Background: The most popular screening test for primary aldosteronism (PAL) is the plasma aldosterone to renin ratio (ARR). Medications, dietary sodium, posture, and time of day all affect renin and aldosterone levels and can result in false-negative or -positive ARR if not controlled. Opinions are divided on whether β-adrenoreceptor blockers significantly affect the ARR.
Methods: Normotensive, nonmedicated male volunteers (n = 21) underwent measurement (seated, midmorning) of plasma aldosterone (by HPLC-tandem mass spectrometry), direct renin concentration (DRC), renin activity (PRA), cortisol, electrolytes, and creatinine and urinary aldosterone, cortisol, electrolytes, and creatinine at baseline, and after 1 wk (25 mg daily) and 4 wk (50 mg daily for three additional weeks) of atenolol.
Results: Compared with baseline, levels of aldosterone, DRC, and PRA were lower (P < 0.001) after both 1 and 4 wk [median (25–75th percentiles): baseline, 189 (138–357) pmol/liter, 40 (30–46) mU/liter, and 4.6 (2.7–5.8) ng/ml · h; 1 wk, 166 (112–310) pmol/liter, 34 (30–40) mU/liter, and 2.6 (2.0–3.1) ng/ml · h; 4 wk, 136 (97–269) pmol/liter, 16 (13–23) mU/liter, and 2.1(1.7–2.6) ng/ml · h, respectively]. ARR was significantly higher after 1 wk compared with baseline when calculated using PRA [61 (30–73) vs. 65 (44–130), P < 0.01] but not DRC [5 (4–7) vs. 5 (4–8)]. At 4 wk, ARR calculated by both PRA [78 (49–125)] and DRC [8 (6–14)] were significantly higher (P < 0.001) compared with baseline, and cortisol levels were significantly lower [92 (68–100) vs. 66 (48–91) ng/ml, P < 0.01]. There were no changes in plasma sodium, potassium, creatinine, or any urinary measurements.
Conclusion: β-Blockers can significantly raise the ARR and thereby increase the risk of false positives during screening for PAL.
Primary aldosteronism (PAL) is a specifically treatable and potentially curable form of secondary hypertension characterized by excessive and autonomous production of the salt-retaining hormone aldosterone (1). Until recently, PAL was thought to be rare, accounting for less than 1% of patients with hypertension, and not worth looking for in the absence of hypokalemia resulting from the kaliuretic action of aldosterone (2). However, studies conducted over the past 18 yr, involving the use of the plasma aldosterone to renin ratio (ARR) to screen for PAL in both hypokalemic and normokalemic hypertensives have shown it to be much more common than previously suspected and to be the commonest specifically treatable and potentially curable form of secondary hypertension (3–8).
A number of factors that affect secretion of aldosterone and/or renin may affect the interpretation of ARR and lead to either false-positive or false-negative results. Among these factors are certain medications (9–12), potassium levels (13, 14), dietary sodium intake (13, 14), coexisting diseases such as chronic renal failure or renal artery stenosis (15, 16), and some physiological factors such as posture and time of the day (17, 18).
Whether β-adrenoreceptor blockers affect the ARR is an important clinical question. Mulatero et al. (9) examined the effects of β-blockers on the ARR in a group of patients suspected of having primary aldosteronism. They reported that commencement of β-blockers resulted in a fall in both plasma aldosterone and plasma renin activity (PRA) and a rise in ARR. However, because the study design involved inclusion only of patients with elevated ARR at baseline, the false-positive ARR rate induced by β-blockers could not be calculated. Young (19) has maintained that β-blockers have minimal effects on the ratio. Clearly, opinions are divided on this issue, and further study is required. If β-blockers do significantly suppress renin levels, because a low renin level can produce a raised ARR even when aldosterone is low or in the lower part of the normal range, β-blockers could reduce the value of the ARR by causing false-positive results. It is therefore essential to now study their effects in individuals who do not have PAL and whose baseline ARR levels are normal and ask the question of whether it is possible that laboratory methodologies used for measuring aldosterone and renin can significantly influence the effect of β-blockers on the ARR.
Aldosterone is most commonly measured in clinical laboratories by RIA methods. However, concerns over the accuracy of these techniques have been raised (20). Schirpenbach et al. (21) found substantially different aldosterone concentrations when aldosterone was measured by four different immunoassay methods. Several very precise and specific aldosterone assays have recently been published using mass spectrometry, including one from our laboratory (22, 23). It is clearly an advantage that this accurate method for aldosterone could be used in the proposed study.
Plasma renin is currently most commonly measured as either 1) PRA by RIA of angiotensin I generated by the action of endogenous renin on endogenous substrate (angiotensinogen) or 2) as direct renin concentration (DRC) by immunometric assay of active renin, with the plasma carefully collected to avoid conversion of inactive to active renin at temperatures ranging from 0–4 C. These are the methods favored for ARR measurement in the recently published Endocrine Society clinical practice guidelines on case detection, diagnosis, and management of PAL (24). Some groups claim superiority of PRA in terms of assay performance, whereas others favor DRC because it can be automated and is therefore less labor intensive.
To avoid the potential for cyclical estrogen and progesterone changes to confound results, we evaluated the effect of atenolol on ARR in healthy male volunteers. We used HPLC-tandem mass spectrometry to measure aldosterone, a method recently developed and validated by Taylor and colleagues (23) and shown to be highly accurate and precise. We compared the ARR calculated using DRC with that calculated using PRA.
Subjects and Methods
Subjects
This study was performed with the approval of the Princess Alexandra Hospital and the University of Queensland Human Ethics Review Committees. Informed written consent was obtained from all participants. Twenty-six healthy men were included according to the following inclusion criteria: consenting, healthy subjects without evidence of renal, liver, or cardiovascular diseases who were not hypertensive or receiving any medications within the previous 2 months and had no anticipated requirement for any during the period of the study. Instructions were given to participants to maintain their usual sodium intake during the period of study, and compliance was assessed by collecting urine specimens to measure urinary sodium excretion at each visit. Three volunteers were excluded after developing dizziness and mild hypotension during the first few days of receiving the study drug. Another two participants withdrew for personal reasons leaving 21 men (mean ± sd age 32 ± 5 yr) who completed the study.
Medication and sampling times
Atenolol (Terry White Chemists, New South Wales, Australia) was administered once daily with breakfast in a dose of 25 mg for 1 wk and then 50 mg for another 3 wk. Nonfasting blood samples were collected into EDTA tubes between 0900 and 1000 h after sitting for 5–15 min at baseline and 1 and 4 wk after commencement of atenolol for measurement of plasma aldosterone, DRC, PRA, cortisol, sodium, potassium, and creatinine and centrifuged immediately at 2500 rpm for 10 min. Plasma was separated and snap frozen in dry ice and stored at −20 C pending assay.
Fasting spot urine samples were collected on the same days for measurement of sodium, potassium, creatinine, and cortisol. Blood pressure and heart rate were recorded during each visit after sitting for 10–15 min.
Analytic methods
Plasma aldosterone and cortisol were measured by HPLC-tandem mass spectrometry using a method recently validated in our laboratory (23). For aldosterone, the interassay coefficient of variation was 9.3% at 242 pmol/liter and 6.0% at 1321 pmol/liter. The intraassay coefficient of variation was 7.3% at 238 pmol/liter and 4.3% at 1344 pmol/liter. The interassay coefficient of variation for cortisol was 2.1% at 55 ng/ml and 1.8% at 462 ng/ml. The intraassay coefficient of variation was 3.7% at 56 ng/ml and 1.6% at 472 ng/ml. DRC was assayed by chemiluminescent immunoassay technology (DiaSorin, Liaison, Italy). The interassay coefficient of variation was 7.4% at 27 mU/liter and 6.0% at 107 mU/liter. The intraassay coefficient of variation was 3.7% at 15 mU/liter, 2.8% at 34 mU/liter, 2.0% at 82 mU/liter, and 1.2% at 258 mU/liter. PRA was assayed by GammaCoat RIA (DiaSorin, Stillwater, MN). The interassay coefficient of variation was 5.6% at 1.6 ng/ml · h, 7.6% at 10.7 ng/ml · h, and 6.8% at 15.2 ng/ml · h. The intraassay coefficient of variation was 10.0% at 1.6 ng/ml · h, 4.6% at 6.2 ng/ml · h, and 9.4% at 17.9 ng/ml · h. Urinary cortisol was measured by HPLC and urinary aldosterone by RIA (Siemens DPC, Los Angeles, CA).
Statistical analysis
SPSS 17.0 for Windows (SPSS, Chicago, IL) was used to analyze the data. Because data were not normally distributed, group data are presented as median (25th and 75th percentiles) unless otherwise stated. Nonparametric testing (Friedman test) was used for multiple comparisons. Pairwise comparisons were performed using Wilcoxon test. A P value <0.05 was considered statistically significant.
Results
Results of measured biochemical and hemodynamic parameters at baseline and after 1 and 4 wk of atenolol are shown in Table 1 and Fig. 1. In Table 1, results of aldosterone and ARR calculated by PRA are shown in both picomoles per liter and nanograms per deciliter. Aldosterone, PRA, DRC, ARR (calculated using DRC and also calculated using PRA), and hemodynamic parameters (blood pressure and heart rate) varied significantly during the period of the study. There were no significant changes in the levels of plasma sodium, potassium, or creatinine.
A, ARR using PRA; B, ARR using DRC; C, PRA; D, DRC; E, aldosterone; F, cortisol at baseline and after 1 wk (25 mg daily) and 4 wk (dose increased to 50 mg daily after the first week) of atenolol. For the purpose of this illustration, group data are presented as means with error bars indicating the sem. *, P < 0.01; +, P < 0.001 for pairwise comparisons (Wilcoxon). n.s., Not significant.
Effects of atenolol after 1 and 4 wk in 21 healthy men
| Baseline | 1 wk (25 mg daily) | 4 wk (25 mg daily for 1 wk and 50 mg daily for 3 wk) | P value (Friedman test) | |
|---|---|---|---|---|
| Aldosterone | ||||
| pmol/liter | 189 (138–357) | 166 (112–310) | 136 (97–269) | <0.001 |
| ng/dl | 6.78 (4.95–12.8) | 5.97 (4.03–11.1) | 4.88 (3.48–9.65) | <0.001 |
| DRC (mU/liter) | 40 (30–46) | 34 (30–40) | 16 (13–23) | <0.001 |
| PRA (ng/ml · h) | 4.6 (2.7–5.8) | 2.6 (2–3.1) | 2.1 (1.7–2.6) | <0.001 |
| ARR using DRC (pmol/liter)/(mU/liter) | 4.8 (4.1–7.2) | 4.7 (4.2–7.6) | 7.8 (6.1–14.5) | <0.001 |
| ARR using PRA | ||||
| (pmol/liter)/(ng/ml · h) | 61 (30–73) | 65 (44–130) | 78 (49–125) | <0.01 |
| (ng/dl)/(ng/ml · h) | 2.1 (1.0–2.6) | 2.3 (1.6–4.7) | 2.7 (1.8–4.5) | <0.01 |
| Plasma cortisol (ng/ml) | 92 (68–100) | 82 (68–100) | 66 (48–91) | <0.001 |
| Plasma Na+ (mmol/liter) | 139 (138–140) | 140 (138–143) | 140 (139–141) | NS |
| Plasma K+ (mmol/liter) | 4.1 (3.9–4.1) | 4.0 (3.9–4.2) | 4.0 (3.7–4.1) | NS |
| Plasma creatinine (μmol/liter) | 70 (65–83) | 76 (67–80) | 73 (67–77) | NS |
| Systolic blood pressure (mm Hg) | 128 (124–132) | 126 (122–130) | 120 (120–126) | <0.001 |
| Diastolic blood pressure (mm Hg) | 88 (84–90) | 84 (82–88) | 82 (80–84) | <0.001 |
| Heart rate (beats/min) | 87 (80–89) | 74 (70–77) | 62 (60–66) | <0.001 |
| Baseline | 1 wk (25 mg daily) | 4 wk (25 mg daily for 1 wk and 50 mg daily for 3 wk) | P value (Friedman test) | |
|---|---|---|---|---|
| Aldosterone | ||||
| pmol/liter | 189 (138–357) | 166 (112–310) | 136 (97–269) | <0.001 |
| ng/dl | 6.78 (4.95–12.8) | 5.97 (4.03–11.1) | 4.88 (3.48–9.65) | <0.001 |
| DRC (mU/liter) | 40 (30–46) | 34 (30–40) | 16 (13–23) | <0.001 |
| PRA (ng/ml · h) | 4.6 (2.7–5.8) | 2.6 (2–3.1) | 2.1 (1.7–2.6) | <0.001 |
| ARR using DRC (pmol/liter)/(mU/liter) | 4.8 (4.1–7.2) | 4.7 (4.2–7.6) | 7.8 (6.1–14.5) | <0.001 |
| ARR using PRA | ||||
| (pmol/liter)/(ng/ml · h) | 61 (30–73) | 65 (44–130) | 78 (49–125) | <0.01 |
| (ng/dl)/(ng/ml · h) | 2.1 (1.0–2.6) | 2.3 (1.6–4.7) | 2.7 (1.8–4.5) | <0.01 |
| Plasma cortisol (ng/ml) | 92 (68–100) | 82 (68–100) | 66 (48–91) | <0.001 |
| Plasma Na+ (mmol/liter) | 139 (138–140) | 140 (138–143) | 140 (139–141) | NS |
| Plasma K+ (mmol/liter) | 4.1 (3.9–4.1) | 4.0 (3.9–4.2) | 4.0 (3.7–4.1) | NS |
| Plasma creatinine (μmol/liter) | 70 (65–83) | 76 (67–80) | 73 (67–77) | NS |
| Systolic blood pressure (mm Hg) | 128 (124–132) | 126 (122–130) | 120 (120–126) | <0.001 |
| Diastolic blood pressure (mm Hg) | 88 (84–90) | 84 (82–88) | 82 (80–84) | <0.001 |
| Heart rate (beats/min) | 87 (80–89) | 74 (70–77) | 62 (60–66) | <0.001 |
Values are presented as medians (25–75th percentiles). P values are the significance levels of multiple comparisons between the three collection time points as determined by Friedman testing. P values of pairwise comparisons between baseline vs. 1 wk, baseline vs. 4 wk, and 1 wk vs. 4 wk are shown in Fig. 1. NS, Not significant.
Effects of atenolol after 1 and 4 wk in 21 healthy men
| Baseline | 1 wk (25 mg daily) | 4 wk (25 mg daily for 1 wk and 50 mg daily for 3 wk) | P value (Friedman test) | |
|---|---|---|---|---|
| Aldosterone | ||||
| pmol/liter | 189 (138–357) | 166 (112–310) | 136 (97–269) | <0.001 |
| ng/dl | 6.78 (4.95–12.8) | 5.97 (4.03–11.1) | 4.88 (3.48–9.65) | <0.001 |
| DRC (mU/liter) | 40 (30–46) | 34 (30–40) | 16 (13–23) | <0.001 |
| PRA (ng/ml · h) | 4.6 (2.7–5.8) | 2.6 (2–3.1) | 2.1 (1.7–2.6) | <0.001 |
| ARR using DRC (pmol/liter)/(mU/liter) | 4.8 (4.1–7.2) | 4.7 (4.2–7.6) | 7.8 (6.1–14.5) | <0.001 |
| ARR using PRA | ||||
| (pmol/liter)/(ng/ml · h) | 61 (30–73) | 65 (44–130) | 78 (49–125) | <0.01 |
| (ng/dl)/(ng/ml · h) | 2.1 (1.0–2.6) | 2.3 (1.6–4.7) | 2.7 (1.8–4.5) | <0.01 |
| Plasma cortisol (ng/ml) | 92 (68–100) | 82 (68–100) | 66 (48–91) | <0.001 |
| Plasma Na+ (mmol/liter) | 139 (138–140) | 140 (138–143) | 140 (139–141) | NS |
| Plasma K+ (mmol/liter) | 4.1 (3.9–4.1) | 4.0 (3.9–4.2) | 4.0 (3.7–4.1) | NS |
| Plasma creatinine (μmol/liter) | 70 (65–83) | 76 (67–80) | 73 (67–77) | NS |
| Systolic blood pressure (mm Hg) | 128 (124–132) | 126 (122–130) | 120 (120–126) | <0.001 |
| Diastolic blood pressure (mm Hg) | 88 (84–90) | 84 (82–88) | 82 (80–84) | <0.001 |
| Heart rate (beats/min) | 87 (80–89) | 74 (70–77) | 62 (60–66) | <0.001 |
| Baseline | 1 wk (25 mg daily) | 4 wk (25 mg daily for 1 wk and 50 mg daily for 3 wk) | P value (Friedman test) | |
|---|---|---|---|---|
| Aldosterone | ||||
| pmol/liter | 189 (138–357) | 166 (112–310) | 136 (97–269) | <0.001 |
| ng/dl | 6.78 (4.95–12.8) | 5.97 (4.03–11.1) | 4.88 (3.48–9.65) | <0.001 |
| DRC (mU/liter) | 40 (30–46) | 34 (30–40) | 16 (13–23) | <0.001 |
| PRA (ng/ml · h) | 4.6 (2.7–5.8) | 2.6 (2–3.1) | 2.1 (1.7–2.6) | <0.001 |
| ARR using DRC (pmol/liter)/(mU/liter) | 4.8 (4.1–7.2) | 4.7 (4.2–7.6) | 7.8 (6.1–14.5) | <0.001 |
| ARR using PRA | ||||
| (pmol/liter)/(ng/ml · h) | 61 (30–73) | 65 (44–130) | 78 (49–125) | <0.01 |
| (ng/dl)/(ng/ml · h) | 2.1 (1.0–2.6) | 2.3 (1.6–4.7) | 2.7 (1.8–4.5) | <0.01 |
| Plasma cortisol (ng/ml) | 92 (68–100) | 82 (68–100) | 66 (48–91) | <0.001 |
| Plasma Na+ (mmol/liter) | 139 (138–140) | 140 (138–143) | 140 (139–141) | NS |
| Plasma K+ (mmol/liter) | 4.1 (3.9–4.1) | 4.0 (3.9–4.2) | 4.0 (3.7–4.1) | NS |
| Plasma creatinine (μmol/liter) | 70 (65–83) | 76 (67–80) | 73 (67–77) | NS |
| Systolic blood pressure (mm Hg) | 128 (124–132) | 126 (122–130) | 120 (120–126) | <0.001 |
| Diastolic blood pressure (mm Hg) | 88 (84–90) | 84 (82–88) | 82 (80–84) | <0.001 |
| Heart rate (beats/min) | 87 (80–89) | 74 (70–77) | 62 (60–66) | <0.001 |
Values are presented as medians (25–75th percentiles). P values are the significance levels of multiple comparisons between the three collection time points as determined by Friedman testing. P values of pairwise comparisons between baseline vs. 1 wk, baseline vs. 4 wk, and 1 wk vs. 4 wk are shown in Fig. 1. NS, Not significant.
Results of pairwise (Wilcoxon) comparisons of measured parameters between each of the three collection time points are shown in Fig. 1. Median aldosterone levels were significantly (P < 0.001) lower after 1 wk of atenolol compared with baseline, as were levels of plasma DRC, PRA, systolic blood pressure, diastolic blood pressure, and heart rate. After 1 wk of atenolol, ARR was higher than at baseline when calculated using PRA [61 (30–73) vs. 65 (44–130), P < 0.01] but not when calculated using DRC [4.8 (4.1–7.2) vs. 4.7 (4.2–7.6), P = 0.31] (Fig. 1). Aldosterone, DRC, and PRA levels at 4 wk were again significantly (P < 0.001) lower compared with baseline. However, unlike after 1 wk, ARR calculated using both PRA [78 (49–125) (pmol/liter)/(ng/ml · h)] and DRC [7.8 (6.1–14.5) (pmol/liter)/(mU/liter)] were significantly higher (P < 0.001). Plasma cortisol levels were also significantly lower after 4 wk treatment compared with baseline.
When levels after 4 wk of atenolol were compared with levels after 1 wk, aldosterone (P < 0.001), DRC (P < 0.001), PRA (P < 0.01), and cortisol (P < 0.01) were all significantly lower. ARR measured with DRC (but not ARR measured with PRA, which had already increased after 1 wk) was significantly (P < 0.001) higher.
Results of measured urinary parameters (corrected for creatinine) are presented in Table 2. There were no significant differences between levels at baseline and those after 1 or 4 wk of treatment for urinary potassium, sodium, aldosterone, or cortisol levels.
Measured urinary parameters (corrected for creatinine) in 21 healthy men
| Baseline | 1 wk (25 mg daily) | 4 wk (25 mg daily for 1 wk and 50 mg daily for 3 wk) | P value (Friedman Test) | |
|---|---|---|---|---|
| Aldosterone (nmol/mmol) | 2.8 (1.7–5) | 2.6 (1.2–3.8) | 2.9 (1.3–5.0) | NS |
| Sodium (mmol/mmol) | 11.6 (9.4–19.2) | 11.5 (7.7–16.1) | 14.6 (7.6–18.7) | NS |
| Potassium (mmol/mmol) | 5.8 (4.3–7.4) | 6.2 (3.7–8.3) | 6.3 (4.4–8.1) | NS |
| Cortisol (nmol/mmol) | 6.8 (3.7–17.2) | 7.9 (3.6–15.5) | 7.7 (3.7–12.7) | NS |
| Baseline | 1 wk (25 mg daily) | 4 wk (25 mg daily for 1 wk and 50 mg daily for 3 wk) | P value (Friedman Test) | |
|---|---|---|---|---|
| Aldosterone (nmol/mmol) | 2.8 (1.7–5) | 2.6 (1.2–3.8) | 2.9 (1.3–5.0) | NS |
| Sodium (mmol/mmol) | 11.6 (9.4–19.2) | 11.5 (7.7–16.1) | 14.6 (7.6–18.7) | NS |
| Potassium (mmol/mmol) | 5.8 (4.3–7.4) | 6.2 (3.7–8.3) | 6.3 (4.4–8.1) | NS |
| Cortisol (nmol/mmol) | 6.8 (3.7–17.2) | 7.9 (3.6–15.5) | 7.7 (3.7–12.7) | NS |
Values are presented as medians (25th–75th percentiles). NS, Not significant.
Measured urinary parameters (corrected for creatinine) in 21 healthy men
| Baseline | 1 wk (25 mg daily) | 4 wk (25 mg daily for 1 wk and 50 mg daily for 3 wk) | P value (Friedman Test) | |
|---|---|---|---|---|
| Aldosterone (nmol/mmol) | 2.8 (1.7–5) | 2.6 (1.2–3.8) | 2.9 (1.3–5.0) | NS |
| Sodium (mmol/mmol) | 11.6 (9.4–19.2) | 11.5 (7.7–16.1) | 14.6 (7.6–18.7) | NS |
| Potassium (mmol/mmol) | 5.8 (4.3–7.4) | 6.2 (3.7–8.3) | 6.3 (4.4–8.1) | NS |
| Cortisol (nmol/mmol) | 6.8 (3.7–17.2) | 7.9 (3.6–15.5) | 7.7 (3.7–12.7) | NS |
| Baseline | 1 wk (25 mg daily) | 4 wk (25 mg daily for 1 wk and 50 mg daily for 3 wk) | P value (Friedman Test) | |
|---|---|---|---|---|
| Aldosterone (nmol/mmol) | 2.8 (1.7–5) | 2.6 (1.2–3.8) | 2.9 (1.3–5.0) | NS |
| Sodium (mmol/mmol) | 11.6 (9.4–19.2) | 11.5 (7.7–16.1) | 14.6 (7.6–18.7) | NS |
| Potassium (mmol/mmol) | 5.8 (4.3–7.4) | 6.2 (3.7–8.3) | 6.3 (4.4–8.1) | NS |
| Cortisol (nmol/mmol) | 6.8 (3.7–17.2) | 7.9 (3.6–15.5) | 7.7 (3.7–12.7) | NS |
Values are presented as medians (25th–75th percentiles). NS, Not significant.
Discussion
After administration of atenolol to healthy normotensive males for 1 month, we found significantly higher ARR levels calculated using either PRA or DRC. In the case of ARR using PRA, levels were already higher after only 1 wk at a low dose. Although aware that atenolol is used in the management of normotensive patients with coronary artery disease, because the young men in this study were normotensive, we introduced atenolol at the low dose of 25 mg daily for 1 wk before increasing to 50 mg daily for 3 wk to avoid symptoms due to significant lowering of blood pressure. The results of this study support the contention that β-blockers can raise the ARR and thereby increase the risk of false-positive results during screening for PAL. Furthermore, our data suggest that the magnitude of this effect may depend not only on the dose and duration of treatment but also on the renin assay method. A daily dose of 25 mg atenolol for only 1 wk was enough to significantly increase the ratio when PRA (but not DRC) was used to calculate the ratio. As shown in Fig. 1, PRA fell significantly within 1 wk, and then more slowly, whereas PRC fell progressively over 4 wk. The more rapid suppressive effect on PRA than DRC may have occurred because atenolol can cause reduction of both renin and angiotensinogen (renin substrate) concentrations (25, 26). The ratio calculated using PRA after an additional 3 wk administration of atenolol at a daily dose of 50 mg tended to be higher than after the first week, but the difference was not statistically significant. Unlike the ARR calculated by PRA, the ratio calculated using DRC was not significantly higher than baseline after 1 wk at the 25-mg/d dose but was significantly higher after the additional 3 wk at 50 mg/d. The effect of progressively falling aldosterone levels on the ratio is a confounding factor in terms of interpreting the effects of atenolol on ARR and cannot be ignored. Because two dependent variables are involved in the calculation of ARR, when both are falling, the rate of fall of each in individual patients will be an important determinant of ARR but was not examined here.
Median ARR calculated by PRA and DRC rose after 4 wk by 27.9 and 62.5%, respectively. Despite this, ARR values obtained in the current study did not exceed the upper limit of normal used by our laboratory. However, the lack of false-positive results could have been explained by 1) the short duration of atenolol administration, 2) the small number of participants, and 3) the fact that participants were young, normotensive, and male (characteristics associated with lower ARR values (27, 28). We would therefore anticipate that in the clinical situation, the propensity for β-blockers to cause false-positive ARR results would be greater.
Atenolol could affect the ARR by several mechanisms. First, β-blockers suppress renin production by inhibiting β-adrenergic receptors in the juxtaglomerular apparatus of the kidney (29). The reduced renin levels after β-adrenergic blockade seen in this study are in accord with findings of other authors (30–32). Second, if β-blockers also reduce renin substrate levels (25, 26), this should reduce the amount of generated angiotensin I in the PRA assay even further. Reduction of cortisol levels by β-blockers seen in this study has been previously reported by other authors (33, 34). Because cortisol levels reflect ACTH, and ACTH is stimulated by stress, falling cortisol levels may have been an indication of a reduction in stress induced by atenolol’s known anxiolytic effect. Decreased stress may have led also to decreased central sympathetic output, which would directly lower renin secretion and levels. Lower ACTH and cortisol levels may also have been contributed to by subjects becoming progressively less stressed by the demands of blood and urine collection. There was no significant change in urinary sodium excretion (as an index of sodium intake) to explain the observed changes in renin and aldosterone. The lack of fall in urinary aldosterone and cortisol levels (despite falling plasma levels) may be due to the fact that spot, and not 24-h, urine samples were collected.
A strength of this study is that aldosterone levels were measured by a recently described very accurate method (23), rather than by immunoassay methods that are less precise (21). Restricting the study to male participants excluded effects of changing levels of estrogen and progesterone during the menstrual cycle (35, 36). To our knowledge, no previous study has compared the use of PRA vs. DRC for calculating the ARR during administration of atenolol.
Limitations of this study include a relatively small number of participants and use of normal healthy volunteers rather than hypertensive patients with normal baseline ARR. However, this approach enabled us to avoid the potentially confounding effects of other antihypertensive medications and of restricted sodium intake.
In conclusion, atenolol therapy is associated with increased ARR values and therefore may increase the risk of false-positive results during screening for PAL. Hence, where possible, β-blockers should be avoided before ARR testing, if necessary replacing them with other antihypertensives that have little or no effect on ARR, such as prazosin and slow-release verapamil to which hydralazine can be added. Of course, it is highly desirable and much simpler to measure ARR before any antihypertensive medications have been commenced.
Acknowledgments
Disclosure Summary: The authors have nothing to disclose.
Abbreviations:
- ARR,
Aldosterone to renin ratio;
- DRC,
direct renin concentration;
- PAL,
primary aldosteronism;
- PRA,
plasma renin activity.
