Raised aldosterone to renin ratio predicts antihypertensive efficacy of spironolactone: a prospective cohort follow-up study

Abstract

Aims

Aldosterone/renin ratio is an index for inappropriate aldosterone activity, and it is increasingly being used to screen for primary aldosteronism within the hypertensive population. It may also be a good index to help predict the response to spironolactone. To assess the blood pressure response to oral spironolactone in hypertensive patients with primary aldosteronism identified with raised aldosterone to renin ratio.

Methods

We conducted a prospective cohort study of hypertensive patients with raised aldosterone/renin ratio, who failed to suppress plasma aldosterone with salt loading and fludrocortisone suppression test. These patients were treated with spironolactone and were followed-up for a period of up to 3 years.

Results

We studied 28 (12 male) subjects with a mean age of 55 (s.d. 10) years who were followed up for a mean period of 12.9 (7) months. At baseline, the patients were taking a mean of 2.1 (1.2) antihypertensive drugs, but despite this 16/28 (57%) had diastolic BP  >90 mmHg, 39% with systolic BP  >160 mmHg. After commencing spironolactone, three patients complained of breast tenderness but continued treatment and one patient was intolerant of spironolactone and had to stop treatment. Of the remaining 27 patients, the mean number of antihypertensive drugs used dropped to spironolactone plus 0.7 (s.d. 0.9). All but one patient (96%) achieved a diastolic BP≤90 mmHg and 78% achieved a systolic BP≤160 mmHg. In total 48% had BP≤140/90 mmHg and 13/27 (48%) were treated with spironolactone monotherapy. Assessing only patients on drug treatment at baseline (n = 24), spironolactone significantly reduced the need for antihypertensive drugs by −0.5 (CI 0.1–1.0), P = 0.02, as well as reducing blood pressure [systolic BP −15 mmHg (CI 5–25), P = 0.007 and diastolic BP (mmHg) by −8 mmHg (CI 4–13), P = 0.001].

Conclusions

Spironolactone was a highly effective antihypertensive agent in hypertensive patients who had a raised aldosterone/renin ratio. As a raised ratio was highly predictive of nonsuppression of plasma aldosterone suggesting primary aldosteronism, it might be worthwhile using spironolactone in this subgroup of hypertensive patients with raised aldosterone/renin ratios, provided that adrenal adenomas are excluded with imaging techniques.

Introduction

The prevalence rate of primary aldosteronism is controversial but it may be more common than previously thought, accounting for up to 15% of clinic hypertensives [1–3]. We have recently reported that in a single GP practice, the prevalence of primary aldosteronism may be equally prevalent [4]. These studies utilized the aldosterone to plasma renin activity ratio (ARR) as a screening test. We believe that a raised ARR indicates inappropriate aldosterone activity, and we found that 94% of subjects with an ARR≥750 failed to suppress plasma aldosterone with salt loading [3]. We have treated a cohort of these patients with oral spironolactone, and we report the response to this intervention.

Methods

Subjects

We screened for primary aldosteronism in all hypertensive patients referred for assessment in our Hypertension Research Centre in Tayside, using the ambulant plasma aldosterone (pmol ml⁻¹) to plasma renin activity (ng ml⁻¹ h⁻¹) ratio of greater than or equal to 750 [3, 5] as our selection criteria (see below). Those subjects with a raised aldosterone to renin ratio were admitted for a standard salt loading and fludrocortisone suppression test (see below). Patients with a positive test (nonsuppressible serum aldosterone ≥140 pmol l⁻¹) had adrenal CT scanning to screen for an adenoma. We also screened for glucocorticoid suppressible hyperaldosteronism (GSH) using genetic testing in patients who exhibited a drop in plasma aldosterone at 12.00 h midday following salt loading. Patients without an adrenal adenoma on CT scanning and GSH on genetic testing were included in this follow-up cohort.

Blood pressure measurements

Seated blood pressure measurements were carried out using standard mercury sphygmomanometer in accordance to the British Hypertension Society guidelines.

Salt loading and fludrocortisone suppression test

The usual antihypertensive medications were withdrawn for between 7 and 10 days. None of these patients was on oral spironolactone which would require a more prolonged drug withdrawal period. Where necessary, an α-adrenoceptor blocker, doxasozin, was prescribed to control the blood pressure if the diastolic blood pressure was greater than or equal to 110 mmHg. All patients were routinely given two tablets of Slow-K (600 mg potassium chloride each) daily to avoid hypokalaemia. On day 1, each patient was given five Slow-Sodium (600 mg sodium chloride each) tablets at 18.00 h and 22.00 h following baseline supine 08.00 h and ambulant 12.00 h blood samplings for electrolytes, plasma cortisol and aldosterone. On day 2, each patient was given four Slow-Sodium tablets with fludrocortisone 0.5 mg at 08.00 h, and further Slow-Sodium tablets, five at 12.00 h, four at 18.00 h and four at 22.00 h. The day 2 regime was repeated on day 3. On day 4, the 08.00 h treatment regime was given following supine blood sampling, and the midday bloods were sampled following 4 h of ambulation at 12.00 h. In total, 48 Slow-Sodium tablets or 28 800 mg of sodium chloride with 1.5 mg of fludrocortisone were given over a 4 day period. Patients had their blood pressures carefully monitored throughout the study period to identify possible complications.

Laboratory testing

Plasma renin activity (PRA) was measured using Biodata Renin MAIA assay (Serono Diagnostics Ltd, Woking, Surrey) with an intra-assay coefficient of variation of <10% between 0.3 and 18 ng ml⁻¹ h⁻¹, and a least detectable concentration of 0.3 mg ml⁻¹ h⁻¹. The interassay batch variation over a year was 11% (QC mean values 2.3 and 6.1 ng ml⁻¹ h⁻¹).

Plasma aldosterone was measured using a solid-phase (coated tube) radioimmunoassay technique, DPC Coat-a-Count assay (DPC, Llanberis, Caernarfon, Gwynedd) with an intra-assay coefficient of variation of <10% between 200 and 3300 pmol l⁻¹, and a least detectable concentration of 70 pmol l⁻¹). The interassay variation as calculated from QC pools run in each assay over a year was 10% (QC mean values 400 and 1050 pmol l⁻¹).

Intervention

Subjects with raised ARR and a positive salt loading and fludrocortisone suppression test had oral spironolactone 50 mg once daily added to their usual antihypertensive regime. The first clinic visit was at 1 month, and subsequent duration between clinic visits varied between 4 weeks to 12 weeks at the discretion of the clinicians based on blood pressure responses and possible adverse effects. The usual antihypertensive drug treatment was withdrawn if blood pressure was ≤150/90 mmHg after the addition of spironolactone. The dose of spironolactone was reduced to 25 mg if adverse effects occurred or blood pressure normalized on a 50 mg monotherapy.

Statistics

Data are presented as mean (s.d.). Paired Student’s t-test was used to assess the effect of spironolactone on the number of drugs used as well as its blood pressure lowering efficacy during follow-up. Statistical significance was defined at a P value of 0.05.

Results

We prospectively studied a cohort of 28 subjects (12 males, mean age 55 years, s.d. 10) These patients were followed up for a mean of 12.9 months (range 3–35 months). Prior to spironolactone, the mean number of drugs the patients were taking was 2.1 (s.d. 1.2), despite which 16/28 (57%) had diastolic BP >90 mmHg (11/28) 39% with systolic BP >160 mmHg. One patient could not tolerate spironolactone and had to stop the drug after 4 months of treatment (nonspecifically unwell). Of the remaining 27 patients, the mean number of antihypertensive drugs required dropped to spironolactone plus 0.7 (s.d. 0.9). All but one patient (96%) achieved a diastolic BP≤90 mmHg and 78% achieved a systolic BP≤160 mmHg. In total 48% had a BP≤140/90 mmHg and 13/27 (48%) needed only spironolactone therapy (Table 1).

Table 1.

Anti-hypertensive treatment and blood pressure readings at baseline and during follow-up.

Table 2 shows that treatment with spironolactone significantly reduced antihypertensive drug treatment as well as improving blood pressure control in patients who were on drug treatment at baseline.

Table 2.

Blood pressure lowering efficacy of spironolactone during follow-up.

Discussion

The favourable blood pressure lowering response to spironolactone in this cohort of patients suggested that a raised ARR was an effective index in identifying patients who would benefit from this drug treatment. About 60% of these patients were treated with nonspironolactone diuretics at baseline, and most of the remaining patients had had diuretics tried at some stage. Blood pressure control remained suboptimal however, suggesting that the favourable blood pressure response with spironolactone in our study was not related to its diuretic effect alone.

Our study has parallels with previous studies indicating a favourable response to spironolactone in subjects with low ‘stimulated’ renin hypertension [6–8]. Further small scale studies however, revealed that other hypotensive agents (methydopa [9] and thiazide diuretics [10, 11]) were equally effective in low renin hypertensives, and this blood pressure lowering effects were not dependant on the renin status [12] (thiazide diuretics [11, 13], nifedipine [14], and methydopa [9]). We believe that the use of ARR is more specific than ‘stimulated’ renin in identifying primary aldosteronism, which forms the bulk of mineralocorticoid related hypertension. We have reported that the frusemide stimulation test, a commonly used test to assess renin responsiveness had poor negative predictive value in screening for subjects with possible primary aldosteronism [15]. In other words, many subjects with primary aldosteronism retain their renin responsiveness. In contrast, a raised ARR in our centre predicted primary aldosteronism in 94% of subjects [3]. The finding of hypertensive subjects with low renin and nonsuppressible serum aldosterone is not new, and was described about 20 years ago [16]. In our hands this test appears reasonably specific for primary aldosteronism if a ratio ≥750 is used.

The use of spironolactone for treating ‘essential’ hypertension was curtailed in Britain following the Committee on Safety of Medicines’ (CSM) report in 1988 that it was associated with monomyelocytic leukaemia in rats [17]. However, spironolactone has been used in humans for more than 30 years [18], with no reported risk of malignancy apart from five cases of breast carcinoma in women [19], this link being tenuous bearing in mind the high prevalence of this form of malignancy in the female population. In contrast to those studies performed in the late sixties and early seventies, we have used lower doses of spironolactone. This may explain the relatively low incidence of adverse effects and the small dropout rate from our cohort.

Based on our results, we believe that ARR may be an effective way of identifying hypertensive subjects who would respond to spironolactone who represent a significant proportion of the hypertensive population. We believe that there is now good evidence to mount a randomised trial of spironolactone vs other therapy in this group of patients. If our results are generalizable, and confirmed by further studies, our management strategy has the potential to make a major impact on the management of hypertension especially as about 1 in 10 hypertensives may fall into this subcategory.

Limitations

We do not think that ‘regression to the mean’ explains our results. Our patients had their blood pressures measured for 3–6 months by their general practitioners before being labelled hypertensive, and were then referred to our hypertension clinic for assessment. Furthermore, the response to spironolactone was remarkable, in that nearly half the patients needed only spironolactone to control their hypertension. Previously, many of these patients needed three or more drugs to achieve blood pressure control. However, we note that in six of the 27 patients who continued on spironolactone either systolic or diastolic blood pressures or both were higher at the follow-up, so this treatment is not uniformly effective in all subjects.

Our study is not free from the inherent biases of an uncontrolled nonrandomized design, and the blood pressure measurements were carried out by at least three different individuals. However, the strength of our study is that it was carried out as part of routine clinical practice and therefore relevant to day-to-day medical care of hypertensive patients. We believe that our results point to the need for a randomised controlled study to assess the response to spironolactone in this subgroup of hypertensives.

Conclusions

Low dose spironolactone was highly effective in reducing blood pressure in hypertensive patients with nonadenomatous primary aldosteronism identified with a raised aldosterone to renin ratio. Treatment with spironolactone significantly reduced the need for antihypertensive drug therapy.