中华急诊医学杂志  2022, Vol. 31 Issue (12): 1685-1690   DOI: 10.3760/cma.j.issn.1671-0282.2022.12.020
高密度脂蛋白对链球菌血流感染预后的预测价值
段晓光 , 石朝阳 , 孙文举 , 张晓娟 , 杜丽娟 , 王海旭 , 孙同文     
郑州大学第一附属医院综合ICU,河南省重症医学重点实验室,郑州市脓毒症重点实验室,郑州 450052
摘要: 目的 探讨高密度脂蛋白(high density lipoprotein HDL)水平对链球菌血流感染患者预后的预测价值。方法 选取2015年1月至2019年12月郑州大学第一附属医院收治的链球菌血流感染患者698例为研究对象。记录患者血培养阳性48 h内血脂水平及其他临床资料。2020年1月至3月电话随访,记录患者终点事件发生情况,终点事件为确诊链球菌血流感染60 d全因死亡。依据HDL水平,将患者分为低HDL组(HDL≤0.84 mmol/L)和高HDL组(HDL>0.84 mmol/L)。采用单因素分析和多因素Cox回归分析影响链球菌血流感染预后的相关因素,绘制患者工作特征曲线(ROC)评估HDL水平对60 d预后的预测价值。Kaplan-Meier生存曲线比较不同HDL水平患者的生存差异。结果 (1)根据纳入、排除标准共纳入491例,成功随访461例(随访率93.89%),60 d存活373例,死亡88例,60 d病死率19.09%(88/461)。(2)单因素分析:年龄、总胆固醇(total cholesterol, TC)、HDL、低密度脂蛋白(low density lipoprotein LDL)、血小板、白蛋白、纤维蛋白原、甘油三脂(triglyceride TG)、肌酐、谷丙转氨酶、谷草转氨酶、白细胞、PCT、总胆红素、直接胆红素、呼吸衰竭、休克在生存组和死亡组之间的差异均有统计学意义。(3)多因素Cox回归结果显示:HDL(RR=1.922, 95%CI: 1.186~3.117, P=0.008), 谷草转氨酶(RR=1.953, 95%CI:1.233~3.094, P=0.004), 休克(RR=15.196, 95%CI: 6.953~33.211, P<0.001),呼吸衰竭(RR=9.509, 95%CI: 4.232~21.367, P<0.001)为影响链球菌血流感染死亡的独立危险因素。(4)ROC曲线下面积显示: HDL单独预测链球菌血流感染的预后有一定价值, ROC曲线下面积为:0.602,HDL联合谷草转氨酶、休克和呼吸衰竭预测预后的ROC曲线下面积为:0.960,敏感度为92%,特异度为92%。(5)Kaplan-Meier生存曲线分析显示:HDL>0.84 mmol/L组链球菌血流感染患者无终点事件发生的累积存活率高于HDL≤0.84 mmol/L组,但差异无统计学意义(Log - Rank检验:χ2=0.843,P < 0.358)。结论 HDL水平低的链球菌血流感染患者60 d死亡风险增加,HDL是链球菌血流感染患者60 d死亡的独立危险因素,可作为评价链球菌血流感然预后的指标。
关键词: 高密度脂蛋白    链球菌    血流感染    预后    
Prognostic value of high density lipoprotein level in patients with streptococcal bloodstream infection
Duan Xiaoguang , Shi Zhaoyang, , Sun Wenjun , Zhang Xiaojuan , Du Lijuan , Wang Haixu , Sun Tongwen     
Department of General ICU, the First Affiliated Hospital of Zhengzhou University, Henan Key Laboratory of Critical Care Medicine, Zhengzhou Key Laboratory of Sepsis, Zhengzhou 450052, China
Abstract: Objective Investigate the prognostic value of high density lipoprotein (HDL) level in patients with streptococcal bloodstream infection. Methods A total of 698 patients with streptococcal bloodstream infection admitted to the First Affiliated Hospital of Zhengzhou University from January 2015 to December 2019 were enrolled. Serum lipid and other clinical data of patients with positive blood culture within 48 h were recorded. The patients were followed up by telephone from January to March in 2020, and the end-point events were recorded, which were all-cause death 60 days after the diagnosis of streptococcal bloodstream infection. The patients were divided into two groups according to the levels of HDL: low HDL group (HDL ≤0.84 mmol/L) and high HDL group (HDL > 0.84 mmol/L). Univariate and multivariate Cox regression analysis were used to analyze the 60-day prognostic factors of patients with streptococcus bloodstream infection. The receiver operating characteristic (ROC) curve was used to explore predictive value of HDL level for 60-day prognosis of patients. Kaplan-Meier survival curve was used to compare the cumulative survival of patients with different HDL levels. Results (1) A total of 491 patients were enrolled according to the inclusion criteria, and 461 patients were followed up successfully, with a follow-up rate of 93.89%. There were 373 survival patients and 88 death patients at 60 days, with a 60-day mortality rate of 19.09% (88/461). (2) There were significant differences in age, total cholesterol (TC), HDL, low density lipoprotein (LDL), platelets, albumin, fibrinogen, triglyceride (TG), creatinine, alanine aminotransferase, aspartate aminotransferase, white blood cell, PCT, total bilirubin, direct bilirubin, and respiratory failure and shock between the survival group and death group. (3) Multivariate Cox regression analysis showed that HDL (RR=1.922, 95% CI: 1.186-3.117, P=0.008), aspartate aminotransferase (RR=1.953, 95%CI: 1.233-3.094, P=0.004), shock (RR=15.196, 95% CI: 6.953-33.211, P < 0.001), and respiratory failure (RR=9.509, 95%CI: 4.232-21.367, P < 0.001) were independent risk factors for 60-day mortality of patients with streptococcal bloodstream infection. (4) The ROC curve analysis showed that HDL alone had a certain value in predicting the 60-day prognosis of patients with streptococcal bloodstream infection. The area under ROC curve (AUC) was 0.602, and the AUC of the combined predictive value of HDL, aspartate aminotransferase, shock and respiratory failure was 0.960, with a sensitivity of 92% and a specificity of 92%.(5)Kaplan-Meier survival curve analysis showed that the cumulative survival rate of patients without endpoint event in the HDL > 0.84 mmol/L group was higher than that in the HDL ≤ 0.84 mmol/L group, but without statistically significant difference (Log-Rank test: χ20.843, P < 0.358). Conclusions Patients with low HDL level of streptococcal bloodstream infection have an increased risk of 60-day death. HDL is an independent risk factor for 60-day death in patients with streptococcal bloodstream infection, and can be used as an indicator to evaluate the prognosis of patients with streptococcal bloodstream infection.
Key words: High density lipoprotein    Streptococcus    Bloodstream infection    Prognosis    

脓毒症队列和大型流行病学研究显示,高密度脂蛋白(high density lipoprotein HDL)和低密度脂蛋白(low density lipoprotein LDL)水平降低发生感染性疾病风险增加,且预后较差[1-5]。Trinder等[6]研究显示HDL和LDL水平与感染住院的风险呈负相关。然而这些流行病学联系容易受混杂因素的影响,目前尚不清楚它们是否存在因果关系, 需更多研究证实。目前关于HDL水平与血流感染预后关系的研究较少,本文旨在研究HDL水平对链球菌血流感染患者预后的预测价值。

1 资料与方法 1.1 一般资料

本研究符合医学伦理要求,已通过郑州大学第一附属医院伦理委员会审查,伦理编号:2020-KY-157。选取2015年1月至2019年12月郑州大学第一附属医院收治的血培养显示为链球菌的患者为研究对象。纳入标准:血培养报告为链球菌,临床表现支持链球菌血流感染。排除标准:恶性肿瘤终末期、脑死亡、失访、放弃治疗、数据缺失的患者。筛选流程见图 1

图 1 HDL对链球菌血流感染的预测价值研究对象筛选流程 Fig 1 Screening process of the predictive value of HDL for streptococcal bloodstream infection
1.2 研究方法 1.2.1 资料收集

记录患者的性别、年龄、既往史、感染部位、呼吸衰竭、休克及血培养阳性48 h内血脂、凝血、肝肾功能、PCT、CRP、白蛋白等指标。

1.2.2 随访与分组

于2020年1月至3月电话随访,记录患者终点事件发生情况,终点事件为确诊链球菌血流感染60 d全因死亡。根据60 d预后分为死亡组和存活组;按照全部患者HDL的中位数为界限分为低HDL组(HDL≤0.84 mmol/L)和高HDL组(HDL>0.84 mmol/L)。

1.3 统计学方法

用SPSS 21.0软件进行数据分析,分类数据用百分率表示。连续性数据用均数±标准差(x±s)表示,组间比较符合正态分布的资料用成组t检验,不符合正态分布的资料用秩和检验(Mann-Whitney U检验)。分类资料比较用χ2检验。Cox比例风险回归模型分析血脂水平对链球菌血流感染预后危险因素。采用Kaplan-Meier生存曲线和log-rank检验比较不同血脂水平患者生存率的区别。所有统计均为双侧检验,检验水平为α=0.05。

2 结果 2.1 患者基本情况

根据纳入、排除标准共491例纳入研究,成功随访461例(男性294例,女性167例),随访率93.89%, 60 d存活373例,死亡88例,60 d病死率19.09%(88/461),生存组: 年龄(48.2±18.98)岁,死亡组:年龄(55.77±19.09)岁。

2.2 影响链球菌血流感染预后的单因素分析

与存活组比较,死亡组患者年龄较大,总胆固醇(TC)、HDL和LDL、血小板、白蛋白、纤维蛋白原水平较低,甘油三酯(TG)、肌酐、谷丙转氨酶、谷草转氨酶、降钙素原(PCT)、总胆红素、直接胆红素水平较高,呼吸衰竭、休克发生率较高,差异均有统计学意义(均P<0.05)。见表 1

表 1 一般资料和实验室指标 Table 1 General information and laboratory indicators
指标 生存组 死亡组 χ2/t/Z P
男性(n,%) 229(49.7) 65(14.1) 4.792 0.029
年龄(岁,x±s 48.20±18.98 55.77±19.09 10.966 0.001
基础疾病(n,%) 199(53.4) 62(70.5) 8.480 0.004
感染部位(n,%) 19.877 0.003
  肺部 112(30.0) 39(44.3)
  肠道 38(10.2) 10(11.4)
  胆道 22(5.9) 3(3.4)
  皮肤软组织 42(11.3) 14(15.9)
  中枢神经 18(4.8) 7(8.0)
  IE 105(28.2) 7(8.0)
  其他 36(9.7) 8(9.1)
TC [MQ1, Q3)] 3.56(2.92, 4.21) 3.21(2.30, 4.07) 3.010 0.003
TG [MQ1, Q3)] 1.03(0.80, 1.44) 1.37(0.87, 2.06) 3.330 0.001
HDL [MQ1, Q3)] 0.86(0.62, 1.25) 0.78(0.43, 1.02) 2.981 0.003
LDL [MQ1, Q3)] 2.17(1.62, 2.72) 1.66(0.82, 2.71) 3.778 <0.001
PCT[MQ1, Q3)] 0.27(0.07, 2.16) 12.18(1.82, 49.17) 8.700 <0.001
CRP[MQ1, Q3)] 53.77(31.00, 216.74) 53.77(8.07, 101.56) 4.481 <0.001
白细胞[MQ1, Q3))] 8.40(5.80, 12.84) 8.82(3.53, 14.40) 0.266 0.791
血小板[MQ1, Q3)] 180.00(123.00, 245.00) 104.00(46.75, 185.75) 5.311 <0.001
白蛋白(x±s 34.04±7.90 29.51±8.95 3.425 0.001
纤维蛋白原[MQ1, Q3)] 4.04(3.06, 4.87) 3.45(2.51, 4.41) 2.446 0.014
谷丙转氨酶[MQ1, Q3)] 20.00(12.00, 39.00) 27.05(16.25, 47.30) 2.658 0.008
谷草转氨酶[MQ1, Q3)] 22.00(16.95, 39.00) 42, .50(23.00, 89.75) 5.704 <0.001
肌酐[MQ1, Q3)] 64.00(53.00, 81.00) 79.00(60.00, 114.00) 4.351 <0.001
总胆红素[MQ1, Q3)] 10.60(7.30,17.10) 14.97(8.90, 31.00) 3.800 <0.001
直接胆红素[MQ1, Q3)] 5.50(3.40, 8.60) 8.30(4.30, 18.13) 4.006 <0.001
休克(n,%) 14(3.8) 75(85.2) 303.387 <0.001
呼吸衰竭(n,%) 21(5.6) 77(87.5) 285.108 <0.001
2.3 多因素Cox回归分析

以60 d结局指标为因变量,选择单因素分析中差异有统计学意义的年龄、性别、基础疾病、感染源、TC、TG、HDL、LDL、血小板、白蛋白、纤维蛋白原、谷丙转氨酶、谷草转氨酶、肌酐、总胆红素、直接胆红素、PCT、CRP、休克、呼吸衰竭作为协变量(各自变量赋值方式见表 2),进行多因素Cox回归分析,结果显示,HDL、谷草转氨酶、休克和呼吸衰竭与链球菌血流感染患者60 d病死率具有相关性,是链球菌血流感染患者短期预后的危险因素(均P<0.05),见表 3

表 2 链球菌血流感染患者60 d死亡的多因素Cox回归分析中各自变量赋值方式 Table 2 Assignment of the independent variables in multivariate Cox regression analysis of 60-day death in patients with streptococcal bloodstream infection
变量 赋值
性别 女=1,男=2
年龄(岁) ≤60岁=1,>60岁=2
基础疾病 无=1,有=2
感染源 无=1,肺部=2,肠道=3,胆道=4,皮肤软组织=5,中枢神经系统=6,IE=7
TC ≤3.51=1,>3.51=2
TG ≤1.07=1,>1.07=2
HDL >0.84=1,≤0.84=2
LDL >2.12=1,≤2.12=2
血小板 ≤350=1,>350=2
白蛋白 >35=1,≤35=2
纤维蛋白原 >4=1,≤4=2
谷丙转氨酶 ≤40=1,>40=2
谷草转氨酶 ≤40=1,>40=2
肌酐 ≤115.00=1,>115.00=2
总胆红素 ≤25=1,>25=2
直接胆红素 ≤10=1,>10=2
PCT ≤0.356=1,>0.356=2
CRP ≤60.369=1,>60.369=2
休克 无休克=1,休克=2
呼吸衰竭 无衰竭=1,衰竭=2
注:以正常参考值上、下限值为界限分组

表 3 链球菌血流感染患者60 d死亡的多因素Cox回归分析 Table 3 Multivariate Cox regression analysis of 60-day death in patients with streptococcal bloodstream infection
变量 RR 95%CI P
HDL 1.922 1.186~3.117 0.008
谷草转氨酶 1.953 1.233~3.094 0.004
休克 15.196 6.953~33.211 <0.001
呼吸衰竭 9.509 4.232~21.367 <0.001
注:RR为相对危险度,95%CI为95%可信区间
2.4 HDL、谷草转氨酶、休克和呼吸衰竭对链球菌血流感染患者60 d死亡的预测价值

HDL对链球菌血流感染患者60 d死亡有一定预测价值,预测作用相对较弱。但HDL联合谷草转氨酶、休克和呼吸衰竭对链球菌血流感染患者60 d死亡的预测价值较高,敏感度为92%,特异度为92%。见图 2表 4)。

图 2 HDL、谷草转氨酶、休克和呼吸衰竭和联合指标预测链球菌血流感染患者60 d死亡的ROC曲线 Fig 2 ROC curves of HDL, aspartate aminotransferase, shock and respiratory failure, and combined indicators predicting 60-day death in patients with streptococcal bloodstream infection

表 4 HDL、谷草转氨酶、休克和呼吸衰竭和联合指标对链球菌血流感染患者60 d死亡的预测价值 Table 4 Predictive value of HDL, aspartate aminotransferase, shock and respiratory failure, and combined indicators for 60-day death in patients with streptococcal bloodstream infection
指标 AUC 95%CI P 最佳截断值 敏感度(%) 特异度(%)
高密度脂蛋白 0.602 0.536~0.669 0.003 25.900 31.8 88.2
谷草转氨酶 0.695 0.631~0.760 0.000 39.500 58.0 75.9
休克 0.907 0.862~0.952 0.000 0.500 85.2 96.2
呼吸衰竭 0.909 0.867~0.952 0.000 0.500 87.5 94.4
联合指标 0.950 0.919~0.982 0.000 92.0 92.0
2.5 Kaplan - Meier生存曲线分析

HDL>0.84 mmol/L组链球菌血流感染患者无终点事件发生的累积存活率高于HDL≤0.84mmol/L组,但Log-Rank检验结果显示两组比较差异无统计学意义(Log-Rank检验:χ2=0.843,P < 0.358)。见图 3

图 3 不同HDL水平两组链球菌血流感染患者无终点事件发生的Kaplan-Meier生存曲线 Fig 3 Kaplan-Meier survival curves of patients with streptococcal bloodstream infection with different HDL levels without end-point events
3 讨论

链球菌尤其是化脓性链球菌引起的血流感染预后较差,其中最为凶险的是中毒休克综合征合并多脏器功能不全,病死率高达50%~70%[7-9]。聚合酶链式反应(polymerase chain reactior PCR)、二代测序技术(next generation sequencing,NGS)虽然能快速、可靠鉴别细菌,但仍有较大局限性。目前临床确诊主要依靠血培养,但培养结果通常要在收到标本后48~72 h后才能获得,即使是比较快速的革兰染色也要12~24 h,这影响了早期血流感染的诊断和治疗。因此仍需探索更多可靠指标,早期发现和评估血流感染[10-13]。本研究结果显示HDL与链球菌血流感染患者60 d病死率具有相关性,HDL≤0.84 mmol/L组60 d病死率是HDL>0.84 mmol/L组的1.922倍;ROC曲线分析显示HDL对链球菌血流感染患者60 d死亡有一定预测价值,HDL联合谷草转氨酶、休克和呼吸衰竭对链球菌血流感染患者60 d死亡的预测价值较高。

脂质水平在早期脓毒症中迅速变化,并可能预测脓毒症预后[4, 14-18]。HDL具有多种作用,包括抗炎、抗凋亡、中和细菌细胞壁脂多糖和脂磷壁酸的特性[19-22]。脂磷壁酸是革兰氏阳性细菌的主要细胞壁成分,是一种结构类似于革兰氏阴性细菌脂多糖的两亲性阴离子糖脂。脂磷壁酸被认为是可能触发全身炎症反应的主要免疫刺激成分之一[21]。Levels等[21]研究显示向健康人全血中注射脂磷壁酸,68%±10%脂磷壁酸与HDL结合、28%±8%脂磷壁酸和LDL结合和5%±4%的脂磷壁酸和极低密度脂蛋白胆固醇结合。Tanaka等[19]评价了重组HDL静脉注射在不同脓毒症模型中的作用,结果显示输注重组HDL可改善脓毒症小鼠模型的存活率。HDL可能是评估感染和脓毒症预后的良好指标。Trinder等[6]研究显示HDL水平每降低1 mmol/L,感染性疾病发生风险增加1.19倍,28 d病死率增加2.7倍。本研究显示死亡组HDL水平较低,差异有统计学意义,多因素Cox回归分析提示HDL是链球菌血流感染60 d死亡的独立危险因素。HDL>0.84 mmol/L组链球菌血流感染患者无终点事件发生的累积存活率高于HDL≤0.84 mmol/L组,但差异无统计学意义,可能与样本量较小有关。

研究表明HDL与病原相关的脂类(如脂多糖、脂磷壁酸)的亲和力最大,这些脂质介导了脓毒症患者的过度免疫激活,HDL与脂多糖、脂磷壁酸结合,阻断由其引起的炎症级联反应,显著减少炎症介质的释放[21-22]。HDL还具有免疫调节作用,抗血栓形成作用和抗氧化作用[23-25]。HDL相关的载脂蛋白,如载脂蛋白A1和载脂蛋白M,与细胞膜上的免疫细胞受体,如巨噬细胞上的toll样受体、T细胞受体和B细胞受体结合,调节免疫反应[26-29],通过活化天然免疫机制发挥抗炎活性。最终减轻感染和脓毒症引起的临床症状。

同时本研究也发现链球菌血流感染合并休克、呼吸衰竭和多脏器功能不全者,预后较差,病死率较高。休克、呼吸衰竭、谷草转氨酶是链球菌血流感染60 d死亡的独立危险因素。一项关于肺炎克雷伯菌血流感染病死率预测指标研究显示,脓毒性休克(OR: 6.42, 95%CI: 1.34~30.69)是肺炎克雷伯菌28 d病死率的独立危险因素[30]。另一项关于肺炎克雷伯菌感染危险因素的研究,对370名肺炎克雷伯杆菌血流感染患者分析显示,呼吸衰竭(OR:5.27,95%CI: 1.40~19.87)是肺炎克雷伯杆菌血流感染死亡的独立危险因素[31]。同时有研究表明脓毒症肝损伤是脓毒症患者常见的死亡原因, 改善脓毒症肝损伤有利于改善脓毒症的预后[32]

利益冲突  所有作者声明无利益冲突

作者贡献申明  段晓光:酝酿和设计实验,实施研究,采集数据,分析解释数据,文章撰写;石朝阳、孙文举、张晓娟、杜丽娟、王海旭:实施研究、采集数据分析,解释数据;孙同文:酝酿和设计实验、对文章的知识性内容作批评性审阅,研究经费,行政、技术或材料支持指导

参考文献
[1] Madsen CM, Varbo A, Tybjærg-Hansen A, et al. U-shaped relationship of HDL and risk of infectious disease: two prospective population-based cohort studies[J]. Eur Heart J, 2018, 39(14): 1181-1190. DOI:10.1093/eurheartj/ehx665
[2] Kaysen GA, Ye XL, Raimann JG, et al. Lipid levels are inversely associated with infectious and all-cause mortality: international MONDO study results[J]. J Lipid Res, 2018, 59(8): 1519-1528. DOI:10.1194/jlr.P084277
[3] Cirstea M, Walley KR, Russell JA, et al. Decreased high-density lipoprotein cholesterol level is an early prognostic marker for organ dysfunction and death in patients with suspected sepsis[J]. J Crit Care, 2017, 38: 289-294. DOI:10.1016/j.jcrc.2016.11.041
[4] Chien YF, Chen CY, Hsu CL, et al. Decreased serum level of lipoprotein cholesterol is a poor prognostic factor for patients with severe community-acquired pneumonia that required intensive care unit admission[J]. J Crit Care, 2015, 30(3): 506-510. DOI:10.1016/j.jcrc.2015.01.001
[5] Lekkou A, Mouzaki A, Siagris D, et al. Serum lipid profile, cytokine production, and clinical outcome in patients with severe sepsis[J]. J Crit Care, 2014, 29(5): 723-727. DOI:10.1016/j.jcrc.2014.04.018
[6] Trinder M, Walley KR, Boyd JH, et al. Causal inference for genetically determined levels of high-density lipoprotein cholesterol and risk of infectious disease[J]. Arterioscler Thromb Vasc Biol, 2020, 40(1): 267-278. DOI:10.1161/ATVBAHA.119.313381
[7] 王澎, 杨启文, 周翔, 等. 成人酿脓链球菌导致中毒休克综合征临床分析[J]. 中华医院感染学杂志, 2016, 26(10) 2251-2253, 2259. DOI:10.11816/cn.ni.2016-160887
[8] Pandey M, Good MF. The quest for GAS vaccine[J]. Oncotarget, 2015, 6(33): 34063-34064. DOI:10.18632/oncotarget.6140
[9] Duan XG, Zhang RF, Zhang XJ, et al. Identification of prognostic factors in patients with streptococcus bloodstream infection[J]. Front Med (Lausanne), 2022, 9: 832007. DOI:10.3389/fmed.2022.832007
[10] 刘峰宇, 孙同文. 脓毒症心肌病的诊疗进展[J]. 中华急诊医学杂志, 2022, 31(7): 858-861. DOI:10.3760/cma.j.issn.1671-0282.2022.07.003
[11] 刘峰宇, 孙同文. 脓毒症定义变迁及病理机制研究进展[J]. 中华卫生应急电子杂志, 2022, 8(1): 39-43. DOI:10.3877/cma.j.issn.2095-9133.2022.01.009
[12] Sun TW, Wang LX, Zhang HB. Intestinal microbiota in sepsis[J]. Intensive Care Res, 2022, 2(1): 1-7. DOI:10.1007/s44231-022-00001-8
[13] Zhang ZH, Chen L, Cai HB, et al. Low quality evidence supporting recommendations in the 2021 sepsis guideline: an indication for precise medicine?[J]. Intensive Care Res, 2022, 2(1): 23-25. DOI:10.1007/s44231-022-00007-2
[14] Chien JY, Jerng JS, Yu CJ, et al. Low serum level of high-density lipoprotein cholesterol is a poor prognostic factor for severe sepsis[J]. Crit Care Med, 2005, 33(8): 1688-1693. DOI:10.1097/01.ccm.0000171183.79525.6b
[15] van Leeuwen HJ, Heezius ECJM, Dallinga GM, et al. Lipoprotein metabolism in patients with severe sepsis[J]. Crit Care Med, 2003, 31(5): 1359-1366. DOI:10.1097/01.CCM.0000059724.08290.51
[16] Grion CM, Cardoso LT, Perazolo TF, et al. Lipoproteins and CETP levels as risk factors for severe sepsis in hospitalized patients[J]. Eur J Clin Invest, 2010, 40(4): 330-338. DOI:10.1111/j.1365-2362.2010.02269.x
[17] Lee SH, Park MS, Park BH, et al. Prognostic implications of serum lipid metabolism over time during sepsis[J]. Biomed Res Int, 2015, 2015: 789298. DOI:10.1155/2015/789298
[18] Liu SH, Liang HY, Li HY, et al. Effect of low high-density lipoprotein levels on mortality of septic patients: a systematic review and meta-analysis of cohort studies[J]. World J Emerg Med, 2020, 11(2): 109-116. DOI:10.5847/wjem.j.1920-8642.2020.02.008
[19] Tanaka S, Genève C, Zappella N, et al. Reconstituted high-density lipoprotein therapy improves survival in mouse models of sepsis[J]. Anesthesiology, 2020, 132(4): 825-838. DOI:10.1097/ALN.0000000000003155
[20] Khovidhunkit W, Kim MS, Memon RA, et al. Effects of infection and inflammation on lipid and lipoprotein metabolism: mechanisms and consequences to the host[J]. J Lipid Res, 2004, 45(7): 1169-1196. DOI:10.1194/jlr.r300019-jlr200
[21] Levels JH, Abraham PR, van Barreveld EP, et al. Distribution and kinetics of lipoprotein-bound lipoteichoic acid[J]. Infect Immun, 2003, 71(6): 3280-3284. DOI:10.1128/IAI.71.6.3280-3284.2003
[22] Levels JH, Abraham PR, van den Ende A, et al. Distribution and kinetics of lipoprotein-bound endotoxin[J]. Infect Immun, 2001, 69(5): 2821-2828. DOI:10.1128/IAI.69.5.2821-2828.2001
[23] de Nardo D, Labzin LI, Kono H, et al. High-density lipoprotein mediates anti-inflammatory reprogramming of macrophages via the transcriptional regulator ATF3[J]. Nat Immunol, 2014, 15(2): 152-160. DOI:10.1038/ni.2784
[24] Pajkrt D, Doran JE, Koster F, et al. Antiinflammatory effects of reconstituted high-density lipoprotein during human endotoxemia[J]. J Exp Med, 1996, 184(5): 1601-1608. DOI:10.1084/jem.184.5.1601
[25] Birjmohun RS, van Leuven SI, Levels JH, et al. High-density lipoprotein attenuates inflammation and coagulation response on endotoxin challenge in humans[J]. Arterioscler Thromb Vasc Biol, 2007, 27(5): 1153-1158. DOI:10.1161/ATVBAHA.106.136325
[26] Fessler MB, Parks JS. Intracellular lipid flux and membrane microdomains as organizing principles in inflammatory cell signaling[J]. J Immunol, 2011, 187(4): 1529-1535. DOI:10.4049/jimmunol.1100253
[27] Yvan-Charvet L, Kling J, Pagler T, et al. Cholesterol efflux potential and antiinflammatory properties of high-density lipoprotein after treatment with niacin or anacetrapib[J]. Arterioscler Thromb Vasc Biol, 2010, 30(7): 1430-1438. DOI:10.1161/ATVBAHA.110.207142
[28] Suzuki M, Pritchard DK, Becker L, et al. High-density lipoprotein suppresses the type Ⅰ interferon response, a family of potent antiviral immunoregulators, in macrophages challenged with lipopolysaccharide[J]. Circulation, 2010, 122(19): 1919-1927. DOI:10.1161/CIRCULATIONAHA.110.961193
[29] Kabouridis PS, Jury EC. Lipid rafts and T-lymphocyte function: implications for autoimmunity[J]. FEBS Lett, 2008, 582(27): 3711-3718. DOI:10.1016/j.febslet.2008.10.006
[30] Zhang Y, Guo LY, Song WQ, et al. Risk factors for carbapenem-resistant K. pneumoniae bloodstream infection and predictors of mortality in Chinese paediatric patients[J]. BMC Infect Dis, 2018, 18(1): 248. DOI:10.1186/s12879-018-3160-3
[31] Xiao TT, Yu W, Niu TS, et al. A retrospective, comparative analysis of risk factors and outcomes in carbapenem-susceptible and carbapenem-nonsusceptible Klebsiella pneumoniae bloodstream infections: tigecycline significantly increases the mortality[J]. Infect Drug Resist, 2018, 11: 595-606. DOI:10.2147/IDR.S153246
[32] Chen SN, Tan Y, Xiao XC, et al. Deletion of TLR4 attenuates lipopolysaccharide-induced acute liver injury by inhibiting inflammation and apoptosis[J]. Acta Pharmacol Sin, 2021, 42(10): 1610-1619. DOI:10.1038/s41401-020-00597-x