# Construction and evaluation of a thrombosis risk prediction model in patients with primary immune thrombocytopenia

**Authors:** 佳慧 毛, 琳 王, 艳 石, 琳琳 邵, 宇 侯, 明 侯

PMC · DOI: 10.3760/cma.j.cn121090-20250507-00213 · Chinese Journal of Hematology · 2026-02-01

## TL;DR

This study identifies risk factors for thrombosis in immune thrombocytopenia patients and builds a prediction model to help identify high-risk individuals.

## Contribution

A novel thrombosis risk prediction model for primary immune thrombocytopenia patients using clinical factors and logistic regression.

## Key findings

- 12% of ITP patients developed thrombosis, with arterial thrombosis being the most common.
- Age, ITP duration >1 year, coronary heart disease, and platelet count were independent risk factors.
- The prediction model showed good accuracy with an AUC of 0.80 and strong calibration.

## Abstract

分析原发免疫性血小板减少症（ITP）患者发生血栓的危险因素，在此基础上构建血栓预测模型并对模型进行评价。

回顾性分析2018年1月至2022年12月于山东大学齐鲁医院住院治疗的334例ITP患者的临床资料，通过Logistic回归分析发生血栓的独立危险因素并构建预测模型。

334例ITP患者中40例（12.0％）发生血栓，其中男18例，女22例。动脉血栓、静脉血栓、混合血栓发生率分别为9.58％（32/334）、1.80％（6/334）和0.60％（2/334）。单因素分析显示，年龄、ITP病程>1年、合并高血压、合并冠心病、合并糖尿病、PLT均是ITP患者发生血栓的危险因素（P值均<0.05）。多因素分析显示，年龄、ITP病程>1年、合并冠心病、PLT均是发生血栓的独立危险因素（P值均<0.05）。列线图预测模型的受试者操作特征曲线的曲线下面积为0.80（95％CI：0.72～0.88），校准曲线显示列线图的预测与实际血栓发生率有良好的一致性，Hosmer-Lemeshow拟合优度检验显示χ2＝5.838，P＝0.665。

本研究构建了ITP患者发生血栓的列线图预测模型，有助于识别易发生血栓的高危患者。

## Linked entities

- **Diseases:** thrombosis (MONDO:0000831), coronary heart disease (MONDO:0005010), diabetes (MONDO:0005015)

## Full-text entities

- **Diseases:** coronary heart disease (MESH:D003327), ITP (MESH:D016553), hypertension (MESH:D006973), thrombosis (MESH:D013927), TIA (MESH:D002546), DVT (OMIM:612862), diabetes (MESH:D003920), venous thrombosis (MESH:D020246)
- **Chemicals:** TC (MESH:D013667), LDL-C (-), TG (MESH:D013866)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC13008554/full.md

## Figures

3 figures with captions in the complete paper: https://tomesphere.com/paper/PMC13008554/full.md

## References

22 references — full list in the complete paper: https://tomesphere.com/paper/PMC13008554/full.md

---
Source: https://tomesphere.com/paper/PMC13008554