Can incomplete adrenal venous sampling data be used in predicting the subtype of primary aldosteronism?

Background. - Adrenal venous sampling (AVS) is the gold standard for preoperative differentiation between unilateral and bilateral primary aldosteronism (PA). However, results are sometimes vitiated by failing to access the right adrenal vein. Materials and methods. - The present study assumed that clinical decisions can be made with incomplete AVS data, by comparing aldosterone/cortisol (A/C) ratio in both left and right adrenal veins with that in the inferior vena cava (LAV/IVC and RAV/IVC). Receiver operation characteristic (ROC) curve and scatterplot were used to certify the upper and lower cutoffs and to analyze the significance of discrimination. One hundred and sixty patients diagnosed with PA from April 2017 to June 2018 underwent AVS in the Urology Department of West China Hospital, Chengdu, China. One hundred and eleven with complete AVS data were divided into 3 groups: left-sided (N= 40), right-sided (N=29) and bilateral (N=42). We also collected patients from September 2018 to April 2019 in our department as validation cohort to test our hypothesis. Results. - On the basis of LAV/IVC, RAV/IVC and diagnostic category, upper cutoff was 1.14 (50% sensitivity and 100% specificity) and lower cutoff 0.07 (27.5% sensitivity and 100% specificity) for LAV/IVC, and 1.04 (55% sensitivity and 100% specificity) and 0.08 (40% sensitivity and 100% specificity), respectively, for RAV/IVC. Conclusion. - The diagnostic model in this study contributes to clinical decision-making in patients with only partial PA with incomplete AVS data. (C) 2019 Elsevier Masson SAS. All rights reserved.