Construction and validation of a risk prediction model for PICC-related venous thrombosis of adult cancer patients based on Meta-analysis
LIU Xiaoling
CHEN Hui
Abstract：Objective:To construct and validate a risk predictive model for peripheral inserted central catheter(PICC)-related venous thrombosis in adult cancer patients based on Meta-analysis and retrospective clinical data.Methods:A systematic literature search was performed in PubMed,EMbase,the Cochrane Library,CNKI,WanFang Database,VIP,and CBM for identifying researches focusing PICC related venous thrombotic risk factors in adult cancer patients published through December 2022,with the latest update in December 2023.Meta-analysis was performed using STATA 14.0 to identify independent risk factors for PICC related venous thrombosis in adult cancer patients,then the effect of each independent risk factor on the development of PICC-related venous thrombosis was determined using β-formation conversion for developing the risk predictive model.Finally,we collected the clinical data of 905 adult cancer patients who underwent PICC catheterization at Chongqing University Cancer Hospital between January 2020 and December 2021 to evaluate the predictive performance of the risk predictive model by drawing the receiver of curve(ROC).Results:A total of 16 articles covering 32 risk factors were included for the final data analysis.Meta-analysis confirmed that 13 of these risk factors were significantly associated with the development of the PICC-related venous thrombosis.We constructed the prediction model as follow:Logit(P)=-2.336+0.978 × tumor metastasis+1.613 × TPN infusion+0.940 × age+1.157 × tumor stage+1.411× catheterized vein+1.157 × catheter tip position+0.802 × APTT level+1.416 × D-dimer+1.340 × pathological type+0.854 × diabetes+1.169 × thrombotic history+1.154 × chemotherapy history during catheterization+0.285 × hyperlipidemia.The clinical data of 905 patients were imported into the prediction model for external validation,obtaining a sensitivity of 67.7％,specificity of 60.4％,Youden index of 0.281,and area under the ROC curve(AUC)of 0.638.Conclusion:This study constructs a risk prediction model for PICC-related venous thrombosis based on Meta-analysis,which includes 13 independent risk factors.The model have good predictive performance,indicating that the risk prediction model for PICC-related venous thrombosis constructed based on evidence integration strategy have good extrapolation,scientific validity,and practicality.
Keywords：cancerperipheral inserted central catheterPICCMeta-analysisrisk factorsprediction model
Publication Date：2024-08-15
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：8( 2681-2688 )