Establishing a prediction model for lower extremity deep venous thrombosis in emergency inpatients in the post epidemic era.

OBJECTIVE

This study aimed to analyze the risk factors of lower extremity deep vein thrombosis (LEDVT) in emergency inpatients in the post-epidemic era, and to establish a prediction model for identifying high-risk patients of LEDVT.

METHODS

Emergency inpatients admitted to our hospital from June 2022 to June 2023 were divided into two groups: the epidemic group and the post-epidemic group. The baseline characteristics, blood routine, liver and kidney function, blood coagulation function, and LE ultrasonography were compared between the two groups. Multivariate logistic analysis and receiver operating character (ROC) curve were used to establish and evaluate the effectiveness of a prediction model for LEDVT in the post-epidemic era.

RESULTS

A total of 967 patients were analyzed, including 388 cases in the epidemic group and 579 cases in the post-epidemic group. The portion of LEDVT cases in the post-epidemic group (33.2%) was significantly higher than that in the epidemic group (26.8%,  = 0.036). Binary Logistic regression analysis showed that age, smoking history, drinking history and glycosylated hemoglobin (HBA1c) were independent risk factors for thrombosis. The prediction model was established as  = 0.863 × age + 0.978 × smoking history + 0.702 × drinking history + 0.104 × HBA1c - 2.439. The area under the ROC curve was 0.718.

CONCLUSION

The incidence of LEDVT in emergency inpatients in the post-epidemic era was significantly higher than that in the epidemic period. Age, smoking and drinking history, and glycosylated hemoglobin are at high risk for thrombosis.