Distribution of pathogenic fungi,related risk factors and establishment of a prediction model for pulmonary fungal infection after heart valve replacement surgery

SUN Chengjun; LYU Xiangni; ZHANG Ting; LI Meng; WANG Li; ZHANG Yi

doi:10.3969/j.issn.1673-3827.2025.05.012

Abstract：Objective To analyze the distribution of pathogenic fungi and related risk factors of pulmonary fungal in-fection after heart valve replacement(HVR),establish a prediction model and verify its predictive efficacy.Methods A total of 23 patients with valvular heart disease who had secondary pulmonary fungal infection after HVR surgery in our hospital from January 2020 to January 2024 were retrospectively included in the infected group(n=23).In addition,pa-tients with valvular heart disease who had no complications after HVR surgery during the same period were included as the non-infected group(n=41).The clinical data of the two groups were collected,and the distribution of pathogenic fungi was analyzed.The logistic regression model was used to analyze the influencing factors of pulmonary fungal infection after HVR surgery,and a prediction model was constructed.Results A total of 32 strains of fungi were detected in 23 patients with postoperative pulmonary fungal infection,and Candida albicans was the most common,accounting for as high as 37.50%.Multivariate logistic regression showed that hypoalbuminemia,arteriovenous intubation,reintubation,duration of antibacterial drug use,length of hospital stay,and level of tumor necrosis factor(TNF-α)were important influencing factors for pulmonary fungal infection after HVR surgery(P<0.05);lymphocyte count(LYM)was a protective factor(P<0.05).Receiver operating characteristic(ROC)curve analysis showed that there were significant differences in the evaluation of pulmonary fungal infection after HVR surgery by the above factors and combined prediction(P<0.05);a-mong them,the area under the curve(AUC)of the combined prediction for evaluating fungal infection after HVR surgery was 0.967,with a 95%confidence interval of 0.907-1.000,a sensitivity of 0.957,and a specificity of 0.976.The decision tree model showed that the duration of antibacterial drug use at the first layer was the most important predictor,and the classification accuracy of the model was 96.40%.The Pearson correlation coefficient showed that pulmonary fungal infec-tion after HVR surgery was significantly positively correlated with hypoalbuminemia,arteriovenous intubation,reintuba-tion,duration of antibacterial drug use,length of hospital stay and level of TNF-α(P<0.05),and significantly negatively correlated with LYM(P<0.05).Conclusion The risk factors for pulmonary fungal infection after HVR surgery include hypoalbuminemia,arteriovenous intubation,reintubation,duration of antibacterial drug use,length of hospital stay,and level of TNF-α;the protective factor is LYM.Therefore,during the perioperative period,close attention should be paid to patients with the above risk factors to prevent postoperative pulmonary fungal infection.