Prevalence and Predictors for Preoperative Deep Vein Thrombosis in Patients with Thoracolumbar Fractures Caused by High-Energy Injuries.

OBJECTIVE

To determine the prevalence of preoperative deep vein thrombosis (DVT) and to identify the predictors for preoperative DVT in patients with thoracolumbar fractures caused by high-energy injuries.

METHODS

A total of 429 patients with thoracolumbar fractures caused by high-energy injuries from October 2016 to November 2019 were included in this retrospective study. The patients underwent ultrasonography before operation and were divided into the DVT group and non-DVT group. Demographic data and laboratory results were collected, and mechanical and chemical prophylaxis were performed to prevent DVT. The incidence of DVT was evaluated and the predictors of preoperative DVT were identified.

RESULTS

The overall incidence of preoperative DVT was 14.45% (62/429). Of all patients, 1 (0.23%) had proximal DVT and 61 (14.22%) had distal DVT. The incidence of preoperative DVT was 24.59% in patients with thoracic fractures and 11.04% in those with lumbar fractures. Multivariate logistic regression identified 3 predictors, including lower extremity motor (P < 0.001), time from injury to operation (P < 0.001), and D-dimer (P = 0.001). Lower extremity motor score <3, time from injury to operation >3 days, and D-dimer >1.81 mg/L showed satisfied accuracy for predicting preoperative DVT. The receiver operating characteristic curve analysis indicated the diagnostic value of D-dimer was the highest when the area under the receiver operating characteristic curve curve was 0.769.

CONCLUSIONS

Despite mechanical and chemical prophylaxis, the incidence of preoperative DVT is still very high, especially in thoracic fractures. Lower extremity motor score <3, time from injury to operation >3 days, and D-dimer >1.81 mg/L are predictors for preoperative DVT. More attention is warranted to prevent preoperative DVT when ≥2 predictors are present.