CLINICAL PRACTITIONER PERSPECTIVE
Primary aldosteronism is a common cause of secondary hypertension, affecting up to 10% of hypertensive patients. Fortunately, we have made progress in the diagnosis and treatment of the disorder, as well as in the elucidation of disease-causing mechanisms. In this TriPoint article, a clinical practitioner discusses the factors to consider when screening patients for the presence of primary aldosteronism; a clinical researcher provides an overview of diagnostic approaches to determine the subtype of primary aldosteronism, appropriate treatment modalities, and outcome measures; and a basic researcher reviews recent advances in our understanding of genetic and molecular mechanisms underlying unopposed aldosterone secretion by the adrenal gland.
The approach to screening for primary aldosteronism (PA) has evolved over the 58 years since Jerome Conn reported the first case of PA. PA was thought to be a rare cause of hypertension for three decades because of strict rules for case-detection testing (e.g., sodium-balanced diet for two weeks, discontinuing most antihypertensive medications) and the erroneous concept that most patients with PA are hypokalemic. In 1981, Hiramatsu and colleagues showed that case-detection testing could be performed with very few restrictions on dietary sodium intake or blood pressure medications. They demonstrated that screening for PA could be accomplished by measuring the morning ambulatory paired plasma aldosterone concentration (PAC) and plasma renin activity (PRA) for the aldosterone-to-renin ratio (ARR). On the basis of studies completed with the ARR as a case-detection test, experts on every continent have recognized that PA is more common than previously thought—affecting 5% to 10% of all hypertensive patients—and that most patients with PA are not hypokalemic.
Who Should Be Tested for PA?
Th e Endocrine Society clinical practice guidelines recommend that casedetection testing for PA should be performed in patient groups with a relatively high prevalence of PA:
• Blood pressure >160/100 mm Hg
• Drug-resistant hypertension
• Hypertension and spontaneous or diuretic-induced hypokalemia
• Hypertension with adrenal incidentaloma
• Hypertension and family history of early-onset hypertension or stroke at a young age (<40 years) • All hypertensive first-degree relatives of patients with PA Hypertension affects 68 million adults in the United States, and approximately 14 million persons in this population (21%) have treatmentresistant hypertension and should be tested for PA. Th us, case-detection testing should occur commonly in the offices of primary care providers and endocrinologists.
Cutoffs and Accuracy of CaseDetection Testing
Th e ARR is widely accepted as the screening test of choice for PA. Importantly, the lower limit of detection varies among different PRA assays and can have a dramatic effect on the ARR. Thus, the cutoff for a high ARR is laboratory-dependent and, more specifically, PRA assay–dependent. At Mayo Clinic, a PAC (in ng/dL) to PRA (in ng/mL per hour) ratio greater than 20 (>555 in SI units) and a PAC ≥15 ng/dL (≥416 pmol/L) are found in more than 90% of patients with surgically confirmed aldosterone-producing adenomas. In patients without PA, most of the variation in the ARR occurs within the reference range.
It is important for the clinician to recognize that the ARR is a case-detection tool with a sensitivity and specificity of approximately 75% and that most positive results should be followed by a confirmatory aldosterone suppression test to verify autonomous aldosterone production before subtype evaluation or treatment is initiated.
Some reference laboratories have changed renin measurement methodology from PRA to measurement of plasma renin concentration (PRC). It is reasonable to consider a result from a PAC:PRC case-detection test to be positive when the PAC is 15 ng/dL or greater (≥416 pmol/L) and the PRC is below the lower limit of detection for the assay.
Impact of Antihypertensive Medications
The ARR may be assessed while the patient is taking antihypertensive medications (with some exceptions [see below]) and without posture stimulation. Mineralocorticoid receptor (MR) antagonists (e.g., spironolactone and eplerenone) and high-dosage amiloride are the only medications that absolutely interfere with interpretation of the ARR, and they should be discontinued at least six weeks before testing.
Angiotensin-converting enzyme (ACE) inhibitors, angiotensin receptor antagonists (ARBs), and diuretics have the potential to falsely elevate PRA. Therefore, in a patient treated with an ACE inhibitor, ARB, or diuretic, the finding of a detectable PRA level or a low ARR does not exclude the diagnosis of PA. However, a very useful clinical point is that when a PRA level is undetectably low in a patient taking an ACE inhibitor, ARB, or a diuretic, PA should be highly suspect. Thus, ACE inhibitors, ARBs, and non–potassium-sparing diuretics do not need to be discontinued. A second important clinical point is that PRA is suppressed (<1.0 ng/mL per hour) in almost all patients with PA. Adrenergic inhibitors (e.g., β-adrenergic blockers and central α2 agonists) suppress renin secretion, but also in turn suppress aldosterone secretion (although to a lesser degree than renin) in normal individuals; thus, although the ARR may rise in hypertensive patients without PA treated with adrenergic inhibitors, the PAC remains less than 15 ng/dL (<416 pmol/L) and the casedetection test is not affected in a clinically important way.
CLINICAL RESEARCHER PERSPECTIVES
Th e correct diagnosis of primary aldosteronism (PA) and its subtypes is essential for optimal treatment. After a positive screening test, the clinician must proceed by A) establishing the diagnosis of PA with confirmatory testing; and B) performing subtype determination (lateralized aldosterone secretion versus bilateral aldosterone secretion) by adrenal imaging and adrenal venous sampling. Th is approach allows a targeted treatment which is highly effective in controlling hypokalemia and hypertension, reducing cardiovascular event incidence to that of matched hypertensive patients, and normalizing overall mortality.
Targeted therapy requires correct subtype identification. Th e great majority of patients (≥98%) have sporadic PA. In these patients unilateral aldosteronism (mostly aldosteroneproducing adenoma, APA) has to be distinguished from bilateral adrenal hyperplasia (IHA). Th e pretest probability for unilateral aldosteronism from an APA is around 70% in spontaneously hypokalemic patients. However, in the much more prevalent normokalemic cases APA is found in approximately 30% of cases. If subtype differentiation is incorrect, the patient will be subjected to inappropriate therapy: adrenalectomy in IHA or mineralocorticoid receptor blockade in APA.
Adrenal imaging by MRI or CT appears to be rather insensitive and nonspecific regarding subtype differentiation. Based on adrenal vein sampling (AVS) as the gold standard, up to 50% of patients may be categorized incorrectly based on imaging. Misleading results are mainly attributable to nonsecreting adrenal tumours (incidentalomas). Furthermore, the majority of hormone-producing adenomas are small (<1.5 cm) and may therefore escape notice even in thinsection images. In bilateral adrenal hyperplasia, imaging frequently shows normal-sized adrenal glands, and only sometimes demonstrates multiple small nodules. Therefore, many institutions, including our own, have adopted AVS as the standard in all patients who are candidates for surgery. There is no universal agreement as to how to perform AVS. Data linking postsurgical outcome with different protocols such as ACTH stimulation during AVS or simultaneous sampling from both adrenal veins are lacking. In the absence of evidence we prefer the simplest protocol: AVS without ACTH stimulation and with sequential sampling from adrenal veins. In our hands rapid cortisol determination is essential to verify correct catheter positioning during AVS. Using rapid cortisol determination routinely in our center has improved technical success rates of AVS to >90%. A standard operational procedure is necessary to avoid protocol violations, and has been in place in our institution since 2008. In addition, a multidisciplinary PA board to discuss AVS results of each patient is important because interpretation of AVS results remains a challenge. In our center, the decision for surgery is based on a lateralization index of ≥4.0. We do not request contralateral suppression for a surgical approach. Based on these criteria patients with suspected APA in Munich subjected to surgery have a 98% chance to enter biochemical remission (normalized serum potassium concentrations, normalized ARR, normalized aldosterone concentrations following sodium loading).
Patients with PA suffer from a multitude of comorbidities. Before targeted therapy is initiated, careful screening for metabolic, psychiatric, and cardiovascular disease is required. Metabolic complications include impaired glucose tolerance or frank diabetes. Patients with PA often have secondary hyperparathyroidism with consequent osteoporosis requiring vitamin D substitution. Longstanding secondary hyperparathyroidism may lead to tertiary hyperparathyroidism requiring parathyroidectomy. PA patients also suffer from increased anxiety and depression levels, which have to be adequately addressed. Prevalent cardiovascular disease such as atrial fibrillation, myocardial infarction, and stroke are more frequent than in matched hypertensive controls. Renal function is often impaired although aldosterone excess may obscure renal insufficiency via increased renal plasma flow.
Subtypes of primary aldosteronism require different therapeutic strategies. In our center, candidates for unilateral adrenalectomy are treated with mineralocorticoid antagonists pre-operatively. Th e rationale is to block aldosterone action, control hypertension, correct potassium losses, and increase plasma renin concentration to avoid postsurgical zona glomerulosa insufficiency with consecutive hyperkalemia. Evidence from randomized controlled trials supporting this strategy is still lacking.
Unilateral minimally invasive laparoscopic adrenalectomy is the standard in unilateral aldosteronism leading to reduced operating time and hospital stay. Morbidity of this procedure is very low, and mortality close to 0%. Nearly all patients will profit from surgery with regard to their postoperative blood pressure. Th e rate of remission from hypertension is higher for younger patients, female patients and patients with a short time between hypertension onset and diagnosis of PA (< 5 years). Whereas early studies using less stringent criteria for normotension reported remission rates of 70% and higher, more recent series reported remission in one-third of patients. The remaining patients have persistent hypertension, which is more easily controlled using fewer drugs. Cases in which surgery is not possible receive life-long long-term mineralocorticoid receptor blockade. Outcome with MR blockers is similar to surgery, and escape from treatment effects have rarely been reported. Bilateral adrenal hyperplasia is treated medically, as bilateral adrenalectomy in early series did not provide convincing improvement with regard to postoperative blood pressure. In addition, such a strategy requires lifelong hormone substitution with inherent risks from adrenal crisis. Medical treatment aims to antagonize aldosterone at the receptor level by the mineralocorticoid antagonists spironolactone, potassium canrenoate or eplerenone, with additional antihypertensives, as needed, to reach target blood pressure levels. Spironolactone is given at a low starting dose of 25 mg and titrated—depending on potassium and blood pressure—to 50 or 75 mg. Higher doses of spironolactone often induce side effects such as gynaecomastia, impotence, and menstrual disturbances, which result from its action on progesterone and androgen receptors. In the RALES study, 10% of male patients receiving 25 mg spironolactone developed gynecomastia/breast pain vs. 1% in the placebo group. Recently, remission from PA has been reported after longterm treatment with spironolactone in IHA. Compared to spironolactone, less adverse effects can be expected from the more selective mineralocorticoid antagonist eplerenone (off -label use for hypertensives in Europe). It has been the experience that eplerenone has to be used in 1.5 to 2 times higher doses than spironolactone to achieve a similar effect on blood pressure. If additional antihypertensive medication is needed, potassiumsparing diuretics such as triamterene or amiloride can be used. Reduction of salt intake supports medical treatment as adverse effects of aldosterone are intensifi ed in a high salt environment. Glucocorticoid-remediable aldosteronism ( familial hyperaldosteronism type I) is treated with low-dose dexamethasone (0.125-0.25 mg/day), which usually corrects hypertension, although biochemical alterations may persist. Patients can alternatively be given aldosterone antagonists or potassium-sparing diuretics, the latter especially being an option when treating children. Familial hyperaldosteronism type II is treated in the same way as sporadic cases of primary aldosteronism. Familial hyperaldosteronism type III caused by germ-line mutations in the potassium channel KCNJ5 often requires bilateral adrenalectomy because of severe intractable hypertension.
BASIC RESEARCHER PERSPECTIVE
Aldosterone plays a major role in regulating sodium and potassium homeostasis, and blood pressure. Primary aldosteronism (PA) results from autonomous aldosterone production from the adrenal cortex. Studies published in the last two years have uncovered the genetic causes of a subset of PA and highlighted the central role of ionic homeostasis and maintenance of zona glomerulosa cell membrane potential in the pathogenesis of the disease. These discoveries may pave the way toward better diagnostic approaches and new therapeutic opportunities, concerning up to 10% of the hypertensive population.
Mechanisms Controlling Aldosterone Biosynthesis
Aldosterone production from the zona glomerulosa (ZG) is tightly controlled to maintain electrolyte and fluid homeostasis by the kidney. Both angiotensin II (AngII) and potassium, the major regulators of aldosterone biosynthesis, act by increasing intracellular calcium concentrations, thus activating the calcium signaling pathway. This triggers a cascade of events ultimately leading to increased transcription of the CYP11B2 gene coding for aldosterone synthase. Th e maintenance of an appropriate zona glomerulosa cell membrane potential and intracellular ionic homeostasis is crucial to this mechanism because membrane depolarization leads to opening of voltage-gated calcium channels and an increase in intracellular calcium. In the zona glomerulosa the main ionic conductance is that of K+ , due to the expression of different types of K+ channels. Thus, the cell membrane potential closely follows the equilibrium potential of K+ over a large range of extracellular K+ concentrations. Th e concentration gradient of K+ between the intracellular and extracellular space that is required for the establishment of the membrane potential is generated by the activity of the Na+ , K+ -ATPase. Alteration of intracellular ionic homeostasis plays an important role in the pathogenesis of PA. Mutations affecting proteins involved in the regulation of zona glomerulosa cell membrane potential and ionic homeostasis have recently been identified in familial and sporadic forms of PA.
Familial Forms of PA
While the majority of cases of PA are sporadic, there exist three familial forms of hyperaldosteronism, accounting for 6%–10% of cases. In two of them, genetic diagnosis allows prompt and optimized therapeutic intervention. Familial hyperaldosteronism type I (FH-I), also called glucocorticoid suppressible aldosteronism (GRA), is characterized by early and severe hypertension, most often before the age of 20 years, in subjects with biochemical abnormalities of PA of variable intensity and in some cases adrenal nodules. FH-I is inherited as an autosomal dominant trait and is characterized by significant production of hybrid steroids (18-hydroxycortisol and 18-oxocortisol), and normalization of aldosterone levels and blood pressure with low doses of dexamethasone. The condition is due to formation of a hybrid gene resulting from unequal crossing over between CYP11B2 and the adjacent highly homologous gene CYP11B1, coding for steroid 11β-hydroxylase. Fusion of the promoter region of CYP11B1 to the coding region of CYP11B2 produces a chimeric gene, with the activity of aldosterone synthase, but tissue specificity and ACTH regulation of 11ß hydroxylase.
Familial hyperaldosteronism type II (FH-II) is also transmitted as an autosomal dominant trait, but is not associated with hybrid gene formation and, thus, hyperaldosteronism is not suppressible by dexamethasone. FH-II is clinically and biochemically undistinguishable from sporadic PA. The anatomic findings are variable from aldosterone-producing adenomas to bilateral adrenal hyperplasia. FH-II is diagnosed on the basis of two or more affected family members, and phenotypic variability within affected families is typical for the disease. In contrast to the other familial forms of PA, the genetic cause of FH-II has not been resolved so far, although a locus associated with the disease has been mapped to chromosome 7p22.
Potassium Channels and
ATPases in the Pathogenesis of
Familial and Sporadic PA
Recent work identified germline mutations in the KCNJ5 gene in a severe familial form of PA, familial hyperaldosteronism type III (FHIII), and similar recurrent somatic KCNJ5 mutations as cause of sporadic aldosterone producing adenomas. Patients with FH-III displayed early-onset resistant hypertension, profound hypokalemia, and very high levels of the hybrid steroids 18-oxocortisol and 18-hydroxycortisol. The disease was due to massive bilateral adrenal hyperplasia requiring bilateral adrenalectomy during childhood to control blood pressure. Subsequently, additional mutations and phenotypic variability of FHIII have been reported, with some patients presenting with mild PA resembling to FH-II.
KCNJ5 encodes the G proteinactivated inward rectifi er potassium channel GIRK4. Th e diff erent mutations identified in aldosteroneproducing adenomas and FH-III are all located near or within the selectivity filter of GIRK4 and affect the ion selectivity of the channel, with increased sodium conductance leading to sodium entry into the cell and chronic membrane depolarization. These changes are responsible for increased intracellular calcium leading to constitutive secretion of aldosterone and possibly cell proliferation. Somatic KCNJ5 mutations are present in 34%–47% of aldosterone producing adenoma samples from Western countries, and as high as 65% of patients from Japan. These mutations are associated with female gender, younger age, and a more severe phenotype of the disease. More recently, whole exome sequencing in APA has identified somatic mutations in two members of the P-type ATPase gene family, namelyATP1A1, encoding the α1 subunit of the Na+ , K+ -ATPase, and ATP2B3, coding for the plasma membrane calcium-transporting ATPase 3 (PMCA3). Mutations in the α1 subunit of the Na+ , K+ -ATPase led to a complete loss of pump activity and a strongly reduced affinity for K+ , while mutations of PCMA3 were predicted to affect intracellular calcium clearance. ATPase mutations were present in ~7% of a large series of tumors, but in contrast to KCNJ5 mutations, were more prevalent in males. Once again, mutation carriers had higher preoperative aldosterone levels compared to patients without mutation and lower serum potassium concentrations. Future studies will indicate whether carriers of somatic KCNJ5, ATP1A1, or ATP2B3 mutations may benefit from specific treatment options. Additional genetic investigation in large cohorts of patients will allow unraveling the missing genetics of the remaining 50% of APA, as well as discovering the susceptibility factors leading to bilateral adrenal hyperplasia, the other most common form of PA.