traditional chinese medicine for epilepsy

8/19/2019 Traditional Chinese Medicine for Epilepsy

1/31

Traditional Chinese medicine for epilepsy (Review)

Li Q, Chen X, He L, Zhou D

This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library 2009, Issue 3

http://www.thecochranelibrary.com

Traditional Chinese medicine for epilepsy (Review)

Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

http://www.thecochranelibrary.com/http://www.thecochranelibrary.com/


2/31

T A B L E O F C O N T E N T S

1HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

8DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

9 AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

9 ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

10REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

11CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

20DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Analysis 1.1. Comparison 1 One formula of TCMH versus single Western medication, Outcome 1 Xiaxingci granule versus

phenytoin. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

Analysis 1.2. Comparison 1 One formula of TCMH versus single Western medication, Outcome 2 Dianxianning pill

versus valproate. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

Analysis 1.3. Comparison 1 One formula of TCMH versus single Western medication, Outcome 3 Tianmadingxiancapsule. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

Analysis 1.4. Comparison 1 One formula of TCMH versus single Western medication, Outcome 4 Zhixian I versus

phenytoin. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

Analysis 1.5. Comparison 1 One formula of TCMH versus single Western medication, Outcome 5 Antiepilepsy capsule

versus phenobarbital. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

Analysis 1.6. Comparison 1 One formula of TCMH versus single Western medication, Outcome 6 Incidence of adverse or

harmful effects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

26 ADDITIONAL TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

27 APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

28HISTORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

28CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

28DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

29SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .29INDEX TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

iTraditional Chinese medicine for epilepsy (Review)



3/31

[Intervention Review]

Traditional Chinese medicine for epilepsy

Qifu Li1

, Xiaoyan Chen2

, Li He1

, Dong Zhou1

1Department of Neurology, West China Hospital, Sichuan University, Chengdu, China. 2Department of Neurology, Shanghai

Changzheng Hospital, Second Military Medical University, Shanghai, China

Contact address: Dong Zhou, Department of Neurology, West China Hospital, Sichuan University, Guo xue xiang No.37, Chengdu,

Sichuan, 610041, China. [email protected].

Editorial group: Cochrane Epilepsy Group.

Publication status and date: New, published in Issue 3, 2009.

Review content assessed as up-to-date: 23 November 2007.

Citation: Li Q, Chen X, He L, Zhou D. Traditional Chinese medicine for epilepsy. Cochrane Database of Systematic Reviews 2009,Issue 3. Art. No.: CD006454. DOI: 10.1002/14651858.CD006454.pub2.


A B S T R A C T

Background

Seizures are poorly controlled in many people with epilepsy, despite current antiepileptic treatments. Some turn to alternative or

complementary therapy to treat their condition and the use of traditional Chinese medicinal herbs (TCMH) is increasingly popular.

However, it remains unclear whether the existing evidence is rigorous enough to support its use.

Objectives

To determine the effectiveness and safety of traditional Chinese medicine in people with epilepsy.

Search methods

Our search included the Cochrane Epilepsy Group’s Specialised Register and the Cochrane Central Register of Controlled Trials

(CENTRAL) (The Cochrane Library 2007, Issue 1), MEDLINE (1950 to 2007) and EMBASE (1974 to 2007).

Selection criteria

Randomised controlled trials evaluating traditional Chinese medicine in people of any age with any type of epilepsy, and comparing

one formula of TCM with no intervention, placebo or single Western medicine (monotherapy).

Data collection and analysis

Two review authors independently extracted trial data and assessed quality. We assessed the following outcomes: (a) seizure freedom

for at least one year; (b) 50% or greater reduction in seizure frequency; (c) percentage reduction in seizure frequency and duration; and

(d) adverse events.

Main results

Five short-term studies involving 1125 participants met the inclusion criteria. All the studies were of poor methodological quality and

had a high probability of selection, detection and performance bias.

Two studies assessed seizure freedom for one year. One found no difference between Xiaxingci granule and phenytoin for primary

generalized tonic-clonic seizures (RR 1.00; 95% CI 0.07 to 14.90).The other study found no difference between Dianxianning pill

and valproate (RR 13.00; 95% CI 0.74 to 227.72) for different types of epilepsy.

1Traditional Chinese medicine for epilepsy (Review)


mailto:[email protected]:[email protected]


4/31

Three studies assessed a 50% or greater reduction in seizure frequency. One found an advantage for Tianmadingxian capsule when

compared to phenytoin (RR 1.37; 95% CI 1.23 to 1.53) in different types of epilepsy, the second an advantage for Zhixian I pill when

compared to phenytoin (RR 1.31; 95% CI 1.16 to 1.48) in primary generalized tonic-clonic seizure, and the third an advantage for

an ’Antiepilepsy capsule’ when compared to phenobarbital (RR 1.21; 95% CI 1.02 to 1.43) for primary and secondary generalized

tonic-clonic seizure. One study reported the incidence of adverse effects and the Peto odds ratio was 0.04 (99% CI 0.01 to 0.12, P <

0.00001) favouring TCMH compared to phenobarbital. Authors’ conclusions

The current evidence is insufficient to support the use of traditional Chinese medicine as a treatment for epilepsy. Much larger, high

quality randomised clinical trials are needed to evaluate the effectiveness and safety of traditional Chinese medicinal herbs for treating

epilepsy.

P L A I N L A N G U A G E S U M M A R Y

Traditional Chinese medicine for epilepsy

There is no strong evidence for the use of traditional Chinese medicine as a treatment for epilepsy.

Patients with epilepsy are currently treated with antiepileptic drugs, but a significant number of people continue to have seizures and

many experience adverse effects. As a result there is increasing interest in alternative or complementary therapies, including traditional

Chinese medicine.

The review authors found five, short-term randomised controlled trials involving 1125 participants. All the studies were of poor

methodological quality and had a high probability of bias. Although some studies included in this review seemed to show some benefit

of traditional Chinese medicine, the current evidence is insufficient to support its use as a treatment for epilepsy. Much larger, high

quality randomised clinical trials are needed to evaluate the effectiveness and safety of traditional Chinese medicinal herbs for treating

epilepsy.

B A C K G R O U N D

Epilepsy is an importantneurological condition, with an estimated

annual incidence of 50 per 100,000 and prevalence of 5 to 10 per

1000 in the developedworld(Sander 1996). Approximately 3% of

the population will sufferfrom epilepsy at some point in their lives

(Hauser 1993). Although the majority of people with epilepsy will

go into remission, up to 30% will become drug-resistant despite

treatment with adequate doses of appropriate antiepileptic drugs

(AEDs) (Cockerell 1995). Their continuing attacks will result in

reduced quality of life, and may also lead to injuries, social isola-tion and depression (Villeneuve 2004). Hence, there is a constant

search for newer modes of treatment. Many people are turning

to alternative or complementary therapy to treat their condition

and the use of traditional Chinese medicinal herbs is becoming

increasingly popular.

Traditional Chinese medicine (TCM) is a 3000-year old theoret-

ical and methodological holistic system for both the treatment

and prevention of disease (Fulder 1996). TCM has unique theo-

ries regarding aetiology, systems of diagnosis and treatment, which

are vital to its practice. TCM herb treatment consists typically of

complex prescriptions of a combination of several components.

The combination is based on Chinese diagnostic patterns (i.e.

inspection, listening, smelling, inquiry and palpation) and has a

completely different rationale compared to many Western drug

treatments (Liu 2000). Historically the Chinese have used tradi-

tional Chinese medicinalherbs (TCMH) to treat epilepsy. The first

known document on epilepsy in China appeared in The Huang

Di Nei Ching, written by a group of TCM physicians between770 and 221 B.C. The description of epilepsy in this book, and

in many others published later, was confined to generalized con-

vulsive seizures. No documentation of absence or simple partial

seizures was noted. The first classification of epilepsy, probably by

Cao Yuan Fang in A.D. 610, listed five types of epilepsy: ’Yang

Dian’, ’Yin Dian’, ’Feng (Wind) Dian’, ’Shih (Wet) Dian’ and ’Ma

(Horse) Dian’. The treatment of epilepsy, based on principles of




5/31

’Yin Yang Wu Xing’, consisted of herbs, acupuncture, massage,

food therapy and therapeutic exercise (Lai 1991).

Traditional Chinese medicinal herbs are taken, or made, from nat-

ural plants. TypicalTCM treatment uses oneherb as thebasicdrug

to treat the major disease problem. This is then mixed with other

herbs to treat additional symptoms, creating a multifunction for-mulation for the disease. In recent decades, many pharmacolog-

ical functions have been suggested by experimental research. For

example, gouteng (Uncaria rhynchophylla , UR) has anticonvulsiveand free radical scavenging activities (Hsieh 1999a ; Hsieh 1999b);

shitei-To was foundto have anticonvulsanteffects (Minami 1999);

qingyangshen (QYS) was found to have antiepileptic properties

and have a therapeutic effect on kainic acid (KA) induced experi-

mental seizures (Guo 1993). Other herbs, such as tianma (Gastro-dia rhizome), changpu ( Acorus calamus ) anddannanxing ( Arisaema cum bile ), were also shown to have anticonvulsant effects ( Wang 1996). Allthese herbs are usually used in Chinese medicine for the

treatment of epilepsy. Although the active ingredient for each herb

is not known, there are both clinical reports and animal studies

that suggest the prescription may inhibit seizures ( Wang 1996).

TW970, a modified formula of the Chinese medicine ’chaihu-

longu-muli-tang’ can reduce the seizure frequency in patients with

refractory epilepsy and this may be due to antioxidant effects

(Hung 2002). Zheng tai instant powder is a complex prescription

of traditional Chinese medicine, indicated for generalized tonic

clonicseizure dueto itseffecton thecentral nervous system, energy

metabolism, neurotransmitters and haemorheology (Hu 2000).

TCM has been widely used for many years to treat epilepsy and

there is a great deal of clinical literature. However, the benefits and

adverse reactions of these treatments have not been systematically

reviewed, partly because there is no standard dose of active ingre-dients and prescriptions vary between patients. This review aims

to summarize the existing evidence on the comparative effective-

ness and safety of traditional Chinese medicinal herbs for treating

epilepsy.

O B J E C T I V E S

To assess whether traditional Chinese medicinal herbs (TCMHs)

(one formulation) for the treatment of people with epilepsy of any

type and syndrome:

1. are associated with a greater probability of patients

becoming seizure free, or having a significant reduction in the

frequency or duration of seizures (or both); or

2. are associated with adverse events

when compared with other antiepileptic drugs (AEDs), placebo

or no treatment.

M E T H O D S

Criteria for considering studies for this review

Types of studies

Randomised studies using adequate or quasi (i.e. day of the week)

methods of randomisation. Studies may be single blinded, double

blind or unblinded.

Types of participants

People of any age, either gender and of any ethnic origin with

epilepsy of any type and syndrome(ILAE 1981; ILEA 1989), at

any stage of diagnosis.

Types of interventionsTraditional Chinese medicinal herbs (TCMHs) are defined as

preparations derived from plants, or parts of plants, including sin-

gle herbs or mixtures of different herbs. We included any types

of preparation, such as decoction, oral liquid, tablet, capsule or

powder. We also included single chemicals extracted from a plant,

or synthetic chemicals based on plant constituents.

We considered the following comparisons:

1. one formula of TCMH versus no interventions;

2. one formula of TCMH versus placebo;

3. one formula of TCMH versus single Western medication

(monotherapy).

We excluded studies of TCMH in combination with antiepileptic

drugs (AEDs) versus AEDs alone. We will include these studies in

the update of this review.

Types of outcome measures

Primary outcomes

1. Seizure freedom for at least 48 weeks (Glauser 2006).

2. Satisfactory seizure control: 50% or greater reduction in

seizure frequency.

3. Absolute or percentage reduction in seizure frequency and

duration.

Secondary outcomes

1. Improved quality of life, if assessed by standardized,

validated, reliable scales.

2. Incidence of adverse or harmful effects:

i) organ damage: irreversible end organ damage;

ii) cognitive side effects;




6/31

iii) allergic reactions (e.g. skin rashes, Steven Johnson

syndrome).

3. Withdrawals due to side effects or due to lack of efficacy.

Search methods for identification of studies

We searched the Cochrane Epilepsy Group’s Specialised Register,

the Cochrane Central Register of Controlled Trials (CENTRAL)

(The CochraneLibrary 2007, Issue1), MEDLINE on OVID (1950to April week 1, 2007), EMBASE (1974 to June 2007), CINAHL

(1982 to April week 1, 2007), AMED (the Allied and Comple-

mentary Medicine Database, 1985 to April 2007), the China Bi-

ological Medicine Database (CBM-disc, 1979 to June 2007) and

the VIP Chinese Science and Technique Journals Database (1989

to June 2007). The search terms used to search MEDLINE are set

out in Appendix 1.

We handsearched the Journal of Chinese Integrative Medicine , Chi-

nese Medical Journal , Chinese Journal of Internal Medicine , Chi-nese Journal of Integrated Traditional and Western Medicine , Chi-nese Journal of Integrative Medicine and other Traditional ChineseMedicine journals.

We searched the reference lists of all relevant papers for further

studies. In addition, we contacted colleagues and experts in the

field to identify any unpublished or ongoing studies. There was

no language restriction either in the search or in the inclusion of

studies.

Data collection and analysis

Selection of studies

To determine the studies to be assessed further, two authors (QL

and DZ) independently scanned the titles, abstract sections and

keywords of every record. The two authors retrieved full articles

for further assessment if the information given suggested that the

study:

1. included participants with epilepsy;

2. compared Chinese herbal medicine with any other active or

placebo intervention;

3. used random or quasi-random allocation to the comparison

groups.

If there was any doubt regarding these criteria from the informa-

tion given in the title and abstract, we retrieved the full article for

clarification. Two authors (QL and DZ) independently assessed

each of these studies for inclusion and there were no recorded dis-

agreements. We excluded studies that did not meet the inclusion

criteria and have stated the reasons in ’Characteristics of excluded

studies’.

Data extraction and management

Two review authors (QL and DZ) independently assessed included

studies for quality and any differences of opinion were resolved by

the mutualconsent.The followingdatawere extracted,if available.

1. General information

i) Published/unpublished; title; authors; reference/source; contact address; country; urban/rural etc.; language of

publication; year of publication; duplicate publications; sponsor;

setting.

2. Study methods i) Method of generation of random list.

ii) Method of concealment of randomisation.

iii) Stratification factors.

iv) Blinding methods.

3. Participants i) Number (total and per group).

ii) Sex and age distribution.

iii) Seizure types and epilepsy syndrome.

iv) Time between first seizure and randomisation.v) Aetiology of epilepsy.

vi) Presence of neurological signs.

vii) Number and types of antiepileptic drugs (AEDs)

taken.

4. Intervention and control i) Type of herbal medicines.

ii) Details of treatment regime (dose, route, timing).

iii) Type of control.

iv) Details of control treatment (dose, route, timing).

5. Follow-up data i) Duration of follow up.

ii) Dates of treatment withdrawal and reasons for

treatment withdrawal.iii) Withdrawal rates.

6. Outcome data i) As described in the types of outcome measures

7. Analysis data i) Methods of analysis (intention-to-treat and per

protocol analysis).

ii) Comparability of groups at baseline (yes or no).

Assessment of risk of bias in included studies

Two review authors independently assessed included studies for

quality using two methods.

The Cochrane approach was used to assess allocationconcealment.

The method of allocation concealment in each trial was scored

using the following grading system:

Grade A: adequate concealment;

Grade B: uncertain concealment;

Grade C: inadequate concealment;

Grade D: allocation concealment not used.




7/31

We also divided studies into the following three categories accord-

ing to the simple approach set out in The Cochrane Handbook for Systematic Reviews of Interventions (Handbook 2008). The quality criteria include blinding and losses to follow up as well as alloca-

tion concealment.

A - all quality criteria met: low risk of bias.B - one or more of the quality criteria only partly met: moderate

risk of bias.

C - one or more criteria not met: high risk of bias.

The description of the quality of each study was given based on

these components. Two review authors (QL and DZ) indepen-

dently assessed each trial selected for inclusion in the review. There

were no recorded disagreements.

Data synthesis

We analysed data using Review Manager 5.0 (RevMan 2008). We

used relative risks with 95% confidence intervals (CI) for binary

outcomes. We used standardised mean differences with 95% CI

for continuous outcomes. Where possible all analyses included all

participants in the treatment groups to which they had been al-

located. For the efficacy outcome (50% or greater reduction in

seizure frequency), we planned to undertake three analyses: (1)

primary (intention-to-treat) analysis; (2) worst case; (3) best case,

to address the potential problem of missing outcome data. How-

ever, not all included studies reported loss to follow up data, so we

were unable to perform these analyses.

Weassessed clinical heterogeneity by comparingthe distributionof

important participant factors between studies (age, gender, seizure

type, duration of epilepsy, number of AEDs taken at time of ran-

domisation) and study factors (randomisationconcealment, blind-

ing, losses to follow up). We assessed statistical heterogeneity with

theI2 statistic; a value greater than 50%was considered to indicate

significant heterogeneity. However, the preparation and composi-

tion of herbal medicines in each study varied and there was clear

heterogeneity of intervention, so meta-analysis was not attempted.

We planned to assess the impact of important patient characteris-

tics including seizure type, duration and aetiology of epilepsy, and

presence of neurological signs upon outcome, assuming sufficient

data were available. We also planned to undertake sensitivity anal-

yses including: (i) all studies; (ii) only those using adequate meth-

ods of allocation concealment, such as sealed opaque envelopes

or telephone randomisation. For the analysis of adverse events, as

stated above, we calculated Peto odds ratios (ORs) with 99% CIs

for the comparison between randomised groups.

R E S U L T S

Description of studies

See: Characteristicsof included studies; Characteristicsof excluded

studies; Characteristics of studies awaiting classification.

Studies identified

We initially found 16 controlled studies of traditional Chinese

medicine (TCM) assessing the treatment of epilepsy. We excluded

nine studies for the following reasons (see also the ’Characteristics

of excluded studies’). In four studies (Chen 1998; Li 1997; Wang

2001; Xu 2005) the diagnostic criteria for epilepsy were unclear

or did not meet the ILAE (International League Against Epilepsy)

classifications. Three studies (Lin 1999; Liu 1994b; Yin 2001)

were excluded according to our preset that AEDs have been used

in combination with traditional Chinese herbs.Twostudies (Chen

2000; Wu 2002) used healthy volunteers as the control group.

Another two studies (Ma 2003a ; Ma 2003b) are currently awaiting

assessment for inclusion in this review. For one study (Ma 2003a )

the study design was unclear and we were not able to contact the

author for further information. The other trial report (Ma 2003b)

mentionedrandomisationbut patients were allocatedin a 6:1 ratio

to treatmentor control. We thereforedecided to contact theauthor

for further information about the randomisation process, but we

have been unable to contact them. Ultimately, five studies (Liu

1994a ; Song 2001; Tian 2006; Xiang 1998; Xin 1999) fulfilled

our inclusion criteria.Details are given in the table ’Characteristics

of included studies’ andare summarised below. Allthe studies were

conducted in China and were published in Chinese. We did not

find any ongoing studies.

Design

All the included studies were of single centre, parallel design and

had a control group. All of them mentioned randomisation but

the randomisation procedures were unclear. None of the studies

were blinded and their duration ranged from two months to three

years.

Participants

The number of participants in the studies ranged from 40 to

401, with a total of 1125 in the five studies in this review. The

age of participants ranged from two to 70 years with mean ages,

where given, ranging from 6.25 to 30.84 years. All studies had

adults and children as participants, except Xin 1999, which in-

cluded only children aged from three to 13 years old. Two studies

(Liu 1994a ; Xiang 1998) only included primary generalized tonic-

clonic seizure and one study ( Xin 1999) included both primary

and secondary generalized tonic-clonic seizure. The remainingtwo

(Song 2001; Tian 2006) included all partial or generalized. All

the included participants had a verified diagnosis of ’Xian Zheng’

according to the TCM system and four studies (Liu 1994a ; Tian

2006; Xin 1999; Song 2001) also required the detailed types in

’Xian Zheng’, according to the criteria of the China Ministry of




8/31

Health. All studies except one ( Xin 1999) reported the time be-

tween first seizure and randomisation, and this ranged from three

months to more than 30 years. None of the studies mentioned

whether the included participants were newly diagnosed. Only

one study (Tian 2006) mentioned that most of the participants

had antiepileptic drug monotherapy or biotherapy history beforeentering the studies, but did not give details. The baseline seizure

frequency was reported in three studies (Tian 2006; Xiang 1998;

Xin 1999), and all the three studies included the patients who

had a seizure at least once a month. Although none of the studies

reported the aetiologies of epilepsy, three (Liu 1994a ; Tian 2006;

Xiang 1998) excluded secondary or symptomatic epilepsy caused

by stroke, trauma or tumour. Neurological signs were not men-

tioned in any of the studies.

Interventions

The preparation and composition of herbal medicines in each

study varied and each trial compared one formulation of TCMH

with a single antiepileptic drug (see Table 1). Several herbs, such as

changpu ( Acorus calamus ), dannanxing ( Arisaema cum bile), banxia (Rhizoma pinelliae ), fu ling (Poria ), yuanzhi (Polygala ) and tianma (Gastrodia rhizome) were tested. TCMH was given in the formof patient prescription in four studies (Liu 1994a ; Song 2001;

Tian 2006; Xin 1999 ). Two studies (Tian 2006; Xin 1999) used

capsules,one granule (Liu 1994a )andonepill(Song 2001).All the

tested herbs were preparedby the trial authors’ hospital, except the

’antiepilepsy capsule’, which was produced by a pharmaceutical

company.

The different intervention and comparators are listed below.

Liu 1994a compared Xiaxingci granule with phenytoin.

Song 2001 compared Dianxianning pill with valproate.

Tian 2006 compared Tianmadingxian capsule with phenytoin.

Xiang 1998 compared Zhixian I with phenytoin.

Xin 1999 compared ’Antiepilepsy capsule’ with phenobarbital.

The duration of treatment was two months ( Xiang 1998), 12

weeks (Liu 1994a ) and three months (Song 2001; Tian 2006; Xin

1999).

Outcomes measures

Two studies (Liu 1994a ; Song 2001) used seizure freedom as an

outcome measure. All studies reported the number of participants

with percentage reduction in seizure frequency (such as 50% or

greater, 75% or greater) as one of the major outcomes. All studies

except one (Tian 2006) also reported the electroencephalogram

(EEG) results before and after treatment. One study ( Xin 1999)

reported the number of participants with good (75% or greater)

or moderate (50% to 74%) reduction in seizure duration. None of

the studies mentioned health-related quality of life as outcomes.

All studies except one (Tian 2006) assessed adverse events and

reported the incidence in detail. None of the studies reported

withdrawal due to side effects or due to lack of efficacy.

Risk of bias in included studies

All of the included studies were of low quality (’C’). An overview of

study quality can be found inthe table ’Characteristics of included

studies’.

Randomisation

Although all the studies mentioned that the patients were ran-

domly allocated to the intervention and control groups, none of

them described the allocation sequence in detail. None of the

included studies mentioned allocation concealment. We tried to

contact theauthors of studies by telephonein order to obtain more

detailed information, however we were unable to reach them. We

also wrote to the authors but did not receive any replies.

Blinding

None of the studies mentioned blinding.

Description of withdrawals, losses to follow up and

intention-to-treat analysis

A follow up of at least two months was used. One study (Liu 1994a )

had a follow up of 12 weeks, but in the outcomes mentioned a

seizure free interval of one year, which seems to be contradictory.

None of the studies described withdrawals and losses to follow up

or performed intention-to-treat analysis.

Similarity of comparison groups at baseline

All studies provided data on important baseline characteristics of

the intervention and control groups to judge the comparability.

Two studies (Tian 2006; Xin 1999) claimed that the two groups

were comparable at baseline. Age and gender appeared to be bal-

anced in all studies and other baseline characteristics were partly

provided. The time between first seizure and randomisation was

reported in all except one ( Xin 1999). The baseline seizure fre-

quency was reported in two studies (Tian 2006; Xin 1999).

Effects of interventionsThe five studies tested five different traditional Chinese medicines

compared with antiepileptic drugs (AEDs), and the reported out-

comes included seizure freedom, 50% or greater reduction in

seizure frequency, percentage reduction in seizure frequency and

duration, improvement in EEG and adverse effects. We did not

perform a meta-analysis due to heterogeneity in interventions and

controls.




9/31

Primary outcome measures

I. Xiaxingci granule versus phenytoin

We found one study (Liu 1994a ) comparing Xiaxingci granule with placebo, which recruited 40 participants. All participants

were diagnosed with primary generalised tonic-clonic seizure.

(1) Seizure freedom

The study described the number of participants who were seizure

free for one year. No statistically significant difference was found

between Xiaxingci granule and phenytoin in this study (RR 1.00;

95% CI 0.07 to 14.90, P = 1.00).

(2) 75% or greater reduction in seizure frequency

Xiaxingci granule was no better or worse than phenytoin (RR 1.50;

95% CI 0.28 to 8.04, P = 0.64).


Xiaxingci granule was no better or worse than phenytoin (RR 1.00;

95% CI 0.81 to 1.23, P = 1.00).

II. Dianxianning pill versus valproate

We found one study (Song 2001) comparing Dianxianning pill

with valproate, which recruited 150 participants. All participants were diagnosed with epilepsy according to the classifications of

ILEA 1989.

(1) Seizure freedom

The study described the number of participants who were seizure

free for one year. There was no statistically significant difference

(RR 13.00; 95% CI 0.74 to 227.72, P = 0.08).


A statistically significant difference was found favouring Dianxi-

anning pill (RR 2.73; 95% CI 2.00 to 3.74, P < 0.00001).


Dianxianning pill was no better or worse than valproate (RR 0.98;

95% CI 0.92 to 1.04, P = 0.52).

III. Tianmadingxian capsule versus phenytoin

We found one study (Tian 2006) comparingTianmadingxian cap-

sule with phenytoin, which recruited 334 participants. All partici-

pants were diagnosed with epilepsy according to the classifications

of ILEA 1989.


There was a statistically significant difference between the treat-

ment and control groups favouring Tianmadingxian capsule (RR

1.37; 95% CI 1.23 to 1.53, P < 0.00001).


The study showed a statistically significant difference between

Tianmadingxian capsule and phenytoin, favouring Tianmadingx-

ian capsule (RR 2.38; 95% CI 1.83 to 3.09, P < 0.00001).



Tianmadingxian capsule and phenytoin, favouring Tianmadingx-

ian capsule (RR 1.19; 95% CI 1.10 to 1.28, P < 0.00001).

IV. Zhixian I versus phenytoin

We found one study ( Xiang 1998) comparing Zhixian I with

phenytoin, which recruited 200 participants. All participants were

diagnosed with primary generalized tonic-clonic seizure.


The study used Zhixian I versus phenytoin and a statistically sig-

nificant difference was found favouring Zhixian I (RR 1.31; 95%

CI 1.16 to 1.48, P < 0.0001).



Zhixian I and phenytoin, favouring Zhixian I (RR 1.42, 95% CI

1.20 to 1.68, P < 0.0001).



Zhixian I and phenytoin, favouring Zhixian I (RR 1.10; 95% CI

1.03 to 1.18, P = 0.006).




10/31

V. ’Antiepilepsy capsule’ versus phenobarbital

We found one study ( Xin 1999) comparing ’Antiepilepsy capsule’

with phenobarbital, which recruited 401 participants. All partici-

pants werediagnosed withprimary generalizedtonic-clonicseizure

and partial secondary tonic-clonic seizure.


We found that there was a slight difference between antiepilepsy

capsule and phenobarbital, favouring ’Antiepilepsy capsule’ (RR

1.21; 95% CI 1.02 to 1.43, P = 0.03).


The study reported a statistically significant difference between

antiepilepsy capsule versus phenobarbital, favouring antiepilepsy

capsule (RR 1.46; 95% CI 1.13 to 1.88, P = 0.003).

(3) 75% or greater reduction in seizure duration

Only this study reported the number of participants with 75% or

50% or greater reduction in seizure duration. We found that there

was no difference between ’Antiepilepsy capsule’ and phenobarbi-

tal for the outcome 75% or greater reduction in seizure duration

(RR 1.29; 95% CI 0.98 to 1.69, P = 0.07).

(4) 50% or greater reduction in seizure duration

The participants treated with’Antiepilepsy capsule’ seemedto have

more likelihood of achieving a 50% or greater reduction in seizure

duration (RR 1.25; 95% CI 1.05 to 1.48, P = 0.01).

Secondary outcome measures

(1) Quality of life measures

Quality of life measures were notrecorded in any studies, therefore

they could not be examined.

(2) Incidence of adverse or harmful effects

All studies except one (Tian 2006) mentioned adverse events (see

’Characteristics of included studies’) and two studies (Liu 1994a ;

Song 2001) showed no obvious adverse effect in the TCM group.

The other two studies ( Xiang 1998; Xin 1999) reported only slight

gastrointestinal discomfort after TCM. Xin 1999 reported the

incidence of adverse effects both in the ’Antiepilepsy capsule’ and

phenobarbital group and the Peto odds ratio was 0.04 (99% CI

0.01 to 0.12, P < 0.00001), favouring the ’Antiepilepsy capsule’.

None of the studies reported any cognitive side effects, allergic

reaction skin rashes or Steven Johnson syndrome.

(3) Withdrawals due to side effects or due to lack of efficacy

Since none of the studies reported loss to follow up, we could not

analyse thewithdrawalsdue to side effectsor dueto lack of efficacy.

Subgroup and sensitivity analysis

None of the studies contained stratified data and individual raw

data on aetiology, duration and frequency were not available. We

were therefore unable to analyse studies according to different ae-

tiology, duration or frequency. Five studies were included and no

study had the highest possible quality, with adequate concealment

of randomisation, therefore we did not perform the planned sen-

sitivity analysis.

D I S C U S S I O N

Summary

In this review we have included data from 1125 patients from five

studies in which participants wererandomised to eithertraditional

Chinese medicine (TCM) or antiepileptic drugs (AEDs). All the

studies were of low methodological quality. No studies were con-

ducted using adequate methods of randomisation, allocation con-

cealment or blinding of both investigators and patients. They pro-

vide only limited descriptions of baseline and follow up data. All

studies stated that random assignment was used, but there was in-

sufficient information to judge whether or not this was conducted

properly. All the included studies therefore had a high probability

of selection, detection and performance bias. Although the studiesdid report some benefit (in terms of patients becoming seizure free

or having a 50% or greater reduction in seizure frequency), we

are unable to draw a reliable conclusion on the effect of TCM for

epilepsy. No adverse effects were found in two studies and only

slight gastrointestinal discomfort was reported in the two other

studies reporting this.

Limitation of the included studies

The included studies were of poor methodological quality in terms

of randomisation, allocation concealment and blinding. We will

make further attempts to contact the relevant authors for more

detailed information on allocation concealment and randomisa-

tion as this review is updated in the future.

The included studies were quite heterogeneous in their popula-

tions. The age of participants varied greatly, ranging from 2 to 70,

but only one study ( Xin 1999) focused on children. The effects of

TCM on different age groups needs further investigation.

None of studies used a placebo control. There is marked clinical

heterogeneity due to differences in TCMformulations.Four tested




11/31

herbs were prepared by the authors’ hospital and there may be

potential conflict of interest in promoting the therapeutic effect of

TCM. There is still a lack of information about quality standards

for the development of herbal preparations and the manufacture

of the herbal products. Future studies should provide information

about standardisation including compositions, quality control anddetailed regimens.

Theprimarygoalof treatmentfor epilepsy isthe control of seizures,

or even for patients to become seizure free, however the follow-up

period used in the studies was quite unclear or short. Most of the

included studies reported end-of-treatment responses and there is

a lack of long-term follow-up data. Some important outcomes,

such as percentage reduction of seizure duration, quality of life

and cognitive side effects, were assessed by only a single study or

none.

Although the study results seem to show TCM is a relatively safe

treatment modality, we cannot be assured of the safety of TCM for

epilepsy patients since the small sample sizes might have limited

power to detect rare adverse effects. The safety of using TCM isstill to be proved.

No studies reported a sample size calculation, which is essential

for ensuring adequate sample size and statistical power. None re-

ported the loss to follow up, and there may be selective reporting

of outcome data. None of included studies mentioned ethical is-

sues or whether the participants gave informed consent.

Limitations of this systematic review

Although we conducted comprehensive searches, only Chinese

language publications were found and included and no multi-centre, large-scale randomised controlled trial (RCT) was identi-

fied. We undertook extensive searches for unpublished material

and few of the studies identified qualified for inclusion, but at the

same time we cannot disregard the fact that studies with negative

findings may remain unpublished.

TCM is widely used for treating epilepsy in China. In this re-

view we only included studies of TCM versus AEDs, according

to our protocol. We excluded some RCTs where TCM in combi-

nation with AEDs was compared with AEDs. In addition, studies

of traditional Chinese medicinal herbs (TCMHs) compared with

acupuncture for epilepsy are not included. These studies may be

included in a future review.

Although TCM is widely accepted in China, there are still somecomplicated problems that need to be noted. Chinese herbal

medicines are quite often composed of mixtures of up to 20 dif-

ferent herbs. They are also sometimes customised by practitioners

for each individual patient, based on the differentiation of the pa-

tient’s ’symptoms’. Study designs could be adapted to this ’individ-

ualised treatment’ by stratification of practitioners or by the pat-

tern of the ’syndromes’. However, most of the constituents of the

pharmacological preparations cannot be specified precisely. There

may be variation between different formulations and batches of

treatment, although the variation is an inevitable consequence of

the nature of TCM and the Chinese government does specify thelimits of variation that are acceptable. This variation is a factor

that may contribute to any heterogeneity between different study

results, making meta-analysis difficult.

A U T H O R S ’ C O N C L U S I O N S

Implications for practice

The current evidence is insufficient to support traditional Chi-

nese medicine (TCM) as a treatment for epilepsy. Much larger,

high quality randomised clinical trials are needed to evaluate the

effectiveness and safety of traditional Chinese medicinal herbs for

treating epilepsy.

Implications for research

More high quality randomised controlled studies following the

CONSORT criteria (Moher 2001) are required to assess the ef-

fects of traditional Chinese medicine for epilepsy. Future clinical

studies should include a broad age range of patients, and have a

very large sample size to delineate the effect of TCM on patients

with different seizure types. Relevant clinical events, such as being

seizure free for at least 48 weeks, a 50% or greater reduction in

seizure frequency and duration, or quality of life, should be in-

cluded as outcomes assessed in these studies. A longer-term fol-low up should be used to determine the genuine effectiveness of

TCM and its long-term effects. Adverse reactions should be rigor-

ously investigated to assess safety. Future studies should improve

the description of herbalmedicines being tested, e.g. plant species,

geographical origin, harvest season, preparation procedures and

quality of the products.

A C K N O W L E D G E M E N T S

We would like to acknowledge the Cochrane Epilepsy Group for

their technical support. We thank Professor Tony Marson andRachael Kelly (Cochrane Epilepsy Group), Dr Sridharan Rama-

ratnam (Department of Neurology, Apollo Hospitals, India), Pro-

fessor Paula Williamson (University of Liverpool) and Professor

Taixiang Wu (Chinese Cochrane Center, China) for their kind

advice and encouragement in the preparation of this review.




12/31

R E F E R E N C E S

References to studies included in this review

Liu 1994a {published data only}

Liu QS, Long JX, He ZY. The clinical observation of

Xiaxingci granule of treatment for primary generalized

tonic-clonic seizure. Hu Nan Zhong Yi Xue Yuan Xue Bao

(Journal of Hunan College of Traditional Chinese Medicine)1994;14(4):33–4.

Song 2001 {published data only}

Song PX. Clinical study of Dianxianning for 100 epilepsy

cases. Shang Dong Zhong Yi Za Zhi (Shandong Journal of

Traditional Chinese Medicine) 2001;20(5):338–9.

Tian 2006 {published data only}

Tian ZH. Clinical observation of Tianmadingxian capsule

on epilepsy. He Bei Zhong Yi (Hebei Journal of Traditional

Chinese Medicine) 2006;28(7):489–91.

Xiang 1998 {published data only}

Xiang JT, Jiang HY. Zhixian I for tonic-clonic seizure. HuNan Zhong Yi Yao Dao Bao (Hunan Guiding Journal of

TCMP) 1998;4(11):19–20.

Xin 1999 {published data only}

Xin L, Wei YL, Ye QF, Qu ZQ. Anti-epilepsy capsule for

301 cases of children epilepsy. Liao Ning Zhong Yi Za Zhi

(Liaoning Journal of Traditional Chinese Medicine) 1999;26

(11):496–7.

References to studies excluded from this review

Chen 1998 {published data only}

Chen YT. Dingxian prescription treatment for 145 epileptic

cases. Nan Jing Zhong Yi Yao Da Xue Xue Bao (Journal of

Nanjing University of TCM) 1998;14(4):246.

Chen 2000 {published data only}

Chen LC, Chou MH, Lin MF, Yang LL. Lack of

pharmacokinetic interaction between valproic acid and a

traditional Chinese medicine, Paeoniae Radix, in healthy

volunteers. Journal of Clinical Pharmacy & Therapeutics

2000;25(6):453–9.

Li 1997 {published data only}

Li Y. 153 cases of epilepsy treated by modified capsule of

five tigers powder fexpelling wing-evill. Guo Yi Lun Tan

(Forum on Traditional Chinese Medicine) 1997;12(4):26–7.

Lin 1999 {published data only}

Lin YL, Zhuang HC, Lin ZX, Xie TL, Zhan JW. The

therapeutic effect of Ningxian capsule treatment for 180epileptic cases. Fu Jian Yi Yao Za Zhi (Fu Jian Medical

Journal) 1999;21(6):67–8.

Liu 1994b {published data only}

Liu YX, Wang MZ, Sun MZ, Niu SX, Sun XF, Li JK.

Clinical analysis of 72 epileptic patients treated with alkaline

extract of Euphorbia Fisheriana. Zhong Guo Zhong Xi Yi Jie

He Za Zhi (Chinese Journal of Integrated Medicine) 1994;4

(5):282–4.

Wang 2001 {published data only}

Wang FL, Zhang MM, Wang FC, Wang ZH. The study of

Dingxian powder treatment for epilepsy. Liao Ning Zhong

Yi Za Zhi (Liaoning Journal of TCM) 2001;28(12):743.

Wu 2002 {published data only}

Wu HM, Liu CS, Tsai JJ, Ko LY, Wei YH. Antioxidant

and anticonvulsant effect of a modified formula of Chaihu-

Longu-Muli-Tang. American Journal of Chinese Medicine

2002;30:339–46.

Xu 2005 {published data only}

Xu Zi. The observation of Ditantongguan potion treatment

for 23 pediatric epilepsy. Xin Jiang Zhong Yi Yao (Xinjiang

TCM) 2005;23(3):20–1.

Yin 2001 {published data only}

Yin HJ, Guo ZY. Clinical study of Ginkgo biloba extract

in the treatment of epilepsy. Yi Yao Dao Bao (Herald of

Medicine) 2001;20(2):98.

References to studies awaiting assessment

Ma 2003a {published data only}

Ma R, Zhang XL. The clinical observation of Xifeng capsule

treatment for 200 pediatric tonic-clonic epilepsy. Journal of

Traditional Chinese Medicine 2004;45(5):363–5.

Ma 2003b {published data only}

Ma R, Li S, Li X, Hu S, Sun X, Liu Y, et al.Clinical

observation on 930 child epilepsy cases treated with anti-

epilepsy capsules. Journal of Traditional Chinese Medicine 2003;23(2):109–12.

Additional references

Cockerell 1995

Cockerell OC, Johnson L, Sander JWAS, Hart YM,

Shorvon SD. Remission of epilepsy: results from the

National General Practice Study of Epilepsy. Lancet 1995;

346:140–4.

Fulder 1996

Fulder S. The Handbook of Alternative and Complementary

Medicine . Oxford: Oxford University Press, 1996.

Glauser 2006

Glauser T, Ben-Menachem E, Bourgeois B, Cnaan A,

Chadwick D, Guerreiro C, et al.ILAE treatment guidelines:

evidence-based analysis of antiepileptic drug efficacy and

effectiveness as initial monotherapy for epileptic seizures

and syndromes. Epilepsia 2006;47(7):1094–120.Guo 1993

Guo Q, Kuang P. Effect of qingyangshen on hippocampal

alpha- and beta-tubulin gene expression during kainic acid

induced epileptogenesis. Journal of Traditional Chinese

Medicine 1993;13(4):281–6.

Handbook 2008

Higgins JPT, Green S, editors. Cochrane Handbook

for Systematic Reviews of Interventions 5.0.0 [updated




13/31

February 2008]. The Cochrane Collaboration, 2008.

Available from www.cochrane-handbook.org.

Hauser 1993

Hauser WA, Annegers JF, Kurland LT. Incidence of epilepsy

and unprovoked seizure in Rochester, Minnesota:1935-

1984. Epilepsia 1993;34(3):453–68.

Hsieh 1999a

Hsieh CL, Chen MF, Li TC, Li SC, Tang NY, Hsieh CT.

Anticonvulsant effect of Uncaria rhynchophylla (Miq) Jack.

in rats with kainic acid-induced epileptic seizure. American

Journal of Chinese Medicine 1999;27:257–64.

Hsieh 1999b

Hsieh CL, Tang NY, Chiang SY, Hsieh CT, Lin JG.

Anticonvulsive and free radical scavenging actions of two

herbs, Uncaria rhynchophylla (MIQ) Jack and Gastrodia

elata Bl., in kainic acid-treated rats. Life Sciences 1999;65(20):2071–82.

Hu 2000

Hu H, Zhou Y, Sui Y, Qi M. An experimental study of

effect of zheng tai instant powder on grand mal epilepsy. Journal of Traditional Chinese Medicine 2000;20:210–5.

Hung 2002

Hung-Ming W, Liu CS, Tsai JJ, Ko LY, Wei Y. Antioxidant

and anticonvulsant effect of a modified formula of chaihu-

longu-muli-tang. American Journal of Chinese Medicine

2002;30:339–46.

ILAE 1981

Commission on Classification and Terminology of the

International League Against Epilepsy. Proposal for revised

clinical and electroencephalographic classification of

epileptic seizures. Epilepsia 1981;22(4):489–501.

ILEA 1989

Commission on Classification and Terminology of theInternational League Against Epilepsy. Proposal for revised

classification of epilepsies and epileptic syndromes. Epilepsia

1989;30:389–99.

Lai 1991

Lai C W, Lai Y H. History of epilepsy in Chinese traditional

medicine. Epilepsia 1991;32(3):299–302.

Liu 2000

Liu JP, McIntosh H, Lin H. Chinese medicinal herbs for

chronic hepatitis B. Cochrane Database of Systematic Reviews 2000, Issue 4. [Art. No.: CD001940. DOI: 10.1002/

14651858.CD001940]

Minami 1999

Minami E, Shibata H, Nunoura Y, Nomoto M, Fukuda

T. Efficacy of shitei-to, a traditional Chinese medicine

formulation, against convulsions in mice. American Journal

of Chinese Medicine 1999;27(1):107–15.

Moher 2001

Moher D, Schulz KF, Altman D. The CONSORT

statement: revised recommendations for improving the

quality of reports of parallel-group randomized trials. JAMA

2001;285(15):1987–91.

RevMan 2008The Nordic Cochrane Centre. The Cochrane Collaboration.

Review Manager. 5.0. Copenhagen: The Nordic Cochrane

Centre. The Cochrane Collaboration, 2008.

Sander 1996

Sander JW, Shorvon SD. Epidemiology of the epilepsies.

Journal of Neurology, Neurosurgery, and Psychiatry 1996;61(5):433–43.

Villeneuve 2004

Villeneuve N. Quality-of-life scales for patients with drug-

resistant partial epilepsy. Reviews of Neurology (Paris) 2004;

160:376–93.

Wang 1996

Wang Q. Advances in treatment of epilepsy with traditionalChinese medicine. Journal of Traditional Chinese Medicine

1996;16(3):230–7.∗ Indicates the major publication for the study




14/31

C H A R A C T E R I S T I C S O F S T U D I E S

Characteristics of included studies [ordered by study ID]

Liu 1994a

Methods DESIGN: randomised controlled trial

RANDOMISATION PROCEDURE: unclear

BLINDING: not used

DURATION OF FOLLOW UP: 12 weeks

Participants COUNTRY AND SETTING: China; Changsha city, Hunan Province.

NUMBER AND SEX: treatment group (males): 20 (11); control: 20 (12).

AGE: treatment: 4 to 53 years old (mean 22.8 years); control: 5 to 53 years old (mean 25.7 years)

SEIZURE TYPES: primary generalized tonic-clonic seizure and verified diagnosis of ’wind phlegm types’

of ’xian zheng’ in TCM

NEWLY DIAGNOSED: unclear

BASELINE SEIZURE FREQUENCY: no mention

TIME BETWEEN FIRST SEIZURE AND RANDOMISATION: treatment: 0.5 to 20 years (mean 6.

5 years); control: 0.5 to 18 years (mean 7.6 years)

Interventions Both groups are monotherapy

Treatment group: Xiaxingci granule 4 g p.o, t.i.d

Control group: phenytoin total 10 mg/kg/d, 3 times a day

Both groups were 4 weeks as a course and 3 courses in total and there was a week for a rest between the

two courses

Outcomes SEIZURE FREEDOM: seizure free for 1 year: treatment 1; control 1

REDUCTION IN SEIZURE FREQUENCY (number of patients): 75% or greater reduction (treatment

3, control 2); 25% to 75% reduction (treatment 14, control 15); less than 25% reduction (treatment 2,

control 2)

IMPROVEMENT RATE IN EEG: treatment group 15%; control group 10%

ADVERSE EVENTS: both groups had no obvious effect on blood routine examination, hepatic function

and renal function

Notes (1) There was potential conflict of interest in the use of Xiaxingci granules made by the authors’ hospital

(2) The course and therapy was irregular

Risk of bias

Item Authors’ judgement Description

Allocation concealment? Unclear B - Unclear




15/31

Song 2001



BLINDING: not used

DURATION OF FOLLOW UP: 90 days to 1 year

Participants COUNTRY AND SETTING: China; Taiyuan city, Shandong Province

NUMBER AND SEX: treatment group (males): 100 (46), control: 50 (28).

AGE: treatment group: 2 to 45 years old (2 to 15 years (60 patients), 20 to 30 years (30 patients), 35 to

45 years (10 patients)). Control: 3 to 45 years old (3 to 15 years (30 patients), 20 to 30 years (15 patients)

, 35 to 45 years (5 patients))

SEIZURE TYPES (number of patients): classifications in ILAE 1989 and verified diagnosis of ’xian zheng’

in TCM. GTCS: treatment (35), control (15); absence: treatment(35), control (15); simple partial seizure:

treatment (10), control (5); complex partial seizure: treatment (10), control (5); myoclonic seizures:

treatment (5), control (5); and atonic seizures: treatment (5), control (5)


BASELINE SEIZURE FREQUENCY: no mention

TIME BETWEEN FIRST SEIZURE AND RANDOMISATION: 1 to 30 years (mean 9.87 years)

unclear in each group

Interventions Both group are monotherapy

Treatment: Dianxianning pill 4 g for adults and 2.5 g for children 3 times daily for 90 days

Control: valproate 0.4 g for adults and 0.2 g for children 3 times daily for 90 days

Outcomes SEIZURE FREEDOM: treatment group seizure free (6), control group (0)

SEIZURE FREQUENCY (number of patients): treatment: 80% or greater reduction (82), 60% to 79%

reduction (12); control: 80% or greater reduction (30), 60% to 79% reduction (66)

IMPROVEMENT IN EEG: treatment: 100 (100%) obviously improved; control: 20 (40%) obviously

improved, 18 (36%) improved and 12 (24%) unchanged

ADVERSE EVENTS: treatment group had no obvious adverse events; control: 5 had gastrointestinalreaction and 2 had fatigue and drowsiness

Notes There was potential conflict of interest in the use of ’Dianxianning pill’ made by the authors’ hospital

Risk of bias



Tian 2006



BLINDING: not used

DURATION OF FOLLOW UP: 90 days

Participants COUNTRY AND SETTING: China; Shijiazhuang city, Hebei Province

NUMBER AND SEX: treatment group (males): 178 (87), Control: 156 (76)AGE: 4 to 70 years old.




16/31

Tian 2006 (Continued)

Treatment group: 4 to 10 years (39), 11 to 20 years (67), 21 to 30 years (51), 31 to 40 years (13), > 40

years (8). Control: 4 to 10 years (32), 11 to 20 years (54), 21 to 30 years (46), 31 to 40 years (19), > 40

years (5)

SEIZURE TYPES (number of patients): GTCS: treatment (91), control (82); absence: treatment (31),

control (40); simple partial seizure: treatment (34), control (23); complex partial seizure treatment (22),

control (11). Verified diagnosis of ’xian zheng’ in TCM, ’wind phlegm obstacle orifice’ types: treatment

(81), control (76); ’pyrophlegm intervene mind’ types: treatment (54), control (47); ’stagnant blood block

brain’ types: treatment (28), control (23); ’collateral and hepatic and renal yin deficiency’ types: treatment

(15), control (10)


BASELINE SEIZURE FREQUENCY: at least once a month and in total more than 3 times

TIME BETWEEN FIRST SEIZURE AND RANDOMISATION (number of patients): treatment: 0.3

to 5 years (85), 6 to 10 years (52), 11 to 20 years (30), 21 to 30 years (6), > 30 years (5); Control: 0.3 to

5 years (67), 6 to 10 years (46), 11 to 20 years (31), 21 to 30 years (9), > 30 years (3)

NUMBER AND TYPES OF AEDs TAKEN: most had monotherapy or biotherapy history including

VAL, CBZ, TPM etc


Treatment group: Tianmadingxian capsule to be taken 3 times daily at age-dependent dosage: < 8 years

old (1.05 g), 8 to 12 years old (1.4 g), > 13 years old (2.1 g)

If the participants had ever taken AEDs, these should be decreased for 1 month, then Tianmadingxianling

taken separately for 90 days

Control group: phenytoin 0.15 g 3 times daily for adults and 3 to 8 mg/kg daily for children for 90 days

Outcomes REDUCTION IN SEIZURE FREQUENCY (number of patients): 75% or greater reduction treatment

(125), control (46); 51% to 75%reduction treatment (44), control (62); 26% to 50%reduction treatment

(6), control group (21); less than 25% reduction treatment (3), control (27)

ADVERSE EVENTS: no mention

Notes (1) There was potential conflict of interest in the use of ’Tianmadingxian capsule’ made by the authors’

hospital

Risk of bias



Xiang 1998

Methods DESIGN: randomised controlled trialRANDOMISATION PROCEDURE: unclear

BLINDING: not used

DURATION OF FOLLOW UP: 60 days to 3 years

Participants COUNTRY AND SETTING: China; Yongzhou city, Hunan Province

NUMBER AND SEX: treatment group (males): 100 (58); control: 100 (53).

AGE: treatment: 8 to 61 years old, mean 30.84 (SD 6.64) years; control: 9 to 53 years old, mean 28.82




17/31

Xiang 1998 (Continued)

(SD 6.52) years

SEIZURE TYPES: primary generalized tonic-clonic seizure and verified diagnosis of ’xian zheng’ in TCM


BASELINE SEIZURE FREQUENCY: at least once a month and in total more than 3 times

TIME BETWEEN FIRST SEIZURE AND RANDOMISATION: treatment: 0.3 to 25 years (mean 7.

66 +/-2.12 years); control: 0.4 to 21 years (mean 7.42 +/-1.84 years)


Treatment group: ’Zhixian I’ mixture to be taken twice daily for 60 days.

Control group: phenytoin total 0.2 to 0.8g, 2 to 3 times a day for 60 days

Outcomes REDUCTION IN SEIZURE FREQUENCY(number of patients): 75% or greater reduction: treatment

(88), control (62); 51% to 75% reduction treatment (9), control (12); 26% to 50% reduction treatment

(2), control (16); less than 25% reduction treatment (1), control (10)

IMPROVEMENT IN EEG (number of patients): treatment group: normal (23), borderline abnormal (2)

, abnormal (75) before treatment, normal (78), abnormal (8), improvement (14) after treatment; control:normal (30), borderline abnormal (11), abnormal (59) before treatment, normal (40), abnormal (33),

improvement (25), aggravate (2) after treatment

RELAPSE RATE: participants who had more than 50% reduction in seizure frequency followed up for 3

years for observation of the relapse rate: treatment group 1 year (13/97, 13.4%), 2 years (20/97, 20.6%)

, 3 years (25/97, 25.8%); control group 1 year (25/74, 39.8%), 2 years (40/74, 20.6%), 3 years (51/74,

68.9%)

ADVERSE EVENTS: treatment group had gastrointestinal reaction (32) and no obvious effect on blood

routine examination, hepatic function and renal function.

Control group (number of participants): ataxia (30), gingival thickening (42), hirsuties (13), neuropathy

(11), gastrointestinal reaction (41), rash (18), hematopoietic system reaction (14)

Notes (1) There was potential conflict of interest in the use of ’Zhixian I’ made by the authors’ hospital

(2) The baseline seizure frequency was not mentioned

Risk of bias



Xin 1999



BLINDING: not usedDURATION OF FOLLOW UP: 90 days

Participants COUNTRY AND SETTING: China; Beijing

NUMBER AND SEX: treatment group (males): 301(167);control group (males): 100 (59)

AGE: treatment: 3 to 13 years old (mean 6.98 years); control: 3 to 13 years old (mean 6.25 years)

SEIZURE TYPES (number of patients): primary generalized tonic-clonic seizure and partial secondary

tonic-clonic seizure and verified diagnosis of ’xian zheng’ in TCM. GTCS: treatment 244, control 76;




18/31

Xin 1999 (Continued)

STCS: treatment 57, control 24


BASELINE SEIZURE FREQUENCY in 3 months before inclusion: 1 to 2 times per month and in total

more than 3 times


Treatment group: ’Antiepilepsy capsule’ to be taken 3 times daily at age-dependent dosage: 3 to 6 years

(2.5 g) and > 6 years (4 g). If the participants had ever taken AEDs for more than 1 month, Antiepilepsy

capsule was taken with AEDs for 1 month, then stopped taking AEDs and took separate ’Antiepilepsy

capsule’ for 90 days.

Control group: phenobarbital 1.5 to 2 mg/kg 3 times daily for 90 days

Outcomes REDUCTION IN SEIZURE FREQUENCY (number of patients): 75% or greater reduction treatment

(180), control (41); 51% to 75% reduction treatment (42), control (20); no reduction treatment (79),

control (39)

IMPROVEMENT IN EEG (number of patients): treatment: obviously improved (113), improved (40)and unchanged (86) ; control: obviously improved (37), improved (8) and unchanged (35)

ADVERSE EVENTS (number of patients): treatment group had gastrointestinal reaction such as poor

appetite (3), vomiting (2), bellyache (1) and diarrhoea (1); total adverse events incidence rate 2.33%,

no obvious effect on blood routine examination, hepatic function and renal function. Control group:

drowsiness (17), insomnia (1), slow reaction (3), vomiting (3), restlessness (2), dizziness and headache (1)

, rash (12); total adverse events incidence rate 37.0% and no obviously effect on blood routine, hepatic

function and renal function except 4 had lower haemoglobin

Notes

Risk of bias



AED = antiepileptic drug

GTCS = generalised tonic-clonic seizure

STCS= secondary generalised tonic clonic seizure

VAL=valproate

CBZ=carbamazepine

TPM= topiramate

p.o. = orally

t.i.d. = three times daily




19/31

Characteristics of excluded studies [ordered by study ID]

Study Reason for exclusion

Chen 1998 The diagnosis criteria for epilepsy were unclear

Chen 2000 Healthy volunteers as the control

Li 1997 The diagnosis criteria for epilepsy were unclear

Lin 1999 TCM (Ningxian capsule) combined with AEDs compared with AEDs

Liu 1994b TCM (alkaline extract of Euphorbia fisheriana ) combined with AEDs compared with AEDsCross-over design

Wang 2001 The participants did not accord with the diagnosis criteria of ILAE

Wu 2002 Healthy volunteers as the control

Xu 2005 The diagnosis criteria for epilepsy were unclear

Yin 2001 TCM (Ginkgo biloba extract) combined with AEDs compared with AEDs


ILAE = International League Against Epilepsy

TCM = traditional Chinese medicine

Characteristics of studies awaiting assessment [ordered by study ID]

Ma 2003a

Methods DESIGN: unclear

BLINDING: not used

DURATION OF FOLLOW UP: 180 day

Participants COUNTRY AND SETTING: China; Tianjin city

NUMBER AND SEX: xifeng capsule group (males): 200 (118); antiepilepsy capsule group: 100 (51); phenobarbital

group: 100 (59)

AGE (number of patients): xifeng capsule group: 2 to 6 years old (36), 6 to 10 years old (54), 11 to 14 years old(110); antiepilepsy capsule group: 2 to 6 years old (21), 6 to 10 years old (40), 11 to 14 years old (39); phenobarbital

group: 2 to 6 years old (20), 6 to 10 years old (36), 11 to 14 years old (44)

SEIZURE TYPES: primary generalized tonic-clonic seizure


BASELINE SEIZURE FREQUENCY (number of patients): less than once a month: Xifenggroup(113), antiepilepsy

capsule group (46), phenobarbital group (47); 1 to 3 times a month: Xifeng group (41), antiepilepsy capsule group

(24), phenobarbital group (30); more than 3 times a month: Xifeng group (46), antiepilepsy capsule group (30),




20/31

Ma 2003a (Continued)

phenobarbital group (23)

TIME BETWEEN FIRST SEIZURE AND RANDOMISATION (number of patients): xifeng group: less than 1

year (84), 1 to 5 years (86), 6 to 10 years (27), 11 to 14 years (3); antiepilepsy capsule group: less than 1 year (36), 1

to 5 years (48), 6 to 10 years (16); phenobarbital group: less than 1 year (54), 1 to 5 years (37), 6 to 10 years (9)

Interventions TREATMENT GROUP:

Xifeng capsule to be taken 3 times daily at age-dependent dosage: < 1 year 1 (1 capsule), 1 to 3 years (2 capsules), 3

to 7 years (5 capsules) and > 7 years (8 capsules)

’Antiepilepsy capsule’ to be taken 3 times daily at age-dependent dosage: < 1 year 1 (1 capsule), 1 to 3 years (2

capsules), 3 to 7 years (5 capsules) and > 7 years (8 capsules)

Phenobarbital group: phenobarbital: 2 mg/kg 3 times daily for 180 days

Outcomes REDUCTION IN SEIZURE FREQUENCY (number of patients)

Seizure freedom for at least 1 year

Xifeng group (49), antiepilepsy group (8), phenobarbital group (3)

75% or greater reduction Xifeng group (113), antiepilepsy group (56), phenobarbital group (38)

50% to 74% reduction


Less than 50% reduction


Notes

Ma 2003b



BLINDING: not used

DURATION OF FOLLOW UP: 180 days

Participants COUNTRY AND SETTING: China; Tianjin city

NUMBER AND SEX: treatment group (males): 930 (518); control: 160 (95)

AGE: treatment: 1 to 5 years old (72), 5 to 10 years old (69), 10 to 14 years old (19), mean 22.8 years; control: 1 to

5 years old (320), 5 to 10 years old (444), 10 to 14 years old (166), mean 25.7 years

SEIZURE TYPES: classification of epilepsy in ILAE 1981


BASELINE SEIZURE FREQUENCY: more than 6 attacks within 6 months before treatment

TIME BETWEEN FIRST SEIZURE AND RANDOMISATION: unclear

Interventions TREATMENT GROUP: ’Antiepilepsy capsule’ to be taken 3 times daily at age-dependent dosage: 1 to 5 years (1

to 5 capsules), 6 to 10 years (7 capsules) and 11 to 14 years (8 capsules). If the participants had ever taken AEDs for

more than 1 month, antiepilepsy capsule was taken with AEDs for 1 month, then AEDs were gradually decreased

from the second month, then stopped taking AEDs and took separate antiepilepsy capsule for 180 days

Control group: phenobarbital 1.5 to 2 mg/kg 3 times daily for 180 days

Outcomes REDUCTION IN SEIZURE FREQUENCY (number of patients): 75% or greater reduction treatment (534),

control (64); 51% to 75% reduction treatment (241), control (19); 25% to 50% reduction treatment (96), control

(38); no reduction or aggravated treatment (13), control group (10)




21/31

Ma 2003b (Continued)

Notes





22/31

D A T A A N D A N A L Y S E S

Comparison 1. One formula of TCMH versus single Western medication

Outcome or subgroup titleNo. of

studies

No. of

participants Statistical method Effect size

1 Xiaxingci granule versus

phenytoin

1 Risk Ratio (M-H, Fixed, 95% CI) Subtotals only

1.1 Seizure freedom 1 40 Risk Ratio (M-H, Fixed, 95% CI) 1.0 [0.07, 14.90]

1.2 75% or greater reduction

in seizure duration

1 40 Risk Ratio (M-H, Fixed, 95% CI) 1.5 [0.28, 8.04]


in seizure frequency


2 Dianxianning pill versus

valproate


2.1 Seizure freedom 1 200 Risk Ratio (M-H, Fixed, 95% CI) 13.0 [0.74, 227.72]2.2 80% or greater reduction






3 Tianmadingxian capsule 1 Risk Ratio (M-H, Fixed, 95% CI) Subtotals only





in seizure duration





4 Zhixian I versus phenytoin 1 Risk Ratio (M-H, Fixed, 95% CI) Subtotals only










5 Antiepilepsy capsule versus

phenobarbital








5.3 75% or greater reductionin seizure duration



in seizure duration


6 Incidence of adverse or harmful

effects

1 Odds Ratio (Peto, Fixed, 99% CI) Subtotals only

6.1 Antiepilepsy capsule

versus phenobarbital

1 401 Odds Ratio (Peto, Fixed, 99% CI) 0.04 [0.01, 0.12]




23/31

Analysis 1.1. Comparison 1 One formula of TCMH versus single Western medication, Outcome 1 Xiaxingci

granule versus phenytoin.

Review: Traditional Chinese medicine for epilepsy

Comparison: 1 One formula of TCMH versus single Western medication

Outcome: 1 Xiaxingci granule versus phenytoin

Study or subgroup Treatment Control Risk Ratio Weight Risk Ratio

n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI

1 Seizure freedom

Liu 1994a 1/20 1/20 100.0 % 1.00 [ 0.07, 14.90 ]

Subtotal (95% CI) 20 20 100.0 % 1.00 [ 0.07, 14.90 ]

Total events: 1 (Treatment), 1 (Control)Heterogeneity: not applicable

Test for overall effect: Z = 0.0 (P = 1.0)

2 75% or greater reduction in seizure duration

Liu 1994a 3/20 2/20 100.0 % 1.50 [ 0.28, 8.04 ]

Subtotal (95% CI) 20 20 100.0 % 1.50 [ 0.28, 8.04 ]

Total events: 3 (Treatment), 2 (Control)

Heterogeneity: not applicable


3 25% or greater reduction in seizure frequency

Liu 1994a 18/20 18/20 100.0 % 1.00 [ 0.81, 1.23 ]

Subtotal (95% CI) 20 20 100.0 % 1.00 [ 0.81, 1.23 ]




0.01 0.1 1 10 100

Favours experimental Favours control




24/31

Analysis 1.2. Comparison 1 One formula of TCMH versus single Western medication, Outcome 2

Dianxianning pill versus valproate.



Outcome: 2 Dianxianning pill versus valproate



1 Seizure freedom

Song 2001 6/100 0/100 100.0 % 13.00 [ 0.74, 227.72 ]

Subtotal (95% CI) 100 100 100.0 % 13.00 [ 0.74, 227.72 ]





Song 2001 82/100 30/100 100.0 % 2.73 [ 2.00, 3.74 ]

Subtotal (95% CI) 100 100 100.0 % 2.73 [ 2.00, 3.74 ]



Test for overall effect: Z = 6.29 (P < 0.00001)


Song 2001 94/100 96/100 100.0 % 0.98 [ 0.92, 1.04 ]

Subtotal (95% CI) 100 100 100.0 % 0.98 [ 0.92, 1.04 ]




0.01 0.1 1 10 100





25/31


Tianmadingxian capsule.



Outcome: 3 Tianmadingxian capsule




Tian 2006 169/178 108/156 100.0 % 1.37 [ 1.23, 1.53 ]

Subtotal (95% CI) 178 156 100.0 % 1.37 [ 1.23, 1.53 ]




2 75% or greater reduction in seizure duration

Tian 2006 125/178 46/156 100.0 % 2.38 [ 1.83, 3.09 ]

Subtotal (95% CI) 178 156 100.0 % 2.38 [ 1.83, 3.09 ]





Tian 2006 175/178 129/156 100.0 % 1.19 [ 1.10, 1.28 ]

Subtotal (95% CI) 178 156 100.0 % 1.19 [ 1.10, 1.28 ]




0.01 0.1 1 10 100





26/31

Analysis 1.4. Comparison 1 One formula of TCMH versus single Western medication, Outcome 4 Zhixian I

versus phenytoin.



Outcome: 4 Zhixian I versus phenytoin




Xiang 1998 97/100 74/100 100.0 % 1.31 [ 1.16, 1.48 ]

Subtotal (95% CI) 100 100 100.0 % 1.31 [ 1.16, 1.48 ]





Xiang 1998 88/100 62/100 100.0 % 1.42 [ 1.20, 1.68 ]

Subtotal (95% CI) 100 100 100.0 % 1.42 [ 1.20, 1.68 ]





Xiang 1998 99/100 90/100 100.0 % 1.10 [ 1.03, 1.18 ]

Subtotal (95% CI) 100 100 100.0 % 1.10 [ 1.03, 1.18 ]




0.01 0.1 1 10 100





27/31


Antiepilepsy capsule versus phenobarbital.



Outcome: 5 Antiepilepsy capsule versus phenobarbital




Xin 1999 222/301 61/100 100.0 % 1.21 [ 1.02, 1.43 ]

Subtotal (95% CI) 301 100 100.0 % 1.21 [ 1.02, 1.43 ]





Xin 1999 180/301 41/100 100.0 % 1.46 [ 1.13, 1.88 ]

Subtotal (95% CI) 301 100 100.0 % 1.46 [ 1.13, 1.88 ]

Total events: 180 (Treatment), 41 (Co

traditional chinese medicine for epilepsy

Documents