3.0-T high-field magnetic resonance imaging of the female pelvis: preliminary experiences.

The purpose of this study was to evaluate if 3.0 T allows for clinically useful pelvic magnetic resonance imaging, i.e. if familiar image quality and tissue contrast can be achieved at 3.0 T as compared with at 1.5 T. Adapting a 1.5-T protocol to the 3.0-T environment is subject to a variety of factors. In order to reduce the number of potential variables, we chose two cornerstones: the 3.0-T sequence should have similar spatial resolution and acquisition time; furthermore, the contrast parameters repetition time (TR) and echo time (TE) were kept identical. Based on this modified 3.0-T T2-weighted turbo spin-echo sequence (TR/TE 2,705/80 ms; 0.7x1.04x4 mm measured voxel size; field of view 360 mm; 4.03-min scan time) we performed an intraindividual study on 19 patients with the 1.5-T sequence as the standard of reference. Two radiologists analyzed the examinations in consensus with regard to tissue contrast (visualization of zonal anatomy of the uterus and/or delineation of pathologic findings) rated on a three-point scale (3 is 3.0 T better; 2 is 3.0 T equal; 1 is 3.0 T worse than 1.5 T). In addition, the signal difference between muscle and bone marrow was measured as a marker for tissue contrast. The analysis of the image quality comprised the level of the artifacts (rated on a five-point scale: 1 is no artifacts; 5 is nondiagnostic study), the visual signal-to-noise ratio (rated on a three-point scale) and detail delineation. Only minor artifacts were observed at both 1.5 and 3.0 T; the difference was not statistically significant. The visual signal-to-noise ratio and the delineation of image details were rated equal for 1.5 and 3.0 T. With regard to image contrast, both qualitative analysis as well as quantitative analysis revealed comparable image contrast for the 1.5- and 3.0-T protocols. Pathological findings were seen equally well with both field strengths. Clinically diagnostic pelvic studies of high image quality can be obtained using a 3.0-T scanner with our modified examination protocol. To fully exploit the capability of the high-field technique, and to point out potential advantages, further intraindividual studies are needed, with the adjustment of other imaging parameters to the high-field environment.