【摘要】 目的：分析数字化导板引导种植修复无牙颌患者种植失败的影响因素，并建立列线图模型。方法：收集2021年5月至2023年4月在我院进行种植修复治疗的136例患者的临床资料。按7∶3原则分为建模组（95例患者、461枚种植体）和验模组（41例患者、197枚种植体），根据种植情况分为失败组和成功组。Logistic多元回归模型分析种植失败的相关危险因素，并建立列线图模型，模型的内部验证与预测效能评价，分别采用校正曲线和决策曲线完成。将验模组各变量输入列线图模型中进行验证。结果：建模组中失败组吸烟比例、牙周病史比例、糖尿病比例、种植体直径<4 mm比例、种植体长度<10 mm比例、骨密度Ⅲ/Ⅳ类骨比例及颈部偏差、根部偏差、角度偏差均高于成功组（P<0.05）。经Logistic多元回归发现：吸烟、牙周病史、糖尿病、种植体直径、种植体长度、骨密度是数字化导板引导种植修复无牙颌患者种植失败的危险因素。本研究所建立的列线图模型C-index达0.892(0.822-0.963)，观测值与预测值较为吻合。当模型的阈值>0.37时，其带来的临床净收益均优于各单一因素。验模组ROC曲线AUC为0.945(95%CI:0.904～0.973),H-L拟合优度检验显示χ~2=2.705,P=0.911(P>0.05)，所建立的列线图模型在区分度和一致性表现上均较为理想。结论：本研究基于吸烟、牙周病史、糖尿病、种植体直径、种植体长度、骨密度所构建的列线图预测模型，在预测数字化导板引导种植修复无牙颌患者种植失败方面表现出较好价值，可为临床开展预防工作提供一定依据。更多还原

【Abstract】 Objective:To analyse influencing factors of implant failure in edentulous patients undergoing digitally guided implant restoration and to establish a nomogram model.Methods:The clinical data of 136 patients who received implant restoration in our hospital from May 2021 to April 2023 were collected.According to the ratio of 7:3,the patients were divided into the modeling group(95 patients,461 implants) and the validation group(41 patients,197 implants).All cases were further categorized into the failure group and the success group based on the implant survival status.Multiple logistic regression model was used to analyze the risk factors related to implant failure,and a nomogram model was established.The internal validation and prediction efficacy evaluation of the model were completed using calibration curves and decision curve analysis,respectively.Variables of the validation group were inputted into the nomogram model for external verification.Results:In the modeling group,the proportions of smoking,history of periodontal disease,diabetes mellitus,implant diameter<4 mm,implant length<10 mm, bone density of type Ⅲ/Ⅳ bone,as well as cervical deviation,apical deviation and angular deviation in the failure group were significantly higher than those in the success group(P<0.05).Multiple logistic regression analysis revealed that smoking,history of periodontal disease,diabetes mellitus,implant diameter,implant length and bone density were independent risk factors for implant failure in edentulous patients undergoing implant restoration guided by digital surgical guides.The concordance index(C-index)of the established nomogram model reached 0.892(95% CI:0.822～0.963),and the observed values were in good agreement with the predicted values.When the threshold probability of the model was greater than 0.37,the clinical net benefit of the model was superior to that of any single risk factor.For the validation group,the area under the receiver operating characteristic curve(AUC) was 0.945(95% CI:0.904～0.973).Hosmer-Lemeshow goodness-of-fit test showed that χ~2=2.705,P=0.911(P>0.05).The established nomogram model exhibited satisfactory discriminative ability and consistency.Conclusions:The nomogram prediction model constructed based on smoking,history of periodontal disease,diabetes mellitus,implant diameter,implant length and bone density shows good value in predicting implant failure in edentulous patients undergoing implant restoration guided by digital surgical guides,which can provide a certain basis for clinical intervention.更多还原