Your SlideShare is downloading. ×
Acupuncture for stroke rehabilitation
Upcoming SlideShare
Loading in...5
×

Thanks for flagging this SlideShare!

Oops! An error has occurred.

×

Introducing the official SlideShare app

Stunning, full-screen experience for iPhone and Android

Text the download link to your phone

Standard text messaging rates apply

Acupuncture for stroke rehabilitation

11,076
views

Published on


0 Comments
1 Like
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total Views
11,076
On Slideshare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
66
Comments
0
Likes
1
Embeds 0
No embeds

Report content
Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
No notes for slide

Transcript

  • 1. Acupuncture for stroke rehabilitation (Review) Wu HM, Tang JL, Lin XP, Lau JTF, Leung PC, Woo J, Li Y This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library 2009, Issue 1 http://www.thecochranelibrary.com Acupuncture for stroke rehabilitation (Review) Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 2. TABLE OF CONTENTS HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Analysis 1.1. Comparison 1 Acupuncture versus open control, Outcome 1 Improvement of global neurological deficit at the end of treatment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 WHAT’S NEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 HISTORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 INDEX TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Acupuncture for stroke rehabilitation (Review) i Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 3. [Intervention Review] Acupuncture for stroke rehabilitation Hong Mei Wu1 , Jin-Ling Tang2 , Xiao Ping Lin3 , Joseph TF Lau3 , Ping Chung Leung4 , Jean Woo5 , Youping Li6 1 Department of Geriatrics, West China Hospital, Sichuan University, Chengdu, China. 2 Dept of Community and Family Medicine, The Chinese University of Hong Kong, Lek Yuen Health Centre, Shatin, China. 3 Centre of Epidemiology & Biostatistics, School of Public Health, The Chinese University of Hong Kong, Shatin, China. 4 Institute of Chinese Medicine, The Chinese University of Hong Kong, Shatin, China. 5 Department of Community and Family Medicine, School of Public Health, The Chinese University of Hong Kong, Shatin, China. 6 Chinese Cochrane Centre, West China Hospital, Sichuan University, Chengdu, China Contact address: Hong Mei Wu, Department of Geriatrics, West China Hospital, Sichuan University, 37 Guo Xue Xiang Street, Chengdu, Sichuan, 610041, China. drwhm@163.com. drwhm@126.com. Editorial group: Cochrane Stroke Group. Publication status and date: Edited (no change to conclusions), published in Issue 1, 2009. Review content assessed as up-to-date: 15 March 2006. Citation: Wu HM, Tang JL, Lin XP, Lau JTF, Leung PC, Woo J, Li Y. Acupuncture for stroke rehabilitation. Cochrane Database of Systematic Reviews 2006, Issue 3. Art. No.: CD004131. DOI: 10.1002/14651858.CD004131.pub2. Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. ABSTRACT Background Stroke is the third leading cause of death in Western society; in China it is the second most common cause of death in cities and the third in rural areas. It is also a main cause of adult disability and dependency. Acupuncture for stroke has been used in China for hundreds of years and is increasingly practiced in some Western countries. Objectives To assess the efficacy and safety of acupuncture for patients with stroke in the subacute or chronic stage. Search strategy We searched the Cochrane Stroke Group Trials Register (November 2005), the Cochrane Complementary Medicine Field Trials Register (November 2005), the Cochrane Central Register of Controlled Trials (The Cochrane Library Issue 3, 2005), MEDLINE (1966 to November 2005), EMBASE (1980 to November 2005), CINAHL (1982 to November 2005), AMED (1985 to November 2005), the Chinese Stroke Trials Register (November 2005), the Chinese Acupuncture Trials Register (November 2005), the Chinese Biological Medicine Database (1977 to November 2005), the National Center for Complementary and Alternative Medicine Register (November 2005), and the National Institute of Health Clinical Trials Database (November 2005). We handsearched four Chinese journals and checked reference lists of all papers identified for further trials. Selection criteria Truly randomised unconfounded clinical trials among patients with ischemic or hemorrhagic stroke, in the subacute or chronic stage, comparing acupuncture involving needling with placebo acupuncture, sham acupuncture or no acupuncture. Data collection and analysis Two review authors independently selected trials for inclusion, assessed quality, extracted and cross-checked the data. Acupuncture for stroke rehabilitation (Review) 1 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 4. Main results Five trials (368 patients) met the inclusion criteria. Methodological quality was considered inadequate in all trials. Although the overall estimate from four trials suggested the odds of improvement in global neurological deficit was higher in the acupuncture group compared with the control group (odds ratio (OR) 6.55, 95% confidence interval (CI) 1.89 to 22.76), this estimate may not be reliable since there was substantial heterogeneity (I2 = 68%). One trial showed no significant improvement of motor function between the real acupuncture group and the sham acupuncture group (OR 9.00, 95% CI 0.40 to 203.30), but the confidence interval was wide and included clinically significant effects in both directions. No data on death, dependency, institutional care, change of neurological deficit score, quality of life or adverse events were available. Authors’ conclusions Currently there is no clear evidence on the effects of acupuncture on subacute or chronic stroke. Large, methodologically-sound trials are required. PLAIN LANGUAGE SUMMARY Acupuncture for stroke rehabilitation There is no clear evidence of the effects of acupuncture on stroke rehabilitation. Acupuncture has biological effects that might improve recovery from stroke or facilitate rehabilitation. This review looked for randomised trials comparing acupuncture with control in patients who had a stroke more than one month previously. Five trials were identified but all the trials were of poor quality and no definite conclusions could be drawn about the effects of acupuncture in such patients. More large, high quality randomised trials are needed. BACKGROUND acupuncture can cause multiple biological responses, including Stroke is the third leading cause of death in the world (Bonita circulatory and biochemical effects. These responses can occur lo- 1990). In China, stroke is the second most common cause of death cally or close to the site of application, or at a distance. They are in cities and the third most common cause of death in rural areas mediated mainly by sensory neurons to many structures within (MOH PRC 1999). Stroke is also a major cause of disability and the central nervous system. This can lead to activation of pathways dependency in the elderly in the world. The overall prevalence of affecting various physiological systems in the brain as well as in stroke is 17.5/1000 per year (95% confidence interval (CI) 17.0 the periphery (Jansen 1989; Johansson 1993; Magnusson 1994; to 18.0) in the population aged 45 years and over. The prevalence Sun 2001; Wang 2001). Whether acupuncture can really improve of stroke-associated dependence is 11.7/1000 per year (95% CI recovery in patients with stroke is still uncertain. 11.3 to 12.1). Approximately 67% of survivors become function- ally dependent (O’Mahony 1999) and 10% require long-term in- Acupuncture is one of the main modalities of treatment in tradi- stitutional care (DPH 1994) thus imposing a great burden on the tional Chinese medicine and can be traced back more than 2000 family and community. Despite considerable research efforts on years in China (Wu 1996). Being a relatively simple, inexpensive multiple treatment modalities, there is still no single rehabilitation and safe treatment compared to other conventional interventions, intervention demonstrated unequivocally to aid recovery. This re- acupuncture has been well accepted by Chinese patients and is ality drives people to search for other modalities of treatment in an widely used to improve motor, sensation, speech and other neu- attempt to further improve the outcome of stroke rehabilitation, rological functions in patients with stroke. As a therapeutic inter- such as acupuncture and Chinese herb medicine. vention, acupuncture is also increasingly practiced in some West- ern countries (Johansson 1993; NIH 1998). However, it remains Many studies in animals and humans have demonstrated that uncertain whether the existing evidence is scientifically rigorous Acupuncture for stroke rehabilitation (Review) 2 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 5. enough so that acupuncture can be recommended for routine use. record numbers, dates of birth or day of the week, or controlled trials using any other non-random allocation methods. There are a large number of studies of the clinical efficacy of acupuncture in stroke rehabilitation published in the world, espe- cially in China, but not all demonstrate a beneficial effect on stroke rehabilitation (Hu 1993; Zhang 1996; Johansson 2001; Sun 2001; Types of participants Sze 2002). Systematic reviews of trials of acupuncture in stroke rehabilitation have been conducted (Ernst 1996; Hopwood 1996; Trials involving patients of any age or sex with ischemic or hem- Park 2001; Smith 2002; Sze 2002). These reviews, however, in- orrhagic stroke in the subacute (one to three months since onset) cluded mainly trials with stroke patients in either acute or subacute or chronic phases (over three months since onset) were eligible. and chronic stages. One recent systematic review conducted by a Stroke must be diagnosed according to the World Health Orga- Chinese research group demonstrated that acupuncture appeared nization definition (rapidly developed clinical signs of focal (or to be safe but without clear evidence of benefit for patients with global) disturbances of cerebral function, lasting more than 24 acute stroke (Liu 2005). Whether or not acupuncture is effective hours or leading to death, with no other apparent cause than of for subacute and chronic stroke alone is still unknown. vascular origin (Asplund 1988)); or confirmed by computerised tomography (CT) or magnetic resonance imaging (MRI). Trials of The aim of this review was to systematically analyse all the ran- patients with subarachnoid hemorrhage and subdural hematoma domised controlled trials of acupuncture for subacute and chronic were also to be included but were to be analysed separately. Trials stroke to provide the best available evidence to inform clinical that included patients in the acute phase (within one month since practice and further research planning on stroke treatment. onset) were excluded. OBJECTIVES Types of interventions The objective of this systematic review was to determine the effi- cacy and safety of acupuncture therapy in patients with subacute Trials evaluating acupuncture therapy which involved needling af- and chronic stroke. We intended to test the following hypotheses: ter stroke onset at subacute or chronic phases were included in the review, regardless of times of treatment and length of treatment (1) acupuncture can reduce the risk of death or dependency in period. Either traditional acupuncture in which the needles were patients with subacute and chronic stroke at the end of treatment inserted in classical meridian points or contemporary acupunc- and at follow up; ture in which the needles were inserted in non-meridian or trig- (2) acupuncture can improve neurological deficit and quality of ger points were included regardless of the source of stimulation life after treatment and at the end of follow up; (for example, hand or electrical stimulation). Trials in which the acupuncture treatment did not involve needling, such as acupres- (3) acupuncture can reduce the numbers of patients requiring sure or laser acupuncture, were excluded. institutional care; The control interventions were placebo acupuncture, sham (4) acupuncture is not associated with any intolerable adverse ef- acupuncture, or other conventional treatment. Placebo acupunc- fects. ture refers to a needle attached to the skin surface (not penetrat- ing the skin but at the same acupoints) (Van Tulder 2000). Sham acupuncture refers to: (1) a needle placed in an area close to but not in the acupuncture METHODS points (Van Tulder 2000); (2) subliminal skin electrostimulation via electrodes attached to the skin (SCSSS 1999). Criteria for considering studies for this review Comparisons investigated were: (1) acupuncture only compared with placebo or sham treatment; (2) acupuncture in addition to baseline medication or treatment Types of studies compared with placebo or sham treatment in addition to baseline Randomised controlled clinical trials (RCTs) comparing acupunc- medication or treatment; ture with at least one control group that used placebo, sham treat- (3) acupuncture in addition to baseline medication or treatment ment or conventional treatment in patients with subacute (one to compared with baseline medication or treatment alone. three months since onset ) or chronic stroke (over three months since onset) were included in the review. We excluded trials using Trials that compared different forms of acupuncture only were quasi-randomisation or allocating patients using alternation, case excluded. Acupuncture for stroke rehabilitation (Review) 3 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 6. Types of outcome measures We searched the Cochrane Stroke Group Trials Register, which Trials were included that used at least one of the following outcome was last searched by the Review Group Co-ordinator in Novem- measures. ber 2005. We also searched the Chinese Stroke Trials Regis- ter (November 2005), the Chinese Acupuncture Trials Register (November 2005) and the Trials Register of the Cochrane Com- Primary outcome measure plementary Medicine Field (November 2005). In addition, we (1) Death or dependency at the end of follow up (at least three searched the following bibliographic databases and trials regis- months or longer after stroke onset). Dependency was defined as ters: the Cochrane Central Register of Controlled Trials (CEN- dependent on others in activities of daily living, based on the cor- TRAL) (The Cochrane Library Issue 3, 2005), MEDLINE (1966 to related definition of the Barthel scores (Activities of Daily Living, November 2005), EMBASE (1980 to November 2005), CINAHL ADL) as a score of less than 60 or an Oxford handicap grade 3 to (1982 to November 2005), AMED (the Allied and Complemen- 6 (Sulter 1999), or the trialist’s own definition. tary Medicine Database, 1985 to November 2005), the Chinese Biological Medicine Database (CBM-disc, 1979 to November 2005), the National Center for Complementary and Alternative Secondary outcome measures Medicine Register (http://nccam.nih.gov/clinical trials/) and Na- (1) The proportion of people requiring institutional care or re- tional Institute of Health Clinical Trials Database (http://clinical- quiring extensive family support at the end of follow up (at three trials.gov) (searched November 2005) (Appendix 1). months or longer after stroke onset). Family care is the main form Four Chinese journals relevant to acupuncture were handsearched of care for severely dependent people in developing countries. (from 1980 to November 2005). (2) Changes of neurological deficit after acupuncture treatment and at the end of follow up (at three months or longer after stroke • Acupuncture Research onset). The measures could focus on specific impairment (for ex- • Chinese Acupuncture and Moxibustion ample, Motricity Index, or Motor Assessment Scale which assess • Journal of Clinical Acupuncture and Moxibustion only motor function) or global neurological deficit (for example, • Shanghai Journal of Acupuncture and Moxibustion the National Institute of Health Stroke Scale, European Stroke We checked reference lists of all relevant papers identified, includ- Scale, the Scandinavian Stroke Scale) or two kinds of Chinese ing two systematic reviews (Park 2001; Sze 2002) for further pub- Stroke Recovery Scales which involve motor, sensory and other lished and unpublished trials. impaired neurological functions. The Chinese Stroke Recovery Scale (CSRS 1) refers to ’the Revised Diagnostic Criteria of Acute Cerebral Infarction’ formulated by the second National Academic Symposium on Cerebrovascular Diseases of the Chinese Medical Data collection and analysis Association in 1986, which is similar to the Revised Scandinavian Stroke Scale (RSSS). The Chinese Stroke Recovery Scale 2 (CSRS 2) refers to ’the Chinese Stroke Recovery Scale based on principles Study selection of traditional Chinese medicine’. Two review authors (Wu HM, Lin XP) independently checked (3) Death from any causes during the entire treatment and follow- the titles and abstracts of trials for inclusion based on selection up period. criteria outlined above. The full text of the article was retrieved (4) Quality of life (QOL) at the end of follow up (at three months if there was any doubt whether the article should be excluded or or longer after stroke onset). This could be measured by the Not- not. In cases of disagreement between the two review authors, a tingham Health Profiles or Spiter Quality of Life Index. third member of the stroke research group (Tang JL) reviewed the (5) Possible adverse events including dizziness, difficulty in toler- information to decide on inclusion or exclusion of an abstract. ating electrostimulation, infection, puncture of a lung, heart tam- ponades, spinal cord injury, disrupted pacemaker function; and presumed to be caused by acupuncture or electrostimulation. The Methodological quality assessment number of patients developing at least one severe adverse event No specific scoring system or checklist was used to assess the listed above was evaluated. methodological quality of the included trials. Rather, the method- Trials reporting only physiological or laboratory parameters were ological quality of studies was documented using the following excluded. internal validity criteria in this review. (1) Method of randomisation: a method to generate the sequence of randomisation was regarded as appropriate if it allowed each Search methods for identification of studies study participant to have the same chance of receiving each inter- See: ’Specialized register’ section in Cochrane Stroke Group vention and the investigators could not predict which treatment Acupuncture for stroke rehabilitation (Review) 4 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 7. was next. Methods of allocation using date of birth, date of ad- Data analysis mission, hospital numbers, or alternation were not regarded as ap- The Cochrane Review Manager software, RevMan 4.2, was used propriate. to calculate treatment effects across trials. Heterogeneity between (2) Concealment of allocation: scored A (adequate), B (unclear), trial results was tested using a standard chi-squared test, with a or C (inadequate) following criteria adopted from the Cochrane threshold value of p < 0.1, and with the I2 statistic. For dichoto- Handbook for Systematic Reviews of Interventions and Schulz et mous outcomes, the results were expressed as odds ratio (OR) with al (Schulz 1995). 95% confidence intervals (CI). Data were pooled using the ran- A: adequate measures to conceal allocations such as central ran- dom-effects model. If appropriate continuous data reporting the domisation; serially numbered, opaque, sealed envelopes; or other same scale for each trial had been available we had planned to use description that contained convincing elements of concealment. the weighted mean difference (WMD) or the standardised mean B: unclearly concealed trials in which the authors either did not difference (SMD) if different scales had been used. report an allocation concealment approach at all, or reported an Because poststroke neurological recovery is known to be obvious approach that did not fall into one of the categories in A. within three months, especially poststroke motor recovery, and is C: inadequately concealed trials in which the method of allocation unlikely beyond six months, mixing patients with interval to stroke was not concealed, such as alteration methods or use of case record onset of less than three months, three to six months, and more numbers. than six months in one sample would make the assessment of the (3) Blinding (both of participants and outcome assessors): a study efficacy of an intervention methodologically unsound. Therefore, was regarded as double blind if the phrase ’double blind’ was used. if appropriate data were available, we planned a subgroup analy- The method was regarded as appropriate if it was stated that neither sis to compare patients with different times to start of treatment the person doing the assessments nor the study participant could (within three months, three to six months, and after six months) identify the intervention being assessed, or if in the absence of using the method outlined by Deeks et al (Deeks 2001). such a statement the use of active placebos, identical placebos, or Similarly, if appropriate data were available, we planned a sensi- dummies was mentioned. tivity analysis to assess the effects of including: (4) Intention-to-treat analysis: whether an intention-to-treat anal- ysis was possible on all patients from the published data (that is, • only those trials which were double blind; whether there were any exclusions from the trial after randomisa- • only those trials with adequate concealment of tion) and the number of patients who were lost to follow up. If randomisation; there were no withdrawals, it should be stated in the article. • only those trials published in a language other than Chinese. Quality assessment was performed by two independent review au- Finally, if sufficient RCTs were identified, we planned to examine thors (Wu HM, Lin XP) and disagreement reported and resolved for potential publication bias (Vickers 1998) using a funnel plot ( by a third member (Tang JL). These criteria did not form exclu- Egger 1997). sion criteria, and are described in the ’Characteristics of included studies’ table. RESULTS Data extraction Information on patients, methods, interventions, outcomes, and Description of studies results was extracted independently by two review authors (Wu See: Characteristics of included studies; Characteristics of excluded HM, Lin XP) using a self-developed data extraction form. Dis- studies. agreements were resolved by a third member (Tang JL) or through Fifty potentially eligible trials were retrieved for further assess- discussion. For dichotomous outcomes (such as death or depen- ment. Of these, five trials including a total of 368 patients met dency, numbers experiencing adverse effects), the number of par- our inclusion criteria. Forty-five trials were excluded because: ticipants experiencing the event and the total number of partici- (1) they compared two different kinds of acupuncture (22 trials) pants in each arm of the trial were extracted. For continuous out- (Che 2002; Ding 2000; Feng 1996a; Feng 1996b; Jiang 1998; comes (impairment scales such as specific neurological disturbance Jin 1993; Lai 1997a; Lai 1998; Lai 2004; Li 1994a; Li 1993; Li and QOL), mean change and standard deviation for the mean 1994b; Liang 1993; Qi 2000; Qu 1991; Sui 2001;Tong 1997; change in each arm of the trial were extracted along with the total Xiao 1996; Yu 2002; Zhang 2002a; Zheng 1996; Zhou 1995); number in each group. Data on the number of patients with each (2) of confounding with drug therapy or rehabilitation treatment outcome event and by allocated treatment group, irrespective of (14 trials) (Chen 2000; Cui 1992; Gao 2001; Guo 1999; Hou compliance or follow up, was sought to allow an intention-to-treat 1998; Huang 2002; Jiang 2000; Lai 1997b; Liu 2004a; Muo 2001; analysis. Zhang 1988; Zhang 1997; Zhang 2002b; Zheng 2000); Acupuncture for stroke rehabilitation (Review) 5 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 8. (3) no usable data were available for analysis (four trials) (Fink included trials. The needle retention time was 15 to 30 minutes 2004; Li 1997b; Liu 1998; Wu 1999); in all included trials. Three trials (Dai 1997; Lun 1999; Wang (4) they included patients with acute stroke (five trials) (Kjendahl 2001) reported the achievement of DeChi, an irradiating feeling 1997; Liu 2004b; Sallstrom 1996; Wang 2004; Wei 2005). said to indicate effective needling. Information on needle type was For details of each included trial please see the ’Characteristics of available in three trials (Dai 1997; Lun 1999; Naeser 1992). The included studies’ table. length of treatment period ranged from 30 to 90 days with the Of the five included trials, four were conducted in China and one number of treatment sessions varying from 30 to 72 sessions and in the USA. The age of participants ranged from 24 to 86 years. the frequency of treatment from five sessions per week to one ses- More men than women were included in three trials (between sion per day. 56% and 75% men) (Dai 1997; Lun 1999; Wang 2001). Two The most commonly reported outcome was proportion of patients trials did not describe the gender characteristics of patients (Li with an improvement of global neurological deficit after treatment 1997a; Naeser 1992). The range of time from stroke onset was rather than the change score of neurological deficit (Dai 1997; from one month to 8.5 years. There was one trial involving patients Li 1997a; Lun 1999; Wang 2001). The measures employed in- with interval to stroke onset between one to three months (Naeser cluded CSRS 1 (Dai 1997; Li 1997a) and CSRS 2 (Lun 1999; 1992), one trial more than three months (Dai 1997) and three Wang 2001). One trial evaluated the effect of acupuncture on the trials including patients with interval to stroke onset of either less improvement of motor function after treatment (Naeser 1992). than three months or more than three months (Li 1997a; Lun None of the five trials provided any information on death or de- 1999; Wang 2001). Two trials (Dai 1997; Naeser 1992) included pendency, proportion of patients requiring institutional care or patients with ischemic stroke only. All other trials included patients extensive family support, quality of life and adverse events. with ischemic and hemorrhagic stroke (Li 1997a; Lun 1999; Wang 2001). There was no trial involving patients with subarachnoid hemorrhage or subdural hematoma. All of the included trials used CT or MRI to confirm the diagnosis of stroke. The severity on Risk of bias in included studies entry was mild to severe in two trials (Dai 1997; Naeser 1992), without a definition of severity, and not stated in the remaining three trials (Li 1997a; Lun 1999; Wang 2001). Method of randomisation One trial compared real acupuncture plus baseline treatment with No trial described the method of randomisation. sham acupuncture plus baseline treatment (Naeser 1992), the re- maining four trials compared acupuncture in addition to base- line medication or treatment with baseline medication or treat- Concealment of allocation ment alone, and none of the trials compared acupuncture only with placebo or sham treatment. Among the included trials there Allocation concealment was unclear in all included trials. were three three-armed trials (Dai 1997; Li 1997a; Wang 2001) comparing acupuncture in addition to baseline medication or treatment with baseline medication treatment alone, one kind of Blinding acupuncture with another kind of acupuncture, or acupuncture Only one trial reported that participants were blinded but did not only with other treatment. In this review, the baseline medication describe the method in detail (Naeser 1992). No information on or treatment included Western medicine (WM), traditional Chi- blinding was available in the remaining four trials. nese medicine (TCM), non-pharmacological therapy or a com- bination. Western medicine included aspirin and other conven- tional drug therapies. None of the included trials reported the acupuncture rationale or acupuncturists’ background, including Intention-to-treat analysis duration of relevant training, length of clinical experience and ex- There was no statement on dropouts or withdrawals in any of the pertise in the specific condition. The acupuncture interventions included trials. For each trial the number of patients randomised used varied considerably across trials. Four trials used only manual was the same as patients analysed, so it appears there were no stimulation (Dai 1997; Li 1997a; Lun 1999; Wang 2001) and one exclusions from the trials after randomisation. We conclude that used only electrical stimulation (Naeser 1992). Acupuncture point although the results appear to be analysed by intention to treat we prescriptions were not consistent, with two trials involving only cannot be certain about this. scalp or body acupoints (Lun 1999; Wang 2001) and three trials In general, the methodological quality of included trials was poor. using both body and scalp acupoints (Dai 1997; Li 1997a; Naeser An insufficient number of trials prohibited us from performing 1992). Numbers of points used ranged from 5 to 12 points in all meaningful sensitivity analysis in relation to study quality. Acupuncture for stroke rehabilitation (Review) 6 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 9. Effects of interventions Quality of life at the end of follow up No data on quality of life at the end of follow up were available in all included trials. Death or dependency at the end of treatment and follow up Possible adverse events No data on death or dependency at the end of treatment and No information on possible adverse events was reported in all follow up were available in all included trials. included trials. Proportion requiring institutional care or requiring Subgroup analysis extensive family support at the end of follow up From the available information, it was not possible to conduct No data on the proportion of patients requiring institutional care subgroup analysis on interval to stroke onset of less than three or extensive family support at the end of follow up were available months, three to six months, and more than six months. in all included trials. Sensitivity analysis Changes of global neurological deficit after From the available information, it was also not possible to perform acupuncture treatment and at the end of follow up a meaningful sensitivity analysis in this systematic review because of insufficient numbers of trials and the general poor quality of Four trials with a total of 352 patients (Dai 1997; Li 1997a; Lun trials. 1999; Wang 2001) measured improvement of global neurolog- ical deficit after acupuncture treatment by using categorical ap- proaches only (CSRS 1 and CSRS 2) rather than continuous scales, Publication bias for example changes of global neurological deficit score. There was significant heterogeneity among the four trials (chi squared P = It was not possible to perform a funnel plot for checking publica- 0.04, I2 = 68%), which was possibly due to differences in outcome tion bias in this systematic review because of limited number of measurements used, times of evaluation from stroke onset, and trials for each outcome. types of stroke. This means that the overall estimate of treatment effect is not reliable. Hence, the apparent improvement of global neurological deficit in the acupuncture group compared with the control group (OR for neurological improvement with acupunc- DISCUSSION ture 6.55, 95% CI 1.89 to 22.76) should be interpreted with cau- tion. Five trials with a total of 368 patients were included in this review. Compared with three previous systematic reviews on acupunc- ture for stroke (Liu 2005; Park 2001; Sze 2002), the present re- view included only patients in the subacute or chronic stage of Changes of specific neurological impairment after stroke. However, there was still no clear evidence on the effects of acupuncture treatment and at the end of follow up acupuncture on subacute or chronic stroke. One trial with 16 patients comparing real acupuncture with sham acupuncture (Naeser 1992) evaluated the effect of acupuncture Although the overall estimate appeared to show there was a signifi- on the improvement of motor function after treatment by using a cant improvement of global neurological deficit after acupuncture categorical approach rather than continuous scales. There was no treatment, this result needs to be interpreted with caution. The statistically significant difference between the groups (OR 9.00, result was based on data from four trials with a limited number 95% CI 0.40 to 203.30), but the confidence interval was very of patients (total of 352 patients); the trials were of very poor wide and included clinically significant effects in both directions. methodological quality with regard to method of randomisation, allocation concealment and blinding of assessment. Methodolog- ically less rigorous studies show larger differences between exper- imental and control groups than do those conducted with better Death from any cause during the whole treatment rigor (Kjaergard 1999; Moher 1998; Schulz 1995). Schulz et al ( and follow up period Schulz 1995) found that odds reductions were exaggerated by up No data on death from any cause during the period of treatment to 30% for trials that did not have clear concealment, and by 41% and follow up were available in all included trials. for inadequately concealed trials. Acupuncture for stroke rehabilitation (Review) 7 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 10. There was no significant improvement of motor function after was that clinical trials of acupuncture must use an optimal form of acupuncture treatment. It should be noted that the result was based treatment, defined by examining standard texts and by surveying on data from only one trial with a small number of participants and consulting experts, and must be reported by using Standards (total of 16 patients). Larger samples would be needed to detect for Reporting Interventions in Controlled Trials of Acupuncture moderate effects with any reliability. (STRICTA) (MacPherson 2001; White 2001). The insufficient number of trials identified and their general low None of the included trials in this systematic review reported methodological quality prohibited meaningful sensitivity analysis whether any adverse events relevant to acupuncture were apparent to illuminate how robust the results of the review were to exclusion in stroke patients. The reason for the insufficient report was possi- of the trials with inadequate methodology. It was also not possible bly that Chinese practitioners perceive acupuncture as free of side to perform sensitivity analysis to illuminate how robust the results effects. A NIH consensus report also stated that one of the advan- of the review were to exclusion of Chinese trials because only one tages of acupuncture was that the incidence of adverse effects is non-Chinese study was included. Smith and his colleagues (Smith substantially lower than that of many other accepted medical inter- 2002) found that sensitivity analyses based on blinding, reporting ventions (NIH 1998). However, some studies also demonstrated quality, validity score and country of origin all showed a higher that serious adverse events have been associated with acupuncture, proportion of positive results for poor quality studies than for such as infections (HIV, hepatitis, subacute bacterial endocarditis) those of higher quality in assessing the evidence of effectiveness of caused by non-sterile needles, or complications (pneumothorax, acupuncture for stroke rehabilitation. cardiac tamponade) caused by tissue trauma, but the incidence of adverse events was unknown (Ernst 1997). A recent systematic Because poststroke neurological recovery is known to be obvious review (Liu 2005) showed that acupuncture appears to be a safe within three months, especially poststroke motor recovery, and un- treatment when used in the acute phase of stroke, with severe ad- likely beyond six months, mixing patients with interval to stroke verse events occurring only very rarely (possibly less than 1.5%). onset of less than three months, three to six months, and more than six months in one sample would make the assessment of the It was not possible to perform a funnel plot to assess the degree of efficacy of an intervention methodologically unsound. From the publication bias in this systematic review because of the limited available information, it was not possible to perform pre-speci- numbers of trials for each outcome. Four of the five included trials fied subgroup analyses comparing patients with different times to in this review were conducted in China and published in Chi- start of treatment (within three months, three to six months, and nese (the remaining trial was published in English). Vickers and after six months) for each outcome. This was due to the limited colleagues (Vickers 1998) found that some countries, including amount of data, the varied outcome measurements, or both, in China, publish unusually high proportions of positive results. Al- this systematic review. though we have undertaken extensive searches for published ma- terial, we still could not exclude the possibility that studies with The long-term goal of treatment for stroke is to reduce mortal- negative findings remain unpublished. ity and disability, and ultimately to prolong survival and improve quality of life. Data available from the included trials were mainly secondary outcomes. There was a lack of data from RCTs on clin- ically relevant outcomes at long-term follow up, such as mortality AUTHORS’ CONCLUSIONS and quality of life. Therefore, we were not able to draw conclu- sions about these important outcomes. Implications for practice The quality of acupuncture treatment is closely related to its effec- This systematic review does not provide evidence to support the tiveness. Misleading results may occur if the treatment schedules routine use of acupuncture for patients with subacute or chronic were inadequate or administered by unskilled practitioners. How- stroke. ever, information on the experience and training of the acupunc- turists who gave the treatments was not available in any of the in- Implications for research cluded trials. Furthermore, the acupuncture techniques, the num- The widespread use of acupuncture, the promising results with less ber of acupoints, the number and duration of sessions and the severe side effects, lower cost, and the insufficient quality of the duration of the intervention period varied across trials. Some tri- available trials warrant further research. Large sham or placebo- als reported that the acupuncture points, the number of sessions controlled trials are needed to confirm or refute the available evi- and the duration of treatment were individualised according to dence. The following features should be addressed in further stud- the practical conditions in each stroke patient. From the scarce ies: description of treatment in all trials, it is difficult to evaluate if the acupuncture treatment was valid or not. The consensus of an (1) detailed reporting of the generation of the allocation sequence international group of experienced acupuncturists and researchers and allocation concealment; Acupuncture for stroke rehabilitation (Review) 8 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 11. (2) application and clear description of blinding; (3) use of placebo or sham acupuncture as the control; (4) clear definition of the modality of acupuncture, acupunc- ture technique based on evidence or a consensus of experts (STRICTA); (5) use of standard validated outcome measures; (6) reporting of clinically important outcome measures at long- term follow up, such as mortality, disability and quality of life; (7) adverse events critically assessed by standardised monitoring or an effective self-report system. Attention should be paid to rare, severe adverse events relevant to acupuncture; and (8) the study should be reported according to the STRICTA cri- teria in conjunction with the CONSORT statement. ACKNOWLEDGEMENTS We would like to thank Mrs Hazel Fraser for providing us with relevant trials and systematic reviews from the Cochrane Stroke Group Trials Register; Mrs Brenda Thomas for her help with de- veloping the search strategy and helpful comments; Kelvin Tsoi and Wilson Tam for providing us with statistical support; and Yanling Zhang, Jun Li and Hongwei Zhang for copying trials. We would also like to express our gratitude to Dr Livia Candelise, Dr Andrew Vickers, Dr Steff Lewis and lead editor Prof Peter Sander- cock for their very helpful comments. REFERENCES References to studies included in this review Wang 2001 {published data only} Wang ZY. Compared analyses among Chinese medicinal herbs, Dai 1997 {published data only} acupuncture and combination of them in the treatment of Dai CY, Lu F. Efficacy observation on the treatment of sequelae of hemiplegia in stroke patients. Modern Rehabilitation 2001;5(6): cerebral infarction in 46 patients by integrating Chinese with 130. modern medicine. The Practical Journal of Integrating Chinese with Modern Medicine 1997;10(1):438. References to studies excluded from this review Li 1997a {published data only} Li X. Efficacy analysis of rehabilitation in the treatment of Che 2002 {published data only} hemiplegia in 112 stroke patients. Chinese Journal of Rehabitative Che JL. Clinical observation on the treatment of sequelae of stroke Theory and Practice 1997;3(1):22–4. in 40 patients with relatively long duration by needling abdominal points. Jiangshu Traditional Chinese Medicine 2002;23(1):31–2. Lun 1999 {published data only} Lun X, Peng ZF, Peng SZ. Clinical observation on the treatment of Chen 2000 {published data only} sequelae of stroke by needling temporal three points. Journal of Chen ZF, Jiang GH, Lai XS. Clinical observation on the electronic Clinical Acupuncture and Moxibustion 1999;15(4):8–9. acupuncture treating vascular dementia. Journal of Clinical Acupuncture and Moxibustion 2000;16(12):18–20. Naeser 1992 {published data only} Naeser MA, Alexander MP, Stiassny-Eder D, Galler V, Hobbs J, Cui 1992 {published data only} Bachman D. Real versus sham acupuncture in the treatment of Cui YY, Zhao JH. Observation on the effects of acupuncture in the paralysis in acute stroke patients: a CT scan lesion site study. treatment of compulsory cry and laugh due to multiple cerebral Journal of Neurorehabilitation 1992;6(4):163–73. infarction. Acupuncture Research 1992;1:19–20. Acupuncture for stroke rehabilitation (Review) 9 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 12. Ding 2000 {published data only} of vascular dementia. Chinese Acupuncture and Moxibustion. 1997; Ding BY, Cui YJ. Observation on the effects of acupuncture and 12:713–6. cupping in the treatment of arthritic spasm in 52 stroke patients. Lai 1997b {published data only} Shanghai Journal of Acupuncture and Moxibustion 2000;19(3):28. Lai XS. Observation on the effects of acupuncture in the treatment Feng 1996a {published data only} of senile vascular dementia. Chinese Acupuncture and Moxibustion. Feng SL, Guo ZQ, Xie GR. Observation on the clinical effects of 1997;4:201–2. hemiplegia from apoplexy treated by needling Feng Fu, Ya Men Lai 1998 {published data only} points. Journal of Hunan College of Traditional Chinese Medicine Lai XS, Muo FZ, Jiang GH, Li JQ, Chen JL, Kuang ZY. 1996;16(3):61–4. Observation of clinical effect of acupuncture on vascular dementia Feng 1996b {published data only} and its influence on superoxide dismutase, lipid peroxide and nitric Feng SL, Guo ZQ, Xie GR. Clinical observation and experimental oxide. Chinese Journal of Integrated Traditional and Western studies on the treatment of cerebral thrombus by needling “Feng Medicine 1998;18(11):648–51. Fu, Ya Men” points. Journal of Guangzhou University of Traditional Lai 2004 {published data only} Chinese Medicine 1996;13(2):20–3. Lai XS, Liu JY, Jiang GM. Effect of combination of acupuncture Fink 2004 {published data only} treatment at Liench’ tian and plum blosson needle treatment at the Fink M, Rollnik JD, Bijak M, Borstadt C, Dauper J, Guergreltcheva tip of tongue on aphasia after stroke and its significance on V, et al.Needle acupuncture in chronic poststroke leg spasticity. hemorheology. Chinese Journal of Clinical Rehabilitation 2004;8 Archives of Physical Medicine and Rehabilitation 2004;85(4):667–72. (19):3818–20. Gao 2001 {published data only} Li 1993 {published data only} Gao HY, Yan LF, Liu BB, Wang Y, Wei XL, Sun LY. Clinical study Li YH, Jin R. Three forms of acupuncture in the treatment of on treatment of senile vascular dementia by acupuncture. Journal of sequelae of cerebrovascular accident. Chinese Acupuncture and Traditional Chinese Medicine 2001;21(2):103–9. Moxibustion. 1993;4:35–38. Guo 1999 {published data only} Li 1994a {published data only} Guo XY, Wei YF. Clinical observation on the effects of acupuncture Li CD, Huang Y, LI YK, Hu KM, Jiang ZY. Treating post-stroke in the treatment of muscular spasm in 124 stroke patients. Chinese depression with “mind-refreshing antidepressive” acupuncture Acupuncture and Moxibustion 1999;9:553. therapy: a clinical study of 21 cases. International Journal of Hou 1998 {published data only} Clinical Acupuncture 1994;5(4):389–93. Hou AL, Wang L, Pu Y. Clinical observation of treatment of multi- infarctional dementia by herbs, oxygen and acupuncture. Shanghai Li 1994b {published data only} Journal of Acupuncture and Moxibustion 1998;17(2):12–3. Li YH, Jin R. Clinical study in the treatment of sequelae of cerebrovascular accident with needling temporal points. Huang 2002 {published data only} Acupuncture Research 1994;19(2):4–6. Huang HS, Huang X. Effects of “Chuan Xiong Qin and Huang Qi” on the treatment of upper and lower limbs pain due to stroke in 50 Li 1997b {published data only} patients. Research of Traditional Chinese Medicine 2002;18(1):9–10. Li YH, Yang WH, Zhuang LX, Zheng L. Clinical study of acupuncture in the treatment of vascular dementia. Acupuncture Jiang 1998 {published data only} Research 1997;3(1):22–4. Jiang GH, Li YH, Yang WH. Clinical observation on treatment of sequelae of stroke with CT localization acupuncture. Shanghai Liang 1993 {published data only} Journal of Acupuncture and Moxibustion 1998;17(2):6–7. Liang R. Clinical observation and experimental studies on the treatment of sequalae of stroke by needling temporal points. Jiang 2000 {published data only} International Journal of Clinical Acupuncture 1993;4(1):19–26. Jiang ZY, Li CD. Needling paravertebral points in treatment of post-stroke thalamic pain. International Journal of Clinical Liu 1998 {published data only} Acupuncture 2000;11(1):7–10. Liu J, Peng XH, Lin DD, Li CD, Jiang ZY, Zeng L, et al.Clinical Jin 1993 {published data only} study of electrical scalp acupuncture in the treatment of vascular Jin R, Lai XS, Li YH, Liang RA. Clinical observation on the dementia. Chinese Acupuncture and Moxibustion 1998;4:197–200. treatment of sequelae of stroke with three temporal needling. Liu 2004a {published data only} Chinese Acupuncture and Moxibustion. 1993;2:11–2. Liu HG, Huang SJ, Zhang HL. Effect of Jiawei simiao yongan tang Kjendahl 1997 {published data only} combined with electroacupuncture in the recovery of motor Kjendahl A, Sallstrom S, Osten PE, Stanghelle JK. A one year function in patients with post-stroke sequelae. Chinese Journal of follow-up study on the effects of acupuncture in the treatment of Clinical Rehabilitation 2004;8(10):1916–7. stroke patients in the subacute stage: a randomized, controlled Liu 2004b {published data only} study. Clinical Rehabilitation 1997;11:192–200. Liu GF, Zhou ZX. Effect of neuromuscular electrical stimulation in Lai 1997a {published data only} various modes on the rehabilitation of upper limb function in Lain XS, Zhang JW, Muo FZ, Jiang GH, Li JQ, Zheng ZC. patients with hemiplegia. Chinese Journal of Clinical Rehabillitation Clinical observation on the effects of acupuncture in the treatment 2004;8(22):4424–5. Acupuncture for stroke rehabilitation (Review) 10 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 13. Muo 2001 {published data only} Zhang 1988 {published data only} Muo FZ, Li JQ, Chu L, Lai XS, Zhang JX, Liu SH. The influence Zhang ZJ. Observation on the effects of acupuncture in the of electroacupuncture on intelligence structure in patients with treatment of post stroke aphasia. Journal of Traditional Chinese vascualr dementia. Shanghai Journal of Acupuncture and Medicine 1988;5:33–4. Moxibustion 2001;20(6):6–8. Zhang 1997 {published data only} Zhang SF, Ye XR, Dan QH, Zhang WF, Cui Y. The influence of Qi 2000 {published data only} acupuncture on SOD, CAT and serum intraproteinal fluorescence Qi GS, Zhao XJ. Clinical observation on the treatment of stroke in in stroke patients. Chinese Acupuncture and Moxibustion 1997;9: rehabilitative stage with “Qing Nao YI Shui” acupuncture therapy. 517–9. Journal of Tianjin College of Traditional Chinese Medicine 2000;19 (2):23–5. Zhang 2002a {published data only} Zhang L, Sheng L. Clinical observation on the effects of magnetic Qu 1991 {published data only} acupuncture in the treatment of sequelae of stroke. Chinese Journal Qu RM, Qu XM. Observation on the treatment of sequelae of of Information in Traditional Chinese Medicine 2002;9(1):63. apoplexy in 105 patients with needling Du Mai points. Jiang XI Zhang 2002b {published data only} Traditional Chinese Medicine 1991;22(5):32–4. Zhang ZZ, Ma FJ, Ma Y. Observation on the effects of acupuncture Sallstrom 1996 {published data only} in the treatment of urinary retention due to cerebral infarction in Sallstrom S, Kjendahl A, Osten PE, Stanghelle JH, Borchgrevink 36 patients. He Long Jiang Medicine and Pharmacy 2002;25(3):71. CF. Acupuncture in the treatment of stroke patients in the subacute Zheng 1996 {published data only} stage: a randomized, controlled study. Complementary Therapies in Zheng ZC. Needling scalp and lumbo-sacral points in the Medicine 1996;4:193–7. treatment of sequelae of stroke in 634 patients. Shanghai Journal of Acupuncture and Moxibustion 1996;15(5):15–6. Sui 2001 {published data only} Sui MH. Clinical observation on the therapeutic effect of Zheng 2000 {published data only} acupuncture of hand-Shaoyang Meridian acupoints for treatment Zheng L, Zhuang LX, Li YH. Observations on the curative effect of of upper-limb spasm in stroke patients. Acupuncture Research 2001; acupuncture and herbs on vascular dementia. Shanghai Journal of 26(2):131–3. Acupuncture and Moxibustion 2000;19(3):8–10. Zhou 1995 {published data only} Tong 1997 {published data only} Zhou JZ, Zhang CS, LI L, Zhou SW, Shi XM. Clinical study in the Tong LG, Li ZG, Ren YX, Yang J, Meng XY, Ma WM, et al.Body treatment of sequelae of apoplexy with “Xing Nao Kai Qiao” quick-needling plus scalp acupuncture in treating apoplexy and its acupuncture therapy. Chinese Acupuncture and Moxibustion. 1995; sequelae. International Journal of Clinical Acupuncture 1997;9(2): 3:6–8. 133–9. Wang 2004 {published data only} Additional references Wang MH, Liang B, Wang M, Chen LQ, Wang YS, Li ZY, et Asplund 1988 al.Clinical study on the treatment of apopletic hemiplegia by Asplund K, Tuomilehto J, Stegmayr B, Wester PO, Tunstall-Pedoe acupuncture plus rehabilitation training. Shanghai Journal of H. Diagnostic criteria and quality control of the registration of Acupuncture and Moxibustion 2004;23(4):7–9. stroke events in the MONICA project. Acta Medica Scandinavica Supplementum 1988;728:26–39. Wei 2005 {published data only} Wei LL. Effect of Shuitu acupoint injection with stellate ganglion Bonita 1990 block on swallow dysfunction after stroke. Chinese Journal of Bonita R. Epidemiology of stroke. Lancet 1992;339:342–4. Clinical Rehabilitation 2005;9(9):106–7. Deeks 2001 Deeks JJ, Altman DG, Bradburn MJ. Statistical methods for Wu 1999 {published data only} examining heterogeneity and combining results from several studies Wu D. Needling eye points combined with electrical scalp points in in meta-analysis. In: Egger M, Davey Smith G, Altman DG editor the treatment of urinary incontinence due to cerebrovascular (s). Systematic Reviews in Health Care: Meta-analysis in Context. accident in 40 patients. Journal of Clinical Acupuncture and 2nd Edition. London: BMJ Publication Group, 2001. Moxibustion 1999;15(9):33–4. DPH 1994 Xiao 1996 {published data only} Director of Public Health. Annual report. Southampton and South Xiao F, Liu W. Observation on treatment of dysphasia due to West Hampshire Health Commission 1994. cerebrovascular accident with electroacupuncture at shegen point. Egger 1997 Journal of Chinese Medicine 1996;50:27–9. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta- Yu 2002 {published data only} analysis detected by a simple graphical test. BMJ 1997;315:629–34. Yu MZ, Yang GZ, Zhan KR, Fan YY, Peng MF. Clinical Ernst 1996 observation on the effects of electrical stimulation at nervous trunk Ernst E, White AR. Acupuncture as an adjuvant therapy in stroke points in the treatment of sequalae of stroke. Chinese Journal of rehabilitation?. Wien Medizinische Wochenschrift 1996;21(22): Basic Medicine in Traditional Chinese Medicine 2002;8(3):55–6. 556–8. Acupuncture for stroke rehabilitation (Review) 11 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 14. Ernst 1997 Park 2001 Ernst E, White A. Life -threatenting adverse reactions after Park J, Hopwood V, White AR, Ernst E. Effectiveness of acupuncture? A systematic review. Pain 1997;71:123–6. acupuncture for stroke: a systematic review. Journal of Neurology Hopwood 1996 2001;248:558–63. Hopwood V. Acupuncture in stroke recovery: a literature review. Schulz 1995 Complementary Therapies in Medicine 1996;4:258–63. Schulz KF, Chalmers I, Hayes R, Altman D. Empirical evidence of Hu 1993 bias. JAMA 1995;273(5):408–12. Hu HH, Chung C, Liu TJ, Chen RC, Chen CH, Chou P, et al.A SCSSS 1999 randomized controlled trial on the treatment for acute partial Swedish Collaboration on Sensory Stimulation in Stroke. Sensory ischemic stroke with acupuncture. Neuroepidemiology 1993;12: stimulation after stroke: a randomized controlled trial. 106–13. Cerebrovascular Diseases 1999;9 (Suppl 1):28. Jansen 1989 Jansen G, Lundeberg T, Kjartansson S, Samuelson UE. Smith 2002 Acupuncture and sensory neuropeptides increase cutaneous blood Smith LA, Moore OA, McQuay HJ, Moore A. Assessing the flow in rats. Neuroscience Letters 1989;97:305–9. evidence of effectiveness of acupuncture for stroke rehabilitation: Johansson 1993 stepped assessment of likelihood of bias. www.jr2.ox.ac.uk/ Johansson K, Lindgren I, Widner H, Wiklund I, Johansson BB bandolier/booth/alternat/ACstroke.html. [electronic database] Johansson K, et al.Can sensory stimulation improve the functional 2002. outcome in stroke patients?. Neurology 1993;43:2189–92. Sulter 1999 Johansson 2001 Sulter G, Steen C, Keyser JD. Use of the Barthel Index and Johansson BB, Haker E, Arbin MV. Acupuncture and Modified Rankin Scale in acute stroke trials. Stroke 1999;30: transcutaneous nerve stimulation in stroke rehabilitation: A 1538–41. randomized, controlled trial. Stroke 2001;32:707–13. Sun 2001 Kjaergard 1999 Sun HL, Li XM. Clinical study on treatment of cerebral apoplexy Kjaergard LL, Villumsen J, Gluud C. Quality of randomised with penetration needling of scalp acupoints. Chinese Acupuncuture clinical trials affects estimates of intervention efficacy (abstract). and Moxibustion 2001;21(5):275–8. VII Cochrane Colloquium. 1999:57. Sze 2002 Liu 2005 Sze FK, Wong E, Or KKH, Lau J, Woo J. Does acupuncture have Zhang SH, Liu M, Asplund K, Li L. Acupuncture for acute stroke. additional value to standard poststroke motor rehabilitation?. Cochrane Database of Systematic Reviews 2005, Issue 2. [Art. No.: Stroke 2002;33:186–94. CD003317. DOI: 10.1002/14651858.CD003317.pub2] MacPherson 2001 Van Tulder 2000 MacPherson H, White A, Cummings M, Jobst K, Rose K, Van Tulder MW, Cherkin DC, Berman B, Lao L, Koes BW. Niemtzow R. Standards for reporting interventions in controlled Acupuncture for low back pain. Cochrane Database of Systematic trials of acupuncture: the STRICTA recommendations. Reviews 2002, Issue 2. [Art. No.: CD001351. DOI: 10.1002/ Complementary Therapies in Medicine 2001;9:246–9. 14651858.CD001351.pub2] Magnusson 1994 Vickers 1998 Magnusson M, Johansson K, Johansson BB. Sensory stimulation Vickers A, Goyal N, Harland R, Rees R. Do certain countries promotes normalization of postural control after stroke. Stroke produce only positive results? A systematic review of controlled 1994;25:1176–80. trials. Controlled Clinical Trials 1998;19:159–66. MOH PRC 1999 White 2001 Ministry of Health, PRC. Chinese Health Statistical Digest 1999: White AR, Filshie J, Cummings TM. Clinical trials of acupuncture: 69–70. consensus recommendations for optimal treatment, sham controls Moher 1998 and blinding. Complementary Therapies in Medicine 2001;9: Moher D, Pham B, Jones A, Cook DJ, Jadad AR, Moher M, et 237–45. al.Does quality of reports of randomised trials affect estimates of intervention efficacy reported in meta-analysis. Lancet 1998;352: Wu 1996 609–13. Wu JN. A short history of acupuncture. Journal of Alternative Complementary Medicine 1996;2:19–21. NIH 1998 NIH Consensus Conference. Acupuncture. JAMA 1998;280: Zhang 1996 1518–24. Zhang X, Yuan Y, Kuang P. The changes of vasoactive intestinal O’Mahony 1999 peptide, somato-statin and pancreatic polypeptide in blood and O’Mahony PG, Thomson RG, Dobon R. The prevalence of stroke CSF of acute cerebral infarction patients and the effect of and associated disability. Journal of Public Health Medicine 1999;21 acupuncture on them. Chen Tzu Yen Chiu 1996;21(4):10–6. (2):1666–71. ∗ Indicates the major publication for the study Acupuncture for stroke rehabilitation (Review) 12 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 15. CHARACTERISTICS OF STUDIES Characteristics of included studies [ordered by study ID] Dai 1997 Methods RCT. Method of randomisation: not stated. Blinding: not stated. ITT analysis: not stated. Losses to FU: none. Participants Country: China. Number of patients included: 136 (46/45/45). Demographics: aged 48 to 86 yrs, 75% male. Type of stroke: cerebral infarction only. Diagnosis: WHO definition and all confirmed by CT. Severity on entry: mild to severe. Time from stroke onset: 3 to 14 months. Setting: unclear. Comparability: unclear. Interventions 3 arms: (1) acupuncture + WM; (2) acupuncture only; (3) WM only. Comparison eligible: acupuncture + WM versus WM. Acupuncture treatment (1) Acupuncture rationale: not stated. (2) Needling details: points used: both body and scalp (three temporal points) acupoints. Numbers of points used: 5 - 11 points. Depths of insertion: 1.5 - 2.0 inch. Deqi elicited: Yes. Needle stimulation: manual. Needle retention time: 30 min. Needle type: stainless steel, gauge 26 - 30# with length 2.0 - 2.6 inch. (3) Treatment regimen: Number of treatment sessions: 30 sessions. Frequency of treatment: 1 session/day. Total course: 30 days. (4) Practitioner background: not stated. (5) Co-intervention: aspirin 25 mg qd orally. Control interventions WM: aspirin 25 mg qd orally. Outcomes Number of patients with improvement in global neurological deficit (CSRS 1 score decrease > 18%) at the end of treatment. Acupuncture for stroke rehabilitation (Review) 13 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 16. Dai 1997 (Continued) FU: 30 days. Notes Conclusion of authors: positive. Risk of bias Item Authors’ judgement Description Allocation concealment? Unclear B - Unclear Li 1997a Methods RCT. Method of randomisation: not stated. Blinding: not stated. ITT analysis: not stated. Losses to FU: none. Participants Country: China. Number of patients included: 112 (42/20/50). Demographics: aged 24 to 76 yrs. Type of stroke: both ischemic and hemorrhagic strokes. Diagnosis: WHO definition and all confirmed by CT or MRI. Severity on entry: not stated. Time from stroke onset: 1 month to 8.5 yrs. Setting: inpatients. Comparability: comorbidity and past history similar. Interventions 3 arms: (1) acupuncture + PT and OT (2) acupuncture only (3) PT and OT. Comparison eligible: acupuncture + PT and OT versus PT and OT. Acupuncture treatment (1) Acupuncture rationale: not stated. (2) Needling details Points used: both body and scalp acupoints. Numbers of points used: 10 - 12 points. Depths of insertion: not stated. Deqi elicited: unclear. Needle stimulation: manual. Needle retention time: 30 min. Needle type: not stated. (3) Treatment regimen Number of treatment sessions: 72 sessions. Frequency of treatment: 6 sessions/week. Total course: 3 months. (4) Practitioner background: not stated. (5) Co-intervention: baseline medication, PT and OT. Acupuncture for stroke rehabilitation (Review) 14 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 17. Li 1997a (Continued) Control interventions: baseline medication plus PT and OT. Outcomes Number of patients with improvement in global neurological deficit (CSRS 1 score) at the end of treat- ment. FU: 3 months. Notes Conclusion of authors: positive. Risk of bias Item Authors’ judgement Description Allocation concealment? Unclear B - Unclear Lun 1999 Methods RCT. Method of randomisation: not stated. Blinding: not stated. ITT analysis: not stated. Losses to FU: none. Participants Country: China. Number of patients included: 109 (61/48). Demographics: aged 35 to 75 yrs, 60% male. Type of stroke: both ischemic and hemorrhagic strokes. Diagnosis: WHO definition and all confirmed by CT before entry. Severity on entry: not stated. Time from stroke onset: 2 months to 5 years. Setting: unclear Comparability: unclear. Interventions Comparison: acupuncture + TCM versus TCM. Acupuncture treatment (1) Acupuncture rationale: not stated. (2) Needling details Points used: scalp acupoints. Numbers of points used: 5 points. Depths of insertion: 2 - 2.6 inch. Deqi elicited: yes. Needle stimulation: manual. Needle retention time: 30 min. Needle type: stainless steel needle with gauge 28 - 30# and length 2.6 inch. (3) Treatment regimen Number of treatment sessions: 30 sessions. Frequency of treatment: 1/day. Total course: 45 days. (4) Practitioner background: not stated. Acupuncture for stroke rehabilitation (Review) 15 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 18. Lun 1999 (Continued) (5) Co-intervention: baseline TCM. Control interventions TCM. Outcomes Number of patients with improvement in global neurological deficit (CSRS 2) at the end of treatment. FU: 45 days. Notes Conclusion of authors: positive. Risk of bias Item Authors’ judgement Description Allocation concealment? Unclear B - Unclear Naeser 1992 Methods RCT. Method of randomisation: not stated. Blinding: patients blinded. ITT analysis: not stated. Losses to FU: none. Participants Country: USA. Number of patients included: 16 (10/6). Demographics: aged 44 to 74 yrs. Type of stroke: cerebral infarction only. Diagnosis: all confirmed by CT. Severity on entry: moderate. Time from stroke onset: 1 - 3 months. Setting: inpatients. Comparability: no significant difference in age or time post onset. Interventions Comparison: real acupuncture + daily PT versus sham acupuncture + daily PT. Acupuncture treatment (1) Acupuncture rationale: not stated. (2) Needling details Points used: both body and scalp acupoints. Numbers of points used: 11 body acupoints; numbers of scalp acupoints unclear. Depths of insertion: unclear. Deqi elicited: unclear. Needle stimulation: electrical stimulation with frequency 1 - 2 Hz and amplitude unknown. Needle retention time: 20 min. Needle type: gauge 34# , length and material unknown. (3) Treatment regimen Number of treatment sessions: 20 sessions. Frequency of treatment: 5 sessions/week. Acupuncture for stroke rehabilitation (Review) 16 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 19. Naeser 1992 (Continued) Total course: 4 weeks. (4) Practitioner background: not stated. (5) Co-intervention: PT. Control interventions: sham acupuncture + PT. Outcomes Number of patients with improvement in motor function (BMIT) within 5 days after completing treat- ment. FU: 35 days. Notes Conclusion of authors: negative. Risk of bias Item Authors’ judgement Description Allocation concealment? Unclear B - Unclear Wang 2001 Methods RCT. Method of randomisation: not stated. Blinding: not stated. ITT analysis: not stated. Losses to FU: none. Participants Country: China. Number of patients included: 90 (34/30/26). Demographics: aged 39 to 75 yrs, 56% male. Type of stroke: both ischemic and hemorrhagic strokes. Diagnosis: WHO definition and all confirmed by CT. Severity on entry: not stated. Time from stroke onset: 2 months to 5 years. Setting: unclear. Comparability: unclear. Interventions 3 arms: (1) acupuncture + TCM (2) acupuncture only (3) TCM only. Comparison eligible: acupuncture + TCM versus TCM only. Acupuncture treatment (1) Acupuncture rationale: not stated. (2) Needling details Acupuncture for stroke rehabilitation (Review) 17 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 20. Wang 2001 (Continued) Points used: body acupoints. Numbers of points used: 6 - 10 points. Depths of insertion: not stated. Deqi elicited: yes. Needle stimulation: manual. Needle retention time: 30 min. Needle type: not stated. (3) Treatment regimen Number of treatment sessions: 60 sessions. Frequency of treatment: 10 sessions/13 days. Total course: 78 days. (4) Practitioner background: not stated. (5) Co-intervention: TCM. Control interventions: TCM. Outcomes Number of patients with improvement in global neurological deficit (CSRS 2) at the end of treatment. FU: 78 days. Notes Conclusion of authors: positive. Risk of bias Item Authors’ judgement Description Allocation concealment? Unclear B - Unclear BMIT: Boston Motor Inventory Test CSRS 1: Chinese Stroke Recovery Scale based on the revised diagnostic criteria of acute cerebral infarction formulated by the second National Academic Symposium on Cerebrovascular Diseases of the Chinese Medical Association in 1986, which is similar to the Revised Scandinavian Stroke Scale (RSSS) CSRS 2: Chinese Stroke Recovery Scale based on principles of traditional Chinese medicine CT: computerised tomography FU: follow up ITT: intention to treat OT: occupational therapy MRI: magnetic resonance imaging PT: physical therapy RCT: randomised controlled trial TCM: Traditional Chinese Medicine WHO: World Health Organization WM: Western medicine yrs: years Acupuncture for stroke rehabilitation (Review) 18 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 21. Characteristics of excluded studies [ordered by study ID] Che 2002 The trial aimed to assess effects of two methods of acupuncture on subacute or chronic stroke. Acupoints were different between the two groups. Chen 2000 Confounded; the trial aimed to assess effects of acupuncture only compared with drug therapy (such as WM or TCM). Cui 1992 It was not possible to include data from this trial in the analysis (see reason for exclusion for Chen 2000). Ding 2000 It was not possible to include data from this trial in the analysis (see reason for exclusion for Che 2002). Feng 1996a It was not possible to include data from this trial in the analysis (see reason for exclusion for Che 2002). Feng 1996b It was not possible to include data from this trial in the analysis (see reason for exclusion for Che 2002). Fink 2004 It was not possible to include data from this trial in the analysis. MAS scores (as a clinical measure of spasticity) were assessed before and after the treatment period but mean change of neurological score after the treatment period was not available. Gao 2001 It was not possible to include data from this trial in the analysis (see reason for exclusion for Chen 2000). Guo 1999 The trial aimed to assess effects of acupuncture only compared with PT and OT. Hou 1998 The trial aimed to assess the effects of the combination of acupuncture, oxygen and herbs compared with acupuncture or oxygen respectively. Huang 2002 It was not possible to include data from this trial in the analysis (see reason for exclusion of Chen 2000). Jiang 1998 It was not possible to include data from this trial in the analysis (see reason for exclusion of Che 2002). Jiang 2000 It was not possible to include data from this trial in the analysis (see reason for exclusion of Chen 2000). Jin 1993 It was not possible to include data from this trial in the analysis (see reason for exclusion of Che 2002). Kjendahl 1997 The trial included patients with stroke less than one month since onset although the median time to post-stroke was more than one month. Lai 1997a The trial aimed to assess the effects of electrical acupuncture compared with manual acupuncture. Acupoints were the same between the two groups. Lai 1997b It was not possible to include data from this trial in the analysis (see reason for exclusion of Chen 2000). Lai 1998 The trial aimed to assess the effects of electrical acupuncture compared with manual acupuncture. Acupoints were the same between two groups. Acupuncture for stroke rehabilitation (Review) 19 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 22. (Continued) Lai 2004 It was not possible to include data from this trial in the analysis (see reason for exclusion of Che 2002). Li 1993 It was not possible to include data from this trial in the analysis (see reason for exclusion of Che 2002). Li 1994a It was not possible to include data from this trial in the analysis (see reason for exclusion of Che 2002). Li 1994b It was not possible to include data from this trial in the analysis (see reason for exclusion of Che 2002). Li 1997b No outcome measures of interest were available in the trial, which aimed to assess the effects of acupuncture on the change of intellectual disturbance in patients with vascular dementia caused by cerebrovascular disease. Liang 1993 It was not possible to include data from this trial in the analysis (see reason for exclusion of Che 2002). Liu 1998 It was not possible to include data from this trial in the analysis. ADL was assessed before and after the treatment period but the number of patients being independent after the treatment period was not available. The trial primarily aimed to assess the effects of acupuncture on the intellectual disturbance in patients with vascular dementia due to stroke. Liu 2004a Confounded (acupuncture plus one kind of Chinese herbs versus another kind of Chinese herbs). Liu 2004b The trial included patients with stroke less than one month since onset (that is, acute stroke). Muo 2001 It was not possible to include data from this trial in the analysis (see reason for exclusion of Chen 2000). Qi 2000 It was not possible to include data from this trial in the analysis (see reason for exclusion of Che 2002). Qu 1991 It was not possible to include data from this trial in the analysis (see reason for exclusion of Che 2002). Sallstrom 1996 The median time to post-stroke was more than one month, but the trial included patients with acute stroke (that is, less than one month). Sui 2001 It was not possible to include data from this trial in the analysis (see reason for exclusion of Che 2002). Tong 1997 It was not possible to include data from this trial in the analysis (see reason for exclusion of Che 2002). Wang 2004 Randomisation was not stated; the trial included patients with acute stroke (less than one month since onset). Wei 2005 The trial included patients with ischemic stroke less than one month since onset (that is, acute stroke). Data for patients with hemorraghic stroke more than 150 days were not separated from data for patients with ischemic stroke. Wu 1999 Data on pre-planned outcome measures of interest were not available from the trial, which aimed to assess the effects of acupuncture on urinary incontinence in stroke patients. Xiao 1996 It was not possible to include data from this trial in the analysis (see reason for exclusion of Che 2002). Acupuncture for stroke rehabilitation (Review) 20 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 23. (Continued) Yu 2002 It was not possible to include data from this trial in the analysis (see reason for exclusion of Che 2002 and Chen 2000). Zhang 1988 It was not possible to include data from this trial in the analysis (see reason for exclusion of Chen 2000). Zhang 1997 It was not possible to include data from this trial in the analysis (see reason for exclusion of Chen 2000). Zhang 2002a The trial aimed to assess the effects of magnetic acupuncture compared with routine acupuncture. Acupoints were similar between the two groups. Zhang 2002b The duration of stroke since onset was not stated in the trial; it was not possible to include data from this trial in the analysis (see reason for exclusion of Chen 2000). Zheng 1996 The trial aimed to assess the effects of combination of body and scalp acupuncture compared with body acupuncture only or scalp acupuncture only. Zheng 2000 Confounded (acupuncture plus Chinese herbs versus WM only). Zhou 1995 It was not possible to include data from this trial in the analysis (see reason for exclusion of Che 2002). ADL: activities of daily living MAS: Modified Ashworth Scale OT: occupational therapy PT: physical therapy TCM: Traditional Chinese Medicine WM: Western medicine Acupuncture for stroke rehabilitation (Review) 21 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 24. DATA AND ANALYSES Comparison 1. Acupuncture versus open control No. of No. of Outcome or subgroup title studies participants Statistical method Effect size 1 Improvement of global 4 352 Odds Ratio (M-H, Random, 95% CI) 6.55 [1.89, 22.76] neurological deficit at the end of treatment Analysis 1.1. Comparison 1 Acupuncture versus open control, Outcome 1 Improvement of global neurological deficit at the end of treatment. Review: Acupuncture for stroke rehabilitation Comparison: 1 Acupuncture versus open control Outcome: 1 Improvement of global neurological deficit at the end of treatment Study or subgroup Treatment Control Odds Ratio Weight Odds Ratio n/N n/N M-H,Random,95% CI M-H,Random,95% CI Dai 1997 44/46 30/45 24.6 % 11.00 [ 2.34, 51.65 ] Li 1997a 38/42 42/50 28.0 % 1.81 [ 0.50, 6.49 ] Lun 1999 59/61 25/48 24.9 % 27.14 [ 5.94, 123.93 ] Wang 2001 32/34 21/26 22.5 % 3.81 [ 0.68, 21.48 ] Total (95% CI) 183 169 100.0 % 6.55 [ 1.89, 22.76 ] Total events: 173 (Treatment), 118 (Control) Heterogeneity: Tau2 = 1.02; Chi2 = 8.18, df = 3 (P = 0.04); I2 =63% Test for overall effect: Z = 2.96 (P = 0.0031) 0.1 0.2 0.5 1 2 5 10 Favours control Favours treatment Acupuncture for stroke rehabilitation (Review) 22 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 25. APPENDICES Appendix 1. MEDLINE search strategy The following search strategy, using a combination of controlled vocabulary and text word terms, was used for MEDLINE and was modified to suit other databases. MEDLINE (Ovid) 1 exp cerebrovascular disorders/ 2 (stroke$ or poststroke$ or cva$).tw. 3 (cerebrovascular$ or cerebral vascular).tw. 4 (cerebral or cerebellar or brainstem or vertebrobasilar).tw. 5 (infarct$ or isch?emi$ or thrombo$ or apoplexy or emboli$).tw. 6 4 and 5 7 (cerebral or intracerebral or intracranial or parenchymal).tw. 8 (brain or intraventricular or brainstem or cerebellar).tw. 9 (infratentorial or supratentorial).tw. 10 7 or 8 or 9 11 (haemorrhage or hemorrhage or haematoma or hematoma).tw. 12 (bleeding or aneurysm).tw. 13 11 or 12 14 10 and 13 15 1 or 2 or 3 or 6 or 14 16 acupuncture/ 17 exp acupuncture therapy/ 18 electroacupuncture/ 19 meridians/ 20 acupuncture points/ 21 acupuncture$.tw. 22 (electroacupuncture or electro- acupuncture).tw. 23 acupoints.tw. 24 ((meridian or non-meridian or trigger) adj10 point$).tw. 25 or/16-24 26 15 and 25 27 limit 26 to human WHAT’S NEW Last assessed as up-to-date: 15 March 2006. 20 October 2008 Amended Contact details updated Acupuncture for stroke rehabilitation (Review) 23 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 26. HISTORY Protocol first published: Issue 2, 2003 Review first published: Issue 3, 2006 15 July 2008 Amended Converted to new review format. CONTRIBUTIONS OF AUTHORS Wu HM: developing search strategy, assessment of studies, data extraction, data analysis, data entry, writing of protocol and review. Lin XP: assessment of studies, data extraction, data entry. Tang JL: developing search strategy, assessment of studies, data analysis, resolution of disagreements, writing protocol and review. Lau J: expertise, suggestions and corrections. Leung PC: expertise, suggestions and corrections. Woo J: expertise, suggestions and corrections. Li YP: expertise, suggestions and corrections. DECLARATIONS OF INTEREST None known SOURCES OF SUPPORT Internal sources • Hong Kong Branch of Chinese Cochrane Center, China. • Department of Community & Family Medicine, Chinese University of Hong Kong, China. • Chinese Cochrane Center, West China Hospital, Si Chuan University, China. External sources • Hong Kong Croucher Foundation, China. Acupuncture for stroke rehabilitation (Review) 24 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  • 27. INDEX TERMS Medical Subject Headings (MeSH) ∗ Acupuncture Therapy; Randomized Controlled Trials as Topic; Stroke [∗ rehabilitation] MeSH check words Humans Acupuncture for stroke rehabilitation (Review) 25 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.