Original article 2217

Low-renin status in therapy-resistant hypertension: a clue to

efficient treatment
Ivar K. Eidea , Peter A. Torjesend , Anders Drolsumb , Almira Babovicc and
Nils P. Lilledahle

Objective Therapy resistance is an enduring problem in treatment. Through the subsequent 6–12 months of open
clinical hypertension. Our aims were to estimate: (1) the treatment, seven patients (18%) showing an escape
contribution of a low-renin status in therapy resistance; (2) phenomenon had their high blood pressure effectively
whether such status could give a clue to more successful treated by extra amiloride. Of the 60 low-renin patients,
treatment; and (3) the contribution by adrenal cortical eight had adrenal adenoma.
adenomas and by primary aldosteronism.
Conclusion A low-renin status characterized two-thirds of
Setting Patients were referred from general and internal patients with treatment-resistant hypertension, who could
medicine practices following written invitations and be treated efficiently by aldosterone inhibition. Patients
included consecutively. Participants were examined and with an escape phenomenon (18%) could effectively be
followed-up on an outpatient basis. treated by increasing the aldosterone inhibitor. Low-renin
hypertensives had high prevalence of adrenocortical
Design and interventions Patients were divided adenomas and primary aldosteronism. J Hypertens
according to renin status. Low-renin patients were treated 22:2217–2226 & 2004 Lippincott Williams & Wilkins.
with an aldosterone inhibitor in a prospective, randomized,
placebo-controlled, double-blind, cross-over study.
Journal of Hypertension 2004, 22:2217–2226
Main outcome measures Prevalence of low-renin status
in therapy resistance. Blood pressure and hormonal Keywords: adenoma, adrenal, aldosterone, aldosteronism, amiloride,
diuretic, escape, hyperaldosteronism, inhibitor, primary
responses to specific treatment. Numbers of
a
adrenocortical adenomas and primary aldosteronism. Departments of Medicine, b Roentgenology, and c Nuclear Medicine, Ullevål
University Hospital, Oslo, d Hormone Laboratory, Aker University Hospital, Oslo
and e Oslo Eye Center, Majorstua, Oslo, Norway.
Results In 90 treatment-resistant hypertensive, 67% had
Conflicts of interest: None
plasma renin activity (PRA) below 0.5 nmol/l per hour. Of
the 60 low-renin patients, 38 were studied on a fixed Correspondence and requests for reprints to Ivar Eide, MD, Head of Nephrology
and Professor, Department of Medicine, Ullevål University Hospital, Kirkeveien
combination of amiloride and hydrochlorothiazide. Three 166, N-0407 Oslo, Norway.
weeks’ treatment reduced blood pressure by 31/15 mmHg Tel: +47 67 53 64 48; fax: +47 22 11 91 81; e-mail: [email protected]
compared to placebo (P < 0.0001). Serum aldosterone and Received 15 March 2004 Revised 24 June 2004
plasma renin activity increased substantially during active Accepted 25 June 2004

Introduction dures may be complicated by the necessity of suspend-

According to recent evidence [1–5], the prevalence of ing ongoing antihypertensive treatment. Laragh9s group
primary aldosteronism and adrenal adenomas may be initially found the hyporeninism autonomous, in as
higher than previously estimated, also in those with much as plasma renin activity (PRA) hardly increased
treatment-resistant hypertension. A closely related during sodium depletion, while such autonomy did not
hypertension variety, originally described by Laragh apply to sodium depletion during antihypertensive
and co-workers [6–8] and coined low-renin hyper- medication with diuretics, which increased PRA consid-
tension, accounts for nearly 30% of unselected essential erably [10]. In contrast to the latter observation, Wein-
hypertensives, while the contribution to treatment berger and Fineberg [11], Gordon and co-workers [1,2]
resistance by these patients has not yet been quantified and recently Gallay et al. [12] found that low-renin
extensively. Since this kind of high blood pressure may hypertension may still be diagnosed despite a wide
need specific treatment [9], the diagnosis of a low-renin range of ongoing antihypertensive treatment.
status might give a clue to more effective therapy, even
in resistant hypertension. Our primary aim was to estimate the prevalence of low
PRA in therapy-resistant hypertension; secondly, to
As with primary aldosteronism, the diagnostic proce- examine whether resistant patients with a low-renin
0263-6352 & 2004 Lippincott Williams & Wilkins

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 2218 Journal of Hypertension 2004, Vol 22 No 11

status may be treated successfully with an aldosterone and night-time dips to 142
22/86
14 mmHg (97–
inhibitor; and, thirdly, to investigate whether adrenal 211/57–114). At night, systolic blood pressure increased
cortical adenomas or primary aldosteronism may be in two only and remained unchanged in seven, diastolic
prevalent in low-renin therapy resistant essential hyper- blood pressure increased in seven and was unchanged
tension. in three. The remaining 17 patients did not have their
24-h blood pressure recorded. They included four with
Methods adrenal adenomas, two with renal artery stenosis (one
Patients
Takayasu’s disease), one excluded due to multiple
In August 2000 and August 2001 all physicians in carotid stenoses, three who did not meet for screening
general and internal medicine practice in Oslo, Norway, or work-up, two normotensive patients, four who failed
were invited in writing to refer their patients with or refused 24-h blood pressure recording and one
therapy-resistant hypertension to our hypertension ser- already on amiloride/hydrochlorothiazide (HCT).
vice. Characteristics of the 90 referred patients are
shown in Table 1. All were examined and followed up The 90 referred patients had been on their therapeutic
on an outpatient basis. Twenty-five had retired but regimens, consisting of at least three simultaneous
otherwise most were still working, and all felt relatively drugs including a diuretic, for 2.7
4.0 years (range 1.2
well despite heavy dosage of antihypertensives and months–10 years). Fifty-one (56%) patients were taking
serious hypertension, except for impotence in most of four or more antihypertensive drugs (Table 2).
the men and crural oedema in some patients. Otherwise
their clinical status was mostly normal, yet 13 had a Clinical and laboratory work-up
thrusting apex beat. Therapy resistance was defined Blood pressure was measured with mercury sphygmo-
according to international convention as blood pressure manometers (Korotkov phase 5) throughout the study.
above 140/90 mmHg despite a medication of at least After screening, patients also had an initial 24-h blood
three different drugs including a diuretic [13]. pressure recording (Spacelabs, Spacelabs Medical, Red-
mond, Oregon, USA), urinalysis, ophthalmology, gener-
In the complete sample, blood pressure was 180
22/ al haematology, serum electrolytes, kidney function
105
11 (means
SD, range 145–244/65–137) as including serum urea, serum creatinine and a 24-h
measured by the referring physicians (pre-study level) creatinine clearance and serum lipids.
and an initial 193
28/105
16 mmHg (132–255/60–
201) in the Hypertension Service (screening). In 73 of Initial work-up PRA was determined at noon, after 30
these patients, 24-h ambulatory recordings showed day- min ambulation but with no prior supine rest, using kits
time pressures of 156
22/97
13 (111–218/68–140) from DiaSorin (Stillwater, Minnesota, USA; Gamma-
Coat PRA 125 I RIA Kit), with an intra-assay coefficient
of variation (CV) of 8%, and inter-assay CV of 17% for
Table 1 Characteristics of referred patients
the low range (, 1.7 nmol/l per h) and 11% for high
concentrations. The detection limit for PRA was
Number 90 patients 0.2 nmol/l per h. The lower normal reference limit for
Age 58
12 years (31–81)
Women 32 [35%] PRA was 0.5 nmol/l per h (mean  2 SD), and the
Referral blood pressure (mmHg) 180
22 (145–244)/105
11 (65–137) upper normal reference limit for serum aldosterone
Hypertension 13.6
11.3 years (0.5–53)
Resistance 2.7
4.0 years (1.2 months–10 years)
400 pmol/l (mean þ 2 SD). As a double-check of low
Weight 89.5
16 kg (53–134) PRA levels, follow-up assays were carried out using a
BMI 29.8
4.2 (20.7–38.3) different technique, i.e. the Active Renin IRMA Kit
Hgb 14.2
1.6 g/dl (9.8–17.9)
Sodium 139
2.4 mmol/l (130–144) from Nichols Institute Diagnostics (San Juan Capistra-
Potassium 3.9
0.4 mmol/l (2.5–4.7) no, California, USA) (reference supine range 5–47 with
Calcium 2.35
0.22 mmol/l (2.40–2.65)
Creatinine clearance 108
34 ml/min (51–189)
Cholesterol 5.4
1.0 mmol/l (3.2–8.4)
HDL 1.3
0.4 mmol/l (0.65–2.62) Table 2 Antihypertensive regimens
Retinopathy K.-W. grade II 19 [21%]
Retinopathy K.-W. grade III 5 [6%] Diuretics 90 (100%)
Micro- or macroalbuminuria 28 [31%] Two simultaneous diuretics 12 (13%)
Casts 14 [16%]
-Blocker 69 (77%)
Reduced creatinine clearance 9 [10%] A1 Blocker 63 (70%)
LVH (ECG* and/or chest X-ray) 46 [51%] Calcium-channel blocker 46 (51%)
Coronary heart disease 14 [16%] Angiotensin-converting enzyme (ACE) inhibitor 27 (30%)
Peripheral arteriosclerosis 4 [5%] Æ-blockers 21 (23%)
Cerebrovascular accident 5 [6%] Three simultaneous agents 35 (39%)
Four simultaneous agents 37 (41%)
Data presented as means
SD where relevant. Values in parentheses are Five simultaneous agents 11 (12%)
ranges; those in square brackets are percentages. BMI, body mass index; Hgb, Six simultaneous agents 3 (3%)
haemoglobin; HDL, high-density lipoprotein; LVH, left ventricular hypertrophy; Did not meet 4 (4%)
ECG, electrocardiogram. *Sokolov–Lyon.

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 Low renin in resistant hypertension Eide et al. 2219

a geometric mean of 14.8 U/ml); the two renin meth- amiloride/HCT or placebo and, if one tablet was
ods correlated according to y ¼ 0.151 þ 0.013x and tolerated but blood pressure was not or only clinically
r ¼ 0.60. Serum aldosterone was assayed at noon, after insufficiently reduced, the dosage was doubled. This
30 min ambulation, using a radioimmunoassay (RIA) kit early visit was selected exclusively as a control for any
from Diagnostic Products Corporation (DPC) (Los kind of intolerance according to current literature.
Angeles, California, USA), with an inter-assay CV of 13 Then followed 2 weeks treatment on permanent dosage
and 6%, respectively, for low and high concentrations. and a further washout period for 1–2 months, through-
Serum aldosterone was double-checked by the RIA out which the original diuretic was again given. The
DSL-8600 kit from Diagnostic Systems Laboratories, patients were thereafter crossed-over to an identical
Inc. (Webster, Texas, USA) (reference range for supine sequence with the alternative active drug/placebo.
adults, 29.4–161.5 pg/ml (75–1080 pmol/l)). Urine aldo- Through 6–12 months following termination of the
sterone was determined by the same DPC-kit as for final washout period, the patients were treated un-
serum, after hydrolysis and ether extraction of free and blinded with amiloride/HCT until a final 24-h blood
conjugated steroids, with an inter-assay CV of 11% over pressure recording.
the complete assay range. The reference range was 10–
50 nmol/24 h. Plasma catecholamines were assayed
using the radioenzymatic method of Peuler and John- Statistical analysis
son [14], as described previously [15]. Throughout, Statistical work-up was carried out with the StatView
laboratory analyses of any kind were undertaken exam- program for MacIntosh. Therapeutic effects were eval-
iner-blinded. uated with two-tailed Student’s t-tests, both paired and
unpaired, and with confidence intervals of blood pres-
Dynamic MAG-3 renal scintigraphy was performed sure reductions (Delta). Correlations were calculated by
using a DS-7 Gammacamera (General Electric, Paris, the Fisher9s r to z test and by confidence limits for
France) with large field, LEAP collimator and 140 KeV correlation coefficients. Results are given as means
energy-window. Contrast-enhanced computed tomogra-
SD with range in brackets, except for in the figures,
phy (CT) examinations of the adrenals and renal where means
SE are given. Statistical significance
arteries/kidneys were performed on a 4-detector helical was defined as P < 0.05.
scanner (CT lights speed QX1; GE Healthcare Tech-
nologies, Milwaukee, Wisconsin, USA) with 2.5 mm
collimation, 3.75 mm slice thickness, 1.5 mm data re- Results
construction interval, 7.5 mm/s table speed and pitch 3. Primary aim
Of the 90 referred patients with therapy-resistant
Prospective therapeutic study on patients with low-renin hypertension, as many as 60 (67%) had a low-renin
status status. The different subgroups are shown in Figure 1.
Eligible low-renin patients went through a prospective,
placebo-controlled, randomized, double-blind, cross- Low-renin patients
over study with Moduretic Mite (MSD) and placebo. Of the 60 patients with a low-renin status, 45 had
Moduretic Mite is a fixed combination of amiloride PRA below the detection limit of 0.2, 8 had 0.3 and
(2.5 mg) and hydrochlorothiazide (25 mg), kindly sup- 7 had 0.4 nmol/l per hour. As also shown in Figure 1,
plied (unconditionally) with placebo by MSD Norge the 60 included: eight adrenal adenomas; 10 white-coat
AS, Drammen, Norway. All patients signed an informed hypertensives, i.e. normotensive by 24-h ambulatory
consent and the study was approved by the Regional recording; one renal artery stenosis; one considered
Committee for Medical Ethics, Health Region East, non-compliant; one in hypertensive crisis; one already
Norway, as well as the Norwegian Medicines Agency on amiloride/HCT; and one notorious salt addict with
on 19 March 1999. Randomization, coding, dispatching plasma renin and serum aldosterone both below the
of active and placebo tablets, opening the codes and reference limit. Including one with bilateral adenoma,
information of results in writing were undertaken by the remaining 38 were judged acceptable for the
the Norwegian Medicinal Depot. However, these prospective study on amiloride/HCT.
bodies would not tolerate any, even a temporary,
reduction of treatment in patients with serious hyper-
tension. Hypertensive complications
K.-W. retinopathy grade III appeared more than twice
Patients remained on their pre-study regimens through- as frequently in the low-renin group than in the total
out, except for treatment periods with amiloride/HCT sample (13 versus 6%), and the prevalence of left
or placebo, when the original diuretic was discontinued. ventricular hypertrophy was higher in the low-renin
After screening, an enrolment period of 2–3 weeks9 group (63 versus 47%). Otherwise the low-renin group
work-up followed, then a 1-week run-in period on was similar to the total sample (Table 1).

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 2220 Journal of Hypertension 2004, Vol 22 No 11

Fig. 1

Normal-renin* 7 Renal artery stenosis* *(All high-renin) 3

All patients
High-renin* 12 White-coat hypertension (2 high-renin) 5
n = 90 Did not meet 4 Diverse group*** 8
Normotensive 5
Normal-renin essential 4
Excluded 1
Normal-renin hyperaldo 1 Did not meet 4
Normotensive 5
Low-renin *) Include 3 ⫹ 2 white-coat, Normal-renin hyperaldo 1
patients respectively
n = 60 **) One with both adenoma and ren. art.
stenosis, counted as adenoma only

***) High-renin essential 4

Hydronephrosis (high renin) 1
Low-renin patients Adenoma 7 Cerebellar hematoma (high renin) 1
for prospective study White-coat 10 Intrarenal hematoma (high renin) 1
including 20 with Renal artery stenosis 1 Carotid stenoses, excluded 1
No compliance 1
hyperaldosteronism
Hypertensive crisis 1
(one with adenoma)
Already on AMIL/HCT 1
n = 38 Salt addiction 1

Subgroups of the total number of referred patients (n ¼ 90). Normal-renin, normal-renin hypertension; High-renin, high-renin hypertension;
hyperaldo, hyperaldosteronism; AMIL/HCT, amiloride/hydrochlorothiazide.

Second aim ment serum creatinine increased from 90.7
15.5

Prospective study mol/l (61–137 mol/l) to 96.3
18.9 mol/l (64–146
As shown in Figure 1, 38 low-renin patients were mol/l), which was statistically significant (P < 0.005)
accepted for the placebo-controlled sub-study. These but could hardly be considered clinically significant.
had initial supine serum aldosterone levels of 317
Serum potassium increased from 4.0
0.3 mmol/l (3.4–
146 (106–700) pmol/l. After 2 weeks on final treatment 4.7 mmol/l) to 4.1
0.3 mmol/l (3.5–5.0 mmol/l); this
plus the run-in week, 34 of 38 showed a substantial fall was statistically non-significant.
in systolic, and 34 of 38 in diastolic blood pressure,
when compared to placebo (Fig. 2), with pressure Escape
reductions of 31
21/15
11 mmHg (13 to 67/9 to During the final part of the study, one elderly patient
38) (P < 0.0001 for both) (Fig. 3). No regression died suddenly at home, yet hyperkalaemia had not
towards the mean occurred throughout the complete been found; four elderly and two younger patients
set of control and washout periods subsequent to the refused further follow-up and demanded open treat-
initial screening (Fig. 2). Amiloride/HCT was doubled ment with amiloride/HCT; and two patients developed
in 26 patients, and their blood pressure fell further, prostatic cancer. The remaining 29 patients were re-
from 164
22/90
13 (128–212/50–110) to 153
22/ examined with an ambulatory 24-h blood pressure
86
13 mmHg (123–210/60–114) (P < 0.002 and recording after 6–12 months on open amiloride/HCT
P < 0.02 for systolic and diastolic pressure). It should treatment following the final washout period. In these
be pointed out here, that K.-W. retinopathy grade III 29 patients, blood pressure decreased from 164
17/
appeared more than twice as frequently in these 38 101
11 (135–218/84–140) to 147
19/90
12 mmHg
patients than in the total sample (13 versus 6%), and (117–187/69–116 mmHg) after the final washout (P <
the prevalence of left ventricular hypertrophy was high- 0.0005), despite seven patients with an escape phenom-
er (71 versus 47%). enon, i.e. blood pressure had again risen to hyper-
tensive daytime levels averaging 171
10/102
11
In none of the patients did amiloride/HCT have to be mmHg (157–187/92–116 mmHg). This escape phe-
stopped because of side-effects. A few had minor nomenon was treated with an extra 5 mg amiloride and,
abdominal cramps, but otherwise both one and two in six patients, systolic blood pressure again fell by an
tablets were well tolerated. Throughout active treat- average of 16.3
11.9 mmHg (8–39 mmHg) (P < 0.02)

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 Low renin in resistant hypertension Eide et al. 2221

Fig. 2

225 120

200
100

Diastolic blood pressure (mmHg)

Systolic blood pressure (mmHg)

175

150 80

125
60
100

75 40

50
20
25

0 0
CTR 1 CTR 2 PLAC CTR 3 AMIL CTR 4 CTR 1 CTR 2 PLAC CTR 3 AMIL CTR 4

Blood pressures through the prospective study with amiloride/hydrochlorothiazide (P < 0.001). AMIL, amiloride/hydrochlorothiazide, CTR 1,
screening; CTR 2, work-up; CTR 3–4, washout periods; PLAC, placebo (means
SE).

and diastolic by 9.2
10.6 mmHg (2–25 mmHg) (NS, 47 nmol/24 h. Of those with increased urine aldosterone
P < 0.09). We lost contact with the seventh patient. excretion, two had increased serum aldosterone.
After 6 months, the mean dose of amiloride/HCT was Amongst the total sample of originally resistant patients
4.25/42.5 mg/day plus 5 mg amiloride extra in the seven the prevalence of adenoma was 8.9%, among the 60
patients with escape. low-renin patients the prevalence reached 13%.

Both serum aldosterone and PRA increased markedly Aldosteronism

during treatment with amiloride/HCT (Fig. 4), while Among the 38 low-renin patients, 17 had primary
plasma noradrenaline and adrenaline remained un- aldosteronism, with urine aldosterone excretion of
changed. During active treatment the aldosterone/renin 68
14 nmol/24 h (50–100 nmol/24 h) but with normal
ratio decreased from 23
19 pg/U (3–107 pg/U) to serum aldosterone in 13 and high in 4. An additional
11
11 pg/U (1–50 pg/U) (P < 0.0001), as followed- three patients had primary aldosteronism with serum
up by the Active Renin IRMA Kit. aldosterone at 476, 537 and 700 pmol/l.

Third aim The 17 patients with increased urine aldosterone excre-

Adrenal adenoma tion had highly significant increments in urine potas-
Within the low-renin group of 60, eight patients had sium excretion, at 98
32 mmol/24 h (43–146 mmol/
adrenal adenoma, including one bilaterally. He was 24 h) versus 67
26 mmol/24 h (25–134 mmol/24 h)
included in the group of 38 and went for the prospec- (P < 0.005) in the remaining 21 with normal urine
tive study (Fig. 1). Thus far, four underwent adrena- aldosterone. There was also a urine potassium/sodium
lectomy, presenting well-defined adenomas with quotient of 0.72 in those with increased urine aldo-
diameters of 19–34 mm that were histologically con- sterone, versus 0.43 in those with normal urine aldo-
firmed. All the adenoma patients had low renin activity, sterone (nearly statistically significant, with P < 0.07).
six with PRA below 0.2 nmol/l per h and two with The quotient in those with increased serum aldosterone
0.3 nmol/l per h. Four had normal supine serum aldo- was far from significant (P < 0.97).
sterone concentrations between 141 and 363 pmol/l,
three had an increased supine serum aldosterone at In the 38 patients as a whole, serum bicarbonate
601, 613 and 783 pmol/l, and one had a marginally averaged 28.7
2.9 mmol/l (24.3–36.1) and a positive
increased supine serum aldosterone of 465 pmol/l. Four linear correlation (r ¼ 0.40, P < 0.02) appeared between
had increased urine aldosterone excretion, between 63 serum aldosterone and serum bicarbonate, with aldo-
and 110 nmol/24 h, and four were normal, within 42– sterone explaining 16% (r 2 ) of the rise in serum

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 2222 Journal of Hypertension 2004, Vol 22 No 11

Fig. 3 Fig. 4

40 70

Active renin concentration (µU/ml)

60
35
50

30 40

30
25
20
mmHg

20
10

0
15 CTR 1 PLAC CTR 2 AMIL CTR 3

10
180

5 160
140

Aldosterone (pg/ml)
120
0
Delta SBT Delta DBT 100
80
Blood pressure (BP) reduction on amiloride/hydrochlorothiazide. Delta
SBT (P < 0.0001), systolic blood pressure reduction; Delta DBT (P < 60
0.0001), diastolic blood pressure reduction (means
SE). 40
20
0
CTR 1 PLAC CTR 2 AMIL CTR 3

bicarbonate. In these 38 patients, serum potassium was Serum aldosterone and plasma renin in the prospective study on
3.9
0.4 mmol/l (3.2–4.7) and eight patients had levels amiloride/hydrochlorothiazide. AMIL, amiloride/hydrochlorothiazide;
CTR 1, work-up; CTR. 2, first wash-out; CTR 3, second wash-out;
below 3.5 mmol/l; but no correlation was found be- PLAC, placebo. AMIL versus PLAC: for aldosterone (P < 0.002) and
tween the serum aldosterone and potassium, serum for plasma renin (P < 0.001). 1 U, microunits/millilitre ¼ 0.6 pg active
potassium and bicarbonate, or urine potassium and renin enzyme concentration/millilitre plasma (Active Renin IRMA Kit).
serum bicarbonate.

Discussion
The principal and primary result of the present study histology or even by excretion or peripheral concentra-
was that: (1) two-thirds of the patients with treatment- tions, these observations raise the possibility that the
resistant hypertension had low PRA; (2) they responded adenomas were actually secreting aldosterone.
well to a fixed combination of amiloride and HCT; and
(3) there was an important contribution to this by In 18% of the patients in the prospective study, a
adrenal cortical adenomas, far ahead of the 0.5% partial treatment escape occurred. The escape was
prevalence previously estimated in unselected essential probably related to a rise in serum aldosterone or PRA,
hypertension [16,17]. Although only four adenomas while plasma catecholamines remained unchanged. In
have been histologically verified thus far, each of the accordance with the rise in serum aldosterone, we
eight was accompanied by the same positive para- added 5 mg extra amiloride, and once more the systolic
meters, i.e. increased serum aldosterone and/or urine blood pressure decreased significantly. We omitted
aldosterone despite angiotensin-converting enzyme ACE inhibitors or A1 blockers combined with ami-
(ACE) inhibition or A1 receptor blockade, and unde- loride/HCT, but this combination might have been
tectable PRA despite long-lasting, massive diuretic effective, as shown recently for an ACE inhibitor in
therapy and ACE inhibition or A1 blockade. Thus, combination with the novel specific aldosterone inhibi-
there may be ample reason to suspect adenomas in the tor, eplerenone, in patients not stratified for renin status
remaining four as well, yet so far not histologically [18], and also with dual ACE inhibitor/A1 blocker
verified. On the other hand, while hypersecretion of treatment combined with aldosterone inhibition in low-
aldosterone cannot be verified by either CT scan, renin yet normotensive subjects [19].

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 Low renin in resistant hypertension Eide et al. 2223

All our patients were either on ACE inhibitors, A1 well as the many cases of primary aldosteronism. Due
blockers or both, which would certainly increase the to general and well-known assay problems with low
PRA. Nevertheless, PRA was below or close to the concentrations, follow-up of PRA with a different
detection limit in most of our low-renin patients. methodology (IRMA) was carried out to double-check
Although a considerable number of our patients were low renin levels found during the initial baseline work-
on
-blocker treatment, it is highly questionable up. The IRMA method, subsequently also utilized
whether
-blockade will reduce PRA to undetectable alone and throughout the prospective part of this study,
levels with the DiaSorin Kit. Thus, in one of the most employs a direct radiometric assay of human renin by
cited studies of the
-blocker effect on PRA, Bühler monoclonal anti-renin antibodies. This procedure was
et al. [20] found only 60% reduced activities in patients developed and thoroughly evaluated by Derkx et al.
with normal plasma renin and 80% reduction of high [26] but criticized for possibly overestimating plasma
levels. According to Esler et al. [21],
-blockers do not renin concentrations, due to cross-reaction with prore-
reduce PRA to a marked degree even by high doses. nin, concentrations of which are at least 20 times higher
Hollifield et al. [22], and a more recent study by than those of renin [27,28]. The criticism also implied
Eichhorn et al. [23], reported similar PRA reductions that the IRMA technique was sufficiently sensitive for
around 60%, while in the recent RESOLVD Study [24] medium and high renin concentrations only, while not
PRA was reduced only by 37%. In a recent and so for low. Derkx and Schalekamp [29] refuted these
comprehensive study on non-resistant hypertensives, criticisms and showed, in experimental settings, that
Mulatero et al. [25], also with kits from Sorin Biomedi- the IRMA method could readily distinguish low-, as
cal Diagnostics, found that atenolol increased the aldo- well as normal- and high-renin hypertension with good
sterone/renin ratio by 62%. With only a small reduction agreement with enzyme-kinetic assays also in the low
in plasma aldosterone, this ratio matches closely that of renin range.
60% PRA reduction found by Bühler et al. [20] and by
Eichhorn et al. [23]. Finally, 20 (53%) of the patients in In spite of a possible overestimation, as well as supine
the present prospective study had aldosteronism that sampling for IRMA and PRA below the detection limit
self-evidently could not have been the result of
- of 0.2 nmol/l per h that could not be further quantified,
blockade. there was a relatively high degree of correspondence
between our initial plasma renin activities (DiaSorin)
During work-up for aldosteronism in seriously hyper- and the initial direct assays (IRMA) on plasma samples
tensive patients Gordon [2] advised the reduction of taken even on different days. All direct renin concen-
high-dose
-blockers, which can cause false-positives trations (IRMA) were below the physiological geo-
by lowering renin more than aldosterone, and to stop metric mean, and as many as 58% of the patients were
diuretics and reduce calcium-channel blockers, which below the lower reference limit. Within these low
can cause false negatives; but, in order to maintain baseline levels, the IRMA method produced consis-
blood pressure control, not to stop medications. Not to tently low renin concentrations at baseline, as well as
stop medications was also demanded by the present responses to treatment with amiloride/HCT (Fig. 4).
Ethics Committee, while Weinberger and Fineberg We therefore feel convinced that the low-renin status
[11] advised discontinuing
-blockers during work-up, in our patients was a valid observation, although it
since these might still markedly reduce the PRA. should be born in mind that the points raised herein
Unfortunately, in our patients, it was necessary not to can be addressed only within the limitations imposed
stop
-blockade, in view of the resistant hypertension. by measuring renin and aldosterone while continuing
In our opinion, however, even if
-blockers had re- medications.
duced PRA to undetectable levels, some factor other
than the resulting low renin levels or
-receptors must Our sample of 90 resistant hypertensives included 21
be responsible for the high pressures, not least since all patients with primary aldosteronism, diagnosed by
our patients, in addition, were either on ACE inhibitors, urine aldosterone and/or high serum aldosterone, in-
A1 blockers or both. To us, the alternative mechanism cluding one with normal-renin aldosteronism (Fig. 1).
indicated by low PRA levels most likely was miner- Seventeen of the 38 patients in the prospective sub-
alocorticoidism, as supported by the high frequency of study had increased urine aldosterone. Although aldo-
primary aldosteronism and adenomas. For the badly sterone concentrations and excretion in the present
hypertensive patients, low PRA thus constituted a study were relatively low-grade, they were high
much-needed key to efficient treatment, and aldo- enough, together with low PRA, to spark the suspicion
sterone inhibition markedly reduced the pressures. of CT diagnoses in eight adenoma patients, challenging
a previous underestimation of adenoma prevalence in
The term low-renin hypertension may be contentious essential hypertension [16,17]. Of the eight patients
and even difficult to define. Still, the simple finding of diagnosed with adrenal adenomas by CT, four had high
a low PRA led us to the diagnosis of eight adenomas, as levels of both serum aldosterone and urine aldosterone,

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 2224 Journal of Hypertension 2004, Vol 22 No 11

and the remaining four had increased urine concentra- information and also compromised the simplicity of the
tions only, while all had almost undetectable PRA. present diagnostic and therapeutic approach. The sim-
Thus, a low PRA appeared more sensitive than isolated ple procedure as outlined here and by Gallay et al. [12]
high serum or urine aldosterone in detecting adenomas. may be highly relevant due to an increasing incidence
of treatment-resistant hypertension, i.e. an increasing
None of our patients with primary aldosteronism, or need for multi-drug regimens [30–32]. In addition, the
even adrenal adenoma, had hypokalaemia. In agree- present procedure may be conducted with considerable
ment with previous observations [1,11,19], this variable cost benefit even in the busy general practice, fully in
may accordingly be a less sensitive sign of adenoma. accordance with Gallay et al. [12].
However, urine potassium excretion was increased in
the patients with high urine aldosterone (P < 0.005), A widely recommended indicator of primary aldosteron-
while not so in those with high serum aldosterone. ism has been the aldosterone/renin ratio [1,12,33]. In
Moreover, the potassium/sodium quotient was in- our patients this parameter was not highly specific for
creased by 68% and nearly statistically significant. genuine primary aldosteronism since the low-renin pa-
Urine aldosterone excretion may thus be the more tients without an increased serum or urine aldosterone
reliable diagnostic of primary aldosteronism. On the invariably had a high ratio as well. For adrenal adeno-
other hand, for the diagnosis of adenoma it seems that mas this ratio would be even less specific than for the
both serum and urine aldosterone must be assayed in broader term primary aldosteronism. This is in agree-
each patient to be as sensitive as a low PRA alone, ment with recent observations of Schwartz et al. [34].
while potassium excretion may be of limited value in On the other hand, the aldosterone/renin ratio seems
patients on diuretics. Nevertheless, statistically there quite specific if one accepts low-renin hypertension as
was a highly significant increase in potassium excretion some variant of primary aldosteronism. Thus, in the
in those with increased urine aldosterone. present study the ratio was highly responsive to amilor-
ide/HCT. As a guide to further clinical work-up and
Serum bicarbonate before treatment with amiloride/ treatment, this ratio may be well suited, albeit unspe-
HCT correlated positively with serum aldosterone and cific for a stricter definition of aldosteronism [12].
might support a diagnosis of primary aldosteronism
given a low-renin status, as in our patients, as may both An advantage of the combined amiloride/HCT medica-
the serum aldosterone/renin ratio and the urine potas- tion may be that HCT increases potassium excretion.
sium/sodium quotient. In a hypothetical case of second- Also, HCT acts more proximally than amiloride, there-
ary aldosteronism, neither aldosteronism nor increased by increasing intratubular flow rate to the distal seg-
urine potassium or serum bicarbonate would have ments and augmenting the natriuretic potential of
meant anything other than the consequences of on- aldosterone blockade [35]. We preferred amiloride to
going thiazide or loop diuretic treatment. Altogether, spironolactone because of the problem with gyneco-
the final diagnosis of adrenocortical adenoma may only mastia in men. While amiloride indirectly inhibits the
be secured by adrenal CT- or magnetic resonance effect of aldosterone on distal basolateral Naþ ,K þ -
(MR)-tomography plus histological examination after ATPases, this agent blocks luminal epithelial sodium
removal. However, given the high prevalence of hyper- channels (ENaC), effectively reducing the intracellular
tension, these demanding techniques cannot be gener- substrate for the sodium pump. The action of the
ally utilized, while PRA, aldosterone and related ENaC is so essential that mice phenotypically devoid
parameters may serve as easily available screening of the
-subunit of ENaCs die of pseudohypoaldoster-
variables. onism [36]. Why the low-renin status was associated
with therapy resistance may not be easily explained.
Since blood pressure fell markedly during inhibition of Nevertheless, it may be related to increased ENaC
the aldosterone effect, a primary mineralocorticoid activity in the collecting ducts of the kidneys, which
excess probably explained a considerable proportion of can be decreased by aldosterone inhibition, as sug-
the treatment resistance in our patients with low PRA. gested recently in preliminary experience by Pratt
Complementary diagnostic procedures might have in- [37]. Also, in accordance with Pratt’s results we found
cluded aldosterone suppression in the 20 patients with no effect of ordinary diuretics on therapy-resistant
aldosteronism [1,11], and stimulation of PRA in pa- hypertension, despite enhancing their effect by the
tients with low renin [11]. The former would mean use of ACE inhibitors or A1 blockers in any of our
saline infusion, which we considered too risky at these patients.
high blood pressures, the latter would mean sodium
depletion by diuretics, ACE inhibition or A1 blockade. In the present study, one might probably comment that
However, as all our patients were already on massive high blood pressure during the placebo period could be
treatment with diuretics, as well as ACE inhibition or due to the absence of a diuretic medication. However,
A1 blockade, these procedures would have added little this is not likely since every patient, except during

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 Low renin in resistant hypertension Eide et al. 2225

placebo periods, was treated with diuretics in rather Acknowledgements

high doses. Thus, all the patients in the present We thank Mrs Anne Elise Larsen and Mrs Roseli
treatment study were on diuretics and as many as 17 Andreasson for expert technical assistance.
patients took diuretics equivalent to at least 25 mg
HCT or a thiazide combined with a loop diuretic. Both
pre-study and in comparison with placebo periods,
References
1 Gordon RD, Ziesak MD, Tunny TJ, Stowasser M, Klemm SA. Evidence
these diuretics had no effect on the seriously hyper- that primary aldosteronism may not be uncommon: 12% incidence among
tensive pressures, while amiloride resulted in massive antihypertensive drug trial volunteers. Clin Exp Pharmacol Physiol 1993;
20:296–298.
pressure reductions. Wallach et al. [38] proposed that 2 Gordon RD. Mineralocorticoid hypertension. Lancet 1994; 344:
truly low-renin patients might be singled out by a 240–242.
diuretic challenge, which would exclude more unspeci- 3 Stewart PM. Mineralocorticoid hypertension. Lancet 1999; 353:
1341–1347.
fic PRA suppression caused by sodium intake. On 4 Calhoun DA, Nishizaka MK, Zaman MA, Thakkar RB, Weissmann P.
chronic, relatively high diuretic doses our patients, as Hyperaldosteronism among black and white subjects with resistant hyper-
tension. Hypertension 2002; 40:892–896.
well as those of Weinberger and Fineberg [11] and 5 Mosso L, Carvajal C, Gonzales A, Thakkar RB, Weissmann P. Primary
Gallay et al. [12], were subject to long-lasting and aldosteronism and hypertensive disease. Hypertension 2003; 42:
massive diuretic challenge, and, moreover, each of our 161–165.
6 Laragh JH, Cannon PJ, Ames RP. Aldosterone secretion and various
90 patients took ACE inhibitors and/or A1 blockers, forms of hypertensive vascular disease. Ann Intern Med 1963; 59:
and still maintained their low-renin status. 117–120.
7 Brunner HR, Laragh JH, Baer L, Newton MA, Goodwin FT, Krakoff LR,
et al. Essential hypertension: renin and aldosterone, heart attack and
As might be expected at such high blood pressures as stroke. N Engl J Med 1972; 286:441–449.
in the present sample, signs of hypertensive complica- 8 Laragh JH. Vasoconstriction-volume analysis for understanding and treat-
ing hypertension: The use of renin and aldosterone profiles. Am J Med
tions were frequent. It might have been of particular 1973; 55:261–274.
interest to examine these in depth and also to follow 9 Carey RM, Douglas JG, Schweikert JR, Liddle GW. The syndrome of
them up during treatment [39,40]. In particular, echo- essential hypertension and suppressed plasma renin activity. Normal-
ization of blood pressure with spironolactone. Arch Intern Med 1972;
cardiography might have been of great interest both 130:849–854.
because of the high frequency (Table 1) and unreliable 10 Vaughan ED Jr, Laragh JH, Gavras I, Bühler FR, Gavras H, Brunner HR,
et al. Volume factor in low and normal renin essential hypertension.
LVH criteria by both ECG and radiology. However, we Treatment with either spironolactone or chlorthalidone. Am J Cardiol
aimed to focus on the contribution of a low-renin status, 1973; 32:523–532.
and whether low PRA might offer a key to more 11 Weinberger MH, Fineberg NS. The diagnosis of primary aldosteronism
and separation of two major subtypes. Arch Intern Med 1993;
efficient treatment. To quantify more closely the 153:2125–2129.
occurrence of hypertensive complications in patients 12 Gallay BJ, Ahmad S, Xu L, Toivola B, Davidson RC. Screening for primary
aldosteronism without discontinuing hypertensive medications: plasma
with low PRA and their response to therapy must, aldosterone–renin ratio. Am J Kidney Dis 2001; 37:699–705.
therefore, await further and more extended study. 13 Sixth Report of the Joint National Committee on Prevention, Detection,
Evaluation and Treatment of High Blood Pressure. Arch Intern Med
1997; 157:2413–2445.
In summary, we feel that our three initial aims were 14 Peuler JD, Johnson GA. Simultaneous single isotope radioenzymatic
reached by the data obtained . Two weeks only on final assay of plasma norepinephrine, epinephrine and dopamine. Life Sci
dosage may be too short to show the full effect of 1977; 21:625–636.
15 Kjeldsen SE, Flaaten B, Eide I, Helgeland A, Leren P. Evidence of
amiloride; however, this period was long enough to increased peripheral catecholamine release in patients with long-standing,
show a striking pressure reduction. The present obser- untreated, essential hypertension. Scand J Clin Lab Invest 1982;
42:217–223.
vations may benefit a lot of patients given the high 16 Berglund G, Andersson O, Wilhelmsen L. Prevalence of primary and
prevalence of hypertension, of therapy resistance, and secondary hypertension: studies in a random population sample. BMJ
of low-renin status. Gallay et al. [12] found 83% low- 1976; 2:554–556.
17 Sinclair AM, Isles CG, Brown I, Cameron H, Murray DG, Robertson JW.
renin patients among their resistant cases, and our Secondary hypertension in a blood pressure clinic. Arch Intern Med
reported 67% is not far from their results, leading to 1987; 147:1289–1293.
18 Pitt B, Reichek N, Metscher B, Phillips R, Roniker B, Kleiman J, et al.
similar conclusions. Our low-renin patients responded
Efficacy and safety of eplerenone, enalapril and eplerenone/enalapril
well to aldosterone inhibition. With respect to the combination therapy in patients with left ventricular hypertrophy. Am J
significant contribution by adrenal adenomas, our data Hypertens 2002; 15:23A–24A.
19 Azizi M, Bissery A, Bura-Rivière A, Menard J. Dual renin–angiotensin
fully support those of Weinberger and Fineberg [11], system blockade restores blood pressure–renin dependency in indivi-
Gallay et al. [12] and Gordon and co-workers [1,2]. One duals with low renin concentrations. J Hypertens 2003; 21:1887–1895.
additional advantage of our study, was a reduction in 20 Bühler FR, Laragh JH, Baer L, Vaughan ED, Brunner HR. Propranolol
inhibition of renin secretion. N Engl J Med 1972; 287:1209–1214.
the number of daily medications. While counting 21 Esler MD, Zweifler AD, Randall O, DeQuattro, V. Pathophysiologic and
amiloride/HCT as two drugs throughout, the number of pharmacokinetic determinants of the antihypertensive response to propra-
nolol. Clin Pharmacol Ther 1977; 22:299–308.
antihypertensive substances given certainly turned out 22 Hollifield JW, Sherman K, Vander Zwagg R, Shand DG. Proposed
to be about the same. However, while counting amilor- mechanisms for propranolol’s antihypertensive effect in essential hyper-
ide/HCT as one medication only, the number of pa- tension. N Engl J Med 1976; 295:68–75.
23 Eichhorn EJ, McGhie AL, Bedotto JB, Corbett JR, Malloy CR, Hatfield
tients taking four or more daily medications was BA, et al. Effects of bucindolol on neurohormonal activation in congestive
reduced from 19 to 4 (i.e. 79% reduction). heart failure. Am J Cardiol 1991; 67:67–73.

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 2226 Journal of Hypertension 2004, Vol 22 No 11

24 RESOLVD Investigators. Effects of metoprolol CR in patients with

ischemic and dilated cardiomyopathy. The randomized evaluation of
strategies for left ventricular dysfunction pilot study. Circulation 2000;
101:378–391.
25 Mulatero P, Rabbia F, Milan A, Paglieri C, Morello F, Chiandussi L, et al.
Drug effects on aldosterone/plasma renin activity ratio in primary aldoster-
onism. Hypertension 2002; 40:897–902.
26 Derkx FHM, de Bruin RJA, van Gool JMG, van den Hoek MJ, Beeren-
donck CC, Rosmalen F, et al. Clinical validation of renin monoclonal
antibody sandwich assays of renin and prorenin, and use of renin inhibitor
to enhance prorenin immunoreactivity. Clin Chem 1996; 42:1051–1063.
27 Sealey JE, Laragh JH. Renin and prorenin: advances and declines in
methodology [Editorial]. Clin Chem 1996; 42:993–994.
28 Sealey JE, Catanzaro D, Laragh JH. Reply. Clin Chem 1997; 43:697.
29 Derkx FHM, Schalekamp MADH. More on renin. Clin Chem 1997;
43:694–697.
30 Hansson L, Zanchetti A, Carruthers SG, Dahlöf B, Elmfeldt D, Julius S,
et al. Effects of intensive blood pressure lowering and low-dose aspirin in
patients with hypertension: principal results of the Hypertension Optimal
Treatment (HOT) randomized trial. Lancet 1998; 351:1755–1762.
31 UK Prospective Diabetes Study Group. Tight blood pressure control and
risk of macrovascular and microvascular complications in type 2 diabetes.
UKPDS 38. BMJ 1998; 317:703–713.
32 Kjeldsen SE, Dahlöf B, Devereux RB, Julius S, de Faire U, Fyhrquist F,
et al. Lowering of blood pressure and predictors of response in patients
with left ventricular hypertrophy: the LIFE study. Am J Hypertens 2000;
13:899–906.
33 Racine M-C, Douville P, Lebel M. Functional tests for primary aldosteron-
ism: Value of captopril suppression. Curr Hypertens Rep 2002; 4:
245–249.
34 Schwartz GL, Chapman AB, Boerwinkle E, Kisabeth RM, Turner ST.
Screening for primary aldosteronism: implications of an increased plasma
aldosterone/renin ratio. Clin Chem 2002; 48:1919–1923.
35 Good DW, Wright FS. Luminal influences on potassium excretion:
sodium concentration and flow rate. Am J Physiol 1979; 236:
F192–F205.
36 McDonald FJ, Yang B, Hrstka RF, Drummond HA, Tarr DE, McCray PB
Jr, et al. Disruption of the beta-subunit of the epithelial Naþ -channel in
mice: hyperkalemia and neonatal death associated with a pseudohypoal-
dosteronism phenotype. Proc Natl Acad Sci USA 1999; 96:1727–1731.
37 Pratt JH. Low-renin hypertension: More common than we think? Cardiol
Rev 2000; 8:202–206.
38 Wallach L, Nyarai I, Dawson KG. Stimulated renin: a screening test for
hypertension. Ann Intern Med 1975; 82:27–34.
39 Brilla CJ, Matsubara LS, Weber KT. Anti-aldosterone treatment and the
prevention of myocardial fibrosis in primary and secondary hyperaldoster-
onism. J Mol Cell Cardiol 1993; 25:563–575.
40 Pitt B, Zannad F, Remme WJ, Cody R, Castaigne A, Perez A, et al. The
effect of spironolactone on morbidity and mortality in patients with severe
heart failure. Randomized Aldactone Evaluation Study Investigators.
N Engl J Med 1999; 341:709–717.

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.