Pulmonary Embolism Risk Higher With Traumatic Chest Injury

BOCA RATON, FLA. (EGMN) - Severe chest injury constitutes a newly recognized independent risk factor for pulmonary embolism in trauma patients.

"At our trauma center we have a new venous thromboembolism risk algorithm, and patients with chest injury now go into the high-risk category. You have to give these patients some type of chemical prophylaxis as early as you feel safe, because if you're developing a pulmonary embolism without a deep vein thrombosis, the methods used to prevent DVT [such as a prophylactic inferior vena cava filter] are not going to prevent the pulmonary embolisms," Dr. Mary M. Knudson said in a presentation at the annual meeting of the American Surgical Association.

This lack of benefit for prophylactic IVC filters in preventing pulmonary embolism (PE) was another one of the key findings in her study, in which she examined risk factors and outcomes for PE and DVT in 888,652 patients who were treated at 326 level I or II trauma centers that were included in the American College of Surgeons' National Trauma Data Bank (NTDB) for 2007-2009.

The incidence of DVT in this very large group of trauma patients was 1.06%, and for PE it was 0.42%. Only 20% of patients with PE also had a reported DVT, but because of how the data in the national registry are collected, it isn't known which came first.

The risk factors for PE and DVT were not the same. For example, patients with severe chest injury (defined as an Abbreviated Injury Scale score of 3 or higher) were 42% more likely to develop PE than were trauma patients without such an injury, but they weren't at increased risk for DVT. In contrast, patients with severe traumatic brain injury were 34% more likely to have DVT than were those without traumatic brain injury, but they were also 13% less likely to be diagnosed with a PE. And patients who were ventilator dependent longer than 3 days were at a 5.3-fold increased risk for DVT, but they were at a 3.8-fold increased risk for PE.

In contrast, several other significant predictors had overlapping risks for both PE and DVT, including shock, pelvic fracture, and spine injury, reported Dr. Knudson, a professor of surgery at the University of California, San Francisco.

She noted that compared with her previously reported analysis of the NTDB experience for 1994-2001, the PE incidence rate is rising, while PE-attributed mortality has declined. The likely explanation for these trends is that in recent years, the liberal use of chest CT after trauma has become routine, with resultant incidental detection of small PEs that are amenable to anticoagulation therapy.

"I think we're recognizing a disease that's in an earlier stage, and one that we probably overlooked in the past," the surgeon said.

The incidence of PE in trauma patients more than doubled, from 0.21% in 1994-2001 to 0.49% in the same trauma centers in 2007-2009. Meanwhile, mortality in trauma patients with PE dropped from 15% in the earlier period to 11% more recently. In 1994-2001, PE was associated with an adjusted fourfold increased risk of mortality, whereas in 2007-2009 PE conferred a 2.4-fold increase in mortality.

Dr. Knudson concluded that prophylactic IVC filters are ineffective in preventing trauma-related PE because the use of such filters doubled between her first and second studies, even as the PE incidence rate more than doubled.

"I'm not suggesting that prophylactic IVC filters cause pulmonary embolisms, but they certainly aren't preventing them," she said.

For this reason, she added, it's "very rare" for prophylactic IVC filters to be utilized at the San Francisco trauma center.

Based upon her new NTDB study and other data, Dr. Knudson's current concept of posttrauma venous thromboembolism is that severely injured patients who arrive at a trauma center in shock are already coagulopathic even before they receive transfusions. During the next 2-3 days, as they receive multiple transfusions, their protein C becomes depleted and they become hypercoagulable.

At that point, patients with a chest injury (with its attendant profound inflammation) are more at risk for PE, whereas those with traumatic brain injury have stasis and may be more at risk for DVT, she explained.

Discussant Dr. David B. Hoyt, executive director of the American College of Surgeons, said that Dr. Knudson's identification of severe chest injury as a novel contributor to PE is a major new observation that will be an important consideration when clinicians are assessing a trauma patient's overall risk.

Dr. Knudson declared having no relevant financial interests.