Abstract: Background　Current studies have shown that smoking is a risk factor not only for the chronic pancreatitis but also for the pancreatic cancer.However，there is still no research on the impact of smoking on the risk of acute pancreatitis（AP）.Objective　To explore the impact of smoking on the risk of AP.Methods　The prospective cohort research method was adopted in this paper.The on-the-job and retired Kailuan Group employees going through the 2006—2007 annual health check-ups with no AP history but complete smoking information were selected as the observational cohort.A total of 98 287 cases were included in the analysis finally.The study population was divided into the smoking group（n=39 635） and non-smoking group（n=58 652）.The incidence of AP was compared between the two groups.The survival curves were created using the Kaplan-Meier method to calculate the cumulative incidence which was verified through the Log-Rank method.Furthermore，a multivariate analysis was done by Cox proportional hazard regression model about the impact of smoking on new-onset AP events.Results　During the mean follow-up period of（9.7±1.4） years，a total of 158 AP cases occurred.The AP incidence density of the total population stood at 1.56/10 000 persons-year.The AP incidence for the smoking and non-smoking groups were 1.98/10 000 persons-year and 1.45/10 000 persons-year，respectively.As verified by the Log-Rank method，the cumulative incidence of the smoking group was higher than that of the non-smoking group〔0.19%（76/39 635） vs. 0.14%（82/58 652），P<0.05〕］．The Cox proportional hazard regression model showed that the onset risk of AP in the smoking group was 1.82 times higher than that in the non-smoking group〔95%CI（1.39，2.38）〕.After adjusting the indicators（gender，age，systolic pressure，waist circumference，hip circumference，total cholesterol，triglyceride，high-density lipoprotein cholesterol，uric acid，creatinine，alcohol consumption，physical exercise，and history of hypertension） with statistical differences between the two groups at baseline，the onset risk of AP in the smoking group was 1.58 times higher than that in the non-smoking group〔95%CI（1.05，2.38）〕.Conclusion　Smoking is a key risk factor for AP.Early smoking cessation is recommended as the part of the clinical treatment of AP patients.

Key words: Smoking；Pancreatitis, acute；Cohort studies

摘要： 背景　现有研究表明，吸烟既是慢性胰腺炎发病的危险因素，也是胰腺癌发病的危险因素，但目前尚缺乏吸烟对急性胰腺炎（AP）发病风险影响的研究。目的　探讨吸烟对AP发病风险的影响。方法　采用前瞻性队列研究方法，以参加开滦集团2006—2007年健康体检、无AP病史且吸烟资料完整的开滦集团在职及离退休职工为观察队列，最终纳入分析者98 287例。研究人群按吸烟（n=39 635）和非吸烟（n=58 652）分为两组，比较两组人群的AP发病率；经Kaplan-Meier法绘制生存曲线，计算累积发病率，以Log-Rank法进行检验，并采用多因素Cox比例风险回归模型分析吸烟对新发AP事件的影响。结果　平均随访（9.7±1.4）年，共发生AP 158例，总人群AP发病密度为1.56/万人年，吸烟组和非吸烟组的AP发病密度分别为1.98/万人年和1.45/万人年。经Log-Rank法检验，吸烟组的累积发病率高于非吸烟组〔0.19%（76/39 635）比0.14%（82/58 652），P<0.05〕。Cox比例风险回归模型显示，吸烟组的AP发病风险是非吸烟组的1.82倍〔95%CI（1.39，2.38）〕；校正了基线时两组间存在统计学差异的指标（性别、年龄、收缩压、腰围、臀围、总胆固醇、三酰甘油、高密度脂蛋白胆固醇、尿酸、肌酐、饮酒、体育锻炼、高血压病史）后，吸烟组的AP发病风险是非吸烟组的1.58倍〔95%CI（1.05，2.38）〕。结论　吸烟是AP发生的重要危险因素，建议将早期戒烟作为AP患者临床治疗管理的一部分。

关键词: 吸烟；胰腺炎, 急性；队列研究