Invisible or negligent—EUS detection for the negative CT or/and MRCP CBD stone

Abstract

Since the advent of endoscopic sphincterotomy, developed in Japan and Germany in 1974, it has become a very common technique used for the treatment of a wide variety of conditions of the biliary system. However, because of the risk of adverse events associated with endoscopic retrograde cholangiopancreatography (ERCP)-guided treatment of bile duct stones, it is important to identify appropriate candidates for this procedure and reserve biliary endoscopy for patients with the highest probability of intraductal stones.

The prevalence of common bile duct (CBD) stones is reported to be 5% to 15% in patients undergoing elective cholecystectomy for symptomatic, uncomplicated cholelithiasis.^1-3 Liver biochemical tests may be most useful in excluding the presence of CBD stones. The negative predictive value of a normal liver function test in a series of more than 1000 patients undergoing laparoscopic cholecystectomy was over 97%, whereas the positive predictive value of any abnormal liver biochemical test was only 15%.⁴ The role of endoscopy in the evaluation of suspected choledocholithiasis, a guideline statement developed by the Standards of Practice Committee of the American Society for Gastrointestinal Endoscopy (ASGE) in 2010, proposed a strategy to assign the risk of choledocholithiasis in patients with symptomatic cholelithiasis based on clinical predictors.⁵ The very strong predictors (major criteria) include CBD stones found on transabdominal ultrasound (US) study, clinical cholangitis, and total bilirubin level >4 mg/dL. The strong predictors (minor criteria) are dilated CBD on US (>6 mm with gallbladder in situ) or total bilirubin level 1.8 to 4 mg/dL. Patients with one major or two minor predictors are considered high risk and should receive preoperative ERCP. However, if patients have intermediate risk, they should undergo endoscopic ultrasonography (EUS) or magnetic resonance cholangiopancreatography (MRCP) examination before surgery to detect the presence of CBD stones. Two years earlier, ASGE released new guidelines for choledocholithiasis management.⁶ Emphasis was increased on using laboratory results and bile duct diameter in deciding when to perform ERCP. We noticed that there was no additional new definition of risk predictors in comparison to 2010. However, in the 2019 guidelines, new criterion requiring both bilirubin >4 mg/dL and biliary dilatation was added, which has a specificity approaching 90%. This is similar to the other two high-risk predictors, cholangitis and bile duct stones on imaging, both of which have a specificity exceeding 90%.⁷ Likewise, computed tomography (CT) is not included in the diagnostic algorithm if CBD stones are diagnosed based on US, liver function test, and clinical information. Most of the invisible CBD stones are small (<5 mm) or less calcified; therefore, they cannot be clearly visualized on CT.^{8, 9} Uyeda et al¹⁰ reported that nearly one-quarter of all gallstones are isoattenuating to the surrounding bile at 120 kVp because of their high cholesterol and low calcium content and are thus not detectable on CT. We had similar experiences. Nevertheless, CT is an important diagnostic tool for CBD stones in oriental countries because of the higher incidence of pigment stones than the Western populations. Consequently, we proposed an algorithm modified from the European Society of Gastrointestinal Endoscopy (ESGE) guideline¹¹ to which CT examination is added before EUS or/and MRCP investigation for the evaluation of suspected choledocholithiasis and it is implemented in our hospital.

Lem et al¹² investigated patients with intermediate or high risk for CBD stones undergoing EUS, according to the ASGE 2010 guidelines, and found a 66.7% (22/33) sensitivity and 84.7% (50/59) specificity for detecting CBD stones. Nine patients with indications for prompt ERCP based on the 2010 guidelines did not undergo the procedure because of negative findings on EUS. These patients did not have gallstone-related symptoms during the follow-up period. Therefore, the ASGE 2019 guideline eliminates two criteria and combines them to achieve a 90% specificity. The roles of EUS and MRCP in the cost-effectiveness model in patients at intermediate risk of choledocholithiasis are the most extensively studied. It appears that EUS and MRCP result in cost savings by avoiding the expense and adverse events of ERCP.^{13, 14} To the best of our knowledge, no RCTs have compared EUS with MRCP, but several prospective observational trials were identified. The evidence for MRCP vs EUS for choledocholithiasis was evaluated by a recent systematic review and meta-analysis by Meeralam et al.¹⁵ The pooled sensitivity of EUS was higher than that of MRCP. However, there was no difference in the specificity between EUS and MRCP. The diagnostic odds ratio (OR) was greater for EUS than for MRCP. EUS has better sensitivity and OR than MRCP. In Lem's study, we suggest that different strategy be used for EUS or MRCP in high-risk patients before preoperative ERCP. EUS or MRCP may be needed for comorbid patients to waive unnecessary ERCP procedures, thus avoiding possible serious post-ERCP complications. However, ordinary high-risk patients may be managed directly by ERCP or surgery in large volume ERCP centers.

In conclusion, Lem et al. demonstrated the utility of EUS in the evaluation of suspected choledocholithiasis. It is critical that patients with very strong predictors still harbor the possibility of the absence of CBD stones. CT is an indispensable modality for investigating the presence of CBD stones. For patients with intermediate risk, EUS or MRCP is needed to clarify whether CBD stones are present. Which is better depends upon the hospital facility and experts good at the procedure. Prompt ERCP procedures, in addition to immediate costs, may result in adverse events associated with delays and prolonged hospitalization. We should have our own guidelines, regardless of ASGE or ESGE recommendations, to manage suspected choledocholithiasis in our daily practice.

The author declares no conflict of interest.