Introduction

Since its original description by Oddi in 1887, the sphincter of Oddi has been the subject of much study and controversy. Its very existence as a distinct anatomic or physiologic entity has been disputed. Not surprisingly, the clinical syndrome of sphincter of Oddi dysfunction (SOD) and its therapy are controversial areas.^1,2 Nevertheless, SOD is commonly diagnosed and treated by physicians. This chapter reviews the epidemiology and clinical presentation of SOD and currently available diagnostic and therapeutic modalities.

Definitions

Postcholecystectomy pain resembling the patient’s preoperative biliary colic occurs in at least 10% to 20% of patients.³ These patients should have appropriate noninvasive and invasive (when clinically appropriate) evaluation to rule out common bile duct (CBD) stones, tumors, or strictures near the cholecystectomy site. Patients in whom these entities are ruled out have a high frequency of SOD. SOD refers to an abnormality of sphincter of Oddi contractility. It is a benign, noncalculous obstruction to flow of bile or pancreatic juice through the pancreaticobiliary junction (i.e., the sphincter of Oddi) resulting from a dyskinetic or stenotic sphincter of Oddi. SOD may be manifested clinically by “pancreaticobiliary” pain, pancreatitis, abnormal liver function tests, or abnormal pancreatic enzymes. Sphincter of Oddi dyskinesia refers to a motor abnormality of the sphincter of Oddi, which may result in a hypotonic sphincter but, more commonly, causes a hypertonic sphincter. In contrast, sphincter of Oddi stenosis refers to a structural alteration of the sphincter, probably from an inflammatory process, with subsequent fibrosis.

Because it is often impossible to distinguish patients with sphincter of Oddi dyskinesia from patients with sphincter of Oddi stenosis, the term SOD has been used to incorporate both groups of patients. In an attempt to deal with this overlap in etiology and to determine the appropriate use of sphincter of Oddi manometry (SOM), a biliary clinical classification system has been developed for patients with suspected SOD (Hogan-Geenen SOD classification system; Table 46.1) based on clinical history, laboratory results, and endoscopic retrograde cholangiopancreatography (ERCP) findings.⁴ A pancreatic classification has also been developed, but it is less commonly used (Box 46.1).⁵ Both the biliary and the pancreatic classification systems have been modified,⁶ making them more applicable for clinical use because biliary and pancreatic drainage times have been generally abandoned. Various less accurate terms—papillary stenosis, ampullary stenosis, biliary dyskinesia, and postcholecystectomy syndrome—are used in the medical literature to describe this entity. The last term, postcholecystectomy syndrome, is a misnomer because SOD may occur with an intact gallbladder.

Anatomy, Physiology, and Pathophysiology

The anatomy, physiology, and pharmacology of the sphincter of Oddi have been reviewed by Bosch and Pena.⁷ The sphincter of Oddi is a small complex of smooth muscles surrounding the terminal CBD, main (ventral) pancreatic duct (of Wirsung), and common channel (ampulla of Vater), when present (Fig. 46.1). It has both circular and figure-eight components. The high-pressure zone generated by the sphincter is variably 4 to 10 mm in length. Its role is to regulate bile and pancreatic exocrine juice flow and to prevent duodenum-to-duct reflux (i.e., maintain a sterile intraductal environment). The sphincter of Oddi possesses both a variable basal pressure and phasic contractile activity. The former seems to be the predominant mechanism, regulating outflow of pancreaticobiliary secretion into the intestine. Although phasic sphincter of Oddi contractions may aid in regulating bile and pancreatic juice flow, their primary role seems to be maintaining a sterile intraductal milieu.

Fig. 46.1 Anatomy of sphincter of Oddi.

Sphincter regulation is under neural and hormonal control. Phasic wave activity of the sphincter is closely tied to the migrating motor complex of the duodenum. Innervation of the bile duct does not seem to be essential because sphincter function has been reported to be preserved after liver transplantation.⁸ Although regulatory processes vary among species, cholecystokinin (CCK) and secretin seem to be most important in causing sphincter relaxation, whereas nonadrenergic, noncholinergic neurons, which at least partially transmit vasoactive intestinal peptide and nitric oxide, also relax the sphincter.⁹ The role of cholecystectomy in altering these neural pathways needs further definition.

Luman and colleagues¹⁰ reported that cholecystectomy, at least in the short-term, suppresses the normal inhibitory effect of pharmacologic doses of CCK on the sphincter of Oddi. The mechanism of this effect is unknown, however. Wedge specimens of the sphincter of Oddi obtained at surgical sphincteroplasty from patients with SOD show evidence of inflammation, muscular hypertrophy, fibrosis, or adenomyosis within the papillary zone in approximately 60% of patients.¹¹ In the remaining 40% with normal histology, a motor disorder is suggested. Less commonly, infections with cytomegalovirus or Cryptosporidium, as may occur in patients with acquired immunodeficiency syndrome (AIDS), or Strongyloides have caused SOD.

How does SOD cause pain? From a theoretical point of view, abnormalities of sphincter of Oddi pressure can give rise to pain by (1) impeding the flow of bile and pancreatic juice resulting in ductal hypertension, (2) inducing ischemia arising from spastic contractions, and (3) resulting in “hypersensitivity” of the papilla. Although unproved, these mechanisms may act alone or in concert to explain the genesis of pain. Patients with SOD have been shown to have lower perception thresholds in the referred pain area. Visceral and referred hyperalgesia may be important features in the pathogenesis of pain in patients with SOD.¹²

Epidemiology

SOD may occur in children or adults of any age; however, patients with SOD are typically middle-aged women.^13,14 Although SOD most commonly occurs after cholecystectomy, it may be present with the gallbladder in situ.¹⁵ In a survey on functional gastrointestinal (GI) disorders, SOD seemed to have a significant impact on quality of life because it was highly associated with work absenteeism, disability, and health care use.¹⁶ Using a brief symptom inventory and the 12-item short form health survey (SF-12), Winstead and Wilcox¹⁷ found that patients with biliary SOD and unexplained recurrent pancreatitis had a significantly worse quality of life than nonpatients, high levels of somatic complaints, and a common history of sexual and physical abuse (20%).

Limited studies of the frequency of manometrically documented SOD in patients before cholecystectomy have been done. Guelrud and colleagues¹⁸ evaluated 121 patients with symptomatic gallstones and a normal CBD diameter (by transcutaneous ultrasound) by SOM before cholecystectomy. An elevated basal sphincter pressure was found in 14 (11.6%) patients. SOD was diagnosed in 4.1% (4 of 96) of patients with a normal serum alkaline phosphatase and in 40% (10 of 25) with an elevated serum alkaline phosphatase. Ruffolo and associates¹⁹ evaluated 81 patients with symptoms suggestive of biliary disease but normal ERCP and no gallbladder stones on transcutaneous ultrasound by scintigraphic gallbladder ejection fraction and endoscopic SOM. Of patients, 53% had SOD, and 49% had an abnormal gallbladder ejection fraction. SOD occurred with a similar frequency in patients with an abnormal gallbladder ejection fraction (50%) and a normal ejection fraction (57%).

The frequency of diagnosing SOD varies considerably in reported series with the patient selection criteria, the definition of SOD, and the diagnostic tools used. In a British report, SOD was diagnosed in 41 (9%) of 451 consecutive patients being evaluated for postcholecystectomy pain.²⁰ Roberts-Thomson and Toouli²¹ evaluated 431 similar patients and found SOD in 47 (11%). In a subpopulation of patients with normal ERCP (except dilated ducts in 28%) and recurrent pain of more than 3 months’ duration, SOD was diagnosed in 68%. Sherman and colleagues⁵ used SOM to evaluate 115 patients with pancreaticobiliary pain with and without liver function test abnormalities. Patients with bile duct stones and tumors were excluded from analysis. Of 115 patients, 59 (51%) showed abnormal basal sphincter of Oddi pressure greater than 40 mm Hg. These patients were categorized further by the Hogan-Geenen SOD classification system based on clinical presentation, laboratory results, imaging tests, and ERCP findings (see Table 46.1). The frequency of abnormal manometry of one or both sphincter segments was 86%, 55%, and 28% for biliary type I, II, and III patients. These abnormal manometric frequencies were very similar to the frequencies reported by others for type I and type II patients.^22,23 In biliary type III patients, the finding of an abnormal basal sphincter pressure has ranged from 12% to 59%.^6,24 Patient selection factors may be one explanation for this great variability.

SOD can involve abnormalities in the biliary sphincter, pancreatic sphincter, or both.^6,25 The true frequency of SOD depends on whether one or both sphincters are studied. Eversman and colleagues⁶ performed manometry of the biliary and pancreatic sphincter segments in 360 patients with pancreatobiliary pain and intact sphincters. In this large series, 19% had abnormal pancreatic sphincter basal sphincter pressure alone, 11% had abnormal biliary basal sphincter pressure alone, and 31% had abnormal basal sphincter pressure in both segments (overall frequency of sphincter dysfunction was 61%). Among the 214 patients labeled type III by a modified Hogan-Geenen SOD classification system, 17%, 11%, and 31% had elevated basal sphincter pressure in the pancreatic sphincter alone, biliary sphincter alone, or both segments (overall frequency of SOD was 59%). In 123 type II patients, SOD was diagnosed in 65%; 22%, 11%, and 32% had elevated basal sphincter pressure in the pancreatic sphincter only, biliary sphincter only, or both sphincter segments. Similar findings were reported by Aymerich and colleagues.²⁶ In a series of 73 patients with suspected SOD, basal pressures were normal in both segments in 19%, abnormal in both segments in 40%, and abnormal in one segment but normal in the other segment in 41%. The negative predictive value of normal biliary basal sphincter pressure in excluding SOD was 0.42; when the pancreatic basal sphincter pressure was normal, the negative predictive value was 0.58. These two studies suggest that both the bile duct and the pancreatic duct should be evaluated when assessing the sphincter by SOM.

Although SOM has traditionally been thought to be reproducible,²⁷ two more recent studies have shown abnormal sphincter pressures in 42% and 60% in symptomatic patients restudied about 1 year after a normal study.^28,29 Dysfunction may occur in the pancreatic duct portion of the sphincter of Oddi and cause recurrent pancreatitis. As noted earlier, a pancreatic SOD classification system has been developed (see Box 46.1), but it has not been widely used.^5,6 Manometrically documented SOD has been reported in 15% to 72% of patients with recurrent pancreatitis, previously labeled as idiopathic^5,24,30; this is discussed later in this chapter.

Clinical Presentation

Abdominal pain is the most common presenting symptom of patients with SOD. The pain is usually epigastric or right upper quadrant, may be disabling, and lasts for 30 minutes to hours. In some patients, the pain is continuous with episodic exacerbations. It may radiate to the back or shoulder and be accompanied by nausea and vomiting. Food or narcotics may precipitate the pain. The pain may begin several years after a cholecystectomy was performed for a gallbladder dysmotility or stone disease and is similar in character to the pain leading to the cholecystectomy. Alternatively, patients may have continued pain that was not relieved by a cholecystectomy. Jaundice, fever, and chills are rarely observed.

A symposium on functional disorders of the pancreas and biliary tree established the Rome III diagnostic criteria³¹ for SOD. These criteria include episodes of severe abdominal pain located in the epigastrium or right upper quadrant or both and all of the following: (1) symptom episodes last 30 minutes or more with pain-free intervals, (2) recurrent symptoms occur at different intervals (not daily), (3) the pain builds up to a steady level, (4) the pain is moderate to severe enough to interrupt the patient’s daily activities or lead to an emergency department visit, (5) the pain is not relieved by bowel movements, (6) the pain is not relieved by postural change, (7) the pain is not relieved by antacids, (8) other structural diseases that would explain the symptoms are excluded. The pain may manifest in one or more of the following ways: pain with nausea and vomiting, pain radiation to the back or right subscapular region or both, and pain awakens patient from sleep.

Physical examination is typically characterized only by mild epigastric or right upper quadrant tenderness. The pain is not relieved by trial medications for acid peptic disease or irritable bowel syndrome. Laboratory abnormalities consisting of transient elevation of liver function tests, typically during episodes of pain, are present in less than 50% of patients. Patients with SOD may present with typical pancreatic pain (epigastric or left upper quadrant radiating to the back) with or without pancreatic enzyme elevation and recurrent pancreatitis. The pain is often indistinguishable from biliary pain.³¹ SOD may exist in the presence of an intact gallbladder.^18,19,32 Because the symptoms of SOD or gallbladder dysfunction cannot be reliably separated, the diagnosis of SOD is commonly made after cholecystectomy or less often after gallbladder abnormalities have been excluded.³¹

Clinical Evaluation

The diagnostic approach to suspected SOD may be influenced by the presence of key clinical features. However, the clinical manifestations of functional abnormalities of the sphincter of Oddi may not always be easily distinguishable from manifestations caused by organic ones (e.g., CBD stones) or other functional nonpancreaticobiliary disorders (e.g., irritable bowel syndrome).

Diagnostic Methods (Invasive)

Endoscopic Retrograde Cholangiopancreatography

Because of their associated risks, invasive testing with ERCP and manometry should be reserved for patients with clinically significant or disabling symptoms. Invasive assessment of patients for SOD generally is not recommended unless definitive therapy (sphincter ablation) is planned if abnormal sphincter function is found. Cholangiography is essential to rule out stones, tumors, or other obstructing processes of the biliary tree that may cause symptoms identical to SOD. Once such lesions are ruled out by a good-quality cholangiographic study, ducts that are dilated or drain slowly suggest obstruction at the level of the sphincter. Various methods to obtain a cholangiogram are available. For noninvasive imaging, MR cholangiography is most promising, but quality varies greatly from center to center. Software development continues, and quality of images continues to evolve and improve. Direct cholangiography can be obtained by percutaneous methods, intraoperative methods, or more conventionally at ERCP.

Although some controversy exists, extrahepatic ducts that are greater than 12 mm in diameter (postcholecystectomy) when corrected for magnification are considered dilated. Drugs that affect the rate of bile flow and relaxation or contraction of the sphincter of Oddi influence drainage of contrast material and must be avoided to obtain accurate drainage times (if drainage time is desired). Because the extrahepatic bile duct angulates from anterior (the hilum) to posterior (the papilla), the patient must be supine to assess gravitational drainage through the sphincter. Although definitive normal supine drainage times have not been well defined,⁶¹ a postcholecystectomy biliary tree that fails to empty all contrast material by 45 minutes is generally considered abnormal. Endoscopic evaluation of the papilla and peripapillary area can yield important information that can influence the diagnosis and treatment of patients with suspected SOD. Occasionally, ampullary cancer may simulate SOD. The endoscopist should do tissue sampling of the papilla (preferably after sphincterotomy) in suspicious cases.⁶²

Radiographic features of the pancreatic duct are also important to assess in a patient with suspected SOD. Dilation of the pancreatic duct (>6 mm in the pancreatic head and >5 mm in the body) and delayed contrast agent drainage time (9 minutes in the prone position) may give indirect evidence for the presence of SOD. ERCP alone is generally not indicated in the evaluation of abdominal pain of obscure origin in the absence of objective findings that suggest a biliary or pancreatic disease (i.e., type III patient).⁶³ Sherman and colleagues⁶⁴ found that only 10% of 197 patients with pancreaticobiliary pain, normal liver function tests, serum amylase, upper GI tract evaluation, and abdominal ultrasound or CT scan had an ERCP finding that might affect their therapy, including chronic pancreatitis (7%), gallbladder stones or sludge (2%), and a choledochal cyst (1%). Most of these findings could have been identified on noninvasive (MR imaging or MRCP) and less invasive (EUS) imaging tests. In view of the high procedure-related complication rate in these patients, the investigators concluded that ERCP alone could not be justified. In the National Institutes of Health State-of-the-Science Conference on ERCP, it was concluded that ERCP, if performed in type III patients, should be coupled with SOM.⁶⁵

Intraductal Ultrasound

Intraductal ultrasound makes it possible to assess sphincter of Oddi morphology during endoscopy. The sphincter appears as a thin hypoechoic circular structure on intraductal ultrasound.⁶⁶ Limited studies so far reveal no correlation between the basal sphincter pressures (as detected at SOM) and the thickness of the hypoechoic layer.⁶⁷ Although intraductal ultrasound may provide additional information at the level of the sphincter, it cannot be used as a substitute for SOM.

Sphincter of Oddi Manometry

The most definitive development in understanding of the pressure dynamics of the sphincter of Oddi occurred with the advent of SOM. SOM is the only available method to measure sphincter of Oddi motor activity directly. Although SOM can be performed intraoperatively and percutaneously, it is most commonly done in the ERCP setting. SOM is considered by most authorities to be the “gold standard” for evaluating patients for SOD.^68,69 The use of manometry to detect motility disorders of the sphincter of Oddi is similar to its use in other parts of the GI tract. However, performance of SOM is more technically demanding and hazardous, with complication rates (in particular, pancreatitis) of 30% reported. Questions remain as to whether these short-term observations (2- to 10-minute recordings per pull-through) reflect the 24-hour pathophysiology of the sphincter. Despite some problems, SOM is gaining more widespread clinical application.

Technique and Indications for Sphincter of Oddi Manometry

SOM is usually performed at the time of ERCP. All drugs that relax (anticholinergics, nitrates, calcium channel blockers, glucagon) or stimulate (most narcotics, cholinergic agents) the sphincter should be avoided for at least 8 to 12 hours before manometry and during the manometric session. Current data indicate that benzodiazepines do not affect the sphincter pressure and are acceptable sedatives for SOM. Meperidine, at a dose of 1 mg/kg or less, does not affect the basal sphincter pressure but does alter phasic wave characteristics.⁷⁰ Because the basal sphincter pressure is generally the only manometric criterion used to diagnose SOD and determine therapy, it was suggested that meperidine could be used to facilitate conscious sedation for manometry. Droperidol⁷¹ and propofol⁷² are increasingly used for SOM, and it appears that these agents also do not affect the basal sphincter pressure. Similarly, ketamine, used in combination with meperidine and diazepam or midazolam, does not significantly alter the biliary and pancreatic basal sphincter pressure.⁷³ If glucagon must be used to achieve cannulation, an 8- to 15-minute waiting period is required to restore the sphincter to its basal condition.

Catheters sized 5-Fr should be used because virtually all standards have been established with catheters of this size. Triple-lumen catheters are state-of-the-art and are available from several manufacturers. Various catheter types can be used. Catheters with a long intraductal tip may help secure the catheter within the bile duct, but such a long nose is commonly a hindrance if pancreatic manometry is desired. Over-the-wire (monorail) catheters can be passed after first securing one’s position within the duct with a guidewire. Whether this guidewire influences basal sphincter pressure is unknown. Some triple-lumen catheters accommodate a 0.018- to 0.021-inch diameter guidewire passed through the entire length of the catheter and can be used to facilitate cannulation or maintain position in the duct. A study in our unit found, however, that stiffer shafted nitinol core guidewires used for this purpose commonly increase basal sphincter pressure by 50% to 100%.⁷⁴ To avoid such artifacts, such wires need to be avoided, the wires need to be pulled back into the catheter during the recording period, or guidewires with a very soft core must be used.

Aspiration catheters, in which one recording port is sacrificed to permit both end and side-hole aspiration of intraductal juice, are highly recommended for pancreatic manometry (Fig. 46.2). Most centers prefer to perfuse the catheters at 0.25 mL/channel using a low-compliance pump. Lower perfusion rates give accurate basal sphincter pressures but do not give accurate phasic wave information. A new water perfused sleeve system, similar to that used in the lower esophageal sphincter, awaits further study in the sphincter of Oddi.⁷⁵ The perfusate is generally distilled water, although physiologic saline needs further evaluation. The latter may crystallize in the capillary tubing of perfusion pumps and must be flushed out often.

Fig. 46.2 Modified triple-lumen aspirating catheter.

SOM requires selective cannulation of the bile duct or pancreatic duct (Fig. 46.3). The duct entered can be identified by gently aspirating on any port. The appearance of yellow fluid in the endoscopic view indicates entry into the bile duct. Clear aspirate indicates that the pancreatic duct was entered. It is preferable to obtain a cholangiogram or pancreatogram or both before performing SOM because certain findings (e.g., CBD stone) may obviate the need for SOM; this can be done simply by injecting contrast material through one of the perfusion ports. Blaut and colleagues⁷⁶ showed that injection of contrast material into the biliary tree before SOM does not significantly alter sphincter pressure characteristics. Similar evaluation of the pancreatic sphincter after contrast agent injection has not been reported. To ensure accurate pressure measurements, one must ensure that the catheter is not impacted against the wall of the duct. When deep cannulation is achieved and the patient is adequately sedated, the catheter is withdrawn across the sphincter at 1- to 2-mm intervals by standard station pull-through technique.

Fig. 46.3 Manometry. A, Normal major papilla. B, Biliary manometry. C, Pancreatic manometry.

Ideally, both the pancreatic and the bile ducts should be studied. Data indicate that an abnormal basal sphincter pressure may be confined to one side of the sphincter in 35% to 65% of patients with abnormal manometry.^6,26,77–80 One sphincter may be dysfunctional, whereas the other is normal. Raddawi and colleagues⁷⁷ reported that an abnormal basal sphincter was more likely to be confined to the pancreatic duct segment in patients with pancreatitis and to the bile duct segment in patients with biliary-type pain and elevated liver function tests. Abnormalities of the basal sphincter pressure ideally should be observed for at least 30 seconds in each lead and be seen on two or more separate pull-throughs. From a practical clinical standpoint, we settle for one pull-through (from each duct) if the readings are clearly normal or abnormal.

During standard station pull-through technique, it is necessary to establish good communication between the endoscopist and the manometrist who is reading the tracing as it rolls off the recorder or appears on the computer screen. Good communication permits optimal positioning of the catheter to achieve interpretable tracings. Alternatively, electronic manometry systems with a television screen can be mounted near the endoscopic image screen to permit the endoscopist to view the manometry tracing during endoscopy. After the baseline study is done, agents to relax or stimulate the sphincter can be given (e.g., CCK), and manometric or pain response can be monitored. The value of these provocative maneuvers for everyday use needs further study before widespread application is recommended.

Criteria for interpretation of a sphincter of Oddi tracing are standard; however, they may vary from center to center. Some areas in which there may be disagreement in interpretation include the required duration of basal sphincter of Oddi pressure elevation, the number of leads in which basal pressure elevation is required, and the role of averaging pressures from the three (or two in an aspirating catheter) recording ports.⁴ Our recommended method for reading the manometry tracings is first to define the zero duodenal baseline before and after the pull-through. Alternatively, intraduodenal pressure can be continuously recorded from a separate intraduodenal catheter attached to the endoscope.

The highest basal pressure (defined as the pressure above the zero duodenal baseline; Fig. 46.4) that is sustained for at least 30 seconds is identified. The mean of the readings from the four lowest amplitude points in this zone is taken as the basal sphincter pressure for that lead for that pull-through. The basal sphincter pressure for all interpretable observations is averaged; this is the final basal sphincter pressure. The amplitude of phasic wave contractions is measured from the beginning of the slope of the pressure increase from the basal pressure to the peak of the contraction wave. Four representative waves are taken for each lead, and the mean pressure is determined. The number of phasic waves per minute and the duration of the phasic waves can also be determined.

Fig. 46.4 A, Abnormal station pull-through at sphincter of Oddi manometry (SOM). The study has been abbreviated to fit onto one page. B, Schematic representation of one lead of the above tracing. Baseline duodenal 0 reference (a). Intraductal (pancreatic) pressure of 20 mm Hg (abnormal) (b). Basal pancreatic sphincter pressure is 45 mm Hg (abnormal) (c). Phasic waves are 155 to 175 mm Hg in amplitude and 6 seconds in duration (normal).

(Redrawn from Fogel EL, Sherman S: Performance of sphincter of Oddi manometry. Clin Perspect Gastroenterol 4:165–173, 2001.)

Most authorities read only the basal sphincter pressure as an indicator of pathology of the sphincter of Oddi. However, data from Kalloo and coworkers⁸¹ suggest that intraductal biliary pressure, which is easier to measure than sphincter of Oddi pressure, correlates with sphincter of Oddi basal pressure. In this study, intrabiliary pressure was significantly higher in patients with SOD than patients with normal sphincter of Oddi pressure (20 mm Hg vs. 10 mm Hg; P < .01). In a similar study, Fazel and colleagues⁸² found that pancreatic duct pressure correlated with pancreatic sphincter pressure (P < .01). Pancreatic duct pressure was significantly higher in patients with SOD compared with patients with normal pressure (20 mm Hg vs. 11 mm Hg; P < .001). These studies must be confirmed, but they support the theory that increased intrabiliary or intrapancreatic pressure is a cause of pain in SOD.

The best study establishing normal values for SOM was reported by Guelrud and associates.⁸³ The study evaluated 50 asymptomatic control patients and repeated the evaluation on two occasions in 10 subjects. This study established normal values for intraductal pressure, basal sphincter pressure, and phasic wave parameters (Table 46.2). The reproducibility of SOM was confirmed. Various authorities interchangeably use 35 mm Hg or 40 mm Hg as the upper limits of normal for mean basal sphincter of Oddi pressure (this is 3 standard deviations above the mean). Although all authorities diagnose SOD when the basal sphincter pressure is 35 to 40 mm Hg or greater, some also make this diagnosis when there are greater than 50% retrograde contractions, tachyoddia (phasic wave frequency >7/min), or a paradoxical contraction response after an intravenous dose of CCK.⁸⁴

Table 46.2 Suggested Standard for Abnormal Values for Endoscopic Sphincter of Oddi Manometry Obtained from 50 Volunteers without Abdominal Symptoms

Basal sphincter pressure*	>35 mm Hg
Basal ductal pressure	>13 mm Hg
Phasic contractions
Amplitude	>220 mm Hg
Duration	>8 sec
Frequency	>10/min

Note: Values were obtained by adding 3 standard deviations to the mean (means were obtained by averaging the results on two to three station pull-throughs). Data combine pancreatic and biliary studies.

* Basal pressures determined by (1) reading the peak basal pressure (i.e., the highest single lead as obtained using a three-lumen catheter) and (2) obtaining the mean of these peak pressures from multiple station pull-throughs.

Adapted from Guelrud M, Mendoza S, Rossiter G, et al: Sphincter of Oddi manometry in healthy volunteers. Dig Dis Sci 35:38–46, 1990.

Several studies have shown that pancreatitis is the most common major complication after SOM.^85–87 Using standard perfused catheters, pancreatitis rates of 31% have been reported. Such high complication rates have limited more widespread use of SOM. These data also emphasize that manometric evaluation of the pancreatic duct is associated with a particularly high complication rate. Rolny and associates⁸⁶ found that patients with chronic pancreatitis were at higher risk of postprocedure pancreatitis after pancreatic duct manometry. They reported an 11% incidence of pancreatitis after manometric evaluation of the pancreatic duct. Pancreatitis developed in 26% of patients with chronic pancreatitis undergoing SOM.

Methods that have been proposed to decrease the incidence of postmanometry pancreatitis include (1) use of an aspiration catheter; (2) gravity drainage of the pancreatic duct after manometry; (3) decrease in the perfusion rate to 0.05 to 0.1 mL/lumen/min; (4) limitation of pancreatic duct manometry time to less than 2 minutes (or avoid pancreatic manometry); (5) use of the microtransducer (nonperfused) or sleeve SOM system^75,88–91; and (6) placement of a pancreatic stent after manometry, sphincterotomy, or both. In a prospective randomized study, Sherman and colleagues⁸⁵ found that an aspirating catheter (catheter that allows for aspiration of the perfused fluid from end and side holes while accurately recording pressure from the two remaining side ports) reduced the frequency of pancreatic duct manometry–induced pancreatitis from 31% to 4%. The reduction in pancreatitis rates with the use of this catheter in the pancreatic duct and the very low incidence of pancreatitis after bile duct manometry lend support to the notion that increased pancreatic duct hydrostatic pressure is a major cause of this complication.

When the pancreatic duct sphincter is studied by SOM, aspiration of pancreatic juice and the perfusate is strongly recommended. In a prospective randomized trial, Wehrmann and colleagues⁸⁸ found that microtransducer manometry was associated with a significantly lower incidence of postmanometry pancreatitis than standard (nonaspirating) perfusion manometry (13.8% vs. 3.1%; P = .04). A sleeve SOM catheter–based system was shown more recently to have similar accuracy as the standard triple-lumen SOM catheter with less artifact.⁹⁰ Because the sleeve assembly is reverse-perfused, no fluid enters the ducts potentially reducing the rate of postprocedure pancreatitis. In another prospective randomized trial, Tarnasky and colleagues⁹² showed that placement of a stent in the pancreatic duct decreased post-ERCP pancreatitis from 26% to 6% in a group of patients with pancreatic sphincter hypertension undergoing biliary sphincterotomy alone.

SOM is recommended in patients with idiopathic pancreatitis or unexplained disabling pancreaticobiliary pain with or without hepatic enzyme abnormalities. An attempt should be made to study both sphincters, but clinical decisions can be made when the first sphincter evaluated is abnormal. However, if the other sphincter is dysfunctional and not treated, the outcome of therapy may be suboptimal. Indications for the use of SOM have also been developed according to the Hogan-Geenen SOD classification system (see Table 46.1). In type I patients, there is a general consensus that a structural disorder of the sphincter (i.e., sphincter stenosis) exists. Although SOM may be useful in documenting SOD, it is not an essential diagnostic study before endoscopic or surgical sphincter ablation. Such patients uniformly benefit from sphincter ablation regardless of the SOM results (see the section on endoscopic therapy). Type II patients exhibit sphincter of Oddi motor dysfunction in 55% to 65% of cases. In these patients, SOM is highly recommended because the results of the study predict outcome from sphincter ablation. Type III patients have pancreaticobiliary pain without other objective evidence of sphincter outflow obstruction. SOM is mandatory to confirm the presence of SOD. Although this has not been well studied, the results of SOM may predict outcome from sphincter ablation in these patients.

Stent Trial as a Diagnostic Test

Placement of a pancreatic or biliary stent on a trial basis in hope of achieving pain relief and predicting the response to more definitive therapy (i.e., sphincter ablation) has received only limited application. Pancreatic stent trials, especially in patients with normal pancreatic ducts, are strongly discouraged because serious ductal and parenchymal injury may occur if stents are left in place for more than a few days.^93,94 Goff⁹⁵ reported a biliary stent trial in 21 patients with normal biliary manometry suspected to have type II and type III SOD. Stents (7-Fr) were left in place for at least 2 months if symptoms resolved and were removed sooner if they were judged ineffective. Relief of pain with the stent was predictive of long-term pain relief after biliary sphincterotomy. Pancreatitis developed in 38% of the patients (14% were graded severe) after stent placement. Because of this high complication rate, biliary stent trials are strongly discouraged. Rolny and colleagues⁹⁶ also reported a series of bile duct stent placement as a predictor of outcome after biliary sphincterotomy in 23 postcholecystectomy patients (7 type II and 16 type III). Similar to the study by Goff,⁹⁵ resolution of pain during at least 12 weeks of stent placement predicted a favorable outcome from sphincterotomy regardless of sphincter of Oddi pressure. In this series, there were no complications related to stent placement.

Therapy for Sphincter of Oddi Dysfunction

The therapeutic approach in patients with SOD is aimed at reducing the resistance to the flow of bile or pancreatic juice, or both, caused by the sphincter of Oddi.¹³ Historically, emphasis has been placed on definitive intervention (i.e., surgical sphincteroplasty or endoscopic sphincterotomy). This approach seems appropriate for patients with high-grade obstruction (type I as per Hogan-Geenen criteria). In patients with lesser degrees of obstruction, the clinician must carefully weigh the risks and benefits before recommending invasive therapy. Most reports indicate that patients with SOD have a complication rate from ERCP, manometry, and endoscopic sphincterotomy of at least twice that of patients with ductal stones.^97,98

Endoscopic Therapy

Endoscopic Sphincterotomy

Endoscopic sphincterotomy is the standard therapy for patients with SOD.¹¹³ Most data on endoscopic sphincterotomy relate to biliary sphincter ablation alone. Clinical improvement after therapy has been reported to occur in 55% to 95% of patients (see Table 46.1). These variable outcomes are reflective of the different criteria used to document SOD, the degree of obstruction (type I biliary patients appear to have a better outcome than type II and type III patients), the methods of data collection (retrospective vs. prospective), and the techniques used to determine benefit. Rolny and colleagues¹¹⁴ studied 17 type I postcholecystectomy biliary patients by SOM (Table 46.3). In this series, 65% had abnormal SOM (although not specifically stated, the biliary sphincter apparently was studied alone). Nevertheless, during a mean follow-up interval of 2.3 years, all patients benefited from biliary sphincterotomy. The results of this study suggested that because type I biliary patients invariably benefit from biliary sphincterotomy, SOM in this patient group not only is unnecessary but also may be misleading. However, the results of this study have never been validated at another center.

Table 46.3 Biliary Sphincter Ablation in Type I Sphincter of Oddi Dysfunction (28-Month Follow-up)*

Basal Sphincter of Oddi Pressure	N	Asymptomatic or Improved after ES or SS
<40 mm Hg	6 (35%)	6 (100%)
>40 mm Hg	11 (65%)	11 (100%)

ES, endoscopic sphincterotomy; SS, surgical sphincterotomy.

* 15 ES, 2 SS.

Adapted from Rolny P, Geenen JE, Hogan WJ: Post-cholecystectomy patients with ‘objective signs’ of partial bile outflow obstruction: Clinical characteristics, sphincter of Oddi manometry findings, and results of therapy. Gastrointest Endosc 39:778–781, 1993.

In contrast, results of several nonrandomized controlled trials^{23,32,58,115,116} suggest that performance of SOM is highly recommended in biliary type II and type III patients because clinical benefit is less certain (Table 46.4). Several other case series have reported symptom improvement in 75% to 100% of type I patients undergoing biliary sphincterotomy.^59,117–120 Although most of the studies reporting efficacy of endoscopic therapy in SOD have been retrospective, three notable randomized trials have been reported. In a landmark study, Geenen and associates¹²¹ randomly assigned 47 postcholecystectomy type II biliary patients to biliary sphincterotomy or sham sphincterotomy. SOM was performed in all patients but was not used as a criterion for randomization. During a 4-year follow-up, 95% of patients with an elevated basal sphincter benefited from sphincterotomy. In contrast, only 30% to 40% of patients with an elevated sphincter pressure treated by sham sphincterotomy or with a normal sphincter pressure treated by endoscopic sphincterotomy or sham sphincterotomy benefited from this therapy. The two important findings of this study were that SOM predicted the outcome from endoscopic sphincterotomy and that endoscopic sphincterotomy offered long-term benefit in type II biliary patients with SOD.

Table 46.4 Biliary Sphincterotomy for Type II and Type III Sphincter of Oddi Dysfunction Documented by Sphincter of Oddi Manometry: Results of Five Nonrandomized Trials

Author (Year)	Clinical Benefit
	Type II	Type III
Choudhry et al (1993)32*	10/18 (56%)	9/16 (56%)
Botoman et al (1994)23	13/19 (68%)	9/16 (56%)
Bozkurt et al (1996)115	14/19 (78%)	5/5 (100%)
Wehrmann et al (1996)116	12/20 (60%)	1/13 (8%)
Rosenblatt et al (2001)58	22/30 (73%)	11/32 (34%)

* Six had cholecystectomy.

Confirming data were seen in a 2-year follow-up study by Toouli and coworkers.^122,123 In this study, postcholecystectomy patients with biliary-type pain (mostly type II) were prospectively randomly assigned to endoscopic sphincterotomy or sham after stratification according to SOM. At 2 years after endoscopic sphincterotomy, 85% (11 of 13) of patients with elevated basal pressure improved, whereas 38% (5 of 13) of patients improved after a sham procedure (P = .041). Patients with normal SOM were also randomly assigned to sphincterotomy or sham. The outcome was similar for the two groups (8 of 13 improved after sphincterotomy, and 8 of 19 improved after sham; P = .47). Sherman and associates¹²⁴ reported their preliminary results of a randomized study comparing endoscopic sphincterotomy and surgical biliary sphincteroplasty with pancreatic septoplasty (with or without cholecystectomy) with sham sphincterotomy for type II and type III biliary patients with manometrically documented SOD. The results are shown in Tables 46.5 and 46.6. During a 3-year follow-up period, 69% of patients undergoing endoscopic or surgical sphincter ablation improved compared with 24% in the sham sphincterotomy group (P = .009). There was a trend for type II patients to benefit more often from sphincter ablation than type III patients (13 of 16 [81%] vs. 11 of 19 [58%]; P = .14).

Table 46.5 Change in the Mean Pain Score*, Number of Hospital Days per Month Required for Pain, and the Percentage Improved in Patients with Manometrically Documented Sphincter of Oddi Dysfunction Randomly Assigned to Endoscopic Sphincterotomy, Sham Sphincterotomy, and Surgical Sphincteroplasty with or without Cholecystectomy

Table 46.6 Clinical Benefit Correlated with Sphincter of Oddi Dysfunction Type

Fig. 46.5 shows the recommended approach of the Rome III committee on functional disorders of the sphincter of Oddi³¹ for the therapy of patients with types I, II, and II biliary SOD. Evidence is now accumulating that the addition of a pancreatic sphincterotomy to an endoscopic biliary sphincterotomy in patients with pancreatic sphincter disease may improve the outcome, as preliminarily reported by Guelrud and coworkers.¹²⁵ Soffer and Johlin¹²⁶ reported that 25 of 26 patients (mostly type II) who failed to respond to biliary sphincterotomy had elevated pancreatic sphincter pressure. Pancreatic sphincter therapy was performed with overall symptomatic improvement in two-thirds of patients. Eversman and colleagues¹²⁷ found that 90% of patients with persistent pain or pancreatitis after biliary sphincterotomy had residual abnormal pancreatic basal pressure. Data from 5 years of follow-up revealed that patients with untreated pancreatic sphincter hypertension were much less likely to improve after biliary sphincterotomy than patients with isolated biliary sphincter hypertension. Elton and colleagues¹²⁸ performed pancreatic sphincterotomy on 43 type I and type II SOD patients who failed to benefit from biliary sphincterotomy alone. During the follow-up period, 72% were symptom-free, and 19% had partial or transient improvement.

Fig. 46.5 Algorithm of the history, diagnostic work-up, and treatment of patients with types I, II, and III biliary sphincter of Oddi dysfunction (SOD).

(From Behar J, Corazziari E, Guelrud M, et al: Functional gallbladder and sphincter of Oddi disorders. Gastroenterology 130:1498–1509, 2006.)

Kaw and colleagues¹²⁹ presented preliminary data showing that response to sphincterotomy also depends on treating the diseased sphincter segment. Patients with pancreatic sphincter hypertension who fail to respond to biliary sphincterotomy can be “rescued” by undergoing pancreatic sphincterotomy (Table 46.7). More recent data from our unit¹³⁰ examined the outcome of endoscopic therapy in patients with SOD with initial pancreatic sphincter hypertension (with or without biliary sphincter hypertension). Patients were followed for a mean of 43.1 months (range 11 to 77 months); reintervention was offered for sustained or recurrent symptoms at a median of 8 months after initial therapy. Performance of an initial dual pancreatobiliary sphincterotomy was associated with a lower reintervention rate (70 of 285 [24.6%]) than biliary sphincterotomy alone (31 of 95 [33%]; P < .05). Confirmatory outcome studies, preferably in randomized trials, are awaited.

Table 46.7 Response to Sphincterotomy in Relation to Sphincter of Oddi Segment Treated (Follow-up 17 Months)

Many authorities argue that the current SOD classification systems might not be a good predictor of outcome.^84,131–133 In a study of 121 patients classified by the modified Milwaukee biliary classification system (18 type I, 53 type II, and 50 type III patients) and treated by biliary sphincterotomy with (49 patients) or without (72 patients) pancreatic sphincterotomy, Freeman and colleagues¹³¹ reported a good to excellent response in 69%. The response was not significantly different between biliary types I, II, and III. The authors found that significant predictors of a poor response to therapy were normal pancreatic manometry, delayed gastric emptying, daily opioid use, and age younger than 40 years. Abnormal liver function tests or dilated bile duct were not significant predictors of outcome. These results indicate that the response rate and enthusiasm for sphincter ablation must be correlated with patient presentation and results of manometry and balanced against the high complication rates reported for endoscopic therapy of SOD.

Most studies indicate that patients undergoing ERCP, manometry, and endoscopic sphincterotomy for SOD have complication rates two to five times higher than patients undergoing ERCP and endoscopic sphincterotomy for ductal stones.^97,98 Pancreatitis is the most common complication, occurring in up to 30% of patients in some series. Several prospective, multicenter studies examining risk factors for post-ERCP pancreatitis identified suspected SOD as an independent factor by multivariate analysis.¹³⁴ A suspicion of SOD tripled the risk of postprocedure pancreatitis to a frequency (23%) that was comparable to that found in other more recent prospective studies.^{92,98,135–137} Placement of a prophylactic pancreatic duct stent has been shown to limit such complications.^92,138–140 These studies have also shown that the risk of pancreatitis is intrinsic to the patient group (patient-related factors) and events occurring during the procedure (procedure-related factors) rather than the SOM when the SOM is performed with an aspirating catheter. In multivariate analysis, SOM has not been shown to be a risk factor for pancreatitis.^141,142

Balloon Dilation and Stent Placement

Balloon dilation of strictures in the GI tract has become a common procedure. In an attempt to be less invasive and possibly to preserve sphincter function, adaptation of this technique to treat SOD has been described. Because of the unacceptably high complication rates, primarily pancreatitis, this technology has little role in the primary management of SOD.¹⁴³ Similarly, although biliary stent placement might offer short-term symptom benefit in patients with SOD and predict outcome from sphincter ablation, it too has unacceptably high complication rates and cannot be advocated in this setting.⁹⁵

Botulinum Toxin Injection

Botulinum toxin (Botox), a potent inhibitor of acetylcholine release from nerve endings, has been successfully applied to smooth muscle disorders of the GI tract such as achalasia. In a preliminary clinical trial, botulinum toxin injection into the sphincter of Oddi resulted in a 50% reduction in the basal biliary sphincter pressure and improved bile flow.¹⁴⁴ This reduction in pressure may be accompanied by symptom improvement in some patients. Although further study is warranted, botulinum toxin may serve as a therapeutic trial for SOD with responders undergoing permanent sphincter ablation. In a small series,¹⁴⁵ 22 postcholecystectomy type III patients with manometric evidence of SOD underwent botulinum toxin injection into the intraduodenal sphincter segment. Of the 12 patients who responded to botulinum toxin injection, 11 later benefited from endoscopic sphincterotomy, whereas only 2 of 10 patients who did not benefit from botulinum toxin injection later responded to sphincter ablation. Such an approach requires two endoscopies to achieve symptom relief. Patients must have relatively frequent episodes of pain to assess the benefit from botulinum toxin. Further studies are needed before this technique can be recommended.

Failure to Achieve Symptomatic Improvement after Biliary Sphincterotomy

There are several potential explanations as to why patients may fail to experience symptom relief after biliary sphincterotomy is performed for well-documented SOD. First, biliary sphincterotomy may have been inadequate, or restenosis may have occurred. Although the biliary sphincter is commonly not totally ablated,¹⁴⁶ Manoukian and coworkers¹⁴⁷ indicated that clinically significant biliary restenosis occurs infrequently. If no “cutting space” remains in such a patient, balloon dilation to 8 to 10 mm may suffice, but long-term outcome from such therapy is unknown, and the risks may be considerable.¹⁴³ Prophylactic pancreatic stent placement may reduce the frequency of post–balloon dilation pancreatitis.¹⁴⁸ Second, as noted previously,^125–130 the importance of pancreatic sphincter ablation is being increasingly recognized. Third, patients may fail to respond to sphincterotomy because they have chronic pancreatitis. Tarnasky and colleagues¹⁴⁹ reported that patients with SOD were four times more likely to have evidence of chronic pancreatitis than patients without SOD (P = .01). Although SOD seems to be associated with chronic pancreatitis, a causal relationship has not been proven. These patients may or may not have abnormal pancreatograms. Intraductal pancreatic juice aspiration after secretin stimulation may help make this diagnosis.^150–152 EUS may show parenchymal and ductular changes of the pancreas in some of these patients suggesting chronic pancreatitis.¹⁵³ Fourth, some patients may be having pain from altered gut motility of the stomach, small bowel, or colon (irritable bowel or pseudo-obstruction variants). There is increasing evidence that upper GI motility disorders may masquerade as pancreatobiliary-type pain (i.e., discrete right upper quadrant pain). Multiple preliminary studies show disordered duodenal motility in such patients.^154–156 Soffer and Johlin¹⁵⁷ found that small bowel dysmotility occurred with greater frequency in type II and type III SOD patients who failed to benefit from sphincterotomy than in patients who did respond. This area needs much more study to determine the frequency, significance, and coexistence of these motor disorders along with SOD.

DeSautels and colleagues¹⁵⁸ suggested that type III patients have duodenal-specific visceral hyperalgesia with pain reproduction by duodenal distention. These patients were also shown to have high levels of somatization, depression, obsessive-compulsive behavior, and anxiety compared with control subjects.¹⁵⁹ Patients with SOD appear to have a higher than expected prevalence of irritable bowel syndrome,¹⁶⁰ and SOD may occur as part of a more generalized functional disorder of the gut. Patients with SOD not only have more somatization than controls, but also they may have an antecedent history of sexual or physical abuse similar to patients with irritable bowel syndrome.¹⁶¹ Wald¹⁵ suggested that selective treatment of the sphincter of Oddi cannot be expected to provide symptom resolution in such patients, and this may account for the high failure rate of sphincterotomy in many patients with type III SOD.

Sphincter of Oddi Dysfunction in Recurrent Pancreatitis

Disorders of the pancreatic sphincter may give rise to unexplained (idiopathic) pancreatitis or episodic pain suggestive of a pancreatic origin.¹¹² Although the pathogenesis of acute pancreatitis in SOD is uncertain, it is believed that the combination of pancreatic duct obstruction and increased exocrine juice flow are needed.⁸⁴ SOD is a frequent cause of recurrent pancreatitis previously labeled as idiopathic acute recurrent pancreatitis (IARP). It has been documented with manometry in 15% to 72% of such patients (Table 46.8).^{6,24,30,162–171} Pancreatic sphincter manometry should be done in patients with IARP, particularly patients with normal biliary manometry and patients who have recurrent attacks after a biliary sphincterotomy. Isolated pancreatic sphincter hypertension is common among patients with IARP found to have SOD.^77,172 In addition, pancreatic sphincter hypertension may explain recurrent pancreatitis despite biliary sphincterotomy or surgical biliary sphincteroplasty.¹⁷²

Table 46.8 Manometrically Documented Sphincter of Oddi Dysfunction Causing Idiopathic Acute Recurrent Pancreatitis

Biliary sphincterotomy alone has been reported to prevent further pancreatitis episodes in more than 50% of patients in some series. From a scientific, but not practical, viewpoint, care must be taken to separate out subtle biliary pancreatitis¹⁷³ that would similarly respond to biliary sphincterotomy. Because IARP is an episodic illness, long-term follow-up is necessary to conclude that a patient is “cured.” Sphincter ablation is the recommended therapy for patients with IARP resulting from SOD. Historically, ablation has been accomplished surgically.¹⁰⁹ However, with increasing experience, endoscopic sphincterotomy has become the treatment of choice. Controversy continues to exist about the type of sphincterotomy that should be performed.¹⁷⁴

The value of ERCP, SOM, and sphincter ablation therapy was studied in 51 patients with idiopathic pancreatitis.⁶⁹ An elevated basal sphincter pressure was present in 24 (47.1%) patients. There were 30 patients treated by biliary sphincterotomy (n = 20) or surgical sphincteroplasty with septoplasty (n =10). Of 18 patients, 15 (83%) with an elevated basal sphincter pressure had long-term benefit (mean follow-up 38 months) from sphincter ablation therapy (including 10 of 11 treated by biliary sphincterotomy), in contrast to only 4 (33.3%; P < .05) of 12 patients with a normal basal sphincter pressure (including 4 of 9 patients treated by biliary sphincterotomy).

Guelrud and colleagues¹²⁵ found, however, that severance of the pancreatic sphincter was necessary to resolve pancreatitis (Table 46.9). In this series, 69 patients with idiopathic pancreatitis resulting from SOD underwent treatment by standard biliary sphincterotomy (n = 18), biliary sphincterotomy with pancreatic sphincter balloon dilation (n = 24), biliary sphincterotomy followed by pancreatic sphincterotomy in separate sessions (n = 13), or combined pancreatic and biliary sphincterotomy in the same session (n = 14). Of patients undergoing pancreatic and biliary sphincterotomy, 81% had resolution of pancreatitis compared with 28% of patients undergoing biliary sphincterotomy alone (P < .005). Sherman and colleagues¹⁶⁸ reported that only 44% of patients with SOD and IARP had no further attacks during a 5-year follow-up interval after biliary sphincterotomy alone. These data are consistent with the theory that many such patients who benefit from biliary sphincterotomy alone have subtle gallstone pancreatitis.

Table 46.9 Pancreatic Sphincter Dysfunction and Recurrent Pancreatitis: Response to Sphincter Therapy

* P < .005 vs. biliary sphincterotomy alone.

Adapted from Guelrud M, Plaz J, Mendoza S, et al: Endoscopic treatment in type II pancreatic sphincter dysfunction. Gastrointest Endosc 41:A398, 1995.

The results of Guelrud and colleagues¹²⁵ also support the anatomic findings of separate biliary and pancreatic sphincters and the manometry findings of residual pancreatic sphincter hypertension in more than 50% of persistently symptomatic patients who undergo biliary sphincterotomy alone. Kaw and Brodmerkel¹⁶⁹ reported that among patients with idiopathic pancreatitis secondary to SOD, 78% had persistent manometric evidence of pancreatic sphincter hypertension despite a biliary sphincterotomy. Toouli and coworkers¹⁷⁵ also showed the importance of pancreatic and biliary sphincter ablation in patients with idiopathic pancreatitis. In this series, 23 of 26 patients (88%) undergoing surgical ablation of both the biliary and the pancreatic sphincter were either asymptomatic or had minimal symptoms at a median follow-up of 24 months (range 9 to 105 months). Okolo and colleagues¹⁷⁶ retrospectively evaluated the long-term results of endoscopic pancreatic sphincterotomy in 55 patients with manometrically documented or presumed pancreatic sphincter hypertension (presumption based on recurrent pancreatitis with pancreatic duct dilation and contrast medium drainage time from the pancreatic duct >10 minutes). During a median follow-up of 16 months (range 3 to 52 months), 34 patients (62%) reported significant pain improvement. Patients with normal pancreatograms were more likely to respond to therapy than patients with pancreatographic evidence of chronic pancreatitis (73% vs. 58%).

Jacob and coworkers¹⁷⁷ postulated that SOD might cause recurrent episodes of pancreatitis, even though SOM was normal, and pancreatic stent placement might prevent further attacks. In a randomized study, 34 patients with unexplained recurrent pancreatitis; normal pancreatic duct SOM, ERCP, and secretin testing; and no biliary crystals were treated with pancreatic stents (n = 19; 5-Fr to 7-Fr, with stents exchanged three times over a 1-year period) or conservative therapy (n = 15). During a 3-year follow-up, pancreatitis recurred in 53% of the patients in the control group and only 11% of the patients with stent placement (P < .02). This study suggests that SOM may be an imperfect test because patients may have SOD, but it may not be detected at the time of SOM. Long-term studies are needed to evaluate the outcome after removal of stents, and concern remains regarding stent-induced ductal and parenchymal changes.^93,94 Because of the concern of stent-induced injury to the pancreas, trial pancreatic duct stent placement to predict outcome from pancreatic sphincterotomy is not recommended.¹⁷⁸

Wehrmann and colleagues¹⁷⁹ evaluated the feasibility and effectiveness of botulinum toxin injection in patients with recurrent pancreatitis resulting from pancreatic sphincter hypertension. No side effects of the injection were noted in any of the 15 treated patients. At 3-month follow-up, 12 patients (80%) remained asymptomatic, but 11 developed a relapse at a follow-up period of 6 ± 2 months. These 11 patients underwent pancreatic or combined pancreatobiliary sphincterotomy with subsequent remission after a median follow-up of 15 months. This study showed that injection of botulinum toxin is safe, may be effective in the short-term, and may predict the outcome from pancreatic sphincter ablation in patients having frequent episodes of pancreatitis, but the need for definitive sphincter ablation in most patients limits its clinical use.

These data show that SOD is the most common cause of IARP when detailed endoscopic evaluation is performed. SOM should be considered the “gold standard” for diagnosing SOD. Complete sphincter evaluation requires manometric assessment of both the biliary and the pancreatic sphincters. Although the best endoscopic therapy of SOD in the setting of IARP warrants further investigation, there is mounting evidence that pancreatic sphincter ablation is necessary in most patients to achieve the best long-term results.

Conclusion

Knowledge of SOD and manometric techniques to assist in this diagnosis is evolving. Successful endoscopic SOM requires good general ERCP skills and careful attention to the main details summarized in this chapter. If SOD is suspected in a type III patient or a type II patient with mild to moderate pain level, medical therapy should generally be tried. If medical therapy fails or is bypassed, ERCP and manometric evaluation should be considered when expertise is available. Although the role of less invasive studies is uncertain, these are generally not advocated because of relatively low sensitivity and specificity. Sphincter ablation is generally warranted in symptomatic type I patients and type II and type III patients with abnormal manometry. Symptom relief ranges from 55% to 95%, depending on the patient presentation and selection. Initial nonresponders after biliary sphincterotomy alone has been performed may benefit from thorough pancreatic sphincter and pancreatic parenchymal evaluation. Data are accumulating to support pancreatic sphincter evaluation and therapy in expert centers, particularly in type III patients, to achieve optimal outcome. Patients with SOD have relatively high complication rates after invasive studies or therapy. A thorough review of the risk-to-benefit ratio with individual patients is mandatory.