25018069
OBJECTIVE	To assess the optimal pancreatic phase delay in terms of parenchymal enhancement and tumor-to-pancreas contrast with a bolus-tracking method .
METHODS	Patients referred for suspicion of pancreatic tumor and undergoing 64-detector computed tomography scanner were randomized to an individualized scan delay of 10 , 20 , or 30seconds of nonionic contrast material ( 370mg I/mL ) after aortic enhancement above 150 Hounsfield units .
METHODS	The volume of contrast was adjusted to patient weight .
METHODS	Pancreatic and tumor enhancements were measured .
METHODS	Statistical analysis included analysis of variance and post hoc Tukey tests .
RESULTS	One hundred and fifty patients were randomized to individualized scan delays of 10 , 20 , or 30seconds .
RESULTS	Pancreatic parenchymal enhancement in all patients ( n = 150 ) was significantly higher with a delay of 20 or 30 seconds than that with 10 seconds ( P < .001 for both ) .
RESULTS	Tumor-to-pancreas contrast for solid tumors ( n = 59 ) was significantly higher with a delay of 30seconds than that with 10seconds ( P = .015 ) .
RESULTS	Adenocarcinoma-to-pancreas contrast during pancreatic phase was significantly higher for a 20 - or 30-second delay than for a 10-second delay ( P = .027 and .011 , respectively ) for one reader .
CONCLUSIONS	With a flow rate of 4mL/s and weight-adjusted contrast volume , an individualized scan delay of 30seconds after aortic transit time revealed higher pancreatic enhancement and tumor-to-pancreas contrast than that with a delay of 10seconds .

