THE ROLE OF SPECT IN TUMOR DIAGNOSIS

Kuhi et al. introduced single photon emission corn puterized tomography (SPECT) in 1963 (1), yet it has been only during the past few years that a concentrated effort has been made to develop systems for clinical ap plications. Preliminary reports have been published evaluating the accuracy of SPECT in the detection of mass lesions of the brain, chest, and abdomen, using technetium-99m, iodine-123 and 131, and gallium-67 labeled compounds. The results indicate that SPECT studies have a diagnostic accuracy competitive with that of other imaging modalities. The purpose of this paper is to evaluate the current status of emission computed tomography and to suggest areas where it may be used advantageously for the examination of patients with known or suspected cancer. For the detection of a lesion by SPECT, a target-to nontarget (T/NT) ratio of 2 is adequate (2), whereas for most planar imaging a ratio of at least 5 to 6 is usually necessary. if the T/NT ratio on the planar image is in creased to a level comparable to that provided by to mography, the detection oflesions by planar imaging can be markedly improved. For example, in a study of pa tients with cancer, background activity from the Ga-67 was reduced by using a second radionuclide (Tc-99m HEDP) and subtraction techniques (3). When the un processed Ga-67 images were compared with those ob tamed using the second-radionuclide subtraction method, 30 % of the lesions originally interpreted as positive and 67 % of those interpreted as suspicious were found to be negative after the background was dimin ished—twice as many true lesions were detected on the processed images as had been found on the original studies. These results demonstrated the increased ac curacy of tumor detection that could be attained if th