Primary aldosteronism: diagnostic accuracy of the losartan and captopril tests.

BACKGROUND To assess whether angiotensin-II receptor blockers (ARBs) offer any additional advantage in confirming the diagnosis of primary aldosteronism (PA) and their use in the differentiation of PA subtypes. METHODS A prospective, cohort, head-to-head study was conducted between July 2003 and July 2006. A total of 135 patients received captopril and losartan tests to confirm the diagnosis of PA in the TAIPAI (Taiwan Primary Aldosteronism Investigation) intervention. RESULTS In total, 71 patients were diagnosed with PA. The area under the receiver-operating characteristic (ROC) curve of the postcaptopril plasma aldosterone concentration (PAC) was significantly less than that of the postlosartan PAC (0.744 vs. 0.829, P = 0.038). Using an aldosterone-renin ratio (ARR, ng/dl per ng/ml/h) >35 with a PAC >10 ng/dl, the specificity was 89.1% vs. 93.8% and the sensitivity was 66.2% vs. 84.5 % for the captopril test vs. the losartan test, respectively. With respect to the losartan test, the accuracy was 88.9%, the agreement was good (k = 0.778), and there was no disagreement with the McNemar test (P = 0.118). Losartan had the advantage of a better negative predictive value to exclude PA when patients were referred with a serum potassium (SK) level <3.8 mmol/l. When a postlosartan ARR >60 was the cutoff value, the positive predictive value was 82% with a negative predictive value of 57% in distinguishing aldosterone-producing adenomas (APAs) from idiopathic hyperaldosteronism (IHA). CONCLUSIONS The postlosartan ARR and PAC were shown to have better accuracy for the diagnosis of PA than the captopril test. With a postlosartan ARR >60, APAs can be adequately differentiated from IHA.