- Email: [email protected]
Gastroduodenal regulation of common duct bile flow in the dog
GASTROENTEROLOGY
1968;94:755-81
Gastroduodenal Regulation of Common Duct Bile Flow in the Dog NOBUO MOCHINAGA, SUSHIL K. SARNA, WYLIE J. DODDS, and TEIJI MATSUMOTO
ROBERT E. CONDON,
Departments of Surgery, Physiology. and Radiology, Medical College of Wisconsin Veterans Administration Medical Center. Milwaukee. Wisconsin
We investigated the relationship between the entry of individual drops of bile into the duodenum and gastroduodenal motor activity in the fasted state in 10 conscious dogs. The common bile duct was transected and a catheter was inserted through each end. The exteriorized catheters were connected to a photometric drop-flow meter. During phase III activity, bile entered the duodenum in single drops, only in between two consecutive contractions, or as a series of drops during transient inhibition of duodenal contractions by antral phase III contractions. During phase II activity bile also entered the duodenum, usually in between contractions or when the duodenum was intermittently quiescent. Bile entered the duodenum during a duodenal contraction only when the contraction amplitude was ~15% -C1% (mean -C SE) of the maximal amplitude during phase III contractions. Bile flow into the duodenum showed a cyclic pattern with a peak during late duodenal phase II activity and a trough during duodenal phase I activity only when phase III activity originated in the duodenum and migrated caudad. There was no cyclic pattern of bile flow when phase III activity originated in the proximal jejunum and migrated caudad. The total volume of bile flow in a migrating motor complex cycle and bile flow rate were greater when phase III activity started in the proximal jejunum than when it started in the duodenum. We conclude that gastroduodenal contractions play an important role in the regulation of bile flow into the duodenum. The cyclic pattern of bile flow is altered when phase III activity starts ectopically in the jejunum.
I
n species such as humans and dogs, the common bile duct descends behind the first portion of the pancreas, enters the dorsal or mesenteric wall of the duodenum, passes obliquely through it, and terminates at the choledochoduodenal junction (1,2). The intramural length of this duct in the dog is 1.5-Z cm.
and Zablocki
The outer layer of the intramural portion of the bile duct is formed by the tunica muscularis of the duodenum. The inner layer, or muscularis proprius, begins in the infundibular portion of the bile duct and extends in the submucosa to the termination of the duct as the sole investing muscle (1).This anatomic structure and a substantial length of the common bile duct and all of the sphincter of Oddi within the duodenal wall suggest that contraction of the duodenal wall would occlude or constrict both the intramural duct and the sphincter of Oddi. Thus, rhythmic duodenal contractions are likely to affect bile flow into the duodenum. The duodenal contractions at the level of the choledochoduodenal junction are, in turn, coordinated with antral contractions (3,~). However, the relationship between antroduodenal contractions and the entry of bile flow into the duodenum has not been examined. Our objectives in this study were (a) to define the patterns of bile entry into the duodenum by the use of a photometric drop-flow meter (in contrast to isotope methods that determine the mean or total bile flow volume in a given period, the drop-flow meter measures the instantaneous entry of bile into the distal common bile duct and the duodenum]; (b) to determine the relationship between bile entry into the duodenum and antroduodenal contractions; and (c) to evaluate whether the ectopic generation of phase III activity in the jejunum has an effect on bile flow into the duodenum.
Materials
and Methods
The experiments were done on 10 healthy conscious dogs, each weighing 17-35 kg. Each animal was anesthetized with pentobarhital sodium and implanted
Abbreviation used in this paper: MMC. complex. 0 1988 by the American Gastroenterological 0016-5085188/$X50
migrating Association
motor
GASTROENTEROLOGYVol.94,No.3
756 MOCHINAGAETAL.
with a set of three strain gauge transducers, two on the antrum and one on the duodenum, and eight bipolar electrodes on the duodenum, jejunum, and ileum. The strain gauge transducer on the duodenum was at the level of the choledochoduodenal junction, about 4 cm distal to the pylorus. The first electrode was at the level of the choledochoduodenal junction and the remaining electrodes were at lo%, 30%, 50%, SO%, 70%, 80%, and 90% of the total length of the small intestine. The strain gauge transducer axis was oriented along the circular muscle axis in the stomach and the duodenum. The lead wires were brought out through a stainless steel cannula as described previously (5). The common bile duct was carefully dissected to prevent damage to the surrounding tissue and then transected 2-3 cm proximal to the choledochoduodenal junction. A silastic catheter (ID 2.6mm, OD 4.9 mm) fitted with an adapter was inserted through each end of the common bile duct and secured. The silastic catheters were tied together, side to side, to stretch the common bile duct to its original length. The catheters were brought out subcutaneously at the subscapular region. The exteriorized catheters were connected to each other for normal bile flow and stored in a jacket that the dog wore. The flow resistance of the silastic catheters and the photometric drop-flow meter (model PTTLS; Grass Instruments Co., Quincy, Mass.) measured with an Arndorfer pump (Arndorfer Medical Specialists, Inc., Greendale, Wise.) before implantation was in the range of 2-3 mmHzO. The recordings were begun lo-14 days after surgery. The dogs were fasted overnight and stood in a modified Pavlov sling. The exteriorized catheters were connected to a photometric drop-flow meter. In separate experiments, the drop-flow meter was calibrated by counting the number of bile drops per milliliter. The mean number of bile drops per milliliter was 27. Bile flow rate was determined by dividing the total volume of bile flow during a specific phase of a migrating motor complex (MMC) cycle or during the entire MMC cycle by the duration of that phase or the period of the MMC cycle, respectively. The four phases of the duodenal MMC cycle were defined in the usual manner (6,7). Phase 1 had motor quiescence except for isolated contractions. Phase 11had intermittent contractions. Phase Ill had regular contractions at their maximal frequency except when they were interrupted concurrent with a single antral contraction or a cluster of antral contractions. Phase IV also exhibited intermittent contractions but was of much shorter duration than phase II activity. A MMC cycle was considered to start at the end of phase Ill activity. The recordings were made on a 12-channel Grass recorder, model 7D. The lower and upper cutoff frequencies were set at 0.04and 15 Hz, respectively, for myoelectric signals, and direct current and 15 Hz, respectively, for strain gauge transducers. The drop counter was connected to the bottom recording channel. Eight channels of data, always including the drop counter signal, were recorded on a Hewlett-Packard FM instrumentation tape recorder (model 3968A; Hewlett-Packard Co., Palo Alto, Calif.) for later electronic filtering and playing back at a different
speed. The data were analyzed by analysis of variance and unpaired t-tests. The mean values were determined for each dog, from which the overall mean value and standard error were computed. The n value in each case represents the number of dogs. A probability of 50.05was considered to be statistically significant.
Results All dogs exhibited MMCs in the small intestine and all but 1 exhibited cyclic motor activity in the stomach (7). However, the point of origin of phase III activity varied in different experiments. Because our aim was to correlate gastroduodenal motor activity with the pattern of bile flow, the MMCs were divided into three groups, depending on the site of origin of phase III activity. In group 1 MMCs, phase III activity occurred first in the antrum and the duodenum and migrated caudad from the duodenum to the terminal ileum. The duodenal MMC period in this group was 92 2 4 min (mean k SE). The antral, duodenal, and jejunal phase III activity durations were 23 2 2, 14 +- 1.3, and 8 + 0.7 min,
respectively.
In group 2 MMCs, phase III activity originated in the jejunum and migrated caudad to the terminal ileum. The jejunal MMC period, 74 ? 9 min, and phase III duration, 6 t 0.4 min, in this group were significantly shorter than those in group 1 (p < 0.05). When phase III activity started in the jejunum, the motor activity in the stomach generally continued in phase II pattern, but a clear quiescent state occurred in the duodenum. This quiescent state was called phase I if its duration fell within the mean ? 2 SD of phase I duration in group 1 experiments.
In group 3, phase III activity also originated in the jejunum, but it did not produce any phase I in the duodenum or the stomach as defined above. The jejunal MMC period, 76 * 6 min, and phase III duration, 6 + 0.5 min, in this group were significantly shorter than those in group 1 (p < 0.002), but not those in group 2 (p > 0.05). A MMC cycle was classified into one of the above groups only if the above characteristics were met for two consecutive phase III activities.
Relationship Between Bile Flow and Migrating Motor Complex Cycle Total bile flow volumes and bile flow rates. The total bile flow volume during a MMC cycle in group 1, 15.9 2 3.0 ml, was significantly less than that in group 2, 30.8 t 8.3 ml (p < 0.05), or group 3, 32.2 +- 8.9 ml (p < 0.05) (Figure 1). The bile flow rate during the entire MMC cycle in group 1, 0.18 5 0.04 mlimin, was also significantly less than that in group
March 1988
DUODENAL REGULATION OF BILE FLOW
-
0 Cl 62 (N=S)(N=5)(1=7) *P
-
61
63
G2 G3
(N=9)(N=S)(N=7)
w.r.t.GROUP
*P-CO.005
1
WP
Figure
w.r.t. GROUP
1
w.r.t, GROUP
1
1. The total bile flow volumes and bile flow rates were significantly greater in experiments in groups z and 3 than in group 1.
2, 0.39 ? 0.06 ml/min (p < 0.005), or group 3, 0.41 * ml/min (p < 0.025) [Figure 1). There was no significant difference between groups 2 and 3 for bile flow rate and bile flow volume. Bile flow during different phases of the migrating motor complex cycle. The volume and rate 0.09
of bile flow during the four phases of the MMC cycle for all three groups are shown in Figure 2. Group 1 had a distinct cyclic pattern of bile flow. Both the bile volume and flow rate during phase II activity were significantly greater than those during phases I, III, and IV (p < 0.025). In groups 2 and 3, the bile volume and bile flow rate during phase I activity were not significantly different from those during phase II activity (p > 0.05). Also, the bile flow volume and flow rate during phase I activity in groups 2 and 3 were significantly greater than those during phase I activity in group 1 (p < 0.025). There was no significant difference between bile flow volumes during other phases of the MMC cycle in groups 1, 2, and 3 (p > 0.05). The bile flow rates during phases I, III, and IV in groups 2 and 3 were significantly greater than those in group 1 (p < 0.025; Figure 2). Bile flow patterns during a migrating motor complex cycle. The total volume of bile flow in 10 percentiles of the MMC cycle in the three groups is shown in Figure 3. A distinct peak in bile volume occurred during late phase II activity of the MMC cycle in group 1. The bile volume entering the duodenum was minimal from 0% to 40% of the MMC cycle and peaked between 80% and 90% of the MMC cycle. The bile flow volume during duodenal phase III activity ~90°~-1000~ of the MMC cycle) was 0.8
w
4 Lo
0’ -u
1
nm
A
n I
GROUP1
GROUP2
GROUP3
(leg)
(w=5)
(N=7)
+P<
0.2
0
urn
0.005wt.GROUP1
4W~
AL w
I
urn
n
I
urn
1 run
GROUP1
GROUP2
GROUP3
(n=9)
(n=5)
(n=7)
tp<0.05 HP<
wt.GtlUUP1
0.025w.ct GROUP 1
x p
2.
Bile
flow volumes
757
and bile flow rates during the four phases of the MMC cycle in the three groups dafimd in the text.
758
MOCHINAGA
GASTROENTEROLOGY
ET AL.
z Du -r
St
r
,
Jt!--r -
I-
I
-
-
-
=
-
-t-
Vol. 94, No. 3
= -
r -
I
+
:
T
% OF MMC
CYCLE
GROUP 2
GROUP1
GROUP 3
(H=S)
(n=9)
(N=7)
mp>o.os
*p<0.005
x p >0.05
The bile flow pattern was cyclic in group 1 with a peak during late phase II activity and a trough during phase I activity. The bile flow fluctuated randomly when phase III activity started ectopically in groups 2 and 3. The migrating motor complex cycle was considered to start and end at the end of phase III activity. Du, duodenum; Je, jejunum; St, stomach.
St
st
-DlJ Je
Du Je
(4 GROUP
W 1
GROUP 1
st
St
Du Je
DU
Je
(Cl GROUP 2
6’) GROUP
3
Representative patterns of bile flow in individual MMC cycles in experiments in groups 1, 2, and 3. Patterns A and U are from group 1,whereas C and Dare from groups 2 and 3, respectively. In most experiments in group 1, bile flow peaked during phase II activity in the duodenum. Bile flow volume fluctuated randomly during the MMC cycle in groups 2 and 3. Du, duodenum; Je, jejunum; St, stomach.
March
1988
significantly less (p < 0.025) than that during late phase II activity, but was significantly greater than that during phase I activity (p < 0.025). Although the mean bile flow column peaked during late phase II activity in group 1 (Figure 3), the pattern of bile flow varied during different MMC cycles in the same dog and in different dogs. Figures 4A and 4B show two typical patterns of bile flow in group 1 MMC cycles. The bile flow generally began to increase with the onset of phase II activity. Multiple peaks of bile flow were usually observed when phase II duration was long. In contrast to group 1, there was no cyclic pattern of bile flow into the duodenum in groups 2 and 3 (Figure 3). The bile flow volume fluctuated randomly throughout the MMC cycle. The maximum bile flow volume was not significantly different than the minimum bile flow volume (p > 0.05), indicating that there was no distinct peak in bile flow. A similar random fluctuation of bile flow volume was observed during individual MMC cycles (Figures 4C and 4D). Relationship Between Bile Flow and Antroduodenal Contractions The entry of bile into the duodenum was closely related to antroduodenal contractions. During phase III activity, bile entered the duodenum in single isolated drops or as a rapid burst of several drops (Figure 5). The single drops entered the duodenum only between two consecutive contractions when there was minimal tension in the duodenal wall at the level of the choledochoduodenal junction (Figure 5). Rapid bursts of drops entered the duodenum during phase III activity only when the duodenal contractions were transiently inhibited (Figure 51. During phase II activity, bile entered the duodenum as single drops in between two consecutive contractions or as a series of drops during prolonged quiescence (Figure 6). In an analysis of 1673 duode-
Relationship between bile flow and phase III contractions. Single drops of bile entered the duodenum in between two successive duodenal contractions. A burst of bile flow occurred when the duodenum was transiently inhibited in concert with phase III contractions in the antrum.
DUODENAL
REGULATION
OF BILE FLOW
I
SG A-4
759
Imv
I
SG D-5 ~ DROPCOUNTER I
Figure
~+.V? 4
1-
*i
ILaiiiicilidi+,,
4,
&
4
,L4
*I,
-i
,
ire 4
6. Relationship between bile flow and phase II contractions Bile entry into the duodenum occurred as a series of drops during intermittent quiescence of the duodenum, and as single drops in between two successive contractions.
nal contractions (1191 drops of bile flow), there were only 69 instances (77 drops] when a drop or drops of bile entered the duodenum while the duodenal wall at the level of the choledochoduodenal junction was in the contracted state. The amplitude of these duodenal contractions was 15.2% t 1.4% of the maximal amplitude of contractions during phase III activity. When bile flow occurred during phase I activity, it usually occurred as a steady series of drops.
Discussion Our findings show that in the fasted state gastroduodenal motor activity plays a major role in the regulation of bile flow into the duodenum. Bile entered the duodenum only when the duodenal wall at the level of the choledochoduodenal junction was not contracting or the amplitude of contraction was small compared with the maximal amplitude of contractions during phase III activity. During phase II activity, bile usually entered the duodenum during intermittent quiescent periods and during phase III activity when the duodenal contractions were transiently inhibited in coordination with a cluster of antral contractions. When bile entered the duodenum in between two consecutive duodenal contractions during phase II or phase III activity, it usually entered as a single drop whose timing coincided with the time of little or no tension in the duodenal wall, These findings suggest that when the duodenal wall contracts, it constricts the common bile duct and the choledochoduodenal junction to impede bile flow. The hepatic secretory pressure and intraductal pressures generated by gallbladder contractions are able to overcome duodenal resistance to outflow only when duodenum is quiet or its contractions are feeble. Our findings are in agreement with the radiographic or manometric observations of others (8,9) who reported that bile flow ceases while the duodenum contracts. Ono et al. (10) reported that bile flow into the
760
MOCHINAGAETAL.
duodenum of humans was independent of duodenal contractions. His conclusions were based on concurrent myoelectric recordings from the sphincter of Oddi and duodenum by electrodes implanted at surgery for gallstone disease. The electrodes to record sphincter of Oddi myoelectric activity were inserted blindly through the duodenal muscle layers. Ono et al. (10) did not include any tracings of myoelectric activity in their paper, nor did they provide evidence that the myoelectric activity they recorded from the sphincter of Oddi muscle was different from that of the adjacent duodenal muscle. The thin sphincter of Oddi muscle layer is anatomically and embryologically distinct, but it is closely integrated with the duodenal muscle. Under these circumstances it would be difficult to record selectively from the sphincter of Oddi muscle by extracellular electrodes. Bile flow into the duodenum was related not only to the absence of duodenal contractions but also to the presence of antral contractions. Matsumoto et al. (11) reported that the antral contractions were closely coupled with gallbladder contractions. Thus bile may enter the duodenum when the propagating gastric contraction arrives in the antrum, the gallbladder contracts, and the duodenum is quiescent. This three-way coordination, wherein the antrum and gallbladder contract and the duodenum relaxes, results in a gastroduodenal-gallbladder-biliary tract triangle that may coordinate the concurrent entry of gastric contents and bile into the duodenum and thereby ensure instant mixing. Our findings show that the cyclic pattern of bile flow in the fasted state previously reported by others (12-14) is present only when phase III activity starts in the proximal duodenum at the level of the choledochoduodenal junction. When phase III activity started ectopically in the jejunum, bile flow fluctuated randomly during the MMC cycle. In such cases, there was no significant or consistent peak in bile flow volume during the MMC cycle. This random pattern of bile flow did not depend on whether phase I activity occurred in the duodenum (group 2) or not (group 3). Interestingly, the volume and rate of bile flow during a MMC cycle were greater when phase III activity started ectopically in the jejunum than when it started in the proximal duodenum. This increase occurred in spite of the fact that the MMC cycle length was shorter when phase III activity started ectopically in the jejunum. The increase in total bile flow was primarily due to an increase in bile flow during phases I and IV of the MMC cycle. One of the reasons for this increase may be that when phase III activity occurs in the proximal duodenum, the forceful persistent contractions may provide a high resis-
GASTROENTEROLOGYVol.94,No.3
tance to bile outflow during this period, thereby diverting the hepatic bile to the gallbladder to be concentrated. When phase III activity does not occur in the duodenum, unconcentrated bile from the liver may continue to flow into the duodenum during the entire MMC cycle without filling the gallbladder. This hypothesis remains to be confirmed by direct measurement of gallbladder filling during normal and ectopic MMC cycles. The use of a drop counter allowed us to correlate the relationship between instantaneous bile entry into the duodenum and duodenal contractions. In contrast, the radioisotope method measures the total bile flow in a given sampling period. Because the internal diameter of the exteriorized catheters was made about the same as that of the common bile duct, the resistance introduced by the catheter assembly was negligible. A catheter inserted into the sphincter of Oddi and common bile duct under direct vision through a duodenal cannula can also be used to measure bile flow volume in drops (15). However, the insertion of a duodenal cannula may modify duodenal contractile activity at the level of the sphincter of Oddi. Furthermore, the insertion of a catheter through the sphincter of Oddi may impair the ability of the sphincter and the duodenal wall to constrict it. So far four factors that contribute to the cycling of bile flow in the fasted state have been identified: (a) migrating motor complexes that transport the bile from the duodenum to the ileum for reabsorption, (b) reabsorption of bile acids in the ileum and their return to the liver for recirculation, (c) gallbladder motor activity that partially empties the gallbladder and (d) resisduring each MMC cycle (11,16,17), tance of the sphincter of Oddi that may impede bile flow (18-21) or propagating contractions in some species (like the opossum), which may propel bile (22). The roles of MMCs and ileal reabsorption are well understood, but those of the gallbladder and the choledochoduodenal junction are understood incompletely. Scott et al. (20) recorded sphincter of Oddi contractions from the dog by manometric methods and concluded that these contractions regulate bile flow during the fasted state. The frequency of contractions recorded from the sphincter of Oddi by them had the same frequency as that of duodenal electrical control activity. Furthermore, there was a 1:1 relationship between sphincter of Oddi contractions and bursts of duodenal electrical response activity representing duodenal contractions. Because of the close integration of the sphincter of Oddi and duodenal muscle layers in the dog, it is possible that the duodenal electrical control activity may drive the sphincter of Oddi electrical control activity so that the two may contract in phase. It is
March
1988
also possible, however, that the canine sphincter of Oddi may not have independent phasic contractions and the intraluminal contractions recorded manometrically from the intramural sphincter of Oddi and common bile duct may be due to duodenal muscle contractions. Because the distal common bile duct transverses the duodenal wall in the caudad direction, the caudad propagation of duodenal contractions at the level of the choledochoduodenal junction would be recorded as caudad propagating contractions in the intramural common bile duct by the manometric method. Regardless of whether the canine sphincter of Oddi has independent contractions or not, our findings indicate that duodenal contractile activity is an important factor in the regulation of intraduodenal bile flow, in addition to the four factors cited above. In summary, bile flow into the duodenum depends on the balance between the driving forces and the resisting forces. The driving forces for bile flow are hepatic secretory pressure, which is dictated by the enterohepatic circulation of bile acids, and gallbladder motor activity. The common bile duct in most species acts as a simple conduit. In some species, like the opossum, the propagating contractions in the extraduodenal sphincter of Oddi may also provide an additional driving force. The resistive forces that deter bile flow into the duodenum are the passive resistance of the sphincter of Oddi and duodenal contractile activity.
DUODENAL
8. 9. 10. 11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
2. 3.
4.
5.
6
HE, Christensen GC. Miller’s anatomy of the dog. 2nd ed. Philadelphia: WB Saunders, 1979. Boyden EA. The anatomy of the choledochoduodenal junction in man. Surg Gynecol Obstet 1957;104:641-52. Sarna SK, Kitai R, Muniappan K, Marzio L, Daniel EE, Waterfall WE. Gastroduodenal coordination: a computer analysis (abstr). In: Duthie HL, ed. Gastrointestinal motility in health and disease. Lancaster: MTP Press, 1978:271-2. Allen GL, Poole EW, Code CF. Relationships between electrical activities of antrum and duodenum. Am J Physiol 1964; 207:906-10. Sarna S, Northcott P, Belbeck L. Mechanisms of cycling of migrating myoelectric complexes-the effect of morphine. Am J Physiol 1982;242:G588-95. Code CF, Marlett JA. The interdigestive myoelectric complex of the stomach and small bowel of dogs. J Physiol (Lond) 1975:246:289-309.
OF BILE FLOW
761
7. Sarna
References 1. Evans
REGULATION
21.
22.
SK. Cyclic motor activity; migrating motor complex: 1985. Gastroenterology 1985;89:894-913. Potter JC, Mann FC. Pressure changes in the biliary tract. Am J Med Sci 1926:171:202-17. Nebesar RA, Pollard JJ, Potsaid MS. Cine cholangiography: some physiologic observations. Radiology 1966;86:475-9. Ono K, Watanabe N. Suzuki K, Tsuchida H, Sugiyama Y, Abo M. Bile flow mechanism in man. Arch Surg 1968;96:869-74. Matsumoto T, Sarna SK, Condon RE. Dodds WJ. Mochinaga N. Canine gallbladder cyclic motor activity. Am J Physiol 1987 (in press). Peeters TL, Vantrappen G, Janssens J, Bile acid output and the interdigestive migrating motor complex in normals and in cholecystectomy patients. Gastroenterology 1980;79:678-81. Keane FB, DiMagno EP, Dozois RR, Go VLW. Relationships among canine interdigestive exocrine pancreatic and biliary flow, duodenal motor activity, plasma pancreatic polypeptide, and motilin. Gastroenterology 1980;78:310-6. Scott RB, Eidt PB, Shaffer EA. Regulation of fasting canine duodenal bile acid delivery by sphincter of Oddi and gallbladder. Am J Physiol 1985;249:G622-33. Scott RB, Strasberg SM, El-Sharkawy TY. Diamant NE. Regulation of the fasting enteroheptic circulation of bile acids by the migrating myoelectric complex in dogs. J Clin Invest 1983;71:644-54. Itoh Z, Takahashi I. Periodic contractions of the canine gallbladder during the interdigestive state. Am J Physiol 1981;240:G183-9. Traynor OJ. Dozois RR, DiMagno EP. Canine interdigestive and postprandial gallbladder motility and emptying. Am J Physiol 1984;246:G426-32. Ryan JP. Motility of the gallbladder and biliary tree. In: LR Johnson, ed. Physiology of the gastrointestinal tract. New York: Raven, 1981:473-94. Becker JM. Duff WM, Moody FG. Myoelectric control of gastrointestinal and biliary motility: a review. Surgery 1981;89:466-77. Scott RB, Strasberg SM, El-Sharkawy TY, Diamant NE. Fasting canine biliary secretion and the sphincter of Oddi. Gastroenterology 1984:87:793-804. Toouli J, Geenen JE, Hogan WJ, Dodds WJ, Arndorfer KC. Sphincter of Oddi motor activity: a comparison between patients with common bile duct stones and controls. Gastroenterology 1982;82:111-7. Toouli J, Dodds WJ, Honda R. Sarna S, et al. Motor function of the opossum sphincter of Oddi. J Clin Invest 1983:71:208-20.
Received February 18. 1987. Accepted October 21, 1987. Address requests for reprints to: Sushi1 K. Sarna, Ph.D., Zablocki Veterans Administration Medical Center, Surgical Research 151. 5000 West National Avenue, Milwaukee, Wisconsin 53295. This work was supported in part by grant 7722-OlP from the Veterans Administration Research Service and DK32346 from the National Institutes of Diabetes and Digestive and Kidney Diseases to Dr. Sarna.
1968;94:755-81
Gastroduodenal Regulation of Common Duct Bile Flow in the Dog NOBUO MOCHINAGA, SUSHIL K. SARNA, WYLIE J. DODDS, and TEIJI MATSUMOTO
ROBERT E. CONDON,
Departments of Surgery, Physiology. and Radiology, Medical College of Wisconsin Veterans Administration Medical Center. Milwaukee. Wisconsin
We investigated the relationship between the entry of individual drops of bile into the duodenum and gastroduodenal motor activity in the fasted state in 10 conscious dogs. The common bile duct was transected and a catheter was inserted through each end. The exteriorized catheters were connected to a photometric drop-flow meter. During phase III activity, bile entered the duodenum in single drops, only in between two consecutive contractions, or as a series of drops during transient inhibition of duodenal contractions by antral phase III contractions. During phase II activity bile also entered the duodenum, usually in between contractions or when the duodenum was intermittently quiescent. Bile entered the duodenum during a duodenal contraction only when the contraction amplitude was ~15% -C1% (mean -C SE) of the maximal amplitude during phase III contractions. Bile flow into the duodenum showed a cyclic pattern with a peak during late duodenal phase II activity and a trough during duodenal phase I activity only when phase III activity originated in the duodenum and migrated caudad. There was no cyclic pattern of bile flow when phase III activity originated in the proximal jejunum and migrated caudad. The total volume of bile flow in a migrating motor complex cycle and bile flow rate were greater when phase III activity started in the proximal jejunum than when it started in the duodenum. We conclude that gastroduodenal contractions play an important role in the regulation of bile flow into the duodenum. The cyclic pattern of bile flow is altered when phase III activity starts ectopically in the jejunum.
I
n species such as humans and dogs, the common bile duct descends behind the first portion of the pancreas, enters the dorsal or mesenteric wall of the duodenum, passes obliquely through it, and terminates at the choledochoduodenal junction (1,2). The intramural length of this duct in the dog is 1.5-Z cm.
and Zablocki
The outer layer of the intramural portion of the bile duct is formed by the tunica muscularis of the duodenum. The inner layer, or muscularis proprius, begins in the infundibular portion of the bile duct and extends in the submucosa to the termination of the duct as the sole investing muscle (1).This anatomic structure and a substantial length of the common bile duct and all of the sphincter of Oddi within the duodenal wall suggest that contraction of the duodenal wall would occlude or constrict both the intramural duct and the sphincter of Oddi. Thus, rhythmic duodenal contractions are likely to affect bile flow into the duodenum. The duodenal contractions at the level of the choledochoduodenal junction are, in turn, coordinated with antral contractions (3,~). However, the relationship between antroduodenal contractions and the entry of bile flow into the duodenum has not been examined. Our objectives in this study were (a) to define the patterns of bile entry into the duodenum by the use of a photometric drop-flow meter (in contrast to isotope methods that determine the mean or total bile flow volume in a given period, the drop-flow meter measures the instantaneous entry of bile into the distal common bile duct and the duodenum]; (b) to determine the relationship between bile entry into the duodenum and antroduodenal contractions; and (c) to evaluate whether the ectopic generation of phase III activity in the jejunum has an effect on bile flow into the duodenum.
Materials
and Methods
The experiments were done on 10 healthy conscious dogs, each weighing 17-35 kg. Each animal was anesthetized with pentobarhital sodium and implanted
Abbreviation used in this paper: MMC. complex. 0 1988 by the American Gastroenterological 0016-5085188/$X50
migrating Association
motor
GASTROENTEROLOGYVol.94,No.3
756 MOCHINAGAETAL.
with a set of three strain gauge transducers, two on the antrum and one on the duodenum, and eight bipolar electrodes on the duodenum, jejunum, and ileum. The strain gauge transducer on the duodenum was at the level of the choledochoduodenal junction, about 4 cm distal to the pylorus. The first electrode was at the level of the choledochoduodenal junction and the remaining electrodes were at lo%, 30%, 50%, SO%, 70%, 80%, and 90% of the total length of the small intestine. The strain gauge transducer axis was oriented along the circular muscle axis in the stomach and the duodenum. The lead wires were brought out through a stainless steel cannula as described previously (5). The common bile duct was carefully dissected to prevent damage to the surrounding tissue and then transected 2-3 cm proximal to the choledochoduodenal junction. A silastic catheter (ID 2.6mm, OD 4.9 mm) fitted with an adapter was inserted through each end of the common bile duct and secured. The silastic catheters were tied together, side to side, to stretch the common bile duct to its original length. The catheters were brought out subcutaneously at the subscapular region. The exteriorized catheters were connected to each other for normal bile flow and stored in a jacket that the dog wore. The flow resistance of the silastic catheters and the photometric drop-flow meter (model PTTLS; Grass Instruments Co., Quincy, Mass.) measured with an Arndorfer pump (Arndorfer Medical Specialists, Inc., Greendale, Wise.) before implantation was in the range of 2-3 mmHzO. The recordings were begun lo-14 days after surgery. The dogs were fasted overnight and stood in a modified Pavlov sling. The exteriorized catheters were connected to a photometric drop-flow meter. In separate experiments, the drop-flow meter was calibrated by counting the number of bile drops per milliliter. The mean number of bile drops per milliliter was 27. Bile flow rate was determined by dividing the total volume of bile flow during a specific phase of a migrating motor complex (MMC) cycle or during the entire MMC cycle by the duration of that phase or the period of the MMC cycle, respectively. The four phases of the duodenal MMC cycle were defined in the usual manner (6,7). Phase 1 had motor quiescence except for isolated contractions. Phase 11had intermittent contractions. Phase Ill had regular contractions at their maximal frequency except when they were interrupted concurrent with a single antral contraction or a cluster of antral contractions. Phase IV also exhibited intermittent contractions but was of much shorter duration than phase II activity. A MMC cycle was considered to start at the end of phase Ill activity. The recordings were made on a 12-channel Grass recorder, model 7D. The lower and upper cutoff frequencies were set at 0.04and 15 Hz, respectively, for myoelectric signals, and direct current and 15 Hz, respectively, for strain gauge transducers. The drop counter was connected to the bottom recording channel. Eight channels of data, always including the drop counter signal, were recorded on a Hewlett-Packard FM instrumentation tape recorder (model 3968A; Hewlett-Packard Co., Palo Alto, Calif.) for later electronic filtering and playing back at a different
speed. The data were analyzed by analysis of variance and unpaired t-tests. The mean values were determined for each dog, from which the overall mean value and standard error were computed. The n value in each case represents the number of dogs. A probability of 50.05was considered to be statistically significant.
Results All dogs exhibited MMCs in the small intestine and all but 1 exhibited cyclic motor activity in the stomach (7). However, the point of origin of phase III activity varied in different experiments. Because our aim was to correlate gastroduodenal motor activity with the pattern of bile flow, the MMCs were divided into three groups, depending on the site of origin of phase III activity. In group 1 MMCs, phase III activity occurred first in the antrum and the duodenum and migrated caudad from the duodenum to the terminal ileum. The duodenal MMC period in this group was 92 2 4 min (mean k SE). The antral, duodenal, and jejunal phase III activity durations were 23 2 2, 14 +- 1.3, and 8 + 0.7 min,
respectively.
In group 2 MMCs, phase III activity originated in the jejunum and migrated caudad to the terminal ileum. The jejunal MMC period, 74 ? 9 min, and phase III duration, 6 t 0.4 min, in this group were significantly shorter than those in group 1 (p < 0.05). When phase III activity started in the jejunum, the motor activity in the stomach generally continued in phase II pattern, but a clear quiescent state occurred in the duodenum. This quiescent state was called phase I if its duration fell within the mean ? 2 SD of phase I duration in group 1 experiments.
In group 3, phase III activity also originated in the jejunum, but it did not produce any phase I in the duodenum or the stomach as defined above. The jejunal MMC period, 76 * 6 min, and phase III duration, 6 + 0.5 min, in this group were significantly shorter than those in group 1 (p < 0.002), but not those in group 2 (p > 0.05). A MMC cycle was classified into one of the above groups only if the above characteristics were met for two consecutive phase III activities.
Relationship Between Bile Flow and Migrating Motor Complex Cycle Total bile flow volumes and bile flow rates. The total bile flow volume during a MMC cycle in group 1, 15.9 2 3.0 ml, was significantly less than that in group 2, 30.8 t 8.3 ml (p < 0.05), or group 3, 32.2 +- 8.9 ml (p < 0.05) (Figure 1). The bile flow rate during the entire MMC cycle in group 1, 0.18 5 0.04 mlimin, was also significantly less than that in group
March 1988
DUODENAL REGULATION OF BILE FLOW
-
0 Cl 62 (N=S)(N=5)(1=7) *P
-
61
63
G2 G3
(N=9)(N=S)(N=7)
w.r.t.GROUP
*P-CO.005
1
WP
Figure
w.r.t. GROUP
1
w.r.t, GROUP
1
1. The total bile flow volumes and bile flow rates were significantly greater in experiments in groups z and 3 than in group 1.
2, 0.39 ? 0.06 ml/min (p < 0.005), or group 3, 0.41 * ml/min (p < 0.025) [Figure 1). There was no significant difference between groups 2 and 3 for bile flow rate and bile flow volume. Bile flow during different phases of the migrating motor complex cycle. The volume and rate 0.09
of bile flow during the four phases of the MMC cycle for all three groups are shown in Figure 2. Group 1 had a distinct cyclic pattern of bile flow. Both the bile volume and flow rate during phase II activity were significantly greater than those during phases I, III, and IV (p < 0.025). In groups 2 and 3, the bile volume and bile flow rate during phase I activity were not significantly different from those during phase II activity (p > 0.05). Also, the bile flow volume and flow rate during phase I activity in groups 2 and 3 were significantly greater than those during phase I activity in group 1 (p < 0.025). There was no significant difference between bile flow volumes during other phases of the MMC cycle in groups 1, 2, and 3 (p > 0.05). The bile flow rates during phases I, III, and IV in groups 2 and 3 were significantly greater than those in group 1 (p < 0.025; Figure 2). Bile flow patterns during a migrating motor complex cycle. The total volume of bile flow in 10 percentiles of the MMC cycle in the three groups is shown in Figure 3. A distinct peak in bile volume occurred during late phase II activity of the MMC cycle in group 1. The bile volume entering the duodenum was minimal from 0% to 40% of the MMC cycle and peaked between 80% and 90% of the MMC cycle. The bile flow volume during duodenal phase III activity ~90°~-1000~ of the MMC cycle) was 0.8
w
4 Lo
0’ -u
1
nm
A
n I
GROUP1
GROUP2
GROUP3
(leg)
(w=5)
(N=7)
+P<
0.2
0
urn
0.005wt.GROUP1
4W~
AL w
I
urn
n
I
urn
1 run
GROUP1
GROUP2
GROUP3
(n=9)
(n=5)
(n=7)
tp<0.05 HP<
wt.GtlUUP1
0.025w.ct GROUP 1
x p
2.
Bile
flow volumes
757
and bile flow rates during the four phases of the MMC cycle in the three groups dafimd in the text.
758
MOCHINAGA
GASTROENTEROLOGY
ET AL.
z Du -r
St
r
,
Jt!--r -
I-
I
-
-
-
=
-
-t-
Vol. 94, No. 3
= -
r -
I
+
:
T
% OF MMC
CYCLE
GROUP 2
GROUP1
GROUP 3
(H=S)
(n=9)
(N=7)
mp>o.os
*p<0.005
x p >0.05
The bile flow pattern was cyclic in group 1 with a peak during late phase II activity and a trough during phase I activity. The bile flow fluctuated randomly when phase III activity started ectopically in groups 2 and 3. The migrating motor complex cycle was considered to start and end at the end of phase III activity. Du, duodenum; Je, jejunum; St, stomach.
St
st
-DlJ Je
Du Je
(4 GROUP
W 1
GROUP 1
st
St
Du Je
DU
Je
(Cl GROUP 2
6’) GROUP
3
Representative patterns of bile flow in individual MMC cycles in experiments in groups 1, 2, and 3. Patterns A and U are from group 1,whereas C and Dare from groups 2 and 3, respectively. In most experiments in group 1, bile flow peaked during phase II activity in the duodenum. Bile flow volume fluctuated randomly during the MMC cycle in groups 2 and 3. Du, duodenum; Je, jejunum; St, stomach.
March
1988
significantly less (p < 0.025) than that during late phase II activity, but was significantly greater than that during phase I activity (p < 0.025). Although the mean bile flow column peaked during late phase II activity in group 1 (Figure 3), the pattern of bile flow varied during different MMC cycles in the same dog and in different dogs. Figures 4A and 4B show two typical patterns of bile flow in group 1 MMC cycles. The bile flow generally began to increase with the onset of phase II activity. Multiple peaks of bile flow were usually observed when phase II duration was long. In contrast to group 1, there was no cyclic pattern of bile flow into the duodenum in groups 2 and 3 (Figure 3). The bile flow volume fluctuated randomly throughout the MMC cycle. The maximum bile flow volume was not significantly different than the minimum bile flow volume (p > 0.05), indicating that there was no distinct peak in bile flow. A similar random fluctuation of bile flow volume was observed during individual MMC cycles (Figures 4C and 4D). Relationship Between Bile Flow and Antroduodenal Contractions The entry of bile into the duodenum was closely related to antroduodenal contractions. During phase III activity, bile entered the duodenum in single isolated drops or as a rapid burst of several drops (Figure 5). The single drops entered the duodenum only between two consecutive contractions when there was minimal tension in the duodenal wall at the level of the choledochoduodenal junction (Figure 5). Rapid bursts of drops entered the duodenum during phase III activity only when the duodenal contractions were transiently inhibited (Figure 51. During phase II activity, bile entered the duodenum as single drops in between two consecutive contractions or as a series of drops during prolonged quiescence (Figure 6). In an analysis of 1673 duode-
Relationship between bile flow and phase III contractions. Single drops of bile entered the duodenum in between two successive duodenal contractions. A burst of bile flow occurred when the duodenum was transiently inhibited in concert with phase III contractions in the antrum.
DUODENAL
REGULATION
OF BILE FLOW
I
SG A-4
759
Imv
I
SG D-5 ~ DROPCOUNTER I
Figure
~+.V? 4
1-
*i
ILaiiiicilidi+,,
4,
&
4
,L4
*I,
-i
,
ire 4
6. Relationship between bile flow and phase II contractions Bile entry into the duodenum occurred as a series of drops during intermittent quiescence of the duodenum, and as single drops in between two successive contractions.
nal contractions (1191 drops of bile flow), there were only 69 instances (77 drops] when a drop or drops of bile entered the duodenum while the duodenal wall at the level of the choledochoduodenal junction was in the contracted state. The amplitude of these duodenal contractions was 15.2% t 1.4% of the maximal amplitude of contractions during phase III activity. When bile flow occurred during phase I activity, it usually occurred as a steady series of drops.
Discussion Our findings show that in the fasted state gastroduodenal motor activity plays a major role in the regulation of bile flow into the duodenum. Bile entered the duodenum only when the duodenal wall at the level of the choledochoduodenal junction was not contracting or the amplitude of contraction was small compared with the maximal amplitude of contractions during phase III activity. During phase II activity, bile usually entered the duodenum during intermittent quiescent periods and during phase III activity when the duodenal contractions were transiently inhibited in coordination with a cluster of antral contractions. When bile entered the duodenum in between two consecutive duodenal contractions during phase II or phase III activity, it usually entered as a single drop whose timing coincided with the time of little or no tension in the duodenal wall, These findings suggest that when the duodenal wall contracts, it constricts the common bile duct and the choledochoduodenal junction to impede bile flow. The hepatic secretory pressure and intraductal pressures generated by gallbladder contractions are able to overcome duodenal resistance to outflow only when duodenum is quiet or its contractions are feeble. Our findings are in agreement with the radiographic or manometric observations of others (8,9) who reported that bile flow ceases while the duodenum contracts. Ono et al. (10) reported that bile flow into the
760
MOCHINAGAETAL.
duodenum of humans was independent of duodenal contractions. His conclusions were based on concurrent myoelectric recordings from the sphincter of Oddi and duodenum by electrodes implanted at surgery for gallstone disease. The electrodes to record sphincter of Oddi myoelectric activity were inserted blindly through the duodenal muscle layers. Ono et al. (10) did not include any tracings of myoelectric activity in their paper, nor did they provide evidence that the myoelectric activity they recorded from the sphincter of Oddi muscle was different from that of the adjacent duodenal muscle. The thin sphincter of Oddi muscle layer is anatomically and embryologically distinct, but it is closely integrated with the duodenal muscle. Under these circumstances it would be difficult to record selectively from the sphincter of Oddi muscle by extracellular electrodes. Bile flow into the duodenum was related not only to the absence of duodenal contractions but also to the presence of antral contractions. Matsumoto et al. (11) reported that the antral contractions were closely coupled with gallbladder contractions. Thus bile may enter the duodenum when the propagating gastric contraction arrives in the antrum, the gallbladder contracts, and the duodenum is quiescent. This three-way coordination, wherein the antrum and gallbladder contract and the duodenum relaxes, results in a gastroduodenal-gallbladder-biliary tract triangle that may coordinate the concurrent entry of gastric contents and bile into the duodenum and thereby ensure instant mixing. Our findings show that the cyclic pattern of bile flow in the fasted state previously reported by others (12-14) is present only when phase III activity starts in the proximal duodenum at the level of the choledochoduodenal junction. When phase III activity started ectopically in the jejunum, bile flow fluctuated randomly during the MMC cycle. In such cases, there was no significant or consistent peak in bile flow volume during the MMC cycle. This random pattern of bile flow did not depend on whether phase I activity occurred in the duodenum (group 2) or not (group 3). Interestingly, the volume and rate of bile flow during a MMC cycle were greater when phase III activity started ectopically in the jejunum than when it started in the proximal duodenum. This increase occurred in spite of the fact that the MMC cycle length was shorter when phase III activity started ectopically in the jejunum. The increase in total bile flow was primarily due to an increase in bile flow during phases I and IV of the MMC cycle. One of the reasons for this increase may be that when phase III activity occurs in the proximal duodenum, the forceful persistent contractions may provide a high resis-
GASTROENTEROLOGYVol.94,No.3
tance to bile outflow during this period, thereby diverting the hepatic bile to the gallbladder to be concentrated. When phase III activity does not occur in the duodenum, unconcentrated bile from the liver may continue to flow into the duodenum during the entire MMC cycle without filling the gallbladder. This hypothesis remains to be confirmed by direct measurement of gallbladder filling during normal and ectopic MMC cycles. The use of a drop counter allowed us to correlate the relationship between instantaneous bile entry into the duodenum and duodenal contractions. In contrast, the radioisotope method measures the total bile flow in a given sampling period. Because the internal diameter of the exteriorized catheters was made about the same as that of the common bile duct, the resistance introduced by the catheter assembly was negligible. A catheter inserted into the sphincter of Oddi and common bile duct under direct vision through a duodenal cannula can also be used to measure bile flow volume in drops (15). However, the insertion of a duodenal cannula may modify duodenal contractile activity at the level of the sphincter of Oddi. Furthermore, the insertion of a catheter through the sphincter of Oddi may impair the ability of the sphincter and the duodenal wall to constrict it. So far four factors that contribute to the cycling of bile flow in the fasted state have been identified: (a) migrating motor complexes that transport the bile from the duodenum to the ileum for reabsorption, (b) reabsorption of bile acids in the ileum and their return to the liver for recirculation, (c) gallbladder motor activity that partially empties the gallbladder and (d) resisduring each MMC cycle (11,16,17), tance of the sphincter of Oddi that may impede bile flow (18-21) or propagating contractions in some species (like the opossum), which may propel bile (22). The roles of MMCs and ileal reabsorption are well understood, but those of the gallbladder and the choledochoduodenal junction are understood incompletely. Scott et al. (20) recorded sphincter of Oddi contractions from the dog by manometric methods and concluded that these contractions regulate bile flow during the fasted state. The frequency of contractions recorded from the sphincter of Oddi by them had the same frequency as that of duodenal electrical control activity. Furthermore, there was a 1:1 relationship between sphincter of Oddi contractions and bursts of duodenal electrical response activity representing duodenal contractions. Because of the close integration of the sphincter of Oddi and duodenal muscle layers in the dog, it is possible that the duodenal electrical control activity may drive the sphincter of Oddi electrical control activity so that the two may contract in phase. It is
March
1988
also possible, however, that the canine sphincter of Oddi may not have independent phasic contractions and the intraluminal contractions recorded manometrically from the intramural sphincter of Oddi and common bile duct may be due to duodenal muscle contractions. Because the distal common bile duct transverses the duodenal wall in the caudad direction, the caudad propagation of duodenal contractions at the level of the choledochoduodenal junction would be recorded as caudad propagating contractions in the intramural common bile duct by the manometric method. Regardless of whether the canine sphincter of Oddi has independent contractions or not, our findings indicate that duodenal contractile activity is an important factor in the regulation of intraduodenal bile flow, in addition to the four factors cited above. In summary, bile flow into the duodenum depends on the balance between the driving forces and the resisting forces. The driving forces for bile flow are hepatic secretory pressure, which is dictated by the enterohepatic circulation of bile acids, and gallbladder motor activity. The common bile duct in most species acts as a simple conduit. In some species, like the opossum, the propagating contractions in the extraduodenal sphincter of Oddi may also provide an additional driving force. The resistive forces that deter bile flow into the duodenum are the passive resistance of the sphincter of Oddi and duodenal contractile activity.
DUODENAL
8. 9. 10. 11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
2. 3.
4.
5.
6
HE, Christensen GC. Miller’s anatomy of the dog. 2nd ed. Philadelphia: WB Saunders, 1979. Boyden EA. The anatomy of the choledochoduodenal junction in man. Surg Gynecol Obstet 1957;104:641-52. Sarna SK, Kitai R, Muniappan K, Marzio L, Daniel EE, Waterfall WE. Gastroduodenal coordination: a computer analysis (abstr). In: Duthie HL, ed. Gastrointestinal motility in health and disease. Lancaster: MTP Press, 1978:271-2. Allen GL, Poole EW, Code CF. Relationships between electrical activities of antrum and duodenum. Am J Physiol 1964; 207:906-10. Sarna S, Northcott P, Belbeck L. Mechanisms of cycling of migrating myoelectric complexes-the effect of morphine. Am J Physiol 1982;242:G588-95. Code CF, Marlett JA. The interdigestive myoelectric complex of the stomach and small bowel of dogs. J Physiol (Lond) 1975:246:289-309.
OF BILE FLOW
761
7. Sarna
References 1. Evans
REGULATION
21.
22.
SK. Cyclic motor activity; migrating motor complex: 1985. Gastroenterology 1985;89:894-913. Potter JC, Mann FC. Pressure changes in the biliary tract. Am J Med Sci 1926:171:202-17. Nebesar RA, Pollard JJ, Potsaid MS. Cine cholangiography: some physiologic observations. Radiology 1966;86:475-9. Ono K, Watanabe N. Suzuki K, Tsuchida H, Sugiyama Y, Abo M. Bile flow mechanism in man. Arch Surg 1968;96:869-74. Matsumoto T, Sarna SK, Condon RE. Dodds WJ. Mochinaga N. Canine gallbladder cyclic motor activity. Am J Physiol 1987 (in press). Peeters TL, Vantrappen G, Janssens J, Bile acid output and the interdigestive migrating motor complex in normals and in cholecystectomy patients. Gastroenterology 1980;79:678-81. Keane FB, DiMagno EP, Dozois RR, Go VLW. Relationships among canine interdigestive exocrine pancreatic and biliary flow, duodenal motor activity, plasma pancreatic polypeptide, and motilin. Gastroenterology 1980;78:310-6. Scott RB, Eidt PB, Shaffer EA. Regulation of fasting canine duodenal bile acid delivery by sphincter of Oddi and gallbladder. Am J Physiol 1985;249:G622-33. Scott RB, Strasberg SM, El-Sharkawy TY. Diamant NE. Regulation of the fasting enteroheptic circulation of bile acids by the migrating myoelectric complex in dogs. J Clin Invest 1983;71:644-54. Itoh Z, Takahashi I. Periodic contractions of the canine gallbladder during the interdigestive state. Am J Physiol 1981;240:G183-9. Traynor OJ. Dozois RR, DiMagno EP. Canine interdigestive and postprandial gallbladder motility and emptying. Am J Physiol 1984;246:G426-32. Ryan JP. Motility of the gallbladder and biliary tree. In: LR Johnson, ed. Physiology of the gastrointestinal tract. New York: Raven, 1981:473-94. Becker JM. Duff WM, Moody FG. Myoelectric control of gastrointestinal and biliary motility: a review. Surgery 1981;89:466-77. Scott RB, Strasberg SM, El-Sharkawy TY, Diamant NE. Fasting canine biliary secretion and the sphincter of Oddi. Gastroenterology 1984:87:793-804. Toouli J, Geenen JE, Hogan WJ, Dodds WJ, Arndorfer KC. Sphincter of Oddi motor activity: a comparison between patients with common bile duct stones and controls. Gastroenterology 1982;82:111-7. Toouli J, Dodds WJ, Honda R. Sarna S, et al. Motor function of the opossum sphincter of Oddi. J Clin Invest 1983:71:208-20.
Received February 18. 1987. Accepted October 21, 1987. Address requests for reprints to: Sushi1 K. Sarna, Ph.D., Zablocki Veterans Administration Medical Center, Surgical Research 151. 5000 West National Avenue, Milwaukee, Wisconsin 53295. This work was supported in part by grant 7722-OlP from the Veterans Administration Research Service and DK32346 from the National Institutes of Diabetes and Digestive and Kidney Diseases to Dr. Sarna.