Abdomen ; Liver ; Spleen ; Magnetic Resonance Imaging ; Diffusion Magnetic Resonance Imaging
Abstract
PURPOSE: To compare the image quality and stability of apparent diffusion coefficient (ADC) in diffusion-weighted MRI (DWI) of the upper abdomen among the breath-hold (BH), free-breathing (FB) and respiratory-triggered (RT) techniques. MATERIALS AND METHODS: We analyzed the qualitative and quantitative parameters of 204 consecutive patients who underwent DWI (BH-DWI, FB-DWI or RT-DWI; n=68 in each technique). Qualitative parameters included liver contour, vascular landmarks, intra-slice homogeneity, and inter-slice discontinuity on DWI with a b-factor of 800 s/mm2 and a four-grade scale. Quantitative parameters included inter-slice or intraslice inhomogeneity of ADC in the spleen. RESULTS: RT-DWI showed better liver contour compared to BH-DWI (p < 0.001) or FB-DWI (p = 0.001). As for the quality of the vascular landmarks, BH-DWI was inferior to FB-DWI (p = 0.025) and RT-DWI (p < 0.001). FB-DWI had the poorest result (p < 0.001) for inter-slice discontinuity compared to the other techniques. FB-DWI showed significantly larger inter-slice differences between the highest and the lowest ADCs in the spleen compared with those of RT-DWI (p < 0.001). Intra-slice homogeneity was significantly better in RT-DWI and FB-DWI than in BH-DWI (p < 0.001). CONCLUSION: Compared with BH or FB techniques, RT-DWI appears to result in the best imaging by providing better anatomic detail without skipping continuous slices, in addition to more homogeneous ADCs