Authors
Cuihua Wang, Haige Zhao, Dandan Guo, Si Zhang, Yafei Yan
Published in
Clinical and applied thrombosis/hemostasis : official journal of the International Academy of Clinical and Applied Thrombosis/Hemostasis. Volume 32. Pages 10760296261508954. Epub Nov 17, 2026.
Abstract
BackgroundShort-term prognostic tools for venous thromboembolic (VTE) deterioration after acute lower-extremity deep vein thrombosis (DVT) are limited. We aimed to develop and internally validate a model integrating clinical, laboratory, and duplex ultrasound characteristics.MethodsThis single-center retrospective cohort included 785 adults with ultrasound-confirmed acute lower-extremity DVT from January 2018 to December 2025. The primary outcome was clinically detected and objectively confirmed ipsilateral DVT progression or new-onset symptomatic pulmonary embolism (PE) within 30 days. Three nested ridge-penalized logistic models included clinical predictors (Model 1), clinical plus laboratory predictors (Model 2), and clinical, laboratory, and ultrasound predictors (Model 3). Missing data were multiply imputed, and the complete modeling procedure was internally validated with 1,000 bootstrap resamples. Discrimination, calibration, Brier score, incremental performance, and decision-curve net benefit were assessed.ResultsThe primary outcome occurred in 102 patients (13.0%): 82 had DVT progression and 27 symptomatic PE, with seven having both. Optimism-corrected AUCs were 0.702, 0.766, and 0.808 (95% CI, 0.769-0.847) for Models 1-3, respectively. For Model 3, the optimism-corrected calibration slope was 0.94 and Brier score 0.087. Optimism-corrected ΔAUCs were 0.064 and 0.042, with corresponding ΔBrier scores of -0.009 and -0.008. Model 3 showed the highest optimism-corrected net benefit across thresholds of approximately 8%-25%. Observed outcome proportions across risk tertiles were 3.1%, 9.6%, and 26.3%.ConclusionsModel 3 provided internally validated prediction of clinically detected VTE deterioration under local treatment and imaging practices. External validation and prospective clinical-impact assessment are required before routine use.
PMID:
43044100
Bibliographic data and abstract were imported from PubMed on 18 Nov 2026.
Read full publication at:
Please sign in
to see all details.
Advertisement
Stats
- Recommendations n/a n/a positive of 0 vote(s)
- Views 4
- Comments 0