Chinese Herbal Medicines For Threatened Miscarriage

Chinese herbal medicines for threatened miscarriage (Review)
Li L, Dou L, Leung PC, Wang CC
This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library
2012, Issue 5
http://www.thecochranelibrary.com

Copyright 2012 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
TABLE OF CONTENTS
HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
AUTHORS CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ACKNOWLEDGEMENTS
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 1.1. Comparison 1 Chinese herbal medicines versus Western medicines (primary outcome and non-specific
secondary outcome), Outcome 1 Effectiveness of intervention (primary outcome). . . . . . . . . . .
secondary outcome), Outcome 2 No relief of clinical signs. . . . . . . . . . . . . . . . . . .
secondary outcome), Outcome 12 Effectiveness of intervention (non-prespecified secondary outcome). . . .
Analysis 3.1. Comparison 3 Combined medicines versus Western medicines (primary outcome and non-specific secondary
outcome), Outcome 1 Effectiveness of intervention (primary outcome). . . . . . . . . . . . . . .
outcome), Outcome 2 No relief of clinical signs. . . . . . . . . . . . . . . . . . . . . . .
outcome), Outcome 12 Effectiveness of intervention (non-prespecified secondary outcome). . . . . . . .
ADDITIONAL TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
HISTORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DIFFERENCES BETWEEN PROTOCOL AND REVIEW . . . . . . . . . . . . . . . . . . . . .
INDEX TERMS
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1
1
2
2
4
4
8
11
12
13
13
17
71
73
74
75
76
77
78
79
81
84
85
85
85
85
85
[Intervention Review]
Chinese herbal medicines for threatened miscarriage

Lu Li1 , Lixia Dou2 , Ping Chung Leung3 , Chi Chiu Wang1
1 Department
of Obstetrics and Gynaecology, The Chinese University of Hong Kong, Shatin, Hong Kong. 2 Cochrane Pregnancy and
Childbirth Group, Department of Womens and Childrens Health, The University of Liverpool, Liverpool, UK. 3 Institute of Chinese
Medicine, The Chinese University of Hong Kong, Shatin, Hong Kong
Contact address: Chi Chiu Wang, Department of Obstetrics and Gynaecology, The Chinese University of Hong Kong, Prince of Wales
Hospital, Shatin, New Terrorities, Hong Kong. ccwang@cuhk.edu.hk.
Editorial group: Cochrane Pregnancy and Childbirth Group.
Publication status and date: New, published in Issue 5, 2012.
Review content assessed as up-to-date: 2 April 2012.
Citation: Li L, Dou L, Leung PC, Wang CC. Chinese herbal medicines for threatened miscarriage. Cochrane Database of Systematic
Reviews 2012, Issue 5. Art. No.: CD008510. DOI: 10.1002/14651858.CD008510.pub2.
ABSTRACT
Background
Threatened miscarriage occurs in 10% to 15% of all pregnancies. Vaginal spotting or bleeding during early gestation is common,
with nearly half of those pregnancies resulting in pregnancy loss. To date, there is no effective preventive treatment for threatened
miscarriage. Chinese herbal medicines have been widely used in Asian countries for centuries and have become a popular alternative
to Western medicines in recent years. Many studies claim to show that they can prevent miscarriage. However, there has been no
systematic evaluation of the effectiveness of Chinese herbal medicines for threatened miscarriage.
Objectives
To review the therapeutic effects of Chinese herbal medicines for the treatment of threatened miscarriage.
Search methods
We searched the Cochrane Pregnancy and Childbirth Groups Trials Register (31 January 2012), Chinese Biomedical Database (1978 to
31 January 2012), China Journal Net (1915 to 31 January 2012), China National Knowledge Infrastructure (1915 to 31 January 2012),
WanFang Database (1980 to 31 January 2012), Chinese Clinical Trial Registry (31 January 2012), EMBASE (1980 to 31 January
2012), CINAHL (31 January 2012), PubMed (1980 to 31 January 2012), Wiley InterScience (1966 to 31 January 2012), International
Clinical Trials Registry Platform (31 January 2012) and reference lists of retrieved studies. We also contacted organisations, individual
experts working in the field, and medicinal herb manufacturers.
Selection criteria
Randomised or quasi-randomised controlled trials that compared Chinese herbal medicines (alone or combined with other pharmaceuticals) with placebo, no treatment (including bed rest), or other pharmaceuticals as treatments for threatened miscarriage.
Data collection and analysis
Two review authors independently assessed all the studies for inclusion in the review, assessed risk of bias and extracted the data. Data
were checked for accuracy.
Main results
In total, we included 44 randomised clinical trials with 5100 participants in the review.
We did not identify any trials which used placebo or no treatment (including bed rest) as a control.
The rate of effectiveness (continuation of pregnancy after 28 weeks of gestation) was not significantly different between the Chinese
herbal medicines alone group compared with the group of women receiving Western medicines alone (average risk ratio (RR) 1.23;
95% confidence interval (CI) 0.96 to 1.57; one trial, 60 women).
Chinese herbal medicines combined with Western medicines were more effective than Western medicines alone to continue the
pregnancy beyond 28 weeks of gestation (average RR 1.28; 95% CI 1.18 to 1.38; five trials, 550 women).
Authors conclusions
There was insufficient evidence to assess the effectiveness of Chinese herbal medicines alone for treating threatened miscarriage.
A combination of Chinese herbal and Western medicines was more effective than Western medicines alone for treating threatened
miscarriage. However, the quality of the included studies was poor. More high quality studies are necessary to further evaluate the
effectiveness of Chinese herbal medicines for threatened miscarriage.
PLAIN LANGUAGE SUMMARY

Miscarriage or spontaneous abortion is the loss of a pregnancy without medical or mechanical means before completion of the 20th week
of gestation. The fetus is not sufficiently developed to have been able to survive outside the mothers womb. Threatened miscarriage
is a very common in early pregnancy. Most threatened miscarriages occur in the first 12 weeks of pregnancy and become evident as
vaginal bleeding, abdominal and low back pain that persist for days or weeks. So far, therapies have limited effectiveness in preventing
early pregnancy loss due to threatened miscarriage. Chinese herbal medicines are a part of Traditional Chinese Medicines and are made
up of products from plants and some animal and mineral substances. They have become very popular and are commonly used as an
alternative treatment for threatened miscarriage.
This review compared the therapeutic effects of Chinese herbal medicines with other pharmaceutical agents. Among the 44 included
randomised trials involving 5100 participants, all from China, no trial used placebo or bed rest as a control intervention. Twenty
trials used a common prescription of Shou Tai Pill as a basic formula, while the other 24 trials used other formulae. The Western
medicines included tocolytic drugs such as salbutamol and magnesium sulfate, hormonal supplementation with human chorionic
gonadotrophin or progesterone and supportive supplements including vitamin E and folic acid. Five trials followed 550 women until
after 28 weeks of gestation and delivery and showed that combined Chinese herbal and Western medicines were more effective than
Western medicines alone in the treatment of threatened miscarriage. The remaining studies looked at the immediate effects of treatment.
Combined treatment was more effective than Western medicines in preventing inevitable miscarriage so that the pregnancy continued.
Many of the trials did not report on side effects during treatment or throughout continuing pregnancy and birth. Chinese medicine
practitioners slightly modify the classical prescriptions depending on the individual womens clinical presentations. All the trials had
poor methodological quality. In conclusion, there is a lack of evidence from randomised controlled trials on the effectiveness of Chinese
herbal medicines for the treatment of threatened miscarriage and to determine if Chinese herbal medicines alone are more beneficial
than Western medicines alone for threatened miscarriage.
Description of the condition

BACKGROUND
Miscarriage
Miscarriage is defined as spontaneous abortion occurring without medical or mechanical means before completion of the 20th
week of gestation so that the fetus is not sufficiently developed to
survive (Cunningham 2005). It denotes pregnancy loss prior to
completion of the 20th gestational week, or 139 days, counting
from the first day of the last normal menses (DeCherney 2007).
The incidence of miscarriage is commonly stated as 10% to 15%
of all pregnancies, and is the most common complication during
pregnancy (Petrozza 2006). However, the incidence is difficult to
determine precisely, since as many as 30% may go unrecognised,
and these can occur very early during a pregnancy.
Threatened miscarriage
Threatened miscarriage presents with vaginal bleeding or any
bloody vaginal discharge during early pregnancy, and although the
bleeding is frequently slight, it may persist for days or weeks without cervical dilatation and fetal loss (Cunningham 2005). Threatened miscarriage will become inevitable when gross rupture of
fetal membranes occurs along with severe vaginal bleeding and
cervical dilatation; imminent fetal loss is almost certain in these
cases (Cunningham 2005). Vaginal bleeding during early gestation occurs in 20% to 25% pregnancies and may last for days or
weeks; nearly half of these pregnancies will result in pregnancy loss
(Cunningham 2005). If the miscarriage is avoided, there is still a
high risk of preterm labour (Batzofin 1984), and low birthweight
(Funderburk 1980), and perinatal death (Weiss 2004).
Major causes of miscarriage

More than 80% of miscarriages occur in the first 12 weeks of pregnancy (Loue 2004). There are many factors that put a pregnancy
at a high risk of miscarriage.
Genetic defects
Chromosomal anomalies contribute to at least half of miscarriages

at the early stage of pregnancy, and it could depend on paternal,
maternal or fetal factors (Cunningham 2005). Chromosomal abnormalities from either parent can be inherited by their offspring.
Numerical abnormality, such as haploid and triploid, and structural abnormality, such as deletion and translocation, are two classical types of chromosomal anomalies for miscarriage. Fetal factors
include abnormal zygotic development and aneuploidy (numerical chromosomal abnormality).
2010). For the mother, a pregnancy is like an allograft, the maternal body may have a rejection reaction to the fetus. Maternal fetal
incompatibility occurs because of ABO and Rhesus (Rh) blood
group antigens (Bandyopadhyay 2010) and excess antiphospholipid antibodies (Suzumori 2010).
Other maternal factors
The risk of miscarriage appears to increase with parity as well as

with maternal and paternal age (Gracia 2005). A variety of maternal diseases and developmental abnormalities have been implicated, including chronic infection, anatomic defects, endocrine
deficiencies, toxin exposure, immunologic disorders, and physical
or emotional trauma, which may lead to recurrent miscarriages and
threatened miscarriages (DeCherney 2007). Some isolated events
such as laparotomy, chronic debilitating disease, nutrition, high
fever, tobacco, drug abuse, or alcoholism, are also considered to
be causes of threatened miscarriage (Lyttleton 2004).
Environmental factors
Over or lasting exposure to some chemicals and metals such as lead,

cadmium, and arsenic, could also cause harm to the pregnancy
and lead to pregnancy loss (Tabacova 1994).
Clinical management
In current clinical practice, surgical and non-surgical interventions are used in the management of miscarriage. However, in
threatened miscarriage, non-surgical interventions, which prevent
the inevitable consequence of pregnancy loss, are rather empirical
(Tien 2007). Most cases will progress to the next stage no matter
what is done. Bed rest and avoidance of sexual intercourse, though
commonly advised, are without sufficient proven benefit (Aleman
2005). Supportive care may reduce inevitable miscarriage (Tien
2007). Progesterone support has been used to reduce the incidence
of miscarriage (El-Zibdeh 2009), but the conclusion is still very
limited (Haas 2008). Hence, to date, no therapy has been confirmed as efficacious for threatened miscarriage.
Description of the intervention

Traditional Chinese medicine (TCM) has been used for more than
3000 years in China (Ma 2006). Chinese herbal medicine, one
of the commonly therapeutic approaches of TCM, mainly refers
to products made of, or from, plants, or parts of plants, and also
includes animal and mineral substances.
Immunological dysfunction
Miscarriage, caused by immunological dysfunction, is also related

to the parents. The paternal factor includes human leukocyte antigens (HLA) (Kano 2007) and antisperm antibodies (Choudhury
How the intervention might work

Unlike mainstream Western medicine, TCM has a unified medical
theory for clinical diagnosis and treatment which focuses on Qi

and blood as the two basic elements of human physiology. Qi is

equivalent to vital energy. There is an overall or essential Qi for
each organ (such as kidney Qi, liver Qi, etc). Qi deficiency (or
Qi vacuity) can lead to aging and weakness, whereas Qi stagnant
can cause pain. Blood is the sustenance of the body; Blood deficiency (or Blood vacuity) can lead to pallor and dizziness, whereas
Blood stasis can result in bleeding tendency.
TCM practitioners use inquiry, inspection, hearing, and palpation to make diagnoses and prescribe Chinese herbal medicines according to the Jun, Chen, Zuo, Shi principle (which stands for the
characters as Monarch, Minister, Assistant and Guide) (Li 2005b).
As each of the herbal medicines has its own properties and potential interactivity, the application of this principle will decrease or
avoid the side effects of the other herbs, enhance the therapeutical
actions of some herbs and collaborate with all the herbs to create
a more harmonious effect on the human body resulting in a more
enhanced and direct impact on treatments. Among all the systems,
deficiency in the kidney and liver functions are particularly important in the pathology and mechanism of miscarriage. Kidney
stores the essential Qi that warms up and activates all the other
systems in the body. It is responsible for growth, development, and
reproduction (Li 2005b). Liver stores the blood that regulates
the flow of Qi and maintains reproductivity (Li 2005b). The
main causes of miscarriage include Qi deficiency, blood heat,
blood deficiency and kidney deficiency (Ma 2006). Amongst
all of these, kidney deficiency is the most frequent clinical type
in miscarriage (Lyttleton 2004); women with kidney deficiency
tend to miscarry earlier in the pregnancy (Lyttleton 2004). Chinese herbal medicines such as Radix rehmanniae (Rehmania root),
Fructus lycii (Lycium fruit) and Semen cuscutae (Cuscuta fruit), are
commonly used to correct kidney deficiency through nourishing
the Kidney and benefiting the fetus in preventing miscarriage
(Liu 2002).
Why it is important to do this review

Chinese medicines and other herbal medicines or botanicals are
principally used in China and South East Asia, but these are increasingly becoming accepted worldwide as an alternative to Western medicines to promote the health of both mothers and fetuses
and to relieve medical problems during pregnancy. Based on our
preliminary literature search, miscarriage, in particular threatened
miscarriage, is one of the most common clinical indications for
Chinese herbal medicines during pregnancy (Li 2011b).
Despite the wide use of Chinese herbal medicines for preventing
pregnancy loss, the therapeutic effects of Chinese herbal medicines
for threatened miscarriage in early pregnancy has not been evaluated systematically. Similarly, medication exposures during early
pregnancy are known to result in birth defects, but currently there
is insufficient information on the risks and safety of Chinese herbal
medicines.
OBJECTIVES
To review the therapeutic effects of Chinese herbal medicines for
the treatment of threatened miscarriage.
METHODS
Criteria for considering studies for this review
Types of studies
Randomised or quasi-randomised controlled trials, as well as cluster-randomised trials, comparing Chinese herbal medicines (alone
or combined with other pharmaceuticals) with placebo, no treatment (including bed rest), or other pharmaceuticals as treatments
for threatened miscarriage.
Trials were included regardless of publication status, with or without full text. We only identified studies that compared Chinese
herbal medicines with other pharmaceuticals (mainly Western
medicines). We did not identify any trials which used placebo or
bed rest as a control.
We did not apply any language restrictions, and most of the papers
were in Chinese or English, which was the mother language or
second language for all the authors, who fully understood the
papers as well as the design and intervention of the trials.
Types of participants
All pregnant women with a viable pregnancy diagnosed with
threatened miscarriage, regardless of underlying causes. No treatment was given before the trial interventions. Fetal viability was assessed by ultrasound to ensure exclusion from the study of women
with inevitable, incomplete, or missed miscarriage. Women with
vaginal bleeding after the 20th week of pregnancy were also excluded. We included women regardless of whether the pregnancy
was singleton or multiple, and irrespective of the maternal age and
parity.
Types of interventions
All types of Chinese herbal medicines in either standard or combined formulae for the treatment of threatened miscarriage, regardless of the dose or duration of administration, were compared with
other pharmaceuticals. The pharmaceuticals were mostly Western
medicines. We planned the following comparisons.
Chinese herbal medicines versus placebo.
Chinese herbal medicines versus no treatment (including
bed rest).
Chinese herbal medicines alone versus other
pharmaceuticals.

Combined Chinese herbal medicines and other

pharmaceuticals versus other pharmaceuticals.
Search methods for identification of studies
Types of outcome measures
Electronic searches
Primary outcomes
(1) Effectiveness of intervention: continuation of pregnancy after

28 weeks of gestation.
Pregnancy after 28 weeks of gestation is generally considered viable, and pregnancy loss before 28 weeks is considered to be non
viable due to extreme low birthweight and underdeveloped structures. In this review, only viable pregnancy at 28 weeks was considered as the primary outcome.
Secondary outcomes
Mother
During treatment
(2) No relief of clinical signs (vaginal bleeding and abdominal
pain).
(3) No improvement in laboratory investigations (urinary and
serum beta human chorionic gonadotropin (-HCG) titre).
After treatment
(4) Repeated threatened miscarriage before 28th week of the same
pregnancy (current miscarriage signs and symptoms remitted after
intervention but relapsed in the same pregnancy).
(5) Preterm labour.
(6) Any other adverse pregnancy outcomes reported, including
side effects, toxicity, etc.
We searched the Cochrane Pregnancy and Childbirth Groups Trials Register by contacting the Trials Search Co-ordinator (31 January 2012).
The Cochrane Pregnancy and Childbirth Groups Trials Register
is maintained by the Trials Search Co-ordinator and contains trials
identified from:
1. monthly searches of the Cochrane Central Register of
Controlled Trials (CENTRAL);
2. weekly searches of MEDLINE;
3. weekly searches of EMBASE;
4. handsearches of 30 journals and the proceedings of major
conferences;
5. weekly current awareness alerts for a further 44 journals
plus monthly BioMed Central email alerts.
Guidelines of the search strategies for CENTRAL, MEDLINE and
EMBASE, the handsearched journals and conference proceedings
and the journals reviewed via the current awareness service were
followed according to the Specialized Register section from the
Cochrane Pregnancy and Childbirth Group.
In addition, we searched:
1. EMBASE (1980 to 31 January 2012) (Appendix 1)
2. Cumulative Index to Nursing and Allied Health Literature
(CINAHL) (31 January 2012) (Appendix 2)
3. PubMed (1980 to 31 January 2012) (Appendix 3)
4. China Journal Net (CJN) (1915 to 31 January 2012)
(Appendix 4)
5. China National Knowledge Infrastructure (CNKI) (1915
to 31 January 2012) (Appendix 4)
6. WanFang Database (Chinese Ministry of Science &
Technology) (1980 to 31 January 2012) (Appendix 5)
7. Chinese Biomedical Database (CBM) (1978 to 31 January
2012) (Appendix 6)
8. Wiley Inter Science (1966 to 31 January 2012) (Appendix
6)
9. Chinese Clinical Trial Registry (31 January 2012)
(Appendix 7)
10. International Clinical Trials Registry Platform (ICTRP) (31
January 2012) (Appendix 7)
Fetus
(7) Live birth (non-prespecified outcomes).
(8) Preterm birth.
(9) Stillbirth.
(10) Neonatal death.
(11) Fetal structural malformations.
(12) Any other adverse perinatal outcomes reported, including
side effects, toxicity, etc.
Searching other resources
Handsearching
We searched 64 Chinese language journals (Appendix 8).

References from published studies
Assessment of risk of bias in included studies
We searched the reference lists of relevant trials and reviews identified.
Two review authors (LL and LD) independently assessed the risk
of bias for each study using the criteria outlined in the Cochrane
Handbook for Systematic Reviews of Interventions (Higgins 2011).
It includes seven parts as follows. Any disagreements were resolved
by discussion or by involving the third assessor (CCW).
Unpublished literature
As some of the trials showed that there would be on-going studies, we tried to contact the authors for more details if the studies were completed and results were available. We contacted the
pharmaceutical companies for more information of the relevant
medicines/products.
Personal communications
We contacted organisations, individual experts working in the

field, and medicinal herb manufacturers in order to obtain additional references.
We did not apply any language restrictions.
Data collection and analysis
Selection of studies
To determine which clinical trials to include, we screened the
titles, abstracts, and keywords of the trials identified by the search.
Two review authors (LL and LD) independently assessed each
trial for inclusion and any disagreements were discussed. If the
disagreements could not be resolved, we contacted the trial authors
for clarification. We did not blind the review authors to the journal
of origin or institution.
(1) Random sequence generation (checking for possible

selection bias)
We describe for each included study the method used to generate

the allocation sequence in sufficient detail to allow an assessment
of whether it should produce comparable groups.
We assessed the method as:
low risk of bias (any truly random process, e.g. random
number table; computer random number generator);
high risk of bias (any non-random process, e.g. odd or even
date of birth; hospital or clinic record number);
unclear risk of bias.
(2) Allocation concealment (checking for possible selection

bias)
We describe for each included study the method used to conceal

allocation to interventions prior to assignment and assess whether
intervention allocation could have been foreseen in advance of, or
during recruitment, or changed after assignment.
We assessed the methods as:
low risk of bias (e.g. telephone or central randomisation;
consecutively numbered sealed opaque envelopes);
high risk of bias (open random allocation; unsealed or nonopaque envelopes, alternation; date of birth);
(3) Blinding (checking for possible performance bias)
Data extraction and management

We designed a form to extract data, and two review authors (LL)
and (LD) extracted the data using the agreed form for study eligibility. We resolved discrepancies through discussion or consulted
the third review author (CCW). We entered data into Review
Manager software (RevMan 2011), and checked for accuracy. We
assessed the abstracts in the same way as full papers, then included
them in the analyses. We excluded trials that did not meet our
eligibility criteria (or as a result of the study authors replies to our
queries) and noted the reasons for exclusion in the Characteristics
of excluded studies table. Studies awaiting classification are those
for which the authors of the original reports have not yet provided
information. However, we will reconsider these trials for inclusion
if the authors provide more information or once the full publications become available to confirm our queries.
We describe for each included study the methods used to blind

study participants and personnel from knowledge of which intervention a participant received. We assessed blinding separately for
different outcomes or classes of outcomes. We judged studies at
low risk of bias if they were blinded. Trials with no blinding or
blinding of participants only were considered at high risk of bias
which may affect the results.
low, high or unclear risk of bias for participants;
low, high or unclear risk of bias for personnel;
low, high or unclear risk of bias for outcome assessors.
(4) Incomplete outcome data (checking for possible attrition

bias due to the amount, nature and handling of incomplete
outcome data)

We describe for each included study, and for each outcome or class
of outcomes, the completeness of data including attrition and exclusions from the analysis. We state whether attrition and exclusions were reported and the numbers included in the analysis at
each stage (compared with the total randomised participants), reasons for attrition or exclusion where reported, and whether missing data were balanced across groups or were related to outcomes.
Where sufficient information was reported, or was supplied by
the trial authors, we re-included the missing data in the analyses
which we undertook.
We assessed methods as:
low risk of bias (e.g. no missing outcome data; missing
outcome data balanced across groups);
high risk of bias (e.g. numbers or reasons for missing data
imbalanced across groups; as treated analysis done with
substantial departure of intervention received from that assigned
at randomisation);
If data for more than 20% of participants were missing, we excluded the outcome or study from the analysis.
(5) Selective reporting (checking for reporting bias)
We describe for each included study how we investigated the possibility of selective outcome reporting bias and what we found.
low risk of bias (where it is clear that all of the studys prespecified outcomes and all expected outcomes of interest to the
review have been reported);
high risk of bias (where not all the studys pre-specified
outcomes have been reported; one or more reported primary
outcomes were not pre-specified; outcomes of interest are
reported incompletely and so cannot be used; study fails to
include results of a key outcome that would have been expected
to have been reported);
(6) Other sources of bias (compliance and baseline

similarity)
We describe for each included study any important concerns we

had about other possible sources of bias. For example, if the trial
stopped early, or if there was a baseline imbalance (e.g. severe blood
loss before intervention) or differential diagnosis (e.g. pattern/
syndrome differentiation for individualised treatment) between
the comparing groups.
We assessed whether each study was free of other problems that
could put it at risk of bias as follows.
Compliance
Good: more than 95% participants received the treatment

exactly following the physicians instructions.
Fair: 95% to 90%.
Poor: 90% to 80%.
Failed: less than 80%.
Baseline similarity
Yes: all participants were inpatients or outpatients, with the
symptoms, signs, examinations or diagnosis related to threatened
miscarriage and suitable for each study, then were randomly
selected for different study groups.
No.
Unclear.
(7) Overall risk of bias
We made explicit judgements about whether the studies were at

high risk of bias, according to the criteria given in the Cochrane
Handbook for Systematic Reviews of Interventions (Higgins 2011).
With reference to (1) to (6) above, we assessed the likely magnitude
and direction of the bias and whether we considered it was likely
to impact on the findings. We planned to explore the impact of
the level of bias through undertaking sensitivity analyses - see
Sensitivity analysis.
Measures of treatment effect

Statistical analysis was performed using (RevMan 2011).
Continuous data
Not applicable in this review.
Dichotomous data
We presented results as summary risk ratio with 95% confidence

intervals for dichotomous data.
Unit of analysis issues

Trials with three arms (Chinese herbal medicines alone, Western medicines alone, combined Chinese herbal and Western
medicines) were included and are described in the Characteristics
of included studies. We input the data separately for the respective
meta-analysis for a better understanding on the effectiveness of
each intervention.

Dealing with missing data

For included studies, we noted levels of attrition, and explored the
impact of included studies with high levels of missing data for the
overall assessment of treatment effect by using sensitivity analysis.
For all outcomes, we carried out analyses, on an intention-to-treat
basis; we attempted to include all participants randomised to each
group in the analyses. The denominator for each outcome in each
trial was the total number of participants randomised minus any
participants whose outcomes were known to be missing.
Assessment of heterogeneity
We assessed statistical heterogeneity in each meta-analysis using
the T, I and Chi statistics. We regarded heterogeneity as substantial if T was greater than zero and either I was greater than
30% or there was a low P value (less than 0.10) in the Chi test
for heterogeneity.
Data synthesis
We carried out statistical analysis using RevMan 2011 and used
fixed-effect inverse variance meta-analysis for combining data
when the studies were estimating the same underlying treatment
effect and the populations and methods of the trials were judged
sufficiently similar. For clinical heterogeneity sufficient to expect
that the underlying treatment effects differed between trials, or if
substantial statistical heterogeneity was detected, we used randomeffects meta-analysis to produce an overall summary if an average
treatment effect across trials was considered clinically meaningful.
We treated the random-effects summary as the average range of
possible treatment effects and we discussed the clinical implications of treatment effects differing between trials. Where the average treatment effect was not clinically meaningful, we did not
combine trials. For random-effects analyses, we presented the results as the average treatment effect with its 95% confidence interval, and the estimates of T and I.
Subgroup analysis and investigation of heterogeneity

Due to lack of data we were unable to carry out any of the prespecified subgroup analysis. In future updates of this review, if more
data become available, we will carry out the following prespecified
subgroup analyses:
1. maternal age below 35 versus 35 and above;
2. primipara versus multipara;
3. threatened miscarriage in first trimester versus second
trimester;
4. referred herbal medicines versus non-referred herbal
medicines, according to the formulary stated in the Chinese
Pharmacopeia;
5. short-term treatment (one course only) versus long-term
treatment (more than one course);
6. quasi-randomised clinical trials versus randomised clinical

trials.
In future updates of this review, we will use the following outcome
in subgroup analysis:
continuation of pregnancy after 28 weeks of gestation.
For fixed-effect meta-analyses, we will conduct planned subgroup
analyses classifying whole trials by interaction tests as described by
Deeks 2001. For random-effects meta-analyses, we will assess differences between subgroups by inspection of the subgroups confidence intervals; non-overlapping confidence intervals indicating
a statistically significant difference in treatment effect between the
subgroups.
Sensitivity analysis
It was not necessary to carry out the planned sensitivity analyses
in this review. In future updates, we will carry out the following
prespecified sensitivity analyses, where necessary. We will carry out
sensitivity analysis to explore the effect of trial quality for important outcomes in the review. Sensitivity analyses will be performed
to examine the effect of:
1. high risk of bias in the allocation of participants to groups
associated with a particular study (Schulz 1995); or
2. high levels of missing data (Higgins 2011).
RESULTS
Description of studies
See: Characteristics of included studies; Characteristics of excluded
studies; Characteristics of studies awaiting classification.
From the 169 clinical trials of Chinese herbal medicines for threatened miscarriage identified by the search, we assessed 63 potentially eligible trials for inclusion in this review. We included 44
trials, excluded 17 trials, and two trials are still awaiting classification, see Characteristics of studies awaiting classification.
Included studies
Forty-four trials (5100 women) were included for assessment and
meta-analysis in this review. All the participants were pregnant
Chinese women, and the studies were carried out in different
provinces of mainland China (middle-income country), mainly in
municipal hospitals and provincial level hospitals. For full details,
see Characteristics of included studies and Table 1 which is a quality assessment of selected randomised clinical trials.

Types of studies
All of the included studies were declared as randomised controlled

trials, although only a few of them (11 trials) did report the details
of randomisation methods.
visiting sequence, visiting date, layered method, randomised

number table and 2:1 ratio randomisation. See Characteristics of
included studies.
Blinding
Types of participants
All of the included trials recruited women with threatened miscarriages before 20 gestational weeks and who had not received any
treatment before the study commenced.
Types of interventions
Five trials compared Chinese herbal medicines with Western

medicines treatments, while 39 trials compared combined Chinese
herbal medicines and Western medicines with Western medicines
alone. We did not identify any trials comparing Chinese herbal
medicines with placebo or no treatment (including bed rest).
Two studies reported that single blinding was applied but without
further details; we considered that no blinding was applied in the
other studies. Blinding was not feasible in this kind of clinical
study, especially to the participants and clinicians.
Incomplete outcome data

No incomplete data were reported in the included studies. After
double checking the total number of participants and the patients
included in each intervention group, there were no missing data
in all the studies.
Follow-up
Selective reporting
All the included trials followed up the participants until the end
of the treatments. Four of the included trials followed up the
participants until delivery, two of which reported the development
of the infants.
All the studies clearly reported the prespecified outcomes and all
expected outcomes of interest, and no incomplete outcomes were
recorded.
Excluded studies
Other potential sources of bias
Seventeen trials were excluded because the control group included

normal pregnant women; different Western medicines in the combined medicines group and Western medicines group; Chinese
medicines versus Chinese medicines; Kampoo medicines versus
Western medicines; low quality of the methodology; unqualified
gestational age. See Characteristics of excluded studies.
All included studies followed up the patients until the end of treatment (which were prespecified in their studies), some studies continually followed up the health conditions of mothers after delivery
and the development of the newborn. Thirty-four studies reported
that there was no significant difference amongst the groups before
treatment, and were considered to have good baseline similarity.
Risk of bias in included studies
Effects of interventions
All the included trials reported that they use randomisation methods when they recruited participants for different intervention
groups, and no missing data were reported. We double checked
each included study and found that the number of participants
were consistent with the numbers reported for outcomes and follow-up. As there were no reports of drop-outs, and no evidence of
reporting biases on the number of participants, the interventions
and the outcomes in the 44 included studies, no subgroup analysis
was carried out.
We did not find any comparisons of Chinese herbal medicines

with placebo or no treatment in the 44 included studies. All included studies compared Chinese herbal medicines with active interventions (pharmaceuticals, mostly Western medicines). So only
two comparisons were included.
Chinese herbal medicines alone versus Western medicines
alone.
Combined Chinese herbal and Western medicines versus
Western medicines alone.
Allocation
All the included clinical trials reported that the participants were
randomised, allocated and grouped. However, only 11 studies reported detailed randomisation and allocation methods, including
Western medicines included tocolytic drugs (e.g. salbutamol and

magnesium sulfate), hormonal supplementations (e.g. HCG and
progesterone), immunotherapy (e.g. IgG immunisation and antiphospholipid antibodies) and supportive supplements (e.g. vitamin E and folic acid).

Chinese herbal medicines alone versus Western

medicines alone
Primary outcomes
1. Effectiveness of the intervention: continuation of

pregnancy after 28 weeks of gestation
The rate of effectiveness (continuation of pregnancy after 28 weeks
of gestation) appeared to be higher in the Chinese herbal medicines
group (90.0%) compared with the group of women receiving
Western medicines (73.3%) but this difference was not statistically significant (risk ratio (RR) 1.23; 95% confidence interval
(CI) 0.96 to 1.57) (Analysis 1.1). Only one trial (Zhong 2002)
was included in the analysis, with 60 participants, half of whom
were treated with Chinese herbal medicines while the other half
received with Western medicines.
2. Subgroup analysis
No data were available for subgroup analyses of maternal age below 35 versus 35 and above, primipara versus multipara, threatened miscarriage in first trimester versus second trimester, referred
herbal medicines versus non-referred herbal medicines, short-term
treatment (one course only) versus long-term treatment (more
than one course), or quasi-randomised clinical trials versus randomised clinical trials.
Secondary outcomes
3. No relief of clinical signs (vaginal bleeding and abdominal

pain)
The incidence of no relief of clinical signs was lower in the women
in the Chinese herbal medicines group than those in the Western
medicines group, 10% versus 26.7%, respectively, however, the
difference was not statistically significant (RR 0.38; 95% CI 0.11
to 1.28) (Analysis 1.2). Only one trial (Zhong 2002), involving
60 women, was included in this analysis.
4. Other prespecified outcomes

No data were available from the included studies for the following prespecified outcomes: no improvement in laboratory investigations; repeated threatened miscarriage; preterm labor; adverse
pregnancy outcomes; livebirth; preterm birth; stillbirth; neonatal
death; fetal structural malformations; and other adverse perinatal
outcomes.
Non-prespecified outcomes
5. Effectiveness of intervention: continuation of pregnancy

after treatment
Fifteen studies with 1807 participants assessed the outcomes immediately after the course of treatment, instead of long-term observations after 28 weeks of gestation. Therefore, as a supplement
to the primary and secondary outcomes, we have included these
clinical trials and analysed the data as non-prespecified outcomes,
i.e., continuation of pregnancy after treatment.
Chinese herbal medicine alone was, on average, significantly more
effective than Western medicines in preventing inevitable miscarriage and continuing the pregnancy, 87.7% versus 71.0%, respectively (average RR 1.22; 95% CI 1.12 to 1.32) (Analysis 1.12).
Substantial heterogeneity was observed (T = 0.02; I = 68%; Chi
2 = 43.10, df = 14 (P less than 0.0001)), so we used a randomeffects model for this analysis.
Combined Chinese herbal medicines and Western

medicines versus Western medicines alone
Primary outcomes

after 28 weeks of gestation
A combination of Chinese herbal medicines and Western
medicines was more effective than Western medicines alone to
continue the pregnancy after 28 weeks of gestation, 94.4% versus 73.6%, respectively (average RR 1.28; 95% CI 1.18 to 1.38)
(Analysis 3.1). In total, five clinical trials with 550 participants
were included (Chen 2002; Feng 1997; Lu 2011; Lv 2007; Zhong
2002).
2. Subgroup analysis
Due to a lack of detailed information, it was not possible to carry
out the planned subgroup analyses. Only the mean values and/or
ranges of maternal age in each group were reported in all the trials, however, a comparison of women below 35 years and women
above 35 years old was not possible. All the clinical trials reported
the parity of the participants but did not provide details about
the parity in each group, so further comparisons of primipara and
multipara were not possible. Data on gestational age at threatened miscarriage were only available in one study (Lu 2011), so a
comparison between first trimester and second trimester was not
possible. All of the Chinese herbal medicines and the supplements
were standard formulae as stated in Chinese Pharmacopeia, so no
subgroup analysis of referred and non-referred herbal medicines

10
was carried out. As to the treatment course, Chens study (Chen

2002) stopped the treatments around 12 weeks to five months of
gestation, Fengs (Feng 1997) and Lus (Lu 2011) studies did not
report the termination of treatment but evaluated the effectiveness
after the relief of clinical signs. Lvs study (Lv 2007) reported one
week as one course but did not give details on the total number
of the courses for the treatment. Zhongs study (Zhong 2002) reported that the treatment was stopped at two weeks after clinical
signs were relieved. Therefore, it is difficult to extract the data
and carry out subgroup analysis. There were no quasi-randomised
clinical trials in the included studies.
Secondary outcomes
3. No relief of clinical signs (vaginal bleeding and abdominal

pain)
The result showed that combined medicines treatment was more
effective in the relief of clinical signs than Western medicines treatment, the incidence rates were 5.6% versus 26.4%, respectively,
(RR 0.21; 95% CI 0.13 to 0.36) (Analysis 3.2). In total, five clinical trials with 550 participants were included (Chen 2002; Feng
1997; Lu 2011; Lv 2007; Zhong 2002).
4. No improvement in laboratory investigations (urinary and

serum 08-HCG titre)
Only one trial (Lv 2007) measured serum -HCG and progesterone levels during the treatments. However, the author did not
report the results, so we could not include any data for further
analysis.
5. Other prespecified events

No data were available from the included studies for the following
prespecified outcomes: repeated threatened miscarriage; preterm
labour; adverse pregnancy outcomes; livebirth; preterm birth; stillbirth; neonatal death; fetal structural malformations; and other
adverse perinatal outcomes.
Non-prespecified outcomes

after treatment
Combined Chinese herbal and Western medicines was significantly more effective than Western medicines alone for preventing
inevitable miscarriage (continuation of pregnancy), than Western
medicines alone 92.4% versus 72.8%, respectively (average RR
1.24; 95% CI 1.18 to 1.30;) (Analysis 3.12). In total, 26 clinical
trials (involving 2809 women) were included in the analysis. Substantial heterogeneity was observed (T = 0.00; I = 37%; Chi
= 39.91, df = 25 (P equal to 0.03)), so we used a random-effects
model in our analysis.
DISCUSSION
Effectiveness
Chinese herbal medicines claim to be effective, and are accepted as
an alternative treatment for threatened miscarriage in most Asian
countries. This review aimed to evaluate the therapeutic effects of
Chinese herbal medicines for threatened miscarriage. The most
valuable and important comparison to evaluate the intervention
is for Chinese herbal medicines to be compared with placebo.
However, no such trials were identified.
We identified five randomised clinical trials comparing Chinese
herbal medicines with Western medicines. Meta-analysis indicates
that a combination of Chinese herbal and Western medicines
are more effective than other pharmaceuticals (Chinese herbal
medicines or Western medicines alone) for treating threatened
miscarriage, to prevent inevitable miscarriage and to continue the
pregnancy beyond 28 weeks gestation. However, we have no evidence to assess whether Chinese herbal medicines alone are more
effective because no controlled trials of placebo versus no treatment (including bed rest) were identified. As a non-prespecified
outcome, we examined the 39 studies that assessed effectiveness
immediately after the course of treatment (rather than long-term
observations after 28 weeks of pregnancy). These studies indicated that Chinese herbal medicines alone were more effective than
Western medicines alone in treating threatened miscarriage. Chinese herbal medicines alone or combined with Western medicines
were more effective than Western medicines alone in relieving the
clinical signs of threatened miscarriage, including vaginal bleeding, low back pain and abdominal pain.
However, it should be emphasized that the use of Western
medicines, such as human chorionic gonadotropin (HCG) which
maintains the luteotrophic effects in supporting continued secretion of placental oestrogen, and progesterone and progesterone,
are also not supported by evidence proving them to be beneficial
(Devaseelan 2010; Haas 2008), yet they are considered as classical
therapies for threatened miscarriage (Devaseelan 2010; Wahabi
2011). Most cases will progress to the next stage no matter what is
done. Even bed rest has no significant effects in altering the course
and progress of miscarriage (Aleman 2005). So any conclusion on
the effectiveness of Chinese medicines for threatened miscarriage
is very tentative.
This review favoured Chinese herbal medicines for threatened
miscarriage. Most Chinese medicine practitioners have slightly
modified the classical prescriptions depending on the individual

11
clinical presentations. In the 44 included clinical trials, 20 trials

used a common prescription of Shou Tai Pill as a basic formula,
while the other 24 trials used other referred formulae. Some herbal
medicines have been added to, or removed from the standard formula during treatment.
Adverse effects and follow-up

Chinese herbal medicines have been commonly used to treat
threatened miscarriage in attempts to promote maternal health
and embryo-fetal development. However, confirmation of safety
claims is still pending, so it is important to study the potential
adverse effects of Chinese medicines on mothers and fetuses. Most
of the clinical trials did not report whether there were side effects
of Chinese herbal medicines during treatment or afterwards. Very
few clinical trials followed up the pregnancy until, and after birth.
So we are unable to draw any conclusions about the safety of Chinese herbal medicines for mothers and fetuses. Xiaos study (Xiao
2008), mentioned that all the participants were followed up after
one month of delivery, but did not report any further information.
Zhangs study (Zhang 2008b), reported that no complications occurred during the deliveries and that all the infants developed well,
both mentally and physically, in the half-year follow-up. Songs
study (Song 2007), reported that there were no congenital malformations of the newborns and that no adverse effects occurred
during the treatment, but did not provide any detailed data. Lis
study (Li 2006), reported detailed information and data on the
percentage of follow-up, term delivery and preterm delivery, Apgar
scores of newborns and average newborn weight. Thus, only one
in four trials was available but no further analysis could be carried
out.
Study design
With regard to the design of the clinical trials of Chinese herbal
medicines, there are still many limitations. Firstly, well-conducted
randomised controlled trials are important for meta-analysis. All
the selected trials in this review had inadequate methodological
quality, compared with foreign literature. For the five trials included for the primary outcome analysis, the authors reported
that the women were randomised in the control and intervention
groups, but none of them reported the detailed methods used for
randomisation. For the other 39 trials, only seven trials reported
adequate sequence generation, layered method (Li 2004), randomised number table (Li 2006; Wang 2011; Xiao 2008; Zhang
2008a) and 2:1 ratio randomisation (Hou 2010; Yang 2006);
while four trials reported inadequate sequence generation, visiting sequence (Chen 2003) and visiting date (Liu 2008; Song
2007; Sun 2003). None of the 44 trials reported the method of
blinding, but from the descriptions of methods reported in each
trial, we believe that all studies were open to both the doctors and
the women. In line with other reports (Wu 2007; Wu 2009; Zhuo
2008), it is not surprising that many problems in the study designs
and methodologies of clinical trials of Chinese herbal medicines
were identified. Trial quality would be greatly improved if the trialists were adequately trained to carry out and report such clinical trials according to the international standard, including sufficient details of randomisation method and having adequate allocation concealment, double-blinded participants, researchers and
outcome assessors, participants classifications, and effects assessments.
Secondly, a good clinical trial should also provide some essential
information, such as the average days or weeks of the treatments,
the changes in medicine dosage and compositions, the number
of women with a successful pregnancy until 28 weeks or afterwards, and the mortality and follow-up of newborns, which would
be helpful to examine the effects of Chinese herbal medicines
in the treatment of this condition. Thirdly, there is a potential
risk that the interventions were delivered in a way that automatically favoured the group receiving Chinese medicines. Detailed
information on the standardisation of the treatment in the control groups is lacking, so there is no way of knowing whether the
treatment periods in the two groups were similar. This leads to
concerns about whether the two groups were treated equivalently.
Finally, the small numbers of qualified clinical trials and insufficient information in this review prevented us from carrying out
planned subgroup analysis which could inform this review on the
effectiveness and safety of Chinese herbal medicines.
In summary, comparisons made between different treatments
of threatened miscarriage suggest that combined Chinese herbal
and Western medicines were more effective than Chinese herbal
medicines alone or Western medicines alone for treatment of
threatened miscarriage.
AUTHORS CONCLUSIONS
Implications for practice
No placebo controlled trials were available for comparison.
Paucity of evidence supported the effectiveness of Chinese herbal
medicines alone for the treatment of threatened miscarriage for
preventing pregnancy loss and continuing the pregnancy after 28
weeks. However, combined treatments were significantly more effective than Western medicines alone for preventing miscarriage.
Systematic information regarding the potential harm and longterm effects to the mother or child, or both, with the use of Chinese herbal medicines in the treatment of threatened miscarriage
was lacking.
Implications for research

It is both interesting and important to identify the active compounds in the Chinese herbal medicines that are beneficial in treat-

12
ing threatened miscarriage.This will involve studying pharmacokinetics as well as the mechanism and the interactions of these active
compounds with Western medicines. In addition, lack of qualified placebo controlled trials, adequate randomisation methods,
potential bias in intervention limit the conclusions of the systematic reviews in Chinese herbal medicines. We strongly recommend
standardisation of clinical trials on the efficacy of Chinese herbal
medicines during pregnancy.
ACKNOWLEDGEMENTS
We appreciate the help given by the Managing Editor (Sonja Henderson), and other associates (Denise Atherton, Frances J Kellie,
Jill Hampson, Lynn Hampson and Therese Dowswell) in the editorial office of the Cochrane Pregnancy and Childbirth Group.
As part of the pre-publication editorial process, this review has
been commented on by three peers (an editor and two referees
who are external to the editorial team), a member of the Pregnancy

and Childbirth Groups international panel of consumers and the
Groups Statistical Adviser.
The preparation of this review was partially supported by the
Health and Health Service Research Grant from Food and Health
Bureau, Hong Kong Special Administration Region (HHSRF
06070511 & 09100031) to CCW and PCL.
LL received the Hop Wai Scholarship and Zi Ying Scholarship
from the Institute of Chinese Culture and Postgraduate Student Grants for Overseas Academic Activities from the Graduate
School, The Chinese University of Hong Kong to attend Cochrane
training workshops in Oxford and Freiburg and to visit the Editorial Base of the Cochrane Pregnancy and Childbirth Group at
University of Liverpool in order to receive one to one training and
support; LL received a Yu To Sang Memorial Scholarship 2008/
2009 and 2009/2010 from The Chinese University of Hong Kong
to pursue her PhD study.
REFERENCES
References to studies included in this review

Chen 2002 {published data only}
Chen JH, Wang J, Shang L. Integrated medicines for 51
cases of threatened miscarriage. Ning Xia Medicine Journal
2002;24(8):506.
Chen LZ. Combined TCM and west medicine to treating
45 cases of threatened miscarriage. Hunan Guiding Journal
of TCMP 2003;9(3):29.
Cui 2002 {published data only}
Cui XP, Yang JB, Deng YC. Tai Er An decoction for
threatened miscarriage. Shan Xi Traditional Chinese
Medicine 2002;23(5):3878.
Deng 2009 {published data only}
Deng WH. Integrated treatment for 100 cases of threatened
miscarriage during early pregnancy. Zhe Jiang Journal of
Traditional Chinese Medicine 2009;44(2):117.
Feng 1997 {published data only}
Feng ZR. Integrated medicines for 61 cases of threatened
miscarriage. Jiang Su Journal of Traditional Chinese Medicine
1997;18(9):23.
Fu 2006 {published data only}

Fu XX. Integrated medicines for threatened miscarriage.
Forum on Traditional Chinese Medicine 2006;21(3):401.
Hou 2010 {published data only}
Hou LL, Shen WW. Chinese medicines for threatened
miscarriage. Si Chuang Journal of Chinese Medicine 2010;28
(11):967.
Hu 2010 {published data only}
Hu HJ, Xie YH. San Huang An Tai decoction for early
threatened miscarriage. Journal of Emergency Traditional
Chinese Medicine 2010;19(4):66970.
Huang 2011 {published data only}
Huang P, Zhang L, Wei YH, Ma BM. Combined Chinese
medicines and western medicines for early threatened
miscarriage. Chinese Journal of Modern Drug Application
2011;5(7):901.
Kuang 2007 {published data only}
Kuang LJ, Kuang JL. Integrated treatment for threatened
miscarriage during early pregnancy. Chinese Archives of
Feng 2010a {published data only}

Feng YQ. The effects on -HCG and CA125 of Bu Shen
An Tai method for threatened miscarriage. Shan Xi Journal
of Chinese Medicine 2010;31(7):7901.
Li 2004 {published data only}

Li WL, Li DJ, Lu Y, Zhou J, Liu CL. Bu Shen An Tai Yin
for threatened miscarriages. Clinical Journal of Traditional
Feng 2010b {published data only}

Feng YQ, Wang L. Clinical observational study of Bu Shen
An Tai method for threatened miscarriage. He Bei Journal of

Li SX, Chen HY. Integrated medicines for threatened
miscarriage. He Bei Journal of Traditional Chinese Medicine
2005;27(3):2167.

13

Li L, Zhou CX, Liu JY. Bu Shen Gu Tai Decoction with
Hermone Therapy for Threatened Miscarriage. Guang Xi
Li 2009a {published data only}
Li YM, Yang Y. Integrated medicines for threatened
miscarriages during early pregnancy. Journal of Changchun
University of Traditional Chinese Medicine 2009;25(2):
2556.
Li 2009b {published data only}
Li Z, Nong YS. Integrated medicines for 91 cases of
threatened miscarriage in early pregnancy. He Bei Journal of
Li GR, Zhang N. An Tai decoction for threatened
miscarriage. Shan Xi Journal of Chinese Medicine 2010;31
(11):14479.
with dark area surrounding pregnancy sac. Chinese Journal

of Integrated Traditional and Western Medicine 2007;27(11):
10258.
Sun 2003 {published data only}
Sun GY, Yin XH. Clinical efficacy on the sex hormone of
combination of Chinese medicine and west medicine for
the treatment of threatened miscarriage. Hunan Guiding
Journal of TCM 2003;9(11):334.
Wang 2005 {published data only}
Wang XF, OuYang XB. Intergrated medicines for threatened
miscarriages. Fujian Journal of Traditional Chinese Medicine
2005;36(1):434.
Wang YB, Li XH. Bu Shen Gu Chong decoction for 48
cases of threatened miscarriages. Liao Ning Journal of
Liu 2008 {published data only}

Liu SM. Integrated medicines for threatened miscarriages.
Family Nurse 2006;6(5c):1370.

Wang ZM, Ji QY. Integrated medicines for early threatened
miscarriage. Traditional Chinese Medicine Research 2010;23
(9):467.
Liu 2009 {published data only}

Liu F, Meng QC, Yao J, Jiang LP. Bu Shen Jian Pi formula
for 30 cases of threatened miscarriage. Journal of New

Wang F, Liu XH. Clinical efficacy of Gushen Antai pill for
early threatened abortion. Evaluation and Analysis of Druguse in Hospitals of China 2011;11(12):11213.
Liu 2011a {published data only}

Liu SL. Bao Chang decoction for early threatened
miscarriage. Guang Ming Journal of Chinese Medicine 2011;
26(10):20234.
Xiao 2008 {published data only}

Xiao ZX. Integrated medicines for 30 cases of threatened
miscarriage. Cuiding Journal of Traditional Chinese Medicine
and Pharmacy 2008;14(6):5960.
Liu 2011b {published data only}

Liu XM. Bao Tai decoction for threatened miscarriage.
Chinese Journal of Misdiagnosis 2011;11(13):3111.
Xu 2005 {published data only}

Xu H, Hu YJ. Integrated medicines for 32 cases of
threatened miscarriages, with progesterone as control. Zhe
Jiang Journal of Traditional Chinese Medicine 2005;9:383.
Lu 2011 {published data only}

Lu YJ. Chinese medicines for threatened miscarriage: 104
cases. Shan Dong Journal of Chinese Medicine 2011;30(3):
176.
Lv 2007 {published data only}
Lv XH, Han YP, Luo Y. Integrated medicines for 58 cases
of threatened miscarriages. Yun Nan Journal of Traditional
She 2008 {published data only}
She XH, Tang JQ. Integrated medicines for threatened
miscarriages. Guang Xi Journal of Traditional Chinese
Medicine 2008;31(3):43.
Shen 2010 {published data only}
Shen YQ, Wu XH, Wei JM. Combined Chinese and
western medicines for early threatened miscarriage. Jiang Su
Journal of Chinese Medicine 2010;2:11.
Xun 2008 {published data only}

Xun AH, Xia QH. Integrated medicines for threatened
miscarriages. Modern Traditional Chinese Medicine 2008;5
(16):7.
Yang 2001 {published data only}
Yang LF. Integrated medicines for threatened miscarriage
(42 cases). Jiang Xi Journal of Traditional Chinese Medicine
2001;32(2):57.
Yang 2006 {published data only}
Yang MQ. Integrated medicines for 100 cases of threatened
miscarriage in early pregnancy. Guangxi Medical Journal
2006;28(12):19845.
Zeng 2011 {published data only}
Zeng JT. Clinical observation of integrated medicines for
threatened miscarriage. Guide of Chinese Medicine 2011;9
(17):2956.
Song 2005 {published data only}

Song YL, Zhao YQ, Zhang XY. Clinical observations on
126 cases of early threatened miscarriage treated by Zhi Xue
Bao Tai Yin. Journal of Traditional Chinese Medicine 2005;
46(2):1168.
Zhang 2007 {published data only}

Zhang XF, Li X, Wang XR. Integrated medicines for
threatened miscarriages. Hu Bei Journal of Traditional
Song 2007 {published data only}

Song YL, Zhu LP. The fetus protection effects of Zhi Xue
Bao Tai decoction on women of early threatened miscarriage
Zhang 2008a {published data only}

Zhang XM. Integrated medicines for 50 cases of threatened
miscarriage in early pregnancy. Cuiding Journal of

14
Traditional Chinese Medicine and Pharmacy 2008;14(10):

389.
Zhang 2008b {published data only}
Zhang H. Integrated treatment to threatened miscarriage
and the effects on estrogen levels. Journal of Sichuan of
Zhong 2002 {published data only}
Zhong DM, Jiang HZ. Jian Pi Gu Shen therapy for 90 cases
of threatened miscarriage. Traditional Chinese Medicine
Research 2002;15(4):246.
Zhou 2010 {published data only}
Zhou BQ. An Tai decoction for threatened miscarriage.
Chinese Journal of Ethnomedicine and Ethnopharmacy 2010;
16:168.
References to studies excluded from this review

Bi 2010 {published data only}
Bi LM. Outcomes of integreated medicines for threatened
miscarriage. Chinese Community Physician 2010;17:99.
Chan 2010 {published data only}
Chen M. Integrated medicines for theatened miscarriage.
Chinese journal of ethnomedicine and ethnopharmacy 2010;
15:1968.
Gao 2011 {published data only}
Gao J, Luo SP. Clinical observation of Shoutai pills for
treatment of early threatened abortion. Journal of New
Guo 2010 {published data only}
Guo YQ, Jiao LM. Integrated medicines for early threatened
miscarriage. He Nan Journal of Traditional Chinese Medicine
2010;30(2):1767.
Hu 2010b {published data only}
Hu YF, Zhang JM. Yun Kang decoction for threatened
miscarriage. Shan Xi Journal of Chinese Medicine 2010;41
(330):49.
Li W. Integrated medicines for threatened miscarriage.
Chinese Journal of Coal Industry Medicine 2011;14(6):
8856.
Lin 2010 {published data only}
Lin GQ. An Tai decoction for early threatened miscarriage.
Family Medicine 2010;9:6656.
Lu 2007 {published data only}
Lu Z, Jin S, Chen H. Clinical observation on Chinese
integrative medicine in the treatment of threatened
abortion. China Medical Herald 2007;4(32):734.
Lu 2011b {published data only}
Lu QB, Ren QL, Haung MH, Lu Y, Ni YY, Xu JY. Clinical
evaluation of An Zi mixtures effects on treating threatened
abortion. Journal of Nan Jing Traditional Chinese Medicine
2011;27(5):4147.
Lu 2011c {published data only}
Lu QY, Shen JH. Chinese medicines for threatened
miscarriage: 30 cases. Hu Nan Journal of Traditional Chinese
Medicine 2011;27(1):63110.
Luo 2010 {published data only}

Luo LH, Gui P, Zhou AL. Shoutai pill combined with
western medicines for threatened miscarriage. Shen Zhen
Journal of Integrated Medicines 2010;20(1):512.
Qin 2010 {published data only}
Qin XR. Chinese medicines for early threatened miscarriage:
150 cases. Si Chuan Journal of Traditional Chinese Medicine
2010;28(12):8890.
Ushiroyama 2006 {published data only}
Ushiroyama T, Araki R, Sakuma K, Nosaka S, Yamashita Y,
Kamegai H. Efficacy of the kampo medicine Xiong-GuiJiao-Ai-Tang, a traditional herbal medicine, in the treatment
of threatened abortion in early pregnancy. American Journal
of Chinese Medicine 2006;34(5):73140.
Wu 2010 {published data only}
Wu JF, Zhao HY. Integrated medicines for threatened
miscarriage. China Naturopathy 2010;8(1):44.
Zhang JF, Zhang YF, Liu GZ, Feng QJ. Clinical and
experimental study on Yunan granule in treating threatened
abortion. Chinese Journal of Integrative Medicine. 2000;20
(4):2514.
Zhang TT, Huang CM, Qin BF, Su LD, Dai DY. Treatment
to 140 cases of threatened miscarriage (kidney deficiency).
Shan Dong Journal of Chinese Medicine 2006;25(4):2357.
Zhang Y. Hu tai decoction combined with western
medicines for threatened miscarriage. Shan Xi Journal of
References to studies awaiting assessment

Chen SQ. Clinical observations on integrative medicines
for 480 cases of threatened miscarriage. Hu Nan Journal of
Traditional Chinese Medicine Mar. 1999;15(2):15.
Guan 2008 {published data only}
Guan HF. Combined Chinese medicines and Western
medicines for 50 cases of threatened miscarriages. Modern
Traditional Chinese Medicine Jul. 2008;28(4):267.
Additional references
Aleman 2005
Aleman A, Althabe F, Belizn JM, Bergel E. Bed rest during
pregnancy for preventing miscarriage. Cochrane Database
of Systematic Reviews 2005, Issue 2. [DOI: 10.1002/
14651858.CD003576.pub2]
Bandyopadhyay 2010
Bandyopadhyay AR, Chatterjee D, Chatterjee M, Ghosh
JR. Maternal fetal interaction in the ABO system: A
comparative analysis of healthy mother and couples with
spontaneous abortion in Bengalee population. American
Journal of Human Biology 2011;23(1):769.

15
Batzofin 1984
Batzofin JH, Fielding WL, Friedman EA. Effect of vaginal
bleeding in early pregnancy on outcome. Obstetrics &
Gynecology 1984;63:515.
Choudhury 2010
Choudhury SR, Knapp LA. Human reproductive failure I:
immunological factors. Human Reproduction Update 2010;
7(2):11334.
Cunningham 2005
Cunningham G, Leveno KL, Bloom SL, Hauth JC, Gilstrap
LC III, Wenstrom KD. Williams Obstetrics. 22nd Edition.
New York: Mcgraw-Hill Professional Publishing, 2005.
DeCherney 2007
DeCherney AH, Nathan L. Current Diagnosis & Treatment
Obstetrics & Gynecology. 10th Edition. New York: McGrawHill Professional Publishing, 2007.
Deeks 2001
Deeks JJ, Altman DG, Bradburn MJ. Statistical methods
for examining heterogeneity and combining results from
several studies in meta-analysis. In: Egger M, Davey Smith
G, Altman DG editor(s). Systematic reviews in health care:
meta-analysis in context. London: BMJ Books, 2001.
Devaseelan 2010
Devaseelan P, Fogarty PP, Regan L. Human chorionic
gonadotrophin for threatened miscarriage. Cochrane
Database of Systematic Reviews 2010, Issue 5. [DOI:
10.1002/14651858.CD007422.pub2]
El-Zibdeh 2009
El-Zibdeh MY, Yousef LT. Dydrogesterone support in
threatened miscarriage. Maturitas 2009;65 Suppl 1:
S43S46.
Funderburk 1980
Funderburk SJ, Guthrie D, Meldrum D. Outcome of
pregnancies complicated by early vaginal bleeding. British
Journal of Obstetrics and Gynaecology 1980;87:100.
Gracia 2005
Clarisa R, Gracia. Risk factors for spontaneous abortion in
early symptomatic first-trimester pregnancies. Obstetrics &
Gynecology 2005;106(5):9939.
Haas 2008
Haas DM, Ramsey PS. Progestogen for preventing
miscarriage. Cochrane Database of Systematic Reviews 2008,
Issue 2. [DOI: 10.1002/14651858.CD003511.pub2]
Higgins 2011
Higgins JPT, Green S, editors. Cochrane Handbook for
Systematic Reviews of Interventions Version 5.1.0 [updated
March 2011]. The Cochrane Collaboration, 2011.
Available from www.cochrane-handbook.org.
Kano 2007
Kano T, Mori T, Furudono M, Ishikawa H, Watanabe H,
Kikkawa E, et al.Human leukocyte antigen may predict
outcome of primary recurrent spontaneous abortion treated
with paternal lymphocyte alloimmunization therapy.
American Journal of Reproductive Immunology 2007;58(4):

3837.
Li 2005b
Li JB. Zhong Yi Xue. 6th Edition. Peoples Health
Publishing House, 2005.
Li 2011b
Li Lu. Efficacy and safety of Chinese medicines for
pregnancy. PhD thesis, The Chinese University of Hong
Kong, Hong Kong 2011.
Liu 2002
Liu RF. Clinical observation of integrated traditional and
western medicine. Chinese Journal of Integrated Traditional
and Western Medicine 2002;22:689.
Loue 2004
Loue S, Sajatovic M. Encyclopedia of Womens Health.
Springer, 2004:4201.
Lyttleton 2004
Lyttleton J. Treatment of Infertility with Chinese Medicine.
Churchill Livingstone, 2004:315.
Ma 2006
Ma B. Zhong Yi Fu Ke Xue. Vol. 8, Shanghai: Science &
Technology Press, 2006:4.
Petrozza 2006
Petrozza JC. Early pregnancy loss. http://
emedicine.medscape.com/article/260495-overview
(accessed 2009) 2006.
RevMan 2011
The Nordic Cochrane Centre, The Cochrane Collaboration.
Review Manager (RevMan). 5.1. Copenhagen: The Nordic
Cochrane Centre, The Cochrane Collaboration, 2011.
Schulz 1995
Schulz KF, Chalmers I, Hayes R, Altman D. Empirical
evidence of bias. JAMA 1995;273:40812.
Suzumori 2010
Suzumori N, Sugiura-Ogasawara M. Genetic factors as a
cause of miscarriage. Current Medicinal Chemistry 2010;17
(29):34317.
Tabacova 1994
Tabacova S, Little RE, Balabaeva L, Pavlova S, Petrov I.
Complications of pregnancy in relation to maternal lipid
peroxides, glutathione, and exposure to metals. Reproductive
Toxicology 1994;8(3):21724.
Tien 2007
Tien JC, Tan TYT. Non-surgical interventions for
threatened and recurrent miscarriages. Singapore Medical
Journal 2007;48(12):107489.
Wahabi 2011
Wahabi HA, Abed Althagafi NF, Elawad M, Al Zeidan RA.
Progestogen for treating threatened miscarriage. Cochrane
Database of Systematic Reviews 2011, Issue 3. [DOI:
10.1002/14651858.CD005943.pub3]
Weiss 2004
Weiss JL, Malone FD, Vidaver J. Threatened abortion:
a risk factor for poor pregnancy outcome, a population-

16
based screening study. American Journal of Obstetrics and

Gynecology 2004;190:745.
Wu 2007
Wu TX, Liu GJ. What about the concepts, design, practice
and report of allocation concealment and blinding exactly?
. Chinese Journal of Evidence-Based Medicine 2007;7(3):
2037.
Wu 2009
WuTX, Li YP, Bian ZX, Liu GJ, Moher D. Randomized
trials published in some Chinese journals:how many are
randomized?. Trials 2009;10:46.
Zhuo 2008
Zhuo Q, Wu TX, Yang XZ, Zeng XX, Yuan Y. Identification
and authentication of claimed RCTs for Cochrane
systematic review. 16th Cochrane Colloquium, Freiburg.
37 October 2008.
Indicates the major publication for the study

17
CHARACTERISTICS OF STUDIES
Characteristics of included studies [ordered by study ID]

Chen 2002
Methods
Randomised controlled trial of combined medicines (Chinese herbal medicines + Western medicines) compared with Western medicines alone
Participants
84 inpatients or outpatients from Peoples Hospital of Ning Xia were recruited. Participants were all diagnosed as threatened miscarriage due to vaginal bleeding and abdominal pains
Interventions
Treatment group received Chinese herbal medicines combined with Western medicines
1) The Chinese medicine formula was Shou Tai Pill, including Chinese Dodder Seed 30
g; Himalayan Teasel Root, Chinese Taxillus Twig 20 g, and Donkey-hide Glue 15 g
2) Formula changes
Qi deficiency: Pilose Asiabell Root, Mongolian Milkcetch Root 15 g each; and
Liquorice Root 6 g were added.
Blood deficiency: Chinese Angelica, and White Paeony Root 10 g were added.
Blood heat: Baical Skullcap Root 10 g was added.
Yin deficiency: Rehmannia Root 15 g, Glossy privet fruit 10 g, and Yerbadetajo
Herb 10 were added.
Severe vomiting: tangerine peel 10 g, Villous Amomrum Fruit 10 g, and Perilla
Stem 10 g were added.
Spleen deficiency: Largehead Atractylodes Rhizome 10 g was added.
Severe bleeding: Hairyvein Agrimonia 15 g, Chinese Arborvitae Twig 12 g,
Dragon Bone 30 g, and Fortune Windmillpalm 10 g were added.
3) Decoction: po, BID until 12 weeks or 5 months of pregnancy
4) Western medicines were received at the same time, including HCG 2000 U, im, qd;
progesterone 20~40 mg, im, bid; vitamin E 100 mg, po, qd; folic acid 5 mg, po, tid;
Salbutamol sulfate 2 pills, po, tid; 25% magnesium sulfate 60 ml ivgtt
Control group was treated with Western medicines alone.
Same as above, HCG 2000U, im, qd; progesterone 20~40 mg, im, bid; vitamin E 100
mg, po, qd; folic acid 5 mg, po, tid; Salbutamol Sulfate 2 pills, po, tid; 25% magnesium
sulfate 60 mL ivgtt
Outcomes
Symptoms subsided and pregnancy maintained until delivery were considered as effective. The effectiveness rate of combined medicines group was 96.07%, and Western
medicines group was 69.69% (P < 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement
Random sequence generation (selection Unclear risk

bias)

Support for judgement

patients were randomly allocated.
18
Chen 2002
(Continued)
Allocation concealment (selection bias)
Unclear risk
Blinding (performance bias and detection High risk

bias)
All outcomes
Blinding of participants and clinicians was

not feasible.
Incomplete outcome data (attrition bias)

All outcomes
Low risk
No exclusion was reported. No losses were

reported. It was an ITT
Selective reporting (reporting bias)
Unclear risk
The protocol of the trial was not available.
Other bias
Unclear risk
No information was provided on the balance of baseline.
Chen 2003
Methods
Participants
83 inpatients from Affiliated Hospital of Heng Dong Nurse School were recruited (19992002). Participants were all diagnosed as threatened miscarriage due to vaginal bleeding
and abdominal pains
Interventions
1) The Chinese medicine formula was Tai Shan Pan Shi Yin, including Pilose Asiabell
Root 12 g, Mongolian Milkcetch Root 10 g, Largehead Atractylodes Rhizome 10 g,
Liquorice Root 5 g, Chinese Angelica 12 g, Szechuan Lovage Rhizome 3 g, White Paeony
Root 9 g, Steamed Rehmannia Root 12 g, Himalayan Teasel Root 10 g, Baical Skullcap
Root 8 g, and Villous Amomrum Fruit 3 g
2) Formula changes
Heat excess: Villous Amomrum Fruit was removed and Baical Skullcap Root was
increased to 16 g.
3) Decoction: po, bid.
4) Western medicines were received at the same time, including HCG 1000 U, im, qd
till bleeding stopped; vitamin E 20 mg, po, tid
Control group was treated with Western medicines alone. Same as above, HCG 1000U,
im, qd till bleeding stopped; vitamin E 20 mg, po, tid
Outcomes
Symptoms such as vaginal bleeding and abdominal pains subsided, and clinical examinations showed pregnancy maintained were considered as effective. The effectiveness
rate of combined medicines group was 93.33%, and Western medicines group was 76.
32% (P < 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement


19
Chen 2003
(Continued)
Random sequence generation (selection High risk

bias)
Randomised into two groups by visiting

sequence.
High risk
Randomised into two groups by visiting

sequence.

bias)
All outcomes

not feasible.

All outcomes
Low risk

reported (but the participants were allocated according to attending sequence and
there was a gap between the numbers of the
2 groups). It was an ITT
Unclear risk
Other bias
Low risk
The intervention groups were comparable,

no significant difference was found between groups
Cui 2002
Methods
Randomised controlled trial of Chinese herbal medicines compared with Western

medicines
Participants
70 inpatients or outpatients from Affiliated Hospital of Shan Xi Chinese Medicine

College were recruited (1995 March-2000 December). Participants were all diagnosed
as threatened miscarriage due to vaginal bleeding and abdominal pains
Interventions
Chinese herbal medicine group:

1) Tai Er An decoction: mainly used Chinese Taxillus Twig, Chinese Dodder Seed,
Himalayan Teasel Root, 15 g each; Cattail Pollen, Trogopterus Dung 9 g each; Danshen
Root, Villous Amomrum Fruit, Perilla Stem, Donkey-hide Glue, White Paeony Root
10 g each; and Liquorice Root 6 g
2) Decoction: po, TID.
The Western medicines group used progesterone 20 mg, im, qd; 3 days as a course,
usually 5 courses
Outcomes
Symptoms such as vaginal bleeding and abdominal pains subsided, and clinical examinations shows pregnancy maintained were considered as effective. The effectiveness rate
of Chinese medicines group was 98%, and Western medicines group was 65% (P < 0.
05)
Notes
RCT with 2 arms.
Risk of bias
20
Cui 2002
(Continued)
Bias
Authors judgement

bias)
Unclear risk
Blinding (performance bias and detection Low risk

bias)
All outcomes
Single blinding was reported in this study,

but the review authors doubt if it was a real
single-blinding, as the study author did not
report the detailed method of blinding of
participant

All outcomes
Low risk

Unclear risk
Other bias
Low risk

Deng 2009
Methods
Participants
200 participants at the Department of Obstetrcis and Gynaecology, Chinese Medicine

Hospital of Foshan (Guangzhou, China) were recruited. Participants were all diagnosed
as threatened miscarriage by HCG and ultrasound examination, and suffered with vaginal
bleeding and abdominal pains
Interventions
1) The Chinese medicine formula: included Donkey-hide Glue, Pilose Asiabell Root,
White Paeony Root, Himalayan Teasel Root 15 g each; Prepared Rhizome of Adhesive
Rehmannia, Chinese Taxillus Twig, 20 g each; Chinese Dodder Seed 30 g; and Chinese
Mugwort leaf and Liquorice Root 10 g each
2) Formula changes
Significant low back pain: Eucommia Bark 15 g was added, Chinese Taxillus Twig
was increased to 30 g.
Hard stool: Desertliving Cistanche 10 g was added.
Combined with habitual miscarriage: Lotus Seed, and Ramie Root 10 g were
added.
3) Decoction: po, qd
4) Western medicines were received at the same time, including HCG 1000U, im, qd
for 10 days then decline the dosage; vitamin E 100 mg, po, tid; folic acid 0.4 mg, qd

21
Deng 2009
(Continued)
im, qd for 10 days then decline the dosage; vitamin E 100 mg, po, tid; folic acid 0.4 mg,
qd
Outcomes
Symptoms such as vaginal bleeding and abdominal pains subsided and pregnancy maintained were considered as effective. The effectiveness rate of combined medicines group
was 95% while Western medicines group was 72% (P < 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)
the patients were randomized divided into

2 groups.

2 groups.
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk

Feng 1997
Methods
Participants
108 outpatients from Second Affiliated Hospital of Hunan University of Chinese

Medicine were recruited (1995 January-1996 Novenber). Participants were all diagnosed
Interventions
1) The Chinese medicine formula was Shou Tai Pill, included Chinese Dodder Seed 30
g, Himalayan Teasel Root 20 g, Chinese Taxillus Twig 20 g, Donkey-hide Glue 10 g;
Pilose Asiabell Root, White Paeony Root, Himalayan Teasel Root 15 g each; Prepared
Rhizome of Adhesive Rehmannia, Chinese Taxillus Twig 20 g each; Chinese Dodder
Seed 30 g; and Chinese mugwort leaf and Liquorice Root 10 g each
2) Formula changes

22
Feng 1997
(Continued)
Qi deficiency: Pilose Asiabell Root 15 g, Mongolian Milkcetch Root 15 g, and

Liquorice Root 6 g were added.
Blood deficiency: Chinese Angelica 10 g and White Paeony Root 10 g were added.
Yin deficiency: Rehmannia Root 15 g, Glossy Privet Fruit 10 g, and Yerbadetajo
Herb 10 g were added.
Increased vomiting: Tangerine Peel 10 g, Villous Amomrum Fruit 10 g, and
Perilla Stem 10 g were added.
Spleen deficiency: Largehead Atractylodes Rhizome 10 g was added.
Increased bleeding: Herb of Hairyvein Agrimonia 15 g, Chinese Arborvitae Twig
10 g, Dragon Bone 30 g were added.
3) Decoction: po, QOD.
4) Western medicines were received at the same time, including vitamin E, folic acid,
progesterone (regular dosage but not listed in the study)
Control group was treated with Western medicines alone. Same as above, vitamin E,
folic acid, progesterone (regular dosage but not listed in the study)
Outcomes
Symptoms subsided and pregnancy maintained till delivery were considered as effective.
The effectiveness rate of combined medicines group was 95.08%, and Western medicines
group was 70.21% (P < 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Unclear risk

23
Feng 2010a
Methods

medicines
Participants
202 outpatients from Chinese Medicine Hospital of Shi Jia Zhuang were recruited.
Participants were all diagnosed as threatened miscarriage due to vaginal bleeding and
abdominal pains
Interventions

1) Bu Shen An Tai decoction: mainly used Chinese Taxillus Twig, Chinese Dodder Seed,
Himalayan Teasel Root, Milkvetch Root, Pilose Asiabell Root, Eucommia Bark, 15 g
each; Medicinal Cornel Fruit, Yerbadetajo Herb, 12 g each; Donkey-hide Glue, Glossy
Privet Fruit, 10 g each; Morinda Root 9 g
2) Decoction: po, BID.
The Western medicines group used dydrogesterone 4 pills, po, once, then 1 pill, po,
TID; till 3 days after bleeding stopped
Outcomes
of Chinese medicines group was 87.5%, and Western medicines group was 77% (P < 0.
01). ?-HCG significantly increased (P < 0.01) while CA-125 significantly decreased (P
< 0.01)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)

2 groups.

2 groups.
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk


24
Feng 2010b
Methods

medicines
Participants
78 outpatients from Chinese Medicine Hospital of Shi Jia Zhuang were recruited. Participants were all diagnosed as threatened miscarriage due to vaginal bleeding and abdominal pains
Interventions

1) Bu Shen An Tai decoction: mainly used Chinese Taxillus Twig, Chinese Dodder Seed,
Himalayan Teasel Root, Milkvetch Root, Pilose Asiabell Root, Eucommia Bark, 15 g
each; Medicinal Cornel Fruit, Yerbadetajo Herb, 12 g each; Donkey-hide Glue, Glossy
Privet Fruit, 10 g each; Morinda Root 9 g
The Western medicines group used dydrogesterone 4 pills, po, once, then 1 pill, po,
TID; till 3 days after bleeding stopped
Outcomes
of Chinese medicines group was 87.5%, and Western medicines group was 76.3% (P <
0.01). CA-125 significantly decreased (P < 0.01)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)

2 groups.

2 groups.
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk


25
Fu 2006
Methods
Participants
87 inpatients from First Affiliated Hospital of Zhong Shan University were recruited
(2004 May-2006 May). Participants were all diagnosed as threatened miscarriage due to
vaginal bleeding and abdominal pains
Interventions
1) The Chinese medicine formula was Si Wu soup, included Steamed Rehmannia Root
12 g, Chinese Angelica 9 g, White Paeony Root 9 g, Szechuan Lovage Rhizome 6 g;
Donkey-hide Glue, Pilose Asiabell Root, White Paeony Root, Himalayan Teasel Root
15 g each; Prepared Rhizome of Adhesive Rehmannia, Chinese Taxillus Twig 20 g each;
Chinese Dodder Seed 30 g; and Chinese Mugwort Leaf and Liquorice Root 10 g each
2) Formula changes
Kidney deficiency: Chinese Dodder Seed 15 g, Chinese Taxillus Twig 15 g,
Himalayan Teasel Root 15 g, and Donkey-hide Glue 10 g were added; Szechuan
Lovage Rhizome was decreased (no information on the dosage).
Qi deficiency: Pilose Asiabell Root 30 g, Mongolian Milkcetch Root 15 g,
Donkey-hide Glue 10 g, and Villous Amomrum Fruit 6 g were added.
Blood deficiency: Chinese Angelica was increased to 18 g, Szechuan Lovage
Rhizome was removed, Donkey-hide Glue 15 g, Himalayan Teasel Root 15 g, and
Chinese Taxillus Twig 15 g were added.
Blood heat: Szechuan Lovage Rhizome removed, Rehmannia Root 15 g replaced
Steamed Rehmannia Root, Dwarf Lilyturf Tuber, and Cochinchinese Asparagus Root
15 g, Donkey-hide Glue 10 g, and Ramie Root 20 g were added.
3) Decoction: po, QD.
4) Western medicines were received at the same time, including progesterone 10 mg,
im, qd; HCG 2000 U, im, qd; Allylestrenol 5 mg, Ritodrine 10 mg, po, tid; vitamin C
100 mg, po, qd; folic acid 0.4 mg, tid
Control group was treated with Western medicines alone. Same as above, progesterone 10
mg, im, qd; HCG 2000U, im, qd; Allylestrenol 5mg, Ritodrine 10mg, po, tid; vitamin
C 100 mg, po, qd; folic acid 0.4 mg, tid
Outcomes
of combined medicines group was 97.87%, and Western medicines group was 87.50%
(P < 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)
Unclear risk

26
Fu 2006
(Continued)

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk

Hou 2010
Methods
Participants
75 inpatients from Womens Hospital of Nan Jing were recruited (2008 Sep-2010 Feb)
. Participants were all diagnosed as threatened miscarriage due to vaginal bleeding and
abdominal pains
Interventions
The Chinese medicine formula was Jianpi Huoxue Bushen Antai decoction, including
Milkvetch Root, Hairyvein Agrimonia Herb and Bud, 30 g each; White Paeony Root,
Ramie Root, Chinese Dodder Seed, 20 g each; Largehead Atractylodes Rhizome, Donkey-hide Glue, Chinese Taxillus Twig, 15 g each; Himalayan Teasel Root, 10 g; Largetrifoliolious Bugbane Rhizome, Chinese Angelica, Bitter Orange, Liquoric Root, 6 g each.
?
Decoction: po, dosing was not listed in the study.
Western medicines were received at the same time, including HCG and progesterone;
vitamin K and P-aminomethyl Benzoic Acid (PAMBA) if necessary. Dosage and dosing
were not listed in the study
Control group was treated with Western medicines alone. Same as above, HCG and
progesterone; vitamin K and P-aminomethyl Benzoic Acid (PAMBA) if necessary. Dosage
and dosing were not listed in the study
Outcomes
5% (P < 0.01)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

27
Hou 2010
(Continued)
Random sequence generation (selection Low risk

bias)
2:1 ratio randomization.
Low risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk

Hu 2010
Methods
Participants
68 inpatients and outpatients from Integrated Medicines Hospital of Wen Zhou were
recruited. Participants were all diagnosed as threatened miscarriage due to vaginal bleeding and abdominal pains
Interventions
The Chinese medicine formula was San Haung An Tai Yin, including Hairyvein Agrimonia Herb and Bud, White Paeony Root, 30 g each; Milkvetch Root, Ramie Root,
Chinese Taxillus Twig, Chinese Dodder Seed, Himalayan Teasel Root, Szechwon Tangshen Root, 15 g each; Honeysuckle Flower, 12 g; Largehead Atractylodes Rhizome 10
g; Dock Root, 9 g; Liquoric Root, 5 g; Figwortflower Picrorhiza Rhizome, SanChi, 3 g
each
Decoction: po, bid, 15 days as a course.
4) Western medicines were received at the same time, including HCG 40 mg, im, qd,
15 days as a course
Control group was treated with Western medicines alone. Same as above, HCG 40 mg,
im, qd. 15 days as a course
Outcomes
71% (P < 0.05). The dark area around gestational sac was significantly reduced (P < 0.
05)
Notes
RCT with 2 arms.

28
Hu 2010
(Continued)
Risk of bias
Bias
Authors judgement

bias)

2 groups.

2 groups.
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
High risk

Unclear risk
Other bias
Low risk

Huang 2011
Methods
Participants
100 inpatients or outpatients from Integrated Medicines Hospital of Gui Zhou were
recruited (2008 Jul-2010 Dec). Participants were all diagnosed as threatened miscarriage
due to vaginal bleeding and abdominal pains (Luteal Phase Defect)
Interventions
1) The Chinese medicine formula was Shou Tai Pill, including Chinese Dodder Seed 30
g; Chinese Taxillus Twig, 20 g; Himalayan Teasel Root, 15 g; Largehead Atractylodes
Rhizome, Donkey-hide Glue, 10 g each; Liquoric Root, 6 g
2) Formula changes
Abdominal pain: White Paeony Root 20 g; Eucommia Bark 15 g; Szechwon
Tangshen Root, 30 g were added.
Blood deficiency: Medicinal Cornel Fruit, 20 g was added.
Vaginal bleeding: Garden Burnet Root, 30 g was added.
3) Decoction: po, bid, 10 days as a course.
4) Western medicines were received at the same time, including HCG 2000 U, im, qd,
progesterone 20 mg, im, qd, vitamin E, 100 mg, po, qd, folic acid, 0.4 mg, po, qd
Control group was treated with Western medicines alone. Same as above, HCG 2000
U, im, qd, progesterone 20 mg, im, qd, vitamin E, 100 mg, po, qd, folic acid, 0.4 mg,
po, qd

29
Huang 2011
(Continued)
Both groups had standard care for pregnancy (prohibit sexual activity, bed rest, and
psychotherapy)
Outcomes
rate of combined medicines group was 96%, and Western medicines group was 80% (P
< 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)

2 groups.

2 groups.
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk

Kuang 2007
Methods
Randomised controlled trial of comparisons among combined medicines, Chinese herbal

medicines alone, and Western medicines alone
Participants
180 inpatients or outpatients from Second Affiliated Hospital of Hunan University of

Chinese Medicine were recruited (2004 May-2006 May). Participants were all diagnosed
Interventions

1) mainly used Pilose Asiabell Root 10 g, Mongolian Milkcetch Root, White Paeony
Root, Indian Buead and Tuber Fleeceflower Root 10 g each; Prepared rhizome of Adhesive
Rehmannia, Chinese Dodder Seed, Himalayan Teasel Root and Chinese Taxillus Twig
15 g each; and Liquorice Root 5 g
2) Formula changes

30
Kuang 2007
(Continued)
vaginal bleeding: Cuttlefish bone 30 g, Dark Plum Fruit 15 g and Tuber of

Hyacinth Bletilla 10 g were added, and prepared Himalayan Teasel Root before add
into the formula.
dry mouth and hard stool: White Paeony Root was increased to 30 g, Liquorice
Root was increased to 10 g, and Chinese Thorowax Root, Baical Skullcap Root, Dwarf
Lilyturf Tuber and Root of Lobed Kudzuvine 10 g each were added.
3) 7-day as a course, and usually took 1-3 courses. (Details of administration were not
available.)
The Western medicines group used vitamin E capsules 0.1 g, po, BID; HCG 1000 U,
im, QD, then alternative injection after vaginal bleeding stopped
All 3 groups had standard care for pregnancy (prohibit sexual activity, psychotherapy,
and regulate ultrasound tests)
Outcomes
of Chinese medicines group was 68.3%, Western medicines group was 63.3%, and combined medicine group was 91.7%. Significant difference was found between combined
medicine and Chinese medicines or Western medicines alone (P < 0.05). No statistic
difference was found between Chinese herbal medicines and Western medicines groups
(P > 0.05)
Notes
RCT with 3 arms.
Risk of bias
Bias
Authors judgement

bias)
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk

no significant difference was found among
groups

31
Li 2004
Methods
Participants
50 inpatients or outpatients from Affiliated Hospital of An Hui University of Chinese

Medicine were recruited (1999 June-2003 December). Participants were all diagnosed
Interventions
1) The Chinese medicine formula was Bu Shen An Tai Yin, included Chinese Dodder
Seed, Eucommia Bark, Chinese Taxillus Twig, Himalayan Teasel Root, Heterophylly
Falsestarwort Root, Mongolian Milkcetch Root, Largehead Atractylodes Rhizome, Baical
Skullcap Root, White Paeony Root, Steamed Rehmannia Root and Ramie Root, each
10 g
2) Formula changes (no information on the dosage)
Blood heat: Baical Skullcap Root was increased.
Kidney deficiency: Chinese Dodder Seed was increased.
3) Decoction: po, bid, 10 days as a course.
im, qd; vitamin E 100 mg, po, qd; folic acid 5 mg, tid
Control group was treated with Western medicines alone. Same as above, progesterone
10 mg, im, qd; vitamin E 100 mg, po, qd; folic acid 5 mg, tid
Outcomes
of combined medicines group was 96%, and Western medicines group was 80% (P < 0.
05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)
Layered method, but no further information was available.
Layered method, but no further information was available.
Low risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk

32
Li 2004
(Continued)
Other bias
Low risk

Li 2005
Methods
Participants
90 inpatients from First Affiliated Hospital of Guang Zhou University of Chinese

Medicine were recruited (2003 April-2004 June). Participants were all diagnosed as
threatened miscarriage due to vaginal bleeding and abdominal pains
Interventions
1) The Chinese medicine formula was Shou Tai Pill, included Chinese Dodder Seed
15 g, Chinese Taxillus Twig 15 g, Himalayan Teasel Root 15 g, Donkey-hide Glue 10
g, Pilose Asiabell Root 15 g, Mongolian Milkcetch Root 15 g, Largehead Atractylodes
Rhizome 10 g, White Paeony Root 15 g, and Liquorice Root 5 g
2) Formula changes
Increased bleeding: Baical Skullcap Root, Lotus Rhizome Node and Garden
Burnet Root, each 10 g were added.
Low back pain: Eucommia Bark and Palmleaf Raspberry Fruit, each 10 g were
added.
Stool dehydration: Desertliving Cistanche and Mulberry Fruit, each 10 g were
added.
4) Western medicines were received at the same time, including vitamin E 50 mg, po,
bid; folic acid 0.4 mg, po, qd; HCG 2000 U, im, qd; Allylestrenol, 5 mg, po, tid, if with
a history of over twice habitual miscarriage
Control group was treated with Western medicines alone. Same as above, vitamin E 50
mg, po, bid; folic acid 0.4 mg, po, qd; HCG 2000 U, im, qd; Allylestrenol, 5 mg, po,
tid, if with a history of over twice habitual miscarriage
Outcomes
of combined medicines group was 95%, and Western medicines group was 76.67% (P
< 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)
Unclear risk

33
Li 2005
(Continued)

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk

Li 2006
Methods
Participants
89 inpatients or outpatients from Affiliated Hospital of Guang Xi College of Chinese

Medicine were recruited (2003 January-2005 September). Participants were all diagnosed
Interventions
1) The Chinese medicine formula was Bu Shen Gu Tai soup, included Chinese Dodder
Seed 30 g, Barbary Wolfberry Fruit 30 g, Chinese Taxillus Twig 20 g, Himalayan Teasel
Root 20 g; Pilose Asiabell Root 30 g; Common Yam Rhizome 15 g, Eucommia Bark 20
g, White Paeony Root 20 g, and Liquorice Root 5 g
2) Formula changes
Dry mouth: Glossy privet fruit 15 g and Yerbadetajo Herb 20 g were added.
3) Decoction: po, BID, for 10 days.
4) Western medicines were received at the same time, including HCG 2000 U and
progesterone 20 mg, im, qod for 10 days
Control group was treated with Western medicines alone. Same as above, HCG 2000U
and progesterone 20 mg, im, qod for 10 days
Outcomes
of combined medicines group was 77.78%, and Western medicines group was 70.50%
(P < 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)

Randomised number table.
34
Li 2006
(Continued)
Low risk
Randomised number table.

bias)
All outcomes

not feasible.

All outcomes
Low risk
No exclusion was reported . No losses were

reported for the main outcomes, but for the
secondary outcomes of this trial the level of
losses to follow up was high and the data
for these outcomes would not be included
in the review. It was an ITT
Unclear risk
Other bias
Low risk

Li 2009a
Methods
Participants
138 inpatients or outpatients from Zhao Qing Peoples Hospital were recruited (2004
May-2006 May). Participants were all diagnosed as threatened miscarriage due to vaginal
Interventions
g, Himalayan Teasel Root 15 g, Chinese Taxillus Twig 15 g, Pilose Asiabell Root 15 g,
Donkey-hide Glue 10 g, Largehead Atractylodes Rhizome 12 g, Steamed Rehmannia
Root 12 g, Common Macrocarpium Fruit 10 g, White Paeony Root 15 g, Eucommia
Bark 15 g, and Liquorice Root 6 g
2) Formula changes
Bleeding: Fineleaf Schizonepeta Herb 10 g, Garden Burnet Root 10 g were added.
Yin deficiency: Pilose Asiabell Root replaced by Heterophylly Falsestarwort Root,
Baical Skullcap Root were added.
Qi deficiency: Pilose Asiabell Root 30 g was added.
Vomiting: Villous Amomrum Fruit, Pinellia Tuber and Bamboo Shavings were
added.
Stool dehydration: Desertliving Cistanche and Hemp Fruit were added.
3) Decoction: po, BID, 7 days as a course.
4) Western medicines were received at the same time, including HCG 2000 U, im, qod,
progesterone 20 mg, im, qd
im, qod, progesterone 20 mg, im, qd

35
Li 2009a
(Continued)
Outcomes
of combined medicines group was 93.1%, and Western medicines group was 80.3% (P
< 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk

Li 2009b
Methods
Participants
189 inpatients from Wu Yi Chinese Medicine Hospital were recruited (2005 September2007 September). Participants were all diagnosed as threatened miscarriage due to vaginal
Interventions
g, Himalayan Teasel Root 10 g, Chinese Taxillus Twig 15 g, Donkey-hide Glue10 g,
Baical Skullcap Root 6 g, Pilose Asiabell Root 15 g, Largehead Atractylodes Rhizome 12
g, White Paeony Root 15 g, and Liquorice Root 6 g
2) Formula changes
Vomiting: Bamboo Shavings 6 g, Villous Amomrum Fruit 6 g and Perilla Stem 6
g were added.
Bleeding: Ramie Root 15 g and Hairyvein Agrimonia 15 g were added.
Low back pain: Eucommia Bark 10 g and Barbary Wolfberry Fruit 15 g were

36
Li 2009b
(Continued)
added.
Stool dehydration: Hemp Fruit 15 g and Desertliving Cistanche 10 g were added.
3) Decoction: details not provided.
4) Western medicines were received at the same time, including vitamin E 50 mg, po,
bid; folic acid, 0.4 mg, po, qd; HCG 2000 IU, im, qd
Control group was treated with Western medicines alone. Same as above, vitamin E 50
mg, po, bid; folic acid, 0.4 mg, po, qd; HCG 2000 IU,im, qd
Outcomes
of combined medicines group was 91.75%, and Western medicines group was 75% (P
< 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)

(In its English abstract, it reported with
randomly divided. But in Chinese version no information regarding this was provided.)

(In its English abstract, it reported with
randomly divided. But in Chinese version no information regarding this was provided.)
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk


37
Li 2010
Methods
Participants
268 inpatients and outpatients from Chinese Medicines Hospital of An Yang were recruited. Participants were all diagnosed as threatened miscarriage due to vaginal bleeding
and abdominal pains
Interventions
1) The Chinese medicine formula included Chinese Dodder Seed, Chinese Taxillus
Twig, 15 g each; Himalayan Teasel Root, Largehead Atractylodes Rhizome, Donkeyhide Glue, Szechwon Tangshen Root, Villous Amomum Fruit, 10 g each
2) Formula changes
Abdominal pain: White Paeony Root 15~30 g; Liquoric Root 6 g were added.
Low back pain: Eucommia Bark 10 g was added.
Dry mouth: Baical Skullcap Root 15 g, Cape Jasmine Fruit 10 g were added.
3) Decoction: po, bid for 2 weeks.
4) Western medicines were received at the same time, including HCG 2000 U, im, qod,
progesterone, 20 mg, im, qod for 2 weeks
Control group was treated with Western medicines alone. Same as above, HCG 2000
U, im, qod, progesterone, 20 mg, im, qod for 2 weeks
Both groups had standard care for pregnancy (prohibit sexual activity and bed rest)
Outcomes
66% (P < 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)

2 groups.

2 groups.
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk

38
Li 2010
(Continued)
Other bias
Low risk

Liu 2008
Methods

medicines
Participants
90 outpatients from Shan Xi Service Centre for Pregnancy were recruited (2006 June2007 October). Participants were all diagnosed as threatened miscarriage due to vaginal
Interventions
Treatment group received Chinese herbal medicines.

1) The Chinese medicine formula was given according to different subtype of diagnosis
of threatened miscarriage
i. Kidney deficiency subtype: used Shou Tai Pill, included Chinese Dodder Seed 12 g,
Chinese Taxillus Twig 12 g, Himalayan Teasel Root 12 g, Donkey-hide Glue 9 g, Pilose
Asiabell Root 9 g, and Mongolian Milkcetch Root 9 g
ii. Qi deficiency subtype: used Tai Yuan Yin, included Pilose Asiabell Root 9 g, Mongolian
Milkcetch Root 9 g, Largehead Atractylodes Rhizome 9 g, Chinese Angelica 6 g, White
Paeony Root 12 g, Chinese Taxillus Twig 12 g, and Villous Amomrum Fruit 6 g
iii. Trauma subtype: used Sheng Yu Soup, included Chinese Angelica 6 g, Szechuan
Lovage Rhizome 3 g, White Paeony Root 12 g, Rehmannia Root 9 g, Pilose Asiabell
Root 9 g, and Mongolian Milkcetch Root 9 g
iv. Blood heat: used Bao Yin Jian, included Rehmannia Root 9 g, Mongolian Milkcetch
Root 12 g, Bark of Chinese Corktree 9 g, and White Paeony Root 9 g
im, and vitamin E 100 mg, po, tid
20 mg, im, and vitamin E 100 mg, po, tid
Outcomes
Symptoms such as vaginal bleeding and abdominal pains subsided, and clinical examinations shows pregnancy maintained at 12 weeks were considered as effective. The effectiveness rate of combined medicines group was 91.4%, and Western medicines group
was 75.5% (P < 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)


Randomization according to visiting date.
39
Liu 2008
(Continued)
High risk
Randomization according to visiting date.

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk

Liu 2009
Methods

medicines
Participants
45 outpatients from Shen Zhen Womens Hospital were recruited (2006 August-2007
September). Participants were all diagnosed as threatened miscarriage due to vaginal
Interventions

1) mainly used Chinese Dodder Seed, Chinese Taxillus Twig, Pilose Asiabell Root each
20 g; Himalayan Teasel Root, Donkey-hide Glue, Largehead Atractylodes Rhizome,
Herb of Hairyvein Agrimonia each 15 g; Villous Amomrum Fruit, Rhizome of East
Asian Tree Fern each 10 g, and Liquorice Root 5 g
2) Decoction: po, BID for 2 weeks.
The Western medicines group used progesterone 20 mg, im, qd; HCG 2000 U, im, qod
for 2 weeks
Both groups had standard care for pregnancy (prohibit sexual activity, psychotherapy,
Outcomes
of combined medicines group was 90%, and Western medicines group was 66.7% (P <
0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

40
Liu 2009
(Continued)

bias)
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Unclear risk
Liu 2011a
Methods

medicines
Participants
155 outpatients from Chinese Medicine Institute of Shan Xi were recruited (2008 Jun2010 Dec). Participants were all diagnosed as threatened miscarriage due to vaginal
Interventions

1) Bao Tai decoction: mainly used Szechwon Tangshen Root, Milkvetch Root, 15 g;
Chinese Dodder Seed, Chinese Taxillus Twig, White Paeony Root, 12 g each; Himalayan
Teasel Root, Donkey-hide Glue, Largehead Atractylodes Rhizome, India Mustard Seed,
10 g each
2) Decoction: po, BID for 2 weeks.
The Western medicines group used progesterone 100 mg, po, qd for 2 weeks
Outcomes
0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)


2 groups.
41
Liu 2011a
(Continued)
Unclear risk

2 groups.

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Unclear risk
Liu 2011b
Methods

medicines
Participants
100 inpatients and outpatients from Peoples Hospital of Xing Tai were recruited (2009
Jan-2012 Jan). Participants were all diagnosed as threatened miscarriage due to vaginal
Interventions

1) Bao Tai decoction: mainly used Milkvetch Root, Ramie Root, 30 g; Hairyvein Agrimonia Herb and Bud, Chinese Taxillus Twig, 24 g; Glossy Privet Fruit, 18 g; Himalayan
Teasel Root, Chinese Dodder Seed, Eucommia Bark, Giant St.Johnswort Herb, Medicinal Cornel Fruit, Fortune Windmillpalm Petiole, Gordon Euryale Seed, 15 g; Largehead
Atractylodes Rhizome, Largetrifoliolious Bugbane Rhizome , 10 g each
2) Decoction: po, BID, 7 days as a course till symptoms subsided
The Western medicines group used progesterone 20 mg, im, qd; vitamin E, 100 mg, po,
TID, 7 days as a course till symptoms subsided
Outcomes
of Chinese medicines group was 80%, and Western medicines group was 72% (P < 0.
05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)


2 groups.
42
Liu 2011b
(Continued)
Unclear risk

2 groups.

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk

Lu 2011
Methods
Participants
200 inpatients from Chinese Medicines Hospital of Ning Bo were recruited. Participants
were all diagnosed as threatened miscarriage due to vaginal bleeding and abdominal pains
Interventions
1) The Chinese medicine formula included Chinese Dodder Seed, Eucommia Bark,
Glossy Privet Fruit, Chinese Taxillus Twig, 100 g each; Medicinal Cornel Fruit, 50 g
2) Decoction: external use at Yong Quan acu-point, qd.
3) Western medicines (hormone treatment) were received at the same time, details were
not listed in this study
Control group was treated with Western medicines alone. Same as above, hormone
treatment but details were not listed in this study
Outcomes
group was 78.2% (P < 0.01)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)


2 groups.
43
Lu 2011
(Continued)
Unclear risk

2 groups.

bias)
All outcomes
Single blinding was reported in this study,

but the review authors doubt if it was a real
single-blinding, as the study author did not
report the detailed method of blinding of
participant

All outcomes
Low risk

Unclear risk
Other bias
Low risk

Lv 2007
Methods
Randomised controlled trials of combined medicines (Chinese herbal medicines + Western medicines) compared with Western medicines alone
Participants
98 inpatients from Tian Jin Chinese Medicine University were recruited (2004 May2006 May). Participants were all diagnosed as threatened miscarriage due to vaginal
Interventions
1) The Chinese medicine formula: included Chinese Dodder Seed 20 g, Himalayan
Teasel Root, Chinese Taxillus Twig, Pilose Asiabell Root, Common Yam Rhizome each 15
g, Largehead Atractylodes Rhizome 12 g, Common Macrocarpium Fruit 10 g, Steamed
Rehmannia Root 10 g, and Liquorice Root 6 g
2) Formula changes
Bleeding: Donkey-hide Glue, Baical Skullcap Root and Ramie Root were added.
Abdomen pain: White Paeony Root was added.
Blood Hhat: Baical Skullcap Root was added.
Insomnia: Spina Date Seed was added.
Low back pain: Eucommia Bark was added.
Vomiting: Villous Amomrum Fruit was added.
im, qd; vitamin E 100 mg, po, tid
20 mg, im, qd; vitamin E 100 mg, po, tid
Outcomes
group was 70% (P < 0.05)

44
Lv 2007
(Continued)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk

She 2008
Methods
Participants
100 inpatients or outpatients from Chinese Medicine Hospital of Gui Lin were recruited.
Participants were all diagnosed as threatened miscarriage due to vaginal bleeding and
abdominal pains
Interventions
g, Chinese Taxillus Twig 10 g, Himalayan Teasel Root 10 g, Donkey-hide Glue 10 g,
Pilose Asiabell Root 20 g, and Largehead Atractylodes Rhizome 15 g
2) Formula changes
Qi and blood deficiency: Mongolian Milkcetch Root 15 g, Steamed Rehmannia
Root 10 g and White Paeony Root 10 g were added.
Blood heat: Rehmannia Root 15 g, Baical Skullcap Root 12 g and Bark of
Chinese Corktree 6 g were added.
3) Decoction: po, BID, 10 days as a course.
im, qd; HCG 2000 U, im, qod; vitamin E 100 mg, po, qd; folic acid 0.4 mg, qd
20 mg, im, qd; HCG 2000U, im, qod; vitamin E 100 mg, po, qd; folic acid 0.4 mg, qd

45
She 2008
(Continued)
Outcomes
rate of combined medicines group was 93.3%, and Western medicines group was 70%
(P < 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk

Shen 2010
Methods
Participants
120 inpatients or outpatients from Chinese Medicines Hospital of Dong Tai were recruited. Participants were all diagnosed as threatened miscarriage due to vaginal bleeding
and abdominal pains
Interventions
1) The Chinese medicine formula was Shou Tai Pill and Bu Shen Jian Pi Gu Chong
decoction, including Chinese Dodder Seed, 20 g; ?Chinese Taxillus Twig, Himalayan
Teasel Root, 15 g each; Largehead Atractylodes Rhizome, Szechwon Tangshen Root,
Donkey-hide Glue, Wingde Yan Rhizome, 10 g each; Liquorice Root 6 g
2) Formula changes
Bleeding: Ramie Root 15 g, Garden Burnet Root 10 g were added.
Low back pain: Eucommia Bark, Palmleaf Raspberry Fruit, 10 g each were added.
3) Decoction: po, bid.
4) Western medicines were received at the same time, including HCG 10 mg, im, qd;

46
Shen 2010
(Continued)
vitamin E 100 mg, po, qd; 7 days as a course

Control group was treated with Western medicines alone. Same as above, HCG 10 mg,
im, qd; vitamin E 100 mg, po, qd; 7 days as a course
Both groups had standard care for pregnancy (bed rest and regulate ?-HCG and ultrasound tests)
Outcomes
< 0.01)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)

2 groups.

2 groups.
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk

Song 2005
Methods

medicines
Participants
243 outpatients from Ning Xia Womens Hospital were recruited (2001 May-2003 May)
abdominal pains
Interventions

1) Zhi Xue An Tai Yin: mainly used Himalayan Teasel Root, Chinese Dodder Seed,
Chinese Taxillus Twig, Perilla Stem, Villous Amomrum Fruit, White Paeony Root,

47
Song 2005
(Continued)
Largehead Atractylodes Rhizome, Common Macrocarpium Fruit, Sharpleaf Galangal

Fruit, Mongolian Milkcetch Root, Heterophylly Falsestarwort Root, Chinese Arborvitae
Twig, Garden Burnet Root, India Madder Root, and Donkey-hide Glue 10 g each
2) Formula changes
Heat sign: Baical Skullcap Root was added.
Cold sign: Chinese Mugwort Leaf and Ginger each 6 g were added.
Vomiting: Pinellia Tuber, Clove, Persimmon Calyx and Receptacle each 10 g, and
Ginger 6 g were added.
3) po, TID, for 7 days.
The Western medicines group used progesterone 20mg, im, qd; vitamin E 100 mg, po,
tid; vitamin K 8 mg, tid; An Luo Xue, 5 mg, po, tid
Outcomes
Symptoms such as vaginal bleeding and abdominal pains subsided, and clinical examinations showed pregnancy maintained (HCG normal and LH > 200 mLU/ml) were
considered as effective. The effectiveness rate of combined medicines group was 76.98%,
and Western medicines group was 43.59% (P < 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk

Song 2007
Methods

48
Song 2007
(Continued)
Participants
105 outpatients from Ning Xia Womens Hospital were recruited (2005 October-2006
August). Participants were all diagnosed as threatened miscarriage due to vaginal bleeding
and abdominal pains
Interventions
1) The Chinese medicine formula was Zhi Xue Bao Tai Yin, included Himalayan Teasel
Root, Chinese Dodder Seed, Chinese Taxillus Twig, Perilla Stem, Villous Amomrum
Fruit, White Paeony Root, Largehead Atractylodes Rhizome, Common Macrocarpium
Fruit, Sharpleaf Galangal Fruit, Mongolian Milkcetch Root, Heterophylly Falsestarwort
Root, Chinese Arborvitae Twig, Garden Burnet Root, India Madder Root, and Donkeyhide Glue, 10 g each
2) Formula changes
Heat sign: Baical Skullcap Root was added.
Cold sign: Chinese Mugwort Leaf and Ginger each 6 g were added.
Vomiting: Pinellia Tuber, Clove, Persimmon Calyx and Receptacle 10 g each, and
ginger 6 g were added.
3) Decoction: po, TID, for 7 days.
im, qd; vitamin E 100 mg, po, tid; vitamin K 8 mg, tid; An Luo Xue, 5 mg, po, tid
20 mg, im, qd; vitamin E 100 mg, po, tid; vitamin K 8 mg, tid; An Luo Xue, 5 mg, po,
tid
Outcomes
Symptoms such as vaginal bleeding and abdominal pains subsided, and clinical examinations showed pregnancy maintained (HCG normal and LH increased) were considered as effective. The effectiveness rate of combined medicines group was 81.5%, and
Western medicines group was 43.1% (P < 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)
Randomization according to first visit

date.
High risk
Randomization according to first visit

date.

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk

49
Song 2007
(Continued)
Other bias
Low risk

Sun 2003
Methods

Participants
105 inpatients or outpatients from Xiang Tan Chinese Medicine Hospital were recruited
Interventions

1) mainly used Heterophylly Falsestarwort Root 15 g, Mongolian Milkcetch Root 12
g, Largehead Atractylodes Rhizome 10 g, White Paeony Root 15 g, Barbary Wolfberry
Fruit 15 g, Chinese Taxillus Twig 15 g, Himalayan Teasel Root 15 g, Chinese Dodder
Seed 15 g, Common Yam Rhizome 15 g, and Liquorice Root 6 g
2) Formula changes
Severe bleeding: Garden Burnet Root 10 g and Yerbadetajo Herb 10 g were added.
Blood heat: Mongolian Milkcetch Root removed, Baical Skullcap Root 6 g were
added.
Abdomen pain: White Paeony Root 20-30 g and Largetrifoliolious Bugbane
Rhizmome 10 g were added.
Vomiting: Perilla Stem 10 g, Bamboo Shavings 10 g and Tangerine Pee 16 g were
added.
3) po, BID, 5-day as a course, and usually 1-3 courses.
The Western medicines group used vitamin E 100 mg, po, tid; folic acid 0.4 mg, po,
qd; HCG 1000 U, im, qd
Outcomes
Symptoms such as vaginal bleeding and abdominal pains subsided, and clinical examinations showed pregnancy maintained were considered as effective. The effectiveness rate of
Chinese medicines group was 94%, Western medicines group was 71%, and combined
medicine group was 98%. Significant difference was found between combined medicine
or Chinese medicines and Western medicines alone (P < 0.05). No statistic difference
was found between Chinese herbal medicines and combined medicines groups (P > 0.
05)
Notes
RCT with 3 arms.
Risk of bias
Bias
Authors judgement

bias)
Randomization according to visit date.
Randomization according to visit date.
High risk

50
Sun 2003
(Continued)

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
High risk
Wang 2005
Methods
Participants
140 outpatients from Hui An Chinese Medicine Hospital were recruited (2002 October2004 September). Participants were all diagnosed as threatened miscarriage due to vaginal
Interventions
1) The Chinese medicine formula was Zhu Ma An Tai soup, included Ramie Root 30
g, Chinese Taxillus Twig 20 g, Himalayan Teasel Root 20 g, Chinese Dodder Seed 20 g,
Largehead Atractylodes Rhizome 10 g, Baical Skullcap Root 10 g, White Paeony Root
12 g, Rehmannia Root 15 g, Yerbadetajo Herb 30 g, and Liquorice Root 3 g
2) Formula changes
Severe bleeding: Chinese Arborvitae Twig and root of Common Euscaphis were
added.
Severe vomiting: Bamboo Shavings and Villous Amomrum Fruit were added.
Stool dehydration: Desertliving Cistanche and Platycladi Seed were added.
im, qd; HCG 2000 U, im, qod; vitamin E 100 mg, po, qd; folic acid 0.4 mg, qd, (add
Zhi Xue Min if bleeding hardly)
20 mg, im, qd; HCG 2000U, im, qod; vitamin E 100 mg, po, qd; folic acid 0.4 mg, qd,
(add Zhi Xue Min if bleeding hardly)
Outcomes
< 0.05)
Notes
RCT with 2 arms.
Risk of bias

51
Wang 2005
(Continued)
Bias
Authors judgement

bias)
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk

Wang 2007
Methods

medicines
Participants
96 inpatients or outpatients from Nan Yang Chinese Medicine College were recruited
(2003 March-2006 March). Participants were all diagnosed as threatened miscarriage
due to vaginal bleeding and abdominal pains
Interventions

1) Bu Shen Gu Chong soup: mainly used Chinese Dodder Seed 30 g, Himalayan Teasel
Root 15 g, Chinese Taxillus Twig 30 g, Largehead Atractylodes Rhizome 18 g, Donkeyhide Glue 10 g, Eucommia Bark 15 g, White Paeony Root 15 g, Pilose Asiabell Root 30
g, Fineleaf Schizonepeta Herb 9 g, and Liquorice Root 6 g
2) Formula changes
QI deficiency: Pilose Asiabell Root removed, Heterophylly Falsestarwort Root
and Mongolian Milkcetch Root, each 30 g were added.
Severe bleeding: Herb of Hairyvein Agrimonia 30 g, Male Fern Rhizome 15 g and
Garden Burnet Root 30 g were added.
Vomiting: Villous Amomrum Fruit 8 g and Perilla Stem 9 g were added.
Heat sign: Baical Skullcap Root 9 g and Ramie Root 30 g were added.
Abdomen pain: White Paeony Root 30 g was added.
3) po, BID, 7-day as a course.
The Western medicines group used HCG 1000U, im, qd, 7 day as a course
Outcomes

52
Wang 2007
(Continued)
5% (P < 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Unclear risk
Wang 2010
Methods
Participants
110 outpatients from He Nan Population Institute were recruited (2007 Jan-2009 Dec)
abdominal pains
Interventions

1) Yi Shen An Tai decoction: mainly used Chinese Dodder Seed, Chinese Taxillus Twig,
15 g each; Largehead Atractylodes Rhizome, Himalayan Teasel Root, Baical Skullcap
Root, Donkey-hide Glue, Szechwon Tangshen Root, 10 g each; Liquoric Root, 6 g
2) Formula changes
Bleeding: Garden Burnet Root 12 g, Argy Wormwood Leaf 6 g were added.
Abdominal pain: White Paeony Root 15g was added.
Vomiting: Bamboo Shavings 10 g, Pinellia Tuber 6 g, Villous Amomum Fruit 6 g
were added.
Low back pain: Eucommia Bark 12 g was added.
Vexation: Lancelesf Lily Bulb 15 g was added.
3) Decoction: po, BID till the 12th week of gestation.
The Western medicines group used HCG 2000 U, im, qod, progesterone 20 mg, im,
qd, vitamin E, 100 mg, po, bid, folic acid, 0.4 mg, po, qd; till the 12th week of gestation.

53
Wang 2010
(Continued)
Outcomes
of Chinese medicines group was 81.82%, and Western medicines group was 58.18% (P
< 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)

2 groups.

2 groups.
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk

Wang 2011
Methods

medicines
Participants
225 outpatients from Xi Dian Company Hospital of Shan Xi were recruited (2010 JanDec). Participants were all diagnosed as threatened miscarriage due to vaginal bleeding
and abdominal pains
Interventions
1) The Chinese medicine formula was Gu Shen An Tai Pill, including Largehead Atractylodes Rhizome, Chinese Dodder Seed, Chinese Taxillus Twig, White Paeony Root,
Rehmannia Root, Himalayan Teasel Root, Tuber Fleeceflower Root, Baical Skullcap
Root, Gambir Plant, Desertliving Cistanche. Details of dosage were not listed in this
study
2) Decoction: po, tid, 2 weeks as a course.
3) Western medicines were received at the same time, including progesterone, 100mg,
po,qd; vitamin E 200 mg, po, tid; 2 weeks as a course

54
Wang 2011
(Continued)
Control group was treated with Western medicines alone. Same as above, progesterone,
100mg, po,qd; vitamin E 200 mg, po, tid; 2 weeks as a course
Both groups had standard care for pregnancy (prohibit sexual activity, psychotherapy,
Outcomes
18% (P < 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)
Randomized number method
Randomized number method
Low risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk

Xiao 2008
Methods
Participants
60 inpatients from Xin Shao Peoples Hospital were recruited (2005 October-2007 October). Participants were all diagnosed as threatened miscarriage due to vaginal bleeding
and abdominal pains
Interventions
1) The Chinese medicine formula was Yan Xue Yi Shen soup, included Chinese Dodder
Seed 10 g, Chinese Taxillus Twig 15 g, Himalayan Teasel Root 10 g, Eucommia Bark
15 g, Mongolian Milkcetch Root 20 g, Ginseng 10 g, Common Yam Rhizome 15 g,
White Paeony Root 15 g, Donkey-hide Glue 15 g, Steamed Rehmannia Root 10 g and

55
Xiao 2008
(Continued)
Tangerine Peel l5 g
2) Decoction: po, BID
im, qd; HCG 2000 U, im, qd for 10 days then decline the dosage; vitamin E 50 mg,
Ritodrine 10 mg, po, tid; folic acid 0.4 mg, QD, An Luo Xue, vitamin K, vitamin C if
bleeding hardly (no detailed administration)
20 mg, im, qd; HCG 2000U, im, qd for 10 days then decline the dosage; vitamin E 50
mg, Ritodrine 10 mg, po, tid; folic acid 0.4 mg, QD, An Luo Xue, vitamin K, vitamin
C if bleeding hardly (no detailed administration)
Outcomes
rate of combined medicines group was 90%, and Western medicines group was 66.7%
(P < 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)
Randomized number method.
Randomized number method.
Low risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk

Xu 2005
Methods
Participants
62 inpatients or outpatients from Affiliated Hospital of Zhe Jiang University were recruited. Participants were all diagnosed as threatened miscarriage due to vaginal bleeding
and abdominal pains

56
Xu 2005
(Continued)
Interventions
1) The Chinese medicine formula was Yun Kang decoction. (Further information of
herbs and dosages were not available.)
2) Decoction: 20 ml, po, BID, for 14 days.
im, qd; vitamin E 1 pill, po, qd
40 mg, im, qd; vitamin E 1 pill, po, qd
Outcomes
6% (P < 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Unclear risk
Xun 2008
Methods
Participants
75 inpatients or outpatients from Jiang Su East West Medicine Hospital were recruited

57
Xun 2008
(Continued)
Interventions
1) The Chinese medicine formula, included Pilose Asiabell Root 10 g, Mongolian
Milkcetch Root 10 g, Largehead Atractylodes Rhizome 10 g, White Paeony Root 10 g,
Steamed Rehmannia Root 10 g, Chinese Taxillus Twig 12 g, Donkey-hide Glue 10 g,
and Tangerine Peel 10 g
im, qd; HCG 2000 U, im, qd, qod if vaginal bleeding stopped
20 mg, im, qd; HCG 2000U, im, qd, qod if vaginal bleeding stopped
Outcomes
rate of combined medicines group was 93.3%, and Western medicines group was 80%
(P < 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Unclear risk
Yang 2001
Methods
Participants
55 inpatients or outpatients from Jiang Xi Ning Dou Chinese Medicine Hospital were
recruited. Participants were all diagnosed as threatened miscarriage due to vaginal bleeding and abdominal pains

58
Yang 2001
(Continued)
Interventions
1) The Chinese medicine formula was Shou Tai Pill, included Chinese Dodder Seed 30 g,
Chinese Taxillus Twig 12 g, Himalayan Teasel Root 12 g, Donkey-hide Glue 15 g, Pilose
Asiabell Root 30 g, Largehead Atractylodes Rhizome 10 g, Common Yam Rhizome 15
g, and Liquorice Root 6 g
im, qd; vitamin K3 4 mg, po, bid.
20 mg, im, qd; vitamin K3 4 mg, po, bid.
Outcomes
5% (P < 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Unclear risk
Yang 2006
Methods
Participants
150 inpatients or outpatients from Guang Zhou Qin Zhou Chinese Medicine Hospital
were recruited (2004 January-2006 May). Participants were all diagnosed as threatened
miscarriage due to vaginal bleeding and abdominal pains

59
Yang 2006
(Continued)
Interventions
1) The Chinese medicine formula was Jiao Ai soup, included Donkey-hide Glue 15 g,
Chinese Mugwort Leaf 12 g, Steamed Rehmannia Root 20 g, Szechuan Lovage Rhizome
5 g, White Paeony Root 10 g, Chinese Angelica 10 g, and Liquorice Root 6 g
2) Formula changes
Qi and blood deficiency: Mongolian Milkcetch Root 30 g and Pilose Asiabell
Root 20 g were added.
Kidney deficiency: Chinese Taxillus Twig 15 g, Eucommia Bark 15 g, Himalayan
Teasel Root 15 g and Chinese Dodder Seed 15 g were added.
Adominal distension: Villous Amomrum Fruit 3 g and Tangerine Peel 6 g were
added.
4) Western medicines were received at the same time, including HCG 1000 U, im, qd
for 10 days then decline the dosage; vitamin E 100 mg, po, tid; folic acid 0.4 mg, qd
im, qd for 10 days then decline the dosage; vitamin E 100 mg, po, tid; folic acid 0.4 mg,
qd
Outcomes
< 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)
Low risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Unclear risk

60
Zeng 2011
Methods
Participants
118 inpatients and outpatients from Dong Keng Hospital of Dong Wan were recruited
(2009 Jun-2011 Feb). Participants were all diagnosed as threatened miscarriage due to
Interventions
1) The Chinese medicine formula was Bu Shen Yang Xue Gu Qi decoction, including
Chinese Taxillus Twig, Eucommia Bark, White Paeony Root, Wingde Yan Rhizome,
Donkey-hide Glue, 15 g each; Chinese Dodder Seed, Steamed Rehmannia Root, Himalayan Teasel Root, 10 g each; Milkcetch Root 20 g; Tangerine Peel, 5 g
2) Decoction: po, bid till 7 days after symptoms subsided.
3) Western medicines were received at the same time, including progesterone, 2 pills,
po, bid till 7 days after symptoms subsided
2 pills, po, bid till 7 days after symptoms subsided
Outcomes
0% (P < 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)

2 groups.

2 groups.
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk


61
Zhang 2007
Methods

medicines
Participants
102 inpatients from Shan Dong Ji Nan Hospital were recruited (2004 May-2006 May)
abdominal pains
Interventions

1) Shou Tai Pill: mainly used Chinese Dodder Seed 10 g, Himalayan Teasel Root 15
g, Chinese Taxillus Twig 15 g, Donkey-hide Glue 10 g, Deer Horn 10 g, and Morinda
Root 10 g
2) Formula changes
Low back pain: Eucommia Bark was added.
Abdominal distension: Mongolian Milkcetch Root and Largetrifoliolious
Bugbane Rhizmome were added.
Bleeding: Herb of Hairyvein Agrimonia, Garden Burnet Root and Lotus Seed Pot
were added.
3) Decoction: po, qd.
The Western medicines group used vitamin E 0.1g, po, qd; progesterone 20~40 mg, im,
qd; folic acid 5 mg, po, qd; HCG, 2000 IU, im, qod
Outcomes
85% (P < 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk


62
Zhang 2008a
Methods
Participants
96 inpatients or outpatients from Liu Yang Central Hospital were recruited (2006 October-2008 January). Participants were all diagnosed as threatened miscarriage due to
Interventions
1) The Chinese medicine formula was Shou Tai Pill, included Chinese Dodder Seed
20 g, Chinese Taxillus Twig 25 g, Donkey-hide Glue, Himalayan Teasel Root, Steamed
Rehmannia Root, Pilose Asiabell Root and Largehead Atractylodes Rhizome, each 15 g;
and Liquorice Root 10 g
2) Formula changes
Qi deficiency: Mongolian Milkcetch Root 30 g was added.
Severe bleeding: Herb of Hairyvein Agrimonia 30 g, Male Fern Rhizome 15 g,
Garden Burnet Root 30 g were added.
Vomiting: Villous Amomrum Fruit 8 g and Perilla Stem 9 g were added.
Heat sign: Baical Skullcap Root 9 g and Ramie Root 30 g were added.
Abdomen pain: White Paeony Root 30 g was added.
Insomnia: Spina Date Seed 8 g was added.
3) Decoction: po, BID, 7-day as a course.
4) Western medicines were received at the same time, including HCG 2000 U, im, qod;
vitamin E 100 mg, po, qd; folic acid 2.5 mg, po, qd
im, qod; vitamin E 100 mg, po, qd; folic acid 2.5 mg, po, qd
Outcomes
1% (P < 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)
Randomized number table.
Randomized number table.
Low risk

bias)
All outcomes

not feasible.

All outcomes

Low risk

63
Zhang 2008a
(Continued)
Unclear risk
Other bias
Low risk

Zhang 2008b
Methods
Participants
100 inpatients or outpatients from Second Affiliated Hospital of Hunan University of

Chinese Medicine were recruited (2004 January-2006 December). Participants were all
diagnosed as threatened miscarriage due to vaginal bleeding and abdominal pains
Interventions
1) The Chinese medicine formula was Bao Tai Yin, included Steamed Rehmannia Root
20 g, Common Macrocarpium Fruit 15 g, Donkey-hide Glue 15 g, Himalayan Teasel
Root 15 g, White Paeony Root 10 g, Chinese Dodder Seed 15 g, Chinese Taxillus Twig
15 g, Eucommia Bark15 g, Largehead Atractylodes Rhizome 15 g, Mongolian Milkcetch
Root 20 g, Pilose Asiabell Root 10 g, and Liquorice Root 6 g
2) Formula changes
Blood heat: Bamboo Shavings, Mulberry Leaf and Towel Gourd Vegetable Sponge
each 10 g were added.
Abdominal distension: Villous Amomrum Fruit 3 g and Tangerine Pee 16 g were
added.
3) Decoction: po, tid, 1 or 2 weeks.
4) Western medicines were received at the same time, including progesterone, 20 mg,
qd; vitamin E 100 mg, po, tid
20 mg, qd; vitamin E 100 mg, po, tid
Outcomes
< 0.05)
Notes
RCT with 2 arms.
Risk of bias
Bias
Authors judgement

bias)
Unclear risk

64
Zhang 2008b
(Continued)

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk

Zhong 2002
Methods

Participants
90 inpatients or outpatients from Guang Zhou Second Peoples Hospital were recruited
Interventions

1) mainly used Pilose Asiabell Root, Largehead Atractylodes Rhizome, Common Yam
Rhizome, Chinese Taxillus Twig, Chinese Dodder Seed, Himalayan Teasel Root and
Baical Skullcap Root (no information on the dosage)
2) Formula changes (no information on the dosage)
Blood deficiency: Donkey-hide Glue was added.
Cold sign: Chinese Mugwort Leaf was added.
Blood heat: Garden Burnet Root was added.
Abdomen pain: White Paeony Root, Nutgrass Galingale Rhizome and Perilla
Stem were added.
Vomiting: Bamboo Shavings, Villous Amomrum Fruit and Tangerine Peel were
added.
Dry mouth: Rehmannia Root, Glossy Privet Fruit and Yerbadetajo Herb were
added.
The Western medicines group used HCG 2000 U, im, qod, then decline the dosage
after 12 weeks of pregnancy
Outcomes
The effectiveness rate of Chinese medicines group was 90%, Western medicines group
was 73.3%, and combined medicine group was 93.3%. Significant difference was found
between Chinese herbal medicines and Western medicines alone (P < 0.05). No statistic
difference was found between Chinese herbal medicines and combined medicines (P >
0.05)
Notes
RCT with 2 arms.

65
Zhong 2002
(Continued)
Risk of bias
Bias
Authors judgement

bias)
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk

Zhou 2010
Methods

medicines
Participants
87 outpatients from Central Hospital of Jia Ling were recruited (2005 May-2010 May)
abdominal pains
Interventions

1) An Tai Yin: mainly used White Paeony Root 30 g; Chinese Dodder Seed, Eucommia
Bark, Steamed Rehmannia Root, 25 g each; Chinese Taxillus Twig, Szechwon Tangshen Root, Largehead Atractylodes Rhizome, Donkey-hide Glue, Chinese Angelica, Himalayan Teasel Root, Liquorice Root, 15 g each
2) Decoction: po, TID till 7 days after vaginal bleeding stopped
The Western medicines group used progesterone 20 mg, im, qd; vitamin E, 100 mg, po,
qd; folic acid, 5 mg, po, tid; till 7 days after vaginal bleeding stopped
Outcomes
0.05)
Notes
RCT with 2 arms.
Risk of bias
66
Zhou 2010
(Continued)
Bias
Authors judgement

bias)

2 groups.

2 groups.
Unclear risk

bias)
All outcomes

not feasible.

All outcomes
Low risk

Unclear risk
Other bias
Low risk

Po (per os): by mouth; Im: (intramuscular); iv: (intravascular); ivgtt: (intravenously guttae, or IVdrop ) referred to different administration methods; QOD: (every other day); QD: (once per day); BID: (twice per day); TID: (3 times per day); PIN: (taken when
necessary) referred to different dosing; HCG: human chorionic gonadotropin; ITT (intention-to-treat); RCT: randomised controlled
trial.
Characteristics of excluded studies [ordered by study ID]
Study
Reason for exclusion
Bi 2010
Comparisons were made between patients with threatened miscarriage and normal pregnant women. No outcomes were reported on the effectiveness of treatments. So the review authors decided to exclude this trial
Chan 2010
This study included some participants with longer than 20 weeks of gestation, which did not meet our inclusion
criteria. So the review authors decided to exclude this paper
Gao 2011
This study did not mention the randomisation methods. Also, the number of participants in each group could
indicate randomisation was not applied to the allocation and intervention. So the review authors decided to
exclude this paper
Guo 2010
Comparison was between combined medicine and Western medicines. However, the women in 2 groups received
different Western medicines. So the review authors considered that these 2 groups were not comparable and
doubt if the trial author could reach his conclusion that combined medicine was better than Western medicines.
We decided to exclude this paper

67
(Continued)
Hu 2010b
Li 2011
Lin 2010
This study did not mention the randomisation methods. And the number of participants in each group could
exclude this paper
Lu 2007
Lu 2011b
exclude this paper
Lu 2011c
Both treatment group and control group received a second Chinese medicine intervention, so the review authors
considered that these 2 groups were not comparable and doubt if the trial author could reach his conclusion that
the test Chinese medicine was effective. We decided to exclude this paper
Luo 2010
Qin 2010
Comparison was between one Chinese medicine and another Chinese medicine. The review authors considered
that the trial author could not reach his conclusion on the benefits of Chinese medicines as treatments. So this
paper was excluded
Ushiroyama 2006
Comparisons were made between a Kampo medicine and Western medicines. Kampo medicine is Japanese
traditional medicine, which may originate from China. To avoid the confusion of readers on the topic of our
review, which is about Chinese herbal medicines, we decided to exclude this trial
Wu 2010
exclude this paper
Zhang 2000
This study included some participants with longer than 20 weeks of gestation, which did not meet our inclusion
criteria. Furthermore, the amount of participants in each group differed greatly. There were 580 participants in 1
Chinese medicine group, but 50 participants in another Chinese medicine group and 50 participants in Western
medicines group. Therefore, the review authors considered the participants were not randomly allocated, and

68
(Continued)
decision was made to exclude this paper

Zhang 2006
Data were collected from 1995 to 2005, using 10 years. So the review authors doubt if this is a real RCT.
Also, the number of participants in each group differed. There were 140 participants in the combined medicine
group, but 128 participants in Western medicines group, and the author did not give further information on
the randomisation method. Thus, we decided to exclude this paper
Zhang 2011
RCT: randomised controlled trial
Characteristics of studies awaiting assessment [ordered by study ID]

Chen 1999
Methods
RCT of Chinese herbal medicines combined with Western medicines compared with Western medicines alone
Participants
720 participants diagnosed with threatened miscarriage were 2:1 divided into 2 groups. (480 for combined medicine
group, and 240 for Western medicines group)
Interventions
Comparison between combined medicines and Western medicines was made
Outcomes
The effectiveness of combined medicine is higher than Western medicines. Chinese herbal medicine could help and
improve the treatment of Western medicines alone
Notes
Dates of data collection started from 1991, and the result was published in 1999, without mentioning the duration
that the 720 patients were involved in the treatments. So the review authors doubt if this was a real RCT, and have
been awaiting response from authors for further information
Guan 2008
Methods
RCT of Chinese herbal medicines combined with Western medicines compared with Western medicines alone
Participants
100 participants diagnosed with threatened miscarriage were divided into 2 groups
Interventions
Comparison between combined medicines and Western medicines was made
Outcomes
The effectiveness of combined medicines is higher than Western medicines. Chinese herbal medicines could help
and improve the treatment of Western medicines alone

69
Guan 2008
Notes
(Continued)
Dates of data collection were from March 1999 to March 2006, using 7 years. So the review authors doubt if this
was a real RCT, and have been awaiting response from authors for further information
RCT: randomised controlled trial

70
DATA AND ANALYSES
Comparison 1. Chinese herbal medicines versus Western medicines (primary outcome and non-specific secondary
outcome)
Outcome or subgroup title

1 Effectiveness of intervention
(primary outcome)
2 No relief of clinical signs
3 No improvement of laboratory
investigations
4 Repeated threatened miscarriage
5 Preterm labour
6 Adverse pregnancy outcomes
7 Preterm birth
8 Stillbirth
9 Neonatal death
10 Fetal structural malformations
11 Adverse neonatal outcomes
(non-prespecified secondary
outcome)
No. of
studies
No. of
participants
60
Risk Ratio (M-H, Random, 95% CI)
1.23 [0.96, 1.57]
1
0
60
0

Risk Ratio (M-H, Fixed, 95% CI)
0.38 [0.11, 1.28]

0.0 [0.0, 0.0]
0
0
0
0
0
0
0
0
15
0
0
0
0
0
0
0
0
1807

0.0 [0.0, 0.0]

0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
1.22 [1.12, 1.32]
Statistical method
Effect size
Comparison 2. Chinese herbal medicines versus Western medicines (subgroup analysis)

(maternal age)
1.1 Maternal age below 35
1.2 35 and above
(parity)
2.1 Primipara
2.2 Multipara
(gestational stage)
3.1 First trimester
3.2 Second trimester
(reference)
4.1 Referred herbal medicines
4.2 Non-referred herbal
medicines
No. of
studies
No. of
participants
0.0 [0.0, 0.0]
0
0
0
0
0
0

0.0 [0.0, 0.0]

0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
0
0
0
0
0
0

0.0 [0.0, 0.0]

0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
0
0
0
0
0
0

0.0 [0.0, 0.0]

0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
0
0
0
0

0.0 [0.0, 0.0]

0.0 [0.0, 0.0]
Statistical method

Effect size
71
(treatment course)
5.1 Short-term treatment
5.2 Long-term treatment
(randomisation)
6.1 Quasi-RCTs
6.2 RCTs
Comparison 3.
outcome)
0.0 [0.0, 0.0]
0
0
0
0
0
0

0.0 [0.0, 0.0]

0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
0
0
0
0

0.0 [0.0, 0.0]

0.0 [0.0, 0.0]
Combined medicines versus Western medicines (primary outcome and non-specific secondary

(primary outcome)
2 No relief of clinical signs
3 No improvement of laboratory
investigations
4 Repeated threatened miscarriage
5 Preterm labour
6 Adverse pregnancy outcomes
7 Preterm birth
8 Stillbirth
9 Neonatal death
10 Fetal structural malformations
11 Adverse neonatal outcomes
(non-prespecified secondary
outcome)
No. of
studies
No. of
participants
550
1.28 [1.18, 1.38]
5
0
550
0

0.21 [0.13, 0.36]

0.0 [0.0, 0.0]
0
0
0
0
0
0
0
0
26
0
0
0
0
0
0
0
0
2809

0.0 [0.0, 0.0]

0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
1.24 [1.18, 1.30]
Statistical method
Effect size
Comparison 4. Combined medicines versus Western medicines (subgroup analysis)

(maternal age)
1.1 Maternal age below 35
1.2 Versus 35 and above
(parity)
2.1 Primipara
2.2 Multipara
(gestational stage)
No. of
studies
No. of
participants
0.0 [0.0, 0.0]
0
0
0
0
0
0

0.0 [0.0, 0.0]

0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
0
0
0
0
0
0

0.0 [0.0, 0.0]

0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
Statistical method

Effect size
72
3.1 First trimester

3.2 Second trimester
(reference)
4.1 Referred herbal medicines
4.2 Non-referred herbal
medicines
(treatment course)
5.1 Short-term treatment
5.2 Long-term treatment
(randomisation)
6.1 Quasi-RCTs
6.2 RCTs
0
0
0
0
0
0

0.0 [0.0, 0.0]

0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
0
0
0
0

0.0 [0.0, 0.0]

0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
0
0
0
0
0
0

0.0 [0.0, 0.0]

0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
0
0
0
0

0.0 [0.0, 0.0]

0.0 [0.0, 0.0]
Analysis 1.1. Comparison 1 Chinese herbal medicines versus Western medicines (primary outcome and
non-specific secondary outcome), Outcome 1 Effectiveness of intervention (primary outcome).
Review:
Comparison: 1 Chinese herbal medicines versus Western medicines (primary outcome and non-specific secondary outcome)
Outcome: 1 Effectiveness of intervention (primary outcome)
Study or subgroup
Zhong 2002
Total (95% CI)
Chinese
herbal
medicines
Western medicines
Risk Ratio
MH,Random,95%
CI
Weight
Risk Ratio
MH,Random,95%
CI
n/N
n/N
27/30
22/30
100.0 %
1.23 [ 0.96, 1.57 ]
30
30
100.0 %
1.23 [ 0.96, 1.57 ]
Total events: 27 (Chinese herbal medicines), 22 (Western medicines)

Heterogeneity: not applicable
Test for overall effect: Z = 1.63 (P = 0.10)
Test for subgroup differences: Not applicable
0.01
0.1
Favours Western medicines

10
100
Favours Chinese herbal
73
non-specific secondary outcome), Outcome 2 No relief of clinical signs.
Review:
Outcome: 2 No relief of clinical signs
Study or subgroup
Zhong 2002
Total (95% CI)
Chinese
herbal
medicines
Western medicines
n/N
n/N
3/30
8/30
100.0 %
0.38 [ 0.11, 1.28 ]
30
30
100.0 %
0.38 [ 0.11, 1.28 ]
Risk Ratio
MH,Random,95%
CI
Weight
Risk Ratio
MH,Random,95%
CI

Heterogeneity: not applicable
Test for overall effect: Z = 1.57 (P = 0.12)
0.01
0.1
Favours experimental

10
100
Favours control
74
non-specific secondary outcome), Outcome 12 Effectiveness of intervention (non-prespecified secondary
outcome).
Review:
Outcome: 12 Effectiveness of intervention (non-prespecified secondary outcome)
Study or subgroup
Chinese
herbal
medicines
Western medicines
Risk Ratio
MH,Random,95%
CI
Weight
Risk Ratio
MH,Random,95%
CI
n/N
n/N
49/50
13/20
4.1 %
1.51 [ 1.09, 2.08 ]
Feng 2010a
89/102
77/100
8.8 %
1.13 [ 0.99, 1.29 ]
Feng 2010b
35/40
29/38
6.4 %
1.15 [ 0.93, 1.42 ]
Kuang 2007
41/60
38/60
5.3 %
1.08 [ 0.83, 1.40 ]
Liu 2008
53/58
37/49
7.3 %
1.21 [ 1.01, 1.45 ]
Liu 2009
27/30
10/15
3.3 %
1.35 [ 0.93, 1.97 ]
Liu 2011a
72/77
65/78
9.2 %
1.12 [ 1.00, 1.26 ]
Liu 2011b
40/50
36/50
6.2 %
1.11 [ 0.89, 1.39 ]
Song 2005
97/126
51/117
6.0 %
1.77 [ 1.41, 2.22 ]
Song 2007
44/54
22/51
3.8 %
1.89 [ 1.34, 2.65 ]
Sun 2003
34/36
25/36
5.9 %
1.36 [ 1.08, 1.71 ]
Wang 2007
46/48
43/48
9.3 %
1.07 [ 0.96, 1.20 ]
Wang 2011
107/115
97/110
10.1 %
1.06 [ 0.97, 1.15 ]
Zhang 2007
67/69
27/33
7.7 %
1.19 [ 1.01, 1.40 ]
Zhou 2010
41/45
31/42
6.7 %
1.23 [ 1.01, 1.51 ]
960
847
100.0 %
1.22 [ 1.12, 1.32 ]
Cui 2002
Total (95% CI)

Heterogeneity: Tau?? = 0.02; Chi?? = 43.10, df = 14 (P = 0.00008); I?? =68%
Test for overall effect: Z = 4.74 (P < 0.00001)
0.02
0.1

10
50
Favours control
75
Analysis 3.1. Comparison 3 Combined medicines versus Western medicines (primary outcome and nonspecific secondary outcome), Outcome 1 Effectiveness of intervention (primary outcome).
Review:
Comparison: 3 Combined medicines versus Western medicines (primary outcome and non-specific secondary outcome)
Outcome: 1 Effectiveness of intervention (primary outcome)
Combined
Chinese/Western
Western medicines
n/N
n/N
Chen 2002
49/51
23/33
11.6 %
1.38 [ 1.09, 1.74 ]
Feng 1997
58/61
33/47
16.4 %
1.35 [ 1.11, 1.65 ]
Lu 2011
99/104
75/96
47.7 %
1.22 [ 1.09, 1.37 ]
Lv 2007
53/58
28/40
13.1 %
1.31 [ 1.05, 1.62 ]
Zhong 2002
28/30
22/30
11.2 %
1.27 [ 1.01, 1.61 ]
304
246
100.0 %
1.28 [ 1.18, 1.38 ]
Study or subgroup
Total (95% CI)
Risk Ratio
MH,Random,95%
CI
Weight
Risk Ratio
MH,Random,95%
CI
Total events: 287 (Combined Chinese/Western), 181 (Western medicines)

Heterogeneity: Tau?? = 0.0; Chi?? = 1.51, df = 4 (P = 0.83); I?? =0.0%
0.01
0.1
Favours Western medicines

10
100
Favours Combined
76
Analysis 3.2. Comparison 3 Combined medicines versus Western medicines (primary outcome and nonspecific secondary outcome), Outcome 2 No relief of clinical signs.
Review:
Outcome: 2 No relief of clinical signs
Study or subgroup
Combined medicines
Western medicines
Risk Ratio
MH,Random,95%
CI
Weight
Risk Ratio
MH,Random,95%
CI
n/N
n/N
Chen 2002
2/51
10/33
12.2 %
0.13 [ 0.03, 0.55 ]
Feng 1997
3/61
14/47
18.3 %
0.17 [ 0.05, 0.54 ]
Lu 2011
5/104
21/96
29.5 %
0.22 [ 0.09, 0.56 ]
Lv 2007
5/58
12/40
27.9 %
0.29 [ 0.11, 0.75 ]
Zhong 2002
2/30
8/30
12.0 %
0.25 [ 0.06, 1.08 ]
304
246
100.0 %
0.21 [ 0.13, 0.36 ]
Total (95% CI)
Total events: 17 (Combined medicines), 65 (Western medicines)

Heterogeneity: Tau?? = 0.0; Chi?? = 1.06, df = 4 (P = 0.90); I?? =0.0%
0.01
0.1
Favours Combined

10
100
Favours Western
77
Analysis 3.12. Comparison 3 Combined medicines versus Western medicines (primary outcome and nonspecific secondary outcome), Outcome 12 Effectiveness of intervention (non-prespecified secondary
outcome).
Review:
Outcome: 12 Effectiveness of intervention (non-prespecified secondary outcome)
Study or subgroup
Combined medicines
Western medicines
Risk Ratio
MH,Random,95%
CI
Weight
Risk Ratio
MH,Random,95%
CI
n/N
n/N
Chen 2003
39/45
24/38
2.3 %
1.37 [ 1.05, 1.79 ]
Deng 2009
95/100
72/100
5.8 %
1.32 [ 1.16, 1.50 ]
Fu 2006
46/47
35/40
6.1 %
1.12 [ 0.99, 1.27 ]
Hou 2010
48/51
15/24
1.7 %
1.51 [ 1.10, 2.07 ]
Hu 2010
30/34
22/34
2.2 %
1.36 [ 1.03, 1.80 ]
Huang 2011
48/50
40/50
5.0 %
1.20 [ 1.03, 1.39 ]
Kuang 2007
55/60
38/60
3.4 %
1.45 [ 1.18, 1.78 ]
Li 2004
24/25
20/25
3.3 %
1.20 [ 0.97, 1.48 ]
Li 2005
57/60
23/30
3.4 %
1.24 [ 1.01, 1.52 ]
Li 2006
35/45
31/44
2.6 %
1.10 [ 0.86, 1.41 ]
Li 2009a
67/72
53/66
5.6 %
1.16 [ 1.01, 1.33 ]
Li 2009b
91/97
69/92
5.9 %
1.25 [ 1.10, 1.42 ]
129/140
93/128
6.4 %
1.27 [ 1.13, 1.43 ]
She 2008
56/60
29/40
3.5 %
1.29 [ 1.05, 1.58 ]
Shen 2010
53/58
45/56
5.0 %
1.14 [ 0.98, 1.32 ]
Sun 2003
35/36
25/36
3.0 %
1.40 [ 1.12, 1.75 ]
Wang 2005
76/80
30/60
2.4 %
1.90 [ 1.47, 2.46 ]
Wang 2010
45/55
32/55
2.5 %
1.41 [ 1.09, 1.82 ]
Xiao 2008
27/30
20/30
2.1 %
1.35 [ 1.02, 1.79 ]
Xu 2005
29/32
26/30
4.1 %
1.05 [ 0.87, 1.25 ]
Xun 2008
42/45
24/30
3.7 %
1.17 [ 0.96, 1.42 ]
Yang 2001
38/42
8/13
1.0 %
1.47 [ 0.95, 2.28 ]
Yang 2006
95/100
36/50
4.1 %
1.32 [ 1.10, 1.58 ]
Zeng 2011
65/68
41/50
5.4 %
1.17 [ 1.01, 1.34 ]
Li 2010
0.01
0.1
10
100
Favours control
(Continued . . . )

78
(. . .
Study or subgroup
Combined medicines
Western medicines
Risk Ratio
MH,Random,95%
CI
Continued)
Risk Ratio
MH,Random,95%
CI
Weight
n/N
n/N
Zhang 2008a
45/50
40/46
5.2 %
1.04 [ 0.90, 1.20 ]
Zhang 2008b
46/50
39/50
4.4 %
1.18 [ 1.00, 1.40 ]
Total (95% CI)
1532
1277
100.0 %
1.24 [ 1.18, 1.30 ]
Total events: 1416 (Combined medicines), 930 (Western medicines)

Heterogeneity: Tau?? = 0.00; Chi?? = 39.91, df = 25 (P = 0.03); I?? =37%
0.01
0.1
10
100
Favours control
ADDITIONAL TABLES
Table 1. Quality assessment of selected randomised clinical trials
Studies
Randomisation
Blindingb
Follow-upc
Follow-up
duration
Complianced
Baseline
Similaritye
Chen 2002
Unclear
Open
Adequate
Good
Adequate
Unclear
Chen 2003
Inadequate
Open
Adequate
Good
Adequate
Yes
Cui 2002
Unclear
single-blinded
Adequate
Good
Adequate
Yes
Deng 2009
Unclear
Open
Adequate
Good
Adequate
Yes
Feng 1997
Unclear
Open
Adequate
Good
Adequate
Unclear
Feng 2010a
Unclear
Open
Adequate
Good
Adequate
Yes
Feng 2010b
Unclear
Open
Adequate
Good
Adequate
Yes
Fu 2006
Unclear
Open
Adequate
Good
Adequate
Yes
Hou 2010
Adequate
Open
Adequate
Good
Adequate
Yes
Hu 2010
Unclear
Open
Adequate
Good
Adequate
Yes
Huang 2011
Unclear
Open
Adequate
Good
Adequate
Yes
Kuang 2007
Unclear
Open
Adequate
Good
Adequate
Yes

79
(Continued)
Li 2004
Adequate
Open
Adequate
Good
Adequate
Yes
Li 2005
Unclear
Open
Adequate
Good
Adequate
Yes
Li 2006
Adequate
Open
Adequate
Good
Adequate
Yes
Li 2009a
Unclear
Open
Adequate
Good
Adequate
Yes
Li 2009b
Unclear
Open
Adequate
Good
Adequate
Yes
Li 2010
Unclear
Open
Adequate
Good
Adequate
Yes
Liu 2008
Inadequate
Open
Adequate
Good
Adequate
Yes
Liu 2009
Unclear
Open
Adequate
Good
Adequate
Unclear
Liu 2011a
Unclear
Open
Adequate
Good
Adequate
Unclear
Liu 2011b
Unclear
Open
Adequate
Good
Adequate
Yes
Lu 2011
Unclear
single-blinded
Adequate
Good
Adequate
Yes
Lv 2007
Unclear
Open
Adequate
Good
Adequate
Yes
She 2008
Unclear
Open
Adequate
Good
Adequate
Yes
Shen 2010
Unclear
Open
Adequate
Good
Adequate
Yes
Song 2005
Unclear
Open
Adequate
Good
Adequate
Yes
Song 2007
Inadequate
Open
Adequate
Good
Adequate
Yes
Sun 2003
Inadequate
Open
Adequate
Good
Adequate
Unclear
Wang 2005
Unclear
Open
Adequate
Good
Adequate
Yes
Wang 2007
Unclear
Open
Adequate
Good
Adequate
Unclear
Wang 2010
Unclear
Open
Adequate
Good
Adequate
Yes
Wang 2011
Adequate
Open
Adequate
Good
Adequate
Yes
Xiao 2008
Adequate
Open
Adequate
Good
Adequate
Yes
Xu 2005
Unclear
Open
Adequate
Good
Adequate
Unclear

80
(Continued)
Xun 2008
Unclear
Open
Adequate
Good
Adequate
Unclear
Yang 2001
Unclear
Open
Adequate
Good
Adequate
Unclear
Yang 2006
Adequate
Open
Adequate
Good
Adequate
Unclear
Zeng 2011
Unclear
Open
Adequate
Good
Adequate
Yes
Zhang 2007
Unclear
Open
Adequate
Good
Adequate
Yes
Zhang 2008a
Adequate
Open
Adequate
Good
Adequate
Yes
Zhang 2008b
Unclear
Open
Adequate
Good
Adequate
Yes
Zhong 2002
Unclear
Open
Adequate
Good
Adequate
Yes
Zhou 2010
Unclear
Open
Adequate
Good
Adequate
Yes
a adequate:
clearly by computer, envelope or telephone; uncertain: reported randomisation but without any approach and methods;
inadequate: no randomisation
b double-blinded, single-blinded, open or unclear with blinding of participants, caregivers and administrating treatment and outcome
assessors.
c adequate: < 5% loss; fair: 5%-10% loss; poor: 10%-20% loss and excluded from this review; inadequate: > 20% loss and exclude from
this review; unclear: not reported.
d good: > 95% participants received the treatment exactly following the physicians instructions; fair: 95%-90%; poor: 90%-80%; failed,
< 80%.
e good: no significant difference (P > 0.05) between the participants in intervention groups and control group; unclear: not reported.
NA: not applicable.
APPENDICES
Appendix 1. Search strategy for EMBASE
1. exp PREGNANCY/
2. (spontaneous adj2 abortion*).af
3. (threat* adj3 (pregnancy ADJ loss)).af
4. (abortion* adj3 threat*).af
5. (spontaneous adj3 (pregnancy ADJ loss)).af
6. miscarriage*.af
7. exp CHINESE HERB/
8. (chin* adj6 herb*).af
9. ((china OR chinese) AND (tradition* adj4 medicine*)).af
10. 2 OR 3 OR 4 OR 5 OR 6
11. 7 OR 8 OR 9
81
12. 1 AND 10 AND 11
Appendix 2. Search strategy for CINAHL

1. exp PREGNANCY/
2. (spontaneous adj2 abortion*).af
3. (threat* adj3 (pregnancy ADJ loss)).af
4. (abortion* adj3 threat*).af
5. (spontaneous adj3 (pregnancy ADJ loss)).af
6. miscarriage*.af
7. (chin* adj6 herb*).af
8. ((china OR chinese) AND (tradition* adj4 medicine*)).af
9. DRUGS, CHINESE HERBAL/
10. 2 OR 3 OR 4 OR 5 OR 6
11. 7 OR 8 OR 9
12. 1 AND 10 AND 11
Appendix 3. Search strategy for PUBMED

1. ((Chin Med[Journal] OR (chinese[All Fields] AND medicine[All Fields]) OR chinese medicine[All Fields]) OR (medicine,
traditional[MeSH Terms] OR (medicine[All Fields] AND traditional[All Fields]) OR traditional medicine[All Fields] OR
(traditional[All Fields] AND medicine[All Fields]))) OR (Trends Cardiovasc Med[Journal] OR Case Manager[Journal] OR
tcm[All Fields])
2. (((therapy[Subheading] OR therapy[All Fields] OR treatment[All Fields] OR therapeutics[MeSH Terms] OR therapeutics[All Fields]) OR (therapy[Subheading] OR therapy[All Fields] OR therapeutics[MeSH Terms] OR therapeutics[All Fields]))
OR clinical[All Fields]) OR application[All Fields]
3. 1 AND 2
4. (abortion, spontaneous[MeSH Terms] OR (abortion[All Fields] AND spontaneous[All Fields]) OR spontaneous abortion[All
Fields] OR miscarriage[All Fields]) OR (abortion, induced[MeSH Terms] OR (abortion[All Fields] AND induced[All Fields])
OR induced abortion[All Fields] OR abortion[All Fields])
5. threatened/abrupt[All Fields] OR threatened/actual[All Fields] OR threatened/actually[All Fields] OR threatened/endangered[All
Fields] OR threatened/exposed[All Fields] OR threatened/forced[All Fields] OR threatened/injured[All Fields] OR threatened/involved[All Fields] OR threatened[All Fields]
6. (therapeutics[MeSH Terms] OR therapeutics[All Fields] OR therapeutic[All Fields]) OR (recurrence[MeSH Terms] OR
recurrence[All Fields] OR recurrent[All Fields]) OR complete[All Fields] OR incomplete[All Fields] OR missed[All Fields] OR
inevitable[All Fields]
7. 4 AND 5 NOT 6
8. 3 AND 7
Appendix 4. Search strategy for CNKI and CJN (Chinese)

1.
(subject =miscarriage) OR (subject =abortion)
2.
(subject= threatened)
3. (subject =therapeutic) OR (subject = recurrent) OR (subject = recurrent) OR (subject = complete) OR (subject= incomplete) OR
(subject= inevitable) OR (subject = missed)
4.
1 AND 2 NOT 3
5.
(subject= Chinese medicine*(therapy application+ clinical use)) OR (subject= traditional medicine) OR (subject=TCM)
6.
4 AND 5

82
Appendix 5. Search strategy for WanFang Database (Chinese)

1.
2.
3.
4.
5.
6.
7.
TCM OR (traditional medicine) OR (Chinese medicine)

application OR (clinical use) OR therapy
1 AND 2
miscarriage OR abortion
(threatened abortion) OR (threatened miscarriage)
4 AND 5
3 AND 6
Appendix 6. Search Strategies for CBM and Wiley InterScience

Searched by subject heading/keyword/abstract with:
Traditional Chinese Medicines
threatened miscarriage treatment
western medicines
comparisons studies
randomized controlled trials
meta-analysis
Or could be included or replaced by similar words:

herbal medicines
pharmaceuticals
miscarriage
spontaneous abortion
therapy
Medical Subject Headings (MeSH)

Miscarriage, Abortion, Threatened [*drug therapy]; Traditional Chinese Medicines, Herbal medicines, [*therapeutic use]; Randomised
Controlled Trials [*methods/topic]; Meta analysis [*method/topic]
Appendix 7. Search Strategy for ICTRP and Chinese Clinical Trials Registry
Keywords of Threatened, Threatened miscarriage, Threatened abortion, abortion, Chinese medicine, herbal medicine were
searched in the title list of registered clinical trials in the databases.
Appendix 8. List of handsearched journals

The following journals were specifically searched for this review.
1. Acta Chinese Medicine and Pharmacology (1980 to Jan 2012)
2. Beijing Journal of Traditional Chinese Medicine (1980 to Jan 2012)
3. Central Plains Medical Journal (1980 to Jan 2012)
4. China Medical Herald (1980 to Jan 2012)
5. Chinas Naturopathy (1980 to Jan 2012)
6. Chinese Archives of Traditional Chinese Medicine (1980 to Jan 2012)
7. Chinese Journal of Information On Tcm (1980 to Jan 2012)
8. Chinese Journal of Medicine (1980 to Jan 2012)
9. Chinese Journal of Obstetrics and Gynecology (1953 to Jan 2012)
10. Chinese Journal of Perinatal Medicine (1999 to Jan 2012)
11. Chinese Journal of Practical Gynecology and Obstetrics (1986 to Jan 2012)
12. Chinese Medicine and Materia Medica (1980 to Jan 2012)
13. Clinical Journal of Anhui Traditional Chinese Medicine (1980 to Jan 2012)
14. Clinical Journal of Traditional Chinese Medicine (1980 to Jan 2012)
83
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
Forum On Traditional Chinese Medicine (1980 to Jan 2012)

Gansu Journal of Traditional Chinese Medicine (1980 to Jan 2012)
Guizhou Medical Journal (1980 to Jan 2012)
Hebei Journal of Traditional Chinese Medicine (1980 to Jan 2012)
Heilongjiang Medicine and Pharmacy (1980 to Jan 2012)
Henan Medical Information (1980 to Jan 2012)
Henan Traditional Chinese Medicine (1980 to Jan 2012)
Hubei Journal of Traditional Chinese Medicine (1980 to Jan 2012)
Hunan Guiding Journal of Traditional Chinese Medicine and Pharmacology (1980 to Jan 2012)
Hunan Journal of Traditional Chinese Medicine (1980 to Jan 2012)
Jiangsu Journal of Traditional Chinese Medicine (1980 to Jan 2012)
Jiangxi Journal of Traditional Chinese Medicine (1980 to Jan 2012)
Jounal of Hubei College of Traditional Chinese Medicine (1980 to Jan 2012)
Jouranl of Guangzhou University of Traditional Chinese Medicine (1980 to Jan 2012)
Journal of Anhui Traditional Chinese Medical College (1980 to Jan 2012)
Journal of Changzhi Medical College (1980 to Jan 2012)
Journal of Chinese Medicinal Materials (1980 to Jan 2012)
Journal of Chinese Rural Physician (1980 to Jan 2012)
Journal of Guangzhou University of Traditional Chinese Medicine (1980 to Jan 2012)
Journal of Guiyang College of Traditional Chinese Medicine (1980 to Jan 2012)
Journal of Handan Medical College (1980 to Jan 2012)
Journal of Jinzhou Medical College (1980 to Jan 2012)
Journal of Nanjing University of Traditional Chinese Medicine (1980 to Jan 2012)
Journal of New Chinese Medicine (1980 to Jan 2012)
Journals of Practical Obstetrics and Gynecology (1986 to Jan 2012)
Journal of Practical Traditional Chinese Medicine (1980 to Jan 2012)
Journal of Tianjin College of Traditional Chinese Medicine (1980 to Jan 2012)
Journal of Traditional Chinese Medicine (1980 to Jan 2012)
Journal of Traditional Chinese Medicine and Chinese Materia Medica of Jilin (1980 to Jan 2012)
Journal of Youjiang Medical College For Nationalities (1980 to Jan 2012)
Liaoning Journal of Traditional Chinese Medicine (1980 to Jan 2012)
Maternal and Child Health Care of China (1980 to Jan 2012)
Modern Journal of Integrated Traditional Chinese and Western medicine (1980 to Jan 2012)
Modern Traditional Chinese Medicine (1980 to Jan 2012)
Nei Mongol Journal of Traditional Chinese Medicine (1980 to Jan 2012)
New Journal of Traditional Chinese medicine (1971 to Jan 2012)
Ningxia Medical Journal (1980 to Jan 2012)
Practical Clinical Medicine (1980 to Jan 2012)
Primary Journal of Chinese Materia Medica (1980 to Jan 2012)
Progress in Obstetrics and Gynecology (1994 to Jan 2012)
Qinghai Medical Journal (1980 to Jan 2012)
Shaanxi Journal of Traditional Chinese Medicine (1980 to Jan 2012)
Shanghai Journal of Traditional Chinese Medicine (1980 to Jan 2012)
Shanxi Journal of Traditional Chinese Medicine (1980 to Jan 2012)
Sichuan Journal of Traditional Chinese Medicine (1980 to Jan 2012)
Sponsored By Gubei College of Traditional Chinese Medicine (1980 to Jan 2012)
The Practical Journal of Integrating Chinese With Modern Medicine (1980 to Jan 2012)
Tianjin Journal of Traditional Chinese Medicine (1980 to Jan 2012)
Traditional Chinese Medicinal Research (1980 to Jan 2012)
Zhejiang Journal of Traditional Chinese Medicine (1964 to Jan 2012)

84
HISTORY
Protocol first published: Issue 5, 2010
Review first published: Issue 5, 2012
CONTRIBUTIONS OF AUTHORS
Dr Li Lu and Prof Wang Chi Chiu both wrote the initial and final versions of the review. Dr Dou Li Xia and Prof Leung Ping Chung
commented on the final version of the review.
DECLARATIONS OF INTEREST
None known.
SOURCES OF SUPPORT
Internal sources
Department of Obstetrics and Gynaecology, The Chinese University of Hong Kong, Hong Kong.
Institute of Chinese Medicine, The Chinese University of Hong Kong, Hong Kong.
Hop Wai Scholarship 2009. Institute of Chinese Culture,The Chinese University of Hong Kong, Hong Kong.
to support attending Cochrane training and workshops in United Kingdom (27 July-23 August 2009)
Zi Ying Scholarship 2010. Institute of Chinese Culture,The Chinese University of Hong Kong, Hong Kong.
to support attending Cochrane workshop in Germany (18-20 Mar 2010)
CUHK Postgraduate Student Grants for Oversea Academic Activities. The Chinese University of Hong Kong, Hong Kong.
to support the study trips in United Kingdom and Germany.
External sources
Health and Health Services Research Fund (HHSRF) from Food and Health Bureau, Hong Kong Special Administration
Region, Hong Kong.
DIFFERENCES BETWEEN PROTOCOL AND REVIEW

In the protocol, we planned to include studies with only abstracts available. After working on the review, we found that there are two
papers with abstract only and related to our topic, however, no detailed information is provided for our further inclusion for metaanalysis study. Therefore, we excluded these trials and all the included papers had full texts. Additionally, we included 44 trials at first
and after having carefully read all the included studies and input data to the extraction forms, we found that 39 of the studies concluded
the outcomes immediately after the courses of treatment, instead of long-term observations until term delivery. Therefore, we still listed
the characteristics of these trials but reported the related data as non-prespecified secondary outcome (continuation of pregnancy after
treatment) to supplement the primary outcome. We have added livebirth as a secondary outcome.

85
INDEX TERMS
Medical Subject Headings (MeSH)
Abortion, Threatened [ drug therapy]; Drugs, Chinese Herbal [ therapeutic use]; Phytotherapy [ methods]; Randomized Controlled
Trials as Topic
MeSH check words

Female; Humans; Pregnancy

86

Chinese Herbal Medicines For Threatened Miscarriage

Enviado por

Dados do documento

Título original

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Chinese Herbal Medicines For Threatened Miscarriage

Enviado por

Direitos autorais:

Formatos disponíveis

Chinese herbal medicines for threatened miscarriage (Review)

Li L, Dou L, Leung PC, Wang CC