5.6 Primary aldosteronism and other steroid-related causes of endocrine hypertension

Mineralocorticoid hypertension is characterized by increased distal renal tubular sodium reabsorption, raised body sodium content, plasma volume expansion, markedly reduced body potassium content, with a metabolic alkalosis and suppression of renin production by the juxtaglomerular cells of the kidney (and correspondingly low levels of angiotensin II). Primary aldosteronism is the most common cause of mineralocorticoid hypertension (1); less frequent causes include the rare inborn errors of adrenal steroid synthesis (11β-hydroxylase and 17α-hydroxylase deficiency), alterations in corticosteroid metabolism (syndrome of apparent mineralocorticoid excess), and constitutive activation of the epithelial sodium channel (Liddle’s syndrome).

Aldosterone is produced from cholesterol in a series of biochemical reactions which involve sequential hydroxylation and dehydrogenation reactions (Fig. 5.6.1). These reactions are performed in cells of the zona glomerulosa of the adrenal cortex. The unique steps in the formation of aldosterone are an 18-hydroxylation of corticosterone to form 18-hydroxycorticosterone which, following a second hydroxylation reaction, is spontaneously dehydrated to form aldosterone. Both of these steps are carried out by a single enzyme, aldosterone synthase, encoded by the gene CYP11B2 (2). This enzyme also converts 11-deoxycorticosterone, the preferred substrate, to corticosterone in zona glomerulosa cells.

CYP11B2 is highly homologous to the gene that encodes 11β-hydroxylase (CYP11B1) (3); in the zona fasciculata, this enzyme converts 11-deoxycortisol to cortisol and 11-deoxycorticosterone to corticosterone. It can also catalyse 18-hydroxylation, thus forming 18-hydroxy-deoxycorticosterone and 18-hydroxycorticosterone in this zone. In normal circumstances, zonation of the adrenal cortex results in distinct separation of functions, so that 17α-hydroxysteroids such as 11-deoxycortisol and cortisol are not available to act as substrates for aldosterone synthase. However, in some circumstances, loss of strict zonation does occur, in which case cortisol becomes available as a substrate, resulting in the production of large quantities of steroids such as 18-hydroxycortisol and 18-oxocortisol, which are normally minor products of 11β-hydroxylase (4). The clearest example of this occurs in glucocorticoid-remediable aldosteronism, where aldosterone synthase is expressed at a high level throughout the zona fasciculata (see Chapter 5.1) (5). In Conn’s adenomas, there is also loss of strict zonal separation of hydroxylase activity, so that aldosterone synthase can convert cortisol to 18-hydroxy- and 18-oxocortisol, which can be detected in both urine and plasma (6). It is unlikely that either of these steroids has a significant effect on the clinical presentation of Conn’s syndrome.

Normally, aldosterone secretion is regulated by changes in body sodium status, through the renin/ angiotensin system (7). Loss of sodium stimulates this system, whereas high intake suppresses it. Potassium is also a powerful direct stimulus to aldosterone secretion; very small increments, which do not alter plasma levels perceptibly, raise aldosterone secretion rate. In normal subjects, ACTH can acutely stimulate aldosterone release, but chronic pharmacological doses of ACTH given over several days results in suppression of aldosterone synthesis by mechanisms that remain incompletely understood. Importantly, electrolyte status affects the sensitivity of the zona glomerulosa to all agonists. Thus, sodium depletion enhances sensitivity to angiotensin II, potassium, and ACTH, while sodium loading has the opposite effect. The converse is true for potassium status, so that the sensitivity of aldosterone to angiotensin II stimulation is set by the prevailing potassium levels.

In primary aldosteronism, aldosterone levels are, by definition, inappropriate for the prevailing angiotensin II and potassium levels. The volume expansion and high extracellular sodium concentration caused by aldosterone suppress release of renin from the juxtaglomerular cells of the kidney; as renin activity determines angiotensin II production, levels of the peptide are low. Furthermore, the aldosterone excess leads to increased potassium loss in the urine, resulting in hypokalaemia. Despite this, aldosterone secretion remains higher than normal. However, but for the hypokalaemia, aldosterone levels would be higher still (8, 9).

In normal subjects, aldosterone levels show diurnal variation, upon which are superimposed fluctuations entrained by acute posture-induced changes in renin release. Levels of the steroid are highest in the morning and have a nadir, which mirrors the pattern of plasma cortisol. In the majority of patients with primary aldosteronism, suppression of renin leads to loss of posture-induced changes. Indeed, paradoxically, concentrations may fall on assuming an upright posture, while the diurnal variation in aldosterone concentration is blunted but not completely absent (10). Thus, aldosterone levels tend to be highest after overnight recumbency and are lower in the later part of the day. In patients with an aldosterone-producing adenoma, aldosterone will respond acutely to administration of ACTH (indeed, the aldosterone response to ACTH may be greater than in normal subjects) but usually shows no response to administration of angiotensin II, which is, at least in part, due to the increased total body sodium content. However, in patients with bilateral adrenal hyperplasia and in a subgroup of subjects with aldosterone-producing adenomas, aldosterone does respond acutely to administration of angiotensin II (11). These patients may also show a small aldosterone rise after ambulation (reflecting their angiotensin II responsiveness). However, apart from the fact that patients with typical solitary Conn’s adenomas are less likely to show responsiveness to angiotensin II than those with bilateral adrenal hyperplasia, there are no real practical advantages arising from this observation.

In normal subjects, a variety of other amines and peptides are reported to influence aldosterone secretion (12). These are summarized in Table 5.6.1. Of these, atrial natriuretic peptide may be one of the more important mechanisms that inhibits aldosterone production by the adrenal cortex. In patients with primary aldosteronism, volume expansion results in increased secretion of atrial natriuretic peptide from the heart. However, despite this, aldosterone secretion remains elevated and patients with primary aldosteronism are reported not to respond to infusion of exogenous atrial natriuretic peptide (13). In some subjects with aldosterone-producing adenomas, other hormones, including gonadotrophins and gastrointestinal-derived peptides such as GIP-1, have, rarely, been reported to regulate aldosterone production (14). In these cases, expression of receptors for the peptides has been demonstrated in the tumour material. However, this appears to be a relatively unusual circumstance.

Production of 18-hydroxy- and 18-oxocortisol is modestly increased in patients with Conn’s adenomas (in contrast with patients with glucocorticoid-remediable aldosteronism, where levels are considerably higher). In addition to excessive aldosterone production, patients with primary aldosteronism often have increased plasma levels of the immediate precursor, 18-hydroxycorticosterone (15). It is possible that hypokalaemia reduces the efficiency of its conversion to aldosterone. Levels of this hormone are greater in patients with Conn’s adenomas than those with bilateral adrenal hyperplasia, but the observation, in itself, is of little practical importance in patient investigation or management. Furthermore, there is no good evidence that steroids other than aldosterone contribute in a major way to the clinical and pathophysiological features of primary aldosteronism.

Some patients with adrenocortical carcinomas produce excessive 11-deoxycorticosterone (16), which, in marked excess, can also cause mineralocorticoid hypertension. In these subjects, a range of other steroids is often present and diagnosis is usually obvious by imaging and measurement of plasma and urinary steroid concentrations.

The mineralocorticoid receptor mediates the classical effects of aldosterone, acting as a ligand-activated transcription factor (17). The receptor is found in the cytosol of epithelial cells, particularly in the renal collecting duct; other major target sites include the colon and the salivary gland. However, mineralocorticoid receptors have also been identified in nonepithelial sites such as heart, brain, vascular smooth muscle, liver, and peripheral blood leucocytes (18). The mineralocorticoid receptor belongs to the nuclear receptor superfamily of proteins and consists of an N-terminal domain, a DNA-binding domain, and a C-terminal ligand-binding domain (19). Aldosterone binds to this latter domain and causes a conformational change to the mineralocorticoid receptor, whereupon it dissociates from various heat-shock proteins and immunophilins, and translocates to the cell nucleus where it binds as a homodimer to the hormone response element of aldosterone-responsive genes in order to activate or repress gene transcription (the classical genomic effect of aldosterone).

The most important physiological action of aldosterone is to increase the reabsorption of sodium in the kidney and other epithelial sites at the expense of potassium and hydrogen ions (20). The cortical collecting tubules and the distal convoluted tubule are the principal sites of aldosterone-mediated sodium and potassium transport. The major determinant of renal sodium reabsorption is the epithelial sodium channel (ENaC) located on the apical membrane of the distal convoluted tubule (21). Its availability in open conformation at the apical membrane of the cell is increased by aldosterone, vasopressin, glucocorticoids, and insulin whilst elevated intracellular levels of calcium and sodium lead to its down-regulation (22).

Aldosterone induces the expression of the ENaC’s α-, β-, and γ-subunits although its major effect appears to be achieved either by increasing the number of channels in the plasma membrane or by increasing the probability that the channels are open to allow the passage of Na⁺. This regulation of ENaC is achieved via the expression of a wide range of aldosterone-induced proteins, some of which appear to act by preventing tonic inhibition of ENaC activity. The best-characterized of these proteins is the serine–threonine kinase, SGK1 (23). Aldosterone causes phosphorylation and activation of SGK1, which in turn increases ENaC activity by an increase in the number of channels at the cell surface (Fig. 5.6.2). The principal ENaC inhibitory accessory protein is Nedd4 (neuronal precursor cells expressed developmentally down-regulated). This ubiquitin protein ligase binds to the C-terminal regions of β- and γ-subunits of ENaC, leading to channel internalization and degradation. It has been demonstrated that the stimulatory action of SGK on ENaC is mediated through phosphorylation of serine residues on Nedd4. Such phosphorylation reduces the interaction between Nedd4 and ENaC, leading to elevated ENaC cell surface expression (24, 25).

Hydrogen ion excretion by the kidney in the distal nephron is also regulated by aldosterone. Hydrogen ion secretion is through a sodium-insensitive route, since it occurs principally in the intercalated cells of the collecting tubule. This segment of the nephron exhibits little or no aldosterone-induced sodium transport, and so aldosterone-induced natriuresis and hydrogen ion secretion appear to be independent events. This effect is mediated via an effect of aldosterone on the activity of the ATP-dependent apical hydrogen ion pump and parallel regulation of the basolateral membrane Cl⁻/HCO₃− exchanger (26).

The net effect of aldosterone on the renal tubule is therefore to promote sodium retention at the expense of potassium and also to promote hydrogen ion excretion by the kidney. This explains the clinical features observed in cases of primary aldosterone excess, i.e. plasma hypokalaemia, alkalosis, a raised exchangeable sodium content, and low total body potassium.

Aldosterone and cortisol have equal affinities for the mineralocorticoid receptor; given that cortisol is found at much higher levels in plasma (up to 1000-fold in comparison to aldosterone levels) the vast majority of these receptors would be expected to be transactivated by glucocorticoid, particularly at times when cortisol levels are highest. This is clearly not the case in vivo, due in part, at least, to the activity of the enzyme 11β-hydroxysteroid dehydrogenase (11β-HSD), which acts as a gatekeeper to prevent activation of the mineralocorticoid receptor by much higher available levels of cortisol (Fig. 5.6.3) (27). The type 2 isoform of this enzyme is found in the renal distal nephron as well as other aldosterone-sensitive target tissues (colon, salivary glands, and placenta) and converts cortisol to its inactive metabolite, cortisone, which has does not transactivate the mineralocorticoid receptor (28). Whenever 11β-HSD2 activity is absent, inhibited, or overwhelmed, this ‘protective’ mechanism is lost and cortisol gains access to the mineralocorticoid receptor to act as a potent mineralocorticoid. However, this may be an oversimplification, as it has been pointed out that the capacity of the enzyme is insufficient to lead complete conversion of cortisol to cortisone, and it appears that cortisol may still be able to bind the mineralocorticoid receptor but fail to transactivate it; the ability of cortisol to act as an agonist may depend on other factors, including availability of NADH, generated as a consequence of the activity of 11β-HSD2, and local (intracellular) redox state (29). In particular, it has been proposed that NADH acts as a corepressor of mineralocorticoid receptor activation when cortisol is a ligand, and may provide an alternate (or additional) way in which reduced 11β-HSD2 activity permits cortisol to transactivate the receptor. The clinical consequences of this are described later in this chapter.

It is now accepted that, as well as classical genomic effects through ligand–receptor binding of DNA regulatory elements, aldosterone also exerts rapid, nongenomic, effects (30). Nongenomic effects are associated with rapid activation (occurring within minutes), in the absence of a need for transcription or protein synthesis (31). Because of the brief response time, it is presumed that nongenomic actions are initiated at the membrane level, and membrane signalling transduction pathways have been intensively studied. These suggest the existence of novel steroid hormone receptors or possibly classical receptors embedded in the membrane that initiate the nongenomic signal cascade, although none have yet been found. However, two recent studies have demonstrated a role for intracellular calcium as well as protein kinase C activity as a potential mechanism of action of the nongenomic receptor (31, 32). In addition, several of these rapid responses can be blocked by the mineralocorticoid receptor antagonist spironolactone, suggesting that at least some nongenomic effects may be mediated via the classic mineralocorticoid receptor. Reports of rapid, nongenomic effects of aldosterone have been described in smooth muscle, cardiac muscle, skeletal muscle, colonic epithelial cells, and myocardial cells (33). These effects have been linked to the development of increased systemic vascular resistance and so could, theoretically, contribute to hypertension and cardiovascular disease.

Classical mineralocorticoid receptors have been localized in a number of nonepithelial tissues, particularly in the cardiovascular system and central nervous system. While the functional properties of the receptors in these tissues are largely similar (in terms of transactivation and downstream signalling), the effects they mediate are extremely diverse. In contrast to its established effects on electrolyte balance in epithelial tissue, aldosterone in the cardiovascular system promotes cardiac hypertrophy, fibrosis, and abnormal vascular endothelial function. In the central nervous system, mineralocorticoid receptor activation appears to regulate blood pressure, salt appetite, and sympathetic tone. In contrast to epithelial tissues, mineralocorticoid receptors in the central nervous system do not appear to colocalize with 11β-HSD2 (34). The lack of 11β-HSD2 in mineralocorticoid receptor-rich areas suggests that the majority of brain receptors are likely to be occupied by glucocorticoid although infusion of aldosterone intracerebroventricularly raises blood pressure in experimental circumstances, suggesting that the hormone may have central actions to influence cardiovascular function (35, 36).

Primary aldosteronism is the most common cause of secondary hypertension (1, 37). By definition, the syndrome is a consequence of excessive autonomous aldosterone production. Solitary benign adenomas of the adrenal cortex (Conn’s adenomas; APA) account for approximately 40% of presentations with primary aldosteronism; bilateral adrenal hyperplasia, in which several autonomous nodules are present throughout the adrenal cortex, is the most common cause, accounting for around 60%. A very small number of patients with primary aldosteronism have an inherited form (glucocorticoid-remediable aldosteronism) due to the presence of a chimeric gene, expression of which in the adrenal cortex is regulated by ACTH but which encodes aldosterone synthase (5); this is discussed in Chapter 5.7. Very rarely, primary aldosteronism can be due to carcinoma of the adrenal cortex and this is discussed below.

In the years following the first description by Conn in 1955, the frequency of primary aldosteronism was thought to be low. This probably reflected the reliance on hypokalaemia as a diagnostic pointer in hypertensive patients. More recently, however, sensitive screening tests have been used to detect primary aldosteronism, principally based on the ratio of aldosterone to renin (ARR) (37–39). The widespread introduction of these tests has undoubtedly led to a rise in the detection rate for primary aldosteronism, worldwide. Interestingly, this has resulted in a change in the pattern of disease seen; in the Mayo clinic, earlier experience was that the majority of patients with primary aldosteronism had an APA, while more recent series have clearly shown that the majority of patients with primary aldosteronism have bilateral adrenal hyperplasia (40). Despite this, the true prevalence of primary aldosteronism remains unclear, probably due to differences in definition and source population. A range of prevalence figures for primary aldosteronism has been reported, with some groups suggesting figures as high as 12% (41). Even higher rates (up to 20%) have been reported in populations of patients with resistant hypertension (42), although such studies rarely categorize the type of primary aldosteronism being found. However, in one large series of patients with hypertension (3900 patients), who were very thoroughly screened for the condition with a series of measurements, including urinary corticosteroid levels as well as plasma measurements of aldosterone and renin, a prevalence of 6.5% was reported, with only half these subjects (3.7%) harbouring adrenal adenomas (43). Another study in Italy, using careful confirmatory tests for the disorder, reported a prevalence of 11% for primary aldosteronism in an unselected hypertensive cohort, with 4.8% of subjects harbouring an aldosterone-producing adenoma (44), while a comprehensive study in Greece of patients with resistant hypertension reported a figure of 11.3% (45). This figure is in keeping with other series and is probably a realistic estimate of the frequency of primary aldosteronism in a hypertensive population.

The frequency of Conn’s adenomas is slightly greater in female patients; the age at diagnosis of patients with adenomas is less than that for patients with bilateral adrenal hyperplasia. For patients with adenomas, most series report that tumours occur more commonly on the left-hand side.

Bilateral adrenal hyperplasia more frequently affects older patients. It has been suggested that this syndrome is part of the spectrum of low renin essential hypertension and does not, in itself, constitute a distinct diagnostic entity (46). Thus, post mortem series of patients with essential hypertension report increased adrenal nodularity and hyperplasia (discussed above), and it is unclear whether patients with low renin essential hypertension differ in any substantial way from those diagnosed as having bilateral adrenal hyperplasia. Indeed, the distinction may be artificial and a consequence of rather arbitrary diagnostic criteria.

The majority of Conn’s tumours are benign. Grossly, they are characteristically around 1 cm in diameter or less and the cut surface has a bright yellow appearance (Fig. 5.6.4), which reflects the lipid-laden nature of the cells (47). On histological examination, the tumour contains cells which are typical of adrenal cortex. In the normal adrenal, it is possible to distinguish zona glomerulosa type cells, which have a high nuclear/cytoplasmic ratio and moderate amounts of lipid. Zona fasciculata type cells have a lower nuclear/cytoplasmic ratio and greater amount of lipid, and cells of the zona reticularis are lipid-depleted and appear eosinophilic. Furthermore, there are distinct differences in the morphological appearance of the mitochondria of these cell types on electron microscopy (48). Conn’s adenomas are often composed of relatively uniform zona fasciculata type cells (Fig. 5.6.5) but may contain a mixture of fasciculata, glomerulosa, and reticularis cell types. Some cells may display features of both fasciculata and glomerulosa, so-called hybrid cells. It has been proposed that these histological differences reflect contrasting responsiveness to angiotensin II (tumours which contain predominantly zona glomerulosa type cells are responsive, whereas those with mainly zona fasciculata type cells are unresponsive (49)). Again, the practical value of this differentiation is limited.

In some cases there may be evidence of diffuse hyperplasia of the zona glomerulosa. The mechanisms underlying this are unknown. Alternatively, the adrenal cortex may be enlarged because it contains multiple nodules, which are histologically typical of zona fasciculata (50). Routine post mortem examinations in patients with essential hypertension and, indeed, in older normotensive subjects, show a high frequency of nodular change in the adrenal cortex (51). Thus, there may be no clear demarcation between essential hypertension and adrenal hyperplasia, either pathologically or clinically (see below).

Aldosterone-producing adrenal carcinomas are rare. In any individual tumour it may be difficult to define malignant potential, and multifactorial analysis of histological features is usually required. In general, malignant tumours tend to be larger (over 100 g) and show a very abnormal pattern of corticosteroids in plasma and urine. Rarely, some tumours will secrete both aldosterone and cortisol.

The molecular basis for development of Conn’s adenomas, adrenal carcinomas, and hyperplasia remains uncertain. In carcinomas, there is evidence that tumours are monoclonal and there may be an association with abnormal expression of p53 protein (52). Furthermore, increased levels of insulin-like growth factor-II have been reported in these tumours. In contrast, aldosterone-producing Conn’s adenomas may be polyclonal, and no single molecular pathology has been identified to account for their development. Increased expression of CYP11B2, leading to increased aldosterone synthase activity, has been reported, but the reason for this is uncertain. There are also reports of increased expression of renin in Conn’s adenomas and it may be that in some the primary fault relates to overactivity of a local intra-adrenal renin/angiotensin system, leading to overexpression of aldosterone synthase. Gross chromosomal rearrangements have been sought in typical Conn’s adenomas and have not been consistently found. Other studies have included screens for mutations in the subunits of the stimulatory G protein, G_s, and in the angiotensin II receptor (53). There is one report of increased RAS oncogene expression in Conn’s adenomas, a report that remains unconfirmed (54). Other reports have demonstrated that a minority of Conn’s adenomas display aberrant expression of G-protein-coupled receptors, including those that act as ligands for gut hormones and other peptides (14). However, these appear to be relatively rare. Finally, overexpression of a potassium channel (TASK) in the mouse leads to development of a syndrome that recapitulates features of primary aldosteronism (55); however, is not clear whether TASK channel abnormalities play any role in the genesis of this syndrome in humans.

Rarely, aldosterone-producing adenomas are associated with other genetic mutations. For example, Conn’s adenomas have been reported in association with multiple endocrine neoplasia type I and also with the Beckwith−Wiedemann syndrome. However, these inherited conditions are extremely rare. Familial aldosterone-producing adenomas, designated FH II (to distinguish them from GRA (FH I)) are described, where an autosomal dominant pattern of inheritance is found (56). These are relatively rare, although detection of kindreds requires assiduous case detection. Family studies have suggested that a locus on chromosome 7 is associated with FH II, but the precise gene responsible has not been identified. None the less, it is prudent to enquire about the family history of hypertension and consider inherited conditions in any apparent sporadic cases of primary aldosteronism.

Aldosterone binds to mineralocorticoid receptors in the distal renal tubule to increase sodium reabsorption (by activating the epithelial sodium−hydrogen exchanger). Activation of mineralocorticoid receptors also results in activation of sodium−potassium pump activity. The precise molecular events that link mineralocorticoid receptor activation to sodium reabsorption and potassium loss are discussed above. Mineralocorticoid receptor activation actions lead to expansion of body sodium and depletion of body potassium content; the excess body sodium results in expansion of both extracellular fluid volume and plasma volume. Although there is a reasonable correlation between body sodium and blood pressure in primary aldosteronism (57), it is likely that the rise in blood pressure reflects mechanisms other than, or in addition to, simple plasma volume expansion with the associated rise in cardiac output. For example, mineralocorticoid receptors are present in vascular smooth muscle and their activation leads to alteration in pressor responsiveness to adrenergic stimulation. Furthermore, there is good evidence that mineralocorticoid receptors in cardiac tissue regulate collagen formation (58, 59), and a similar action in the peripheral vasculature might be expected to result in remodelling which would help sustain blood pressure. Thus, there is good evidence that aldosterone levels are inversely related to arterial compliance in essential hypertension, while therapy with a mineralocorticoid receptor antagonist (eplerenone) in patients with essential hypertension leads to a reduction in arteriolar media thickness, in comparison with treatment using a β-blocker (60). Patients with primary aldosteronism would, by analogy, be expected to have vascular remodelling, reduced vascular compliance, effects which will increase systolic hypertension. This concept is supported by a study which shows that the outcome of surgical removal of an aldosterone-producing adenoma is directly related to the degree of vascular remodelling in resistance arterioles removed preoperatively (61).

Finally, receptors for aldosterone are present in the central nervous system and may regulate central sympathetic outflow as well as thirst and sodium appetite (62). It is known that central administration of aldosterone raises blood pressure without altering circulating concentrations of the hormone and that the rise in blood pressure is not associated with sodium retention. Thus, sustained excessive aldosterone is likely to raise blood pressure through a variety of different mechanisms, all of which are dependent on activation of mineralocorticoid receptors. The fact that blood pressure can be lowered effectively and specifically by a mineralocorticoid receptor antagonist, such as spironolactone, confirms this notion without giving any major insight into the relative importance of the pressor mechanisms involved.

Increased aldosterone secretion causes expansion of body sodium content, with a consequent rise in both extracellular fluid volume and plasma volume. However, unless water is restricted, plasma sodium is generally within the normal range. The increased distal sodium reabsorption and potassium loss are associated with hydrogen ion depletion, resulting in a systemic metabolic alkalosis.

The excess sodium reabsorption is invariably associated with total body potassium depletion, although this need not result in hypokalaemia. Potassium levels are generally at the lower end of the normal range or frankly subnormal; in only 50% of patients is plasma potassium distinctly low (40). Several factors may account for the relative normality of plasma potassium in this syndrome. First, relatively mild aldosterone excess may be less likely to lead to profound hypokalaemia. Secondly, other factors, including intercurrent drug therapy, may determine the prevailing potassium level in this syndrome. Thus, calcium-channel antagonist treatment can reduce aldosterone secretion in Conn’s syndrome, leading to normalization of serum potassium levels. Conversely, drugs that increase sodium delivery to the distal renal tubule (for example, thiazide diuretics) will increase the tendency to hypokalaemia. Thirdly, hypokalaemia is more likely to be observed in circumstances of increased sodium intake. Thus, relative restriction of dietary sodium may result in a reduced tendency to develop hypokalaemia.

There is a tendency for magnesium concentrations to be reduced in primary aldosteronism, although this is generally not a major therapeutic issue. It is important to bear in mind that profound hypokalaemia may be associated with magnesium deficiency and that both ions should be replaced in such circumstances. A proportion of patients with primary aldosteronism have impaired glucose tolerance: in a small minority, frank diabetes mellitus may develop. This may be a consequence of potassium deficiency.

A few of the above changes give rise to characteristic physical findings. Occasionally, when hypokalaemia is severe, patients may develop muscular weakness or a frank proximal myopathy. In a small number of patients, the alkalosis which accompanies the other electrolyte abnormalities can become sufficiently severe to result in tetany.

The rise in blood pressure in primary aldosteronism is generally mild or moderate, but rare patients have been described with malignant-phase hypertension. Under this unusual circumstance plasma renin concentrations will not be suppressed, due to the severe renal ischaemia present in the malignant phase. The blood pressure in patients with primary aldosteronism is often resistant to conventional antihypertensive drug treatment (63). This gives a clue to the need to investigate patients further. As noted above, there is a reported increase in frequency of primary aldosteronism in patients with resistant hypertension, suggesting that the degree of blood pressure elevation in patients with aldosterone excess is particularly severe.

There are no clear distinguishing features of cardiovascular function in primary aldosteronism to differentiate patients with essential hypertension. For example, baroreflex activity is normal, although recent studies have suggested that aldosterone may alter function of the autonomic nervous system (64). Studies of blood pressure variability in primary aldosteronism have been performed and show no distinct pattern on ambulatory recording over a 24-h period (65).

As mentioned above, aldosterone has effects on both vascular contractility and vascular structure. For example, administration of aldosterone to animals increases cardiac collagen content and there is a good correlation between aldosterone levels and cardiac collagen content in humans. Furthermore, there are changes in left ventricular mass and left ventricular function in patients with primary aldosteronism. Such analyses have been difficult to perform, as it is important to ensure that patients and controls are carefully matched for age, gender, body mass index, and other factors that can affect left ventricular hypertrophy. Nonetheless, several studies have shown that primary aldosterone excess leads to more severe left ventricular hypertrophy than is seen in patients with similar levels of blood pressure due to essential hypertension (66, 67). Moreover, abnormalities of left ventricular function, including abnormal diastolic relaxation, have been described. Whether these changes regress with effective aldosterone receptor antagonism or with removal of the source of aldosterone has not been thoroughly evaluated in humans. However, in animal studies, aldosterone-related cardiac hypertrophy can be blocked effectively by spironolactone, while experiments in hypertensive rats show that aldosterone is responsible for severe vascular damage and that this can be prevented by mineralocorticoid receptor antagonism (68). In these animal models the damage caused by mineralocorticoid excess is dependent on concomitant sodium loading and, often, partial nephrectomy; there is not only structural change but development of marked inflammatory change including infiltrate with lymphocytes and evidence of local synthesis of proinflammatory cytokines (69). It is not clear whether similar changes are seen in humans with primary aldosteronism, although a careful comparison of cardiovascular outcomes in patients with aldosterone excess with essential hypertension shows that primary aldosteronism is associated with a substantial excess of risk of left ventricular hypertrophy, atrial fibrillation, myocardial infarction, and stroke (70).

There are no detailed studies of the effect of primary aldosteronism renal structure. Severe hypokalaemia, which can occur in primary aldosteronism, results in vacuolation within the kidney. However, such severe potassium depletion in primary aldosteronism is very uncommon. In animal models of hypertension, aldosterone excess causes significant renal damage due to deoxycorticosterone, while mineralocorticoid excess is associated with substantial histological evidence of inflammation and glomerular damage (71). Although primary renal impairment is not commonly reported in patients with aldosterone excess, there is evidence that aldosterone can determine the rate of progression of other forms of renal disease. For example, in patients with essential hypertension, aldosterone appears to interact with sodium intake to determine the excretion of protein loss in the urine (42), while analysis of renal function in the large Italian study of prevalence of primary aldosteronism (the PAPY study) showed that patients with primary aldosteronism had higher urinary albumen excretion subjects with essential hypertension (72). Thus, aldosterone excess appears to cause significant increased renal damage beyond that anticipated for the level of blood pressure.

The diagnosis of primary aldosteronism falls into two distinct parts. First, aldosterone excess needs to be suspected and the primary nature of the disorder established. Secondly, the cause must be identified. If one accepts that primary aldosteronism may affect around 11% of the hypertensive population and that just less than half of those subjects may harbour a Conn’s adenoma, it is reasonable to consider which screening procedures are appropriate in patients with hypertension. A comprehensive guideline on the detection and classification of primary aldosteronism is of particular value in this regard (73).

Some authors have advocated very widespread screening for primary aldosteronism (74). However, it is difficult to justify screening of all hypertensive patients for a condition that may affect only around 10%; in this circumstance, it would be more appropriate to screen selected subgroups at high risk. Clearly, hypokalaemia (either spontaneous or provoked by diuretic therapy) is an important diagnostic clue. Patients who are resistant to conventional antihypertensive therapy (generally defined as not achieving target blood pressure despite use of three appropriate agents) are another group in whom screening is justified. Furthermore, although the true frequency of familial primary aldosteronism (either due to glucocorticoid-remediable aldosteronism or other less well-defined entities) remains uncertain, patients with hypertension who have a positive family history of primary aldosteronism should be screened for the condition. Finally, screening is reasonable in subjects developing hypertension at a young age (<40 years).

Simultaneous measurement of aldosterone and renin (either renin activity or active renin concentration) provides the most reliable single screening test for primary aldosteronism. Either measure on its own is prone to the influence of drug therapy, posture, or other confounding factors. The ARR circumvents many of these problems, as both measurements change in a parallel manner in response to most manoeuvres and is therefore of value as an initial screen (38). Additionally, there is no need to control dietary sodium intake.

The cut-off value for an abnormal ratio that merits further investigation must be determined using local assay conditions. A figure of 750 has been suggested as sufficiently sensitive and specific when plasma aldosterone is expressed in pmol/l and renin activity in ng/ml per h. Some screening algorithms demand not only a raised ARR but a cut-off of a minimal level of aldosterone (e.g. greater than 300 pmol/l) as this substantially increases the positive predictive value of the ARR in detecting the syndrome, a practice that we endorse (73). Finally, the performance of many assays (particularly those for renin) varies and it is necessary to establish a ratio for the normal population locally. It should be noted that in the ARR, renin, as the denominator, has an undue weight on the derived value (75). For this reason, care must be taken in the interpretation using new assays; the great majority of studies defining the prevalence of primary aldosteronism have used renin activity assays, and it is not safe to assume that similar data would be achieved using high throughput renin concentration assays. Factors that affect renin, including gender, age, and body mass index, must also be taken into account.

Drug therapy also has a substantial influence on the ARR, mainly through effects on renin (76). For example, β-blockers depress renin in plasma leading to raised (and therefore false-positive) levels of the ARR. Angiotensin-converting enzyme inhibitors will raise renin and lower aldosterone and, for that reason, reduce the ARR. Diuretics will raise renin and aldosterone and tend to reduce the ARR or have no significant effect. For these reasons, confirmation of the abnormal screening measurement should be made under more stringent conditions, where drug treatment has been either discontinued or altered to avoid confounding agents—α-blockers are unlikely to influence the ratio and can be safely used in this circumstance. Once a positive screening test using reliable methodology, and where it is clear that this is not an artefact caused by interfering antihypertensive agents, confirmation of the diagnosis is then required using a range of possible methods outlined below.

It is important to demonstrate that aldosterone secretion is autonomous to confirm the diagnosis of primary aldosteronism. It should be noted, of course, that all of the tests described show that aldosterone is independent of control of the renin/angiotensin system and do not provide information about other regulatory mechanisms (such as ACTH in GRA). Four main tests are described; the most appropriate needs to be selected to suit local circumstances and investigation facilities. In all of the sodium-loading tests (including the fludrocortisone test) it is important to maintain plasma potassium levels as near normal as possible, both for safety reasons and as hypokalaemia itself will reduce aldosterone secretion and affect the performance of the test being used. Oral potassium supplements are likely to be required in each instance (e.g. Slow K 600 mmol three times per day).

This test can be simply performed in an outpatient setting (40). Patients need to be given low sodium tablets to raise intake to 200 mmol/day for a 4-day period; for the final 24 h a 24-h urine collection should be made to measure aldosterone excretion, and blood taken for measurement of renin and aldosterone. In normal subjects, aldosterone excretion should be suppressed to less than 5 µg/24 h. As described above, potassium supplements should be used to maintain plasma potassium levels within the normal reference range if possible during the period of sodium loading.

The simplest test to confirm the presence of primary aldosteronism is infusion of normal saline (2 L over a 4-h period) (77). If plasma aldosterone levels remain elevated (in practice above 140 pmol/l) at the end of this manoeuvre, the diagnosis is confirmed. However, there is a small risk of provoking cardiac failure, particularly in elderly patients, and the test should be performed with caution.

A more elaborate version of the sodium-loading test is the administration of the synthetic mineralocorticoid fludrocortisone (0.5 mg four times daily for 2 days), with measurements of aldosterone at the beginning and end of this manoeuvre. Some authorities regard this as a definitive test in primary aldosteronism. Although doubtless reliable, it does necessitate admission of patients to hospital, which may not be cost-effective. In the test described by Gordon, fludrocortisone is given in a dose of 400 mg/day in association with additional sodium chloride tablets (90 mmol daily) (78). In normal subjects, aldosterone should be fully suppressed, and failure to suppress is diagnostic of primary aldosteronism. However, the test carries with it a substantial risk of significant potassium depletion and profound hypokalaemia, with the attendant dangers of cardiac dysrythmia.

Administration of captopril (25 mg), with measurement of renin and aldosterone before and 2 h after drug therapy, is described as a diagnostic manoeuvre for primary aldosteronism (79). In normal subjects, aldosterone levels will be suppressed by a single dose of captopril (as a consequence of inhibition of angiotensin II formation), while this is not the case in patients with Conn’s adenomas.

In summary, confirmation of primary aldosteronism can often be had simply by careful measurements of aldosterone and renin in patients in whom dietary sodium intake is not restricted and in whom confounding drug therapy (principally calcium-channel blockers, which can lower aldosterone levels in Conn’s adenoma patients) has been eliminated. In such circumstances, it may be justifiable to proceed to definitive tests for the differential diagnosis of primary aldosteronism.

When primary aldosteronism is confirmed, the principal problem is to distinguish between a Conn’s adenoma and bilateral adrenal hyperplasia. The distinction is important since subsequent treatment of the two variants is different. A small number of patients will have glucocorticoid-remediable aldosteronism; this can be readily diagnosed on the basis of a simple genetic test (80). The distinction is important since subsequent treatment of the two variants is different.

Aldosterone levels are generally higher in patients with Conn’s adenomas than in those with bilateral adrenal hyperplasia, but this is not, in itself, a reliable discriminant. Similarly, concentrations of other corticosteroids, including 18-hydroxycorticosterone, 18-hydroxycortisol, and 18-oxocortisol, are higher in patients with adenomas but are not routinely measured in the diagnostic workup of patients with primary aldosteronism and, in any case, do not reliably improve discrimination.

Dynamic tests of aldosterone responsiveness do not help discriminate accurately between Conn’s adenomas and bilateral adrenal hyperplasia. Although aldosterone does not respond to the administration of angiotensin II, in the majority of patients with adenomas (in contrast to patients with bilateral hyperplasia, where a very brisk response may be seen), a positive response is reported in a substantial minority of patients with adenomas. For this reason, reliance on aldosterone response to upright posture or angiotensin II is not a secure means of discriminating between the two main causes of primary aldosteronism.

Imaging of the adrenal glands is a key step in differential diagnosis. Ultrasonography is of no value, although large adrenal carcinomas will be readily identified. In all patients with confirmed primary aldosteronism, careful imaging of the adrenal glands with either CT or MRI is necessary. CT scans of the abdomen with 3 to 5 mm slices of the adrenal regions will provide accurate identification of the adrenal glands and should demonstrate the majority of adenomas, although very small lesions (in practice, those less than 5 mm) may be missed by this technique. A typical lesion identified by CT scanning is shown in Fig. 5.6.6. MRI scanning of the abdomen also gives good resolution of the adrenal glands but offers no advantage over carefully performed CT imaging. Radiolabelled cholesterol scanning (generally carried out after dexamethasone suppression to reduce normal adrenal gland uptake of cholesterol) has been used to identify adrenal adenomas in patients with primary aldosteronism. However, this is not a sensitive technique. In some patients with bilateral adrenal hyperplasia, CT scanning may show enlargement of the glands; small nodules can be visualized. Due to the heterogeneity in nodule size, the CT scan appearance in these patients may be confused with that of a single adenoma, and the definitive diagnosis of a solitary aldosterone-producing adenoma requires selective adrenal vein sampling.

This technique is indicated in any patient in whom surgical adrenalectomy is contemplated and in whom the presence of a unilateral adenoma is not clear cut. As small adrenal incidentalomas are commonly seen in normal subjects, particularly with advancing years, it is therefore not safe to assume that a lesion on CT scanning is responsible for the syndrome of primary aldosteronism. A reasonable approach is to consider sampling in any patient over the age of 40 in whom surgery is indicated (there is clearly no need to perform sampling if the patient or clinician does not feel that surgery, regardless of the diagnosis, is appropriate). Furthermore, the procedure is absolutely necessary when radiology is uncertain and when adrenal gland surgery is being considered in patients with no definite radiological abnormality.

In performing the technique, simultaneous measurement of both aldosterone and cortisol in the adrenal effluent is required (81). It is necessary to measure cortisol in order to confirm the technical success of the procedure by demonstrating a concentration gradient between adrenal vein and low inferior vena cava. Unfortunately, it is not always possible to achieve bilateral adrenal vein catheterization (the failure rate may be up to 25%, with greatest difficulty occurring in cannulation of the right adrenal vein) and this limits the value of the procedure. The confirmation of lateralization is achieved by demonstrating a ratio of aldosterone:cortisol that is at least twofold when comparing right with left (or vice versa). Some authors recommend use of ACTH during the procedure to stimulate secretion of aldosterone from an adenoma, and improve the sensitivity of the test; we suggest that this adds to the complexity of what is already a technically demanding test (82). Finally, real-time measurement of cortisol during the test has been advocated as a means of improving technical success rates for cannulation of the adrenal veins; this is not a widely available assay.

Although it has been suggested that adrenal vein sampling is not necessary in patients with a clear adenoma visualized on CT scanning, there are reports of removal of nonfunctional adrenal ‘incidentalomas’ which were not responsible for aldosterone excess. Thus, where there is any doubt, adrenal vein sampling should be performed.

The diagnosis of primary aldosteronism can be problematic but has been assisted by the recent publication of a clear investigative strategy by the Endocrine Society (73). Figure 5.6.7 summarizes a coherent diagnostic approach to investigate the patient with suspected primary aldosteronism, which is based upon both Endocrine Society and Mayo Clinic guidelines (40). It should always be noted that, before performing specific diagnostic tests, serum potassium should be normalized, if necessary by oral supplementation, and patients should be encouraged to maintain a liberal dietary salt intake prior to ARR testing and throughout subsequent investigations for primary aldosteronism.

The mineralocorticoid receptor antagonist, spironolactone, is effective as an antihypertensive agent in primary aldosteronism. Fairly high dosage may be required (historically use of up to 400 mg/day was reported), although it is appropriate to start with low doses (25 mg/day) and increase gradually until blood pressure control is achieved. The high dose necessary to cure hypertension may, however, limit use of spironolactone, particularly in male patients. Principal side effects of spironolactone include gynaecomastia, diminished libido, and impotence, and reflect transactivation of the androgen receptor. The alternative mineralocorticoid receptor antagonist, eplerenone, does not have significant affinity for the androgen receptor and is free of these unwanted effects. It can be used in primary aldosteronism, although it appears less effective as a mineralocorticoid antagonist compared with spironolactone; high doses (up to 150 mg twice daily) may be required.

Amiloride, which blocks the epithelial sodium channel in the distal renal tubule, is also effective in lowering blood pressure in primary aldosteronism. Indeed, earlier studies which compared amiloride with spironolactone show that the drugs were equally effective in lowering body sodium content and reducing blood pressure in this condition. As with spironolactone, amiloride must be given in relatively high dosage (up to 40 mg/day).

In treating patients with either drug, it is important to monitor plasma potassium concentrations. In patients with renal impairment, there is a risk of hyperkalaemia and dosage of both drugs should be kept to the minimum under these circumstances. Plasma renin concentrations give some guide to the effectiveness of drug therapy in patients with primary aldosteronism. Thus, persistent suppression of renin levels suggests that the drug is not being given at an effective aldosterone-antagonist dosage.

Patients with primary aldosteronism are often resistant to other antihypertensive drug therapy. Clearly, angiotensin-converting enzyme inhibitor treatment is illogical in patients in whom renin and angiotensin II levels are suppressed. Calcium-channel blockers, particularly of the dihydropyridine class, are reported to reduce aldosterone secretion in patients with Conn’s adenomas. They can be combined safely with spironolactone or amiloride and may provide effective blood pressure control in patients resistant to single-drug therapy.

Surgical removal of an aldosterone-producing adenoma is normally the most appropriate treatment for patients with a unilateral lesion. However, before surgery is performed it is necessary to optimize blood pressure control and to correct any significant electrolyte disturbance. Previous studies have shown that the blood pressure response to either spironolactone or amiloride can predict the blood pressure outcome following surgical adrenalectomy. One practical consequence of this may be to help predict those patients in whom surgical treatment may not be curative; in these circumstances or where surgery is contraindicated, combination therapy with aldosterone antagonist drugs, with or without other antihypertensive treatments, may be appropriate.

Treatment with either spironolactone (or amiloride) will normally fully correct potassium depletion in primary aldosteronism before surgery. Effective therapy with either drug will also minimize the risk of postoperative hypoaldosteronism, which can occur due to atrophy of the normal zona glomerulosa caused by the excessive autonomous aldosterone secretion. Potassium supplementation may also be given, although administration of adequate doses of either spironolactone or amiloride is normally sufficient over a longer period of time to maintain a normal body potassium content.

The surgical approach to the adrenal gland in patients with unilateral adenomas was previously either by an anterior or a lateral open operation but laparoscopic adrenalectomy is the now the surgical approach of choice. In experienced hands the operation has a relatively low morbidity and a high success rate (83). It is important to inform patients about the likely success rates of adrenal surgery; this is often poorly documented in series of surgical adrenalectomy, and some do not fully differentiate between ‘cure’ of hypertension (where patients require no antihypertensive therapy) and ‘improvement’, where patients may need less medication than before the procedure. It is likely that surgery will cure the tendency to hypokalaemia if the adenoma is correctly removed; however, absolute cure rates of blood pressure elevation may be less than 30%, with improvement in blood pressure in a further 30% (84). It is possible that the duration of the syndrome before diagnosis influences the ultimate outcome following surgery; there is evidence that vascular structural changes predict the achieved blood pressure level after successful adrenalectomy.

Bilateral adrenal hyperplasia should be treated medically using either spironolactone or amiloride in conjunction with other antihypertensive drugs, as necessary. Although it has been suggested that partial adrenalectomy can improve blood pressure control, it is difficult to justify this when effective drug therapy is available.

Suspected adrenal carcinomas should be surgically resected at the earliest opportunity. It may not be possible to diagnose, with certainty, the malignant nature of a lesion on histological grounds alone, and the presence of recurrent or metastatic disease is the only certain way of doing so. Malignant adrenal lesions do not respond to external radiotherapy and are generally resistant to combination chemotherapy. Some patients with adrenocortical malignancy may show a response to the use of mitotane (ortho, para′, dichlorodiphenyldichloroethane (o,p′DDD)), but consistent good responses are unusual.

Other causes of mineralocorticoid hypertension, listed in Table 5.6.2, are uncommon and are discussed briefly below.

This syndrome was first described by Grant Liddle, in 1963, in a family in which the siblings appeared to have features of aldosterone excess (early onset hypertension and hypokalaemia) but with suppressed plasma renin and aldosterone levels (85). It is now known that this syndrome is inherited as an autosomal dominant trait and occurs due to mutations in the genes encoding the β or γ subunits of the ENaC. (Fig. 5.6.2). Thirteen mutations in the β ENaC and four in γ ENaC subunits have been identified in patients with Liddle’s syndrome so far (86, 87). Most either alter or delete a highly conserved PY-motif at the C-terminal end of the channel that is involved in its normal regulation by virtue of its interaction with Nedd4 (see above). The effect of the mutations is to alter the interaction so that trafficking of ENaC to the proteosome is disrupted, and the likelihood of the channel being in open conformation in the apical membrane is greatly increased. The exception to this is one isolated mutation in γ ENaC (Asn530Ser) which is located in the extracellular loop of the gamma subunit and does not affect the PY-motif (88). These various mutations all lead to constitutive activation of the sodium channel, resulting in excessive sodium reabsorption in the distal nephron irrespective of circulating mineralocorticoid levels, which are suppressed. The laboratory findings include increased urinary potassium excretion, hypokalaemia, and suppression of plasma renin activity and of circulating levels of angiotensin II and aldosterone.

Interestingly, in the proband of one of Liddle’s original cases, renal transplantation resulted in normalization of blood pressure and electrolyte abnormalities. In practice, however, blockers of the sodium channel, such as amiloride or triamterene, effectively treat the electrolyte abnormalities and hypertension. Mineralocorticoid antagonists such as spironolactone are ineffective, as this disorder is not a consequence of activation of the mineralocorticoid receptor.

This rare disorder was first described in 2000 and is characterized by constitutive activation of the mineralocorticoid receptor as well as an alteration in receptor sensitivity (89). A missense mutation in the hormone binding domain of the mineralocorticoid receptor has been identified as the cause, leading to the substitution of leucine for serine at codon 810 (S810L). The S810L mutation alters mineralocorticoid receptor sensitivity; most significantly, both progesterone and spironolactone, which usually act as antagonists at the mineralocorticoid receptor, become potent agonists. Subjects with this mutation are characterized by early onset of severe hypertension with suppression of aldosterone and plasma renin. This mutation and the resulting phenotype were described in eight out of 23 of the index patient’s family, suggesting an autosomal dominant mode of transmission. Progesterone levels increase by up to 100-fold in pregnancy, and carriers of the S810L tend to develop severe pregnancy-associated hypertension.

Apparent mineralocorticoid excess (AME) is a rare syndrome of hypertension and hypokalaemia associated with suppression of plasma renin activity and low plasma concentrations of aldosterone and other known mineralocorticoids (90). As described above, 11β-HSD2 normally oxidizes cortisol to cortisone, which does not transactivate the mineralocorticoid receptor. In this manner, 11β-HSD2 acts as a ‘gatekeeper’ to prevent the mineralocorticoid receptor becoming saturated with cortisol which is present at a much higher level than aldosterone. The molecular basis of this syndrome was described in 1995; 11β-HSD2 activity is reduced or absent such that cortisol overwhelms the mineralocorticoid receptor causing cortisol-mediated mineralocorticoid hypertension.

Classically, this syndrome, inherited in an autosomal recessive manner, usually presents in childhood with failure to thrive, short stature, significant hypertension, and hypokalaemia. The potassium depletion may be severe, leading to nephrogenic diabetes insipidus and rhabdomyolysis. Biochemical diagnosis of AME can be made by measuring the ratio of cortisol (compound F) to cortisone (compound E) as indicated by the ratios of their tetrahydro (allo)-urinary metabolites (THF + alloTHF:THE) (91). Normal subjects excrete two- to threefold more urinary free cortisone than urinary free cortisol, reflecting the significant activity of renal 11β-HSD2. In AME, however, urinary free cortisone excretion is extremely low, leading to an increased THF + alloTHF:THE ratio in urine. Despite a marked increase in the half life of plasma cortisol, AME patients are not cushingoid since the normal negative feedback system remains intact, leading to a marked reduction in cortisol secretion rates.

The gene encoding 11β-HSD2 is 6.2 kb long, comprises five exons, and is located on chromosome 16q22 (Fig. 5.6.8) (92). Less than 100 cases of AME have been reported, with more than 35 different nonsilent mutations identified clustered in exons 1–5 (Fig. 5.6.8). Complete abolition of enzymatic activity results in the classical and severe AME phenotype described above. Milder cases of AME, so-called ‘type II apparent mineralocorticoid excess’ with isolated hypertension and normal or low-normal potassium have been described in Italian patients (93). In this kindred, a homozygous mutation in the 11β-HSD2 gene (R279C) has been identified which causes a reduction but not complete abolition of 11β-HSD2 activity. It can be seen, therefore, that AME comprises a spectrum of mineralocorticoid hypertension with a good correlation between genotype and phenotype.

AME can be effectively treated with amiloride, although high doses (up to 40 mg daily) can be required for therapeutic benefit. Mineralocorticoid receptor antagonism with spironolactone offers an alternative therapy, but its use may be limited by the relatively high doses required to competitively antagonize the agonist effects of cortisol in this circumstance; this consideration is particularly important in young male patients where unwanted androgen effects can limit use of this drug. Dexamethasone has also been used therapeutically but its use is limited by the need to employ a dose sufficiently high to inhibit endogenous cortisol production, exposing patients to unwanted glucocorticoid side effects. As well as improving blood pressure, a major aim of treatment is correction of hypokalaemia, which contributes to the poor growth rate seen in children with this disorder. As is common in secondary hypertension, definitive therapy may not always normalize blood pressure (or potassium levels) and additional antihypertensive agents may be required. Deficiency of 11β-HSD and consequent mineralocorticoid hypertension can also occur as a result of ingestion of liquorice or carbenoxolone (previously used for the treatment of peptic ulcer disease). The active component of liquorice is glycyrrhizic acid and its hydrolytic product glycyrrhetinic acid, which have been shown to inhibit the activity of 11β-HSD2 in the renal tubule allowing cortisol-driven mineralocorticoid hypertension (94). Carbenoxolone is a semisynthetic hemisuccinate derivative of glycyrrhetinic acid and has its effect through a mechanism analogous to that of liquorice.

Subjects consuming excessive quantities of liquorice may present with hypertension and hypokalaemia associated with suppression of plasma renin activity and aldosterone as well as an increase in exchangeable sodium levels. This condition responds to treatment with spironolactone or amiloride, but is best dealt with by cessation of liquorice ingestion.

Other mineralocorticoids rarely circulate in sufficient levels to cause hypertension. The aldosterone precursor deoxycorticosterone, which binds and activates the mineralocorticoid receptor, circulates at concentrations around 2% of those of aldosterone and so, under normal circumstances, does not contribute to electrolyte and blood pressure regulation. However, excessive plasma levels of deoxycorticosterone can be found, rarely, in patients with adrenal carcinomas (16). The result is hypertension similar to that caused by excess aldosterone, and is associated with sodium retention and potassium loss leading to hypokalaemia. Less commonly, raised deoxycorticosterone levels are found in adult patients with the rare inborn errors of adrenal steroid synthesis due to defective 17α-hydroxylase or 11β-hydroxylase activity. In both of these circumstances, increased ACTH drive to the adrenal causes chronic excess deoxycorticosterone secretion (95). The resultant sodium retention causes suppression of renin release and, as a consequence, aldosterone levels are generally low. Most presentations occur shortly after birth or in early childhood. Adrenal androgens are produced in excessive amounts in patients with 11β-hydroxylase deficiency, leading to virilization of female subjects. In 17-hydroxylase deficiency, there is inability to synthesize sex hormones, with the result that affected males fail to develop normal masculine external genitalia, while females fail to progress through adrenarche or puberty. Diagnosis is confirmed by measurement of corticosteroid metabolite excretion in the urine. A more complete description of these autosomal recessive disorders and their management is given in Chapter 000.

Approximately 80% of patients with Cushing’s syndrome have hypertension, increasing to 95% in subjects with Cushing’s syndrome due to ectopic ACTH production. Ectopic ACTH syndrome is generally associated with hypokalaemic alkalosis (in 95–100%) consistent with mineralocorticoid hypertension. Several studies have demonstrated that the mineralocorticoid excess state is explained by saturation of 11β-HSD2 by the very high cortisol concentrations seen in the ectopic ACTH syndrome. Both the urinary ratios of tetrahydrocortsol and allotetrahydrocortisol/ tetrahydrocortisone and free cortisol/ cortisone are elevated, not because of impaired 11β-HSD2 function but due to saturation of the enzyme by high levels of cortisol (96). Thus, the enzyme is overwhelmed by substrate and cortisol cannot be inactivated to cortisone in the renal tubule leading to activation of the mineralocorticoid receptor by cortisol.

Gordon’s syndrome (also known as pseudohypoaldosteronism type II), is a rare autosomal dominant disorder characterized by hypertension, hyperkalaemia, hyperchloraemia, acidosis, and sodium retention leading to suppression of plasma renin and aldosterone (97). The molecular basis of this disorder has been found to be explained by mutations in the WNK (with no K (lysine)) kinases. These are a family of protein kinases with unusual protein kinase domains due to the unusual placement of the catalytic lysine when compared to all other protein kinases (98). Pseudohypoaldosteronism type II develops due to mutations in either WNK1 or WNK4 (99)

WNK4 normally inhibits the thiazide sensitive Na–Cl cotransporter of the distal nephron; thus missense mutations increase the activity of the Na–Cl cotransporter, leading to thiazide-sensitive hypertension; systolic and diastolic blood pressure fall by approximately 45 mm Hg and 25 mm Hg respectively after treatment with 25 mg of hydrochlorothiazide per day (100). The mechanism of hypertension in subjects with WNK1 mutations is less clear. It has been demonstrated that WNK1 mutations abolishes the WNK4-mediated inhibition of the Na–Cl cotransporter in the distal convoluted tubule. However, WNK1-mediated Gordon’s syndrome is less sensitive to thiazide diuretic treatment, suggesting that other mechanisms may be involved (101). There are reports that WNK1-mediated hypertension may also occur through activation of ENaC and inhibition of ROMK (inwardly rectifying K) channel, which controls potassium secretion in the renal distal nephron.

Primary aldosteronism is the commonest cause of mineralocorticoid hypertension, although other rare causes should be considered as discussed above and an algorithm outlining a potential approach to the investigation of mineralocorticoid excess is illustrated in Fig. 5.6.9. Importantly, primary aldosteronism is now considered to be the commonest cause of secondary hypertension; reported prevalence in the hypertensive population ranges from 6–12%. Much of this increase in detection of primary aldosteronism is due to more widespread screening of hypertensive populations using the ARR (which is driven by the level of renin) as a first step. The subsequent investigation of suspected mineralocorticoid hypertension can follow a logical pattern thereafter, and the algorithm shown in Fig. 5.6.9 offers one such simple approach, although it should be stressed that it relies on availability of reliable endocrine biochemical and imaging services. The publication of the recent consensus clinical guideline for the investigation of primary aldosteronism by the Endocrine Society (summarized in Fig. 5.6.7) provides a clear approach, thereafter, to the investigation and management of the patient suspected of having primary aldosteronism.