BACKGROUND/AIM: ERCP has been considered an accurate means in the diagnosis of choledocholithiasis. However, small bile duct stone can be difficult to diagnose during ERCP. The aim of this study was to prospectively evaluate the ability of intraductal US for detecting small choledocholithiasis in patients with highly suspected bile duct stones demonstrating no filling defect on ERC. METHODS: Fifty nine consecutive patients who underwent ERCP within 48 hours after admission for highly suspected choledocholithiasis with negative results were evaluated with intraductal US (20 MHz) for the presence of stones or sludge. Patients showing any filling defect on biliary tree on ERCP excluded. Reference standard for choledocholithiasis was endoscopic extraction of stone or sludge. RESULTS: 1) Of 59 patients, IDUS found evidence of bile duct stone in 11(18.6%) patients. IDUS findings were confirmed by endoscopic extraction of stone in 10 of 11 (16.9%) patients. The sensitivity and specificity of IDUS in detecting choledocholithiasis were 100% and 97.9%. 2) IDUS showed evidence of biliary sludge in 10 patients and confirmed by extraction of sludge in 7 patients. 3) Mean diameter of CBD stones diagnosed by IDUS was 2.3 mm (range: 1-4 mm). 4) In dilated bile ducts, detection rate of bile duct stone/sludge missing on ERC were significantly higher than that of nondilated bile duct (p<0.05). CONCLUSION: Intraductal US is highly accurate in the detection of small bile duct stones that missed on ERCP. In patients with suspicion of bile duct stones but normal ERCP, intraductal US could be performed as a useful adjunct to ERCP.