ISSN 1470-3947 (print) | ISSN 1479-6848 (online)

Endocrine Abstracts (2019) 63 P34 | DOI: 10.1530/endoabs.63.P34

Aldosterone measurement in the Diagnosis of primary aldosteronism - comparison between two automated immunoassays and two liquid chromatography tandem mass-spectrometry methods

Valentina Fagotto1,2, Olaf Uhl1, Daniel Heinrich1, Nina Nirschl1, Lisa Sturm1, Jürgen Kratzsch3, Yoon Ju Bae3, Charlotte Fries4, Wiebke Fenske4,5, Martin Reincke1 & Martin Bidlingmaier1

1Medizinische Klinik und Poliklinik IV, Klinikum der Universität München, Munich, Germany; 2Department of Experimental and Clinical Sciences, University of Udine, Udine, Italy; 3Institute of Laboratory Medicine, Clinical Chemistry, and Molecular Diagnostics, University of Leipzig, Leipzig, Germany; 4Medical Department III, Endocrinology, Nephrology, Rheumatology, University Hospital of Leipzig, Leipzig, Germany; 5Integrated Research and Treatment Centre for Adiposity Diseases, Department of Medicine, University of Leipzig, Leipzig, Germany.

Poor agreement between different analytical methods measuring aldosterone concentrations impedes applicability of uniform cut-offs in the diagnostic work-up of primary aldosteronism (PA). We compared 4 analytical methods (2 immunoassays and 2 LC-MS/MS) in plasma samples before and after saline infusion test (SIT). 80 hypertensive patients underwent the standardized diagnostic work-up within the German Conn’s registry. Patients (PA n=39/non-PA n=41) were classified by a experienced endocrinologists panel based on suppression of aldosterone as measured by our routine immunoassay (LIAISON, Diasorin, LSN) to concentrations <50 pg/mL, imaging, adrenal vein sampling findings, immunohistochemistry and treatment response. Aliquots from all blood samples were used to also measure aldosterone by the IDS iSYS immunoassay (iSYS) and 2 different LC-MS/MS methods (Munich LC-MS, Agilent 1290/Sciex 6500+, utilizing a Chromsystems commercial kit, and Leipzig LC-MS, an in-house method published before). Aldosterone concentrations measured by the 3 alternative methods moderately correlated to those obtained by LSN (Pearson’s r from 0.917 to 0.959), but agreement was limited (slope vs. LSN from 0.6486 to 0.9635, intercept from −24.3 to −10.0). Correlation was best between Leipzig LC-MS and iSYS. Mean and median aldosterone concentrations of all samples (n=160) were significantly higher with LSN as compared to all other assays. Bland-Altman plots showed significant mean bias (−81.15% to 14.78%) with higher scatter at concentrations <100 pg/mL. Compared to the expert panel classification, LSN post SIT aldosterone was falsely high (>50 pg/mL) in 7 cases (17%), but there were no false negatives. Because all other methods reported lower absolute aldosterone concentrations, applying the same cut-off would have resulted in a high number of false negatives (iSYS 38.5%, Leipzig LC-MS 28.2%, Munich LC-MS 56.4%). Accordingly, applying the cut-off of 50 pg/mL to ROC analysis AUC was 0.967 for LSN, but lower for iSYS (0.838), Munich LC-MS (0.883) and Leipzig LC-MS (0.916). Using the expert panel classification as ‘gold standard’, the best cut-offs for aldosterone during SIT are 61 pg/mL (LSN), 37 pg/mL (iSYS), 34 pg/mL (Leipzig LC-MS) and 20 pg/mL (Munich LC-MS). Aldosterone concentrations measured by different analytical methods differ significantly, particularly at lower concentrations, crucial for interpretation of SIT. Discrepancies occur between different immunoassays, between immunoassays and LC-MS methods, but also between different LC-MS methods. Our data emphasize the need for method-specific aldosterone cut-offs, even when LC-MS methods are used. Compared to the cut-off of 50 pg/mL traditionally used with LSN, other methods evaluated require significantly lower cut-offs to achieve the same sensitivity in detecting PA.