Reduction in the incidence of pancreatitis in patients undergoing sphincter of Oddi manometry: A successful quality improvement project. (Original Article).

Objective: Acute pancreatitis is a recognized complication of sphincter of Oddi manometry (SOM). Its frequency of occurrence has been reported in the range of 4 to 31%. In an earlier retrospective study performed at this institution, the incidence of pancreatitis was 9.3% in patients who only had SOM compared with 26.1% in those patients who had SOM and endoscopic retrograde cholangiopancreatography (ERCP) with or without sphincterotomy at the same session. On the basis of these data, a quality-improvement project was initiated at two university-affiliated hospitals. This involved performance of SOM without ERCP. If ERCP was required, it was performed at a different session. The purpose of this project was to decrease the incidence of pancreatitis associated with SOM.

Methods: This study involved prospective patient identification and retrospective chart review of patients who underwent SOM without ERCP between May 1998 and December 2000. SOM was performed using a triple-lumen catheter with water perfusion at a rate of 0.25 ml/min using an Arndorfer pneumohydraulic capillary perfusion system. The data recorded included pancreatitis after SOM, pancreatitis after ERCP and sphincterotomy, average days in the hospital after pancreatitis, and time between SOM and ERCP.

Results: Forty-one patients were studied. Three (7.32%) patients had pancreatitis after SOM. Five patients subsequently underwent ERCP and sphincterotomy and one (20%) patient had pancreatitis. The overall frequency of pancreatitis after SOM and any subsequent ERCP or sphincterotomy was 4 (9.78%) of 41(95% confidence interval, 3.9-22.5%). The odds ratio for pancreatitis with ERCP and SOM at the same time compared with the SOM-only strategy was 3.26 (P 0.05). The average stay in the hospital after pancreatitis ranged from 2 to 4 days, with a mean length of stay of 2.75 days. The time between SOM and subsequent ERCP ranged from 6 to 20 days, with a mean of 10.4 days.

Conclusion: By adopting a protocol to perform diagnostic SOM, separate from ERCP and sphincterotomy, we were able to decrease the incidence of pancreatitis considerably at our institutions.

**********

The sphincter of Oddi is a complex muscular structure that regulates the flow of bile and pancreatic secretions into the duodenum. Sphincter of Oddi manometry (SOM) is a useful diagnostic procedure for evaluating patients with unexplained biliary pain or idiopathic recurrent pancreatitis. (1,2) It is the standard for evaluating patients with sphincter of Oddi dysfunction (SOD) because it can predict which patients will respond to endoscopic biliary sphincterotomy. (3) Acute pancreatitis can occur in up to 31% of patients after SOM. (4,5) Its cause is multifactorial and includes trauma, edema, increased intraductal pressure, overfilling of the ductal system, and spasm of the sphincter resulting from repeated cannulation during endoscopic retrograde cholangiopancreatography (ERCP), endoscopic biliary sphincterotomy, and SOM. (6) In an earlier retrospective study performed at this institution, (7) the incidence of pancreatitis was lower in those patients who had only SOM compared with those patients who had SOM and ERCP (9.3 versus 26.1%, P < 0.026). In an attempt at quality improvement and on the basis of data from the prior retrospective study, SOM has been performed without ERCP since 1998 at our institutions. If ERCP or sphineterotomy is required, it is performed at a different session separated from SOM by at least 24 hours. This study reviews the effects of this protocol on the incidence of pancreatitis in SOM performed between May 1, 1998, and December 31, 2000.

Patients and Methods

During a 32-month period between May 1998 and December 2000, a prospective patient identification and retrospective chart review of patients who underwent SOM without ERCP was performed at two university-affiliated hospitals. Inclusion criteria were patients referred to the University of South Florida for suspected SOD, and SOM performed without ERCP. Patients with Type I SOD were excluded.

All patients underwent extensive evaluation to exclude other causes of abdominal pain before undergoing SOM. SOM was performed using a side-viewing upper endoscope through which a triple-lumen, 5-French catheter was passed into the sphincter of Oddi to measure pressure. The catheter was perfused with bubble-free deionized water at a rate of 0.25 ml/min, using an Arndorfer pneumohydraulic capillary perfusion system (Arndorfer Medical Specialties, Greendale, WI). (8) The basal pressure was considered abnormal if the mean of all recorded pressures was higher than 40 mm Hg. The ductal system entered was confirmed by aspiration of clear (pancreatic) or yellow (bile) fluid. All patients who underwent SOM had free cannulation of the ductal system without the aid of contrast injection.

Patients were categorized according to the Geenen-Hogan classification of SOD into biliary types I, II, and III and pancreatic types I, II, and III. (9) Clinical pancreatitis was diagnosed in patients who developed upper abdominal pain associated with an elevation in amylase at least three times normal. (10) Pancreatitis was graded as mild, moderate, or severe, depending on the length of hospitalization according to the criteria previously proposed. (10)

Data recorded included age, sex, length of hospital stay, type of SOD, sphineterotomy, time between SOM and ERCP, and incidence of pancreatitis. The University of South Florida Institutional Review Board approved the study.

Statistical Analysis

Prior work (7) established the absence of association between the incidence of pancreatitis and age, sex, duration of procedure, dose of midazolam, sphincter of Oddi pressure, and type of SOD. The current project compared the incidence of pancreatitis in patients undergoing SOM and ERCP when the procedures are performed at separate sessions with historical controls undergoing both SOM and ERCP during the same endoscopy session. A [chi square] test was used to assess significance. Controls were 46 patients selected from our previous study (7) who had undergone SOM and ERCP at the same session. Of these 46 patients, 12 (26.1%) had pancreatitis.

Results

Forty-one patients were studied (Table 1). All were female patients except one. Patients' ages ranged from 14 to 77 years, with a mean age of 40.9 years. Thirty-four patients (82.9%) had biliary type SOD and 7 (17%) had pancreatic type SOD. Three patients (7.32%) had pancreatitis after SOM. Five patients subsequently underwent ERCP and sphineterotomy, and one (20%) of these patients had pancreatitis. The overall frequency of pancreatitis after SOM and any subsequent ERCP or sphineterotomy was 4 (9.78%) of 41(95% confidence interval, 3.9-22.5%). In the historical control group, 12 patients had pancreatitis and 34 did not have pancreatitis; in the present study, 4 patients had pancreatitis and 37 did not have pancreatitis (odds ratio for pancreatitis with ERCP and SOM compared with SOM only, 3.26; P = 0.05). All patients who had pancreatitis had biliary type SOD. The average days in the hospital after pancreatitis ranged from 2 to 4 days (mean, 2.75 days), and all patients had mild pancreatitis. The time betwee n SOM and subsequent ERCP or sphineterotomy, if performed, ranged from 6 to 50 days (mean, 10.4 days).

Other than pancreatitis, no other significant complications occurred, after SOM, ERCP, or sphineterotomy. Specifically, there were no adverse drug reactions, biliary or pancreatic sepsis, perforations, or episodes of bleeding. No pulmonary or cardiovascular events were noted.

Discussion

Pancreatitis is a well-recognized complication after SOM, ranging in incidence from 4 to 31 %. (4,5) Different methods have been proposed to decrease the incidence of pancreatitis. This includes aspiration catheters, which may decrease the incidence of pancreatitis to 4% (11); however, one recording port is sacrificed. Pancreatic stenting after biliary sphineterotomy in patients with pancreatic SOD may decrease the incidence of pancreatitis to 7% (12) Gravity draining of the pancreatic duct has also been suggested to decrease the incidence of pancreatitis, as is limiting the time of pancreatic duct manometry to less than 2 minutes. Decreasing the perfusion rate to 0.1 mi/channel/mm may also decrease the incidence of pancreatitis. This may not affect basal sphincter pressure measurements; however, phasic wave pressure measurements will not be accurate. The use of a microtransducer, which is a nonperfusion system, may decrease the incidence of pancreatitis to as low as 2%. (13)

In our study, the incidence of pancreatitis was 7.32% after SOM and 9.78% with combined SOM and subsequent ERCP and sphincterotomy. Although the incidence of pancreatitis using the aspiration catheter was suggested to be as low as 4% in one study, (2) it was reported to be 8% in a second study, (14) which is not significantly different from our technique.

Usually, ERCP is performed just before SOM at the same session and sphincterotomy is then performed if SOM shows high pressures consistent with SOD. This practice has been shown to be associated with an increased incidence of pancreatitis. (7) Separating SOM from diagnostic ERCP and sphincterotomy reduces overall procedure time, numbers of cannulations, and amount of fluid (contrast and water) injected into the pancreatic ductal system. This may reduce trauma to the pancreatic duct and injury resulting from acinar filling with contrast or water. Also, performing ERGP and sphincterotomy at a different session allows time for the pancreatic duct to recuperate from any subclinical injury caused by the initial SOM.

By adopting a protocol to perform SOM only and to perform ERCP with sphincterotomy on selected patients who have abnormal SOD, we were able to decrease significantly the incidence of pancreatitis to 9.78%. Patients who have SOD will undergo a second procedure for sphincterotomy, which will increase the total cost of the treatment. However, by decreasing the incidence of pancreatitis and avoiding the cost of hospital stay, the increased cost of two procedures is more than offset. On the basis of our findings, we recommend performing SOM as a single procedure temporally separated from diagnostic ERCP and sphincterotomy.

Table 1

Summary of results (a)

Parameter              Value

Total no. of patients  41
Age                    14-77 yr (mean age, 40.9 yr)
Sex                    40 F, 1M
Type of SOD            Biliary 34, pancreatic 7
Days between SOM       6-20 d (mean interval, 10.4 d)
  and ERCP
Pancreatitis           4 (9.78%; 3 after SOM and 1 after ERCP)
Days in hospital       2-4 d (mean hospital stay, 2.75 d)
  after pancreatitis

(a)SOD, sphincter of Oddi dysfunction; SOM, sphincter of Oddi manometry;
ERCP, endoscopic retrograde cholangiopancreatography.

Accepted September 19, 2002.

References

(1.) Tanaka M, Ikeda S. Sphincter of Oddi manometry: Comparison of microtransducer and perfusion methods. Endoscopy 1988;20:184-188.

(2.) Hogan WJ, Sherman S, Pasricha P, et al. Position paper: Sphincter of Oddi manometry. Gastrointest Endosc 1997;45:342-348.

(3.) Geenen JE, Hogan WJ, Dodds WJ, et al. The efficacy of endoscopic sphincterotomy after cholecystectomy in patients with sphincter of Oddi dysfunction. N Engl J. Med 1989;320:82-87.

(4.) Rolny P, Anderberg B, Ishe I, et al. Pancreatitis after sphincter of Oddi manometry. Gut 1990;31:821-824.

(5.) Albert MB, Steinberg WM, Irani SK. Severe acute pancreatitis complicating sphincter of Oddi manometry. Gastrointest Endosc I 988;34:342-345.

(6.) LaFerla G, Gordan S, Archibald M, et al. Hyperamylasemia and acute pancreatitis following endoscopic retrograde cholangiopancreatography. Pancreas 1986;1:160-163.

(7.) Maldonado ME, Brady PG, Mamcl J, et al. Incidence of pancreatitis in patients undergoing sphincter of Oddi manometry. Am J Gastroenterol 1999;94:387-390.

(8.) Geenen JE, Hogan WJ, Dodds WJ, et al. Intraluminal pressure recording from the human sphineter of Oddi. Gastroenterology 1980;78:3 17-324.

(9.) Sherman S, Troiano FP, Hawes RH, et al. Frequency of abnormal sphincter of Oddi manometry compared with the clinical suspicion of sphincter of Oddi dysfunction. Am J Gastroenterol 1991;86:586-590.

(10.) Cotton PB, Lehman G, Vennes J, et al. Endoscopic sphincterotomy complications and their management: An attempt at consensus. Gastrointest Endosc 1991;37:383-393.

(11.) Sherman S, Hawes RH, Troiano FP, et al. Pancreatitis following bile duct sphincter of Oddi manometry: Utility of the aspirating catheter. Gastrointest Endosc 1992;38:347-350.

(12.) Tarnasky PR, Palesch YY, Cunningham JT, et al. Pancreatic stenting prevents pancreatitis after biliary sphincterotomy in patients with sphincter of Oddi dysfunction. Gastroenterology 1998; 115:1518-1524.

(13.) Wehrmann T, Schmitt T, Schonfeld A, et al. Endoscopic sphincter of Oddi manometry with a portable electronic microtransducer system: Comparison with the perfusion manometry method and routine clinical application. Endoscopy 2000;32:444-451.

(14.) Walters DA, Geenen JE, Catalano MF, et al. A randomized controlled trial comparing the use of an aspirating catheter vs a standard perfused catheter on the incidence of pancreatitis following sphincter of Oddi manometry. Gastrointest Endosc 1997;45:AB152 (abstr).

RELATED ARTICLE: Key Points

* Sphincter of Oddi manometry (SOM) is the standard technique for evaluating sphincter of Oddi dysfunction.

* Acute pancreatitis is a common complication of SOM.

* By performing SOM separate from ERCP and sphincterotomy, the incidence of pancreatitis was considerably decreased at our institutions.

From the Division of Digestive Diseases, Department of Internal Medicine, University of South Florida, Tampa, FL.

Reprint requests to Syed T. Bin-Sagheer, MD, Bassett HealthCare, 1 Atwell Road, Cooperstown, NY 13326.

Copyright [C] 2003 by The Southern Medical Association

0038-4348/03/9603-0223