Perioperative medicine & outcomes
A Risk Assessment Model for Predicting Perioperative Venous Thromboembolism in Patients Receiving Surgery under Anesthesia Care
Authors' conclusion
We developed a clinically intuitive risk assessment model that predicts perioperative VTE across diverse surgical populations, based on the Agency for Healthcare Research and Quality's definition. This model demonstrates superior performance compared to existing instruments
Study context
- Population
- Surgical patients at two US tertiary care hospitals (Montefiore Medical Center, Beth Israel Deaconess Medical Center)2,647 (0.8%) with venous thromboembolism (VTE)
- Participants
- 319,134 patients
- Exposure
- New VTE risk model vs Caprini score and Rogers risk assessment model
- Outcome
- Perioperative VTE (new ICD code for DVT or PE plus VTE-related imaging order), from admission through discharge; model discrimination (AUC)
Methodological detailsprediction model development and validation, two tertiary care hospitals, internal temporal validation, external validation, stepwise backward logistic regression, bootstrap resampling, Performance assessed by ROC curves and Brier score, Development: Montefiore 2016–2021; internal temporal validation: 2021–2023
- Prediction model development and validation
- Two tertiary care hospitals
- Internal temporal validation
- External validation
- Stepwise backward logistic regression
- Bootstrap resampling
- Performance assessed by ROC curves and Brier score
- Development: Montefiore 2016–2021; internal temporal validation: 2021–2023
Main finding
As reported in the abstract
Discrimination for perioperative VTE (AUC)
| Group | AUC, development cohort | AUC, internal temporal validation cohort | AUC, external validation cohort | Predictive accuracy |
|---|---|---|---|---|
| Model | 0.87 (95% CI, 0.86 to 0.89) | 0.84 (95% CI, 0.81 to 0.87) | 0.76 (95% CI, 0.75 to 0.77) | — not stated in the abstract |
| Caprini score | — not stated in the abstract | — not stated in the abstract | — not stated in the abstract | 0.66 |
| Rogers risk assessment model | — not stated in the abstract | — not stated in the abstract | — not stated in the abstract | 0.51 |
AUC, development cohort
Model0.87 (95% CI, 0.86 to 0.89)
Caprini score— not stated in the abstract
Rogers risk assessment model— not stated in the abstract
AUC, internal temporal validation cohort
Model0.84 (95% CI, 0.81 to 0.87)
Caprini score— not stated in the abstract
Rogers risk assessment model— not stated in the abstract
AUC, external validation cohort
Model0.76 (95% CI, 0.75 to 0.77)
Caprini score— not stated in the abstract
Rogers risk assessment model— not stated in the abstract
Predictive accuracy
Model— not stated in the abstract
Caprini score0.66
Rogers risk assessment model0.51
— not stated in the abstract
Observational data: associations, not effects.
Further findings
Prediction score performance before and after surgery
- AUC in patients before surgery
- 0.91; 95% CI, 0.89 to 0.93
- AUC in patients after surgery
- 0.84; 95% CI, 0.82 to 0.86
Original publication
A Risk Assessment Model for Predicting Perioperative Venous Thromboembolism in Patients Receiving Surgery under Anesthesia Care. Anesthesiology. Published 3 April 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →