Distribution of Pathogenic Bacteria in Patients with Lung Infection after Radical Operation for Lung Cancer and Construction of Early Risk Prediction Model
Objective To analyze the distribution of pathogenic bacteria in patients with lung infection after radical op-eration for lung cancer,and to construct an early risk prediction model.Methods 72 patients with lung cancer underwent radical operation were enrolled,and patients were classified into 2 groups according to the postoperative presence or absence of postopera-tive lung infection,infection group(n=18)and non-infection group(n=54).Analyze the distribution of pulmonary infection pathogens in patients in the infection group.Compare the clinical data of 2 groups of patients,and use a multivariate logistic re-gression model to screen out independent risk factors for postoperative infection in patients.Create a column chart prediction mod-el based on independent risk factors,and verify the predictability and accuracy of the column chart.Results 72 patients had 18 cases of pulmonary infection,with an infection rate of 25.00％.A total of 35 strains of pathogenic bacteria were isolated from 18 infections,including 22 strains(62.85％)of Gram negative bacteria,mainly Pseudomonas aeruginosa,Klebsiella pneumoniae,and Escherichia coli;9 strains(25.71％)of Gram positive bacteria,mainly Staphylococcus aureus and Staphylococcus hemolyticus;There were 4 strains of fungi(11.42％),mainly Candida albicans.Compared with the uninfected group,the infected group had a higher proportion of patients with age ≥ 60 years,a history of smoking,preoperative FEV1 ≤80％,surgical time ≥150 minutes,multi lobectomy,and intraoperative bleeding ≥200 mL(P＜0.05);Age,smoking history,preoperative FEV1,surgical time,resec-tion range,and intraoperative bleeding were all independent influencing factors for postoperative pulmonary infection(P＜0.05);The column chart results indicated that age ≥60 years old was 52 points,smoking history was 65 points,preoperative FEV180％was 68 points,surgical time ≥150 minutes was 49 points,multi lobectomy was 78 points,and intraoperative bleeding ≥200 mL was 70 points.The accuracy and differentiation rate of predictive value of the mode was higher.Conclusion The pathogens of lung infection after radical resection for lung cancer are mainly Gram-negative bacteria,furthermore,age,smoking history,preoper-ative FEV1,duration of surgery,extent of resection,and intraoperative blood loss are influencing factors for postoperative lung in-fection,and the accuracy and differentiation rate of predictive value of the mode is higher.
Radical surgery for lung cancerLung infectionPathogen distributionRisk factorsPredictive model
NETL
NSTL
万方数据
