目的:探讨血清学生物标志物的动态变异性与脓毒症患者死亡风险的相关性,构建基于血清学生物标志物的预测模型并评估其对脓毒症死亡风险的预测效能.方法:纳入2022年10月-2023年12月在海南医学院第一附属医院接受治疗的脓毒症患者138例,男71例,女67例.收集患者临床基线资料,于确诊第1天、第3天检测患者血清学指标[C-反应蛋白(C-reactive protein,CRP)、降钙素原(procalcitonin,PCT)、血乳酸(lactic acid,LAC)、白介素-6(interleukin-6,IL-6)、IL-8、IL-10、可溶性分化簇 163(soluble cluster of differentiation 163,sCD163)、细胞间黏附分子-1(intercellular adhesion molecule-1,ICAM-1)],计算血清学指标变异率;采用序贯器官衰竭评分(sequential organ failure score,SOFA评分)对患者病情严重程度进行评估;随访28 d,根据患者28 d死亡情况分为死亡组(38例)和存活组(100例).对比两组患者临床实验室资料,logistic回归分析脓毒症死亡风险的独立相关因素.使用R语言构建基于独立相关因素的列线图模型,绘制校正曲线和受试者工作特征曲线,评估模型在早预测脓毒症患者死亡风险的效能.结果:28 d随访结果显示,138例患者中,38例死亡,死亡率为27.54％.据此分组后对比临床资料显示,死亡组SOFA评分、IL-6变异率、IL-8变异率、LAC变异率、sCD163变异率高于存活组,差异有统计学意义(P＜0.05);logistic回归分析显示,IL-6变异率(OR=1.444)、SOFA评分(OR=1.559)、LAC变异率(OR=1.295)、sCD163变异率(OR=1.208)是脓毒症患者28 d死亡的独立相关因素(P＜0.05).使用R语言基于独立相关因素绘制列线图模型,模型的C-指数为0.979(95％CI:0.965～0.998),校正曲线与理想曲线走形接近;列线图模型在早期预测脓毒症患者死亡的AUC为0.974,灵敏度为92.11％,特异度为92.00％.结论:通过分析血清学标志物的早期变异率能够准确预测脓毒症患者死亡风险,提高临床医生预警能力.
Diagnostic efficacy of dynamic variability of serum markers in predicting the risk of death in sepsis patients
Objective:To explore the correlation between the dynamic variability of serum biomarkers and the risk of mortality in patients with sepsis,to construct a predictive model based on serum biomarkers,and to evalu-ate its predictive efficacy for the risk of death in sepsis.Methods:A total of 138 patients with sepsis treated in our hospital from October 2022 to December 2023 were included.Clinical baseline data of the patients were collected.On the first and third day of diagnosis,the following serum markers were measured:C-reactive protein(CRP),procalcitonin(PCT),lactic acid(LAC),interleukin-6(IL-6),IL-8,IL-10,soluble cluster of differentiation 163(sCD163),and Intercellular adhesion molecule-1(ICAM-1).The variability rate of the serum markers was calcu-lated.The severity of the patients'conditions was assessed using the sequential organ failure score(SOFA).The patients were followed up for 28 days,and based on their 28-day mortality,they were divided into a deceased group(38 cases)and a survival group(100 cases).The clinical and laboratory data of the two groups were com-pared,and logistic regression analysis was performed to identify independent factors associated with the risk of death from sepsis.A nomogram model based on the independent factors was constructed using the R program-ming language,and calibration curves and receiver operating characteristic curves were plotted to evaluate the model's efficacy in early prediction of the risk of death in sepsis patients.Results:After a 28-day follow-up,38 out of 138 patients died,resulting in a mortality rate of 27.54％.After grouping,the clinical data showed that the SOFA score,IL-6 variability rate,IL-8 variability rate,LAC variability rate,and sCD163 variability rate were significantly higher in the deceased group compared to the survival group(P＜0.05).logistic regression analysis revealed that the IL-6 variability rate(OR=1.444),SOFA score(OR=1.559),LAC variability rate(OR=1.295),and sCD163 variability rate(OR=1.208)were independent factors associated with 28-day mortality in pa-tients with sepsis(P＜0.05).Based on the independent factors,a nomogram model was constructed using the R programming language.The model's C-index was 0.979(95％CI:0.965-0.998),and the calibration curve close-ly resembled the ideal curve.The AUC of the nomogram model for early prediction of death in sepsis patients was 0.974,with a sensitivity of 92.11％and a specificity of 92.00％.Conclusion:Analysis of the early variability of serum markers can accurately predict the risk of death in patients with sepsis,enhancing the clinical physician's a-bility to provide early warnings.
NETL
NSTL
万方数据
