Original Research MRI of the Female Pelvis at 3T Compared to 1.5T: Evaluation on High-Resolution T2-Weighted and HASTE Images Masako Kataoka, MD, PhD, 1,2 * Aki Kido, MD, PhD, 1 Takashi Koyama, MD, PhD, 3 Hiroyoshi Isoda, MD, PhD, 1 Shigeaki Umeoka, MD, PhD, 1 Ken Tamai, MD, 1 Yuji Nakamoto, MD, PhD, 1 Yoji Maetani, MD, PhD, 1 Nobuko Morisawa, MD, 1 Tsuneo Saga, MD, PhD, 1 and Kaori Togashi, MD, PhD 1 Purpose: To evaluate the feasibility of MRI of the female pelvis using high-resolution T2-weighted imaging (T2WI) and the half-Fourier acquisition single-shot turbo spin- echo (HASTE) technique at 3 Tesla (T) compared to 1.5T, while focusing on the uterine body and cervical anatomy. Materials and Methods: A total of 19 healthy women un- derwent pelvic MR scans on 3T and 1.5T scanners. Axial and sagittal T2W (voxel size of 0.6  0.8  2 mm) and sagittal HASTE images were obtained. The images were evaluated qualitatively for overall image quality, contrast in the uterine zonal appearance and cervical structure, image inhomogene- ity, and artifacts. A quantitative evaluation was performed regarding zonal contrast and image inhomogeneity. Results: On T2WI, the image contrast in the uterine cervix and vagina were significantly higher at 3T than at 1.5T, although there was no significant difference in the overall image quality or contrast in the uterine zonal appearance. Image inhomogeneity was more prominent at 3T, and mo- tion artifact was more severe at 1.5T. Conclusion: Our results suggest that MRI of the female pelvis at 3T may potentially provide excellent images of the uterine cervix on high-resolution T2WI. New techniques to reduce inhomogeneity are thus called for. Key Words: MRI; ultrahigh magnetic field; pelvis; uterus; ovary J. Magn. Reson. Imaging 2007;25:527–534. © 2007 Wiley-Liss, Inc. MAGNETIC RESONANCE IMAGING (MRI) is now con- sidered to be a reliable tool for evaluating gynecological conditions. MRI can exhibit the zonal anatomy of the uterus on T2-weighted imaging (T2WI) and thus help to evaluate uterine tumors and other gynecological condi- tions. With the advent of fast MRI, kinematic imaging of the uterus using half-Fourier acquisition single-shot turbo spin-echo (HASTE) can demonstrate uterine peri- stalsis (1,2) and uterine contraction (3,4), which are related to the menstrual cycle (1), hormonal status (5), dysmenorrhea (4), and uterine pathology (6). However, current 1.5-Tesla (T) MRI has some limitations due to its limited signal-to-noise ratio (SNR). For instance, its ability to provide a detailed evaluation of the extent of cervical cancer is not sufficient. Sheu et al (7) pointed out the difficulty of detecting microscopic tumor exten- sion. In a systematic review Bipat et al (8), reported that the sensitivity of MRI for parametrial invasion and lymph node involvement remains unsatisfactory (74% and 60%, respectively). Lower values have been ob- tained in studies with a sample size greater than 50. In a recent study of 1.5T MRI in 115 patients, Choi et al (9) reported that the sensitivity for parametrial status and lymph node metastasis was as low as 38% and 36%, respectively. Some researchers have tried to use an endovaginal coil in patients with early cervical cancer to obtain a higher SNR (10); however, this technique pro- vides only a limited field of view (FOV) and requires the invasive process of inserting a probe into the vagina. Recently, MR scanners with field strengths of 3T or higher have become available for both clinical or re- search use. Theoretically, increasing the magnetic field on a 3T MR scanner would allow us to double the SNR in comparison to that obtained using a 1.5T MR scan- ner. However, a doubling of the SNR is still not obtain- able, most likely due to signal inhomogeneity or other artifacts (11). Nevertheless, the gain in SNR in turn may be used to reduce the scanning time, or to reduce the voxel size for high-resolution images on the order of less than 1 mm. Numerous reports have documented the advantages of the 3T MR system over 1.5T for morpho- logical evaluations, functional imaging, and spectros- 1 Department of Diagnostic Imaging and Nuclear Medicine, Graduate School of Medicine, Kyoto University, Kyoto, Japan. 2 Department of Radiology, University of Cambridge, Addenbrooke’s Hospital, Cambridge, UK. 3 Department of Radiology, Graduate School of Medicine, Kyoto Univer- sity, Kyoto, Japan. Contract grant sponsor: Ministry of Education, Culture, Sports, Sci- ence and Technology; Contract grant number: 16209035. *Address reprint requests to: M.K., Department of Radiology, University of Cambridge, Box 219 Addenbrooke’s Hospital, Hills Road, Cambridge, CB2 2QQ, UK. E-mail: makok@kuhp.kyoto-u.ac.jp Received February 7, 2006; Accepted September 28, 2006. DOI 10.1002/jmri.20842 Published online 00 Month 2006 in Wiley InterScience (www. interscience.wiley.com). JOURNAL OF MAGNETIC RESONANCE IMAGING 25:527–534 (2007) © 2007 Wiley-Liss, Inc. 527