2014-09-16 18:59:33 UTC

Sphincter of Oddi Dysfunction: Still Alive?

Sept. 16, 2014


Peter B. Cotton, MD, FRCS, FRCP

Professor of Medicine, Digestive Disease Center, Medical University of South Carolina

Like many long-time ERCPists and sphincterotomizers, I was attracted to the concept of sphincter of Oddi dysfunction (SOD) as a cause of biliary pain, and, for many years, was convinced that my treatments were helpful in many if not most cases. This despite an early comment from a respected mentor, Professor Solly Marks (grandfather of South African gastroenterology), who kindly remarked after my eloquent lecture on the subject: “Peter, that sounds like sewing flatus to moonbeams.”

Probably like many other “experts” seeing patients referred from considerable distances, I did not keep careful track of their outcomes, and counted failure subconsciously only by those few patients brave enough to complain or return. One well-known center suggested a success rate of 75 percent simply because only 25 percent underwent re-intervention at the same center;1 what about all those who went elsewhere or battled on in pain?

Sphincter dysfunction is traditionally divided into three types.2 Type I includes patients with biliary pain, a dilated bile duct and abnormal liver function tests. Type II patients have a dilated bile duct or abnormal labs, but not both. Type III patients have none of those criteria. Type I patients mostly have an organic cause (sphincter stenosis or small stones) and seem to be well-treated by endoscopic sphincterotomy. Nowadays those patients can be best diagnosed by endoscopic ultrasound.

Over the last two decades, type II and III patients have been referred increasingly to approximately 25 centers in the U.S. performing sphincter manometry, because the “experts” continued to espouse its value and community physicians believed it to be dangerous. Manometry itself is actually not dangerous, but ERCP is indeed so in these patients. As a result of this trend, my own clinic and ERCP practice became dominated by complex and often desperate patients with “suspected SOD,” expecting to be cured quickly with a simple cut. My colleagues and I began to wonder if we were indeed tilting at moonbeams, and decided to explore the problem objectively. It has taken almost 10 years. A lot of preparatory work and a planning grant led to the EPISOD research study funded by NIDDK, the results of which were published recently in JAMA.3

SOD, at least type III, dissolves like moonbeams under the searchlight of science.

We took patients aged 18 to 65 with burdensome post-cholecystectomy biliary-type pain, no evidence for pancreatitis and no prior sphincter intervention. We excluded patients with bile ducts larger than 9mm in diameter, those taking narcotics daily and those with marked psychological problems. They were essentially “SOD type III” except that we did allow some minor elevations of liver and pancreas lab tests. After consent, application of multiple questionnaires and some other resulting exclusions, 214 underwent ERCP with manometry (of both sphincters) and were randomized (2:1), irrespective of the manometry results, to sphincterotomy or to sham. Those randomized to sphincterotomy and who had elevated pancreatic sphincter pressures (PSH) were randomized again (1:1) to biliary sphincterotomy or to both biliary and pancreatic sphincterotomy. All patients, including those treated with sham, received a temporary small pancreatic stent. Patients, caregivers and research staff were blinded to the treatment allocation during the follow-up of one year. Success was defined as a marked reduction of the pain score as measured at nine and 12 months, without any sphincter re-intervention and no narcotics in the last three months.

The results were striking and definitive. The success rate for sham treated patients was 37 percent, but only 23 percent in the sphincterotomy arm. In those with PSH, 30 percent responded to dual sphincterotomy, and 19 percent to biliary alone. Re-interventions occurred in 26 percent of treated patients and 34 percent of the controls. Post-procedure pancreatitis occurred after 11 percent of primary sphincterotomies and 15 percent of controls. There were two perforations, one requiring surgery, no bleeds and no deaths.

It is noteworthy that three-quarters of the patients had abnormal manometry (64 percent pancreatic, with or without biliary, and 12 percent biliary alone). The results of sphincter manometry did not correlate with the outcomes. Furthermore, we could not identify any clinical features that made success more or less likely (e.g., presence or absence of daily pain, lab abnormalities, reasons for the cholecystectomy and response to it, presence of other functional digestive disorders, or psychological factors).

Seventy-two patients eligible for the study but who declined randomization entered into an observational study (EPISOD 2) where sphincterotomy was directed by the manometry results, as in current standard practice. The success rates were equally unimpressive, 31 percent for dual sphincterotomy, 24 percent biliary and 17 percent for none. We conclude that most if not all of the ”successes” in the study were due to a powerful placebo effect in very distressed patients coming to known “experts” and supported for a year by contact with research staff.

It appears that Solly Marks was right. SOD, at least type III, dissolves like moonbeams under the searchlight of science. Where does that leave the overall concept of SOD? The justification for doing manometry-directed sphincterotomy in type II patients is based on many unblinded cohort studies with results that are not totally convincing,2 and from three small and old randomized trials.4–6 Pooling the data from the randomized studies (no doubt not an appropriate thing to do) showed that 78 percent (29/37) of the patients with post-cholecystectomy pain and abnormal biliary manometry were relieved by biliary sphincterotomy. While that figure is impressive, it is noteworthy that success was reported also after biliary sphincterotomy in a third (11/35) of patients with normal manometry, and in no fewer than half (13/25) of those with elevated pressures and no sphincterotomy. What is also a little odd is the high rate of success when the studies focused exclusively on the biliary sphincter. Yet we know now that pancreatic sphincter hypertension is just as common in these patients. Despite those reservations, these studies gained a lot of traction and have bolstered the field effectively for years.

If sphincterotomy does indeed help some type II patients, we certainly need more research into optimal methods for identifying them. Which clinical factors are really predictive? Is dynamic biliary scanning really worthwhile? Is a Botox trial helpful?

Since SOD type III does not exist, and many type I patients have an organic explanation that can be revealed by EUS, it is time to ditch the old I, II, III classification. We are left with one problem syndrome called “suspected SOD,” whose complexities and components will be revealed gradually by further research.

SOD is also postulated as a cause of recurrent acute pancreatitis. The recent report that dual sphincterotomy is no more effective than biliary sphincterotomy alone also raises doubts about the relevance of SOD in that context.7

This is indeed the end of an era, but also, hopefully, the beginning of a new one. It is time for a comprehensive reappraisal of “functional” biliary (and pancreatic) disorders. The clinical problems are important and need our best attention. There are plenty of very genuine patients with burdensome biliary pain (before and after cholecystectomy). It is time to delve further into the mechanisms of pain and to improve our diagnostic and therapeutic approaches. We need to reduce the number of unnecessary (and potentially dangerous) interventions.

Dr. Cotton is a consultant for Olympus and receives royalties from Cook. He is also a board member for Apollo Endosurgery and SF Health Care Quality Consulting.


1. Park SH, Watkins JL, Fogel EL, Sherman S, Lazzell L, Bucksot L, et al. Long-term outcome of endoscopic dual pancreatobiliary sphincterotomy in patients with manometry-documented sphincter of Oddi dysfunction and normal pancreatogram. Gastrointest Endosc 2003;57:483–491

2. Petersen BT. An evidence-based review of sphincter of Oddi dysfunction: part 1, presentations with “objective” biliary findings (types I and II). Gastrointest Endosc 2004;59:525-34.

3. Cotton PB, Durkalski V, Romagnuolo j, et al. Effect of endoscopic sphincterotomy for suspected sphincter of Oddi dysfunction on pain-related disability following cholecystectomy – the EPISOD randomized clinical trial. JAMA 2014; 311 (20).

4. Sherman S, Lehman G, Jamidar P, et al. Efficacy of endoscopic sphincterotomy and surgical sphincteroplasty for patients with sphincter of Oddi dysfunction (SOD); randomized controlled study. Gastrointest Endosc 1994; 40: P125

5. 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;320:82-7.

6. Toouli J, Roberts-Thomson IC, Kellow J, et al. Manometry based randomized trial of endoscopic sphincterotomy for sphincter of Oddi dysfunction. Gut 2000;46:98-102.

7. Cote G, Imperiale TF, Schmidt SE et al. Similar Efficacies of Biliary, With or Without Pancreatic, Sphincterotomy in Treatment of Idiopathic Recurrent Acute Pancreatitis. Gastroenterology 2012; 143: 1502-1509

More on ERCP

AGA Supports New Government Scope Safety Protocols

Feb. 28, 2018

While voluntary, AGA encourages GIs to review and consider adopting sampling and culturing of reprocessed duodenoscopes.

Attend the AGA Postgraduate Course for Practical, Take Home Information

Feb. 12, 2018

Join us on June 2-3 in Washington, D.C. This course is held in conjunction with Digestive Disease Week® (DDW) 2018.

Voluntary Recall of Pentax ED-3490TK Duodenoscopes

Feb. 8, 2018

Recall designed to reduce the potential for leakage of patient fluids into the closed elevator channel and under the distal cap.