头颈癌患者的饮食行为与生存:头颈5000研究结果
该研究基于头颈5000队列,采用Cox比例风险模型分析水果、蔬菜和油炸食品摄入与头颈癌总生存的关联,并校正临床、社会及生活方式变量(含吸烟、饮酒和HPV状态)。充分校正吸烟和饮酒后,水果摄入与生存的关联消失(HR 0.91,95%CI 0.67,1.23;趋势P=.55),蔬菜摄入的关联减弱(HR 0.79,95%CI 0.61,1.03;趋势P=.04);油炸食品摄入与生存无关联(HR 0.88,95%CI 0.72,1.07;趋势P=.13)。结论认为蔬菜摄入与头颈癌生存存在适度关联,但存在吸烟和饮酒的混杂。
要点
- 充分校正吸烟和饮酒后,水果摄入与头颈癌生存的关联消失(HR 0.91,95%CI 0.67,1.23)。
- 蔬菜摄入与头颈癌生存存在适度关联,充分校正后HR 0.79(95%CI 0.61,1.03)。
- 油炸食品摄入与头颈癌生存无关联(HR 0.88,95%CI 0.72,1.07)。
核心结论
- 蔬菜摄入与头颈癌生存存在适度关联,但存在吸烟和饮酒的混杂。
研究设计与样本
- 基于头颈5000队列,采用Cox比例风险模型分析水果、蔬菜和油炸食品摄入与头颈癌总生存的关联,校正临床、社会及生活方式变量(含吸烟、饮酒和HPV状态)。
主要发现
- 在最小校正分析中,水果和蔬菜摄入与生存改善相关。
- 充分校正吸烟和饮酒后,水果摄入与生存的关联消失(HR 0.91,95%CI 0.67,1.23;趋势P=.55)。
- 充分校正吸烟和饮酒后,蔬菜摄入与生存的关联减弱(HR 0.79,95%CI 0.61,1.03;趋势P=.04)。
- 在最小校正或充分校正分析中,油炸食品摄入与生存均无关联(HR 0.88,95%CI 0.72,1.07;趋势P=.13)。
局限性
- 存在吸烟和饮酒的混杂。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
BACKGROUND: The association between diet and head and neck cancer (HNC) survival is unclear. METHODS: Cox proportional hazard models measured the association between fruit, vegetable, and deep-fried food intake and HNC overall survival adjusting for clinical, social and lifestyle variables including smoking, alcohol, and HPV status. RESULTS: Fruit and vegetable intake and improved survival were associated in minimally adjusted analyses. Following adjustment for smoking and alcohol consumption (fully adjusted analyses), the association with survival disappeared for fruit (HR 0.91, 95% CI 0.67, 1.23; P for trend = .55) and attenuated for vegetables (HR 0.79, 95% CI 0.61, 1.03; P for trend = .04). We observed no association between survival and deep-fried food intake in minimally adjusted or fully adjusted analyses (HR 0.88 95% CI 0.72, 1.07; P for trend = .13). CONCLUSIONS: Vegetable intake and HNC survival are modestly associated. There is some confounding by tobacco and alcohol consumption.
引用
- Lang S, Schimansky S, Beynon R, et al. Dietary behaviors and survival in people with head and neck cancer: Results from Head and Neck 5000. Head Neck. 2019. PMID: 30698303. doi:10.1002/hed.25660
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。