肺源性脓毒症致病菌耐药性危险因素及早期预测分析

Abstract: Objective To explore the risk factors for drug resistance of pathogenic bacteria in patients with pulmonary sepsis and conduct early prediction analysis, providing reference for clinical diagnosis and treatment decisions. Methods A retrospective collection was made of 267 patients with sepsis caused by pathogenic bacteria identified through sputum culture, admitted to the emergency department of Jinling Hospital, Nanjing Medical University from January 2019 to December 2023. These patients were divided into multidrug-resistant and non-multidrug-resistant groups based on antimicrobial susceptibility test results. Binary Logistic stepwise regression was used to analyze the related influencing factors. The receiver operating characteristic (ROC) curve and the area under the ROC curve (AUC) were used to analyze the early predictive value of D-dimer (D-D), procalcitonin (PCT), and platelet count (PLT) for multidrug resistance. Results The white blood cell count (WBC), neutrophil count (NEUT), and PCT in the multidrug-resistant group were lower than those in the non-multidrug-resistant group, while PLT and D-D were higher (P＜0.05). Decreased PCT and PLT and increased D-D suggest a higher possibility of drug-resistant pathogenic bacteria in patients with pulmonary sepsis. The AUC for D-D in predicting multidrug resistance was 0.826, for PCT was 0.824, for PLT was 0.644, and for the combined prediction of PCT and D-D was 0.889. Conclusion The combined detection of D-D and PCT has good application value in the early prediction of drug resistance in pathogenic bacteria of pulmonary sepsis, which is superior to the single detection of either. In clinical practice, when dealing with patients with pulmonary sepsis, the optimal cut off values of D-D and PCT can be used for early prediction of drug resistance, early intervention, and improvement of therapeutic efficacy, providing an Objective basis for clinical treatment and management.