A nomogram model can predict the risk of venous thromboembolism in postoperative patients with gynecological malignancies

Objective To explore independent factors influencing the risk of lower extremity deep vein thrombosis during the postoperative period in patients with gynecological malignancies by constructing a predictive model. Methods In our study, we collected 573 patients with gynecological malignancies in the postoperative period between September 2016 and September 2020, who were divided into a modeling (n = 402) and verification group (n = 171) according to a ratio of 7:3. Univariate and multivariate regression analyses were used to determine independent factors influencing deep vein thrombosis (DVT). A nomogram model was created and a risk score was calculated. Results Multivariate regression analysis showed that the independent factors affecting DVT among these patients included age, hyperlipidemia, abnormal uterine bleeding, degree of anemia, D-dimer, operation time, and intraoperative blood loss. By incorporating these factors into a nomogram, we determined that the C-index and calibration curve of the two groups both showed that the model distinguishes and fits well. Further comparing between the high- and low-risk groups, we found that the model has favorable predictive performance. Conclusion The predictive nomogram for the risk of DVT in patients with gynecological malignancies in the postoperative period demonstrated good calibration and recognition accuracy. Further independent research is necessary to verify our results.