This study showed that patients with a CBD stone detected at IOC and left in situ had a statistically significant higher risk for retained CBD stones requiring ERCP than that of patients who underwent a planned perioperative intervention for stone removal. A subgroup analysis in patients with small and asymptomatic CBD stones also suggested that these should be removed intraoperatively or shortly after surgery with ERCP. When comparing risk for retained stones between the non-IOC group and the IOC group, the risk was higher in the non-IOC group (HR 1.4, 95 per cent c.i. 1.3 to 1.6).

The study was based on data from the GallRiks registry that has nearly complete (90 per cent) coverage of patients undergoing gallstone surgery or ERCP in Sweden⁸.

The outcome measures only included patients who, for any reason, underwent an unplanned ERCP and had a retained CBD stone. The indication for the ERCP was unknown and the outcome measure could thus include asymptomatic stones discovered during ERCP undertaken for other reasons.

Retained CBD stones of minor clinical relevance and CBD stones that rapidly passed spontaneously were not included, nor were patients with retained stones not undergoing ERCP for other reasons such as CBD stone-related death, too frail to undergo ERCP, and those managed conservatively. This could lead to underestimation of the risk for retained stones. Further studies where these data are cross-referenced with the national patient registry could shed light on this potential bias.

Another weakness of the definition of retained CBD stones was the inclusion of patients with a bile duct stent left in situ and who were planned for ERCP later than 7 days after surgery. During ERCP and removal of the stent, minor retained stones, otherwise asymptomatic and never identified if the ERCP had not been undertaken, would have been registered as a terminal event. This is a potential source of bias since delayed ERCP in this case was part of the primary strategy and not due to clinical relevance of the stones. Unfortunately, it was not possible to identify these patients from the registry data. Since this group was included as ‘intervention’ and registered with the terminal event, it could make the intervention appear less favourable.

Although IOC is an accurate diagnostic method for identification of CBD stones¹², false-negative or -positive test results may occur. Also, the estimation of CBD stone diameter may be inaccurate. However, it is believed that the data from this study are externally valid as the study represents current clinical practice across a nation. Finally, there is a potential selection bias towards the group where no IOC was performed, but there are no data available on the reason why IOC was omitted, other than when it was attempted and unsuccessful.

The decision to choose intervention or surveillance of CBD stone was made by the surgeon and probably depended on previous symptoms and stone size. This could also have caused selection bias. However, the subgroup analysis of small and asymptomatic stones still showed that, regardless of selection and assuming a favourable course, CBD stones believed to be harmless if left in place benefit from removal, thus avoiding the risk for retained stones becoming symptomatic.

Since interventions range from simply flushing or manipulating the CBD stone through the papilla Vateri to open choledochotomy with extraction of the stone, the intervention group was heterogeneous. In the present study, focus on stones left to surveillance compared with any intervention was considered relevant because surveillance is a fundamentally different form of management. Furthermore, surveillance was the preferred strategy in 30.5 per cent (661 of 2168) of the smallest asymptomatic stones.

Findings of the present study are in line with a previous study based on GallRiks data where risk for unfavourable 30-day outcome was greater if CBD stones were left in situ without intervention¹³. The British Society of Gastroenterology guidelines also recommend that all CBD stones be removed when identified at cholecystectomy¹⁴.

However, recently published data suggest a conservative approach for asymptomatic CBD stones, partly because of the risk for complications of ERCP itself¹⁵. In that study, however, the safer strategy of rendezvous ERCP¹⁰ was not employed and CBD stone size was not reported.

The results are applicable in the international setting where IOC may not always be a routine examination during cholecystectomy. In many countries IOC is undertaken selectively such as when the anatomy is unclear or when there is strong suspicion of CBD stones¹⁴. The findings of this study suggest that a substantial proportion of CBD stones that are asymptomatic would remain undetected if IOC were omitted, and that many patients would require subsequent ERCP for extraction of retained CBD stones. In other words, ERCP for retained CBD stones could be avoided if IOC is performed routinely and any stones detected removed intraoperatively or removed shortly after primary surgery using rendezvous ERCP.

The risk of leaving a CBD stone in situ must always be weighed against the risk of the intervention performed to remove the stones. It was not within the scope of this study to compare risk for complications, other than retained stones, between intervention and surveillance. What this study does show, however, is that the risk of having to undergo late postoperative ERCP because of retained stone is higher if IOC is omitted or if CBD stones are not managed immediately. Further studies are needed to compare the overall complication risk between surveillance and various methods for clearing CBD stones.