目的:探讨血乳酸（Lac）、乳酸清除率（LCR）和降钙素原（PCT）水平在评估脓毒症患者病情严重程度及预后中的临床意义。方法:回顾性选择2009年4月至2019年12月在首都医科大学附属北京友谊医院重症医学科住院治疗的年龄在18～80岁的脓毒症患者。收集患者临床资料，包括性别、年龄、基础疾病、感染部位、器官损伤情况、急性生理学与慢性健康状况评分Ⅱ（APACHEⅡ）、序贯器官衰竭评分（SOFA）、入重症监护病房（ICU）即刻的Lac和PCT水平、入ICU 24 h Lac、24 h LCR及28 d预后情况等。根据Sepsis-3脓毒性休克诊断标准将患者分为脓毒症组和脓毒性休克组；再根据28 d预后将脓毒性休克患者分为存活组和死亡组，比较组间患者各指标的差异。采用多因素Logistic回归分析筛选脓毒性休克患者死亡的危险因素；用受试者工作特征曲线（ROC）分析Lac、24 h LCR、PCT、SOFA、APACHEⅡ评分预测脓毒性休克患者预后的价值。结果:共入组998例脓毒症患者，男性642例，女性356例；年龄（59.56±13.22）岁。脓毒性休克患者478例，其中28 d死亡180例，存活298例。①与脓毒症组比较，脓毒性休克组患者年龄更大（岁：60.49±12.31比58.72±13.97），APACHEⅡ评分、SOFA评分、即刻Lac及PCT、24 h Lac水平更高〔APACHEⅡ（分）：24.57±7.04比19.37±6.93，SOFA（分）：7.78±3.31比4.38±3.42，即刻Lac（mmol/L）：3.00（1.70，5.00）比1.40（1.00，2.30），即刻PCT（μg/L）：0.05（0.00，4.00）比0.00（0.00，1.10），24 h Lac（mmol/L）：2.60（1.60，4.40）比1.40（1.00，2.20）〕，28 d病死率更高〔41.63%（199/478）比19.42%（101/520）〕，差异均有统计学意义（均n P<0.05）。②与脓毒性休克存活组比较，死亡组患者APACHEⅡ评分、SOFA评分、即刻Lac、24 h Lac明显升高，24 h LCR明显下降n 〔APACHEⅡ（分）：26.19±6.52比22.25±6.07，SOFA（分）：9.07±2.90比7.50±3.10，即刻Lac（mmol/L）：3.80（2.50，5.10）比2.80（2.00，3.90），24 h Lac（mmol/L）：3.20（2.20，5.60）比2.10（1.60，3.30），24 h LCR：1.43（-37.50，30.77）%比16.67（0.00，33.98）%，均n P<0.05〕。在器官功能评估方面，死亡组中心静脉压（CVP）、氧合指数（PaOn 2/FiOn 2）更低n 〔CVP（mmHg；1 mmHg＝0.133 kPa）：5.00（2.00，8.00）比6.00（2.00，9.00），PaOn 2/FiOn 2（mmHg）：184.21±84.57比199.20±86.98〕，丙氨酸转氨酶（ALT）、血肌酐（SCr）更高〔ALT（U/L）：376.56±41.43比104.17±14.10，SCr（μmol/L）：213.53±8.06比181.91±5.03〕，差异均有统计学意义（均n P<0.05）。③多因素Logistic回归分析显示，PaOn 2/FiOn 2、SCr、Lac、SOFA评分是脓毒性休克患者死亡的独立危险因素〔PaOn 2/FiOn 2：优势比（n OR）＝0.997，95%可信区间（95%n CI）为0.996～0.999，n P＝0.001；SCr：n OR＝1.001，95%n CI为1.000～1.002，n P＝0.041；Lac：n OR＝0.925，95%n CI为0.871～0.982，n P＝0.011；SOFA评分：n OR＝1.178，95%n CI为1.110～1.251，n P＝0.000〕。ROC曲线分析显示，SOFA评分、SOFA+APACHEⅡ、Lac+24 h LCR+PCT+SOFA+APACHEⅡ可以预测脓毒性休克患者死亡，ROC曲线下面积（AUC）分别为0.769（95%n CI为0.740～0.798）、0.787（95%n CI为0.759～0.815）、0.800（95%n CI为0.773～0.827），Lac、24 h LCR、PCT、SOFA、APACHEⅡ5项指标联合预测的AUC最大。n 结论:Lac是脓毒性休克患者死亡的独立危险因素，但不能预测预后，需要综合LCR、PCT、SOFA、APACHEⅡ评分及临床器官功能情况进行分析。“,”Objective:To explore the value of lactic acid (Lac), lactate clearance (LCR) and procalcitonin (PCT) in assessing the severity and predicting the prognosis in sepsis.Methods:18-80-year-old patients with sepsis admitted to the department of critical care medicine of Beijing Friendship Hospital, Capital Medical University from April 2009 to December 2019 were enrolled. The gender, age, basic illness, infection site, organ function, acute physiology and chronic health evaluationⅡ (APACHEⅡ), sequential organ failure assessment (SOFA), Lac and PCT were collected on admission to intensive care unit (ICU), as well as Lac after 24 hours, 24-hour LCR, and 28-day prognosis. The patients were divided into sepsis group and septic shock group according to Sepsis-3 criteria. According to the 28-day prognosis, the septic shock patients were divided into survival group and death group, and the differences of each index between the two groups were compared. Multivariate Logistic regression was used to analyze the risk factors of death in septic shock patients. The receiver operating characteristic (ROC) curve was used to analyze the role of Lac, LCR, PCT, SOFA score and APACHEⅡscore in predicting prognosis of the patients with septic shock.Results:A total of 998 patients with sepsis were enrolled, including 642 males and 356 females; with (59.56±13.22) years old. There were 478 patients with septic shock, among which 180 died and 298 survived during the 28-day observation. ① Compared with the sepsis group, the age of the sepsis shock group was significantly higher (years old: 60.49±12.31 vs. 58.72±13.97), APACHEⅡ score, SOFA score, Lac, PCT and 24 h Lac increased [APACHEⅡ: 24.57±7.04 vs. 19.37±6.93, SOFA: 7.78±3.31 vs. 4.38±3.42, Lac (mmol/L): 3.00 (1.70, 5.00) vs. 1.40 (1.00, 2.30), PCT (μg/L): 0.05 (0.00, 4.00) vs. 0.00 (0.00, 1.10), 24-hour Lac (mmol/L): 2.60 (1.60, 4.40) vs. 1.40 (1.00, 2.20)], and the 28-day mortality was significantly higher [41.63% (199/478) vs. 19.42% (101/520)], with significant statistic differences (alln P < 0.05).② Compared with the survival group, APACHEⅡ score, SOFA score, Lac, 24-hour Lac significantly increased in the septic shock death group, and 24-hour LCR decreased [APACHEⅡ: 26.19±6.52 vs. 22.25±6.07, SOFA: 9.07±2.90 vs. 7.50±3.10, Lac (mmol/L): 3.80 (2.50, 5.10) vs. 2.80 (2.00, 3.90), 24-hour Lac (mmol/L): 3.20 (2.20, 5.60) vs. 2.10 (1.60, 3.30), 24-hour LCR: 1.43 (-37.50, 30.77)% vs. 16.67 (0.00, 33.98)%, all n P < 0.05]. In assessment of organ function, central venous pressure (CVP) and oxygenation index (PaO n 2/FiOn 2) were lower in death group [CVP (mmHg; 1 mmHg = 0.133 kPa): 5.00 (2.00, 8.00) vs. 6.00 (2.00, 9.00), PaOn 2/FiOn 2 (mmHg): 184.21±84.57 vs. 199.20±86.98], alanine aminotransferase (ALT) and serum creatinine (SCr) increased [ALT (U/L): 376.56±41.43 vs. 104.17±14.10, SCr (μmol/L): 213.53±8.06 vs. 181.91±5.03], with significant statistic differences (alln P < 0.05).③ Multivariate Logistic regression analysis showed that PaO n 2/FiOn 2, SCr, Lac and SOFA were independent risk factors of prognosis in septic shock [PaOn 2/FiOn 2: odds ratio (n OR) = 0.997, 95% confidence interval (95%n CI) was 0.996-0.999, n P = 0.001; SCr: n OR = 1.001, 95%n CI was 1.000-1.002, n P = 0.041; Lac:n OR = 0.925, 95%n CI was 0.871-0.982, n P = 0.011; SOFA: n OR = 1.178, 95n %CI was 1.110-1.251, n P = 0.000]. ROC curve analysis showed that SOFA, SOFA+APACHEⅡ, Lac+24-hour LCR+PCT+SOFA+APACHEⅡcould predict mortality in septic shock patients, and the area under the ROC curve (AUC) was 0.769 (95%n CI was 0.740-0.798), 0.787 (95%n CI was 0.759-0.815), 0.800 (95%n CI was 0.773-0.827), respectively. The joint of the five indicators, Lac, 24-hour LCR, PCT, SOFA and APACHEⅡ has the largest AUC.n Conclusions:Lac is an independent risk factor for death in patients with septic shock, however, the prognosis cannot be predicted. Comprehensive analysis of LCR, PCT, SOFA, APACHEⅡand the clinical organ functions are required for analysis.