Article Text


Recurrent bile duct stones after endoscopic sphincterotomy
  1. S Sultan,
  2. J Baillie
  1. Duke University Medical Center, Durham, NC, USA
  1. Correspondence to:
    Dr J Baillie
    Duke University Medical Center, Rm 0339 Orange Zone, South Hospital, Trent Drive, Durham, NC 27710, USA;

Statistics from

Multiple endoscopic retrograde cholangiopancreatographies (even three or more) for recurrent bile duct stones appear to be safe and effective. Late complications are relatively frequent but are endoscopically manageable

In 1974, Professors Meinhard Classen (Germany) and Keiichi Kawai (Japan) independently published their pioneering experience of endoscopic biliary sphincterotomy (EBS).1,2 This bold innovative work was performed in the face of considerable antagonism from the surgical community, which considered that endoscopists had no business in the biliary tree, traditionally regarded as surgical “turf”. In the 30 years since, countless thousands of patients have benefited from EBS, being spared major surgical and radiological procedures. That an electrosurgical incision of a centimetre or less can have such beneficial effect is testament to the ingenuity of the pioneer endoscopists and their partners in industry. EBS is now an everyday event in endoscopy units around the world, the great majority being completed safely and achieving the desired goal. Indeed, we tend to take EBS for granted. The recent National Institutes of Health Consensus Conference Panel deemed endoscopic retrograde cholangiopancreatography (ERCP) and EBS to be essential tools in the management of patients with acute cholangitis, pancreatitis, and choledocholithiasis.3 EBS has been shown to be safe and effective in experienced hands but it is undoubtedly associated with both short and long term complications.4

Recurrent bile duct stone formation is not uncommon following EBS. The literature suggests an incidence ranging from 4 % to 24%.5,6 Further ERCP with extension of the previous sphincterotomy is often the treatment of choice in patients with recurrent choledocholithiasis but little is know of the long term results and complications of this treatment. In this issue of Gut, Sugiyama and colleagues7 attempt to address these questions (see page 1856). In a retrospective review of ERCP cases seen between 1977 and 1998, the authors identified 84 patients who required repeat ERCP (with or without EBS) for recurrent bile duct stones. Over a mean follow up period of 11.7 years, 31 of 84 patients (37%) developed complications, including choledocholithiasis, cholangitis, and cholecystitis. The rate of recurrent stone formation was 31%. Twenty one of 26 patients (81%) with recurrent choledocholithiasis and four patients with acute cholangitis required repeat EBS. Three of these patients (10%) went on to develop a further recurrence of biliary stones, two being treated by further EBS. Using univariate and multivariate analysis, the authors found that “late” biliary complications of EBS were associated with the following risk factors: interval between initial EBS and second EBS <5 years, bile duct diameter >15 mm, and the presence of a periampullary diverticulum. They concluded that multiple ERCPs (even three or more) were safe and effective. They advocated careful follow up for patients with identifiable risk factors.

The authors’ data are interesting but are their findings in agreement with the existing literature on EBS? The rate of late complications of EBS in this series was 31%, significantly higher than previously reported. Indeed, the same group of authors previously reported a long term complication rate of 9.7% in a study that included 100 patients followed for a mean of 14.2 years.8 In a separate study, with a mean follow up of 15 years, late complications were seen in 24% of EBS patients.4 A possible explanation for this discrepancy is selection bias: inclusion of patients requiring a second ERCP may have biased the study in favour of biliary pathology. There may also have been unrecognised patient factors that predisposed some individuals to be “stone formers”. At the time of repeat ERCP, the authors noted the presence of a “non-enlarged biliary orifice” in 49 of 84 patients (58%). We believe that it is wrong to equate this appearance with papillary stenosis. A small sphincterotomy site may still be sufficient to allow bile to drain unimpeded into the duodenum. However, if we accept that these EBS sites were indeed stenotic, then we must ask why this complication was so common. Were the initial sphincterotomies too small? It is difficult to assess this, as 31 of the patients described had their EBS performed at other institutions. The technical ability and expertise of the endoscopists involved is unknown, further confounding the results. Many years ago, Dr Steven Silvis (Minneapolis) demonstrated in dogs that following EBS, the size of the orifice shrunk by approximately 50% over the first year and then remained relatively stable (personal communication). Referral centres occasionally see patients with symptomatic papillary stenosis following a timid sphincterotomy but we believe that in the last decade this has become an increasingly rare occurrence.

The authors examined a total of 15 risk factors and ultimately identified five that were significant in univariate analysis. These were then entered into a multivariate model which demonstrated three significant risk factors for stone recurrence. Rather than a hypothesis driven approach based on prior published studies, the authors chose to use a more exploratory approach. In the end, their findings were consistent with previous studies. They demonstrated three risk factors for recurrent choledocholithiasis after EBS, two of which (bile duct diameter >15 mm and periampullary diverticulum) had been described previously.9,10 A new risk factor identified by this study was interval between initial EBS and repeat ERCP <5 years.

The authors state that none of the patients with bile duct diameters >15 mm had choledochal cysts. One has to ask: how did they know? It can be very difficult to distinguish a dilated bile duct from congenital dilatation of the extrahepatic biliary tree (type I choledochal cyst).11 Choledochal cysts typically exhibit poor drainage, even after EBS, and are a “hot bed” for recurrent choledocholithiasis and cholangitis. The authors talk about periampullary diverticula causing “compression” of the distal bile duct. This is at odds with the current consensus that periampullary diverticula promote stone formation by a combination of bacterial overgrowth (with ascending infection) and motility disturbance (sphincter of Oddi dysfunction). This study is limited by its retrospective nature and the heterogeneity of the patient population. One wonders how many sphincterotomies are “enough”? Urban legend (but no hard data) holds that extending a prior sphincterotomy risks perforation. It seems that repeat EBS does the job (allowing access to remove stones, improving biliary drainage) although in the present study three patients had early recurrence of biliary symptoms after their second EBS.

What about balloon sphincteroplasty as an alternative to repeat EBS? Unfortunately, it appears that dilating the papilla rather than cutting it does not reduce the incidence of recurrent stones.12 The overall complication rates of EBS and balloon sphincteroplasty are approximately equal but the balloon technique carries an increased risk of post-procedure pancreatitis.13 That being the case, most ERCP endoscopists reserve balloon dilation of the papilla for special circumstances, such as coagulopathy that renders EBS dangerous. It would be interesting to know if balloon sphincteroplasty is equivalent to EBS in efficacy for managing recurrent choledocholithiasis after sphincterotomy.

Despite its limitations, the study of Sugiyama and colleagues7 suggests that late complications of EBS are more common than we imagine. The good news is that repeat EBS manages almost all of these episodes of late cholangitis and recurrent choledocholithiasis effectively. Patients who recurrently develop biliary sludge and stones despite EBS most likely have difficulty emptying a dilated bile duct. Sometimes the best results are obtained by surgery. Typically, a biliary bypass of some variety (for example, choledochoduodenostomy, choledochojejunostomy) will be performed at the end of open bile duct exploration and stone clearance. The presence of a choledochal cyst usually requires resection of the affected segment with a high Roux-type biliary diversion. Elderly patients often sail through anaesthesia and biliary surgery without turning a hair, and should be evaluated by an experienced biliary surgery before being labelled as “unsuitable for surgery/anaesthesia”. That EBS predisposes to bile duct cancer (cholangiocarcinoma) is another urban legend that this study effectively debunks: none was seen in this study population. Endoscopists since 1974 have been convinced of the old surgical maxim that “a chance to cut is a chance to cure”. We believe that this study supports that belief, with the caveat that recurrent choledocholithiasis and its complications may be more common than we care to believe.

Multiple endoscopic retrograde cholangiopancreatographies (even three or more) for recurrent bile duct stones appear to be safe and effective. Late complications are relatively frequent but are endoscopically manageable


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