Comparison of multi-echo and single-echo gradient-recalled echo sequences for SPIO-enhanced liver MRI at 3 T

Abstract

AIM: To assess the utility of a T2*-weighted, multi-echo data imaging combination sequenced on superparamagnetic iron oxide (SPIO)-enhanced liver magnetic resonance imaging (MRI) using a 3 T system.

MATERIALS AND METHODS: Fifty patients underwent SPIO-enhanced MRI at 3 T using T2*-weighted, single-echo, gradient-recalled echo (GRE) sequences [fast imaging with steady precession; repetition time (TR)/echo time (TE), 126 ms/9 ms; flip angle, 30°] and multi-echo GRE (multi-echo data image combination) sequences (TR/TE, 186 ms/9 ms; flip angle, 30°). Three radiologists independently reviewed the images in a random order. The sensitivity and accuracy for the detection of focal hepatic lesions (a total of 76 lesions in 33 patients; 48 solid lesions, 28 non-solid lesions) were compared by analysing the area under the receiver operating characteristic curves. Image artefacts (flow artefacts, susceptibility artefacts, dielectric artefacts, and motion artefacts), lesion conspicuity, and overall image quality were evaluated according to a four-point scale: 1, poor; 2, fair; 3, good; 4, excellent. The signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of the lesions were compared.

RESULTS: Image artefacts were more frequent with single-echo GRE (p<0.05). The mean scale of image quality assessment for flow, susceptibility, dielectric, and motion artefacts were 2.76, 3.13, 3.42, and 2.89 with single-echo, respectively, compared with 3.47, 3.43, 3.47, and 3.39, respectively, with multi-echo GRE. There was no significant difference in lesion conspicuity between single-echo (3.15) and multi-echo (3.30) GRE sequences. The overall image quality was significantly (p<0.05) better with multi-echo (3.37) than with single-echo GRE (2.89). The mean SNR and CNR of the lesions were significantly (p<0.05) higher on multi-echo (79±23 and 128±59, respectively) images than on single-echo (38±11 and 102±44, respectively) images. Lesion detection accuracy and sensitivity were not significantly different between the two sequences. Mean accuracies and sensitivities were 0.864 and 0.785 for single-echo and 0.847 and 0.785 for multi-echo GRE, respectively.

CONCLUSION: At 3 T, the T2*-weighted, multi-echo data image combination sequence performs comparably to the T2*-weighted, single-echo GRE sequence for SPIO-enhanced MRI with good overall image quality and a decrease in undesired artefacts.