CAN WE USE FREE METANEPHRINE TO IMPROVE INTERPRETATION OF ADRENAL VENOUS SAMPLING ?

Abstract

Objective: Adrenal venous sampling (AVS) is the gold standard to assess lateralization of aldosterone secretion in primary aldosteronism (PA). The selectivity index (SIcortisol, adrenal:peripheral cortisol ratio) determines correct catheter positioning. The lateralization index (LIcortisol, aldosterone:cortisol ratios between adrenal veins) distinguishes unilateral aldosterone producing adenoma (APA) from bilateral adrenal hyperplasia (BAH). Cortisol is not always reliable and metanephrine has been alternatively suggested for SI determination. The aim of our study was to assess the performance of SImetanephrine and LImetanephrine. Design and method: This retrospective study conducted at the HEGP ESH excellence center included 245 patients who had 1) AVS with simultaneous adrenal and peripheral vein samplings without cosyntropin stimulation between 07/2013 and 05/2017, 2) adrenal and peripheral vein samples stored at −80°C to measure retrospectively free metanephrine. Receiver operating characteristic (ROC) curves were used to assess the performance of SImetanephrine and LImetanephrine. Results: Based on SIcortisol and LIcortisol, 198/245 patients had successful AVS (SIcortisol above 2) among whom 108 had APA (LIcortisol above 4) and 90 BAH (LIcortisol below 4). 48/245 patients had failed AVS. Among the 245 AVS, 434 adrenal samples had SIcortisol above 2. A SImetanephrine of 16 was the optimal threshold for successful AVS. This cutoff confirms the selectivity of 22 AVS among the 48 considered as non-selective with SIcortisol. 65 patients underwent unilateral adrenalectomy with 6-month follow up confirming PA biochemical cure (PASO criteria). ROC curve analysis of LImetanephrine plotted APA (n = 65) versus BAH (n = 52) (AUC = 0.954). There was a large overlap of LImetanephrine between APA and BAH. A LImetanephrine threshold of 5 had 80% sensitivity and 95% specificity to distinguish between APA from BAH. Conclusions: We confirmed that SImetanephrine is superior to SIcortisol in assessing the selectivity of AVS. A threshold of 16 for SImetanephrine decreased the rate of AVS falsely considered as failed. In contrast, LImetanephrine did not have enough sensitivity/specificity to distinguish between APA and BAH. The lack of a gold standard to confirm BAH may participate in this finding. A prospective study would be needed to confirm these results