Adrenal Venous Sampling in Young Patients with Primary Aldosteronism. Extravagance or Irreplaceable?

Abstract

Context

Current clinical guidelines suggest that adrenal venous sampling (AVS) may not be mandatory in young patients with primary aldosteronism (PA) and a solitary adrenal adenoma on imaging.

Objective

The aim of this study was to further elucidate whether conventional imaging alone is sufficient to distinguish unilateral from bilateral PA among patients aged 40 years or younger.

Methods

This was a retrospective study where data from 45 patients with PA, aged between 26 and 40 years, who underwent successful AVS between 2005 and 2019, were analyzed. Results concerning laterality on imaging studies and AVS were recorded. Outcome in surgically treated patients was assessed according to the Primary Aldosteronism Surgical Outcomes criteria.

Results

In 4 of 25 patients with unilateral aldosterone production according to AVS, computed tomography inaccurately suggested bilateral disease. Following unilateral adrenalectomy, all 4 patients showed complete clinical success. Five of 20 patients with bilateral aldosterone production according to AVS had a solitary adrenal nodule (8-19 mm) on imaging. Two of these 5 patients were treated with unilateral adrenalectomy, neither having complete biochemical and/or clinical success postoperatively. Two of 16 patients younger than 35 years had discordant results, 1 with unilateral and 1 with bilateral aldosterone production, according to AVS.

Conclusion

Imaging studies inaccurately predicted laterality in a significant number of young patients with PA. In contrast to current clinical guidelines, our results support AVS for subtype evaluation in young adults with PA, including patients 35 years or younger.

primary aldosteronism, aldosterone, adrenal venous sampling, Conn’s disease, adrenal adenoma

Primary aldosteronism (PA) is caused by autonomous hypersecretion of aldosterone. Although various forms of PA exist, the vast majority is caused by bilateral idiopathic adrenal hyperplasia and unilateral aldosterone-producing adenoma (1, 2). Distinction between these 2 forms is of major importance since the management differs: Lifelong medical therapy with mineralocorticoid receptor antagonist for patients with bilateral disease and curative adrenalectomy in patients with unilateral disease.

Adrenal venous sampling (AVS) is considered to be the gold standard to distinguish unilateral from bilateral PA (3). With conventional imaging, including abdominal computed tomography (CT) and magnetic resonance imaging, false-positive and false-negative results are of concern since nonfunctioning adrenal adenomas are common in the general population, and small aldosterone-producing adenomas may not be detected. In a systematic review of 38 studies, including 950 patients with PA, adrenal imaging misdiagnosed the source of aldosterone excess in approximately 40% of the patients. Relying on imaging alone, inappropriate adrenalectomy would have been performed in 15% of patients with bilateral disease, and 19% of patients with unilateral disease would have been treated with mineralocorticoid receptor antagonist instead of curative surgical treatment (4).

Current clinical guidelines from the Endocrine Society suggest that AVS may not be necessary in patients 35 years or younger who have marked aldosterone excess (>831 pmol/L), spontaneous hypokalemia, and a solitary adrenal lesion with radiological features consistent with cortical adenoma (3). Moreover, it has even been questioned whether patients with PA who are 40 years or younger actually need AVS in the presence a visible unilateral adrenal lesion (5, 6). These suggestions are, however, based on a small number of observations that have only generated low-grade evidence (3), and further studies are needed. The aim of this study was to further elucidate whether conventional imaging alone is sufficient to distinguish unilateral from bilateral PA among patients aged ≤40 years.

Materials and Methods

This was a retrospective study based on data from a total of 418 patients with confirmed PA who underwent AVS between September 2005 and October 2019 at the Sahlgrenska University Hospital. For the current analysis, data from 45 of 47 patients with successful AVS (96% success rate), aged 40 years or younger when the AVS was performed, were included. Data on laboratory, imaging, and AVS results was collected, as well as information on background characteristics, management and histopathological findings. The outcome in surgically treated patients was assessed according to the Primary Aldosteronism Surgical Outcomes criteria as complete, partial, or absent clinical and biochemical success (7).

Diagnosis and Subtype Studies

All patients had been screened for PA by measuring aldosterone and renin in serum/plasma. Autonomous aldosterone secretion was confirmed by showing lack of aldosterone suppression following a 4-hour intravenous saline infusion test in a recumbent position in 43 of 45 patients. In the 2 remaining patients, PA was considered proven because of marked aldosterone excess in combination with low renin and severe hypokalemia. Autonomous cosecretion of cortisol was excluded in 28 of 45 (62%) patients by dexamethasone suppression test (n = 14), urinary-free cortisol (n = 11), or both (n = 3).

All 45 patients with confirmed PA underwent either CT (n = 43) or magnetic resonance tomography (n = 2) prior to AVS. Forty patients were investigated with both nonenhanced and contrast enhanced CT (slice thickness 0.5 mm [n = 1], 0.9 mm [n = 1], 2 mm [n = 2], 2.5 mm [n = 10], 3 mm [n = 19], 5 mm [n = 6], and 6.5 mm [n = 1]). Three patients were examined by nonenhanced CT only (slice thickness 1.25 and 3 mm respectively). The remaining 2 patients underwent contrast enhanced magnetic resonance (slice thickness 2.5 and 3.5 mm, respectively).

Subsequently, all patients underwent AVS as previously described (8). In short, sequential cannulation of both adrenal veins, under continuous cosyntropin stimulation, was performed through a percutaneous femoral vein approach. Selectivity index was used to assess the adequacy of the adrenal vein cannulations. Successful AVS was defined as selectivity index >5, in other words when the cortisol concentrations collected from an adrenal vein were at least 5 times greater than the concentration from a peripheral vein. For differentiation between uni- and bilateral PA, the corresponding cortisol and aldosterone concentrations were used to calculate the lateralization index according to the following formula: (aldosterone/cortisol dominant adrenal vein)/(aldosterone/cortisol nondominant adrenal vein). A lateralization index greater than 4.0 was considered to be compatible with unilateral PA and index lower than 3.0 with bilateral disease. A lateralization index between 3 and 4 was considered to indicate unilateral disease if the contralateral index (aldosterone/cortisol nondominant adrenal vein)/(aldosterone/cortisol peripheral vein) was below 1.0.

Following AVS, all patients were referred to a multidisciplinary adrenal conference so as to decide the therapeutic approach, based on clinical, imaging, and AVS findings.

Biochemical Analyses

During the study period, aldosterone was measured with 3 radioimmunoassays. Between 2005 and October 2008, serum samples were analyzed by Adaltis MAIA with a coefficient of variation (CV) of 11% to 14%. Between November 2008 and October 2014, Siemens Coat-A-Count was used (CV 6-10%). Since November 2014 to the end of the study, plasma samples were analyzed by DiaSorin Liaison (CV 8-13%). Between 2005 and 2008, serum cortisol was measured with Bayer Centaur (CV 5-6.5%), between 2008 and 2011 with Roche Modular E (CV 5-7%), between 2011 and 2015 with Roche Cobas, Cortisol 1 (CV 3-4%), and since 2015 with Roche Cobas, Cortisol-II (CV 2-3%).

Reevaluation of Imaging Studies

For the current report, 2 experienced radiologists (with 21 and 10 years’ experience respectively), independently re-evaluated the imaging studies performed prior to treatment for PA. Both were blinded to the original CT report, as well as the clinical, laboratory, and histopathological findings. The appearance of each adrenal gland was documented, denoting the presence or absence of adrenal lesions.

Ethical Considerations

Ethics approval for the study was obtained from the Regional Ethical Review Board in Gothenburg (reference number 862/16; approved February 9, 2017).

Statistical Analyses

Continuous variables are summarized as mean (± SD) or median (interquartile range), and categorical data as number (%). For comparison between groups, the unpaired t-test was used for normally distributed data and the Mann–Whitney U-test for non-normally distributed data. For categorical variables the chi-squared or the Fisher exact test was used. All tests were 2-tailed and P < .05 was considered statistically significant. The statistical analyses were performed using SPSS statistics, version 24 (SPSS Inc., Chicago, IL, USA).

Inter-reader agreement, expressed as k factor, was assessed quantitatively with the following range of k-values: values ≤0 indicating no agreement, values 0.01 to 0.20 none to slight, 0.21 to 0.40 fair, 0.41 to 0.60 moderate, 0.61 to 0.80 substantial, and 0.81 to 1.00 almost perfect agreement (9, 10).

Results

Baseline Characteristics

In total, 45 patients (34 women and 11 men) with PA, aged between 26 and 40 years, who underwent successful AVS between 2005 and 2019 at our institution, were included in the analysis (Table 1). Sixteen of 45 patients were 35 years or younger. The median duration of hypertension was 2 (interquartile range 1-5) years. At referral, 27 (60%) patients had potassium supplementation and 21 (47%) had uncontrolled hypertension despite treatment with multiple antihypertensive medications. The median morning plasma aldosterone level was 765 (interquartile range 431-1300) pmol/L and the median plasma aldosterone following saline infusion test was 552 (interquartile range 552-813) pmol/L.

Table 1.

Baseline characteristics of the study cohort

	All (n = 45)	Unilateral PAa (n = 25)	Bilateral PAa (n = 20)	P
Sex				.5
Women	34	20	14
Men	11	5	6
Age (years)	36 ± 4	36 ± 4	37 ± 4	.5
Duration of hypertension (years)	2 (1-5)	2 (1-5)	2 (1-5)	.8
Systolic BP (mmHg)	145 ± 15	143 ± 14	146 ± 15	.5
Diastolic BP (mmHg)	91 ± 11	91 ± 9	92 ± 12	.6
Antihypertensive drugs, no	1 (1-2)	1 (1-2)	1 (1-2)	.14
Antihypertensive drugs, DDD	2 (1-3)	2 (1-3.3)	2 (0.6-2.5)	.3
Potassium supplementation, n	27	21	6	<.0001
Potassium supplementation (g/day)	6.0 (4.5-11.2)	6.0 (4.5-10.1)	6.8 (2.6-21.6)	.6
PA (pmol/L)	765 (431-1300)	947 (530-1345)	534 (359-830)	.026
PA post-SIT (pmol/L)	552 (350-813)	727 (390-1093)	370 (280-673)	.023

Data are presented as mean ± standard deviation or median (interquartile range)

Abbreviations: BP, Blood pressure; DDD, daily defined dose; PA, primary aldosteronism; SIT, saline infusion test.

^aAccording to adrenal vein sampling.

Subtype Classification

According to AVS, 25 of 45 (55%) patients had unilateral PA (11 right, 14 left) with a mean lateralization index (LI) of 31 (range 3.2-127). Two of these 25 patients had LI between 3 and 4, both with a contralateral index <1. The remaining 20 patients had bilateral disease with LI <3. No patient had a procedure-related complications in association with the AVS. Patients with unilateral PA had higher morning and postsaline infusion plasma aldosterone, and more often oral potassium supplementation than patients with bilateral disease (Table 1).

Patients with Bilateral Disease According to Imaging Studies

Fourteen patients had normal adrenal glands and 2 had bilateral lesions on imaging. Of these 16 patients, 4 had unilateral disease according to AVS, 2 with normally appearing glands and 2 patients with bilateral lesions (Table 2). One of these 4 patients was younger than 35 years and 3 underwent unilateral adrenalectomy. Complete clinical success, according to the Primary Aldosteronism Surgical Outcomes criteria, was achieved in all. Postoperative biochemical evaluation was performed in 2 patients, showing complete biochemical success in both.

Table 2.

Summary of patients with discordant lateralization results on imaging, AVS patients with inconclusive lateralization on imaging, and patients with concordant results

No	Age/Gender	Potassium supplementation (g/day)	LI/Side	CI	Radiological finding/side/size (mm)	Treatment	Biochemical success	Clinical success	Histopathology
Patients with unilateral disease on imaging and bilateral disease according to AVS (false-positive unilateral aldosteronism)
0	36/F	2.3	1/L		Adenoma/L/8	MRA
23	40/M	18.8	2.9/R		Adenoma/R/10	ADX/R	Absent	Partial	Adenoma
7	33/F	0	2.1/R		Adenoma/R/19	MRA
13	37/F	6	2.5/L		Adenoma/L/14	MRA
28	40/M	30	3.9/L	1.4	Adenoma/L/12	ADX/L	Partial	Partial	Diffuse AH
Patients with bilateral disease on imaging and unilateral disease according to AVS (false-negative unilateral aldosteronism)
15	38/M	15	50/R	<1	Adenoma/Bil/22, 17	ADX/R	NA	Complete	Adenoma
18	39/F	3	3.2/R	<1	Normal	MRA
32	34/F	0	6/R	1	Normal	ADX/L	Complete	Complete	Multinodular AH
45	40/F	0	9/L	<1	Adenoma/Bil/10, 7	ADX/L	Complete	Complete	Adenoma
Patients with inconclusive lateralization on imaging
43	37/M	0	1.6/R		Enlarged/R	MRA
21	39/M	1.5	2.3/L		Enlarged/Bil	MRA
19	39/F	6	2.5/R		Small nodules/Bil/5, 7	MRA
14	38/F	6	17/L	<1	Small nodule/L/7	ADX/L	Complete	Partial	Adenoma
Patients with bilateral disease according to imaging and AVS
29	40/M	0	1.2/R		Normal	MRA
41	39/F	0	1.2/R		Normal	MRA
5	31/F	0	1.2/R		Normal	MRA
36	33/F	0	1.3/L		Normal	MRA
26	40/F	6	1.3/L		Normal	MRA
42	38/F	0	1.3/L		Normal	MRA
39	36/F	0	1.6/R		Normal	MRA
31	31/F	0	1.6/R		Normal	MRA
22	39/F	0	1.8/R		Normal	MRA
44	40/F	0	2.3/L		Normal	MRA
1	26/F	0	2.7/L		Normal	MRA
12	36/M	0	2.9/L		Normal	MRA
Patients with unilateral disease according to imaging and AVS
6	32/F	18	4	<1	Adenoma/L/17	ADX/L	Complete	Complete	Adenoma
20	39/F	1.5	10/L	<1	Adenoma/L/10	ADX/L	Complete	Complete	Adenoma
37	38/F	0	12/R	<1	Adenoma/ R/25	ADX/R	Complete	Complete	Adenoma
8	34/F	6	13/R	<1	Adenoma/R/15	ADX/R	Complete	Complete	Multinodular AH
25	40/F	9	14/R	<1	Adenoma/R/29	ADX/R	Complete	Partial	Adenoma
17	39/M	4.5	14/L	<1	Adenoma/R/22	ADX/R	Complete	Partial	Adenoma
9	35/F	3	15/R	<1	Adenoma/R/26	ADX/R	NA	Partial	Adenoma
16	38/F	18.8	15/L	<1	Adenoma/L/15	ADX/L	Complete	Partial	Adenoma
11	35/F	4.5	19/R	<1	Adenoma/R/10	ADX/R	NA	Complete	NA
10	35/M	7.5	21/L	<1	Adenoma/L/15	ADX/L	Complete	Partial	Adenoma
2	28/F	6	28/L	<1	Adenoma/L/14	ADX/L	Complete	Complete	Adenoma
3	29/F	6	28/L	<1		MRA
24	40/F	2,3	32/R	<1	Adenoma/R/15	ADX/R	Complete	Complete	Adenoma
34	36/F	0 despite hypokalemia	40/L	<1	Adenoma/L/11	ADX/L	Complete	Complete	Adenoma
4	30/F	9	40/L	<1	Adenoma/L/10	ADX/L	Complete	Complete	Adenoma
27	40/F	6	43/L	<1	Adenoma/L/22	ADX/L	NA	Partial	Adenoma
40	38/F	13.5	61/R	<1	Adenoma/R/25	ADX/R	Complete	Complete	Adenoma
38	36/M	15.8	72/L	<1	Adenoma/L /10	ADX/L	Complete	Complete	Adenoma
33	34/M	4.2	80/R	<1	Adenoma/R/11	ADX/R	Complete	Partial	Adenoma
30	30/F	3	127/L	<1	Adenoma/L/12	ADX/L	Complete	Complete	Adenoma

Abbreviations: ADX, adrenalectomy; AH, adrenal hyperplasia; AVS, adrenal vein sampling; bil, bilateral; CI, contralateral index; F, female; L, left; LI, lateralization index; M, male; MRA, Mineralocorticoid receptor antagonist; NA, not assessed; R, right.

Patients with Unilateral Disease According to Imaging Studies

Among 25 patients with unilateral adrenal lesion on imaging, 20 lateralized to the ipsilateral side according to the AVS. Imaging inaccurately suggested unilateral disease in 5 patients with unilateral adrenal adenoma (Table 2). One of these 5 patients was younger than 35 years (Fig. 1, subject 7).

Figure 1.

CT from a 33-year-old woman with primary aldosteronism showing (A) a 19-mm adenoma in the right adrenal nodule (arrow) and (B) a normal-appearing left adrenal gland. Based on AVS, showing bilateral aldosterone production, the patient received treatment with mineralocorticoid receptor antagonist.

Despite being considered to have bilateral disease according to the AVS, 2 of 5 patients were still treated with unilateral adrenalectomy. Postoperatively, 1 patient had partial biochemical and clinical success. The other patient had partial clinical success but absent biochemical success.

Patients with Inconclusive Lateralization According to Imaging Studies

Four patients had inconclusive lateralization on imaging: 1 with unilateral “enlarged” adrenal gland, 1 with bilaterally “enlarged” adrenal glands, 1 with a small (7 mm) adrenal nodule, and 1 with bilateral small nodules (5 and 7 mm, respectively) (Table 2). One of these 4 patients, the 1 with a 7 mm unilateral nodule, had ipsilateral aldosterone overproduction according to AVS and was treated with adrenalectomy with complete biochemical and partial clinical success.

Re-evaluation of Imaging Studies: Inter-reader Variability

Re-evaluation of the pretreatment imaging studies by 2 radiologists was in agreement with the original report in 37 patients; 23 patients with unilateral adenoma, 12 patients with normal adrenal glands, 1 patient with bilateral adenoma, and 1 patient with a small (7 mm) unilateral nodule. Of the 23 patients with unilateral adenoma according to all 3 reports, 4 had bilateral disease according to AVS (Table 2, subjects 0, 7, 13, and 28). Also, the patient considered to have bilateral adenomas by all 3 readers had unilateral aldosteronism according to AVS (subject 45).

Discordant results between AVS and imaging were reported in 8 patients (Table 3). On the basis of the imaging re-evaluation by 2 radiologists, 5, respectively 4, patients with bilateral PA according to AVS, would have been considered to have unilateral disease. Similarly, 1, respectively 5, patients with unilateral PA according to AVS, would have been considered to have bilateral disease.

Table 3.

Summary of discordant cases comparing lateralization according to AVS and different reading performance

No	Age/ Gender	Laterality according to AVS	Radiological finding/side/size (mm) original report	Radiological finding/side/size (mm) re-evaluation #1	Radiological finding/side/size (mm) re-evaluation #2
Patients with discordant AVS and imaging results from all 3 radiological reports
0	36/F	Bilateral	Adenoma/L/8	Adenoma/L/9	Adenoma/L/9
7	33/F	Bilateral	Adenoma/R/19	Adenoma/R/17	Adenoma/R/17
13	37/F	Bilateral	Adenoma/L/14	Adenoma/L/13	Adenoma/L/13
28	40/M	Bilateral	Adenoma/L/12	Adenoma/L/15	Adenoma/L/18
45	40/F	Unilateral/L	Adenoma/Bil/10, 7	Adenoma/Bil/13, 11	Adenoma/Bil/9, 8
Patients with discordant AVS and imaging results from at least 1 radiological report
34	36/F	Unilateral/L	Adenoma/L/11	Adenoma/L/11	Normal
32	34/F	Unilateral/L	Normal	Adenoma/L/9	Normal
18	39/F	Unilateral/R	Normal	Adenoma/R/9	Normal
15	38/M	Unilateral/R	Adenoma/Bil/ 22, 17	Adenoma/R/20	Adenoma/Bil/ 22, 17
23	40/M	Bilateral	Adenoma/R/10	Adenoma/R/9	Enlarged/Bil
43	37/M	Bilateral	Enlarged/R	Small nodule/R/7	Small nodule/R/7
21	39/M	Bilateral	Enlarged/Bil	Normal	Normal
19	39/F	Bilateral	Small nodules/Bil/5, 7	Small nodule/R/5	Enlarged/R

Abbreviations: AVS, adrenal vein sampling; bil, bilateral; F, female; L, left; M, male; R, right.

The level of agreement between the 2 radiologists was substantial (k = 0.71) when all lesions defined as adenomas, irrespective of size, were taken into account. Almost perfect agreement was found (k = 0.90) when only lesions ≥10 mm were included.

Discussion

In this study we have analyzed data from 45 patients who were 40 years or younger when they were diagnosed with PA and underwent AVS. In total, 9 (20%) patients had discordant findings on imaging compared with AVS: 5 patients with unilateral adenoma and AVS-verified bilateral aldosterone production, 2 patients with normal adrenal glands and AVS-verified unilateral aldosterone production, and 2 patients with bilateral adenoma and AVS-verified unilateral aldosterone production. Thus, the source of the aldosterone excess was incorrectly predicted by imaging in one-fifth of the patients, and, if the treatment decision was based on imaging alone, 5 unnecessary adrenalectomies would have been performed.

Adrenal tumors are uncommon among the young. Based on data from 5 large autopsy series, only 1% of patients aged between 30 and 39 years had adrenal adenoma compared with 6% of patients aged between 60 and 69 years (11). This grounded the suggestion that young patients with PA, and unilateral adrenal adenoma on imaging, do not have to proceed with AVS to decide laterality (3). However, the data supporting this position consists of small studies with various AVS protocols, and definitions of cure, resulting in a limited body of scientific evidence regarding their management (12-16). To our knowledge, only 5 studies have specifically investigated the accuracy of adrenal imaging in young patients with PA (Table 4) (12, 13). The first 2 reports showed that the accuracy of adrenal imaging in patients younger than 40 years with unilateral adenoma was 100% (12, 13). Three following studies demonstrated significant discordance in young patients (14-16). In the study by Lim, 6 of 21 patients younger than 40 years with unilateral PA had bilateral disease according to imaging (14). All 6 cases with discordant findings were older than 35 years. Subsequently, a cut-off age of 35 years was implemented in the current clinical guidelines. Similarly, in the study by Riester, 8 out of 28 young patients had discordant results, mostly patients without typical adenoma on imaging and unilateral PA on AVS (15). One patient, 36 years old, had unilateral adenoma on imaging and bilateral aldosterone production according to AVS (15).

Table 4.

Summary of studies including patients with PA, aged ≤40 years, comparing lateralization according to adrenal imaging and AVS

Author, year (ref)	Origin	Design	N total/ ≤40 year./≤35 years	Main findings	Comments
Mulatero, 2008 (12)	Italy	Prospective	70/5/n.s.	11 of 38 patients aged >40 with unilateral lesion on imaging had bilateral PA according to AVS 4 of 32 patients without a typical adenoma on imaging had unilateral PA according to AVS None of 5 patients 40 years or younger with unilateral disease on imaging had bilateral PA according to AVS	Adenoma defined as unilateral lesion >10 mm
Küpers, 2012 (13)	France	Retrospective	87/25/n.s.	None of 9 patients 40 years or younger with unilateral disease on imaging had bilateral subtype according to AVS 6 of 16 young patients without a typical adenoma on imaging had unilateral PA according to AVS Appearance of a typical adenoma in combination with hypokalemia or glomerular filtration rate ≥100 mL/min/1.73 m2 predicts unilateral PA	Adenoma defined as unilateral lesion>10 mm
Lim, 2014 (14)	USA	Retrospective	143/21/5	6 of 21 adrenalectomized patients 40 years or younger had inaccurate imaging In all 5 patients 35 years or younger, imaging and AVS results were concordant	Only adrenalectomized patients were included Patients with unilateral lesions that did not lateralize on AVS may be excluded
Riester, 2014 (15)	Germany	Retrospective	194/28/n.s.	One of 16 patients 40 years or younger with unilateral disease on imaging had bilateral PA according to AVS 7 of 12 young patients with bilateral disease on imaging had unilateral PA according to AVS The clinical prediction score proposed by Küpers failed to predict unilateral PA	Adenoma defined as unilateral lesion >10 mm AVS performed without ACTH stimulation
Umakoshi, 2017 (16)	Japan	Retrospective	358/69/30	8 of 39 patients aged between 35 and 40 with unilateral disease on imaging had bilateral PA according to AVS 3 of 30 patients aged between 30 and 35 years with unilateral disease on imaging had bilateral disease according to AVSa	Included only subjects with unilateral adrenal lesion >10 mm and severe PA Patients with bilateral disease on imaging were not included Some patients with unilateral lesions that did not lateralize on AVS may not have been included

Abbreviations: AVS, adrenal vein sampling; n.s., not specified; PA, primary aldosteronism.

^aNo subjects lateralized on the contralateral gland when a unilateral lesion was detected by imaging in all studies.

To our knowledge, the study by Umakoshi et al. presents the largest cohort of young adults with PA (16). All patients had severe PA and unilateral adenoma on imaging, and all underwent AVS and then adrenalectomy. Patients with bilateral disease on imaging were excluded from the analysis. Nevertheless, concordance between CT and AVS was found in 79% (31/39) of patients aged between 35 and 40 years and 90% (27/30) of patients aged <35 years. All 3 patients aged <35 years, with discordant findings according to the authors, had LI between 3 and 4, and a contralateral index (CI) <1, and benefited from surgery. The authors, therefore, supported the current recommendations to omit AVS in ages <35.

It has been proposed that certain clinical variables, such as severe hypokalemia and high aldosterone concentrations, may be useful to predict the source of the aldosterone excess. Riester et al. (15) analyzed the utility of the prediction score proposed by Küpers (13) and found it to be applicable in young patients. The vast majority of our patients with unilateral PA had hypokalemia and most of the patients with bilateral disease presented with normal potassium concentrations. Nevertheless, 2 patients with unilateral PA according to AVS, and bilateral disease on imaging, presented with normokalaemia. Both were underwent surgery, resulting in complete clinical and biochemical success. Also, 3 patients with profound hypokalemia had bilateral PA according to AVS, and unilateral adenoma on imaging. Two of these underwent adrenalectomy, resulting in either partial or absent success.

During the study period, 4 patients proceeded directly to unilateral adrenalectomy, without an AVS, based on imaging showing unilateral adenoma, and a severe phenotype with profound hypokalemia. Thus, the majority of patients in our region with PA that are considered to be candidates for surgery undergo AVS. Additionally, 2 patients with unsuccessful AVS proceeded to adrenalectomy without further attempts to investigate laterality. The postoperative outcome is known in 4 of these 6 cases, showing complete clinical success in all and complete biochemical success in 3.

This is the first study that evaluated the inter-reader agreement of imaging studies in patients with PA. The interpretation differed between the 3 radiological reports (the original and the 2 re-evaluations) in 8 patients. In most cases, the difference concerned whether a small adenoma was present or not. It is well known that detection of small adenomas (<1 cm) can be difficult. Thus, as anticipated, inter-reader agreement increased from substantial to almost perfect when lesions ≤10 mm were classified as an inconclusive finding. Nevertheless, in 1 patient an 11-mm adenoma was not recognized by 1 reviewer and another patient was considered to have bilateral adenomas by 2 reviewers and 1 adenoma and an extra adrenal lesion by the third reviewer. In both of these cases the therapeutic strategy, if only based on imaging and not on AVS, would have been completely different.

Deceptive AVS results, despite successful cannulation, cannot be ruled out in some of our patients as a possible reason for the discordance between imaging and AVS. Indeed, some authors consider surgical outcome, and not AVS, to be the ultimate measure to confirm laterality in patients with PA (17). In our study, adrenalectomy was performed in 2 cases based on AVS finding and not on imaging, which had suggested bilateral disease. In both cases, complete clinical and biochemical success was achieved. Similarly, Citton et al. reported treatment failure in 2 young patients with unilateral adenoma on imaging and persistent PA following unilateral adrenalectomy and declared that additional lateralizing methods are needed even in young patients with PA (18).

A limitation of this study is the retrospective design, illustrated by the lack of information on some variables, such as the clinical and/or biochemical outcome in a subset of patients. Also, in cases with unilateral adenoma on imaging and bilateral aldosterone production according to AVS who received medical treatment, the inferiority of imaging cannot be proved since information on surgical outcome, for obvious reasons, are lacking. Another limitation is that not all patients were screened for concomitant autonomous cortisol overproduction. Since cosecretion of cortisol in patients with PA may influence the interpretation of AVS (19), incorrect laterality cannot be excluded in all patients. On the other hand, a major strength is the fairly large and well-characterized cohort of young patients with PA, as well as the standardized AVS protocol used throughout the study, performed by a single radiologist.

In conclusion, this retrospective study demonstrates that imaging studies inaccurately predict laterality in a significant number of young patients with PA. Our results, therefore, question current clinical guidelines suggesting that AVS may not be mandatory in young adults with PA, including patients 35 years or younger. Given the low rate of complications with AVS, and the small number of patients that can be spared from AVS, our findings instead encourage liberal use of AVS in young patients with PA, especially subjects with bilateral adrenal lesions, apparently normal adrenal glands, or unequivocal findings on imaging.

Abbreviations

Abbreviations
 
• AVS

  adrenal venous sampling

 
• CV

  coefficient of variation

 
• PA

  primary aldosteronism

Acknowledgments

Financial Support: This work has not received any specific grants.

Additional Information

Disclosures: The authors have nothing to declare.

Data Availability

The data that support the findings of this study are available from the corresponding author upon request.