In the current study, LBD without EST was as effective and safe as EST plus LBD in patients with large bile duct stones. To our knowledge, this is the first study to compare the efficacy and safety of LBD alone with EST plus LBD for the treatment of large bile duct stones.
EPBD was originally devised to extract CBD stones while minimizing damage to the sphincter of Oddi. However, the drawback of EPBD compared with EST is the more limited size of the papillary opening. Approximately 10% of bile duct stones are difficult to remove using conventional techniques, and for these patients, ML is generally the next step [17–19]. However, ML is time-consuming, has a potential for injury of the EST site or bile duct, and may be complicated by impaction of the stone-capturing basket. Moreover, because small stone fragments after ML may act as nidi for stone recurrence, ML is one of the risk factors for recurrent bile duct stones after endoscopic stone extraction . The main purpose of LBD is to avoid or lessen the use of ML for removal of large CBD stones and to reduce complications that may be related to ML.
Ersoz et al.  first reported the use of EST followed by EPBD with large-diameter (12–20 mm) balloons as an alternative technique for bile duct stones that are difficult to remove by standard methods. Complete stone retrieval without ML was successful in 54 (93.1%) of 58 patients, and stone clearance was achieved by ML in 4 (6.9%) patients. Complications occurred in nine patients (15.5%), including two (3.4%) with mild pancreatitis. Several studies have since been conducted using EST plus LBD for large, difficult bile duct stones [10–13, 21, 22]. Figures for overall stone clearance have ranged from 95% to 100%, with ML required for 1% to 27%. Complication rates have also varied from 0% to 8.3%, with pancreatitis between 0% and 4.5%. In most previous studies using EST plus LBD for removal of large CBD stones [9–13, 21, 22], the authors suggested that this technique may be associated with a lower risk of pancreatitis because EST prior to LBD may result in separation between the pancreatic and biliary orifices, and it can guide the direction of balloon dilation toward the bile duct rather than the pancreatic duct during LBD. However, recent two studies indicate that LBD without EST may be safe and effective in patients with large CBD stones [14, 15]. In a retrospective preliminary study , overall successful stone removal was achieved in 37 (97.4%) of 38 patients, and ML was required in 8 (21.1%) patients. A mild degree of postprocedure pancreatitis developed in only one (2.6%) patient. The authors proposed that a prior EST before LBD may not play an important role in the guidance of balloon dilation toward the bile duct. They also suggested that ML may induce papillary edema or spasms that may obstruct the pancreatic duct orifice. Thus, LBD may lower the incidence of pancreatitis by reducing the need for ML when removing large bile duct stones. In addition, because LBD is not performed on a nondilated CBD, which is one of the risk factors for post-ERCP pancreatitis, LBD may not carry the same risk of postprocedure pancreatitis as EPBD with a balloon catheter diameter of ≤10 mm for the removal of CBD stones . In another retrospective study , overall complete stone clearance was achieved in 229 (92.7%) of 247 patients, and ML was needed in 39 (15.8%) patients while retrieving the stones. There were nine (3.6%) complications, including two (0.8%) cases of mild pancreatitis. In the present study, the rates of overall stone clearance and complete stone removal without ML were similar between the two groups (96.8% vs. 95.7%, P = 0.738; and 80.6% vs. 73.9%, P = 0.360, respectively). The pancreatitis rates were similar between the two groups (6.5% vs. 4.3%, P = 0.593), and all cases were mild and self-limiting. The progressive decline in pancreatic exocrine function with aging may protect older patients from pancreatic injury, and one meta-analysis comparing EST and EPBD for bile duct stones demonstrated that age of <60 years was one of the factors related to a higher rate of pancreatitis in patients with EPBD [23, 24]. Therefore, the relatively old age of the patients in the current study may explain these results.
An additional purpose of LBD is to reduce complications by avoiding full-incision EST (major EST) in patients with large CBD stones. Although the reported bleeding rates from previous studies involving LBD range from 0% to 9% [9–15, 21], several reports on the performance of major EST before LBD showed a relatively high incidence of bleeding (8.3%–9%) [9, 11]. In this study, minor EST was performed before LBD in the EST plus LBD group and clinically significant hemorrhage was not noted in either group.
Other complications occurred in two patients in the EST plus LBD group. Perforation resulting from a duodenal wall tear opposite the major papilla occurred in one patient. It occurred during stone removal with a basket after LBD and ML and was caused by the tip of the duodenoscope. This complication was found shortly after complete stone removal. A basket impaction occurred in the other patient. The basket capturing the stone was broken during ML, and a portion of the broken basket remained in the bile duct. These complications were not related to LBD, and both patients recovered with surgical intervention.
Previous definitions of technical success have varied by publication [9–13, 21]. To define technical success, the frequency of required examinations may be used, but is often subject to the endoscopist. Moreover, the goal of LBD in managing large CBD stones is to avert ML and its potential complications. In the present study, we defined technical success as complete removal of CBD stones by performing LBD without an additional procedure such as ML, and we did not take into account the number of endoscopic sessions. In a retrospective study of LBD alone for retrieval of large CBD stones , patients in the treatment failure group showed a tendency to have a greater transverse stone diameter and smaller balloon/stone diameter ratio than patients in the treatment success group (20.8 ± 6.5 mm vs. 16.7 ± 3.9 mm P = 0.077] and 0.80 ± 0.23 vs. 0.96 ± 0.19 P = 0.066], respectively). In another retrospective multicenter study of EST plus LBD for bile duct stone removal , the median maximum stone size in patients undergoing ML was significantly larger than that in patients who did not undergo ML (16.7 vs. 13.3 mm, P < 0.01). In this study, treatment failure was associated with larger transverse stone diameters compared with treatment success and smaller balloon/stone diameter ratios. These results suggest that ML is more frequently used with larger stone sizes and that using a balloon catheter with a diameter smaller than the maximum transverse diameter of the stone causes resistance at the ampullary opening during stone removal with a basket or retrieval balloon catheter. Thus, the diameter of the balloon should exceed the maximum transverse diameter of the stone, but not the diameter of the bile duct.
Our patient group is small and the study was limited by its retrospective nature. Moreover, our study included many older patients who may be related to a lower rate of postprocedure pancreatitis. Therefore, the efficacy and safety of LBD alone in relatively young patients with large CBD stones remains uncertain. Randomized studies comparing LBD alone and EST plus LBD should be conducted in order to confirm our results.