【摘要】 目的 探讨缺血性脑卒中(IS)患者入院时的残余胆固醇(RC)水平与出院后死亡风险的关系。方法 选取35～80岁的IS患者2 021例为研究对象，收集出院后死亡终点数据。采用限制性立方样条(RCS)回归，分析患者入院时RC水平与死亡结局发生风险的剂量-反应关系。利用Cox回归计算风险比(HR)与95%CI,分析IS患者入院时RC水平与出院后死亡风险的关联。结果 根据RCS模型，RC水平与IS死亡和其他原因死亡具有非线性关联(P<0.001)。以RC水平的中位数为截断值，将研究对象分为低水平RC组(RC<0.72 mmol/L)和高水平RC组(RC≥0.72 mmol/L)。与高水平RC组相比，低水平RC组年龄、男性比率显著升高，空腹血糖(GLU)、总胆固醇(TC)、三酰甘油(TG)、低密度脂蛋白胆固醇(LDL-C)、非高密度脂蛋白胆固醇(HDL-C)、载脂蛋白A-1(ApoA-1)、载脂蛋白B(ApoB)水平及TG/HDL-C、TG/HDL-C、LDL-C/HDL-C、糖尿病比率均显著降低(P<0.05～0.01)。Cox回归分析显示，未调整协变量时，与低水平RC组相比，高水平RC组呈现出较低的全因死亡风险(HR=0.765,95%CI:0.619～0.946,P=0.013)和较低的IS死亡风险(HR=0.638,95%CI:0.435～0.936,P=0.022);调整性别、年龄、吸烟史、饮酒史、高血压病史、糖尿病病史后，高水平组仍然呈现出较低的全因死亡风险(HR=0.760,95%CI:0.614～0.941,P=0.012)和较低的IS死亡风险(HR=0.653,95%CI:0.444～0.961,P=0.031)。男性(HR=0.753,95%CI:0.572～0.990,P=0.042)、年龄≥65岁(HR=0.755,95%CI:0.594～0.959,P=0.021)、非吸烟(HR=0.746,95%CI:0.590～0.943,P=0.014)、非饮酒(HR=0.735,95%CI:0.588～0.919,P=0.007)、合并高血压病(HR=0.738,95%CI:0.580～0.940,P=0.014)、不合并糖尿病(HR=0.724,95%CI:0.561～0.934,P=0.013)患者中，高水平RC(≥0.72 mmol/L)与全因死亡风险降低关联具有统计学意义。年龄≥65岁(HR=0.598,95%CI:0.391～0.916,P=0.018)、非吸烟(HR=0.628,95%CI:0.408～0.967,P=0.035)、非饮酒(HR=0.656,95%CI:0.439～0.979,P=0.039)、不合并高血压病(HR=0.321,95%CI:0.108～0.957,P=0.041)、不合并糖尿病(HR=0.607,95%CI:0.389～0.947,P=0.028)患者中，RC≥0.72 mmol/L与IS死亡风险降低关联具有统计学意义。调整年龄、性别、吸烟史、饮酒史、高血压病史、糖尿病史多因素后，在男性、年龄≥65岁、不合并糖尿病患者中，RC≥0.72 mmol/L与全因死亡风险增加均无显著关联关系(均P>0.05);在年龄≥65岁、非吸烟、非饮酒、不合并高血压病、不合并糖尿病患者中，RC≥0.72 mmol/L与IS死亡风险增加均无显著关联关系(均P>0.05)。与高水平RC组相比，低水平RC组IS患者预后全因死亡、IS死亡、其他原因死亡的发生率更低，生存率更高。结论 IS患者入院时RC<0.72 mmol/L会增加患者出院后远期的全因死亡和IS死亡风险。更多还原

【Abstract】 Objective Exploring the relationship between residual cholesterol(RC) levels at admission and the risk of death after discharge in patients with ischemic stroke(IS). Methods Select 2 021 IS patients aged 35-80 as the research subjects, and collect endpoint data on mortality after discharge. Using restricted cubic spline(RCS) regression, analyze the dose-response relationship between RC levels at admission and the risk of mortality outcomes in patients. Using Cox regression to calculate hazard ratio(HR) and 95%CI, analyze the association between RC levels at admission and mortality risk after discharge in IS patients. Results According to the RCS model,there is a non-linear correlation between RC levels and deaths from IS and other causes( P < 0. 001). Using the median RC level as the cutoff value, the study subjects were divided into low-level RC group( RC < 0. 72 mmol/L) and high-level RC group( RC≥0. 72 mmol/L). Compared with those in high level RC group,the age and male ratio in the low level RC group were significantly increased,and the levels of fasting blood glucose( GLU),total cholesterol( TC),triglyceride( TG),low-density lipoprotein cholesterol( LDL-C),non high-density lipoprotein cholesterol( HDL-C),apolipoproteins A-1( Apo A-1),apolipoproteins B( Apo B),TG/HDL-C,TG/HDLC,LDL-C/HDL-C and diabetes ratio were significantly decreased( P < 0. 05-0. 01). Cox regression analysis showed that,when the covariates were not adjusted,compared with the low level RC group,the high level RC group showed a lower risk of all-cause death( HR = 0. 765,95% CI: 0. 619-0. 946,P = 0. 013) and a lower risk of IS death( HR = 0. 638,95% CI: 0. 435-0. 936,P = 0. 022); after adjusting for gender,age,smoking history,drinking history,hypertension history and diabetes history,the high level RC group still showed a lower risk of all-cause death( HR = 0. 760,95% CI: 0. 614-0. 941,P = 0. 012) and lower IS mortality risk( HR = 0. 653,95% CI: 0. 444-0. 961,P = 0. 031). Male( HR = 0. 753,95% CI: 0. 572-0. 990,P = 0. 042),age≥65 years old( HR = 0. 755,95% CI: 0. 594-0. 959,P = 0. 021),non-smoking( HR = 0. 746,95% CI: 0. 590-0. 943,P = 0. 014),nondrinking( HR = 0. 735,95% CI: 0. 588-0. 919,P = 0. 007),hypertension( HR = 0. 738,95% CI: 0. 580-0. 940,P = 0. 014), without diabetes( HR = 0. 724, 95% CI: 0. 561-0. 934, P = 0. 013), high levels of RC( ≥0. 72 mmol/L) were statistically associated with risk of all-cause death reduced. Among patients with age ≥65years old( HR = 0. 598,95% CI: 0. 391-0. 916,P = 0. 018),non-smoking( HR = 0. 628,95% CI: 0. 408-0. 967,P = 0. 035),non-drinking( HR = 0. 656,95% CI: 0. 439-0. 979,P = 0. 039),without hypertension( HR = 0. 321,95% CI: 0. 108-0. 957,P = 0. 041),without diabetes( HR = 0. 607,95% CI: 0. 389-0. 947,P = 0. 028),RC≥0. 72 mmol/L was statistically associated with risk of death of IS reduced. After adjusting for age,gender,smoking history,drinking history,hypertension history and diabetes history,among men,age ≥ 65 years old,without diabetes,RC ≥ 0. 72 mmol/L was not significantly associated with the increased risk of all-cause death( all P >0. 05); there was no significant correlation between RC ≥ 0. 72 mmol/L and the increased risk of death of IS in patients with age ≥ 65 years old,non-smoking,non-drinking,non-hypertension and non-diabetes( all P > 0. 05).Compared with the high-level RC group,the low-level RC group had a lower incidence of all-cause death,IS death,and other causes of death,and a higher survival rate. Conclusion The RC level of IS patients which is lower than0. 72 mmol/L at admission will increase the risk of all-cause death and IS death in the long term after discharge.更多还原