In this study, we retrospectively analyzed the outcomes of therapeutic BE-ERCP for CBD stones in RYG patients treated at three tertiary institutions. Initial biliary intervention, including complete stone extraction or biliary stent placement, was successful in 78% (62/79), complete stone extraction was initially achieved in 53% (42/79) and ultimately in 63% (50/79), and adverse events occurred in 5% (5/79). In addition, we identified two independent factors affecting complete stone extraction: the largest CBD diameter ≥ 14 mm was a negative factor, and the retroflex position was a positive factor, especially in difficult cases with a large CBD diameter or stone size. Thus, this study was the first to clarify the efficacy and safety of therapeutic BE-ERCP for CBD stones in RYG patients and identify the factors affecting complete stone clearance.

To achieve successful endoscopic extraction of CBD stones in patients who had had surgically altered anatomies due to having undergone gastrectomy, such as Billroth-II reconstruction or R-Y anastomosis, four processes needed to be carried out: reaching the papilla of Vater endoscopically, performing selective biliary cannulation, conducting an ampullary procedure (e.g., sphincterotomy or balloon dilation) and performing stone extraction. The first step was considered the most challenging in ERCP for RYG patients due to the excessive length or rigid adhesion of the R-Y limb, especially when using a conventional side-viewing duodenoscope [6] or a forward-viewing colonoscope [7,8,9]. Indeed, successfully reaching the papilla has been reported in 92% (54/59) of Billroth-II patients but only 33–67% of RYG patients. However, due to recent advances in enteroscopes, such as the advent of single-balloon enteroscopy (SBE) as well as DBE, the successful approach rate has remarkably improved to 91–96% in RYG patients with short-type SBE [18,20,22] and 95–98% with short-type DBE [15,19,21]. Similarly, successful scope insertion to the papilla of Vater was obtained in 92% (73/79) of RYG patients using short-type DBE in this study.

Selective biliary cannulation was also challenging due to the difficulty of positioning the scope from a front view of the papilla of Vater and the limited controllability of the catheter through elevator-unequipped enteroscopes, in contrast to standard duodenoscopes. Previous studies reported the success rate of biliary cannulation in RYG patients to be 74–95% [15,18,19,20,21,22]. One of the tips for biliary cannulation is to perform the procedure using the retroflex position, which can facilitate direct visualization of the papilla from the front [20]. Recently, the position was reported to be a potential favorable factor for successful biliary cannulation [22]. In the present study, 12% (9/73) of patients had failed biliary cannulation, but alternative approaches, such as PTBD which might induce pain and discomfort associated with the external transhepatic catheter [11,12,13,14], EUS-BD (including EUS-guided antegrade intervention) which required complicated process for stone extraction and had a risk of biliary peritonitis [5,30,31,32,33], surgery or conservative therapy, improved the situation. Depending on the patient condition and capabilities of the institution, an immediate decision to alter the treatment plan may also be crucial.

The basic strategy for an ampullary procedure and subsequent stone extraction is considered to be the same as for managing patients with normal anatomy. In the present study, several combinations of an ampullary procedure were performed, as shown in Table 3. Given the difficulty of sphincterotomy and the precutting method due to the inverted view, EPBD or EPLBD alone may be acceptable for RYG patients [35], as it is for normal anatomies [36,37] and Billroth-II gastrectomy patients [27]. Regarding stone extraction, the latest enteroscope with a 3.2 mm working channel can utilize most devices, including an ML, and complete stone extraction was ultimately achieved in 88% of patients in the present study in whom such a procedure was attempted. For difficult cases, stone extraction in two sessions following drainage was recommended [4,5]. In addition, in some cases with large CBD stones, electrohydraulic lithotripsy (EHL) using cholangioscopy [4,38,39,40], percutaneous transhepatic cholangioscopy (PTCS) [14], or EUS-guided antegrade cholangioscopy [31,32,33] might need to be considered for complete extraction.

This was the first study to reveal the factors affecting complete stone extraction in RYG patients who underwent BE-ERCP. First, we identified an interesting risk factor for incomplete stone extraction: the largest CBD diameter ≥ 14 mm. Dilation of the CBD in RYG patients is often seen post-cholecystectomy [41]. In patients with a large CBD stone size and large CBD diameter, sufficient papillary dilation by EPLBD with or without crushing stones using an ML was usually required for successful stone extraction [4,16,17,25,26,27,28,29,36,37,42], but achieving stone clearance is not easy. A previous study reported that a large stone size was a risk factor for incomplete stone extraction by ERCP, in patients with a history of Billroth-II [42] as well as those with normal anatomy [43]. In contrast, small stones floating into larger diameter CBD are often difficult to grasp, even when using available devices, such as a basket or retrieval balloon catheter. This was also an issue when large stones were crushed with an ML. Thus, regardless of the CBD stone size, a large CBD diameter can make complete stone extraction difficult. Dedicated devices that can easily catch small stones floating in large diameter CBDs are desired.

In addition, a retroflex position was identified as a positive factor affecting complete stone extraction. As mentioned above, this position was reported to be useful for successful biliary cannulation in RYG patients [20,22]. The retroflex position can be obtained by advancing the endoscope without releasing the looped scope and forming a J-turn at the IDA. Thereby, a coaxial relationship between the devices and distal CBD and maintaining a proper distance from the tip of the scope to the papilla of Vater with a better view of the papilla can thus be obtained. Such a situation can facilitate stones to be removed along the axis of the CBD. In the present study, in a sub-analysis of the difficult-to-manage cohorts—i.e., those with a large CBD diameter (≥14 mm) or stone size (≥10 mm)—the retroflex position was also a significant factor affecting successful complete stone extraction. In fact, we experienced several cases where stone extraction could not be completed initially in the non-retroflex position, whereas the retroflex position allowed complete stone extraction to be easily performed in the second session, as shown in Figure 1. Taken together, these findings suggest that the retroflex position may be recommended for complete stone extraction as well as successful biliary cannulation in RYG patients. However, this technique should be performed carefully due to the risk of perforation at the IDA.

Adverse events occur in 5–18% of patients treated with this procedure [15,20,22,44], and the incidence rate was 5% in the present study. In contrast to conventional ERCP, perforation is one of the most common adverse events for this procedure [45] and occurs mainly during scope insertion or stone extraction. Immediately noticing the issue and thoroughly performing intraluminal drainage is important, as the situation can sometimes be managed if minor perforation occurs, as shown in one of our cases. Acute pancreatitis occurred in a patient who had a 15 mm diameter CBD stone in an 18 mm diameter CBD and was treated with precutting, a 10 mm diameter EPBD, and an ML. In a systematic review, EPLBD with EST is reported to carry a low risk of pancreatitis compared with EST or EPBD alone (2.4%, 4.3%, and 8.6%, respectively; p < 0.001) [46], therefore, a sufficient EPLBD for a dilated CBD may be important to avoid a risk of procedure-related pancreatitis, but further prospective studies will be needed, as described above.

In addition, patients enrolled in this study were relatively elderly, showing a median age of 79 years old. A previous study also reported that both technical success rates and the rates of adverse events were similar between elderly (≥75 years old) and non-elderly groups (<75 years old), suggesting that BE-ERCP is a feasible procedure for elderly individuals with a surgically altered anatomy [44]. Repeated BE-ERCP may carry a risk for elderly patients, so middle-term stent placement may be an option for the treatment of cases of complicated CBD stones, although caution against life-threatening cholangitis should be practiced [4,47].

Several limitations associated with the present study warrant mention. First, this was a retrospective study with a relatively small cohort, but three tertiary hospitals participated in it. Further prospective studies are needed to validate the present findings. Second, most of the patients were unable to be followed at the hospital, instead visiting family doctors. Therefore, an analysis based on long-term follow-up data, such as the stone recurrence rate, was not conducted.

In conclusion, therapeutic BE-ERCP for CBD stones in RYG patients, with a relatively elderly cohort, was effective and safe using short-type DBE. The largest CBD diameter ≥ 14 mm was an independent risk factor for failed complete stone extraction, but the use of a retroflex position may be considered as a recommended technique to achieve complete stone clearance.