INTRODUCTION — Endoscopic balloon dilatation of the biliary sphincter can be a valuable adjunct to the therapeutic arsenal of the biliary endoscopist for removal of bile duct stones during endoscopic retrograde cholangiopancreatography (ERCP) in highly selected patients.

This topic will review endoscopic balloon dilatation of the biliary sphincter. ERCP with endoscopic sphincterotomy is discussed separately. (See "Overview of endoscopic retrograde cholangiopancreatography (ERCP) in adults" and "Endoscopic biliary sphincterotomy".)

RATIONALE — Endoscopic balloon dilatation (EBD) permits effective stone removal with a success rate similar to that of endoscopic sphincterotomy (EST), the standard treatment for bile duct stones. In patients with small stones (≤10 mm), EBD permits successful stone extraction in virtually all cases without the need for an additional sphincterotomy or mechanical lithotripsy. In patients with more complicated stones (diameter >10 mm or number >3), the success rates of EBD and EST are comparable, but lithotripsy is required in approximately 50 percent of patients, and an additional sphincterotomy or repeat ERCP in 15 to 30 percent of patients [1].

EBD has a number of potential advantages compared with standard sphincterotomy.

●After EBD, the function of the biliary sphincter is preserved, whereas it is permanently lost after EST [2]. This has the potential benefit of preventing the chronic reflux of gastroduodenal contents into the biliary system observed after EST. However, although patients who have undergone EST may have bacterial colonization and chronic inflammation of the biliary epithelium, its clinical significance is uncertain since long-term follow-up studies after EST have not shown a high rate of serious complications [3,4].

●Compared with EST, EBD reduces the risk of bleeding after the ERCP and is therefore especially suited for the treatment of patients with hemostatic disorders [1,5,6].

However, despite these advantages, some reports have found a high rate of pancreatitis following EBD compared with EST, although the data have been conflicting [1,7-9]. As a result, most endoscopists are reluctant to perform EBD for stone removal.

Because of the above considerations, the main indications for stone removal after EBD are for patients who are at risk of bleeding after sphincterotomy, and patients in whom the local anatomy makes a sphincterotomy impossible or dangerous (eg, patients with a periampullary diverticulum or Billroth II gastrectomy) [10-13]. (See "Endoscopic retrograde cholangiopancreatography (ERCP) after Billroth II reconstruction".)

Another indication for EBD is to widen the sphincterotomy opening after EST to allow for removal of large bile duct stones [14-16]. For this purpose, large diameter EBD balloons (eg, 15 to 20 mm diameter) are used [14].

TECHNIQUE — After a diagnostic cholangiogram has confirmed the presence of bile duct stones and deep cannulation of the bile duct has been obtained, a 0.035 inch guidewire should be advanced through the diagnostic cannula and positioned high in the biliary tree [17]. A balloon catheter is then passed over the guidewire and positioned in the biliary orifice with the middle portion of the balloon located at the site of the biliary sphincter. A variety of balloon catheters are commercially available that differ in the maximum inflated balloon diameter, the length of the balloon, and/or maximum tolerated balloon pressure ("burst pressure"). The length of the balloon and its maximum inflated diameter are important issues that will be discussed below. The burst pressure may be important for dilatation of tight biliary strictures, but the amount of radial force exerted by inflation of the balloon is less important in dilating the relatively flexible and non-stenosed biliary sphincter.

Maximum inflated diameter — The first studies of endoscopic balloon dilatation (EBD) for stone removal were conducted in 1982 using a balloon catheter with a maximum inflated diameter of 15 mm [18]. In the late 1980s, interventional radiologists and surgeons were dilating the biliary sphincter to a diameter well above 10 mm [19]. However, the reintroduction of EBD for removal of bile duct stones was accompanied by a trend toward using smaller balloon catheters, with the aim of performing a less traumatizing and sphincter-preserving procedure. One group, for example, used a 10 mm dilatation balloon in their first series of 28 patients and subsequently switched to a balloon catheter with a maximum inflated diameter of 8 mm [9].

We have used 8 mm dilatation balloons from the beginning of our experience [1]. This catheter has become the standard dilatation balloon catheter for EBD of the biliary sphincter and is used in most studies on this technique. EBD of the biliary sphincter to a diameter of 8 mm preserves sphincter function, which has not been demonstrated for dilatation of the sphincter to 10 mm or more [2].

For very small stones (eg, <5 mm), dilatation of the sphincter to a diameter of 6 mm probably suffices. However, since spontaneous passage usually occurs with such small stones, only a few patients fall into this category. Furthermore, most 6 mm dilatation balloons are only available with a relatively short balloon length and are difficult to maintain in position during balloon dilatation since the balloon has the tendency to move either into the bile duct or out into the duodenum. Thus, we prefer dilatation balloon catheters with a minimum balloon length of 3 cm.

Despite the reports of successful dilatation with balloons ≥12 mm [20-22], in our opinion, dilating the biliary sphincter to ≥10 mm may be beyond the scope of minimizing the trauma applied to the sphincter of Oddi. As a result, for extraction of larger stones, we prefer to dilate the biliary sphincter to 8 mm and to crush the stones by mechanical lithotripsy instead of dilating the sphincter to a larger diameter.

Sphincter dilatation — Once the dilatation balloon is correctly positioned in the biliary orifice, sphincter dilatation is a simple procedure compared with endoscopic sphincterotomy for which many maneuvers (all important for the safety and efficacy of the procedure) require significant expertise [23,24]. During dilatation of the sphincter, the balloon is gradually filled with diluted contrast under endoscopic and fluoroscopic guidance to maintain the correct position of the dilatation balloon and to observe the gradual disappearance of the waist in the balloon, indicating the progress of dilatation of the biliary sphincter (image 1). Once the waist in the balloon has disappeared, we prefer to maintain the inflated dilatation balloon in position for 45 to 60 seconds before deflation and removal.

Other dilatation procedures have been suggested, including manometry-guided control [2,25] and multiple dilatation cycles [9]. Manometry is not imperative since the non-stenosed and relatively flexible biliary sphincter easily permits dilatation well before the burst pressure of the dilatation balloon is reached, in contrast to dilatation of the tight biliary strictures in patients with sclerosing cholangitis or postcholecystectomy stenoses.

Cannulation of the bile duct — In the majority of patients, cannulation of the bile duct is easily permitted after balloon dilatation of the biliary sphincter; a situation comparable to the straightforward cannulation sometimes encountered in patients with recent spontaneous stone passage through the papilla. However, repeated cannulation of the bile duct may occasionally prove difficult after EBD, especially in patients in whom the initial deep cannulation of the bile duct, prior to the balloon dilatation, was difficult. In our series, this problem was encountered in only 2 of 101 patients (2 percent) who underwent EBD for removal of bile duct stones [1]. This occasional difficulty with repeated entrance to the bile duct after EBD can be overcome by exchanging catheters and wire baskets over a guidewire left in situ after the removal of the dilatation balloon.

Stone extraction — Extraction of bile duct stones after EBD should be performed according to standard guidelines using wire baskets [23]. (See "Endoscopic management of bile duct stones: Standard techniques and mechanical lithotripsy".) When the basket reaches the biliary sphincter, some resistance is usually encountered that can be overcome by using the up/down knob on the endoscope and by inserting the endoscope slightly further into the duodenum. Extraction of stones up to 10 mm in diameter is typically safe and successful using this technique while applying traction force in the axis of the bile duct.

For larger stones, the amount of space created by dilatation of the biliary sphincter is usually insufficient for controlled stone extraction; thus, mechanical lithotripsy should be available. To prevent impaction of the wire basket, we prefer to start with the Olympus lithotriptor crushing basket to capture these larger stones (eg, >10 mm diameter). In most cases, this permits effective mechanical lithotripsy when the maneuvers described above fail, without having to release the stone from the basket [23].

When stone removal is unsuccessful after EBD, an additional sphincterotomy can be performed as an "escape procedure." Further attempts can be made to remove the remaining stones through the further enlarged biliary orifice.

Residual stones are less likely to pass spontaneously after EBD compared with endoscopic sphincterotomy. Thus, an occlusion cholangiogram should always be obtained to check for residual stones once stone extraction has been completed.

SUCCESS RATE — Success rates of stone extraction following endoscopic balloon dilatation (EBD) range from 85 to 100 percent in various reports [1,2,8,9,18,26]. A systematic review concluded that EBD was slightly less successful than endoscopic sphincterotomy (EST) [27]. However, comparison of results with known success rates of EST is hampered by several problems concerning the definition of success and selection of patients.

The definition of successful stone removal after EBD varies among studies. Some studies report the rate of successful stone removal after EBD only, considering the procedure to have failed if a sphincterotomy with additional stone extraction is required. On the other hand, others argue that only the presence or absence of bile duct stones at the end of the endoscopic retrograde cholangiopancreatography (ERCP) should determine the qualification "success" or "failure," and that the relative contributions of EBD and additional EST are secondary outcomes.

Some endoscopists are reluctant to perform a sphincterotomy in patients in whom stone removal was unsuccessful after EBD, and reschedule such patients for a separate procedure [8]. In these settings, the success rate of EBD may be overestimated since multiple endoscopic procedures were required.

Another problem in comparing the success rate of stone removal after EBD and EST concerns the criteria by which patients are selected for these procedures. Most studies on EBD have restricted the maximum stone diameter and sometimes also the maximum number of stones for inclusion in the study. It is inappropriate to compare the success rate of EBD in such a selected group of patients with that reported in most series on EST in which an "all-comers" policy is usually followed.

Randomized trials are imperative to compare the results and complications of EBD and EST. At least nine such trials have now been completed [1,2,7,26,28-32]. In addition, two systematic reviews [27,32], have summarized available studies.

The following conclusions can be drawn from these studies:

●For removal of stones of all sizes, EBD is slightly less effective than EST (RR 0.90, 95% CI 0.84-0.97). For patients with smaller and fewer bile duct stones, the overall success rates are comparable.

●Mechanical lithotripsy is more often required after EBD than after EST (RR 1.34, 95% CI 1.08 to 1.66), probably because EBD does not enlarge the papillary orifice to the same extent as EST. However, in patients with relatively "easy" bile duct stones (diameter ≤10 mm and number ≤3), mechanical lithotripsy after EBD is required in only 5 percent of cases [1].

In patients with smaller and fewer bile duct stones, EBD therefore almost always permits successful stone removal. In patients with larger or multiple stones, the bile duct can be cleared successfully in the majority of patients after EBD, but lithotripsy is required in about 50 percent of patients and an additional sphincterotomy or repeat ERCP in 15 to 30 percent.

EARLY COMPLICATIONS — The risk of bleeding appears to be decreased in endoscopic balloon dilatation (EBD) compared with endoscopic sphincterotomy (EST). Bleeding has been reported in 2 to 5 percent of patients undergoing EST for bile duct stones [24] and was observed in 2 percent of the EST patients in the aforementioned eight randomized trials. By contrast, no significant bleeding has been observed following EBD in a total of over 1000 reported patients [1,2,7-10,18,26]. Furthermore, two of these series included patients with cirrhosis and known hemostatic disorders [6,8], who have a six- to eightfold higher risk for bleeding post-EST [6,24]. This very low frequency of bleeding suggests that EBD is especially suited for patients with an increased risk of bleeding.

EBD has not evolved into widespread use because of concern related to complications, especially pancreatitis, which occurs more often after EBD than after EST (RR 1.96, 95% CI 1.34 to 2.89).

There are several theoretical reasons why EBD is associated with an increased risk for pancreatitis:

●Balloon dilatation produces circumferential trauma to the biliary sphincter and therefore partially in the direction of the pancreatic duct. This in contrast to EST where, ideally, the incision of the biliary sphincter is directed away from the pancreas.

●During balloon dilatation of the biliary sphincter the ampullary sphincter is also dilated. This sphincter, which is shared by common bile duct and main pancreatic duct at their most distal part, has a variable length and may measure several millimeters in patients with a long common channel (figure 1). Studies in pigs have demonstrated that transmural inflammation and intramucosal bleeding of the biliary sphincter (and possibly the ampullary sphincter) occurs immediately after EBD. This may result in a relative outflow obstruction of the pancreatic duct causing an increase in the rate of hyperamylasemia and/or pancreatitis. This may be especially relevant in patients in whom prior cannulation of the bile duct was difficult, with multiple unintentional cannulations of the pancreas.

●Hyperamylasemia and/or pancreatitis may occur more frequently after EBD because stone removal takes more time and involves more manipulation of the papillary complex.

LATE COMPLICATIONS — Most studies on endoscopic balloon dilatation (EBD) have reported only short-term follow-up. In two reports with one-year follow-up [8,9], no hepatobiliary problems were identified except for residual stones in 2 percent of patients in one report [8].

In the Amsterdam trial, the rate of complications was similar for patients who underwent EBD or endoscopic sphincterotomy (EST) (18 versus 23 percent) [1]. The majority of complications were due to symptomatic recurrent bile duct stones. Acute cholecystitis was observed significantly more often after EST than after EBD (10 versus 1 percent) [1]. This latter finding may be explained by preservation of sphincter function after EBD, preventing reflux of duodenal contents and bacteria. The origin of these "recurrent" bile duct stones is difficult to ascertain. Stone formation de novo in the bile duct seems unlikely given the relatively short period of follow-up. It is more likely that these stones passed from gallbladders in situ or were left behind at the initial endoscopic retrograde cholangiopancreatography.

One of the largest series with relatively long-term follow-up included 1000 patients who had undergone EBD for common bile duct stones; 837 were followed for a mean of 4.4 years [33]. Biliary complications occurred in 104 patients (12 percent), mainly in those in whom the gallbladder was left in situ and who were known to have residual gallstones. The most common complications in such patients were recurrence of common bile duct stones and cholecystitis. Three patients developed cholangitis. No long-term follow-up is yet available of any of the randomized studies, making it difficult to relate these results to the rate of late complications after EST.

Sphincter preservation — While the early complications after endoscopic sphincterotomy have been extensively studied, less is known about the long-term effects of endoscopic balloon dilatation of the biliary sphincter. We found that biliary sphincter function was absent up to 17 years after endoscopic sphincterotomy [3]. Permanent loss of the physiological barrier between duodenum and biliary tract was associated with bacterial colonization of the biliary tract.

The presence of bacteria in the biliary system (which is sterile under physiological conditions), reflux of gastroduodenal contents, and biochemical changes in bile composition after EST may cause late complications [3]. These complications may include recurrence of bile duct stones due to deconjugation of bilirubin by bacterial enzymes, inflammatory changes of the hepatobiliary system, and, in theory, malignant degeneration.

In a retrospective study of 100 patients, 15 to 17 years after EST for bile duct stones, late complications, primarily recurrent bile duct stones, occurred in as many as 24 percent of patients [4]. However, the related mortality was low (1 percent) and only one patient required surgical treatment; the remaining patients were effectively managed endoscopically or conservatively. More serious complications such as malignant degeneration were not observed. However, the low incidence of hepatobiliary malignancies in the general population suggests that the power of retrospective studies may be too small to detect an increased risk for this complication after EST.

Although long-term data are not yet available, EBD may reduce the risk of some long-term complications related to EST since sphincter function appears to be preserved following EBD. In one report, for example, sphincter manometry revealed that sphincter function was intact after EBD while it was absent after EST [2]. Similar findings were noted in a study in pigs in which the architecture of the biliary sphincter remained intact after EBD without signs of sphincter fibrosis [34].

TIME REQUIRED FOR ENDOSCOPIC STONE REMOVAL — In the Amsterdam trial, the median duration of endoscopic retrograde cholangiopancreatography (ERCP) was slightly longer for endoscopic balloon dilatation (EBD) compared with endoscopic sphincterotomy (EST) (50 versus 35 minutes) [1]. Multivariate analysis identified four significant parameters associated with the duration of ERCP: stone diameter, stone number, EBD or EST, and prior Billroth-II gastrectomy. In the EBD group, the median duration of ERCP markedly increased for stone diameters over 9 mm; in the EST group this increase was observed only when stone diameters exceeded 15 mm. Thus, the difference in time required for ERCP between EBD and EST was most pronounced for patients with stone diameters between 10 and 15 mm. In these patients, EBD usually required additional lithotripsy, whereas EST usually permitted stone extraction without prior lithotripsy. For patients with stones <10 mm in diameter, stone removal after EBD required about five minutes more than EST.

PRIOR BILLROTH-II GASTRECTOMY — Endoscopic sphincterotomy (EST) can be difficult and dangerous in patients who have undergone a Billroth-II gastrectomy. Several reports have described the use of EBD in this setting [11,25,35,36]. In the only randomized trial, complete stone removal was achieved in 14 of 16 EBD patients (88 percent) versus 15 of 18 EST patients (83 percent) [11]. Mechanical lithotripsy was required in three and four patients, respectively, and only one EBD patient required an EST for stone removal after unsuccessful EBD. The median time required for stone removal was 30 minutes in both groups. Complications occurred in seven EST patients (39 percent, including three patients with bleeding) and in three EBD patients (19 percent, including one patient with pancreatitis). These data and other reports suggest that EBD in patients who have undergone Billroth-II gastrectomy appears to be successful and safe and may be superior to EST.

ENDOSCOPIC PAPILLARY LARGE BALLOON DILATATION — Endoscopic papillary large balloon dilatation (EPLBD) with large-diameter (12 to 20 mm) dilatation balloons has been introduced as an adjunctive technique to enlarge the papillary orifice after endoscopic sphincterotomy (EST) to facilitate the removal of large or difficult stones.

The concept is to combine the advantages of EST with those of EBD. Theoretically, the risk of perforation or bleeding is reduced by performing a less than maximal EST, while the risk of pancreatitis from balloon dilatation is reduced by first separating the biliary and pancreatic orifices with EST [20].

In a randomized comparative study between EPLBD and EST, 200 consecutive patients with bile duct stones were treated with EST plus EPLBD (12 to 20 mm balloon diameter) or EST alone [21]. Outcomes were similar in terms of overall successful stone removal (97 versus 98 percent), large (15 mm) stone removal (94 versus 97 percent), and the use of mechanical lithotripsy (8 versus 9 percent). Complications were similar between the two groups (5 versus 7 percent). There was no difference in the rate of postprocedure pancreatitis.

Data have suggested that a shorter balloon dilation time lowered the risk of post-ERCP pancreatitis for patients undergoing EST combined with balloon dilation. In a trial including nearly 2000 patients with common bile duct stones, balloon dilation for 30 seconds reduced the rate of post-ERCP pancreatitis compared with dilation for 300 seconds (7 versus 15 percent; odds ratio 0.40, 95% CI 0.25-0.66), while the rates of successful stone extraction were not significantly different between groups [37].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Biliary infection and obstruction".)

SUMMARY AND RECOMMENDATIONS

●Endoscopic balloon dilatation (EBD) of the biliary sphincter can be a valuable adjunct to the therapeutic arsenal of the biliary endoscopist for removal of bile duct stones during endoscopic retrograde cholangiopancreatography (ERCP) in highly selected patients. EBD permits effective stone removal with a success rate similar to that of endoscopic sphincterotomy (EST), the standard treatment for bile duct stones.

●In our opinion, EST should remain the standard procedure for removal of bile duct stones. This is not because the results of EBD are disappointing, but because many of the arguments that led to interest in EBD (mainly concern related to complications from EST) have been put into better perspective. Prospective studies have shown that EST for bile duct stones is much safer than was once believed. The most important complication of endoscopic stone removal is pancreatitis, the one complication for which EBD may carry an increased risk. (See "Overview of endoscopic retrograde cholangiopancreatography (ERCP) in adults".)

●Concern related to long-term complications after EST was another consideration that spurred interest in EBD since EBD preserves the biliary sphincter. EST results in a permanent loss of sphincter function, leading to chronic duodenobiliary reflux with bacterial colonization and low-grade inflammation of the biliary system. However, the clinical consequences of these events remain unclear since long-term follow-up studies have not revealed any severe late complications.

●We still consider EBD to be a reasonable option in young patients (<40) with small stones who do not have additional risk factors for pancreatitis. Generally accepted risk factors for post-ERCP pancreatitis include younger age, female sex, pancreas divisum, sphincter of Oddi dysfunction, prior ERCP-induced pancreatitis, difficulty of cannulation, precut sphincterotomy, pancreatic duct injection, sphincter of Oddi manometry, and a small common bile duct (<5 mm), particularly in the setting of sphincter of Oddi dysfunction (see "Post-endoscopic retrograde cholangiopancreatography (ERCP) pancreatitis"). Stones in such patients can be removed easily without lithotripsy while preserving sphincter function. As noted in this review, however, there are no data available to substantiate this policy.

●We also consider EBD in settings where EST is more dangerous or difficult, such as in patients with risk factors for post-sphincterotomy bleeding and those who have undergone a Billroth-II gastrectomy (see "Post-endoscopic retrograde cholangiopancreatography (ERCP) bleeding" and "Endoscopic retrograde cholangiopancreatography (ERCP) after Billroth II reconstruction"). This may also hold true for the occasional patient with a periampullary diverticulum where the position of the papilla prohibits a safe sphincterotomy. Although no series are available on its use in this situation, EBD may be a safe and easy alternative to EST.

●The vast majority of patients with bile duct stones do not fall in one of the small subgroups outlined above. EST therefore remains the cornerstone of the endoscopic management of bile duct stone. EBD is a technique that should be reserved for special indications only.