Michael Muzinic

Quantitative Comparison of Yttrium-90 (90 Y)-Microspheres and Technetium-99m (99m Tc)-Macroaggregated Albumin SPECT Images for Planning 90 Y Therapy of Liver Cancer

Yttrium-90 (90 Y)-microspheres administered via the hepatic artery has been used for the treatment of unresectable primary or metastatic cancer in the liver. Prior to 90 Y therapy, however, the 90 Y administered activity and the percent shunting to lungs must be determined, most commonly by gamma camera imaging of technetium-99m (99m Tc)-macroaggregated albumin (MAA). The purpose of the current study was to identify and evaluate an objective measure of the correlation of 90 Y and MAA activity distributions and thus assess the reliability of MAA imaging for evaluation of 90 Y administered activity and tumor and liver radiation doses. The MAA study consisted of two acquisitions. After administration of 185 MBq of MAA, a partial-body or so-called breakthrough scan was performed in order to determine the percent shunting to lungs. Immediately after a breakthrough scan, a combined single-photon emission computed tomography (SPECT)/transmission computed tomography (CT) scanner was used to image MAA distribution in order to derived the prescribed 90 Y administered activity based on tumor and liver dosimetry. 90 Y SPECT/CT was performed 2–4 weeks later and activities used were in the range of 777–2,442 MBq. In order to compare 90 Y and MAA SPECT images, first the respective CT image sets were registered using a transform based on normalized mutual information. The transform thus derived was used to align the 90 Y and MAA SPECT image sets, and the Spearmans (rho) rank correlation as well as image distance (L2-norm) between the registered SPECT images were then calculated. The Spearmans rank correlation values ranged from 0.451 to 0.818 and the L2 distances from 0.626 to 2.889. Based on visual inspection, the registration of the 90 Y and MAA SPECT images appeared reasonably accurate. The regression coefficient (r) between visual scoring and the Spearmans rank correlation was 0.65 and between visual scoring and L2 distance 0.61. The Spearmans rank correlation thus appears to be more reliable than the image distance for assessing the correlation of the 90 Y and MAA images.

Role of in-111 labeled leukocyte scintigraphy in the diagnosis of intracerebral lesions.

The differential diagnosis of intracerebral enhancing lesions on contrast computed tomography includes tumors, abscesses, and cerebrovascular accidents. Particularly important is the differentiation between tumor and abscess. While ln-111 labeled leukocyte imaging is an accurate test for identification of foci of infection in general, the role of this procedure in the evaluation of the intracranial lesion is not well established. We undertook a retrospective review of 16 patients with contrast enhancing intracerebral lesions identified on computed tomography, who were also studied with labeled leukocyte imaging. Final diagnoses were: abscess (n=2), primary brain tumor (n=6), metastasis (n=4), dermoid cyst (n=1), and cerebral infarct (n=3). There were two positive labeled leukocyte studies; both were cerebral abscesses. No labeled leukocyte activity was identified in any of the tumors or infarcts. We conclude that ln-111 labeled leukocyte imaging is an accurate method of differentiating infectious from noninfectious causes of intracerebral lesions identified on computed tomography.