INTRODUCTION — Risk categories for venous thromboembolism (VTE; deep vein thrombosis [DVT] or pulmonary embolism [PE]) defined by the American College of Chest Physicians (table 1) include moderate risk (baseline risk in the absence of prophylaxis of ≥3 percent) and high risk (ie, baseline risk in the absence of prophylaxis ≥6 percent) . A systematic review reported an overall incidence of VTE at 12 percent in trauma patients who received no prophylaxis and 7 percent in those who received only mechanical prophylaxis . However, estimates of VTE risk vary widely in this population [3-12]. Higher rates (60 percent in one study ) likely reflect patients with multiple serious injuries (eg, traumatic brain and spinal injury) or need for multiple surgeries. Nevertheless, patients sustaining traumatic injuries are at least at moderate risk for VTE with severely injured patients at high risk for VTE [3,13,14]. As such, thromboprophylaxis is indicated for hospitalized trauma patients. Even with appropriate prophylaxis, VTE remains a leading cause of mortality . (See "Prevention of venous thromboembolic disease in adult nonorthopedic surgical patients", section on 'Baseline thrombosis risk'.)
While pharmacologic thromboprophylaxis is desirable for decreasing the rates of DVT and PE, a clinical decision to initiate VTE prophylaxis must balance the risk of bleeding related to the patient's injuries with the risk of thrombus formation .
The impact of severe traumatic injury on risk, clinical evaluation, and decision making for VTE prevention in this population is reviewed here. General considerations and recommendations are reviewed separately. (See "Prevention of venous thromboembolic disease in adult nonorthopedic surgical patients" and "Prevention of venous thromboembolism in adults undergoing hip fracture repair or hip or knee replacement".)
PATHOPHYSIOLOGY AND RISK FACTORS — The hereditary and other acquired causes of venous thrombosis are similar to those of the general population. (See "Overview of the causes of venous thrombosis".)
However, the occurrence of trauma introduces additional risk factors for VTE. Traumatic injury is a classic example of Virchow's triad (endothelial injury, venous stasis, hypercoagulability). But there is evidence that traumatic injury, particularly severe injury, also activates a prothrombotic state, which has been attributed to many factors, including decreased levels of functional protein C or abnormal antithrombin levels [17,18].
Risk factors — The Caprini model (calculator 1), while useful for many surgical populations, has not been validated in the trauma population. Several models have sought to determine the risk factors for VTE specific to the trauma population [3,19,20]. While knowledge of risk factors is important, prediction of events in trauma patients remains difficult. (See 'Surveillance' below.)
Accepted risk factors for thromboembolism in injured patients include spinal cord injury, head trauma, lower extremity fracture, pelvic fracture, need for surgical intervention, increasing age, femoral vein line insertion, surgical repair venous injuries, prolonged immobilization, prolonged hospital stay, and high injury severity score [3,13,21-25]. Other risk factors may also be important [19,20,25-27]. Risk factors identified in two of the larger observational studies are presented below.
●In a review of 1602 VTE episodes from the National Trauma Data Bank, the following were identified as independent risk factors for VTE :
•Ventilator days >3 (odds ratio [OR] 8.08, 95% CI 6.86-9.52)
•Venous injury (OR 3.56, 95% CI 2.22-5.72)
•Age ≥40 years (OR 2.01, 95% CI 1.74-2.32)
•Lower extremity fracture with abbreviated injury score (AIS) ≥3 (OR 1.92, 95% CI 1.64-2.26)
•Major operative procedure (OR 1.53, 95% CI 1.30-1.80)
•Head injury with AIS ≥3 (OR 1.24, 95% CI 1.05-1.46)
●In a later review of 1233 patients, the original Greenfield risk assessment profile (RAP) score was reevaluated and simplified to include only five risk factors . Patients were separated into those with and without VTE. The groups were similar in age, sex, mechanism, and mortality, but injury severity and RAP scores were higher in those with VTE. The VTE group also had more transfusions and a longer time to VTE prophylaxis. The risks for VTE relative to no VTE for the individual factors were as follows:
•≥4 transfusions in the first 24 hours (OR 2.60, 95% CI 1.64-4.13)
•Glasgow Coma Score <8 for >4 hours (OR 2.13, 95% CI 1.28-3.54)
•Pelvic fracture (OR 2.26, 95% CI 1.44-3.57)
•>2 hour operation (OR 1.80, 95% CI 1.14-2.85)
•40 to 59 years of age (OR 1.70, 95% CI 1.10-2.63)
There is often poor correlation between incidence of deep vein thrombosis (DVT) and pulmonary embolism (PE), with many patients presenting with PE having no evidence of extremity thrombosis [4,20,28-30]. Whether the risk factors for DVT differ from those associated with PE in the trauma population is not well studied. Studies evaluating chest imaging (admission computed tomography scan or others early in admission) have found a fairly high incidence of PE early following traumatic injury [31-33]. Some evidence suggests that some early PEs identified on imaging may be a result of the chest injury rather than a true thromboembolic event.
●In one multicenter cohort study, differing risk factors were found for DVT versus PE among 1822 severely injured blunt trauma patients [29,30].
•For DVT, independent risk factors were delay of thromboprophylaxis >48 hours after injury (OR 0.57, 95% CI 0.36-0.90) and thoracic AIS score ≥3 (OR 1.82, 95% CI 1.12-2.95).
•For PE, independent risk factors were serum lactate >5 (OR 2.33, 95% CI 1.43-3.79) and male sex (OR 2.12, 95% CI 1.17-3.84).
●In a smaller study of 110 patients with PE following injury, long bone fracture, admission to a ward (not intensive care unit), and female sex were associated with an increased risk for early PE (<4 days), whereas severe head injury, severe chest injury, and major surgery within 48 hours of admission were associated with later PE.
CLINICAL FEATURES AND DIAGNOSIS — Maintaining a high level of suspicion for VTE is important when caring for trauma patients. Symptoms and signs of VTE in trauma patients are often masked by those related to the patient's injuries, which complicates detection. Extremity pain and swelling are commonly encountered. Similarly, tachycardia and desaturation events are common in trauma patients due to atelectasis, pulmonary contusion, or other injuries.
When VTE is suspected, further evaluation is always warranted. The diagnosis of VTE is discussed elsewhere. (See "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity" and "Clinical presentation, evaluation, and diagnosis of the nonpregnant adult with suspected acute pulmonary embolism".)
THROMBOPROPHYLAXIS — As a population at high risk for VTE, early, aggressive thromboprophylaxis is warranted for trauma patients [2,34]. We continue prophylaxis until hospital discharge for all high-risk patients and beyond hospital discharge for patients with significant orthopedic injuries that limit mobility. The typical strategies for preventing VTE (pharmacologic and nonpharmacologic methods) are similarly used in trauma patients. These methods are described in detail separately. It is important for the clinician to be aware that VTE prophylaxis reduces but does not eliminate the risk of VTE entirely. (See "Prevention of venous thromboembolic disease in adult nonorthopedic surgical patients", section on 'Administration'.)
Approach — The optimal approach to thromboprophylaxis in patients with specific traumatic injuries remains incompletely defined [25,35,36]. The method to use for a particular patient depends greatly upon the nature and severity of injuries (algorithm 1) and the presence of conditions that may contraindicate their use [2,37-43]. When possible, all hospitalized patients with traumatic injuries should receive at least one mode of prophylaxis. We generally use a combination of lower extremity mechanical compression and low-molecular-weight (LMW) heparin, depending upon the perceived risk. Patients at risk who do not have a contraindication to antithrombotic therapy should receive pharmacologic prophylaxis irrespective of their mobility. We suggest not routinely placing prophylactic inferior vena cava (IVC) filters; however, it is reasonable to use IVC filters in very-high-risk patients, particularly if there are ongoing contraindications to anticoagulation. (See 'Prophylactic IVC filters' below.)
A systematic review and meta-analysis of four randomized trials of thromboprophylaxis in trauma patients found a significantly reduced risk for deep vein thrombosis (DVT) in patients receiving prophylaxis compared with no prophylaxis (4 versus 9 percent; relative risk [RR] 0.52, 95% CI 0.32-0.84) . The reduction in pulmonary embolism (PE) was not statistically significant (1.7 versus 3.3 percent; RR 0.65, 95% CI 0.29-1.43). Mortality was low in both groups (1 versus 1.4 percent; RR 0.59, 95% CI 0.2-1.70). Other findings included:
●Mechanical prophylaxis reduced the risk of DVT relative to no prophylaxis (five trials; RR 0.43, 95% CI 0.25-0.73).
●Pharmacological prophylaxis reduced the risk of DVT relative to mechanical prophylaxis (six trials; RR 0.48, 95% CI 0.25-0.95) but had a higher risk for minor (RR 2.37, 95% CI 1.13-4.98), but not major, bleeding.
●LMW heparin reduced the risk of DVT relative to unfractionated heparin (two trials; RR 0.68, 95% CI 0.50-0.94) with similar rates of bleeding.
●Combined (mechanical and pharmacological prophylaxis) reduced the risk of DVT relative to pharmacologic prophylaxis alone (three trials; RR 0.34, 95% CI 0.19-0.60).
The risk of VTE rises sharply if treatment is delayed beyond 72 to 96 hours . If pharmacologic prophylaxis is contraindicated for a period that will extend beyond this time frame, placement of a prophylactic IVC filter may be appropriate . (See 'Prophylactic IVC filters' below.)
While new agents for VTE prophylaxis are emerging (eg, antiplatelet agents), these remain under study and are not routinely used in the trauma population . There are no randomized controlled trials comparing traditional anticoagulants with newer anticoagulants or antiplatelet agents in the trauma population, but these are expected in the future. (See "Prevention of venous thromboembolic disease in adult nonorthopedic surgical patients".)
Agent and dose — For most trauma patients, we suggest LMW heparin rather than unfractionated heparin (UFH) for VTE prophylaxis. Data suggest that LMW heparin may be superior to UFH for the prevention of DVT in patients with major trauma [46-52]. This recommendation is consistent with the experience at the authors’ institution guidelines from the American College of Chest Physicians (ACCP) that state a preference for LMW heparin in the trauma population . UFH is preferred in those with severe renal insufficiency (eg, creatinine clearance <20 to 30 mL/min) (table 2), while fondaparinux is preferred in those with heparin-induced thrombocytopenia. (See "Prevention of venous thromboembolic disease in adult nonorthopedic surgical patients".)
●Low-dose UFH, with activity against factor Xa and thrombin, was one of the first pharmacologic agents used in the prevention of VTE and has traditionally been considered a safe and effective option in trauma patients, particularly those with high risk of bleeding, due to its reversibility.
●LMW heparin is derived from UFH and has better bioavailability and a longer half-life than UFH. Its activity is more effective against factor Xa, and it has less effect on platelets and microvascular permeability, potentially resulting in less bleeding .
LMW heparin is more effective than the traditional twice-daily dosing of UFH for reducing VTE rates in high-risk populations [2,46]. In a meta-analysis of two randomized trials in patients with major trauma, LMW heparin reduced the risk of DVT compared with UFH (25 versus 37 percent; RR 0.68, 95% CI 0.50-0.94) with similar rates of bleeding. Observational studies have similarly reported lower rates of DVT, lower rates of PE, and, in one study, lower mortality in patients treated with LMW heparin compared with UFH [48,49]. Most of these studies used standard twice-daily UFH dosing of LMW heparin in the trauma population. In one randomized trial comparing LMW heparin with UFH given three times per day, rates of DVT were not significantly different (5.1 versus 8.2 percent; RR 0.62, 95% CI 0.3-1.28) . Adverse events were similar. In an observational study from a single institution, rates of DVT were similar before and after changing the VTE prophylaxis protocol from twice-daily LMW heparin to three times per day UFH (7 percent in both groups) . Rates of PE and bleeding were also similar.
Monitoring and dose adjustment — Achieving adequate VTE prophylaxis remains challenging in the trauma population for a variety of reasons. For LMW heparin, monitoring anti-Xa levels has been proposed to aid dose adjustments; however, studies suggest that the efficacy of using of anti-Xa levels, like monitoring and adjustment of UFH, may be variable across subpopulations [56-61]. In a systematic review, overall 37 percent of patients achieved prophylactic anti-Xa levels using a variety of dosing regimens . The average time to achieving prophylactic anti-Xa levels was 28 hours. While there is likely a role for anti-Xa-based monitoring and LMW heparin dose adjustment, the optimal protocol and specific trauma population that would benefit have not been determined.
●In a systematic review that included 45 studies of trauma or surgical patients with obesity, a variety of prophylactic (predominantly LMW heparin) dosing regimens were used, including weight-based, weight- or body mass index-stratified, and fixed-dose regimens . Target anti-Xa concentrations were more frequently achieved with weight-based or high fixed-dose regimens compared with standard fixed-dose regimens, but the incidence of DVT was similar between the groups. The high fixed-dose approach was associated with an increased risk of bleeding complications. In this study, anti-Xa levels higher than the target range were not associated with an increased incidence of bleeding, rather, most bleeding complications occurred when anti-factor Xa was within the desired range or below.
●In a later systematic review, three separate analyses were performed using 24 studies limited to trauma patients (excluding neurosurgery, burns and single system orthopedic injury) . Patients who received VTE prophylaxis and attained prophylactic anti-Xa levels had a reduced risk of developing VTE; however, titrating dose-adjustment strategies to achieve prophylactic anti-Xa levels paradoxically did not reduce VTE. The study authors speculated that adjustments targeting anti-Xa levels may be overly delayed to prevent VTE.
•Achieving prophylactic anti-Xa levels significantly reduced the odds of experiencing a clinical VTE compared with lower levels (3.1 versus 4.0 percent; odds ratio [OR] 0.52, 95% CI 0.28–0.95).
•Overall, 75 percent of patients receiving dose-adjusted enoxaparin based on anti-Xa levels achieved adequate prophylaxis (anti-Xa ≥0.2 IU/mL).
•Compared with standard enoxaparin dosing, dose-adjusted enoxaparin achieved prophylactic anti-Xa levels more frequently (66 versus 45 percent; OR 4.05, 95% CI 1.80–9.03), but without a significant reduction in VTE. Patients who received dose adjustment were four times more likely to have supraprophylactic anti-Xa level, but the observed difference in bleeding was not statistically significant.
Compression — When the risk of bleeding is judged to be too high for anticoagulation, we suggest a mechanical method for VTE prophylaxis, principally lower extremity intermittent pneumatic compression (IPC), unless there is a specific contraindication to its use (eg, lower extremity fractures, ischemia). IPC alone is effective and is better than no prophylaxis for reducing the incidence of VTE [2,63-65]. In a meta-analysis of five randomized trials, mechanical prophylaxis reduced the risk of DVT relative to no prophylaxis (4.5 versus 8.8 percent; RR 0.43, 95% CI 0.25-0.73) .
However, compliance with IPC is overall poor in the trauma population, underscoring the need to initiate pharmacologic prophylaxis as soon as feasible [66-68]. Switching to or adding a pharmacologic agent, such as LMW heparin, should be done as soon as the bleeding risk becomes acceptably low (eg, 48 to 72 hours following neurosurgery) or the coagulopathy has been reversed. (See 'Specific trauma populations' below.)
Concomitant mechanical prophylaxis can also be recommended for patients who are treated with pharmacologic prophylaxis. In a meta-analysis of three trials, combined mechanical and pharmacological prophylaxis reduced the risk of DVT relative to pharmacologic prophylaxis alone (RR 0.34, 95% CI 0.19-0.60) .
Prophylactic IVC filters — Based on large observational studies [69-71] and one randomized trial , for most severely injured trauma patients, we suggest not routinely placing prophylactic IVC filters in patients without known VTE. However, it is reasonable to use IVC filters in very-high-risk patients, particularly if there are ongoing contraindications to anticoagulation.
Prophylactic IVC filters (figure 1) (typically retrievable) are placed via percutaneous access into the inferior vena cava. The filter serves as a mechanical obstruction to thrombus traveling toward the pulmonary vasculature. The goal of prophylactic placement is to decrease the incidence of future PE. The possibility that prophylactic IVC filter placement may reduce PE gained early support given the high-risk nature and bleeding risks in the severely injured trauma population, particularly polytrauma patients. But while IVC filters potentially block thrombus originating in the lower half of the body from traveling proximally, they are ineffective at blocking thrombus originating from the upper extremities or neck. In addition, several complications have been described with indwelling IVC filters over time, including thrombus accumulation potentially leading to complete IVC thrombosis, filter migration, fracture, filter tilt, and vena cava perforation. (See "Placement of vena cava filters and their complications", section on 'Complications'.)
A clear consensus governing the use of prophylactic filters in trauma patients has yet to be reached. As examples, the Eastern Association for the Surgery of Trauma and Society of Interventional Radiology guidelines suggest prophylactic IVC filter use for high-risk patients with a contraindication to anticoagulation, but the ACCP guidelines do not [1,25,73]. A later multidisciplinary guideline from an expert panel of clinicians from radiology, cardiology, pulmonology, trauma surgery and vascular surgery also suggested avoiding routine prophylactic IVC filter placement in patients without known VTE . Evidence of the efficacy and safety of prophylactic IVC filter placement in trauma patients has predominantly come from observational studies [69,70,75-78], two small pilot studies [79,80] and a later randomized trial . The data highlight the lack of clear survival benefit for prophylactic IVC filter placement [69,70,72,75-77]. The incidence of PE is reduced when prophylactic IVC filters are used in patients who do not receive pharmacologic thromboprophylaxis, and while observational studies have noted an increased incidence of DVT in patients receiving IVC filters [69,70,78,81,82], this was not confirmed in a systematic review , or in the randomized trial described below . (See "Prevention of venous thromboembolic disease in adult nonorthopedic surgical patients", section on 'Prophylactic vena cava filters' and "Placement of vena cava filters and their complications".)
●In a multicenter trial that included 240 patients (median Injury Severity Score [ISS] of 27), the incidence of symptomatic PE or death at 90 days was similar in patients who underwent early IVC filter placement (within 72 hours of admission) compared with controls (13.9 versus 14.4 percent, hazard ratio [HR] 0.99, 95% CI 0.51-1.94) . The use and timing of prophylactic anticoagulation and dosing were at the discretion of the clinicians. Among the subgroup of patients who survived 7 days and did not receive prophylactic anticoagulation within 7 days after injury (46 patients in the prophylactic IVC filter group, 34 patients in the control group), the incidence of symptomatic PE between day 8 and day 90 was significantly lower for patients in the IVC filter group compared with controls (0 versus 14.7 percent; HR 0, 95% CI 0.00-0.55). All the PEs occurred between day 9 and day 19 after the injury. The incidence of lower extremity DVT was similar for prophylactic IVC filter and controls (11.4 versus 10.1 percent, relative risk 1.1, 95% CI 0.6-2.3).
Large observational studies provide further support for not routinely placing prophylactic IVC filters in trauma patients.
●In a study that analyzed data from a large multicenter trauma registry, hospitals were stratified into four quartiles according to the frequency of IVC filter use in trauma patients (ranging from 0.7 percent in the lowest quartile to 4.6 percent in the highest quartile) . Risk-adjusted mortality was similar in all four quartiles. In this study, rates of DVT were higher in patients who received prophylactic IVC filters compared with those who did not (5.5 versus 1.1 percent), and this association remained statistically significant after adjusting for known potential confounders such as injury severity and use of pharmacologic VTE prophylaxis (odds ratio [OR] 1.83, 95% CI 1.15-2.93).
●A retrospective single-institution study of 451 trauma patients who underwent IVC filter placement and 1343 matched controls reported that among patients who survived 72 hours, hospital mortality was similar in both groups after adjusting for age, sex, and injury severity (adjusted OR 1.13, 95% CI 0.65-1.97) . Mortality rates remained similar in both groups at 6 and 12 months postinjury.
●Lastly, a study analyzing data from several large clinical and administrative databases from 2003 to 2015 found that despite a precipitous decline in use of IVC filters (>90 percent prophylactic) during this period, rates of PE remained essentially unchanged .
Vena cava filters can be permanent or retrievable. Early studies involved treatment with permanent filters, but later studies in the trauma population reflect the growing popularity of retrievable filters, driven by the temporary nature of the post-traumatic prothrombotic state. However, the rate of filter retrieval is overall low. When a prophylactic IVC filter is placed, the trauma team should ensure that a protocol is in place to ensure that follow-up has been arranged for patients who receive a temporary IVC filter . (See "Placement of vena cava filters and their complications", section on 'Filter retrieval'.)
SURVEILLANCE — While trauma patients are at high risk for VTE, and many centers and some older trauma guidelines advocate surveillance (typically duplex ultrasound), we suggest not routinely obtaining imaging studies or other tests in an effort to identify occult deep vein thrombosis (DVT) in trauma patients [36,85]. Surveillance incurs additional expense, and there is no conclusive evidence that outcomes are improved . However, some argue that surveillance programs increase awareness of the potential for VTE but can lead to overdiagnosis [87-89]. In one comparative study between two hospitals with similar prophylaxis guidelines, one implemented a protocol of standard lower extremity ultrasound surveillance and the other did not . Although the rates of VTE based on clinical parameters were no different between the hospitals, the odds of receiving a diagnosis of DVT were 5.3 times higher in the hospital with the ultrasound surveillance program, suggesting a surveillance bias in centers with routine surveillance programs.
Some have suggested other tests to identify trauma patients at the highest risk for VTE based upon specific risk factors [26,90-93], including D-dimer [94,95], or thromboelastography [96-98]. Whether such identified patients would benefit from these methods needs to be validated in prospective studies. (See 'Risk factors' above.)
As an example, one predictive model, which was created using risk factors derived from an institution with a liberal duplex protocol (intensive care unit length of stay, transfusion of blood products, spinal cord injury, pelvic fracture), was applied to the entire trauma population . The most sensitive threshold would detect only about one half of DVTs and would require testing one fourth of all trauma patients, costing nearly $600,000 to implement during the study period. The authors concluded that even using a protocol for high-risk patients would be cost prohibitive and an inappropriate utilization of resources.
SPECIFIC TRAUMA POPULATIONS — Pharmacologic thromboprophylaxis is desirable but may need to be delayed if there is any ongoing bleeding or severe coagulopathy. Data to precisely estimate the risk of bleeding in trauma patients are limited. The risk is highest in patients with major trauma, particularly with multiple serious injuries, or injuries involving the brain or spine; other injuries at risk for bleeding include solid organ injury being managed conservatively, liver injuries that have been packed, and pelvic fractures, depending upon the grade of injury and whether there is active bleeding. The approach to and timing of prophylaxis for selected injuries is reviewed in the sections below.
Traumatic brain injury — Given the risk associated with even minor hemorrhagic expansion of intracranial trauma, pharmacologic prophylaxis for VTE after traumatic brain injury (TBI) remains one of the most controversial topics in this field. Patients with TBI are at increased risk of VTE, likely due to such factors as hypercoagulability and prolonged immobility [10-12,26,71]. The goal for VTE prophylaxis is to start pharmacologic therapy as soon as feasible based on bleeding risk. (See "Management of acute moderate and severe traumatic brain injury".)
Early pharmacologic VTE prophylaxis after head trauma has received growing support in both safety and efficacy [100-104]. Early initiation of VTE prophylaxis with low-molecular-weight (LMW) heparin within 72 hours of injury in patients with stable head computed tomography (CT) scan is supported by the American College of Surgeons Trauma Quality Improvement Program (TQIP) Best Practice Guidelines on traumatic brain injury . These guidelines refer to the modified Berne-Norwood criteria to stratify injury risk of TBI progression . The 2016 update of the Brain Trauma Foundation Guidelines also supports LMW heparin or unfractionated heparin (UFH) in combination with mechanical prophylaxis, though with no statement on timing, though acknowledging that there is an increased risk for expansion of intracranial hemorrhage . A systematic review addressed the question of timing, concluding that, based on current available literature, initiation of VTE prophylaxis within 72 hours after a stable head CT reduces VTE without a corresponding increase in intracranial hemorrhage .
In a retrospective study using the TQIP database that included over 3000 patients with severe TBI, early initiation of pharmacologic VTE prophylaxis within 72 hours of injury halved the rate of VTE compared with patients who received late prophylaxis (≥72 hours), without an increase in risk of death or neurosurgical intervention for new or expanding intracranial hemorrhage (ICH) . In another retrospective study in patients with ICH from TBI, the administration of early pharmacologic prophylaxis did not alter the rate of VTE or intracranial hemorrhage compared with patients treated after 48 hours . Other studies suggest similar efficacy and safety when started within 24 to 36 hours of blunt traumatic brain injury or spinal surgery [103,111].
LMW heparin may be preferable to UFH in patients with isolated TBI. In a retrospective study of 20,417 patients with isolated TBI, patients treated with LMW heparin were less likely to develop VTE compared with those treated with UFH .
Spinal cord injury — Deep vein thrombosis (DVT) is a common complication of traumatic spinal cord injury, occurring in 50 to 100 percent of untreated patients, with the greatest incidence between 72 hours and 14 days. The increased risk for VTE continues into the rehabilitation phase [7-9]. Based upon observational data, most recommend consideration of early initiation of pharmacologic prophylaxis using LMW heparin within 72 hours after spinal cord injury [36,113-116].
Patients with traumatic spinal cord injury, especially those with new-onset paraplegia or quadriplegia, are a subset of the trauma population considered very high risk for VTE [3,22,117,118]. The level and severity of injury does not clearly have an impact on the risk for DVT [119,120]. (See "Acute traumatic spinal cord injury", section on 'Venous thromboembolism and pulmonary embolism'.)
Early initiation of VTE prophylaxis is widely regarded as a treatment priority , and, although there is significant variability among practitioners, this population is considered by some to be an indication for prophylactic inferior vena cava (IVC) filter placement, particularly in those patients who are not candidates for anticoagulation [25,82,113,122]. Whether use of prophylactic IVC filters contributes to increased DVT risk in spinal cord injured patients is uncertain. Studies have reported increased rates of DVT in patients with prophylactic IVC filters compared with those without filters (20 versus 5 percent, in one study) . However, given the retrospective nature of these data, it is possible that the observed differences may have been due in part to selection bias and/or other confounding factors (eg, pharmacologic VTE prophylaxis usage). (See 'Prophylactic IVC filters' above.)
Early initiation of pharmacologic VTE prophylaxis, at less than 72 hours, is associated with a reduced incidence of DVT in patients with spinal cord injury [118,123]. This desire for early prophylaxis, however, is tempered by concern for expansion of intraspinal hematoma. However, in a single-center retrospective study of 501 patients with spinal cord injury, expansion of intraspinal hematoma occurred rarely (1 percent) with similar rates seen in patients who received early pharmacologic prophylaxis compared with those who did not . In a retrospective cohort study of 1432 patients with spinal fractures, 14 percent of whom had operative fixation, early pharmacologic VTE prophylaxis initiated within 24 hours was not associated with an increased risk of postoperative bleeding or epidural hematoma .
LMW heparin may be preferred in patients with spinal cord injury. In a small randomized trial comparing LMW heparin with UFH three times daily in 41 patients with spinal cord injury, overall VTE rates and bleeding complications were similar . However, patients who received LMW heparin had a lower incidence of pulmonary embolism (PE). In another trial, 107 patients were randomized to enoxaparin or UFH in combination with intermittent pneumatic compression . Rates of DVT were similar (65.6 versus 63.3 percent). However, the incidence of PE was lower for enoxaparin (5.2 versus 18.4 percent), as was the incidence of major bleeding (2.6 versus 5.3 percent).
Extremity injury — Patients with severe extremity injury (ie, multiple fracture, fracture associated with nerve or vascular injury), particularly lower extremity injuries, are at high risk for VTE. Patients should receive mechanical and pharmacologic prophylaxis for VTE as soon as is feasible . Similarly, severe pelvic fractures also increase the risk for VTE, and early prophylaxis is associated with lower rates of VTE and improved mortality . (See "Prevention of venous thromboembolism (VTE) in adults undergoing non-major extremity orthopedic surgery".)
The most important risk factors are likely related directly to the extremity injury and immobilization. Venous repair does not appear to increase the incidence of venous thromboembolic complications based on a review of 103 venous injuries . In this study, DVT occurred at 10 sites remote from the vascular injury site in 82 total patients. There were three vein thromboses in 34 repairs. In a review of 237 venous injuries, only a delay in starting thromboprophylaxis independently predicted VTE after controlling for injury severity score, hemodynamics, injured vessel, and surgical subspecialty .
With rare exception, anticoagulation should be continued perioperatively during subsequent procedures given the high risk of thromboembolic complications in this population. (See "Surgical management of severe lower extremity injury".)
Solid organ injury — As with all trauma patients, early pharmacologic prophylaxis is considered a goal in management even among those with solid organ injury, but due to the potential hemorrhagic complications of anticoagulation, many practitioners are hesitant to initiate early pharmacologic prophylaxis after solid organ injury. Despite the growing body of evidence to support early VTE prophylaxis after solid organ injury, definitive recommendations are awaiting prospective evaluation. Based on observational studies showing that thromboprophylaxis does not increase the rate of bleeding or failure of nonoperative management in patients with solid organ injury [42,130-135], we suggest early initiation of VTE prophylaxis within 48 hours, typically initiating treatment when the hemoglobin has stabilized with minimal change over a 24 hour period of time . However, pharmacologic prophylaxis may be contraindicated by associated injuries. (See "Management of splenic injury in the adult trauma patient" and "Management of hepatic trauma in adults" and "Management of blunt and penetrating renal trauma".)
Most of the literature has focused on the safety of VTE prophylaxis, especially in the setting of nonoperative management. Small retrospective studies have suggested that early thromboprophylaxis (within 48 hours) may be safe following splenic trauma, but there are currently no standards for the initiation of prophylaxis in patients who are managed nonoperatively [42,132,137]. In a retrospective review of 162 patients with blunt solid organ injury (spleen, liver, kidney), initiation of pharmacologic prophylaxis within 48 hours was safe and was not associated with an increased need for operative management for bleeding . Similarly, in a retrospective case-matched series of patients with blunt splenic injury treated nonoperatively, early initiation of VTE within 48 hours of injury was not associated with increased need for additional interventions . In a single-center study from Alabama in the United States that included 328 patients, early VTE prophylaxis was associated with fewer VTE complications, and there were no hemorrhagic complications associated with its use . These results were similar to an earlier study for which there was no significant difference in the rate of failure of nonoperative management or the volume of transfusion required for early compared with later prophylaxis .
Pediatric trauma considerations — In contrast to the adult population, pediatric trauma patients (age <14) have a lower incidence of VTE, ranging from 0.3 to 1.2 percent [139-142]. Thus, the risk:benefit ratio for pharmacologic prophylaxis is less favorable in this population, and it may not be appropriate to apply guidelines used in adult trauma patients. Major risk factors for VTE include the presence of a central venous catheter [CVC], adolescent (postpubertal) age, increasing injury severity, and operative intervention [139-142]. (See "Venous thrombosis and thromboembolism (VTE) in children: Risk factors, clinical manifestations, and diagnosis", section on 'Risk of VTE in children versus adults'.)
Our approach to VTE prophylaxis in children is based primarily on the age of the child [143,144]:
●Prepubertal children (age <15 years) – Based on the available data, we suggest not routinely initiating pharmacologic VTE prophylaxis in prepubertal trauma patients. However, pharmacologic prophylaxis may be appropriate if there are multiple additional risk factors that place the child at particularly high risk for VTE (eg, high severity of injury, critical illness, major surgery, transfusion, need for mechanical ventilation, previous history of VTE, severe obesity, prolonged immobility, CVC). (See "Venous thrombosis and thromboembolism (VTE) in children: Treatment, prevention, and outcome", section on 'Approach to VTE prophylaxis'.)
●Postpubertal adolescents (age ≥15 years) – The risk of VTE in postpubertal adolescent patients with major trauma approaches that in adult populations [141,142]. Thus, we suggest using the adult recommendations outlined above in this age group.
Our approach is generally consistent with the guidelines of the Eastern Association for the Surgery of Trauma and the Pediatric Trauma Society [143,144].
Pediatric VTE risk prediction models and decision algorithms have been developed, though none have been prospectively validated [145-148]. Factors included in these risk prediction models vary slightly based on the specific model. Most models include age, Glasgow Coma Score, injury severity score, sex, major surgery, transfusion, and critical illness (eg, intensive care unit admission or need for intubation or inotropic agent); some also include sex, presence of a CVC, immobilization, and pelvic or lower extremity fracture [145,146,148]. The models all appear to accurately identify patients at increased risk for VTE (ie, among patients with the highest predicted risk, actual rates of VTE range from 5 to 10 percent).
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Superficial vein thrombosis, deep vein thrombosis, and pulmonary embolism" and "Society guideline links: General issues of trauma management in adults".)
SUMMARY AND RECOMMENDATIONS
●Patients with trauma are, at minimum, at moderate risk for venous thromboembolism (VTE; deep vein thrombosis [DVT], pulmonary embolism [PE]). Severely injured patients are at high risk for VTE. In addition to typical risk factors for VTE, factors that increase the risk for VTE in trauma patients include spinal cord injury, head trauma, lower extremity fracture, pelvic fracture, and high injury severity score. (See 'Introduction' above and 'Risk factors' above.)
●For trauma patients, associated injuries can mask presenting symptoms/signs of VTE and complicate their detection. A high level of suspicion is needed, and further evaluation is warranted whenever VTE is suspected. For most trauma patients, we do not suggest routine surveillance to identify occult VTE. Many centers do routinely image trauma patients, but there is no conclusive evidence that outcomes are improved. (See 'Clinical features and diagnosis' above and 'Surveillance' above.)
●As a population at risk for VTE, severely injured patients should receive early, aggressive VTE prophylaxis. We continue prophylaxis until hospital discharge for all high-risk patients and beyond hospital discharge with patients with significant orthopedic injuries that limit mobility. It is important for the clinician to be aware that VTE prophylaxis reduces but does not eliminate the risk of VTE entirely. Even with appropriate prophylaxis, VTE remains a leading cause of mortality in hospitalized trauma patients. Our approach to VTE prophylaxis in adult trauma patients with severe injury is as follows:
•For patients with a low risk for bleeding, we recommend combined pharmacologic and mechanical thromboprophylaxis rather than mechanical prophylaxis alone (Grade 1B), and further suggest combined thromboprophylaxis rather than pharmacologic prophylaxis alone (Grade 2B). Combined pharmacologic and mechanical thromboprophylaxis is more effective than either modality alone. (See 'Thromboprophylaxis' above.)
•For patients with bleeding risk deemed too high to permit pharmacologic thromboprophylaxis, we suggest mechanical methods, rather than no prophylaxis (Grade 2B). This typically consists of intermittent pneumatic compression (IPC) provided there are no contraindications (eg, external fixation, extremity ischemia). IPC alone is more effective than no prophylaxis for preventing DVT and can generally be applied to at least one extremity. Pharmacologic prophylaxis should be started as soon as the bleeding risk has resolved. (See 'Compression' above and 'Specific trauma populations' above.)
•When pharmacologic prophylaxis is used, we suggest low-molecular-weight (LMW) heparin rather than unfractionated heparin (UFH) for most patients (Grade 2B). UFH is preferred in those with severe renal insufficiency (eg, creatinine clearance <30 mL/min). For most patients, we suggest initiating pharmacologic thromboprophylaxis within 72 hours rather than initiating it later (Grade 2C). The nature of the specific injuries dictates where in the provided time range is more appropriate. (See 'Anticoagulation' above and 'Specific trauma populations' above.)
●The recommendations given above for pharmacologic prophylaxis are also appropriate for high-risk postpuberty pediatric trauma patients. (See 'Pediatric trauma considerations' above.)
●For most severely injured trauma patients, we suggest not placing prophylactic inferior vena cava (IVC) filters (Grade 2B). Most patients can be started on pharmacologic thromboprophylaxis reasonably early in the postinjury period, and early placement of an IVC filter does not appear to be beneficial in such patients. However, it is reasonable to use IVC filters in very-high-risk patients, particularly if there are ongoing contraindications to anticoagulation. When retrievable IVC filters are used, a protocol for follow-up should be used to ensure that the filter is removed in a timely fashion. (See 'Prophylactic IVC filters' above and "Placement of vena cava filters and their complications", section on 'Filter retrieval'.)