E-ISSN: 1019-5157
ISSN: 2651-5024
Research
Development of an Early Predictive Model for Lower Extremity Deep Vein Thrombosis Following Intracerebral Hemorrhage Surgery
Wuchang Wang✉ ,
Youjie Yan ,
Wenyong Gao ,
Xiaodong Wang
Neurosurgery, Northern Jiangsu People's Hospital Affiliated to Yangzhou University; Neurosurgery, Northern Jiangsu People's Hospital
DOI: 10.5137/1019-5149.JTN.50525-26.7
Accepted: 20/06/2026
Article in Press
Corresponding Author:
Wuchang Wang (wwc15530856310@163.com)
Abstract
Aim
We sought to clarify the relationships between postoperative lower extremity deep vein thrombosis (LEDVT) and perioperative laboratory and clinical indicators in patients with intracerebral hemorrhage (ICH). We aimed to identify independent determinants of LEDVT and develop a predictive model to guide early preventive interventions.
Material and Methods
This retrospective analysis included 371 patients who underwent surgical treatment for ICH in our hospital between January 2019 and January 2025. Based on the results of postoperative venous ultrasound, patients were classified into a deep vein thrombosis (DVT) group (n = 67) and a non-DVT group(n = 304). We examined the differences in perioperative clinical variables between the two groups. Candidate risk factors were initially explored through univariate analysis. We then performed multivariate logistic regression to identify independent predictors. We established a nomogram-based prediction model and evaluated its predictive ability using calibration curves and decision curve analysis.
Results
Postoperative LEDVT occurred in 18.1% of the patients. Univariate analysis identified significant associations with age, body weight, the presence of ventricular hemorrhage, preoperative activated partial thromboplastin time (APTT), and preoperative levels of D-dimer, C-reactive protein (CRP), and fibrinogen (FIB). Multivariate logistic regression identified preoperative D-dimer, CRP, and FIB levels, APTT values, and the presence of ventricular hemorrhage as independent risk factors. Our predictive nomogram that incorporated these variables showed good discrimination, and a receiver operating characteristic area under the curve of 0.829.
Conclusion
We found preoperative ventricular hemorrhage, APTT, and D-dimer, CRP, and FIB levels to be independent predictors of LEDVT after ICH surgery. The nomogram developed in this study may serve as a useful tool for early risk assessment and individualized prevention of postoperative LEDVT.
We sought to clarify the relationships between postoperative lower extremity deep vein thrombosis (LEDVT) and perioperative laboratory and clinical indicators in patients with intracerebral hemorrhage (ICH). We aimed to identify independent determinants of LEDVT and develop a predictive model to guide early preventive interventions.
Material and Methods
This retrospective analysis included 371 patients who underwent surgical treatment for ICH in our hospital between January 2019 and January 2025. Based on the results of postoperative venous ultrasound, patients were classified into a deep vein thrombosis (DVT) group (n = 67) and a non-DVT group(n = 304). We examined the differences in perioperative clinical variables between the two groups. Candidate risk factors were initially explored through univariate analysis. We then performed multivariate logistic regression to identify independent predictors. We established a nomogram-based prediction model and evaluated its predictive ability using calibration curves and decision curve analysis.
Results
Postoperative LEDVT occurred in 18.1% of the patients. Univariate analysis identified significant associations with age, body weight, the presence of ventricular hemorrhage, preoperative activated partial thromboplastin time (APTT), and preoperative levels of D-dimer, C-reactive protein (CRP), and fibrinogen (FIB). Multivariate logistic regression identified preoperative D-dimer, CRP, and FIB levels, APTT values, and the presence of ventricular hemorrhage as independent risk factors. Our predictive nomogram that incorporated these variables showed good discrimination, and a receiver operating characteristic area under the curve of 0.829.
Conclusion
We found preoperative ventricular hemorrhage, APTT, and D-dimer, CRP, and FIB levels to be independent predictors of LEDVT after ICH surgery. The nomogram developed in this study may serve as a useful tool for early risk assessment and individualized prevention of postoperative LEDVT.
Keywords
lower extremity deep vein thrombosis
intracerebral hemorrhage
independent predictor analysis
prediction model