PURPOSE: To determine relative image qualities and to evaluate their ability to visualize biliary trees and pancreatic ducts, we compared the breath-hold fast spin echo (FSE) and respiratory triggered FSE technique in magnetic resonance cholangiopancreatography (MRCP). MATERIALS AND METHODS: Forty-seven patients with suspected of hepatic disease but no pancreatic or biliary ductal dilatation, as determined by other imaging techniques('group of pathologic pancreatobiliary tree') underwent MRCP. Heavily T2-weighted FSE coronal images were obtained by both breath-hold and respiratory triggered techniques. These two images were 3D-reconstructed using a maximal intensity projection algorithm. Three radiologists scored the image qualities of anatomic structures in each set of image, then directly compared the image quality of the images obtained by the two techniques. RESULTS: For the visualization of common hepatic ducts and common bile dvcts, FSE MRCP images obtained using the respiratory-triggered technique were triggered technique were significantly better than those obtained using the breath-hold technique (P<0.05). Fifty-nine to 88% of breath-hold images of the biliary tree and 63-95% of respiratory triggered images were optimal. For the pancreatic duct, however, 24% of breath-hold images and 15% of respiratory-triggered images provided optimal image quality. In direct comparison, respiratory triggered images were better in 25 cases (52.1%), both images were comparable in 12 cases (25.0%), and in 11 cases (22.9%), breath-hold images were better. These differences were statistically significant (p<0.05). CONCLUSION: For the vizualization of extrahepatic bile ducts, the respiratory triggered FSE sequence was better than the breath-hold sequence; for the evaluation of both a non-dilated and dilated pancreatobiliary system, however, both techniques need further development.