Institutional members access full text with Ovid®

Magnetization Transfer Imaging of the Pituitary: Further Insights into the Nature of the Posterior Bright Spot

Holder, Chad A.; Elster, Allen D.

Journal of Computer Assisted Tomography: March/April 1997 - Volume 21 - Issue 2 - pp 171-174

Purpose: After more than a decade of investigation, the chemical nature of the posterior pituitary “bright spot” remains elusive. Speculations into the source of this high signal have included relaxation of water by phospholipid vesicles, vasopressin, paramagnetic substances, and membrane-associated proteins. We hypothesized that if the T1 shortening observed in this structure were caused by water/macromolecular interactions, this interaction could be modulated by the use of magnetization transfer (MT) saturation.

Method: Twenty-five normal subjects were recruited over a 2 month period who were identified on routine T1 sagittal head images to have pituitary bright spots with cross-sectional area of >2 mm2. Thin section (4 mm), T1-weighted (SE 450/20) sagittal MR images were obtained both with and without the use of an MT suppression pulse (1,000 Hz offset, 200 Hz bandwidth, peak amplitude 7.3 μT). Region-of-interest measurements were made of the posterior pituitary lobe, anterior pituitary lobe, genu of corpus callosum, and pons, with MT ratios (MTRs) calculated for each structure.

Results: Relatively low (and similar) MTRs were observed in both parts of the pituitary gland: anterior lobe, 12.3%; posterior lobe 10.8%. Paired t test analysis demonstrated no statistically significant difference between the MTRs of the anterior and posterior pituitary lobes (p = 0.23). Considerable suppression of signal was noted in the genu (MTR = 25.0%) and pons (MTR = 21.9%). The MTRs of both portions of the pituitary differed significantly from those of the genu and pons (p < 0.00001).

Conclusion: The high signal of the posterior pituitary gland suppresses only slightly on MT images, having a behavior similar to that in the anterior lobe but significantly different from the rest of the brain. These findings suggest that direct water/macromolecule, water/membrane, or water/phospholipid interactions are not likely to be responsible for the appearance of the bright spot. The experimental results are more consistent with water interacting with a paramagnetic substance or low molecular weight molecule (e.g., vasopressin, neurophysins).

From the Department of Radiology, Bowman Gray School of Medicine, Wake Forest University, Medical Center Boulevard, Winston-Salem, NC 27157-1022, U.S.A.

Address correspondence and reprint requests to Dr. A. D. Elster.

© Lippincott-Raven Publishers