Norio Hayashi PhD, Presenter: Nothing to Disclose

Tosiaki Miyati PhD, Abstract Co-Author: Nothing to Disclose

Naoki Ohno MS, Abstract Co-Author: Nothing to Disclose

Yasuji Ryu MD, Abstract Co-Author: Nothing to Disclose

Yumie Takeshita, Abstract Co-Author: Nothing to Disclose

Osamu Matsui MD, Abstract Co-Author: Research Consultant, Bayer AG Research grant, Bayer AG Research Consultant, Eisai Co, Ltd Research Consultant, Kowa Company, Ltd

Tsuyoshi Matsuda BS, Abstract Co-Author: Employee, General Electric Company

Toshinari Takamura, Abstract Co-Author: Nothing to Disclose

Shigeru Sanada PhD, Abstract Co-Author: Stockholder, Hologic, Inc

Takashi Hamaguchi MS, Abstract Co-Author: Nothing to Disclose

Hiroji Iida, Abstract Co-Author: Nothing to Disclose

Masako Takanaga, Abstract Co-Author: Nothing to Disclose

To compare and evaluate the accuracy of fat fraction and iron content measurement by 2-, 3-, and 6-point Dixon methods and magnetic resonance (MR) spectroscopy with and without T2 relaxation correction in phantom and clinical studies.

Institutional review board approval and written informed consent were obtained. To evaluate the influence of T2 relaxation on water and lipid spectral peaks, and subsequent fat fraction measurements, seven individual phantoms with known lipid and iron contents were constructed. Hepatic fat fraction was measured in patients (22 males and 13 females; mean age 54.2 ± 12.5 years; range, 19 – 71 years) with fatty liver disease. Fat fractions in phantom and patients were determined using a 3.0 Tesla MRI scanner (Signa EXCITE HDxt 3.0T: GE Healthcare, WI) using the multiecho (2,3,6 echoes) Dixon method with and without T2* correction. Single-voxel MR spectroscopy with T2 correction that was coregistered with imaging was performed for comparison.

In the phantom study, the R2* value measured by the 6-echo water-fat separation method showed a strongly positive correlation with the actual iron content (r > 0.996). In addition, MRI-derived fat fractions measured by the 6-point Dixon method and MRS with T2 correction showed good agreement with the actual value. On the other hand, MRI-derived fat fraction measured by the 2- and 3-point Dixon methods without T2* correction did not show good agreement with the actual value. In clinical cases, hepatic fat fraction determined by the Dixon method showed a positive correlation with that determined by MR spectroscopy (2-point: r = 0.823; 3-point: r = 0.406; 6-point: r = 0.814). When iron content was different in clinical studies, hepatic fat fraction determined by the 6-point Dixon method and MR spectroscopy showed no difference. However, hepatic fat fraction determined by the 2- or 3-point Dixon method was significantly larger than that determined by the 6-point method or MR spectroscopy (P < 0.001).

To accurately evaluate fat fraction by MRI, it is necessary to perform T2 or T2* correction. Therefore, using the 6-point Dixon method or MR spectroscopy with T2 correction, it is possible to accurately evaluate fat fraction by MRI.

The multiecho Dixon method with T2* correction or MR spectroscopy with T2 correction is a potentially useful non-invasive tool for assessment of fat fraction and iron content.

Hayashi, N, Miyati, T, Ohno, N, Ryu, Y, Takeshita, Y, Matsui, O, Matsuda, T, Takamura, T, Sanada, S, Hamaguchi, T, Iida, H, Takanaga, M, Accuracy of Hepatic Fat Fraction Measurement Using the Multiecho Dixon Method with and without T2* Correction: Phantom and Clinical Study.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11000611.html