Cholelithiasis and cholecystitis are among the most common surgical conditions, with approximately 19,000 cases of cholecystectomy performed each year in Australia. Ten percent of patients with cholelithiasis will have associated CBD stones, and while management of gallstones and cholecystitis have clear guidelines, the management of CBD stones has historically been more controversial.

The management options of CBDS without cholangitis can be operative, endoscopic or radiological. Operative approaches include open (trans-cystic or choledochotomy), laparoscopic (trans-cystic or choledochotomy) or endoscopic (ERCP), which was first described in 1974 [1]. Percutaneous trans hepatic decompression, is a radiological approach that is utilized less frequently and carries significant risk.

In a Cochrane review from 2013, 16 randomised clinical trials with a total of 1758 patients were examined, comparing laparoscopic and open CBD exploration with ERCP in the management of CBD stones. There was no significant difference between these approaches, in both morbidity and mortality. However, open CBD exploration was associated with a higher clearance rate (94% vs 84%) [2].

In the setting of cholangitis, the management approach is mainly ERCP [2, 6]. In a single centre study by Poh et al., it was found that ERCP was more efficient in clearing the bile duct. This paper reported a total of 182 patients that had an emergency laparoscopic cholecystectomy, and received intervention for choledocholithiasis. The duct clearance rate was found to be 63% in the laparoscopic bile duct exploration group, compared to 86% in the ERCP group [7]. However, stone clearance rates by laparoscopic and open techniques have been reported to be as high as 98% in other papers. In a retrospective cohort study by Lee et al., data from a single centre was collected between 1997 and 2011. Laparoscopic bile duct exploration was attempted in 157 patients with conversion to open in 5 (3.2%). Bile duct exploration was done through a choledochotomy with a clearance rate of 98% [8]. The wide variation between studies, can stem from differences in local experience and resource availability. Despite this, ERCP is still the most commonly used modality of treatment for CBD stones in Australia, as reported in a recent narrative review by March et al. [9].

There are several studies that report on complications of ERCP, especially in association with a sphincterotomy [4, 10, 11]. These include pancreatitis, bleeding, perforation and cholangitis in the short term and recurrent CBDS in the long term. Physiologically, the predominant theory behind the development of recurrent primary CBDS post ES, is the reflux of duodenal contents into the biliary system leading to bacterial colonization and subsequent stone formation [12]. Additionally, though there is no high level evidence of ES causing atypia or malignancy, there is some evidence it causes reactive and proliferative changes in the CBD [13].

As an alternative to sphincterotomy, the ampulla can be dilated endoscopically. A recent meta-analysis by Jin et al. demonstrated no short term difference in outcome when comparing endoscopic sphincterotomy to balloon dilatation. Both techniques showed equal rates of stone retrieval, with no statistically significant difference in the incidence of overall adverse events (7.9% vs 10.7%, P = 0.25), post-ERCP pancreatitis (4.0% vs 5.0%, P = 0.54), haemorrhage (1.7% vs 2.8%, P = 0.32), perforation (0.3% vs 0.9%, P = 0.35) or acute cholangitis (1.3% vs 1.3%, P = 0.92). Long-term complications were not reported in that meta-analysis [14], however, Tsujino et al. reported that balloon dilatation alone was associated with recurrent stone disease in 13.5% over a 10-year follow up period [15].

In general, the recurrence rate of CBDS quoted in the literature ranges between 4 and 25% [16, 17]. This wide discrepancy likely reflects selection bias and differences in follow up periods. Kim et al. retrospectively reviewed 101 patients who underwent an ERCP for CBDS with a mean follow up period of 25 months. The CBDS recurrence rate was found to be 5.8–6.9%. There was no statistical difference between stone recurrence associated with ES and balloon dilatation, and time to recurrence ranged from 10 to 42 months. Univariate analysis demonstrated that the number of ERCP sessions to clear CBD stones, the angle between proximal and distal CBD of < 135 degrees, use of mechanical lithotripsy, and the presence of a peri- ampullary diverticulum were statistically significant risk factors. Multivariate analysis revealed that the presence of a peri-ampullary diverticulum was the only independent risk factor for CBD stone recurrence [18]. In a further study by Costamagna et al., 529 patients underwent an ERCP + ES with a minimum 5 year follow up. Stone recurrence was reported as 11.1%, with a statistically significant association with a dilated CBD duct > 22 mm [4].

Our study has demonstrated that ERCP + ES has an inherent long term complication rate of recurrent primary CBDS formation. The rate of 8.9% is in keeping with those reported in the literature. A significantly increased risk is associated with the presence of a type A peri-ampullary diverticulum and a dilated common bile duct of > 15 mm, also in keeping with the literature. Sphincterotomy or dilatation of the ampulla is generally undertaken at the time of ERCP to facilitate CBDS extraction. This can damage the physiological barrier provided by the sphincter of Oddi, allowing reflux of enteric contents and bacteria into the CBD, predisposing to the formation of recurrent primary CBD stones [19, 20].

An alternative minimally invasive method to manage CBDS and avoid the long term complication of ES, is to extract CBDS laparoscopically either trans-cystically or via a choledochotomy. This operative approach has its own complications, with choledochotomy reported to have an increased risk of bleeding, CBD stricturing and bile leak [21, 22]. Hong et al compared laparoscopic cholecystectomy and ERCP (LC+ERCP) to laparoscopic cholecystectomy and operative CBD exploration (LC+BDE). They randomised 234 patients with no statistically significant difference in outcome, including operating time, success rate, postoperative complications, retained stone rate, length of stay or hospital cost [23]. However, all bile duct explorations were undertaken via a choledochotomy. Use of choledochoscopy or imaging guided baskets via the transcystic approach may lead to different results.

There were several limitations to our study. Firstly, it is retrospective, carrying an inherent bias in data collection, and being limited by the quality of the available clinical notes. Information concerning late complications would have ideally been obtained from multiple sources to ensure complete and accurate data collection. Secondly, there are several hospital-based ERCP services in close proximity. Consequently, there is a possibility of under representing the true rate of recurrent primary CBDS, as patients may have presented to other surrounding centres. Lastly, the study group consisted of a relatively elderly cohort, which may have also impacted on the rate of recurrent primary CBDS.