| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
BRIEF REPORT |
Divisions of Endocrinology and Metabolism (R.J.A., S.S., N.C.), and of Hypertension (W.V., S.D.N., R.G.V.), Department of Internal Medicine, Division of Gastrointestinal and Endocrine Surgery, Department of Surgery (F.E.N., S.A.H.), Division of Vascular and Interventional Radiology, Department of Radiology (B.L.D., S.C.J., C.K.T., J.L.), and Departments of Pathology (F.H.W.) and of Clinical Sciences (D.L.), University of Texas Southwestern Medical Center, Dallas, Texas 75390-8857; and Endocrinology Laboratory (D.W.C.), Esoterix, Incorporated, Calabasas Hills, California 91301
Address all correspondence and requests for reprints to: Dr. Richard J. Auchus, Division of Endocrinology and Metabolism, Department of Internal Medicine, University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, Texas 75390-8857. E-mail: richard.auchus{at}UTSouthwestern.edu.
| Abstract |
|---|
|
|
|---|
Objectives: We measured 18OHB, A, and cortisol (F) in blood from adrenal vein sampling (AVS) studies. We compared the discriminatory power of gradients in 18OHB/A and 18OHB/F ratios with A/F ratio gradients for distinguishing APA from IHA.
Design, Setting, and Subjects: We measured 18OHB and A in excess serum from 23 AVS studies performed at our university hospitals.
Main Outcome Measures: We calculated the ratios 18OHB/A, 18OHB/F, and A/F for all specimens, and determined the adrenal vein gradients for these ratios.
Results: The 18OHB/A ratios were much lower in blood draining APAs (2.17 ± 0.62) than in blood draining the contralateral adrenals (12.96 ± 12.76; P < 0.001) but similar to blood draining IHA adrenals (4.69 ± 4.32; P = 0.02). In contrast, the 18OHB/F ratios were elevated in specimens from APAs (26.03 ± 11.51) compared with IHA adrenals (9.22 ± 5.18; P < 0.001) or the contralateral adrenals (6.23 ± 2.97; P < 0.001). Using 18OHB/F gradient greater than two or 18OHB/A gradient less than 0.5 as criteria for lateralization, interpretations agreed with lateralizations based on A/F gradients in 21 of 23 cases.
Conclusions: High serum 18OHB in APA reflects augmented production of both 18OHB and A, not disproportionate 18OHB secretion relative to A. The 18OHB/A and 18OHB/F gradients are useful adjuncts but not as reliable as A/F gradients for A lateralization during AVS.
| Introduction |
|---|
|
|
|---|
Various strategies have been used to distinguish unilateral A production from an aldosterone-producing adenoma (APA) from bilateral or idiopathic hyperaldosteronism (IHA) (9). Computed tomography is very sensitive, but the small size of APAs and the high prevalence of incidental adrenal tumors compromise the predictive value of computed tomography scans (10, 11, 12). The failure of A to increase with upright posture (13) and high serum 18-hydroxycorticosterone (18OHB) (14, 15) both suggest APA, but posture studies are cumbersome, and clearly discriminating limits of serum 18OHB are not established. Consequently, adrenal vein sampling (AVS) is the definitive procedure for distinguishing APA from IHA (10, 11, 12).
A greater understanding of why serum 18OHB tends to be elevated in APA cases might improve the diagnostic use of 18OHB in PA. The conversion of 18OHB to A is the final step in A synthesis, and although the 18OHB/A ratio in mixed venous blood remains constant for APA patients (16), it is not known if the efficiency of this final step is altered in APAs. If so, the measurement of 18OHB in AVS specimens might help to distinguish APA from IHA. To test this hypothesis, we retrospectively measured 18OHB with A and cortisol (F) in AVS specimens obtained during cosyntropin infusion.
| Subjects and Methods |
|---|
|
|
|---|
Subjects were identified at the University of Texas Southwestern Medical Center during 20022005. The diagnosis of PA required: suppressed plasma renin (activity < 1 ng/ml·h = <0.3 ng/liter·sec) or direct immunoreactive renin (<5 µU/ml); serum A more than 15 ng/dl (>416 pmol/liter); and failure to suppress A production with either oral salt loading (24 h urinary A > 14 µg/24 h = >39 nmol/24 h) or saline infusion (serum A > 10 ng/dl = >278 pmol/liter after 2 liters iv over 4 h). Eligible PA subjects underwent technically successful AVS with measurement of A and F as part of their evaluation. Written informed consent was obtained to measure additional analytes from excess serum obtained from AVS. The study protocol was approved by the institutional review board at the University of Texas Southwestern.
Assays and calculations
Blood samples were transported on ice and promptly centrifuged for 10 min at 4 C and 2200 rpm. Serum F was measured by immunochemiluminometric assay with an ADVIA Centaur immunoassay analyzer (Bayer Diagnostics, Tarrytown, NY), neat and with dilutions. The assay range was 0.275 µg/dl with between-run precision (coefficient of variation), 1.95.4% in the 3.937.2 µg/dl range. Serum aliquots were sent to Mayo Medical Laboratories (Rochester, MN) for A assay by their proprietary RIA.
The A/F ratio in each sample was used to localize the source(s) of A (10, 11, 12). AVS was technically successful if the F concentrations in both the adrenal vein specimens were more than three times the F concentration in the IVC specimen ("adrenal vein F step-up"). Unilateral A production (APA or primary adrenal hyperplasia) was localized to the adrenal gland from which the higher A/F ratio derived if this A/F ratio was at least four times the A/F ratio of the other adrenal vein sample ("A/F gradient"); presumed IHA was deduced if the A/F ratios in the two adrenal vein samples differed by less than four (all less than two).
Serum 18OHB was measured by HPLC-tandem mass spectrometry after solvent extraction with [2H4]-dexamethasone internal standard. Serum controls and duplicate sets of calibrators were analyzed in each batch. The 18OHB was eluted using a methanol/water gradient through a 50 x 2.1-mm reverse phase analytical HPLC column (particle size, 5 µm). An MDS-SCIEX API4000 triple quadrupole mass spectrometer (Applied Biosystems, Foster City, CA), operating in negative ion atmospheric pressure chemical ionization mode, was used for detection using selected reaction monitoring. The imprecision of the assay (coefficient of variation) ranged from 1.76.3% when measured at four levels from 55000 ng/dl. Accuracy was demonstrated by spiking 18OHB into serum with recovery of 101112%. A was measured, neat and with dilution, in the same aliquots by RIA after methylene chloride extraction. Intraassay precision ranged from 7.29.6%, while interassay precision ranged from 6.79.2% in serum controls with A concentrations of 3.837 ng/dl.
AVS procedure
AVS was performed under a continuous infusion of 50 µg/h cosyntropin (Amphastar, Rancho Cucamonga, CA) in 5% dextrose as described (12). All 27 subjects with successful AVS and retrievable specimens gave their consent for the study. An interpretation was rendered despite a F step-up less than three in four cases, but only data from the 23 AVS studies with F step-ups more than three were included for the study to prevent misclassifications. Excess serum was stored in less than 0.5-ml aliquots at 80 C until batch assayed.
Statistical analyses
Statistical analyses used SAS/STAT software, version 9.1 (SAS Institute Inc., Cary, NC). P values for comparisons between APA and contralateral 18OHB, A, F, 18OHB/A, and 18OHB/F adrenal vein samples were based on signed rank tests. P values for comparisons of 18OHB, A, F, 18OHB/A, and 18OHB/F between APA and IHA adrenal vein samples were based on rank sum tests. Tests of differences and ratios yielded the same judgment of significance.
| Results |
|---|
|
|
|---|
|
The 18OHB (5867 ± 3050 ng/dl) and A (2009 ± 1465 ng/dl) values for adrenal vein blood from IHA adrenals were both lower than for APAs (both P < 0.001), but F values (695 ± 330 µg/dl) were similar (P = 0.5). The 18OHB/A ratios for blood draining IHA adrenals (mean 4.69 ± 4.32; range 1.1414.37) clustered near values similar to those obtained for APA specimens, except for much higher values in subject No. 18 (Table 1
; P = 0.02 IHA vs. APA). The 18OHB/A ratios for the right and left adrenal vein specimens from individual IHA subjects were strikingly similar, within a factor of 1.7 except for subject No. 1. Using an 18OHB/A gradient less than 0.5 as criterion for lateralization, the 18OHB/A gradients agreed with the A/F gradients in 21 of 23 cases (Table 1
). These data demonstrate that APAs actually produce abundant 18OHB relative to A.
The 18OHB/F ratios were elevated in blood draining APAs (26.03 ± 11.51) compared with blood draining IHA adrenals (9.22 ± 5.18; P < 0.001) or the contralateral adrenals (6.23 ± 2.97; P < 0.001). Using an 18OHB/F gradient of more than two as criterion for lateralization, the 18OHB/F gradients also agreed with the A/F gradients in 21 of 23 cases (Table 1
).
| Discussion |
|---|
|
|
|---|
We propose that 18OHB from adrenals contralateral to APAs primarily derives from the zona fasciculata (Fig. 1
). Under cosyntropin stimulation, the "suppressed" adrenal still synthesizes a trace of 18OHB and A from the zona glomerulosa. Cosyntropin simultaneously increases intraadrenal corticosterone, which provides substrate for the minor 18-hydroxylase activity of CYP11B1 (18), yielding 18OHB. This model is consistent with previous indirect demonstrations that the zona fasciculata still produces 18OHB when A synthesis is suppressed (19). Consequently, the residual 18OHB production from the adrenal contralateral to an APA attenuates the 18OHB/F gradients mustered by the tumor and generates inverted 18OHB/A gradients, which are both useful for confirming APA lateralization.
|
We used strict criteria for PA diagnosis, adrenal vein F step-ups, and A/F lateralization gradients to prevent misclassification in this pilot study. AVS was performed with cosyntropin stimulation to mitigate fluctuations in F production and to maximize adrenal vein F step-ups. Our results might be different if less stringent criteria for PA or AVS lateralization were used. Furthermore, bilateral simultaneous AVS without cosyntropin infusion (17) would most accurately reflect relative 18OHB and A production from the APAs. Some specimens might have been altered by storage at 80 C for many months. Nevertheless, given the remarkable internal consistency of results and P values, these limitations should not detract from our main conclusions.
Our retrospective pilot study cannot demonstrate whether 18OHB measurements during AVS either improved A localization or patient outcomes. We suggest that 18OHB measurements might prove particularly useful for AVS studies with unsuccessful access of the right adrenal vein or suboptimal adrenal vein F step-ups, which ordinarily preclude conclusive interpretations.
| Acknowledgments |
|---|
| Footnotes |
|---|
Conflicts of Interest: D.W.C. declares stock ownership in LabCorp; all other authors declare no conflicts of interest.
First Published Online May 1, 2007
Abbreviations: A, Aldosterone; APA, aldosterone-producing adenoma; AVS, adrenal vein sampling; F, cortisol; HTN, hypertension; IHA, idiopathic hyperaldosteronism; 18OHB, 18-hydroxycorticosterone; PA, primary aldosteronism.
Received November 29, 2006.
Accepted April 24, 2007.
| References |
|---|
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |