The major aldosterone metabolite 3 alpha,5 beta tetrahydroaldosterone reflects up to 45% of the aldosterone secretion. Its 24-h urinary excretion is likely to provide an accurate index of the daily aldosterone production and to be an indicator for primary aldosteronism (PA). In a prospective study, the validity of tetrahydroaldosterone as a screening test for PA was evaluated in comparison to serum potassium, plasma aldosterone, plasma renin activity, plasma aldosterone/renin activity ratio (PARR), as well as 24-h urinary aldosterone-18-glucuronide and free aldosterone. A total of 111 normotensive individuals, 412 PA patients and 1453 essential hypertensive patients, were studied. The effect of blood sampling technique on potassium level was also investigated. Tetrahydroaldosterone differentiated PA from essential hypertension with a sensitivity of 96% and a specificity of 95%. The sensitivity was 89% for plasma aldosterone, 87% for free aldosterone, 85% for PARR, 71% for aldosterone-18-glucuronide and 51% for renin activity. Specificities varied between 91% and 85%. The combined use of the parameters plasma aldosterone > or =9.0 ng/dL and PARR > or =25 resulted in a sensitivity of 82% and specificity of 95%. Forearm exercise proved to be a source of erroneous elevations of potassium sufficient to obscure the suspicion of PA. The data suggest that tetrahydroaldosterone is the most reliable screening test for PA. Tetrahydroaldosterone determination in combination with aldosterone-18-glucuronide and free aldosterone increases diagnostic specificity for PA. Potassium, renin, plasma aldosterone, and basal PARR are inadequate screening procedures because they are subject to high rates of false-positive and false-negative results.