Sphincter of Oddi Dysfunction

Emerg Med 37(3):14-17, 2005

The author discusses clinical manifestations of functional problems with the biliary and pancreatic sphincters and the specialized techniques used to treat them.

By John Baillie, MB, ChB, FRCP

The sphincter of Oddi is a valve comprised of circular muscle that controls the opening and closing of the bile duct within the duodenal papilla (ampulla of Vater). Its resting state is contracted—that is, closed. When neural or humoral signals cause it to relax, bile is released into the duodenum to assist in the digestion of dietary fat. There is a separate but less well-defined sphincter controlling the release of pancreatic juice. Typically, ablation of the sphincter of Oddi by sphincterotomy does not destroy the pancreatic sphincter.

Sphincter of Oddi dysfunction encompasses various conditions in which the biliary or pancreatic sphincter is considered to be malfunctioning. This article will focus principally on biliary sphincter dysfunction.
 

What is sphincter of Oddi dysfunction (SOD)?

The term sphincter of Oddi dysfunction was first used in a paper published in the New England Journal of Medicine in 1989 by Joseph Geenen, Walter Hogan, and colleagues. Because their work was done in part in Milwaukee, Wisconsin, their proposed classification of SOD became known as the Milwaukee classification. Geenen and Hogan were trying to make sense of the underlying pathology in a heterogeneous group of patients with biliary-type pain following cholecystectomy. They developed their classification system based on the following criteria: typical biliary pain; liver function test (LFT) abnormalities (specifically, elevations in serum transaminase levels of 1.5 times normal or higher on two occasions related to pain, with resolution between attacks); dilatation of the common bile duct of 12 mm or more; and delayed drainage of contrast medium from the biliary tree. Over time, the last criterion was repeatedly shown to be an unreliable sign of SOD and was dropped.

Using these criteria, Geenen, Hogan, and colleagues divided SOD patients into three types (see table below). Type I patients exhibit all three criteria. They constitute a subgroup of post-cholecystectomy patients with a mechanical outlet obstruction to biliary flow (papillary stenosis). About 95% of type I patients are cured by endoscopic biliary sphincterotomy (EBS). Biliary pressure measurement does not affect the outcome and is not indicated in these patients.

Type II patients have typical biliary pain and either abnormal liver serology or dilatation of the bile duct, but not both. Type II patients with abnormal sphincter of Oddi manometry have an 80% or better chance of improvement following EBS. Those whose pressures are normal benefit only 30% of the time. For this reason, it is strongly recommended that suspected type II patients undergo biliary manometry.

Type III patients have typical biliary pain alone. These patients have the least likelihood of benefiting from EBS. In most centers, the improvement rate is in the range of 30% to 40% if sphincter of Oddi manometry is positive, but less than 20% if it is negative. These patients also have the highest complication rate from endoscopic retrograde cholangiopancreatography (ERCP). Type III patients should never have EBS for suspected SOD without prior manometry showing elevated sphincter of Oddi pressures.
 

Does this mean that EBS should not be performed empirically for biliary pain?

Exactly. As a risk management strategy, endoscopists would do well to steer clear of type III SOD patients. They have the highest risk of post-ERCP pancreatitis and the lowest yield in terms of long-term benefit from intervention. The guidelines of the American Society for Gastrointestinal Endoscopy have clearly stated for more than a decade that ERCP "is generally not indicated" in the investigation of pain syndromes without other evidence of biliary obstruction. With the literature now quite clear about the significant risks and limited benefits of ERCP in this setting, it is difficult to defend oneself medicolegally against severe pancreatitis complicating such procedures.

In type III patients, the extrahepatic bile duct is not dilated, making cannulation difficult. Prolonged unsuccessful instrumentation of the duodenal papilla in an attempt to obtain a cholangiogram makes pancreatitis likely. Sometimes damage is done with a needle-knife catheter, which is used to make a blind cut into the duodenal papilla to expose the bile duct. Needle-knife papillotomy is a dangerous procedure in this setting and should be avoided.
 

What is sphincter of Oddi manometry and where can patients have it done?

Sphincter of Oddi manometry is a specialized technique for measuring biliary and sometimes pancreatic sphincter pressure. Given the cost of the equipment and the technical difficulty of performing the procedure, it is typically limited to tertiary referral centers where hepatobiliary and pancreatic disorders are managed. At my institution, we use general anesthesia in about one-third of these procedures. Because many SOD patients are habituated to narcotic analgesics and benzodiazepines, they can be very difficult to sedate using standard techniques.

In the early days, sphincter of Oddi manometry was performed using a water-perfusion pump system. Most centers now use solid-state pressure transducers that are read and interpreted by specialized computer software. Details regarding the test results are beyond the scope of this article, but we look for sustained elevations (lasting 20 to 30 seconds or more) in sphincter pressure in multiple leads of more than 40 mm Hg. Certain drugs, such as morphine and anticholinergics, must be avoided during ERCP for sphincter of Oddi manometry because they can affect the pressure readings.

We do not offer direct-to-procedure access for sphincter of Oddi manometry. Suspected SOD patients need to be evaluated in a specialist clinic first because many of them do not fit the criteria for this diagnosis. At my institution, only about a third of patients referred as having type III SOD end up being offered ERCP with sphincter of Oddi manometry. The remainder are thought to have some variant of chronic functional abdominal pain, gastroesophageal reflux disease, chest wall pain syndrome (from costochondritis or fibromyalgia, for example), or orthopedic pain.

Referring physicians should be careful to avoid promising suspected SOD patients that certain procedures will be done. These patients typically arrive with unrealistic expectations of what can be done for them and are often angry and disappointed when told that they do not fit the criteria for SOD.
 

Outside the United States, SOD is considered an American disease. Sphincter of Oddi manometry is rarely offered in Europe, Asia, or Africa. Is SOD really just a myth?

That is a difficult question to answer. Type I SOD (papillary stenosis) certainly exists, and it is the most satisfying to treat. We never see it in referral centers because it can be cured in the hospital or clinic setting with EBS. Most patients diagnosed as having type III SOD do not have a demonstrable sphincter abnormality. Unfortunately, they comprise 80% to 90% of the patients referred for work-up. Many of them are anxious, depressed, frustrated, and angry. They are best assessed in a specialist clinic by physicians and other staff who have experience in investigating and managing this condition.

In my experience, type II patients present the greatest challenge. When one returns to the original Milwaukee criteria, it is clear that the LFT abnormalities have to normalize between attacks and that bile duct dilatation is defined as 12 mm or more. It is not uncommon, however, to see patients referred as having possible type II SOD who have mild elevations in LFTs (including alkaline phosphatase levels) that never normalize or enlargement of the bile duct that is less than 12 mm. (The upper limit of normal is generally considered to be 7 mm.)

Patients with persistently abnormal LFTs are usually obese and often have hepatic steatosis (fatty liver) on biopsy. Chronic drug-related LFT elevations are another source of confusion. It is often hard to interpret mild dilatation of the extrahepatic bile duct. On its own, this finding has little clinical significance. The duct could contain a stone or perhaps a small distal tumor that is causing the enlargement. If there is sufficient cause for concern, endoscopic ultrasound is a noninvasive way to explore this possibility. Magnetic resonance cholangiography used to be relatively insensitive for detecting small biliary stones, but this is no longer the case. For many patients in this setting, using another form of imaging to avoid ERCP makes sense.

Undoubtedly, some patients with idiopathic acute recurrent pancreatitis have a hypertensive pancreatic sphincter. For these individuals, pancreatic sphincterotomy may cure the problem. Pancreatic sphincter dysfunction is a rare diagnosis that requires specialist investigation and management.
 

Sphincterotomy seems like a blunt tool to use to manage a fairly subtle abnormality. What progress has been made in the pharmacologic management of SOD?

Since the late 1970s, when Staritz in Germany claimed success using topical nitrates, endoscopists have been looking for a local or systemic agent to relax the sphincter of Oddi. So far, the results have been disappointing. There is some rationale for using nitrates, since nitric oxide is an important neurotransmitter regulating sphincter of Oddi tone. Calcium channel blockers, such as nifedipine, have also been tried, with limited success.

Kalloo and others have tried inhibiting sphincter of Oddi contraction using a local injection of botulinum toxin into the duodenal papilla. The initial data were encouraging, but these results have not been confirmed by larger studies. It now appears that such an injection may provoke a chronic inflammatory response that could actually cause papillary stenosis over time. Other pharmacologic agents, including neurotoxic venoms, are currently being evaluated for treating SOD.
 

What about balloon dilation of the sphincter? Is this worth a trial in SOD?

The U.S. multicenter trial of endoscopic sphincterotomy versus EBS for management of bile duct stones, recently published in Gastroenterology, showed that EBS carries a very significant risk of procedure-related pancreatitis. Most experts believe that the risk of EBS is even greater in SOD patients, who already have a sensitive sphincter. In my opinion, SOD patients should never be managed with EBS or a trial of biliary stenting.
 

Is there any way to reduce the risk of post-ERCP pancreatitis in SOD patients?

Yes. There are good data now available that prophylactic stenting of the pancreatic duct orifice, using small-caliber, unflanged, single-pigtail stents, greatly reduces the risk of post-ERCP pancreatitis and appears to virtually abolish the risk of severe, necrotizing pancreatitis. Whenever possible, we place a 6- to 8-cm long, #3 French pancreatic stent when performing sphincter of Oddi manometry in potential SOD patients, regardless of whether sphincterotomy is performed. The majority of these stents spontaneously migrate out of the pancreas within days after the procedure. We also recommend prophylactic stenting of the pancreatic duct whenever there has been repeated or otherwise traumatic instrumentation of the papilla during ERCP.
 

What progress has been made toward a noninvasive radiologic test for SOD?

This is the holy grail of ERCP. Most endoscopists who perform sphincter of Oddi manometry would be delighted never to have to do another one. Unfortunately, no noninvasive test has clearly emerged as a sensitive and specific alternative. A number of studies of so-called gated radionuclide biliary scans have claimed to show good correlation with sphincter of Oddi manometry and the results of sphincterotomy. However, none has withstood the test of time. Magnetic resonance cholangiography-based dynamic testing is currently being evaluated. At present, ERCP with sphincter of Oddi manometry remains the gold standard for diagnosing SOD.

back to top

Printable version of this article

E-mail this article