机器人辅助脊柱畸形手术中压力性损伤风险预测模型的构建与验证OA
Construction and verification of a risk prediction model for pressure in-jury in robot-assisted spinal deformity surgery
目的 构建机器人辅助脊柱畸形手术中压力性损伤的风险预测模型,并验证其临床效能.方法 回顾性分析2020 年 1 月至 2025 年 1 月惠州市第三人民医院收治的 487 例行机器人辅助脊柱畸形手术患者的临床资料,依据 7∶3 比例将其分为训练集(341 例)和验证集(146 例).依据压力性损伤发生情况将训练集患者分为损伤组和未损伤组,比较两组患者临床资料,采用多因素 logistic 回归分析机器人辅助脊柱畸形手术中压力性损伤的影响因素.基于R语言构建列线图模型,并通过校准曲线、ROC曲线、DCA曲线评估模型效能.结果 训练集、验证集中机器人辅助脊柱畸形手术患者压力性损伤发生率分别为 14.96%(51/341)、15.75%(23/146).损伤组的体重指数(BMI)＜18.5 kg/m2 或≥28 kg/m2、糖尿病、营养不良、手术时间≥3 h、术中体温异常占比高于未损伤组,手术固定融合节段数多于未损伤组,差异有统计学意义(P＜0.05).多因素 logistic 回归分析结果显示,BMI＜18.5 kg/m2或≥28 kg/m2、糖尿病、营养不良、手术时间≥3 h、术中体温异常均是机器人辅助脊柱畸形手术中压力性损伤的独立危险因素(P＜0.05).基于多因素回归分析结果构建列线图预测模型,经验证显示训练集、验证集的一致性指数分别为 0.841、0.827,校正曲线与理想曲线拟合度较好;ROC 分析结果显示,训练集、验证集预测机器人辅助脊柱畸形手术中压力性损伤的 AUC、灵敏度、特异度分别为 0.900(95%CI:0.863～0.930)、84.31%、86.21%和 0.872(95%CI:0.807～0.922)、82.61%、84.55%;DCA 曲线显示,训练集阈值概率在 0.13～0.83 和 0.97～1.00 内、验证集阈值概率在 0.11～0.72 和 0.92～0.96 内模型有较高的净获益值.结论 BMI、糖尿病、营养不良、手术时间、术中体温均是机器人辅助脊柱畸形手术中压力性损伤的风险因素,基于上述影响因素构建的列线图风险预测模型具有良好的临床应用价值.
Objective To construct a risk prediction model for pressure injury in robot-assisted spinal deformity surgery,and to verify its clinical efficacy.Methods The clinical data of 487 patients underwent robot-assisted spinal deformity surgery in the Third People's Hospital of Huizhou from January 2020 to January 2025 were retrospectively analyzed.According to the ratio of 7∶3,they were divided into the training set(341 cases)and the validation set(146 cases).According to the occurrence of pressure injury,the patients in the training set were divided into the injury group and the non-injury group.The clinical data of the two groups were compared.Multivariate logistic regression was used to analyze the influencing factors of pressure injury in robot-assisted spinal deformity surgery.The nomogram model was constructed based on R language,and the effectiveness of the model was evaluated by calibration curve,ROC curve and DCA curve.Results The incidence rates of pressure injury in patients underwent robot-assisted spinal deformity surgery in the training set and the validation set were 14.96%(51/341)and 15.75%(23/146)respectively.The proportions of body mass index(BMI)＜18.5 kg/m2 or≥28 kg/m2,diabetes,malnutrition,operation time≥3 h and intraoperative abnormal body temperature in the injury group were higher than those in the non-injury group,and the number of fusion segments of surgical fixation was more than that of the non-injury group,with statistically significant differences(P＜0.05).Multivariate logistic regres-sion analysis showed that BMI＜18.5 kg/m2 or≥28 kg/m2,diabetes,malnutrition,operation time≥3 h and intraoper-ative abnormal body temperature were all independent risk factors for pressure injury during robot-assisted spinal deformity surgery(P＜0.05).The nomogram prediction model was constructed based on the results of multi-factor regression analysis,while the verification showed that the consistency indexes of the training set and the validation set were 0.841 and 0.827 respectively,and the calibration curve was well fitted with the ideal curve.ROC analysis results showed that the AUC,sensitivity and specificity of the training set and the validation set for predicting pressure injury in robot-assisted spinal deformity surgery were 0.900(95%CI:0.863-0.930),84.31%,86.21%and 0.872(95%CI:0.807-0.922),82.61%,84.55%respectively.The DCA curve showed that the model had a higher net benefit value when the threshold probability of the training set was within 0.13-0.83 and 0.97-1.00,and the threshold probability of the validation set was within 0.11-0.72 and 0.92-0.96.Conclusion BMI,diabetes,malnutrition,operation time and intraoperative body temperature are all risk factors for pressure injury in robot-assisted spinal deformity surgery.The nomogram risk prediction model based on the above influencing factors has good clinical application value.
邓永清;黄媚;胡世勋
广东省惠州市第三人民医院手术室,广东 惠州 516002广东省惠州市第三人民医院手术室,广东 惠州 516002广东省惠州市第三人民医院手术室,广东 惠州 516002
医药卫生
脊柱畸形机器人辅助手术压力性损伤影响因素列线图
Spinal deformityRobot-assisted surgeryPressure injuryInfluencing factorsNomogram
《中国当代医药》 2026 (12)
4-9,6
广东省惠州市科技计划项目(2022CZ010451).
评论