Multicenter Registry for Comparative Effectiveness Analysis of Venous Thromboembolism in Trauma Patients

Consortium of Leaders in the Study Of Traumatic Thromboembolism (CLOTT)

Venous thromboembolism (VTE) remains a leading cause of death in trauma patients. Based on the EAST Management Guidelines for the prevention of VTE in trauma patients, a number of research questions could be addressed by a thorough current literature review combined with a multicenter concurrent analysis. This proposal seeks to create a data registry of trauma patients from multiple trauma centers around the United States that will serve as a platform for the study of VTE.

The VTE related research questions, identified by the EAST Management Guidelines as needing further study, are:

1. While numerous risk factors for VTE have been identified and reported, there are only 2 that have Level 1 evidence to support them-spine fracture and spinal cord injury. Many of the other reported risk factors were identified in studies that did not have protocols for surveillance, reported only on patients with symptoms, or used data from large administrative databases such as the National Trauma Data Bank4. It has also been shown that the frequency of diagnosis of deep venous thrombosis (DVT) increases when more duplex imaging of the extremities is undertaken5. Thus, if an aggressive concurrent surveillance program for DVT and aggressive work up of suspected pulmonary embolism (PE) is undertaken:

   1. Are there unknown or under-reported risk factors?
   2. Is there a hierarchy of risk among the factors? For example, is a long bone fracture associated with a higher incidence of VTE than a closed head injury?
   3. Do all closed head injuries with an AIS>3 have the same degree of risk or is an epidural hematoma more risky than a subdural hematoma?
   4. What constitutes high and very high risk?

2. What is the comparative efficacy and value6 of unfractionated heparin (UH), low molecular weight heparin (LMWH) and direct thrombin inhibitors (DTI) with or without associated mechanical prophylaxis (MP) in preventing VTE in trauma patients? In examining this question, the morbidity of these agents must be carefully determined, particularly with respect to intracranial injury and intra-abdominal solid viscus injury.

3. In trauma patients who develop VTE what is the best treatment and for how long?

4. Should all trauma patients who develop proximal deep venous thrombosis (pDVT) undergo a workup for thrombophilia or should this be based up the degree of "inverse risk" (inverse risk meaning a patient with a minor injury, who is thought to be at low risk, who develops pDVT)? Similarly, should trauma patients who develop pulmonary embolism (PE) undergo a workup for thrombophilia or should this be based up the degree of "inverse risk" [question 4 adds importance to answering questions 1a and 1b]?

5. A recent meta-analysis of observational studies suggests that prophylactic inferior vena cava filters (PICVF) significantly reduce the risk of PE (OR: 0.09-0.49)7. The authors of the meta-analysis point out a number of methodological flaws with the studies selected for the analysis. Most importantly, all were retrospective studies with historical controls and the use of pharmacologic prophylaxis was not uniform among the studies evaluated. Despite this report, there is continued significant practice variation in the use of PIVCF and the role of PIVCF in the care of trauma patients remains at question. Thus, if an aggressive concurrent surveillance program for DVT and aggressive work up of suspected PE is undertaken:

   1. Are all patients perceived to be at high risk for VTE who also have a relative or absolute contraindication to pharmacologic prophylaxis receiving a PIVCF?
   2. Do PIVCF reduce the risk of PE in trauma patients in whom risk adjustment has been done?
   3. Do PIVCF have value for the trauma patient at very high risk for VTE, the high risk trauma patient or the moderate risk trauma patient? In examining this question, the morbidity of PIVCF must be carefully determined, particularly with respect to their placement (including local complications at the insertion site, their migration and degree of tilt) and long-term complications.

6. Because most DVT is asymptomatic the clinical examination is extremely insensitive. Therefore, surveillance of the lower extremity venous systems is necessary. However, the use of routine surveillance is controversial. It is also known that venous duplex exams, while very accurate in the symptomatic patient, can be quite insensitive in the asymptomatic patient. In addition, surveillance scanning is hampered by external fixators and wounds (as many as 30% of extremities cannot be scanned for this reason10. Finally, it is not uncommon for trauma patients to have a PE (even a fatal PE) with negative followup surveillance of the lower extremity (LE) and upper extremity (UE) venous systems. This would suggest either that the surveillance duplex was inaccurate, or the clot arose in a vein that could not be well interrogated by duplex (i.e., the hypogastric), or that the clot arose de novo in the pulmonary artery11. Thus, if an aggressive concurrent surveillance program for DVT and aggressive work up of suspected PE is undertaken:

   1. Does "protocolized" surveillance (surveillance of asymptomatic patients) have value? In other words, does the quality of the outcome (determining the presence of asymptomatic thrombus in the vein of an extremity) justify the cost?
   2. What constitutes the most effective protocol in terms of when the initial scan should be done and the frequency of subsequent scans?
   3. What is the fate of peroneal vein and tibial vein clot? Does it vary depending on the degree of risk (i.e., more likely to propagate in patients who are at the highest risk compared to patients with lower risk)? The same question could be posed with respect to soleal vein or gastrocnemius vein clot.
   4. Does magnetic resonance venography have a role in high risk trauma patients as a surveillance modality-particularly those with pelvic fractures or severe LE fractures that cannot be reliably scanned?
   5. Does contrast venography have a role in high risk trauma patients as a surveillance modality, particularly in patients who have had a PE and who have a negative duplex?
   6. Since it has been shown by numerous authors that there is surveillance bias with regard to the diagnosis of DVT (i.e., the diagnosis is more often reported at hospitals doing more frequent scanning5, does the frequency of the diagnosis of PE increase as the number of chest CT angiograms are performed?

7. Recent studies suggest an association between local14 or systemic11 inflammation and pulmonary thrombosis (perhaps not embolism). Is there an increased risk of DVT or primary pulmonary thrombosis for trauma patients who have an associated inflammatory process such as sepsis, urinary tract infection (UTI) or pneumonia (PNA)? What constellation of local injury puts a patient at risk for pulmonary thrombosis?

We are proposing to perform an extensive literature review on each of these questions to produce a state of the art publication in follow-up of the EAST Management Guidelines and to provide a platform for a multicenter study of each or all of these questions.

Inclusion Criteria:

• Admitted to the hospital for care of injuries
• Have a greater than minimal (moderate to highest) level of VTE risk

Exclusion Criteria:

• Discharged prior to 24 hours in hospital
• Minimal VTE risk