Thromboprophylaxis in patients with acute spinal injuries: an evidence-based analysis.

BACKGROUND: The increased risk for venous thromboembolic events following spinal trauma is well established. The purpose of the present study was to examine the literature in order to determine the optimum thromboprophylaxis regimen for patients with acute spinal injuries with or without spinal cord injury. METHODS: EMBASE, MEDLINE, and Cochrane databases were searched from the earliest available date to April 2008 for clinical trials comparing different methods of thromboprophylaxis in adult patients following acute spinal injuries (with or without spinal cord injury). Outcome measures included the prevalences of deep-vein thrombosis and pulmonary embolism and treatment-related adverse events. RESULTS: The search yielded 489 studies, but only twenty-one of them fulfilled the inclusion criteria. The prevalence of deep-vein thrombosis was significantly lower in patients without spinal cord injury as compared with patients with spinal cord injury (odds ratio = 6.0; 95% confidence interval = 2.9 to 12.7). Patients with an acute spinal cord injury who were receiving oral anticoagulants had significantly fewer episodes of pulmonary embolism (odds ratio = 0.1; 95% confidence interval = 0.01 to 0.63) than those who were not receiving oral anticoagulants (either untreated controls or patients managed with low-molecular-weight heparin). The start of thromboprophylaxis within the first two weeks after the injury resulted in significantly fewer deep-vein-thrombosis events than delayed initiation did (odds ratio = 0.2; 95% confidence interval = 0.1 to 0.4). With regard to heparin-based pharmacoprophylaxis in patients with spinal trauma, low-molecular-weight heparin significantly reduced the rates of deep-vein thrombosis and bleeding episodes in comparison with the findings in patients who received unfractionated heparin, with odds ratios of 2.6 (95% confidence interval = 1.2 to 5.6) and 7.5 (95% confidence interval = 1.0 to 58.4) for deep-vein thrombosis and bleeding, respectively. CONCLUSIONS: The prevalence of deep-vein thrombosis following a spine injury is higher among patients who have a spinal cord injury than among those who do not have a spinal cord injury. Therefore, thromboprophylaxis in these patients should start as early as possible once it is deemed safe in terms of potential bleeding complications. Within this population, low-molecular-weight heparin is more effective for the prevention of deep-vein thrombosis, with fewer bleeding complications, than unfractionated heparin is. The use of vitamin K antagonists appeared to be effective for the prevention of pulmonary embolism.

Traumatic spinal cord injury produced by controlled contusion in mouse.

Previous work from this laboratory has described a rat spinal cord injury (SCI) model in which the mid-thoracic spinal cord is subjected to a single rapid and calibrated displacement at the site of a dorsal laminectomy. Injury is initiated at the tip of a vertical shaft driven by an electromagnetic shaker. Transducers arranged in series with the shaft record the patterns of displacement and force during the impact sequence. In the present study, this device and the relevant surgical procedures were adapted to produce a spinal contusion injury model in laboratory mice. The signal generator for the injury device has also been converted to a computer-controlled interface to permit extension of the model to other laboratories. Mice were subjected to SCI across a range of severities by varying the amplitude of displacement and the magnitude of measured preload force on the dural surface. A moderate injury produced by displacement of 0.5 mm over 25 msec resulted in initial paralysis and recovery of locomotion with chronic deficits in hindlimb function. The magnitude of the peak force, impulse, power, and energy generated at impact were correlated with behavioral outcome at 1 day postinjury, while peak displacement and impulse were the best predictors of behavioral outcome at 28 days postinjury. The shape of the force recording proved to be a highly sensitive measure of subtle variations in the spinal compartment that were otherwise difficult to detect in this small species. The results demonstrate that the electromagnetic spinal cord injury device (ESCID) can be used to produce a well-controlled contusion injury in mice. The unique features of controlled displacement and monitoring of the biomechanical parameters at the time of impact provide advantages of this model for reducing outcome variability. Use of this model in mice with naturally occurring and genetically engineered mutations will facilitate understanding of the molecular mechanisms of pathophysiology following traumatic spinal cord injury.