基于列线图模型分析CT增强高压注射外渗的风险因素及预防对策分析

Risk Factors and Preventive Measures of CT Enhanced High Pressure Injection Extravasation Were Analyzed Based on the Nomogram Model

目的基于列线图模型分析CT增强高压注射外渗的风险因素,并分析预防CT增强高压注射外渗发生的对策。方法回顾2020年1月~2023年6月期间于本院进行CT增强扫描发生高压注射外渗的患者的临床资料(外渗组,n=72),另回顾未发生高压注射外渗患者的临床资料(无外渗组,n=216)。采用Logistic回归模型分析明确导致CT增强高压注射外渗的危险因素,并根据独立危险因素构建列线图预测模型,采用受试者工作特征曲线(ROC)、Bootstrap法验及Calibration曲线对模型进行评价。结果外渗组年龄≥60岁、BMI≥24kg/m^(2)、手腕注射、未对比剂加热、注射速率≥3mL/s、化疗史人数占比均高于无外渗组(P<0.05);Logistic回归模型分析结果显示,年龄≥60岁、手腕注射、注射速率≥3mL/s、化疗史是影响高压注射外渗的独立危险因素(P<0.05);ROC曲线显示,预测模型的AUC为0.828(95%CI:0.780~0.870),灵敏度90.28%、特异度78.70%、约登指数0.690;Bootstrap验证结果显示,C-index值为0.762(95%CI:0.651~0.816),Calibration曲线显示,Hosmer-Lemeshowχ^(2)=0.628,P=0.317。结论年龄≥60岁、BMI≥24 kg/m^(2)、手腕注射、未对比剂加热、注射速率≥3 mL/s、化疗史是CT增强扫描发生高压注射外渗的独立危险因素,基于上述因素构建的预测模型对CT增强高压注射外渗具有较高的预测效能。

Objective To analyze the risk factors of CT enhanced high pressure injection extravasation based on the column graph model,and analyze the countermeasures to prevent CT enhanced high pressure injection extravasation.Methods The clinical data of patients with high-pressure injection extravasation who underwent enhanced CT scan in our hospital from January 2020 to June 2023 were reviewed(extravasation group,n=72).The clinical data of patients without high pressure injection extravasation were also reviewed(no extravasation group,n=216).Logistic regression model was used to analyze and identify the risk factors leading to CT enhanced highpressure injection extravusion,and a nomogram prediction model was constructed according to independent risk factors.Receiver operating characteristic curve(ROC),Bootstrap test and Calibration curve were used to evaluate the model.Results The proportion of patients in extravasation group with age≥60 years,BMI≥24 kg/m^(2),wrist injection,no heating with contrast agent,injection rate≥3 mL/s and history of chemotherapy were higher than those in nonextravasation group(P<0.05).Logistic regression model analysis showed that age≥60 years old,wrist injection,injection rate≥3 mL/s,and chemotherapy history were independent risk factors for high pressure injection extravasation.The ROC curve showed that the AUC of the prediction model was 0.828(95%CI:0.780~0.870),the sensitivity was 90.28%,the specificity was 78.70%,and the Yoden index was 0.690.The Calibration results from Bootstrap verification showed that the C-index value was 0.762(95%CI:0.651~0.816).The calibration curve showed that Hosmer-Lemeshowχ^(2)=0.628,P=0.317.Conclusion Age≥60 years,BMI≥24kg/m^(2),wrist injection,heating without contrast agent,injection rate≥3mL/s,and history of chemotherapy are independent risk factors for high pressure injection extravasation in enhanced CT scan.The prediction model based on the above factors has high predictive efficiency for CT enhanced high-pressure injection extravasation.