What is the clinical evidence from this study?

Study design: Cohort. Population: Postpartum venous thromboembolism (n=279). Intervention: Low-molecular-weight heparin (nadroparin). Primary outcome: Imaging-based thrombus regression (AUC-ROC 0.831, 95% CI 0.696-0.967).

What question did this study set out to answer?

The aim is to develop a predictive model for thrombus regression in postpartum patients with venous thromboembolism (VTE).

April 10, 2026Open Access

A novel predictive model for evaluating thrombus regression in postpartum venous thromboembolism patients

Key Result

The random forest model optimally predicted thrombus regression in postpartum VTE patients receiving LMWH, achieving an AUC-ROC of 0.831.

Key Points

The aim is to develop a predictive model for thrombus regression in postpartum patients with venous thromboembolism (VTE).
Data analysis from three cohorts of postpartum VTE patients
Development of a random forest (RF) model
Identification of key predictors including anti-Xa activity and D-dimer levels
The RF model was identified as the optimal predictive model for thrombus regression events
Key predictors include anti-Xa activity, antithrombin III levels, D-dimer levels, and BMI
The model may assist in assessing changes in the thrombotic state of postpartum VTE patients

Study Design

Type

Cohort (n=279)

Multicenter

Yes

Structured PICO

Does a Random Forest model predict thrombus regression events in patients with postpartum venous thromboembolism?

Population

Patients with postpartum venous thromboembolism (VTE) from three cohorts

Intervention

Random Forest (RF) predictive model

Outcome

Thrombus regression eventssurrogate

A novel Random Forest model utilizing anti-Xa activity, antithrombin III, D-dimer, and BMI can predict thrombus regression in postpartum VTE patients to potentially guide precision medication.

Main Result

Effect estimate: AUC-ROC 0.831 (95% CI 0.696-0.967)

Limitations

Small sample size, particularly for the external validation cohort (n = 29)
Training data came from a single-center retrospective cohort, which may introduce selection bias
Cannot eliminate the possibility of other unmeasured factors, such as haemoglobin levels
Model was developed in a Chinese population, and its applicability to other ethnic groups requires further investigation

Abstract

Based on data from the three cohorts of patients with postpartum VTE, the RF model was identified as the optimal model for predicting thrombus regression events, with anti-Xa activity, antithrombin III levels, D-dimer levels, and BMI serving as key predictors. This study may help assess changes in the thrombotic state of postpartum VTE patients and guide clinical precision medication.

Bookmark

View Full Paper