We have compared prospectively the sensitivity, specificity and accuracy of CT and MRI in a series of 27 consecutive patients (age 23-76 yrs, 17 females, 10 males) with clinically suspected primary aldosteronism. We found 13 patients with a unilateral aldosterone-producing adenoma (11 on the left and 2 on the right side), 6 with idiopathic hyper-aldosteronism and 8 with primary hypertension, which in two cases was associated with a nonfunctioning adrenal adenoma. The diagnosis of aldosterone-producing adenoma was confirmed at surgery and pathology in all cases. Idiopathic hyper-aldosteronism was diagnosed on the basis of the results of dexamethasone-suppressed adrenal scintigraphy and/or selective adrenal vein sampling. MRI correctly identified all cases of aldosterone-producing adenoma, but gave false positive results in five cases: one had idiopathic hyper-aldosteronism with bilateral nodular hyperplasia and four primary hypertension, which in two patients was associated with a nonfunctioning adrenal adenoma. Therefore, the sensitivity of MRI was 100%, its specificity 64% and overall diagnostic accuracy 81%. In comparison, CT correctly recognized only eight of the 13 patients with aldosterone-producing adenoma and gave false positive results in three primary hypertensives, including the two patients with a nonfunctioning adrenal adenoma. Therefore, its sensitivity, specificity and accuracy were 62, 77 and 69%, respectively. Based on these results, it could be anticipated that about four of every ten patients with aldosterone-producing adenoma would not be correctly diagnosed by CT.(ABSTRACT TRUNCATED AT 250 WORDS)