Background: Endoscopic sphincterotomy is sometimes done unnecessarily in patients with suspected choledocholithiasis. Our aims were to assess the diagnostic accuracy of endoscopic ultrasonography and endoscopic retrograde cholangiography (ERC) and to find out whether endoscopic ultrasonography may help to prevent unnecessary sphincterotomy or surgical explorations.
Methods: We recruited 119 patients aged 70.4 (SD 16.1) years with strongly suspected choledocholithiasis who presented to our endoscopy unit between January, 1994, and January, 1995. During the same spell of sedation or within 2 h of each other, endoscopic ultrasonography and ERC were carried out by investigators unaware of the patient's history. Endoscopic sphincterotomy with instrumental exploration was then done as the gold standard for the presence or the absence of stones.
Findings: 78 (66%) patients had choledocholithiasis; 17 (14%) had other bileduct diseases; 24 (20%) had a clear bileduct or did not require an invasive endoscopic procedure. The sensitivity of endoscopic ultrasonography was 93%, specificity 97%, positive predictive value 98%, and negative predictive value 88%. The corresponding values for ERC were 89%, 100%, 100%, and 83%. There were five false-negative cases by endoscopic ultrasonography (of which three were also negative with ERC) and one false-positive. The morbidity rate was 4.1%.
Interpretation: We conclude that endoscopic ultrasonography is at least as sensitive as ERC. Endoscopic ultrasonography may prevent inappropriate invasive explorations of the common bileduct.