Introduction
Primary aldosteronism (PA) is a disorder of dysregulated aldosterone production that
occurs independently from the physiological renin-angiotensin system. PA is a common
and modifiable cause of secondary hypertension [1]. Compared with equivalent primary hypertension, PA has been associated
with a disproportionately greater risk of cardiovascular, metabolic, and renal
complications [2]
[3]
[4]
[5]
[6]
[7], mediated via inappropriate mineralocorticoid receptor (MR) activation
within target tissues. Directed therapies, such as unilateral adrenalectomy or MR
antagonists, have been shown to mitigate unfavorable outcomes in patients with PA
[3]
[8]
[9]
[10]
[11]. While determining the true prevalence
of PA hinges on a number of variables, such as the population studied, clinical
setting, tests, and criteria used to define PA [12]
[13]
[14]
[15]
[16], compelling evidence
suggests that PA spans a spectrum of severity, of which only the very late stages
are typically recognized. In this review, we discuss the reported frequency of PA
across its pathophysiological spectrum and various populations at risk.
The evolving definition of primary aldosteronism
In 1955, Dr. Jerome Conn published his clinical journey to solving the first
documented case of PA in a 34-year-old woman with severe hypertension, profound
hypokalemia, metabolic alkalosis, and intermittent tetany, which resolved after
unilateral adrenalectomy [17].
Although hypokalemia and metabolic alkalosis became hallmarks of PA, Dr. Conn
acknowledged that such findings only capture the most severe forms of the
disease. Moreover, as his expertise in PA evolved, Dr. Conn correctly
anticipated that most patients with PA have normokalemic hypertension [18] and that the pathophysiology of PA
might emerge in normotensive stages prior to clinically advanced manifestations
[19]
[20]. In the years that followed,
numerous studies revealed the presence of autonomous aldosterone production
among both hypertensive and normotensive individuals, supporting a broad
spectrum of PA severity [21]
[22]
[23]
[24]. Following the introduction of the
aldosterone-to-renin ratio (ARR) as a screening tool for PA in 1981, however,
this initial paradigm of a PA severity continuum shifted towards a more
pragmatic, binary disease classification [25]. Although intended to help standardize the diagnostic process,
the diagnostic criteria for both ARR and the subsequently implemented
confirmatory tests have been arbitrary and heterogeneous. An ARR of≥30
(ng/dL)/(ng/mL/h), along with a plasma
aldosterone concentration≥15 ng/dL, constitute the most
conservative criteria for a positive PA screening test [12]. Less stringent criteria increase
the detection rates of both advanced and mild forms of PA [26]
[27]
[28]
[29]. Two recent systematic reviews of
ARR testing found that the clinical performance of ARR varied widely with the
patient population, clinical setting, and thresholds used, rendering
sensitivities from 10% to 100% [30]
[31].
Considering the dynamic fluctuations of renin and aldosterone in response to a
large number of determinants [32]
[33]
[34], the current Endocrine Society
guideline proposes confirmatory testing in most patients with positive screening
results to solidify the PA diagnosis [12]. Notably, however, PA has been diagnosed in individuals with ARR
values lower than the commonly used threshold for PA, indicating the limitations
of arbitrary screening criteria in capturing the broad continuum of
inappropriate aldosterone production. Using fludrocortisone-dexamethasone
suppression tests, Markou et al. diagnosed PA in 13 of 100 normotensive women,
regardless of baseline plasma aldosterone or ARR values, and 11 of these women
developed hypertension within five years, supporting an incipient PA
pathophysiology stage in these participants [35]. In a similar study, elevated 24-h
urine aldosterone excretion suggestive of PA was found following oral sodium
loading in a substantial number of individuals across the entire blood pressure
spectrum, including in individuals with ARR considered normal [36]. Nevertheless, the definition of
inappropriate aldosterone production relies, yet again, on arbitrary thresholds
for each confirmation test. Variability in diagnostic tests and criteria
explains inconsistencies in the reported prevalence of PA, which varied from
3.2% to 12.7% in the primary care setting, and from 1%
to 29.8% in referral centers [37].
Beyond the heterogeneity of the diagnostic tests, PA prevalence reports also
differ across various populations. Generally, the risk of PA is highest in
patients with severe hypertension, and hypertension and hypokalemia. High PA
prevalence has also been reported in individuals with hypertension and adrenal
nodules, obstructive sleep apnea (OSA), or a family history of PA, and several
expert guidelines recommended PA screening in these populations [12]
[38]
[39]
[40].
The reported prevalence of overt primary aldosteronism in treatment-resistant
hypertension
Although reports of PA prevalence among patients with hypertension have been
heterogeneous due to the vast differences in PA definition between studies,
findings consistently suggest that PA prevalence increases in parallel with
hypertension severity, ranging from 11–22% in the individual
with treatment-resistant hypertension ([Table 1]) [14]
[27]
[36]
[41]
[42]
[43]
[44]
[45]
[46]. On average, only under
3–4% of individuals with treatment-resistant hypertension are
screened for PA [14]
[16]
[47]
[48]. When conducted, PA screening
usually targets patients with multiple risk factors and those who have already
developed cardiovascular and renal complications [47]. Physicians with secondary
hypertension expertise, including endocrinologists and nephrologists, who
typically also perform confirmatory testing and oversee the management of PA,
are more likely to consider and screen for PA than primary care providers [48]. As very few patients with
hypertension are referred to specialists, the true prevalence of PA across
hypertension stages in the general population remains uncertain. In a
meta-analysis of 39 studies of adults with hypertension, the weighted mean
prevalence of overt PA was 5.5% for high-normal blood pressure,
4.2% for stage 1-, 10.2% for stage 2-, and 16.4% for
stage 3 hypertension [37].
Table 1 Studies of overt primary aldosteronism prevalence
across the blood pressure spectrum.
|
Reference
|
Criteria
|
Study population
|
Prevalence of overt PA
|
|
Calhoun 2002 [41]
|
PRA<1 ng/mL/h AND
Positive oral sodium suppression test
|
Resistant hypertension
|
20%
|
|
Mosso 2003 [42]
|
ARR>25
(ng/dL)/(ng/mL/h) AND
Positive fludrocortisone suppression test
|
Hypertension (overall)
|
6.10%
|
|
Stage 1 hypertension
|
2%
|
|
Stage 2 hypertension
|
8%
|
|
Stage 3 hypertension
|
13.20%
|
|
Rossi 2006 [43]
|
ARR≥40
(ng/dL)/(ng/mL/h) AND
Positive captopril challenge test and/or LDF
score#≥0.5
|
Hypertension (overall)
|
11.20%
|
|
Stage 1 hypertension
|
6.60%
|
|
Stage 2 hypertension
|
15.50%
|
|
Stage 3 hypertension
|
19%
|
|
Douma 2008 [44]
|
ARR>20
(ng/dL)/(ng/mL/h)*
with ALD>15 ng/dL*
AND Positive saline and fludrocortisone
suppression test AND Responsive to spironolactone
|
Resistant hypertension
|
11.30%
|
|
Ito 2011 [49]
|
ARR>20
(ng/dL)/(ng/mL/h) AND
Positive captopril challenge test
|
Prehypertension
|
6.80%
|
|
Stage 1 hypertension
|
3.30%
|
|
Markou 2013 [35]
|
Positive fludrocortisone-dexamethasone suppression test
|
Normotension
|
13%
|
|
Galati 2016 [45]
|
ARR>20
(ng/dL)/(ng/mL/h) with
ALD>10 ng/dL with
PRA<1 ng/mL/h
|
Hypertension (overall)
|
4.70%
|
|
Resistant hypertension
|
10.50%
|
|
Monticone 2017 [27]
|
ARR>30
(ng/dL)/(ng/mL/h) with
ALD>10 ng/dL AND Positive
saline infusion or captopril challenge test
|
Hypertension (overall)
|
5.90%
|
|
Stage 1 hypertension
|
3.90%
|
|
Stage 2 hypertension
|
9.70%
|
|
Stage 3 hypertension
|
11.80%
|
|
Brown 2020 [36]
|
Positive oral sodium suppression test with post-test urine
ALD≥12 µg/day
|
Normotension
|
11.30%
|
|
Stage 1 hypertension
|
15.70%
|
|
Stage 2 hypertension
|
21.60%
|
|
Stage 3 hypertension
|
22%
|
|
Xu 2021 [46]
|
ARR≥30
(ng/dL)/(ng/mL/h)*
with ALD>10 ng/dL AND
Positive saline infusion or captopril challenge test
|
Hypertension (overall)
|
3.30%
|
|
Stage 1 hypertension
|
0.70%
|
|
Stage 2 hypertension
|
3.40%
|
|
Stage 3 hypertension
|
12.60%
|
|
Zekarias 2022 [14]
|
ARR≥30
(ng/dL)/(ng/mL/h) with
ALD≥10 ng/dL
|
Resistant hypertension
|
16.90%
|
#Logistic discriminant function (LDF) score was described
elsewhere. [97];
*ARR or ALD values have been converted to common
units of measurement for ease of comparison.; Abbreviation; ARR,
aldosterone-renin-ratio; ALD, aldosterone; PA, primary aldosteronism;
PRA, plasma renin activity; LDF, logistic discriminant function.
The reported prevalence of overt primary aldosteronism in normotension and
mild hypertension
Current guideline-directed PA screening may overlook milder cases of autonomous
aldosterone production in the pathophysiological continuum of PA, which results
in an underestimation of its prevalence among individuals with mild
hypertension. Ito and colleagues performed systematic screening of PA during
annual health examinations in adults without hypertension and found that
22.7% of prehypertensive subjects (BP
120–139/80–89 mmHg) had a positive ARR
screening, and 6.8% had confirmed PA with captopril suppression test
[49]. Other observational studies
have provided evidence for PA in a subset of individuals with normal or mildly
elevated blood pressure ([Table 1]).
Importantly, these unrecognized yet biochemically overt PA forms have been
linked to unfavorable outcomes. During a four-year observation period in the
Framingham Offspring Study, normotensive participants with higher aldosterone
levels were more likely to develop incident and worsening hypertension [50]. Similar studies revealed that
normotensive participants with suppressed plasma renin activity had a
significantly higher risk of developing incident hypertension when compared with
those without suppressed renin [51]
[52]. These results
indirectly suggest that early detection of autonomous aldosterone secretion and
targeted therapeutic measures might mitigate cardiovascular complications
mediated by inappropriate MR activation. However, the paucity of data on the
cost-effectiveness of early screening and directed therapy make strategic
interventions of uncertain value in early hypertension stages.
The reported prevalence of overt primary aldosteronism in patients with
hypokalemia
Numerous reports have assessed the prevalence of hypokalemia in patients with
overt PA. In a meta-analysis of 39 observational studies, hypokalemia was
presented in 0–37.5% of patients with PA seen in primary care
and 0–67% of those from referral centers [37], with the highest proportion in
patients with aldosterone-producing adenoma [43], In contrast, the prevalence of PA
among individuals with hypertension and hypokalemia has been poorly studied.
Burrello and colleagues [53]
retrospectively reviewed the medical records of 840 hypertensive patients with
at least a single episode of hypokalemia (potassium
level<3.7 mmol/L) and found that the rates of overt PA
increased with the decreasing serum potassium levels. Overt PA was diagnosed in
28.1% of the overall cohort and in 88.5% of patients with severe
hypokalemia (potassium level<2.5 mmol/L). Additionally,
patients with spontaneous hypokalemia had overt PA more often than those with
diuretic-induced hypokalemia (37.4% versus 16.5%, respectively)
[53].
Hypokalemia was previously viewed as a hallmark of advanced PA [12]. Several reports, however, present
normotensive patients diagnosed with hypokalemia prior to recognized PA [54]
[55]
[56]
[57]. The mechanism underlying this
entity is poorly understood. High vascular compliance, and an adaptative
increase in renal sodium excretion in response to aldosterone excess, to
maintain normal blood pressure, have been proposed as possible mechanisms [57]. Since PA has not been
systematically screened in patients with isolated hypokalemia without
hypertension, the prevalence of normotensive hypokalemic PA remains unknown.
The reported prevalence of overt primary aldosteronism in patients with
adrenal incidentaloma
In observational studies of patients with adrenal incidentaloma, the reported
prevalence of overt PA ranges between 1 and 30%, depending on the study
design and the applied diagnostic criteria ([Table 2]) [58]
[59]
[60]
[61]
[62]
[63]
[64]
[65]
[66]
[67]. Although overt PA is more
frequently found in patients with adrenal incidentaloma and hypertension than in
normotensives [12]
[68]
[69]
[70], it should be emphasized that
adrenal incidentalomas are not always the source of aldosterone. Several studies
have found high discrepancy rates between imaging findings and adrenal vein
sampling (AVS) results [71]
[72]
[73]. Moreover, immunohistochemistry
targeting aldosterone synthase (CYP11B2) has demonstrated that macroscopic
nodules do not always produce aldosterone [74]. Conversely, microscopic CYP11B2-positive areas are not uncommon
in patients with overt PA, including when cross-sectional imaging detects no
adrenal abnormalities [75]
[76].
Table 2 Studies of overt primary aldosteronism prevalence
in patients with adrenal incidentaloma.
|
Reference
|
Criteria
|
Study population
|
Prevalence of overt PA
|
|
Mantero 2000 [58]
|
ARR elevation and suppressed PRA
|
Adrenal incidentaloma with exclusion of severe hypertension
and hypokalemia
|
1.6% (all were hypertensive)
|
|
Bernini 2002 [59]
|
ARR>20
(ng/dL)/(ng/mL/h)*
AND Positive saline suppression or captopril
challenge test
|
Adrenal incidentaloma with exclusion of hypokalemia
|
4% (all were hypertensive), 5.5% of
hypertensive patients
|
|
Kim 2005 [60]
|
Not specified
|
Adrenal incidentaloma
|
10% (all were hypertensive), 25% of
hypertensive patients
|
|
Comlekci 2010 [61]
|
ARR>25
(ng/dL)/(ng/mL/h) AND
Positive saline Suppression test
|
Adrenal incidentaloma with hypertension
|
4.4%
|
|
Kim 2013 [62]
|
ARR>20
(ng/dL)/(ng/mL/h)
|
Adrenal incidentaloma
|
4.6%
|
|
Patrova 2015 [63]
|
ARR (not specified threshold)
|
Adrenal incidentaloma with hypertension
|
1.4%
|
|
Tabuchi 2016 [64]
|
ARR>20
(ng/dL)/(ng/mL/h) AND
Positive one of the confirmation tests
|
Adrenal incidentaloma
|
10.7% (93.3% were hypertensive)
|
|
Stavropoulos 2018 [65]
|
Positive saline suppression test with post-test
ALD≥6 ng/dL
|
Adrenal incidentaloma
|
3.35% (all were hypertensive)
|
|
Jing 2022 [66]
|
ARR (not specified threshold) AND Positive saline
suppression or captopril challenge test
|
Adrenal incidentaloma
|
11.8% (76% were hypertensive and 12%
were hypokalemia), 23.2% of hypertensive patients,
75% of hypokalemia patients
|
|
Kmieć 2022 [67]
|
ARR>20
(ng/dL)/(ng/mL/h)*
with ALD>4 ng/dL
|
Adrenal incidentaloma with exclusion of overt PA
|
25.4% (56.3% were hypertensive) 24.3%
of hypertensive patients
|
*ARR or ALD values have been converted to common units
of measurement for ease of comparison.; Abbreviation; ARR,
aldosterone-renin-ratio; ALD, aldosterone; PRA, plasma renin activity;
PA, primary aldosteronism.
An unresolved question regarding the evaluation of adrenal incidentaloma is
whether we should screen for autonomous aldosterone production in patients who
do not have hypertension. It has been suggested that biochemically overt PA is
not rare in patients with adrenal incidentaloma and normal blood pressure [77]
[78]. Kmieć and colleagues
conducted a prospective study attempting to detect subclinical autonomous
aldosterone secretion in patients with adrenal incidentaloma by using permissive
criteria of ARR>1.2 (ng/dL)/(mIU/L)
(approximately 10 (ng/dL)/(ng/mL/h)) and plasma
aldosterone concentration>4 ng/dL. Normotensive patients
with urine aldosterone excretion>10 µg/day were
considered to have mild autonomous aldosterone secretion, while hypertensive
patients with urine aldosterone>12 µg/day were
diagnosed with overt PA. Overall, 25.4% of individuals with adrenal
incidentaloma were found to have mild autonomous aldosterone secretion
(24.3% among hypertensive and 26.9% in normotensive
participants) [67].
The reported prevalence of overt primary aldosteronism in patients with
obstructive sleep apnea
The pathophysiology of OSA has been associated with alterations in the
renin-angiotensin-aldosterone system [79]. Studies in animal models suggest that intermittent hypoxia and
acute hypercapnia, as observed in OSA, are linked to elevated plasma
aldosterone, independently of plasma renin activity [80]. Nonetheless, whether these
changes contributed to a higher prevalence of overt PA in patients with OSA
remains debatable ([Table 3]). In an
Italian study of individuals newly diagnosed with hypertension, patients with
OSA confirmed with polysomnography had a much greater prevalence of overt PA
(34%) than those without OSA (9.8%) [81]. Another single-center study in
Poland showed a 21.3% PA prevalence in 94 patients with hypertension and
moderate to severe OSA [82]. In
contrast, a substantially lower prevalence of overt PA (8.9%) was
reported in the HYPNOS trial, a multi-ethnic cohort study of 203 patients with
OSA [83], suggesting that the
prevalence of biochemically overt PA among individuals with OSA is comparable to
other difficult-to-control hypertensive populations, questioning the current
Endocrine Society recommendation to screen patients with OSA for PA. An
important caveat of the HYPNOS trial, however, is the heterogeneity of
laboratory assays and confirmatory tests used across the seven participating
centers from four countries. In addition, the Apnea-Hypopnea Index employed in
the HYPNOS trial was derived from cardiorespiratory polygraphy, which is not the
gold standard for OSA diagnosis [84].
In an Australian study, only 2 (5%) of 40 participants with OSA and
hypertension were identified to have "likely PA" based on ARR.
Additionally, 30% of patients were thought to have "possible
PA" based on low renin and normal ARR in the context of interfering
medications [85]. Nevertheless,
Chomsky et al. estimated that screening for PA in patients with OSA remains
cost-effective even if it detects only 3% of new PA cases [86]. Therefore, in light of the
currently available evidence, the recommendation to screen for PA in patients
with OSA is still reasonable. Further studies with standardized protocols are
required to characterize the association between PA and OSA.
Table 3 Studies of overt primary aldosteronism prevalence
in patients with obstructive sleep apnea (OSA) and
hypertension.
|
Reference
|
Criteria
|
Study population
|
Prevalence of overt PA
|
|
Di Murro 2010 [81]
|
ARR>40
(ng/dL)/(ng/mL/h) with
ALD>15 ng/dL with a suppressed
renin
|
Hypertension with and without OSA
|
34% in OSA, 9.8% in non-OSA
|
|
Buffolo 2019 [83]
|
According to the Endocrine Society guideline
|
Hypertension with OSA
|
8.9%
|
|
Dobrowolski 2021 [82]
|
Positive saline suppression test with post-test
ALD>10 ng/dL
|
Hypertension with and without moderate to severe OSA
|
21.3% in moderate to severe OSA, 8% in
non-moderate to severe OSA
|
|
Chee 2021 [85]
|
ARR>20
(ng/dL)/(ng/mL/h)*
or renin is suppressed or low, together with an adrenal
adenoma
|
Hypertension with OSA
|
Likely PA; 5%, Possible PA; 30%
|
*ARR or ALD values have been converted to common units
of measurement for ease of comparison.; Abbreviation; ARR,
aldosterone-renin-ratio; ALD, aldosterone; OSA, obstructive sleep apnea;
PA, primary aldosteronism.
The reported prevalence of overt primary aldosteronism in young adults with
hypertension
Secondary hypertension should be considered in individuals with early-onset
hypertension [87]. PA testing,
however, is not recommended routinely in young adults [12]
[88]. As such, data on the prevalence
of overt PA in young adults are scarce. A prospective observational study
conducted in India among adults under age 40 at hypertension onset revealed that
18.8% of participants had positive ARR screening results, and
17.8% overt PA was confirmed by saline suppression test. The recorded
prevalence of overt PA was highest in people with hypertension onset at
30–39 years [89]. A 10-year
delay between the initial hypertension diagnosis and PA confirmation was noted,
suggesting that PA screening should be conducted in patients with early-onset
hypertension.
Other conditions associated with primary aldosteronism
Atrial fibrillation has emerged as one of the cardiovascular complications
associated with PA. A meta-analysis found atrial fibrillation to be 3.52 times
more likely in patients with overt PA than in those with primary hypertension
[2]. Nguyen and colleagues
performed PA testing among 300 consecutive patients with hemorrhagic or ischemic
stroke and found overt PA in 30% of patients with hypertension and
atrial fibrillation, compared to 11.1% in the subgroup with
treatment-resistant hypertension and 13.3% in those with hypertension
and hypokalemia [90]. Another
observational study found that overt PA was highly prevalent (42.5%) in
participants with atrial fibrillation and no other identifiable causes of
arrhythmia, such as structural heart defect or coronary artery disease,
suggesting that overt PA might have led to atrial fibrillation in these patients
[91]. Current guidelines do not
yet endorse screening for PA in patients with atrial fibrillation [12]
[38]. Stroke [92]
[93], diabetes mellitus [94]
[95], and obesity [96] are other disorders associated
with overt PA. Further epidemiologic and mechanistic studies are needed to
understand these associations prior to expanding PA screening
recommendations.
