Online Submissions: http://www.journaltcm.
com J Tradit Chin Med 2016 October 15; 36(5): 555-563
info@journaltcm.com
Review of systematic reviews and Meta-analyses investigating
Traditional Chinese Medicine treatment for type 2 diabetes mellitus
Liu Meijun, Liu Zhicheng, Xu Bin, Zhang Wei, Cai Jianwei
aa
Liu Meijun, Zhang Wei, Sichuan Suining City First People's
Hospital, Suining 629000, China; the Second Clinical Medi-
cal College of Nanjing University of Chinese Medicine, Nan-
jing 210023, China
Liu Zhicheng, The Second Clinical Medical College of Nan-
jing University of Chinese Medicine, Nanjing 210023, China
Xu Bin, Key Laboratory of Integrated Acupuncture and
Drugs (Nanjing University of Chinese Medicine), Ministry of
Education, Nanjing 210023, China
Cai Jianwei, The Second Clinical Medical College of Nanjing
University of Chinese Medicine, Nanjing 210023, China
Supported by the National Natural Science Foundation of
China (No. 30873307, 81373749, 81574071), and the People
Programme (Marie Curie Actions) of the European Union's
Seventh Framework Programme under REA grant agree-
ment (No. PIRSES-GA-2013-612589)
Correspondence to: Prof. Xu Bin, Key Laboratory of Inte-
grated Acupuncture and Drugs (Nanjing University of Chi-
nese Medicine), Ministry of Education, Nanjing 210023, Chi-
naxuuuux@sina. com
Telephone: +86-25-86798095
Accepted: September 27, 2015
Abstract
OBJECTIVE: To conduct a review of the systematic
reviews and meta-analyses on the Traditional Chi-
nese Medicine (TCM) treatment of type 2 diabetes
mellitus (T2DM).
METHODS: We searched PubMed, Cochrane, Web
of Science, Chinese Biomedical Literature Database
(CBM), China National Knowledge Infrastructure
(CNKI), China Science and Technology Journal Data-
base (VIP), Wanfang, and other databases from da-
tabase inception to May 2014 for systematic re-
views and meta-analyses on TCM treatment of
T2DM. Manuscripts were read by two investigators
JTCM | www. journaltcm. com
ISSN 0255-2922
© 2016 JTCM. All rights reserved.
SYSTEMATIC REVIEW
if they met the inclusion criteria, and data were ex-
tracted. A Measurement Tool to Assess Systematic
Reviews (AMSTAR) was used to classify research
quality, and the evidence quality was graded by
the Grade of Recommendation, Assessment, Devel-
opment, and Evaluation (GRADE) system.
RESULTS: Eighteen systematic reviews and me-
ta-analyses were considered. Fifteen analyzed the
efficacy of Chinese herbal medicines, and three in-
vestigated the efficacy of acupuncture. AMSTAR
evaluation ranged from 3-10, and re-evaluation of
the main results implied that treatment of T2DM
with TCM has certain advantages when compared
with conventional Western medicine. However, the
evidence quality was generally low.
CONCLUSION: This work shows favorable evidence
for the clinical treatment of TCM on T2DM. Howev-
er, it is recommended that evidence-based deci-
sions are made based on clinical trials because of
the GRADE scores of the studies. To achieve high-
er quality of clinical research, clinical research on
TCM requires specific and suitable research meth-
ods. Further trials may increase the quality of evi-
dence to evaluate the clinical efficacy of TCM for
T2DM.
© 2016 JTCM. All rights reserved.
Key words: Medicine, Chinese traditional; Diabetes
Mellitus, type 2; Review; Meta-analysis
INTRODUCTION
Type 2 diabetes mellitus (T2DM) is a common disease
555 October 15, 2016 | Volume 36 | Issue 5 |
 Liu MJ et al. / Systematic Review
with a complex etiology. World Health Organization
reported that in 2000 there were 151 million people
suffering from diabetes worldwide, and that an estimat-
ed 300 million people will suffer from diabetes in
2025.1,2 The current pace of development of diabetes
has far exceeded expectations. The prevalence of diabe-
tes and prediabetes among Chinese adults is 9.7% and
15.5% , respectively,3,4 up from less than 1% in 1980,
2.5% in 1994, and 5.5% in 2000. Diabetes is a risk fac-
tor for cardiovascular disease and stroke, which caused
29 million deaths by 2010.5,6 T2DM can cause a vari-
ety of complications if blood glucose levels are not con-
trolled, and results in the use of medications, high out-
patient and inpatient costs, and requires expensive diag-
nostic tests. Furthermore, absenteeism from recurrent
symptoms can cause indirect economic losses. Modern
treatment for T2DM is mainly via Western Medicine.
However, severe side-effects of medications make com-
plementary and alternative therapies more appealing
among patients and clinicians. These complementary
treatments include herbs, acupuncture, and other Tradi-
tional Chinese Medicine (TCM).7
A systematic review re-evaluates collected systematic
reviews on the treatment, diagnosis, and other aspects
of specific disease or health problems, and can provide
a higher quality of evidence for readers.8 Numerous sys-
tematic reviews and meta-analyses on TCM manage-
ment of T2DM have been published. However, the
methodological quality and limitations of individual
studies has resulted in a lack of quality evidence. There-
fore, this study aims to re-evaluate the systematic re-
views and meta-analyses, and classify their quality of
evidence using the Grade of Recommendation, As-
sessment, Development, and Evaluation (GRADE)
system.
MATERIALS AND METHODS
Search strategy
PubMed, Cochrane, Web of Science, Chinese Biomedi-
cal Literature Database (CBM), China National
Knowledge Infrastructure (CNKI), China Science and
Technology Journal Database (VIP), and Wanfang da-
tabases were searched on computer from database in-
ception to May 2014. The grey literature was also
searched using the CNKI and Wanfang doctor/master
thesis database. Different combinations of key words
and random words (Cochrane Library [MeSH],
PubMed [MeSH], CBM [MeSH]) were chosen based
on different databases. Meanwhile, the references at-
tached to the article or related reviews were tracked,
and interventions were excluded from the retrieval
strategy. Search terms included: type 2 diabetes melli-
tus, systematic reviews, and meta-analysis. The search
strategy was presented as follows, taking the Cochrane
Library as an example.
#1 Type 2 diabetes mellitus
JTCM | www. journaltcm. com
#2 Type 2 diabetes mellitus [MeSH Terms]
#3 T2DM
#4 Meta-Analysis
#5 Meta-Analysis [MeSH Terms]
#6 Meta-Analysis [Publication Type]
#7 Systematic review
#8 #1 OR #2 OR #3
#9 #4 OR #5 OR #6 OR #7
#10 #8 AND #9
Inclusion criteria
(a) Systematic Review or meta-analysis on TCM treat-
ment for T2DM as a primary intervention with at least
one randomized controlled trial (RCT) either in Eng-
lish or Chinese. (b) The original study objects were pa-
tients with T2DM or in accordance with the relevant
domestic or overseas diagnostic criteria, not limited by
sex and ethnicity. (c) The interventions were mainly
TCM therapy, including herbal medicine (e. g. herb
compounds, single herb, Chinese patent medicine, or
herbal extracts), acupuncture (e. g. acupuncture, moxi-
bustion, acupoint injection, acupoint application, or
auricular acupuncture), and massage. (d) The latest or
the most detailed edition was chosen for repeatedly
published literature. (e) The grey literature like confer-
ence papers meeting the above criteria was considered.
Exclusion criteria
(a) Comparative studies on different Chinese medicine
therapies (e. g. Chinese herbal medicine vs acupunc-
ture or herb compounds vs Chinese patent medicine);
(b) quality assessment and methodological studies for
systematic reviews; and (c) abstracts or comments for
conference papers.
Literature screening and review data extraction
Two individual investigators read all titles and ab-
stracts, and research that was obviously in accordance
with the exclusion criteria was eliminated, while those
initially meeting the inclusion criteria were read in full.
Discrepancies were resolved by discussion or involve-
ment of the third investigator. Information extracted
from the review data included: title, author, source, ob-
ject of study, inclusion criteria, exclusion criteria, treat-
ment group interventions, control group interventions,
course of treatment, quality assessment methods, index-
es for therapeutic effects, and adverse events. After
completing the data extraction form, crosschecking
was executed between the two investigators.
Assessment of literature quality and evidence quality
The methodological quality of included systematic re-
views or meta-analyses was evaluated by A Measure-
ment Tool to Assess Systematic Reviews (AMSTAR).
The AMSTAR tables include 11 items, each with com-
ment options of "YES," "NO," "NOT CLEAR," or
"NOT USED." Each item with the answer "YES" is re-
corded as 1 point, while answers to the rest are record-
ed as 0 points, for a potential total of 11.9 The quality
556 October 15, 2016 | Volume 36 | Issue 5 |
 Liu MJ et al. / Systematic Review

classification by AMSTAR scores is as follows: 0-4 is
low, 4-8 is medium, and 9-11 is high. The evidence
quality of the systematic reviews was assessed by
GRADE. Evidence classification is as follows: (a) high
quality (very confident that the real effects are close to
those estimated; (b) medium quality (medium confi-
dence that the real effects are close to those estimated);
(c) low quality (limited confidence that the real effects
are close to those estimated); and (d) very low quality
(almost no confidence that the real effects compare
with those estimated).
RESULTS
Literature search results
Eighty papers were searched after preliminary retrieval.
Ten repeated studies were excluded by literature man-
agement software (EndNote X6), while five did not
meet the inclusion criteria. Therefore, 65 papers were
initially included. After reading the full text of each pa-
per, the following were excluded: 15 systematic re-
views, three repeated conference papers/journal arti-
cles, three papers on literature quality assessment, five
papers on the systematic reviews or meta-analyses
whose interventions did not meet the exclusion crite-
ria, and one research comment. Finally, 18 papers were
included.10-27 There were six manuscripts published in
English,22-27 including one Cochrane review,26 and 12
Chinese manuscripts10-21 published from 2003 to 2013.
The literature screening procedure is shown in Figure 1.
Basic characteristics and quality assessment of
included literature
There were 15 systematic reviews and meta-analyses10,
12-20, 22, 23, 25-27
on the efficacy and safety of Chinese herbal
treatment for T2DM, among the 18 papers. The herb-
al medicines used included: (a) Chinese patent medi-
cines, including Liuweidihuang pills; (b) herbal ex-
tracts like berberine and green tea extracts; and (c)
point injection formulations (e. g. astragalus injection,
puerarin injection, and Shengmai injection). The re-
maining three papers11,21,24 evaluated the efficacy of acu-
puncture therapy on T2DM. Two papers11,21 assessed
the efficacy of point injection, and the final paper24
studied the efficacy of acupuncture. The methodologi-
cal quality of the studies was evaluated by AMSTAR,
with total points ranging from 3 to 10. Six papers were

Obtain relative literature by searching

the databases (n = 80)
PubMed (n = 5), Cochrane (n = 10), Obtain relative literature by other
Web of Science (n = 6), CNKI (n = 21), sources
VIP (n = 11), Wanfang (n = 9), (n = 0)
CBM (n = 18)

Exclude the repeated

literature
(n = 65)

Preliminary screening by
reading the tiles and
abstracts (n = 65)
Re-screening by reading
the full paper
(n = 35)
Incorporate qualitative
synthesized literature
(n = 18)
Incorporate quantitative
synthesized (Meta analysis)
literature (n = 18)
Exclude (n = 30)
Non-research literature like journals items
(n = 6), column introduction (n = 5),
submission instructions (n = 8),
conference notices (n = 9), news reports
Not in accord with the inclusive criteria
(n = 2)
Exclude (n = 17)
Systematic review (n = 5)
Interventions not in accord with
the inclusive criteria (n = 5)
Comment paper about the
literature quality (n = 3)
Repeated conference paper and
journal published paper (n = 3)
Research comment (n = 1)

Figure 1 Literature screening procedure and results

CBM: Chinese Biomedical Literature Database; CNKI: China National Knowledge Infrastructure; VIP: China Science and Technology
Journal Database.

JTCM | www. journaltcm. com 557 October 15, 2016 | Volume 36 | Issue 5 |
 Liu MJ et al. / Systematic Review
classified as high quality and five as medium quality. In-
formation on the included studies is presented in Table 1.
Results of re-evaluation
Ten papers12-16,19-20,22-23,26 evaluated the efficacy and safety
of Chinese patent medicine treatment for T2DM.
There was one26 Cochrane systematic review. The re-
sults suggested some clinical efficacy of Chinese patent
medicine when compared with conventional Western
medicine or placebo treatment. However, the GRADE
classification ranged from low quality to very low quali-
ty, because of the low quality of eligible research. Three
papers14,15,22 assessed the efficacy of Liuweidihuang pills
in the treatment of T2DM compared with convention-
al Western medicine. The AMSTAR scores were 4, 4,
and 9, respectively. Meta-analysis showed superior effi-
cacy of Liuweidihuang pills for T2DM compared with
conventional Western Medicine. However, limitations
in the original research resulted in a GRADE classifica-
tion of low quality.
Three papers10,25,27 evaluated the clinical efficacy of herb-
al extracts (berberine and the green tea extracts) for
T2DM. The research to evaluate the clinical efficacy of
berberine was based on randomized double-blind place-
bo-controlled trials. The results showed that herbal
medicines have some clinical efficacy when compared
with placebo treatment. Despite the high methodologi-
cal quality, factors like small sample size and publica-
tion bias resulted in a GRADE classification of medi-
um quality. One systematic review27 included seven
clinical trials on using green tea extracts for T2DM.
Meta-analysis of the trials found that green tea extracts
were not significantly different in overall clinical effica-
cy and other indicators in patients when compared
with the placebo treatment.
Two systematic reviews11,21 analyzed the efficacy of as-
tragalus injection, puerarin injection, and Shengmai in-
jection treatment for T2DM2. Meta-analysis showed
that astragalus injection was clinically effective in
T2DM patients [RR = 1. 58, 95% CI (1. 20, 2. 09)]
when compared with the control group, and point in-
jection of puerarin was significantly better than the
control group [RR = 1. 28, 95% CI (1. 11, 1. 47)].
However, the GRADE classification was very low quali-
ty because of limitations, wide confidence intervals,
small sample size, and publication bias.
One systematic review on the treatment of T2DM by
acupunctures published by Tae-Hun Kim24 studied the
efficacy and safety of acupunctures therapy vs conven-
tional Western Medicine interventions. The meta-anal-
ysis suggested that acupuncture improved clinical effi-
cacy compared with conventional Western Medicine.
No severe side-effects were observed. However, the bias
risk for the original study resulted in a GRADE classifi-
cation of low quality.
DISCUSSION
This study found that there are certain clinical effects
JTCM | www. journaltcm. com
that TCM treatment can exert on T2DM, and there
are few adverse reactions, such as hypoglycemia, when
compared with Western drugs alone. Furthermore,
treatment by Chinese medicine plus Western Medicine
was superior to that of Western Medicine alone.
T2DM is categorized as "Xiao Ke" in TCM, and was
first mentioned in the Chinese book Su Wen Qi Bing
Lun.28 An alternative name for "Xiao Ke" is described
in Huang Di Nei Jing29 as "Xiao Dan, " "Fei Xiao," or
"Ge Xiao" based on different pathogeneses and symp-
toms. TCM theorizes that the internal organs weak-
ness, excessive intake of fat and sugar, and mood disor-
ders are the main causes of "Xiao Ke" The pathogenesis
is based on body fluid loss dryness-heat, according to
TCM. Furthermore, a lack of Yin is its intrinsic factor,
while dryness-heat is its extrinsic factor. The most com-
monly evaluated drug among the Chinese patent drugs
was Liuweidihuang pills, which is composed of pre-
pared prepared Dihuang (Radix Rehmanniae), Shan-
zhuyu (Fructus Macrocarpii), Shanyao (Rhizoma Di-
oscoreae Oppositae), Zexie (Rhizoma Alismatis), Mudan-
pi (Cortex Moutan Radicis), Fuling（Poria）. The herbs
can help to improve liver and kidney function. In dia-
betes patients treated with Liuweidihuang pills for
12 weeks, fasting blood glucose and HbA1c concentra-
tions decreased, blood NO levels increased, endothelin
(ET) contents decreased, and superoxide dismutase
and glutathione peroxidase increased significantly the
treatment group than the control group patients.com-
pared with before intervention.30 These results suggest
that the use of Liuweidihuang pills in the early stages
of diabetes can help to reduce blood glucose levels, im-
prove oxidative stress, and recover NO-ET levels.
Therefore, it can protect injured endothelial cells and
prevent the subsequent macrovascular complications of
diabetes. We found that TCM has advantages in the
treatment of T2DM when compared with the clinical-
ly used drugs. However, the original research method-
ological limitations resulted in low GRADE classifica-
tions for quality of evidence and low credibility of the
assessment results. The relevant herbal extracts or herb-
al medicine compounds have attracted the attention of
global researchers because of developments in the for-
mulation and preparation, and the controllability of
clinical studies. Among them was berberine, which can
control blood glucose and improve diabetes symptoms
by improving the insulin resistance, inhibiting gluco-
neogenesis, and reducing cholesterol.31-33 Additional in-
vestigations that compare Western and Chinese medi-
cine are required to meet the development needs of the
next era.34
The eligible 18 systematic reviews and Meta-analyses
on the treatment of T2DM by TCM were published
between 2002 and 2013. There were 12 Chinese pa-
pers with AMSTAR from 3 to 10, and six English pa-
pers with AMSTAR from 5 to 10. The methodological
quality was relatively low for Chinese papers when
558 October 15, 2016 | Volume 36 | Issue 5 |
 Table 1 Basic features, methodology, and evidence quality of included studies
Eligible Interventions for the Interventions for AMSTAR GRADE
n Quality Outcome index Main results of the systematic review and meta analysis
study treatment group the control group scores classification
Naren 10 Chinese medicine Western drugs Low FBG, 2hPBG, Berberine may reduce the fasting blood glucose of T2DM, but 10 1. Low
QMG (herbal compounds (glipizide, HbA1c, adverse is not superior to glipizide, metformin, rosiglitazone, or other [1, 4]
et al or herbal extracts) metformin, effects, drugs 2. Low
201210 rosiglitazone, Side-effects of berberine treatment for T2DM was less severe, [1, 5]
etc. ) and placebo mainly including gastrointestinal symptoms, and no
hypoglycemia

Li KJ 4 Point injection Blank control Low Clinical efficacy Astragalus injection treatment for T2DM showed a significant 3 Very low

JTCM | www. journaltcm. com

et al Jadad scores clinical effect, compared with the control group [RR = 1. 58, [1, 2, 4]
2008 11 for 1 95% CI (1. 20, 2. 09)]
Peng JL 6 Chinese patent Western drugs, Low Clinical efficacy, Jinqijiangtang troche treatment for T2DM showed a significant 7 Low
et al medicine placebo FBG, 2hPBG, clinical effect compared with the placebo group [RR = 0. 65, [1, 4, 5]
201312 (Jinqijiang Tang HbA1c, adverse 95% CI (0. 41, 1. 03)]
troche) effects
Li JB 7 Chinese medicine Western drugs No FBG, 2hPBG, Chinese medicine (alone or combined with Western Medicine) 4 Cannot
et al (herbal compounds assessment HbA1c, adverse treatment for T2DM showed significant clinical effects without judge
201313 or Chinese patent effects severe adverse effects
medicine) or
combine with

559
Western Medicine
Chen YJ 5 Chinese patent Blank control Low Clinical efficacy Liuweidihuang pill treatment for T2DM showed a significant 4 Low
201114 medicine Jadad scores clinical effect compared with the blank control group [RR = 1. [1, 4, 5]
(Liuweidihuang pills) for 1 21, 95% CI (1. 11, 1. 31)]
Li KJ 5 Chinese patent Blank control Low Clinical efficacy Liuweidihuang pill treatment for T2DM showed a significant 4 Low
Liu MJ et al. / Systematic Review

et al medicine Jadad scores clinical effect compared with the blank control group [RR = 1. [1, 4, 5]
200915 (Liuweidihuang pills) for 1 24, 95% CI (1. 07, 1. 43)]
Xu ZY 12 Chinese patent Western drugs Low FBG, HbA1c, Xiaoke Wan treatment for T2DM could improve total efficacy 6 Low [1, 4]
et al medicine (Xiaoke (Metformin, Jadad scores adverse effects when compared with conventional medicine [OR = 3. 11, 95%
201316 Wan) glyburide, between 1-2 CI (2. 38, 4. 22)]
gliclazide)
Li KJ 4 Chinese patent Blank control Low Clinical efficacy Xiaoke Wan treatment for T2DM showed a significant clinical 4 Low
200917 medicine (Xiaoke Jadad scores effect compared with the blank control group [RR = 1. 18, [1, 2, 4]
Wan) for 1 95% CI (1. 07, 1. 30)]
There was a low incidence of adverse events [OR = 0. 33, 95%
CI (0. 02, 4. 13)]
Li KJ 18 Chinese medicine Placebo, blank Low Clinical efficacy Chinese patent medicine showed better clinical efficacy 3 Low
et al (herbal compounds, control Jadad scores compared with placebo and control groups [RR = 1. 28, 95% [1, 2, 4]
201018 single herbs, less than 3 CI (1. 11, 1. 47)]
Chinese patent
medicine, and
herbal extracts)

October 15, 2016 | Volume 36 | Issue 5 |

 Table 1 Basic features, methodology, and evidence quality of included studies (continuted)
Eligible Interventions for the Interventions for AMSTAR GRADE
n Quality Outcome index Main results of the systematic review and meta analysis
study treatment group the control group scores classification
Zhang Y 87 Chinese medicine Western drugs, Medium Clinical efficacy, For Chinese medicine compared with placebo and 9 Medium [4]
201019 (herbal compounds, placebo, blank FPG, 2hPBG, Qidantongmai compared with placebo, there was no significant
decoctions, control HbA1c, insulin difference in FPG [MD = －0. 38, 95% CI (－0.92, 0.15)],
free-decoctions, sensitive index Huoxuejiangtangpingzhi Fang showed significant effects
Chinese patent compared with the placebo [MD = －1. 10, 95% CI (－1.63,
Medicine or herbal －0.57)]
extracts) or diet For YiSuLing compared with metformin, there was no

JTCM | www. journaltcm. com

therapy significant difference in FPG [MD = －0. 15, 95% CI
(－0.46, 0.15)]; Shengmai hypoglycemic decoction showed
significant effects when compared with gliclazide [MD =
－1.40, 95% CI (－1. 97, －0. 83)]; Shengqingjiangtang Fang
showed significant effects compared with acarbose [MD =
－0.45, 95% CI (－0.74, －0.16)]; Tongyu No. I had a
significant difference compared with aspirin [MD = －1.43,
95% CI (－2.78, －0.08)]; Yin and moistening Yiqihuoxue
Conger method was significantly different than metformin
[MD = －1.10, 95% CI (－2.17, －0.03)]; Tangzhiqing Fang
was significantly different than fluvastatin [MD = －1.50, 95%

560
CI (－2.82, －0.18)]

Chen 17 Chinese medicine Western drugs, 1 RCT as Clinical efficacy, Jinqijiang Tang troche treatment decreased FBG, PBG, and 6 Medium [4]
WN et al (herbal compounds, placebo High, the FBG, 2hPBG, HbA1c when compared with the placebo group, but no
Liu MJ et al. / Systematic Review

201220 Chinese patent rest as Low HbA1c, adverse significant difference exists
medicine) effects Chinese medicine could improve the clinical efficacy when
compared with Western Medicine [RR = 1.17, 95% CI (1.11,
1.22)]
There was a low incidence of adverse effects

Li KJ 4 Point injection Blank control Low Clinical efficacy Point injection showed superior clinical efficacy compared with 4 Very low [1,
et al Jadad scores the control group [RR = 1.28, 95% CI (1. 11, 1.47)] 2, 4, 5]
200921 for 1
Pu Run 18 Chinese patent Western drugs 1 RCT as FBG, 2hPBG, When compared with the Western drug group, Liuweidihuang 9 Medium [4]
et al medicine High, the HbA1c pills improved FBG [MD = 0.54, 95% CI (0.15, 0.93)];
201322 (Liuweidihuang rest as Low improved 2hPBG [MD = 1.05, 95% CI (0.29, 1.81)]; and
pills) or combined improved HbA1c [MD = 0.23, 95% CI (0.02, 0.45)]
with Western
medicine

October 15, 2016 | Volume 36 | Issue 5 |

 Table 1 Basic features, methodology, and evidence quality of included studies (continuted)
Eligible Interventions for the Interventions for AMSTAR GRADE
n Quality Outcome index Main results of the systematic review and meta analysis
study treatment group the control group scores classification
Dong H 14 Chinese medicine Western drugs, Low FBG, 2hPBG, Chinese medicine combined with lifestyle intervention 9 Medium [4]
et al (herbal compounds, placebo, lifestyle HbA1c, adverse decreased FBG significantly when compared with the placebo
201223 herbal extracts), or effects group [MD = －0.87, 95% CI (－1.2, －0.54)]; improved
combined Western 2hPBG [MD = －1.72, 95% CI (－2.32, －1.11)]; and
medicine and improved HbA1c [MD = －0.72, 95% CI (－0.97, －0.47)]
lifestyle Chinese medicine intervention showed no significant difference
in FBG when compared with Western drugs [MD = 0.20, 95%

JTCM | www. journaltcm. com

CI (－0.11, 0.51)]; or HbA1c [MD = －0. 11, 95% CI
(－0. 32, 0. 09)]
Chinese medicine combined with Western drugs decreased
FBG significantly compared with Western drugs alone [MD =
－0.59, 95% CI (－0.83, －0.35)]; improved 2hPBG [MD =
－1.05, 95% CI (－1.62, －0.48)]; improved HbA1c [MD =
－0.53, 95% CI (－0.95, －0.11)]
No severe adverse effects were observed

Kim TH 5 Acupuncture and Western drugs Low FBG, 2hPBG, Treatment of T2DM by moxibustion showed significant clinical 5 Low [1, 4,
et al moxibustion HbA1c, adverse efficacy by qualitative analysis, without any severe adverse effects 5]

561
201124 effects
Kim S 4 Chinese medicine Placebo Low FPG, FPI, Chinese medicine showed no significant difference in 5 Low [1, 4,
et al (herbal compounds, 2hPBG, HbA1c decreasing FBG compared with the placebo group [WMD = 5]
201125 herbal extracts) －0.43, 95% CI (－1.16, 0. 30)]; or FPI [WMD = －8.43,
95% CI (－19. 54, 2.68)]
Liu MJ et al. / Systematic Review

There was no significant difference in FPG, HbA1c, or PP2h

compared with the non-treatment group
Liu JP 66 Chinese medicine Western drugs Low FBG, 2hPBG, Chinese medicine interventions showed significant clinical 7 Low [1, 5]
et al (glyburide, HbA1c, adverse efficacy in FBG, 2hPBG, and HbA1c compared with placebo
200226 tolbutamide, effects group and the single Western drug intervention group. No
gliclazide), severe side-effects were observed
placebo
Wang X 7 Chinese medicine Placebo 3 RCT as FBG, FINS, Chinese medicine showed no significant differences in each 9 Medium [4]
et al (herbal compounds, High, the 2hPBG, HbA1c, index compared with placebo groups, FBG [SMD = 0.04, 95%
201327 herbal extracts) rest as Low Homa-IR CI (－0.15, 0.24)]; FINS [SMD = －0.09, 95% CI (－0.30,
0. 11)]; 2hPBG [SMD = －0. 14, 95% CI (－0.63, 0.34)];
HbA1c [SMD = 0.10, 95% CI (－0.13, 0.33)]; and Homa-IR
[SMD = －0.06, 95% CI (－0.35, 0.23)]
Notes: T2DM: type 2 diabetes mellitus; FBG: fasting blood glucose; FPI: fasting plasma insulin; 2hPBG: 2-h postprandial blood glucose; HbA1c: hemoglobin A1c; PP2h: 2-h blood glucose after a meal; FINS:
fasting insulin; Homa-IR: homeostasis model assessment-insulin resistance index; GRADE: Grade of Recommendation, Assessment, Development, and Evaluation; RCT: randomized controlled trial. GRADE
downgrade factors: 1: Research limitations; 2: Inconsistency; 3: Indirect; 4: Inaccuracies; 5: Publication bias.

October 15, 2016 | Volume 36 | Issue 5 |

 Liu MJ et al. / Systematic Review
compared with English papers, according to the
GRADE classifications. Numerous systematic evalua-
tions of Chinese medicine interventions have been per-
formed by Chinese researchers.35-37 The results generally
found that incomplete literature retrieval, poor control
of selection bias, and low accuracy of data combination
methods are present in Chinese papers, which affected
the credibility of the results. High quality of systematic
reviews and meta-analyses are one of the most impor-
tant sources of evidence for evidence-based medicine
practice, and are the best basis from which to make the
clinical decisions. The methodological quality of sys-
tematic reviews on TCM needs improvement, so
high-quality clinical studies with large prospective,
multi-center, double-blind and randomized designs
should be conducted.
Comprehensive literature retrieval and quality assess-
ment were strictly executed in this work, and the
GRADE classification system was used to classify the
evidence quality. However, there are some limitations
of our study. Most research trials were executed in Chi-
na, and the limitations of research design quality and
sample sizes affect the extrapolation and the strength of
evidence. Furthermore, stratified analysis of disease
course, syndromes, and complications were excluded
by the eligible research. Therefore, it is difficult to pre-
cisely evaluate the efficacy of TCM treatment for
T2DM.
In summary, the present study can act as a reference for
the clinical use of TCM for T2DM, but caution is ad-
vised for clinicians because of GRADE classifications
of the included studies. Nevertheless, this study can
engage researchers involved in evidence-based medi-
cine, statistical methods, and the clinical health field.
Professional research teams should be trained to ex-
plore the evaluation of TCM, and analyze the litera-
ture on the use of TCM in treatments. Meanwhile,
quality supervision on the systematic evaluation of
TCM should be carried out, and an evaluation method-
ology should be established along with a TCM litera-
ture database. This will ensure that TCM treatments
can be evaluated objectively, which will improve the
evidence quality of Chinese medicine evidence-based
health policy.
REFERENCES
1 Yang W, Lu J, Weng J, et al. Prevalence of diabetes among
men and women in China. N Engl J Med 2010; 362(12):
1090-1101.
2 Yu X, Li MW, Jiang H, et al. Prevalence and control of di-
abetes in Chinese adults. JAMA 2013; 310(9): 948-958.
3 Lozano R, Naghavi M, Foreman K, et al. Global and re-
gional mortality from 235 causes of death for 20 age
groups in 1990 and 2010. Lancet 2012; 380(9859):
2095-2128.
4 Murray CJ, vos T, Lozano R, et al. Disability-adjusted life
years (DALYs) for 291 diseases and injuries in 21 regions,
JTCM | www. journaltcm. com
1990-2010. Lancet 2012; 380(9859): 2197-2223.
5 Mima Y, Kuwashiro T, Yasaka M, et al. Impact of Metfor-
min on the Severity and Outcomes of Acute Ischemic
Stroke in Patients with Type 2 Diabetes Mellitus. J Stroke
Cerebrovasc Dis 2016; 25(2): 436-446.
6 Zinman B, Wanner C, Lachin JM, et al. Empagliflozin,
cardiovascular outcomes, and mortality in Type 2 diabetes.
N Engl J Med 2015; 373(22): 2117-2128.
7 Nahas R, Moher M. Complementary and alternative med-
icine for the treatment of type 2 diabetes. Can Fam Physi-
cian 2009; 55(6): 591-596.
8 Yang JH, Liu YL, Yuan JQ, et al. Development and im-
provement of systematic reviews. Zhong Guo Xun Zheng
Yi Xue Za Zhi 2011; 6(1): 54-57.
9 SeoHJ, KimKU. Quality assessment of systematic reviews
or meta-analyses of nursing interventions conducted by
Korean reviewers. BMC Med Res Methodol 2012; 28(12):
129.
10 Qimuge NR, Zhao TY, Tian C. Systematic review of effi-
cacy and safety of berberine treatment of type 2 diabetes.
Zhong Guo Xun Zheng Yi Xue Za Zhi 2012; 12(1):
81-91.
11 Li KJ, Ma LH, Li DM. Systematic review of randomized
controlled trials by astragalus injection treatment of Type
2 diabetes. Liaoning Zhong Yi Za Zhi 2008; 35(12):
1798-1800.
12 Peng JL, Yin P, Wang B, et al. Efficacy and safety for Jinqi-
jiang Tang troche treatment of type 2 diabetes. Yi Yao Dao
Bao 2013; 32(6): 796-800.
13 Li JB, Wang DK, Lu FE, et al. Efficacy and safety evalua-
tion for Jinlida particle treatment of type 2 diabetes.
Zhong Guo Yi Yuan Yong Yao Ping Jia Yu Fen Xi 2013; 13
(7): 591-594.
14 Chen YJ. Systematic review of randomized controlled tri-
als for Liuweidihuang pills treatment of type 2 diabetes.
Zhong Guo Zhong Yi Yao Xian Dai Yuan Cheng Jiao Yu
2011; 9(7): 148-149.
15 Li KJ, Ma LH. Systematic review of randomized con-
trolled trials for LiuWeidihuang pills treatment of type 2
diabetes. Shandong Zhong Yi Za Zhi 2009; 28(10):
684-686.
16 Xu ZY, Xia W, Zhu MJ, et al. Meta-analysis of random-
ized controlled trials for Xiaoke pills to treat type 2 diabe-
tes. Liaoning Zhong Yi Yao Da Xue Xue Bao 2013; 15(1):
137-140.
17 Li Kj. Systematic review of randomized controlled trials
for Xiaoke pills to treat type 2 diabetes. Yi Yao Dao Bao
2009; 28(2): 257-258.
18 Li KJ, Ma LH. Systematic review of randomized con-
trolled trials for Yiqiyangjing Fang to treat type 2 diabetes.
Zhong Guo Zhong Yi Ji Chu Yi Xue Za Zhi 2010; 16(4):
310-313.
19 Zhang Y. Systematic review of randomized controlled tri-
als for Chinese medicine treatment on type 2 diabetes. Bei-
jing: Beijing University of Traditional Chinese Medicine,
2010: 31-55.
20 Chen WN, Liu ZL, Zhang C, et al. Systematic review of
randomized controlled trials for Chinese medicine treat-
ment on type 2 diabetes. Zhong Guo Zhong Yi Yao Xin
Xi Za Zhi 2012; 19(1): 21-28.
562 October 15, 2016 | Volume 36 | Issue 5 |
 Liu MJ et al. / Systematic Review
21 Li KJ, Ma LH. Systematic review of randomized con-
trolled trials for Shengmai San formulations treatment on
type 2 diabetes. Zhong Cheng Yao 2009; 31(1): 20-23.
22 Pu R, Geng XN, Yu F. Evidence-based integrative medi-
cine. Chin J Integr Med 2013; 19(10): 783-791.
23 Hui D, Nan W, Li Z. Berberine in the treatment of type 2
diabetes mellitus: a systemic review and Meta-analysis. Ev-
id Based Complement Alternat Med 2012; 2012: 1-11.
24 Kim TH, Choi TY, Shin BC, et al. Moxibustion for man-
aging type 2 diabetes mellitus: a systemic review. Chin J
Integr Med 2011; 17(8): 575-579.
25 Kim S, Shin BC, Lee MS, et al. Red ginseng for type 2 di-
abetes mellitus: a systemic review of randomizedcontrolled
trials. Chin J Integr Med 2011; 17(12): 937-944.
26 Liu JP, Zhang M, Wang W, et al. Chinese herbal Meidi-
cines for type 2 diabetes mellitus. The Cochrane Datebase
syst Rev 2004; (3): CD003642.
27 Wang X, Tian J, Jiang J. Effects of green tea or green tea
extract on insulin sensitivity and glycaemic control in pop-
ulations at risk of type 2 diabetes mellitus: a systemic re-
view of randomized controlled trials. J Hum Nutr Diet
2014; 27(5): 501-512.
28 Yao CP. Huang Di Nei Jing. Beijing: Metallurgical Indus-
try Press, 2012: 193.
29 Yao CP. Huang Di Nei Jing. Beijing: Metallurgical Indus-
try Press, 2012: 580.
30 Lou JY, Cao FQ, Zhang Z, et al. Effects of Liuweidi-
huang pills on vascular endothelial function in type 2 dia-
JTCM | www. journaltcm. com
betic patients. Zhong Guo Lin Chuang Yao Li Xue Yu Zhi
Liao Xue 2013; 18(12): 1392-1396.
31 Kong W, Wei J, Abidi P, et al. Berberine is a novel choles-
terol lowering drug working through a unique mechanism
distinct from statins. Nat Med 2004; 10(12): 1344-1351.
32 Lee YS, Kim WS, Kim KH, et al. Berberine, a natural
plant produnct, activates amp-activated protein kinase
with beneficial metabolic effects in diabetic and insulin-re-
sistant States. Diabetes 2006; 55(8): 2256-2264.
33 Xia X, Yan J, Shen Y, et al. Berberine improves glucose
metabolism in diabetic rats by inhibition of hepatic gluco-
neogenesis. PLoS One 2011; 6(2): e16556.
34 Wang ML, Zhang YD. Methodology and trends for com-
bination of Chinese medicine and Western Medicine. Shi
Yong Zhong Yi Nei Ke Za Zhi 2013; 27(5): 168-169.
35 Li Q, Xia Y, Mou YJ, et al. Re-evaluation for the quality
of systematic review and Meta-analysis of literature pub-
lished in domestic Chinese medicine journals. Beijing
Zhong Yi Yao Da Xue Xue Bao (Zhong Yi Lin Chuang
Ban) 2012; 19(3): 28-33.
36 Hu D, Kang DY, Wu YX, et al. Methodological quality as-
sessment for the Chinese medicine related systematic re-
views published in domestic journals. Zhong Guo Zhong
Xi Yi Jie He Za Zhi 2011; 31(3): 402-406.
37 Xiong J, Du YH, Li B, et al. Evaluation for the quality of
systematic review/Mata analysis methodology and publish
of the Chinese acupuncture and moxibustion. Zhong Guo
Zhen Jiu 2009; 29(9): 763-769.
563 October 15, 2016 | Volume 36 | Issue 5 |