Access to this full-text is provided by Wiley.
Content available from International Journal of Food Science & Technology
This content is subject to copyright. Terms and conditions apply.
Original article
Systematic review and meta-analysis of anti-hyperglycaemic
effects of Pu-erh tea
Hsin-Cheng Lin,
1,2,3
Chien-Tai Lee,
4
Yea-Yin Yen,
5
Chiao-Lee Chu,
6
Yen-Ping Hsieh,
6
Chi-Sheng Yang
1
&
Shou-Jen Lan
1,7
*
1 Department of Healthcare Administration, Asia University, Taiwan No.500, Lioufeng Rd., Wufeng, Taichung 41354, Taiwan
2 Taichung Armed Forces General Hospital, Taichung, Taiwan No.348, Sec. 2, Zhongshan Rd., Taiping Dist., Taichung City 411, Taiwan
3 National Defense Medical Center, Taipei, Taiwan No.161, Sec. 6, Minquan E. Rd., Neihu Dist., Taipei City 11490, Taiwan
4 Tsaotun psychiatric center ministry of Health and Welfare, Nantou, Taiwan No.161, Yuping Rd., Caotun Township, Nantou County 542, Taiwan
5 Department of Oral Hygiene, College of Dental Medicine, Kaohsiung Medical University, Taiwan No.100, Shih-Chuan 1st Road,
Kaohsiung 80708, Taiwan
6 Department of Long Term Care, National Quemoy University, Kinmen County, Taiwan No. 1, University Rd., Jinning Township, Kinmen
County 892, Taiwan
7 Department of Medical Research, China Medical University Hospital, China Medical University, Taichung, Taiwan No.91, Hsueh-Shih
Road, Taichung 40402, Taiwan
(Received 18 June 2018; Accepted in revised form 3 September 2018)
Summary Pu-erh tea was presumed to have anti-hyperglycaemic effects with limited evidence. This study uses meta-
analysis to investigate anti-hyperglycaemic effect of Pu-erh tea. Five English databases and three Chinese
ones were systematically searched up to July 31, 2018. Those databases were searched to identify studies
containing keywords of ‘Pu-erh’, ‘Pu’er’, ‘blood sugar’, ‘blood glucose’ and ‘hyperglycaemia’. RevMan 5
and Stata were then utilized to conduct meta-analysis. Systematic reviews collected two mice studies with
sixteen records for meta-analysis. Meta-analysis results showed that Pu-erh tea has significant anti-hyper-
glycaemic effect on mice. Pooled weight mean difference of blood sugar on mice studies were 71 and
116 mg dL
1
at 21st day and 28th day respectively. Meta-regression disclosed over a longer intervention
period showed that Pu-erh tea can reduce fasting blood glucose. Secondly, a higher dose of Pu-erh is
shown to lower fasting blood glucose more significantly.
Keywords Blood sugar, meta-analysis, Pu-erh tea, systematic review.
Introduction
Tea drinking has long been part of Chinese culture.
Tea was believed to have a multitude of benefits on
health. Green tea, black tea or oolong tea is shown
to be good for rats with high fructose diet. These teas
can ameliorate plasma triglyceride and total choles-
terol concentration. Extracts of Pu-erh tea can signifi-
cantly lower plasma concentrations of glucose,
insulin, triglycerides and free fatty acids (Yang et al.,
2001; Wu et al., 2004; Hsu, 2009; Yamashita et al.,
2012). In recent years, many people drink Pu-erh tea
because of its potential health benefits. Recent studies
have confirmed that catechins, caffeine, polyphenols,
amino acids and polysaccharides in extracts of Pu-erh
tea have beneficial effect on the in vivo glucose home-
ostasis for people with type 2 diabetes (Du et al.,
2012). The carbohydrates derived from Pu-erh tea
polysaccharides (PTPS) can inhibit alpha-glucosidase
(Deng et al., 2015). Some studies have shown that
Pu-erh tea can even take up the roles as a new natu-
ral neuroprotective agent via a biological compound
that binds with transcription factors so as to inhibit
metabotropic glutamate Receptor 5 (mGluR5) to pro-
tect neurons from glutamate (Li et al., 2017). To pro-
vide diabetes patients with better choices on
controlling their blood sugar, researchers need to
understand the effect of Pu-erh tea on regulating dia-
betes mellitus. This study has investigated the effects
of Pu-erh tea on lowering blood glucose via both a
systematic literature review and a meta-analysis
method to analyse all relevant researches. This study
also attempted to make comprehensive conclusions to
*Correspondent: E-mail: shoujenlan@gmail.com or
sjlan@asia.edu.tw
[The copyright line for this article was changed on November 1,
2018 after first online publication].
International Journal of Food Science and Technology 2019, 54, 516–525
doi:10.1111/ijfs.13966
©2018 The Authors International Journal of Food Science & Technology published by John Wiley & Sons Ltd
on behalf of Institute of Food, Science and Technology (IFSTTF)
This is an open access article under the terms of the Creative Commons Attribution License, which permits use,
distribution and reproduction in any medium, provided the original work is properly cited.
516
the effectiveness of Pu-reh tea on lowering blood
sugar.
Pu-erh tea has long been considered as a health bev-
erage. ‘Compendium of Materia Medica’of 1596 A.D.
in China has recorded below passages: ‘Pu-erh tea bit-
ter moment, the solution for oily cattle and sheep poi-
soning, phlegm, including Tong vent, Pu-erh tea paste
black paint, sober first, the green is better, digestion
Sputum, stomach and fluid.’
Some studies have investigated the effect of Pu-erh
tea on physical health using scientific methods. Seven-
week-old SD rats were used as study samples, The study
showed that Pu-erh tea can prevent hyperlipidaemia of
SD rats on high-fat diet, while the cooked tea has an
even more significant effect (Xu et al., 2015). Pu-erh tea
protects the vascular endothelium, with possible mecha-
nism related to its antioxidant, PGI2 (Prostaglandin I2)
synthesis and the inhibition of TXA2 (Thromboxane
A2) synthesis. Li explored the protective mechanism of
Pu-erh tea on nervous system (Li et al., 2017). Pu-erh
tea contains biological compounds that bind with tran-
scription factors and subsequently inhibit mGluR5
(mluR5) expression. Inhibition of mGluR5 can lead to
nerve cells protection. As such, Pu-erh tea can be
regarded as a new natural neuroprotective agent. The
effective components of Pu-erh tea include tea polysac-
charide, tea-browning, tea polyphenols, Gallic acid and
trace elements (Hsu, 2009).
Method
Systemic literature review and meta-analysis were con-
ducted to investigate the effect of Pu-erh tea on reduc-
ing blood glucose and other related factors.
Information sources and searches
Eight bibliographic databases were searched to include
PubMed, EBSCO, SCOPUS, Cochrane Library, Web of
science, Airiti Library, CKND (China Known Network
Database) and Google Scholar. Chinese literatures on
Airiti Library and China Knowledge Network database
were also searched. The study used Google Scholar to
search other studies and restricted search criteria on
publications in English or Chinese only.
Three independent reviewers (Lee CT, Yen YY and
Lin HC) have examined and confirmed the adequacy
of literatures searched from these bibliographies data-
base, using keywords of (Pu-erh OR Pu’er OR Pu-
erh*) AND (blood sugar*OR blood glucose OR
hyperglycemia).
Study selection
Firstly, the authors of this study have unanimously
determined that only articles meeting inclusion criteria
as defined by PICO principle (population, interven-
tions, compare and outcome) will be selected and
included in the studies. Secondly, based upon search
criteria of research topic and abstract, obviously irrele-
vant studies were initially filtered out. Investigators
then examined full-text articles and abstracts deter-
mined to be compliant with inclusion criteria. Thirdly,
searched trials were considered fully eligible only when
specific research has allocated mice to a Pu-erh tea
intervention versus a control group. Additionally, the
study has further established excluding criteria that
screen out review articles, duplicate articles, non-full
text, articles without data of blood glucose changes
and non-animal trial studies from being included in
meta-analysis. Outcome from systemic review and
meta-analysis of this study has supported that Pu-erh
tea intervention can lead to decreases in blood sugar
levels.
Data analysis
Data extraction
Eligible articles of this study were primarily interven-
tional researches with statistical data recording before
and after blood glucose changes in both experimental
group and control group. As some studies showed
research results only graphically and due to the way the
results were presented by related literatures, our study
has obtained the results in two ways as shown below:
Firstly, for studies with average blood glucose and
standard deviation values of before and after the inter-
vention, we calculated the pooled standard deviation
and the difference in average blood glucose value, as
shown by below formula:
dE¼
XEpos
XE
SEdif ¼
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ðNEpos 1ÞðSDEposÞ2þðNEpre 1ÞðSDEpreÞ2
ðNEpos 1ÞðNEpre 1Þ
s
Where d
E
is the average difference in the blood glu-
cose value between before and after Pu-erh tea inter-
vention in the trial group; and S
E_dif
is the standard
deviation of the average value of blood glucose before
and after the experiment.
Secondly, for those studies where blood glucose data
were only graphically represented, the author used
Screen Ruler Pro program tools (http://www.wonder
webware.com/screen-ruler-pro/) to measure average val-
ues of blood glucose and associated standard deviation.
The effect size
In the past, results of consolidation analysis were
mainly expressed by a significant level combination
©2018 The Authors International Journal of Food Science & Technology published by John Wiley & Sons Ltd
on behalf of Institute of Food, Science and Technology (IFSTTF)
International Journal of Food Science and Technology 2019
Anti-hyperglycaemic effects on Pu-erh tea H.-C. Lin et al. 517
method. Some scholars argued that the effect size (ES)
in those studies should be analysed to understand the
size and significance of impact differences (Glass,
1976). Meta-analysis takes two additional steps than
primary study, namely clarifying research methods first
and then quantifying results of studies. By following
these two additional steps, standard common unit or
‘ES’ between different studies can be determined
accordingly. After calculating respective ES in each
study, we subsequently determined the ‘overall aver-
age ES’ to validate the intervention effect. Because
most studies used different measurement tools or
scales, the effect of averaging differences needed to
be standardised. In case if all studies used the same
measurement tools or scales, then the mean differ-
ence can be used as an effective surrogate of consoli-
dation analysis value, rendering standardisation
processes unnecessary.
Risk bias and quality assessment
Two authors (Chang Lee SN and Lan SJ) assessed
studies on the presence of a high, low or unclear risk
of bias. The Cochrane’s risk of bias table was used to
determine the methodological quality of each trial
(Green, 2017). A total of seven potential risks of bias
were evaluated: random sequence generation, alloca-
tion concealment, blinding of participants, blinding of
outcome assessment, incomplete outcome data, selec-
tive reporting and other bias. Studies involving three
or more high risks of bias were considered as poor
methodological quality (Figs 1 and 2).
Analysis
This study used Stata version 13 (Stata Corp LP, Col-
lege Station, TX, USA) and Review Manager(Rev-
Man) version 5.3 (The Cochrane Collaboration 2014,
NordicCochrane Centre Copenhagen, Denmark) in
analysing those studies by a random effect meta-analy-
sis for high heterogeneity. SMD and 95% confidence
intervals (CIs) were used as ES measurement.
Researchers then used Forest Plot to describe the
weighted mean difference and 95% Confidence Inter-
val (95% CI) and simultaneously compiled the statistic
report of Begg’s test. Meta-regression was conse-
quently used to evaluate the effect of intervention per-
iod (time) and dose.
Results
Search results
A total of 132 studies were identified in the first stage
of the literature search. Among the forty-one articles
written in English, thirteen studies were from PubMed
dataset, six from EBSCO database, thirteen from
SCOPUS dataset, one from Cochrane Library and
eight from Web of Science. As for the remaining
Figure 1 Risk of bias graph: review
authors’ judgements about each risk of bias
item presented as percentages across all
included studies. Green for low risk of bias,
yellow for unclear risk of bias and red for
high risk of bias. [Colour figure can be
viewed at wileyonlinelibrary.com]
Figure 2 Risk of bias summary: review
authors’ judgements about each risk of bias
item for each included study. Green for low
risk of bias, yellow for unclear risk of bias
and red for high risk of bias. [Colour figure
can be viewed at wileyonlinelibrary.com]
©2018 The Authors International Journal of Food Science & Technology published by John Wiley & Son sL td
on behalf of Institute of Food, Science and Technology (IFSTTF)
International Journal of Food Science and Technology 2019
Anti-hyperglycaemic effects on Pu-erh tea H.-C. Lin et al.518
eligible ninety-one articles written in Chinese, forty-
three articles were from Airiti Library, twenty-three
from China Known Network database and twenty-five
from Google Scholar. The second stage of search was
designed to screen out duplicates, reviewing articles
and those unsuitable for meta-analysis. A total of eigh-
teen articles were identified as potentially relevant as a
result of the second stage search. During the subse-
quent full-text review stage (the third stage), all eigh-
teen articles deemed potentially relevant were closely
examined by researchers, resulting in ten articles being
screened out for not meeting PICO criteria (Fig. 3). Of
the remaining eight articles left after aforementioned
screening rounds, only two articles were determined to
have met the criteria for meta-analysis (Table 1).
A systematic review of the literature
Main observation indexes include types of animals,
gender, quantity, methods of interventions, time and
body weight (Table 1). After intervention of Pu-erh
tea, most studies have shown significant level of blood
sugar decrease. One study has focused specifically on
the effect of Pu-erh tea on streptozotocin (STZ), sub-
stance responsible for inducing diabetes mellitus in rats
with blood glucose and lipid conditions. Results of the
abovementioned study have shown that Pu-erh tea
extract can significantly lower the concentrations of
plasma glucose, insulin, triglycerides and free fatty
acid, and in turn significantly ameliorate clinical out-
comes for subjects in the treatment group (Hsu, 2009).
Study by Zhou et al. (2009) have also demonstrated
that polysaccharide in Pu-erh can significantly reduce
blood glucose level of diabetic mice (Zhou et al.,
2009). Zhou, Kong and Chen have further concluded
that, in terms of efficacy in reducing blood sugar, a
high dose of Pu-erh tea polysaccharide (160 mg kg
1
)
was more potent than that of a low dose
(80 mg kg
1
). Results from study using 8-week-old
Wistar spontaneous mutant rats as sample have also
shown that Pu-erh tea extract contains catechins, caf-
feine, polyphenols, amino acids and polysaccharides,
ingredients capable of maintaining glucose homeostasis
for type 2 diabetes patients (Du et al., 2012). Deng
et al. (2015) used 4-week-old ICR (Institute of Cancer
Research) mice to examine the mechanism of PTPS on
reducing blood glucose. Results discovered by Deng
et al. has indicated that carbohydrates in Pu-erh tea
polysaccharide can reduce blood glucose via a-glucosi-
dase inhibition, while the inhibition effect on a-amy-
lase was found to be less significant. As part of results
derived from their study, Deng et al. further concluded
that PTPS of Pu-erh tea have a more potent hypergly-
caemic effect than acarbose (Deng et al., 2015).
Meta-analysis
The authors used data derived from two articles (Du
et al., 2012; Li et al., 2014) to conduct meta-analysis.
To better understand the effect of intervening dura-
tions, the study has divided the intervention periods
into 7, 14, 21 and 28 days, respectively.
Heterogeneity test
A total of sixteen data of two articles were analysed.
The RevMan5 software was used to examine whether
the time-to-attendance was affected or whether it was
heterogeneous (Fig. 4). The statistical results are
Figure 3 Flowchart of search results and article retrieval. [Colour figure can be viewed at wileyonlinelibrary.com]
©2018 The Authors International Journal of Food Science & Technology published by John Wiley & Sons Ltd
on behalf of Institute of Food, Science and Technology (IFSTTF)
International Journal of Food Science and Technology 2019
Anti-hyperglycaemic effects on Pu-erh tea H.-C. Lin et al. 519
Table 1 System review results
Study Animals Experiment Subjects Intervention
Experiment
duration Outcome
Cai et al.
(2017)
(Japan)
C57BL/6J male
mice, 8 weeks
old
Divided into four groups
and fed either a normal
diet or a high fat diet, with
or without 5 mg mL
1
Pu-
erh tea in the drinking
water for 16 weeks
5mgmL
1
Pu-erh tea in the drinking
water
16 weeks Pu’erh tea extract significantly reduced the gain
of body weight and subcutaneous adipose
tissue(SAT), but not visceral adipose tissue(VAT)
adiposity, in mice fed the high-fat diet and
induced adipogenesis in VAT.
The expression of DNL-related genes, including
Glut4, encoding an important insulin-regulated
glucose transporter (GLUT4), were highly
elevated in VAT.
The expression of DNL-related genes in VAT was
inversely correlated with hepatosteatosis and
systemic insulin resistance.
Deng et al.
(2015)
(Taiwan)
ICR (Institute
of Cancer
Research)
mice(4 weeks
old) 18–20
g/Male
Randomly divided into 4
groups of 5
acarbose (5 mg kg
1
) TPS of 5-year old
Pu-erh tea (1 mg kg
1
)
TPS of 5-year old Pu-erh tea (5 mg kg
1
)
Tea extraction acarbose (5 mg kg
1
)
TPS of 5-year old Pu-erh tea (1 mg kg
1
)
TPS of 5-year old Pu-erh tea (5 mg kg
1
)
Blood glucose
levels were
measured at 0,
0.5, 1, 2 and 3 h
The carbohydrates found in Pu-erh tea
polysaccharide (PTPS) inhibit a-glucosidase but
have a less inhibitory effect on a-amylase.
Mice experiments show that tea polyphenols of
PTPS may be better than acarbose on inhabiting
blood sugar. It inhibits a-glucosidase, and has a
smaller inhibition on a-amylase. It also has
fewer side effects.
Du et al.
(2012)
(China)
Spontaneous
mutation
Wistar rat
(8 weeks
old)/Male
48 obese diabetic rats and
eight non-diabetic rats
were randomly divided
into six groups
The water extraction and composition of
Pu-erh tea:
100 mg kg
1
200 mg kg
1
400 mg kg
1
acarbose: 10 mg kg
1
rosiglitazone(Avandia):5 mg kg
1
4 weeks Catechins, caffeine, polyphenols, amino acids and
polysaccharides in Pu-erh tea extracts have
beneficial effects on glucose homeostasis in
type 2 diabetes and in modification of balancing
glucose in rats. Pu-erh tea extract inhibits
intestinal intestine sucrose, maltase and
pancreatic amylase activity in rats. After 4 weeks
gavaging of Pu-erh extract, rats had impaired
glucose tolerance and improved relief of insulin
response. Pu-erh extract has an effect on
glucose homeostasis in type 2 diabetes and
improves insulin resistance.
Fang et al.
(2015)
(China)
SPF Balb/c
rate (4–8
weeks
old)/Male
A. Single dose: 48 mice
were divided randomly
into control group, Pu-erh
tea group, green tea group,
EGCG, EGCG +caffeine,
caffeine
B. long-term study: 32 mice
were divided randomly
into four groups, control,
GTE, EGCG, or caffeine
A. Single dose:
Pu-erh tea group:
Pu-erh tea extract 800 mg kg
1
Green tea group:
green tea extract 800 mg kg
1
EGCG group:
epigallocatechin gallate 240 mg kg
1
EGCG +caffeine group: EGCG
240 mg kg
1
+caffeine 80 mg kg
1
Caffeine group: caffeine 80 mg kg
1
B. long-term study: GTE group: green tea
extract 5 mg mL
1
A. 2 h
B.4 weeks
A. Mice that were given gavage feeding which
consumed Green Tea Extract (GTE), Pu-erh Tea
Extract (PTE), or Caffeine showed a significant
reduction in blood glucose concentration
compared to the controlled group.
B. Epigallocatechin gallate (EGCG) had no effect
on hyperglycaemic activity. After removing
caffeine from both GTE and PTE, these groups
all lost their effect of reducing blood glucose.
After re-adding caffeine, the authors discovered
their ability to reduce blood glucose
©2018 The Authors International Journal of Food Science & Technology published by John Wiley & Son sL td
on behalf of Institute of Food, Science and Technology (IFSTTF)
International Journal of Food Science and Technology 2019
Anti-hyperglycaemic effects on Pu-erh tea H.-C. Lin et al.520
Table 1 (Continued)
Study Animals Experiment Subjects Intervention
Experiment
duration Outcome
EGCG group: EGCG 1.5 mg mL
1
Caffeine group: caffeine 0.5 mg mL
1
Hsu (2009)
(Taiwan)
Sprague
Dawley (6
weeks old)
SD rate
Two types of experiment: A.
Prevention experiment:
According to body weight
randomly divided into 5
groups of 5.
B. Treatment experiment:
According to body weight
randomly divided into 6
groups of 5.
A. Prevention experiment: Pu-erh tea
extracts 0.1 g kg
1
B.W.
B. Treatment experiment: Pu-erh tea
extract
0
0.1 g kg
1
B.W
0.2 g kg
1
B.W.
A. Prevention
experiment:
29 days.
B. Treatment
experiment:
22 days.
Prevention experiment: the addition of Pu-erh tea
extracts before and after evoking diabetes have
significantly effect in lower the level of plasma
glucose, insulin, triglyceride and free fatty acid.
Treatment experiment: the addition of Pu-erh tea
extract can significantly improve the clinical
symptoms and adipose tissue weight in the
treatment group.
Li et al.
(2014)
(China)
C57BL/6J
mice,
Hereditary
diabetic mice:
KKAy mice,
70 female,
5–6 weeks
old, and
Diabetic
model by
STZ: 70 SD
(Sprague-
Dawley)
male rats,
female,
8–10 weeks
old
A. Hereditary diabetic
experiment: (KKAy) mice
70 KKAy with fasting blood
glucose >16 mmol L
1
were selected as the
experiment samples, were
randomly divided into four
groups of 15. Four groups:
model group, positive drug
group, Pu-erh tea group,
and green tea group.
C57BL/6J mice (n=11)
served as control group.
B. Diabetic model by STZ:
70 SD male rats were
randomly divided into a
normal control group and
a diabetic model group. 10
normal control groups and
60 diabetic model groups.
The diabetic model was
divided into five groups: a
positive group, a Pu-erh
tea extract group, a green
tea extract group, 12 in
each group.
A. Hereditary diabetic experiment: a.
Model group: distilled water
10 mL kg
1
day
1
b. Positive drug group:
1.33 mg kg
1
day
1
c. Pu-erh tea group: Pu-erh tea extract
1.0 g kg
1
d. Green tea group: green tea extract
1.0 g kg
1
B. Diabetic model by STZ: a. Normal
control group: distilled water
10 mL kg
1
day
1
b. Positive drug group:
1.33 mg kg
1
day
1
c. Pu-erh tea group: Pu-erh tea extract
1.0 g kg
1
d. Green tea group: green tea extract
1.0 g kg
1
4 weeks Pu-erh tea extract and green tea extract had an
effect on lowering fasting blood glucose (FBG)
in both KKAy mice and STZ (streptozocin-
diabetic) rats. At the same time, both of them
reduced the area under the blood glucose curve,
but the difference is that the effect of Pu-erh tea
extract is superior to green tea extract. While
Pu-erh tea extract significantly reduced the
fasting serum insulin in the model group. Green
tea extract did not significantly make any
difference compared to the model group.
Yamashita
et al. (2012)
(Japan)
Male ICR
mice, 6
weeks old
The mice divided into four
groups to receive oolong
tea, black tea, Pu-erh tea,
or water
2 g of oolong, black, and Pu-erh tea leaves
was extracted in boiled water for 2 min
7 days Intake of oolong, black, or Pu-erh tea for 7 days
enhanced GLUT4 translocation to the plasma
membrane of skeletal muscle. Each type of
fermented tea stimulated the phosphorylation of
phosphoinositide 3-kinase (PI3K), Akt/protein
kinase B, and AMP-activated protein kinase
(AMPK). Fermented tea also increased the
©2018 The Authors International Journal of Food Science & Technology published by John Wiley & Sons Ltd
on behalf of Institute of Food, Science and Technology (IFSTTF)
International Journal of Food Science and Technology 2019
Anti-hyperglycaemic effects on Pu-erh tea H.-C. Lin et al. 521
highly heterogeneous, and the results indicate that Pu-
erh tea has a statistically significant effect on lowering
blood glucose.
Meta-analysis results
A total of sixteen data sets from two articles of litera-
tures were included (forty in experimental group and
twenty in controlled group). The follow-up time points
were 7, 14, 28 days and up. Subsequent population
meta-analysis has shown that there was a high hetero-
geneity of the literature (I²= 91.9%, P<0.01). Using
random effect model analysis, diabetic mice were con-
tinuously fed with Pu-erh tea. After 7, 14, 21 and
28 days, fasting blood glucose was observed to
decrease by 0.83 (95% CI: 0.27–1.62), 3.52 (95% CI:
2.80 to 9.84), 3.94 (95% CI: 0.71–7.16) and 6.43
(95% CI: 3.07–9.78) respectively. The study has found
that the statistically significant effect of Pu-erh tea on
decreasing blood glucose after 21 and 28 consecutive
days is equivalent to a decrease in mean fasting glucose
of 71 mg dL
1
(95% CI: 13–129) and of 116 mg dL
1
(95% CI: 55–179) respectively. There was no post error
(the P-value of Begg’s test is =0.499; Fig. 1).
Meta-regression result
Numerical moderator analysis was used for two vari-
ables: intervention period and dose. Per data derived
from meta-regression the study has discovered that the
longer the intervention period, the more significant
effect of reduced fasting blood glucose level (slope:
0.088; 95% CI: 0.142~0.0347; P<0.05). Simi-
larly, this study has also revealed that the bigger the
dose of Pu-erh, the more significant effect of reduced
fasting blood glucose (slope: 0.002; 95% CI: 0.003~
0.001; P<0.05) (Fig. 5; Table 2).
Other biomarkers results of the two papers that met meta-
regression criteria
Du et al. (2012) also found that fasting blood insulin
levels of diabetic db/db mice given 400 mg kg
1
Pu-erh
tea (32 mIU mL
1
) were lower than those of type 2 dia-
betic db/db mice (53mIU mL
1
) on day 28. Further-
more, the increased blood glucose values of db/db mice
given 400 mg kg
1
Pu-erh were lower than those of the
type 2 diabetic db/db mice group at 1 and 3 h post-
OGTT. Li et al. (2014) also indicated that fasting blood
insulin levels in diabetic mice (34 IU mL
1
), given
1000 mg kg
1
, The sample group given Pu-erh tea were
lower than those (41 IU mL
1
) group without Pu-erh
tea. These two papers have shown that Pu-erh tea can
decrease fasting blood insulin of diabetic mice.
Discussion
A total of eighteen articles (seven in English and the
other eleven in Chinese) of literatures were collected
Table 1 (Continued)
Study Animals Experiment Subjects Intervention
Experiment
duration Outcome
protein expression of insulin receptor. These
results strongly suggest that fermented tea
activates both PI3K/Akt- and AMPK-dependent
signalling pathways to induce GLUT4
translocation and increases the expression of
insulin receptor to improve glucose intolerance.
Zhou et al.
(2009)
(China)
Diabetes
mellitus mice/
(20 2g)
The mice divided into four
groups: normal control
group, diabetes mellitus
model for the control
group, low dose group,
high dose group, six mice
in each group
Pu-erh tea extract tea polysaccharides
High dose 160 mg kg
1
Low dose 80 mg kg
1
3 weeks Pu-erh tea polysaccharide promoted the recovery
of body weight of diabetic mice, but there was
not a significant difference in the weight of the
mice.
Pu-erh tea polysaccharide effectively reduced the
blood sugar level in diabetic mice. The high
dose of Pu-erh tea polysaccharide had a better
hypoglycaemic effect than the low dose of Pu-
erh tea polysaccharide, and the statistical
difference was significant (P<0.05). The results
showed a relationship between dose and
response.
©2018 The Authors International Journal of Food Science & Technology published by John Wiley & Son sL td
on behalf of Institute of Food, Science and Technology (IFSTTF)
International Journal of Food Science and Technology 2019
Anti-hyperglycaemic effects on Pu-erh tea H.-C. Lin et al.522
and included in this study. Among the eighteen articles
collected, only two were determined by researchers as
adequate for meta-analysis, given that the two studies
were pre and post-test designed with 7, 14, 21 and
28 days tracking time. The researchers have conducted
meta-analysis and subsequently concluded that Pu-erh
tea can effectively control blood sugar in diabetes,
based upon data derived from those meta-analyses.
There are four articles related to cell line experi-
ments. Results from those four articles have generally
supported that Pu-erh tea extract can significantly
lower blood sugar. Consequently, Pu-erh tea can be
used as one effective alternative for controlling blood
sugar and diabetes. Some studies have focused on the
effect of Pu-erh tea ingredients on decreasing both
blood glucose and lipid. The results have shown that
uracil, gallic acid in peroxisome proliferator-activated
receptor gamma (PPARc), farnesoid X receptor
(FXR), liver X receptor (LiXR) each has a noticeable
lipid-lowering effect, with gallic acid on PPARcactiva-
tion having an effect similar to the positive control ago-
nistic effect of drugs. Regarding hypoglycaemic effect,
ethanol extract of Pu-erh tea had the highest hypogly-
caemic activity (Zhang et al., 2006, 2009). In recent
years, study scope of Pu-erh tea has been shifted from
Pu-erh tea types and extraction methods towards the
effect of specific Pu-erh tea ingredients. Liu (2013) indi-
cated the mechanism of Pu-erh tea polysaccharide in
promoting 3T3-L1 adipocyte differentiation and
glucose intake (Liu, 2013). Using molecular biology to
investigate the treatment of diabetes, the results showed
that Pu-erh polysaccharides simulate PPARcand glu-
cose transporter type 4 (GLUT4) protein performance
to promote adipocyte differentiation and glucose
intake. This mechanism improves insulin resistance effi-
ciency and hypoglycaemic effect. The other hypogly-
caemic mechanism of PTPS may involve the regulation
of PIP3/Akt signal pathway. PTPS was founded to
enhance the expression of PI3Kp85/p-Akt/GLUT4 in
type 2 Diabetes mellitus mice (Li et al., 2015). The
intake in those Diabetes mellitus mice was administered
orally with PTPS being dissolved in normal saline at
the dose of 200, 400, 800 mg kg
1
body weight per day
for 28 days. The expression of PI3Kp85, p-Akt and
GLUT4 increased with the dose-dependent effect. With
the increase of PTPS, glucose level has in turn
decreased. Another study indicated that polyphenol-
rich and caffeine-rich Pu-erh tea improved diet-induced
metabolic syndrome. This effect was likely associated
with the remodelling of gut microbiota (Gao et al.,
2018).
Regarding results from the systematic review, a
majority of the literatures reviewed have shown that
Pu-erh tea makes a significant difference in hypogly-
caemic effect. Different experiment designs such as
intervention concentration of PTPS, intervention
time, among other factors will nevertheless affect
respective final results of the study. Furthermore, we
Figure 4 Forest plot of lower blood sugar
involved in the timing of mouse trials. The
overall I
2
:81.1%. [Colour figure can be
viewed at wileyonlinelibrary.com]
©2018 The Authors International Journal of Food Science & Technology published by John Wiley & Sons Ltd
on behalf of Institute of Food, Science and Technology (IFSTTF)
International Journal of Food Science and Technology 2019
Anti-hyperglycaemic effects on Pu-erh tea H.-C. Lin et al. 523
have also examined the effect of intervention dura-
tion on heterogeneity, with results showing that
heterogeneity exists with different lengths of interven-
tion time.
The results gleaned from meta-analysis have shown
that the intake of Pu-erh tea in diabetic mice has a sig-
nificant effect on lowering blood glucose after 21 days.
Based upon conclusive findings derived from this
research, the authors recommend with professional
confidence that people drink Pu-erh tea for more than
3 weeks in order to maximise the effect on lowering
blood sugar.
Conclusions
Research findings from articles selected for systemic
review have generally supported that a number of
ingredients extracted from Pu-erh tea have the health
benefit of lowering blood sugar, with a few studies
conducting experiments and validating the argument
that drinking Pu-erh tea can help regulate and main-
tain adequate level of blood sugar. Unfortunately,
none of those articles have identified specific ingredient
(s) responsible for lowering hyperglycaemia and their
actual mechanism(s).
Two factors have restricted the number of literatures
selected and included in meta-analyses of our study.
Firstly, the difference of Pu-erh tea extraction method
and dose might be the important reason for the incon-
gruous results of the Pu-erh tea effect on blood sugar.
Inconsistent extraction method might result in different
component and concentration in Pu-erh tea extraction.
The sample group given low dose of Pu-erh tea
extraction compared to the control group show non-
significant effect, even the experiment through 1
month. Secondly, different animal model might show
unequal response effect to Pu-erh tea extraction.
Via systemic literature reviews and subsequent meta-
analyses, using data furnished by those selected arti-
cles, our research has validated the claimed health
benefits of Pu-erh tea in decreasing blood sugar level
with additional evidence-based information derived
from statistical analyses. This study has further identi-
fied ‘intervention concentration’ and ‘intervention
time’ as the two key factors in determining the efficacy
of Pu-erh tea in decreasing hyperglycaemia. With real-
world data from our study, the authors are pleased to
validate the widely claimed health benefits of Pu-erh
tea in regulating blood sugar.
As all existing researches published and accessible to
us still fail to identify the exact ingredient(s) at work,
their respective optimal intervention concentration and
duration, regulating mechanisms and relationship or
interactions with blood biomarkers such as insulin, the
authors unanimously agree that more studies are
required to explore those unanswered questions,
Figure 5 Meta-regression for intervention period (a) and dose (b).
(a) Longer intervention period, the effect of reduced fasting blood
glucose (slope: 0.088; 95% CI: 0.142~0.0347; P<0.05). SMD,
standardised mean difference; Intervention periods (time: days). (b)
Higher dose of Pu-erh, the effect of reduced fasting blood glucose
(slope: 0.002; 95% CI: 0.003~0.001; P<0.05, Pu-erh tea dose
unit: mg kg
1
). SMD: Standardised Mean Difference. [Colour figure
can be viewed at wileyonlinelibrary.com]
_ES Coef. Std. Err. tP>|t|95% Conf. Interval
Intervention period (days) 0885874 0.0249096 3.56 0.004 0.1424013 0.0347735
Dose (mg kg
1
)0.0024948 0.0005599 4.46 0.001 0.0037043 0.0012853
_cons 1.416394 0.5136038 2.76 0.016 0.3068203 2.525967
*Number of obs =16
†
s
2
=0.1734
‡
I-squared_res=43.37%
§
Adj R-squared=89.59%.
Table 2 Numerical moderator analysis
was used to two variables, including inter-
vention period and dose
©2018 The Authors International Journal of Food Science & Technology published by John Wiley & Son sL td
on behalf of Institute of Food, Science and Technology (IFSTTF)
International Journal of Food Science and Technology 2019
Anti-hyperglycaemic effects on Pu-erh tea H.-C. Lin et al.524
preferably in the forms of animal studies or human
clinical trials. We would recommend that interested
researchers may consider further studies on tea
polysaccharide of Pu-erh tea and its role in lowering
blood sugar.
Conflicts of interest
All the authors read and approved the final manu-
script. There are no conflicts of interest to declare.
Funding
The authors received no financial support for the
research, authorship and/or publication of this article.
Ethical approval
This review article did not require informed consent or
approval by the ethics committee.
References
Cai, X., Hayashi, S., Fang, C. et al. (2017). Pu’erh tea extract-
mediated protection against hepatosteatosis and insulin resistance
in mice with diet-induced obesity is associated with the induction
of de novo lipogenesis in visceral adipose tissue. Journal of Gas-
troenterology,52, 1240–1251.
Deng, Y.T., Lin-Shiau, S.Y., Shyur, L.F. & Lin, J.K. (2015). Pu-
erh tea polysaccharides decrease blood sugar by inhibition of
alpha-glucosidase activity in vitro and in mice. Food Funct,6,
1539–1546.
Du, W.H., Peng, S.M., Liu, Z.H., Shi, L., Tan, L.F. & Zou, X.Q.
(2012). Hypoglycemic effect of the water extract of Pu-erh tea.
Journal of Agriculture and Food Chemistry,60, 10126–10132.
Fang, C.-Y., Wang, X.-J., Huang, Y.-W., Hao, S.-M. & Sheng, J.
(2015). Caffeine is responsible for the bloodglucose-lowering effects
of green tea and Puer tea extractsin BALB/c mice. Chinese Journal
of Natural Medicines,13, 595–601.
Gao, X., Xie, Q., Kong, P. et al. (2018). Polyphenol- and caffeine-
rich postfermented Pu-erh tea improves dietinduced metabolic syn-
drome by remodeling intestinal homeostasis in mice. Infection and
Immunity,86, e00601–17.
Glass, G.V. (1976). Primary, secondary, and meta-analysis of
research. Educational Researche,5,3–8.
Green, J.P.H.S. (2017). Cochrane Handbook for Systematic Reviews
of Interventions. www.cochrane-handbook.org: The Cochrane Col-
laboration. Chichester: John Wiley & Sons.
Hsu, S.-H. (2009). Effect of Pu-erh tea on blood sugar and lipid in
STZ-induced diabetic rats. In: Department of Food and Nutrition.
Taichung: Providence University.
Li, Z., Wang, Y., Wang, G., Chen, Y., Li, C. & Ma, X. (2014).
Study of the hypoglycemic effect of Pu0er tea and green tea
extracts. Jounral of Tea Science,34, 428–434.
Li, S., Chen, H., Wang, J., Wang, X., Hu, B. & Lv, F. (2015).
Involvement of the PI3K/Akt signal pathway in the hypoglycemic
effects of tea polysaccharides on diabetic mice. International Jour-
nal of Biological Macromolecules,81, 967–974.
Li, C., Chai, S., Ju, Y. et al. (2017). Pu-erh tea protects the nervous
system by inhibiting the expression of metabotropic glutamate
receptor 5. Molecular Neurobiology,54, 5286–5299.
Liu, Z.-H. (2013). Pu-erh tea polysaccharides stimulate adipogenesis
and glucose uptake in 3T3-L1 adipocyte. In: Biochemistry and
Molecular Biology. Master degree thesis. Pp. 1–49. Taipei: College
of Medicine, National Taiwan University.
Wu, L.Y., Juan, C.C., Hwang, L.S., Hsu, Y.P., Ho, P.H. & Ho,
L.T. (2004). Green tea supplementation ameliorates insulin resis-
tance and increases glucose transporter IV content in a fructose-
fed rat model. European Journal of Nutrition,43, 116–124.
Xu, X., Wang, P., Luo, S., Hou, Y., Shao, W. & Yang, J. (2015).
Hypoglycemic effects of fermented Puer tea extracts -Theabrownins
in mice of type 2 diabetes. Chinese Journal of Ethnomedicine and
Ethnopharma,20,9–10.
Yamashita, Y., Wang, L., Tinshun, Z., Nakamura, T. & Ashida,
H. (2012). Fermented tea improves glucose intolerance in mice
by enhancing translocation of glucose transporter 4 in skeletal
muscle. Journal of Agriculture and Food Chemistry,60, 11366–
11371.
Yang, M., Wang, C. & Chen, H. (2001). Green, oolong and
black tea extracts modulate lipid metabolism in hyperlipidemia
rats fed high-sucrose diet. Journal of Nutritional Biochemistry,
12,14–20.
Zhang, D.-Y., Liu, Z.-H., Shi, Z.-P., Liu, Y.-L. & Fu, D.-H. (2006).
Study on blood glucose-depressing and blood lipid-depressing func-
tion of Pu’er tea by high throughput screening method. Jounral of
Tea Science,26,49–53.
Zhang, D.-Y., Shao, W.-F., Liu, Z.-H., Liu, Y.-L. & Huang, Y.-W.
(2009). Research on the anti-diabetes and anti-hyperlipidenmia
function of monomers in Pu-erh tea. Jounral of Tea Science,29,
41–46.
Zhou, B.-X., Kong, L.-B. & Chen, J.-X. (2009). Study on Extraction
of Pu’er Tea Polysaccharides and Hypoglycemic. Chinese Agricul-
tural Science Bulletin,25,55–59.
©2018 The Authors International Journal of Food Science & Technology published by John Wiley & Sons Ltd
on behalf of Institute of Food, Science and Technology (IFSTTF)
International Journal of Food Science and Technology 2019
Anti-hyperglycaemic effects on Pu-erh tea H.-C. Lin et al. 525
Available via license: CC BY
Content may be subject to copyright.