Objective To investigate the relationships of tumor necrosis factor-α (TNF-α) and glycated hemoglobin (HbA1c) with the occurrence of painful diabetic neuropathy (PDN) in patients with type 2 diabetes mellitus (T2DM).

Methods A total of 225 patients with T2DM were enrolled and divided into PDN group (52 patients) and non-PDN group (173 patients) based on the presence of PDN. Clinical data of the two groups were collected and compared. A multivariate Logistic regression model was used to analyze risk factors for PDN, and a receiver operating characteristic (ROC) curve was plotted to assess the predictive value of TNF-α and HbA1c for PDN in T2DM patients.

Results Patients in the PDN group had a longer duration of T2DM, higher levels of fasting blood glucose (FBG), HbA1c, C-reactive protein (CRP), and TNF-α compared with the non-PDN group (P < 0.05). Multivariate Logistic regression analysis showed that a longer duration of T2DM, higher TNF-α levels, and higher HbA1c levels were independent risk factors for PDN in T2DM patients (P < 0.05, OR>1). ROC curve analysis revealed that the area under the curve for the combined prediction of PDN in T2DM patients by TNF-α and HbA1c was 0.877(95%CI, 0.826 to 0.916), which was greater than that of TNF-α and HbA1c alone 0.684(95%CI, 0.619 to 0.745) and 0.764(95%CI, 0.703 to 0.818), respectively.

Conclusion High serum levels of TNF-α and HbA1c are independent risk factors for PDN in T2DM patients, and the combined detection of two biomarkers has high predictive efficacy for the risk of PDN in T2DM patients.