A Risk Assessment Model for Predicting Perioperative Venous Thromboembolism in Patients Receiving Surgery under...

Area of Science:

• Medical research
• Clinical informatics
• Patient safety

Background:

• Perioperative venous thromboembolism (VTE) significantly increases patient morbidity, mortality, and healthcare costs.
• Accurate risk assessment is crucial for identifying at-risk surgical patients.
• Existing VTE risk models require enhancement for improved clinical utility.

Purpose of the Study:

• To develop and validate a novel risk assessment model for perioperative VTE.
• To align the VTE prediction model with the Agency for Healthcare Research and Quality's Patient Safety Indicator 12 criteria.
• To enhance early identification and targeted prevention strategies for VTE.

Main Methods:

• Retrospective analysis of registry data from 319,134 surgical patients across two US tertiary care hospitals.
• Development of the prediction model using data from 2016-2021 and internal temporal validation using data from 2021-2023.
• Classification of perioperative VTE based on ICD codes and VTE-related imaging orders; model development via stepwise backward logistic regression and bootstrap resampling.

Main Results:

• The developed model demonstrated robust discriminatory performance with an AUC of 0.87 in the development cohort and 0.84 in the internal validation cohort.
• The model significantly outperformed existing risk assessment tools, including the Caprini score (AUC 0.66) and Rogers model (AUC 0.51).
• The prediction score showed strong performance for VTE prediction both pre-operatively (AUC 0.91) and post-operatively (AUC 0.84).

Conclusions:

• A clinically intuitive risk assessment model for perioperative VTE has been successfully developed and validated.
• The new model exhibits superior predictive accuracy compared to current instruments for VTE risk stratification.
• This advanced model holds significant potential for improving VTE prevention strategies in hospitalized surgical patients.