mean baseline BPs, baseline serum and urinary laboratory
characteristics, and antihypertensive medication (Table 1).
Baseline mean serum aldosterone and ARR tended to be
nonsignificantly higher in the placebo group (122 ng/L and
32.3) compared to the spironolactone group (94 ng/L and
15.2) (P
ϭ0.075 for aldosterone and Pϭ0.406 for ARR).
The mean age of patients was approximately 61 years, heart
rate was 69 bpm, and body mass index was 32.3 kg/m
2
. Mean
office BP was 154/92 mm Hg, daytime ABPM BP was 142/
82 mm Hg, and 24-hour ABPM BP was 141/80 mm Hg.
Isolated systolic hypertension (office systolic BP
Ͼ140 mm Hg
and diastolic BP
Ͻ90 mm Hg) was present in 36.4% of patients
in the spironolactone group and in 39.3% of patients in the
placebo group. Patients were using a mean of 4.6 antihyperten-
sive drugs in the spironolactone group and 4.5 in the placebo
group; the median was 4 antihypertensive drugs in each group.
Most patients used either hydrochlorothiazide or indapamide. A
small number of patients used a combination of more diuretics,
such as a fixed combination of hydrochlorothiazide with
amiloride (
Ͻ22% in each group, mean dose of amiloride of 3.75
mg/day), and a few used a combination of hydrochlorothizide
with furosemide or indapamide with furosemide.
The change of BP values after 8 weeks of treatment
compared to baseline is shown in Table 2. The difference
between the fall of mean ABPM daytime systolic BP between
the spironolactone and placebo groups was
Ϫ5.4 mm Hg
(95% CI
Ϫ10.0; Ϫ0.8) (Pϭ0.024). The difference between
the groups in the fall of ABPM daytime diastolic BP was not
significant
Ϫ1.0 mm Hg (Ϫ4.0; 2.0) (Pϭ0.358).
A significantly greater reduction of systolic BP was ob-
served in all measured systolic BP parameters, whereas the
difference of diastolic BP reduction was not significant
between the spironolactone and placebo groups (Table 2).
Spironolactone significantly reduced pulse pressure in all of
the ABPM measurements (Table 2). When a stratification
analysis according to diastolic BP at entry was applied, the
reduction of BP was not significantly different between the
study groups in patients with isolated systolic hypertension,
systolic-diastolic hypertension, or with ABPM daytime dia-
stolic BP above or below 85 mm Hg (see online supplemental
Table S1 at http://hyper.ahajournals.org).
A small comparable weight gain was observed in both study
groups (Table 2). Also, serum sodium did not change signifi-
cantly between groups. Serum potassium increased by a median
0.3 mmol/L and serum creatinine by a median 7
mol/L in the
spironolactone group. The mean serum potassium increased
during the 8 weeks of spironolactone treatment from 4.15 to
4.52 mmol/L, and the highest serum potassium value reached at
8 weeks was 5.53 mmol/L. No patient was excluded from the
study because of severe hyperkalemia, progression of renal
insufficiency, or inadequate drug compliance.
The goal office systolic BP
Ͻ140 mm Hg at 8 weeks was
reached in 30 (54.5%) patients using spironolactone and in 24
(42.9%) patients using the placebo (P
ϭ0.257). The respective
office diastolic BP goal
Ͻ90 mm Hg was reached in 38
(69.1%) patients using spironolactone and 36 patients using
the placebo (64.3%) (P
ϭ0.688).
To evaluate the BP response to treatment, both the spirono-
lactone and placebo groups were divided into tertiles. As
168 patients with resistant hypertension
initially screened for study eligibility
51 patients ineligible
11 declined participation in the trial
22 had significant white coat
syndrome and normal BP values on
ABPM
8 non-compliance
4 secondary cause of hypertension
found
4 did not meet inclusion or exclusion
criteria
117 patients randomized and treated
59 assigned to
spironolactone
58 assigned to
placebo
4 patients
discontinued
2 due to adverse
events
2 due to protocol
violation
2 patients
discontinued
1 due to adverse
event
1 withdrew
consent
55 patients
completed
study
56 patients
completed
study
Figure 1. Trial profile.
Va´clavík et al
Spironolactone in Resistant Hypertension
1071
by guest on October 5, 2017
http://hyper.ahajournals.org/
Downloaded from
expected, in the placebo group, the changes of BP from
baseline to 8 weeks between tertiles of baseline serum
potassium, baseline aldosterone, ARR, and PRA were insig-
nificant. The only baseline parameter that significantly pre-
dicted both systolic and diastolic BP response to spironolac-
tone treatment after 8 weeks was the baseline ARR; mean
change of systolic/diastolic BP in the first tertile (ARR
Ͻ7)
was
Ϫ4.0/0.0 mm Hg, in the second tertile (ARR 7 to 45) was
Ϫ13.0/Ϫ5.0 mm Hg, and in the third tertile (ARR Ͼ45) was
Ϫ15.0/Ϫ7.0 mm Hg (Pϭ0.019 for systolic and Pϭ0.049 for
diastolic 24-hour ABPM BP) (Table 3). Baseline PRA
significantly predicted systolic 24-hour BP response (with the
greatest BP response in the lowest tertile of baseline PRA,
P
ϭ0.006), but not diastolic BP response (Pϭ0.107) (Table
3). The baseline aldosterone value did not significantly
predict BP response to spironolactone treatment.
In 28 patients (24%) enrolled into the trial, the secondary
cause of hypertension was found during subsequent evalua-
tion after trial completion with comparable distribution in
both study arms: primary aldosteronism (8 in the spironolac-
tone and 9 in the placebo groups), renovascular hypertension
(3 and 3), obstructive sleep apnea (1 and 2), and nephrogenic
hypertension (1 and 1).
The frequency of adverse events was comparable in both
study arms (see online supplemental Table S2). Serious adverse
events leading to study medication discontinuation occurred in 2
patients using spironolactone and in 1 patient using the placebo
(P
ϭ0.618). The total number of adverse events was 24 in the
spironolactone group and 26 in the placebo group.
Discussion
This randomized trial showed that the addition of spironolac-
tone in patients with resistant arterial hypertension using a
Table 1.
Patient Demographics and Baseline Characteristics
(Completed Study Set)
Patient Characteristics
Spironolactone
Group (n
ϭ55)
Placebo
Group (n
ϭ56)
Demographic
characteristics
Age, years
61.4 (
Ϯ9.6)
60.1 (
Ϯ9.4)
Sex (female)
18 (32.7%)
24 (42.9%)
Height, cm
173.1 (
Ϯ8.9)
170.7 (
Ϯ8.3)
Weight, kg
96.9 (
Ϯ17.1)
94.1 (
Ϯ17.3)
BMI, kg/m
2
32.3 (
Ϯ5.1)
32.3 (
Ϯ5.3)
Heart rate, bpm
67.8 (
Ϯ10.4)
70.0 (
Ϯ9.2)
Mean baseline BP
Office systolic
BP, mm Hg*
154.9 (
Ϯ10.4)
153.5 (
Ϯ12.0)
Office diastolic
BP, mm Hg*
92.6 (
Ϯ10.7)
90.6 (
Ϯ10.9)
ABPM systolic daytime
BP, mm Hg
144.7 (
Ϯ14.8)
140.1 (
Ϯ16.2)
ABPM diastolic
daytime BP, mm Hg
83.6 (
Ϯ11.1)
80.9 (
Ϯ10.4)
ABPM systolic
nighttime BP, mm Hg
136.4 (
Ϯ19.0)
134.4 (
Ϯ20.4)
ABPM diastolic nighttime
BP, mm Hg
76.7 (
Ϯ13.4)
74.1 (
Ϯ11.8)
24-h ABPM systolic
BP, mm Hg
143.1 (
Ϯ13.5)
139.8 (
Ϯ16.4)
24-h ABPM diastolic
BP, mm Hg
81.1 (
Ϯ10.2)
79.3 (
Ϯ10.2)
Baseline serum
laboratory characteristics
Na, mmol/L
140.4 (
Ϯ2.8)
140.9 (
Ϯ3.0)
K, mmol/L
4.2 (
Ϯ0.5)
4.2 (
Ϯ0.5)
Cl, mmol/L
104.0 (
Ϯ3.8)
103.8 (
Ϯ3.2)
Urea, mmol/L
6.2 (3.8; 10.4)
5.8 (4.0; 10.0)
Creatinine,
mol/L
81.0 (56.0; 128.0)
83.0 (55.0; 128.0)
Glycemia, mmol/L
6.0 (4.6; 17.3)
6.5 (4.5; 12.5)
PRA, ng/ml/h
0.4 (0.1; 5.8)
0.3 (0.1; 8.2)
Aldosteronem ng/L
94 (23; 297)
122 (34; 430)
ARR†
15.2 (1.6; 235.0)
32.3 (0.9; 322.0)
Metanephrine, ng/L
30.0 (15.0; 95.8)
30.0 (15.0; 61.2)
Normetanephrine, ng/L
60.0 (30.0; 147.7)
60.0 (30.0; 184.3)
TSH, mIU/L
1.7 (0.3; 5.0)
1.7 (0.4; 7.4)
Cortisol, nmol/L
458 (273; 767)
475 (266; 822)
Baseline urinary
laboratory characteristics
Total urinary cortisol
(nmol/day)
250.0 (38.4; 750.0)
234.0 (50.9; 559.0)
Free urinary cortisol,
nmol/day
69.7 (0.0; 1 036.0)
52.4 (27.9; 810.6)
Microalbuminuria,
mg/day
12.8 (0.0; 347.0)
15.0 (2.5; 221.0)
Proteinuria, g/day
0.1 (0.0; 2.8)
0.2 (0.0; 4.4)
(Continued)
Table 1.
Continued
Patient Characteristics
Spironolactone
Group (n
ϭ55)
Placebo
Group (n
ϭ56)
Medication at
randomization
Angiotensin-converting
enzyme inhibitor
42 (76.4%)
43 (76.8%)
-blocker
41 (74.5%)
47 (83.9%)
Calcium channel blocker
49 (89.1%)
43 (76.8%)
Diuretics
55 (100.0%)
56 (100.0%)
Angiotensin II receptor
blocker
25 (45.5%)
27 (48.2%)
␣-blocker
8 (14.5%)
5 (8.9%)
Centrally acting
antihypertensives
32 (58.2%)
31 (55.4%)
Other antihypertensives
2 (3.6%)
0 (0.0%)
Median no. of
antihypertensives
4 (3; 6)
4 (3; 6)
Data are mean (SD) when normally distributed and median (5th and 95th
percentile range) when they have non-normal distributions. Categorical vari-
ables are number (percentage). None of the baseline parameters is statistically
significantly different between the groups.
*Average of second and third office BP measurement.
†Calculated as serum aldosterone (ng/L)/
͓10 ϫ PRA (ng/ml/h)͔.
1072
Hypertension
June 2011
by guest on October 5, 2017
http://hyper.ahajournals.org/
Downloaded from