Systematic review and meta-analysis of the bone protective effect of phytoestrogens on osteoporosis in ovariectomized rats Song-wen Fu, Gao-feng Zeng, Shao-hui Zong, Zhi-yong Zhang, Bin Zou, Ye Fang, Li Lu, De-qiang Xiao PII: DOI: Reference:
S0271-5317(14)00073-6 doi: 10.1016/j.nutres.2014.05.003 NTR 7340
To appear in:
Nutrition Research
Received date: Revised date: Accepted date:
19 November 2013 8 April 2014 12 May 2014
Please cite this article as: Fu Song-wen, Zeng Gao-feng, Zong Shao-hui, Zhang Zhi-yong, Zou Bin, Fang Ye, Lu Li, Xiao De-qiang, Systematic review and meta-analysis of the bone protective effect of phytoestrogens on osteoporosis in ovariectomized rats, Nutrition Research (2014), doi: 10.1016/j.nutres.2014.05.003
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Systematic review and meta-analysis of the bone protective effect of
RI P
T
phytoestrogens on osteoporosis in ovariectomized rats
Authors: Song-wen Fua, Gao-feng Zengb,*, Shao-hui Zongc, Zhi-yong Zhangb, Bin Zoua, Ye Fanga, Li
SC
Lub, De-qiang Xiaob
Affiliations: a Department of Nutrition and Food Hygiene, Graduate School of Guangxi Medical
b
MA NU
University, Nanning, Guangxi, China.
Department of Nutrition and Food Hygiene, College of Public Hygiene of Guangxi Medical University, Nanning, Guangxi, China. Department of Spine Osteopathia, the First Affiliated Hospital of Guangxi Medical
ED
c
PT
University, Nanning, Guangxi, China.
CE
NOTE: Song-wen Fu and Gao-feng Zeng contributed equally to this work and should be considered
*
AC
as co-first authors.
Corresponding Author. Department of Nutrition and Food Hygiene, College of Public Hygiene of
Guangxi Medical University, No.22 Shuangyong Road, Nanning, Guangxi, P.R. China. 530021. Email address: [email protected] (G.-F. Zeng).
1
ACCEPTED MANUSCRIPT Abbreviations: BMD; bone mineral density
RI P
T
BV/TV; trabecular bone volume fraction Tb.N; trabecular number
SC
Tb.Sp; trabecular separation
S-ALP; serum alkaline phosphatase S-OC; serum osteocalcin WMD; weighted mean differences
ED
SMD; standardized mean difference
MA NU
Tb.Th; trabecular thickness
PT
C-OVX; control ovariectomized group CAMARADES;
AC
SD; Sprague-Dawley
CE
Collaborative Approach to Meta-analysis and Review of Animal Data in Experimental Studies
N (C); number of animals in OVX control group N (PEs); number of animals in phytoestrogens treatment group MD; mean difference
2
ACCEPTED MANUSCRIPT Abstract Phytoestrogens are candidate drugs for the treatment of osteoporosis. Many experiments have been
RI P
T
designed to investigate the preventive effects of phytoestrogens for osteoporosis; however, it is easy for a single dissenting result from animal experiments to mislead clinical investigations. Herein, we
SC
use meta-analysis to assess the evidence for a protective effect of phytoestrogens on ovariectomized rat models of osteopenia. With respect to osteoporosis, PubMed and Web of Science were searched
MA NU
from January 2000 to March 2013 for relevant studies of phytoestrogens in ovariectomized rats. Two reviewers independently selected and assessed the studies. Data were aggregated using a random effects model. Meta-analysis revealed that the phytoestrogens treatment group demonstrated a
ED
significantly higher femur bone mineral density (BMD) and trabecular bone and lower bone turnover
PT
markers (serum alkaline phosphatase (S-ALP) and serum osteocalcin (S-OC)) compared to the control ovariectomized group(C-OVX), thus showing a bone protective effect of phytoestrogens in
CE
ovariectomized rats. Subsequent sensitivity analyses indicated that the effect of phytoestrogens on
AC
S-ALP and S-OC are not robust. Despite the high heterogeneity in the systematic review of animal experiments, the present results indicated that phytoestrogens may offer the most potential for the prevention of bone loss by reducing the expected loss of trabecular bone and BMD. Their effects are likely due to inhibition of bone resorption, but their benefits on bone formation are still unclear. Further studies are needed to assess the effect of phytoestrogens on bone formation and the efficacy and safety of individual phytoestrogens.
Keywords phytoestrogens; ovariecomized rat; osteoporosis; meta-analysis; bone mineral density; bone histomorphometric parameters; bone turnover markers
3
ACCEPTED MANUSCRIPT 1. Introduction Osteoporosis is a metabolic bone disease characterized by low bone mass and microstructure
RI P
T
degeneration of bone tissue, which leads to enhanced fragility. As a result, individuals with osteoporosis are at a high risk of fracture. Women have a higher risk of osteoporosis than men. In
SC
2005-2006, 10% of older women in the United States had osteoporosis at the femur neck, while only 2% of men did [1]. In China, it is estimated that 23.9% and 12.5% of women and 3.2% and 5.3% of
MA NU
men aged ≥50 years have osteoporosis at the femur neck and in their lumbar spine [2].
Estrogen
deficiency due to ovarian dysfunction is the dominant cause of osteoporosis in postmenopausal women [3], and studies have established that the postmenopausal decrease in bone mass can be
ED
attenuated by estrogen [4, 5]. However, data also suggests that estrogen therapy increases the
PT
incidence of breast cancer, coronary heart disease, stroke, and venous thromboembolism [6, 7]. The serious side effects limit its administration; therefore, alternatives that reduce the risk of estrogen in
CE
the treatment of osteoporosis are needed.
AC
Recently, there is much enthusiasm for applying a class of bioactive compounds, i.e. phytoestrogens, to improve human health. Phytoestrogens are polyphenolic, non-steroidal plant compounds with estrogen-like biological activity. The main classes of phytoestrogens are isoflavones (genistein, daidzein, and biochanin A), lignans (enterolactone, enterodiol), coumestanes (coumestrol), flavonoids (quercetin, kaempferol), and stilbenes (resveratrol) [8, 9]. Isoflavones are mainly derived from legumes, meat products, cereals, and soy products [10]. Lignans are widely found in oilseeds, cereals, legumes, fruit, vegetables, and flaxseed [11]. Coumestrol is present in soy nuts, mung bean sprouts, white beans, flaxseed, and is particularly concentrated in alfalfa, clover sprouts, and cabbage [12, 13]. Epidemiological studies indicate that high consumption of soy products is associated with
4
ACCEPTED MANUSCRIPT increased bone mass and may reduce the risk of fracture in postmenopausal women [14 -16]. Phytoestrogens have the ability to exert various estrogenic and antiestrogenic effects, usually by
RI P
T
binding to estrogen receptors [17, 18]. Due to their structural similarity to estrogen, many studies show that phytoestrogens positively affect postmenopausal symptoms, cardiovascular disease, bone
SC
health problems, and breast cancers [19, 20]. Based on this data, phytoestrogens appear to offer great potential for preventing bone loss.
MA NU
Many experiments have been designed to investigate the preventive effects of phytoestrogens for osteoporosis in ovariectomized rats. However, it is easy for a single disparate result to mislead clinical investigation, which could lead to the waste of limited research resources. Any decision to
ED
proceed to a clinical drug trial should be based on a full and unbiased assessment of animal
PT
experimentation, and the limitations of these data should be fully considered [21], such as poor
populations [22].
CE
quality of the experiments or differences of inter-species that might not be applicable to patient
AC
Systematic review and meta-analysis have contributed greatly to the interpretation and aggregation of data in clinical science, particularly when the potential benefit of an intervention is uncertainty. Systematic reviews are secondary studies that attempt to collate all empirical evidence, using a methodical approach to minimize the risk of bias in the selection of studies for inclusion. Meta-analysis, in contrast, is mainly referred to as a statistical method that is used to combine results of individual studies to produce a better estimate of a treatment effect [23]. Systematic reviews of animal studies can ensure that the best possible use is made of existing animal data. Such reviews provide important insights into the validity of animal research, improve the precision of effect estimates from animal experiments, and provide insights regarding which data from animal
5
ACCEPTED MANUSCRIPT experiments can be generalized to humans [24]. Herein, we have investigated the bone protective effect of phytoestrogens in ovariectomized rats using systematic review and meta-analysis in
RI P
T
conjunction.
2.1. Literature search and inclusion criteria
SC
2. Methods for the selection of studies
MA NU
Studies of phytoestrogens in ovariectomized rats of osteoporosis were identified from PubMed (January 2000 to March 2013) and Web of Science (January 2000 to march 2013) using the following keywords: phytoestrogens, isoflavones, soy isoflavones, genistein, daidzein, biochanin A,
ED
flavonoids, quercetin, kaempferol, lignans, stilbenes, resveratrol, coumestrol, coumestanes, icariin,
PT
puerarin, bone, osteoporosis, osteopenic, antiosteoporosis, and bone formation. Studies were chosen for meta-analysis if they met all of the following inclusion criteria: 1)
CE
included ovariectomized rats as subjects; 2) administered phytoestrogens as an intervention; 3)
AC
reported outcomes for the effects of at least one of the bone histomorphometric parameters (expressed as trabecular bone volume fraction (BV⁄TV), trabecular number (Tb.N), trabecular thickness (Tb.Th), and trabecular separation (Tb.Sp)), bone turnover markers (including S-ALP and S-OC), or femur BMD. Duplicate reports or subgroup analysis (different kinds of phytoestrogens) of the primary study were excluded [25]. Two reviewers independently reviewed and assessed the inclusion and exclusion of phytoestrogens in ovariectomized rats with osteoporosis for the systematic review and meta-analysis. Consensus was reached by discussion when disagreements arose.
6
ACCEPTED MANUSCRIPT 2.2. Data extraction and methodological quality appraisal Through a custom data collection form, the following data were extracted: rate type, age at which
RI P
T
rats were ovariectomized, number of rats in group treated with phytoestrogens, number of rats in the C-OVX, intervention, timing of intervention, duration of intervention, and outcomes. From these
SC
data, the mean outcome, standard deviation, and number of rats in the phytoestrogens treatment group and C-OVX were extracted for meta-analysis. For experiments using multiple dosages of
MA NU
phytoestrogens compared to C-OVX, we combined all the phytoestrogens treatment groups into a single control group for each of the experiments [26]. When outcomes were measured at several points, only the final measure was included.
ED
Methodological quality of individual studies was assessed according to the
PT
Collaborative Approach to Meta-analysis and Review of Animal Data in Experimental Studies (CAMARADES) [23] guidelines and the ARRIVE guidelines [27]. These guidelines evaluated
CE
sample size calculation, whether random allocation to treatment or control group was used,
AC
husbandry conditions (e.g. breeding program, light/dark cycle, temperature, type of food, access to food, access to water, environmental enrichment), whether the investigators were blinded to the outcome, compliance with animal welfare regulations, statement of potential conflicts of interests, and whether the study appeared in a peer reviewed publication. The quality scale ranged from 0 to 7 points.
2.3. Statistical analyses Meta-analysis was completed using statistical software provided by the Cochrane Collaboration (RevMan 5.0). Data were aggregated using a random effects model due to the high heterogeneity in
7
ACCEPTED MANUSCRIPT the systematic review of animal experiments [28]. We calculated weighted mean differences (WMD) since the histomorphometric parameters (expressed as BV⁄TV, Tb.N, Tb.Th, and Tb.Sp) and femur
RI P
T
BMD in all strains were made on the same scale. Then, we calculated standardized mean difference (SMD), since the serum biochemical parameters (including S-ALP and S-OC) were measured in
SC
several ways and their mean differed greatly in all strains [29]. Funnel plot analysis was used as a
MA NU
publication bias indicator [30].
3. Interpretation of data 3.1. Search and selection
ED
The electronic search provided 2,849 articles: 1,736 from PubMed and 1,113 from Web of Science.
PT
Of those, 115 studies were scrutinized in full-text form, and of these, 18 studies were included in the
CE
review (Fig. 1).
AC
3.2. Design characteristics and methodological quality of included studies The study design characteristics are shown in Table 1. All of the 18 included studies were written in English. Two of the studies utilized Wistar rats; the remaining studies used Sprague-Dawley (SD) rats. In total, 32 Wistar rats and 231 SD rats were treated with phytoestrogens; 19 Wistar rats and 163 SD rats were used as controls. Concerning study quality, nine studies received a quality score of four or greater. The lowest quality score was two [36] and the highest quality score was five [31, 33, 35] (Table 1). No study described a sample size calculation, and none used a blinding method during either drug administration or outcome assessment. Except for three, all studies mentioned the random allocation
8
ACCEPTED MANUSCRIPT of rats to groups [36, 40, 46]; however, none stated the precise method of randomization. All studies mentioned the husbandry conditions of the rats except for three studies [32, 36, 40], and all studies
RI P
T
explained their compliance with animal welfare regulations except for two studies [37, 41]. Only
SC
four studies contained a statement of potential conflict of interest [31, 33, 35, 40].
3.3. Data analysis
MA NU
3.3.1. Femur BMD
Femur BMD was measured in seven studies [31, 32, 34, 36, 41, 42, 46]. Meta-analysis using a weighted mean difference method with the random effects model revealed that the phytoestrogens
ED
treatment group evidenced a significantly higher femur BMD by 16.80 mg/cm2 (95% CI: 5.83, 27.77,
CE
3.3.2. BV/TV
PT
P=0.003) compared to the C-OVX (Fig. 2).
AC
BV/TV was measured in 10 studies [33, 35, 37, 38, 40, 43, 44, 46, 47, 48]. Meta-analysis using a weighted mean difference method with the random effects model revealed that the phytoestrogens treatment group significantly exhibited higher BV/TV by 7.05% (95% CI: 3.41, 10.70, P=0.0001) compared to the C-OVX (Fig. 3).
3.3.3. Tb.Th Tb.Th was measured in 8 studies [33, 35, 37, 43, 44, 46-48]. Meta-analysis using the random effects model showed a non-significant value of Tb.Th by 3.96 µm in the phytoestrogens treatment group compared to the C-OVX (95% CI, -1.78, 9.70; P=0.18) in Tb.Th (Fig. 4).
9
ACCEPTED MANUSCRIPT 3.3.4. Tb.N
RI P
the phytoestrogens treatment group than in the C-OVX (Fig. 5).
T
Tb.N [33, 35, 37, 43, 46-48] was significantly higher by 0.9 mm-1 (95% CI: 0.45, 1.35, P=0.0001) in
SC
3.3.5. Tb.Sp
Tb.Sp [33, 35, 37, 43, 46-48] was significantly lower by 110.05µm (95% CI, -160.22, -59.89;
MA NU
P
S0271-5317(14)00073-6 doi: 10.1016/j.nutres.2014.05.003 NTR 7340
To appear in:
Nutrition Research
Received date: Revised date: Accepted date:
19 November 2013 8 April 2014 12 May 2014
Please cite this article as: Fu Song-wen, Zeng Gao-feng, Zong Shao-hui, Zhang Zhi-yong, Zou Bin, Fang Ye, Lu Li, Xiao De-qiang, Systematic review and meta-analysis of the bone protective effect of phytoestrogens on osteoporosis in ovariectomized rats, Nutrition Research (2014), doi: 10.1016/j.nutres.2014.05.003
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Systematic review and meta-analysis of the bone protective effect of
RI P
T
phytoestrogens on osteoporosis in ovariectomized rats
Authors: Song-wen Fua, Gao-feng Zengb,*, Shao-hui Zongc, Zhi-yong Zhangb, Bin Zoua, Ye Fanga, Li
SC
Lub, De-qiang Xiaob
Affiliations: a Department of Nutrition and Food Hygiene, Graduate School of Guangxi Medical
b
MA NU
University, Nanning, Guangxi, China.
Department of Nutrition and Food Hygiene, College of Public Hygiene of Guangxi Medical University, Nanning, Guangxi, China. Department of Spine Osteopathia, the First Affiliated Hospital of Guangxi Medical
ED
c
PT
University, Nanning, Guangxi, China.
CE
NOTE: Song-wen Fu and Gao-feng Zeng contributed equally to this work and should be considered
*
AC
as co-first authors.
Corresponding Author. Department of Nutrition and Food Hygiene, College of Public Hygiene of
Guangxi Medical University, No.22 Shuangyong Road, Nanning, Guangxi, P.R. China. 530021. Email address: [email protected] (G.-F. Zeng).
1
ACCEPTED MANUSCRIPT Abbreviations: BMD; bone mineral density
RI P
T
BV/TV; trabecular bone volume fraction Tb.N; trabecular number
SC
Tb.Sp; trabecular separation
S-ALP; serum alkaline phosphatase S-OC; serum osteocalcin WMD; weighted mean differences
ED
SMD; standardized mean difference
MA NU
Tb.Th; trabecular thickness
PT
C-OVX; control ovariectomized group CAMARADES;
AC
SD; Sprague-Dawley
CE
Collaborative Approach to Meta-analysis and Review of Animal Data in Experimental Studies
N (C); number of animals in OVX control group N (PEs); number of animals in phytoestrogens treatment group MD; mean difference
2
ACCEPTED MANUSCRIPT Abstract Phytoestrogens are candidate drugs for the treatment of osteoporosis. Many experiments have been
RI P
T
designed to investigate the preventive effects of phytoestrogens for osteoporosis; however, it is easy for a single dissenting result from animal experiments to mislead clinical investigations. Herein, we
SC
use meta-analysis to assess the evidence for a protective effect of phytoestrogens on ovariectomized rat models of osteopenia. With respect to osteoporosis, PubMed and Web of Science were searched
MA NU
from January 2000 to March 2013 for relevant studies of phytoestrogens in ovariectomized rats. Two reviewers independently selected and assessed the studies. Data were aggregated using a random effects model. Meta-analysis revealed that the phytoestrogens treatment group demonstrated a
ED
significantly higher femur bone mineral density (BMD) and trabecular bone and lower bone turnover
PT
markers (serum alkaline phosphatase (S-ALP) and serum osteocalcin (S-OC)) compared to the control ovariectomized group(C-OVX), thus showing a bone protective effect of phytoestrogens in
CE
ovariectomized rats. Subsequent sensitivity analyses indicated that the effect of phytoestrogens on
AC
S-ALP and S-OC are not robust. Despite the high heterogeneity in the systematic review of animal experiments, the present results indicated that phytoestrogens may offer the most potential for the prevention of bone loss by reducing the expected loss of trabecular bone and BMD. Their effects are likely due to inhibition of bone resorption, but their benefits on bone formation are still unclear. Further studies are needed to assess the effect of phytoestrogens on bone formation and the efficacy and safety of individual phytoestrogens.
Keywords phytoestrogens; ovariecomized rat; osteoporosis; meta-analysis; bone mineral density; bone histomorphometric parameters; bone turnover markers
3
ACCEPTED MANUSCRIPT 1. Introduction Osteoporosis is a metabolic bone disease characterized by low bone mass and microstructure
RI P
T
degeneration of bone tissue, which leads to enhanced fragility. As a result, individuals with osteoporosis are at a high risk of fracture. Women have a higher risk of osteoporosis than men. In
SC
2005-2006, 10% of older women in the United States had osteoporosis at the femur neck, while only 2% of men did [1]. In China, it is estimated that 23.9% and 12.5% of women and 3.2% and 5.3% of
MA NU
men aged ≥50 years have osteoporosis at the femur neck and in their lumbar spine [2].
Estrogen
deficiency due to ovarian dysfunction is the dominant cause of osteoporosis in postmenopausal women [3], and studies have established that the postmenopausal decrease in bone mass can be
ED
attenuated by estrogen [4, 5]. However, data also suggests that estrogen therapy increases the
PT
incidence of breast cancer, coronary heart disease, stroke, and venous thromboembolism [6, 7]. The serious side effects limit its administration; therefore, alternatives that reduce the risk of estrogen in
CE
the treatment of osteoporosis are needed.
AC
Recently, there is much enthusiasm for applying a class of bioactive compounds, i.e. phytoestrogens, to improve human health. Phytoestrogens are polyphenolic, non-steroidal plant compounds with estrogen-like biological activity. The main classes of phytoestrogens are isoflavones (genistein, daidzein, and biochanin A), lignans (enterolactone, enterodiol), coumestanes (coumestrol), flavonoids (quercetin, kaempferol), and stilbenes (resveratrol) [8, 9]. Isoflavones are mainly derived from legumes, meat products, cereals, and soy products [10]. Lignans are widely found in oilseeds, cereals, legumes, fruit, vegetables, and flaxseed [11]. Coumestrol is present in soy nuts, mung bean sprouts, white beans, flaxseed, and is particularly concentrated in alfalfa, clover sprouts, and cabbage [12, 13]. Epidemiological studies indicate that high consumption of soy products is associated with
4
ACCEPTED MANUSCRIPT increased bone mass and may reduce the risk of fracture in postmenopausal women [14 -16]. Phytoestrogens have the ability to exert various estrogenic and antiestrogenic effects, usually by
RI P
T
binding to estrogen receptors [17, 18]. Due to their structural similarity to estrogen, many studies show that phytoestrogens positively affect postmenopausal symptoms, cardiovascular disease, bone
SC
health problems, and breast cancers [19, 20]. Based on this data, phytoestrogens appear to offer great potential for preventing bone loss.
MA NU
Many experiments have been designed to investigate the preventive effects of phytoestrogens for osteoporosis in ovariectomized rats. However, it is easy for a single disparate result to mislead clinical investigation, which could lead to the waste of limited research resources. Any decision to
ED
proceed to a clinical drug trial should be based on a full and unbiased assessment of animal
PT
experimentation, and the limitations of these data should be fully considered [21], such as poor
populations [22].
CE
quality of the experiments or differences of inter-species that might not be applicable to patient
AC
Systematic review and meta-analysis have contributed greatly to the interpretation and aggregation of data in clinical science, particularly when the potential benefit of an intervention is uncertainty. Systematic reviews are secondary studies that attempt to collate all empirical evidence, using a methodical approach to minimize the risk of bias in the selection of studies for inclusion. Meta-analysis, in contrast, is mainly referred to as a statistical method that is used to combine results of individual studies to produce a better estimate of a treatment effect [23]. Systematic reviews of animal studies can ensure that the best possible use is made of existing animal data. Such reviews provide important insights into the validity of animal research, improve the precision of effect estimates from animal experiments, and provide insights regarding which data from animal
5
ACCEPTED MANUSCRIPT experiments can be generalized to humans [24]. Herein, we have investigated the bone protective effect of phytoestrogens in ovariectomized rats using systematic review and meta-analysis in
RI P
T
conjunction.
2.1. Literature search and inclusion criteria
SC
2. Methods for the selection of studies
MA NU
Studies of phytoestrogens in ovariectomized rats of osteoporosis were identified from PubMed (January 2000 to March 2013) and Web of Science (January 2000 to march 2013) using the following keywords: phytoestrogens, isoflavones, soy isoflavones, genistein, daidzein, biochanin A,
ED
flavonoids, quercetin, kaempferol, lignans, stilbenes, resveratrol, coumestrol, coumestanes, icariin,
PT
puerarin, bone, osteoporosis, osteopenic, antiosteoporosis, and bone formation. Studies were chosen for meta-analysis if they met all of the following inclusion criteria: 1)
CE
included ovariectomized rats as subjects; 2) administered phytoestrogens as an intervention; 3)
AC
reported outcomes for the effects of at least one of the bone histomorphometric parameters (expressed as trabecular bone volume fraction (BV⁄TV), trabecular number (Tb.N), trabecular thickness (Tb.Th), and trabecular separation (Tb.Sp)), bone turnover markers (including S-ALP and S-OC), or femur BMD. Duplicate reports or subgroup analysis (different kinds of phytoestrogens) of the primary study were excluded [25]. Two reviewers independently reviewed and assessed the inclusion and exclusion of phytoestrogens in ovariectomized rats with osteoporosis for the systematic review and meta-analysis. Consensus was reached by discussion when disagreements arose.
6
ACCEPTED MANUSCRIPT 2.2. Data extraction and methodological quality appraisal Through a custom data collection form, the following data were extracted: rate type, age at which
RI P
T
rats were ovariectomized, number of rats in group treated with phytoestrogens, number of rats in the C-OVX, intervention, timing of intervention, duration of intervention, and outcomes. From these
SC
data, the mean outcome, standard deviation, and number of rats in the phytoestrogens treatment group and C-OVX were extracted for meta-analysis. For experiments using multiple dosages of
MA NU
phytoestrogens compared to C-OVX, we combined all the phytoestrogens treatment groups into a single control group for each of the experiments [26]. When outcomes were measured at several points, only the final measure was included.
ED
Methodological quality of individual studies was assessed according to the
PT
Collaborative Approach to Meta-analysis and Review of Animal Data in Experimental Studies (CAMARADES) [23] guidelines and the ARRIVE guidelines [27]. These guidelines evaluated
CE
sample size calculation, whether random allocation to treatment or control group was used,
AC
husbandry conditions (e.g. breeding program, light/dark cycle, temperature, type of food, access to food, access to water, environmental enrichment), whether the investigators were blinded to the outcome, compliance with animal welfare regulations, statement of potential conflicts of interests, and whether the study appeared in a peer reviewed publication. The quality scale ranged from 0 to 7 points.
2.3. Statistical analyses Meta-analysis was completed using statistical software provided by the Cochrane Collaboration (RevMan 5.0). Data were aggregated using a random effects model due to the high heterogeneity in
7
ACCEPTED MANUSCRIPT the systematic review of animal experiments [28]. We calculated weighted mean differences (WMD) since the histomorphometric parameters (expressed as BV⁄TV, Tb.N, Tb.Th, and Tb.Sp) and femur
RI P
T
BMD in all strains were made on the same scale. Then, we calculated standardized mean difference (SMD), since the serum biochemical parameters (including S-ALP and S-OC) were measured in
SC
several ways and their mean differed greatly in all strains [29]. Funnel plot analysis was used as a
MA NU
publication bias indicator [30].
3. Interpretation of data 3.1. Search and selection
ED
The electronic search provided 2,849 articles: 1,736 from PubMed and 1,113 from Web of Science.
PT
Of those, 115 studies were scrutinized in full-text form, and of these, 18 studies were included in the
CE
review (Fig. 1).
AC
3.2. Design characteristics and methodological quality of included studies The study design characteristics are shown in Table 1. All of the 18 included studies were written in English. Two of the studies utilized Wistar rats; the remaining studies used Sprague-Dawley (SD) rats. In total, 32 Wistar rats and 231 SD rats were treated with phytoestrogens; 19 Wistar rats and 163 SD rats were used as controls. Concerning study quality, nine studies received a quality score of four or greater. The lowest quality score was two [36] and the highest quality score was five [31, 33, 35] (Table 1). No study described a sample size calculation, and none used a blinding method during either drug administration or outcome assessment. Except for three, all studies mentioned the random allocation
8
ACCEPTED MANUSCRIPT of rats to groups [36, 40, 46]; however, none stated the precise method of randomization. All studies mentioned the husbandry conditions of the rats except for three studies [32, 36, 40], and all studies
RI P
T
explained their compliance with animal welfare regulations except for two studies [37, 41]. Only
SC
four studies contained a statement of potential conflict of interest [31, 33, 35, 40].
3.3. Data analysis
MA NU
3.3.1. Femur BMD
Femur BMD was measured in seven studies [31, 32, 34, 36, 41, 42, 46]. Meta-analysis using a weighted mean difference method with the random effects model revealed that the phytoestrogens
ED
treatment group evidenced a significantly higher femur BMD by 16.80 mg/cm2 (95% CI: 5.83, 27.77,
CE
3.3.2. BV/TV
PT
P=0.003) compared to the C-OVX (Fig. 2).
AC
BV/TV was measured in 10 studies [33, 35, 37, 38, 40, 43, 44, 46, 47, 48]. Meta-analysis using a weighted mean difference method with the random effects model revealed that the phytoestrogens treatment group significantly exhibited higher BV/TV by 7.05% (95% CI: 3.41, 10.70, P=0.0001) compared to the C-OVX (Fig. 3).
3.3.3. Tb.Th Tb.Th was measured in 8 studies [33, 35, 37, 43, 44, 46-48]. Meta-analysis using the random effects model showed a non-significant value of Tb.Th by 3.96 µm in the phytoestrogens treatment group compared to the C-OVX (95% CI, -1.78, 9.70; P=0.18) in Tb.Th (Fig. 4).
9
ACCEPTED MANUSCRIPT 3.3.4. Tb.N
RI P
the phytoestrogens treatment group than in the C-OVX (Fig. 5).
T
Tb.N [33, 35, 37, 43, 46-48] was significantly higher by 0.9 mm-1 (95% CI: 0.45, 1.35, P=0.0001) in
SC
3.3.5. Tb.Sp
Tb.Sp [33, 35, 37, 43, 46-48] was significantly lower by 110.05µm (95% CI, -160.22, -59.89;
MA NU
P
