To analyze the optimal timing strategy for the detection of hypervascular l
iver tumors during the arterial phase of magnetic resonance (MR) imaging. a
test examination after injection of 2 mL of gadopentetate dimeglumine was
performed in 47 patients. The time course of the tumor-to-liver contrast-to
-noise ratio (CNR) for all studies together was determined relative to the
start of injection, the time of peak aortic enhancement, and the time of pe
ak enhancement in the tumor. All studies were grouped together and the high
est CNR was transiently observed at the time of peak tumor enhancement. Thi
s CNR was significantly higher than those observed at fixed delays after pe
ak aortic enhancement, However, the CNRs at peak tumor enhancement +/-1.5 s
econds did not differ significantly from those obtained after peak aortic e
nhancement. Finally, the CNRs obtained at fixed delays after the start of i
njection remained significantly lower, In hypervascular liver tumors, a hig
her CNR can be obtained during the arterial phase when the MR imaging delay
is determined relative to the time of peak enhancement in the tumor or the
aorta rather than being fixed after the start of contrast material injecti
on. Timing based on the enhancement profile in the tumor rather than in the
aorta should be performed only if rapid MR imaging is available with a tim
e resolution of about 1.5 seconds to image the whole liver. J, Magn, Reson,
Imaging 1999;9:562-567, (C) 1999 Wiley-Liss, Inc.