# Development of a prediction model for lower limb deep vein thrombosis in critically ill patients after intracranial hemorrhage at high altitude: a retrospective study

## TL;DR

This study identifies risk factors for deep vein thrombosis in ICU patients after brain hemorrhage at high altitude and creates a predictive model.

## Contribution

A novel predictive model for LDVT in high-altitude ICU patients, incorporating hypoalbuminemia, D-dimer levels, and aboriginal residency.

## Key findings

- Hypoalbuminemia and elevated D-dimer levels are independent risk factors for LDVT.
- Aboriginal residency at high altitudes (≥ 4,500 m) is a protective factor against LDVT.
- The predictive model shows strong performance with an AUC of 0.815.

## Abstract

To analyze the risk factors for lower limb deep vein thrombosis (LDVT) in critically ill patients admitted to the intensive care unit (ICU) following intracranial hemorrhage at high altitude, and to establish a predictive model.

A retrospective analysis was conducted on 359 patients who underwent surgery for intracranial hemorrhage and were admitted to the ICU of Xizang Autonomous Region People’s Hospital between August 1, 2021, and December 31, 2023. Patients were categorized into an LDVT group (n = 86) and a non-LDVT group (n = 273) based on the occurrence of LDVT during their ICU stay. Demographic data, comorbidities, laboratory results, and treatment approaches were compared between the groups. Independent risk factors were identified through univariate and multivariate logistic regression analyses. A risk prediction model was developed using R software, and its performance was internally validated.

Compared to the non-LDVT group, patients in the LDVT group had significantly longer ICU and overall hospital stays (p < 0.05). Multivariate logistic regression revealed hypoalbuminemia and elevated D-dimer levels as independent risk factors for LDVT, while aboriginal residency in high-altitude areas (≥ 4,500 m) was identified as a protective factor. A nomogram incorporating these variables was constructed. Internal validation demonstrated strong agreement between predicted and observed outcomes, with the area under the receiver operating characteristic (ROC) curve reaching 0.815 (95% CI [0.761–0.870]). The Hosmer-Lemeshow test indicated a good model fit (p = 0.088).

LDVT significantly prolongs ICU and hospital stays in critically ill patients. Hypoalbuminemia and elevated D-dimer levels are independent risk factors for LDVT, whereas aboriginal residency at high altitudes (≥ 4,500 m) serves as a protective factor. The developed risk prediction model shows strong predictive performance.