牙源性颌骨囊肿术后复发的影响因素分析及风险预测研究OA
Analysis of influencing factors and risk prediction study on postoperative recurrence of odontogenic jaw cysts
目的:探讨牙源性颌骨囊肿术后复发的影响因素,并构建其风险预测模型.方法:回顾性收集 2023 年1 月~2024 年 12 月于本院行手术治疗的 461 例牙源性颌骨囊肿患者的临床资料,按照 7∶3 的比例将其随机分为训练集(n=323)和验证集(n=138).根据术后 1 年的复发情况将训练集患者分为复发组与未复发组,收集并比较两组患者的临床资料.采用单因素分析及多因素 Logistic 回归分析筛选复发的影响因素,并基于此构建列线图风险预测模型.采用受试者工作特征曲线(ROC)评估模型的区分度,计算曲线下面积(AUC)、灵敏度及特异度;通过校准曲线及 Hosmer-Lemeshow 检验评估模型校准度.结果:牙源性颌骨囊肿术后 1 年复发率为 18.22%(84/461).多因素 Logistic 回归分析结果显示,角化囊肿[比值比(OR)=1.898,95%置信区间(95%CI)1.329~2.712]、囊肿直径(OR=1.697,95%CI 1.142~2.522)、单纯刮治术(OR=1.795,95%CI 1.237~2.605)以及中性粒细胞/淋巴细胞比值(NLR)(OR=1.626,95%CI 1.151~2.296)是牙源性颌骨囊肿术后复发的独立危险因素(P<0.05),拔除受累牙(OR=0.572,95%CI 0.381~0.860)、辅助治疗(OR=0.621,95%CI 0.452~0.853)是牙源性颌骨囊肿术后复发的独立保护因素(P<0.05);根据多因素 Logistic 回归分析结果构建牙源性颌骨囊肿术后复发的风险预测模型,其中角化囊肿、囊肿直径、单纯刮治术、拔除受累牙、辅助治疗、NLR 因素对应的风险评分分别为91 分、82 分、79 分、65 分、85 分、83 分.该模型预测训练集牙源性颌骨囊肿术后复发的 AUC 为 0.965(95%CI 0.939~0.982),灵敏度和特异度为95.24%、85.38%;预测验证集牙源性颌骨囊肿术后复发的 AUC 为 0.937(95%CI 0.882~0.971),灵敏度和特异度为90.48%、89.74%;模型预测训练集和验证集牙源性颌骨囊肿术后复发的校准曲线均趋近于理想曲线.Hosmer-Lemeshow 拟合优度检验结果显示,训练集χ2=4.126、P=0.127,验证集χ2=2.851、P=0.094.结论:角化囊肿、囊肿直径、单纯刮治术及 NLR 是牙源性颌骨囊肿术后复发的独立危险因素,拔除受累牙及辅助治疗是其独立保护因素.基于上述因素构建的列线图预测模型预测效能良好,有助于识别高危患者并指导个体化治疗.
Objective:To investigate the influencing factors for postoperative recurrence of odontogenic jaw cysts and to construct a risk prediction model.Methods:Clinical data of 461 patients with odontogenic jaw cysts who underwent surgi-cal treatment in our hospital from January 2023 to December 2024 were retrospectively collected.The patients were randomly divided into a training set(n=323)and a validation set(n=138)at a ratio of 7∶3.Patients in the training set were divided into recurrence group and non-recurrence group based on the recurrence status at 1 year postoperatively,and clinical data were collected and compared between the two groups.Univariate analysis and multivariate Logistic regression analysis were used to screen independent influencing factors for recurrence,and a nomogram risk prediction model was constructed accord-ingly.The receiver operating characteristic(ROC)curve was used to evaluate the discrimination of the model,and the area under the curve(AUC),sensitivity,and specificity were calculated.Calibration curve and Hosmer-Lemeshow test were used to evaluate the calibration of the model.Results:The 1-year postoperative recurrence rate of odontogenic jaw cysts was 18.22%(84/461).Multivariate Logistic regression analysis showed that keratocyst[odds ratio(OR)=1.898,95%confidence interval(95%CI)1.329~2.712],cyst diameter(OR=1.697,95%CI 1.142~2.522),simple curettage(OR=1.795,95%CI 1.237~2.605),and neutrophil-to-lymphocyte ratio(NLR)(OR=1.626,95%CI 1.151~2.296)were independent risk fac-tors for postoperative recurrence of odontogenic jaw cysts(P<0.05),while extraction of involved teeth(OR=0.572,95%CI 0.381~0.860)and adjuvant therapy(OR=0.621,95%CI 0.452~0.853)were independent protective factors(P<0.05).A risk prediction model for postoperative recurrence of odontogenic jaw cysts was constructed based on the results of multivariate Logistic regression analysis.The risk scores corresponding to keratocyst,cyst diameter,simple curettage,extraction of involved teeth,adjuvant therapy,and NLR factors were 91 points,82 points,79 points,65 points,85 points,and 83 points,respectively.The AUC of the model for predicting postoperative recurrence in the training set was 0.965(95%CI 0.939~0.982),with a sensitivity of 95.24%and a specificity of 85.38%,the AUC for predicting postoperative recurrence in the validation set was 0.937(95%CI 0.882~0.971),with a sensitivity of 90.48%and a specificity of 89.74%.The predicted probabilities of re-currence in both the training set and the validation set were close to the ideal curve.The Hosmer-Lemeshow test showed χ2=4.126,P=0.127 for the training set and χ2=2.851,P=0.094 for the validation set.Conclusion:Keratocyst,cyst diameter,simple curettage,and NLR are independent risk factors for postoperative recurrence of odontogenic jaw cysts,while extraction of involved teeth and adjuvant therapy are independent protective factors.The nomogram prediction model constructed based on the above factors demonstrates good predictive performance,which is helpful for identifying high-risk patients and guiding individualized treatment.
曾小勇;李淑林;王思豪;陈凌君
永州市中心医院口腔颌面外科 湖南 永州 425000永州市中心医院口腔颌面外科 湖南 永州 425000永州市中心医院口腔颌面外科 湖南 永州 425000永州市中心医院麻醉科 湖南 永州 425000
医药卫生
牙源性颌骨囊肿复发影响因素列线图
Odontogenic jaw cystsRecurrenceInfluencing factorsNomogram
《临床口腔医学杂志》 2026 (9)
545-550,6
2025年永州市指导性科技计划项目(2025YZ012)
评论