Construction and validation of risk early warning model for elderly patients with chronic obstructive pulmonary disease complicated with ventilator-associated pneumonia
Objective To analyze the risk factors of ventilator-associated pneumonia(VAP)in elderly patients with chronic obstructive pulmonary disease(COPD),and to construct and verify the risk warning model.Methods A total of 170 elderly COPD patients admitted to the Department of Critical Care Medicine of our hospital from August 2020 to December 2022 were retrospectively selected into the modeling group,and were divided into the VAP group and the non-VAP group according to whether the patients were complicated with VAP during hospitalization.Univariate and logistic regression analysis were used to analyze the inde-pendent influencing factors of VAP,and the risk early warning model was constructed according to the inde-pendent influencing factors.The receiver operating characteristic(ROC)curve was used to evaluate the dis-crimination of the model,and the Hosmer-Lemeshow test was used to judge the goodness of fit of the model.Another 73 elderly patients with COPD in the same period were selected as the validation group for external verification of the model.Results In the modeling group,there were 41 cases of the 170 elderly COPD pa-tients were complicated with VAP,with an incidence of 24.1％.The results of univariate and multivariate a-nalysis showed that history of diabetes[odds ratio(OR)=14.821,95％confidence interval(95％CI)2.462-89.232],combined use of antibiotics(OR=6.352,95％CI 1.391-29.008),indwelling gastric tube(OR=5.454,95％CI 1.213-24.521),history of aspiration(OR=7.745,95％CI 1.569-38.224),mechanical venti-lation time＞5 d(OR=8.014,95％CI 2.026-31.695),hospitalization time≥ 14 d(OR=6.494,95％CI 1.371-30.755)and acute physiology and chronic health evaluation Ⅱ(APACHE Ⅱ)score＞18(OR=16.714,95％CI 3.326-83.983)were independent risk factors for VAP in elderly patient with COPD.The ROC results of the risk warning model based on independent influencing factors showed that the area under the curve(AUC)was 0.901(95％CI 0.843-0.958,P＜0.001),the sensitivity was 79.5％,and the specificity was 91.3％.The sensitivity,specificity and total accuracy of model validation were 80.0％,92.5％and 89.0％,respectively.Conclusion The history of diabetes,combined use of antibiotics,indwelling gastric tube,his-tory of aspiration,mechanical ventilation time＞5 d,hospitalization time≥ 14 d and APACHE Ⅱ score＞18 are in-dependent risk factors for VAP in elderly patients with COPD.The risk early warning model constructed by the above factors in this study has been verified to have good predictive performance and has certain clinical application value.
国家科技期刊平台
NSTL
万方数据
