Cerebral function imaging of acupuncture treatment for stroke: A protocol for systematic review and meta-analysis

Background: Stroke is one of the most common causes of death and is the main cause of persistent and acquired disability in adults worldwide. Acupuncture is recommended as an alternative and complementary strategy for stroke treatment by the World Health Organization as it can signicantly improve patients' quality of life. However, the central nervous system (CNS) mechanism of acupuncture treatment of stroke is unclear. The aim of this study is explore the effective pathway and action mechanism of acupuncture treatment for stroke on the CNS. Methods: The following databases will be searched by electronic methods: PubMed; Medline; Embase; Cochrane Library; Chinese National Knowledge Infrastructure; VIP Database; Wan-fang Data; Chinese Biomedical Database. All of them will be retrieved from the establishment date of the electronic database to December 2020, all included studies will be evaluated risk of bias by the Cochrane Handbook. Spatial coordinates of the Montreal Neurological Institute of activated brain regions will be the primary outcome. The systematic review will be conducted with the use of SDM v5.141 software for voxel meta-analysis in this study. Results: This study will obtain the correlation between the activated brain regions of acupuncture treatment for stroke. Conclusion: This study will explore the effective pathway and action mechanism of acupuncture, and provide a reliable scientic basis for the treatment of stroke by acupuncture.


Introduction
Stroke is de ned as an acute focal injury of the central nervous system (CNS) arising from a vascular cause such as cerebral infarction, or intracerebral hemorrhage, or subarachnoid hemorrhage [1] . It is one of the most common causes of death and is the main cause of persistent and acquired disability in adults worldwide. The World Health Organization (WHO) refers to stroke as the incoming epidemic of the 21st century, and it is expected to increasingly affect younger patients. In Europe, stroke cause over 1 million deaths, and worldwide, stroke cause 6.5 million deaths [2] . According to a report published in The Lancet, stroke has become the leading cause of death in China since 2017, and stroke burden in China has increased over the past 30 years, especially remains particularly high in rural areas [3,4] . However, it has been estimated that over 90% of the burden of stroke is attributable to modi able factors [5] . Among these, interventions targeting modi able factors have shown their signi cant impact in reducing the incidence and mortality of stroke [6] .
Acupuncture is one of the oldest and most studied techniques in Chinese medicine, the mode of operation is insertion of a ne needle into the skin or deeper tissues at acupoints of the body, which is recommended as an alternative and complementary strategy for stroke treatment by the WHO [7,8] . This needling involving manipulated manually, electrically, or by heat. In the CNS, endogenous opioids are the principal biological mediators of the therapeutic actions of this ancient technique [7] . Clinical trials have shown that acupuncture can enhance balance [9] , reduce spasticity [10] , and increase muscle strength [11] . A systematic review and meta-analysis of randomized trials conducted in 2010 has indicated that acupuncture may be effective in improving post-stroke impairment, as measured by its motor rehabilitation, increased perfusion within peri-infarcts, and the stimulation of neuronal reorganization, amongst other ndings [12] . However, the poor study quality and publication bias hinder the strength of this recommendation.
Functional magnetic resonance image (fMRI) is nonpolluting, noninvasive, and has high temporal and spatial resolution, neuronal activity was measured by monitoring the hemodynamic response. It can objectively and visually evaluate different cerebral functional areas activated by acupuncture under physiological and pathological conditions [13] . It has become a reliable technique to reveal the CNS mechanism of acupuncture because of high repeatability and reliability [13,14] .
In recent years, researchers have study the CNS mechanism of acupuncture in the treatment of stroke by fMRI, and show that acupuncture can activate several relevant regions in the brain. However, the activation regions found in various studies are not completely consistent, and the exact mechanisms underlying the bene cial effects of acupuncture in the treatment of stroke remain unclear. At present, there is no systematic review to conduct statistical analysis on the studies in this area. Therefore, the aim of this study is collected clinical research with cerebral function image as primary outcome, comparison the in uence of acupuncture treatment for stroke on the CNS, in order to understand the role of acupuncture in the central level, and explore the effective pathway and action mechanism of acupuncture treatment for stroke.

Study registration
This protocol has been registered on the International Prospective Register of Systematic Reviews in PROSPERO (registration number is CRD42021231329), and was performed in accordance with the preferred reporting items for systematic reviews and meta-analysis protocol (PRISMA-P) [15] .

Inclusion criteria
2.2.1 Study type. All Clinical randomized controlled trials (RCTs), semi-randomized controlled trials, observational studies will be included, without restrictions on country but language will be limited on English and Chinese.

Participants
. This study will include participants with stroke. Regardless of gender, age, occupation, education, etiology, and severity, etc.
2.2.3 Type of intervention. Studies using acupuncture in experimental groups will be included,acupuncture may be manipulated manually, electrically or by heat.

Type of outcome measures.
2.2.4.1 Primary outcome. Spatial coordinates of the Montreal Neurological Institute (MNI) [16] of activated brain regions will be the primary outcome.
2.2.4.2 Secondary outcome. Secondary outcomes mainly include the following aspects: effective rate, associated symptoms, quality of life, adverse events incidence.

Exclusion criteria
The exclusion criteria contain the following items: ·Participants with the unclear diagnosis; ·The intervention combined with any complementary therapy such as Chinese herb; ·reviews, animal experiments, theory discussion, case reports, conference articles, and other non-RCTs study; ·Incomplete data or information; ·Repeatedly checked or published literature.

Search strategy
We plan to search the following databases: PubMed; Medline; Embase; Cochrane Library; Chinese National Knowledge Infrastructure (CNKI); VIP Database; Wan-fang Data; Chinese Biomedical Database (CBM). All of them will be retrieved from the establishment date of the electronic database to December 2020. The retrieval mode used will be a combination of medical subject words and free words, including "strokes", "apoplexy", "cerebral infarction", "cerebral stroke", "cerebrovascular accident", "brain vascular accident", "acupuncture", "needle", "electroacupuncture", "scalp acupuncture", "manual acupuncture", "warm needle", "temperature needle", "magnetic resonance imaging", "magnetic resonance image", "NMR imaging", "zeugmatography", "MRI scans", "functional MRI", "fMRI", "spin echo imaging", "randomized controlled trial", "random allocation", "semi-randomized controlled trials", "allocation, random", "RCT randomized controlled", "randomized, controlled", "clinical trial". The search strategy takes PubMed as an example, and shown in Table 1. We plan to conduct a systematic review between March 30, 2021 and December 30, 2022. The selection of study, which includes literature screening, data extraction, management and examination, will be conducted by 2 reviewers. They will independently screen the titles, abstracts, and keywords of all retrieved studies and determine which trials meet the inclusion criteria, full texts of all possible relevant studies will be obtained for further evaluation. Any disagreements will be resolved by discussion between the 2 reviewers and the third author. The details of the selection process are shown in Fig. 1.

Data extraction and management
The two data managers independently extracted the relevant data included in the study according to a uni ed data extraction table. The data extraction table mainly includes: (1) the basic information of the study (name of the researcher, type of study, year of publication, country, language, publication status); (2) study characteristics (sample size, source of cases, age, course of disease, diagnostic criteria, inclusion and exclusion criteria); (3) intervention and control measures (acupoints, manipulation); (4) research methodology (generation of random scheme, allocation concealment, blind method, baseline comparability, loss of follow-up, intentionality analysis); (5) measurement data of outcome indicators.
Any divergences will be discussed and resolved between the data managers, and further differences will be arbitrated by the third manager.

Assessment of risk of bias in included studies
All included studies will be evaluated risk of bias by the Cochrane Handbook, the judgment of risks will be evaluated and described from the following 6 items, including generation of random sequences, allocation concealment, blinding, incomplete outcome data, selective reporting, and other biases. Risk will be divided into 3 levels: "low risk of bias", "high risk of bias" and "unclear risk of bias".

Dealing with missing data
If trial data is insu cient or missing, the corresponding author will be contacted by telephone or email. If missing data is not be available or the author cannot be contacted, these studies will be excluded. We will conduct a limited analysis based on available data and discuss the potential impact of missing data.

Statistical analysis
The systematic review will be conducted with the use of SDM v5.141 (Seed-based D Mapping) [17,18] software for voxel meta-analysis in this study. Data analysis shall be carried out in MNI spatial coordinates. If the research results are not reported in MNI spatial coordinates, Lancaster conversion method shall be used to convert them into MNI space coordinates. The statistical analysis threshold of this voxel meta-analysis was P < 0.005, Z > 1, clustering degree ≥ 10; this corresponds to a corrected P < 0.05.

Heterogeneity assessment
We will use I 2 to evaluate statistical heterogeneity between trials. If the I 2 < 50%, there was little or no heterogeneity in the evidence for the combined outcome, while heterogeneity was evaluated as signi cant (I 2 ≥ 50%) and the trials included were su cient, we will conduct a subgroup analysis to explore the potential source of the heterogeneity.

Subgroup analysis
If there are a large number of subgroup studies, a subgroup analysis is performed to determine heterogeneity. It may be includes the method of intervention, the choice of acupoints, the improvement of concomitant symptoms, the duration and severity of the disease, etc.

Sensitivity analysis
We will conduct a sensitivity analysis to verify the robustness of the preliminary results by a revaluation of risk of bias, methodological quality, missing data or other possible factors. If there is a large difference, sensitivity analysis will be employed careful interpretation.

Publication bias
If more than 10 studies are included in the meta-analysis, we will evaluate the publication bias by EGGER regression test, and the evaluation will be presented in the form of funnel plots. 2.7 Grading the quality of evidence We will evaluate the quality of evidence according to the Grading of Recommendations Assessment, Development and Evaluation (GRADE) [19] . It is divided into four grades: high, moderate, low, very low.

Ethics and dissemination
Ethical approval is not necessary because our study is not linked to individual patient data. The study ndings will be published in a peer reviewed journal or conference presentations to explore the effective pathway and action mechanism of acupuncture treatment for stroke on the CNS.

Discussion
Stroke is a major chronic non-infectious disease that seriously endangers health. It has ve characteristics: high morbidity, high disability rate, high mortality rate, high recurrence rate and high economic burden [20] . With the acceleration of social aging and urbanization, the popularity of unhealthy lifestyles of residents, and risk factors of cerebrovascular diseases are extensive exposure, the stroke burden has an explosive growth trend characterizing by rapid growth of low-income groups, obvious gender and regional differences, and younger trend [21] . At present, the standardized incidence of the rst stroke among residents with 40 to 74 years old in China has increased by an annual rate of 8.3% [20] .
Acupuncture, as an alternative and complementary strategy for stroke treatment recommended by the WHO [8] , has been promoted and applied as it can signi cantly improve the life quality of patients. This could not only provide effective alternative therapies for patients, but also reduce the burden on public health.
A basic study indicated that the mechanism of acupuncture in the treatment of ischemic cerebrovascular diseases may be related to neurovascular units including blood-brain barrier, astrocytes, microglia cells, neurons [22] . Another clinical study showed that acupuncture can promote the remodeling of white matter function after stroke and thus improve motor dysfunction [23] . Despite this, there is still a lack of highquality research evidence on acupuncture treatment of stroke, and the CNS mechanism of acupuncture treatment of stroke is not clear. Therefore, this study conducted a systematic review and meta-analysis of cerebral functional imaging in acupuncture treatment for stroke based on the existing literature, and will obtain the correlation between the activated brain regions of acupuncture, so as to explore the effective