目的 调查住院老年患者因经外周静脉置入中心静脉导管（PICC）而引发的血栓情况，并分析其影响因素，为老年患者置入PICC产生的相关性血栓（PICC-CRT）和症状性血栓的评估与干预提供借鉴。方法 在2023年1月—2023年12月期间，选取广州市第一人民医院接受PICC置管的317例住院老年患者。采用包括患者一般情况调查表、运动功能评定、肌力检测、 Barthel指数评定、Padua评分等多种工具进行综合评估。采用多因素 Logistic 回归构建 PICC-CRT 及症状性血栓的预测模型，并应用逐步回归法优化变量筛选过程。模型性能通过 ROC 曲线进行评估。结果 去除临床资料不完整的患者40例，最终纳入277例患者的完整资料，其中123例患者出现了PICC-CRT，发生率为44.40%（123/277）。血栓分级中，I级78例，Ⅱ级37例，Ⅲ级8例。无症状血栓83例，占67.48%，发生率29.96%（83/277）；症状性血栓40例，占32.52 %，发生率14.44%（40/277）。单因素分析联合多因素Logistic回归显示，卒中史、凝血酶原时间（PT）、导管留置时间是住院老年患者PICC-CRT的关键因素（P＜0.05），预测模型ROC曲线下面积为0.719；置管史、恶性肿瘤史、导管留置时间、置管后并发症数量是住院老年患者PICC症状性血栓的独立影响因素（P＜0.05），预测模型ROC曲线下面积为0.812。结论 文章总结了PICC-CRT和症状性血栓独特的影响因素，基于关键因素构建了预测模型预测其发生，为护理人员预防PICC-CRT和症状性血栓的发生提供了参考。
Objective To explore the incidence of thrombosis associated with peripherally inserted central catheter（PICC）placement in hospitalized elderly patients and to analyze its influencing factors，in order to provide a reference for the assessment and prevention of PICC-catheter related thrombosis（PICC-CRT）and symptomatic thrombosis in this population．Methods A total of 317 elderly inpatients who underwent PICC placement at a tertiary hospital in Guangzhou between January and December 2023 were enrolled．Comprehensive assessments were conducted using general patient information forms，motor function evaluation，muscle strength testing，Barthel Index，and Padua score．Multivariate Logistic regression analysis was used to construct predictive models for PICC-CRT and symptomatic thrombosis，with variable selection optimized via stepwise regression．Model performance was evaluated using receiver operating characteristic（ROC）curve analysis．Results After excluding 40 patients with incomplete clinical data，277 cases were included in the final analysis．Among them，123 patients developed PICC-CRT，with an incidence rate of 44.40%（123/277）．Thrombosis was graded as Grade I in 78 cases，Grade II in 37 cases，and Grade III in 8 cases．Asymptomatic thrombosis occurred in 83 cases（29.96%），accounting for 67.48% of PICC-CRT；symptomatic thrombosis occurred in 40 cases（14.44%），accounting for 32.52%．Univariate and multivariate Logistic regression analyses identified history of stroke，prothrombin time（PT），and catheter dwell time as key risk factors for PICC-CRT（P＜0.05），with the area under the curve（AUC）of 0.719．History of catheterization，malignancy，catheter dwell time，and number of post-catheterization complications were independent predictors of symptomatic thrombosis（P＜0.05），with an AUC of 0.812．Conclusions This study identified distinct risk factors for PICC-CRT and symptomatic thrombosis in elderly inpatients．Predictive models based on key variables may provide useful references for clinical staff in preventing the occurrence of PICC-related and symptomatic thrombosis．
