二甲双胍预防2型糖尿病患者结直肠肿瘤发病的Meta分析

发布时间：2018-03-03 01:06

本文选题：二甲双胍　切入点：结直肠癌　出处：《遵义医学院》2017年硕士论文　论文类型：学位论文

【摘要】：目的:通过全面收集有关二甲双胍与2型糖尿病患者结直肠肿瘤发病关系的原始资料,系统评价二甲双胍的使用与2型糖尿病患者结直肠肿瘤的发病关系,为二甲双胍用于结直肠肿瘤初级预防提供循证医学依据。方法:两名检索员独立检索Pubmed、Embase、Cochrane Library及万方(wanfang)、中国知网(CNKI)、中国生物医学(CBM)、维普中文期刊(VIP)等数据库,检索时间限定为建库至2016年10月。两名评价员严格依据纳入及排除标准独立进行文献筛选并提取相关信息及数据,以纽卡斯尔-渥太华量表(NOS)评分标准对纳入研究进行质量评价,NOS评分7分以上为高质量研究。运用Rev Man 5.3及Stata SE 12.0软件进行统计学处理。通过Q统计量检验及I2统计量检验评估各研究间异质性,若研究间存在异质性则进行敏感性及亚组分析。同时运用漏斗图、Begg秩相关法及Egger线性回归分析法进行发表偏移评估。结果:经筛选最终纳入17篇文献,共包含20个独立研究,依据NOS评分标准3篇文献为中质量研究,14篇文献为高质量研究。Meta分析结果显示,2型糖尿病患者使用二甲双胍组与使用其他降糖药物组相比,结直肠肿瘤发病风险降低了15%,两者间差异具有统计学意义(RR=0.85,95%CI:0.80-0.91,P0.00001);以结直癌与结直腺瘤为研究结局指标分别进行统计学分析,结果显示使用二甲双胍组分别降低了13%结直肠癌、24%结直肠腺瘤的发病风险,且均具有统计学意义(RR=0.87,95%CI:0.81-0.93,P0.0001)、(RR=0.76,95%CI:0.66-0.86,P0.0001)。各研究间异质性检验结果为(Q检验P=0.005;I2=51%),提示纳入研究有异质性存在。利用亚组分析异质性存在原因,亚组分析结果显示以西方人群研究为亚组进行异质性检验结果为(Q检验P=0.03;I2=47%),以东方人群研究为亚组进行异质性检验结果为(Q检验P=0.50;I2=0%),此结果表明东、西方人群各亚组内研究不存在明显异质性;而两组间异质性检验结果(Q检验P=0.005;I2=87.6%)提示存在重度异质性表现,故考虑异质性来源主要是东、西方研究人群间差异所致。进一步采用随机及固有效应模型分别合并统计量进行敏感性分析,其结果均保持稳健,提示本研究结论可信。运用漏斗图、Begg秩相关法及Egger线性回归分析法对纳入研究进行定性及定量的偏倚评估,提示可能存在发表偏倚。结论:当前研究数据支持二甲双胍使用能降低2型糖尿病患者结直肠肿瘤发病的风险,二甲双胍有望成为2型糖尿病患者结直肠肿瘤初级预防的药物。
[Abstract]:Objective: to systematically evaluate the relationship between metformin use and colorectal tumor in type 2 diabetes mellitus by collecting the original data about the relationship between metformin and colorectal tumor in type 2 diabetes mellitus. Methods: two searchers independently searched the databases of Pubmedmor Embase Cochrane Library and Wan Fang Wan Fang, CNKII, Chinese Biomedicine, Weipu Chinese Journal (VIPs), and so on, to provide evidence based medical evidence for the primary prevention of colorectal cancer by metformin. The retrieval time was limited to October 2016. The two evaluators conducted literature screening and extracting relevant information and data strictly according to the inclusion and exclusion criteria. Using the Newcastle Ottawa scale scores (NOSs) to evaluate the quality of the inclusion study and the NOS score above 7 is a high quality study. The statistical processing is carried out by using Rev Man 5.3 and Stata SE 12.0 software. Through the Q statistical test and I2, we use the software of Rev Man 5.3 and Stata SE 12.0. Statistical tests assess heterogeneity among studies, Sensitivity and subgroup analysis were performed if there was heterogeneity among the studies. Meanwhile, Begg rank correlation of funnel map and Egger linear regression analysis were used to evaluate the migration. Results: 17 papers were selected and 20 independent studies were included. The results of Meta-analysis showed that patients with type 2 diabetes were treated with metformin as compared with those with other hypoglycemic drugs. The risk of colorectal cancer was reduced by 15%, and the difference between them was statistically significant. The results showed that metformin group reduced the risk of colorectal adenoma by 24% in 13% colorectal cancer patients. All of them have statistical significance. RRX 0.8795 CI: 0.81-0.93 / P0.0001. The results of heterogeneity test among the various studies are: P 0.005 / I _ 2N _ (51), indicating that heterogeneity exists by subgroup analysis, and the reason of heterogeneity exists by using subgroup analysis, and the results of heterogeneity test among all studies are as follows: (1) there is heterogeneity in the study of RRX (0. 81-0. 93%) and 0. 001% (P 0. 0001), which indicates that heterogeneity exists by subgroup analysis. The results of subgroup analysis showed that the results of heterogeneity test of western population study as subgroup were as follows: Q test, P0. 03, I2: 47, and oriental population, as subgroup. The result of heterogeneity test was Q test, P0. 50 and I2 + 0. 0. The results showed that there was no significant difference between the two groups. There was no significant heterogeneity in the subgroups of the western population, but the results of the heterogeneity test between the two groups showed that there was severe heterogeneity, so the source of the heterogeneity was mainly from the east. The sensitivity analysis using random and intrinsic effect models combined with statistics respectively showed that the results were robust. The results suggest that the results of this study are credible. The qualitative and quantitative bias of the study was evaluated by using the rank correlation method of Begg and Egger linear regression analysis. Conclusion: metformin can reduce the risk of colorectal cancer in patients with type 2 diabetes, and metformin is expected to be the primary prophylaxis of colorectal cancer in patients with type 2 diabetes.
【学位授予单位】：遵义医学院
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R587.1;R735.34

【相似文献】