Diabetic Peripheral Neuropathy (DPN) is a common and debilitating complication of Type 2 Diabetes Mellitus (T2DM), leading to significant morbidity and reduced quality of life. This study aimed to investigated the feasibility of combining heart rate variability (HRV) and skin autofluorescence (SAF) as a novel, non-invasive approach for the detection and severity assessment of DPN in T2DM. By leveraging complementary autonomic and metabolic biomarkers, the proposed method aims to overcome limitations of conventional screening techniques in outpatient clinical settings.

This cross-sectional study included 92 patients (≥20 years, 57 male, 35 female) diagnosed with T2DM, recruited from outpatient clinics. Each patients underwent a comprehensive evaluation for DPN: Toronto Clinical Neuropathy Scoring System (TCNS), HRV analysis with emphasis on LF/HF ratio, and SAF measurement using the AGE Reader device. Statistical analyses included correlation analysis, multiple linear regression, and ROC curve analysis. They were performed to evaluate the discriminative performance of SAF and HRV in detecting DPN. Additionally, a subgroup analysis was conducted on 46 patients with HbA1c levels ≥7%. This analysis aimed to assess the model's effectiveness in patients with suboptimal glycemic control.

In a cohort of 92 participants, our analysis revealed a clear link between DPN severity and two key factors: age (r = 0.375, P < 0.001) and SAF (r = 0.232, P = 0.026). Specifically, older participants and those with elevated SAF levels tended to exhibit more severe DPN. In the subgroup 46 patients (27 male, 19 female) with suboptimal glycemic control (HbA1c ≥7%). DPN severity showed significant associations with both the LF/HF ratio (r = 0.298, P = 0.045) and SAF (r = 0.376, P = 0.010). The multiple linear regression analysis results confirmed that both LF/HF ratio (P = 0.030) and SAF (P = 0.029) were independently and significantly linked to DPN severity. Finally, we combined LF/HF ratio and SAF in a diagnostic model using ROC analysis, the outcome was compelling: the model achieved a sensitivity of 57.5% and a specificity of 100%, offering a highly accurate method for identifying DPN cases while minimizing false positives.

This integrated approach—combining LF/HF ratio and SAF measurement—may be uesd as the early screening tool of DPN, particularly those with suboptimal glycemic control, making it especially valuable in fast-paced outpatient settings.