AUTHOR=Wang Qian , Ma Hui , Jiang Qiang , Guo Lubo 

TITLE=Development of predictive nomogram for clinical use of special-grade antimicrobial agents in patients with diabetes foot infections

JOURNAL=Frontiers in Endocrinology

VOLUME=Volume 16 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/endocrinology/articles/10.3389/fendo.2025.1578767

DOI=10.3389/fendo.2025.1578767

ISSN=1664-2392

ABSTRACT=ObjectiveTo develop a predictive model to quantify the possibility of special-grade antimicrobial agents (SGAs) usage in patients with diabetes foot infections (DFIs), providing reference and guidance for clinical practice.MethodsThis is a cross-sectional study of 328 type 2 diabetes patients with DFIs. General clinical characteristics and biochemical indicators were extracted from the Hospital Information System (HIS) of Jinan Central Hospital in Shandong Province, China. Logistic regression analysis was performed to select predictors, and the nomogram was established based on selected viables visually. Then, the receive operating characteristic (ROC) curve, the calibration curve and the decision curve analysis (DCA) were used to evaluate the performance of this prediction model.Results5 predictors were selected by univariate analysis from 21 variables, including duration of hospitalization, Neutrophil, DBIL, ALB and Wagner grade. The multivariate logical regression analysis illustrated that these 5 factors were independent risk factors for SGAs usage in patients with DFIs. The nomogram model developed by these 5 risk predictors exhibited good prediction ability, as shown by the area under curve (AUC) of ROC curve was 0.884 in the training set and 0.825 in the validation set. Calibration curve showed a good calibration degree of the predictive nomogram model. Moreover, DCA curve showed that the nomogram exhibited greater clinical application values when the risk threshold was between 3% and 63%.ConclusionOur novel nomogram model showed that duration of hospitalization, Neutrophil, DBIL, ALB and Wagner grade were the independent risk factors of SGAs usage in patients with DFIs. This prediction model behaved a great accurate value and provide reference of SGAs usage in clinic. Further validations are still needed to evaluate and improve the performance of this model.