- Email: [email protected]
Sincalide-aided ultrasonography of the common bile duct as a predictor of biliary obstruction determined by ERCP and biliary manometry
0016-5107/90/3605-0467$02.00 GASTROINTESTINAL ENDOSCOPY Copyright © 1990 by the American Society for Gastrointestinal Endoscopy
Sincalide-aided ultrasonography of the common bile duct as a predictor of biliary obstruction determined by ERCP and biliary manometry c. A. Aronchick, MD, w. G. M. Ritchie, MD S. M. Kaplan, MD, S. H. Wright, MD
J. N. Retig, MD, W. H. Lipshutz, MD Philadelphia, Pennsylvania
Sonographically observed changes in common bile duct caliber following intravenous sincalide injection were correlated with distal common duct pathology as defined by endoscopic retrograde cholangiopancreatography and biliary manometry. Thirty-two patients, 17 with prior cholecystectomies, were studied. In post-cholecystectomy patients, a 1-mm or greater diminution of duct caliber within 5 min represented a normal response. When gallbladder contraction occurred in normal patients with an intact gallbladder, no change or a diminution in duct caliber was observed. When gallbladder contraction was not observed, a normal response was considered to be the same as that in post-cholecystectomy patients with a diminution in duct caliber occurring. By using these criteria two false negatives, both with hypertensive sphincters of Oddi that responded normally to sincalide injection, were encountered. This technique was valuable in defining non-obstructed post-cholecystectomy dilated bile ducts which demonstrated a prompt diminution in caliber following sincalide injection. (Gastrointest Endosc 1990;36:467-471)
ERCP and sphincter of Oddi manometry (SOM) are invasive, specialized techniques used to diagnose obstruction and motility disorders of the biliary tree. These diagnostic modalities have certain drawbacks. They require technical and interpretive expertise which is not universally available and they have a small but significant associated morbidity. By comparison ultrasonography is widely available and has become an accepted non-invasive modality for assessing common bile duct (CBD) dilation. Recent studies have indicated that changes in CBD diameter measured by real-time ultrasound following a fatty meal can predict CBD obstruction. 1• 2 However, the reReceived August 22, 1988. For revision November 1, 1988. Accepted June 11, 1990. From the Gastrointestinal Section, Departments of Medicine and Diagnostic Radiology, Pennsylvania Hospital, Philadelphia, Pennsylvania. Reprint requests: C. A. Aronchik, MD, Pennsylvania Hospital, Gastroenterology/Medicine, 8th and Spruce Streets, Philadelphia, Pennsylvania 19107. VOLUME 36, NO.5, 1990
sponse of the gallbladder and sphincter of Oddi to a fatty meal depends upon several factors such as the rate of gastric emptying and the amount and timing of endogenous cholecystokinin (CCK) release. Parenterally administered sincalide (Kinevac,® Squibb Diagnostics, New Brunswick, N.J.), the synthetic terminal octapeptide of CCK, has been shown to contract the gallbladder3 and relax the sphincter of Oddi in man 4 and induce an increase in bile duct flow or choleresis. The purpose of this study was to evaluate the changes in CBD diameter measured by real-time ultrasound in response to intravenous sincalide and to correlate these responses with ERCP and sphincter of Oddi manometry in patients with suspected biliary tract disease. PATIENTS AND METHODS
Thirty-two consecutive patients (5 men, 27 women) (average age 40 years) referred for biliary tract-type pain or 467
suspected biliary tract obstruction were evaluated with sincalide-aided ultrasonography (sincalide-US) and ERCP. In those patients in whom no organic obstruction was defined, SOM was performed. The patients without evidence of organic obstruction also had normal radiographic upper gastrointestinal and small bowel series, barium enemas, and CT of the abdomen. Seventeen patients had had previous cholecystectomies. In all patients, real-time sincalide-aided ultrasonography of the biliary tree was performed. With the patient in the supine or left posterior oblique position, baseline CBD diameters were measured at three levels, at the porta hepatis at the level of the right hepatic artery; at the greatest duct diameter as the CBD lies in the hepatogastric ligament; and in the retropancreatic portion. Baseline gallbladder size was measured in three dimensions. After two baseline measurements at -15 min and 0 min, 0.04 J.Lg/kg of sincalide was given by slow intravenous injection over 30 sec. The CBD was observed continuously for the first 5 min, then at 10, 15, and 30 min post-injection. The presence or absence of gallbladder contraction was also assessed over this period of time. ERCP and SOM were performed separately. Sedation for ERCP was meperidine and diazepam, and diazepam alone for SOM to avoid alteration in basal sphincter of Oddi pressure. For SOM, an Olympus JF-LTlO duodenoscope was advanced to the ampulla. A triple lumen, water-perfused (0.25 ml/min) Arndorfer biliary manometry catheter, attached to a Beckman rectilinear recorder, was passed into the duodenal lumen and a stable zero reference tracing was obtained. The catheter was then freely passed into the pancreatic-biliary ductal system and ductal pressure was recorded. The catheter was then pulled back into the sphincter zone where characteristic phasic activity was seen. Baseline sphincter of Oddi pressure profiles were recorded for 1 min, after which sincalide (0.04 J.Lg/kg/intravenous injection) was administered and the sphincter response noted. Sphincter of Oddi basal pressure was taken as the mean of the nadir of phasic activity over 1 min compared with the duodenal baseline. The highest pressure of all three leads was used. The final pressure being taken as the mean of two pull-throughs. Normal basal sphincter of Oddi pressure was determined in 10 healthy volunteers and found to be less than 40 mm Hg above duodenal pressure. Phasic wave frequency over 1 min was recorded and propagation of phasic activity was measured as the "take off" point of phasic pressure waves in the three leads (proximal, middle, and distal) at a paper speed of 2.5 mm/sec. A normal manometric response of the sphincter of Oddi to intravenous sincalide was defined as an obliteration of phasic activity or a decrease in basal tone. A persistence of phasic activity or an increase in basal tone was considered abnorma1. 4 The change in size of the CBD during sincalideUS was correlated with the ERCP and manometric studies. All ultrasound examinations were read blindly by two radiologists (W. G. M. R. and S. M. K.) without knowledge of the results ofthe ERCP and biliary manometry. All biliary manometric tracings were read blindly by C. A. without knowledge of the ultrasound results. This study was approved by the Human Research Committee of Pennsylvania Hospital on May 21, 1985.
468
RESULTS
The change in CBD diameter measured sonographically in response to intravenous sincalide was dependent on the presence or absence of the gallbladder and, if present, whether or not the gallbladder contracted. Post-cholecystectomy patients
Seventeen post-cholecystectomy patients were evaluated (Table 1). Eleven had a normal ERCP and sphincter of Oddi manometry. Sincalide-US demonstrated a diminution in CBD diameter of greater than 1 mm within 5 min in all of these patients. Six patients had an abnormal ERCP or manometry. Four of these had a sincalide-US study showing no change or an increase in diameter of the CBD within 5 min. One had a non-impacted CBD stone and three had sphincter of Oddi spasm on baseline tracings, with all three having an abnormal sphincteric response to sincalide. Two patients that had baseline sphincter of Oddi spasm had normal sincalide-US studies. Both of these patients had a normal manometric response to intravenous sincalide. Seven of the post-cholecystectomy patients had a dilated CBD (>8 mm) on the baseline sonogram. Five had normal ERCP and manometry and all five demonstrated a marked decrease in CBD diameter after sincalide injection. Four of five had a greater than 3 mm decrease in CBD diameter. Two patients had sphincter of Oddi spasm demonstrated by manometry with an abnormal manometric response of the sphincter to sincalide and both of these patients had abnormal sincalide-US with failure of diminution in size of the CBD. Intact gallbladder patients
Fifteen patients had intact gallbladders (Table 2). Nine had normal ERCP and manometry. In eight of these nine patients with normal results, gallbladder contraction was observed and no change or a diminution in CBD caliber was noted. The remaining normal patient had a non-contracting gallbladder and demonstrated a decrease in CBD size. Six patients had obstructive abnormalities detected on ERCP or manometry. Three had nonimpacted CBD stones, one had pancreatic carcinoma, one had cholangiocarciTable 1. Post-cholecystectomy (17 patients)
Sincalide-US
Normal ERCP or SOM Abnormal
Normal
Abnormal
11
o
2
4
GASTROINTESTINAL ENDOSCOPY
Table 2. Intact gallbladders (15 patients) Sincalide-US
Normal ERCP or SOM Abnormal
Normal
Abnormal
9
o
o
6
Table 3. All cases (32 patients) Sincalide-US Normal Normal ERCPor SOM Abnormal
Abnormal
20
o
2
10
noma, and one had sphincter of Oddi spasm. In these six patients, the CBD diameter was unchanged and gallbladder contraction was not observed following sincalide injection. The comparison between sincalide-US and ERCP and sphincter of Oddi manometry in both groups is shown in Table 3. All patients who had an abnormal sincalide-US had bile duct obstruction found on ERCP or manometry. Two patients whose sincalideUS was normal had elevated basal sphincter of Oddi pressures with an appropriate reduction in pressure in response to intravenous sincalide. Twelve patients had abnormalities detected on ERCP or manometry. All six patients with organic pathology (four CBD stones, one pancreatic carcinoma, and one cholangiocarcinoma) had abnormal liver function tests and abnormal sincalide-US. Six patients were found to have sphincter of Oddi spasm, four of whom had normal liver function tests, but abnormal sincalide-US. DISCUSSION
Although ultrasonography is widely accepted as an accurate method of detection of biliary tract obstruction,5 there are still significant false positive and false negative rates.6~9 To improve the diagnostic accuracy, several authors have performed ultrasound examinations before and after a fatty meal. 1,2, 10 Their findings suggest that in healthy patients with non-dilated and non-obstructed ducts there is a decrease or no change in caliber of the duct when observed 45 min after fatty meal ingestion. In patients with obstruction, an increase in bile duct caliber was observed. These results stimulated us to see if we could improve on their technique. The use of an intravenously administered CCK analogue rather than a fatty meal might decrease the ultrasound examination time and guarantee a VOLUME 36, NO.5, 1990
hormonal effect since the variability of fat absorption would be eliminated. This was confirmed as all significant US changes of bile duct caliber occurred within 5 min of sincalide injection. These changes, however, were more complex than those described in response to a fatty meal and were dependent on the presence or absence of the gallbladder. In the post-cholecystectomy state, a normal sincalide-US was defined by a decrease of common bile duct caliber by > 1 mm within 5 min of sincalide injection. In patients who had intact gallbladders, the normal common bile duct response to sincalide depended on whether the gallbladder contracted. If the gallbladder did not contract, then the normal common bile duct response was similar to the post-cholecystectomy state. If the gallbladder did contract, then the normal common bile duct response was either a decrease or no change in common bile duct caliber. The last finding is in keeping with the observations of Fein et al. l l in a population of normal volunteers using sincalide and also those of Simeone et al. 2 using a fatty meal as the provocative agent. Demonstration of a dilated common bile duct after cholecystectomy6,9 is a frequent clinical problem and may either reflect distal bile duct pathology or a physiologic response following cholecystectomy. Currently, this problem can only be answered by ERCP and manometry. Previous observations 1,2,10 and our data indicate that a decrease in common bile duct caliber after ingestion of a fatty meal or sincalide injection predicts with a high degree of accuracy the absence of organic pathology or sphincter of Oddi dysfunction. We demonstrated a marked decrease in common bile duct size after sincalide injection in four of five normal post-cholecystectomy patients with baseline dilated common bile ducts (Fig. 1). We conclude that sincalide-US is able to predict the presence of the physiologic distention of the bile duct which occurs in many post-cholecystectomy patients. Simeone et al. 2 described an abnormal response as an increase in size of a normal caliber duct or an increase or no change in size in a dilated duct when measured at 45 min after a fatty meal. In our series, an abnormal response consisted of a failure of diminution of bile duct caliber in post-cholecystectomy patients and in patients in whom the gallbladder did not contract following sincalide injection. Although the sincalide-US observations were not extended to 45 min, bile duct caliber usually showed a slow return to pre-injection level in patients with normal responses by 30 min and no consistent pattern of increase in size was observed in those cases with obstruction. This difference in response may reflect a difference in action of sincalide and a fatty meal. It has been shown that injection of CCK produces a shortlasting supraphysiologic level of plasma CCK, whereas ingestion of a fatty meal produces physiologic levels 469
Figure 1. A, Fifty-three-year-old male post-cholecystectomy patient with a 9-mm bile duct. B, Following sincalide injection, the bile duct caliber rapidly diminished to 3 mm. ERCP and biliary manometry confirmed the absence of obstruction.
of CCK of a much longer duration. 3 Although this does not appear to cause significant differences in gallbladder contraction, a fatty meal may produce a more prolonged choleresis and this in the face of an organic or functional obstruction could account for the duct dilation observed by Simeone et al. 2 at 45 min. The major difference in our study as compared with earlier studies where either a fatty meap,2,l0 or sincalide l l was used as the provocative agent is in the correlation of our results with ERCP findings and biliary manometry in those patients in whom no organic obstruction was identified. Retrograde sphincter of Oddi manometry yields a direct measurement of basal sphincter of Oddi tone and propagation of phasic waves. The normal sphincter of Oddi profile and the normal response of the sphincter to CCK has been described. 12 Many patients with sphincter of Oddi dysfunction defined by biliary manometry have normal liver function tests and nondilated ducts. 8 Indeed, we found that four of six patients with sphincter of Oddi spasm had normal liver function tests and abnormal sincalide-US. However, 470
sincalide-US may be beneficial in diagnosing some but not all cases of sphincter of Oddi dysfunction. Two distinct subsets of patients with sphincter of Oddi spasm have been defined by their response to CCK. One group responds normally with a diminution of an elevated sphincter pressure and phasic waves while another group demonstrates a paradoxical increase in sphincter tone. 4 Two of our six patients with elevated sphincter of Oddi pressures demonstrated a normal manometric response to CCK and these cases also demonstrated a normal sincalide-US response of the bile duct. This inability to define a normally responding hypertensive sphincter of Oddi would appear to be a shortcoming of sincalide ultrasonography. The addition of biliary scintigraphy with technetium 99mlabeled iminodiacetic acid derivatives may assist in the recognition of biliary dysfunction in this subset of patients with sphincter of Oddi spasm. 13-15 Sincalide ultrasonography improved the accuracy of ultrasound in the detection of biliary obstruction or sphincter dysfunction. In post-cholecystectomy patients with bile duct dilation, sincalide injection produces a marked reduction in bile duct caliber in the absence of distal bile duct pathology. This observation may have clinical utility in the selection of patients for further invasive endoscopic studies. An abnormal sincalide response always predicted distal bile duct pathology even in the presence of a normal caliber bile duct. The only limitation of sincalide sonography is its inability to detect sphincter of Oddi dysfunction in those patients with a hypertensive sphincter that responds normally to sincalide. Certainly larger numbers of patients will need to be studied before sincalide ultrasonography can be used as a reliable test to determine the need for ERCP and sphincter of Oddi manometry.
REFERENCES 1. Simeone JF, Mueller PR, Ferrucci JT, et al. Sonography of the
2. 3. 4. 5. 6. 7. 8.
bile ducts after a fatty meal: an aid in detection of obstruction. Radiology 1982;143:211-5. Simeone JF, Butch RJ, Mueller PR, et al. The bile ducts after a fatty meal: further sonographic observation. Radiology 1985;154:763-8. Hopman WPM, Rosenbusch G, Jansen JBMJ, Lamers CBHW. Gallbladder contraction: effects of fatty meals and cholecystokinin. Radiology 1985;157:37-9. Toouli J, Roberts-Thomson IC, Dent J, Lee J. Manometric disorders in patients with suspected sphincter of Oddi dysfunction. Gastroenterology 1985;88:1243-50. Cooperberg PL, Li D, Wong P, Cohen MM, Burhenne HJ. Accuracy of common hepatic duct size in the evaluation of extrahepatic biliary obstruction. Radiology 1980;135:141-4. Weinstein BJ, Weinstein DP. Biliary tract dilatation in the non-jaundiced patient. AJR 1980;134:899-904. Muhletaler CA, Gerlock AJ Jr, Fleischer AC, James AE Jr. Diagnosis of obstructive jaundice with non-dilated bile ducts. AJR 1980;134:1149-52. Mueller PR, Ferrucci JT, Simeone JF, Wittenberg J, et al. Postcholecystectomy bile duct dilatation: myth or reality. AJR 1981;136:355-8.
GASTROINTESTINAL ENDOSCOPY
9. vanSonnenberg E, Ferrucci JT, Neff CC, Mueller PR, Simeone JF, Wittenberg J. Biliary pressure: manometric and perfusion studies at percutaneous transhepatic cholangiography and percutaneous biliary drainage. Radiology 1983;148:41-50. 10. Willson SA, Gosink BB, vanSonnenberg E. Unchanged size of a dilated common bile duct after a fatty meal: Results and significance. Radiology 1986;160:29-31. 11. Fein AB, Rauch RF, Bowie JD, Halvorsen RA, Rosenberg ER. Intravenous cholecystokinin octapeptide: its effect on the sonographic appearance of the bile ducts in normal subjects. Radiology 1984;153:499-501. 12. Geenen JE, Hogan WJ, Dodds WJ, Stewart ET, Arndorfer RC.
VOLUME 36, NO.5, 1990
Intraluminal pressure recordings from the human sphincter of Oddi. Gastroenterology 1980;8:317-24. 13. Lee RGL, Gregg JA, Koroshetz AM, Hill TC, Clouse ME. Sphincter of Oddi stenosis: Diagnosis using hepatobiliary scintigraphy and endoscopic manometry. Radiology 1985;156:7936. 14. Lieberman DA, Krishnamurthy GT. Intrahepatic cholestasis: discrimination with biliary scintigraphy combined with ultrasound. Gastroenterology 1986;90:734-43. 15. Shaffer EA, Hershfield NB, Logan K, Kloiber R. Cholescintigraphic detection of functional obstruction of the sphincter of Oddi. Gastroenterology 1986;90:728-33.
471
Sincalide-aided ultrasonography of the common bile duct as a predictor of biliary obstruction determined by ERCP and biliary manometry c. A. Aronchick, MD, w. G. M. Ritchie, MD S. M. Kaplan, MD, S. H. Wright, MD
J. N. Retig, MD, W. H. Lipshutz, MD Philadelphia, Pennsylvania
Sonographically observed changes in common bile duct caliber following intravenous sincalide injection were correlated with distal common duct pathology as defined by endoscopic retrograde cholangiopancreatography and biliary manometry. Thirty-two patients, 17 with prior cholecystectomies, were studied. In post-cholecystectomy patients, a 1-mm or greater diminution of duct caliber within 5 min represented a normal response. When gallbladder contraction occurred in normal patients with an intact gallbladder, no change or a diminution in duct caliber was observed. When gallbladder contraction was not observed, a normal response was considered to be the same as that in post-cholecystectomy patients with a diminution in duct caliber occurring. By using these criteria two false negatives, both with hypertensive sphincters of Oddi that responded normally to sincalide injection, were encountered. This technique was valuable in defining non-obstructed post-cholecystectomy dilated bile ducts which demonstrated a prompt diminution in caliber following sincalide injection. (Gastrointest Endosc 1990;36:467-471)
ERCP and sphincter of Oddi manometry (SOM) are invasive, specialized techniques used to diagnose obstruction and motility disorders of the biliary tree. These diagnostic modalities have certain drawbacks. They require technical and interpretive expertise which is not universally available and they have a small but significant associated morbidity. By comparison ultrasonography is widely available and has become an accepted non-invasive modality for assessing common bile duct (CBD) dilation. Recent studies have indicated that changes in CBD diameter measured by real-time ultrasound following a fatty meal can predict CBD obstruction. 1• 2 However, the reReceived August 22, 1988. For revision November 1, 1988. Accepted June 11, 1990. From the Gastrointestinal Section, Departments of Medicine and Diagnostic Radiology, Pennsylvania Hospital, Philadelphia, Pennsylvania. Reprint requests: C. A. Aronchik, MD, Pennsylvania Hospital, Gastroenterology/Medicine, 8th and Spruce Streets, Philadelphia, Pennsylvania 19107. VOLUME 36, NO.5, 1990
sponse of the gallbladder and sphincter of Oddi to a fatty meal depends upon several factors such as the rate of gastric emptying and the amount and timing of endogenous cholecystokinin (CCK) release. Parenterally administered sincalide (Kinevac,® Squibb Diagnostics, New Brunswick, N.J.), the synthetic terminal octapeptide of CCK, has been shown to contract the gallbladder3 and relax the sphincter of Oddi in man 4 and induce an increase in bile duct flow or choleresis. The purpose of this study was to evaluate the changes in CBD diameter measured by real-time ultrasound in response to intravenous sincalide and to correlate these responses with ERCP and sphincter of Oddi manometry in patients with suspected biliary tract disease. PATIENTS AND METHODS
Thirty-two consecutive patients (5 men, 27 women) (average age 40 years) referred for biliary tract-type pain or 467
suspected biliary tract obstruction were evaluated with sincalide-aided ultrasonography (sincalide-US) and ERCP. In those patients in whom no organic obstruction was defined, SOM was performed. The patients without evidence of organic obstruction also had normal radiographic upper gastrointestinal and small bowel series, barium enemas, and CT of the abdomen. Seventeen patients had had previous cholecystectomies. In all patients, real-time sincalide-aided ultrasonography of the biliary tree was performed. With the patient in the supine or left posterior oblique position, baseline CBD diameters were measured at three levels, at the porta hepatis at the level of the right hepatic artery; at the greatest duct diameter as the CBD lies in the hepatogastric ligament; and in the retropancreatic portion. Baseline gallbladder size was measured in three dimensions. After two baseline measurements at -15 min and 0 min, 0.04 J.Lg/kg of sincalide was given by slow intravenous injection over 30 sec. The CBD was observed continuously for the first 5 min, then at 10, 15, and 30 min post-injection. The presence or absence of gallbladder contraction was also assessed over this period of time. ERCP and SOM were performed separately. Sedation for ERCP was meperidine and diazepam, and diazepam alone for SOM to avoid alteration in basal sphincter of Oddi pressure. For SOM, an Olympus JF-LTlO duodenoscope was advanced to the ampulla. A triple lumen, water-perfused (0.25 ml/min) Arndorfer biliary manometry catheter, attached to a Beckman rectilinear recorder, was passed into the duodenal lumen and a stable zero reference tracing was obtained. The catheter was then freely passed into the pancreatic-biliary ductal system and ductal pressure was recorded. The catheter was then pulled back into the sphincter zone where characteristic phasic activity was seen. Baseline sphincter of Oddi pressure profiles were recorded for 1 min, after which sincalide (0.04 J.Lg/kg/intravenous injection) was administered and the sphincter response noted. Sphincter of Oddi basal pressure was taken as the mean of the nadir of phasic activity over 1 min compared with the duodenal baseline. The highest pressure of all three leads was used. The final pressure being taken as the mean of two pull-throughs. Normal basal sphincter of Oddi pressure was determined in 10 healthy volunteers and found to be less than 40 mm Hg above duodenal pressure. Phasic wave frequency over 1 min was recorded and propagation of phasic activity was measured as the "take off" point of phasic pressure waves in the three leads (proximal, middle, and distal) at a paper speed of 2.5 mm/sec. A normal manometric response of the sphincter of Oddi to intravenous sincalide was defined as an obliteration of phasic activity or a decrease in basal tone. A persistence of phasic activity or an increase in basal tone was considered abnorma1. 4 The change in size of the CBD during sincalideUS was correlated with the ERCP and manometric studies. All ultrasound examinations were read blindly by two radiologists (W. G. M. R. and S. M. K.) without knowledge of the results ofthe ERCP and biliary manometry. All biliary manometric tracings were read blindly by C. A. without knowledge of the ultrasound results. This study was approved by the Human Research Committee of Pennsylvania Hospital on May 21, 1985.
468
RESULTS
The change in CBD diameter measured sonographically in response to intravenous sincalide was dependent on the presence or absence of the gallbladder and, if present, whether or not the gallbladder contracted. Post-cholecystectomy patients
Seventeen post-cholecystectomy patients were evaluated (Table 1). Eleven had a normal ERCP and sphincter of Oddi manometry. Sincalide-US demonstrated a diminution in CBD diameter of greater than 1 mm within 5 min in all of these patients. Six patients had an abnormal ERCP or manometry. Four of these had a sincalide-US study showing no change or an increase in diameter of the CBD within 5 min. One had a non-impacted CBD stone and three had sphincter of Oddi spasm on baseline tracings, with all three having an abnormal sphincteric response to sincalide. Two patients that had baseline sphincter of Oddi spasm had normal sincalide-US studies. Both of these patients had a normal manometric response to intravenous sincalide. Seven of the post-cholecystectomy patients had a dilated CBD (>8 mm) on the baseline sonogram. Five had normal ERCP and manometry and all five demonstrated a marked decrease in CBD diameter after sincalide injection. Four of five had a greater than 3 mm decrease in CBD diameter. Two patients had sphincter of Oddi spasm demonstrated by manometry with an abnormal manometric response of the sphincter to sincalide and both of these patients had abnormal sincalide-US with failure of diminution in size of the CBD. Intact gallbladder patients
Fifteen patients had intact gallbladders (Table 2). Nine had normal ERCP and manometry. In eight of these nine patients with normal results, gallbladder contraction was observed and no change or a diminution in CBD caliber was noted. The remaining normal patient had a non-contracting gallbladder and demonstrated a decrease in CBD size. Six patients had obstructive abnormalities detected on ERCP or manometry. Three had nonimpacted CBD stones, one had pancreatic carcinoma, one had cholangiocarciTable 1. Post-cholecystectomy (17 patients)
Sincalide-US
Normal ERCP or SOM Abnormal
Normal
Abnormal
11
o
2
4
GASTROINTESTINAL ENDOSCOPY
Table 2. Intact gallbladders (15 patients) Sincalide-US
Normal ERCP or SOM Abnormal
Normal
Abnormal
9
o
o
6
Table 3. All cases (32 patients) Sincalide-US Normal Normal ERCPor SOM Abnormal
Abnormal
20
o
2
10
noma, and one had sphincter of Oddi spasm. In these six patients, the CBD diameter was unchanged and gallbladder contraction was not observed following sincalide injection. The comparison between sincalide-US and ERCP and sphincter of Oddi manometry in both groups is shown in Table 3. All patients who had an abnormal sincalide-US had bile duct obstruction found on ERCP or manometry. Two patients whose sincalideUS was normal had elevated basal sphincter of Oddi pressures with an appropriate reduction in pressure in response to intravenous sincalide. Twelve patients had abnormalities detected on ERCP or manometry. All six patients with organic pathology (four CBD stones, one pancreatic carcinoma, and one cholangiocarcinoma) had abnormal liver function tests and abnormal sincalide-US. Six patients were found to have sphincter of Oddi spasm, four of whom had normal liver function tests, but abnormal sincalide-US. DISCUSSION
Although ultrasonography is widely accepted as an accurate method of detection of biliary tract obstruction,5 there are still significant false positive and false negative rates.6~9 To improve the diagnostic accuracy, several authors have performed ultrasound examinations before and after a fatty meal. 1,2, 10 Their findings suggest that in healthy patients with non-dilated and non-obstructed ducts there is a decrease or no change in caliber of the duct when observed 45 min after fatty meal ingestion. In patients with obstruction, an increase in bile duct caliber was observed. These results stimulated us to see if we could improve on their technique. The use of an intravenously administered CCK analogue rather than a fatty meal might decrease the ultrasound examination time and guarantee a VOLUME 36, NO.5, 1990
hormonal effect since the variability of fat absorption would be eliminated. This was confirmed as all significant US changes of bile duct caliber occurred within 5 min of sincalide injection. These changes, however, were more complex than those described in response to a fatty meal and were dependent on the presence or absence of the gallbladder. In the post-cholecystectomy state, a normal sincalide-US was defined by a decrease of common bile duct caliber by > 1 mm within 5 min of sincalide injection. In patients who had intact gallbladders, the normal common bile duct response to sincalide depended on whether the gallbladder contracted. If the gallbladder did not contract, then the normal common bile duct response was similar to the post-cholecystectomy state. If the gallbladder did contract, then the normal common bile duct response was either a decrease or no change in common bile duct caliber. The last finding is in keeping with the observations of Fein et al. l l in a population of normal volunteers using sincalide and also those of Simeone et al. 2 using a fatty meal as the provocative agent. Demonstration of a dilated common bile duct after cholecystectomy6,9 is a frequent clinical problem and may either reflect distal bile duct pathology or a physiologic response following cholecystectomy. Currently, this problem can only be answered by ERCP and manometry. Previous observations 1,2,10 and our data indicate that a decrease in common bile duct caliber after ingestion of a fatty meal or sincalide injection predicts with a high degree of accuracy the absence of organic pathology or sphincter of Oddi dysfunction. We demonstrated a marked decrease in common bile duct size after sincalide injection in four of five normal post-cholecystectomy patients with baseline dilated common bile ducts (Fig. 1). We conclude that sincalide-US is able to predict the presence of the physiologic distention of the bile duct which occurs in many post-cholecystectomy patients. Simeone et al. 2 described an abnormal response as an increase in size of a normal caliber duct or an increase or no change in size in a dilated duct when measured at 45 min after a fatty meal. In our series, an abnormal response consisted of a failure of diminution of bile duct caliber in post-cholecystectomy patients and in patients in whom the gallbladder did not contract following sincalide injection. Although the sincalide-US observations were not extended to 45 min, bile duct caliber usually showed a slow return to pre-injection level in patients with normal responses by 30 min and no consistent pattern of increase in size was observed in those cases with obstruction. This difference in response may reflect a difference in action of sincalide and a fatty meal. It has been shown that injection of CCK produces a shortlasting supraphysiologic level of plasma CCK, whereas ingestion of a fatty meal produces physiologic levels 469
Figure 1. A, Fifty-three-year-old male post-cholecystectomy patient with a 9-mm bile duct. B, Following sincalide injection, the bile duct caliber rapidly diminished to 3 mm. ERCP and biliary manometry confirmed the absence of obstruction.
of CCK of a much longer duration. 3 Although this does not appear to cause significant differences in gallbladder contraction, a fatty meal may produce a more prolonged choleresis and this in the face of an organic or functional obstruction could account for the duct dilation observed by Simeone et al. 2 at 45 min. The major difference in our study as compared with earlier studies where either a fatty meap,2,l0 or sincalide l l was used as the provocative agent is in the correlation of our results with ERCP findings and biliary manometry in those patients in whom no organic obstruction was identified. Retrograde sphincter of Oddi manometry yields a direct measurement of basal sphincter of Oddi tone and propagation of phasic waves. The normal sphincter of Oddi profile and the normal response of the sphincter to CCK has been described. 12 Many patients with sphincter of Oddi dysfunction defined by biliary manometry have normal liver function tests and nondilated ducts. 8 Indeed, we found that four of six patients with sphincter of Oddi spasm had normal liver function tests and abnormal sincalide-US. However, 470
sincalide-US may be beneficial in diagnosing some but not all cases of sphincter of Oddi dysfunction. Two distinct subsets of patients with sphincter of Oddi spasm have been defined by their response to CCK. One group responds normally with a diminution of an elevated sphincter pressure and phasic waves while another group demonstrates a paradoxical increase in sphincter tone. 4 Two of our six patients with elevated sphincter of Oddi pressures demonstrated a normal manometric response to CCK and these cases also demonstrated a normal sincalide-US response of the bile duct. This inability to define a normally responding hypertensive sphincter of Oddi would appear to be a shortcoming of sincalide ultrasonography. The addition of biliary scintigraphy with technetium 99mlabeled iminodiacetic acid derivatives may assist in the recognition of biliary dysfunction in this subset of patients with sphincter of Oddi spasm. 13-15 Sincalide ultrasonography improved the accuracy of ultrasound in the detection of biliary obstruction or sphincter dysfunction. In post-cholecystectomy patients with bile duct dilation, sincalide injection produces a marked reduction in bile duct caliber in the absence of distal bile duct pathology. This observation may have clinical utility in the selection of patients for further invasive endoscopic studies. An abnormal sincalide response always predicted distal bile duct pathology even in the presence of a normal caliber bile duct. The only limitation of sincalide sonography is its inability to detect sphincter of Oddi dysfunction in those patients with a hypertensive sphincter that responds normally to sincalide. Certainly larger numbers of patients will need to be studied before sincalide ultrasonography can be used as a reliable test to determine the need for ERCP and sphincter of Oddi manometry.
REFERENCES 1. Simeone JF, Mueller PR, Ferrucci JT, et al. Sonography of the
2. 3. 4. 5. 6. 7. 8.
bile ducts after a fatty meal: an aid in detection of obstruction. Radiology 1982;143:211-5. Simeone JF, Butch RJ, Mueller PR, et al. The bile ducts after a fatty meal: further sonographic observation. Radiology 1985;154:763-8. Hopman WPM, Rosenbusch G, Jansen JBMJ, Lamers CBHW. Gallbladder contraction: effects of fatty meals and cholecystokinin. Radiology 1985;157:37-9. Toouli J, Roberts-Thomson IC, Dent J, Lee J. Manometric disorders in patients with suspected sphincter of Oddi dysfunction. Gastroenterology 1985;88:1243-50. Cooperberg PL, Li D, Wong P, Cohen MM, Burhenne HJ. Accuracy of common hepatic duct size in the evaluation of extrahepatic biliary obstruction. Radiology 1980;135:141-4. Weinstein BJ, Weinstein DP. Biliary tract dilatation in the non-jaundiced patient. AJR 1980;134:899-904. Muhletaler CA, Gerlock AJ Jr, Fleischer AC, James AE Jr. Diagnosis of obstructive jaundice with non-dilated bile ducts. AJR 1980;134:1149-52. Mueller PR, Ferrucci JT, Simeone JF, Wittenberg J, et al. Postcholecystectomy bile duct dilatation: myth or reality. AJR 1981;136:355-8.
GASTROINTESTINAL ENDOSCOPY
9. vanSonnenberg E, Ferrucci JT, Neff CC, Mueller PR, Simeone JF, Wittenberg J. Biliary pressure: manometric and perfusion studies at percutaneous transhepatic cholangiography and percutaneous biliary drainage. Radiology 1983;148:41-50. 10. Willson SA, Gosink BB, vanSonnenberg E. Unchanged size of a dilated common bile duct after a fatty meal: Results and significance. Radiology 1986;160:29-31. 11. Fein AB, Rauch RF, Bowie JD, Halvorsen RA, Rosenberg ER. Intravenous cholecystokinin octapeptide: its effect on the sonographic appearance of the bile ducts in normal subjects. Radiology 1984;153:499-501. 12. Geenen JE, Hogan WJ, Dodds WJ, Stewart ET, Arndorfer RC.
VOLUME 36, NO.5, 1990
Intraluminal pressure recordings from the human sphincter of Oddi. Gastroenterology 1980;8:317-24. 13. Lee RGL, Gregg JA, Koroshetz AM, Hill TC, Clouse ME. Sphincter of Oddi stenosis: Diagnosis using hepatobiliary scintigraphy and endoscopic manometry. Radiology 1985;156:7936. 14. Lieberman DA, Krishnamurthy GT. Intrahepatic cholestasis: discrimination with biliary scintigraphy combined with ultrasound. Gastroenterology 1986;90:734-43. 15. Shaffer EA, Hershfield NB, Logan K, Kloiber R. Cholescintigraphic detection of functional obstruction of the sphincter of Oddi. Gastroenterology 1986;90:728-33.
471