Endoscopic papillary balloon dilatation (EPBD) has been suggested as an alternative to endoscopic sphincterotomy (EST) for the treatment of choledocholithiasis [1,2], despite its possible association with an unacceptably high risk of pancreatitis [3]. Because this technique dilates the biliary sphincter without requiring surgical incision, acute complications such as hemorrhage and perforation are less likely [2,4,5] and biliary sphincter function is preserved [6-8]. Despite the frequent use of EPBD, studies have rarely reported the appropriate ballooning time (BT). A single recent EPBD study demonstrated a decreased tendency for post-procedural pancreatitis with a short BT (vs. long BT), with no differences in the efficacy of bile duct stone extraction [9]. Because patients experience pain during an EPBD despite adequate pre-procedural medication, the reduction in BT is desirable if the outcome is not affected. We therefore evaluated the safety and efficacy of EPBD performed with short and long BTs.

The conventional, long BT group included 35 patients (19 men, 16 women; mean age, 66.2 ± 17.4 years), as did the short BT group (16 men, 19 women; mean age, 63.3 ± 13.6 years). No significant differences were found in bile duct diameter, stone number or size, or previous cholecystectomy between groups (Table 1). The EPBDs of three patients (8.5%) in each group used a large (> 15 mm diameter) balloon catheter.

EPBD achieved complete duct clearance in 33 patients (94.3%) in the long BT group and 34 patients (97.1%) in the short BT group; this difference was not significant (p = 0.808). All CBD stones were completely retrieved during the initial EPBD in 32 patients (91.4%) in the long BT group and 31 patients (88.6%) in the short BT group (p = 0.478). Complete stone removal was achieved during the second EPBD in one patient (2.9%) in the long BT group and two patients (5.7%) in the short BT group (p = 0.174). The mean numbers of sessions required for complete duct clearance were 1.2 and 1.1, respectively, for the long and short BT groups. No significant difference was observed in the rate of complete bile duct clearance between groups, independent of the use of mechanical lithotripsy (long BT, 97.1% vs. short BT, 100%; p = 0.811). One patient (2.9%) in each group required mechanical lithotripsy (Table 2); in these cases, the dilated biliary outlet was unexpectedly smaller than the largest stone after EPBD, preventing extraction. The stone of one patient in the long BT group could not be extracted because it was impacted in the CBD lumen, and the basket could not open properly. This stone was extracted using peroral transpapillary cholangioscopy with electrohydraulic lithotripsy.

Pancreatitis was noted in four patients (11.4%) in the long BT group and two patients (5.7%) in the short BT group (p = 0.393). All post-ERCP pancreatitis observed in this study was mild and resolved without complications. Asymptomatic hyperamylasemia was present in four patients (11.4%) in the long BT group and three patients (8.6%) in the short BT group (p = 0.690). No other complications, such as hemorrhage, perforation, or cholangitis, occurred in either group. Serum amylase levels measured the morning after ERCP were 346.8 ± 55.6 IU/L in the long BT group and 198.2 ± 39.5 IU/L in the short BT group; this difference was significant (p = 0.037) (Table 3).

EPBD is technically easy and can be readily performed by endoscopists trained in standard therapeutic ERCP techniques [11]. Concerns regarding the increased incidence of ascending cholangitis and the possible recurrence of bile duct stones after EST [12,13], especially in younger patients, have led to the revival of EPBD. However, no agreement has been reached on the appropriate BT, which has been determined primarily by the operator's experience with the procedure. Balloon dilation has been used on the strictures of the esophagus, pylorus, and colon. BT must be sufficient for the dilation of the stricture and the reduction of restenosis risk. However, the papilla is a physiologically narrow area and should lack fibrosis unless a benign or malignant biliary stricture is also present. Reduction in BT may therefore help to preserve the physiological function of the papilla after EPBD, and may also reduce the risk of post-ERCP pancreatitis by curtailing the obstruction time of the pancreatic duct orifice. The sphincter is thought to be adequately dilated when the waist of the balloon disappears completely. In our experience, the waist disappeared within 20 seconds of achieving the target diameter in most EPBD procedures. We therefore found it reasonable to determine the effective minimum BT as the time required for the waist to disappear (20 seconds), and further believed that this minimum BT would be sufficient to dilate the papilla.

Patients may experience abdominal pain during EPBD, and the procedure has been associated with an increased risk of pancreatitis. Longer BTs further exacerbate these disadvantages [14]; BT reduction may conversely help alleviate these problems. Previous studies have reported BTs ranging from 30 to 120 seconds, with most operators using a 60-second BT [11,15-20]. The present study found comparable outcomes in the short and long BT groups, measured by the overall technical success of stone removal. Our results suggest that BT may be reduced when EPBD is performed.

This study found lower rates of post-procedural pancreatitis and hyperamylasemia in the short BT group than in the long BT group, but these differences were not significant. Post-procedural serum amylase levels were significantly lower in the short BT group than in the long BT group. The compression of the papilla or the pancreatic duct orifice by ballooning may cause papillary edema or a sphincter of Oddi spasm during the EPBD procedure, which obstructs the flow of pancreatic juice and eventually results in a pancreatic insult [21-23]. Many factors may act independently or in concert to cause post-ERCP pancreatitis [24-26]; high-risk predictors include suspected sphincter of Oddi dysfunction (SOD), young age, history of post-ERCP pancreatitis, number of contrast injections into the pancreatic duct, and experience of the endoscopist. This study did not assess risk factors for post-ERCP pancreatitis, but our data suggest that a shorter BT may reduce the frequency of pancreatic insults during an EPBD, thus leading to a lower incidence of pancreatitis. A review of eight EPBD studies in which BT was specified (Table 4) found that BT ranged from 30 to 120 seconds, based on the operator's personal experience, and that a single EPBD was performed in most cases [11,15-20]. Two or three EPBDs were performed in two studies [17,20]. A comparison of efficacy and early complications according to BT showed no significant differences in the results of previous studies. Several confounding variables, however, prevented us from ruling out the possibility of correlations. These variables included the endoscopist's experience and the number of pancreatic duct cannulations.

Tsujino et al. [27] reported immediate and long-term complications associated with EPBD in 1000 patients; they initially inflated the balloon with air or diluted contrast material for 2 minutes, but reduced BT after one patient developed severe post-procedural pancreatitis. In the modified procedure, the balloon was inflated slowly for 1 to 2 minutes with diluted contrast, until the waist disappeared. The pressure was maintained for 15 seconds and they achieved complete removal of CBD stones with EPBD. The results of this study suggested that a short BT was sufficient for the removal of CBD stones, and that the ballooning speed may have been an important factor associated with complications; a slowly inflated balloon appeared to be less harmful to the ampulla of Vater, as it decreased post-procedural complications.

In conclusion, this study found that EPBD with a short, 20-second BT was safe and effective. The outcomes of procedures using a short BT were comparable to those using a conventional, 60-second BT for the treatment of bile duct stones. In addition, a short BT tended to be associated with a decreased incidence of post-ERCP pancreatitis and hyperamylasemia. Large prospective population studies are required to confirm the safety and efficacy of EPBD with a short BT, and to empirically define the appropriate BT for EPBD.