Previous studies have shown that the risk of acute pancreatitis (AP) is increased in obesity population, while obese patients are often combined with abnormal fasting plasma glucose (FPG). It still remians controversial whether FPG independently increases the risk of AP and the relationship between FPG and the risk of AP in non-obese patients has been rarely reported in China and abroad.

To explore the association between baseline FPG level and the risk of AP in non-obese population.

Using a prospective cohort study method, a total of 102 512 non-obese cases from the Kailuan study cohort who completed physical examination for the first time in KaiLuan General Hospital and its 10 affiliated hospitals from 2006 to 2009 were enrolled as study subjects. Epidemiological data, anthropometric data, laboratory test indicators and other information of the subjects were collected. The study subjects were divided into 4 groups according to the FPG quartile: the first quartile group (group Q1, FPG≤4.66 mmol/L, n=25 929) ; the second quartile group (group Q2, 4.66 mmol/L≤FPG<5.10 mmol/L, n=25 797) ; the third quartile group (group Q3, 5.10 mmol/L≤FPG<5.67 mmol/L, n=25 162) ; the fourth quartile group (group Q4, FPG≥5.67 mmol/L, n=25 624). The Kaplan-Meier method was used to plot the survival curves of new-onset AP in non-obese population. The cumulative incidence of AP in non-obese population in different FPG level groups were calculated and Log-rank method was used for inter-group test. The Cox proportional hazard regression model was used to analyze the influencing factors for the new-onset AP in non-obese population and the correlation between different FPG level groupings and new-onset AP in non-obese population.

The median follow-up time in this study was (12.8±2.4) years with the cumulative incidence of 320 cases and incidence density of 2.44 cases per 10 000 person-years in AP. There were statistically significant differences in the cumulative incidence of AP among the 4 FPG level groups (χ²=13.96, P<0.001). The results of Cox proportional hazard regression analysis showed that advanced age〔HR=1.02, 95%CI (1.01, 1.03), P=0.001〕, high triacylglycerol (TG) level〔HR=1.22, 95%CI (1.13, 1.30), P<0.001〕, history of cholithiasis〔HR=2.79, 95%CI (1.88, 4.13), P<0.001〕were risk factors for new-onset AP in non-obese population. Years of education ≥9 years〔HR=0.65, 95%CI (0.47, 0.90), P<0.001〕was the protective factor for new-onset AP in non-obese population. The HR for new-onset AP in group Q4 was 1.40 〔95%CI (1.02, 1.92), P=0.038〕. After excluding the population applying hypoglycemic drugs, the conclusions were unchanged, the HR for new-onset AP in group Q4 was 1.40 〔95%CI (1.02, 1.92), P=0.036〕.

Advanced age, high TG levels, and history of cholithiasis are risk factors for new-onset AP, years of education ≥9 years is the protective factor for new-onset AP. And the risk of AP increases when FPG ≥5.67 mmol/L in non-obose population.

研究表明肥胖人群急性胰腺炎（AP）的发病风险增加，而肥胖患者常伴有空腹血糖（FPG）异常，FPG是否独立增加AP的发病风险尚存在争议，且国内外关于非肥胖人群FPG与AP发病风险的关系鲜见报道。

探究开滦研究队列中非肥胖人群基线FPG水平与AP发病风险的关联。

采用前瞻性队列研究方法，选取2006—2009年首次在开滦总医院及其下属10家医院完成体检的开滦研究队列非肥胖人群102 512例为研究对象。收集研究对象的流行病学资料、人体测量学数据、实验室检查指标等资料，将研究对象按FPG四分位数分为4组：第一分位组（Q1组，FPG≤4.66 mmol/L，n=25 929），第二分位组（Q2组，4.66 mmol/L≤FPG<5.10 mmol/L，n=25 797），第三分位组（Q3组，5.10 mmol/L≤FPG<5.67 mmol/L，n=25 162），第四分位组（Q4组，FPG≥5.67 mmol/L，n=25 624）。采用Kaplan-Meier法绘制非肥胖人群新发AP的生存曲线图，计算不同FPG水平分组非肥胖人群AP累积发病率，并采用Log-rank法进行组间检验。采用Cox比例风险回归模型分析非肥胖人群新发AP的影响因素及不同FPG水平分组与非肥胖人群新发AP的相关性。

本研究中位随访时间为（12.8±2.4）年，累计发生AP 320例，AP的发病密度为2.44例/万人年，4组非肥胖人群AP累积发病率比较，差异有统计学意义（χ²=13.96，P<0.001）。Cox比例风险回归模型分析结果显示，高龄〔HR=1.02，95%CI（1.01，1.03），P=0.001〕、高三酰甘油（TG）水平〔HR=1.22，95%CI（1.13，1.30），P<0.001〕、有胆石症病史〔HR=2.79，95%CI（1.88，4.13），P<0.001〕是非肥胖人群新发AP的危险因素；受教育年限≥9年〔HR=0.65，95%CI（0.47，0.90），P<0.001〕是非肥胖人群新发AP的保护因素；Q4组新发AP的HR为1.40〔95%CI（1.02，1.92），P=0.038〕。剔除应用降糖药人群后，结论未发生改变，Q4组新发AP的HR为1.40〔95%CI（1.02，1.92），P=0.036〕。

高龄、高TG水平、有胆石症病史是非肥胖人群新发AP的危险因素；受教育年限≥9年是非肥胖人群新发AP的保护因素。在非肥胖人群中，当FPG≥5.67 mmol/L时AP的发病风险会增加。