[Prevention of venous thromboembolism in polytraumatized patients. Epidemiology and importance]

Unfractionated heparin, low-molecular-weight heparin (LMWH), mechanical compression, and vena cava filters are part of a large panel of chemical or physical methods proposed to trauma patients as prophylaxis against venous thromboembolism. This prophylactic strategy is based on a 1961 autopsy survey showing a 16.6% rate of pulmonary embolism in this population. The objective of this study was to assess the incidence of deep venous thrombosis (DVT) and pulmonary embolism (PE) in multiple trauma patients.

A Medline and Current Contents search was conducted for prospective studies including trauma patients with ISS > 9 whose incidence of DVT and PE was evaluated by contrast venography and/or duplex scan and by ventilation/perfusion lung scan and/or pulmonary angiography and/or autopsy, respectively.

Twelve studies met the selection criteria for DVT. Among the global population of 2,374 trauma patients (14% of all admissions) 47% suffered lower limb injury and 17% had severe head injury. Overall incidence of DVT was about 20%. It was about 38% in patients without prophylaxis (range 2 to 61.5%) and about 13% in patients with prophylaxis (range 0.8 to 37%). Similar variations were observed for proximal DVT. In comparative studies (unfractionated heparin versus LMWH or versus mechanical compression devices), the incidence of thromboembolic events varied from 3.2 to 44% in patients given unfractionated heparin, 0.8 to 31% in those given LMWH, and 3.1% to 12% with mechanical compression. Thirteen studies met the selection criteria for PE and included an overall population of 4,245 trauma patients where the diagnosis of PE was suspected only if the patient had clinical signs. The incidence of PE and fatal PE was about 1.4% and 0.3% respectively. Only one study systematically studied the presence of PE and showed an incidence of 18.7% for PE in a population of 32 patients.

The current literature reports wide variability in methodology, characteristics of the study population and prophylaxis. These differences explain the wide variability from one study to another in the risk factors for venous thrombosis identified by univariate and multivariate analysis.

Current methodological imperfections make it impossible to correctly assess the incidence of DVT and PE in the multiple trauma population. Studies with a rigorous methodology using a precise stratification of the trauma injuries are required to determine the real risk for DVT/PE in trauma patients and to assess the impact of early systematic prophylaxis.