Adrenal Vein Sampling: Rethinking the use of Cortisol for Selectivity Index Calculations. Results of Mulitplex Analysis of 6 Adrenal Steroids

Adrenal venous sampling (AVS) is the gold standard for localization of aldosterone (A) producing adenoma. Normally, AVS samples are analyzed for both A and cortisol (F) only. Concentrations of F are used to calculate the selectivity index (SI), the ratio of the adrenal venous F to peripheral F, to establish successful cannulation. F is then used to normalize A results and to calculate the lateralization index (LI), defined as LI=(A/F)dominant/(A/F)non-dominant. Recently there has been some interest in measuring other adrenal steroids in AVS collections (1,2). We have developed a 6-plex Liquid Chromatography and Tandem Mass Spectrometry (LC-MS/MS) assay for adrenal steroids in the mineralocorticoid/gluccorticoid pathway to investigate alternative means of calculating the SI and LI.

The LC-MS/MS method concomitantly measures A, F, 11-deoxycorticosterone (DOC), corticosterone (B), 18-hydroxy-11-deoxycorticosterone (18OHDOC), and 18-hydroxycorticosterone (18OHB). Using 250μL of sample, serum is extracted by supported liquid extraction using SLE+ 400μL plates (Biotage, Charlotte, NC) with methyl-tertbutyl ether. Gravimetric standards were prepared in steroid free serum (Golden West Biologicals, Temecula, CA). Internal standards employed were d7-A, d4-F, d8-B and d8-DOC. Following dry-down, samples were analyzed on a Shimadzu Prominence LC20AC using a Gemini-NX 100x2.1mm, 3.5m column (Phenomenex, Torrance, CA) maintained at 55^oC and an AB SCIEX API5000 triple quadrupole MS/MS in positive ESI mode. The method was applied to 22 adrenal vein sampling cases (6 bilateral adrenal hyperplasia, 6 Lt adenoma, 9 Rt adenoma, 1 non-diagnostic) both pre and post 250 μg ACTH stimulation.

After log-transform, based on the weakness of correlation to adrenal vein A concentrations, results demonstrate that F is the most appropriate normalizing steroid for LI calculations (r=0.50,p<0.001), followed by B (0.64,p<0.001). However, F has the smallest adrenal to peripheral gradient indicating that it is the least sensitive marker of successful cannulation. Using alternative steroids for SI-calculation, median SIs were: 3.9, 14.2, 17.4, 19.4, 21.6, 26.9 for F, DOC, A, B, B18OH, DOC18OH respectively. All SI values for alternative steroid markers were significantly higher than those of F (p<0.001). Post-ACTH results similarly demonstrate that F and B are the most appropriate normalizing steroids and other steroids are more sensitive markers of cannulation than F by ~2.5-5-fold in the median.