Early (<24 hours) venous thromboembolism prophylaxis in traumatic brain injury: development of the EViTBI algorithm using a national cohort
Heather X Rhodes-Lyons, Lucy Martinek, Christopher Ruiz, Antonio PepeBackground
Optimal timing of venous thromboembolism (VTE) prophylaxis after traumatic brain injury (TBI) remains uncertain because of competing concerns regarding thromboembolism prevention and hemorrhagic progression. National recommendations remain inconsistent, and prophylaxis is frequently delayed beyond 72 hours.
Objective
To develop and internally validate the early VTE prophylaxis in TBI (EViTBI) algorithm for identifying a reproducible low-risk cohort potentially eligible for early (<24 hours) pharmacologic prophylaxis.
Methods
A retrospective cohort study was performed using the American College of Surgeons Trauma Quality Improvement Program Participant Use File (2017 to 2021). Adults with isolated TBI and complete prophylaxis-timing data were included. Multivariable logistic regression generated adjusted ORs for mortality, deep vein thrombosis (DVT), and pulmonary embolism (PE). Adjusted ORs were converted to weighted log-odds (mortality×2; VTE outcomes×1 negative) and summed to create the EViTBI score. Expert-derived thresholds were compared with receiver operating characteristic (ROC) and Youden Index analyses. Discrimination was assessed using the c-statistic.
Results
Among 58 418 patients, 7590 (12.9%) received early prophylaxis. EViTBI classified 44.5% as low risk, 29.9% as moderate risk, and 25.5% as high risk. Compared with high-risk patients, low-risk patients had lower mortality (OR 0.15; 95% CI 0.13 to 0.17), DVT (OR 0.19; 95% CI 0.16 to 0.23), and PE (OR 0.23; 95% CI 0.17 to 0.31). Multivariable model discrimination was strong (c=0.740.75). The composite EViTBI score demonstrated acceptable discrimination for mortality (AUC=0.68). ROC and Youden analyses supported the expert-derived thresholds.
Conclusion
EViTBI identified nearly half of isolated TBI patients as a reproducible low-risk cohort in whom early prophylaxis was associated with lower mortality and VTE rates. Findings are associative and hypothesis-generating; prospective validation incorporating serial imaging and prophylaxis fidelity is required before clinical implementation.
Level of evidence
Prognostic study, level III.