The goal of this study was to improve assessment of diagnostic measures for lateral localization of aldosterone-producing adrenal adenomas preparatory to retroperitoneoscopic removal, in view of the fact that this technique allows for only unilateral access. A retrospective study was carried out of the medical records of 64 patients (38 women, 26 men, average age 46.8+/-11.2) who underwent surgery at University Hospital, Münster, between 1969 and 1998. Seventeen of the 64 patients presented with hyperplasia and 47 had adrenal adenoma. In cases of hyperplasia, computerized tomography imaged a false-positive unilateral tumor 10 times, a false-negative 3 times, and a unilateral hyperplasia 1 time (ultrasonography: tumor 2 times, false-negative 3 times; 131I-Iodomethylnorcholesterol scintigraphy: tumor 5 times, false-negative 1 time, correct 1 time). In cases of adenoma, computerized tomography yielded accurate results 40 times, imaged a false-negative 2 times, and indicated the incorrect side 1 time (Ultrasonography: false-negative 12 times, correct side 9 times, incorrect side 1 time; 131I-Iodomethylnorcholesterol scintigraphy: correct side 19 times, false-positive (both sides) 5 times, negative 3 times, incorrect side 2 times). Venous sampling, which was carried out seven times, yielded accurate results six times, and failed technically one time. Venous sampling appears to be the method of choice for preoperative lateral localization. Thus, retroperitoneoscopic treatment of Conn's syndrome should not be carried out unless venous sampling is carried out first.