SUN-364 In Search of Predictors of Concordance between Imaging and Adrenal Vein Sampling in Unilateral Primary Aldosteronism

Abstract

Background: Correct subtyping of primary aldosteronism (PA) is critical for guiding clinical management. Adrenal imaging is less accurate than adrenal vein sampling (AVS); nonetheless, AVS is invasive, technically challenging and limited to specialty centers. Objective: We aimed to identify predictors of concordance between computed tomography (CT) and AVS in patients with unilateral PA, which could circumvent the need for AVS in a subset of PA patients. Methods: We retrospectively studied all patients with confirmed PA who underwent AVS in our institution between 2009-2018. Simultaneous AVS samples were obtained both before and 10-30 minutes after cosyntropin stimulation. Only patients with unilateral (U) PA, based on a lateralization index (LI) ≥ 4 pre- and/or post-cosyntropin were included from this study. Demographic, laboratory and imaging data were collected. Aldosterone synthase (CYP11B2)-guided targeted next generation sequencing was performed in available cases. Chi-square test was used to compare proportions between two groups. Logistic regression was performed to identify predictors of CT-AVS lateralization concordance. Results: In total, 177 patients, 107 (60.5%) men, with a mean age of 52 years (range 25-79), met our study requirements. Of these, 124 (73%) patients were Caucasian (CA), 41 (24%) African American (AA) and 5 (3%) Asian (AS). AVS subtyping was concordant both pre- and post-cosyntropin (UU) in 118 (67%) patients and discordant in 59 (23%) patients (UB, 36; BU, 23). The mutation status was analyzed in 62 patients who underwent adrenalectomy and identified: 30 KCNJ5, 18 CACNA1D, 8 ATP1A1, 5 ATP2B3 and 1 CTNNB1 mutations. AVS-CT agreement was higher in patients with KCNJ5 than in those with other aldosterone-driver somatic mutations (90% vs. 59%, p<0.01); in AS and CA than in AA patients (80%, 70% and 32%, respectively); in UU patients than in the discordant group (69% vs. 44%, p<0.01); and for single left vs. right adrenal mass on CT (74% vs 46%, p<0.01). After adjusting for sex, LI, contralateral suppression, aldosterone concentrations, and dominant side, the size of the lesion and young age were also associated with a higher likelihood of AVS-CT agreement. Two patients age <40 y (25 y and 36 y) had CT-AVS discrepancy. LI, aldosterone concentrations in the dominant adrenal vein and periphery, contralateral suppression, and sex did not predict concordance of CT- and AVS-based PA lateralization. Conclusions: Agreement between AVS and CT for lateralizing PA is higher with young age, CA and AS race, larger adrenal lesions, and CT abnormalities in the left adrenal gland. Nonetheless, CT-based predictors remain poor compared to AVS, even in young patients with U PA.