Dengzhanhua Preparations For Acute Cerebral Infarction (Review)

Cochrane Database of Systematic Reviews
Dengzhanhua preparations for acute cerebral infarction

(Review)
Cao W, Liu W, Wu T, Zhong D, Liu G
Cao W, Liu W, Wu T, Zhong D, Liu G.

Dengzhanhua preparations for acute cerebral infarction.
Cochrane Database of Systematic Reviews 2008, Issue 4. Art. No.: CD005568.
DOI: 10.1002/14651858.CD005568.pub2.
www.cochranelibrary.com
Dengzhanhua preparations for acute cerebral infarction (Review)

Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
TABLE OF CONTENTS
HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Analysis 1.1. Comparison 1 Dengzhanhua versus other drugs, Outcome 1 Marked neurologic improvement. . . . 17
APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
WHAT’S NEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
HISTORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
INDEX TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Dengzhanhua preparations for acute cerebral infarction (Review) i

[Intervention Review]
Wenzhai Cao2 , Weimin Liu3 , Taixiang Wu4 , Dechao Zhong2 , Guanjian Liu1
1 Department of Evidence-Based Medicine and Clinical Epidemiology, West China Hospital, Sichuan University, Chengdu, China.
2 Department of Internal Medicine,
Zigong No. 1 People’s Hospital, Zigong, China. 3 China Academy of Traditional Chinese Medicine,
4
Beijing, China. Chinese Cochrane Centre, Chinese EBM Centre, West China Hospital, Sichuan University, Chengdu, China
Contact address: Guanjian Liu, Department of Evidence-Based Medicine and Clinical Epidemiology, West China Hospital, Sichuan
University, No. 37 Guo Xue Xiang, Chengdu, Sichuan, 610041, China. ceuliu@hotmail.com.
Editorial group: Cochrane Stroke Group.

Publication status and date: New, published in Issue 4, 2008.
Review content assessed as up-to-date: 9 March 2008.
Citation: Cao W, Liu W, Wu T, Zhong D, Liu G. Dengzhanhua preparations for acute cerebral infarction. Cochrane Database of
Systematic Reviews 2008, Issue 4. Art. No.: CD005568. DOI: 10.1002/14651858.CD005568.pub2.
ABSTRACT
Background
Dengzhanhua preparations are widely used in China. Many controlled trials have been undertaken to investigate the efficacy of
dengzhanhua preparations in the treatment of acute cerebral infarction.
Objectives
To assess whether dengzhanhua preparations are effective and safe at improving outcomes in patients with acute cerebral infarction.
Search methods
We searched the Cochrane Stroke Group Trials Register (last searched October 2007), the Chinese Stroke Trials Register (last searched
June 2006), the trials register of the Cochrane Complementary Medicine Field (last searched June 2006), the Cochrane Central Register
of Controlled Trials (CENTRAL) (The Cochrane Library, Issue 2 2006), MEDLINE (1966 to June 2006), EMBASE (1980 to June
2006), AMED (the Allied and Complementary Medicine Database, 1985 to June 2006), the China Biological Medicine Database
(CBM-disc, 1979 to June 2006), and Chinese Knowledge Infrastructure (CNKI,1994 to October 2007). We also searched the reference
lists of relevant articles.
Selection criteria
Randomised and quasi-randomised controlled clinical trials of dengzhanhua preparations regardless of duration, dosage and route of
administration in patients with confirmed acute cerebral infarction.
Data collection and analysis
Two review authors independently applied the inclusion criteria, assessed trial quality, and extracted the data.
Main results
We included nine trials, all conducted in China, involving 723 participants. The method of randomisation and concealment was poorly
described. The included trials compared dengzhanhua injection plus routine therapy with routine therapy alone. Patients were enrolled
up to one week after the onset of stroke. No trials reported data on the pre-specified primary or secondary outcomes. In a post-hoc
comparison of dengzhanhua injection plus routine therapy versus routine therapy alone, dengzhanhua injection showed a statistically
significant benefit on the outcome ’marked neurologic improvement’ (relative risk 1.53; 95% confidence interval 1.36 to 1.72). No
serious adverse effects were reported.
Dengzhanhua preparations for acute cerebral infarction (Review) 1
Authors’ conclusions
Due to the generally low methodological quality and small sample size of the included trials in this systematic review, we could not
draw a firm conclusion.
PLAIN LANGUAGE SUMMARY
There is no clear evidence that dengzhanhua injections benefit patients with acute cerebral infarction. Dengzhanhua preparations are a
traditional herbal drug that are commonly used in China to treat disorders of the blood supply to the heart and brain, including stroke.
These compounds have a number of actions which might help reduce disability after stroke. The most common type of stroke is cerebral
infarction, which is due to blockage of the blood supply to one part of the brain. This review aimed to include randomised or quasi-
randomised trials of dengzhanhua preparations in the treatment of patients with recent cerebral infarction. Nine studies involving 723
participants were included. The studies were of poor quality. Although treatment with dengzhanhua injections appeared to improve
neurological function, there was no evidence that treatment improved the chance of being alive and free of disability. This review
therefore did not find evidence to support the routine use of dengzhanhua for patients with recent stroke. Further well-designed trials
are needed.
BACKGROUND ischaemic stroke (Gubitz 2004). Therefore, we have to seek a more

effective therapy for these patients.
A stroke is an acute impairment of focal brain function which
can be due to a variety of pathologic alterations in intracranial Dengzhanhua (Breviscapine) injection is extracted from Erigeron
or extracranial blood vessels and can result in death or physical breviscapus (Vant.) (Erigeron breviscapus is a plant which mainly
disability (Goldman 2000). In China, stroke is the second most grows in southwest China; ’Vant.’ means compositae). Dengzhan-
common cause of death in urban areas and the third most com- hua injection is a traditional herbal drug for cardio-cerebral vas-
mon in rural areas (MOH PRC 1999). It is estimated that 15 cular diseases recorded by the Chinese Drug Dictionary 1977 edi-
million new cases of stroke and five million deaths occur each year tion. The main active components of dengzhanhua injection are
worldwide (WHO 2005). In the United States, stroke is the third Scutellarin and Pyromeconic acid (Yang 2001; Zhang 2000).
most common cause of death, accounting for more than one in Recent pharmacological experiments proved that dengzhanhua
every 15 deaths in 2001. Approximately 700,000 Americans suffer has several possible modes of action. The compound can inhibit
a new or recurrent stroke each year (AHA 2004). Acute ischaemic platelet 5-HT release and platelet destruction could be reduced
stroke accounts for 80% of all strokes (Jeffrey 2005). Neurological in vivo (Wang 1989). In addition, dengzhanhua can also reduce
symptoms and signs evolve after stroke onset in 25% to 40% of brain oedema, inhibit myeloperoxidase (MPO) activity and the
patients. Severe clinical manifestations in the acute phase usually expression of intercellular adhesion molecule 1 (ICAM-1). Brevis-
indicate an increased risk of death and physical disability (Vila capine attenuates brain oedema and neutrophil infiltration after
2000). cerebral ischaemia reperfusion (He 2004). More important, there
is evidence that dengzhanhua can protect against reperfusion in-
No medication has yet been confirmed to have neuroprotective ef-
jury after ischaemia through inhibiting protein kinase C (PKC),
fects for acute ischaemic stroke (Adams 2003). Management of pa-
raised blood flow and maintaining the activity of Na+, K+ -A TPase
tients with acute ischaemic stroke with thrombolytic therapy car-
and Ca2+-ATPase in the brain (Chen 1998). Dengzhanhua can
ries the risk of catastrophic intracerebral haemorrhage (Hommel
therefore have effects on the occurrence of, and further neuronal
1995). Although intravenous recombinant tissue plasminogen ac-
impairment after acute cerebral infarction.
tivator (rt-PA) in cerebral infarction can improve functional out-
come of patients within three hours of stroke onset, it cannot be There are several preparations available such as troche (buccal
used as a routine therapy outside special units (NINDS 1995). tablet) (20 mg) and injection (5 mg per 2 ml). It can be taken orally
Heparin has no benefit in reducing mortality in patients with acute 40 mg three times per day, or 5 mg twice a day intramuscular, or
5 mg to 10 mg in 5% to 10% glucose solution intravenously for Types of interventions
acute ischaemic stroke (Ren 2002). We included trials of dengzhanhua regardless of duration, dosage
and route of administration. We compared dengzhanhua with
Though dengzhanhua treatment and its method of manufacture
placebo or no treatment. However, trials of dengzhanhua plus
are widely accepted in China, the constituents of the pharmaco-
another treatment versus the other treatment alone in order to
logical preparations used in the trials cannot be specified precisely.
assess dengzhanhua were also included.
The main component is Breviscapine, but there may be still other
components which are not clear yet. This is in marked contrast to
industrially manufactured pharmacological agents used in Western Types of outcome measures
medicine, in which the chemical constituents and their quantities
(and the percentage of any impurities or contaminants) are very
precisely known, and the variation between different production Primary outcome measure
batches is kept within specified limits. Variation between formu-
(1) Death from any cause at the end of the scheduled follow-up
lations and batches of treatments is an inevitable consequence of
period.
the nature of Chinese Traditional Medicine; it is a factor that may
contribute to any heterogeneity between different study results.
Secondary outcome measures
The effects, both beneficial and adverse, of dengzhanhua on acute
cerebral infarction need to be reviewed systematically and ap- (1) Death or dependence at the end of follow-up period.
praised critically to inform current practice and direct the contin- (2) Quality of life if used in the included trials.
ued search for new treatment regimens. (3) Adverse events: bleeding, nausea, vomiting, abdominal pain,
diarrhoea, allergic reaction, or other serious adverse event caused
by dengzhanhua. We evaluated the number of patients developing
at least one severe adverse event listed above.
OBJECTIVES
To assess the effect of dengzhanhua preparations for acute cerebral
infarction compared with placebo or open control, and to evaluate Search methods for identification of studies
the side effects and adverse events of dengzhanhua preparations. See: ’Specialized register’ section in the Cochrane Stroke Group
We searched the Cochrane Stroke Group Trials Register, which
was last searched by the Review Group Co-ordinator in Octo-
ber 2007, the Chinese Stroke Trials Register (last searched June
METHODS 2006), and the trials register of the Cochrane Complementary
Medicine Field (last searched June 2006). In addition, we searched
the Cochrane Central Register of Controlled Trials (CENTRAL)
(The Cochrane Library, Issue 2, 2006), MEDLINE (1966 to June
Criteria for considering studies for this review 2006) (Appendix 1), EMBASE (1980 to June 2006), AMED
(the Allied and Complementary Medicine Database, 1985 to June
2006) and the China Biological Medicine Database (CBM-disc,
1979 to June 2006), which is a database of Chinese biomedical re-
Types of studies
search literature. At the peer review stage of this review, a reviewer
We included randomised and quasi-randomised controlled clini- suggested we search the Chinese Knowledge Infrastructure for ad-
cal trials (that is, allocation using alternation, the sequence of additional studies. Though this was not included in the protocol, we
mission, case record numbers, dates of birth, or day of the week). searched CNKI (1994 to October 2007).
We also searched the reference lists of relevant articles.
Types of participants
Participants were male or female of any age or ethnic origin with
Data collection and analysis
acute cerebral infarction in either the carotid or vertebral artery ter-
ritory, defined by brain computerised tomography (CT) or mag-
netic resonance imaging (MRI) scan. Trials restricted to patients Study selection
with transient ischaemic attacks (TIA) or intracranial haemorrhage Two authors independently screened every title, abstract, and full
were excluded. text of study reports. We included those studies that met the pre-

determined inclusion criteria. We resolved any disagreements on (1) General information: published or unpublished, title, authors,
the selection of studies for inclusion by consensus discussion or reference or source, contact address, country, urban or rural, lan-
with a third party if necessary. We tried to obtain missing data guage of publication, year of publication, duplicate publications,
from the study authors where possible. sponsor, setting.
(2) Trial characteristics: design, duration of follow up, method of
randomisation, allocation concealment, blinding (patients, people
Quality assessment of studies administering treatment, outcome assessors).
According to the empirical evidence, we assessed the methodolog- (3) Intervention(s): intervention(s) (dose, route and timing), com-
ical quality as described by The Cochrane Handbook for System- parison intervention(s) (dose, route and timing) and co-medica-
atic Reviews of Interventions 4.2.5 (Higgins 2005). tion (dose, route and timing).
(4) Patients: exclusion criteria, total number and number in com-
parison groups, age (adults), baseline characteristics, diagnostic
Minimisation of selection bias
criteria, similarity of groups at baseline (including any co-morbid-
• Was the randomisation procedure adequate? ity), assessment of compliance, withdrawals and losses to follow
• Was the allocation concealment adequate? up (reasons, description), subgroups.
(5) Outcomes: outcomes specified above, any other outcomes as-
Minimisation of performance bias sessed, other events, length of follow up, quality of reporting of
outcomes.
• Were the patients and people administering the treatment
(6) Results: for outcomes and times of assessment (including a
blind to the intervention?
measure of variation) converted to measures of effect specified
below if necessary, intention-to-treat analysis.
Minimisation of attrition bias We resolved disagreements in data extraction by consensus, refer-
ring to the original article. If necessary, we sought information
• Were withdrawals and dropouts completely described?
from the study authors. If there was continued disagreement we
• Was analysis by intention-to-treat?
consultated a third author (Wu).
Minimisation of detection bias

• Were outcome assessors blind to the intervention? Data analysis
We included data in a meta-analysis if they were available, of suf-
Based on these criteria, we broadly subdivided the studies into the
ficient quality and sufficiently similar. We expected both dichoto-
following three categories.
mous data and continuous data. Only dichotomous data were ex-
A - all quality criteria met: low risk of bias.
tracted from included studies and expressed as relative risk (RR).
B - one or more of the quality criteria only partly met: moderate
We did not perform combination analyses for overall results since
risk of bias.
different comparators were used in variety studies. We assessed
C - one or more criteria not met: high risk of bias.
heterogeneity by using the I-square (I2 ) statistic. We used a ran-
This classification was to be used as the basis of a sensitivity analy-
dom-effects model for combination analysis in each subgroup. We
sis. Since the studies were all of the same poor quality, we did not
did not perform a publication bias analysis as there were no more
perform a sensitivity analysis.
than three studies included in a subgroup.
Two authors (Cao, Liu) assessed each trial independently. We re-
solved disagreements, if necessary, by recourse to a third author
Subgroup analyses
(Wu). In cases of further disagreement, we consulted the rest of
the group and a judgment was made based on consensus. We had planned to perform the following subgroup analyses.
• Treatment form (injection, tablet, granules, extract, capsule,
oral liquid, dripping pills, buccal tablets).
Data extraction • Course of treatment, combined medication, route of
For binary outcomes, we extracted the number of events and the administration, dosage.
total number in each group. For continuous outcomes, we ab- • The subgroup analyses were to be the time from stroke
stracted or imputed means, standard deviations and sample sizes onset to treatment, the duration of follow up (four weeks versus
for each group. Two authors (Cao, Liu) independently extracted more than four weeks), patients with Asian ethnic origin
data concerning details of study population, intervention and out- compared with ones non-Asian origin, dose (low, medium, high
comes using a data extraction form specifically designed for this based on the data). We planned to explore reasons for
review. The data extraction form included the following items. heterogeneity in the included studies and, if necessary, we also

planned to do sensitivity analyses to examine the effects of post-hoc analysis on the proportion of patients with marked neu-
excluding study subgroups, such as those studies with lower rologic improvement after treatment. The measures could con-
methodological quality. centrate on either specific neurologic dysfunction (such as motor
or cognition deficit) or overall neurologic deficit (such as the Na-
However, it was not possible to perform the subgroup analyses
tional Institute of Health Stroke Scale, which involves conscious-
because of the lack of data.
ness, vision, gaze, motor and other impaired neurological func-
tion).
As the studies were all of poor quality, we did not perform a sen-
sitivity analysis.
RESULTS
Risk of bias in included studies

Description of studies Two trials (Wang 2005; Liu 2004a) described allocated treatment
See: Characteristics of included studies; Characteristics of excluded according to the sequence of admission and hence were quasi-
studies. randomised.The remaining seven trials (Feng 2005; Huang 2005;
Through electronic searches and screening of reference lists of ar- Li 1996; Peng 2005; Wang 2004; Wen 2003; Yu 2005) did not
ticles we identified 43 potentially relevant references. At the peer describe the method of allocation or allocation concealment. We
review stage of this review, we searched the Chinese Knowledge tried to contact the original authors to check the randomisation
Infrastructure for additional studies and found other 30 poten- method in detail, but none of authors were contacted successfully.
tially relevant references. Among these, we excluded 52 irrelevant Therefore, the overall quality rating of all the included trials was
references (mainly experimental studies), which left us with 21 graded C (high risk of bias). The baseline characteristics of the
potentially eligible trials, of which nine (Feng 2005; Huang 2005; patients in the included trials showed no significant imbalance
Li 1996; Liu 2004a; Peng 2005; Wang 2004; Wang 2005; Wen between groups. The time from stroke onset to randomisation in
2003; Yu 2005) were included, containing 723 patients with acute all the trials was within one week from onset. No loss to follow up
ischaemic stroke. Details of the included trials can be found in the was reported in any trial.
Characteristics of included studies section. We excluded 12 studies
(Bai 2002; Chu 2003; Fang 1996; Liu 1999; Li 2000; Liu 2004;
Lu 2002; Sun 2000; Wang 1987; Xu 1999; Yang 2000; Zhang Effects of interventions
2002) because they compared dengzhanhua with unproven drugs
(see Characteristics of excluded studies).
The included trials were conducted in China. The average age of Death from any cause at the end of the scheduled
participants ranged from 54 to 85 years. All the trials enrolled more follow-up period
males than females, except one trial (Li 1996), which included
None of the trials reported the continued follow up after the ter-
more females. All the trials described the inclusion criteria. All the
mination of the treatment period. None of the trials reported any
trials routinely performed a CT or MRI head scan in patients to
deaths at all.
exclude haemorrhage before randomisation.
All included trials studied dengzhanhua injection to assess the
effect of dengzhanhua, and compared dengzhanhua and routine Death or dependence at the end of follow-up period
therapy in the treatment group with routine therapy alone in the No trials reported any death or dependence at the end of follow-up
control group. The doses of dengzhanhua ranged from 16 ml to 40 period. Therefore, death or dependence at the end of the follow-
ml, 30 mg to 75 mg respectively. The course of treatment ranged up period was not assessed in the included trials.
from 10 days to 30 days. The dengzhanhua injection used in the
trials was administered once a day for one course of treatment.
None of the trials reported follow up after the termination of the Quality of life if used in the included trials
treatment period. Neither deaths nor adverse events were reported. None of the included trials undertook the assessment of quality
None of the trials included quality of life as an outcome measure. of life.
None of the trials reported the outcome measures that we used in

our study protocol. The outcome measures used in the included Adverse events
trials were the measurement of neurologic deficit using the nation- None of the trials provided any data regarding adverse events,
ally approved outcome measures which were similar to the Na- and no case of serious adverse events such as major extracranial
tional Institute of Health Stroke Scale. We therefore performed a bleeding was reported during the treatment period.

Comparison 1.1: Marked neurologic improvement Secondly, the included trials were generally of small sample size.
(post-hoc comparison) None of the trials reported the method of determination of the
Although we were not able to extract any data on the pre-specified sample size. The shortcomings of a small sample size is to imply
primary and main secondary outcomes, we were able to extract low power of a test. However, no trial appeared to apply any power
data on the proportion of participants with marked neurologic analysis or mentioned the possibility of a type-II error occurring.
improvement after treatment. This was a post-hoc outcome, and Thirdly, the scheduled treatment period ranged from 10 days to
this analysis must therefore be interpreted with caution. All nine 30 days in the included trials. All the follow ups terminated at the
trials including randomised data from 723 participants evaluated end of the treatment period. The recovery of neurologic deficit
neurologic dysfunction (divided into six grades from death or de- occurred quickly in the first three months after symptom onset
terioration to cure), and we converted these outcomes to dichoto- (Kotila 1984). Three months after the onset of stroke would be an
mous data (the proportion of patients with at least 45% neurologic appropriate time to measure the neurologic outcome. Therefore
improvement). The included trials appeared to show a statistically we could not determine the long-term effect because of the short
significant benefit from dengzhanhua compared with control (RR period of follow up.
1.53; 95% CI 1.36 to 1.72); no statistical heterogeneity was de-
tected (I2 = 0%). Fourthly, acute stroke trials should use the outcome measures
which are relevant to patients to influence clinical practice (Roberts
1998). The included trials in this review used the nationally ap-
proved outcome measures alone to evaluate the patients neurologic
impairment, which was the least clinically relevant to the patients.
DISCUSSION Handicap or even health-related quality of life outcome measures
might be more relevant. However, none of the included trials used
Though dengzhanhua treatment and its method of manufacture these outcome measures.
are widely accepted in China, the effects, both beneficial and ad-
verse, of dengzhanhua on acute cerebral infarction needed to be Furthermore, we have to mention the effect on death. No deaths at
appraised critically. We included nine completed randomised or the end of the scheduled follow-up period were reported in these
quasi-randomised controlled trials of dengzhanhua preparations trials, which is more important than the neurologic improvement
for acute ischaemic infarction. The result of our meta-analysis for the stroke patients. In the recent review of trials of traditional
suggests that dengzhanhua could have potential therapeutic value Chinese medicines in acute stroke (Wu 2007), the frequency of
in the treatment of acute ischaemic infarction. However, due to death was also extremely low. Therefore, this almost certainly is
the generally poor methodological quality of the included trials, due to under-reporting, and highlights another major concern
we could not make firm conclusions. There are several substantial about the inadequate quality of these studies.
limitations.
Firstly, published studies from China were found to be differ- Finally, herbal medicine used to be perceived as being natural
ent from typical articles published in the Western literature, with and harmless in China, but recent literature on the adverse effect
key details concerning randomisation and blinding omitted. In of herbal medicine, including dengzhanhua injection, reported
our meta-analysis, although all included trials reported the use of allergic reactions, and toxic effects on the liver or renal function
randomisation, two described a method of allocation which was (Zhou 2000). Therefore we tried to review both the beneficial and
not true randomisation and the other seven did not describe the adverse effects of dengzhanhua for stroke. However, none of the
method of randomisation or concealment; our overall rating of included trials reported adverse events, which provided insufficient
the quality of all the included studies was therefore C, that is asso- data for us to evaluate the safety of dengzhanhua. From the possible
ciated with a high risk of bias. From a recent telephone interview pharmacological action of dengzhanhua, we could not draw a firm
with the clinical investigators by the Chinese Cochrane Center, conclusion that it does not cause bleeding.
only 7% of them undertook correct randomisation, while most of
them misunderstood the concept of randomisation (Wu 2006).
Through contact with several trialists before inclusion of the tri-
AUTHORS’ CONCLUSIONS
als, we found that some investigators had little knowledge of ran-
domisation design. One trial described as randomised was actually
a case-control study. Therefore we could not determine that allo-
Implications for practice
cation was truly random and well concealed. Empirical research Based on this systematic review, we found no clear evidence that
had proved that inadequate allocation concealment is associated dengzhanhua injection benefits patients with acute cerebral infarc-
with bias (Moher 1998).The result of this meta-analysis could be tion. Even the apparent improvement in neurologic impairment
confounded by possible bias. was not reliable because of inadequate randomisation and other

methodological weaknesses. Adverse effects were not reported in Furthermore, investigators conducting randomised trials should
the trials, so the safety of dengzhanhua is unclear. Due to the gen- receive formal training in clinical trial design. From the results of
erally poor methodological quality and small sample sizes of the this systematic review, it would be necessary to compare dengzhan-
included trials, we could not draw any reliable implications for hua with placebo or no intervention to discover the definite effect
practice. of dengzhanhua for acute cerebral infarction. In addition to the
outcome measures for beneficial effects, it is essential to establish
a clear monitoring and reporting system for the adverse effects of
Implications for research herbal medicines.
Clinical trials with both high methodological quality and large

sample sizes are needed. Sample size should be estimated by the
proper statistical method and power or type-II errors should be
ACKNOWLEDGEMENTS
evaluated. Trials of dengzhanhua for acute cerebral infarction
should use more clinically relevant outcome measures and suit- We extend our thanks to the Cochrane Stroke Group Editorial
able end-points identified to undertake outcome measurements. Board for their helpful comments.
REFERENCES
References to studies included in this review Wen 2003 {published data only}
Wen TY. Clinical observation on dengzhanhua injection for
Feng 2005 {published data only} 39 patients with acute ischemic stroke. China Traditional
Feng ZW, Hu ZC. Curative effect of dengzhanhua injection Chinese Medicine Emergency 2003;12(3):239.
for acute ischemic stroke. Journal of Guangxi Medical
Yu 2005 {published data only}
University 2005;22(3):438–9.
Yu XQ, Zhou J. Clinical observation of dengzhanhua
Huang 2005 {published data only} injection for 26 patients with acute ischemic stroke.
Huang XY, Ding YJ, Fang HW. Investigation on the Cardiovascular Journal of Chinese Traditional Medicine and
curative effect of dengzhanhua for acute cerebral infarction. Western Medicine 2005;3(10):925.
Modern Journal of Integrated Traditional Chinese and Western
Medicine 2005;14(15):1964–5. References to studies excluded from this review
Li 1996 {published data only}
Li F, Gu DX, Yang SJ, Lu XG. Clinical investigation of Bai 2002 {published data only}
combination therapy of Chinese Traditional Medicine & Bai GH, Li CJ, Xu B. Clinical observation on the effect
Western Medicine for acute cerebral infarction. Zhong Xi of dengzhanhua injection for acute cerebral infarction.
Yi Jie He Shi Yong Lin Chuang Ji Jiu Zha Zhi 1996;3(5): Clinical Medicine 2002;22(12):56.
209–11.
Chu 2003 {published data only}
Liu 2004a {published data only}
Chu HB, Lian QR, Zhao GF. Curative effect of
Liu YY, Liu XL, Yang W. Curative effect of dengzhanhua
dengzhanhua injection for 60 cases with acute cerebral
injection for acute ischemic stroke. Journal of Jinzhou
infarction. Journal of Practical Traditional Chinese lnternal
Medical College 2004;25(6):68.
Medicine 2003;17(5):416–7.
Peng 2005 {published data only}
Peng H, Wen XH. Dengzhanxixin for acute ischemic stroke. Fang 1996 {published data only}
Journal of Clinical Neurology 2005;18(1):73. Fang DW. Curative effect of Yunnan dengzhanhua injection
on 106 cases with acute cerebral infarction. Zhejiang Zhong
Wang 2004 {published data only} Yi Za Zhi 1996;8:381.
Wang X. Clinical observation on dengzhanhua injection for
acute ischemic stroke. Henan Journal of Practical Nervous Li 2000 {published data only}
Diseases 2004;7(4):54. Li Q, Guo YJ. Curative effect of naloxone plus dengzhanhua
for acute cerebral infarction. Yunnan Zhong Yi Zhong Yao
Wang 2005 {published data only} Za Zhi 2000;21(3):17.
Wang J, Gu W, Tan F. Effect of erigeron injection on platelet
level of CD62p and serum content of TNF-alpha and IL-6 Liu 1999 {published data only}
in patients with acute cerebral infarction. Chinese Journal Liu RH, Liu HL. Clinical investigation on curative effect of
of Integrated Traditional and Western Medicine/Chung-kuo dengzhanhua plus citicoline for acute cerebral infarction.
Chung Hsi 2005;25(4):324–6. Lin Chuang Yi Xue 1999;19(7):36–7.
Liu 2004 {published data only} Systematic Reviews 2004, Issue 2. [DOI: 10.1002/
Liu FR, Bian LZ. Curative effect of dengzhanhua on 106 14651858.CD000024.pub2]
cases with acute cerebral infarction. Central Plains Medical He 2004
Journal 2004;31(12):38–9. He W, Liu YM, Chen H, Zeng FD. Effect of breviscapine
Lu 2002 {published data only} on brain edema and neutrophil infiltration induced by focal
Lu YF. The clinical observation of erigeron breviscapus cerebral ischemia and reperfusion in rats. Chinese Journal of
hand-mazz injection in the treatment of cerebral infarction. Pharmacology and Toxicology 2004;18(3):161–5.
Journal of Qiqihar Medical College 2002;23(4):362–4. Higgins 2005
Sun 2000 {published data only} Higgins JPT, Green S, editors. Cochrane Handbook for
Sun CP, Ma JJ, XU J. Investigation on the curative effect Systematic Reviews of Interventions 4.2.5 [updated May
of dengzhanhua for 93 cases with acute cerebral infarction. 2005]. The Cochrane Library. Chichester, UK: John Wiley
Henan Yi Yao Xin Xi 2000;8(2):50–1. & Sons, Ltd, 2005.
Hommel 1995
Wang 1987 {published data only}
Hommel M, Boissel JP, Cornu C, Boutitie F, Lees KR,
Wang H, Shi M, Yang C, Li J. Clinical observation
Berson G, et al.Termination of trial of streptokinase in
on the therapeutic effect of deng-zan-hua-su-pian on
severe acute ischaemic stroke. Lancet 1995;345:57.
cerebrovascular diseases. Journal of Apoplexy and Nervous
Diseases 1987;4(4):247–8. Jeffrey 2005
Jeffrey LA. Stroke, Ischemic. http://www.emedicine.com/
Xu 1999 {published data only} EMERG/topic558.htm (accessed 9 May 2005).
Xu YL, Chen XY. Investigation on curative effect of
dengzhanhua for 34 cases with acute cerebral infarction. Kotila 1984
Zhong Guo Xiang Chun Yi Sheng Za Zhi 1999;3:23–4. Kotila M, Waltimo O, Niemi ML, Laaksonen R, Lempinen
M. The profile of recovery from stroke and factors
Yang 2000 {published data only} influencing outcome. Stroke 1984;15:1039–44.
Yang SH. Investigation on curative effect of dengzhanhua
MOH PRC 1999
plus shenmai injection for acute cerebral infarction. Journal
Chinese Health Statistical Digest. Ministry of Health,
of Jingganshan Medical College 2000;7(3):95.
Peoples Republic of China 1999:69–70.
Zhang 2002 {published data only} Moher 1998
Zhang SF, He MD, Li C. Clinical curative effect of Moher D, Pham B, Jones A, Cook DJ, Jadad AR, Moher
dengzhanhua injection on acute cerebral infarction: a report M, et al.Does quality of reports of randomised trials affect
of 100 cases. Hunan Yi Ke Da Xue Xue Bao 2002;27(3): estimates of intervention efficacy reported in meta-analyses?
235–8. . Lancet 1998;352:609–13.
Additional references NINDS 1995
National Institute of Neurological Disorders and Stroke
Adams 2003 rt-PA Study Group. Tissue plasminogen activator for the
Adams HP Jr, Adams RJ, Brott T, del Zoppo GJ, Furlan A, treatment of acute ischaemic stroke. New England Journal of
Goldstein LB, et al.Guidelines for the early management of Medicine 1995;333:1581–7.
patients with ischemic stroke: a scientific statement from Ren 2002
the Stroke Council of the American Stroke Association. Ren DQ. Breviscapine tablets. Clinical Practical Using
Stroke 2003;34:1056–83. Chinese Herbal Preparations. 1st Edition. Beijing: People’s
AHA 2004 Health Publish House, 2002:669–70.
American Heart Association. New stats show heart Roberts 1998
disease still America’s No. 1 killer, stroke No. 3. http: Roberts L, Counsell C. Assessment of clinical outcomes in
//www.americanheart.org/presenter.jhtml?identifier= acute stroke trials. Stroke 1998;29:986–91.
3018015 (accessed 15 February 2005). Vila 2000
Chen 1998 Vila N, Castillo J, Davalos A, Chamorro A. Proinflammatory
Chen XX. Protection effect of dengzhanhua on ATPase cytokines and early neurological worsening in ischemic
activity in the ischemic-reperfusion rat brain. Zhong Yao stroke. Stroke 2000;31(10):2325–9.
Xin Yao Yu Lin Chuang Yao Li 1998;9(4):214. Wang 1989
Wang ZY. Inhibitory effects of new breviscapine on
Goldman 2000 thrombosis in vivo. Zhong Xi Yi Jie He Za Zhi 1989;9(1):
Goldman L, Ausiello D. Cecil Textbook of Medicine. WB 26–8.
Saunders Company, 2000.
WHO 2005
Gubitz 2004 World Health Organization. Global burden of stroke. http:
Gubitz G, Sandercock P, Counsell C. Anticoagulants //www.who.int/cardiovascular_diseases/en/cvd_atlas_15_
for acute ischaemic stroke. Cochrane Database of burden_stroke.pdf (accessed 21 April 2005).

Wu 2006 of flavonoid glycoside in dengzhanhua. Chinese Herbal
Wu TX, Li YP, Yao X, Li J. Clinical trial registration:to Medicine 2000;31(8):565–6.
improve the quality of clinical research in China. Chinese
Zhou 2000
Journal of Evidence-based Medicine 2006;6(3):153–6.
Zhou XH, Hua W. Dengzhanhua injection cause allergic
Wu 2007 reaction. Adverse Effect of Drug Journal 2000;2(1):61.
Wu B, Liu M, Liu H, Li W, Tan S, Zhang S, et al.Meta-
analysis of traditional Chinese patent medicine for ischemic References to other published versions of this review
stroke. Stroke 2007;38(6):1973–9.
Yang 2001 Cao 2006
Yang WY. Measurment of scutellarin in the erigeron Cao W, Liu W, Wu T. Dengzhanhua preparations for
breviscapus by HPLC. Journal of Chengdu University of acute cerebral infarction. (Protocol). Cochrane Database of
Traditional Chinese Medicine 2001;24:37–8. Systematic Reviews 2006, Issue 1. [Art. No.: CD005568.
Zhang 2000 DOI: 10.1002/14651858.CD005568]
Zhang WD, Kong DM. Research on chemical component ∗
Indicates the major publication for the study

CHARACTERISTICS OF STUDIES
Characteristics of included studies [ordered by study ID]
Feng 2005
Methods RCT, method of randomisation and concealment not stated

Losses to FU: none
Blinding: not stated
Participants Country: China

108 patients (acute ischaemic stroke)
100% CT scan before entry
Comparability: age similar, more males than females
Numbers of severity of stroke not stated
Interventions T: dengzhanhua 40 ml once a day for 30 days + routine treatment

C: routine treatment
Outcomes The proportion of patients with marked neurologic improvement at 30 days
Notes FU: 30 days

Quality grading: C (high risk of bias)
Risk of bias
Item Authors’ judgement Description
Allocation concealment? Unclear B - Unclear
Huang 2005

Losses to FU: none

Numbers of severity of stroke: mild - 4 participants in each group; moderate - 13 participants in treatment
group,12 participants in control group; severe - 13 participants in treatment group,14 participants in
control group


Huang 2005 (Continued)
Notes FU: 14 days

Risk of bias
Li 1996

Losses to FU: none

100% CT or MRI scan before entry
Comparability: age similar, more females than males

Notes FU: 10 days

Risk of bias
Liu 2004a
Methods Quasi-RCT, method of randomisation: the sequence of admission, concealment not stated
Losses to FU: none


Liu 2004a (Continued)
Interventions T: dengzhanhua 75 mg once a day for 14 days + routine treatment

Notes FU: 14 days

Risk of bias
Allocation concealment? No C - Inadequate
Peng 2005

Losses to FU: none


Notes FU: 14 days

Risk of bias

Wang 2004

Losses to FU: none

Numbers of severity of stroke: mild - 20 participants in treatment group,18 participants in control group;
moderate - 32 participants in treatment group, 28 participants in control group; severe - 12 participants
in treatment group,10 participants in control group
Interventions T: dengzhanhua 30 mg once a day for 30 days + routine treatment

Notes FU: 30 days

Risk of bias
Wang 2005
Methods Quasi-RCT, method of randomisation: the sequence of admission, concealment not stated
Losses to FU: none


Notes FU: 15 days

Risk of bias

Wang 2005 (Continued)
Allocation concealment? No C - Inadequate
Wen 2003

Losses to FU: none

Numbers of severity of stroke: mild - 14 participants in treatment group,11 participants in control group;
moderate - 16 participants in treatment group,12 participants in control group; severe - 9 participants in
treatment group,7 participants in control group

Notes FU: 28 days

Risk of bias
Yu 2005

Losses to FU: none

Comparability: age similar, more females than males


Yu 2005 (Continued)
Notes FU: 30 days

Risk of bias
Routine treatment/therapy: mannitol; management of blood glucose, blood pressure, and antibiotics; no thrombolytic therapy was
applied
C: control group
CT: computerised tomography
FU: follow up
MRI: magnetic resonance imaging
RCT: randomised controlled trial
T: treatment group
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Bai 2002 Dengzhanhua compared with unproven drug
Chu 2003 Dengzhanhua compared with unproven drug
Fang 1996 Dengzhanhua compared with unproven drug
Li 2000 Confounded with naloxone
Liu 1999 Dengzhanhua compared with unproven drug
Liu 2004 Dengzhanhua compared with unproven drug
Lu 2002 Dengzhanhua compared with unproven drug
Sun 2000 Dengzhanhua compared with unproven drug
Wang 1987 Dengzhanhua compared with unproven drug
Xu 1999 Dengzhanhua compared with unproven drug

(Continued)
Yang 2000 Confounded with Shenmai
Zhang 2002 Dengzhanhua compared with unproven drug

DATA AND ANALYSES
Comparison 1. Dengzhanhua versus other drugs
No. of No. of
Outcome or subgroup title studies participants Statistical method Effect size
1 Marked neurologic improvement 9 Risk Ratio (M-H, Random, 95% CI) Subtotals only
1.1 Dengzhanhua plus routine 9 723 Risk Ratio (M-H, Random, 95% CI) 1.53 [1.36, 1.72]
therapy versus routine therapy
Analysis 1.1. Comparison 1 Dengzhanhua versus other drugs, Outcome 1 Marked neurologic improvement.
Review: Dengzhanhua preparations for acute cerebral infarction
Comparison: 1 Dengzhanhua versus other drugs
Outcome: 1 Marked neurologic improvement
Study or subgroup Treatment Control Risk Ratio Weight Risk Ratio

M- M-
H,Random,95% H,Random,95%
n/N n/N CI CI
1 Dengzhanhua plus routine therapy versus routine therapy

Feng 2005 30/58 15/50 5.6 % 1.72 [ 1.06, 2.82 ]
Huang 2005 22/30 15/30 7.8 % 1.47 [ 0.97, 2.23 ]
Li 1996 25/41 13/39 5.3 % 1.83 [ 1.10, 3.04 ]
Liu 2004a 43/56 19/50 9.3 % 2.02 [ 1.38, 2.96 ]
Peng 2005 20/33 9/33 3.5 % 2.22 [ 1.19, 4.14 ]
Wang 2004 61/64 37/56 35.7 % 1.44 [ 1.19, 1.75 ]
Wang 2005 25/35 13/33 6.1 % 1.81 [ 1.13, 2.91 ]
Wen 2003 36/39 21/30 21.6 % 1.32 [ 1.03, 1.70 ]
Yu 2005 17/26 10/20 5.0 % 1.31 [ 0.78, 2.20 ]
Subtotal (95% CI) 382 341 100.0 % 1.53 [ 1.36, 1.72 ]

Total events: 279 (Treatment), 152 (Control)
Heterogeneity: Tau2 = 0.0; Chi2 = 7.65, df = 8 (P = 0.47); I2 =0.0%
Test for overall effect: Z = 7.17 (P < 0.00001)
0.1 0.2 0.5 1 2 5 10

Favours control Favours treatment

APPENDICES
Appendix 1. MEDLINE search strategy

The following search strategy was used for MEDLINE (Ovid) and was modified to suit the other databases.
1. cerebrovascular disorders/ or exp basal ganglia cerebrovascular disease/ or exp brain ischemia/ or exp carotid artery diseases/ or
exp cerebrovascular accident/ or exp hypoxia-ischemia, brain/ or exp intracranial arterial diseases/ or exp “intracranial embolism and
thrombosis”/
2. ((brain or cerebr$ or cerebell$ or vertebrobasil$ or hemispher$ or intracran$ or intracerebral or infratentorial or supratentorial
or middle cerebr$ or mca$ or anterior circulation) adj5 (isch?emi$ or infarct$ or thrombo$ or emboli$ or occlus$ or hypoxi$ or
apoplexy)).tw.
3. (isch?emi$ adj6 (stroke$ or apoplex$ or cerebral vasc$ or cerebrovasc$ or cva or attack$)).tw.
4. 1 or 2 or 3
5. plant extracts/ or drugs, Chinese herbal/
6. phytotherapy/ or plants/ or plants, medicinal/ or asteraceae/
7. (dengzhanhua$ or deng zhan hua$ or deng-zhan-hua$).tw.
8. (dengzhanxixin$ or deng zhan xi xin$ or deng-zhan-xi-xin$).tw.
9. (erigero$ or breviscap$ or yimaikang or scutellarin or fleabane).tw.
10. 5 or 6 or 7 or 8 or 9
11. 4 and 10
WHAT’S NEW
Last assessed as up-to-date: 9 March 2008.
Date Event Description
17 April 2008 Amended Converted to new review format.
HISTORY
Protocol first published: Issue 1, 2006
Review first published: Issue 4, 2008
CONTRIBUTIONS OF AUTHORS
All five authors contributed to this review.
• Search strategy: Liu, Wu

• Study selection: Cao, Liu
• Data extraction: Cao, Liu
• Quality assessment: Cao, Liu, Wu
• Statistical and methodological input: Wu, Liu
• Interpretation of data: Cao, Zhong
• Providing general advice on the review: Liu
• Writing the review: Cao, Wu
DECLARATIONS OF INTEREST
None known
SOURCES OF SUPPORT
Internal sources
• Chinese Cochrane Center, Chinese Centre of Evidence-Based Medicine, West China Hospital of Sichuan University, China.
• China Medical Board of New York, USA.
External sources
• Cochrane Stroke Group, UK.
INDEX TERMS
Medical Subject Headings (MeSH)

Acute Disease; Anticoagulants [∗ therapeutic use]; Cerebral Infarction [∗ drug therapy]; Flavonoids [∗ therapeutic use]; Randomized
Controlled Trials as Topic
MeSH check words

Humans


Dengzhanhua Preparations For Acute Cerebral Infarction (Review)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Dengzhanhua Preparations For Acute Cerebral Infarction (Review)

Uploaded by

Copyright:

Available Formats

Cochrane Database of Systematic Reviews

Dengzhanhua preparations for acute cerebral infarction

Cao W, Liu W, Wu T, Zhong D, Liu G

Cao W, Liu W, Wu T, Zhong D, Liu G.

Dengzhanhua preparations for acute cerebral infarction (Review)

Dengzhanhua preparations for acute cerebral infarction (Review) i

Dengzhanhua preparations for acute cerebral infarction

Editorial group: Cochrane Stroke Group.

PLAIN LANGUAGE SUMMARY

Dengzhanhua preparations for acute cerebral infarction

BACKGROUND ischaemic stroke (Gubitz 2004). Therefore, we have to seek a more

Dengzhanhua preparations for acute cerebral infarction (Review) 3

Minimisation of detection bias

Dengzhanhua preparations for acute cerebral infarction (Review) 4

Risk of bias in included studies

None of the trials reported the outcome measures that we used in

Dengzhanhua preparations for acute cerebral infarction (Review) 5

Dengzhanhua preparations for acute cerebral infarction (Review) 6

Clinical trials with both high methodological quality and large

Dengzhanhua preparations for acute cerebral infarction (Review) 8

Dengzhanhua preparations for acute cerebral infarction (Review) 9

Characteristics of included studies [ordered by study ID]

Methods RCT, method of randomisation and concealment not stated

Participants Country: China

Interventions T: dengzhanhua 40 ml once a day for 30 days + routine treatment

Outcomes The proportion of patients with marked neurologic improvement at 30 days

Notes FU: 30 days

Item Authors’ judgement Description

Allocation concealment? Unclear B - Unclear

Methods RCT, method of randomisation and concealment not stated

Participants Country: China

Interventions T: dengzhanhua 40 ml once a day for 14 days + routine treatment

Outcomes The proportion of patients with marked neurologic improvement at 14 days

Dengzhanhua preparations for acute cerebral infarction (Review) 10

Notes FU: 14 days

Item Authors’ judgement Description

Allocation concealment? Unclear B - Unclear

Methods RCT, method of randomisation and concealment not stated

Participants Country: China

Interventions T: dengzhanhua 16 ml once a day for 10 days + routine treatment

Outcomes The proportion of patients with marked neurologic improvement at 10 days

Notes FU: 10 days

Item Authors’ judgement Description

Allocation concealment? Unclear B - Unclear

Participants Country: China

Dengzhanhua preparations for acute cerebral infarction (Review) 11

Interventions T: dengzhanhua 75 mg once a day for 14 days + routine treatment

Outcomes The proportion of patients with marked neurologic improvement at 14 days

Notes FU: 14 days

Item Authors’ judgement Description

Allocation concealment? No C - Inadequate

Methods RCT, method of randomisation and concealment not stated

Participants Country: China

Interventions T: dengzhanhua 40 ml once a day for 14 days + routine treatment

Outcomes The proportion of patients with marked neurologic improvement at 14 days

Notes FU: 14 days

Item Authors’ judgement Description

Allocation concealment? Unclear B - Unclear

Dengzhanhua preparations for acute cerebral infarction (Review) 12

Methods RCT, method of randomisation and concealment not stated

Participants Country: China

Interventions T: dengzhanhua 30 mg once a day for 30 days + routine treatment

Outcomes The proportion of patients with marked neurologic improvement at 30 days