Epidemiology, Risk Factors, and External Validation of Predictive Models for Multidrug-Resistant Bacterial Infections
Xiangheng Dai1, Fuzhou Xu2, Weiqi Lu2
1Department of Spinal Surgery, Shaoguan First People's Hospital, Guangdong Medical University, Shaoguan, People's Republic of China.
Objective:
This study aimed to investigate the epidemiological characteristics and risk factors of multidrug-resistant bacterial (MDRB) infections in diabetic foot ulcer (DFU) and externally validate existing predictive models for MDRB, Pseudomonas aeruginosa (PSA), and methicillin-resistant Staphylococcus aureus (MRSA) infections.
Methods:
A retrospective analysis of 344 patients diagnosed with DFU identified key clinical and laboratory variables associated with outcomes. Univariate and multivariate logistic regression identified risk factors, while receiver operating characteristic (ROC) curves and calibration analyses assessed model validity.
Results:
MDRB infections were linked to longer hospitalization, lower hemoglobin, higher hs-CRP, and higher osteoporosis prevalence. Significant risk factors included prolonged hospitalization, osteomyelitis, osteoporosis, prior antimicrobial use, and respiratory rate. PSA infections correlated with myocardial infarction and peripheral vascular disease, while no significant risk factors were identified for MRSA. External validation showed poor discriminatory ability (AUC: 0.501 for MDRB, 0.505 for PSA, 0.569 for MRSA) and calibration performance, indicating poor generalizability of existing models in our cohort.
Conclusion:
MDRB infections in DFUs are associated with specific clinical characteristics and risk factors. However, existing predictive models demonstrated limited utility in our dataset, emphasizing the need for model refinement and inclusion of additional risk factors. Future research should focus on improving predictive models and developing targeted antimicrobial strategies to enhance clinical outcomes in diabetic foot infections.
