ObjectiveTo investigate the dynamic changes of nodule volume in benign thyroid tumors after radiofrequency ablation (RFA), and to analyze the predictive value of risk factors for nodule regeneration. MethodsA total of 165 patients with benign thyroid nodules who received RFA treatment in the People’s Hospital of Yuechi County from June 2019 to June 2021 were retrospectively collected and divided into small nodule volume group (≤15 mL, n=116) and large nodule volume group (>15 mL, n=49) according to the median nodule volume at admission. The clinical data and serological data of the two groups were compared. Multivariate Cox proportional hazard regression model was used to adjust confounding factors to explore the relationship between initial nodule volume, vascular density, nodule location near critical structure and postoperative nodule regeneration in patients with benign thyroid nodules. According to the proposed Nomogram of the model, Bootstrap method was adopted for sampling verification, calibration curve was adopted to evaluate the calibration degree of the model, and area under the curve (AUC) of receiver operating characteristics (ROC) curve was adopted to evaluate the model differentiation. ResultsIn the small volume nodule group, the proportion of unilateral nodule was higher, and the preoperative beauty score, preoperative symptom score, radiofrequency power, ablation time, total energy, operative time, intraoperative blood loss and hospital stay were lower or short, P<0.05. The change value of thyroid stimulating hormone (TSH), free triiodothyronine (FT3), catalase (CAT) after operation in small volume nodule group were higher than those in the large nodule volume group (P<0.05), but change value of the free thyroxine (FT4), malondialdehyde (MDA), superoxide dismutase (SOD) and glutathione peroxidase (GSH-Px) were lower than those in the large nodule volume group (P<0.05). At 1, 3, 6, 12, and 24 months after RFA operation, the nodule volume of the two groups decreased successively, and volume reduction rate (VRR) increased successively. The changes of nodule volume and VRR in the small nodule volume group were better than those in the large nodule volume group. In the large nodule group, the nodule volume was larger at 1, 3, 6, 12, and 24 months after operation, and the VRR was higher at 1 month after operation, the regeneration time was shorter, the number of residual nodules was higher, and the initial ablation rate (IAR) was lower (P<0.05). After adjusting for age, gender and other factors, the correlation effect value increased with the increase of initial volume and blood vessel density, and the differences in trend test were statistically significant (Ptrend <0.05). Nodules located near the critical structure had an increased risk of regeneration (OR=1.76, P<0.001). The Nomogram constructed according to the multi-factor model has good differentiation (AUC before and after ROC curve validation were 0.854 and 0.814, respectively) and accuracy (mean absolute error of 0.023). ConclusionsRFA achieved clinically relevant volume reduction in both ≤15 mL and >15 mL of single benign thyroid nodules, lasting for at least 2 years. However, the nodule VRR and cosmetic effect were better in the small volume nodule group, and the initial nodule volume, blood vessel density and location were closely related to nodule regeneration. The Nomogram model showed good differentiation and accuracy in predicting the risk of nodule regeneration, providing strong support for clinical decision-making.