Department of Gastrointestinal Surgery, Sichuan Academy of Medical Sciences and Sichuan Provincial People's Hospital, University of Electronic Science and Technology of China, Chengdu, China.
Department of Obstetrics and Gynecology, Sichuan Academy of Medical Sciences & Sichuan Provincial People's Hospital, University of Electronic Science and Technology of China, Chengdu, China.
Adv Nutr. 2023 May;14(3):451-464. doi: 10.1016/j.advnut.2023.02.004. Epub 2023 Feb 26.
Dietary factors may be associated with the occurrence of pancreatic cancer. This umbrella review aimed to review and grade the evidence for the associations between dietary factors and pancreatic cancer risk. We searched PubMed, EMBASE, Web of Science, Scopus, Cochrane Database of Systematic Reviews, and CINAHL for eligible literature. We included meta-analyses of randomized controlled trials (RCTs) or prospective observational studies. We used AMSTAR-2, a measurement tool to assess systematic reviews, to evaluate the methodological quality of the included meta-analyses. For each association, we calculated the summary effect size, 95% CI, heterogeneity, number of cases, 95% prediction interval, small-study effect, and excess significance bias. The protocol for this review was registered in the PROSPERO database (CRD42022333669). We included 41 meta-analyses of prospective observational studies describing 59 associations between dietary factors and pancreatic cancer risk. None of the retrieved meta-analyses included RCTs. No association was supported by convincing or highly suggestive evidence; however, there was suggestive evidence of a positive association between fructose intake and pancreatic cancer risk. There was weak evidence for an inverse association of nuts intake or adherence to the Mediterranean diet with pancreatic cancer incidence, and for positive associations between a higher intake of red meat or heavy alcohol intake and pancreatic cancer incidence. The remaining 54 associations were nonsignificant. Consistent with the American Institute for Cancer Research review, this umbrella review found that regular consumption of nuts and reduced intake of fructose, red meat, and alcohol were associated with a lower risk of pancreatic cancer. Emerging weak evidence supported an inverse association between adherence to the Mediterranean diet and pancreatic cancer risk. As some associations were rated as weak and most were considered nonsignificant, further prospective studies are needed to investigate the role of dietary factors and risk of pancreatic cancer.
饮食因素可能与胰腺癌的发生有关。本 umbrella 综述旨在回顾和评估饮食因素与胰腺癌风险之间关联的证据，并对其进行分级。我们检索了 PubMed、EMBASE、Web of Science、Scopus、Cochrane 系统评价数据库和 CINAHL 以获取符合条件的文献。我们纳入了随机对照试验 (RCT) 或前瞻性观察性研究的荟萃分析。我们使用 AMSTAR-2 评估系统综述的方法学质量，这是一种评估系统综述的工具。对于每种关联，我们计算了汇总效应大小、95%CI、异质性、病例数、95%预测区间、小样本效应和过度显著性偏倚。本综述的方案已在 PROSPERO 数据库中注册（CRD42022333669）。我们纳入了 41 项前瞻性观察性研究的荟萃分析，这些研究描述了 59 种饮食因素与胰腺癌风险之间的关联。检索到的荟萃分析均未纳入 RCT。没有一种关联得到令人信服或高度提示性证据的支持；然而，有提示性证据表明果糖摄入与胰腺癌风险呈正相关。坚果摄入或遵循地中海饮食与胰腺癌发病率呈负相关，以及红肉摄入较高或大量饮酒与胰腺癌发病率呈正相关，这些关联有较弱的证据支持。其余 54 种关联无统计学意义。与美国癌症研究所的综述一致，本 umbrella 综述发现，经常食用坚果和减少果糖、红肉和酒精的摄入与降低胰腺癌风险有关。新出现的弱证据支持地中海饮食的依从性与胰腺癌风险呈负相关。由于一些关联被评为弱，且大多数被认为无统计学意义，因此需要进一步的前瞻性研究来调查饮食因素与胰腺癌风险之间的关系。
