口腔医学研究

口腔医学研究 ›› 2026, Vol. 42 ›› Issue (6): 529-533.DOI: 10.13701/j.cnki.kqyxyj.2026.06.012

• 口腔颌面外科学研究 • 上一篇    下一篇

腮腺良性肿瘤切除术后面神经损伤的回顾性研究与预测模型

陈雅萍*, 梁爽, 贾亚斌, 刘燕, 王福丽   

  1. 信阳市中心医院口腔科 河南 信阳 464000
  • 收稿日期:2025-10-27 出版日期:2026-06-28 发布日期:2026-06-23
  • 通讯作者: *陈雅萍,E-mail:cyp147742@126.com
  • 作者简介:陈雅萍(1995~)女,安徽宿州人,硕士,住院医师,研究方向:口腔颌面外科疾病的诊治。

Retrospective Study and Predictive Model of Nerve Injury after Benign Parotid Tumor Resection

CHEN Yaping*, LIANG Shuang, JIA Yabin, LIU Yan, WANG Fuli   

  1. Department of Stomatology, Xinyang Central Hospital, Xinyang 464000, China
  • Received:2025-10-27 Online:2026-06-28 Published:2026-06-23

PDF (PC)

3

摘要： 目的:统计腮腺良性肿瘤切除术后面神经损伤发生率,分析其相关因素,构建预测模型。方法:选取信阳市中心医院口腔颌面外科2022年7月~2025年7月收治的104例腮腺良性肿瘤切除术患者,对其临床资料进行回顾性分析。记录术后面神经损伤发生情况,采用Logistic多因素回归模型分析面神经损伤的相关独立因素,并构建预测模型。结果:104例患者中发生面神经损伤者32例,发生率为30.77%。Logistic多因素分析结果显示,肿瘤深度、是否游离面神经、是否保留筋膜是腮腺良性肿瘤切除术后面神经损伤的独立影响因素(P<0.05)。据此构建Nomogram预测模型。受试者工作特征曲线分析结果显示,该预测模型判断腮腺良性肿瘤切除术后面神经损伤的曲线下面积为0.810(SE=0.056,95%CI=0.700~0.920,P<0.001),敏感度为0.844,特异度为0.722。结论:腮腺良性肿瘤切除术后存在面神经损伤风险,其发生与肿瘤深度、是否游离面神经及是否保留筋膜相关,据此构建的预测模型有助于面神经损伤的早期筛查。

关键词: 腮腺良性肿瘤切除术, 面神经损伤, 影响因素, 预测模型

Abstract: Objective: To investigate the incidence of facial nerve injury after parotidectomy for benign parotid tumors, analyze its related factors, and construct a predictive model. Methods: A retrospective analysis of clinical data was performed on 104 patients who underwent parotidectomy for benign parotid tumors and were admitted in the Department of Oral and Maxillofacial Surgery, Xinyang Central Hospital from July 2022 to July 2025. The occurrence of postoperative facial nerve injury was recorded, the logistic multiple regression model was used to analyze the independent factors associated with facial nerve injury and a predictive model was constructed. Results: Among the 104 patients, facial nerve injury occurred in 32 cases, with an incidence rate of 30.77%. The Logistic multiple regression analysis showed that the tumor depth, whether the facial nerve was dissected, and whether the fascia was preserved were independent influence factors for facial nerve injury after parotidectomy for benign parotid tumors (P<0.05). A Nomogram prediction model was constructed accordingly. The ROC analysis showed that the AUC of the predictive model for judging facial nerve injury after parotidectomy for benign parotid tumors was 0.810 (SE=0.056, 95%CI=0.700-0.920, P<0.001), with a sensitivity of 0.844 and a specificity of 0.722. Conclusion: There is a risk of facial nerve injury after parotidectomy for benign parotid tumors. The occurrence of this injury is associated with tumor depth, whether the facial nerve is dissected, and whether the fascia is preserved. The predictive model constructed based on those factors is conducive to the early screening of facial nerve injury.

Key words: benign parotid tumor resection, facial nerve injury, influence factors, predictive model