Medline].
20.
Stein, HJ,
Liebermann-Meffert D,
DeMeester T,
and
Siewert JR.
Three-dimensional pressure image and muscular structure of the human lower esophageal sphincter.
Surgery
117:
692-698,
1995[Web of Science][Medline].
21.
Sugarbaker, DJ,
Rattan S,
and
Goyal RK.
Mechanical and electrical activity of esophageal smooth muscle during peristalsis.
Am J Physiol Gastrointest Liver Physiol
246:
G145-G150,
1984
22.
Tsuiki, K,
and
Ritman EL.
Direct evidence that left ventricular myocardium is incompressible through systole and diastole.
Tohoku J Exp Med
132:
119-120,
1980[Web of Science][Medline].
23.
White, FM.
Fluid Mechanics. New York: McGraw Hill, 1986.
24.
Yamamoto, Y,
Liu J,
Smith TK,
and
Mittal RK.
Distension-related responses in circular and longitudinal muscle of the human esophagus: an ultrasonographic study.
Am J Physiol Gastrointest Liver Physiol
275:
G805-G811,
1998
This article has been cited by other articles:
|
|
 |
|
  S. K. Ghosh, P. J. Kahrilas, and J. G. Brasseur Liquid in the gastroesophageal segment promotes reflux, but compliance does not: a mathematical modeling study Am J Physiol Gastrointest Liver Physiol, November 1, 2008; 295(5): G920 - G933. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Liao, D. Lelic, F. Gao, A. M. Drewes, and H. Gregersen Biomechanical functional and sensory modelling of the gastrointestinal tract Phil Trans R Soc A, September 28, 2008; 366(1879): 3281 - 3299. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Korsapati, A. Babaei, V. Bhargava, and R. K. Mittal Cholinergic stimulation induces asynchrony between the circular and longitudinal muscle contraction during esophageal peristalsis Am J Physiol Gastrointest Liver Physiol, March 1, 2008; 294(3): G694 - G698. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. A. Tipnis, J. Liu, J. L. Puckett, and R. K. Mittal Common cavity pressure during gastroesophageal reflux: reassessment using simultaneous pressure, impedance, and ultrasound imaging Am J Physiol Gastrointest Liver Physiol, June 1, 2006; 290(6): G1149 - G1156. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. K. Mittal, B. Padda, V. Bhalla, V. Bhargava, and J. Liu Synchrony between circular and longitudinal muscle contractions during peristalsis in normal subjects Am J Physiol Gastrointest Liver Physiol, March 1, 2006; 290(3): G431 - G438. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. K. Ghosh, P. Janiak, W. Schwizer, G. S. Hebbard, and J. G. Brasseur Physiology of the esophageal pressure transition zone: separate contraction waves above and below Am J Physiol Gastrointest Liver Physiol, March 1, 2006; 290(3): G568 - G576. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. K. Ghosh, P. J. Kahrilas, T. Zaki, J. E. Pandolfino, R. J. Joehl, and J. G. Brasseur The mechanical basis of impaired esophageal emptying postfundoplication Am J Physiol Gastrointest Liver Physiol, July 1, 2005; 289(1): G21 - G35. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Shi, J. E. Pandolfino, Q. Zhang, I. Hirano, R. J. Joehl, and P. J. Kahrilas Deglutitive inhibition affects both esophageal peristaltic amplitude and shortening Am J Physiol Gastrointest Liver Physiol, April 1, 2003; 284(4): G575 - G582. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Pehlivanov, J. Liu, G. S. Kassab, C. Beaumont, and R. K. Mittal Relationship between esophageal muscle thickness and intraluminal pressure in patients with esophageal spasm Am J Physiol Gastrointest Liver Physiol, June 1, 2002; 282(6): G1016 - G1023. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
