Summary of findings
Description of the condition
Uterine fibroids are the most common, non-cancerous uterine growths in women of childbearing age. Alternative names are uterine leiomyomata, fibromyoma, myoma, or fibroids. The lifetime risk of fibroids in a woman over the age of 45 years has been estimated to be more than 60%, including symptomatic and non-symptomatic conditions (Okolo 2008). Around 30% of women of childbearing age have clinically symptomatic uterine fibroids (Newbold 2000; Stewart 2001). Common symptoms may include heavy or painful periods; prolonged menstrual periods; bleeding between periods; pelvic pain or low back pain; 'fullness' in the lower abdomen, with or without urinary or rectal symptoms due to compression; and reproductive problems, such as infertility, multiple miscarriages, or early onset of labour during pregnancy. Many women with uterine fibroids do not have any symptoms. A recent investigation of 21,479 women across eight countries showed that the prevalence of fibroids was from 9.4% (UK) to 17.8% (Italy) in the age population of 40 to 49 years (Zimmermann 2012). Uterine fibroids constitute the main reason for hysterectomies to be carried out, based on data between 1990 and 1997 in the United States (Farquhar 2002).
Uterine fibroids are growths of muscular and fibrous cells within, or attached to, the wall of the uterus. According to the location of the growth, they can be categorised as submucosal when they grow just underneath the uterine lining, intramural when they are in between the muscles of the uterus, and subserosal when they are on the outside of the uterus. Fibroids may grow as a single tumour or in clusters. A single fibroid can be less than one inch in size or can grow to eight inches or more. A group of fibroids can also vary in size. The cause of uterine fibroids remains unknown, however genetic, hormonal, immunological, and environmental factors may play a role in starting the growth of fibroids, or in continuing that growth (Munro 2011). Several risk factors for uterine fibroids have been identified. African-American women are at three- to five-times greater risk than white women. Women who are overweight or obese for their height (based on body mass index (BMI)) are also at slightly higher risk than women who are average in weight for their height. Women who have given birth appear to be at lower risk (Marshall 1997).
Recommended treatment for uterine fibroids depends on the severity of symptoms, the woman's age, pregnancy status, desire for future pregnancies, general health, and the characteristics of the fibroids (Stewart 2001). If a woman shows no symptoms, or the fibroids are small, she may not need any treatment. If a woman has serious symptoms or pain, medical therapy can be used to relieve symptoms. Such treatment may include gonadotropin-releasing hormone agonists (GnRHa) (Lethaby 2001); synthetic steroids with antiprogesterone activity, such as mifepristone, to slow or stop the growth of fibroids (Tristan 2012); and the use of progesterone and its derivatives for short-term treatment of bleeding and for inhibiting the fibroids' growth (Grigorieva 2003; Maruo 2004).
Surgical therapy is considered to be an effective treatment and includes myomectomy to remove only the fibroids and leave the healthy uterus, or hysterectomy to remove the entire uterus (Falcone 2002; Griffiths 2006). Another accepted treatment is uterine artery embolization (UAE), which is used to block off the blood supply to the uterus and so make the fibroids shrink (Gupta 2012; McLucas 2001; Tranquart 2002; Watson 2002). However, few women with uterine fibroids prefer surgery and women may seek less invasive options, such as pain medication, medical therapy, or other alternative therapies.
Description of the intervention
Among alternative therapies, herbal treatments for fibroids are used in several medical traditions and countries (Fugh-Berman 2004). For example, in China the use of traditional Chinese herbal medicines for treating uterine fibroids is a common clinical practice. In this review, herbal preparations are defined as any formulation of medicinal herbs including extracts, raw herbs, or herbal decoctions prescribed by practitioners. These could include herbal products such as Chinese proprietary medicine or self-prepared herbal decoctions. In Chinese medicine, herbal medicine has been used for many years for different diseases or conditions. For example, the herbal medicine Guizhi Fuling formula has been described in historical classics in ancient China for treatment of women's symptoms, and it is still used in China (Li J 2008). However, there are huge variations in the herbal preparations used, which will depend on the practitioners themselves and on the individualised treatment of different women.
How the intervention might work
According to the theory of Chinese medicine, practitioners recognise uterine fibroids as a condition of imbalance between yin and yang in the body (in allopathic terms, disturbances of the endocrine system and blood circulation). Therefore, it is important that the practitioners make a diagnosis based on the symptoms and signs from observing the tongue and taking the pulse, and this practice is called 'pattern differentiation' (Chinese medicine diagnosis). The practitioners prescribe a herbal formula according to the pattern of the syndrome (in Chinese, Zheng). Clinical studies from the Chinese literature show that Chinese herbal preparations might relieve symptoms and shrink the fibroid tumours without significant adverse effects (Huang 2003; Xiong 2002). One of the commonly used herbal medicines is Guizhi Fuling formula, and basic studies showed that Guizhi Fuling formula might work on fibroids by promoting qi flow and blood circulation, immune regulation, and softening and resolving hard lumps (Ji 2011; Li J 2008; Sang 2004). However, the exact mechanisms of the therapeutic effect are not fully understood.
Why it is important to do this review
Is the practice of using herbs for fibroids supported by well-designed clinical evidence? We aim to review the clinical research studies systemically and inform practice by presenting comprehensive, critically appraised evidence.
The primary objective was to evaluate the effectiveness and safety of Chinese herbal medicine for treatment of uterine fibroids.
Criteria for considering studies for this review
Types of studies
Published and unpublished randomised controlled trials were eligible for inclusion, regardless of blinding, publication status, or language. We planned to include cross-over randomised trials but to use only the data from the first phase. We excluded quasi-randomised trials or 'randomised' trials with false methods for random allocation of participants, or where a trial was not stated to be randomised.
Types of participants
Women with uterine fibroids diagnosed by clinical symptoms and physical signs,and confirmed by ultrasound scanning, computed tomography (CT), magnetic resonance imaging (MRI), or a combination of more than one of these procedures. We planned to include women with fibroid related symptoms and palpable uterine fibroids, without confirmation by imaging technology, and to compare these in subgroup analyses. We also planned to include women without any symptoms who were found to have uterine fibroids during routine gynaecological examination, which were confirmed by imaging techniques.
Types of interventions
Experimental interventions included Chinese patented herbal medicines, other patented herbal products pertaining to different traditional medicines, extracts of a single herb or a compound of herbs, or other individualised herbal remedies. We did not limit the administration or formulation of herbal preparations, such as capsule, tablet, granule, decoction, or injection. The control interventions included no treatment, placebo, medical therapy, or surgical procedures.
Types of outcome measures
- Uterine fibroid related symptoms such as heavy, irregular, or prolonged menstrual periods; bleeding between periods; pelvic or low back pain; and low abdominal pressure symptoms such as frequent or urgent urination, or constipation. Symptoms could be measured by either patient reporting or an instrument, regardless of blinding.
- Adverse effects of herbal preparations.
3. Number of women undertaking surgery (myomectomy, hysterectomy, embolization) due to failure of medical prevention or management of the above symptoms.
4. Incidence of complications including anaemia, infertility, miscarriage, premature labour and delivery, abnormal fetal position.
5. Quality of life (measured by a validated scale or instrument).
6. Number and size of the fibroids, the volume of the uterus, or both.
Search methods for identification of studies
We searched the following electronic databases for published and unpublished randomised trials of herbal medicine, without language restriction and in consultation with the Mentrual Disorders and Subfertility Group (MDSG) Trials Search Coordinator:
- Trials Registers of the Cochrane MDSG and the Cochrane Complementary Medicine Field;
- Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2012, Issue 9);
- MEDLINE (1966 to September 2012), EMBASE (1998 to September 2012), AMED, and LILACS (www.bireme.br/bvs/I/ibd.htm) from their date of inception onwards.
The following five major Chinese biomedical databases were searched:
1. Chinese Biomedical Literature database (CBM) (http://www.imicams.ac.cn/);
2. Chinese Medical Current Content (CMCC) (http://www.cmcc.org.cn);
3. China National Knowledge Infrastructure (CNKI-CAJ) (www.cnki.net);
4. VIP information/Chinese Scientific Journals Database (CSJD-VIP) (http://dx3.cqvip.com/);
5. WanFang database/Chinese Medicine Premier (hppt://www.wanfangdata.com.cn/).
We used the search terms: uterine fibroids, hysteromyoma, uterine leiomyomata, fibromyoma, myoma; and combined with traditional medicine, alternative medicine, plant extracts, medicinal plants, non-prescription drugs, herbs, complementary medicine, Chinese medicine, phytodrug or phytopharmaceutical. We had no restriction on publication type. The detailed search strategies are listed in Appendix 1 and Appendix 2.
Searching other resources
- We checked the reference lists of identified randomised controlled trials and review articles in order to find further trials not identified by the electronic searches.
- We searched for ongoing trials through the National Research Register and the website www.controlled-trials.com.
- We also checked the 'grey' literature, including unpublished conference proceedings or abstract books, and contacted pharmaceutical companies which produce herbal medicines for uterine fibroids to identify unpublished trials.
Data collection and analysis
Selection of studies
After an initial screen of titles and abstracts retrieved by the search, conducted by H Yang, the full texts of all potentially eligible studies were retrieved. Two review authors (JP Liu and H Yang) independently selected the trials to be included in the review according to the prespecified selection criteria. Any disagreements were resolved by discussion. Y Xia confirmed the randomisation through phone calls to Chinese trialists.
Data extraction and management
Two review authors (JP Liu and H Yang) independently extracted data from eligible studies using a data extraction form designed and pilot-tested by the authors. Any disagreements were resolved by discussion or by a third review author (Y Xia). Papers not in Chinese, English, Japanese, or Italian were translated with the help of the Cochrane Menstrual Disorders and Subfertility Group. We extracted the following characteristics and data from each included trial: primary author, study setting, methodology, age, gender, and ethnicity of participants, number of participants randomised and analysed, participant inclusion and exclusion criteria, symptoms and methods for measurement, the diagnostic criteria, type of herb or herbs, quality of the products, route of delivery, dosage and duration of intervention, details of the comparison regime, duration of follow up, reasons for and number that dropped out or were lost during follow up, outcome measures (end of treatment and at follow up), and number and type of adverse events.
We sought data on the number of participants with each outcome by allocated treatment group, irrespective of compliance or follow up, to allow an intention-to-treat analysis. For three-arm trials, the data from the control group would be split in half so that half of the participants and half of the events would be used in each comparison.
Assessment of risk of bias in included studies
Two review authors (JP Liu and H Yang) independently assessed the included studies for risk of bias using the Cochrane risk of bias assessment tool (www.cochrane-handbook.org) to assess: selection bias (random sequence generation and allocation concealment); performance bias (blinding of participants and personnel); detection bias (blinding of outcome assessors); attrition bias (incomplete outcome data); reporting bias (selective reporting); and other bias. Disagreements were resolved by discussion or by a third review author (F Cardini). We described all judgements fully and presented the conclusions in the 'Risk of bias' table.
Generation of the allocation sequence
Low risk of bias: if the allocation sequence was generated by a computer or random number table. Drawing of lots, tossing of a coin, shuffling of cards, or throwing dice may also be considered as low risk if a person who was not otherwise involved in the recruitment of participants performed the procedure.
Low risk of bias: if the allocation of participants involved a central independent unit, on-site locked computer, identical appearing numbered drug bottles or containers prepared by an independent pharmacist or investigator, or sealed envelopes. Envelopes should be serially numbered, sealed, and opaque. However, this information is rarely provided, indicating an increased risk of bias.
Unclear risk of bias: if the trial was described as randomised but the method used to conceal the allocation was not described, or the sealed envelopes were not described as opaque.
High risk of bias: if the allocation sequence was known to the investigators who assigned participants, or if the study was quasi-randomised.
Blinding (or masks)
Low risk of bias: double blinding, if the trial was described as double blind and both the participant and physician were blinded, or participant and outcome assessor.
Unclear risk of bias: single blinding, if the participants, or physicians, or outcome assessors were blinded.
High risk of bias: open-label, if blinding was not applied.
Incomplete data reporting
Low risk of bias: if the dropout numbers were low (e.g. less than 20%) and they were evenly distributed among different groups, or if it was specified that there were no withdrawals or losses to follow up.
Unclear risk of bias: if the report gave the impression that there had been no withdrawals or losses to follow up but this was not specifically stated.
High risk of bias: if the number of, or reasons for, withdrawals or losses to follow up were not described.
Selective reporting bias
Selective reporting is a type of reporting bias that affects the internal validity of an individual study. It refers to the selective reporting of some outcomes (for example positive outcomes) and the failure to report others (for example adverse events). If the trial protocols were not available, we would compare the outcome measures in the method section with the actual reported outcomes in the results for the assessment of selective reporting bias.
We considered baseline comparability as an important factor for other bias. If baseline data were comparable, the study would be at low risk of other bias. Otherwise, no information or insufficient information would be considered as either high or unclear risk of bias.
Measures of treatment effect
We presented dichotomous data as risk ratios (RR) and continuous outcomes as mean differences (MD), both with 95% confidence intervals (CI). If similar outcomes were reported on different scales (for example change in weight) we would calculate the standardised mean difference (SMD) with 95% CI.
Unit of analysis issues
The primary analysis would be per woman randomised. Only first-phase data from cross-over trials would be included.
Dealing with missing data
We would perform analyses by intention to treat where possible and attempt to obtain missing data from the original trialists. For dichotomous outcomes, participants with incomplete or missing data were to be included in a sensitivity analysis by counting them as treatment failures to explore the possible effect of loss to follow up on the findings ('worst-case' scenario). For continuous data, we took a 'carry forward' approach, in which we used the last observed patient data, if available, as the missing data to conduct data analysis.
Assessment of heterogeneity
We considered whether the clinical and methodological characteristics of the included studies were sufficiently similar for meta-analysis to provide a clinically meaningful summary. We would assess statistical heterogeneity by the I
Assessment of reporting biases
If there were 10 or more studies in an analysis, we would use a funnel plot to explore the possibility of small study effects (a tendency for estimates of the intervention effect to be more beneficial in smaller studies) (Egger 1997; Vickers 1998).
If the studies were sufficiently similar, we combined the data using a fixed-effect model in the following comparisons:
1. herbal medicine versus no treatment;
2. herbal medicine versus placebo;
3. herbal medicine versus pharmacological treatment;
4. herbal medicines versus a surgical procedure; or
5. herbal medicines plus conventional therapy versus conventional therapy.
Furthermore, if a combined analysis showed significant heterogeneity (defined as P < 0.1 for the heterogeneity test), we would use a random-effects model for the analysis.
Subgroup analysis and investigation of heterogeneity
If a sufficient number of randomised trials was identified and data were available, we would have performed subgroup analyses according to symptoms (presence or absence of), diagnosis with or without imaging confirmation, and the location of uterine fibroids (submucosal, intramural, or subserosal fibroids). Whenever there was significant heterogeneity we used a random-effects model and investigated heterogeneity in both the clinical characteristics and methodological differences between studies. We would carry out subgroup analyses in Review Manager 5.1.7 (RevMan 2012) to see if any differences were explained by differences between the studies.
If a sufficient number of randomised trials were identified for the same interventions, we would conduct sensitivity analyses for the primary outcomes to determine whether the conclusions were robust to arbitrary decisions made regarding the eligibility of studies and analysis. These analyses would include consideration of whether the review conclusions would have differed if:
1. eligibility was restricted to studies without high risk of bias;
2. alternative imputation strategies had been implemented;
3. the summary effect measure was odds ratio rather than risk ratio.
Overall quality of the body of evidence: summary of findings table
Summary of findings tables were generated using GRADEPRO software. These tables evaluated the overall quality of the body of evidence for the main review outcomes using GRADE criteria (study limitations (that is risk of bias), consistency of effect, imprecision, indirectness, and publication bias). Judgements about evidence quality (high, moderate, or low) were justified, documented, and incorporated into reporting of results for each outcome.
We intend to complete an update of the review every 24 months.
Description of studies
Results of the search
Our initial electronic searches identified 984 citations, with a further 35 from additional handsearches. After reading titles and abstracts we excluded 733 of these because they were either duplicates, non-clinical studies, review articles, case reports, case series, or had study objectives different from this review. A total of 158 references published in Chinese or in English were retrieved for further assessment. We excluded 93 of these studies because they did not meet our inclusion criteria. We contacted the trial authors of 13 trials to confirm the randomisation methods and missing data, and this allowed us to exclude, by phone calls, 136 trials that had claimed to be 'randomised'. This was due to inadequate randomisation methods or failure to provide required data (Figure 1).
|Figure 1. Study flow diagram.|
In the previous version of the review, only two trials were included. Our updated searches in September 2012 identified 132 trials, and 19 randomised trials were eligible to be included taking the total number of included trials to 21.
Study design and setting
We were able to include 21 parallel-group, randomised controlled trials (RCTs) involving 2222 participants in this review (Deng XL 2010; Dong M 2011; Fu WJ 2005; Gu HH 2011; Hazlina 2005; Lai XL 2010; Liu Y 2009; Liu LY 2010; Lu JX 2007; Lu HJ 2010; Luo SQ 2010; Ma R 2010; Mao CX 2012; Mao XG 2012; Ni XP 2012; Wang XR 2011; Wen Q 2005; Wu JH 2011; Wu YF 2011; Yan LQ 2000; Zhu FH 2006). These RCTs reported random allocation of participants with uterine fibroids to herbal medicines or placebo, mifepristone, or GnRH agonist. Twelve trials compared the herbal medicine Guizhi Fuling formula plus medication with medication alone (including eight trials of Guizhi Fuling capsule plus mifepristone versus mifepristone, one trial of Guizhi Fuling capsule plus leuprolide versus mifepristone, one trial of Gongliuqing capsule plus mifepristone versus mifepristone, one trial of Lenge Xiaozheng Tang plus mifepristone versus mifepristone, and one trial of Jiliu Tang plus mifepristone versus mifepristone). The 21 RCTs are listed in the table 'Characteristics of included studies'. Twenty trials were published in Chinese and one trial in English. No trial had a pre-trial sample size estimation (power calculation) or was presented as a multicentre trial.
A total of 2222 women with uterine fibroids were randomised into herbal treatment (n = 1118) or control (n = 1104). Twenty trials were conducted in China and one trial in Malaysia. The 21 trials included women of childbearing age with uterine fibroids, diagnosed through routine gynaecological examination and confirmed by B-mode ultrasound (most commonly used diagnostic method for uterine fibroids, which can show a clear two (or three) dimensional image of the size and location of uterine fibroids). As available outcome data were limited, we could not perform prespecified subgroup analyses, that is of symptom type or location of fibroids.
Ten herbal preparations were tested in the 21 trials ( Table 1). The controls were placebo (one trial), pharmaceutical medicines including mifepristone and GnRH agonist (eight trials). Twelve trials tested herbal medicine plus medication versus medication (including four herbal medicines: Guizhi Fuling capsule, Qingliuqing capsule, Lenge Xiaozheng Tang, and Jiliu Tang). The average treatment duration was 3.6 months (ranging from three to six months).
No trial reported the primary outcome for effectiveness, that is uterine fibroid related symptoms measured by a validated instrument. Thirteen of 21 trials reported the outcome of adverse events in relation to herbal medicines ( Table 2). Among secondary outcomes, no trials reported the need for surgical treatment, quality of life, or incidence of complications such as infertility. The outcomes reported were volume of the fibroids or size of the uterus. The volume of fibroids or the size of the uterus was measured by B-mode ultrasound. However, the method of calculating volume was different in some of the trials. Some studies reported the average volume of the fibroids by calculating the maximum fibroid size in each woman, while others reported the average volume by calculating the totality of multiple fibroids. Three trials reported follow up after the completion of treatment, ranging from three to six months.
The reasons for exclusion of 137 studies are listed in the table 'Characteristics of excluded studies'.
Risk of bias in included studies
We had made phone calls to the authors of included 'randomised' trials to confirm the randomisation methods and enquire about missing information. This led to us excluding some 'randomised' trials we planned to include. In general, the included trials had high or unclear risk of bias, and therefore, they were evaluated as low methodological quality (Figure 2; Figure 3).
|Figure 2. Methodological quality graph: review authors' judgements about each methodological quality item presented as percentages across all included studies.|
|Figure 3. Methodological quality summary: review authors' judgements about each methodological quality item for each included study.|
Our contact with trial authors by phone resulted in our exclusion of eight trials that we originally planned to include (see Characteristics of excluded studies). The main reasons were inadequate methods for allocation of participants, such as alternate allocation, or the data for this review were not available and the authors failed to provide them. We were not able to perform meaningful sensitivity or funnel plot analysis due to the limited number of trials.
Eight out of 21 included trials reported generation of allocation sequence. However, only one trial described adequate allocation concealment (Hazlina 2005). The other trials did not report the method for allocation concealment.
Two trials applied blinding (Hazlina 2005; Ma R 2010), including one trial using blinding of participants and personnel (Ma R 2010) and one trial using blinding of the outcome assessor (Hazlina 2005).
Incomplete outcome data
All included trials analysed all or most (> 95%) women randomised and we judged the majority of trials to be at low risk of bias. However, due to the non-availability of raw data from the missing participants for continuous data, we were not able to do an intention-to-treat analysis.
As we were not able to get access to trial protocols for any of the 21 trials, we made our judgement by comparing the outcome measures mentioned in the method section with the reporting in results: 13/21 trials reported all outcome measures described in the methods and, therefore, were evaluated as at low risk of bias; 7/21 partially reported the outcomes in the results, and were evaluated as at high risk of bias.
Other potential sources of bias
All trials reported baseline comparability between the two groups, and were considered to be at low risk of bias. We found no potential sources of within-study bias in the included trials.
Effects of interventions
See: Summary of findings for the main comparison Herbal preparations versus placebo; Summary of findings 2 Herbal preparations versus medication; Summary of findings 3 Herbal preparations plus medication versus medication
1. Herbal medicine versus placebo
No trial in this category reported the primary outcome, that is fibroids related symptoms, using validated methods. One trial tested the herbal medicine Guihong turtle shell pill against placebo for three-month treatment of 136 women with uterine fibroids (Ma R 2010). The participants were diagnosed as having qi stagnation and blood stasis, without an indication for surgery.
Since this trial evaluated non-relevant outcomes such as disappearance or shrinkage of uterine fibroids, we were not able to evaluate the efficacy of herbal medicine in this comparison.
In the Ma R 2010 study, there was one case in the herbal group who developed abnormal serum blood urea nitrogen (BUN) and creatinine levels after three months of treatment, which might represent potential impairment to kidney function. No case with abnormal kidney function occurred in the placebo group. Liver function was monitored and no trial participant developed abnormal levels of serum alanine transaminase (ALT) or aspartate aminotransferase (AST).
2. Herbal medicine versus medication
No trial in this category reported the primary outcome, that is fibroids related symptoms, using validated methods. Seven trials compared herbal medicines versus medication (Fu WJ 2005; Hazlina 2005; Lai XL 2010; Lu JX 2007; Ni XP 2012; Wen Q 2005; Yan LQ 2000). Six different herbal preparations were tested: Gongliuqing capsule, Nona Roguy herbal product, Huoxue Huayu Ruanjian Sanjie, Huoxue Sanjie decoction, Tripterygium wilfordii, and Xiaozheng decoction. Since no two trials tested the same herbal medicine except for Tripterygium wilfordii, a meta-analysis was performed for Tripterygium wilfordii compared with mifepristone on the average volume of uterine fibroids.
2.1 Average volume of uterine fibroids
Among four herbal medicines compared to mifepristone, there was no significant difference between herbal medicine and mifepristone in the average volume of uterine fibroids ( Analysis 1.1). The tested herbal medicines were Gongliuqing capsule (Lai XL 2010), Huoxue Huayu Ruanjian Sanjie (Yan LQ 2000), Huoxue Sanjie decoction (Lu JX 2007), and Xiaozheng decoction (Ni XP 2012). Similarly, there was no significant difference between the Nona Roguy herbal product and GnRH agonist in average volume of uterine fibroids (Hazlina 2005). However, herbal extracts of Tripterygium wilfordii showed a significantly better effect than mifepristone in the average volume of uterine fibroids (MD -23.03 cm
2.2 Average size of uterus
Six trials reported on the size of the uterus after three to six months of treatment. Extracts of Tripterygium wilfordii showed significantly better effect than mifepristone in reducing the size of the uterus (MD -51.25 cm
There was no significant difference between the herbal preparation Nona Roguy and the GnRH agonist regarding uterus size (Hazlina 2005).
2.3 Adverse effects
Five trials in this category reported on adverse effects ( Table 2) and the reported adverse effects included amenorrhoea, menopausal symptoms, gastrointestinal discomfort in the herbal treatment group. However, similar adverse effects were reported in the control group. No serious adverse events such as death or disability were reported.
3. Herbal medicine plus medical treatment versus medical treatment
No trial in this category reported the primary outcome, that is fibroids related symptoms, using validated methods. Thirteen randomised trials tested herbal medicine plus medication against medication alone. Nine trials compared the herbal medicine Guizhi Fuling formula plus mifepristone versus mifepristone (Deng XL 2010; Gu HH 2011; Liu Y 2009; Lu HJ 2010; Mao CX 2012; Mao XG 2012; Wang XR 2011; Wu JH 2011; Wu YF 2011). One trial tested Guizhi Fuling capsule plus leuprolide acetate against mifepristone (Dong M 2011). Guizhi Fuling formula was taken in either capsule or decoction form. Three other herbal medicines were tested with mifepristone against mifepristone alone, including Gongliuqing capsule (Liu LY 2010), Lenge Xiaozheng Tang (Zhu FH 2006), and Jiliu Tang (Luo SQ 2010).
3.1 Average volume of maximum fibroids
Compared with mifepristone alone, Guizhi Fuling formula combined with mifepristone showed a significantly better effect in reducing the average volume of maximum fibroids (MD -1.72 cm
The combination therapy of Guizhi Fuling capsule and leuprolide acetate was more effective than mifepristone alone in one trial (Dong M 2011). Similarly, Gongliuqing capsule plus mifepristone was more effective than mifepristone alone in one trial (Liu LY 2010). In another trial, Lenge Xiaozheng Tang plus mifepristone showed no significant difference compared to mifepristone alone in one trial (Zhu FH 2006).
3.2 Average volume of total multiple fibroids
Three trials measured and reported the average volume of the total multiple fibroids (Liu Y 2009; Lu HJ 2010; Luo SQ 2010). Guizhi Fuling capsule plus mifepristone showed a better effect in reducing the average volume of total multiple fibroids (MD -16.58 cm
3.3 Average size of uterus
Four trials in this category reported on the average size of the uterus after treatment (Deng XL 2010; Liu Y 2009; Luo SQ 2010; Wu JH 2011). A pooled analysis of data from three trials showed a significant beneficial effect of Guizhi Fuling capsule plus mifepristone in reducing the average size of uterus compared to mifepristone alone (MD -31.63 cm
3.4 Adverse effects
Eight trials in this category reported on the adverse effects in relation to the herbal treatments (Deng XL 2010; Dong M 2011; Liu Y 2009; Liu LY 2010; Mao CX 2012; Mao XG 2012; Wu JH 2011; Zhu FH 2006) ( Table 2). The adverse effects included gastrointestinal discomfort, itching, and hot flushes. Similar adverse effects were reported in the control group. No serious adverse events from herbal preparations were reported.
4. Other analyses
Our specified sensitivity analyses, subgroup analyses, and test for publication bias were unable to be performed due to significant heterogeneity of the herbal interventions and the limited number of trials under each comparison.
Although no trial reported on the primary outcome of symptoms, we summarised the major findings in the summary of findings tables ( Summary of findings for the main comparison; Summary of findings 2; Summary of findings 3).
Summary of main results
This systematic review included 21 randomised trials of herbal preparations for the treatment of uterine fibroids. The majority of the herbal preparations were tested in single trials. Only the herbal medicines Guizhi Fuling formula (either capsule or decoction) and Tripterygium wilfordii were tested in two or more trials, with or without mifepristone against mifepristone alone. One trial with good quality was published in English (Hazlina 2005) and the other trials, published in Chinese, were of low quality. No trials reported menstrual symptoms related to fibroids, quality of life, incidence of complications, or the need for a surgical procedure. Most of the trials reported the volume of fibroids or size of the uterus, or both. The trials in this review showed a similar effect of herbal preparations combined with medication in terms of reduced volume of uterine fibroids. However, due to the small sample of the trials and methodological flaws in the majority of the trials, any indicated benefit is not conclusive. Further large and rigorous trials are needed.
Overall completeness and applicability of evidence
The included studies tested 10 different herbal medicines with conventional therapy, including mifepristone or GnRHa. In general, there was no significant difference between herbs and medical treatment for the reduced volume of fibroids. However, the lack of statistical significant difference does not mean equal effectiveness as none of the trials were designed as equivalence or non-inferiority trials and the sample size was no more than 100 in each arm in the majority of the trials.
With regard to the conventional medical treatments used in the included studies, the efficacy of mifepristone for reducing fibroid volume has not been firmly established (Tristan 2012). A Cochrane systematic review demonstrated that mifepristone reduced heavy menstrual bleeding and improved fibroid-specific quality of life (Tristan 2012). Unfortunately, none of the trials in our review reported these outcomes and we don't know whether herbal medicines can be helpful in relieving fibroid related symptoms or not.
The evidence from this review is not sufficiently convincing to support a clinical recommendation due to the following aspects of the trials.
- There is a lack of evidence on the clinical effect of individual herbal preparations for menstrual symptom improvement in uterine fibroids. In clinical practice, Chinese herbal medicine is used mainly for symptom improvement, but this was not confirmed from the included trials due to the lack of a validated measurement, or reporting of symptoms or quality of life. Almost all reported outcomes were surrogate outcomes and may not reflect the clinical effectiveness. Comparisons with placebo are needed, as there is no clear evidence of the efficacy of the comparators used in the trials.
- Although Chinese herbal extracts of Tripterygium wilfordii showed a promising effect compared with mifepristone, the findings are not confirmed as we only had two small trials that were of poor quality and uncertain evidence on safety. The Chinese herbal medicine Guizhi Fuling showed a promising benefit when combined with mifepristone versus mifepristone alone. However, the findings need to be verified in large, rigorous trials.
- The trials reported outcomes by the end of treatment or at short-term follow up. For those women with asymptomatic fibroids or mild symptoms, the use of herbal therapies is intended to prevent fibroid growth or to manage the mild symptoms. For this (wide) subgroup of women the main outcome is the avoidance of surgical treatment, measured through long-term follow up. Future trialists are encouraged to adopt this outcome, as women may simply reach their menopause without needing surgery. In addition, reproductive outcomes related to uterine fibroids, such as the relationship between submucosal, intramural, or subserosal fibroids and pregnancy rates, miscarriage, and malpresentation, should be addressed in future trials (Klatsky 2008).
- Reporting of adverse events in relation to herbal preparations was not sufficient in the included trials, and one trial suggested potential kidney function impairment after three-month herbal treatment. Therefore, the safety of herbal medicine is still undetermined.
Quality of the evidence
This systematic review has several methodological limitations. Firstly, there is a lack of high quality trials and we had to exclude some of the trials that claimed to be randomised because of an unexplainable skew in the distribution of participants among the compared groups or an inadequate method for sequence generation for randomisation, which means they were highly prone to selection bias (Liu J 2002).
Thirdly, the trials had a small sample size. Although some data analyses did not demonstrate a statistically significant difference between herbal medicines and conventional medicine, the results are likely to have been underpowered. Therefore, the size of the trials may mean that the analyses may not establish with confidence that the two interventions have equivalent effects.
Fourth, the trials failed to report clinically useful outcomes such as symptoms or quality of life, which may suggest evidence of selective reporting bias. We could not differentiate the participants with symptoms from those without symptoms in the included studies. It is difficult to justify the herbal medicine treatment as some women without symptoms may not need any treatment. Therefore, we suggest that future trials should measure and report clinical symptoms as one of the major outcomes.
The above limitations mean that potential bias may have been present in the selection of participants, administration of treatment, and assessment of outcomes in the primary studies. Methodologically less rigorous trials show significantly larger intervention effects than more robust trials (Egger 2003; Kjaergard 2001; Moher 1998; Schulz 1995). An empirical study has shown that Chinese trials are significantly affected by publication bias (Vickers 1998). When interpreting the present findings, publication bias should be taken into consideration accordingly.
In summary, the findings of this review should be interpreted with caution due to the small sample sizes, low methodological quality in the majority of the 21 trials, and the limited number of trials included for each individual herbal preparation.
Potential biases in the review process
Although we conducted comprehensive searches in both English and Chinese databases, we may have missed some studies published in the non-English or non-Chinese literature, such as in the Japanese or Korean language. Second, we endeavoured to contact trial authors to clarify the methods for randomisation and obtain missing data, but the response was not satisfactory and leaves some trials with unclear randomisation. This may cause selection bias, and may not reflect the whole picture in using herbal medicine for the treatment of uterine fibroids.
Agreements and disagreements with other studies or reviews
As far as we know, there is no other systematic review or meta-analysis published on the same topic. We also could not identify any large, multicentre trials for a comparison of our findings with other types of evidence.
Implications for practice
Current evidence does not support the use of herbal preparations for treatment of uterine fibroids. There is no conclusive evidence of benefit due to a limited number of trials conducted for individual herbal preparations, the methodological quality of the primary studies, and their insufficient power to meet robust conclusions.
Implications for research
Further well-designed, randomised, double blind, placebo-controlled trials are needed to evaluate herbal preparations for uterine fibroids. To improve quality, trials needs to use appropriate allocation concealment; blinding of participants, researchers and outcome assessors; and clarify the number of participants randomised and the number analysed. Clinically relevant outcomes, such as symptoms, quality of life, infertility, and anaemia, should be addressed and measured using validated patient-reported instruments. Potentially promising herbal preparations require further trials with large samples. As in this systematic review, the quality of herbal medicines was not reported in detail and for future trials it is important to investigate herbal medicines according to a set of criteria which include a preparation consistent with the description in the pharmacopoeia, chemical standardisation, biological assays, animal models, and clinical testing (Yuan 2000). It will be necessary to improve the description of the herbal medicines being tested, for example plant species, geographical origin, harvest season, preparation procedures, and the quality of the products. Furthermore, future trials should pay more attention to the adverse effects of herbal medicines, especially for long-term use. Adverse events should be fully recorded and reported. Finally, trial reports should follow international standards, such as the CONSORT statement (http://www.consort-statement.org/), and the trial protocol should be registered and accessible.
The authors thank the Cochrane Menstrual Disorders and Subfertility Group (MDSG) for their expertise and editorial input. We would like to specifically thank the Trials Search Coordinator of the MDSG, Marian Showell, for her help with the literature searches. We thank Dr Nik Hazlina, Nik Hussain for providing us with additional data from their study. We also thank Ms Nini Chen for helping with validating data extraction and analyses in the updating of the review.
This work was funded by the Grant Number 2011ZX09302-006-01(5) and 101207007 from the Ministry of Science and Technology of China. The contents of this article are solely the responsibility of the authors and do not necessarily represent the official views of the funders.
Jianping Liu's work was supported by the Programme of Innovative Research Team Project (2011-CXTD-09) of Beijing University of Chinese Medicine, and the "111" Project (B08006).
Data and analyses
- Top of page
- Summary of findings [Explanations]
- Authors' conclusions
- Data and analyses
- What's new
- Contributions of authors
- Declarations of interest
- Sources of support
- Index terms
Appendix 1. Detailed search strategies for English literature
MDSG keywords Sept 2012
Keywords CONTAINS "fibroid" or "Leiomyoma" or "myoma" or "myomas" or "myomata" or "uterine fibroids" or "uterine leiomyomas" or "uterine myoma" or "uterine myomas" or "Uterine Neoplasms" or "fibroids" or Title CONTAINS "fibroid" or "Leiomyoma" or "myoma" or "myomas" or "myomata" or "uterine fibroids" or "uterine leiomyomas" or "uterine myoma" or "uterine myomas" or "Uterine Neoplasms" or "fibroids"
Keywords CONTAINS "Chinese herbal medicine" or "chinese herbal preparations" or "Chinese herbal remedy" or "Chinese traditional medicine" or "plant extracts" or "herbal preparations" or "herbal remedy", "herbal supplement" or "herbal supplements" or Title CONTAINS "Chinese herbal medicine" or "chinese herbal preparations" or "Chinese herbal remedy" or "Chinese traditional medicine" or "plant extracts" or "herbal preparations" or "herbal remedy", "herbal supplement" or "herbal supplements"
AMED 1985 to Sept 2012
1 traditional medicine$.tw. (5881)
2 exp plant extracts/ or exp drugs, chinese herbal/ (15672)
3 chinese herb$.tw. (1492)
4 plant extract$.tw. (9931)
5 chinese medicine$.tw. (1026)
6 exp Plants, Medicinal/ (14154)
7 (Plant$ adj2 Medicin$).tw. (13318)
8 herb$.tw. (9770)
9 exp Phytotherapy/ (1049)
10 Phytotherap$.tw. (1346)
11 alternative medicine$.tw. (1201)
12 exp ethnopharmacology/ or exp remedies/ or exp traditional medicine chinese/ (4761)
13 exp herbal drugs/ (6344)
14 or/1-13 (26465)
15 exp uterine neoplasms/ (20)
16 (uterine adj5 neoplasm$).tw. (28)
17 fibroid$.tw. (23)
18 (fibroma$ or leiomyom$).tw. (48)
19 (myoma$ or hysteromyom$).tw. (15)
20 fibroid$.tw. (23)
21 or/15-20 (87)
22 14 and 21 (17)
23 from 22 keep 1-17 (17)
CENTRAL Issue 4, 2012
1 exp Fibroma/ (1)
2 fibroma$.tw. (16)
3 leiomyom$.tw. (148)
4 exp Myoma/ (7)
5 myoma$.tw. (150)
6 hysteromyom$.tw. (7)
7 fibroma$.tw. (16)
8 fibroid$.tw. (153)
9 exp Leiomyoma/ (247)
10 or/1-9 (442)
11 exp medicine, traditional/ or exp medicine, oriental traditional/ (351)
12 traditional medicine$.tw. (69)
13 exp plant extracts/ or exp drugs, chinese herbal/ (2840)
14 chinese herb$.tw. (331)
15 plant extract$.tw. (72)
16 chinese medicine$.tw. (428)
17 exp Plants, Medicinal/ (742)
18 (Plant$ adj2 Medicin$).tw. (35)
19 herb$.tw. (1117)
20 exp Phytotherapy/ (1566)
21 Phytotherap$.tw. (48)
22 alternative medicine$.tw. (69)
23 or/11-22 (4822)
24 10 and 23 (7)
25 from 24 keep 1-7 (7)
CINAHL 1982 to Sept 2012
1 exp Fibroma/ (0)
2 fibroma$.tw. (184)
3 leiomyom$.tw. (177)
4 exp Myoma/ (36)
5 myoma$.tw. (72)
6 hysteromyom$.tw. (0)
7 fibroma$.tw. (184)
8 fibroid$.tw. (297)
9 exp Leiomyoma/ (569)
10 or/1-9 (879)
11 exp medicine, traditional/ or exp medicine, oriental traditional/ (10482)
12 traditional medicine$.tw. (294)
13 exp plant extracts/ or exp drugs, chinese herbal/ (2691)
14 chinese herb$.tw. (313)
15 plant extract$.tw. (123)
16 chinese medicine$.tw. (678)
17 exp Plants, Medicinal/ (13658)
18 (Plant$ adj2 Medicin$).tw. (236)
19 herb$.tw. (4109)
20 exp Phytotherapy/ (3397)
21 Phytotherap$.tw. (93)
22 alternative medicine$.tw. (2179)
23 or/11-22 (26022)
24 10 and 23 (14)
25 exp clinical trials/ (57427)
26 Clinical trial.pt. (29998)
27 (clinic$ adj trial$1).tw. (13150)
28 ((singl$ or doubl$ or trebl$ or tripl$) adj (blind$3 or mask$3)).tw. (7801)
29 Randomi?ed control$ trial$.tw. (11197)
30 Random assignment/ (17445)
31 Random$ allocat$.tw. (1211)
32 Placebo$.tw. (10846)
33 Placebos/ (4145)
34 Quantitative studies/ (3735)
35 Allocat$ random$.tw. (73)
36 or/25-35 (79254)
37 24 and 36 (2)
38 from 37 keep 1-2 (2)
EMBASE to Sept 2012
1 exp traditional medicine/ or exp chinese medicine/ or exp herbal medicine/ or exp oriental medicine/ (21306)
2 exp Plant Extract/ (57548)
3 exp Medicinal Plant/ (40872)
4 (traditional adj2 medicin$).tw. (6936)
5 chinese herb$.tw. (2504)
6 plant extract$.tw. (2889)
7 chinese medicine.tw. (3872)
8 (herbal adj2 medicin$).tw. (3621)
9 (oriental adj2 medicine).tw. (354)
10 (Medicin$ adj2 Plant$).tw. (4699)
11 herb$.tw. (27378)
12 exp Phytotherapy/ (3672)
13 Phytotherap$.tw. (1297)
14 alternative medicine$.tw. (2950)
15 or/1-14 (109922)
16 exp benign uterus tumor/ or exp leiomyoma/ or exp uterus myoma/ (8988)
17 exp Fibroma/ (2830)
18 (Fibroma$ or leiomyom$).tw. (9914)
19 (myoma$ or hysteromyom$).tw. (2258)
20 fibroid$.tw. (2241)
21 or/16-20 (17218)
22 15 and 21 (35)
23 Clinical trial/ (495185)
24 Randomized controlled trials/ (155511)
25 Random Allocation/ (25203)
26 Single-Blind Method/ (7410)
27 Double-Blind Method/ (68576)
28 Cross-Over Studies/ (20046)
29 Placebos/ (111054)
30 Randomi?ed controlled trial$.tw. (28060)
31 RCT.tw. (2194)
32 Random allocation.tw. (605)
33 Randomly allocated.tw. (9592)
34 Allocated randomly.tw. (1314)
35 (allocated adj2 random).tw. (552)
36 Single blind$.tw. (7066)
37 Double blind$.tw. (81296)
38 ((treble or triple) adj blind$).tw. (127)
39 Placebo$.tw. (104327)
40 Prospective Studies/ (73142)
41 or/23-40 (651841)
42 Case study/ (5369)
43 Case report.tw. (110903)
44 Abstract report/ or letter/ (461484)
45 or/42-44 (575754)
46 41 not 45 (629234)
47 animal/ (18235)
48 human/ (6058876)
49 47 not 48 (14465)
50 46 not 49 (629138)
51 or/23-50 (6253078)
52 22 and 51 (33)
53 from 52 keep 1-33 (33)
MEDLINE 1950 to Sept 2012
1 exp Fibroma/ (9610)
2 fibroma$.tw. (7047)
3 leiomyom$.tw. (7602)
4 exp Myoma/ (1626)
5 myoma$.tw. (3317)
6 hysteromyom$.tw. (26)
7 fibroma$.tw. (7047)
8 fibroid$.tw. (2244)
9 exp Leiomyoma/ (13308)
10 or/1-9 (29717)
11 exp medicine, traditional/ or exp medicine, oriental traditional/ (17249)
12 traditional medicine$.tw. (2357)
13 exp plant extracts/ or exp drugs, chinese herbal/ (59586)
14 chinese herb$.tw. (2614)
15 plant extract$.tw. (2601)
16 chinese medicine$.tw. (3950)
17 exp Plants, Medicinal/ (44249)
18 (Plant$ adj2 Medicin$).tw. (4189)
19 herb$.tw. (29907)
20 exp Phytotherapy/ (17380)
21 Phytotherap$.tw. (781)
22 alternative medicine$.tw. (3563)
23 or/11-22 (132460)
24 10 and 23 (43)
25 randomised controlled trial.pt. (251334)
26 controlled clinical trial.pt. (77422)
27 randomised controlled trials as topic/ (53023)
28 random allocation/ (60395)
29 double blind method/ (96065)
30 single blind method/ (11789)
31 or/25-30 (424467)
32 animals/ not (animals/ and humans/) (3189559)
33 31 not 32 (397756)
34 clinical trial.pt. (446433)
35 exp clinical trials as topic/ (201557)
36 (clinic$ adj25 trial$).ti,ab. (142061)
37 cross-over studies/ (21493)
38 (crossover or cross-over or cross over).tw. (40169)
39 ((singl$ or doubl$ or trebl$ or tripl$) adj25 (blind$ or mask$)).ti,ab. (95360)
40 placebos/ (26962)
41 placebo$.ti,ab. (108242)
42 random$.ti,ab. (401463)
43 research design/ (51618)
44 or/34-43 (910204)
45 44 not 32 (843399)
46 33 or 45 (865071)
47 24 and 46 (8)
48 from 47 keep 1-8 (8)
PsycINFO 1806 to Sept 2012
1 exp Fibroma/ (0)
2 fibroma$.tw. (15)
3 leiomyom$.tw. (2)
4 exp Myoma/ (0)
5 myoma$.tw. (12)
6 hysteromyom$.tw. (1)
7 fibroma$.tw. (15)
8 fibroid$.tw. (15)
9 exp Leiomyoma/ (0)
10 or/1-9 (44)
11 exp medicine, traditional/ or exp medicine, oriental traditional/ (0)
12 traditional medicine$.tw. (205)
13 exp plant extracts/ or exp drugs, chinese herbal/ (0)
14 chinese herb$.tw. (65)
15 plant extract$.tw. (41)
16 chinese medicine$.tw. (212)
17 exp Plants, Medicinal/ (0)
18 (Plant$ adj2 Medicin$).tw. (68)
19 herb$.tw. (3468)
20 exp Phytotherapy/ (0)
21 Phytotherap$.tw. (14)
22 alternative medicine$.tw. (790)
23 or/11-22 (4537)
24 10 and 23 (1)
25 from 24 keep 1 (1)
Appendix 2. Search strategies for Chinese biomedical databases
Since the Chinese databases have different indexing and search functions, we listed below generic string search terms for use in different databases:
1 Zi Gong ji liu (Chinese spelling, in English 'uterine fibroids')
2 Zhong yao (Chinese materia medica)
3 Zhong cheng yao (Chinese patent medicine)
4 Zhong cao yao (Chinese herbal drug)
5 Tang yao (herbal decoction)
6 Lin chuang yan jiu (clinical studies)
7 Lin chuang shi yan (clinical trials)
Last assessed as up-to-date: 28 February 2013.
Protocol first published: Issue 2, 2005
Review first published: Issue 2, 2009
Contributions of authors
Jianping Liu conceived the review, wrote the protocol, performed quality assessment and data analyses, wrote and updated the review.
Hong Yang identified studies, extracted data, performed quality assessment, and analysed data.
Yun Xia contacted trial authors for confirmation of randomisation and to obtain missing data.
Francesco Cardini revised the protocol and the review.
Declarations of interest
Sources of support
- Beijing University of Chinese Medicine, China.
- National Research Centre in Complementary and Alternative Medicine (NAFKAM), Norway.
- The '111' Project (B08006), China.
- The Programme for Innovative Research Team (No. 2011-CXTD-09) of Beijing University of Chinese Medicine, China.
Medical Subject Headings (MeSH)
Drugs, Chinese Herbal [therapeutic use]; Gonadotropin-Releasing Hormone [therapeutic use]; Hormone Antagonists [therapeutic use]; Leiomyoma [*drug therapy; pathology]; Mifepristone [therapeutic use]; Phytotherapy [*methods]; Plant Preparations [adverse effects; *therapeutic use]; Randomized Controlled Trials as Topic; Uterine Neoplasms [*drug therapy; pathology]
MeSH check words