The purpose of this study was to determine how the gray-to-white matter contrast in healthy subjects changes on high-b-value diffusion-weighted imaging (DWI) acquired at 3 T and evaluate whether high-b-value DWI at 3 T is useful for the detection of cortical lesions in inflammatory brain diseases.
Ten healthy volunteers underwent DWI at b = 1000, 2000, 3000, 4000, and 5000 s/mm2 on a 3-T MRI unit. On DW images, 1 radiologist performed region-of-interest measurements of the signal intensity of 8 gray matter structures. The gray-to-white matter contrast ratio (GWCR) was calculated. Ten patients with inflammatory cortical lesions were also included. All patients underwent conventional MRI and DWI at b = 1000 and 3000 s/mm2. Using a 4-point grading system, 2 radiologists independently assessed the presence of additional information on DW images compared with fluid-attenuated inversion recovery images. Interobserver agreement was assessed by κ statistics.
In the healthy subjects, the b value increased as the GWCR decreased in all evaluated gray matter structures. On DW images acquired at b = 3000 s/mm2, mean GWCR was less than 1.0 in 7 of 8 structures. For both reviewers, DWI at b = 3000 s/mm2 yielded significantly more additional information than did DWI at b = 1000 s/mm2 (P < 0.05). Interobserver agreement for DWI at b = 1000 s/mm2 and b = 3000 s/mm2 was fair (κ = 0.35) and excellent (κ = 1.0), respectively.
At 3-T DWI, the gray-to-white matter contrast in most gray matter structures reverses at b = 3000 s/mm2. In the evaluation of cortical lesions in patients with inflammatory brain diseases, 3-T DWI at b = 3000 s/mm2 was more useful than b = 1000 s/mm2.
From the Departments of *Diagnostic Radiology and †Neurology, Graduate School of Medical Sciences, and ‡Department of Medical Imaging, Faculty of Life Sciences, Kumamoto University, Kumamoto, Japan.
Received for publication October 1, 2012; accepted January 18, 2013.
Reprints: Toshinori Hirai, MD, PhD, Department of Diagnostic Radiology, Graduate School of Medical Sciences, Kumamoto University, 1-1-1 Honjo, Kumamoto 860-8556 Japan (e-mail: firstname.lastname@example.org).
The authors declare that they have no conflict of interest.