Octreotide scintigraphy has been advocated as the principal imaging modality for localizing ectopic ACTH-secreting tumors in Cushing's syndrome. To assess its usefulness we reviewed the course of 18 consecutive patients with ectopic ACTH-producing tumor. Imaging included (111)In-pentetreotide scintigraphy, computed tomography (CT), and/or magnetic resonance imaging (MRI). Tumor was detected initially in 7/18 patients, and in 3/18 during follow-up. No ACTH-secreting tumor was detected by octreotide scintigraphy when CT/ MRI were negative. Seventeen of forty octreotide scintigrams were abnormal. CT and/or MRI confirmed tumors in 10, but demonstrated nonendocrine lesions in association with 6 false positive octreotide scintigrams. Hepatic venous sampling for ACTH refuted one lesion detected by octreotide and CT scans. Twenty-three of forty octreotide scintigrams were normal. Of these, 8 were false negative, as CT and/or MRI detected tumor; 10 agreed with negative CT and MRI, and 5 correctly refuted false positive CT and/or MRI scans. Repeated CT/ MR, but not octreotide scintigraphy, led to tumor resection in 2 patients. We conclude that octreotide scintigraphy does not offer greater sensitivity than CT/MRI and that false positive scans are common. Although octreotide scintigraphy may be helpful in selected cases, it is not a significant advance over conventional imaging for ectopic ACTH-secreting tumors.