Abstract
OBJECTIVE: Saline infusion testing (SIT) for confirmation of primary aldosteronism (PA) is based on impaired aldosterone suppression in PA compared to essential hypertension (EH). In the past, aldosterone was quantified using immunoassays (IA). Liquid chromatography tandem mass spectrometry (LC-MS/MS) is increasingly used in clinical routine. We aimed at a method-specific aldosterone threshold for the diagnosis of PA during SIT and explored the diagnostic utility of steroid panel analysis. DESIGN: Retrospective cohort study of 187 paired SIT samples (2009-2018). Diagnosis of PA (n = 103) and EH (n = 84) was established based on clinical routine workup without using LC-MS/MS values. SETTING: Tertiary care center. METHODS: LC-MS/MS using a commercial steroid panel. Receiver operator characteristics analysis was used to determine method-specific cut-offs using a positive predictive value (PPV) of 90% as criterion. RESULTS: Aldosterone measured by IA was on average 31 ng/L higher than with LC-MS/MS. The cut-offs for PA confirmation were 54 ng/L for IA (sensitivity: 95%, 95% CI: 89.0-98.4; specificity: 87%, 95% CI: 77.8-93.3; area under the curve (AUC): 0.955, 95% CI: 0.924-0.986; PPV: 90%, 95% CI: 83.7-93.9) and 69 ng/L for LC-MS/MS (79%, 95% CI: 69.5-86.1; 89%, 95% CI: 80.6-95.0; 0.902, 95% CI: 0.857-0.947; 90%, 95% CI: 82.8-94.4). Other steroids did not improve SIT. CONCLUSIONS: Aldosterone quantification with LC-MS/MS and IA yields comparable SIT-cut-offs. Lower AUC for LC-MS/MS is likely due to the spectrum of disease in PA and previous decision making based on IA results. Until data of a prospective trial with clinical endpoints are available, the suggested cut-off can be used in clinical routine.