Abstract:
Background　It is generally believed that hypertriglyceridemia （HTG） increases the risk of developing acute pancreatitis （AP）． However，the relationship between hypercholesterol and AP has not yet been confirmed.Objective　To investigate the effects of serum total cholesterol （TC） on the risk for new-onset AP.Methods　 A prospective cohort study design was applied.Study cohort consisted of 125 210 current employees and retired employees from Kailuan Group who participated in the health check-ups between 2006 and 2009 launched by the group，the starting point of follow-up was when the subjects completed the health examination, and the end point was new AP events，deaths or the end of follow-up （2014-12-31）.and had complete data on serum TC with no AP history at baseline.Participants were divided into normal blood lipid group（TC<5.2 mmol/L），borderline high blood lipid group （TC 5.2-<6.2 mmol/L） and high blood lipid group （TC≥6.2 mmol/L） according to the baseline fasting serum TC levels.Baseline data and incidence of new-onset AP data were collected.The survival curve was plotted by Kaplan-Meier method.The cumulative incidence was determined by Log-Rank method.Multivariate Cox proportional hazards regression model was used to explore the effect of baseline fasting serum TC level on new-onset AP.Results　The sex ratio，average age，BMI，systolic blood pressure，fasting blood glucose，high-density lipoprotein cholesterol （HDL-C），triglyceride，and length of education，rates of smoking，drinking，physical exercising，history of hypertension，and diabetes history differed significantly between the groups （P<0.05）． All of them received an average follow-up period of （7.4±1.2） years，during which 195 cases were found with AP.The density rates of AP were 1.99/10 000，1.81/10 000，and 3.64/10 000 persons-years，in normal，borderline high and high blood lipid groups，respectively.The cumulative incidence in normal，borderline high and high blood lipid groups，was 0.16%，0.14%，0.27%，respectively，showing a significant inter-group difference by Log-Rank test （χ2=12.37，P=0.002）． After adjusting for age，gender，BMI，HDL-C，TG，smoking，drinking，length of education，physical exercising，history of hypertension and history of diabetes，Cox regression analysis showed that compared with the normal blood lipid group，borderline high blood lipid group 〔HR=0.84，95%CI（0.59，1.19）〕 and high blood lipid group 〔HR=1.56，95%CI （1.06，2.31）〕 had a higher risk for AP.Conclusion　Baseline fasting serum TC level greater than or equal to 6.2 mmol/L increases the risk of developing AP.

Key words:
Pancreatitis, Cholesterol, Hypercholesterolemia, Hypertrigly ceridemia, Root cause analysis, Cohort studies

摘要： 背景　高三酰甘油血症作为急性胰腺炎（AP）发病的危险因素已经成为共识，高胆固醇血症是否引起AP发病风险增加尚存在争议。目的　探讨血清胆固醇对新发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—2009年健康体检、无AP病史及总胆固醇（TC）水平资料完整的开滦集团在职及离退休职工作为观察队列，以观察对象完成健康体检时点为随访起点，以新发AP事件、死亡或至随访结束（2014-12-31）为随访终点，最终纳入统计分析者125 210例。研究人群按基线空腹血清TC水平分为3组，血脂合适水平组（TC<5.2 mmol/L）、血脂边缘升高组（TC 5.2~<6.2 mmol/L）、血脂升高组（TC≥6.2 mmol/L）。观察各组人群基线资料和新发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基线空腹血清TC水平对新发AP事件的影响。结果　3组性别、年龄、BMI、收缩压、空腹血糖、高密度脂蛋白胆固醇（HDL-C）、三酰甘油（TG）及吸烟、饮酒、受教育程度、体育锻炼、高血压病史、糖尿病病史者占比比较，差异有统计学意义（P<0.05）。随访（7.4±1.2）年，共发生AP 195例，血脂合适水平组、血脂边缘升高组、血脂升高组的AP发病密度分别为1.99/万人年、1.81/万人年、3.64/万人年。AP的累积发病率分别为0.16%、0.14%、0.27%，经Log-Rank检验，差异有统计学意义（χ2 =12.37，P=0.002）。校正了年龄、性别、BMI、HDL-C、TG、吸烟、饮酒、受教育程度、体育锻炼、高血压病史、糖尿病病史后，Cox回归分析结果显示，与血脂合适水平组比较，血脂边缘升高组和血脂升高组发生AP的HR（95%CI）分别为0.84（0.59，1.19）、1.56（1.06，2.31）。结论　基线空腹血清TC水平≥6.2 mmol/L时增加了AP的发病风险。

关键词: 胰腺炎, 胆固醇, 高胆固醇血症, 高甘油三酯血症, 影响因素分析, 队列研究