Therapeutic outcomes with surgical and medical management of primary aldosteronism

Abstract

Purpose of Review:

Primary aldosteronism (PA) is the most common cause of secondary hypertension, and is significantly under-diagnosed. Delays in diagnosis and treatment can lead to cardiovascular morbidity and mortality. The goal of this study is to review the management of PA, with a focus on medical and surgical treatment outcomes.

Recent Findings:

PA causes cardiovascular dysfunction in excess of degree of hypertension. Adrenalectomy demonstrates a therapeutic advantage over mineralocorticoid antagonist (MRA) therapy, after controlling for degree of hypertension and subtype of PA. Higher rates of atrial fibrillation, heart failure, stroke, and incidence of chronic kidney disease are observed in subjects treated with MRAs than in subjects treated with adrenalectomy. The therapeutic benefit of surgery may reflect definitive resolution of excess aldosterone. Complete mineralocorticoid blockade may achieve similar benefit to adrenalectomy.

Summary:

Adrenalectomy is the most effective treatment for unilateral PA. Biomarkers for MRA therapy might inform optimal medical therapy of bilateral adrenal hyperplasia.

Introduction

Once thought to be rare, primary aldosteronism (PA) is now recognized as the most common cause of secondary hypertension.[1, 2] PA affects 3–13% of hypertensive patients, and 5.5–20% of patients with resistant hypertension in population-based studies.[2–5] Aldosterone causes cardiac and renal dysfunction in excess of degree of hypertension, increasing morbidity and mortality.[6–12] In PA, duration of hypertension is a risk factor for irreversible end-organ-dysfunction, and therefore early diagnosis and optimal management are of paramount importance.[13–17] The goal of this review it to summarize key points in diagnosis and evaluation of PA, and to discuss evidence-based medical and surgical management strategies.

Establishing a diagnosis: who to screen, and how to screen

First described by Conn in 1955, PA is characterized by autonomous hypersecretion of aldosterone.[18] Hyperaldosteronism causes inappropriate retention of sodium and fluid with concurrent potassium excretion. Although the historic presentation includes both hypertension and hypokalemia, current studies suggest that hypokalemia is present in a minority of patients.[2, 19, 20] Early PA may present with mild or minimal blood pressure elevation, and be misdiagnosed as primary (essential) hypertension. The non-unique clinical presentation and misperceptions of disease rarity likely contribute to substantial under-diagnosis of PA. One recent study suggests that only 1–2% of patients with PA are appropriately screened and diagnosed.[21]

One of the major limitations in determining which populations may benefit from screening for PA is the wide variability in prevalence estimates.[22] A recent meta-analysis suggested that study heterogeneity is too great to allow accurate estimates of PA prevalence.[5] In the absence of precise prevalence figures, Endocrine Society Guidelines suggest screening for PA in high-risk groups, including patients with moderate to severe hypertension, medication-resistant hypertension, hypertension with hypokalemia, hypertension with an adrenal mass, and hypertension with sleep apnea.[23] Studies suggest limited awareness of screening guideline amongst general practitioners; this may present an opportunity to increase rates of PA screening.[24]

Initial screening for PA consists of a plasma aldosterone renin ratio, calculated as the plasma aldosterone concentration (PAC) over the plasma renin activity (PRA). Although laboratory values may vary, a diagnosis of PA is suspected in the setting of an elevated PAC, a suppressed PRA, and an ARR which is >20 (ng/dl)/(ng/ml/hr) or >555 (pmol/l)/(ng/ml/hr).[25] The list of medications which may affect renin and aldosterone secretion is extensive, however these typically cause false-negative rather than false-positive results.[26] A PAC >20 ng/dl with PRA<1 ng/ml/hr in a patient with hypertension and hypokalemia is diagnostic of PA; confirmatory testing is suggested for all other patients. The most commonly utilized confirmatory studies include the oral sodium loading test and the intravenous saline infusion test; the fludrocortisone suppression test and captopril stimulation test are less commonly used. The oral sodium loading test has a sensitivity of 96% and specificity of 93%, and has the advantage of being performed as an outpatient, and therefore is the preferred confirmatory test for many centers.[27] Intravenous saline infusion has a sensitivity of 88% and specificity of 84%, and requires a 4 hour continuous infusion, but may have utility in specific clinical scenarios.[28]

Key to management: adrenal vein sampling

Etiologies of PA include unilateral and bilateral adrenal hypersecretion. Unilateral causes of PA include aldosterone secreting adenoma (APA) and unilateral adrenal hyperplasia. Bilateral disease is typically due to bilateral adrenal hyperplasia (BAH). The standard of care for subtype differentiation in PA is adrenal vein sampling (AVS), due to the high prevalence of concomitant non-functional adrenal adenomas which may be indistinguishable from APA on cross-sectional imaging.[23, 29] One subgroup who may present an exception are young PA patients (<35 years of age) with unilateral adrenal adenomas >1 cm in size and a normal contralateral adrenal gland. As the prevalence of adrenal adenomas is low in this population, AVS is unlikely to change management and therefore this subgroup may proceed directly to adrenalectomy.[30] Although technically demanding, complication rates in AVS are low (0.6%) and the procedure provides key functional data regarding the source of aldosterone hypersecretion. [31] Subjects with lateralization on AVS are likely to benefit from adrenalectomy, while those who do not lateralize, consistent with BAH, are managed medically. Lower cure rates are seen when adrenalectomy is performed based on imaging findings rather than AVS, likely reflecting inaccurate subtype differentiation resulting in inappropriate surgery.[32]

Resecting the adrenal gland: How, and how much?

Laparoscopic adrenalectomy is the standard of care for APA and unilateral hyperplasia, and the overwhelming majority of subjects experience clinical improvement with adrenalectomy.[23, 33] Treatment outcomes will be discussed separately, below. Numerous studies have demonstrate that laparoscopic adrenalectomy is safe, with mean complication rate of 4.7%; in one systematic review of 1056 patients there were no reported high-grade complications or deaths.[34] Rates of conversion to open surgery range from 0–2% in modern series.[34–36] Laparoscopic adrenalectomy may be performed via transabdominal or retroperiotoneoscopic approach; the techniques appear equivalent with regard to safety and efficacy [37], and therefore the approach may be chosen based on patient factors and surgeon expertise.

Cortical sparing or partial adrenalectomy is not standard of care for APA, however it appears safe and may be considered in select patients. The primary concern is that partial adrenalectomy may not resect micro- or multinodular disease, leading to persistent hypertension. The clinical data are ambiguous. In one retrospective study, 10% of subjects undergoing partial adrenalectomy demonstrated persistent hyperaldosteronism, and 27% of excised adrenal glands contained multiple nodules.[38] Two randomized control trials demonstrate equivalent [39, 40] clinical outcomes for partial adrenalectomy when compared with total adrenalectomy for PA. One recent multi-center retrospective study suggested higher rates of complete clinical resolution in a retrospective cohort of patients undergoing minimally invasive surgery.[41] Of note, the retrospective study cohort demonstrated significantly larger tumors in the total adrenalectomy group, and subgroups were not controlled for comorbid diseases which have been shown to contribute to poorer clinical outcomes; therefore these findings must be interpreted with care. Future research may clarify these questions. Based on currently available data, we would suggest consideration of partial adrenalectomy only in subjects with a unilateral, well-defined tumor and a contraindication (such as prior contralateral adrenalectomy) to total adrenalectomy.

Several studies have described radiofrequency or thermal ablation of the adrenal gland as a treatment for PA.[42–44] One retrospective study of 44 subjects demonstrated similar outcomes between adrenal ablation and adrenalectomy,[45] while a retrospective study of 63 patients had poorer blood pressure control compared to laparoscopic adrenalectomy.[46] No randomized control data exists to support this practice, and the techniques are not widely available. At specialized centers with access to advanced proceduralists, adrenal ablation may be considered as an alternative treatment for subjects with APA who are not surgical candidates.

Medical treatment

Optimal medical treatment for BAH is focused on blood pressure control and management of serum potassium. Although treatment must be tailored to the individual, the recommended approach includes a multi-drug regimen including a mineralocorticoid antagonist (MRA). Both selective (eplerenone) and nonselective (spironolactone) MRAs are effective, although spironolactone has been the mainstay of therapy for the past five decades due to cost and availability.[23, 47] MRAs are typically titrated to normalization of blood pressure, however titration to a normal renin level may be associated with better outcomes.[48] MRAs must be dosed with care in the setting of chronic kidney disease, as they may precipitate hyperkalemia.[49] Epithelial sodium channel blockers, calcium channel blockers, and other anti-hypertensive medications may be used in conjunction with MRAs.[47] Similar to primary hypertension, dietary sodium restriction is recommended in PA, and salt restriction may act synergistically with MRAs.[50–52] However, the marked degree of dietary restriction required to normalize aldosterone to renin ratios may be unsustainable on a practical level and therefore should be considered an adjunctive rather than primary treatment for PA.[53]

Outcomes: How do medical and surgical management compare?

A lack of standardized outcome measures has historically presented a significant barrier to assessing response to therapy in PA. Although the majority of subjects benefit from both medical and surgical therapy in observational studies, the absence of unifying criteria for therapeutic response has limited comparative investigations. In 2017, the PASO trial published explicit criteria for biochemical and clinical outcomes after adrenalectomy, including standards for systolic and diastolic blood pressure improvement.[33] The complexity of the PASO criteria significantly limits their utility in routine clinical settings, however they may provide standardized benchmarks for future trials. Moreover, the PASO criteria do not reflect the secondary outcomes associated with PA, including cardiovascular and renal dysfunction, which are arguably of significant clinical import.

Perhaps reflecting the ambiguity in blood pressure outcomes, many recent investigations of therapeutic outcomes in PA evaluate alternative endpoints, including target organ function, quality of life, and overall survival. Several recent studies evaluating adrenalectomy versus MRA therapy are discussed below, and summarized in Table 1. Due to a dearth of randomized clinical trials, the best available evidence is based on predominantly retrospective and a few prospective observational studies.

Table 1.

Summary of selected recent studies (2016–2020) comparing medical and surgical treatment of PA.

Study	Year	Study Design	Primary Outcome	Study Groups	Results
Wu et al, Sci. Rep. [69]	2016	Retrospective cohort	All-cause mortality	PA +ADX (n=822), matched 1:1 on comorbidities with PA +MRA (n=822)	ADX demonstrated survival benefit (HR: 0.23, p<0.001 with residual hypertension; HR: 0.21, p<0.001 without residual hypertension)
Hundemer et al, Hypertension [67]	2018	Retrospective cohort	Renal function	PA +ADX (n=120), PA +MRA (n=400), age and GFR-matched PH (n=15474)	ADX was not associated with worse renal outcomes compared to EH; MRA treatment in PA associated with annual decline in GFR (−1.2 ml/min/1.73 m2/y, p<0.001) and higher risk of CKD (HR 1.63) compared to PH
Rossi et al, Hypertension [64]	2018	Longitudinal observation cohort	Atrial fibrillation	PA with lateralizing AVS +ADX (n=41), PA with idiopathic hyperaldosteronism +MRA (n=66)	ADX associated with lower risk of atrial fibrillation compared to MRA (HR: 1.82, p=0.025)
Hundemer et al, JAMA Cardiol. [62]	2018	Retrospective cohort	Atrial fibrillation	PA +ADX (n=201), PA +MRA (n=195), age-matched PH (n=40,092)	PA +MRA with suppressed renin (<1 ng/ml/hr) had higher risk of atrial fibrillation (HR: 2.55, 95% CI: 1.75–3.71) compared to PA +ADX, PA +MRA with non-suppressed renin, and PH
Velema et al, JCEM [73]	2018	Post hoc comparative effectiveness study within randomized clinical trial (SPARTACUS)	Quality of life (QoL)	PA +ADX (n=92), PA +MRA (n=92)	Both ADX and MRA treatment improved QoL after 1 year. ADX associated with better QoL compared to MRA despite equivalent blood pressure control
Katabami et al, J Hypertension [68]	2019	Retrospective cohort	Renal function, blood pressure, anti-hypertensive medication use	PA with APA +ADX (n=276), PA with APA +MRA (n=63)	After 6 months, ADX associated with lower number of antihypertensive medications, higher rates of normal blood pressure, and improved GFR compared with patients on MRA (−12.6 versus −9.0 ml/min per 1.73 m2, p=0.003).
Chang et al, Surgery [76]	2020	Retrospective cohort	Stroke	PA with APA +ADX (n=1047), PA +MRA (n=3167), each subgroup matched 1:4 with PH	ADX associated with lower risk of stroke compared to subjects with PH (HR: 0.53, p<0.001); MRA was not (HR: 1.66, p<0.001)
Pan et al, J Amer. Heart Assoc. [61]	2020	Retrospective cohort	New onset atrial fibrillation	PA +ADX (n=1101), PA +MRA (n=1101), matched with PH (N=8808)	ADX associated with lower risk of atrial fibrillation compared to subjects with PH (HR: 0.29, p=0.014); MRA was not (HR: 1.19, p=0.240)

ADX – Adrenalectomy; APA – Aldosterone producing adenoma; CKD – Chronic kidney disease; GFR – Glomerular filtration rate; HR – Hazard Ratio; MRA – Mineralocorticoid antagonist; PA – Primary aldosteronism; PH – Primary hypertension; QoL – Quality of life

Cardiovascular outcomes.

The cardiovascular system responds dynamically to aldosterone. PA is associated with vascular and cardiac remodeling beyond the degree of hypertension, which is the presumed mechanism for the higher rates of cardiovascular events observed in PA. [6, 7, 10–13] Numerous retrospective and several prospective studies have shown that PA is associated with higher risks of atrial fibrillation,[11, 12, 54–57] coronary artery disease,[11, 12, 54–59] left ventricular hypertrophy and heart failure [11, 54, 55, 57, 58] when compared to primary hypertension. One meta-analysis of 31 studies demonstrated higher risk of atrial fibrillation (OR: 3.52), coronary artery disease (OR: 1.77), heart failure (OR: 2.05), left ventricular hypertrophy (OR: 2.29) and stroke (OR: 2.58) in untreated PA compared to primary hypertension.[60]

Several recent retrospective studies have evaluated cardiovascular outcomes in medical versus surgical therapy. One investigation showed that compared to matched hypertensive controls, patients with PA (n=1101) who underwent adrenalectomy had lower rates of new onset atrial fibrillation (HR: 0.29, p=0.014). MRA treatment (n=1101) was not associated with this protective effect (HR: 1.19, p=0.240).[61] This finding may reflect insufficient blockade of the mineralocorticoid receptor with MRAs, as MRA titration to normal renin is associated with increased efficacy. One observational study demonstrated a 2.5-fold increase in atrial fibrillation in PA subjects treated with MRAs with a suppressed renin, while PA subjects treated with adrenalectomy or MRA to unsuppressed renin were comparable to hypertensive controls.[62] These data support the hypothesis that persistent renin suppression may serve as a biomarker for inadequate mineralocorticoid receptor blockade, leading to ongoing aldosterone exposure and consequent cardiometabolic effects.[63]

Although randomized control trials are lacking, some limited prospective data is available for atrial fibrillation.[64] In one Italian study, 1125 sequential hypertensive patients were prospectively enrolled and screened with aldosterone and renin levels. Approximately 10% (107 subjects) were diagnosed with PA. On AVS, 41 subjects had APA and were treated with adrenalectomy, while 66 did not lateralize on AVS and were treated with MRA. At a median of 11.8 years follow up, adrenalectomy was associated with lower risk of atrial fibrillation compared to MRA (HR: 1.82, p=0.025). There was no significant difference in survival in subjects with PA and those with primary hypertension.

Renal outcomes.

The renal system is a direct target organ of hyperaldosteronism, and PA is associated with decreased glomerular filtration (GFR) and chronic kidney disease (CKD).[9, 65, 66] Several studies have investigated the extent to which treatment with MRA or adrenalectomy mitigates risk of kidney disease. In one large observational cohort study, despite comparable blood pressure control, subjects with PA treated with MRA had higher risk of incident chronic kidney disease compared to patients with primary hypertension (HR 1.63, 95% CI 1.33–1.99).[67] GFR declined at greater annual rates in the MRA treated PA patients (−1.6 ml/min per 1.73 m²/y) than the subjects with primary hypertension (−0.9 ml/min per 1.73 m²/y, p<0.001). In contrast, PA subjects treated with adrenalectomy were not at higher risk of incident CKD or annual decline in GFR.

As standard of care is adrenalectomy for APA and MRA treatment for BAH,[23] the majority of studies comparing medical and surgical outcomes are inherently also comparing subtypes. A major critique of observational studies therefore is that the benefits observed with adrenalectomy may reflect an underlying phenotypic difference between APA and BAH. One recent study sought to isolate the treatment effect by including only subjects with confirmed APA based on AVS.[68] Of 339 APA subjects, 276 underwent adrenalectomy and 63 were treated with MRA therapy. After 6 months, subjects who underwent adrenalectomy had significantly lower utilization of antihypertensive medications, higher rates of normalization of blood pressure, and improved GFR compared with patients on MRA (−12.6 versus −9.0 ml/min per 1.73 m², p=0.003). There was no difference between groups in rates of proteinuria.

Quality of life and survival.

PA is associated with increased risk of death in observational studies, even after controlling for degree of hypertension.[48, 59] Although direct causality cannot be established due to the retrospective nature of observational studies, the presumptive cause would be increased cardiovascular disease, as discussed above. The data are somewhat limited regarding survival outcomes in medical versus surgical management. One retrospective cohort study of PA subjects (n=822) who underwent adrenalectomy were matched 1:1 with control patients treated with MRAs, showed significant survival benefit in the adrenalectomy group (HR: 0.23, p<0.001 with residual hypertension; HR: 0.21, p<0.001 without residual hypertension).[69] Interestingly, adrenalectomy may improve the quality of life as well as prolong it.[70, 71] Observational studies have shown an association between health related quality of life, mental health, sleep disorders and PA.[72] One recent study examining quality of life utilized a post-hoc analysis of the Subtyping Primary Aldosteronism: A Randomized Trial Comparing Adrenal Vein Sampling and Computed Tomography Scan (SPARTACUS) prospective trial.[73] In this investigation, both adrenalectomy and MRAs improved quality of life after one year of treatment; however PA patients treated with adrenalectomy had better quality of life despite equivalent blood pressure control. This difference appeared to be particularly significant in female subjects. A PA-specific QoL questionnaire has recently been developed to further investigate quality of life and response to therapy.[74] This line of investigation may help guide patient counseling and expectations.

Summary and future directions.

These studies and others in the growing body of literature support the therapeutic advantage of adrenalectomy in PA. The observed benefit may reflect more complete elimination of excess aldosterone, limiting ongoing cardiometabolic injury. This hypothesis is consistent with our evolving understanding of aldosterone as a metabolically active hormone with complex systemic effects. Although hypertension is the most prominent sign of PA, targeting therapy to hypertensive control may fail to address concomitant manifestations of the disease, including atrial fibrillation, heart failure, stroke, and renal insufficiency. The early observational data suggests that complete MRA blockade to non-suppressed renin levels may achieve similar therapeutic benefit to surgery in cardiovascular outcomes. Further prospective study of this finding might be of substantial benefit to patients with BAH. Somewhat controversially, the logical extension of the concept of minimizing aldosterone exposure has led some authors to propose managing selected bilateral disease with unilateral adrenalectomy.[63] Proponents of this approach suggest that it may be of particular benefit in young patients where attenuating the severity of their disease may prevent long-term morbidity. While it seems unlikely that any randomized control trials will be developed to study this management strategy, the concept of a combined medical and surgical approach is an interesting one, and has parallels in other endocrine disorders such as hyperparathyroidism where debulking surgery is routinely performed for multigland hyperplasia.[75] Given the very safe profile of laparoscopic adrenalectomy this may be a topic of future studies.

Conclusions

Primary aldosteronism is the most common cause of secondary hypertension and is significantly under-diagnosed. Aldosterone hypersecretion causes cardiometabolic injury in excess of degree of hypertension. Although target organ damage is partially reversible with therapy, the duration of untreated primary aldosteronism is a risk factor for cardiovascular and renal morbidity, including atrial fibrillation, heart failure, stroke, and chronic kidney disease. A high clinical index of suspicion is therefore essential for early diagnosis. Optimal management is predicated on accurate subtype differentiation with adrenal vein sampling. The standard of care for aldosterone producing adenomas is laparoscopic adrenalectomy, while bilateral adrenal hyperplasia is treated with mineralocorticoid receptor antagonists. A growing body of literature supports the therapeutic advantage of surgery over medical management; this may reflect definitive resolution of excess aldosterone exposure. Recent research suggests that complete mineralocorticoid blockade may achieve similar therapeutic benefit, and renin might serve as a biomarker for adequate mineralocorticoid receptor antagonism. Future prospective studies might inform optimal medical therapy, and develop novel combined medical and surgical approaches.