Key result
Eight-variable perioperative model predicts post-neurosurgery VTE risk with an AUC of 0.85.
Why the study?
Venous thromboembolism occurs in 3.0% to 26% of neurosurgical patients, motivating the development of a clinical prediction model using machine learning and statistical methods.
Cohort (n=8,658)
Yes
Effect estimate: AUC 0.85 (95% CI 0.815-0.885)
A novel 8-variable clinical prediction model incorporating clinical factors and biomarkers accurately predicts the risk of venous thromboembolism following neurosurgery, potentially aiding in standardized primary VTE prevention.
No takes yet. Share an insight, caveat, or question.
May support targeted VTE prophylaxis after neurosurgery; leaves open whether implementation improves outcomes in prospective trials.
Liu et al. (2023) conducted a cohort in Venous thromboembolism following neurosurgery (n=8,658). Clinical prediction model for VTE (8 perioperative variables) was evaluated on Prediction of objectively confirmed venous thromboembolism (VTE) after neurosurgery (AUC 0.85, 95% CI 0.815-0.885). A clinical prediction model incorporating eight perioperative variables successfully predicted the risk of venous thromboembolism following neurosurgery, achieving an AUC of 0.85 in the external validation cohort.
Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context: