Introduction: Endoscopic retrograde cholangiopancreatography (ERCP) has become the gold standard for diagnosis and treatment of choledocholithiasis. The American Society for Gastrointestinal Endoscopy (ASGE) in 2010 proposed stratifying patients into 3 risk levels; however, studies have found controversial results about the predictive parameters of these diagnostic criteria. The objective of this study is to determine the performance of the high-risk predictive criteria of the ASGE 2010 in the diagnosis of choledocholithiasis in a Colombian Caribbean population. Methods: Retrospective cross-sectional study, which included patients with suspected choledocholithiasis, and who were taken for evaluation by ERCP, meeting the criteria proposed by the ASGE of high probability. The result obtained was compared with the presence of choledocholithiasis on ERCP, from which values and 95% confidence intervals were estimated for sensitivity, specificity, positive and negative predictive value, and positive and negative likelihood ratio. Results: A total of 118 patient data were included in this study. The median age of the patients was 46 years (IQR= 31; 17- 89); 78% (n= 92) were female. 65.3% (n= 77) were older than 55 years. The ERCP result was positive in 81.4% (n= 96) of the patients. The presence of an altered liver profile (90%) was found to be the most sensitive test, clinical cholangitis (86%) the most specific, the presence of duct lithiasis by US (85%) was the test with the highest positive predictive value, and the presence of duct lithiasis by US (35%) was the test with the highest negative predictive value. Conclusions: The predictive parameters of the ASGE 2010 criteria for the diagnosis of choledocholithiasis show variability with respect to the performance proposed in the guidelines.