The computed tomographic enhancement pattern of the normal canine pituitary gland.

Abstract

Dynamic computed tomography (CT) of the pituitary gland was performed on four healthy male dogs of similar size, weight and age. The pituitary gland region was first identified on lateral scout and transverse non-contrast images. After localization, water soluble iodinated contrast medium was administered intravenously as a bolus at a dose of 1 ml/lb using a pressure injector at an injection rate of 10 ml/sec and a total of 40 post contrast images of the pitutary gland were acquired. No images were made after 400 seconds. The same pituitary region was imaged in each slice. The slice thickness was 1.5 mm, with a two second scan time and an eight second delay between images (resulting in images every ten seconds). The contrast medium injection and initial image were acquired simultaneously, resulting in a non-contrast enhanced initial image. At the completion of the CT scan, a region of interest (ROI) was drawn around the pituitary gland and time density data were obtained. The mean pituitary Hounsfield number was plotted as a function of time. A bi-exponential least squares model was used to derive the best fitting line through the data. The mean relative peak increase in pituitary Hounsfield Units (HU) was 65.9% +/- 2.1%. After the initial increase there was a decrease in pituitary Hounsfield number with a half-time of 16.1 seconds, followed by a slower phase with a half-time of 16.5 minutes. The mean pituitary gland HU value during the period of gradual opacity decline was 35.0% +/- 4.4% above that of the pre-contrast image. Establishing the enhancement pattern in the canine pituitary gland is the precursor to the clinical application of dynamic CT of the pituitary gland to diagnose pituitary microadenomas and/or small macroadenomas before they become exceptionally large.