Prevalence, Clinical, and Molecular Correlates of KCNJ5 Mutations in Primary Aldosteronism

Transcription

1 Prevalence, Clinical, and Molecular Correlates of KCNJ5 Mutations in Primary Aldosteronism Sheerazed Boulkroun, Felix Beuschlein, Gian-Paolo Rossi, José-Felipe Golib-Dzib, Evelyn Fischer, Laurence Amar, Paolo Mulatero, Benoit Samson-Couterie, Stefanie Hahner, Marcus Quinkler, Francesco Fallo, Claudio Letizia, Bruno Allolio, Giulio Ceolotto, Maria Verena Cicala, Katharina Lang, Hervé Lefebvre, Livia Lenzini, Carmela Maniero, Silvia Monticone, Maelle Perrocheau, Catia Pilon, Pierre-François Plouin, Nada Rayes, Teresa M. Seccia, Franco Veglio, Tracy Ann Williams, Laura Zinnamosca, Franco Mantero, Arndt Benecke, Xavier Jeunemaitre, Martin Reincke, Maria-Christina Zennaro See Editorial Commentary, pp Downloaded from by guest on April 9, 2018 Abstract Primary aldosteronism is the most common form of secondary hypertension. Mutations in the KCNJ5 gene have been described recently in aldosterone-producing adenomas (APAs). The aim of this study was to investigate the prevalence of KCNJ5 mutations in unselected patients with primary aldosteronism and their clinical, biological and molecular correlates. KCNJ5 sequencing was performed on somatic (APA, n 380) and peripheral (APA, n 344; bilateral adrenal hyperplasia, n 174) DNA of patients with primary aldosteronism, collected through the European Network for the Study of Adrenal Tumors. Transcriptome analysis was performed in 102 tumors. Somatic KCNJ5 mutations (p.gly151arg or p.leu168arg) were found in 34% (129 of 380) of APA. They were significantly more prevalent in females (49%) than males (19%; P 10 3 ) and in younger patients ( versus years; P 10 3 ) and were associated with higher preoperative aldosterone levels ( versus ng/l; P 0.012) but not with therapeutic outcome after surgery. Germline KCNJ5 mutations were found neither in patients with APA nor those with bilateral adrenal hyperplasia. Somatic KCNJ5 mutations were specific for APA, because they were not identified in 25 peritumoral adrenal tissues or 16 cortisol-producing adenomas. Hierarchical clustering of transcriptome profiles showed that APAs with p.gly151arg or p.leu168arg mutations were indistinguishable from tumors without KCNJ5 mutations. In conclusion, although a large proportion of sporadic APAs harbors somatic KCNJ5 mutations, germline mutations are not similarly causative for bilateral adrenal hyperplasia. KCNJ5 mutation carriers are more likely to be females; younger age and higher aldosterone levels at diagnosis suggest that KCNJ5 mutations may be associated with a more florid phenotype of primary aldosteronism. (Hypertension. 2012;59: ) Online Data Supplement Key Words: aldosterone mineralocorticoids endocrine hypertension adrenal cortex diseases potassium channels mutation genotype-phenotype correlation Continuing medical education (CME) credit is available for this article. Go to to take the quiz. Received October 20, 2011; first decision November 3, 2011; revision accepted December 18, From the Institut National de la Santé et de la Recherche Médicale (S.B., L.A., B.S.-C., M.P., P.-F.P., X.J., M.-C.Z.), Unité Mixte de Recherche Scientifique 970, Paris Cardiovascular Research Center, Paris, France; University Paris Descartes (S.B., L.A., B.S.-C., M.P., P.-F.P., X.J., M.-C.Z.), Paris, France; Medizinische Klinik Innenstadt (F.B., E.F., M.R.), Ludwig-Maximilians-University, Munich, Germany; Dipartimento di Medicina Clinica e Sperimentale G. Patrassi University of Padova Medical School (G.-P.R., G.C., L.L., C.M., T.M.S.), University Hospital, Padova, Italy; Institut des Hautes Etudes Scientifiques and Centre National de la Recherche Scientifique (J.-F.G.-D., A.B.), Bures sur Yvette, France; Assistance Publique-Hôpitaux de Paris (L.A., P.-F.P., X.J., M.-C.Z.), Hôpital Européen Georges Pompidou, Paris, France; Division of Internal Medicine and Hypertension (P.M., S.M., F.V., T.A.W.), Department of Medicine and Experimental Oncology, University of Torino, Torino, Italy; Department of Medicine I (S.H., B.A., K.L.), Endocrine and Diabetes Unit, University Hospital Würzburg, Würzburg, Germany; Clinical Endocrinology (M.Q.), Campus Mitte, University Hospital Charité, Berlin, Germany; Department of Medical and Surgical Sciences (F.F., C.P.), University of Padova, Padova, Italy; Department of Clinical Sciences (C.L., L.Z.), Secondary Hypertension Unit, University Sapienza, Policlinico Umberto I, Viale del Policlinico, Rome, Italy; Endocrine Unit (M.V.C., F.M.), Department of Medical and Surgical Sciences, University of Padua, Padua, Italy; Institut National de la Santé et de la Recherche Médicale (H.L.), U982, Mont-Saint Aignan, France; University of Rouen (H.L.), Mont-Saint Aignan, France; University Hospital of Rouen (H.L.), Rouen, France; Department of General, Visceral and Transplantation Surgery (N.R.), Campus Virchow, University Hospital Charité, Berlin, Germany; Institut de Recherche Interdisciplinaire (A.B.), Centre National de la Recherche Scientifique USR3078, University of Lille I and Lille II, Villeneuve d Ascq, France. S.B., F.B., G.P.-R., J.-F.G.-D.; E.F., L.A., P.M.; B.S.-C., S.H., M.Q.; F.M., A.B., X.J., M.R., and M.-C.Z. contributed equally to this work. The online-only Data Supplement is available with this article at /-/DC1. Correspondence to Maria-Christina Zennaro, Institut National de la Santé et de la Recherche Médicale, U970, Paris Cardiovascular Research Center-PARCC, 56 rue Leblanc, Paris, France. maria-christina.zennaro@inserm.fr 2012 American Heart Association, Inc. Hypertension is available at DOI: /HYPERTENSIONAHA

2 Boulkroun et al KCNJ5 Mutations in Aldosterone-Producing Adenoma 593 Downloaded from by guest on April 9, 2018 Hypertension is a major cardiovascular risk factor that affects between 10% and 40% of the population in industrialized countries. Detection of secondary forms of hypertension is particularly important because it allows for the targeted management of the underlying disease. Primary aldosteronism (PA) is the most common form of secondary hypertension, with an estimated prevalence between 6% and 12% of hypertensives and as high as 20% in patients with resistant hypertension. 1 5 PA occurs as the result of a dysregulation of the mechanisms controlling adrenal aldosterone production, ultimately leading to hypertension with low plasma renin and elevated aldosterone sometimes associated with hypokalemia. Among subtypes of PA, aldosteroneproducing adenoma (APA) and bilateral adrenal hyperplasia (BAH; also known as idiopathic hyperaldosteronism) together account for 95% of cases. 1 3 Aldosterone production from the adrenal zona glomerulosa is tightly controlled to maintain electrolyte and fluid homeostasis by the kidney. Thus, the two most important physiological stimuli of aldosterone secretion are angiotensin II and serum potassium. Glomerulosa cell membrane depolarization leads to opening of voltage-activated Ca 2 channels with activation of the calcium signaling pathway, the major mediator of aldosterone production. 6 Potassium channels play a central role in regulating the membrane potential of glomerulosa cells, as demonstrated in mouse models where the genetic deletion of TWIK related Acid Sensitive K (TASK) channels TASK-1 and TASK-3 removes the background K current, which results in a marked depolarization of adrenal zona glomerulosa cell membrane potential and low-renin hypertension attributed to the autonomous overproduction of aldosterone. 7,8 Along the same line, recent work by Choi et al 9 has identified somatic KCNJ5 mutations in 8 out of a series of 22 APAs. The authors identified 2 recurrent KCNJ5 mutations, G151R and L168R, affecting 2 residues in and near the selectivity filter of the Kir3.4 potassium channel. A similar inherited mutation (T158A) was also found in a family with a severe Mendelian form of PA associated with massive BAH causing early onset resistant arterial hypertension, which was cured by bilateral adrenalectomy. The 3 mutations all result in loss of channel selectivity, with increased sodium conductance leading to membrane depolarization; these changes are presumed to be responsible for constitutive aldosterone secretion and cell proliferation. 9 However, the frequency and clinical and molecular correlates of KCNJ5 mutations, as well as their role in the onset and/or maintenance of PA, remained to be established. Taking advantage of a very large collection of APA samples, endowed with comprehensive clinical and biological information, collected through reference centers organized within the European Network for the Study of Adrenal Tumors (ENS@T), we explored the prevalence of KCNJ5 mutations across the spectrum of APA. We also tested the hypothesis that germinal KCNJ5 mutations may be involved in BAH, the most common form of PA. The clinical and biochemical correlates of the mutation status were investigated, with particular attention to severity and follow-up data. We further investigated the specificity of KCNJ5 mutations relative to cortisol-producing adenomas, their relation to adrenal cortex remodeling, and their molecular consequences in terms of gene expression to provide insight into the mechanisms underlying cell proliferation and aldosterone production. Subjects and Methods Patients Patients with PA were recruited between 1994 and 2010 from 9 different centers belonging to the European Network for the Study of Adrenal Tumors APA working group ( Case detection and subtype identification of PA were performed according to institutional guidelines and in accordance with the Endocrine Society guidelines In patients diagnosed as having PA, a thin-section computed tomography scan or MRI of the adrenal and/or an adrenal venous sampling (AVS) was performed to differentiate between unilateral and bilateral aldosterone hypersecretion. AVS selectivity and lateralization criteria were different among centers, with 7 centers (Paris, Munich, Torino, Berlin, Würzburg, Padova-A, and Padova-C) using a selectivity index for successful AVS (adrenal vein/inferior vena cava cortisol gradient) 2 and a lateralization index (aldosterone/cortisol ratio from high to low side) 4. Two centers (Padova-B and Rome) use AVS selectivity index 1.1 and lateralization index 2.0. Baseline clinical and biochemical characteristics of the patients are indicated in Table S1 (please see the online-only Data Supplement). The diagnosis of adrenocortical adenoma was histologically confirmed after surgical resection. PA patients not diagnosed as having an APA were classified as having BAH. A final diagnosis of APA, diagnosed by computed tomography scanning and AVS, was considered proven when the following conditions were satisfied: (1) histological demonstration of adenoma; (2) normalization of hypokalemia, if present; (3) cure or improvement of hypertension; and (4) normalization of ARR and/or suppressibility of aldosterone under saline load. 3,15 All of the patients gave written informed consent for genetic and clinical investigation within each individual institution. Procedures were in accordance with institutional guidelines. DNA and RNA Isolation and RT-PCR DNA or RNA was extracted from a total of 424 APAs, 25 paired peritumoral adrenal cortices, 10 control adrenals (obtained from radical nephroadrenalectomies), 16 cortisol-producing adenomas, and 518 peripheral DNA samples (344 from patients with APA and 174 from patients with BAH). For DNA and RNA isolation and RT-PCR procedures please see the online-only Data Supplement. KCNJ5 Sequencing The KCNJ5 coding sequence spanning amino acids 122 to 199 was sequenced. For KCNJ5 sequencing please see the online-only Data Supplement. Microarray Analysis For microarray analysis please see the online-only Data Supplement. Statistical Analyses If not stated otherwise, group results are expressed as mean SEM or median and interquartile range for nonnormally distributed variables. Skewed variables were log transformed before statistical testing. Within-group changes from baseline to follow-up were assessed by paired t test. Data between groups were compared using ANOVA test followed by post hoc analysis (Tamhanes T2 test) or 2 test. For nonparametric testing, Mann-Whitney U test was applied. For multivariate analysis, factors significantly associated with the presence of mutations in univariate logistic regression were inserted into a multivariate model. P 0.05 was considered to be statistically significant. Statistical analysis used SPSS 18.

3 594 Hypertension March 2012 Table 1. Type and Frequency of KCNJ5 Mutations in APA Across Different Centers Downloaded from by guest on April 9, 2018 Results Somatic KCNJ5 Mutations Account for a Substantial Proportion of APA Successful sequencing of 380 somatic APA DNA samples demonstrated the occurrence of both the p.gly151arg and the p.leu168arg mutations at the heterozygous state in samples from all of the centers. Two different nucleotide substitutions accounted for the p.gly151arg mutations, c.451g A and c.451g C (Figure S1 and Table 1). The frequencies of the mutations varied among centers (P 0.023), with p.gly151arg being more frequent than p.leu168arg. The overall prevalence of somatic KCNJ5 mutations in APA was 20% for p.gly151arg and 14% for p.leu168arg among samples, with a total prevalence of 34%. For 39 APA Table 2. Center No. With Successful Sequencing c.451g A/G C p.gly151arg c.503t G p.leu168arg No. With Mutations % With Mutations Paris 134* Munich Torino Berlin Würzburg Padova-A Padova-B Padova-C Rome Total APA indicates aldosterone-producing adenoma. *For 216 APA DNA samples, the corresponding paired blood DNA samples were also sequenced. Twenty-five paired peritumoral DNA samples were also sequenced. All of the mutations were in the heterozygous state, and none concurred in the same patient. Clinical and Biochemical Correlates of KCNJ5 Mutations in APA samples, pathological confirmation of the diagnosis was available through in situ hybridization of CYP11B2, coding for aldosterone synthase, in the resected APA. In this subset, the frequency of KCNJ5 mutations was 46% (18 of 39), not significantly different from that observed in the overall cohort (P 1.00). The absence of KCNJ5 mutations in 344 peripheral DNA samples from patients with APA, 216 of which were paired with tumor DNA (72 harboring KCNJ5 mutations), confirmed the somatic nature of the genetic alteration. Importantly, no KCNJ5 mutations were found in 174 germinal DNA samples from patients with BAH. Adrenal cortex remodeling, with increased nodulation, reduced vascularization, and functional zona glomerulosa hyperplasia, is a major feature of adrenals with APA. 16 This finding raises the questions as to whether KCNJ5 mutations Variable Mutation Present No Mutation P P Age at diagnosis of PA, y (n 128) (n 249) No. of females, % 97 (74) 101 (41) Preoperative SBP, mm Hg 154 2(n 120) 157 1(n 234) Preoperative DBP, mm Hg 96 1(n 120) 96 1(n 234) Preoperative No. of antihypertensive medications (n 118) (n 201) Lowest recorded potassium level, mmol/l (n 122) (n 214) Preoperative plasma aldosterone concentration, ng/l* (248.5; 619.1) (n 112) (200.5; 478.8) (n 207) Preoperative PRA, ng*ml/h* 0.30 (0.12; 0.49) (n 47) 0.27 (0.15; 0.60) (n 101) Preoperative plasma renin concentration, mu/l* 2.2 (1.5; 4.2) (n 64) 2.7 (1.4; 4.9) (n 98) Adenoma size, mm 16.0 (12.0; 20.0) (n 116) 15.0 (10.0; 20.0) (n 183) Postoperative SBP, mm Hg 129 1(n 107) 131 1(n 214) Postoperative DBP, mm Hg 84 1(n 107) 83 1(n 214) Postoperative No. of antihypertensive medications (n 105) (n 195) Postoperative serum potassium level, mmol/l (n 111) (n 223) Postoperative plasma aldosterone concentration, ng/l* (64.0; 191.2) (n 81) (71.0; 210.0) (n 161) Values represent mean SEM unless otherwise specified. PA indicates primary aldosteronism; PRA, plasma renin activity; DBP, diastolic blood pressure; SBP, systolic blood pressure; APA, aldosterone-producing adenoma; n, No. of subjects for each group. *Data are median (interquartile range). P value was adjusted for age and sex.

4 Boulkroun et al KCNJ5 Mutations in Aldosterone-Producing Adenoma 595 Downloaded from by guest on April 9, 2018 Figure 1. KCNJ5 mutations do not lead to specific transcriptional profiles of aldosterone-producing adenomas (APAs). A, Hierarchical clustering of 102 APA samples with all genes. APA harboring the p.gly151arg (G151R) or p.leu168arg (L168R) mutations are distributed among non mutated tumors (neg). B, Heat-maps of the statistically significantly (post hoc P 0.01) differentially expressed genes between samples having a G151R (p.gly151arg; left) or an L168R (p.leu168arg; right) mutation with respect of those from patients lacking them (neg). L indicates the log2-based fold change of gene expression between groups. C, Overlap of the upregulated (top part of the Venn diagram) and downregulated (bottom) genes found to have a common regulatory signal with respect to the patients without either p.gly151arg or p.leu168arg genotype. D, Classification of APA samples according to the 96 overlapping statistically significantly (post hoc P 0.01) differentially expressed genes found in C. E, Upstream regulatory genes associated with calcium signaling. F, Target genes implicated in steroid metabolism. may take their origin in the peritumoral adrenal cortex, leading to the formation of autonomous buds eventually developing into APA. Sequencing of KCNJ5 in somatic DNA from 25 paired peritumoral cortices, from which 8 presented somatic KCNJ5 mutations in the corresponding APA, demonstrated that no mutation was present in peritumoral tissue. Finally, KCNJ5 mutations appeared to be specifically associated with APA, because none of 16 somatic DNA samples from cortisol-producing adenomas nor DNA from 10 control adrenal cortices were found to harbor the p.gly151arg or p.leu168arg mutation. Clinical and Biochemical Phenotypes Correlate With KCNJ5 Mutation Status To investigate whether KCNJ5 mutations may be correlated with a specific subgroup of APAs, we analyzed the mutation status with regards to the patient s clinical and biochemical characteristics. Remarkably, somatic KCNJ5 mutations in APA were more frequent in females than in males (49% versus 19%; P 10 3) and were also correlated with clinical and biochemical patient characteristics (Table 2). Compared with patients without mutations, those harboring mutations were younger ( versus years; P 0.001) in both males ( versus years; P 0.001) and females ( versus years; P 0.046). Mutation carriers also had higher plasma aldosterone levels under neutral medication measured at diagnosis than patients with KCNJ5 negative APA (387.4 ng/l [248.5; ng/l] versus ng/l [200.5; ng/l]; P 0.012; Table 2). This difference remained significant after adjustment for age and sex, two factors influencing aldosterone and renin levels.17 In contrast, no significant differences between KCNJ5 mutation carriers and noncarriers were detected with regard to systolic and diastolic blood pressures, serum potassium, and plasma renin levels, as well as number of medications taken before surgery. The mean adenoma size in KCNJ5 mutation carriers was 18 mm, typical for APA and comparable in both groups of patients. Follow-up after adrenalectomy showed similar outcome between carriers and noncarriers of KCNJ5 mutations, in terms of biochemical correction of PA and blood

5 % mutation 596 Hypertension March 2012 Downloaded from by guest on April 9, 2018 pressure outcome (Table 2). Although patients with mutations appeared to require fewer postoperative medications, this difference was no longer significant after correction for age and sex, two major contributors to postsurgical outcome in patients with APA. 18,19 Finally, in univariate models, age at diagnosis (in years or decades), sex, and preoperative aldosterone levels were associated with the presence of KCNJ5 mutations (Table S2). In multivariate analyses, the association of mutations with age (P ), sex (P ), and plasma aldosterone (P ) remained significant. KCNJ5 Mutations Are Not Correlated With Unique Transcriptional Profiles of APA Next we asked whether KCNJ5 mutations would lead to a specific molecular phenotype of APA. Hierarchical clustering of transcriptome profiles from 102 APAs clearly demonstrated that carriers of p.gly151arg or p.leu168arg are not distinguishable at the whole transcriptome level from those without mutation (Figure 1A). Even when analyzing a limited set of genes, that is, regulatory genes associated with calcium signaling (Figure 1E and Table S3) or downstream target genes implicated in steroid metabolism (Figure 1F and Table S4), carriers of KCNJ5 mutations were not separated from noncarriers. Indeed, the APA carrying KCNJ5 mutations are distinguished from noncarriers only by a limited set of genes: for p.gly151arg, 241 statistically significantly (post hoc P 0.01) differentially expressed genes were identified, 106 of which were upregulated and 135 downregulated. For p.leu168arg, 349 differentially expressed genes were identified, 90 upregulated, and 259 downregulated (Figure 1B and 1C). Classification of APA samples according to the 96 overlapping, statistically significantly (post hoc P 0.01) differentially expressed genes enriches the group of APAs harboring somatic mutations with respect to patients not presenting either p.gly151arg or p.leu168arg genotype (Figure 1D). However, classification of differentially expressed genes according to gene ontologies did not allow identification of major pathways or biological functions that could contribute to APA development (Table S5). Taken together, these results imply that abnormal Kir3.4 channel function does not lead to activation of unique signaling cascades but rather indicates that KCNJ5 mutation may represent one of several different mechanisms to activate the calcium signaling pathway. Discussion Despite the recent publication of guidelines for case detection, diagnosis, and treatment of PA, reliable diagnostic and prognostic biomarkers are lacking to achieve a more sensitive and specific screening, as well as to open new therapeutic strategies. 14,20,21 Indeed, the early detection of PA has an enormous impact on clinical outcome and survival, given the severe cardiovascular adverse effects of aldosterone excess that are independent of high blood pressure levels and predict achievement of normotension after surgical treatment The recent identification of somatic KCNJ5 mutations in a limited number of subjects with APA could represent a breakthrough in our understanding of the pathogenic mechanisms underlying the disease. 9 Here we report that somatic * Female Male < >70 Age group (years) Figure 2. Prevalence of KCNJ5 mutations as a function of age. Overall frequency of p.gly151arg or p.leu168arg KCNJ5 mutations stratified by age group and gender. *Significant differences in percentage of mutation carriers between sexes in the respective age group. Numbers in columns indicate the number of subjects for each group. KCNJ5 mutations are present in 34% of unselected APA samples across a large series of patients with PA recruited at major referral centers in Europe. Although a germline KCNJ5 mutation had also been reported in a familial form of PA associated with massive BAH, our results demonstrate that similar germline mutations are not involved in the development of BAH, the most common form of PA. 9,24 Notably, KCNJ5 mutation carriers were younger in comparison with noncarriers, suggesting that they had an earlier onset of the disease or earlier diagnosis because of a more severe phenotype, the latter being also supported by higher preoperative plasma aldosterone levels in mutation carriers. Another striking observation was the higher frequency of KCNJ5 mutations in females compared with males. Although the overall frequency of KCNJ5 mutations declined with age at diagnosis, this sex difference remained significant across different age groups (Figure 2). This finding is reminiscent of the peculiar phenotype of TASK-1 potassium channel knockout animals, which show severe hyperaldosteronism in mature mice in females only, because of abnormal expression of aldosterone synthase in the adrenal zona fasciculata. 8 The absence of hyperaldosteronism in male animals is explained by the differential sex-dependent expression of TASK-3 channels in the adrenal cortex, where TASK-3 expression in male animals can compensate for the lack of its normal heterodimerization partner, TASK-1. This supports a mechanism whereby women with KCNJ5 mutations may be more prone to membrane depolarization and subsequent APA development because of their inability to compensate for Kir3.4 dysfunction via other potassium channels, like TASK-3. Other than allowing evaluation of the overall frequency of KCNJ5 mutations across a large series of patients, this multicenter study has allowed for appreciating some heterogeneity in the prevalence of KCNJ5 mutations across centers. This might reflect the clinical, biological, and genetic diversity of patients with APA, depending also on the different strategies for patient recruitment. Indeed, the screening and confirmatory procedures varied among centers, with different * *

6 Boulkroun et al KCNJ5 Mutations in Aldosterone-Producing Adenoma 597 Table 3. Type and Frequency of KCNJ5 Mutations in APA According to Screening Procedures Used in Different Centers Centers Successful Sequencing (n) c.451g A/G C p.gly151arg c.503t G p.leu168arg TOT % Mutations SI 2; LI 4* SI 1.1; LI TOT % APA indicates aldosterone-producing adenoma; LI, lateralization index; SI, selectivity index; TOT, total. *Adrenal venous sampling selectivity and lateralization criteria is as follows: SI 2; LI 4: Paris, Munich, Torino, Berlin, Wur zburg, Padova-A, and Padova-C. Adrenal venous sampling selectivity and lateralization criteria is as follows: SI 1.1; LI 2: Padova-B and Rome. Downloaded from by guest on April 9, 2018 criteria used for selecting the patients for adrenalectomy. 10,13,25 27 These differences cover the clinical and diagnostic spectrum of PA and, thus, provide a reliable estimate of the true prevalence of KCNJ5 mutations in APA. The frequency of KCNJ5 mutations was found to be higher in patients selected on more conservative criteria (Table 3); together with the younger age and higher aldosterone levels at diagnosis, this suggests that KCNJ5 mutations may be associated with a more florid phenotype. Although we cannot exclude that some of the resected APAs were not BAHs, given the diagnostic complexity of PA, therapeutic outcome was similar among centers, excluding a major recruitment bias because of different screening procedures. Furthermore, when considering patients with postoperative aldosterone levels 180 ng/l and no improvement in terms of postoperative medications, potentially corresponding to cases with BAH, only 17 (4.5%) of 380 patients were identified. Among those, KCNJ5 mutations were found in 5 patients (P 0.80 for mutation frequency compared with the entire cohort), thus excluding that the different mutation frequency across centers reflects a higher prevalence of misdiagnosed BAH. Although adrenals from patients with APA show increased nodulation and adrenal cortex remodeling, suggesting a potential continuum and a common pathogenic mechanism between BAH and APA formation, KCNJ5 mutations appear to be isolated events in the progression toward APA, because they are not found in the peritumoral cortex. The fact that KCNJ5 mutations were not found in cortisol-producing adenoma further suggests independent pathways of adrenal cortex tumorigenesis, with APA originating from the zona glomerulosa expressing KCNJ5. Alternatively, KCNJ5 mutations may be associated with aldosterone excess, with additional, as yet unknown, mechanisms being responsible for tumorigenesis. Remarkably, KCNJ5 mutations do not induce a unique molecular phenotype of APA at the transcriptome level, as could have been expected from mutations acting on a key signaling pathway or master transcriptional regulators. Rather, it appears that they represent one if several different possible mechanisms resulting in activation of calcium signaling in the zona glomerulosa, ultimately leading to autonomous aldosterone production and (possibly) cellular proliferation. In the absence of formal demonstration that KCNJ5 mutations and membrane depolarization lead to increased cell proliferation, our results also raise the possibility that KCNJ5 mutations are not causative of APA development but are a consequence of tumorigenesis, playing only a contributory role toward aldosterone overproduction and tumor growth. Perspectives Although the causal relationship between KCNJ5 mutations and aldosterone overproduction and adrenal cell proliferation remains to be established, more than one third of patients with APA present somatic KCNJ5 mutations. Importantly, similar germinal mutations are not responsible for BAH, the most common form of PA. Future studies need to address the genetic determinants of the remaining 70% of cases of sporadic PA. Although unilateral adrenalectomy represents the treatment of choice for lateralized PA in specialized departments, it is tempting to speculate that Kir3.4 potassium channels may become interesting new drug targets for a subset of APAs not eligible for surgery. Acknowledgments In Paris, we thank Christophe Simian and Fabio Fernandes-Rosa (Genetics department and Institut National de la Santé et de la Recherche Médicale U970, Paris, France) for helpful assistance, as well as the French COMETE (COrtico et MEdullo-surrénale: les Tumeurs Endocrines) Network. In Munich, we thank Brigitte Mauracher, Igor Shapiro, and Susanne Schmid for technical assistance and patient care, respectively, and Stephan Endres (all Medizinische Klinik Innenstadt, Munich, Germany) for statistical support. In Torino, we thank Franco Rabbia and Chiara Bertello (Division of Internal Medicine, University of Torino, Torino, Italy) for selection and clinical study of the patients. In Berlin, we thank Sabine Wricke for technical assistance and Gregor Hanslik for clinical workup (Clinical Endocrinology, Campus Mitte, University Hospital Charité, Berlin, Germany). In Würzburg, we thank Martina Zink for technical assistance. In Padova, we thank Maniselvan Kuppusamy, Valentina Gallina, Beatrice Rubin, and Raffaele Pezzani (University of Padova Medical School, University Hospital) for helpful assistance. Sources of Funding This work was funded through institutional support from Institut National de la Santé et de la Recherche Médicale and by the Agence Nationale pour la Recherche (ANR Physio 2007, No , to M.C.-Z., A.B., and H.L.), the Fondation pour la Recherche sur l Hypertension Artérielle (AO 2007, to M.C.-Z., A.B., and H.L.), and the Genopole Evry (to A.B.). J.-F.G.-D. is the recipient of a Consejo Nacional de Ciencia y Tecnología-Mexico PhD Fellowship (No /302245). The French COMETE (COrtico et MEdullosurrénale: les Tumeurs Endocrines) network is supported in part by PHRC (Programme Hospitalier de Recherche Clinique) grant AOM and by grants from Institut National de la Santé etdela Recherche Médicale and Ministère Délégué à la Recherche et des Nouvelles Technologies. The work was also funded by grants from the Italian Ministry of University, Scientific and Technological

7 598 Hypertension March 2012 Downloaded from by guest on April 9, 2018 Research (COFIN, Cofinanziamento progetti di ricerca di interesse nazionale, No ), Compagnia di San Paolo, and Fondi Regione Piemonte Ricerca Finalizzata 2008 I and II. The German Conn s Registry-Else-Kröner Hyperaldosteronism Registry is supported by a grant of the Else Kröner-Fresenius Stiftung to M.R., who is also supported by the Deutsche Forschungsgemeinschaft (Re 752/17-1). S.H. is also supported by the Else Kröner-Fresenius Stiftung. The European Network for the Study of Adrenal Tumors registry is supported by a grant of the European Science Foundation and has received funding from the Seventh Framework Programme (FP7/ ) under grant agreement No None. Disclosures References 1. Mulatero P, Stowasser M, Loh KC, Fardella CE, Gordon RD, Mosso L, Gomez-Sanchez CE, Veglio F, Young WF Jr. Increased diagnosis of primary aldosteronism, including surgically correctable forms, in centers from five continents. J Clin Endocrinol Metab. 2004;89: 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: Rossi GP, Bernini G, Caliumi C, Desideri G, Fabris B, Ferri C, Ganzaroli C, Giacchetti G, Letizia C, Maccario M, Mallamaci F, Mannelli M, Mattarello MJ, Moretti A, Palumbo G, Parenti G, Porteri E, Semplicini A, Rizzoni D, Rossi E, Boscaro M, Pessina AC, Mantero F. A prospective study of the prevalence of primary aldosteronism in 1,125 hypertensive patients. J Am Coll Cardiol. 2006;48: Calhoun DA, Nishizaka MK, Zaman MA, Thakkar RB, Weissmann P. Hyperaldosteronism among black and white subjects with resistant hypertension. Hypertension. 2002;40: Douma S, Petidis K, Doumas M, Papaefthimiou P, Triantafyllou A, Kartali N, Papadopoulos N, Vogiatzis K, Zamboulis C. Prevalence of primary hyperaldosteronism in resistant hypertension: a retrospective observational study. Lancet. 2008;371: Spat A, Hunyady L. Control of aldosterone secretion: a model for convergence in cellular signaling pathways. Physiol Rev. 2004;84: Davies LA, Hu C, Guagliardo NA, Sen N, Chen X, Talley EM, Carey RM, Bayliss DA, Barrett PQ. TASK channel deletion in mice causes primary hyperaldosteronism. Proc Natl Acad Sci U S A. 2008;105: Heitzmann D, Derand R, Jungbauer S, Bandulik S, Sterner C, Schweda F, El Wakil A, Lalli E, Guy N, Mengual R, Reichold M, Tegtmeier I, Bendahhou S, Gomez-Sanchez CE, Aller MI, Wisden W, Weber A, Lesage F, Warth R, Barhanin J. Invalidation of TASK1 potassium channels disrupts adrenal gland zonation and mineralocorticoid homeostasis. Embo J. 2008;27: Choi M, Scholl UI, Yue P, Bjorklund P, Zhao B, Nelson-Williams C, Ji W, Cho Y, Patel A, Men CJ, Lolis E, Wisgerhof MV, Geller DS, Mane S, Hellman P, Westin G, Akerstrom G, Wang W, Carling T, Lifton RP. K channel mutations in adrenal aldosterone-producing adenomas and hereditary hypertension. Science. 2011;331: Letavernier E, Peyrard S, Amar L, Zinzindohoue F, Fiquet B, Plouin PF. Blood pressure outcome of adrenalectomy in patients with primary hyperaldosteronism with or without unilateral adenoma. J Hypertens. 2008;26: Reincke M, Rump LC, Quinkler M, Hahner S, Diederich S, Lorenz R, Seufert J, Schirpenbach C, Beuschlein F, Bidlingmaier M, Meisinger C, Holle R, Endres S. Risk factors associated with a low glomerular filtration rate in primary aldosteronism. J Clin Endocrinol Metab. 2009;94: Mulatero P, Monticone S, Veglio F. Diagnosis and treatment of primary aldosteronism. Rev Endocr Metab Disord. 2011;12: Rossi GP. Diagnosis and treatment of primary aldosteronism. Endocrinol Metab Clin North Am. 2011;40: Funder JW, Carey RM, Fardella C, Gomez-Sanchez CE, Mantero F, Stowasser M, Young WF Jr, Montori VM. Case detection, diagnosis, and treatment of patients with primary aldosteronism: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2008;93: Mulatero P, Bertello C, Verhovez A, Rossato D, Giraudo G, Mengozzi G, Limerutti G, Avenatti E, Tizzani D, Veglio F. Differential diagnosis of primary aldosteronism subtypes. Curr Hypertens Rep. 2009;11: Boulkroun S, Samson-Couterie B, Dzib JF, Lefebvre H, Louiset E, Amar L, Plouin PF, Lalli E, Jeunemaitre X, Benecke A, Meatchi T, Zennaro MC. Adrenal cortex remodeling and functional zona glomerulosa hyperplasia in primary aldosteronism. Hypertension. 2010;56: Meneton P, Galan P, Bertrais S, Heudes D, Hercberg S, Menard J. High plasma aldosterone and low renin predict blood pressure increase and hypertension in middle-aged Caucasian populations. J Hum Hypertens. 2008;22: Sawka AM, Young WF, Thompson GB, Grant CS, Farley DR, Leibson C, van Heerden JA. Primary aldosteronism: factors associated with normalization of blood pressure after surgery. Ann Intern Med. 2001;135: Stowasser M, Klemm SA, Tunny TJ, Storie WJ, Rutherford JC, Gordon RD. Response to unilateral adrenalectomy for aldosterone-producing adenoma: effect of potassium levels and angiotensin responsiveness. Clin Exp Pharmacol Physiol. 1994;21: Matrozova J, Steichen O, Amar L, Zacharieva S, Jeunemaitre X, Plouin PF. Fasting plasma glucose and serum lipids in patients with primary aldosteronism: a controlled cross-sectional study. Hypertension. 2009;53: Rossi GP, Bolognesi M, Rizzoni D, Seccia TM, Piva A, Porteri E, Tiberio GA, Giulini SM, Agabiti-Rosei E, Pessina AC. Vascular remodeling and duration of hypertension predict outcome of adrenalectomy in primary aldosteronism patients. Hypertension. 2008;51: Milliez P, Girerd X, Plouin PF, Blacher J, Safar ME, Mourad JJ. Evidence for an increased rate of cardiovascular events in patients with primary aldosteronism. J Am Coll Cardiol. 2005;45: Rossi GP, Sechi LA, Giacchetti G, Ronconi V, Strazzullo P, Funder JW. Primary aldosteronism: cardiovascular, renal and metabolic implications. Trends Endocrinol Metab. 2008;19: Geller DS, Zhang J, Wisgerhof MV, Shackleton C, Kashgarian M, Lifton RP. A novel form of human mendelian hypertension featuring nonglucocorticoid-remediable aldosteronism. J Clin Endocrinol Metab. 2008;93: Amar L, Plouin PF, Steichen O. Aldosterone-producing adenoma and other surgically correctable forms of primary aldosteronism. Orphanet J Rare Dis. 2010;5: Mulatero P, Bertello C, Sukor N, Gordon R, Rossato D, Daunt N, Leggett D, Mengozzi G, Veglio F, Stowasser M. Impact of different diagnostic criteria during adrenal vein sampling on reproducibility of subtype diagnosis in patients with primary aldosteronism. Hypertension. 2010;55: Vonend O, Ockenfels N, Gao X, Allolio B, Lang K, Mai K, Quack I, Saleh A, Degenhart C, Seufert J, Seiler L, Beuschlein F, Quinkler M, Podrabsky P, Bidlingmaier M, Lorenz R, Reincke M, Rump LC. Adrenal venous sampling: evaluation of the German Conn s Registry. Hypertension. 2011;57:

8 Downloaded from by guest on April 9, 2018 Prevalence, Clinical, and Molecular Correlates of KCNJ5 Mutations in Primary Aldosteronism Sheerazed Boulkroun, Felix Beuschlein, Gian-Paolo Rossi, José-Felipe Golib-Dzib, Evelyn Fischer, Laurence Amar, Paolo Mulatero, Benoit Samson-Couterie, Stefanie Hahner, Marcus Quinkler, Francesco Fallo, Claudio Letizia, Bruno Allolio, Giulio Ceolotto, Maria Verena Cicala, Katharina Lang, Hervé Lefebvre, Livia Lenzini, Carmela Maniero, Silvia Monticone, Maelle Perrocheau, Catia Pilon, Pierre-François Plouin, Nada Rayes, Teresa M. Seccia, Franco Veglio, Tracy Ann Williams, Laura Zinnamosca, Franco Mantero, Arndt Benecke, Xavier Jeunemaitre, Martin Reincke and Maria-Christina Zennaro Hypertension. 2012;59: ; originally published online January 23, 2012; doi: /HYPERTENSIONAHA Hypertension is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX Copyright 2012 American Heart Association, Inc. All rights reserved. Print ISSN: X. Online ISSN: The online version of this article, along with updated information and services, is located on the World Wide Web at: Data Supplement (unedited) at: Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Hypertension can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: Subscriptions: Information about subscribing to Hypertension is online at:

9 Prevalence, clinical and molecular correlates of KCNJ5 mutations in primary aldosteronism Online Supplement Sheerazed Boulkroun 1,2* PhD, Felix Beuschlein 3* MD, Gian-Paolo Rossi 4* MD, José-Felipe Golib-Dzib 5* MSci, Evelyn Fischer 3$ MD, Laurence Amar 1,2,6$ MD, Paolo Mulatero 7$ MD, Benoit Samson-Couterie 1,2 BSci, Stefanie Hahner 8 MD, Marcus Quinkler 9 MD, Francesco Fallo 10 MD, Claudio Letizia 11 MD, Bruno Allolio 8 MD, Giulio Ceolotto 4 PhD, Maria Verena Cicala 12 MD, Katharina Lang 8 MD, Hervé Lefebvre 13,14,15 MD, PhD, Livia Lenzini 4 PhD, Carmela Maniero 4 MD, Silvia Monticone 7 MD, Maelle Perrocheau 1,2 PhD, Catia Pilon 10 PhD, Pierre-François Plouin 1,2,6 MD, Nada Rayes 16 MD, Teresa M Seccia 4 MD, PhD, Franco Veglio 7 MD, Tracy Ann Williams 7 PhD, Laura Zinnamosca 11 PhD, Franco Mantero 12^ MD, Arndt Benecke 5,17^ PhD, Xavier Jeunemaitre 1,2,6^ MD, PhD, Martin Reincke 3^ MD and Maria-Christina Zennaro 1,2,6^# MD, PhD *, $,, ^ equal contributions # Corresponding author Affiliations: 1 INSERM, UMRS_970, Paris Cardiovascular Research Center, Paris, France 2 University Paris Descartes, Paris, France 3 Medizinische Klinik Innenstadt, Ludwig-Maximilians-University, Munich, Germany 4 DMCS G. Patrassi University of Padova Medical School, University Hospital, Padova, Italy 5 Institut des Hautes Etudes Scientifiques, & Centre National de la Recherche Scientifique, Bures sur Yvette, France 6 Assistance Publique-Hôpitaux de Paris, Hôpital Européen Georges Pompidou, Paris, France 7 Division of Internal Medicine and Hypertension, Department of Medicine and Experimental Oncology, University of Torino, Torino, Italy 8 Department of Medicine I, Endocrine and Diabetes Unit, University Hospital Würzburg, Würzburg, Germany 9 Clinical Endocrinology, Campus Mitte, University Hospital Charité, Berlin, Germany 10 Dept of Medical and Surgical Sciences, University of Padova, Padova, Italy 11 Department of Clinical Sciences, Secondary Hypertension Unit, University "Sapienza", Policlinico Umberto I, Viale del Policlinico 155, 00161, Rome, Italy 12 Endocrine Unit, Dept of Medical and Surgical Sciences, University of Padua,Italy 13 INSERM, U982, Mont-Saint Aignan, France 14 University of Rouen, Mont-Saint Aignan, France 15 University Hospital of Rouen, Rouen, France 16 Dept. of General, Visceral and Transplantation Surgery, Campus Virchow, University Hospital Charité, Berlin, Germany 17 Institut de Recherche Interdisciplinaire, CNRS USR3078, University of Lille I and Lille II, Villeneuve d Ascq, France Address correspondence to: Maria-Christina Zennaro, MD, PhD INSERM, U970 Paris Cardiovascular Research Center PARCC 56, rue Leblanc,

10 75015 Paris France Tel : +33 (0) Fax : + 33 (0) maria-christina.zennaro@inserm.fr

11 Supplementary Materials and Methods DNA and RNA isolation and RT-PCR Tumor DNA was extracted using QIAamp DNA midi kit (Qiagen, Courtaboeuf Cedex, France); DNA from peripheral blood leukocytes was prepared using salt-extraction. Total RNA was isolated from frozen tissue using Trizol (Invitrogen, Carlsbad, CA) and then purified on silica columns using RNeasy Mini Kit (Qiagen). The integrity and quality of the RNA were systematically checked using an Agilent 2100 bioanalyzer with the RNA6000 Nano Assay (Agilent Technologies, Santa Clara, CA). After DNaseI treatment (Invitrogen), 500 ng of total RNA were retro-transcribed using Superscript II RT (Invitrogen) and random hexamers (Promega, Madison, WI). KCNJ5 sequencing KCNJ5 cdna or DNA was amplified using intron-spanning primers as previously reported 1. PCR was performed on 100 ng of DNA (4 µl of RT product) in a final volume of 25 µl containing 0.75 mm MgCl 2 (without MgCl 2 for cdna), 400 nm of each primer, 200 µm deoxynucleotide triphosphate and 1.25 U Platinum Taq DNA Polymerase (Invitrogen). Cycling conditions were as previously described with an annealing temperature of 58 C (56 C for cdna). Direct sequencing of PCR products was performed using the ABI Prism Big Dye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA) on an ABI Prism 3700 DNA Analyzer (Applied Biosystems). Sequencing primers were as previously described 1. Microarray analysis Transcriptome analyses were performed on 102 samples collected through the COMETE network from patients operated on APA between 1994 and 2008 in the Hypertension Unit at the Hôpital Européen Georges Pompidou in Paris. Complementary crna was synthesized from 500 ng total RNA using the NanoAmpTM RT-IVT Labeling Kit (Applied Biosystems). Samples were hybridized to microarrays using Applied Biosystems chemiluminescence detection kit according to the manufacturer s protocol. Acquisition of the chemiluminescence images and primary data analysis were done on an AB1700 apparatus using the Applied Biosystems Expression Array System Software v Data were then median centred, log2- transformed and model-adjusted. Inter-array and inter-group normalizations were carried out using the NeONORM method as described previously 2-4. Multiple probes for a single gene, cross-reactivity of a single probe to several genes, as well as the resolution of probe- annotations were accounted for as defined previously 5. Hierarchical clustering based on Euclidean distance and complete-linkage was carried out using the R-packages hclust and gplot. Differentially expressed genes among groups were calculated using a fold-change z-test with positive false discovery rate correction. Integrated GO, KEGG and PANTHER annotations were used to assign each probe to the different ontologies (biological process, molecular function, signalling pathway) from the PANTHER Database 6. The relative representation of those probes was detected as significantly regulated as compared to a random set of probes drawn from the ensemble of available probes. P values for over- and under-representation of pathways were calculated using a binominal distribution and a Bonferroni correction for multiple testing.

12 Bibliography 1. Choi M, Scholl UI, Yue P, Bjorklund P, Zhao B, Nelson-Williams C, Ji W, Cho Y, Patel A, Men CJ, Lolis E, Wisgerhof MV, Geller DS, Mane S, Hellman P, Westin G, Akerstrom G, Wang W, Carling T, Lifton RP. K+ channel mutations in adrenal aldosterone-producing adenomas and hereditary hypertension. Science. 2011;331: Noth S, Brysbaert G, Benecke A. Normalization using weighted negative second order exponential error functions (NeONORM) provides robustness against asymmetries in comparative transcriptome profiles and avoids false calls. Genomics Proteomics Bioinformatics. 2006;4: Noth S, Brysbaert G, Pellay FX, Benecke A. High-sensitivity transcriptome profiles exhibit distinct signal distribution properties having significant implications for data analysis and biologic interpretation. Genomics Proteomics Bioinformatics. 2006;4: Wilhelm E, Kornete M, Targat B, Vigneault-Edwards J, Frontini M, Tora L, Benecke A, Bell B. TAF6delta orchestrates an apoptotic transcriptome profile and interacts functionally with p53. BMC Mol Biol. 2010;11: Brysbaert G, Pellay FX, Noth S, Benecke A. Quality assessment of transcriptome data using intrinsic statistical properties. Genomics Proteomics Bioinformatics. 2010;8: Mi H, Lazareva-Ulitsky B, Loo R, Kejariwal A, Vandergriff J, Rabkin S, Guo N, Muruganujan A, Doremieux O, Campbell MJ, Kitano H, Thomas PD. The PANTHER database of protein families, subfamilies, functions and pathways. Nucleic Acids Res. 2005;33:D

13 Supplementary Figures Figure S1: Mutations identified in patients with APA. A. Results of KCNJ5 sequencing on APA DNA showing the two missense mutations p.gly151arg and p.leu168arg found in patients described in this report. Two different nucleotide substitutions accounted for the p.gly151arg mutation, c.451g>a and c.451g>c.

14 Table S1. Clinical and biochemical data of patients included in this study. Variable (mean±sem) Preadrenalectomadrenalectomy Post- p-value Age, y 45.8±0.6 (377) 46.4±0.7 (261) n. d. Sex (male/female) 178/197 - n. d. SBP, mm Hg (n=314) 156± ±0.8 < DBP, mm Hg (n=314) 96±0.7 83±0.5 < Serum K + level, mmol/l (n=307) Active renin supine, mu/l * (n=105) 3.2± ±0.35 < (1.5;5.2) 11.3 (5.2; 20.3) < Renin activity ng*ml/h * (n=123) 0.28 (0.15;0.57) 1.40 (0.90;2.37) < Aldosterone supine, ng/l * (n=239) (241.2;570.7) (68.2;209.7) < Number of antihypertensive medications (n=292) 2.22± ±0.07 < SBP indicates systolic blood pressure; DBP, diastolic blood pressure. * median (interquartile range); n indicates the number of subjects for each group.

15 Table S2. Factors associated with the presence of KCNJ5 mutations. Variable Univariate analysis Multivariate analysis Model 1*, p< n = 319 Multivariate analysis Model 2*, p< n=319 n (n mut) OR [95%CI] p value OR [95%CI] p value n (n mut) OR [95%CI] p value Age at diagnosis of PA (years) [0.94 to 0.98] < [0.95 to 0.99] Age at diagnosis of PA (decades) < >70 28 (17) 89 (40) 125 (42) 80 (19) 43 (9) 9 (1) [0.22 to 1.26] 0.32 [0.14 to 0.73] 0.20 [0.08 to 0.50] 0.17 [0.06 to 0.49] 0.08 [0.01 to 0.74] (16) 81 (35) 113 (37) 63 (14) 29 (9) 8 (1) [0.20 to 1.45] 0.42 [0.16 to 1.11] 0.22 [0.07 to 0.63] 0.41 [0.12 to 1.37] 0.13 [0.01 to 1.32] Gender M F 178 (33) 197 (96) [2.61 to 6.68] < [2.26 to 6.33] < (30) 165 (82) [2.27 to 6.41] < Preoperative plasma aldosterone concentration (ng/l) [1.10 to 1.27] [1.07 to 2.28] [1.06 to 2.32] *Multivariable models include age at diagnosis (years, Model1) or age at diagnosis (decades, Model2), together with gender and preoperative plasma aldosterone as predictors. The number of KCNJ5 mutation carriers (n mut) is indicated when subcategories are defined. Odds ratios with 95% confidence interval (OR [95%CI]) are per unit increment in the natural log of plasma aldosterone.