针刺镇痛作用的meta流行病学研究和临床证据评价

发布时间：2018-09-01 08:51

【摘要】：针刺作为一项拥有2000多年历史的治疗手段,近几十年受到了越来越多的关注。针刺的治疗领域广泛,其中其最主要和广受关注的一项作用就是镇痛。虽然目前对于针刺镇痛的机制还未得出科学结论,针刺镇痛的临床疗效已经在世界范围内得到认可。近年来针刺镇痛的临床研究数量不断增长,也暴露出一些研究设计和实施过程中的问题,这些问题在以“针刺治疗疼痛”为主题的系统评价和meta分析中进一步放大,在一定程度上影响了对针刺真实效应的客观评价以及针刺镇痛高质量临床证据的产生。系统评价和meta分析被认为是评价一种疗法是否有效的最高级别证据。为了更好地评估针刺治疗疼痛的临床证据,十分有必要对目前针刺治疗疼痛的系统评价和meta分析的研究状况作系统的了解和分析,即这些系统评价和meta分析是否严谨地评价了针刺治疗疼痛的随机对照试验(randomized controlled trials,RCTs),是否存在影响meta分析结果和结论的偏倚因素、以及这些系统评价和meta分析所提供的临床证据的等级如何。本课题着眼以上三个问题开展研究,以期对目前针刺治疗疼痛的临床证据情况进行全面深入的评价和分析。本课题分为三个部分。第一部分系统检索和分析了针刺治疗疼痛的系统评价中偏倚风险评价的情况,包括是否进行了偏倚风险评价、所使用的评价工具、评价的结果是否被合并进系统评价的分析中、是否影响了系统评价的结论等,目的是对目前针刺镇痛系统评价中偏倚风险评价的情况进行综合评估。第二部分使用meta流行病学的研究方法对于符合纳入标准的针刺治疗疼痛的meta分析,以第一作者国家来源、单中心或多中心、样本量、研究的偏倚风险等方面作为研究特征对所纳入的研究进行分类,分别对连续性结局指标和非连续性结局指标进行分析,目的是筛选可能影响针刺镇痛meta分析效应量的特征因素。第三部分使用GRADE临床证据评级系统对针刺治疗疼痛的系统评价中所提供的临床证据进行临床证据质量分级,目的是筛选针刺治疗疼痛的高质量临床证据以及分析低质量和极低质量临床证据的相关因素。本课题第一部分共纳入91篇系统评价,其中有85篇进行了偏倚风险评估,超过一半(n=59,64.8%)使用了标准工具如Jadad评分、Cochrane偏倚风险评价工具等对所纳入的研究进行了偏倚风险评价。在所有进行了偏倚风险评价的系统评价中,超过三分之一(n=29,34.1%)进行了分域评价,超过一半(n=48,56.5%)将偏倚风险评价的结果并入了其数据分析中。虽然大部分针刺治疗疼痛的系统评价进行了偏倚风险评价,近一半却未将这一结果合并至其数据分析中,也就是偏倚风险评价的过程并未对整个系统评价的结论产生任何影响。此外,这些系统评价所使用的偏倚风险评价工具以基于“研究质量”的量表为主,其更看重“报告”而非“实施”,而报告的不足并不能够反映一个研究真实的方法学质量或者作者实施该研究的真实情况。本课题第二部分总计纳入meta分析31篇,共计包括170个临床试验,受试患者总数19952人。分析国家来源对结局变量的影响,来自发展中国家和来自发达国家的试验的效应量之间的差别没有统计学意义,而来自亚洲地区的试验较其他地区更易报告针刺有益的治疗效果;分析单中心或多中心试验对结局变量的影响,单中心试验较多中心试验更易得出针刺镇痛有效的结论;分析样本量对结局变量的影响,每组样本量小于100较每组样本量大于100的试验更易得出针刺干预有效的结论;分析偏倚对结局变量的影响,高偏倚风险或未知偏倚风险的RCT较低偏倚风险的RCT更易得出针刺镇痛治疗有效的结论。本研究的结果与之前发表的meta流行病学研究的结果基本一致。值得注意的是,大部分有意义的结果都出现在对连续型变量的分析中,二分类变量的分析所得到的有统计学意义的结果十分有限,这与之前发表的很多篇使用二分类变量进行meta流行病学分析的研究结果不相一致。本课题第三部分共纳入meta分析23篇。共筛选出44条临床证据,其中高质量证据16条(36.3%),中质量证据11条(25.0%),低质量证据8条(18.1%),极低质量证据9条(20.6%)。在高质量证据中,有11条对应“疼痛程度”这一结局指标,4条对应“反应率”这一结局指标。对于低质量证据和极低质量证据,影响其证据评级的因素主要为高偏倚风险或不确定的偏倚风险、不一致性、发表偏倚。本课题通过以上三个部分的研究,主要得出如下几点结论:1.针刺治疗疼痛的系统评价和meta分析中的偏倚风险评价的实施情况不容乐观,如果系统评价未能很好地实施偏倚风险评价或偏倚风险评价的结果未能对系统评价产生任何影响,则系统评价本身的可信度大大降低;2.地域、试验设计、样本量、偏倚风险等因素在一定程度上影响着针刺治疗疼痛系统评价中对于针刺效应量的估计;3.虽然针刺治疗疼痛的RCT在研究的设计和实施方面有很多亟待解决的问题,但是针刺治疗疼痛也已经存在相当数量的高质量的临床研究证据。本课题对未来针刺临床研究以及针刺相关系统评价和meta分析提出以下建议:1.对于系统评价和meta分析,需要严格按照Cochrane系统评价员手册进行偏倚风险评价。最大限度地检索纳入研究的研究方案,或者尝试联系论文作者,以期对纳入的研究的真实实施情况有最全面的了解;2.无论是对于RCT还是meta分析,都应审慎地评判其结果以指导临床实践。对于来自亚洲国家的研究、单中心研究、小样本量研究、偏倚风险评价被评为较高的偏倚风险和不确定的偏倚风险的研究,应当留意其对效应量的夸大;3.中医针刺的临床研究设计要根据实际情况进行合理的考虑,不能盲目追求“大样本”、“随机”、“双盲”。只要设计合理、实施严谨,非RCT也可以成为高质量的临床证据。综上所述,本课题全面而系统地评价和分析了针刺治疗疼痛的系统评价的研究现状以及存在的问题,第一次使用meta流行病学的研究方法深入挖掘了这一领域的系统评价和meta分析中可能存在的偏倚因素,并且第一次使用GRADE临床证据评级系统对这一领域的系统评价和meta分析中所提供的临床证据进行了评级。本课题的研究方法和研究结果对于今后更合理的设计和实施针刺治疗疼痛的临床研究及系统评价和meta分析有重要的指导意义。
[Abstract]:Acupuncture, as a treatment method with a history of more than 2000 years, has attracted more and more attention in recent decades. Acupuncture therapy has a wide range of fields, of which the most important and widely concerned role is analgesia. In recent years, the number of clinical studies on acupuncture analgesia has been increasing, and some problems in the design and implementation of the study have been exposed. These problems have been further amplified in the systematic evaluation and meta-analysis on the theme of "acupuncture for pain", which to a certain extent affect the objective evaluation of the true effect of acupuncture and the objective evaluation of acupuncture. Systematic evaluation and meta-analysis are considered to be the highest level of evidence for the effectiveness of a therapy. In order to better assess the clinical evidence of acupuncture for pain, it is necessary to systematically understand and analyze the current status of systematic evaluation and meta-analysis of acupuncture for pain. Whether these systematic evaluations and meta-analyses rigorously evaluated randomized controlled trials (RCTs) of acupuncture for pain, whether there were bias factors affecting the results and conclusions of meta-analysis, and how the clinical evidence provided by these systematic evaluations and meta-analysis ranked. The subject is divided into three parts. The first part systematically searches and analyzes the bias risk assessment in the systematic evaluation of acupuncture pain, including whether the bias risk assessment has been conducted and the assessors used. Whether the results of the evaluation are incorporated into the analysis of the system evaluation and whether the conclusions of the system evaluation are affected is a comprehensive assessment of the bias risk assessment in the current acupuncture analgesia system evaluation. The study was classified according to the national origin of the first author, single-center or multi-center, sample size, and bias risk of the study. Continuous and discontinuous outcome indices were analyzed respectively. The aim was to screen the characteristic factors that might affect the effect of acupuncture analgesia meta-analysis. GRADE Clinical Evidence Rating System was used to classify the quality of clinical evidence provided in the systematic evaluation of acupuncture treatment of pain. The purpose was to screen high-quality clinical evidence of acupuncture treatment of pain and analyze the related factors of low-quality and very low-quality clinical evidence. More than half (n = 59,64.8%) used standard tools such as Jadad score, Cochrane bias risk assessment tools to assess the bias risk of the included studies. More than half (n = 48,56.5%) incorporate the bias risk assessment results into their data analysis. Although most systematic assessments of acupuncture pain have conducted bias risk assessment, nearly half have not incorporated the results into their data analysis, i.e., the bias risk assessment process has not been responsible for the conclusions of the overall system assessment. What's the effect? In addition, the bias risk assessment tools used in these systematic assessments are mainly based on the "research quality" scale, which emphasizes "report" rather than "implementation", and the lack of reports does not reflect the true methodological quality of a study or the authenticity of the author's implementation of the study. There were 31 meta-analyses, including 170 clinical trials and 19 952 subjects. There was no statistically significant difference in the effects of trials from developing and developed countries on outcome variables, but trials from Asia were more likely to report beneficial acupuncture treatments than those from other regions. Analysis of single-center or multi-center test on the impact of outcome variables, single-center test more easily than multi-center test to draw an effective conclusion of acupuncture analgesia; analysis of sample size on the impact of outcome variables, the sample size of each group less than 100 than the sample size of each group greater than 100 test more easily to draw an effective conclusion of acupuncture intervention; analysis of bias on the outcome of the trial; analysis of bias The results of this study are consistent with those of previous meta-epidemiological studies. It is noteworthy that most of the significant results appear in the analysis of continuous variables. In the third part of this topic, 23 meta-analysis papers were included. A total of 44 clinical evidences were screened out, of which 16 (36.3%) were of high quality. 11 (25.0%) were quality evidence, 8 (18.1%) were low quality evidence, and 9 (20.6%) were very low quality evidence. Uncertainty bias risk, inconsistency, publication bias. Through the above three parts of the study, the main conclusions are as follows: 1. Acupuncture pain systematic evaluation and meta-analysis of the implementation of bias risk assessment is not optimistic, if the system assessment can not be well implemented bias risk assessment or bias risk assessment. The results of the evaluation did not have any impact on the system evaluation, and the reliability of the system evaluation itself was greatly reduced. 2. Regional, experimental design, sample size, bias risk and other factors to a certain extent affect the estimation of acupuncture effect in the system evaluation of acupuncture treatment pain. 3. Although the RCT of acupuncture treatment pain in the study design and design. There are many problems to be solved urgently in the aspect of implementation, but there is a considerable amount of high-quality clinical research evidence for acupuncture treatment of pain. The following suggestions are proposed for future acupuncture clinical research, acupuncture-related systematic evaluation and meta-analysis: 1. For systematic evaluation and meta-analysis, Cochrane systematic evaluation should be strictly followed. Bias Risk Assessment in the Members'Manual. Search the included research program to the maximum extent, or try to contact the author of the paper in order to have the most comprehensive understanding of the actual implementation of the included research. 2. Whether for RCT or meta-analysis, the results should be carefully evaluated to guide clinical practice. For Asian countries Study, single-center study, small sample size study, bias risk assessment as a higher bias risk and uncertainty bias risk study, should pay attention to the exaggeration of its effect; 3. Chinese medicine acupuncture clinical research design should be based on the actual situation reasonable consideration, can not blindly pursue "large sample", "random", "double blind" As long as the design is reasonable and the implementation is rigorous, non-RCT can also become high-quality clinical evidence. In summary, this topic comprehensively and systematically evaluates and analyzes the research status and existing problems of the systematic evaluation of acupuncture treatment of pain. For the first time, meta-epidemiological research methods were used to dig out the systematic evaluation in this field. And the possible bias in meta-analysis, and for the first time, GRADE Clinical Evidence Rating System was used to rank the clinical evidence provided in the field of systematic evaluation and meta-analysis. Systematic evaluation and meta analysis have important guiding significance.
【学位授予单位】：第二军医大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：R245

【相似文献】