Diagnosing choledocholithiasis: better to be European or American?

The paper from Jagtap et al. in this issue of Endoscopy compares the respective value of European Society of Gastrointestinal Endoscopy (ESGE) and American Society for Gastrointestinal Endoscopy (ASGE) criteria for predicting choledocholithiasis [1]. Both ESGE and ASGE recommendations for the management of common bile duct (CBD) stones were published in 2019 [2] [3]. Despite the publication of these guidelines and randomized controlled trials, many issues are still a matter of debate, not only for therapeutic management but also for CBD stone diagnosis and predicting the presence of choledocholithiasis [2] [3]. If clinicians could obtain a reliable and efficient preoperative diagnosis of choledocholithiasis, they could avoid, in many cases, the debate between endoscopic, surgical or combined approaches to manage these patients.

Both ESGE and ASGE guidelines used clinical presentation (cholangitis), abnormal liver function tests, and imaging (initial ultrasound for detection, and endoscopic ultrasound [EUS] or magnetic resonance cholangiopancreatography [MRCP] to identify suspected CBD stones) [2] [3]. As none of these parameters alone has demonstrated complete accuracy, an algorithm using a combination of these measures is provided to the clinician based on the probability of showing CBD stones (high, intermediate or low likelihood). These algorithms differ slightly between the two guidelines, yielding different accuracy rates, and the Jagtap et al. paper attempts to clarify these differences [1] [2] [3].

“…the main problem is probably not whether to choose between ASGE or ESGE guidelines, which show similar diagnostic performance, but to understand the risk of assigning an incorrect diagnosis based on likelihood criteria.”

The first issue to resolve is to identify what clinicians expect from the available algorithms. Clinicians want to avoid invasive procedures with non-negligible morbidity such as endoscopic retrograde cholangiopancreatography (ERCP). A desirable procedure has a high positive predictive value (PPV) for diagnosing CBD stones and a high negative predictive value (NPV) for eliminating the presence of CBD stones. On the one hand, the ERCP morbidity rate cannot be accepted if there is no choledocholithiasis; on the other hand, undiagnosed CBD stones can lead to a complicated post-cholecystectomy outcome, such as cystic duct leakage, cholangitis or pancreatitis. The Jagtap et al. paper mainly focuses on PPV and specificity, so what do these parameters mean to clinicians? One could say that sensitivity and specificity describe the theoretical capacity of the diagnostic test, that is, to be positive in patients presenting the anomaly (sensitivity) and to be negative in patients without the anomaly (specificity). This is useful when choosing a test in order to get the best performance. However, in clinical practice, we look to the actual results when using the test (i. e. the PPV and NPV). When the test yields a positive result, does that mean that the patient definitely has a CBD stone? When the test yields a negative result, does that mean you can definitely rule out a CBD stone? Consequently, we should compare the respective PPV and NPV of the ESGE and ASGE criteria, as shown in Table 3 of the Jagtap paper.

For instance, with the high likelihood criteria, NPVs are similar between the two guidelines at about 91.4 %. This means that there is a 9 % chance of missing a residual choledocholithiasis. Despite this low rate, the consequence of a missed diagnosis could be severe, as mentioned above. The PPV of high likelihood criteria is 96.2 % with the ESGE guidelines and 89.5 % with the ASGE guidelines. This means that for patients classified as high likelihood, choledocholithiasis could be misdiagnosed, resulting in unnecessary ERCP in 4 % of the cases when following the ESGE guideline and in 10 % when following the ASGE guideline. The 4 % value is close to the ERCP morbidity rate [4] and the 10 % value is twice this rate. Of course, nothing is perfect, and mistakes are possible but the patient should be informed that ERCP could be performed unnecessarily in 4 % – 10 % of cases that present with a high likelihood criteria for CBD stones. This explains why the authors found that use of the ASGE guidelines resulted in 7.4 % (17 /230) more patients being classified as high likelihood compared with ESGE criteria, only one of whom actually had choledocholithiasis.

The second issue is to decide what to do when patients are classified as presenting with intermediate or low likelihood criteria for CBD. It is easy to understand that a PPV of about 10 % with both ESGE and ASGE criteria and an NPV ranging from 42.1 % to 48.1 % are not conclusive for a CBD diagnosis and therefore EUS or MRCP are required in all cases with intermediate criteria. The respective performances of intermediate likelihood scores with ASGE and ESGE criteria are not very different and these low values limit the usefulness of these criteria. The main problem relates to the low likelihood group. If a patient is classified in the low likelihood group, the clinician would prefer to avoid further imaging investigations (EUS/MRCP) and would refer the patient directly to laparoscopic cholecystectomy. In this low probability group, 2.6 % using the ESGE score and 3.7 % using the ASGE score finally had choledocholithiasis. The resulting message is that one could miss 2 % – 3 % of patients with CBD stones who were incorrectly classified into the low probability group. Therefore, the question is whether to accept this risk or to perform EUS/MRCP unnecessarily in 98 % of the cases. Again, this risk should appear to be acceptable if the patient gives informed consent.

To conclude, the main problem is probably not whether to choose between ASGE or ESGE guidelines, which show similar diagnostic performance, but to understand that: 1) in cases of high likelihood criteria, the risk of performing unnecessary ERCP ranges between 4 % and 10 %; 2) in cases of intermediate likelihood criteria, EUS or MRCP are always required; and 3) in cases of low likelihood criteria, the risk of missing a choledocholithiasis ranges between 2 % and 3 %. Therefore, in all cases, the patient should be informed about the exact risk of excessive investigation or of missing a CBD stone according to the high/intermediate/low likelihood criteria of ESGE/ASGE guidelines.

Publication History

Article published online:
24 June 2020

• References

• 1 Jagtap N, Yashavanth HS, Tandan M. et al. Clinical utility of ESGE and ASGE guidelines for prediction of suspected choledocholithiasis in patients undergoing cholecystectomy. Endoscopy 2020; 52: 569-573
• 2 Manes G, Paspatis G, Aabaken L. et al. Endoscopic management of common bile duct stones: European Society of Gastrointestinal Endoscopy (ESGE) guideline. Endoscopy 2019; 51: 472-491
• 3 Buxbaum JL, Abbas Fehmi SM, Sultan S. et al. ASGE guideline on the role of endoscopy in the evaluation and management of choledocolithiasis. Gastrointest Endosc 2019; 89: 1075-1105
• 4 Dumonceau JM, Kapral C, Aabakken L. et al. ERCP-related adverse events: European Society of Gastrointestinal Endoscopy (ESGE) guideline. Endoscopy 2020; 52: 127-149