Article Text


Can endoscopic papillary balloon dilation really preserve sphincter of Oddi function?
  1. Division of Gastroenterology, Brigham and Women's Hospital
  2. Boston, Massachusetts, USA
  3. dcarrlocke{at}

Statistics from

See article on page 686

The technique of endoscopic balloon dilation of the major duodenal papilla with extraction of bile duct stones was first reported in 19821 but an unacceptably high rate of acute pancreatitis following the procedure impeded its acceptance until further reports in 19942 and 19953 reported its safety. A number of randomised controlled trials have demonstrated its effectiveness in clearing the bile ducts of stones compared with endoscopic sphincterotomy4-8 with variable short term complication rates similar to sphincterotomy. Currently, this technique has not become popular worldwide, but for reasons that are not clear has been adopted much more readily in Japan and Korea than elsewhere.

Those who advocate this technique over sphincterotomy do so on the premise that: (1) it is less traumatic, (2) it is simpler to perform, (3) it is as effective, (4) it avoids the immediate complications, such as bleeding and perforation, (5) it is cheaper or at least equivalent cost, (6) it preserves sphincter of Oddi function, and (7) it avoids the long term complications.

Many of these apparent advantages have been questioned in terms of technical performance and short term outcome and now there is new evidence from the report by Yasuda et al in this issue of Gut 9 that the long term advantages may also be in doubt (see page 686).

The above selection criteria in favour of balloon dilation of the papilla have been criticised and are discussed below under the same categories.

(1) Forcible balloon dilation of the papilla well beyond its normal resting diameter may be just as traumatic or more traumatic than its incision by sphincterotomy and indeed, the description of the degree of trauma is “in the eye of the beholder”. (2) The technique is not necessarily simpler to perform than sphincterotomy, especially as current over-the-guidewire hydrostatic balloons are not designed for this application and may be difficult to keep in position during the dilation process. (3) The overall effectiveness of clearing the bile ducts is statistically equivalent to sphincterotomy although this is at the expense of a far greater need for mechanical lithotripsy and a small proportion of patients requiring additional procedures to achieve clearance. Recent use of intraductal ultrasound has cast doubt on the completeness of ductal clearance after balloon dilation with a retained stone rate of up to 33%. (4) The immediate complication rates, particularly acute pancreatitis, seem to be equivalent to sphincterotomy4-8 although the disturbing occurrence of severe pancreatitis and even death after balloon dilation in one recent randomised study6 calls safety into question if patients with unsuspected sphincter of Oddi dysfunction are treated in this way. (5) In the absence of prospective comparisons of costs, it seems unlikely that balloon dilation is cheaper. Even if a disposable balloon and sphincterotome are equivalent costs, the increased use of mechanical lithotripsy and some need for secondary procedures will have an adverse impact on this. (6 and 7) Studies to date4 10 have suggested recovery of sphincter function after balloon dilation within a few weeks of the procedure but the study by Yasuda et al in this issue ofGut 9 has extended manometric follow up to one year and shown some recovery but to significantly subnormal pressures compared with pre-procedure values. Although the authors discussed the possibility that the degree of “injury” might be dependent on the size and number of stones after balloon dilation, they were unable to show any correlation because of the small numbers involved in each stone size category. The implication is that the trauma of extracting stones or fragments of a size close to or at the limit of the degree of dilation of the papilla induces further injury which may not be able to recover completely. The clinical significance of this is uncertain but may be reflected in differences in long term complications between balloon dilation and sphincterotomy. In their combined retrospective and prospective study of a larger number of patients,9 a significantly lower incidence of biliary complications for balloon dilation included recurrent bile duct stones, cholangitis, and cholecystitis. Individual complications were significant for cholangitis which however did not occur in post-balloon dilation patients but was present in 3.2% of post-sphincterotomy patients, and cholecystitis occurred in 2% and 8.8% of patients, respectively. The numbers were too small to distinguish whether or not the presence of cholelithiasis had a significant effect on this. The incidence of recurrent bile duct stones was 10% and 14%, respectively, with no significant difference. In the smaller prospective part of their study, there were no significant differences in biliary complications at one year between the balloon dilation and sphincterotomy groups, although the recurrent stone rate was only 5.7% and 8.6%, respectively, and the cholecystitis rate 3.3% and 3.8%, respectively, with no reported incidence of cholangitis. As in previous studies, prospectively randomised patients showed no differences in acute complications which averaged 7%, 80% of which were pancreatitis. The groups were comparable with respect to number of stones (l–16), diameter of the largest stones (4–24 mm), and diameter of the bile duct (6–30 mm) but the use of mechanical lithotripsy was significantly higher in the balloon dilation group (17% compared with 9% in sphincterotomy patients).

With this new additional long term evidence it seems likely that endoscopic balloon dilation of the papilla for the treatment of choledocholithiasis would be considered as an alternative to sphincterotomy in certain circumstances but not as a replacement for it on a routine basis. Many centres have already adopted the practice of using balloon dilation in patients with previous surgically altered anatomy where access to the papilla is from the distal side and also in patients where there may be an uncorrectable coagulopathy which might carry increased risks of post-sphincterotomy bleeding. The converse of this discussion is the question of whether endoscopic sphincterotomy, particularly in the young patient, is all that undesirable to consider avoidance. The arguments were discussed in a leading article by Tham and colleagues11 indicating the expected rates for effectiveness, acceptable short and long term morbidity, apparent lack of serious consequences following sphincterotomy, general high standard of training in these techniques, together with the growth in laparoscopic management of choledocholithiasis during cholecystectomy.

The possible unsubstantiated differences in patient populations with choledocholithiasis in different countries may account for some of the differences in early morbidity with balloon dilation between the Far East and the West. I would currently advocate caution in the use of endoscopic dilation of the papilla for routine treatment of choledocholithiasis, except for patients in special circumstances, as detailed above, and then only for those with stone diameters up to the size of the balloon being employed to avoid excessive use of mechanical lithotripsy which in itself may have some influence on the complication rate. Unquestionably, further long term follow up studies are required to substantiate the findings by Yasuda and colleagues9 but for the present time, sphincterotomy, a tried and tested technique now well into its third decade, would still seem to be the standard endoscopic modality to be offered to most patients with choledocholithiasis.

See article on page 686


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