【作者】 张芷毓（Chang Chihyu）；

【导师】 秦叔逵；

【作者基本信息】 南京中医药大学 ， 中西医结合临床， 2013， 博士

【摘要】 研究背景恶性浆膜腔积液(Malignant serous effusions,MSE)是中晚期恶性肿瘤的常见并发症,干扰呼吸、循环以及消化功能,往往严重地影响抗肿瘤治疗和患者的生存质量,部份患者治疗困难,预后较差,生存期短。有文献报道,恶性积液确诊后,平均存活期为(3.1±0.5)个月,6个月病死率为84%。在临床上有些恶性浆膜腔积液的治疗十分棘手,尽管目前局部治疗的方法较多,如利尿、浆膜腔穿刺放液或引流、腔内注射化疗药物、或生物制剂等疗效常不尽人意、复发率高且有不同程度的毒副反应。榄香烯(Elemene)是从温莪术中提取的具有抗癌活性的倍半萜烯化合物,为现代中药制剂。该药疗效确切,毒副作用轻微,在延长生存期,改善生活质量,提高病人生存受益和抗转移复发方面与化疗比较有明显优势,并具有高效安全的双向性,可以增强疗效,联合用药既可有效增加疗效,又能在一定程度上减轻不良反应。并可用于介入、腔内化疗及癌性胸、腹水、心包积液的治疗。经循征医学系统评价证实安全有效的广谱靶向抗肿瘤植物药[3]。现阶段单药应用榄香烯或联合化疗对恶性胸、腹、心包腔积液治疗均对其临床疗效进行了大量的研究与报道。但由于肿瘤的特殊性,迄今尚未见到多中心、大样本的临床试验结果,亦未见这种增效减毒作用的循证医学证据。根据临床流行病学和循证医学评价文献指导原则,本研究收集1996～2012年间已发表的榄香烯乳注射液(Elemene Emulsion Injection)单药与联合化疗治疗恶性浆膜腔积液的临床随机对照试验(RCT),主要采用系统评价和Meta分析的方法,严格评价研究证据的真实性,进行方法学质量和临床疗效及安全性评价,客观评估研究质量的总体水平及治疗上的优势与不足,并基于GRADE系统作出证据水平分级及推荐等级,以期为临床治疗提供更可靠的循证医学证据。目的系统评价(systematic review, SR)榄香烯乳注射液对恶性浆膜腔积液患者治疗的临床疗效、生活质量、毒副反应的影响,并严格评价研究证据的真实性、可靠性及临床适用性,为临床实践决策提供可靠的科学证据。方法1.纳入研究文献1.1.系统检索中文科技期刊数据库(VIP),和中国期刊全文数据库(CNKI)、万方数据知识平台和文献追溯的方法,收集国内1996～2012年间公开发表的关于榄香烯乳注射液治疗恶性浆膜腔积液的临床研究文献。1.2制定检索策略：采用主题词、关键词相结合的检索方法进行检索：#1榄香烯、#2治疗、#3疗效、#4积液、#5胸水、#6腹水、#7:#1AND#4、#8:#1AND#5、#9:#1AND#6。1.3确定纳入和排除标准：纳入文献的研究目标为在榄香烯治疗上或联合化疗药物对恶性浆膜腔积液临床观察其治疗获益。排除重复发表、未设立对照组的研究和动物实验,并排除无法判断疗效综述、经验总结、理论探讨、个案报道等类型的研究,并剔除不符合要求的文献。本研究只纳入随机对照试验(RCT)。1.4干预措施：①榄香烯乳(大连金港制药有限公司制品-国药准字H10960114)vs.化疗药;②榄香烯乳+化疗药vs.化疗药。对照组使用和治疗组相同的常规化疗(药物种类不限)。2组患者采取腔内积液置管闭式引流,然后作腔内注药,同时辅以止吐及其他对症处理的药物,如地塞米松及利多卡因等。2.判效结局指标2.1疗效评定标准：采用世界卫生组织(WHO)制定的标准评价：分为CR、PR、SD、NC4组,总有效率=(CR例数+PR例数)/总例数×100%。2.2生活质量评价指数(quality of life,QOL):采用karnofsky performance status (KPS)评分标准,分为好转、稳定、恶化。2.3安全性评价：按照WHO抗癌药物急性与亚急性毒副反应分级标准。3.文献质量评价3.1严格评价各个相关纳入研究的方法学质量,采用按照Juni等对随机对照试验的4条质量评价标准进行分析评价,质量等级为A级；B级;C级;3.2按照Cochrane系统评价员手册(5.1.0版)中RCT的偏倚风险评价标准评价纳入研究的方法学质量;3.3并采用GRADE系统对证据质量和等级推荐进行分级、循证医学(EBM)证据质量进行评价。4..数据录入及统计分析在严格质量评价后,对收集的各相关研究进行结局测量指标的数据提取,采Cochrane协作网提供的Revman5.0统计软件进行Meta分析。a.异质性检验(齐性检验)。根据异质性结果选用进一步的系统评价方法。b.统计合并效应量(加权合并以相对危险度(Relative Risk, RR)表示),计算效应尺度及95%的置信区间(Confidence Interval,CI)并进行统计推断。c.图示(森林图)单个试验的结果和合并后的结果。以P<0.05表示具有显著的统计学意义,以P<0.01表示具有非常显著的统计学意义。d.敏感性分析：测试各个临床试验的疗效结果经Meta整合之后,所得出总体的关于干预的疗效结果,如果各临床试验结果之间差异很大,异质性测试显示有显著性差异,则排除某纳入研究,重新进行Meta分析,其结果与未排除前的结果进行比较,以检验个别研究结论对合并效应量的影响。d.采用“倒漏斗图”或通过“失安全数”的计算了解潜在的发表偏倚。e.结果解释、作出结论及评价。结果本研究共纳入符合榄香烯乳注射液治疗恶性浆膜腔积液的临床随机对照试验(RCT)31篇中文文献,共纳入1763名患者,其中治疗组903例,对照组860例。治疗组的联合用药或对照组化疗药物以顺铂、金葡素、卡铂、5-FU、IL-2等为主进行治疗。各研究纳入治疗组和对照组的基线条件和基础治疗一致。1.榄香烯乳注射液治疗恶性浆膜腔积液的Meta分析纳入的17篇文献,对榄香烯乳单药治疗与单用化疗治疗进行对比,并能获得近期疗效、生活质量、不良反应及生存数据的随机对照试验。967例入组。其中仅1篇文献能提供1年生存率。纳入文献研究质量均为C级。1.1近期疗效的Meta分析1.1.1对纳入分析的17项研究近期疗效的Meta分析,综合结果显示,榄香烯组治疗恶性浆膜腔积液疗效优于化疗组RR＝1.03,95%CI：(0.93～1.14),P=0.61。但各组之间差异无统计学意义。1.1.2对纳入的17项研究近期疗效的总有效率(CR+PR)的Meta分析,两组有效率为84.31%vs.63.44%,RR=1.32,95%CI(1.22,1.43),P<0.01。差异有统计学意义。揭示榄香烯治疗恶性浆膜腔积液的临床疗效明显优于化疗组。1.2生存质量的Meta分析对纳入4项研究的Karnofsky评分改善率比较,两组生质存量改善率为69.52%vs.35.23%,RR=1.97,95%CI(1.48,2.64),P<0.01,2组差异有统计学意义,说明榄香烯对恶性浆膜腔积液患者生活质量的改善率明显高于化疗组。1.3生存率的Meta分析对纳入1项研究1年生存率的Meta分析,2组1年生存率分别60.00%,31.03%,RR=1.93,95%CI(1.04,3.58),P<0.05。差异有统计学意义。榄香烯合并化疗组的1年生存率高于化疗组。1.4不良反应的Meta分析17项研究分为6个组的不良反应发生率的Meta分析,主要为胸痛、腹痛、发热、胃肠道反应、骨髓抑制、肝臀功能损害等。1.4.1对纳入14项研究胸痛发生率的Meta分析,两组胸痛发生率为29.41%vs.8.97%,RR=2.62,95%CI(1.28,5.37),P<0.01。差异有统计学意义。显示注射榄香烯引起胸痛发生率明显高于化疗药物腔内注射治疗。1.4.2对纳入2项研究腹痛发生率的Meta分析,两组发生率为8.45%vs.2.73%,RR=3.15,95%CI(0.69,14.42),P=0.14。显示注射榄香烯组与化疗组,两组间腹痛发生率相似,差异无统计学意义。1.4.3对纳入10项研究发热发生率的Meta分析,两组发生率为21.15%vs.4.64%,RR=3.32,95%CI(1.24,8.84),P<0.05。注射榄香烯治疗恶性浆膜腔积液发热发生率高于化疗药物腔内注射治疗,差异有统计学意义。1.4.4对纳入15项胃肠道不良反应发生率的Meta分析,两组发生率为6.26%vs.40.52%,RR=0.20,95%CI(0.10,0.37),P<0.01。差异有统计学意义。显示榄香烯组胃肠道反应发生率明显低于化疗组。1.4.5对纳入14项研究骨髓抑制反应率的Meta分析,两组发生率为0.24%vs.20.66%,RR=0.08,95%CI(0.04,0.17),P<0.01。显示榄香烯组骨髓抑制发生率明显低于化疗组,差异有统计学意义。1.4.6对纳入9项研究肝、肾功能损害发生率的Meta分析,两组发生率为0.33%vs.8.44%,RR=0.17,95%CI(0.07,0.43),P<0.01。差异有统计学意义。显示榄香烯组肝、肾功能损害低于化疗组。2.榄香烯乳注射液联合化疗药治疗恶性浆膜腔积液的Meta分析纳入的14篇文献,是对化疗同时加用榄香烯治疗与单用化疗治疗进行对比,并能获得近期疗效、生活质量、不良反应及生存数据的随机对照试验。796例入组。其中无1篇文献能提供生存率数据。纳入文献研究质量：仅1篇为B级,其馀皆为C级。2.1近期疗效的Meta分析2.1.1对纳入分析的14项研究近期疗效的Meta分析,综合结果显示,榄香烯与化疗药物联合组和单纯化疗组治疗恶性浆膜腔积液疗效优于单纯化疗组[RR=0.98,95%CI：(0.79-1.22),P=0.86]。但各组间差异无统计学意义。2.1.2对纳入的14项研究近期疗效的总有效率(CR+PR)的Meta分析,两组有效率为82.52%vs.58.85%,RR=1.41,95%CI(1.19,1.68),P<0.01。差异有统计学意义。揭示榄香烯可以明显提高化疗药物对恶性浆膜腔积液治疗的临床疗效。2.2生存质量的Meta分析对纳入4项研究的Kamofsky评分改善率比较,两组生质存量改善率为65.80vs.33.33%,RR=2.06,95%CI(1.40,3.04),P<0.01。说明榄香烯联合化疗对恶性浆膜腔积液患者生活质量的改善率明显高于单纯化疗组。2.3不良反应的Meta分析14项研究分为5个组的不良反应发生率的Meta分析,主要为胸痛、发热、胃肠道反应、骨髓抑制,以及肝肾功能损害等。2.3.1对纳入7项研究胸痛发生率的Meta分析,两组胸痛发生率为28.70%vs.19.41%,RR=3.15,95%CI(0.81,12.24),P=0.10。差异无统计学意义。显示注射榄香烯与化疗药物联合组与单纯化疗组引起胸痛发生率无明显差异。2.3.2对纳入5项研究发热发生率的Meta分析,两组发生率为39.39%vs.17.28%,RR=3.71,95%CI(0.80,17.19),P=0.09。差异无统计学意义。显示注射榄香烯与化疗药物联合组与单纯化疗组引起发热发生率无明显差异。2.3.3对纳入8项胃肠道不良反应发生率的Meta(?)分析,两组发生率为35.39%vs.53.36%,RR=0.66,95%CI(0.54,0.81),P<0.01。2组差异有统计学意义。显示注射榄香烯与化疗药物联合组胃肠道不良反应发生明显低于单纯化疗组。2.3.4对纳入4项研究骨髓抑制反应率的Meta分析,两组发生率为13.79%vs.50.00%,RR=0.24,95%CI(0.08,0.76),P<0.05。显示榄香烯与化疗药物联合组骨髓抑制发生率低于单纯化疗组。差异有统计学意义。2.3.5对纳入3项研究肝、肾功能损害发生率的Meta分析,两组发生率为4.65%vs.10.11%,RR=0.49,95%CI(0.17,1.44),P>0.05。差异无统计学意义。显示榄香烯与化疗药物联合组和单纯化疗组肝、肾功能发生无明显差异。结论本系统评价结果表明,榄香烯乳注射液单药及联合化疗治疗恶性浆膜腔积液具有明显的优势。治疗组患者的总有效率(CR+PR)、生活质量(KPS)、生存期明显高于对照组患者。榄香烯乳注射的不良反应主要为低热和局部胸痛,同时对症处理可能减轻或避免不良反应的发生。该药避免了应用顺铂等化疗药物所产生的胃肠道反应,骨髓抑制及肝肾功能损害等,增加了疗效。两者联合治疗比单一应用更能有效地控制浆膜腔积液,可以减轻患者的症状,提高生活质量,延长患者生存期,显示出明显的治疗作用。经本研究Meta分析提示榄香烯乳注射液单药及联合化疗对恶性浆膜腔积液是一种安全、有效的治疗方法,鉴于系统评价和Meta分析属于二次证据,其重要性按级别划分属于一级证据,临床参考价值最大。纳入研究分析的原始文献的异质性明显影响较大,基于纳入研究的文献质量评价为C及GRADE系统的证据质量为2B,可能存在部份偏倚等局限性,尚需进一步开展大规模、高质量的基础和临床研究取得高级别的循证医学证据,指导临床实践。更多还原

【Abstract】 BackgroundMalignant serous effusions are common complications caused by Intermediate and advanced-stage malignant tumor, interfering with their respiratory, circulatory and digestive functions and affecting clinical anti-tumor treatment and the patients’ quality of life. Some of these patients are hard to treat with poor prognosis and short survival time. According to the literature we collected, once the definite diagnosis of MSE is made, the average survival time is3.1±0.5months, and morality rate within6months is84percent. It’s rather difficult to find an appropriate treatment for MSE in the clinic. Although methods of local treatment are quite selective, such as diuresis, piercing, drainage of serous effusions and intracavitary injection of chemo-drugs or biological drugs, the effect is not quite satisfying with high recurrence rate and toxicity and side effects in various degrees.Elemene emulsion is an anticancer activity of extract-Sesquiterpenes, derived from Rhizoma Curcumae. With reliable curative effects and mild side effects, this new drug has its obvious advantage in prolonging life expectancy, improving life quality and survival benefits, decreasing the rate of metastasis and recurrence comparing with a single treatment with chemotherapy drugs. Meanwhile, its bidirectional treating effect not only intensifies treating efficacy, but also decreases side effects to some extent. It also applies to the intervention operation, intracavitary injections of chemo-drugs and treatment of malignant pleural fluids, ascites and pericardial effusions. Besides, after the systematic review of evidence-based medicine, elimene emulsion is confirmed as a broad-spectrum and targeting anticancer botanical extract, which is also safe and effective.At present stage there are plenty of studies and publications on the clinical effects of elimene emulsion in treating malignant pleural, abdominal and pericardial effusions solely or with combination chemotherapy. However, with the particularity of tumor, there’s no result of clinical trials based on multi-center and large sample and its ability to enhance clinical effects and reduce toxicity still remains unconfirmed in an evidence-based manner.This study collected randomized controlled trials (RCT) of elemene injections in treating malignant serous effusions with or without chemotherapy from1996to2012according to the documents evaluation rules set by epidemiology and EBM, mainly adopting methods of systematic reviews and meta-analysis to objectively evaluate the authenticity of research evidence, quality of methodologies, clinical efficacy and safety, overall quality of research and strengths and weakness of treatment. Besides, research evidence was classified, graded and recommended based on GRADE system in order to be more reliable in guiding clinical practice.ObjectiveThis research aims to assess the effects of elimene injection in treating malignant serous effusions from such aspects as clinical efficacy, quality of life, toxicity and side effects, and to critically evaluate the authenticity, reliability and clinical applicability of research evidence for the purpose of providing scientific support for clinical practice.Methodology1. Inclusion of Research Literature1.1To collect clinical research documents about elimene injection in treating malignant serous effusions published during years between1996and2012from databases as VIP Database for Chinese Technical Periodicals (VIP), China National Knowledge Infrastructure (CNKI) and WANFANG data platform.1.2Search StrategyTo search research literature by grouping index words:category1elimene injection, category2treatment, category3effectiveness, category4effusion, category5pleural fluid, category6ascites, category7elimene injection and effusion, category8elimene injection and pleural fluid, category9elimene injection and ascites.1.3Inclusion and Exclusion CriteriaThe objective of this research is to observe the clinical effectiveness of elimene injections in treating malignant serous effusions with or without chemotherapy drugs. Duplicated researches or publications, studies without control groups or animal experiments, and reviews, summaries of clinical experience, theoretical investigations or case reports with indefinite efficacy descriptions were excluded. Only randomized controlled trials (RCT) were included.1.4Interventions①elimene injection (Dalian Jiangang Pharmaceutical Co., Ltd. Product, China’s medical permitment NO.H10960114) versus chemo-drugs②elimene injection plus chemo-drugs versus chemo-drugsStandard chemo-drugs were used both in the control group and the experimental group. Both groups applied drainage of serous effusions and intercavitary injection. According to indications of patients other drugs would be used such as dexamethasone, lidocaine, etc..2.Outcome Indexes for Clinical Effect Assessment2.1Effect Evaluation Standard:the WHO Evaluation Standard has four groups:Complete Response (CR), Partial Response (PR), Stable Disease (SD) and No Change (NC). Overall effectiveness=(CR sample size+PR sample size)/total sample size×100%.2.2Quality of Life Assessment Index ((QOL):karnofsky performance status (KPS) rating scales: bettering, stable and worsening.2.3Safety Evaluation Index:Toxicity of Anticancer Drugs Grading Scale made by WHO3Literature Quality Evaluation Assessment3.1To assess the quality of methodologies that involved in the research and evaluate random controlled trials based on four quality assessment criteria professor Juni proposed with rating grade A, B and C;3.2To assess the quality of methodologies that involved in the research according to Risk of Bias Criteria of RCT in Cochrane Reviewers’Handbook (5.1.0version);3.3To grade evidence and recommendations and evaluate evidence quality of EBM via GRADE system.4. Data Entry and Statistical AnalysisAfter strict quality assessment, meta-analysis of outcome indexes extracted from each studies was carried out by statistical software Revman5.0provided by Cochrane Collaboration.a. Homogeneity testFurther systematic review was needed for results of heterogeneity test;b. To calculate combined effect size (weighted combination is represented by relative risk (RR)), effect magnitude and95%confidence interval (CI) for statistical reference;c. To display results of each test and combined results with forest plots; P<0.05indicates statistical significance, and P<0.01means significant difference.d. Sensitivity AnalysisUsing sensitivity analysis to analyze overall effectiveness after medical interventions based on the integration of effectiveness of every single clinical trial. If the result of each single test was different and the result of each single test via homogeneity test also indicated the significant difference, this specific single test was excluded and meta-analysis was re-applied for the total results after exclusion. Comparing results before and after exclusion to test the influence some single tests could have on combined effect size.e. To analyze potential publication bias via "fail-safe number" or "funnel plots";f. Results interpretation, conclusion and evaluationResultsThere were31literature about clinical RCT studying the effects of elemene injections in treating malignant serous effusions in this research. This sample contained1,763control subjects, 903cases of the experimental group and860cases of the control group. The chemotherapy drugs used in the experimental group and the control group mainly included cisplatinum, staphylococcin, carboplatinum,5-FU and IL-2. The baseline conditions and initial therapy for each test were identical.1. Meta-analysis of the effectiveness of Elemene injections in the treatment of malignant serous effusionsThere were17research documents comparing the treating effects of malignant serous effusions with elemene injections or chemotherapy drugs respectively and providing valuable data of short-term effects, quality of life, adverse reactions and survival time via RCT with967cases included. Only one could provide the rate of one-year survival time. The literature quality was grade C.1.1Meta-analysis of short-term effects1.1.1The meta-analysis of short-term effects listed in17literature showed that the experimental group with elemene injections had better treating effects than the control group with chemotherapy drugs. The relative risk reported as1.03with a95%CI (0.93to1.14), P=0.61. The observed difference is not statistically significant.1.1.2According to meta-analysis of overall effectiveness (CR+PR) of short-term effects, the efficacy of the experimental group versus the efficacy of the control group was84.31%versus63.44%, and the relative risk was1.32(95%CI1.22to1.43), P<0.01. The differences were statistically significant. This result showed that elemene injections had better clinical effects in treating malignant serous effusions than chemotherapy drugs.1.2Meta-analysis of quality of lifeThere were4documents studying the improvement rate via Kamofsky index. The rate of improvement of the experimental group was69.52%, while that of the control group was35.23%, and the relative risk was1.97(95%CI1.48to2.64), P<0.01. The differences were statistically significant. This result showed that the group treated with elemene injections was more effective than the control group from the aspect of improving patients’quality of life.1.3Meta-analysis of survival timeOnly one included document provided valuable data for the meta-analysis of one-year survival time of MSE patients. The rate of survival time of the experimental group was60.00%, while that of the control group was31.03%, and the relative risk was1.93(95%CI1.04to3.58), P<0.05. The differences were statistically significant. The rate of one-year survival time of the experimental group was higher than the control group.1.4Meta-analysis of adverse reactions The adverse reactions mentioned in17included documents could be divided into six categories:chest pain, abdominal pain, fever, gastrointestinal tract side effects, bone marrow suppression and liver and kidney impairment. The method of meta-analysis was applied to each group.1.4.1There were14documents related to the incidence of chest pain. Via meta-analysis, the incidence rate of chest pain in the experimental group was29.41%, while that in the control group was8.97%, and relative risk was2.62(95%CI1.28to5.37), P<0.01. The differences were statistically significant. It was more likely to cause chest pain when treating with elemene injections instead of chemotherapy.1.4.2Two studies about the incidence of abdominal pain were included for meta-analysis. The incidence rate of abdominal pain in the experimental group was8.45%, while that in the control group was2.73%, and relative risk was3.15(95%CI0.69to14.42), P=0.14. According to the data, the incidence rate of abdominal pain of both groups had no significant difference.1.4.3There were10documents providing data about the incidence of fever for meta-analysis, the incidence rate of fever in the experimental group was21.51%, while that in the control group was4.64%, and RR was3.32(95%CI1.24to8.84),P<0.05. The differences were statistically significant. It was more likely to cause fever when treating malignant serous effusions with elemene injections instead of chemotherapy.1.4.4Fifteen documents mentioned the incidence of gastrointestinal tract side effects. After meta-analysis, the incidence rate of gastrointestinal tract side effects in the experimental group was6.26%, while that in the control group was40.52%, and relative risk was0.20(95%CI0.10to0.37), P<0.01. The differences were statistically significant. The result indicated that the incidence rate of gastrointestinal tract side effects was lower in the group treated with elemene injections than in the chemotherapy group.1.4.5Fourteen documents were collected to study the incidence of bone marrow suppression with the method of meta-analysis. The incidence rate of bone marrow suppression in the experimental group was0.24%, while that in the control group was20.66%, and relative risk was0.08(95%CI0.04to0.17),P<0.01. This result indicated that the incidence rate of bone marrow suppression was lower in the experimental group than in the control group. The differences were statistically significant.1.4.6We collected9documents to study the incidence rate of liver and kidney impairment of both treating method. Based on the data of meta-analysis, the incidence rate of liver and kidney impairment in the experimental group was0.33%, while that in the control group was8.44%, and relative risk was0.17(95%CI0.07to0.43), P<0.01. The differences were statistically significant. It was less likely to cause liver and kidney impairment when treating with elemene injections than with the chemotherapy.2. Meta-analysis of the effectiveness of Elemene injections combined with chemotherapy in treating malignant serous effusionsThere were14research documents comparing the treating effects between chemo-drugs only and with elemene injections combined, which provided valuable data to assess short-term effects, quality of life, adverse reactions and survival time via RCT with796cases included. No research could provide the rate of1-year survival time. Only one document’s quality reached B, the rest included in this research were C.2.1Meta-analysis of short-term effects2.1.1The meta-analysis of short-term effects extracted from14included documents showed that better treating effects could be gained when combining elemene injections with chemotherapy in treating malignant serous effusions than the control group with chemotherapy drugs only. The relative risk reported as0.98with a95%CI (0.79to1.22), P=0.86. There was no significant difference among groups.2.1.2According to meta-analysis of overall effectiveness (CR+PR) of short-term effects, the effects of the experimental group versus the effects of the control group was82.52%versus58.85%, and the relative risk was1.41(95%CI1.2219to1.68), P<0.01. The differences were statistically significant. According to the analysis, elemene injections could significantly improve clinical effectiveness of chemotherapy in treating malignant serous effusions.2.2Meta-analysis of quality of lifeThere were4documents comparing the rate of improvement via Kamofsky index. The improvement rate of the experimental group was65.80%, while that of the control group was33.33%, and the relative risk was2.06(95%CI1.40to3.04), P<0.01. The differences were statistically significant. This result showed that the improvement rate of patients’QOL was obviously higher when treating with elemene injections and chemotherapy combined than the control group adopting chemotherapy only.2.3Meta-analysis of adverse reactionsThe adverse reactions mentioned in14included documents could be divided into five categories, chest pain, fever, gastrointestinal tract side effects, bone marrow suppression and liver and kidney impairment. Meta-analysis was done respectively.2.3.1There were7documents related to the incidence of chest pain. Via meta-analysis the incidence rate of chest pain in the experimental group was28.70%, while that in the control group was19.41%, and relative risk was3.15(95%CI0.81to12.24), P=0.10. There was no significant difference in decreasing the incidence rate of chest pain between the experimental group and the control group.2.3.2There were5documents providing data about the incidence of fever for meta-analysis, the incidence rate of fever in the experimental group was39.39%, while that in the control group was17.28%, and relative risk was3.71(95%CI0.80to17.19), P=0.09. There was no significant difference in decreasing the incidence rate of fever when treating with elemene injections and chemotherapy combined or chemotherapy only.2.3.3Eight documents were collected to study the incidence rate of gastrointestinal tract side effects with the method of meta-analysis. The incidence rate of gastrointestinal tract side effects in the experimental group was35.39%, while that in the control group was53.36%, and relative risk was0.66(95%CI0.54to0.81), P<0.01. The differences were statistically significant. This indicated that the incidence rate of gastrointestinal tract side effects was lower when treating with elemene injections and chemotherapy combined than with chemotherapy only.2.3.4There were4documents focusing on the incidence of bone marrow suppression. The incidence rate of bone marrow suppressions in the experimental group was13.79%, while that in the control group was50.00%, and relative risk was0.24(95%CI0.08to0.76), P<0.05. The differences were statistically significant, indicating that the incidence rate of bone marrow suppressions was lower when treating with elemene injections and chemotherapy combined than with chemotherapy only.2.3.5We collected3documents to study the incidence rate of liver and kidney impairment of both treating method. Based on the result of meta-analysis, the incidence rate of liver and kidney impairment in the experimental group was4.65%, while that in the control group was10.11%, and relative risk was0.49(95%CI0.17to1.44), P>0.05. There was no significant difference in decreasing the incidence rate of liver and kidney impairment in both groups.ConclusionThis research suggests that elemene injections have obvious advantages in treating malignant serous effusions with or without chemo-drugs over the control group treating with chemotherapy only. Indexes like overall treating effectiveness (CR+PR), QOL (KPS) and survival time in the experimental group were significantly higher than in the control group. The main side effects of elemene injections are low fever and chest pain, but such side effects could be avoided or reduced when dealing with indications carefully. Elemene injections decreased those adverse reactions as gastrointestinal tract side effects, bone marrow suppression and impairment of liver and kidney when treating with chemotherapy like cisplatinum, at the same time elemene injections improved clinical effectiveness. When elemene injections and chemotherapy were used together, the treating effectiveness could be significant. For example, serous effusions could be controlled more effectively, symptoms of patients could be reduced, quality of life could be improved and their survival time could be lengthened.This research indicates that it’s safe and effective to treat malignant serous effusions with elemene injections with or without chemotherapy. Systematic review and meta-analysis was regarded as the indirect evidence, but it was graded as the first-rate evidence considering its importance and was of great value for clinical practice. Besides, considering the homogeneity of original literature (literature quality was C and EBM evidence quality was2B according to GRADE system), limitations like partial biases might be existed. Therefore, large-scale and high quality research should be further carried out for more reliable EBM evidence to guide clinical practice.更多还原