# Establishment and evaluation of a novel tool based on inflammation-nutrition derived biomarkers for early diagnosis of diabetic foot ulcers

## TL;DR

A new tool using inflammation and nutrition biomarkers helps predict diabetic foot ulcers in Chinese adults, showing strong predictive power.

## Contribution

A novel predictive model based on the neutrophil percentage-to-albumin ratio (NPAR) for early diagnosis of diabetic foot ulcers is established and validated.

## Key findings

- NPAR and other factors like age, BMI, and peripheral neuropathy are significant predictors of diabetic foot ulcers.
- The model showed good predictive discrimination with ROC AUC values of 0.892 and 0.877 in training and validation cohorts.
- Calibration curves confirmed the model's accuracy, and decision curve analysis showed clinical usefulness.

## Abstract

The study aimed to investigate the relationship between neutrophil percentage-to-albumin ratio (NPAR) and diabetic foot ulcer (DFU) in Chinese adults, further establish a clinical predictive model, and verify its effectiveness.

We retrospectively collected and analyzed clinical data from a total of 1,002 diabetic patients at Honghui Hospital of Xi’an Jiaotong University between January 2024 and January 2026. The association between the NPAR and DFU risk was assessed using a logistic regression. Moreover, the nonlinear relationship was further characterized through smooth curve fitting and generalized additive model analysis. The predictors were identified via the least absolute shrinkage and selection operator and multivariate logistic analysis. The discrimination and calibration of the nomogram were validated by receiver operating characteristic (ROC) curve and calibration curve. Decision curve analysis (DCA) was used to evaluate clinical usefulness and net benefits of the prediction model.

The multivariate logistic regression analysis demonstrated that NPAR (odds ratio [OR] =1.303, 95% confidence intervals [CI]: 1.212–1.402), age (OR = 1.058, 95% CI: 1.032–1.083), sex (female vs. male, OR = 0.475, 95% CI: 0.281–0.802), body mass index (20–25 kg/m² vs. <20 kg/m², OR = 0.184, 95% CI: 0.094–0.359; ≥25 kg/m² vs. <20 kg/m², OR = 0.445, 95% CI: 0.252–0.788), smoke (yes vs. no, OR = 1.735, 95% CI: 1.023–2.941), peripheral vascular disease (yes vs. no, OR = 5.522, 95% CI: 3.428–8.896), peripheral neuropathy (yes vs. no, OR = 6.914, 95% CI: 4.114–11.618), and hemoglobin (OR = 0.981, 95% CI: 0.967–0.996) were risk-associated indicators for DFU. The calibration curves for the training and validation cohorts both revealed good agreement. In addition, the area under the ROC curve values in the training and validation cohorts were 0.892 (95% CI: 0.864–0.919) and 0.877 (95% CI: 0.831–0.922), respectively, indicating good predictive discrimination. The DCA showed that the nomogram could provide clinical usefulness and net benefit.

This study indicated a positive relationship between DFU risk and the integrated inflammatory-nutritional status represented by NPAR in the Chinese diabetic population. The DFU prediction model incorporating NPAR was validated for its effectiveness and clinical utility, providing evidence for the potential of NPAR as a risk-associated indicator measured at DFU diagnosis.