Article by Walter J. Hogan, MD

..

Sphincter of Oddi Dysfunction
At the request of one of my readers, I've posted some more information about the Sphincter of Oddi and its dysfunction.

It’s a long read, the first part is a writeup on some background info, and the second is about treatment options. Enjoy!



Clinical manifestations and diagnosis of sphincter of Oddi dysfunction

Walter J Hogan, MD

UpToDate performs a continuous review of over 350 journals and other resources. Updates are added as important new information is published. The literature review for version 14.2 is current through April 2006; this topic was last changed on May 15, 2006. The next version of UpToDate (14.3) will be released in October 2006.

INTRODUCTION — The sphincter of Oddi is a muscular structure that encompasses the confluence of the distal common bile duct and the pancreatic duct as they penetrate the wall of the duodenum (show figure 1). The term “sphincter of Oddi” dysfunction has been used to describe a clinical syndrome of biliary or pancreatic obstruction related to mechanical or functional abnormalities of the sphincter of Oddi.

The clinical manifestations and diagnosis of sphincter of Oddi dysfunction will be reviewed here. The treatment of this disorder is discussed separately. (See “Treatment of sphincter of Oddi dysfunction”).

ANATOMY — The sphincter of Oddi (SO) is composed of small circular and longitudinal muscular segments that are approximately 6 to 10 mm in total length and are contained mostly within the wall of the duodenum (show figure 1) [1]. The muscle fibers surround the intraduodenal segment of the common bile duct and the ampulla of Vater. A circular aggregate of muscle fibers known as the sphincter choledochus (or sphincter of Boyden) maintains resistance to bile flow, and thereby permits filling of the gallbladder during fasting and prevents retrograde reflux of duodenal contents into the biliary tree. A separate structure, called the sphincter pancreaticus, encircles the distal pancreatic duct. The muscle fibers of the sphincter pancreaticus are interlocked with those of the sphincter choledochus in a figure eight pattern. Although the pancreatic and biliary sphincter portions of the SO can be distinguished anatomically, their manometric features are similar and a direct anatomic and manometric correlation has not been established.

The muscle fibers of the SO function independently from those of the duodenal musculature. The motility of the SO is complex and not completely understood, but is known to vary in the fasting and fed states.

During fasting, SO motility is integrated with the migrating motor complex (MMC), permitting coordinated release of bile into the duodenum. Myoelectrical potentials within the SO increase during phase one of the MMC, reach a maximum during phase three, and then decrease rapidly.
..
During the fed state, myoelectrical potentials within the SO vary depending upon the type and quantity of nutrients ingested and may be influenced by endogenous hormones such as cholecystokinin [2,3].
..
DEFINITIONS — A great deal has been written about the sphincter of Oddi and its dysfunction, but the literature is often difficult to interpret because of differences in nomenclature [4]. The terms papillary stenosis, sclerosing papillitis, biliary spasm, biliary dyskinesia, and postcholecystectomy syndrome have been used synonymously with sphincter of Oddi dysfunction (SOD). Despite this source of confusion, two separate pathologic entities are widely recognized based upon their distinct pathogenic mechanisms [5,6].

• Sphincter of Oddi stenosis
• Sphincter of Oddi dyskinesia

Sphincter of Oddi stenosis is an anatomic abnormality associated with narrowing of the SO. It can result from any process leading to inflammation or scarring such as pancreatitis, passage of a gallstone through the papilla, intraoperative trauma, infection, and adenomyosis. Sphincter of Oddi stenosis is associated with abnormal SO motility and elevated basal pressure.

Sphincter of Oddi dyskinesia refers to a functional disturbance of the SO, leading to intermittent biliary obstruction. The cause of SO dyskinesia is not well understood. Spasm and relaxation of the SO can be induced pharmacologically with agents known to affect smooth muscle function (such as nitroglycerin), suggesting that the spasm may be influenced by local hormonal or neurologic disturbance.

CLINICAL MANIFESTATIONS — Sphincter of Oddi dysfunction has been associated with two clinical syndromes: biliary pain and pancreatitis. The prevalence of SOD among patients with these conditions is difficult to estimate because of several potential sources of bias among studies evaluating SOD in these settings:

• The amount of investigation for other causes of symptoms differed.
• The diagnostic “gold-standard” for SOD varied.
• The diagnosis of SOD usually involves invasive testing. As a result, most studies have not included an adequate control group.

Biliary pain — Sphincter of Oddi dysfunction is suspected in patients who have biliary-type pain without other apparent causes. In this setting:

• SOD is most commonly recognized in patients who have undergone Cholecystectomy (hence the name postcholecystectomy syndrome).
• The reasons for this are not well understood, but may be related to unmasking of preexisting SOD due to removal of the gallbladder, which may have served as a reservoir to accommodate increased pressure in the biliary system occurring during sphincter spasm [7].
• Another possible explanation is alteration of SO motility due to the severing of nerve fibers that pass between the gallbladder and SO via the cystic duct [8].
• However, SOD also occurs in patients whose gallbladders are intact, suggesting that other pathophysiologic mechanisms are involved [9].

Despite the association of SOD with cholecystectomy, SOD is an uncommon occurrence following cholecystectomy [10]. In one series of 454 patients who had undergone cholecystectomy, the prevalence of SOD dysfunction was estimated to be less than 1 percent [11].

Pancreatitis — The cause of pancreatitis escapes routine diagnostic studies in some patients. (See “Clinical manifestations and diagnosis of acute pancreatitis”, section on Diagnosis of the etiology of acute pancreatitis). SO dysfunction has been hypothesized as a cause of idiopathic recurrent pancreatitis and pancreatitis occurring after endoscopic retrograde cholangiopancreatography (ERCP) [12,13]. (See “Post-ERCP pancreatitis”).

Abnormalities of the pancreatic or biliary sphincter in association with recurrent idiopathic pancreatitis have been demonstrated in several studies [14-16]. However, whether SOD gives rise to, or is a result of pancreatitis has not been proven. In one study, for example, among 32 patients with a known cause for chronic pancreatitis (alcohol, annular pancreas, and biliary causes), seven (22 percent) had elevated basal SO pressures compatible with SOD [14].

Another study evaluated 104 patients who had SO manometry to investigate unexplained upper abdominal pain; all had previously undergone investigation for chronic pancreatitis by pancreatic ductography, endoscopic ultrasound, and pancreatic fluid bicarbonate determination [16]. Of the 68 patients found to have SOD, 29 percent had structural evidence of chronic pancreatitis, while SOD dysfunction was present in 20 of 23 (87 percent) patients found to have chronic pancreatitis.

Evidence that SOD can cause pancreatitis was suggested in an animal model in which transient sphincter contraction induced by application of topical carbachol abolished trans-sphincteric flow and increased pancreatic exocrine secretion and pancreatic duct pressure to levels comparable with that seen in pancreatic duct ligation [17]. The addition of cholecystokinin/secretin stimulation of pancreatic secretion plus carbachol application caused pancreatic tissue damage and an increase in serum amylase levels.

Natural history — Few studies have addressed the long-term natural history of SOD. The available data suggest that the clinical course is variable depending in part upon the initial biliary classification (see “Classification systems” belowsee “Classification systems” below). In a one-year follow-up study, seven type II patients with abnormal SO pressure treated by a sham procedure continued to have symptoms, which resolved only after subsequent sphincterotomy. All patients continued to do well four years later. Five other type II patients with abnormal SO pressure refused sphincterotomy. At four-year follow-up, three were unimproved while two had “fair” improvement.

The clinical course is unpredictable after a sham or endoscopic sphincterotomy in patients with type III biliary pain. In one report, 11 such patients were followed for two years after sphincterotomy. Four improved symptomatically while seven had no change. Eleven other patients had a sham procedure of whom five improved while six had no change in symptoms during two years of follow-up [18].

DIAGNOSIS — The diagnosis of SOD is established by SO manometry, which is performed during ERCP. Several less invasive methods have also been evaluated for establishing the diagnosis, but none has proven to be consistently more accurate in controlled clinical trials [19,20]. This section will first review clinical criteria in which the diagnosis should be considered and then discuss the tests for biliary and pancreatic SOD and SO manometry, which remains the gold standard for the diagnosis.

Rome III criteria — Clinical criteria for diagnosis of functional gallbladder and sphincter of Oddi disorders have been proposed based mainly upon expert consensus [21]. These criteria (known as the Rome III criteria) specify three subsets of functional gallbladder and sphincter of Oddi disorders and also recognize a pancreatic sphincter of Oddi disorder.

The three gallbladder and biliary sphincter disorders are:

• Functional gallbladder disorder
• Functional biliary sphincter of Oddi disorder
• Functional pancreatic sphincter of Oddi disorder

The guidelines stress that patients with upper abdominal pain who do not meet the Rome III symptom-based criteria should not be submitted to ERCP or other invasive procedures. Those who fulfill the criteria should be assessed initially with noninvasive procedures and eventually with therapeutic trials that will more likely identify the majority of patients whose pain is not biliopancreatic in origin and will therefore not require further investigation.

Functional gallbladder and sphincter of Oddi disorders — The following are the Rome III criteria for functional gallbladder and sphincter of Oddi disorders (functional gallbladder, biliary and pancreatic sphincter of Oddi disorders described above are subsets of this grouping)

Must include episodes of pain located in the epigastrium and/or right upper quadrant and ALL the following:

• Episodes lasting 30 minutes or longer
• Recurrent symptoms occurring at different intervals (not daily)
• The pain builds up to a steady level
• The pain is moderate to severe enough to interrupt the patients’ daily activities or lead to an emergency department visit
• The pain is not relieved by bowel movements
• The pain is not relieved by postural change
• The pain is not relieved by antacids
• Exclusion of other structural disease that would explain the symptoms

Supportive criteria include: The pain may be present with one or more of the following:

• pain associated with nausea and vomiting; pain radiates to the back and/or right infrasubscapular region;
• pain awakens from sleep in the middle of the night
  

Functional gallbladder disorder — The following are the Rome III criteria for functional gallbladder disorder:

• Criteria for functional gallbladder and sphincter of Oddi disorders fulfilled
• Gallbladder is present
• Normal liver enzymes, conjugated bilirubin, and amylase/lipase

Functional biliary sphincter of Oddi disorder — The following are the Rome III criteria for functional biliary sphincter of Oddi disorder.

Must include both of the following:

• Criteria for functional gallbladder and sphincter of Oddi disorder
• Normal amylase/lipase
  

Supportive criteria include:

• elevated serum aminotransferases, alkaline phosphatase or conjugated bilirubin temporally related to at least two pain episodes.

Functional pancreatic sphincter of Oddi disorder — The following are the Rome III criteria for functional pancreatic sphincter of Oddi disorder.

Must include both of the following:

• Criteria for functional gallbladder and sphincter of Oddi disorder
• Elevated amylase/lipase
  

Biliary SOD — The presence of biliary SOD has been based upon a variety of parameters, including:

• dilation of the common bile duct,
• provocation tests,
• hepatobiliary scintigraphy, and
• a classification system.

Dilation of the common bile duct — Otherwise unexplained dilation of the common bile duct on ultrasound is associated with SOD and may predict a favorable response to sphincterotomy in patients with other clinical evidence of biliary obstruction (eg, pain, abnormal liver function tests) [22]. However, common bile duct dilation (more than 6 mm) may be observed in up to one-third of patients after cholecystectomy [23-25]. Furthermore, the size of the common bile duct increases with age [26]. As a result, dilation of the common bile duct alone is insufficient evidence for establishing the diagnosis of SOD. It should be considered in the context of symptoms, liver and pancreatic biochemical tests, and a history of multiple gallbladder stones or past removal of common bile duct stones. In most instances, mild dilation is an incidental finding warranting only observation.

Biliary provocation tests — To increase the specificity of common bile duct diameter measurement for determining SOD, several provocation tests have been developed that use either a fatty meal (fatty meal ultrasonography) or cholecystokinin to increase bile flow [27,28]. In patients who have normal SO function, the bile duct diameter remains constant or decreases following stimulation; an increase of more than 2 mm is considered to be pathologic [27].

Correlation of provocation tests with sphincter of Oddi manometry is poor and abnormal findings may be seen in non-biliary diseases. As an example, an abnormal response of the SO to CCK stimulation or the Nardi test has been observed in patients with irritable bowel syndrome [29,30].

Hepatobiliary scintigraphy — Hepatobiliary scintigraphy using technetium-99m labeled dyes can provide a standardized, semiquantitative assessment of delayed biliary drainage in patients whose gallbladder is absent [7,31-33]. In one study, 26 consecutive patients underwent hepatobiliary imaging, ERCP, and SO manometry [31]. A scintigraphic score (referred to as the Hopkin’s score) was constructed from quantitative and visual criteria that successfully discriminated all patients with SOD.

Substantially different conclusions were reached in another report in which scintigraphy results were compared with sphincter of Oddi manometry in 27 patients with suspected SOD following cholecystectomy [34]. Scintigraphy with cholecystokinin infusion was performed within one month of manometry. Scoring of the scans and measurement of the transit time from the hepatic hilum to the duodenum (HDTT) was performed by independent, blinded observers.

Eight patients had abnormal SO manometry (basal SO pressure >40 mmHg). Scintigraphy scoring had a sensitivity of 25 to 38 percent, a specificity of 86 to 89 percent, a positive predictive value of 40 to 60 percent, and a negative predictive value of 75 to 79 percent. The coefficient of variation between observers was 0.72 (ie, moderately good correlation). The sensitivity, specificity, positive and negative predictive values of the HDTT were 13, 95, 50, and 74 percent, respectively. The authors concluded that scintigraphy correlated poorly with manometry in postcholecystectomy patients with suspected SOD.

Clearance rates in patients with SOD overlap with those in a normal population [7]. In addition, scintigraphy may be falsely positive in patients who have extrahepatic biliary obstruction from a variety of causes, or falsely negative in patients who have SO dyskinesia in whom obstruction to bile flow may be intermittent. Furthermore, hepatobiliary scintigraphy cannot detect obstruction to flow arising from the pancreatic portion of the SO. Thus, scintigraphy should have only a supportive role in the evaluation of suspected SOD.

Classification systems — Several investigators have constructed criteria by which the likelihood of finding SOD and its response to treatment can be predicted. The best studied classification system for biliary SOD (known as the Milwaukee Biliary Group Classification) is based upon the number of laboratory, clinical, and radiologic features suggesting SOD in an individual patient. These have been revised in the Rome III statement as described below. The original Milwaukee classification recognizes three groups of patients who have biliary type pain without an identifiable cause prior to manometry [35]:

Type I patients fulfill all of three criteria:

• (a) pain associated with abnormal serum aminotransferases (ALT and AST more than two times normal on at least two occasions);
• (b) a dilated common bile duct more than 10 mm on ultrasound or 12 mm on ERCP; and
• (c) delayed drainage of contrast from the common bile duct after more than 45 minutes in the supine position).
  

Type II patients have one or two of the above criteria.

Type III patients have none of the above criteria.

These criteria have been evaluated in a number of studies [7,36-39]. In one report, for example, 213 patients with pancreaticobiliary pain were evaluated by sphincter of Oddi manometry and ERCP [36]. Abnormal sphincter of Oddi manometry was found in 86, 55, and 28 percent of patients classified as Milwaukee groups I, II, and III, respectively.

However, the ability of these criteria to predict a favorable outcome in patients treated by sphincterotomy has varied among studies [37-41], and their role in management has been criticized [42]. In most series, patients in class I benefit from sphincterotomy; optimal therapy and the value of the Milwaukee criteria are less clear for patients who are class II and III. One study found no differences in outcomes among 73 Milwaukee type II and III patients with intractable biliary-type pain treated by sphincterotomy [38]. In contrast, in another series that included 108 patients with recurrent biliary-type pain after cholecystectomy, elevated SO pressure was found in 63 and 50 percent of Milwaukee type II and III patients, respectively [40]. Improvement in symptoms following sphincterotomy was noted in 70 and 39 percent of patients in the two groups.

Rome III revision — The Rome III consensus statement revised the classification to make it more applicable to clinical practice and, whenever possible, avoid invasive procedures such as ERCP. Thus, the revised system emphasizes noninvasive methods to evaluate common bile duct diameter and suggests that contrast drainage times are not practical currently.

• Type I patients present with biliary-type pain; abnormal aminotransferases, bilirubin or alkaline phosphatase >2 times normal values documented on two or more occasions and a dilated bile duct greater than 8 mm diameter on ultrasound. Approximately 65 to 95 percent of these patients have manometric evidence of biliary SOD.
• Type II patients present with biliary-type pain and one of the previously mentioned laboratory or imaging abnormalities. Approximately 50 to 63 percent of these patients have manometric evidence of biliary SOD.
• Type III patients complain only of recurrent biliary-type pain and have none of the previously mentioned laboratory or imaging criteria. Approximately 12 to 59 percent of these patients have manometric evidence of biliary SOD.

Correlation of noninvasive tests with sphincter of Oddi manometry — As noted above, several studies have compared the accuracy of the noninvasive tests with SOD with manometry. One of the largest studies comparing the most commonly used methods involved 304 patients with suspected SOD who underwent manometry, fatty meal ultrasonography, and hepatobiliary scintigraphy [43]. The following observations were made:

• 73 patients (24 percent) were diagnosed with SOD by manometry.
  Compared to sphincter of Oddi manometry as the gold-standard, the sensitivity and specificity of fatty meal ultrasonography was 21 and 97 percent, respectively.
  The sensitivity and specificity of hepatobiliary scintigraphy were 49 and 78 percent, respectively.
• Hepatobiliary scintigraphy and fatty meal ultrasonography were both abnormal in 90, 50, and 40 percent of patients with Type I, II, and II SOD dysfunction, respectively.
• A durable clinical response was observed in 40 of 73 patients (55 percent) who underwent sphincterotomy. Of those with SOD, 11 of 13 (85 percent) with abnormal hepatobiliary scintigraphy and fatty meal ultrasonography had a good long-term response.
• These data suggest that the correlation of fatty meal ultrasonography and hepatobiliary scintigraphy with sphincter of Oddi manometry was poor. Combined use of hepatobiliary scintigraphy and fatty meal ultrasonography increased sensitivity compared to either alone. The accuracy of these tests decreased in relation to the probability of SOD as determined by the Milwaukee Biliary Group Classification. The authors suggest that despite the poor test characteristics, fatty meal ultrasonography and hepatobiliary scintigraphy may provide assistance in predicting the long-term response to sphincterotomy in patients with manometrically proven SOD.

Pancreatic SOD — Tests of SOD involving the pancreatic segment of the SO have focused on pancreatic outflow obstruction due to SO stenosis.

Pancreatic provocation tests — Provocation tests for evaluating pancreatic SOD are based upon a similar principle as provocation tests for biliary SOD. An increase in pancreatic duct diameter following secretin stimulation of more than 1.5 mm (assessed by transabdominal ultrasound or CT) lasting for more than 30 minutes is considered to be pathologic [44]. In one report evaluating this technique, a positive test was associated with good operative outcome in 90 percent of patients with suspected SO dysfunction [45]. However, these results have not been duplicated in other centers. In another study, the accuracy of the secretin ultrasound test was compared to manometry as the gold standard in 47 patients with acute pancreatitis; ultrasound findings in 35 healthy patients were used as controls [46]. The sensitivity and specificity of the secretin ultrasound was 88 and 82 percent, respectively.

• The addition of endoscopic ultrasound (EUS) to the secretin stimulation test has been attempted to improve accuracy. In one report, for example, 20 patients underwent dynamic imaging of the pancreas using real-time endoscopic ultrasonography with secretin stimulation [47]. Of the seven patients who had SO dysfunction based upon manometric results, only four (57 percent) had an abnormal endoscopic ultrasound measurement. Only 1 of 13 patients with normal SO manometry had an abnormal endoscopic ultrasound study. The results suggested excellent negative and positive predictive values but poor overall sensitivity for this test (show table 1). (See “Glossary of common biostatistical and epidemiological terms”).
• The secretin stimulation test has also been performed during magnetic resonance cholangiopancreatography (MRCP). A pilot study compared the results with manometry in 15 patients with idiopathic recurrent pancreatitis of whom six (40 percent) had documented sphincter dysfunction by manometry [48]. SOD was suggested by secretin MRCP in four patients (27 percent). Secretin MRCP and manometry were concordant in 13 of 15 patients (82 percent in those diagnosed with sphincter dysfunction by manometry, and 100 percent in those in whom sphincter dysfunction was excluded). Further studies are needed.

Classification systems — A classification system similar to the Milwaukee Biliary Group Classification has been developed for pancreatic SOD [36]. It recognizes three groups of patients who have recurrent pancreatitis and/or typical pancreatic pain of uncertain etiology:

Type I patients have all three of the following criteria:

• (a) elevation of pancreatic enzymes (more than 1.5 times the upper limit of normal) associated with pain;
• (b) a dilated pancreatic duct (greater than 6 mm in the head and more than 5 mm in the body by ERCP); and
• (c) delayed drainage of contrast after ERCP (more than nine minutes).

Type II patients have or two of the above criteria.

Type III patients have none of the above criteria.

The accuracy of these criteria for predicting pancreatic SOD was evaluated in one study in which elevated basal sphincter pressure was found in 92, 58, and 35 percent in groups I, II, and III, respectively [36].

Sphincter of Oddi manometry — Sphincter of Oddi manometry remains the gold standard for diagnosis of SOD. The most common method involves retrograde intubation of the sphincter of Oddi with a pressure-transducing manometry catheter during ERCP. Basal pressure and phasic wave contractions are routinely recorded from the common bile duct and pancreatic duct segments of the sphincter of Oddi; the mechanical and electrical activity are similar between the two segments (show figure 2) [49]. Measurement of basal pressures from either the biliary or pancreatic duct alone may miss up to one-quarter of patients with abnormal sphincter pressure [50].

Patients with SO dysfunction have been divided into two groups based upon manometric findings:

• Patients with structural alterations of the SO zone (stenosis)
• Patients with functional abnormalities (dyskinesia)

Patients with stenosis are identified by an abnormally elevated basal SO pressure (>40 mmHg) (show figure 3). This finding is reproducible, and the elevated SO pressure does not relax following administration of smooth muscle relaxants.

Patients with SO dyskinesia may also have elevated basal SO pressure. However, in contrast to SO stenosis, the elevated pressure decreases dramatically following amyl nitrite inhalation or glucagon bolus injection, which relax smooth muscles. Other manometric characteristics of this group are: rapid SO contraction frequency (>7/min), an excess in retrograde phasic contractions (>50 percent), and a substantial basal SO pressure increase (paradoxical response) following administration of cholecystokinin-octapeptide (CCK-8) (show figure 4). Manometric findings in patients with SO dyskinesia are less well reproduced upon repeat measurement compared to patients with SO stenosis.

Limitations — Although SO manometry remains the gold standard for diagnosis of SOD, it is invasive, technically demanding, and has several limitations:

• Most series have suggested that it is associated with an increased risk of pancreatitis. In an illustrative series of 100 consecutive patients who underwent sphincter of Oddi manometry, the overall incidence of pancreatitis (defined as upper abdominal pain associated with an elevation in serum amylase or lipase 24 hours after manometry) was 17 percent [51]. (This compares to an incidence of approximately 5 percent in patients undergoing ERCP for other indications.) The incidence was significantly increased in patients who had manometry plus ERCP during one session compared to those who had manometry alone (26 versus 9 percent). (See “Post-ERCP pancreatitis”).
• No clinical or manometric criteria were able to predict the development of pancreatitis. Furthermore, on multivariate analysis, sphincterotomy was not associated with an increased risk of pancreatitis compared to ERCP alone. Placement of a pancreatic stent following biliary sphincterotomy may reduce the incidence of pancreatitis in patients with pancreatic sphincter hypertension [13]. In one report in which stenting was used, the level of the serum amylase obtained two hours after SO manometry predicted the development of pancreatitis [52]. (See “Treatment of sphincter of Oddi dysfunction”).

On the other hand, not all reports have reached this conclusion. A retrospective study of 268 patients who underwent elective ERCP categorized patients into two major groups:

• those with suspected sphincter of Oddi dysfunction (cases) and
• those with a bile duct stone (controls) [53].

The case group was further subclassified into two groups

• (those who underwent sphincter of Oddi manometry followed by immediate ERCP, and
• those who had an ERCP without manometry).

Similar to other reports, the rate of pancreatitis was much higher in the group with suspected sphincter of Oddi dysfunction (27 versus 3.2 percent). However, there was no significant difference in the rate of pancreatitis in those with sphincter of Oddi dysfunction who underwent manometry and ERCP compared with those who had and ERCP without manometry (OR 0.72, 95 percent CI 0.08 to 9.2). On multivariable analysis, biliary sphincterotomy and pancreatography were independent predictors of pancreatitis. Thus, the authors concluded that the higher risk of pancreatitis was not due to the manometry but rather to the underlying presence of sphincter of Oddi dysfunction.

The technique, equipment, and method of sedation used can affect the results. (See “Role of propofol and options for patients who are difficult to sedate for gastrointestinal endoscopy”).
The interpretation of results can vary among observers depending in part upon experience [54].
Prediction of response to treatment — Despite these limitations, the litmus test by which SO manometry can be judged is in its ability to predict a favorable response to treatment in patients with clinical manifestations of SOD.

Biliary-type pain — The most reliable finding predicting a favorable response to sphincterotomy in patients with biliary-type pain is elevated basal pressure. In one study, for example, 47 patients thought to have SOD based upon clinical findings were randomized to sphincterotomy or sham sphincterotomy prior to undergoing manometry [37]. At one-year follow-up, sphincterotomy improved pain scores and objective parameters in significantly more patients who had elevated basal sphincter pressures who underwent sphincterotomy (90 versus 25 percent). In contrast, pain scores were similar in patients without elevated pressure, regardless of treatment. Furthermore, the majority of patients with elevated basal pressures who underwent sham sphincterotomy benefited from subsequent sphincterotomy; these results persisted at four years’ follow-up.

Similar findings were noted in a study that included 81 patients who were randomly assigned to sphincterotomy or a sham procedure based upon the results of manometry [18]. The manometric record was categorized as SO stenosis, SO dyskinesia, or normal. In the SO stenosis group (comprised of 26 patients), symptoms improved significantly more often in those randomized to sphincterotomy (85 versus 39 percent). In contrast, results were not different in the sphincterotomy or sham groups in patients categorized as dyskinesia or normal. Interestingly, symptomatic outcome was not predicted by bile duct dilation, alteration of liver enzymes, or results of the morphine/neostigmine provocation test, a dissociation also observed by other investigators. Complications included mild pancreatitis in seven patients (14 episodes), one patient developed a fluid collection in the right upper quadrant, presumably related to a minor perforation. The endoscopic incision was extended in 19 patients after three months because of manometric evidence of incomplete division of the sphincter. The above observations suggest that patients classified as type II or higher who have manometric features characteristic of SO stenosis have a significant chance of improving following sphincterotomy [55].

Pancreatitis — Manometric findings in patients with idiopathic recurrent pancreatitis suggest that impedance to flow of pancreatic secretions may cause pancreatitis (show figure 3). This relationship was illustrated in a study of 35 patients with recurrent pancreatitis and no identifiable cause who were evaluated with SO manometry [56]. Twenty-six patients underwent total division of the pancreatic sphincter via open sphincterotomy and septectomy based upon abnormal SO manometry. After a median follow-up of 24 months, 15 (62 percent) of these patients were classified as “cured,” the majority of whom had SO stenosis.
Another report demonstrating the relationship of SOD with pancreatitis focused on the observation that a hypertensive pancreatic portion of the SO increases the risk of pancreatitis in patients undergoing ERCP with sphincterotomy [12]. Eighty patients with manometrically proven pancreatic sphincter hypertension were randomly assigned to stenting of the pancreatic SO segment after biliary sphincterotomy or sphincterotomy alone [12]. Patients who received a stent were significantly less likely to develop pancreatitis (7 versus 26 percent) [13]. The authors strongly suggested that stenting of the hypertensive pancreatic duct segment should be considered after biliary sphincterotomy in patients with SOD. (See “Overview of pancreatic stenting and its complications”).

The remainder of this discussion is continued elsewhere. (See “Clinical manifestations and diagnosis of sphincter of Oddi dysfunction”).

RECOMMENDATIONS — The evaluation of patients with suspected SOD should consider the availability of local expertise since the accuracy of the various diagnostic tests is highly operator dependent.

Evaluation of suspected biliary SOD — As discussed above, patients who are classified as Milwaukee Biliary Group Classification I usually respond to sphincterotomy, which can be performed without prior biliary manometry.

• For Class II patients in whom sphincterotomy is being considered, we recommend documentation of SOD with biliary manometry. Hepatobiliary scintigraphy (in patients whose gallbladder is intact) or a fatty meal ultrasound study (in patients with or without a gallbladder) is a reasonable alternative provided that local standards for these tests have been established (show algorithm 1). On the other hand, some authorities recommend empiric sphincterotomy in such patients, a strategy that was supported by a cost-effectiveness analysis [57].
• Patients who are classified as Milwaukee Biliary Group Classification III are commonly encountered and represent a difficult diagnostic challenge. A statement issued during the National Institutes of Health State of the Science conference on ERCP suggested that the diagnosis and management of type III SOD was the most difficult [58]. Invasive procedures should be delayed or avoided in such patients if possible. ERCP with SO manometry and endoscopic sphincterotomy should ideally be performed at specific referral centers and preferably in randomized controlled trials. Symptoms in these patients may be due to SOD or functional bowel diseases such as dyspepsia or irritable bowel syndrome. To further complicate matters, these conditions may coexist and represent generalized smooth muscle dysfunction or duodenal hyperalgesia [59]. The following recommendations apply to this group of patients.

Our approach begins with evaluation focused on detecting clinical features associated with irritable bowel syndrome or functional dyspepsia (show algorithm 1). Patients whose clinical features are more consistent with one of these diagnoses are treated accordingly. (See “Clinical manifestations and diagnosis of irritable bowel syndrome” and see “Approach to the patient with dyspepsia”).

For other patients we proceed based upon whether or not the gallbladder is intact. In patients who have not undergone cholecystectomy, we obtain a gallbladder ejection fraction to determine whether the gallbladder may be the source of symptoms. For those who have an ejection fraction <40>40 mmHg in all three recording channels). Criteria for SO “dyskinesia” included: an increased incidence of phasic SO contractions (>7/min), an increased incidence of retrograde propagated SO phasic contractions (>50 percent) and a paradoxical contraction response of the SO to cholecystokinin octapeptide bolus injection. So manometry was repeated at three and 24 months. A clinician unaware of the results of manometry or randomization followed the patients for two years.

Twenty-six of the 81 patients (32 percent) had SO stenosis based upon the manometric results. In the SO stenosis group (predominantly Type II patients) 11 of 12 (85 percent) showed long-term improvement after sphincterotomy. In contrast, in the sham group with stenosis only 5 of 13 patients (30 percent) improved, an insignificant number. The other manometric categories (SO normal and SO dyskinesia) showed no significant improvement.

Recurrent pancreatitis — The benefit of endoscopic pancreatic sphincterotomy for patients with pancreatitis thought to be related to SOD was evaluated in a series of 160 patients, of whom 64 percent had complete and long-lasting resolution of symptoms [20]. Some patients required repeat sphincterotomy for recurrent SO stenosis.

Pancreatitis is a potential complication of sphincterotomy. The risk may be greater after pancreatic than biliary sphincterotomy and most often occurs in patients with pancreatic sphincter hypertension [21]. Placement of a pancreatic stent following biliary sphincterotomy may reduce the incidence of pancreatitis in these patients. This was illustrated in a prospective, randomized, controlled trial of 80 patients with manometrically proven pancreatic sphincter hypertension; stenting of the pancreatic SO segment after biliary sphincterotomy significantly reduced the risk of pancreatitis (7 versus 26 percent with no stenting) [22]. Patients who did not receive a stent were significantly more likely to develop pancreatitis if their pancreatic duct was not patent (33 versus 0 percent), supporting the hypothesis that adequate drainage of the pancreas is necessary to prevent pancreatitis following sphincterotomy. (See “Overview of pancreatic stenting and its complications”).

A problem with pancreatic duct stenting is the requirement for another procedure to remove the stent. Nasopancreatic drainage may be an alternative, permitting noninvasive removal of the drain following recovery [20].

Surgical sphincterotomy for recurrent pancreatitis is associated with worse outcomes than for recurrent biliary pain (see below). A possible explanation is that patients with recurrent pancreatitis have developed irreversible structural pancreatic disease that does not improve after sphincterotomy. Although this observation has not been proven for endoscopic pancreatic sphincterotomy, it also probably applies.

Botulinum toxin injection — Endoscopic injection of botulinum toxin for biliary sphincter of Oddi dysfunction has been used successfully by some groups to determine if a symptomatic response might predict a successful outcome to subsequent sphincterotomy [23]. The influence of botulinum toxin on the pancreatic segment of the SO is unknown. Whether this material could be injected successfully into the pancreatic segment without a preceding biliary sphincterotomy is also not known.

SURGERY — Biliary and pancreatic sphincterotomy can also be accomplished by a transduodenal surgical approach. Surgical sphincterotomy has two potential advantages compared to standard endoscopic approaches:

It is difficult to sever the transampullary septum during conventional endoscopic sphincterotomy without risking duodenal perforation. As a result, endoscopic sphincterotomy may not completely relieve pancreatic duct obstruction [24]. Endoscopic sphincterotomy of the biliary segment of the SO may not effect the pancreatic duct segment at all [25].
Surgery may reduce the chance of recurrent stenosis due to scarring.
Despite these potential advantages, endoscopic therapy is less invasive, has similar outcomes, and is the preferred approach at most centers with experience in this technique.

Improvement in biliary pain and recurrent pancreatitis have been demonstrated in approximately 50 to 60 percent of patients treated by surgical sphincterotomy [26-32]. However, surgical sphincterotomy and septoplasty for pancreatitis may be associated with worse outcomes than for recurrent biliary pain [32]. One possible explanation is that some patients have unappreciated pancreatic parenchymal disease that continues to be the source of pain following surgery. For similar reasons, patients with acute recurrent pancreatitis may have better results than those with chronic pancreatitis [30].

Published in: Medicine on July 28th, 2006

To view information on another disease, click on Pancreatitis SOD Library.!

Pancreatitis SOD Library

..