Outcome of Adrenal Vein Sampling Performed During Concurrent Mineralocorticoid Receptor Antagonist Therapy

Context:

Pharmacological inhibition of mineralocorticoid receptor (MR) signaling in patients with primary aldosteronism (PA) reestablishes aldosterone synthesis by nondiseased zona glomerulosa cells through activation of the renin-angiotensin-aldosterone system. In this context, current guidelines recommend discontinuing MR blockade for diagnostic procedures, including adrenal vein sampling (AVS). Discontinuation of MR blockade in high-risk patients may be harmful because of uncontrolled hypertension and severe hypokalemia. We hypothesize that MR antagonist therapy can be continued during AVS as long as renin levels remain suppressed.

Objective:

The objective of this study was to assess the validity of AVS results in the context of MR antagonistic therapy.

Design:

We retrospectively analyzed all AVS studies in Munich (since 2008) and Düsseldorf (since 2011) and identified four of 237 (1.7%) patients with PA who underwent AVS while treated with an MR antagonist. Adrenalectomy was recommended based on the results of AVS in all four patients. After adrenalectomy, follow-up data were obtained to confirm improvement or remission of PA. Main outcome measures included blood pressure values, daily defined doses of antihypertensive medication, as well as levels of aldosterone, renin, and potassium, and the aldosterone/renin ratio.

Results:

In all patients, renin remained low or suppressed during AVS despite MR antagonist treatment. AVS clearly demonstrated unilateral aldosterone excess in each case. After adrenalectomy, all patients showed remission of PA as demonstrated by blood pressure values, potassium levels, and the aldosterone/renin ratio.

Conclusion:

In selected cases of PA, MR antagonist therapy might be continued during AVS, provided that renin values remain low.

Primary aldosteronism (PA) is commonly caused by an aldosterone-producing adenoma (APA) or bilateral idiopathic adrenal hyperplasia, resulting frequently in difficult-to-treat arterial hypertension. In addition to high blood pressure values, sodium retention and low plasma potassium values are characteristic signs of the underlying mineralocorticoid excess (1). Because mineralocorticoid excess exerts a negative feedback on the renin-angiotensin-aldosterone system (RAAS), renin concentrations are usually suppressed. Adrenal vein sampling (AVS) is used as the “gold standard” to identify the subtype of PA and to differentiate unilateral aldosterone production from idiopathic adrenal hyperplasia (2–4). The rationale of this approach is based on the assumption that the dominant side secretes excess aldosterone, whereas the nondominant side characteristically secretes cortisol-corrected aldosterone levels below that of peripheral blood samples.

Treatment with MR antagonists, including spironolactone and eplerenone, is an effective tool to control the symptoms and to reduce the complications of PA (5). Unfortunately, MR antagonist therapy interferes with the diagnostic procedures for screening, confirmation, and subtype classification of PA (6). In patients with APA, aldosterone synthesis from nonadenomatous aldosterone-producing cells of the zona glomerulosa is thought to be low as a consequence of the suppressed systemic renin levels. This inactivity of normal zona glomerulosa cells may explain the occurrence of temporary hyperkalemia and hypoaldosteronism that may be observed in a few cases after removal of an APA and provide a theoretical basis for the calculation of the contralateral suppression index used in the interpretation of AVS (2, 7–9). MR antagonists may reverse, at least in part, the inactivity of the nondominant or contralateral zona glomerulosa cells, sometimes associated with increased renin and aldosterone levels. As a consequence, lateralized aldosterone secretion in AVS may be obscured. In this light, it is widely accepted that MR antagonists should be withdrawn 4 to 6 weeks before AVS (2).

However, discontinuation of such a therapy may put the patient at risk of severe hypokalemia and complications of uncontrolled hypertension. From a theoretical point of view, therapy with MR antagonists may be trivial in terms of masking aldosterone secretion on AVS, provided that normal zona glomerulosa cells are still relatively inactive as indicated by low or suppressed renin values and persistent signs of PA. Whether this hypothesis is correct has not—to the best of our knowledge—been studied to date. To address this question, we retrospectively analyzed a case series of four patients with PA who underwent AVS under MR antagonist therapy.

Patients and Methods

Rationale and definition

The rationale of this study was to assess the reliability of AVS in patients under current MR antagonist treatment. As the outcome parameter, we used remission of PA after adrenalectomy defined as normalization of serum potassium, aldosterone/renin ratio (ARR), and improvement or normalization of blood pressure.

Patients

We screened the local databases at the Hospitals of the Heinrich-Heine University Duesseldorf and the Ludwig-Maximilians-Universität München. Both centers are participants of the German Conn's registry (www.conn-register.de), a prospective database to study long-term outcome of PA (10). Four patients with PA and continuing treatment with an MR antagonist during AVS were identified. This represents 1.7% of all AVS procedures in our centers, 1.2% of AVS procedures since 2008 in Munich (two of 162), and 2.7% of AVS procedures since 2011 in Duesseldorf (two of 75).

Ethics

Patients agreed to be included in the German Conn's Registry or the ENS@T (European Network for the Study of Adrenal Tumours)-Register as approved by the local ethics committees of the Heinrich-Heine-University and the Ludwig- Maximilians-Universität München. Written informed consent was obtained from each patient.

Recorded data and diagnostic procedures

Baseline data included age, gender, blood pressure (manual or automated office blood pressure), daily defined doses of antihypertensive medication (according to the WHO ATC/DDD Index), serum potassium levels, plus serum aldosterone and plasma renin concentrations. Moreover, information on confirmatory testing is given, and results of AVS and imaging are shown. Screening and confirmation testing were generally performed according to The Endocrine Society practice guidelines (6). Variations and specifications are reported in each patient's description. Imaging was done using chemical-shift magnetic resonance imaging (MRI) or multidetector computed tomography (CT). Homogeneous nodules with a mean density <10 Hounsfield units in noncontrast CT or a signal loss in the “opposed phase” T1w images, when compared to the “in-phase” images in chemical-shift MRI, indicated a suspected adrenal adenoma.

AVS was done sequentially and without cosyntropin stimulation. The selectivity index (SI) was defined by the ratio of the adrenal venous cortisol concentration to the peripheral cortisol concentration on each side. A SI greater than (or equal to) 2 indicated successful catheterization of adrenal veins. The lateralization index (LI) was defined by the aldosterone/cortisol ratio on the dominant side divided by the aldosterone/cortisol ratio on the contralateral side. A LI greater than (or equal to) 4 indicated lateralization of aldosterone secretion. The contralateral suppression index was defined as the adrenal venous aldosterone/cortisol ratio of the nonaffected side divided by the peripheral venous aldosterone/cortisol ratio.

Treatment and follow-up

Patients were admitted to surgery for adrenalectomy due to suspected unilateral disease.

Patients were examined during follow-up at least 6 months after adrenalectomy. Follow-up data included blood pressure, daily defined doses of antihypertensive medication (according to the WHO ATC/DDD Index), serum potassium levels, serum aldosterone levels, and plasma renin concentration (PRC). Normalization or improvement of these parameters after adrenalectomy confirmed the diagnosis of unilateral PA. Histological examination of tissue samples was done in each case and confirmed adrenocortical adenoma.

Results

Case 1

A 68-year-old male presented in 2012 with severe arterial hypertension, complicated by symptomatic atrial fibrillation and hypokalemia. The diagnosis of PA was established, and the patient was treated with different antihypertensive agents, including a MR antagonist, leading to improvement of hypokalemia but persistence of hypertension (Tables 1 and 2). The patient was acutely hospitalized because of an intracerebral hemorrhage and in the context of the severe hypertension. CT revealed an adrenal adenoma with a diameter of 16 mm in the right adrenal gland (Figure 1A). Because of his clinically unstable condition, the patient was subjected to AVS without confirmatory testing or adjustment of medication. PRC remained suppressed before AVS (Table 1). At this time, the patient was treated with spironolactone 50 mg, ramipril 10 mg, bisoprolol 10 mg, phenoxybenzamine 80 mg, and molsidomine 24 mg per day. The results of AVS indicated that secretion of aldosterone was lateralized to the right adrenal gland (Table 3). Consequently, the patient was adrenalectomized on the right side. The histological examination revealed two adenomas. At follow-up 6 months after surgery, the patient had an office blood pressure of 125/77 mm Hg, on reduced antihypertensive medication including bisoprolol 7.5 mg, ramipril 10 mg, hydrochlorothiazide 12.5 mg, and amlodipine 5 mg per day without MR antagonist therapy. Serum potassium levels were in the normal range, the ARR decreased from 543 to 15, and aldosterone suppressed normally (below the limit of detection) on sodium loading.

Figure 1.

MRI or CT indicated adrenal morphology (arrows).

A, CT in patient 1 demonstrated an adrenal adenoma with a diameter of 16 mm in the right adrenal gland. B, In patient 2, a tumor of 8 mm suggestive for an adenoma was identified in the right adrenal gland using MRI. C, In patient 3, CT demonstrated remnant adrenal tumor tissue on the left side after the first operation. D, In contrast to the results from AVS, an enlargement of the left adrenal gland was observed in patient 4.

Table 1.

Patient Characteristics Before and After Surgery

Patient No.	Systolic/Diastolic BP, mm Hg		A (11–160 ng/L)		Renin (1.5–18 ng/L)		ARR (<20)		Serum Potassium, mmol/L
	Before Surgery	After Surgery	Before Surgery	After Surgery	Before Surgery	After Surgery	Before Surgery	After Surgery	Before Surgery	After Surgery
1	188/96	125/77	543	44	<1	3	543	15	3.2	4.1
2	161/92	130/70	354	40	4	9	89	4	3.1	4.3
3	180/110	141/94	1043	26	<1	10	1043	3	3.2	4.1
4	175/105	130/80	144	13	2	11	72	1	3.3	4.8

Abbreviations: A, aldosterone; BP, blood pressure. ARR was rounded to integers. Data in parentheses indicate normal values.

Table 2.

Daily Defined Doses of Antihypertensive Medication During AVS (A) and Follow-Up (B)

Medication	Patient 1	Patient 2	Patient 3	Patient 4
MR antagonist
A	0.66	1.66	0.66	0.5
B	—	—	—	—
Other potassium-sparing diuretics
A	—	—	0.5	—
B	—	—	—	—
Thiazide diuretics
A	—	—	—	—
B	0.5	—	—	—
High-ceiling diuretics
A	—	—	—	—
B	—	—	0.66	—
Angiotensin II antagonists
A	—	—	1	—
B	—	—	0.5	—
ACE inhibitors
A	4	—	2	—
B	4	—	—	—
β-Blocking agents
A	1	—	0.33	—
B	0.75	—	0.33	—
Calcium channel blockers
A	—	2	—	—
B	1	—	—	—
α-Adrenoreceptor antagonists
A	2.66	1	1.5	—
B	—	—	—	—
Imidazoline receptor agonists
A	—	—	0.66	—
B	—	—	—	—
Sum of DDDs
A	8.32	4.66	6.65	0.5
B	6.25	0	1.49	0

Abbreviations: ACE, angiotensin-converting enzyme; DDD, daily defined dose; —, absence of medication. Patient 1 was also preoperatively treated with molsidomine 24 mg per day that was not assigned with a DDD. In addition, patient 2 was treated with a nitroglycerin-containing standby medication that was stopped after adrenalectomy.

Table 3.

Results of AVS

Patient No.	A-AV, ng/L		A-IVC, ng/L		F-AV, μg/dL		F-IVC, μg/dL		SI		LI	SuI
	Right	Left	Right	Left	Right	Left	Right	Left	Right	Left
1	2798	1899	78	184	33	153	16	18	2	9	7	1.2
2	11 463	796	142	153	677	246	13	13	52	19	6	0.3
3	857	12 000	917	949	147	227	17	16	9	14	9	0.1
4	128 351	345	409	236	266	30	16	15	17	2	6	0.7

Abbreviations: A, aldosterone; F, cortisol; AV, adrenal vein; IVC, inferior vena cava (or external iliac vein); LI, lateralization index as defined by A/F (AV-dominant)/A/F (AV-nondominant); SI, selectivity index as defined by F (AV-side)/F (IVC-side); SuI, contralateral suppression index as defined by A/F (AV-nondominant)/A/F (IVC-nondominant).

Case 2

A 50-year-old male had severe hypokalemic hypertension and recurrent hypertensive crises. Pre-existing conditions included coronary artery disease, several strokes, and type 2 diabetes mellitus. The ARR and saline infusion test indicated primary PA. MRI revealed a small nodule of 8 mm within the right adrenal gland, suggestive of an adrenal adenoma (Figure 1B). AVS was done twice in an external hospital without success, and MR antagonist therapy was initiated. Subsequently, the patient was admitted to our hospital for subtype differentiation with severe symptomatic hypertension caused by intractable PA. The ARR was 89 on daily antihypertensive therapy including spironolactone 50 mg, eplerenone 50 mg, amlodipine 10 mg, doxazosin 4 mg, and a nitroglycerin-containing standby medication (Tables 1 and 2). Discontinuation of therapy was not done because of recurrent hypertensive crisis and multiple comorbidities in the patient's history. AVS was performed while MR antagonist therapy was continued (Table 2). The results showed lateralization of aldosterone secretion to the right side, leading to unilateral adrenalectomy (Table 3). At follow-up 6 months after adrenalectomy, the patient was in remission from PA (post sodium load test aldosterone below the limit of detection) and with normal blood pressure without antihypertensives (Tables 1 and 2).

Case 3

A 54-year-old female had hypertension since the age of 16 years, which progressed into drug-resistant hypertension associated with hypokalemia in recent years. The diagnosis of PA was established externally. Subsequently, the patient presented with alkalosis, rhabdomyolysis, hypokalemia, and a hypertensive crisis. MR antagonist therapy with spironolactone and with the potassium sparing diuretic amiloride was initiated. Subsequently, the patient was adrenalectomized on the left side for an adrenal tumor of 4 cm. After surgery, the blood pressure improved only for a few days, and continuation of extensive antihypertensive therapy was necessary. The patient was referred to our outpatient unit. The ARR was 1043, PRC was fully suppressed, and serum potassium was low at 3.2 mmol/L. CT suggested a residual tumor of 1.8 cm on the left side (Figure 1C). Because the patient was already adrenalectomized on the left side without success, we decided to substantiate the results. ¹²³Iodometomidate-scintigraphy indicated residual steroidogenic tissue on the left side. MR antagonistic therapy with spironolactone was reduced before AVS from 200 to 50 mg per day. Lateralization of aldosterone secretion to the left side was demonstrated, and reoperation of the left-sided adrenal adenoma was performed (Table 3). Six months after surgery, the patient was in remission from PA with improved blood pressure (Tables 1 and 2).

Case 4

A 72-year-old female with PA had prior unsuccessful AVS and suffered from severe hypertension, hypokalemia, and atrial fibrillation. MR antagonist therapy was then started with eplerenone 50 mg per day. CT showed an enlargement of the left adrenal gland (Figure 1D). Because of severe hypertension and in the light of suppressed renin levels, AVS was performed while treatment with eplerenone 25 mg per day was continued. AVS indicated lateralization of aldosterone secretion to the right adrenal gland, which was followed by right-sided adrenalectomy (Table 3). After surgery, the patient was in remission from PA, and blood pressure was 130/80 mm Hg without antihypertensive drugs (Tables 1 and 2).

Discussion

Treatment with MR antagonists and other potassium-sparing diuretics may cause significant changes in the activity of the RAAS (11). The use of these agents may result in a false-negative ARR or false-positive confirmatory testing (6). Two recent consensus statements on AVS recommend against MR antagonist therapy during AVS because of the risk of renin levels high enough to mask lateralization of aldosterone secretion (2, 3) but mention that some expert centers “use plasma renin measurement to decide whether to perform AVS without withdrawing … the MR antagonists” and consider “the finding of low renin (eg, a direct renin <25 mU/L or a plasma renin activity <0.60 ng/ml/h) as evidence for unlikely stimulation of the contralateral adrenal cortex at a level sufficient to confuse interpretation of lateralization” (2). Assuming that significant aldosterone synthesis of nonadenomatous adrenal aldosterone-producing cells requires measurable renin levels, we aimed at studying this hypothesis in a series of four highly selected patients considered unfit for withdrawal of MR antagonist treatment. The persistence of hypokalemia and hypertension despite MR antagonist treatment in our patients confirm the severity of aldosterone excess, together with severe comorbidities it was also the rationale for continuation of MR antagonist treatment. Our data demonstrate that reliable lateralization can be obtained in patients with suppressed renin levels on MR antagonist treatment. The predictive value of AVS for remission after adrenalectomy in unilateral aldosterone excess appears to be sustained in such a setting. The postoperative course of our patients justifies such an assumption: all four had a favorable clinical response to surgery characterized by normalized potassium levels and normal or improved blood pressure. Two of the four patients were free of blood pressure medications, and the other two had a significant reduction in daily defined dose. In all patients, the ARR was normalized after surgery.

To date, imaging cannot be recommended in general as an alternative for AVS because the diagnostic accuracy of MRI and CT is rather low, especially in the elderly (12, 13). Therefore, we did not employ imaging as a proof of unilateral disease in our patients. In line with this, CT in patient 4 showed an enlargement of the left adrenal gland, whereas this patient was operated on the right side based on the results of AVS. However, in younger patients (below age 35 y), the appearance of hormonally inactive adrenal adenomas is rare, and AVS may be omitted if only one adrenal adenoma is visible and the contralateral gland looks normal even at second view.

Because a number of surgeons prefer partial adrenalectomy to total adrenalectomy, it is necessary to bring out that patient 3 was not cured after the first operation, indicating that partial adrenalectomy might be insufficient to cure patients with unilateral PA. This is in line with observations from other authors (14). In contrast, a recent retrospective study demonstrated a similar clinical outcome in patients with unilateral PA who were either treated with partial or total adrenalectomy (15). Unfortunately, and to the best of our knowledge, there are no randomized and prospective studies comparing partial and total adrenalectomy. As long as criteria for the selection of patients with unilateral PA that may profit from partial adrenalectomy are not clearly defined, we rather tend to recommend total adrenalectomy. In summary, these data suggest that AVS can be performed in selected patients who are unable to withdraw from MR antagonist therapy. However, several caveats have to be taken into account. First, the four cases presented here had long-standing and severe PA, as indicated by the level of aldosterone excess, very low potassium concentration, and the number of blood pressure drugs necessary to control hypertension. Therefore, our assumption may not be correct for less severe cases of PA. Second, the case series presented here is small, and although we do not have evidence against our hypothesis, the statistical basis for our conclusions remains weak. Third, our patients had renin levels during AVS equal to or below 4 ng/L. Whether data obtained with these very low renin levels can be extrapolated to higher renin concentrations (ie, 15 ng/L or 25 mU/L, as discussed by Rossi et al [Ref. 2]) remains to be shown. In light of these limitations, we recommend offering AVS only to selected patients on MR antagonists and only to those patients at high risk for complications when MR treatment is withdrawn. Moreover, clear signs of persistent aldosterone excess should be present, as shown by fully suppressed renin levels and low potassium. In all other patients, the suggestions of the consensus guidelines should be followed. Finally, we encourage other centers to publish their experience to increase the evidence for successful AVS under continued MR antagonist therapy.

Acknowledgments

The study was only feasible due to the support of our PA team and the Endocrine Laboratory team.

The German Conn's Registry–Else Kröner-Fresenius Hyperaldosteronismus Registry is supported by a grant from the Else Kröner-Fresenius-Stiftung (to M.R.). This study was also supported by the Deutsche Forschungsgemeinschaft (WI 3660/1-1, KFO252) (to H.S.W.).

Disclosure Summary: The authors have nothing to disclose.

Abbreviations

 
• APA

  aldosterone-producing adenoma

 
• ARR

  aldosterone/renin ratio

 
• AVS

  adrenal vein sampling

 
• CT

  computed tomography

 
• LI

  lateralization index

 
• MR

  mineralocorticoid receptor

 
• MRI

  magnetic resonance imaging

 
• PA

  primary aldosteronism

 
• PRC

  plasma renin concentration

 
• RAAS

  renin-angiotensin-aldosterone system

 
• SI

  selectivity index.