机构地区:[1]浙江中医药大学第四临床医学院,310053 [2]杭州市老年病医院,310022 [3]杭州市萧山区第一人民医院,311202 [4]浙江大学医学院,310009 [5]西湖大学医学院附属杭州市第一人民医院,310006
出 处:《浙江临床医学》2025年第1期85-88,共4页Zhejiang Clinical Medical Journal
基 金:杭州市卫生科技计划重大项目(Z20230132)。
摘 要:目的分析重症医学科(ICU)长期机械通气(PMV)患者血流感染病原菌分布、药物敏感性及死亡危险因素,并构建临床应用预测模型。方法回顾性分析2019年9月至2021年9月ICU PMV血流感染71例,采用单因素和多因素Logistic回归分析方法,确定影响患者死亡的因素,构建预测列线图。通过受试者操作特征(ROC)曲线、校准曲线、决策曲线分析(DCA)和临床影响曲线(CIC)对列线图进行评估。结果71例患者血液中共培养出病原菌71株:革兰阳性菌55株(77.47%);革兰阴性菌15株(21.12%);真菌1株(1.41%);前3位病原菌为耳葡萄球菌、头葡萄球菌及金黄色葡萄球菌,构成比分别为23.94%、15.49%、14.08%。单因素分析显示,合并肾功能不全、压疮、气管切开对患者预后结果差异有统计学意义(P<0.05)。多因素Logistic回归分析显示,患者合并肾功能不全(OR=7.942;95%CI:1.591~39.634;P=0.012)、压疮(OR=7.202;95%CI:2.582~56.869;P=0.025)、气管切开(OR=12.116;95%CI:1.275~40.699;P=0.002),是导致ICU PMV患者血流感染不良预后的独立危险因素。以此构建列线图模型预测死亡曲线下面积(AUC)为0.825。校准曲线、DCA与CIC显示足够的拟合效应和理想的净效益。结论合并肾功能不全、压疮、气管切开是ICU PMV患者发生血流感染患者死亡危险因素,革兰阳性菌为ICU PMV患者发生血流感染主要病原菌。基于肾功能不全、压疮、气管切开构建的列线图预测模型在全因早期死亡和肿瘤特异性早期死亡预测中有较好预测效果和临床应用价值。Objective To analyze the distribution of pathogenic bacteria,drug sensitivity,and risk factors for mortality in patients undergoing long-term mechanical ventilation(PMV)in the intensive care unit(ICU),and construct a clinical application prediction model.Methods A retrospective analysis was conducted on 71 ICU patients with PMV bloodstream infections from September 2019 to September 2021.Univariate and multivariate Logistic regression analysis methods were used to determine the factors affecting patient mortality and construct a predictive nomogram.The nomogram was evaluated using receiver operating characteristic(ROC)curves,calibration curves,decision curve analysis(DCA),and clinical impact curves(CIC).Results A total of 71 patients with bloodstream infections were included in this study.A total of 71 strains of pathogens were cultured from blood samples,including 15(21.12%)Gram-negative bacteria,55(77.47%)Gram-positive bacteria,and 1(1.41%)fungus.The top three pathogens were Staphylococcus epidermidis,Staphylococcus aureus,and Staphylococcus aureus,accounting for 23.94%,15.49%,and 14.08%,respectively.Univariate analysis revealed that comorbidities such as renal insufficiency,pressure ulcers,and tracheostomy were statistically significant prognostic factors for patients(P<0.050).Multivariate Logistic regression analysis incorporating these factors showed that comorbidities such as renal insufficiency(OR=7.942;95%CI:1.591~39.634;P=0.012),pressure ulcers(OR=7.202;95%CI:2.582~56.869;P=0.025),and tracheostomy(OR=12.116;95%CI:1.275~40.699;P=0.002)were independent risk factors for death in patients with bloodstream infections,and a predictive nomogram for death was constructed based on these factors.The area under the curve(AUC)of the nomogram was 0.825,indicating good predictive value.The calibration curve,DCA,and CIC also showed good fit and ideal net benefits,indicating good clinical value.Conclusion Renal insufficiency,pressure ulcers,and tracheostomy are risk factors for death in patients with bloodstream infecti
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...