Abstract: Background Post-traumatic brain injury consciousness disorders are associated with a poor prognosis and a high risk of secondary complications. There is a clinical absence of specific intervention methods. Non-invasive neuromodulation techniques have emerged as a research focus due to their high safety profile. However, existing evidence is lacking in systematic comparisons of different stimulation modalities and target sites, and the therapeutic efficacy of these modalities remains insufficiently demonstrated. There is an urgent need to identify optimal intervention strategies.Objective To evaluate and compare the awakening efficacy and safety of different non-invasive neuromodulation modalities in patients with disorders of consciousness after traumatic brain injury, providing evidence-based guidance for selecting the optimal intervention in clinical practice. Methods Randomized controlled trials on the effects of non-invasive neuromodulation on disorders of consciousness after traumatic brain injury were searched in PubMed, Embase, Web of Science, Cochrane Library, CNKI, CBM, and Wanfang Data from the inception of each database to July 2025. The control group received conventional treatment or sham stimulation, while the experimental group was given non-invasive neuromodulation in addition to the control group's treatment. The Cochrane Risk of Bias Tool will be used for quality assessment. Network meta-analysis will be performed using Stata 16.0, and the GRADE approach will be applied to evaluate the quality of evidence for outcome measures. Results A total of 49 randomized controlled trials were included, among which 12 were rated as low risk of bias and 37 as some concerns; no study was judged to be at high risk. The pooled sample comprised 3 140 participants and encompassed 12 distinct non-invasive neuromodulation modalities. Network meta-analysis showed that, compared with conventional treatment: for improvement in CRS-R scores, the most effective intervention was transcranial direct current stimulation- dorsolateralprefrontal cortex (SMD=3.01, 95%CI=2.21 to 3.80, P<0.05; SUCRA=90.3%) followed by high-frequency repetitive transcranial magnetic stimulation- dorsolateralprefrontal cortex (SMD=2.77, 95%CI=1.97 to 3.56, P<0.05; SUCRA=83.5%) > high-frequency repetitive transcranial magnetic stimulation-primary motor cortex (SMD=2.44, 95%CI=1.73 to 3.14, P<0.05; SUCRA=70.6%) > transcutaneous auricular vagus nerve stimulation (SMD=2.38, 95%CI=1.51 to 3.25, P<0.05; SUCRA=69.9%) > median nerve electrical stimulation (SMD=2.31, 95%CI=1.13 to 3.49, P<0.05; SUCRA=67.7%); For improvement in GCS scores, the most effective intervention was high-frequency repetitive transcranial magnetic stimulation-primary motor cortex (SMD=2.94, 95%CI=2.21 to 3.67, P<0.05; SUCRA=94.1%), followed by transcutaneous auricular vagus nerve stimulation (SMD=2.46, 95%CI=1.73 to 3.19, P<0.05; SUCRA=78.8%) > high-frequency repetitive transcranial magnetic stimulation- dorsolateralprefrontal cortex (SMD=2.27, 95%CI=1.58 to 2.96, P<0.05; SUCRA=69.9%) > median nerve electrical stimulation (SMD=2.12, 95%CI=1.68 to 2.55, P<0.05; SUCRA=62.7%) > transcranial direct current stimulation- dorsolateralprefrontal cortex (SMD=1.51, 95%CI=0.72 to 2.30, P<0.05; SUCRA=40.1%). Conclusion The existing moderate evidence indicates that various types of non-invasive neuromodulation can improve consciousness after brain injury. For enhancing CRS-R scores, transcranial direct current stimulation- dorsolateralprefrontal cortex is the most effective modality, whereas for improving GCS scores, high-frequency repetitive transcranial magnetic stimulation-primary motor cortex is optimal.

Key words: Consciousness disorders, Brain injury, Non-invasive neuromodulation, Network meta-analysis

摘要： 背景 颅脑损伤后意识障碍预后欠佳，易继发并发症，临床缺乏特效干预手段。非侵入性神经调控技术因安全性高成为研究热点，但现有证据缺乏不同刺激方式及靶点的系统比较，疗效证据不充分，亟需明确最优干预策略。目的 通过网状Meta分析系统评价并比较不同非侵入性神经调控方式对颅脑损伤后意识障碍的促醒疗效及安全性，为临床选择最优干预方式提供循证依据。方法 检索PubMed、EMbase、Web of Science、Cochrane Library、中国知网、中国生物医学文献数据库（CBM）和万方数据知识服务平台中关于非侵入性神经调控方式对颅脑损伤后意识障碍影响的随机对照试验，检索时限为建库至2025年7月。对照组采用常规治疗或假刺激，试验组在对照组的基础上加用非侵入性神经调控方式。采用采用Cochrane偏倚评估工具进行质量评价，采用Stata 16.0软件进行网状Meta分析，采用GRADE对结局指标进行证据等级评价。结果 共纳入49项随机对照试验，其中12项研究被评为低风险，37项研究被评为有一定风险，无高风险研究，共3 140例受试者，涉及12种不同的无创神经调控方式。网状Meta分析结果显示：与常规治疗相比，在改善昏迷恢复量表修订版（CRS-R）评分方面，经颅直流电刺激背外侧前额叶皮质（SMD=3.01，95%CI=2.21~3.80，P<0.05；SUCRA=90.3%）疗效最佳，而后依次为高频重复经颅磁刺激背外侧前额叶皮质（SMD=2.77，95%CI=1.97~3.56，P<0.05；SUCRA=83.5%）> 高频重复经颅磁刺激第一运动皮质（SMD=2.44，95%CI=1.73~3.14，P<0.05；SUCRA=70.6%）> 经皮耳迷走神经电刺激（SMD=2.38，95%CI=1.51~3.25，P<0.05；SUCRA=69.9%）> 正中神经电刺激（SMD=2.31，95%CI=1.13~3.49，P<0.05；SUCRA=67.7%）；在改善格拉斯哥昏迷量表（GCS）评分方面，与常规治疗相比，高频重复经颅磁刺激第一运动皮质（SMD=2.94，95%CI=2.21~3.67，P<0.05；SUCRA=94.1%）疗效最佳，而后依次为经皮耳迷走神经电刺激（SMD=2.46，95%CI=1.73~3.19，P<0.05；SUCRA=78.8%）> 高频重复经颅磁刺激背外侧前额叶皮质（SMD=2.27，95%CI=1.58~2.96，P<0.05；SUCRA=69.9%）> 正中神经电刺激（SMD=2.12，95%CI=1.68~2.55，P<0.05；SUCRA=62.7%）> 经颅直流电刺激背外侧前额叶皮质（SMD=1.51，95%CI=0.72~2.30，P<0.05；SUCRA=40.1%）。结论 现有的中等证据表明，不同类型的非侵入性神经调控方式均能改善脑损伤后意识障碍，在改善 CRS-R 评分方面，经颅直流电刺激背外侧前额叶皮质疗效最佳；在改善 GCS 评分方面，高频重复经颅磁刺激第一运动皮质是最佳的刺激方式。

关键词: 意识障碍, 颅脑损伤, 非侵入性神经调控方式, 网状 Meta 分析