Bile duct stones occurring after CC excision serve as indicators of chronic inflammation in the biliary tract and perpetuate conditions such as cholangitis or pancreatitis. Ultimately, they can worsen ductal stricture and elevate cancer risk. Therefore, long-term follow-up is required in patients diagnosed them. IHD stones emerge as the most common late complication following CC excision, significantly increasing after > 10 years postoperatively. Takimoto reported that IHD stones developed in 1.7% of patients who underwent CC excision in 2009 but confessed that incidence may increase in their recent report.8 In our study, stones were diagnosed 122.2 ± 70.6 months after surgery, and the patient age was 17.4 ± 6.2 years (Table 1). This aligns with previous reports. Ono et al. reported cases of a 14-year-old girl and a 26-year-old man succumbing to cholangiocarcinoma 2 and 26 years after CC excision, respectively, in 2009.10, 12 Nishiyama et al. reported a case of cholangiocarcinoma 33 years following complete excision of CC, highlighting the necessity of > 30 years of long-term follow-up.13 Nonetheless, considering Ono’s report of a 96% overall survival rate, early diagnosis and treatment of CC can yield favorable outcomes.10
In 2022, Takimoto et al. reported that postoperative IHD stone occurred in 10.2% of patients. In our study, IHD stones were identified in 4.6% of patients. Takimoto observed that patients who developed IHD stones were significantly younger at surgery, with type IVa being significantly higher in this group. In our study, type IVa was present in 17 (81.0%) patients. When analyzing patients with clearly identified types of choledochal cysts (n = 332), it was significantly higher than the 34 patients (10.9%) without IHD stones (p < 0.001). According to Kemmotsu et al. and Tsuchida et al., anatomical features, such as stricture at the hepatic duct bifurcation site, are associated with the formation of IHD stones.14–16 Based on this data, Urushihara et al. and Nakagawa et al. suggested preventive hilar ductoplasty, citing promising results and feasibility.17, 18 After securing the long-term outcomes of plasty and the complications associated with it, such as postoperative leaks and re-stenosis at the plasty site, it may be necessary to introduce this procedure in the future. However, it is known that in patients with Type IVa before surgery, intrahepatic duct (IHD) dilatation resolves within one year after surgery for choledochal cysts. Considering this, it is also necessary to establish evidence regarding which patients should be considered candidates for plasty. Based on this, it is also considered necessary to establish criteria for determining the optimal timing for performing ductoplasty to maximize patient outcomes. Although conventional R-en-Y anastomosis was consistently performed in our center, we tried to minimize damage to original structure of the common hepatic duct bifurcation and lymphovascular structures during dissection around the duct and cyst in order to prevent anastomotic stricture and postoperative structural deformation of the main hepatic duct.
Age at surgery was significantly lower in patients who did not develop IHD stones compared with those who did. Subgroup analysis between patients with and without stone formation revealed that IHD stone occurrence was lower in patients who received CC excision (p < 0.05). Patients who developed IHD stones were significantly older at surgery (p < 0.001), and linear regression analysis confirmed an association between age at surgery and occurrence of IHD stones (p < 0.001). Considering the results of our study, we hypothesize that diagnosis and operation at a younger age would be advantageous for prognoses related to IHD stones. Although early diagnosis and surgery have a positive effect on prognosis, waiting until six months of age for surgery is suggested considering the technical aspects of surgery, such as HJ stricture. Ryu et al. reported the outcomes of 43 neonates who underwent CC excision via open and laparoscopic approaches and were followed for an average of 37 months.19 Except for cholangitis and ileus in three patients who underwent open surgery, no known complications occurred in all patients. This demonstrated the technical feasibility of CC excision in neonates, emphasizing the need for consideration of prompt surgery after diagnosis.
In terms of treatment, oral ursodeoxycholic acid (UDCA) may be beneficial for asymptomatic small-sized stones.8 Takimoto reported the resolution of IHD stones in two patients included in the study, with unchanging in three patients. Among patients diagnosed with IHD stones, 16 (76.2%) were taking UDCA. At our center, we suggest UDCA supplementation not only for patients with events related to stones or asymptomatic small stones but also for adolescents at high risk of stone formation, as indicated by follow-up imaging showing IHD dilatation or pneumobilia. Senyuz reported the use of extracorporeal shock wave lithotripsy for stone removal. However, PTCS is the most longstanding therapeutic approach.20 In our study, PTCS was performed in 11 patients. ERCP for stone removal has been recently introduced, in addition to the approach reported by Takimoto, involving accessing the hepaticojejunostomy through the Y-limb, we performed hepaticogastro/jejunostomy via direct access by puncturing through the stomach and duodenal wall under endoscopic ultrasound guide. Stone removal and balloon dilatation was performed via stomy and a stent was maintained for repetitive procedures.7 Two patients have exhibited favorable outcomes without complications and were satisfied with the aesthetic aspect of the procedure. Considering that patients developing stones are adolescents in their 10s to 20s, endoscopic ultrasound-guided stone removal would be a promising modality if technical barriers are overcome. However, this approach is challenging, therefore, only a few gastroenterologists from our center can perform it. Hepatectomy can be considered in cases of recurrent IHD stones or severe ductal stenosis. At our center, left hepatectomy was performed on two patients, with no recurrence of stones and favorable outcomes.
No CCA case was identified in our study population. Mukai’s 2018 report, which conducted a 20-year follow-up, also did not identify any case of biliary carcinoma in pediatric patients.21 Considering that stone occurrence increases after at least 10 years and CCA is reported to occur after 20 years from the operation, a longer-term follow-up seems necessary for patients who have experienced or are at risk of stone formation.
RIPD/CCD stones are poorly understood, with factors such as APBDU and pancreas divisum being considered influential, however, independent reports from patients with CC are limited.1, 9 Komi asserted that even after radical operation, in patients with complicated types, such as type Ib, IIb, and IIIc3, chronic pancreatitis can progress due to protein plugs or pancreatic calculus in the dilated duct.1, 9 Except two patients, whose MRCP was not available, we could identify the type of APBDU in the rest of patients. Two patients were included in Komi type 1A and the rest in complicated type. Among them, Komi type IIIC accompanying pancreas divisum was identified in nine patients. Typically, pancreas divisum is believed to be associated with pancreatitis. However, Terui reported that pancreas divisum is found in 1.4% of APBDU cases and does not always lead to pancreatitis.22 In patients with pancreas divisum, if the minor duct serves as the main drainage pathway, sludge in the common channel may have a lower likelihood of inducing pancreatitis, whereas if the major duct is involved, the risk of pancreatitis may increase. Moreover, given that most cases of pancreas divisum associated with CC are incomplete type, this introduces additional variables to consider.23, 24 Therefore, it seems difficult to establish a direct association between pancreas divisum and increased incidence of pancreatitis. Thus, prospective studies conducting subgroup analysis based on presence of stones is warranted to elucidate the relationship with RIPD/CCD stones. In our study, stones occurred at various sites, including the common channel, major duct, and minor duct. All patients exhibited obstructive symptoms at CC diagnosis, and 11 had preoperative ERCP, suggesting the need to consider the contribution of preoperative obstructive lesions and ERCP to long-term stone formation in RIPD/CCD.25 Stones occurred at 65.6 ± 72.3 months postoperatively, significantly earlier than IHD stones (Table 3). Subgroup analysis and linear regression results (p = 0.05) revealed no association between age at surgery and stone (Table IV). Considering previous reports of biliary epithelial hyperplasia in CC excision specimens, early surgery might reduce inflammation in the biliary system, potentially decreasing IHD stone occurrence, especially in younger patients.26–28 Treatment primarily involved ERCP in all patients except three asymptomatic patients with favorable outcomes. One patient underwent pylorus-preserving pancreaticoduodenectomy and later required lateral pancreaticojejunostomy due to recurrent stones and resultant severe pancreatic duct stricture. Although many patients had repeated ERCP, none showed dilatation of the main pancreatic duct.
The retrospective nature of this study posed limitations in comparing imaging results between the stone occurrence and non-occurrence groups. Most patients with stones got imaging tests since symptom onset, with subsequent tight follow-up. Conversely, asymptomatic patients were primarily monitored with ultrasonography, leading to an insufficient retrospective comparison of anatomy. Additionally, early MRCP images lacked precision, hindering interpretation and direct comparison of anatomical features, potentially influencing stone formation. A prospective study, including MRCP follow-up, is necessary to analyze whether Tsuchida type 2 or 3 persists after surgery and to determine the characteristics of patients in whom changes in RPID/CCD occur in cases of pancreas divisum and complicated ABPDU types.