The Association of Serum Aldosterone with Sex, Age, and Sodium Status

Keywords

Aldosterone • Age • Sex • Sodium

Introduction

Primary aldosteronism (PA) is now recognized as the most common cause of secondary hypertension, occurring in at least 5%, and perhaps more than 10%, of individuals with arterial hypertension [1]. Although the aldosterone-to-renin ratio (ARR) is widely recommended for screening or case detection, the rate of screening for primary aldosterone remains very low [1,2]. One cause for the low rate of screening may be the difficulty that providers have in interpreting the results of the ARR, as its interpretation may not be straightforward [3]. Further, even in cases of confirmed PA, there is a wide range of serum aldosterone levels that are found; circulating aldosterone levels may be in the high-normal range, or elevated above the reference limits. This results in a low level of diagnostic accuracy of the ARR that, in most cases, does not lead to a firm diagnosis of PA. The diagnosis then requires confirmatory tests, which usually require potentially harmful salt loading, as well as additional expense and burden [4]. An increase in the diagnostic accuracy of the serum aldosterone concentration might simplify and improve the diagnosis of PA.

In a recent study, in which we postulated that circulating aldosterone levels in low-renin hypertension would have a bimodal distribution, we observed that the frequency distribution of unadjusted serum aldosterone conformed to a pattern that was suggestive of a bimodal distribution, but not statistically different from a unimodal pattern (P=0.11). We hypothesized that a possible bimodal distribution of aldosterone might be obscured by factors that are known to affect the serum aldosterone concentration. When we adjusted the serum aldosterone concentration for three such factors that were readily available in the population we studied, namely age, sex, and the urine sodium:creatinine ratio, the pattern was clearly bimodal, and the P value for lack of unimodality was 0.008 [5].

This observation suggested the following hypothesis: Aldosterone, as commonly measured, may be affected by confounders such as age, sex, and sodium status, causing a distribution of serum aldosterone that differs from a normal distribution; adjustment for these factors may result in a more normal distribution of serum aldosterone. Because the measured aldosterone level may be decisive in the interpretation of tests for primary aldosteronism, such as the aldosterone-renin ratio (ARR) and the saline infusion test, the effect of this statistical adjustment would be of interest, and possible clinical value. We selected a population of normotensive individuals in which to study the effect of this statistical adjustment.

Methods

We analyzed data from the cohort of 3,345 individuals who attended the sixth examination cycle of the Framingham Offspring Study. These individuals were the children, and their spouses, of the original members of the Framingham Heart Study [6]. The sixth examination cycle was performed from 1995 to 1998. In addition to medical history, physical examination, and standard laboratory tests, blood was drawn for later measurement of serum aldosterone and plasma renin concentrations. This blood was drawn between 8:00 and 10:00 AM, in subjects who were fasting and ambulatory, but remained recumbent for 5-10 minutes before blood was drawn. Aldosterone and renin were measured in 2004. Further details are presented in reference (5)

Subjects were excluded if they had a condition, or were taking medication, deemed likely to affect the levels of circulating aldosterone or renin. Exclusions were congestive heart failure, eGFR less than 60 ml/minute, diabetes mellitus, bilateral edema, and present treatment with any antihypertensive medication . For this analysis we excluded all individuals with hypertension (systolic BP ≥140 mm Hg, or diastolic BP ≥90 mm Hg.). We also excluded 42 subjects who lacked a measurement of the urine Na:creatinine ratio. After exclusions, the final sample for the present analysis consisted of 1,468 normotensive participants, 652 men and 816 women, aged 29 to 85 years.

We used independent two sample t-tests, regression of natural- log-transformed values of serum aldosterone concentration on age, sex, and urinary sodium:creatinine ratio, the dip test for unimodality, dotplots, and Q-Q plots to evaluate the normality of the distribution of serum aldosterone levels with and without statistical adjustment for these key covariates.

Conclusion

In this investigation we aimed to better understand the factors responsible for an observation made in our previous report [5]. In that report our hypothesis that serum aldosterone concentrations would have a bimodal distribution in individuals with low-renin hypertension was statistically confirmed when we adjusted aldosterone concentrations for age, sex, and the urine sodium:creatinine ratio, but was not supported without this statistical adjustment.

In the present investigation we confirm, in the same large, community-based population, looking at normotensive individuals, that age and sex were weakly associated with serum aldosterone concentrations, but, as expected, urine sodium was more strongly associated [8]. When we adjusted for all three variables several statistical indicators revealed a more normal distribution of the adjusted serum aldosterone concentrations.

In the present investigation we confirm, in the same large, community-based population, looking at normotensive individuals, that age and sex were weakly associated with serum aldosterone concentrations, but, as expected, urine sodium was more strongly associated [8-10]. When we adjusted for all three variables several statistical indicators revealed a more normal distribution of the adjusted serum aldosterone concentrations [1].

It is recognized that an elevated ARR is more likely to be associated with a low renin level than with an elevated aldosterone level [4]. Many investigators therefore recommend that a minimal level of serum aldosterone concentration be present before the ARR is considered to be positive, that is, suggestive of a diagnosis of primary aldosteronism. We suggest that an adjustment of the measured serum aldosterone concentration that provides a more accurate approximation of the actual physiologic activity of the hormone (or at least a more normal distribution) may improve the test.

Characteristics of the ARR. Such a correction might also improve the performance of confirmatory tests for PA, several of which also depend on precise estimation of the levels of serum aldosterone.

Our study does have some limitations. We utilized available urinary indices for sodium status (spot urine sodium measurements). Although there may be inherent variability in spot urine sodium measurements, it does reflect a variable that would be readily available to a clinician at the point of care. We recognize the role of potassium on aldosterone activity; however urine potassium measurements were not available in the Framingham Offspring Study

These considerations lead to the following hypothesis: Adjustment of serum aldosterone levels for age, sex, and a measure of urine sodium excretion may result in a more normal distribution of the serum aldosterone concentration. This may lead to an improvement in the diagnostic performance of the ARR, and of other tests whose interpretation depends on the serum aldosterone level, such as the saline infusion test and Captopril challenge test.

In populations similar to those described in the sixth examination cycle of the Framingham Offspring Study, i.e. white men and women older than 30 years; we suggest that the coefficients and constants shown in Table 2 may be used to test this hypothesis. In other populations appropriate normative data would be required. Widely used statistical software makes this adjustment relatively simple.

Acknowledgments

This work was supported by contracts NO1-HC-25195 and HHSN268201500001I from the National Heart, Lung and Blood Institute and grant K24 HL 4334. Dr. Vasan is also supported in part by the Evans Medical Foundation and the Jay and Louis Coffman Endowment from the Department of Medicine, Boston University School of Medicine. We also acknowledge the dedication of the FHS study participants without whom this research would not be possible.

References

1. Funder JW,Carey RM, and Mantero F, et al. “The management of primary aldosteronism: Case detection, diagnosis, and treatment: An Endocrine Society Clinical Practice Guideline”J Clin Endocrinol Metab. 101 (2016):1889-1916.
2. Brown MJ, Drake WM. “Splitting atoms: the Endocrine Society guideline for the management of primary aldosteronism” Lancet Diabetes Endocrinol. 4 (2016):805-807.
3. Stowasser M, Taylor PJ, and Pimento Eduardo, Ahmed AHA, Gordon RD. “Laboratory investigation of primary aldosteronism” Clin Biochem Rev. 31 (2010):39-56.
4. Montori VM, Schwartz GL, and Chapman AB, Boerwinkle E, Turner ST. “Validity of the aldosterone-renin ratio used to screen for primary aldosteronism” Mayo Clin Proc. 76 (2001):877-882.
5. Adlin EV, Braitman LE, and Vasan RS. “Bimodal aldosterone distribution in low-renin hypertension” Am J Hypertens. 26 (2013): 1076-1085.
6. Kannel WB, Feinleib N, and McNamara PM, Garrison RJ, Castelli WP. “An investigation of coronary heart disease in families: the Framingham Offspring Study” Am J Epidemiol. 110 (1979): 281- 290.
7. Hartigan JA, Hartigan PM. “The dip test of unimodality” Ann Statistics. 13 (1985):70-84.
8. Kathiresan S, Larson MG, and Benjamin EJ, et al. “Clinical and genetic correlates of serum aldosterone in the community: the Framingham Heart Study” Am J Hypertens. 18 (2005) (5 Pt 1): 657- 65.
9. Laragh JH, Sealey J, and Brunner HR. “Control of aldosterone secretion in normal and hypertensive man: Abnormal renin- aldosterone patterns in low renin hypertension.” Am J Med. 53 (1972):649-663.
10. Brunner HR, Laragh JH, and Baer L., et al. “Essential hypertension: Renin and aldosterone, heart attack and stroke” N Engl J Med. 286 (1972):441-449.