ARTICLE IN PRESS
Phytomedicine 17 (2010) 389–396
Contents lists available at ScienceDirect
Phytomedicine
journal homepage: www.elsevier.de/phymed
A randomized, double-blind, placebo-controlled, cross-over pilot study on
the use of a standardized hop extract to alleviate menopausal discomforts
R. Erkkola a, S. Vervarcke b, S. Vansteelandt c, P. Rompotti d, D. De Keukeleire e, A. Heyerick e,n
a
Department of Obstetrics and Gynecology, Turku University Central Hospital, 20521 Turku, Finland
Metagenics Belgium, 8400 Ostend, Belgium
c
Department of Applied Mathematics and Computer Science, Ghent University-UGent, 9000 Ghent, Belgium
d
MaxMedica Oy, 24910 Halikko, Finland
e
Laboratory of Pharmacognosy and Phytochemistry, Department of Pharmaceutics, Ghent University-UGent, Harelbekestraat 72, 9000 Ghent, Belgium
b
a r t i c l e in fo
Keywords:
Menopause
Phytoestrogens
Hop (Humulus lupulus L.)
Kupperman index
Visual analogue scale
Menopause rating scale
abstract
Objectives: To examine the efﬁcacy of a hop extract (standardized at 100 mg 8-prenylnaringenin per
day) for relief of menopausal discomforts.
Methods: A 16-week randomized, double-blind, placebo-controlled, cross-over study was conducted
with 36 menopausal women. The participants were randomly allocated to either placebo or active
treatment (hop extract) for a period of eight weeks after which treatments were switched for another
eight weeks. The Kupperman Index (KI), the Menopause Rating Scale (MRS) and a multifactorial Visual
Analogue Scale (VAS) were assessed at baseline, and after eight and sixteen weeks.
Results: After 8 weeks, both active treatment and placebo signiﬁcantly improved all outcome measures
when compared to baseline with somewhat higher average reductions for placebo than for the active
treatment. After 16 weeks only the active treatment after placebo further reduced all outcome
measures, whereas placebo after active treatment resulted in an increase for all outcome measures.
Although, the overall estimates of treatment efﬁcacy (active treatment-placebo) based on linear mixed
models do not show a signiﬁcant effect, time-speciﬁc estimates of treatment efﬁcacy indicate
signiﬁcant reductions for KI (P = 0.02) and VAS (P = 0.03) and a marginally signiﬁcant reduction
(P = 0.06) for MRS after 16 weeks.
Conclusions: Whereas the ﬁrst treatment period resulted in similar reductions in menopausal
discomforts in both treatment groups, results from the second treatment period suggest superiority
of the standardized hop extract over placebo. Thus, phytoestrogen preparations containing this
standardized hop extract may provide an interesting alternative to women seeking relief of mild
vasomotor symptoms.
& 2010 Elsevier GmbH. All rights reserved.
Introduction
The interest in the use of phytoestrogens for the management
of menopausal complaints has increased considerably since the
widely publicized and discussed results of the Women’s Health
Initiative and the One Million Women Study. Phytoestrogens are
mostly non-steroidal polyphenolic plant-derived compounds
that functionally mimic the activity of the human estrogen,
17b-estradiol and typical sources include soy and red clover
(isoﬂavones), ﬂaxseed (lignans) and hops (prenylﬂavonoids)
(Cos et al., 2003). A large body of scientiﬁc evidence from both
epidemiological and experimental studies suggests that the
n
Corresponding author. Tel.: + 32 9 264 8058; fax: +32 9 264 8192.
E-mail address: [email protected] (A. Heyerick).
0944-7113/$ - see front matter & 2010 Elsevier GmbH. All rights reserved.
doi:10.1016/j.phymed.2010.01.007
consumption of phytoestrogen-rich diets may have protective
effects on estrogen-related conditions, such as menopausal
symptoms (Huntley and Ernst, 2004), and estrogen-related
diseases, such as prostate and breast cancers (Holzbeierlein
et al., 2005; Wu et al., 2008), osteoporosis (Messina et al., 2004;
Ye et al., 2006), and cardiovascular diseases (Clerici et al., 2007).
On the other hand, systematic reviews of randomized, controlled
trials (RCTs) show contradictory results and meta-analyses failed
to demonstrate a statistically signiﬁcant reduction of vasomotor
symptoms for phytoestrogens (Krebs et al., 2004; Kronenberg and
Fugh-Berman, 2002; Lethaby et al., 2007). In more selected
patient populations, however, such as in women in early natural
menopause with mild to moderate vasomotor symptoms, a
systematic review found a signiﬁcant reduction of hot ﬂashes in
ﬁve out of ﬁve RCTs (Tempfer et al., 2007). It should be noticed
that intervention studies with phytoestrogens suffer from large
ARTICLE IN PRESS
390
R. Erkkola et al. / Phytomedicine 17 (2010) 389–396
variabilities in outcome measures and very limited reproducibility between studies. A major factor that could signiﬁcantly
contribute to the highly variable clinical results is the large
interindividual variability in bioavailability with a critical role for
the gut microbiota in the bioactivation of phytoestrogens (Nielsen
and Williamson, 2007; Possemiers et al., 2007). As there is a
scientiﬁc rationale for the efﬁcacy of phytoestrogens and as no
serious side effects have been associated with short term use, the
position statement of the North American Menopause Society
(NAMS) for the treatment of menopause-associated vasomotor
symptoms suggests that changes in lifestyle, either alone or
combined with the use of phytoestrogen preparations should be
considered for the relief of mild vasomotor symptoms (Santoro
et al., 2004). A complete vasomotor symptom treatment algorithm was recently proposed by Panay (Panay, 2007).
Although hops have been known to be estrogenic from both
traditional medicine and anecdotal reports, phytoestrogen preparations based on hops (Humulus lupulus L.) are relatively new,
as the estrogenic principle, 8-prenylnaringenin (8-PN), was only
identiﬁed in 1999 (Milligan et al., 1999). In contrast, most
commercially available hop-derived preparations (often in combination with other plants, such as Valeriana ofﬁcinalis L. and
Melissa ofﬁcinalis L.) are focused on the tranquilizing (sedative)
effect of the plant and lack estrogenic acitivity. Comparisons with
well-known phytoestrogens such as coumestrol (from clover or
alfalfa), genistein and daidzein (from soy) shows that 8-PN is one
of the most potent phytoestrogens currently known (Milligan
et al., 1999). In addition, it was recently discovered that
isoxanthohumol, another hop-derived prenylﬂavonoid that is
usually present in much larger quantities, can act as proestrogen
as it can be metabolized into 8-PN by the intestinal microbiota
(Possemiers et al., 2006, 2005). Animal studies have reported a
variety of beneﬁcial effects related to menopausal complaints and
diseases, including the reduction in tail-skin-temperature as a
model for hot ﬂashes (Bowe et al., 2006), the reduction in serum
LH and FSH and in gonadotropin-releasing hormone (GnRH)
receptor transcripts (Christoffel et al., 2006), and the protection of
ovariectomy induced bone loss with a minimal trophic effect on
¨
the uterus (Humpel
et al., 2005).
Until now, only one RCT on the use of a standardized hop
extract to alleviate menopausal discomforts has been carried out
(Heyerick et al., 2006). This randomized, double-blind, placebocontrolled study over 12 weeks with 67 menopausal women
showed a signiﬁcant reduction in menopausal discomforts and
complaints as assessed by the Kupperman index (KI) and by a
simpliﬁed patients’ questionnaire in both the placebo and
treatment groups. However, the effect of the hop extract
standardized at 100 mg 8-PN per day was only signiﬁcantly
superior to placebo after 6 weeks (P = 0.023), but not after 12
weeks (P = 0.086). Furthermore, no dose-response relationship
could be established, as the higher dose (250 mg 8-PN per day)
appeared less active than the lower dose both after 6 weeks and
after 12 weeks. Both the relatively small sample scale and factors
affecting the bioavailability may have contributed to the absence
of a clear dose-response relationship, as large interindividual
differences in metabolic activation of the proestrogen IX to 8-PN
exist in a typical menopausal population (Bolca et al., 2007). Still,
a trend for a more rapid decrease of the KI was noticed for both
active groups as compared to placebo, while most interestingly,
especially the hot ﬂash score (as an isolated item of the KI)
was signiﬁcantly reduced in both treatment groups after 6 weeks
(P o 0.01).
The current study aimed to further investigate the effect of the
daily intake of a hop phytoestrogen containing dietary supplement standardized on 8-PN on menopausal discomforts using a
randomized, double-blind, placebo-controlled, cross-over design.
Menopausal discomforts were scored by the medical professional
using the KI (Kupperman et al., 1953), while the Menopause
Rating Scale (MRS) (Schneider et al., 2000) and a multifactorial
Visual Analogue Scale (VAS) were used as a self-reporting tool for
the participants.
Materials and methods
Participants
Healthy postmenopausal volunteers were recruited from the
Turku area (Finland) mainly via an announcement in the local
newspaper. In total, fourty women aged between 45 and 60 years
were considered for inclusion in the study. All subjects were
healthy postmenopausal Caucasian females. Subjects were eligible for inclusion in the study if they had an intact uterus and had
not experienced menses for at least 12 months and were willing
to sign informed consent forms after being informed about the
study in detail. Exclusion criteria were history of breast cancer,
endometrial cancer or other hormone-dependent cancers, use of
hormone therapy during the last 3 months before the trial and the
use of a dietary supplement containing phytoestrogens during the
last month before the trial. Subjects were assessed for eligibility
by a gynecologist (RE). Four women did not comply with the
inclusion-exclusion criteria, so ﬁnally 36 women were enrolled
and assigned to the treatments according to the study protocol.
Study protocol
A randomized, double-blind, placebo-controlled, cross-over
study was designed to compare the effects of either placebo or
daily doses of a standardized hop extract on menopausal
complaints as assessed by the KI, the MRS and a multifactorial
VAS. The study was designed for a total of 50 participants and
randomization was carried out by distributing random numerical
codes to the two treatment regimens, ie. 25 test-placebo and 25
placebo-test items, respectively (based on an on-line randomizer,
available at: http://www.random.org/sform.html, last accessed on
September 20, 2009). After randomization, the codes were
arranged in a list according to increasing numbers and blinded
to both the gynecologist and the participant.
The treatments were assigned chronologically to the eligible
participants, who received a box containing 60 capsules of
either placebo (12 participants) or standardized hop extract
(24 participants) for the ﬁrst treatment period, as shown in
Fig. 1. At the next visit after 8 weeks the participants were asked
to return the container and the other box containing the opposite
treatment was then distributed. At the last visit after 16 weeks
after baseline examination. The containers were again returned.
At week 8 and week 16, the returned capsules were counted in
order to check for compliance.
The compositions of the capsules are given in Table 1. Placebo
capsules contained maltodextrin (Lab 2509, La Roquette, Lille,
France) instead of the hop extract. The hop extract in the active
treatment capsules was identical to that contained in the food
supplement MenoHops (PL 162/447, Biodynamics, Ostend,
Belgium, Lifenols-extract (Naturex, Avignon, France)). It is
derived from hydroalcoholic extraction of spent hops, the
material remaining after extraction of hops with liquid or
supercritical carbon dioxide. The dark-green opaque gelatin
capsules size 3 (Capsugel, Bornem, Belgium) additionally
contained dicalcium phosphate (Rhodia, Chicago Heights Illinois,
USA), silicium dioxide (Degussa, Rheinfelden, Germany), and
magnesium stearate (Undesa Union Derivan, Barcelona, Spain).
ARTICLE IN PRESS
R. Erkkola et al. / Phytomedicine 17 (2010) 389–396
391
Fig. 1. The participant’s ﬂow diagram.
Table 1
Composition (mg) of the capsules.
Hop extract
Maltodextrin
Dicalcium phosphate
Silicium dioxide
Magnesium stearate
Total
Placebo
Active treatment
0.0
75.1
124.9
1.0
1.0
202
75.1 (100 mg 8-PN)
0.0
124.9
1.0
1.0
202
(Hop extract: 8-prenylnaringenin: 0.13%, xanthohumol: 2.72%, isoxantohumol:
1.60%, 6-prenylnaringenin: 0.12%).
Capsules were prepared in a semi-automatic capsulation machine
(Ultra 8 II, Capsugel-Divison of Warner Lambert, Greenwood,
South Carolina, USA) and the hop extract capsules were controlled
for content of 8-prenylnaringenin (8-PN or ( 7)-2,3-dihydro5,7-dihydroxy-2-(4-hydroxyphenyl)-8- (3-methylbut-2-enyl)-4Hbenzopyran-4-one) by applying HPLC-MS (in analogy to Rong
et al., 2000). In short, the powder obtained from the capsules was
extracted with methanol under sonication for 15 min at room
temperature and, after ﬁltration, an aliquot of 20 ml was injected
onto an XTerras MS reversed-phase C18 column (250 x 4.6 mm,
5 mm) (Waters, Milford, MA, USA). A linear gradient at a ﬂow rate
of 1 ml.min-1 comprised of 250 ppm formic acid in water (solvent
A) and 250 ppm formic acid in acetonitrile (solvent B) ranging
from 40% B in A to 95% B in A over 40 min was applied. All
analyses were carried out using an Agilent 1200 Series LC/MSD SL
system (Agilent Technologies, Santa Clara, CA, USA). External
calibration was carried out using pure standards obtained via
chemical synthesis (8-PN and 6-prenylnaringenin) or preparative
HPLC of hop cone extracts (isoxanthohumol and xanthohumol).
The capsules were found to contain a total of 3.4470.10 mg of
prenylﬂavonoids per capsule, composed of 1.2070.04 mg
of isoxanthohumol, 0.11 70.01 mg of 8-PN, 0.0970.01 mg of
6-prenylnaringenin, and 2.0470.06 mg of xanthohumol.
Participants in this study were evaluated by three different
tools. The KI was scored by the gynecologist (RE), while both the
MRS and a multifactorial VAS were used as self-assessment
questionnaires. For the KI each item is scored from 0 (not present
or not disturbing) to 3 (very severe) depending on the intensity
(Kupperman et al., 1953). Each score is multiplied by a given
factor depending on the contribution to the index. The maximal
score for the KI is 51. The MRS is a scale based on 11 items divided
into three domains: somatic, psychological, and urogenital
(Schneider et al., 2000). Scoring is based on a ﬁve-point Likert
scale ranging from no symptoms to mild, moderate, marked
or severe complaints. The maximal score for the MRS is 44.
A multifactorial VAS presenting three items, respectively hot
ﬂashes, night sweats and sleep disturbances, was scored by the
women themselves. These items were chosen because hot ﬂashes
and night sweats are the primary menopausal complaints. Sleep
disturbances were added to this VAS because hop extracts have
traditionally been used as sleeping aids. Participants were asked
to rate the intensity of the symptoms for each item by placing a
vertical mark a on a 0-100 mm scale (0 =no complaints,100 mm=
very severe complaints). The scores of the individual items were
combined and the maximal combined score for the VAS is thus
300. The participants ﬁlled in the MRS and VAS before the
gynecologist scored the KI.
Statistical analysis
Comparison of baseline characteristics was based on means
and standard deviations for continuous outcome measures,
provided these were approximately normally distributed according to QQ-plots. If not, they were based on medians and
interquartile range (IQR) otherwise. Raw estimates of treatment
efﬁcacy at each measurement occasion were based on the
intention-to-treat principle, using a Welch Two-sample t-test to
allow for different measurement variance in different treatment
groups. Normality of the outcomes at each time in each treatment
arm was assessed and found satisfactory according to QQ-plots.
In order to obtain more powerful treatment comparisons, the
primary analysis of the data was based on a linear mixed effects’
model for the outcome measurements at the different occasions
jointly, including the baseline outcome (Verbeke and Molenberghs,
2000). We allowed for a nonlinear time evolution under placebo
and for treatment-by-period interaction (different treatment
effects allowed in different periods). Detailed model checking
revealed substantial evidence for the presence of random intercepts (there was substantial evidence that the overall level of
ARTICLE IN PRESS
392
R. Erkkola et al. / Phytomedicine 17 (2010) 389–396
postmenopausal symptoms varies between patients) and a residual
autoregressive correlation structure (informally, there was substantial evidence that measurements taken more closely in time
are more strongly correlated than measurements taken further
apart), both of which were accounted for in all analyses.
All reported results are based on models which ﬁt the observed
data well according to scaled residual plots (Fitzmaurice et al.,
2004). Reported p-values for treatment efﬁcacy are based on Wald
tests. Paired Student’s t-tests (2-tailed) were used to make
comparisons between different time points for all outcome
measures within a treatment group.
Analyses were conducted in SAS Version 9.1 and SPSS 15.0.
Ethics approval for the study
The protocol is consistent with the principles of the Declaration of Helsinki and the Joint Ethical Committee of the University
of Turku and University Central Hospital of Turku approved the
study protocol. An oral and written informed consent was
received from all participants.
Results
Although the study was designed for 50 participants, only 40
participants, of which 36 were found eligible, could be recruited
within a reasonable time frame. Three volunteers were excluded
because of the use of HT and one because of the use of a dietary
supplement containing phytoestrogens. As the randomization list
was set up without block-randomization for an estimated 50
participants, the random allocation to the two treatment groups
coincidentally resulted in considerable differences with 24
participants receiving ﬁrst the active treatment followed by
placebo (test-placebo), while only 12 participants received
placebo ﬁrst followed by the active treatment (placebo-test). No
adverse events were reported and there were no drop-outs.
The baseline characteristics of the women in each treatment
arm were similar (Table 2). As no substantial differences with
respect to these characteristics were found, no adjustments were
made in the primary analysis of treatment effectiveness. Fig. 2
shows the box-and-whisker diagrams for the 3 outcome measures
at baseline, after 8 weeks and after 16 weeks, separately for both
patient groups. Table 3 shows the mean outcome measures
( 7 standard deviations) for both treatment groups after 8 weeks
and after 16 weeks.
After 8 weeks, both active treatment and placebo signiﬁcantly
improved all outcome measures compared to baseline with
somewhat higher average reductions for placebo than for the
active treatment. On the other hand, more pronounced and
opposite effects can be observed for all outcome measures at
Table 2
Characteristics of the patients at baseline (mean 7 SD; time since menopause
measured in median 7 IQR).
Characteristic
Active-Placebo
(n= 24)
Placebo-Active
(n = 12)
Age, yr
Time since menopause, yr
Height, cm
Weight, kg
BMI, kg/m2
Baseline KI
Baseline MRS
Baseline VAS
52.8 7 3.5
4.3 7 3.8
164.3 7 4.7
67.2 7 9.1
24.8 7 3.2
24.2 7 7.2
23.0 7 4.0
173.1 7 61.5
54.3 7 3.7
4.75 7 4.3
163.9 7 3.8
67.7 7 6.3
25.2 7 2.4
23.3 7 9.8
21.3 7 4.9
163.4 7 78.6
SD, standard deviation; IQR, interquartile range; BMI, body mass index; MRS,
menopause rating scale; KI, Kupperman index; VAS, visual analogue scale.
week 16, as the active treatment after placebo further reduced all
outcome measures, whereas placebo after the active treatment
resulted in a slight increase for all outcome measures. Additionally, it can be observed that the measurement variability stays
more or less constant over time.
Table 4 shows the raw estimates of treatment efﬁcacy (active
treatment – placebo) at each time and for each outcome. After 8
weeks, placebo shows somewhat higher efﬁciency on all outcome
measures in comparison with active treatment. After 16 weeks,
the effects are opposite and more pronounced. The results after 16
weeks with p-values of 0.05, 0.06 and 0.07, for VAS, MRS and KI,
respectively, indicate marginal evidence of a treatment effect of
the active treatment over placebo.
In order to enhance power, a joint analysis of the repeated
measurements at each time was performed using linear mixed
models. Although the treatment effects were opposite at months
2 and 4, there was insufﬁcient evidence in the observed data to
distinguish whether this is systematic or just the result of
sampling error. Speciﬁcally, only marginal evidence of treatment-by-period interaction (P = 0.08 for KI, P = 0.09 for MRS, and
P = 0.11 for VAS) was found and, henceforth, it was chosen to base
inference on the overall treatment effect (i.e. averaged over the 2
occasions). Table 5 summarizes the results showing no signiﬁcant
effect of active treatment over placebo. Speciﬁcally, it was found
that the experimental treatment yields non-signiﬁcant reductions
(as compared to placebo and over a 2 month period) in KI with 2.3
(P = 0.10, 95% CI: -5.1; 0.5) on average, in MRS with 0.6 (P = 0.40,
95% CI: -1.9; 0.8) on average, and in VAS with 15.3 (P = 0.28, 95%
CI: -43.2; -12.7) on average. To correct for possible baseline
imbalances and to reduce outcome variation, a further secondary
analysis was performed and adjustment for baseline covariates
was carried out. This did not affect the treatment effect estimates
on MRS and KI because none of the baseline covariates were
signiﬁcantly associated with the outcomes. Treatment effect
estimates on VAS changed slightly from -15.3 to -16.0 with a
corresponding p-value of 0.25 instead of 0.28.
Given the fact that opposite treatment effects were observed at
the 2 follow-up measurement occasions, we subsequently
estimated treatment effects at the 2 time points separately.
Table 6 summarizes the results and reveals signiﬁcant treatment
effects on the average KI (P = 0.02) and the average VAS at week
16 (P = 0.03), whereas the treatment effect on the average MRS
was only marginally signiﬁcant (P = 0.06). The active treatment
lowered the KI by 5.9 (95% CI: -10.8; -1.1), on average, at week 16
as compared to placebo. Likewise, the active treatment lowered
the VAS with 45.4 (95% CI -86.0; - 4.7) on average at week 16 as
compared to placebo.
Finally, all outcome measures within each group were
compared at each time point using paired Student’s t-tests
(2-tailed) (for P-values see Fig. 2). Although both the active
treatment and the placebo show signiﬁcant reductions in all
outcome measures after 8 weeks when compared to baseline
(P o 0.05), only the active treatment after placebo retains
signiﬁcance for all outcome measures after 16 weeks when
compared to baseline, whereas placebo after active treatment
loses signiﬁcance for both KI and MRS when compared to
baseline.
Discussion
Although the overall estimates of treatment efﬁcacy in this
randomized, double-blind, placebo-controlled, cross-over study
do not indicate any signiﬁcant effect of the treatment in
comparison to placebo, more detailed analysis uncovers interesting indications of the superiority of the active treatment over
ARTICLE IN PRESS
R. Erkkola et al. / Phytomedicine 17 (2010) 389–396
393
Active Placebo
Placebo Active
60
60
P = 0.244
50
P = 0.017
P = 0.066
30
40
KI
KI
40
20
10
10
0
0
After 8 weeks After 16 weeks
Time
Baseline
40
P = 0.021
After 8 weeks After 16 weeks
Time
P = 0.277
25
30
20
15
15
10
10
After 8 weeks After 16 weeks
Time
Baseline
P = 0.030
300
P = 0.007
300
P = 0.306
VAS
200
100
0
0
After 8 weeks After 16 weeks
Time
P = 0.786
After 8 weeks After 16 weeks
Time
P = 0.008
P = 0.038
P = 0.448
200
100
Baseline
P = 0.002
25
20
Baseline
P = 0.019
35
MRS
30
MRS
P = 0.408
40
P = 0.143
35
VAS
P = 0.008
30
20
Baseline
P = 0.008
50
Baseline
After 8 weeks
Time
After 16 weeks
Fig. 2. Box-and-whisker diagrams for both treatment groups at each time and for each outcome (Kuppermann index (KI), Menopause Rating Scale (MRS), Visual analogue
Scale VAS) with P-values for paired t-test comparisons between different time points within the treatment group (n = outlier).
Table 3
Mean outcome measures (7 SD) for both treatments groups.
Outcome
measure
KI
MRS
VAS
Active-Placebo
Placebo-Active
8 weeks
8 weeks
16 weeks
16 weeks
19.9 7 10.1
22.4 7 7.5
18.2 7 8.5 15.9 7 10.3
20.6 7 4.6
21.4 7 4.9
18.3 7 4.3 17.9 7 4.9
135.1 7 87.4 146.6 7 76.0 115.8 7 77.7 95.2 7 68.9
SD, standard deviation; KI, Kupperman index; MRS, menopause rating scale; VAS,
visual analogue scale.
placebo. Cross-over trials have some inherent issues. For example,
the order of the treatment may have a considerable effect on
the overall outcome, especially when a placebo effect is likely.
When such a set-up is being used in studies on menopausal
complaints, it should be taken into account that menopausal
complaints are typically sensitive to a large and persistent
placebo effect, especially over shorter time periods (Lethaby
et al., 2007). In addition, menopausal complaints also naturally
decrease over longer time periods. As expected, also in this study,
placebo results in a signiﬁcant reduction in menopausal complaints over the ﬁrst eight weeks (20%-35% depending on the
speciﬁc outcome). The reduction in menopausal complaints for
placebo is even more pronounced than for the active treatment
after eight weeks.
In sharp contrast, switching from active treatment to placebo
after eight weeks resulted in a slight increase in menopausal
complaints after sixteen weeks, whereas switching from placebo
to active treatment further reduced the menopausal complaints.
ARTICLE IN PRESS
394
R. Erkkola et al. / Phytomedicine 17 (2010) 389–396
Table 4
Raw estimates of treatment efﬁcacy (active treatment - placebo) at each time and
for each outcome.
Outcome
Week
Effect
95% CI
P
KI
8
16
8
16
8
16
1.7
-6.5
2.3
-3.5
19.4
-51.5
-4.9 ; 8.3
-13.6 ; 0.6
-0.9 ; 5.5
-7.1 ; 0.1
-39.7 ; 78.5
-103.5 ; 0.6
0.60
0.07
0.15
0.06
0.50
0.05
MRS
VAS
CI, conﬁdence interval; KI, Kupperman index; MRS, menopause rating scale; VAS,
visual analogue scale.
Table 5
Overall estimates of treatment efﬁcacy (active treatment - placebo) for each
outcome, based on linear mixed model analysis.
Outcome
Effect
95% CI
P
KI
MRS
VAS
-2.3
-0.6
-15.3
-5.1 ; 0.5
-1.9 ; 0.8
-43.2 ; 12.7
0.10
0.40
0.28
CI, conﬁdence interval; KI, Kupperman index; MRS, menopause rating scale; VAS,
visual analogue scale.
Table 6
Time-speciﬁc estimates of treatment efﬁcacy (active treatment - placebo) for each
outcome, based on linear mixed model analysis.
Outcome
Week
Effect
95% CI
P
KI
8
16
8
16
8
16
1.1
-5.9
1.3
-2.8
13.3
-45.4
-3.1 ; 5.3
-10.8 ; -1.1
-0.9 ; 3.5
-5.6 ; 0.1
-30.1 ; 56.7
-86.0 ; - 4.7
0.60
0.02
0.24
0.06
0.61
0.03
MRS
VAS
CI, conﬁdence interval; KI, Kupperman index; MRS, menopause rating scale; VAS,
visual analogue scale.
This observation is conﬁrmed by the secondary analyses taking
into account the time-speciﬁc estimates of treatment efﬁcacy,
showing signiﬁcant effects of active treatment in comparison to
placebo for both KI and VAS and a marginally signiﬁcant effect for
MRS, suggesting that the standardized hop extract is superior to
placebo with respect to the potency to reduce menopausal
complaints.
The results of this study should be interpreted with caution as
this study is subject to a number of limitations. Although the
number of participants seems low in comparison with the
previous study, it should be noticed that the cross-over design
resulted in a total of 72 interventions divided evenly between
placebo and active treatment, whereas in the previous study only
67 interventions were carried out, including 26 placebo, 20
normal dosage treatments and 21 higher dosage treatments
(Heyerick et al., 2006). Unfortunately, the populations in the two
different treatment regimens were imbalanced (12 vs 24) as block
randomization was not used during the preparation of this study.
The signiﬁcant effects observed in the second treatment period
are, therefore, mostly based on the results of the smaller group
and may thus be more prone to statistical error. Furthermore, the
active treatment regimen was only given for 8 weeks, whereas
it is customary to use a 12 week study time for vasomotor studies
(Lethaby et al., 2007). In addition, no wash out phase between
the cross-over strategy has been used. However, based on
circumstantial evidence and results of the ﬁrst study it was
anticipated that the standardized hop extract would exert its
activity within a short period after treatment, while pharmacokinetic data also indicate wash-out of the active ingredients
within a relatively short time (Rad et al., 2006).
As it is known that menopausal symptoms are self-limiting
with time, it could be argued that the study population was not
ideally suited to study the effect on menopausal complaints as the
mean time since menopause was more than 4 years on average.
Women indeed typically experience hot ﬂashes for 6 months up
to 2 years, but some women experience hot ﬂashes for 10 years or
longer (Santoro et al., 2004). In a Swedish population, it was found
that the maximal prevalence of hot ﬂashes was about 60% at 52 to
54 years of age, declining to 30% at 60 and 9% at 72 years
(Rodstrom et al., 2002) and more recently, in a 13 year follow-up
study it was found that the mean duration of vasomotor
symptoms was 5.2 7 3.8 years (Col et al., 2009). Given the mean
age of the population under study and the mean time since
menopause, it can be concluded that this population is still
relevant with respect to the study carried out. This is corroborated
by the fact that the mean baseline parameters for menopausal
symptoms (Menopause Rating Scale and Kupperman Index) were
indicative of signiﬁcant menopausal symptoms, in line with the
baseline parameters of many other studies investigating the effect
of a treatment on menopausal symptoms (e.g. raloxifene+
estradiol: (Valiati et al., 2009), soy phytoestrogens: (Nahas et al.,
2007), physical exercise: (Moriyama et al., 2008)).
The scales used to measure menopausal discomforts (KI, MRS
and VAS) are not direct measures of vasomotor symptoms, but
rather summaries of quality-of-life in relation to overall menopausal complaints. More direct results on vasomotor symptoms
could be derived from the use of a hot ﬂash diary with daily
scoring of frequency and intensity of hot ﬂashes and night sweats.
With respect to the VAS scores, it should be noticed that both the
vasomotor symptoms and sleep disturbances contributed equally
to the reduction of the VAS parameter, indicating that the
standardized hop extract could also exert improved sleep quality,
which is in correspondence with its traditional use as mild
sedative most often in combination with valerian (Koetter et al.,
2007). Although further studies are required to conﬁrm the
observations made in this and the previous study (Heyerick et al.,
2006), phytoestrogen preparations containing this standardized
hop extract may provide an interesting alternative to women
seeking relief of mild vasomotor symptoms, as suggested in the
NAMS position statement (Santoro et al., 2004).
Herbal treatments for menopausal complaints have been
studied extensively over the last decade. Although epidemiological studies suggest a relationship between the intake of soyderived phytoestrogens (isoﬂavones) and reduced menopausal
complaints, randomized intervention trials with either increased
soy intake or isolated soy-derived phytoestrogens have reported
inconsistent results (Lethaby et al., 2007). Analogously, also
intervention studies with red clover or black cohosh extracts
failed to produce consistent results (Booth et al., 2006; Borrelli
and Ernst, 2008). In most studies, a large heterogeneity in
individual responses has been observed, with some participants
responding only weakly to the treatment or even in the opposite
direction, while others respond strongly to the treatment. In
addition to the highly variable response to the placebo effect, it is
also quite likely that interindividual differences in exposure to
the active ingredients (eg. soy isoﬂavones) after oral intake may
co-determine the individual response to the treatment. Apart
from interindividual differences in absorption, distribution,
metabolism en excretion, it is now recognized that also the gut
microﬂora may have a signiﬁcant impact on the ﬁnal exposure to
both the type and the concentration of the active ingredients
ARTICLE IN PRESS
R. Erkkola et al. / Phytomedicine 17 (2010) 389–396
(Atkinson et al., 2005; Nielsen and Williamson, 2007; Tamura,
2006). In both the previous (Heyerick et al., 2006) and this present
study, similar large differences in individual responses were
observed. Differences in metabolism may in part account for these
differences. More speciﬁcally, isoxanthohumol can be converted
into 8-PN by the intestinal microbiota (Possemiers et al., 2005).
This transformation is subject to a high interindividual variability,
as in a small population of menopausal women only 35% were
categorized as high or moderate 8-PN producers, whereas 65%
was categorized as weak or non-producers (Bolca et al., 2007;
Possemiers et al., 2006). As isoxanthohumol is present in the
standardized hop extract at a level that is ten times higher than
8-PN, the ﬁnal exposure to 8-PN may be subject to a high
interindividual variability. Therefore, it can be speculated that
speciﬁc phytoestrogen preparations may have beneﬁcial properties only in selected patient groups. As a consequence, it would be
advisable that future intervention studies estimating the effect of
phytoestrogens on menopausal complaints incorporate the
analysis of a marker for the individual exposure to the active
ingredients (eg. steady-state concentration in serum or the daily
excretion in urine) in order to correlate the effective exposure
with the observed effects.
In this study, no adverse effects have been recorded, while the
hop extract has shown a good tolerability as there were no dropouts. Hop-derived extracts have traditionally been used with a
high safety proﬁle. The safety of phytoestrogen preparations in
general, given their estrogenic potency, form the focus of recent
investigations. Intervention studies with phytoestrogen preparations including soy isoﬂavones and black cohosh did not show any
adverse effects on the endometrium (Palacios et al., 2007; Reed
et al., 2008). Although animal experiments have shown possibly
negative effect of 8-PN on the uterus (Overk et al., 2008), the
endometrial safety of the standardized hop extract was investigated in a recently ﬁnished study with good results (results to be
published).
In conclusion, the results obtained in this randomized, double
blind, placebo-controlled, cross-over study showed that administration of a hop extract standardized on 8-PN (100 mg per daily
dose) to post-menopausal women during 8 weeks can reduce
discomforts and complaints associated to the menopause as
measured by the KI, MRS and a multifactorial VAS. Although both
placebo and active treatment similarly reduced the menopausal
complaints after the ﬁrst eight weeks, only the active treatment
could further reduce the menopausal complaints after switching
the treatments, whereas for placebo an increase of menopausal
complaints was observed. Signiﬁcant reductions in menopausal
complaints for the active treatment in comparison with placebo
were observed after 16 weeks, suggesting that the standardized
hop extract is superior to placebo with respect to its potency to
reduce menopausal complaints. Although the observed effects
still need to be conﬁrmed in further studies, phytoestrogen
preparations containing this standardized hop extract may
provide an interesting alternative to women seeking relief of
mild vasomotor symptoms.
Source of funding and conﬂict of interest
This study was ﬁnanced by MaxMedica, Finland and Metagenics Belgium, Ostend, Belgium, with the principal aim of
evaluating their food supplement MenoHops. The study and the
analysis of the results, however, were done independently of
MaxMedica and Metagenics Belgium, apart from occasional
advice on matters of protocol. The supporting source had no
inﬂuence on the decision to submit the results of the study for
publication.
395
Acknowledgements
The participating women are thanked for their kind
co-operation. The logistics support of Metagenics Belgium,
Ostend, Belgium, is kindly acknowledged.
References
Atkinson, C., Frankenfeld, C.L., Lampe, J.W., 2005. Gut bacterial metabolism of the
soy isoﬂavone daidzein: exploring the relevance to human health. Exp. Biol.
Med. 230, 155–170.
Bolca, S., Possemiers, S., Maervoet, V., Huybrechts, I., Heyerick, A., Vervarcke, S.,
Depypere, H., De Keukeleire, D., Bracke, M., De Henauw, S., Verstraete, W.,
Van de Wiele, T., 2007. Microbial and dietary factors associated with the
8-prenylnaringenin producer phenotype: a dietary intervention trial with ﬁfty
healthy post-menopausal Caucasian women. Br. J. Nutr. 98, 950–959.
Booth, N.L., Piersen, C.E., Banuvar, S., Geller, S.E., Shulman, L.P., Farnsworth, N.R.,
2006. Clinical studies of red clover (Trifolium pratense) dietary supplements in
menopause: a literature review. Menopause 13, 251–264.
Borrelli, F., Ernst, E., 2008. Black cohosh (Cimicifuga racemosa) for menopausal
symptoms: a systematic review of its efﬁcacy. Pharmacol. Res. 58, 8–14.
Bowe, J., Li, X.F., Kinsey-Jones, J., Heyerick, A., Brain, S., Milligan, S., O’Byrne, K.,
2006. The hop phytoestrogen, 8-prenylnaringenin, reverses the ovariectomyinduced rise in skin temperature in an animal model of menopausal hot
ﬂushes. J. Endocrinol. 191, 399–405.
Christoffel, J., Rimoldi, G., Wuttke, W., 2006. Effects of 8-prenylnaringenin on
the hypothalamo-pituitary-uterine axis in rats after 3-month treatment.
J. Endocrinol. 188, 397–405.
Clerici, C., Setchell, K.D.R., Battezzati, P.M., Pirro, M., Giuliano, V., Asciutti, S.,
Castellani, D., Nardi, E., Sabatin, G., Orlandi, S., Baldoni, M., Morelli, O.,
Mannarino, E., Morelli, A., 2007. Pasta naturally enriched with isoﬂavone
aglycons from soy germ reduces serum lipids and improves markers of
cardiovascular risk. J. Nutr. 137, 2270–2278.
Col, N.F., Guthrie, J.R., Politi, M., Dennerstein, L., 2009. Duration of vasomotor
symptoms in middle-aged women: a longitudinal study. Menopause 16,
453–457.
Cos, P., De Bruyne, T., Apers, S., Berghe, D.V., Pieters, L., Vlietinck, A.J., 2003.
Phytoestrogens: recent developments. Planta Med. 69, 589–599.
Fitzmaurice, G., Laird, N., Ware, J., 2004. Applied Longitudinal Analysis. Wiley.
Heyerick, A., Vervarcke, S., Depypere, H., Bracke, M., De Keukeleire, D., 2006. A ﬁrst
prospective, randomized, double-blind, placebo-controlled study on the use of
a standardized hop extract to alleviate menopausal discomforts. Maturitas 54,
164–175.
Holzbeierlein, J.M., McIntosh, J., Thrasher, J.B., 2005. The role of soy phytoestrogens
in prostate cancer. Curr. Opin. Urol. 15, 17–22.
¨
Humpel,
M., Isaksson, P., Schaefer, O., Kaufmann, U., Ciana, P., Maggi, A.,
Schleuning, W.D., 2005. Tissue speciﬁcity of 8-prenylnaringenin: protection
from ovariectomy induced bone loss with minimal trophic effects on the
uterus. J. Steroid Biochem. Mol. Biol.
Huntley, A.L., Ernst, E., 2004. Soy for the treatment of perimenopausal symptoms –
a systematic review. Maturitas 47, 1–9.
Koetter, U., Schrader, E., Kaufeler, R., Brattstrom, A., 2007. A randomized, double
blind, placebo – Controlled, prospective clinical study to demonstrate clinical
efﬁcacy of a ﬁxed valerian hops extract combination (Ze 91019) in patients
suffering from non-organic sleep disorder. Phytother. Res. 21, 847–851.
Krebs, E.E., Ensrud, K.E., MacDonald, R., Wilt, T.J., 2004. Phytoestrogens for
treatment of menopausal symptoms: a systematic review. Obstet. Gynecol.
104, 824–836.
Kronenberg, F., Fugh-Berman, A., 2002. Complementary and alternative medicine
for menopausal symptoms: a review of randomized, controlled trials. Ann.
Intern. Med. 137, 805–813.
Kupperman, H.S., Blatt, M.H.G., Wiesbader, H., Filler, W., 1953. Comparative
evaluation of estrogenic preparations by the menopausal and amenorrheal
indices. J. Clin. Endocrinol. Metab. 13, 688–703.
Lethaby, A.E., Brown, J., Marjoribanks, J., Kronenberg, F., Roberts, H., Eden, J., 2007.
Phytoestrogens for vasomotor menopausal symptoms. Cochrane Database
Syst. Rev.
Messina, M., Ho, S., Alekel, D.L., 2004. Skeletal beneﬁts of soy isoﬂavones: a review
of the clinical trial and epiderniologic data. Curr. Opin. Clin. Nutr. Metab. Care
7, 649–658.
Milligan, S.R., Kalita, J.C., Heyerick, A., Rong, H., De Cooman, L., De Keukeleire, D.,
1999. Identiﬁcation of a potent phytoestrogen in hops (Humulus lupulus L.) and
beer. J. Clin. Endocrinol. Metab. 84, 2249–2252.
Moriyama, C.K., Oneda, B., Bernardo, F.R., Cardoso, C.G., Forjaz, C.L.M., Abrahao,
S.B., Mion, D., Fonseca, A.M., Tinucci, T., 2008. A randomized, placebocontrolled trial of the effects of physical exercises and estrogen therapy
on health-related quality of life in postmenopausal women. Menopause 15,
613–618.
Nahas, E.A.P., Nahas-Neto, J., Orsatti, F.L., Carvalho, E.P., Oliveira, M., Dias, R., 2007.
Efﬁcacy and safety of a soy isoﬂavone extract in postmenopausal women: A
randomized, double-blind, and placebo-controlled study. Maturitas 58,
249–258.
ARTICLE IN PRESS
396
R. Erkkola et al. / Phytomedicine 17 (2010) 389–396
Nielsen, I.L.F., Williamson, G., 2007. Review of the factors affecting bioavailability
of soy isoﬂavones in humans. Nutr. Cancer 57, 1–10.
Overk, C.R., Guo, J., Chadwick, L.R., Lantvit, D.D., Minassi, A., Appendino, G., Chen,
S.N., Lankin, D.C., Farnsworth, N.R., Pauli, G.F., van Breemen, R.B., Bolton, J.L.,
2008. In vivo estrogenic comparisons of Trifolium pratense (red clover)
Humulus lupulus (hops), and the pure compounds isoxanthohumol and
8-prenylnaringenin. Chem. Biol. Interact. 176, 30–39.
Palacios, S., Pornel, B., Bergeron, C., Chantre, P., Nogales, F., Aubert, L., Vazquez, F.,
Eden, J., Mares, P., 2007. Endometrial safety assessment of a speciﬁc and
standardized soy extract according to international guidelines. Menopause 14,
1006–1011.
Panay, N., 2007. Integrating phytoestrogens with prescription medicines – A
conservative clinical approach to vasomotor symptom management. Maturitas
57, 90–94.
Possemiers, S., Bolca, S., Eeckhaut, E., Depypere, H., Verstraete, W., 2007.
Metabolism of isoﬂavones, lignans and prenylﬂavonoids by intestinal bacteria:
producer phenotyping and relation with intestinal community. FEMS Microbiol. Ecol. 61, 372–383.
Possemiers, S., Bolca, S., Grootaert, C., Heyerick, A., Decroos, K., Dhooge, W., De
Keukeleire, D., Rabot, S., Verstraete, W., Van de Wiele, T., 2006. The
prenylﬂavonoid isoxanthohumol from hops (Humulus lupulus L.) is activated
into the potent phytoestrogen 8-prenylnaringenin in vitro and in the human
intestine. J. Nutr. 136, 1862–1867.
Possemiers, S., Heyerick, A., Robbens, V., De Keukeleire, D., Verstraete, W., 2005.
Activation of proestrogens from hops (Humulus lupulus L.) by intestinal
microbiota; Conversion of isoxanthohumol into 8-prenylnaringenin. J. Agric.
Food Chem. 53, 6281–6288.
Rad, M., Humpel, M., Schaefer, O., Schoemaker, R.C., Schleuning, W.D., Cohen, A.F.,
Burggraaf, J., 2006. Pharmacokinetics and systemic endocrine effects of the
phyto-oestrogen 8-prenylnaringenin after single oral doses to postmenopausal
women. Br. J. Clin. Pharmacol. 62, 288–296.
Reed, S.D., Newton, K.M., LaCroix, A.Z., Grothaus, L.C., Grieco, V.S., Ehrlich, K., 2008.
Vaginal, endometrial, and reproductive hormone ﬁndings: randomized,
placebo-controlled trial of black cohosh, multibotanical herbs, and dietary
soy for vasomotor symptoms: the Herbal Alternatives for Menopause (HALT)
Study. Menopause 15, 51–58.
Rodstrom, K., Bengtsson, C., Lissner, L., Milsom, I., Sundh, V., Bjorkelund, C., 2002. A
longitudinal study of the treatment of hot ﬂushes: the population study of
women in Gothenburg during a quarter of a century. Menopause 9, 156–161.
Rong, H., Zhao, Y., Lazou, K., De Keukeleire, D., Milligan, S.R., Sandra, P., 2000.
Quantitation of 8-prenylnaringenin, a novel phytoestrogen in hops (Humulus
lupulus L.), hop products, and beers, by benchtop HPLC-MS using electrospray
ionization. Chromatographia 51, 545–552.
Santoro, N.F., Clarkson, T.B., Freedman, R.R., Fugh-Berman, A.J., Loprinzi, C.L.,
Reame, N.K., Board Trustees, N.A.M.S., 2004. Treatment of menopauseassociated vasomotor symptoms: position statement of The North American
Menopause Society. Menopause 11, 11–33.
Schneider, H.P.G., Rosemeier, H.P., Schnitker, J., Gerbsch, S., Turck, R., 2000.
Application and factor analysis of the menopause rating scale [MRS] in a postmarketing surveillance study of Climen (R). Maturitas 37, 113–124.
Tamura, M., 2006. Effects of intestinal ﬂora on the metabolism and absorption of
isoﬂavones. Japan Agri. Res. Quart. 40, 45–50.
Tempfer, C.B., Bentz, E.K., Leodolter, S., Tscherne, G., Reuss, F., Cross, H.S., Huber,
J.C., 2007. Phytoestrogens in clinical practice: a review of the literature. Fertil.
Steril. 87, 1243–1249.
Valiati, B., Capp, E., Edelweiss, M.I., de Freitas, F.M., Wender, M.C.O., 2009. Effect of
raloxifene and low-dose percutaneous 17 beta-estradiol on menopause
symptoms and endometrium – A randomized controlled trial. Maturitas 62,
81–84.
Verbeke, G., Molenberghs, G., 2000. Linear Mixed Models for Longitudinal Data.
Springer-Verlag.
Wu, A.H., Yu, M.C., Tseng, C.C., Pike, M.C., 2008. Epidemiology of soy exposures and
breast cancer risk. Br. J. Cancer 98, 9–14.
Ye, Y.B., Tang, X.Y., Verbruggen, M.A., Su, Y.X., 2006. Soy isoﬂavones attenuate bone
loss in early postmenopausal Chinese women – A single-blind randomized,
placebo-controlled trial. Eur. J. Nutr. 45, 327–334.