糖尿病足患者溃疡创面感染病原菌及其预测模型构建

Pathogens of Ulcer Wound Infection in Patients with Diabetic Foot and Construction of Its Prediction Model

摘要: 目的：分析糖尿病足患者并发溃疡创面感染病原菌分布特征及其感染的危险因素，并构建预测模型。方法：回顾性收集2021年1月~2022年12月安徽医科大学第二附属医院整形与创面修复外科收治的56例糖尿病足发生溃疡感染患者(感染组)及10例未发生溃疡感染患者(未感染组)的临床资料。分析糖尿病足并发溃疡感染病原菌分布，多因素Logistic回归分析糖尿病足并发溃疡感染的危险因素，并构建预测模型，其预测价值采用受试者工作特征(receiver operating characteristic, ROC)曲线分析。结果：66例糖尿病足患者中有56例发生创面感染，56例糖尿病足并发溃疡感染患者共检出92株病原菌，主要感染病原菌为大肠埃希菌、铜绿假单胞菌、金黄色葡萄球菌、无乳链球菌、阿萨希丝孢酵母菌。多因素Logistic分析结果显示，合并糖尿病肾病、合并骨髓炎均为糖尿病足并发溃疡感染的危险因素(P < 0.05)。据此构建的模型预测糖尿病足并发溃疡感染的曲线下面积(area under curve, AUC)为0.740，敏感度为69.87%，特异度为90.20%。结论：糖尿病足患者并发溃疡感染主要感染病原菌种类包括大肠埃希菌、铜绿假单胞菌、金黄色葡萄球菌、无乳链球菌、阿萨希丝孢酵母菌，主要危险因素包括合并糖尿病肾病、合并骨髓炎。且据此建立危险因素模型对糖尿病足患者发生创面感染具有中等预测价值，临床可据此给予伴有以上情况的患者针对性干预措施，从而降低糖尿病足患者创面感染的发生风险。

Abstract: Objective: Analyze the distribution characteristics of pathogenic bacteria in ulcerative wound infections among diabetic foot patients and their risk factors for infection, and construct a predictive model. Methods: A retrospective collection of clinical data from January 2021 to December 2022 on 56 patients with diabetic foot ulcers and infections (the infected group) and 10 patients without ulcer infections (the non-infected group) treated by the Department of Plastic Surgery and Wound Repair Department at the Second Affiliated Hospital of Anhui Medical University. Analyzing the distribution of pathogenic bacteria in diabetic foot ulcer complications, a multifactorial Logistic regression analysis identifies the risk factors for infections associated with diabetic foot ulcers, and constructs a predictive model. The predictive value of this model is assessed using the receiver operating characteristic (ROC) curve analysis. Result: Among 66 patients with diabetic foot ulcers, 56 developed wound infections. From these 56 infected wounds, a total of 92 pathogenic strains were isolated. The predominant pathogens were Escherichia coli, Pseudomonas aeruginosa, Staphylococcus aureus, Streptococcus agalactiae, and Trichosporon asahii. Multivariate logistic regression analysis revealed that concomitant diabetic nephropathy and osteomyelitis were independent risk factors for ulcer infection in diabetic foot (P < 0.05). A predictive model incorporating these factors demonstrated an area under curve (AUC) of 0.740, with a sensitivity of 69.87 % and a specificity of 90.20 %. Conclusion: The main pathogens responsible for ulcer infection in diabetic foot patients are Escherichia coli, Pseudomonas aeruginosa, Staphylococcus aureus, Streptococcus agalactiae, and Trichosporon asahii. Diabetic nephropathy and osteomyelitis are key risk factors. The constructed risk-factor model offers moderate predictive value for wound infection in this population. Clinicians can use this model to deliver targeted interventions for high-risk patients, thereby reducing the incidence of diabetic foot ulcer infections.