Article Text


Morphine and biliary pain revisited
  1. E A SHAFFER, Professor and Head
  1. Department of Medicine, University of Calgary and
  2. Clinical Regional Department Head, Internal Medicine
  3. Calgary Regional Health Authority and
  4. Department of Medicine, Foothills Hospital
  5. 1403–29th Street NW, Calgary, Alberta T29 T2N, Canada
  6. Email: Eldon.Shaffer{at}

    Statistics from

    See article on page 838

    Pain in the biliary tract is assumed to originate from either an obstructive event (the gall bladder contracting on a closed cystic duct, blocked by a gall stone) or inflammation (cholecystitis). Neither situation should be present after the gall bladder is extricated yet pain may persist.

    The “post cholecystectomy syndrome” is a poorly defined entity which includes many symptoms that range from the trivial (mild dyspepsia) to severe attacks of abdominal pain and jaundice.1 Depending on the original indication for surgery, 20–50% of patients after cholecystectomy continue to have a variety of non-specific abdominal symptoms such as flatulence and dyspepsia.2-4 True biliary type pain occurs as a relatively small percentage (14%) of symptoms which follow cholecystectomy.5 From a different perspective, such pain following cholecystectomy was a complaint in only 0.6% of men and 2.3% of women in a household survey.6 When symptoms persist following cholecystectomy and no organic basis is evident (from a stricture, retained common duct stone, pancreatitis, tumour, or congenital anomaly) by default the basis becomes “functional”. With the gall bladder gone and the bile ducts (which lack a smooth muscle layer) acting as a simple conduit, the sphincter of Oddi (SO) becomes the prime suspect for the pain through spasm or biliary obstruction. The sphincter, strategically located at the junction of the biliary tract, pancreas, and duodenum, is like a mysterious sphinx, an enigma wrapped in a puzzle. Involvement of the SO may take the form of: (1)stenosis (also termed “papillary stenosis”) from glandular hyperplasia, muscle hyperplasia/hypertrophy or fibrosis; or (2) dyskinesia due to muscular incoordination or muscular hypertonicity (spasm).

    The multinational (Rome II) consensus classified sphincter of Oddi dysfunction into: (a) biliary-type and (b) pancreatic-type.7 The diagnostic criteria for SO dysfunction consist of episodes of severe steady pain located in the epigastrium and right upper quadrant in addition to all of the following: (i) episodes lasting 30 minutes or more; (ii) symptoms occurring on one or more occasions over the previous 12 months; (iii) constant pain which interrupts daily activity or requires consultation with a physician and; (iv) no evidence of structural abnormalities to explain the symptoms. Elevated serum aminotransferases, alkaline phosphate, or conjugated bilirubin, and/or pancreatic enzymes (amylase/lipase) support the diagnosis. Although SO dysfunction can exist in the presence of an intact gall bladder, it is more frequently diagnosed following cholecystectomy.

    Based on the assumption that these symptoms are caused by abnormal motor function of the SO, the only direct assessment is SO manometry which requires endoscopic retrograde cholangiopancreatography (ERCP). Unfortunately, this invasive procedure is associated with an incidence of up to 24% of developing pancreatitis. SO manometry, pioneered by Hogan and Geenan from Milwaukee, classified SO dysfunction into three types (table 1).8 9 These were differentiated according to the presence of objective evidence of biliary obstruction, as identified by: (a) laboratory tests—elevated aminotransferase, alkaline phosphatase, or bilirubin (and for pancreatic SO dysfunction, amylase/lipase); and (b) cholangiographic features—delayed contract drainage and a dilated common duct. The predictability of SO dysfunction and a good response to sphincterotomy varies from a high of 65–95% in type I to 50–63% in type II, but decreases to 12–28% in type III.7 9 The increased likelihood of sphincteric dysfunction in patients with type I has led to the recommendation that manometry is not required; they should proceed directly to endoscopic sphincterotomy.7 The Rome II consensus recommended biliary manometry for those with type II and suggested that it be considered in type III.

    Table 1

    Milwaukee classification of biliary sphincter of Oddi dysfunction4 5

    Dysfunction of the pancreatic sphincter of Oddi differs. Classification into three comparable types is not yet possible. Pancreatic SO dysfunction may be responsible for recurrent pain and pancreatitis with elevated pancreatic enzymes. Manometry of the pancreatic sphincter is necessary for its diagnosis.7

    There is an obvious need for non-invasive screening criteria to detect dysfunction of the biliary sphincter. Pain provocation tests using morphine with or without prostigmine have limited sensitivity and specificity.10 Similarly, the use of a fatty meal or infusion of cholecystokinin to unmask a partially obstructed bile duct on ultrasonography has not gained acceptance.7 Because SO tone predominantly regulates bile flow into the duodenum following cholecystectomy, nuclear medicine studies have used radiopharmaceutical markers of bile to time their clearance from the biliary tree. Difficulties arise in the presence of intrahepatic cholestasis. Such computer assisted choledochoscintigraphy identifies SO dysfunction as a delayed clearance of marker from the biliary system11 or a prolonged transit from the hepatic hilum to the duodenum.12 Although the Rome II consensus recommended this as a valid screening test, its precision is less than ideal.

    Thomas et al in this issue13(see page 838) used morphine provocation to elicit the hypersensitivity of the dysfunctional SO to opiates and so increase the detection accuracy of biliary scanning. In 34 patients with type II (n=21) or type III (n=12) SO dysfunction, routine 99m Tc-IDA quantitative biliary scanning could not distinguish between those with normal (n=16) and those with abnormal (n=18) sphincteric tone (abnormal >40 mm Hg). With low dose morphine provocation (0.04 mg/kg intravenously over five minutes), differences became evident with a sensitivity of 83% and a specificity of 81%. Further, 14 of the 18 patients with abnormal SO manometry experienced biliary pain after morphine compared with two of 16 with normal manometry. Whether or not those patients with positive tests would have benefited from sphincterotomy is unknown as the study did not examine this therapeutic end point.

    One must be careful in attributing all biliary-type pain to a motility disorder affecting only the sphincter of Oddi. The abnormality could reside elsewhere as a more generalised motor disorder of the gut, from the oesophagus14 through the duodenum-jejunum,15 or perhaps as part of the spectrum of the irritable bowel syndrome.16 Altered sensation in the form of visceral hypersensitivity is a potential basis for pain in functional gastrointestinal disorders. An abnormal sensitivity to a relatively innocuous stimulus could result from modified receptor sensitivity at the level of the sphincter or adjacent viscus, increased excitability of neurones in the dorsal horn of the spinal cord, and/or altered central modulation of sensory inputs. The bile ducts per se could be the pain trigger zone.17 Conversely, a recent study in patients with type III SO dysfunction implicated duodenal hyperalgesia as the basis for the biliary-type pain.18 The rich neural connections between the gall bladder and SO, and the biliary tract and the stomach and small intestine orchestrate the interplay between motility and sensitivity of the biliary system and the foregut.

    Use of biliary scans with morphine provocation should provide a more effective non-invasive screening test before embarking on manometry and/or sphincterotomy. If combined with barostat-type evaluations of duodenal hypersensitivity, the mystery of the sphincter of Oddi—a motility disorder or a supersensitive sphincter/viscus as the cause of the pain—might be solved.

    See article on page 838


    View Abstract

    Request permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Linked Articles