Supplementary Material for: A Meta-Analysis of the Impact of Obesity, Metabolic Syndrome, Insulin Resistance, and Microbiome on the Diagnosis of Barrett’s Esophagus

<b><i>Background and Aim:</i></b> The etiology and pathogenesis of Barrett’s esophagus (BE) have been widely studied during recent decades. However, the association between BE and possible risk factors, including abdominal obesity (AO), metabolic syndrome (MetS), insulin resistance (IR), and the microbiome has not reached a consensus and lacks a systematic assessment. The purpose of our study is to evaluate, quantify, and summarize the association between these factors and BE risk. <b><i>Methods:</i></b> A systematic search of Pubmed, Embase, and Cochrane Library databases was performed to identify relevant studies before September 2018. Studies were estimated with the OR, the weighted mean difference (WMD), and the 95% CI by using a random effects model. Subgroup analysis and publication bias were also performed. <b><i>Results:</i></b> A total of 46 citations were included in the analysis, and 119,273 subjects were analyzed (AO 13, MetS 15, IR 9, and microbiome: 9). The pooled results showed that AO (<i>p</i> &lt; 0.01, OR 1.30, 95% CI 1.11–1.52, <i>I</i>² = 31.9%), MetS (<i>p</i> &lt; 0.01, OR 1.68, 95% CI 1.40–2.01, <i>I</i>² = 87.6%), and IR (<i>p</i> &lt; 0.01, WMD 0.23, 95% CI 0.11–0.35, <i>I</i>² = 55.8%) were all significantly associated with an increased risk of BE, but except for the microbiome (<i>p</i> &gt; 0.05, OR 1.27, 95% CI 0.66–2.43, <i>I</i>² = 46.7%). In addition, subgroup analyses were stratified by waist-to-hip ratio, waist circumference, body mass index, diagnosis criteria, strain type, geographical region, and study design, respectively. Moreover, we observed no evidence of publication bias in Egger’s and Begg’s tests. <b><i>Conclusions:</i></b> Our study reveals that AO, MetS, and IR are significantly associated with BE risk, except for the microbiome. The mechanism of BE induced by 3 risk factors should be further explored.

**背景与研究目的：** 近数十年来，学界已针对巴雷特食管（Barrett’s esophagus, BE）的病因与发病机制开展了广泛研究，但关于其与腹型肥胖（abdominal obesity, AO）、代谢综合征（metabolic syndrome, MetS）、胰岛素抵抗（insulin resistance, IR）及微生物组等潜在危险因素的关联尚未达成共识，且缺乏系统性评估。本研究旨在评估、量化并总结上述危险因素与巴雷特食管发病风险之间的关联。
**方法：** 系统检索PubMed、Embase及Cochrane图书馆数据库，筛选2018年9月之前发表的相关研究。采用随机效应模型，以比值比（OR）、加权均数差（WMD）及95%置信区间（CI）对研究结果进行合并分析，并开展亚组分析与发表偏倚检验。
**结果：** 最终共纳入46篇文献进行分析，涉及119273名研究对象（其中腹型肥胖相关研究13项、代谢综合征相关研究15项、胰岛素抵抗相关研究9项、微生物组相关研究9项）。合并分析结果显示，腹型肥胖（p < 0.01，OR=1.30，95%CI 1.11~1.52，I²=31.9%）、代谢综合征（p < 0.01，OR=1.68，95%CI 1.40~2.01，I²=87.6%）及胰岛素抵抗（p < 0.01，WMD=0.23，95%CI 0.11~0.35，I²=55.8%）均与巴雷特食管发病风险升高显著相关；而微生物组与巴雷特食管发病风险无显著关联（p > 0.05，OR=1.27，95%CI 0.66~2.43，I²=46.7%）。此外，本研究按腰臀比、腰围、体质量指数、诊断标准、菌株类型、地理区域及研究设计进行了分层亚组分析。同时，Egger检验与Begg检验均未提示存在发表偏倚。
**结论：** 本研究表明，腹型肥胖、代谢综合征及胰岛素抵抗与巴雷特食管发病风险显著相关，而微生物组未发现此类关联。上述三种危险因素诱发巴雷特食管的具体机制仍有待进一步探究。