乳腺癌患者植入式静脉输液港相关感染风险预测模型的构建

Construction of a Risk Prediction Model for Implanted Venous Access Port-related Infection in Breast Cancer Patients

目的分析影响乳腺癌患者发生植入式静脉输液港(IVAP)相关感染的危险因素,并构建风险预测模型。方法2020年6月至2023年6月,采用方便抽样方法选取南昌市人民医院收治的使用IVAP治疗的200例乳腺癌患者作为研究对象。根据是否出现IVAP相关感染并发症将其分为感染组及非感染组。回顾性分析两组患者的临床资料,使用二元logistic回归分析乳腺癌患者发生IVAP相关感染的影响因素,并根据独立影响因素建立风险预测模型,采用受试者工作特征性曲线对模型进行评估。结果200例患者中,共有27例发生IVAP相关感染,发生率为13.50%。治疗方案、糖尿病史、肠外营养支持均为乳腺癌患者发生IVAP相关感染的独立危险因素(P<0.05)。构建的风险预测模型为Logit(P)=2.001×治疗方案+1.969×糖尿病史+1.495×肠外营养水平-3.842,模型的受试者工作特征性曲线下面积为0.832(95%CI为0.751~0.913,P<0.001),灵敏度为77.8%,特异度为74.0%。结论乳腺癌患者发生IVAP相关感染并发症的影响因素较多,本研究构建的风险预测模型有较好的预测性能,有一定的临床应用价值。

Objective To analyze the risk factors of implanted venous access port(IVAP)-related infection in breast cancer patients and establish a risk prediction model.Methods From June 2020 to June 2023,200 breast cancer patients were selected by convenience sampling who were treated with IVAP in Nanchang People's Hospital.The patients were divided into infected and non-infected groups depending on whether they had IVAP-related infectious complications.The clinical data of the two groups were retrospectively analyzed and the influencing factors of IVAP-related infection in breast cancer patients were analyzed using binary logistic regression.A risk prediction model was established based on independent influencing factors.Then,the model was evaluated using the receiver operating characteristic(ROC)curve.Results Of the 200 patients,27 patients(13.50%)developed IVAP-related infection.Treatment regimen,diabetes history,and parenteral nutrition support were independent risk factors for IVAP-related infection in breast cancer patients(P<0.05).The constructed risk prediction model was Logit(P)=2.001×treatment regimen+1.969×diabetes history+1.495×parenteral nutrition level−3.842.The area under the ROC curve of the model was 0.832(95%CI:0.751-0.913,P<0.001),with a sensitivity of 77.8%and specificity of 74.0%.Conclusion There are many influencing factors for IVAP-related infection complications in breast cancer patients,and the risk prediction model constructed in this study holds certain clinical application value due to its robust prediction performance.