咖啡与子宫内膜癌发病风险的Meta分析

Coffee consumption and risk of endometrial cancer:A systematic review and Meta-analysis of prospective studies

目的：探讨咖啡摄入与子宫内膜癌发病风险的关系。方法：计算机检索CBM、CNKI、万方数据、PubMed、EMBASE和Cochrane图书馆等数据库,收集2012-10以前发表的关于子宫内膜癌与咖啡关系的符合纳入标准的前瞻性研究,由2位评价员按纳入与排除标准独立筛选文献、提取资料并用卡斯尔-渥太华量表（NOS）评价质量后,用STATA12.0软件进行统计分析,采用Q检验和I2进行异质性检验。以咖啡摄入4杯/d为截点,分别合并计算纳入研究的相对危险度（RR）或风险比值（HR）及其95%可信区间（CI）,并按研究地区行亚组分析。另对咖啡摄入与子宫内膜癌发病风险关系进行剂量-效应Meta分析。结果：共纳入10个队列研究,参与者512 386人,包括4 484例子宫内膜癌患者。异质性检验未发现各研究间存在明显异质性。Meta分析结果显示,与很少饮用或不饮用咖啡的人群相比,饮用咖啡≥4杯/d（RR=0.74,95%CI：0.67~0.83）或〈4杯/d（RR=0.87,95%CI：0.81~0.93）的女性人群罹患子宫内膜癌的风险降低。按研究地区的亚组分析与总体结果一致。剂量-效应关系分析显示,饮用咖啡与子宫内膜癌风险之间存在线性剂量-效应关系（P非线性=0.720 7）,饮用咖啡每增加2杯/d,患子宫内膜癌的风险减少了15%（RR=0.85,95%CI：0.75~0.98）。结论：子宫内膜癌的发病风险随着饮用咖啡量的增加而逐渐降低。

OBJECTIVE：To assess the association between coffee consumption and risk of endometrial cancer.METHODS：We searched references from CBM,CNKI,Wangfang data,PubMed,EMBASE and the Cochrane Library before October 2012 to select articles on the association between coffee consumption and risk of endometrial cancer which met initial criteria.Two evaluators independently reviewed and selected articles based on predetermined selection criteria and assess articles with Newcastle-Ottawa Quality Scale,NOS.Stata 12.0was used to conduct meta-analysis.Q test and I2were used to have heterogeneity test.We set coffee drinking 4cups/day as a cut-point and combined relative risk or hazard ratio and corresponding 95%confidence intervals for coffee drinking.We also performed subgroup analyses by geographic region.Moreover,we conducted a dose-response meta-analysis on the association between coffee consumption and risk of endometrial cancer.RESULTS：A total of 10cohort studies were included in the final analyses,involved 512 386participants and 4 484cases of endometrial cancer.No significant heterogeneity was found between studies.Compared with the non/lowest coffee consumption,coffee drinking≥4cups/day or4cups/day was associated with reduction of endometrial cancer risk.The summary RR of endometrial cancer was 0.74（95%CI：0.67-0.83）for≥4cups/day of coffee consumption and 0.87（95%CI：0.81-0.93）for4cups/day of coffee consumption.Subgroup analyses were consistent with the overall analyses.Dose-response analysis indicated that a dose-response relationship was present between coffee drinking and endometrial cancer risk（Pnon-linearity=0.720 7）and increment 2cups/day of coffee consumption was associated with reduced 15%endometrial cancer risk（RR=0.85,95%CI：0.75-0.98）.CONCLUSION：Our meta-analysis suggests that coffee consumption is associated with reduced endometrial cancer risk.