EUS yield in evaluating biliary dilatation in patients with normal serum liver enzymes

The finding of common bile duct (CBD) dilatation on abdominal imaging frequently results in additional testing. It has been our impression that endoscopic ultrasound (EUS) evaluation of a dilated CBD is a low-yield examination in the setting of normal serum liver enzymes. We therefore sought to evaluate the EUS yield in evaluating CBD dilatation in patients with normal as compared to elevated serum liver enzymes. A retrospective review was performed to identify patients referred for EUS evaluation of a dilated CBD in the absence of obvious pathology on prior imaging. Charts were reviewed for patient symptoms, presence of elevated serum liver enzymes, imaging studies before EUS, and EUS findings. Exclusion criteria included clinical jaundice, known biliary stricture, mass lesion or stone, and previously sphincterotomy and/or stent placement. Forty-seven patients were identified: 32 with normal and 15 with elevated serum liver enzymes. There was no difference in mean CBD diameter between these two groups (8.51 vs. 8.79 mm, p=0.854). Of the entire group, 15 patients had undergone prior magnetic resonance cholangiopancreatography (MRCP); an additional 7 patients had undergone prior endoscopic retrograde cholangiopancreatography (ERCP). EUS findings to explain CBD dilatation were found more commonly in patients with elevated compared with normal serum liver enzymes (53% vs. 6%, p=0.001). Periampullary diverticula and choledocholithiasis were the most common findings; of 32 patients with normal serum liver enzymes, one periampullary diverticulum and one CBD stone were found, respectively. The CBD stone had been missed by prior MRCP examination. Of 15 patients with elevated serum liver enzymes, there were 3 cases of choledocholithiasis, 4 periampullary diverticula, and 1 ampullary tumor. EUS should be the test of choice for further evaluation of CBD dilatation when index imaging is normal. Although the EUS yield is low in cases of biliary dilatation in the setting of normal serum liver enzymes, its preferential use would potentially avoid unnecessary MRCP and ERCP.