急性农药中毒院内肺部感染预测模型的构建  

Construction of nosocomial pulmonary infection prediction model for acute pesticide poisoning

作　　者：赵红 邢晶晶 潘惠卿 陈蒙 魏娜 唐晓磊 叶胜 ZHAO Hong;XING Jingjing;PAN Huiqing;CHEN Meng;WEI Na;TANG Xiaolei;YE Sheng(Emergency Department,the Second Affiliated Hospital of Wannan Medical College,Wuhu,Anhui 241000,P.R.China;Graduate School,Wannan Medical College,Wuhu,Anhui 241000,P.R.China;Emergency Department,the First Affiliated Hospital of Wannan Medical College,Wuhu,Anhui 241000,P.R.China;Translational Medicine Centre,the Second Affiliated Hospital of Wannan Medical College,Wuhu,Anhui 241000,P.R.China)

机构地区：[1]皖南医学院第二附属医院急诊医学科,安徽芜湖241000 [2]皖南医学院研究生学院,安徽芜湖241000 [3]皖南医学院第一附属医院急诊医学科,安徽芜湖241000 [4]皖南医学院第二附属医院转化医学中心,安徽芜湖241000

出　　处：《华西医学》2023年第11期1655-1661,共7页West China Medical Journal

摘　　要：目的 探寻急性农药中毒院内肺部感染的危险因素并构建风险预测模型。方法 回顾性分析2021年1月1日—2023年9月30日在皖南医学院第一附属医院及皖南医学院第二附属医院急诊医学科住院诊治的急性农药中毒患者的临床资料。根据患者在院期间是否发生肺部感染分为肺部感染组及非肺部感染组。采用多因素logistic回归分析急性农药中毒患者院内肺部感染的独立危险因素并构建风险预测模型(列线图),采用受试者操作特征曲线分析列线图及独立预测因子对院内肺部感染的预测效能,采用校准曲线及决策曲线分析评估模型的区分度及临床应用价值。结果 共纳入急性农药中毒患者189例。其中,男98例(51.85%),女91例(48.15%);平均年龄为(58.12±18.45)岁;院内出现肺部感染患者36例(19.05%)。多因素logistic回归分析显示,年龄[比值比(odds ratio, OR)=1.030,95%置信区间(confidence interval, CI)(1.001,1.060),P=0.040]、2型糖尿病[OR=2.770,95%CI(1.038,7.393),P=0.042]、缺血性脑血管病[OR=3.213,95%CI(1.101,9.376),P=0.033]、白细胞计数[OR=1.080,95%CI(1.013,1.152),P=0.019]、日常生活能力评分[OR=0.981,95%CI(0.965,0.998),P=0.024]是急性农药中毒院内肺部感染的独立预测因子。基于上述因素构建的列线图预测急性农药中毒患者院内肺部感染的曲线下面积为0.813(P<0.001),校准曲线对列线图进行拟合优度检测提示预测概率与实际发生概率较为一致(P=0.912),决策曲线提示列线图有较好的临床应用价值。结论 年龄、日常生活能力评分、2型糖尿病、缺血性脑血管病和白细胞计数是急性农药中毒院内肺部感染的独立预测因子,基于此构建的列线图具有较好的区分度及一致性,可为临床医务人员及早识别急性农药中毒患者院内肺部感染及早期干预提供依据。Objective To explore the risk factors of nosocomial pulmonary infection in acute pesticide poisoning.Methods The clinical data of patients with acute pesticide poisoning hospitalized in the Emergency Department of the First Affiliated Hospital of Wannan Medical College and the Second Affiliated Hospital of Wannan Medical College between January 1,2021 and September 30,2023 were retrospectively analyzed.Patients were divided into pulmonary infection group and non-pulmonary infection group according to whether they had pulmonary infection during hospital.Multiple logistic regression was used to analyze the independent risk factors of nosocomial pulmonary infection in patients with acute pesticide poisoning,and a risk prediction model(nomogram)was constructed.The predictive efficacy of nomogram and independent predictors in nosocomial pulmonary infection were analyzed by using the receiver operating characteristic curve.Calibration curve and decision curve were used to evaluate the differentiation and clinical application value of the model.Results A total of 189 patients with acute pesticide poisoning were included in the study,with an average age of(58.12±18.45)years old,98 males(51.85%)and 91 females(48.15%).There were 36 cases(19.05%)of pulmonary infection.Multiple logistic regression analysis showed that age[odds ratio(OR)=1.030,95%confidence interval(CI)(1.001,1.060),P=0.040],type 2 diabetes mellitus[OR=2.770,95%CI(1.038,7.393),P=0.042],ischemic cerebrovascular disease[OR=3.213,95%CI(1.101,9.376),P=0.033],white blood cell count[OR=1.080,95%CI(1.013,1.152),P=0.019],activities of daily living score[OR=0.981,95%CI(0.965,0.998),P=0.024]were independent predicting factors for nosocomial pulmonary infection in acute pesticide poisoning.The area under the curve of nosocomial pulmonary infection in patients with acute pesticide poisoning predicted by nomogram based on the above factors was 0.813(P<0.001).The calibration curve showed that the prediction probability was consistent with the actual occurrence probability

关 键 词：急性农药中毒 院内肺部感染 年龄 日常生活能力评分 预测效能 

分 类 号：R595.4[医药卫生—内科学] R563.1[医药卫生—临床医学]