Scholarly article on topic 'Glucocorticoid-Remediable Aldosteronism'

Glucocorticoid-Remediable Aldosteronism Academic research paper on "Biological sciences"

Share paper
Academic journal
The Endocrinologist
OECD Field of science

Academic research paper on topic "Glucocorticoid-Remediable Aldosteronism"

CME Review Article #18

0021-972X/2001/1104-0263 The Endocrinologist Copyright © 2001 by Lippincott Williams & Wilkins

CHIEF EDITOR'S NOTE: This article is the 18th of 36 that will be published in 2001 for which a total of up to 36 Category 1 CME credits can be earned. Instructions for how credits can be earned appear following the Table of Contents.

Glucocorticoid-Remediable Aldosteronism

Robert G. Dluhy, M.D.*

Glucocorticoid-remediable aldosteronism (GRA) represents a rare, heriditary form of primary aldos-teronism which is inherited in an autosomal dominant fashion. GRA is characterized by early onset of moderate-to-severe hypertension and suppressed plasma renin activity. The family history is often positive for a history of early hemorrhagic stroke. However, the clinical and biochemical features that define mineralcorticoid excess states, such as hy-pokalemia, are not consistently present in GRA. Accordingly, recognition of this syndrome can be difficult. In GRA, aldosterone secretion is solely regulated by adrenocorticotropin. As a result, the administration of exogenous glucocorticoids will suppress the hypothalamic-pituitary axis and suppress aldosterone levels, thereby relieving the min-

Learning Objectives:

• Identify the demographic, pathophysiologic, and genetic features of glucocorticoid-remediable aldosteronism (GRA), emphasizing those that distinguish it from other forms of aldosteronism.

• Recall the phenotypic expression of GRA and the best way to diagnose it.

• Describe and contrast options for treating GRA.

Historical Background

n 1966, Sutherland and Laidlaw described a syndrome of hypertension, hyperaldosteronism, and hypokalemia that was found in a family and was reversed by exogenous glucocorticoid therapy [1]. The disorder was called dexamethasone-

Professor of Medicine, Harvard Medical School, Associate Director, Endocrine Hypertension Division, Brigham and Women's Hospital, Boston, Massachusetts. Address correspondence to: Robert G. Dluhy, M.D., Brigham and Women's Hospital, 221 Longwood Avenue, Boston, MA 02115.

*The author has disclosed that he has no significant relationships with or financial interest in any commercial companies that pertain to this educational activity.

eralocorticoid-excess state. GRA is caused by a chimeric gene duplication that results from unequal crossing over between the highly homologous 11 p-hydroxylase (CYP11B1) and aldosterone synthase (CYP11B2) genes. The chimeric gene represents a fusion of the 5'adrenocorticotropin-responsive regulatory region of the 11p-hydroxylase gene and the 3' coding sequence of the aldosterone synthase gene. This results in ectopic expression of aldos-terone synthase activity in the zona fasciculata, the zone of the adrenal gland that normally secretes cortisol. This mutation explains the physiology and genetics of GRA and provides the basis for a simple direct genetic test for this disorder. ■

The Endocrinologist 2001; 11: 263-268

suppressible hyperaldosteronism (more recently renamed glucocorticoid-remediable aldosteronism (GRA), or familial hyperaldosteronism type I). GRA was found to have an autosomal dominant pattern of inheritance, and earlier physiologic studies revealed that plasma renin activity (PRA) was suppressed and aldosterone secretion was solely regulated by adrenocorticotrophic hormone (ACTH).


GRA is considered a rare cause of primary aldostero-nism with an estimated incidence of 1% to 3% of cases. However, the diagnosis of a new case of GRA can yield a large number of affected individuals in a family if the pedigree is expanded and at-risk individuals are tested.

Cases of GRA have been reported worldwide. Many affected families in North America are of Celtic ancestry; no known cases are reported among blacks. Unlike other etiologies of primary aldosteronism, which are usually diagnosed in the third to fifth decades of life, GRA is present from birth onward, occurring equally among males and fe-

males. Co-investigators at Harvard Medical School in Boston, Massachusetts and Yale University School of Medicine in New Haven, Connecticut who published the initial discovery of the mutation causing GRA (see below) have established an International Registry for GRA. Free genetic screening is provided (telephone number: 800-722-5520 ext. 28481 or and the clinical characteristics of affected pedigrees have been recorded in an effort to characterize the GRA pheno-type (see below).


Aldosterone production in normal subjects is regulated by the renin-angiotensin system and potassium. Thus, aldosterone secretion is normally positively regulated by angiotensin II (Ang II) and by potassium balance, with high levels stimulating and low levels reducing aldosterone secretion. In normal subjects, ACTH transiently stimulates aldosterone secretion but a continuous, prolonged ACTH infusion (over 24-48 hours) leads to a return of aldosterone levels to baseline. In contrast, in GRA aldosterone secretion is solely regulated by ACTH with GRA subjects showing an exaggerated response to infused ACTH and a failure to exhibit the expected normal decline after continuous ACTH administration [2]. In GRA, the renin-angiotensin system is suppressed and there is an absence of the normal potassium-induced increase in aldosterone secretion [3]. It is likely that this dysregulation of aldosterone secretion solely by a hormone (ACTH) that is not sensitive to sodium balance results in this mineralocorticoid excess state. As a corollary of the sole regulation of aldosterone by ACTH in GRA, the administration of exogenous gluco-corticoids will suppress the hypothalamic-pituitary axis, suppress aldosterone levels, reactivate suppressed PRA levels and reverse this mineralocorticoid excess state [1].

The adrenal cortex in GRA also produces large quantities of novel 18-oxygenated cortisol compounds (18-oxocortisol [18-OXO-F]; 18-hydroxycortisol [18-OH-F]) [4] (Fig. 1). These so-called "hybrid" steroids share structural features of both the cortisol-producing zona fascicu-lata and the aldosterone-producing zona glomerulosa (that is, these compounds are both 17 and 18 oxidized). GRA is easily distinguished from aldosterone-producing adenoma (APA), the only other condition in which there is overproduction of 18-OXO-F and 18-OH-F, because the levels of these compounds are 20 to 30 times higher than normal in GRA compared with only modest elevations in APA. It is not clear whether 18-OXO-F and 18-OH-F possess sodium-retaining properties and contribute to the pheno-typic variability of this mineralocorticoid excess state.

Figure 1. Steroidogenesis in the normal adrenal cortex and in the glu-cocorticoid-remediable aldosteronism (GRA) adrenal. A. The biosyn-thetic pathway of aldosterone in the normal zona glomerulosa. B. In the normal zona fasciculata cortisol is the normal end product; in the GRA adrenal cortisol is the further 18-oxygenated to the compounds, 18-hydroxycortisol (18-OH-F) and 18-oxocortisol (18-OXO-F). These compounds are a unique biochemical phenotype to diagnose GRA. (Reprinted with permission: Rich GM, Ulick S, Cook S, et al.: Glucocorti-coid-remediable aldosteronism in a large kindred: clinical spectrum and diagnosis using a characteristic biochemical phenotype. Ann Intern Med 1992; 116: 813-20.)

Clinical Phenotype

Blood Pressure

New index cases of GRA continue to be discovered, most commonly in hypertensive children. As a result, a possible diagnosis of GRA should be considered in all hypertensive children, especially those with suppressed PRA. A recent study of 20 children with GRA underscores the importance of considering the diagnosis of GRA in the setting of severe hypertension [5]. Fifty percent (8 of 16) of the GRA children in this study were classified as having severe hypertension (>99th centile for age and sex) (Fig. 2); the presence of severe diastolic and systolic hypertension together particularly called for an evaluation for GRA. The diagnosis of hypertension in childhood or adolescence is often delayed, which may reflect the failure of clinicians to appreciate the normative blood pressure levels in these patient groups (blood pressure classification in children and adolescents requires reference to sex- and age-specific blood pressure centile nomograms) [6]. Thus the blood pressure levels in GRA-affected children are typically lower than values used to define hypertension in adults, but are usually higher than age- or sex-matched controls.

Hypertension associated with GRA across all age groups is often refractory to conventional antihyperten-sive agents. In addition, the blood pressure in GRA-affected subjects within and between pedigrees is often highly variable; some affected individuals are normoten-sive, whereas others have only mild hypertension. This variability in blood pressure levels in GRA may relate to

Figure 2. Blood pressure centiles for females (0-13 years) and blood pressure levels in GRA-positive hypertensive females (n = 9). Centile = percentile for age, sex, and height; SBP = systolic blood pressure; DBP = diastolic blood pressure; yrs. = years.

either other hereditary factors that regulate blood pressure or variation in dietary sodium intake. Thus the family history in GRA does not invariably reveal a history of severe hypertension in first-degree relatives of affected subjects.

Potassium Levels

Although GRA is a mineralocorticoid excess state associated with hyperaldosteronism and suppressed PRA, normokalemia is the rule in affected individuals unless potassium-wasting diuretics, such as hydrochlorothiazide, are administered [7]. The mechanism(s) for the absence of hypokalemia in GRA is unknown, but there does not appear to be impairment in renal responsiveness to potassium loading or mineralocorticoid treatment [3]. Alternatively, the hyperaldosteronism in GRA is mild, with random al-dosterone levels usually in the normal or high-normal range, albeit inappropriate for the suppressed levels of PRA. There is also a distinct plasma aldosterone (PA) diurnal variation in GRA, with low levels during the evening hours because of the sole regulation of aldosterone by ACTH. Both factors may contribute to a mild miner-alocorticoid excess state in GRA with minimal/modest potassium-wasting compared with that seen in APA.

Hemorrhagic Stroke

A retrospective review of 27 pedigrees with genetically proven GRA has documented an increased prevalence of early cerebrovascular complications, primarily cerebral hemorrhage, which is associated with high mortality (61%) [8]. In this study, cerebrovascular complications were present in 48% of all GRA pedigrees and 18% of all GRA patients, and the mean age at the time of stroke was 32 years. As a result, a prominent history of early he-morrhagic stroke in a family is also a clue to the diagnosis of GRA. The underlying mechanism of these cerebral hemorrhages in many patients was documented to be secondary to intracranial aneurysm. This report recommended screening of asymptomatic GRA patients with magnetic resonance angiography, beginning at puberty and every 5 years thereafter (as in adult polycystic kidney disease where there is a similar frequency of aneurysm) [8].


GRA is inherited as an autosomal dominant trait that follows classic Mendelian genetics. Lifton et al. discovered the genetic basis of GRA in 1992 [9]. In a large pedigree, GRA-affected subjects were shown to possess a chim-eric gene duplication that resulted from unequal crossing over between the highly homologous 1 ^-hydroxylase (CYP11B1) and aldosterone synthase (CYP11B2) genes (Fig. 3). These genes are located in close proximity to each other on chromosome 8 and share more than 90% base-pair homology. The chimeric gene represents a fusion of the 5' adrenocorticotropin-responsive regulatory region of 11^-hydroxylase gene (expressed normally in the cortisol-

Figure 3. The chimeric gene duplication in GRA resulting from unequal crossing over between the homologous aldosterone synthase and 11 ß-hy-droxylase genes that are located in close proximity on chromosome 8. This chimeric gene fuses the 5' regulatory sequences of the 11ß-hydroxylase gene and the 3' coding sequences of the aldosterone synthase gene (depicted as occurring in the intron between exons 3 and 4). (Reprinted with permission: Lifton RP, Dluhy RG, Powers M, et al.: A chimeric 11-hy-droxylase/aldosterone synthase gene causes glucocorticoid-remediable aldosteronism and human hypertension. Nature 1992; 355: 262-5.)

producing zona fasciculata) and the 3' coding sequences of the aldosterone synthase gene. The result of this gene duplication is ectopic expression of aldosterone synthase activity in the cortisol-producing zona fasciculata (Fig. 4). As a result, aldosterone and the novel steroids 18-OH-F and 18-OXO-F are all produced ectopically in the zona fas-ciculata under the regulation of ACTH from cortisol and steroid precursors.

In 11 additional GRA pedigrees, all affected subjects proved to have chimeric gene duplications [10]. In these subjects, a minimum of eight independent mutations with five different sites of crossing over were identified, indicating that these mutations arose independently and not from a single ancestral mutation. Confirmation of the presence of such chimeric gene mutations has also been reported in another study of four additional patients from unrelated GRA pedigrees [11].

DNA sequence analyses of the duplicated genes from these unrelated pedigrees indicate that the sites of fusion are variable, but in all cases are upstream of exon 5, suggesting that encoded amino acids in exon 5 are essential for aldos-terone synthase enzymatic function [10, 11]. Similarly, by expression of various chimeric genes constructed in vitro, Pascoe et al. have demonstrated that when these genes are fused after exon 5, aldosterone synthase enzymatic activity is undetectable [11]. Transfection studies with cDNA encoding hybrids between the highly homologous CYP11B1 (11P-hydroxylase) and CYP11B2 (aldosterone synthase) enzymes have also shown that two amino acid changes, Ser288Gly and Val320Ala, are sufficient to convert CYP11B1 into an efficient aldosterone-producing enzyme [12]. These results suggest that a gene conversion involving exons 5 and 6, in which these residues are encoded, could cause a novel form of GRA. However, to date no conversions of the CYP11B1

Figure 4. Left, normal adrenal gland. Aldosterone is produced in the zona glomerulosa under the regulation of angiotensin II and potassium; the production of cortisol in the zona fasciculata is regulated by adreno-corticotropic hormone (ACTH). Aldosterone synthase activity is expressed only in the zona glomerulosa. Right, GRA adrenal. Ectopic expression of aldosterone synthase activity in the zona fasciculata results in production of aldosterone and the novel steroids 18-OH-F and 18-OXO-F under the regulation of ACTH. (Reprinted with permission: Lifton RP, Dluhy RG, Powers M, et al.: Hereditary hypertension caused by chimaeric gene duplications and ectopic expression of aldosterone synthase. Nat Genet 1993; 2: 66-74.)

gene expected to cause GRA (involving exons 4,5, and 6 from CYP11B2) have been found in a sample of low renin hypertensive patients, patients with aldosteronism, and 90 normotensive individuals [13].

Diagnosing GRA

Importance of the Medical and Family Histories

Screening is obviously indicated for all at-risk individuals in diagnosed GRA pedigrees. Diagnosis of GRA without knowledge of the presence of the syndrome in a family member can be difficult, due to the variable GRA phenotype (see above). For example, affected subjects are usually normokalemic and their blood pressures can range from normotension to severe hypertension. Nevertheless, the medical history is often the cornerstone for making the diagnosis of GRA. The diagnosis of GRA should strongly be considered in a patient with hypertension of early onset, especially in children or an individual with a family history of juvenile hypertension in first-degree relatives. Many patients are refractory to conventional antihyper-tensive agents or have a tendency to become hypokalemic if potassium-wasting diuretics are administered. There often is a prominent family history of early mortality or morbidity from hemorrhagic stroke [8].

Biochemical Assessment

PRA will be suppressed in GRA patients unless aldos-terone antagonists have been used as therapeutic agents. Therefore, nonsuppressed PRA levels in the absence of such therapy strongly argues against a diagnosis of GRA. However, the presence of a suppressed PRA level is nonspecific, with up to 20% of adult essential hypertensives having a PRA level <2 ng/mL/hour. GRA patients also have abnormal PA/PRA ratios (>30). Aldosterone blood levels and urinary aldosterone excretion rates are usually normal or mildly elevated but this is abnormal in the setting of suppressed levels of PRA. Serum potassium is usually normal and accordingly hypokalemia lacks sensitivity as a screening test for GRA [7].

The diagnosis of GRA is supported by dexamethasone suppression testing (DST) (0.5 mg every 6 hours during 2 days). A fall in aldosterone to nearly undetectable levels after DST in GRA is expected and reflects the sole control of aldosterone by ACTH in this disorder [1]. As a result, a post-DST PA below 4ng/dL will correctly diagnose GRA patients with high sensitivity and specificity [14]. Significant suppression of aldosterone levels after DST also occurs in APA patients reflecting the well-recognized regulation of aldosterone by ACTH in this disorder. However, au-

tonomous production of aldosterone in APA accounts for the failure of aldosterone levels to fall to very low or nearly undetectable levels.

A Unique Biochemical Phenotype

Urinary excretion of the 18-oxygenated cortisol compounds 18-OH-F and 18-OXO-F are markedly elevated (up to 30 times normal levels) and provide a specific means to diagnose GRA (see above) [4]. However, measurement of urinary 18-OH-F and 18-OXO-F is difficult and not usually available.

Genetic Testing

Biochemical testing, such as DST and 24-hour urine collections, are often inconvenient and especially difficult to perform in pediatric patients. As a result genetic testing is preferred and is 100% sensitive and specific to diagnose GRA, requiring only a single blood collection for leukocyte DNA assessment. After DNA extraction, the hybrid or chimeric gene can be detected by the Southern blot approach (developed by Lifton et al. [9]) or more recently by the long-polymerase-chain-reaction-based approach [15]. The advantage of the "long-polymerase chain reaction" method is that it is considerably faster and cheaper than Southern blotting.


The hypertension in GRA is often refractory to conventional antihypertensive therapy with most patients inadequately controlled by maximal doses of two or three antihypertensive agents. Moreover, the gratifying reduction in blood pressure in response to directed monotherapy underscores the importance of diagnosing GRA [5].

Glucocorticoid Suppression

Traditionally, suppression of adrenocorticotropin with dexamethasone has been used to treat the hypertension associated with GRA [1]. First, suppression of the hypothal-amic-pituitary-adrenal axis does not always result in normalization of blood pressure in GRA; this may relate to end-organ injury (e.g., renal nephrosclerosis), concomitant essential hypertension or, rarely, autonomous production of aldosterone in patients with longstanding GRA. A more important issue is potential toxicity (Cushing syndrome) associated with excessive glucocorticoid dosing, especially in children. When a decision to use glucocorticoids is made, the smallest effective dose should be used; suppressive steroid dosing should also be determined relative to body surface area (8 to 10 mg of hydrocortisone/ m2/day, or bio-

equivalent dosing with a Cortisol analogue). Target blood pressure in children should be guided by age-specific blood pressure percentiles (Fig. 2) [6]. Children should be followed by pediatricians with careful attention paid to linear growth to detect any diminution as a result of overtreatment.

Rarely glucocorticoid-treated patients have exhibited symptoms of hypoaldosteronism with salt wasting, hypotension, and hyperkalemia soon after such treatment is instituted. Hypoaldosteronism occurs because ectopic aldosterone production in the zona fasciculata falls to nearly undetectable levels while the zona glomerulosa is hypo-functional as a result of the prior suppression of the renin-angiotensin system.

Mineralocorticoid Antagonists

Spironolactone, a competitive antagonist of the miner-alocorticoid receptor that is usually the agent of first choice in the medical treatment of the multiple etiologies of primary aldosteronism, is also effective in treating GRA patients. Adult GRA patients should be started at 50 mg twice daily with meals, with subsequent upward titration as tolerated until blood pressure is controlled. Potassium-wasting diuretics (hydrochlorothiazide 12.5 to 25 mg or furosemide 40 to 80 mg/day) may be added in an attempt to further achieve sodium depletion; close monitoring of serum potassium is important if these diuretics are used. Because spironolactone binds to androgen receptors and also blocks testosterone biosynthesis, erectile dysfunction, decreased libido, and gy-necomastia are side effects in men. Menstrual irregularities are seen in women because spironolactone also binds to progesterone receptors. A promising new agent, epleronone, a selective inhibitor of the mineralocorticoid receptor that does not bind to androgen or progesterone receptors, may ultimately be the agent of choice to treat GRA patients.

Amiloride blocks the aldosterone-regulated epithelial sodium channel in the distal nephron and is an alternative to spironolactone treatment in GRA. Divided dosing should be used starting at 5 mg twice daily with a maximum dose of 15 mg twice daily. Triamterene also inhibits sodium-potassium exchange in the distal nephron by blocking sodium conductance channels. Like amiloride, its effects are independent of aldosterone receptor blockade. A divided dosing regiment should be used with maximum doses of 300 mg/day. Adverse effects of this drug are uncommon but include rash, weakness, fatigue, and abnormalities of liver enzymes.

Dihydropyridine Calcium Channel Blockers

The dihydropyridine calcium channel blockers, such as the extended-release formulation of nifedipine and amlodip-ine, have been advocated in the medical management of pri-

mary aldosteronism because this class of medication has been shown to inhibit aldosterone biosynthesis in vitro. However, the aldosterone reduction in response to these agents in vivo in various etiologies of primary aldosteronism has generally been disappointing. The antihypertensive response to di-hydropyridine calcium channel blockers, however, can be gratifying in many patients with primary aldosteronism, including GRA (R.G.D., unpublished data). However, these medications should be viewed as second-line agents.

A sodium-restricted diet (<2g/day) is also recommended in conjunction with pharmacologic treatment because it will minimize potassium wasting and may lower blood pressure. To date, no randomized studies have been performed that compare various treatment regimens in


1. Sutherland DJ, Ruse JL, Laidlaw JC: Hypertension, increase aldosterone secretion and low plasma renin activity relieved by dexam-ethasone. Can Med Assoc J 1996; 95: 1109-19.

2. Ganguly A, Weinberger MH, Guthrie GP, Fineberg NS: Adrenal steroid responses to ACTH in glucocorticoid-suppressible aldostero-nism. Hypertension 1984; 6: 563-7.

3. Litchfield WR, Coolidge C, Silva P, et al.: Impaired potassium-stimulated aldosterone production: a possible explanation for nor-mokalemic glucocorticoid-remediable aldosteronism. J Clin Endo-crinol Metab 1997; 82: 1507-10.

4. Gomez-Sanchez CE, Montgomery M, Ganguly A, et al.: Elevated urinary excretion of 18-oxo-cortisol in gluccocorticoid-remediable

aldosteronism. J Clin Endocrinol Metab 1984; 59: 1022-4.

5. Dluhy RG, Anderson BF, Harlin B, Ingelfinger J, Lifton RP: Gluco-

corticoid-remediable aldosteronism (GRA) is associated with severe hypertension in early childhood. J Pediatr 2001; 138: 715-20.

6. Lieberman E: Pediatric hypertension: clinical perspective. Mayo

Clin Proc 1994; 69: 1098-107.

7. Rich GM, Ulick S, Cook S, Wang JZ, Lifton RP, Dluhy RG: Glu-cocorticoid-remediable aldosteronism in a large kindred: clinical spectrum and diagnosis using a characteristic biochemical pheno-type. Ann Intern Med 1992; 116: 813-20.

8. Litchfield WR, Anderson BF, Weiss RJ, Lifton RP, Dluhy RG: In-tracranial aneurysm and hemorrhagic stroke in glucocorticoid-re-mediable aldosteronism. Hypertension 1998; 31: 445-50.

9. Lifton RP, Dluhy RG, Powers M, et al.: A chimeric 11 beta-hy-droxylase/aldosterone synthase gene causes glucocorticoid-remedi-able aldosteronism and human hypertension. Nature 1992; 355: 262-5.

10. Lifton RP, Dluhy RG, Powers M, et al.: Hereditary hypertension caused by chimeric gene duplications and ectopic expression of al-dosterone synthase. Nat Genet 1993; 2: 66-74.

11. Pascoe L, Curnow KM, Slutsker L, et al.: Glucocorticoid-suppress-ible hyperaldosteronism results from hybrid genes created by unequal crossovers between CYP11B1 and CYP11B2. Proc Natl Acad Sci U S A 1992; 89: 8327-31.

12. Curnow KM, Mulatero P, Emeric-Blanchouin N, Aupetit-Faisant B, Corvol P, Pascoe L: The amino acid substitutions Ser288Gly and Val320Ala convert the cortisol producing enzyme, CYP11B1, into an aldosterone producing enzyme. Nat Struct Biol 1997; 4: 32-5.

13. Mulatero P, Curnow KM, Aupetit-Faisant B, et al.: Recombinant CYP11B genes encode enzymes that can catalyze conversion of 11-deoxycortisol to cortisol, 18-Hydroxycortisol, and 18-Oxocortisol. J Clin Endocrinol Metab 1998; 83: 3996-4001.

14. Litchfield WR, New MI, Coolidge C, Lifton RP, Dluhy RG: Evaluation of the dexamethasone suppression test for the diagnosis of glu-cocorticoid-remediable aldosteronism. J Clin Endocrinol Metab

1997; 82: 3570-3.

15. Stowasser M, Gartside MG, Gordon RD: A PCR-based method of screening individuals of all ages, from neonates to the elderly, for familial hyperaldosteronism type I. Aust N Z J Med 1997; 27: 685-90.