Am J Physiol Gastrointest Liver Physiol 285: G1014–G1027, 2003.
First published July 3, 2003; 10.1152/ajpgi.00138.2003.
Longitudinal and circumferential spike patches in the
canine small intestine in vivo
Wim J. E. P. Lammers, Luc Ver Donck, Jan A. J. Schuurkes, and Betty Stephen
Faculty of Medicine and Health Sciences, Department of Physiology, United Arab Emirates University, Al Ain,
United Arab Emirates; and the Department of Gastrointestinal Pharmacology, Johnson and Johnson
Pharmaceutical Research and Development, Division of Janssen Pharmaceutica, B-2340 Beerse, Belgium
Submitted 25 March 2003; accepted in final form 27 June 2003
slow waves; spike patches; duodenum; jejunum; ileum
along the small intestine. Recently, it was shown that
spikes propagate for a limited extent in time and space
before terminating spontaneously, thereby activating a
relatively small area termed a spike “patch” (11). These
patches were demonstrated in isolated tissues in vitro,
in a single species, the cat, and in one part of the
intestine, the duodenum. The question therefore arises
as to whether spike patches also occur in another
species, in the whole organism, and in other parts of
the small intestine.
We are now presenting an approach enabling us to
record in vivo electrical signals from 240 extracellular
sites simultaneously from the serosal surface of the
intact canine small intestine using an open-abdomen
anesthetized animal model. With a custom-designed
240-electrode array, we sampled the electrical activities in a 2 ⫻ 5-cm area in different parts of the small
intestine, ranging from the duodenum to the distal
parts of the ileum. In all recordings, slow waves were
visible that were often followed by one or more spikes.
Analysis of the propagation of these spikes revealed
the presence of spike patches at all levels in the small
intestine. Two types of spikes were found; the first
type, previously described in the cat duodenum (11,
12), propagated in the longitudinal direction, originated anywhere around the intestinal tube, and occurred predominantly in the duodenum. The second
type, which occurred much more frequently in the
jejunum and the ileum, originated in most cases along
the antimesenteric line and propagated in the circumferential direction. Both types of spikes conducted over
limited space and time before terminating spontaneously.
METHODS
by slow waves and by
action potentials (spikes) that may or may not occur in
the wake of the slow wave (6, 18, 19, 23). Several
studies have investigated the temporal relationship
between slow waves and spikes and between spikes
and motility (2, 3, 5, 7, 16, 25, 26). Not much attention,
however, has been given to the spatial pattern of spike
propagation (6, 17) or to possible regional variations
This study required the use of 12 female beagles (12.2 ⫾
2.1 kg) that had been fasted the day before the experiments.
Anesthesia was induced with scopolamine (0.015 mg/kg),
lofentanil (0.070 mg/kg), and succinylcholine (1 mg/kg) and
maintained throughout the duration of the experiments (3–4
h) with etomidate (1.5 mg 䡠 kg⫺1 䡠 h⫺1) and fentanyl (0.025
mg 䡠 kg⫺1 䡠 h⫺1; see Ref. 27). The animals were ventilated
through a tracheal tube, and the left femoral artery was
Address for reprint requests and other correspondence: W. J. E. P.
Lammers, Dept. of Physiology, Faculty of Medicine and Health
Sciences, P. O. Box 17666, Al Ain, United Arab Emirates (E-mail:
[email protected]).
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.
INTESTINAL MOTILITY IS INITIATED
G1014
0193-1857/03 $5.00 Copyright © 2003 the American Physiological Society
http://www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.3 on June 17, 2017
Lammers, Wim J. E. P., Luc Ver Donck, Jan A. J.
Schuurkes, and Betty Stephen. Longitudinal and circumferential spike patches in the canine small intestine in vivo.
Am J Physiol Gastrointest Liver Physiol 285: G1014–G1027,
2003. First published July 3, 2003; 10.1152/ajpgi.00138.
2003.—In an open-abdominal anesthetized and fasted canine
model of the intact small intestine, the presence, location,
shape, and frequency of spike patches were investigated.
Recordings were performed with a 240-electrode array (24 ⫻
10, 2-mm interelectrode distance) from several sites sequentially, spanning the whole length of the small intestine. All
240 electrograms were recorded simultaneously during periods of 5 min and were analyzed to reconstruct the origin and
propagation of individual spikes. At every level in the small
intestine, spikes propagated in all directions before stopping
abruptly, thereby activating a circumscribed area termed a
“patch.” Two types of spikes were found: longitudinal spikes,
which propagated predominantly in the longitudinal direction and occurred most often in the duodenum, and a second
type, circumferential spikes, which propagated predominantly in the circular direction and occurred much more
frequently in the jejunum and ileum. Circumferential spikes
conducted faster than longitudinal spikes (17 ⫾ 6 and 7 ⫾ 2
cm/s, respectively; P ⬍ 0.001). Circumferential spikes originated in ⬎90% of all cases from the antimesenteric border,
whereas longitudinal spikes were initiated all around the
circumference of the intestinal tube. Finally, the spatial
sequence of spike patches after the slow wave was very
irregular in the upper part of the intestine but much more
regular in the lower part. In conclusion, spikes and spike
patches occur throughout the small intestine, whereas their
type, sites of origin, extent of propagation, and frequencies of
occurrence differ along the length of the small intestine,
suggesting differences in local patterns of motility.
G1015
SPIKE PATCHES IN THE CANINE SMALL INTESTINE
In contrast to previous mapping studies (13, 11), the electrode tips were flush with the bottom surface of the assembly
(Fig. 1B) so that there were no protruding sharp edges,
avoiding the risk of damage to the underlying tissue. Care
was also taken not to put too much pressure on the intestinal
tube. After the electrode assembly was positioned (Fig. 1, C
and D), a temperature probe was located alongside the intestine, and the whole region was packed with moist cotton pads
to prevent drying of the tissue (Fig. 1E). A heat lamp kept the
experimental area at body temperature during subsequent
control and recording periods (36.4 ⫾ 0.6°C). After recordings
from one area had been performed, another part of the
intestine was chosen, and the procedure was repeated. The
order in which the areas were chosen was changed in every
experiment. Figure 1G shows the location of the recording
sites along the small intestine in all dogs.
After the positioning of the electrode assemblies at a location,
a 10-min stabilization period was allowed, followed by 5 min of
Fig. 1. Overview of the experimental
field. A: opened canine abdomen with
an intestinal loop positioned on a wet
cotton pad without any fixation. B:
view on the 240 electrode array with
the electrode tips, arranged in a 24 ⫻
10 array, flush with the bottom of the
assembly. Interelectrode distance 2
mm, total dimensions 46 ⫻ 18 mm. C
and D: controlled lowering of the electrode assembly on the intestinal loop
until electrode tips made contact with
the serosal surface. Dashed white line
is drawn to indicate the antimesenteric
border. E: covering of the experimental
area with additional wet cotton pads. A
heat lamp (not shown) keeps the area
at body temperature, as monitored by a
thermistor (cable visible in bottom left).
F: diagram of the intestinal loop with
the 24 ⫻ 10 electrode array superimposed. G: histogram showing the frequency of the recording sites along the
(normalized) length of small intestine
in 12 dogs.
AJP-Gastrointest Liver Physiol • VOL
285 • NOVEMBER 2003 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.3 on June 17, 2017
cannulated to record systemic blood pressure. The average
systolic and diastolic pressures in all experiments were
156.8 ⫾ 23.4 and 90.8 ⫾ 15.0 mmHg, respectively. There was
a small but not significant drop in the average blood pressure
during the course of the experiments from 118 ⫾ 15 to 105 ⫾
10 mmHg over the 3- to 4-h experiments, accompanied by a
reduction in heart rate during the same time period (83.5 ⫾
20.2 to 59.9 ⫾ 8.5 beats/min).
After a median laparotomy, the abdominal walls were
carefully retracted, and a loop of the small intestine was
identified (Fig. 1A). A wet cotton pad was positioned under
the loop, and a 240-electrode assembly was gradually lowered on the serosal surface until physical contact was made
between the tissue and bottom surface of the assembly. In
eight dogs, as shown in Fig. 1, the electrodes covered one-half
of the intestinal circumference, from the mesenteric to the
antimesenteric borders. In a second series of four dogs, the
area around the antimesenteric border was recorded.
G1016
SPIKE PATCHES IN THE CANINE SMALL INTESTINE
recording. Electrical recordings were performed unipolarly with
a subcutaneous needle in the back right leg as the indifferent
pole. All electrodes were connected through shielded wires to
240 AC preamplifiers were the signals were amplified (⫻4,000),
filtered (2–400 Hz), digitized (1-kHz sampling rate), and stored
on the hard disk of a laptop. In each dog, this recording procedure was repeated at four other sites along the length of the
small intestine. The location of each recording site was marked
with a loose ligature looped around the intestinal tube. At the
end of the experiment, the animals were killed, the small
intestines were removed in toto, and the total length and the
distance of the recording sites from the pylorus were measured.
The average length of the canine small intestine in 12 animals
was 290 ⫾ 48 cm (range 222–366 cm).
Analysis was performed by choosing at random a 16-s
window from the 1st, 3rd, and 5th min of each 5-min recording. From each 16-s window, the propagation of one slow
wave and all spikes after that slow wave were analyzed. The
signals were filtered digitally (20-point moving average) and
displayed on-screen in sets of 20–24 electrograms at a time
(Fig. 2A). The local activation time of a slow wave was
identified by the moment of maximum negative slope (4, 13)
and marked with a cursor. The local activation times of the
spikes were marked by the steepest slope of the signal, which
was usually negative (11). All local activation times are
related to a common reference time, determined by the timing of the first detected slow wave in the mapped area. After
time marking all slow waves and spikes, their activation
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.3 on June 17, 2017
Fig. 2. Reconstructing the pattern of
propagation of slow waves and spikes
(canine jejunum). A: plot of 23 electrograms recorded from a single line of
electrodes (oral to aboral orientation)
with the individual locations shown in
B–E. A slow wave is displayed, propagating aborally, followed by spikes. The
spikes occurred in several clusters, as
indicated by ovals. The timing of both
the slow waves and the spikes was determined by the most rapid negative
slope in each signal. These times, in
milliseconds (relative to the occurrence
of the first slow wave in channel 1), are
plotted in activation maps in B–E. B: all
individual slow-wave activation times,
measured during this particular cycle,
are plotted in a grid representing the
original electrode array. The background color at each electrode site is
determined by the local activation time
in time steps of 200 ms. Electrodes located outside of the intestinal loop did
not record any signal, and these sites
were left empty. C: final activation map
of the slow wave, with the actual activation times left out for the sake of
clarity. Isochrones were drawn in time
steps of 200 ms, and additional details
such as an arrow are added to demonstrate the major propagation pathway.
D and E: procedure for a single spike
with an isochrone time step of 25 ms. In
this case, the spike originated from an
area located in the top right region,
close to the antimesenteric border, and
propagated predominantly in the circumferential direction. The spike did
not propagate ⬎6–8 mm in the longitudinal direction. In this recording, the
contact of the electrode assembly was
not optimal at the aboral end of the
intestinal segment, as indicated by a
lack of slow waves in electrograms 21–
23. Therefore, this part of the map was
not analyzed.
AJP-Gastrointest Liver Physiol • VOL
285 • NOVEMBER 2003 •
www.ajpgi.org
G1017
SPIKE PATCHES IN THE CANINE SMALL INTESTINE
times were displayed in the format of a grid of the original
recording array (24 ⫻ 10; Fig. 2B). For slow waves, isochrones were drawn manually around areas activated in
steps of 50–1,000 ms (Fig. 2, B and C) and for spikes in steps
of 10–100 ms (Fig. 2, D and E).
Housing, care, type of anesthesia, and experimental procedures were approved by the institutional ethics committee.
All pooled data are given as averages and SDs. Significance
was tested with Student’s t-test.
RESULTS
Two types of spike patches were found in the canine
small intestine (Fig. 3). The first type, the longitudinal
spike patch, resembled that described earlier in the
isolated cat duodenum in vitro (11) and showed predominant propagation in the longitudinal direction. A
second type, not described before, showed mostly propagation from the antimesenteric border in the circumferential direction and toward the mesenteric border of
the intestinal tube. Both types of spikes stopped propagating spontaneously in all directions after spreading
over a relatively small area, termed a patch (11).
Spike patches occurred throughout the length of the
small intestines but showed regional variations in type
and frequencies. Figures 4–7 display representative
samples of the recorded electrograms (A) and the reconstructed activation maps of slow waves (B) and of
AJP-Gastrointest Liver Physiol • VOL
285 • NOVEMBER 2003 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.3 on June 17, 2017
Fig. 3. Origin, size, and electrograms of
longitudinal and circumferential spikes.
A: propagation of a slow wave along a
canine duodenum. This was followed by
several spike patches of which two examples are shown. B: map on left shows
a longitudinal spike originating from a
location approximately halfway between the mesenteric and the antimesenteric border, and propagating in the
aboral direction for ⬎100 ms before terminating spontaneously. In map on
right, a patch was initiated along the
antimesenteric border and propagated
predominantly in the circular direction
before stopping spontaneously, after
⬃30 ms. The central isochrone bar relates to the longitudinal patch in steps
of 50 ms and to the circumferential
patch in steps of 10 ms. In C, a and b,
the 2 patches are repeated, and electrograms, recorded along the dominant
propagation pathway (nos. 1 to 10 along
the longitudinal direction for the longitudinal spike and letters a to h along
the circular direction for the circumferential spike), are displayed on right.
From the time latency and the covered
distance, conduction velocities were calculated, showing that circumferential
spike conduction is faster than longitudinal spikes.
G1018
SPIKE PATCHES IN THE CANINE SMALL INTESTINE
longitudinal or circumferential spike patches (C and E)
from the duodenum, the jejunum, the proximal, and
the distal ileum, respectively.
At all sites, slow waves conducted as uniform homogeneous broad waves of excitation. Most of the slow
waves propagated in the aboral direction, although
sometimes oral propagation did occur, as shown in Fig.
4. In addition, the speed of slow wave propagation
declined from the duodenum toward the distal ileum
(1, 15). Therefore, in the duodenum, it took the slow
wave ⬃0.5 s to propagate along a distance of 5 cm (Fig.
AJP-Gastrointest Liver Physiol • VOL
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.3 on June 17, 2017
Fig. 4. Propagation of a slow wave and
several spikes in the canine duodenum
(18 cm from pylorus). A: 24 electrograms recorded from sites located in the
longitudinal direction, as shown by corresponding circled numbers in B. Most
traces showed several spikes that were
clustered in groups 1–6 as indicated by
ovals. Amplifier 9 was defective during
this recording. B: uniform and homogeneous propagation of the slow wave,
which in this case conducted from aboral to oral. Slow-wave isochrones were
plotted every 50 ms. C: sequence of activation of 3 individual spikes (nos. 2, 4,
and 6; isochrones every 50 ms). Spikes
originated at circumscribed locations,
indicated by the stars, and propagated
predominantly in the longitudinal direction before terminating abruptly.
Some spikes conducted in the oral,
orthodromic direction, such as nos. 2
and 4, whereas others propagated in
the aboral, antidromic direction (i.e.,
no. 6). Note that the spikes from patch 4
are not visible in the electrograms in A
since that conduction had already
stopped before reaching that row of
electrodes. D: composite spike map displays the location and shape of all 8
spike patches that occurred after this
particular slow wave. All patches were
oriented in the longitudinal direction
with considerable variation in location,
size, and overlap between patches. The
majority of the mapped area, however,
was not activated at all (83.7%).
4B), whereas in the distal ileum, the slow wave needed
⬎6 s to cover the same distance (Fig. 7B).
At all locations throughout the small intestine,
spikes occurred after the slow wave in varying numbers and at various sites. The pattern of propagation of
945 individual spikes (Table 1) was analyzed, and
representative samples, obtained at different levels in
the small intestine, are presented in Figs. 4C–7C. In
all cases, spikes propagated in limited areas and
stopped conducting abruptly and spontaneously,
thereby activating a circumscribed area (11). In addi285 • NOVEMBER 2003 •
www.ajpgi.org
G1019
SPIKE PATCHES IN THE CANINE SMALL INTESTINE
tion, there were significant variations in the direction
of spike propagation and the shape of spike patches
along the small intestine. In the duodenum (Fig. 3C),
spikes propagated predominantly in the longitudinal
direction, either in the oral or in the aboral direction,
before stopping spontaneously. Hence these patches
tended to be longer in the longitudinal direction and
narrow in the circumferential direction. The reverse
was true in the jejunum and ileum, where spikes propagated predominantly in the circumferential direction,
AJP-Gastrointest Liver Physiol • VOL
and the patches were wider in the circular and smaller
in the longitudinal direction (Figs. 5C, 6C, and 7C). In
the vast majority of cases (⬎90%), circumferential
spikes propagated from the antimesenteric toward the
mesenteric border.
Table 1 details the dimensions of all 945 analyzed
spike patches. Both types of spikes could occur anywhere along the small intestine, but the majority of
longitudinal spikes was located in the duodenum,
whereas the majority of circumferential spikes was
285 • NOVEMBER 2003 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.3 on June 17, 2017
Fig. 5. Typical propagation of slow
waves and spikes in the canine jejunum (61 cm from pylorus). Orientation is similar to that in Fig. 4. A: plot
of an aborally propagating slow wave
followed by several spike clusters (1–
6). B: homogeneous propagation of
the slow wave in the aboral direction.
Slow-wave isochrones were plotted
every 200 ms. C: sequence of activation of 3 individual spikes (nos. 2, 4,
and 6; isochrones every 20 ms) showing predominant conduction in the
circumferential direction and limited
propagation in the longitudinal direction. D: composite spike map displays
the boundaries of all 6 spike patches
that occurred after this particular
slow wave. There is considerable variation in size between patches,
whereas the location and sequence of
spike patches follow the direction of
propagation of the slow wave. In contrast to patches in the duodenum,
spike patches in the jejunum tend to
be oriented in the circular direction
and originate from the antimesenteric
border.
G1020
SPIKE PATCHES IN THE CANINE SMALL INTESTINE
found in jejunum and ileum. In addition, circumferential patches are not the same as longitudinal patches,
which have been rotated by 90 degrees. For example, in
the duodenum, the distance in propagation of both
spikes was similar (10.7 mm in the longitudinal direction for longitudinal spikes and 11.5 mm in the circular
direction for circumferential spikes), but the width of
the longitudinal spike patches was significantly narrower (3.1 mm) than that of the circumferential spike
patches (11.0 mm; P ⬍ 0.001).
The conduction velocity of both types of spikes is also
different. In 80 large patches, an estimate of the conAJP-Gastrointest Liver Physiol • VOL
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.3 on June 17, 2017
Fig. 6. Typical propagation of a slow
wave and several spikes in the canine
proximal ileum (104 cm from pylorus).
Orientation is similar to Fig. 4. A: plot of
an aborally propagating slow wave followed by 15 spike clusters indicated by
ovals. B: propagation of the slow wave
in the aboral direction (isochrones every
500 ms). C: sequence of propagation of 3
individual spikes (nos. 2, 6, and 14; isochrones every 50 ms). All 3 spikes originated from the antimesenteric border.
D: composite spike map displays the
boundaries of all 15 spike patches that
occurred after this particular slow
wave, showing, as at other locations,
considerable variation in size between
patches. Like in the jejunum, spikes
originated in most cases along the antimesenteric border, and the patches
tended to be larger in the circumferential than in the longitudinal direction.
duction velocity of the spikes was obtained by measuring the time difference between the initiation and the
termination of the spike and the distance between
those two sites (as shown in Fig. 3C). This was performed in the longitudinal direction for longitudinal
spikes (n ⫽ 20) and in the circular direction for circumferential spikes (n ⫽ 60). The conduction velocity of
circumferential spikes, 17.1 ⫾ 6.1 cm/s, is significantly
higher than that of longitudinal spikes, 6.9 ⫾ 1.7 cm/s
(P ⬍ 0.01).
In several cases, apparent deviations from the usual
pattern of propagation of circumferential spikes were
285 • NOVEMBER 2003 •
www.ajpgi.org
G1021
SPIKE PATCHES IN THE CANINE SMALL INTESTINE
observed, and Fig. 8 presents examples of two types of
events. In the first type, described in A, it was sometimes possible to record spike signals propagating in
the opposite direction, from the mesenteric toward the
antimesenteric border. This always occurred in conjunction with a circumferential spike propagating from
the antimesenteric line. Our explanation for this phenomenon is that the circumferential spike, initiated at
the antimesenteric border, had also propagated along
the reverse, not mapped, part of the intestinal tube. It
then crossed the mesenteric border and continued to
AJP-Gastrointest Liver Physiol • VOL
propagate along the mapped side of the tube. In other
words, the spike conducted along the back wall of the
intestine for more than half the circumference and
“wrapped” itself around to the other side. Such an
event occurred 30 times in the ileum but never in the
jejunum or the duodenum (Table 1).
The second event is shown in Fig. 8B and shows the
initiation of a circumferential spike at a site other than
at the antimesenteric border. In this case, the second
spike originated approximately halfway between the
antimesenteric and the mesenteric border.
285 • NOVEMBER 2003 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.3 on June 17, 2017
Fig. 7. Typical slow-wave and spike
propagation in the canine distal ileum
(297 cm from pylorus). Orientation is
similar to that in Fig. 4. A: low-amplitude slow waves are followed by numerous spikes. Corresponding spikes at
neighboring traces are indicated by
short vertical lines. B: slow, homogeneous aboral propagation of the slow
wave. Slow-wave isochrones were plotted every second. C: sequence of activation of 3 individual spikes (nos. 12, 22,
and 33; isochrones every 10 ms). Spikes
originated at circumscribed locations,
mostly located along the antimesenteric border, and propagated predominantly in the circumferential direction
before terminating abruptly. In some
cases, the spikes propagated over more
than half the intestinal circumference
(as in nos. 12 and 33), whereas in other
cases (such as in no. 22) the activated
distance was smaller. D: composite
spike map of the boundaries of all 40
spike patches that occurred after this
particular slow wave (several patches
are not visible since they underlie
others).
G1022
SPIKE PATCHES IN THE CANINE SMALL INTESTINE
Table 1. Characteristics of intestinal spike patches
Longitudinal Patches
Duodenum
Jejunum
Proximal ileum
Middle ileum
Distal ileum
Totals
Totals
n
83
56
186
357
263
945
63
31
21
7
4
126
Longitudinal
dimension,
mm
Circular
dimension,
mm
10.7 ⫾ 7.1
11.8 ⫾ 5.8
6.9 ⫾ 2.5
6.3 ⫾ 1.8
4.8 ⫾ 1.9
3.1 ⫾ 1.9*
3.3 ⫾ 1.6*
5.2 ⫾ 3.1
4.6 ⫾ 1.5
6.0 ⫾ 2.8
Circumferential Patches
n
20
25
165
350
259
819
Longitudinal
dimension,
mm
Circular
dimension,
mm
11.0 ⫾ 5.8
7.6 ⫾ 2.8
7.0 ⫾ 3.5
6.5 ⫾ 2.3
6.3 ⫾ 2.9
11.5 ⫾ 3.2†
9.7 ⫾ 3.3†
9.1 ⫾ 3.3†
8.4 ⫾ 3.3†
10.3 ⫾ 4.2†
Wraps
0
0
5
12
13
30
Location from
pylorus, cm
11.9 ⫾ 5.0
50.5 ⫾ 14.7
124.0 ⫾ 34.1
201.4 ⫾ 44.1
267.5 ⫾ 40.8
All data are means ⫾ SD; n, no. of spike patches analyzed in each part of the small intestine. Location from pylorus plots the average values
of the recording positions for each part as measured from the pylorus. Significance levels were determined as follows: * the circular dimension
of longitudinal patches is significantly smaller than their longitudinal dimension (P ⬍ 0.001) in duodenum and jejunum; † the circular
dimension of circumferential patches is not significantly different from their longitudinal dimension (P ⫽ not significant) at any level.
AJP-Gastrointest Liver Physiol • VOL
example of this difference in behavior is shown in Fig.
10. Figure 10, left, shows the situation in the duodenum, whereas Fig. 10, right, displays the situation in
the distal ileum, as measured in three dogs. Maps in
Fig. 10A display the origin of individual spikes that
occurred after a single slow wave. The chronological
sequence of the origin of every spike is indicated. The
locations of the origin of the first spike and of the last
spike are also indicated. The propagation of the slow
wave in all instances was uniform and occurred from
oral to aboral.
In the duodenum (Fig. 10, maps on left), successive
spikes originated often far away from each other, and
the distances between successive origins were therefore often large. This is also shown in Fig. 10B, top left,
wherein the distances are plotted sequentially for all
three cases (mean distance 17.8 ⫾ 10.9 mm). The
situation is very different in the distal ileum, as shown
in Fig. 10, maps on right. There, the origins of successive spikes follow each other closely, and the distances
between origins were usually quite small, as is also
visible in Fig. 10B, top right (mean distance 3.5 ⫾ 2.8
mm; P ⬍ 0.001). If a large step occurred between
successive spike origins, then it usually took place in
the circumferential direction and rarely in the longitudinal direction. Moreover, in the ileum, nearly all
spikes originated along the antimesenteric border,
whereas this was not the case in the duodenum.
This spatial difference in spike origin behavior between the proximal and the distal small intestines
does, however, not take place in the temporal dimension. In Fig. 10B, bottom, the timing of each spike was
plotted in relation to the time of the leading edge of the
slow wave at that site. As indicated by the histograms,
spikes occurred in a relatively broad time period ranging from 0 to 3 s after the passage of the slow wave
(1.47 ⫾ 0.43 and 1.42 ⫾ 0.73 s for duodenum and distal
ileum, respectively; P ⫽ not significant).
DISCUSSION
In previous studies (11, 14), the demonstration of
spike patches was limited to the feline duodenum in
vitro. This study now extends these findings to support
285 • NOVEMBER 2003 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.3 on June 17, 2017
In the four dogs that were mapped along the antimesenteric border (see METHODS), signal analysis of
the spikes revealed several more striking features,
displayed in Fig. 9. As shown in the three isochrone
maps from examples recorded in the duodenum, jejunum, and ileum (Fig. 9, A–C), the initial activation,
which colored the isochrone red, was not only located in
the neighborhood of the antimesenteric border, as expected, but was also relatively large compared with
later isochrones. In fact, in the first 10 ms (the width of
the first isochrone), no pattern of propagation could be
determined. This seems to indicate near-simultaneous
activation of the whole area. It is only later, in the
following isochrones, that a pattern of conduction becomes clear as the spike propagates away from the
antimesenteric border. This last point is also visible in
Fig. 9D, where the intestine was mapped along one
side of the tube. Electrodes 1, 2, and 3, located close to
the antimesenteric border, were activated virtually
simultaneously, whereas electrodes 4–7 were activated
progressively later in time, as also shown by the electrograms displayed in Fig. 9D, right. In conjunction
with this feature, the morphology of the spikes also
differed. In the area of initiation, the spike waveform
starts with an initial negativity (indicated in electrograms 1–3 in Fig. 9D), whereas an initial positive
upstroke appears in electrograms 4–7, in conjunction
with propagation. Such an initial negativity was also
visible in the electrograms recorded in the examples
displayed in Fig. 9, A–C, as shown by the signals
displayed on the right side of each corresponding map.
In all areas of initial activity, initial negativity was
found in all signals. Further away from this area, a
positive signal appeared, as shown in Fig. 9, A and C,
indicating propagation (23). Such large areas of initial
negativity were never seen in longitudinal spike
patches.
Aside from the two types of spike patches in the
small intestine, another feature was consistently observed, and this relates to the spatiotemporal sequence
of spike patches after the slow wave. In the lower
intestine, spike patches tended to follow each other
regularly in time and space, whereas in the upper part
of the duodenum, such an order was often absent. An
G1023
SPIKE PATCHES IN THE CANINE SMALL INTESTINE
the concept that spike patches also occur in vivo, in
another species, and that they occur at all levels along
the small intestine.
In addition, this study reveals the existence of two
types of spikes. The first type, longitudinal spikes, has
been shown previously in the isolated feline duodenum
(11), and current results are in agreement with previous descriptions as follows: 1) these spikes propagate
predominantly in the longitudinal direction, 2) spike
propagation stops spontaneously in all directions,
thereby activating a limited area, termed a patch, 3)
several patches may occur after a slow wave (Fig. 4A),
AJP-Gastrointest Liver Physiol • VOL
and 4) within a patch, spikes may propagate in the oral
or in the aboral direction (Fig. 4, A and C).
The second type of spikes, labeled circumferential
spikes based on their dominant direction of propagation, has not been described before. The following
properties characterize circumferential spikes and
differentiate them from longitudinal spikes: 1) spike
propagation occurs in the circumferential direction,
2) spike conduction velocity in the circumferential
direction is faster than longitudinal spike velocity in
the longitudinal direction, 3) the large majority of
circumferential spikes are initiated along the an285 • NOVEMBER 2003 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.3 on June 17, 2017
Fig. 8. Special patterns of propagation
of circumferential spikes. In A, wrapping of a circumferential spike is
shown. The electrograms, the locations
of which are shown in the map, show a
sequence of four spikes, demonstrating
propagation from the antimesenteric
border toward but terminating before
reaching the mesenteric border. Halfway along the tracing, another deflection, indicated by the red arrow, occurred slightly after the second spike,
from the other side of the intestinal
tube, and propagated toward the antimesenteric border. The corresponding activation map is shown below the
tracings together with a diagram. The
map and the diagram propose that the
circumferential spike had also propagated along the other, unmapped side
of the intestinal tube, reached the mesenteric border, and continued to propagate into the mapped region before
terminating spontaneously. Such a
wrap of the spike was seen 30 times, all
located in the ileum (Table 1). In B, a
relatively rare occurrence of a circumferential spike originating from a site
distant from the antimesenteric border
is demonstrated. The electrograms, recorded along a circular line as indicated in the map, recorded a normal
circumferential spike followed by an
abnormal spike. The first spike was
initiated along the antimesenteric border and propagated in the circumferential direction all the way to the mesenteric border. The second spike was initiated halfway between both borders
and propagated in the circular direction toward both borders.
G1024
SPIKE PATCHES IN THE CANINE SMALL INTESTINE
timesenteric border (Figs. 3 and 5–10), whereas longitudinal spikes may be initiated anywhere along the
intestinal tube’s circumference (Figs. 3, 4, and 10),
and 4) at its site of initiation, the waveform of the
circumferential spikes shows prominent initial negativity over relatively large areas (Fig. 9) in contrast
to longitudinal spikes, where such areas were never
seen (Figs. 3 and 4).
The topographical distributions of the two spike
patches are also different. Longitudinal spike patches
occur mainly in the duodenum, whereas circumferential patches occur much more frequently in the jejuAJP-Gastrointest Liver Physiol • VOL
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.3 on June 17, 2017
Fig. 9. The shape of circumferential
spike waveforms near to and distant
from the antimesenteric border. In A, B,
and C, the antimesenteric areas of the
duodenum, jejunum, and the ileum are
mapped. Activation maps of circumferential spike patches are shown on left,
whereas the waveform of the spike electrograms (duration 120 ms) recorded in
that region are on right. In all 3 locations in the small intestine, spikes originated along the antimesenteric border
(first red isochrone) before subsequent
propagation occurred away from that
area. The electrograms show uniform
and large-amplitude signals recorded
close to the antimesenteric border, all of
which showed initial negativity. Further away from the area of initiation, a
positive deflection was sometimes visible in some areas, indicated by filled
arrowheads, suggesting active propagation (23). This is seen most clearly in D,
mapped between the mesenteric and
the antimesenteric border. A circumferential spike was initiated close to electrode 1 and activated a relatively large
area within 10 ms (as indicated by electrograms 1–3). Thereafter, propagation
occurred toward the mesenteric border
before spontaneously terminating after
60 ms. D, right, shows the gradual transition in the shape of the spike. In electrodes 1–3, the spike waveforms occur
simultaneously and are preceded by an
initial negativity (indicated by the asterisks), whereas further away, in electrodes 4–7, the waveform occurs later,
and a positive deflection is visible (indicated by filled arrowheads), suggesting
active propagation.
num and the ileum (Table 1) so that a countergradient
in the probability of both types is present along the
length of the small intestine. Furthermore, circumferential spike patches are not simply longitudinal
patches turned 90 degrees around. Circumferential
patches are much wider in the longitudinal dimension
than longitudinal patches are in the circular direction.
In the duodenum, for example, the circumferential
spike patches are 11.5 mm long in the longitudinal
direction, whereas the longitudinal spike patches are
much more narrow and measure only 3.1 mm in the
circular direction (Table 1).
285 • NOVEMBER 2003 •
www.ajpgi.org
G1025
SPIKE PATCHES IN THE CANINE SMALL INTESTINE
The pattern of spike propagation within the two
types of spike patches is also different. Circumferential
spikes are predominantly initiated along the antimesenteric border; therefore, conduction may occur along
both sides of the intestinal wall toward the mesenteric
border, effectively straddling the intestine to varying
degrees. In several cases, conduction along one side of
the “saddle” may even reach the opposite mesenteric
border and continue its propagation along the other
side, as shown by the occurrence of several spike
“wraps” (Fig. 8A). Such a pattern of propagation could
not occur with longitudinal spikes.
In relation to the preceding slow wave, the sequence
of both types of spike patches is also very different.
With the longitudinal spike patches, these may occur
anywhere along the intestine’s circumference and
AJP-Gastrointest Liver Physiol • VOL
within a relatively wide excitable area following the
leading edge of the slow wave. Because of this, sequential spike patches may jump from one location to another, as shown in Fig. 10. In contrast, in the distal
small intestine, there is a much tighter relationship
between the propagating slow wave and the sequence
of circumferential spike patches. This relationship,
however, does not extend to the time domain, and, as
shown in Fig. 10, bottom, the initiation of spikes may
occur in a relative broad time period of 3 s after the
slow wave in both the upper and the lower part of the
intestines. This lack of difference in the timing of the
spikes is because of the fact that, although the excitable area available for spikes in the ileum is much
more limited, the velocity of propagation of the slow
wave is also severely reduced (1). In other words, the
285 • NOVEMBER 2003 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.3 on June 17, 2017
Fig. 10. Spatial and temporal analysis
of the sites of initiation of longitudinal
and circumferential spikes in 3 dogs, as
analyzed in the duodenum (left) and in
the distal ileum (right). In A, the sequential locations of the origin of spikes
are indicated by filled circles with the
lines and arrows connecting the circles
showing the temporal sequence of each
spike patch. The locations of the first (1)
and the last site of spike origins (11, 17,
and 9 in the duodenum and 41, 28, and
25 in the distal ileum) are also indicated. The direction of the slow waves
was in all cases from oral to aboral. In
the duodenum, successive spikes occurred at relatively large distances from
each other and were distributed all
along the intestinal circumference. In
contrast, in the distal ileum, most of the
spike origins were located close to the
antimesenteric region, and the distances between successive origins were
much shorter than in the duodenum. In
B, top, the distances between successive
spike origins were plotted for the duodenum and for the distal ileum. In the
duodenum, the distance graph shows
large variations between successive
spike origins, whereas in the distal ileum this was reduced to much lower
values. In B, bottom, histograms of the
timing of each spike origin in relation to
the timing of the local slow wave are
plotted. Spikes originated both in the
duodenum and in the distal ileum during a period of ⬃3 s after the slow wave.
G1026
SPIKE PATCHES IN THE CANINE SMALL INTESTINE
AJP-Gastrointest Liver Physiol • VOL
reported here in anesthetized dogs (28). Furthermore,
spike patches similar to those reported here have been
shown to occur in tissues in vitro (11). Nevertheless,
some modulating effect of the used opioids cannot be
completely ruled out.
Nevertheless, within these limitations, the results
are clear. Spikes and spike patches occur throughout
the canine small intestine. Significant topographical
variations in morphology and origin of spikes were
demonstrated based on the presence of two types of
spikes, possibly propagating separately in each of the
two muscle layers. These facts seem to suggest that the
function of spikes in the generation of patterns of
motility in various parts of the small intestine are
different, which may induce different patterns of motility and underlie different functions along the length
of this organ.
We thank all those involved in Beerse and in Al Ain for contributions to this work and especially Jan Eelen, Patrick Van Bergen,
Dirk Smets, Diane Verkuringen, Kholoud Arafat, and S. Dhanasekaran. In addition, we thank Prof. J. R. Slack for editorial help.
DISCLOSURES
This work was supported by Johnson and Johnson Pharmaceutical Research and Development, a Division of Janssen Pharmaceutica
NV, Beerse, Belgium.
REFERENCES
1. Armstrong HIO, Milton GW, and Smith AWM. Electropotential changes of the small intestine. J Physiol 131: 147–153, 1956.
2. Bass P, Code CF, and Lambert EH. Motor and electric activity of the duodenum. Am J Physiol 201: 287–291, 1961.
3. Berkson J, Baldes EJ, and Alvarez WC. Electromyographic
studies of the gastrointestinal tract. I. The correlation between
mechanical movement and changes in electrical potential during
rhythmic contraction of the intestine. Am J Physiol 102: 683–
692, 1932.
4. Bortoff A. Intestinal slow-wave propogation velocity as a function of longitudinal muscle impedance. In: Physiology of Smooth
Muscle, edited by Bulbring E and Shuba MF. New York: Raven,
1976, p. 111–119.
5. Bortoff A and Ghalib E. Temporal relationship between electrical and mechanical activity of longitudinal and circular muscle during intestinal peristalsis. Am J Dig Dis 17: 317–325, 1972.
6. Bülbring E, Burnstock G, and Holman ME. Excitation and
conduction in the smooth muscle of the isolated taenia coli of the
guinea-pig. J Physiol 142: 420–437, 1958.
7. Daniel EE, Carlow DR, Wachter RT, Sutherland WH, and
Bogoch A. Electrical activity of the small intestine. Gastroenterology 37: 268–281, 1959.
8. Fioramonti J, Fargeas MJ, and Bueno L. Stimulation of
gastrointestinal motility by loperamide in dogs. Dig Dis Sci 32:
641–646, 1987.
9. Fox JE and Daniel EE. Exogenous opiates: their local mechanisms of action in the canine small intestine and stomach. Am J
Physiol Gastrointest Liver Physiol 253: G179–G188, 1987.
10. Konturek SJ, Thor P, Krol R, Dembinski A, and Schally
AV. Influence of methionine-enkephalin and morphine on myoelectric activity of small bowel. Am J Physiol Gastrointest Liver
Physiol 238: G384–G389, 1980.
11. Lammers WJEP. Propagation of individual spikes as “patches”
of activation in the isolated feline duodenum. Am J Physiol
Gastrointest Liver Physiol 278: G297–G307, 2000.
12. Lammers WJEP and Slack JR. Of slow waves and spike
patches. News Physiol Sci 16: 138–144, 2001.
13. Lammers WJEP, Stephen B, Arafat K, and Manefield GW.
High resolution mapping in the gastrointestinal system: initial
results. Neurogastroenterol Motil 8: 207–216, 1996.
285 • NOVEMBER 2003 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.3 on June 17, 2017
narrow ring of excitability propagates much more
slowly in the distal ileum, offering more time for spike
initiation, thereby counterbalancing the smaller area
available for initiation.
The fact that circumferential spikes originate along
the antimesenteric border may have consequences for
the motility in the jejunum and the ileum, similar to
the situation with peristaltic contraction (21, 22). Because these origins are clustered along the antimesenteric line, and as the zone of excitability propagates as
a ring down the intestinal tube, propagation of individual spikes may occur predominantly in the circular
direction. As discussed previously, the spike may propagate in two wavelets from its point of origin, activating simultaneously both sides of the tubes for varying
distances and “straddling” both sides of the intestinal
tube. In addition, because the zone of excitability is
relatively narrow, especially in the ileum, these saddle
patches must follow each other very closely behind the
front of the slow wave (Fig. 7D). All this could impose
a type of “ripple” or squeezing activity on the tube and
possibly its contents. The fact that most of these contractions occur at the opposite side of the tube from the
anchoring effects of the mesenteric attachments may
also be helpful in this regard.
In which muscle layers do these two types of spikes
propagate? In the isolated guinea pig ileum, Stevens et
al. (24) showed that Ca2⫹ waves, which are induced by
spikes, could occur in both layers. In the current study,
longitudinal spikes seem to propagate in the longitudinal muscle layer. This is indicated by spike propagation occurring predominantly in the longitudinal direction. Also, the longitudinal spike velocity (6.9 ⫾ 1.7
cm/s) is comparable with that measured in the cat
(7.7 ⫾ 4.5 cm/s), and its anisotropic pattern of propagation is very similar to that in the duodenum of the
cat (14). Similar arguments seem to indicate that circumferential spikes would propagate in the circular
muscle layer. The predominant direction of propagation is in the circular direction, whereas oral or aboral
propagation of circumferential spikes in the longitudinal direction, from its origin along the antimesenteric
border, was never seen. The velocity of propagation of
circumferential spikes is faster than that of longitudinal spikes, the reasons for which are not clear. It could
be because of the larger mass of available circular
tissue, because of a possible longer length of the individual circular muscle cells, because of better connections between cells, or because of other physical or
electrical properties of the muscle layer.
The limitations of this study must be quite clear.
Although the small intestines were studied in situ, the
animals were anesthetized, and the intestines were
empty. Some published data show that the intestinal
spike activity (10, 20) or phasic activity (8, 9) could be
enhanced in dogs exposed to opioids, such as morphine
or loperamide. Use of opioid anesthesia could have
therefore increased the incidence of spike activity during the course of the experiment. However, preliminary
observations in awake dogs implanted with electrode
arrays indicate similar patterns of spike activity, as
SPIKE PATCHES IN THE CANINE SMALL INTESTINE
AJP-Gastrointest Liver Physiol • VOL
22. Schulze-Delrieu K, Brown B, Herman B, Brown CK, Lawrence D, Shirazi S, Palmieri T, and Raab J. Preservation of
peristaltic reflex in hypertrophied ileum of guinea pig. Am J
Physiol Gastrointest Liver Physiol 269: G49–G59, 1995.
23. Specht PC. Propagation of stimulated slow waves in cat intestinal muscle. Am J Physiol 231: 228–233, 1976.
24. Stevens RJ, Publicover NG, and Smith TK. Propagation and
neural regulation of calcium waves in longitudinal and circular
muscle layers of guinea pig small intestine. Gastroenterology
118: 892–904, 2000.
25. Summers RW and Dusdieker NS. Patterns of spike burst
spread and flow in the canine small intestine. Gastroenterology
81: 742–750, 1981.
26. Tamai T and Prosser CL. Differentiation of slow potentials
and spikes in longitudinal muscle of cat intestine. Am J Physiol
210: 452–458, 1966.
27. Van de Water A, D’Aubioul J, Van Gerven W, Van Ammel
K, and De Clerck F. Selective vasoconstriction by alniditan in
the carotid vascular bed of anaesthetized dogs. Eur J Pharmacol
299: 127–137, 1996.
28. Ver Donck L, Lammers WJEP, and Schuurkes JAJ. A
model to record slow wave activity in the duodenum of the
conscious dog (Abstract). Gastroenterology 124: A164, 2003.
285 • NOVEMBER 2003 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.3 on June 17, 2017
14. Lammers WJEP, Stephen B, Slack JR, and Dhanasekaran
S. Anisotropic propagation in the small intestine. Neurogastroenterol Motil 14: 357–364, 2002.
15. McCoy EJ and Baker RD. Intestinal slow waves: decrease in
propagation velocity along upper small intestine. Am J Dig Dis
14: 9–13, 1969.
16. Mendel C, Pousse A, and Grenier JF. Relationship of electrical slow wave and spike bursts in the dog jejunum in vivo. Can
J Physiol Pharmacol 62: 1315–1319, 1984.
17. Nagai T and Prosser CL. Patterns of conduction in smooth
muscle. Am J Physiol 204: 910–914, 1963.
18. Sancholuz AG, Croley II TE, Christensen J, Macagno EO,
and Glover JR. Phase lock of electrical slow waves and spike
burst in cat duodenum. Am J Physiol Endocrinol Metab Gastrointest Liver Physiol 229: E608–E612, 1975.
19. Sancholuz AG, Croley II TE, Glover JR, Macagno EO, and
Christensen J. Distribution of spike bursts in cat duodenum.
Am J Physiol 229: 925–929, 1975.
20. Sarna SK and Otterson MF. Small intestinal amyogenesia
and dysmyogenesia induced by morphine and loperamide. Am J
Physiol Gastrointest Liver Physiol 258: G282–G289, 1990.
21. Schulze-Delrieu K. Visual parameters define the phase and
the load of contractions in isolated guinea pig ileum. Am J
Physiol Gastrointest Liver Physiol 276: G1417–G1424, 1999.
G1027