Petrosal sinus sampling was performed 171 times in 157 patients with known or suspected Cushing disease. In all cases, samples for measuring adrenocorticotropic hormone (ACTH) levels were obtained from both inferior petrosal sinuses and a peripheral vein, both before and after intravenous corticotropin-releasing hormone (CRH) administration. In seven examinations in seven patients, a ratio, or gradient, between the ACTH levels in the inferior petrosal sinuses existed in the samples obtained prior to CRH stimulation; however, poststimulation samples demonstrated reversal of the gradient, suggesting lateralization to the contralateral side of the pituitary gland. Data from surgical exploration of the pituitary gland were available for six of these patients, all of whom had adenomas. Prestimulation samples provided correct lateralization in three patients, and in two patients the poststimulation samples provided correct lateralization. One patient had a midline adenoma. When the results of pre- and poststimulation petrosal sinus sampling conflict, neither can be relied on uniformly to provide correct lateralization of ACTH-secreting pituitary adenomas.