DISCREPANCIES IN THE OUTCOME OF PLASMA ALDOSTERONE LEVELS ASSESSED BY IMMUNOASSAY AND MASS SPECTROMETRY - CLINICAL RELEVANCE FOR THE DIAGNOSIS OF PRIMARY HYPERALDOSTERONISM

Abstract

Objective: To assess potential discrepancies in the clinical interpretation of the salt loading test (SLT) resulting from the assessment of aldosterone concentrations by two different methods - immunoassay and mass spectrometry. Design and method: Eighty-one patients (47F, 34M, mean age 47.7 ± 14.1 years) with suspected primary hyperaldosteronism (PA) who underwent an SLT were included in the ongoing study. Patients had their treatment modified according to current guidelines prior to the test. The SLT was performed in the sitting position and blood samples were taken before and after 4 hours of administration of 2 litres of saline. An aldosterone concentration higher than 61 ng/dL (170 pmol/L) was considered indicative of a diagnosis of PA. Plasma aldosterone concentrations from the same samples were assessed by: (1) chemiluminescence-based immunoassay (CLIA), LIAISON® by DiaSorin, and (2) liquid chromatography-mass spectrometry (LC-MS), MassChrom Chromsystems kit, UHPLC technique using a QTRAP 6500+ instrument (Sciex). Renin concentration was assessed by CLIA. Results: Median plasma aldosterone concentrations after saline administration assessed by CLIA were significantly higher than those assessed by LC-MS (64.4 vs 44.0 ng/dL; p < 0.001). The correlation coefficients for plasma aldosterone concentration after saline administration was 0.69 (p < 0.001). PA was diagnosed by one of the methods in 22 of patients and by both methods in 16 patients. Conflicting results were obtained in 6 patients (7.4%): in 5 patients CLIA was suggestive of PA but LC-MS not and in one patient LC-MS was suggestive of PA but CLIA not. In four cases, plasma aldosterone concentrations were slightly above or below the cut-off values (CLIA vs LC-MS: 62.9 vs 53; 61.9 vs 50; 97.8 vs 61 and 112 vs 58 ng/dL) in at least one of the methods. In the other two cases, the discrepancies were more pronounced (50 vs 80 ng/dL and 126 vs 24 ng/dL). Conclusions: Discrepant results of aldosterone assessment by CLIA and LC-MS methods were obtained in 7.4% of patients. Discrepancies were generally related to borderline results. If a plasma aldosterone concentration value close to the cut-off values is obtained, a repeat determination using another method should be considered.