Adrenal Cortex Remodeling and Functional Zona Glomerulosa Hyperplasia in Primary Aldosteronism
Primary aldosteronism is the most common form of secondary hypertension with hypokalemia and suppressed renin-angiotensin system caused by autonomous aldosterone production. Our aim was to compare zona glomerulosa (ZG) structure and function between control adrenals and the peritumoral tissue from patients operated on for aldosterone-producing adenoma. ZG morphology and CYP11B1, CYP11B2, and disabled 2 expression were studied in 15 control adrenals and 25 adrenals with aldosterone-producing adenoma. A transcriptome analysis was done using publicly available data sets. In control adrenals, ZG was discontinuous, and CYP11B2 expression was focal or partly continuous and localized to 3 structures, foci, megafoci, and aldosterone-producing cell clusters. CYP11B2 expression was restricted to a limited number of ZG cells expressing Dab2 but not CYP11B1; aldosterone-producing cell clusters were composed of cells with an intermediate phenotype expressing CYP11B2 but not disabled 2 or CYP11B1. In peritumoral tissue, large remodeling of the adrenal cortex was observed with increased nodulation and decreased vascularization that were not correlated with CYP11B2 expression. In 17 out of 25 adrenals, hyperplasia of adjacent ZG was observed with persistent expression of CYP11B2 that was extended to the entire ZG. In all of the adrenals from patients with aldosterone-producing adenoma, CYP11B2 expression was present in foci, megafoci, and aldosterone-producing cell clusters. Transcriptome profiling indicates a close relationship between peritumoral and control adrenal cortex. In conclusion, adrenal cortex remodeling, reduced vascularization, and ZG hyperplasia are major features of adrenals with aldosterone-producing adenoma. Transcriptional phenotyping is not in favor of this being an intermediate step toward the formation of aldosterone-producing adenoma.
- zona glomerulosa hyperplasia
- aldosterone-producing adenoma
- secondary hypertension
- adrenal cortex remodeling
Primary aldosteronism (PAL) is the most common form of endocrine hypertension with a prevalence of 7% to 10% of hypertensive patients.1 In PAL, autonomous production of aldosterone is associated with suppressed renin activity, hypokalemia, and hypertension. The two principal forms are unilateral aldosterone-producing adenoma (APA), a benign tumor of the adrenal cortex, and bilateral adrenal hyperplasia, also known as idiopathic hyperaldosteronism. Whereas in APA the adjacent zona glomerulosa (ZG) is believed to be nonfunctional because of negative feedback regulation, bilateral adrenal hyperplasia involves the entire ZG of both adrenal glands.2 Patients with PAL operated on the basis of an APA, and/or a lateralized aldosterone hypersecretion can be cured by unilateral adrenalectomy. However, only ≈70% of patients with lateralized aldosterone secretion are cured by surgery,3 suggesting that, in some cases, residual aldosterone secretion from the contralateral adrenal gland, although within the normal range, may be inappropriate for salt status and blood pressure levels.
Aldosterone is synthesized from cholesterol in the ZG of the adrenal cortex by a series of specific enzymatic reactions.4 Histology of the mammalian adrenal cortex was originally described in 1866.5 The ZG is the outer layer of the cortex adjacent to the capsule composed of small and clear cells. In rodents, 5 to 10 cell layers (≈50 μm) form a continuous zone6,7; in humans, in contrast, the ZG is discontinuous.8,9 The zona fasciculata (ZF) is the thickest zone of the cortex and is composed of columns of large and clear cells (because of the presence of many intracellular lipid droplets). Finally, the zona reticularis (ZR) is composed of small and eosinophilic cells. The 3 zones present distinct roles in steroid hormone production: ZG produces mineralocorticoids, ZF produces glucocorticoids, and ZR produces androgens. In human and rodents, the functional zonation involves the zone-specific expression of 2 cytochrome p450 isoenzymes: 11β-hydroxylase (encoded by the CYP11B1 gene) is responsible for the final steps of cortisol biosynthesis in human (corticosterone in rodents), whereas aldosterone synthase (CYP11B2) catalyzes the final steps of aldosterone biosynthesis. The expression of CYP11B2 is limited to the ZG, whereas CYP11B1 is expressed in ZF and ZR,10,11 although until very recently the exact pattern of expression of the 2 isozymes in humans was still a matter of debate. Recently, Nishimoto et al12 showed that CYP11B2 was sporadically detected in ZG, whereas CYP11B1 was expressed in ZF and ZR. Interestingly, in addition to conventional zonation, the authors also described a variegated zonation consisting of subcapsular cell clusters expressing CYP11B2 that they named aldosterone-producing cell clusters (APCCs). In samples from patients with APA, the nontumor portions of the adrenals frequently bear 1 or multiple of these APCCs strongly expressing CYP11B2. Based on these findings, the authors proposed 2 patterns of aldosterone production, 1 regulated by the renin-angiotensin system and 1 autonomous in APCCs. They also propose that APCCs eventually develop into APA.
To further investigate the cellular phenotype of aldosterone-producing cells in normal adrenal glands and PAL and to explore the nature of APCCs, we have analyzed the expression of CYP11B2 in relation to specific markers of ZG (disabled 213) and ZF/reticularis (CYP11B1). Furthermore, we investigated the morphology, nodulation, and vascularization of the perinodular adrenal cortex and correlated these findings with CYP11B2 expression. These results were complemented by transcriptome analyses of peritumoral tissue inferred from publicly available data sets.
Subjects and Methods
We obtained formalin-fixed and paraffin-embedded adrenals through the COMETE (Cortico- et médullo-surrénale, les tumeurs endocrines) Network from 25 patients who had undergone surgery for lateralized PAL at Hôpital Européen Georges Pompidou between 2002 and 2008. Methods for screening and criteria for diagnosing PAL were in accordance with institutional guidelines and have been described recently.3 The clinical and biological characteristics of the patients are described in Tables 1 and S1 (available in the online Data Supplement at http://hyper.ahajournals.org). Fifteen control adrenals were obtained from enlarged nephrectomies, consisting of removal of one kidney together with the neighboring adrenal gland and neighboring lymph nodes (kindly provided by the Department of Pathology, University Hospital of Rouen, and by Dr M. Sibony, Hôpital Tenon). Two cortisol-producing adenomas were obtained through the COMETE Network. All of the patients gave written informed consent to participate to the study.
Histological annotations were performed on control adrenals, APA, and peritumoral adjacent tissue. Cellular compositions of APA and the ZG hyperplasia were determined on hematoxilin-eosin-safran (HES) staining. The main criterion used to determine ZG hyperplasia was the continuous character of the ZG; the second criterion, in case of discontinuity of the ZG, was its thickness (≥200 μm) measured with a micrometer integrated in the microscope.8,9 Nodulation was analyzed by Sirius red staining and converted into a semiquantitative score using the formula: nodulation score=∑i=13ni×i, where “i” is the score attributed to each type of nodulation (1=partial nodulation, 2=near complete nodulation, 3=complete nodulation; Figure 4A), and “ni” is the number of nodules of the corresponding type. All of the microscopic examinations were done on a Leica microscope.
In Situ Hybridization
In situ hybridization was performed on the entire adrenal gland (APA or nodules and peritumoral adjacent tissue) from PAL patients and from control adrenals. For detailed procedures, please see the online Data Supplement at http://hyper.ahajournals.org.
Determination of CYP11B2 Expression Area
CYP11B2 expression area was defined as the percentage of the adrenal gland expressing the enzyme. Analysis was performed at ×25 (ocular lens: ×10.0; objective lens: ×2.5) on all of the adrenal glands. The CYP11B2 expression area was determined on the entire adrenal gland (APA or nodules and peritumoral adjacent tissue) from PAL patients and from control adrenals using ImageJ software. For peritumoral adjacent tissues and control adrenals, CYP11B2 expression area was calculated as the percentage of measured adrenal cortex surface. For APA, CYP11B2 expression area was calculated as the percentage of APA surface; this measure was then reported to APA volume calculated using measured nodule dimensions (CYP11B2 expression area=% CYP11B2 expression×APA volume).
Immunohistochemistry was performed on serial sections of the same tissues used for in situ hybridization. For the detailed procedure please see the online Data Supplement.
Analysis of Vascularization
Vascularization was analyzed by CD34 immunostaining in control adrenal and adjacent tissues of APA. Microvessel surface analysis was performed at ×100 (ocular lens: ×10; objective lens: ×10), 4 different areas were measured, and the average was then calculated. The vessel surface was determined using ImageJ software.
Transcriptome Data Sets Used and Nomenclature
Transcriptome data set form Giordano et al16 (GEO accession No. GSE10927) consists of 4 groups defined according to sample tissue type and clinical annotation (Table S2): (1) G_ACA, adrenal cortex adenomas other than aldosterone-producing adenoma; (2) G_APA, adrenal cortex aldosterone-producing adenoma; (3) G_CA, normal adrenal cortex adjacent to adrenal adenomas other than aldosterone-producing adenoma; and (4) G_ADJ, normal adrenal cortex adjacent to aldosterone-producing adenoma.
Control adrenal versus APA subtraction profiles was used to define a gene signature of statistically significantly differentially expressed common probes for both data sets with P<0.01. The signature identified through this procedure was used to compute the correlation coefficients between samples. We calculated average Pearson correlation coefficients sample by sample for the control adrenal versus APA differentially expressed genes (P<0.01) with the average expression values of the same genes in the control adrenal and APA subpopulations.
For details please see the online Data Supplement.
Functional Zonation of the Human ZG Differs From Rodents
To characterize more precisely the cellular composition of areas expressing CYP11B2, we investigated expression of CYP11B2, CYP11B1, and Dab2 (disabled 2, a specific marker of the ZG in rodent13) in 15 control adrenals. We considered that this large number could give a reliable estimate of CYP11B2 expression if considering variables possibly affecting aldosterone biosynthesis, including sex, age, salt intake, anesthesia, and medication. Adrenals were examined using HES staining (Figures 1A and 1E and S1A and S1E) to distinguish the different cortical cell layers. Human ZG consisted of a noncontinuous layer of cells located in the most external part of the cortex (black arrows; Figures 1A and 1E and S1A and S1E8,9). Immunohistochemistry for Dab2 (Figures 1B and 1F and S1B and S1F) showed labeling in the external cortex, which was superposable to HES staining, demonstrating that Dab2 is a specific marker of the ZG not only in rodents13 but also in humans. In situ hybridization showed CYP11B2 expression in a limited number of ZG cells (Figures 1C and 1G and S1C and S1G), as described previously.12 Two distinct patterns of expression were distinguished, focal (Figures 1C and 1G and S1C) or partly continuous (Figure S1G). In both patterns, CYP11B2 expression was less extended than Dab2 expression, indicating that part of the ZG cells was silent in terms of aldosterone production. In situ hybridization of CYP11B1 showed expression in the entire ZF and ZR (Figures 1D and 1H and S1D and S1H), with no signal detected in CYP11B2 and Dab2-positive cells. Assessing the focal expression of CYP11B2 disclosed 2 types of clusters that coexisted in the same adrenal. Clusters of cells morphologically and functionally identified as ZG cells (Figure 2A and 2E), coexpressing Dab2 and CYP11B2 (Figure 2B, 2C, 2F, and 2G), were defined according to their size as foci (<300 μm) or megafoci (>300 μm). We also found large clusters of cells composed of morphological ZG cells in contact with the capsule and inner columnar ZF-like cells (Figure 2I through 2L), resembling the APCCs described by Nishimoto et al.12 In contrast to megafoci, APCCs did not express Dab2 (Figure 2J). In all of the cases, no expression of CYP11B1 was detected in CYP11B2-positive cells (Figure 2D, 2H, and 2L). This result strongly suggests that APCCs contain cells presenting an intermediate phenotype between ZG and ZF cells.
ZG Adjacent to APA Presents Structural and Functional Hyperplasia
The morphological examination of 25 APAs revealed a mixed cellular composition, with, in two thirds of cases, a majority of ZF-like cells (Table S3 and Figures 3A and S2A). Despite their different cellular compositions, all of the APAs expressed CYP11B2 in all (Figures 3C and S2C) or in a large number of cells (data not shown). Interestingly, Dab2 was coexpressed with CYP11B2 in APAs (Figures 3B and S2B), indicating that APAs consist of ZG cells that may have acquired morphological characteristics of ZF cells because of increased steroid production. In APAs, CYP11B1 was detected only in cells that did not express CYP11B2 (Figures 3D and S2D). Eleven adrenals presented secondary nodules that were also positively stained for CYP11B2.
Morphological analyses of peritumoral adjacent tissue revealed major changes in ZG, which appeared continuous (Figure 3E) or presented extended focal thickening associated with expression of CYP11B2 (Figure 3G) and Dab2 (Figure 3F) but negative for CYP11B1 (Figure 3H). Quantification of ZG hyperplasia8,9 showed that these changes were significantly different compared with control adrenals (Table 2). In peritumors that did not present ZG hyperplasia, focal expression of CYP11B2 was detected (Figure S2G and S2K) in all but 1 patient. As in control adrenals, megafoci composed of ZG cells (Figure S2E) expressing Dab2 (Figure S2F) and CYP11B2 (Figure S2G) but not CYP11B1 (Figure S2H), and APCCs formed of a mixed population of cells (Figure S2I), expressing CYP11B2 (Figure S2K) but not Dab2 (Figure S2J) and CYP11B1 (Figure S2L), were present. Foci, megafoci, and APCCs were also detected in peritumoral adjacent tissues with ZG hyperplasia (data not shown). No differences were observed between control adrenals and peritumoral adjacent tissues in terms of number of foci (P=0.30) or APCCs (P=0.86), whereas number of megafoci was significantly increased in peritumoral adjacent tissues compared with control adrenals (control adrenals: 0.45±0.25; peritumoral adjacent tissues: 2.58±0.63; P=0.011). When considering adrenals presenting ZG hyperplasia, CYP11B2 expression area, as quantified by image analysis, was only slightly and not significantly increased in peritumoral adjacent tissue compared with control adrenal (control adrenal: 1.28±0.3; peritumoral adjacent tissue: 1.92±0.3; P=0.083). Finally, there was no difference between adrenals with or without ZG hyperplasia in terms of age, sex, size of the APA, systolic and diastolic blood pressure, duration of hypertension, plasma potassium, aldosterone and renin, or urinary aldosterone levels (Table S4).
Tissue Remodeling and Decreased Vascularization of the Adrenal Cortex Adjacent to APA
To evaluate factors possibly promoting APCC and megafoci formation, as well as ZG hyperplasia, we next investigated nodulation and vascularization, factors that have been related to adrenal tissue remodeling.14 Control adrenals all presented some degree of incomplete or near complete nodulation, and in some adrenals complete nodulation was observed (Figure 4). Interestingly, the peritumoral adjacent tissues presented a major increase of tissue remodeling compared with control adrenals (score of nodulation in control adrenals: 11.19±1.95; peritumoral adjacent tissue: 56.00±8.92; P<0.0001; Figure 4B). However, there was no correlation between nodulation and any of the measured parameters related to aldosterone production, including number of foci, megafoci, and APCC; CYP11B2 expression area in peritumoral adjacent tissues and APA; as well as clinical and biological parameters (urinary aldosterone, plasma potassium, systolic and diastolic blood pressure, or duration of hypertension; Table S5).
Because tissue vascularization might be related to nodulation or to the presence of foci, megafoci, or APCC, we analyzed vascular surface in control adrenals and peritumoral adjacent tissues by CD34 immunohistochemistry (Figure 4C and 4D). Vascular architecture was similar in both tissues (Figure 4C). Although there was no difference in vascularization area between APA and control adrenals (not shown), as already described,15 vascularization surface was significantly decreased (≈33%) in peritumoral adjacent tissues compared with control adrenals (Figure 4D; control adrenals, 9.67±0.94 versus peritumoral adjacent tissues, 6.47±0.42; P=0.0010). Again, no correlation was observed between vascular surface and nodulation score, number of foci, megafoci, APCC, or CYP11B2 expression area in both control adrenals and peritumoral adjacent tissues (Table S5).
Transcriptional Phenotyping Does Not Suggest Adrenal Cortex Nodulation Being an Intermediate Toward APA Formation
To gain insight into the sequence of events linking adrenal cortex remodeling to APA, we inferred gene expression in the peritumoral adjacent tissue from published data by Giordano et al.16 To this end we determined differentially expressed genes (P<0.01) between the control adrenals and APA groups compiled from the data set (Table S2). When the covariance of gene expression of those selected genes is used as a distance measure in a principal component analysis of the control adrenals, peritumoral adjacent tissues, ACA, and APA (Figure 5A), the peritumoral adjacent tissue samples are very closely related to the control adrenal samples, even if they are shifted slightly along the first principal component toward the APA samples. The ACA samples are truly intermediate between control adrenals/peritumoral adjacent tissues and APA in their molecular phenotype. To further explore the nature of peritumoral adjacent tissue, we calculated specific gene signatures for control adrenal and APA likeliness. Again, the peritumoral adjacent tissue samples from the data set of Giordano et al16 correspond in their molecular phenotype much more to control adrenal than APA (Figure 5B). Hence, the molecular profiling done supports a general close relationship between peritumoral adjacent tissues and control adrenals.
Correlation of CYP11B2 Expression With Aldosterone Biosynthesis and Function
An important issue in PAL is whether there is a correlation between CYP11B2 expression and aldosterone biosynthesis and function. In PAL, total CYP11B2 expression area, including APA(s) and peritumoral adjacent tissue, was significantly increased as compared with control adrenals (control adrenals: 1.28±0.3; PAL: 39.86±3.5; P<0.0001). A significant negative correlation was observed between plasma potassium levels and 24-hour urinary aldosterone concentration (r=−0.75; P<0.0001; Figure S3A) in PAL patients. CYP11B2 expression area in APA was significantly correlated with plasma potassium levels (r=−0.51; P=0.016), whereas it was not correlated with 24-hour urinary aldosterone levels (r=0.32; P=0.16). Interestingly, when APA volume was taken into account for CYP11B2 expression, a significant correlation was observed between CYP11B2 expression*volume and 24-hour urinary aldosterone (r=0.52; P=0.012; Figure S3B). CYP11B2 expression*volume in APA was again negatively correlated with plasma potassium concentration, although less significantly (r=−0.41; P=0.056). When considering CYP11B2 expression area in peritumoral adjacent tissue, a positive, although not significant, correlation with plasma potassium levels was observed (r=0.39; P=0.06; Figure S3C).
In the present study, we report the occurrence of increased adrenal cortex remodeling and ZG hyperplasia in PAL. We found a dramatic increase in the nodulation of the cortex adjacent to the tumor in a large series of patients operated on for APA, as well as decreased vascularization. The ZG adjacent to APA expressed CYP11B2 mRNA, suggesting that it was functional despite autonomous aldosterone production from APA and suppressed renin-angiotensin system.
We first extensively characterized aldosterone-producing structures in the adrenal ZG. CYP11B2 was detected in all of the normal adrenal glands; however, its expression was restricted to a limited number of ZG cells. Only in rare cases was CYP11B2 expression semicontinuous. In most adrenals, CYP11B2 localized to 3 types of focal structures, the already described APCCs,12 which we show are specific cell clusters composed of a mixed population of Dab2− and CYP11B1− cells, and 2 newly described structures, foci or megafoci, composed of ZG cells that were Dab2+ and CYP11B1−. Although discontinuous or discrete CYP11B2 expression has been reported previously,12,17 our results first relate this expression pattern to ZG structure and support a model in which glomerulosa cells, silent for aldosterone production, have the potential to express CYP11B2, depending on physiological demand (eg, salt intake).18
APCCs have been described as morphologically isolated structures with a defined boundary from the neighboring areas, but without an apparent fibrous capsule, and it has been proposed that they could eventually develop into APA.12 However, no difference was observed in terms of the number of APCCs between peritumoral adjacent tissues and control adrenals; moreover, there was no correlation between the number of APCCs and adrenal cortex remodeling. In contrast, a significant increase of the number of megafoci was observed in peritumoral adjacent tissue. Furthermore, 14 of 25 adrenals from patients with APA presented continuous ZG hyperplasia that was not observed in control adrenals. In these patients, ZG was functional, with CYP11B2 expression detected in all of the cells. Although we confirm the presence of APCCs in both control adrenals and peritumoral adjacent tissues, our data do not support the evidence for constitutive and autonomous aldosterone production from these structures, which may eventually develop into APA. Indeed, morphological resemblance between APCCs and cells composing APA is not corroborated by shared molecular characteristics, because APCCs do not express the ZG marker Dab2, in contrast to APA. Furthermore, although persistence of CYP11B2 expression in the ZG or in APCCs has been observed previously in adrenals with APA,12,17 our results demonstrate that not only is the adjacent ZG not suppressed in the presence of autonomous aldosterone production, but that in 50% of patients, the entire ZG has acquired the capacity to express CYP11B2.
Our study demonstrates that adrenal cortex remodeling, decreased vascularization, and ZG hyperplasia are major features of adrenals with APA. A question arising from these results is whether nodulation of adjacent cortex precedes development of APA. In favor of nodulation being a primary event is the fact that we observe reduced vascularization in peritumoral adjacent tissue. Indeed, in most adult adrenals, a certain degree of nodularity can be observed, of which the frequency increases with age and severity of hypertension19; based on these findings, some authors propose that adrenocortical nodules may occur as a result of localized compensatory overgrowth of adrenocortical cells in response to localized ischemic changes attributed to atherosclerosis or hypertension.14 Accordingly, a certain degree of nodulation was observed in our control adrenals. In our study, the mean age of cases with PAL was 45±7 years compared with 61±14 years for patients operated with normal adrenals (P<0.0001), and a majority of control patients were also hypertensive, excluding age and hypertension as major determinants for the large remodeling observed in peritumoral adjacent tissue. Furthermore, there was no evidence for any correlation between vascularization and nodulation of the adjacent adrenal cortex, as well as between nodulation and levels or duration of hypertension. Although we could not perform expression studies on our peritumoral adjacent tissue, data inferred from publicly available transcriptomes indicate that the molecular phenotype of peritumoral adjacent tissue is not intermediary between control adrenals and APA, implying that abnormalities of the cortex adjacent to APA do not precede APA formation. Nodulation of the adrenal cortex is a feature specific to APA, because it was absent from adrenals with cortisol-producing adenomas (Figure S4). However, the natural history of the development of APA is unknown; therefore, we cannot exclude that tissue remodeling initiated a long time before detection of the clinical phenotype of PAL is underlying the formation of an autonomous bud with autonomous aldosterone production and then developing into APA.
Although we cannot exclude that hypertension attributed to APA is initiating tissue remodeling at some time point, another mechanism whereby APA may lead to increased nodulation is a local effect of aldosterone itself. Indeed, it is now clearly established that aldosterone is able to induce important tissue remodeling in many organs, including heart, kidney, and blood vessels.20 A prerequisite for a possible local effect of aldosterone in the adrenal cortex adjacent to APA is the expression of the mineralocorticoid receptor. Indeed, we found strong mineralocorticoid receptor expression in the ZG of control adrenals, as well as in peritumoral adjacent tissue (Figure S5). Also, local aldosterone effects have been described in adrenal H295R cells, where aldosterone induces, through an autocrine mineralocorticoid receptor–mediated action, the expression of T-type calcium channels.21 Future studies should address the relationship between increased aldosterone levels in APA and adrenal cortex remodeling.
The present study provides evidence for adrenal cortex remodeling, reduced vascularization, and ZG hyperplasia being major features associated with APA. Our results may have some clinical implications for the treatment of patients with PAL. Extensive remodeling of the adrenal cortex adjacent to APA, whether primary or secondary to APA itself, and the presence of secondary functional nodules in a large number of adrenals suggest that, despite a curative effect of enucleation of the adenoma associated with normalization of adrenal endocrine function,22 long-term follow-up is required because of the possibility that buds with autonomous aldosterone production, with the potential to develop into APA, may be present in the remaining adrenal tissue. This work provides the basis for future studies addressing the sequence of events leading to the development of APA.
We thank Dr Mathilde Sibony for kindly providing tissue samples from control adrenal glands, Prof Patrick Bruneval for adrenal samples with cortisol-producing adenoma, and Dr Celso Gomez-Sanchez for the anti-mineralocorticoid receptor antibody 1D5.
Sources of Funding
This work was funded through support received from the Agence Nationale pour la Recherche (ANR-07-PHYSIO-013-01), the Fondation pour la Recherche sur l’Hypertension Artérielle (AO 2007), and the Genopole Evry (to A.B.). J.-F.B.G. is recipient of a CONACYT-Mexico PhD Fellowship (No. 207676/302245).
- Received June 25, 2010.
- Revision received July 12, 2010.
- Accepted September 2, 2010.
Plouin PF, Amar L, Chatellier G. Trends in the prevalence of primary aldosteronism, aldosterone-producing adenomas, and surgically correctable aldosterone-dependent hypertension. Nephrol Dial Transplant. 2004; 19: 774–777.
Conn JW, Louis LH. Primary aldosteronism: a new clinical entity. Trans Assoc Am Physicians. 1955; 68: 215–231;discussion, 231–213.
Connell JM, Davies E. The new biology of aldosterone. J Endocrinol. 2005; 186: 1–20.
Lack EE. Adrenal cortical adenoma. In: Tumors of the Adrenal Glands and Extraadrenal Paraganglia (Afip Atlas of Tumor Pathology), 3rd series. Fascicle 19 ed. Washington, DC 1997: 91–122.
Lack EE. Developmental, physiologic, and anatomic aspects of adrenal cortex and medulla. In: Tumors of the Adrenal Glands and Extraadrenal Paraganglia (Afip Atlas of Tumor Pathology), 3rd series. Fascicle 19 ed. Washington, DC 1997: 13.
Sasano H, Suzuki T, Moriya T. Differential diagnosis and molecular advances. Endocr Pathol. 2003: 211–216.
Giordano TJ, Kuick R, Else T, Gauger PG, Vinco M, Bauersfeld J, Sanders D, Thomas DG, Doherty G, Hammer G. Molecular classification and prognostication of adrenocortical tumors by transcriptome profiling. Clin Cancer Res. 2009; 15: 668–676.
Enberg U, Volpe C, Hoog A, Wedell A, Farnebo LO, Thoren M, Hamberger B. Postoperative differentiation between unilateral adrenal adenoma and bilateral adrenal hyperplasia in primary aldosteronism by mRNA expression of the gene CYP11B2. Eur J Endocrinol. 2004; 151: 73–85.
Marney AM, Brown NJ. Aldosterone and end-organ damage. Clin Sci (Lond). 2007; 113: 267–278.
Walz MK, Peitgen K, Diesing D, Petersenn S, Janssen OE, Philipp T, Metz KA, Mann K, Schmid KW, Neumann HP. Partial versus total adrenalectomy by the posterior retroperitoneoscopic approach: early and long-term results of 325 consecutive procedures in primary adrenal neoplasias. World J Surg. 2004; 28: 1323–1329.