In the present study, we developed a new chemiluminescent enzyme immunoassay (CLEIA) using a two-step sandwich method to measure aldosterone concentrations. We investigated serum and plasma aldosterone concentrations in 75 blood samples from 27 patients using a radioimmunoassay (RIA) and the CLEIA (with current and newly improved reagents) as well as liquid chromatography-tandem mass spectrometry (LC-MS/MS). Based on the results of the Passing–Bablok regression analysis, the aldosterone levels measured using CLEIA with the new reagents and those measured by LC-MS/MS were found to be significantly correlated (slope, 0.984; intercept, 0.2). However, aldosterone levels varied depending on the measurement method (i.e., CLEIA with the new reagent, CLEIA with the current reagent, and RIA). Aldosterone levels were lower with the improved CLEIA method than with RIA and CLEIA using the current reagent. Therefore, the cutoff values of the screening test as well as those of the confirmatory test for primary aldosteronism (PA) should be adjusted to follow current clinical practice guidelines for PA. The formula that can be used to obtain the aldosterone level (pg/mL) when using CLEIA with the new reagent is 0.765 × RIA (pg/mL) − 33.7. This formula will enable PA cutoff values to be set for provisional screening and confirmatory tests.