Objective: We undertook this study to evaluate pulse-sequence performance in terms of liver signal-to-noise ratio (SNR) and lesion-liver contrast-to-noise ratio (CNR) on T1- and T2-weighted echoplanar MR imaging.
Subjects and methods: Forty-nine patients referred for MR imaging of the liver were examined at 1.5 T with echoplanar MR imaging using spin-echo, inversion-recovery, and gradient-echo pulse sequences. T2-weighted spin-echo (TE = 25, 50, 100, and 150 msec), T1-weighted inversion-recovery (T1 = 100, 380, 600, and 800 msec), and T2*-weighted gradient-echo (TE = 20 msec) images were acquired after one excitation (TR = infinite), using a 128 x 128 data matrix. T2-weighted spin-echo (TE = 20, 50, 100, and 150 msec) images were also obtained with two excitations (TR = 6 sec), resulting in a 128 x 256 data matrix. Signal intensity measurements were made to calculate liver SNR and lesion-liver CNR.
Results: Single-excitation, T2-weighted, spin-echo images at a minimum TE of 25 msec provided the highest liver SNR (p < .05). Single-excitation, T2-weighted, spin-echo images at TEs of 50 and 100 msec, and T1-weighted inversion-recovery images at Tls of 100 and 380 msec provided the highest lesion-liver CNR (p < .05). However, the latter two pulse sequences had considerably inferior liver SNR (p < .05).
Conclusion: Single-excitation, T2-weighted, spin-echo images provide both superior liver SNR and superior lesion-liver CNR. These results can be used to guide technique selection when echoplanar MR imaging is used to examine the liver.