目的:探讨累积空腹血糖(cumFPG)暴露与全因死亡的关系。方法:采用前瞻性队列研究方法，以连续参加2006—2007、2008—2009、2010—2011年度健康体检且资料完整的开滦集团56 845名职工为观察对象，平均随访7.77年，以发生全因死亡为终点事件。按cumFPG四分位分组，比较各组全因死亡发生密度的差异；采用n Cox比例风险模型和自然样条函数分析cumFPG对全因死亡的影响。n 结果:平均随访(7.77±1.05)年，全因死亡发生密度随cumFPG增加呈上升趋势(分别为4.93/千人年、5.87/千人年、8.48/千人年、14.02/千人年)，经n Log-n rank检验，差异有统计学意义(n P<0.001)。n Cox比例风险模型显示，校正年龄、性别、体重指数、高血压、基线空腹血糖等混杂因素，与n Q1组比较，n Q4组发生全因死亡的n HR(95%n CI)为1.28(1.15~1.42)，cumFPG每增加1个标准差，发生全因死亡的风险增加17%。自然样条函数分析显示cumFPG与全因死亡风险呈现类似J形曲线关系。n 结论:高cumFPG是发生全因死亡的危险因素。“,”Objective:To investigate the relationship between cumulative fasting blood glucose(cumFPG)exposure and all-cause mortality.Methods:The prospective cohort study included 56 845 subjects of Kailuan Group who participated in physical examinations from 2006 to 2007, 2008 to 2009, and 2010 to 2011 with complete data and a median of 7.77 years of follow up. The end point event was all-cause mortality. The incidence of all-cause mortality was compared in various groups divided by four quartile of cumFPG. The Cox proportional hazards model and natural spline were used to analyze the effect of cumFPG on all-cause mortality.Results:During the average(7.77±1.05)years of follow-up, the incidence density of all-cause mortality was on the rise with the increase of cumFPG(4.93, 5.87, 8.48, and 14.02 per 1 000 person-years), with statistically significant difference by n Log-n rank test(n P<0.001).n Cox proportional hazards model showed that after adjusting for potential confounding factors(age, sex, body mass index, hypertension, baseline fasting plasma glucose), the n HR value(95%n CI)of all-cause mortality in the fourth quartile group was 1.28(1.15-1.42)compared to the first quartile. When cumFPG increased every standard deviation, the risk of all-cause mortality increased by 17%. Natural spline analysis exhibited a similar J curve relationship between cumFPG and all-cause mortality.n Conclusion:High cumFPG is a risk factor of all-cause mortality.