Currently, multiple guidelines from major societies suggest that choledocholithiasis, whether symptomatic or not, should be treated [1]. The standard-of-care in the US and much of the world for patients with choledocholithiasis is to perform an ERCP with sphincterotomy to remove the stones, possibly followed by laparoscopic cholecystectomy to minimize the risk of developing recurrent choledocholithiasis or cholecystitis. Although considered generally safe and successful, of the nearly 500,000 ERCPs performed in the United States annually, ERCP-related adverse events and short- and long-term procedural morbidity is substantial [2], including up to 10% of patients who suffer from post-ERCP pancreatitis (PEP) [3]. Other significant procedure-related complications include sphincterotomy bleeding, perforation, cholangitis, and procedure-related cardiopulmonary events. Despite significant improvement in procedural technique and optimization of patient selection over the last four decades, the principal prophylaxis against ERCP-related complications remains avoidance of unnecessary procedures through careful patient selection. Although patients with elevated procedural risk and limited benefit are among those most frequently excluded, a risk-stratified approach to pursuing or deferring biliary intervention and/or surgery is lacking. A selective risk-based approach may be appropriate, especially in asymptomatic individuals with incidentally found choledocholithiasis who have small (≤ 4 mm) stones that are more likely to pass spontaneously [4].

In this issue of Digestive Diseases and Sciences, Kayashima et al. [5] tested the hypothesis that the standard-of-care (SOC) approach to managing choledocholithiasis may have inferior outcomes to an on-demand approach (ERCP performed only in patients who eventually develop symptoms) in asymptomatic patients with stones ≤ 4 mm, possibly due to a higher rate of spontaneous stone passage coupled with fewer procedural complications. The study was performed in a single center as a retrospective cohort study in Keio University School of Medicine in Tokyo, Japan. Of over a thousand consecutive patients screened, 148 were selected for study who were asymptomatic with no prior biliary intervention or surgery, with subjects stratified by stone size (≤ 4 mm vs. > 4 mm). The conclusion was that the overall adverse event-free survival of asymptomatic patients with diminutive stones with the on-demand ERCP approach was superior to the SOC approach (adjusted adverse event-free 3-year survival 97.4% in the on-demand group and 70.1 in the SOC group) whereas the opposite was true in those with larger stones (54.3% in the on-demand group and 87.5% in the SOC group). Patients in the on-demand group with diminutive stones had lower rates of delayed cholangitis than patients in the SOC group, possibly due to incompletely cleared stones in a no-longer sterile biliary tree whereas the absence of a sphincterotomy in the on-demand group may have been protective. The patients in the SOC group also incurred an added risk of post-ERCP pancreatitis (11.8%). Overall, a complication rate of 5.9 vs. 35.3% favored the on-demand approach for treating these patients. On the other hand, patients with larger stones had markedly higher rate of spontaneous cholangitis in the on-demand group compared with no cholangitis in the SOC group. The authors postulate that in the SOC group, the sphincter of Oddi is cut completely, which may facilitate spontaneous stone passage in patients with recurrent or missed stones. Overall, in patients with larger stones, a complication rate of 11.5 vs 35.2% favored the SOC group.

The principal limitation of the study is its retrospective design with evidence of selection bias, since the reasons patients elected to delay procedures (on-demand group) included: patient request, older age, significant comorbidities, and physician’s discretion, enriching this group with subjects with significant comorbidities and an unhealthier overall condition. Another limitation included a lack of standardization in determining the stone size, since multiple imaging modalities were used across the study subjects including EUS, MRCP and CT scan. MRCP, for example, may underestimate stone size and the number of ductal stones, potentially biasing patients undergoing this imaging modality towards inferior outcomes and a higher risk of choledocholithiasis recurrence [6]. This study also included multiple ERCP providers who were assisted in some cases by trainees, with no stated standardized procedural technique. For example, therapeutic modalities such as papillary balloon dilation without antecedent sphincterotomy were used in some patients [6], which may elevate procedure-related complications such as PEP. Furthermore, there was no set protocol for clinical and imaging follow-up. The cutoff of ≤ 4 mm may be reasonable in predicting which stones are more likely to pass spontaneously, supported by literature references [4, 7] and the authors’ ROC analysis.

The study was not able to address some key clinical questions such as the optimal patient demographic for on-demand strategy. This strategy appeared to be appropriate in patients at elevated procedural and sedation risk such as patients who are elderly and frail and have significant comorbidities. It is unclear if this can be generalized to all patients with asymptomatic choledocholithiasis, including young healthy patients with long life expectancy who have many more years ahead of them to develop gallstone related complications. Furthermore, the authors noted that of sixteen patients with diminutive stones who had follow-up imaging, eleven had confirmed spontaneous stone passage. Would surveillance imaging in these asymptomatic patients be clinically meaningful and would it be cost-effective? Are there other factors that may influence the risk of asymptomatic choledocholithiasis turning symptomatic? Does a larger duct diameter or the presence of intact gallbladder in patients with diminutive stones protect against symptom development?

The study appears to be the first publication supporting the concept that risk-stratification of incidentally found common bile duct stones may improve outcomes and avoid unnecessary ERCPs in patients with small duct stones. Though there are some limitations due to the retrospective study design and the aforementioned issues, the findings generate important conclusions that, if confirmed, may revolutionize the everyday clinical management of patients with choledocholithiasis. Looking forward, prospective randomized studies are needed to better clarify the optimal management strategy of patients with incidentally found asymptomatic common duct stones. Moreover, guidelines may need to better focus on an individualized risk-stratified approach to managing these patients—as some stones may be best left unturned!