组织工程与重建外科杂志

组织工程与重建外科杂志 ›› 2024, Vol. 20 ›› Issue (3): 305-.

• • 上一篇    下一篇

四肢创伤性软组织缺损患者游离皮瓣移植失败的危险因素及预测模型构建

  

  • 出版日期:2024-06-01 发布日期:2024-07-05

Risk factors analysis and predictive model construction for the failure of free flap transplantation in patients with traumatic soft tissue injury of limbs

  • Online:2024-06-01 Published:2024-07-05

PDF (PC)

16

摘要：

　目的　探讨四肢创伤性软组织缺损（STI）患者游离皮瓣移植（FFT）修复失败的危险因素，并构建预测模型。方法　选取2017年1月至2023年6月期间行FFT的575例四肢创伤性STI患者为研究对象，7︰3随机分为建模组（n=403）和验证组（n=172）。多因素 Logistic 回归分析确定四肢创伤性 STI 患者 FFT 修复失败的危险因素。利用 R3.6.3 软件中rms程序包构建预测四肢创伤性STI患者FFT修复失败的列线图模型，并采用受试者工作特征（ROC）曲线和霍斯默-莱梅肖（H-L）拟合度曲线，对列线图模型预测区分度和一致性进行评估。结果多因素Logistic回归分析显示，受伤至手术时间≥6 h（OR=2.179）、吻合血管选择不合理（OR=6.770）、并发感染（OR=4.980）、术中止血不彻底（OR=3.054）、术后固定不牢固（OR=2.132）等，均为四肢创伤性 STI 患者 FFT 修复失败的独立危险因素（P＜0.05）；建模组 H-L 检验 χ2 =11.889、P=0.546，ROC 曲线下面积为 0.809（95%CI：0.748~0.869）；验证组 H-L 检验 χ2 =6.990、P=0.322，ROC 曲线下面积为 0.827（95%CI：0.722~0.932）。结论　基于受伤至手术时间、吻合血管选择、并发感染、术中止血不彻底、术后固定不牢固等5个危险因素构建的预测四肢创伤性STI患者FFT修复失败的列线图模型具有较好的区分度和一致性。

关键词:

Abstract:

Objective To explore the risk factors for failure of free flap transplantation （FFT） in patients with traumatic
soft tissue injury （STI） of limbs， and to construct a predictive model. Methods A total of 575 patients with traumatic STI in
the limbs who underwent FFT from January 2017 to June 2023 were regarded as the study subjects. The patients were
randomly separated into a modeling group （n=403） and a validation group （n=172） at 7：3. Multivariate Logistic regression
analysis was applied to determine the risk factors for FFT repair failure in patients with traumatic STI of the limbs. The RMS
program package in R3.6.3 software was applied to construct a column graph model for predicting the failure of FFT repair in
patients with traumatic STI of limbs， receiver operating characteristic （ROC） curve and the Hosmer-Lemeshow （H-L） fit
curve were applied to evaluate the discrimination and consistency of the column plot model prediction. Results The results
of multivariate Logistic regression analysis showed that the time from injury to surgery ≥ 6 hours （OR=2.179）， unreasonable
selection of anastomosis vessels （OR=6.770）， concurrent infection （OR=4.980）， incomplete intraoperative hemostasis （OR=
3.054）， and unstable postoperative fixation （OR=2.132） were all independent risk factors for FFT repair failure in patients
with traumatic STI of limbs （P<0.05）
.
The H-L test of the modeling group showed χ2 =11.889，P=0.546， and the area under
the ROC curve was 0.809（95% CI：0.748-0.869）； The H-L test of the validation group showed χ2 =6.990，P=0.322， and
the area under the ROC curve was 0.827（95% CI：0.722-0.932）
.
Conclusion Based on five risk factors： Time from injury
to surgery， selection of anastomotic vessels， concurrent infection， incomplete intraoperative hemostasis， and unstable
postoperative fixation， the column chart model for predicting the failure of FFT repair in patients with traumatic STI of limbs
has good discrimination and consistency

Key words: