[Influencing factors analysis and prediction model establishment of toe-amputation in patients with diabetic foot]

Objective: To analyze the influencing factors of toe-amputation in diabetic foot patients and construct a predictive model. Methods: The clinical data of 437 diabetic foot patients who were hospitalized in Air Force Medical Center from January 2017 to January 2024 were retrospectively analyzed, including 327 males and 110 females, with a median age[M(Q₁,Q₃)] of 63.0 (55.0, 69.0) years.The patients were divided into amputation group(n=344) and non-amputation group (n=93) according to whether toe-amputation surgery was performed within 3 months. The differences of clinical indicators between the two groups were analyzed.The patients were randomly divided into the training set and the validation set in a 7∶3 ratio using R 4.2. Multivariate logistic regression model was used to screen the influencing factors of toe-amputation in diabetic foot patients, and the nomogram of toe amputation prediction model was drawn. The predictive ability, accuracy and clinical applicability of the model were assessed using the area under the curve (AUC) of the receiver operating, calibration curve and decision curve analysis(DCA), respectively. Results: The patients in the toe amputation group had higher proportions of diabetes duration, lower extremity arterial disease (LEAD), white blood cell count, total bilirubin, fibrinogen, erythrocyte sedimentation rate, C-reactive protein level, and positive bacterial cultures in wound secretions compared to those in the non-toe amputation group. Conversely, the levels of high-density lipoprotein cholesterol, blood uric acid, the internal diameter and blood flow of dorsalis pedis artery were lower in the toe amputation group than those in the non-toe amputation group (all P<0.05). Multivariate logistic regression model analysis indicated that LEAD (OR=4.19,95%CI:1.88-9.34), blood uric acid (OR=0.99,95%CI:0.99-0.99), fibrinogen (OR=1.86,95%CI:1.19-2.91), and wound secretion culture positive for staphylococcus aureus (OR=5.77,95%CI:2.72-12.25) or non-staphylococcus aureus infection (OR=4.74,95%CI:1.43-15.77), were influencing factors of diabetic foot toe-amputations (all P<0.05). The prediction model of toe amputation in patients with diabetes foot was established using the above parameters. In the training set, the AUC, sensitivity and specificity of the prediction model for predicting toe amputation in diabetic foot patients were 0.86(95%CI:0.80-0.91)、90% and 72%, respectively. In the validation set, the AUC, sensitivity and specificity of the prediction model for predicting toe amputation in diabetic foot patients were 0.85(95%CI:0.78-0.92), 77% and 71%, respectively. The calibration curves of the training set and the validation set indicates that the model has good calibration ability. The DCA curve of the training set and the validation set suggests that the predictive model has good clinical applicability. Conclusions: LEAD, blood uric acid, fibrinogen, and positive bacterial culture of wound secretion are the influencing factors of diabetes toe amputation. The nomogram model including the above factors can evaluate the risk of diabetic foot amputation more intuitively.