Background: Peripherally inserted central catheter-related venous thrombosis (PICC-RVT) is an important complication in patients with gastrointestinal (GI) malignancies. Established catheter-related factors may interact with patient-level hypercoagulability and inflammation, but their combined predictive contribution in patients undergoing interventional therapy remains uncertain. This study aimed to estimate the cumulative incidence of symptomatic PICC-RVT and develop an internally validated prediction model accounting for competing events.
Methods: This single-center retrospective cohort included 782 patients with GI malignancies who underwent PICC placement. The primary outcome was symptomatic PICC-RVT confirmed by ultrasonography. Death and unplanned catheter removal for non-thrombotic causes were treated as competing events. Seven candidate predictor parameters were prespecified and evaluated using Fine-Gray subdistribution hazard regression, followed by Akaike information criterion-based backward reduction. Model discrimination and calibration were assessed using the C-index and calibration curves. Internal validation was performed with 1,000 bootstrap resamples, repeating the complete model-development procedure.
Results: During a median potential follow-up of 72 days, 74 patients developed symptomatic PICC-RVT. The cumulative incidence was 7.8% at 1 month and 9.2% at 3 months. Four predictors were retained in the final model: CVR >0.45, non-cavoatrial junction catheter tip position, higher baseline D-dimer, and NLR ≥3.0. The apparent C-index was 0.76 (95% CI, 0.71-0.81), and the optimism-corrected C-index was 0.75. Calibration was acceptable at 1, 2, and 3 months.
Conclusions: The model quantified the combined associations of established catheter-related factors and patient-level coagulation and inflammatory markers with symptomatic PICC-RVT in patients with GI malignancies undergoing interventional therapy. These findings reinforce standardized catheter sizing and tip positioning. External validation and clinical impact assessment are required before the model is considered for clinical implementation.
Reference:Cui T, Li J, Chen W, Hou Y, Lv H. Development and validation of a competing risk nomogram for predicting PICC-related venous thrombosis in patients with gastrointestinal malignancies. Front Oncol. 2026 Aug 5;16:1822555. doi: 10.3389/fonc.2026.1822555. PMID: 42620322; PMCID: PMC13485509.