Related Experiment Video
Updated: Jun 29, 2026
Establishing a Competing Risk Regression Nomogram Model for Survival Data
Published on: October 23, 2020
Risk Factors and Nomogram for Predicting the Recurrence of Conventional Ameloblastoma
Shengqi Ouyang1,2,3, Jinqi Zhang1,2,3, Linlin Ren1,2,3
1Hospital of Stomatology, Sun Yat-Sen University, Guangzhou, China.
Objective:
To identify risk factors for the recurrence of conventional ameloblastoma (AM) and develop a predictive nomogram model for postoperative recurrence.
Methods:
Clinical, pathological, and radiographic data from 235 patients treated at the Hospital of Stomatology, Sun Yat-sen University (2005-2022) were retrospectively analyzed. Logistic regression identified independent prognostic factors, and a nomogram model was constructed using R software.
Results:
Multilocularity, cortical bone perforation, and inferior alveolar nerve involvement significantly increased recurrence risk in conventional AM (p < 0.05). Logistic regression analysis revealed that multilocular tumors (OR = 3.68, p = 0.002), honeycomb patterns (OR = 10.80, p < 0.001), and cortical bone perforation (OR = 2.62, p = 0.011) were significantly associated with increased recurrence risk. Surgical approaches significantly impacted recurrence rates, with curative resection surgery (CRS) resulting in the lowest recurrence risk as compared with fenestration decompression (FD) and local curettage (LC). CRS and curettage and fenestration decompression (CFD) were identified as effective protective factors against recurrence (p < 0.001). A nomogram model with high predictive accuracy (AUC = 0.867) was developed.
Conclusion:
Multilocularity, cortical bone perforation, and inferior alveolar nerve involvement significantly increased the risk of recurrence. The nomogram model provided effective risk assessment for clinical decision-making.
