INTRODUCTION: Microvascular disease is a common complication of type 2 diabetes mellitus (T2DM). As a typical microvascular complication, diabetic kidney disease (DKD) interacts with peripheral atherosclerosis (AS) and significantly increases the risk of adverse cardiovascular events. OBJECTIVES: To investigate the risk factors for peripheral AS in early-stage T2DM-DKD patients, and to construct an individualized nomogram prediction model. METHODS: A single-center, retrospective study was conducted involving 392 T2DM patients with early-stage DKD admitted to Peking University International Hospital from March 2015 to August 2021. According to the results of peripheral arterial ultrasound examination, they were divided into peripheral AS group (AS+DKD group) (n = 294) and DKD-only group (n = 98). Clinical data and laboratory parameters were collected, and CVAI was calculated. Univariate and multivariate logistic regression analyses were used to identify independent factors associated with peripheral AS. A nomogram model was constructed using R software, and the predictive performance of the model was evaluated by receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA). RESULTS: The incidence of peripheral AS in early-stage T2DM-DKD patients was 75.00%. Univariate analysis showed that the AS+DKD group had higher Age, longer duration of diabetes, higher CVAI, HDL-C and sCr level, as well as lower BMI, DBP, fasting C-peptide (FCP), 2-hour C-Peptide, TC, TG, LDL-C, ALT, AST, and eGFR levels (all P < 0.05). Multivariate logistic regression analysis revealed that elevated CVAI (OR = 1.034, 95% CI: 1.010- 1.058), increased Age (OR = 1.054, 95% CI: 1.018- 1.090), and decreased FCP (OR = 0.771, 95% CI: 0.608- 0.978) were independent risk factors for peripheral AS in early-stage T2DM-DKD patients (all P < 0.05). The AUC values for the three models, included Age only, FCP+ Age, and CVAI+ FCP+ Age, were 0.814 (95% CI: 0.765-0.863), 0.825 (95% CI: 0.776-0.874), and 0.837 (95% CI: 0.790-0.884), with a sensitivity of 87.4%, 85.0%, 87.1% and a specificity of 62.2%, 68.4%, 68.4%, respectively. Pairwise DeLong tests revealed that the AUC of the combined model (Age + FCP + CVAI, AUC = 0.837) was significantly larger than those of both the Age + FCP model (AUC difference = 0.0120, 95% CI: 0.000- 0.024, p = 0.0431) and Age alone (AUC difference = 0.0229, 95% CI: 0.003- 0.043, p = 0.0274). No significant difference was observed between Age alone and the Age + FCP model (p = 0.2295). A nomogram model was constructed based on the above three indicators to visualize the risk of AS in early-stage T2DM-DKD patients. The bootstrap-corrected calibration curve showed acceptable agreement between predicted probabilities and observed frequencies. The DCA curve confirmed significant clinical net benefit of the model. CONCLUSIONS: CVAI alone shows limited predictive value for peripheral AS in early-stage T2DM-DKD patients. However, a combined nomogram incorporating CVAI, Age, and FCP demonstrates good discrimination, calibration, and clinical utility, facilitating early identification of high-risk patients and improving prognosis.