Data_Sheet_1_The impact of smoking and alcohol consumption on rosacea: a multivariable Mendelian randomization study.docx

BackgroundsObservational studies have shown that cigarette smoking is inversely associated with risk of rosacea, However, it remains uncertain whether this association is causal or it is a result of reverse causation, and whether this association is affected by drinking behaviors.
MethodsThis study utilized the summary-level data from the largest genome-wide association study (GWAS) for smoking, alcohol consumption, and rosacea. The objective was to investigate the effect of genetically predicted exposures to smoking and alcohol consumption on the risk of developing rosacea. Two-sample bidirectional Mendelian randomization (MR) was applied, accompanied by sensitive analyses to validate the robustness of findings. Furthermore, multivariable MR was conducted to evaluate the direct impact of smoking on rosacea.
ResultsA decreased risk of rosacea was observed in individuals with genetically predicted lifetime smoking [odds ratio (OR)MR − IVW = 0.53; 95% confidence interval (CI), 0.318–0.897; P = 0.017], and number of cigarettes per day (ORMR − IVW = 0.55; 95% CI, 0.358–0.845; P = 0.006). However, no significant associations were found between initiation of regular smoking, smoking cessation, smoking initiation, alcohol consumption and rosacea. Reverse MR analysis did not show any associations between genetic liability toward rosacea and smoking or alcohol drinking. Importantly, the effect of lifetime smoking and the number of cigarettes per day on rosacea remained significant even after adjusting for alcohol consumption in multivariable MR analysis.
ConclusionSmoking was causally related to a lower risk of rosacea, while alcohol consumption does not appear to be associated with risk of rosacea.

背景：既往观察性研究显示，吸烟与酒渣鼻（rosacea）发病风险呈负相关。然而，目前仍不确定这一关联是否为因果关联，或是反向因果所致，同时该关联是否受饮酒行为影响亦尚不明确。
方法：本研究利用针对吸烟、饮酒及酒渣鼻的最大规模全基因组关联研究（Genome-Wide Association Study, GWAS）的汇总级数据，旨在探究遗传预测的吸烟与饮酒暴露对酒渣鼻发病风险的影响。本研究采用两样本双向孟德尔随机化（Mendelian randomization, MR）分析，并辅以敏感性分析以验证研究结果的稳健性；此外，还通过多变量孟德尔随机化分析评估吸烟对酒渣鼻的直接影响。
结果：遗传预测的终身吸烟量[孟德尔随机化-逆方差加权比值比（MR-IVW OR）=0.53；95%置信区间（CI）：0.318~0.897；P=0.017]及每日吸烟支数（MR-IVW OR=0.55；95%CI：0.358~0.845；P=0.006）可使酒渣鼻发病风险降低。然而，规律吸烟起始、戒烟、吸烟启动与饮酒均未与酒渣鼻呈现显著关联。反向孟德尔随机化分析未显示酒渣鼻遗传易感性与吸烟或饮酒存在任何关联。值得注意的是，即使在多变量孟德尔随机化分析中校正饮酒因素后，终身吸烟量及每日吸烟支数对酒渣鼻的影响仍具有统计学显著性。
结论：吸烟与酒渣鼻发病风险降低存在因果关联，而饮酒似乎与酒渣鼻发病风险无显著关联。