Study population
A randomized controlled trial was conducted for 3 months, at the Sendai Medical Center in Sendai City, Miyagi Prefecture, located in North Eastern Japan. The study implementation period in the protocol was during 2017; however, due to recruitment issues, it was delayed until the beginning of 2018. Recruitment process started on September 25, 2017 using pamphlets, targeting post-menopausal women who had planned to undergo medical check-up in January 2018. Until the scheduled date of closure on November 30, 2017, eighty five women voluntarily applied to participate in the study. Among them, 63 were selected based on the inclusion and exclusion criteria as follows.
Inclusion criteria: (1) Postmenopausal women (natural absence of menstruation over at least 12 months since the last menstrual period or a bilateral oophorectomy procedure performed in a woman (surgical menopause); (2) Those who could visit the clinic every month for interviews or investigations during the study period; (3) Those who could remain compliant with the daily supplementation regimen and record daily notes throughout the study period comprising 12 weeks.
Exclusion criteria: (1) Those with history of allergy to soy foods, dairy products, or Brewer's yeast; (2) Those who reported intake of medications or functional foods that could affect our study results; (3) Those administered hormone therapy or medications that could affect blood hormone levels; (4) Those considered ineligible by the investigators.
The details of the study were explained to these women, and the first 60 women to give consent to participate in the study were selected at the end of December 2017. They were listed in chronological order and assigned into two groups by simple randomization. Odd number group A received a 10 mg equol supplement and lactobionic acid daily, while even number group B received no supplement. The participants received pre-intervention investigations during January 16 to 18, 2018. The equol-supplementation group started taking the supplement on the day after completion of these investigations. All the participants had a regular follow-up every month, for 3 months until April 20, 2018. In the equol-supplementation group, one woman dropped out and two women reported that they had occasionally used the equol supplement before the intervention. Consequently, 57 women (48 to 69 years of age) were eventually included for analysis in the study (Figure 1). The nurses at the Sendai Medical Center generated the random allocation sequence, enrolled participants, and assigned participants to interventions. The registered dietitian was responsible for explanation, distribution and monitoring supplements to the intervention group.
Study treatments
Twenty seven women belonging to the equol-supplementation group received 10 mg of oral equol-containing supplement per day, composed of 98% S-equol, 2% daidzein, 0.2% glycitein, and 0.1% genistein extracted from fermented soy beans (product name: FlavoCel EQ-5, Daicel Corporation, Tokyo, Japan).
Determination of equol-producer status
Prior to the intervention, early morning urine samples were collected from all participants. Urinary equol was measured using an immunochromatographic strip (Soy Check, Healthcare Systems Co., Ltd), as described in a previous study [15]. Individuals were considered as equol-producers if their urinary equol level was higher than 1.0 μM, as described in previous studies [3, 4].
Primary outcome measures
Body height and weight were measured using a height weight scale (A & D Company Limited, Tokyo, Japan). Overnight fasting blood samples were obtained to determine the levels of triglycerides (TG), total cholesterol (TC), low density lipoprotein (LDL) cholesterol, high density lipoprotein (HDL) cholesterol, hemoglobin A1c (HbA1c), uric acid (UA), intact parathyroid hormone (PTH), and 25-hydroxy vitamin D. To assess the degree of arterial stiffness (arteriosclerosis), brachial-ankle pulse wave velocity (baPWV) was measured using vascular ultrasound (Fukuda Denshi, Tokyo, Japan). Visceral fat area was measured using computed tomography (CT). The level of AGEs was determined by measuring the skin autofluorescence on the volar side of the forearm using an AGE reader ((DiagnOptics, Groningen, Netherlands) as described previously [26]. Autofluorescence was defined as the average fluorescence per nm over the entire emission spectrum (420-600 nm) as ratio of the average fluorescence per nm over the 300-420-nm range [26]. Age-adjusted SAF levels (z scores) were calculated for each woman, based on the total population. The above measurement of parameters was performed at the baseline and 3-months post intervention.
Secondary outcome measures
Self-administered questionnaires were used for the assessment of climacteric symptoms, using the Climacteric Scale developed by the Japan Society of Obstetrics and Gynecology (table 1). The questionnaire contained 21 items scored on a 4-point scale (Never=0, mild=1, moderate=2, severe=3). The total score represented the overall severity of the symptoms. These questionnaires were administered during the monthly follow-ups, for 3 months. Starting from the first month of the intervention, the overall improvement in symptoms was also assessed using the following four responses: ‘a lot’, ‘somewhat’, ‘no change’, and ‘worse’. This technique was validated and widely used in the Japanese population.
Table 1. Climacteric symptom rating scale by the Japan Society of Obstetrics and Gynecology
|
Item
|
Score
|
Q1
|
1.Facial skin blushing and upper body (hot flashes)
|
Never=0, mild=1, moderate=2, severe=3
|
Q2
|
2.Easy to sweat (sweating)
|
Never=0, mild=1, moderate=2, severe=3
|
Q3
|
3.Difficulty getting to sleep (insomnia)
|
Never=0, mild=1, moderate=2, severe=3
|
Q4
|
4.Difficulty staying asleep (light sleep)
|
Never=0, mild=1, moderate=2, severe=3
|
Q5
|
5.Irritability
|
Never=0, mild=1, moderate=2, severe=3
|
Q6
|
6.Anxiety
|
Never=0, mild=1, moderate=2, severe=3
|
Q7
|
7.Often irritated by trifles (anxious trifles)
|
Never=0, mild=1, moderate=2, severe=3
|
Q8
|
8.Feeling unhappy or depressed (depressive mood),
|
Never=0, mild=1, moderate=2, severe=3
|
Q9
|
9.Fatigue
|
Never=0, mild=1, moderate=2, severe=3
|
Q10
|
10. Eye strain
|
Never=0, mild=1, moderate=2, severe=3
|
Q11
|
11.Memory problems (forgetfulness)
|
Never=0, mild=1, moderate=2, severe=3
|
Q12
|
12.Dizziness
|
Never=0, mild=1, moderate=2, severe=3
|
Q13
|
13.Palpitations
|
Never=0, mild=1, moderate=2, severe=3
|
Q14
|
14.Chest tightness
|
Never=0, mild=1, moderate=2, severe=3
|
Q15
|
15.Headache
|
Never=0, mild=1, moderate=2, severe=3
|
Q16
|
16.Neck stiffness
|
Never=0, mild=1, moderate=2, severe=3
|
Q17
|
17.Backache and low back pain
|
Never=0, mild=1, moderate=2, severe=3
|
Q18
|
18.Joint pain
|
Never=0, mild=1, moderate=2, severe=3
|
Q19
|
19.Cold hands and feet
|
Never=0, mild=1, moderate=2, severe=3
|
Q20
|
20.Numbness in the legs or arms
|
Never=0, mild=1, moderate=2, severe=3
|
Q21
|
21.Sensitive to sounds
|
Never=0, mild=1, moderate=2, severe=3
|
Q22
|
Overall improvement in symptoms (not included in the baseline questionnaire)
|
Worse=-1, No change=0, Somewhat=1, A lot=2
|
Treatment adherence and monitoring of adverse effects
During the monthly follow-ups, the staff interviewed the adherence to equol supplement such as frequency and dose, as well as clinical signs and symptoms of adverse effects in order to ensure compliance as well as make timely decision on cessation of the supplement.
Statistical analyses
All statistical analyses were performed using the IBM SPSS 19 statistical software (IBM Japan, Minato-ku, Tokyo, Japan). Mann-Whitney test was used to compare the differences in continuous data, and chi-square test was used to compare categorical data between equol-supplementation group and control group. For the assessment of quantitative changes in the metabolic and aging biomarkers before and after intervention, we use the Wilcoxon sign rank test. For comparing proportions of people with improved metabolic and aging biomarkers between the control and equol-supplementation groups, we use the chi-square test and the Fisher exact test. Additionally, we assessed the change in results with respect to equol exposure by categorizing the groups into four categories: 1) equol producers consuming equol supplement, 2) equol non-producers consuming equol supplement, 3) quol producers without equol supplement, and 4) equol non-producers without supplement. The extended Mantel-Haenszel chi square for linear trend was used to examine the association between post-intervention improvement (dependent variable) and equol exposure (independent variable).
Changes in the severity (total score) of climacteric symptoms over time, i.e., at baseline, 1 month, 2 months and 3 months post equol intervention., were analyzed using a two-way repeated measure analysis of variance (ANOVA). Analysis of the studentized residuals showed that there was normality, as assessed by the Shapiro-Wilk test of normality and no outliers, as assessed by no studentized residuals greater than ± 3 standard deviations. There was sphericity for the interaction term, as assessed by Mauchly’s test of sphericity (p > .05). Also, post hoc pairwise comparisons between equol intervention and control groups as well as between equol producers and non-producers were conducted using the Bonferroni correction. Improvements in climacteric symptoms, in the control and equol-supplementation groups, were compared using the chi-square test and the Fisher exact test for proportions. All tests were two-sided, and the statistical significance was set to p<0.05.