AT THE AAST ANNUAL MEETING
WAIKOLOA, HAWAII – Venous thromboembolism prophylaxis with low molecular weight heparin (LMWH), instead of unfractionated heparin (UH), is associated with lower risk of pulmonary embolism (PE) in patients with major trauma, results from a large study have shown.
The results of the study, based on data from the American College of Surgeons (ACS) Trauma Quality Improvement Program, suggest that LMWH-based strategies for thromboprophylaxis should be preferred after major trauma.
“Patients with major injury are at high risk for developing venous thromboembolism,” James Byrne, MD , said at the annual meeting of the American Association for the Surgery of Trauma. “Deep vein thrombosis frequently complicates the clinical course, and pulmonary embolism remains a leading cause of delayed mortality. We know that pharmacologic prophylaxis reduces the risk of DVT. For this reason, timely initiation of either low molecular weight or unfractionated heparin is indicated for all patients.”
Dr. Byrne, a general surgery resident at Sunnybrook Health Science Center, Toronto, Ontario, Canada, went on to note that LMWH is often favored because of a randomized controlled trial which showed that LMWH was associated with fewer deep vein thromboses ( N Engl. J. Med. 1996;335:701-7 ). However, significant practice variability continues to exist.
“Practitioners might favor the shorter half-life of unfractionated heparin in patients where they perceive the risk for hemorrhagic complications is high,” he said. “There’s also recent evidence to suggest that dosing may be all important and that unfractionated heparin dosed three times daily may be equivalent to low molecular weight heparin. If this is true, it might suggest that the historically higher cost of low molecular weight heparin could favor the use of unfractionated heparin.”
Furthermore, there is a is a lack of evidence comparing either agent to prevent PE, he added. “This is an important gap in our knowledge, because PE frequently occurs in the absence of an identified DVT and carries a significant risk of death. At present, it is not known how practice patterns with respect to choice of prophylaxis type influence risk of PE at the patient or hospital levels.”
Due to a lack of evidence comparing agents to prevent PE, the researchers set out to compare the effectiveness of LMWH versus UH to prevent PE in patients with major trauma who were treated at trauma centers participating in the ACS Trauma Quality Improvement Program from 2012 to 2015. They included all adults with severe injury who received LMWH or UH and excluded those who died or were discharged within five days, and those with a bleeding disorder or chronic anticoagulation. The exposure was defined as thromboprophylaxis with LMWH versus UH, and the primary outcome was PE confirmed on radiologic imaging. Potential confounders were considered, including patient baseline characteristics, anatomic and global injury severity, presenting characteristics in the emergency department, acute intracranial injuries, orthopedic injuries, early surgical interventions, and timing of prophylaxis initiation.
Dr. Byrne and his associates then used three analytic approaches in the study: a propensity score matching methodology, a multivariable logistic regression model for PE, and a center-level analysis examining the influence of LMWH utilization on hospital rates of PE.
They identified 153,474 trauma patients from 217 trauma centers. Their median age was 50 years and 67% were male. Blunt trauma was most common (89%), with a mean Injury Severity Score score of 20. LMWH was the most common type of thromboprophylaxis used (74%), and PE was diagnosed in 2,722 patients (1.8%).
Compared with patients who received LMWH, those who received UH were older and were significantly more likely to have been injured by falling (42% vs. 28%), with higher rates of severe head injuries (43% vs. 24%) and intracranial hemorrhage (38% vs. 19%). Conversely, LMWH was most favored in patients with orthopedic injuries.
After propensity score matching, patients on LMWH suffered significantly fewer PEs (1.4% vs. 2.4%; odds ratio, 0.56). This result was consistent within propensity-matched subgroups, including for patients with blunt multisystem injuries (OR, 0.60), penetrating truncal injuries (OR, 0.65), shock in the ED (OR, 0.68), isolated severe traumatic brain injury (OR, 0.49), and isolated orthopedic injuries (OR, 0.28).
Results of a sensitivity analysis in which each propensity-matched pair was matched within the same trauma center yielded similar results. Specifically, patients who received LMWH were at significantly lower risk for developing PE (OR, 0.64). “Importantly, this analysis minimized residual confounding due to differences in hospital-level processes of care, such as prophylaxis dosing or frequency, mechanical prophylaxis use, and thromboembolism screening practices,” Dr. Byrne noted.
Multivariable logistic regression also showed that patients who received LMWH had lower odds of PE (OR, 0.59). Other significant predictors of PE included obesity (OR, 1.54), severe chest injury (OR, 1.31), femoral shaft fracture (OR, 1.60), and spinal cord injury (OR, 1.60). Delays in prophylaxis initiation beyond the first day in the hospital were associated with significantly higher rates of PE, with an 80% increased risk of PE for patients who had their prophylaxis initiated after the fourth day.
The researchers conducted a center-level analysis in an effort to answer the question whether practice patterns with respect to choice of prophylaxis type influence hospital rates of PE. Across all 217 trauma centers in the study, the median rate of LMWH use was 80%, while the mean rate of PE was 1.6%. When trauma centers were grouped into quartiles based on their unique rate of LMWH use, trauma centers in the highest quartile (median LMWH use: 95%) were 50 times more likely to use LMWH, compared to those in the lowest quartile (median LMWH use: 39%) after adjusting for patient case mix. Compared with the lowest quartile, trauma centers that used the greatest proportion of LMWH had significantly lower rates of PE (1.2% vs. 2.0%). After adjusting for patient baseline and injury characteristics, patients who were treated at trauma centers in the highest quartile had significantly lower odds of PE (OR, 0.59).
Dr. Byrne acknowledged certain limitations of the study, including the potential for residual confounding and the inability to account for the dosing and frequency of prophylaxis that was given. “We were only able to measure the type and timing of prophylaxis initiation. We don’t know what doses of prophylaxis were used, and it is possible that the trauma centers included in this study favored use of UH twice daily,” he said.
Therefore, it is possible that the results might have been different if they had been able to directly compare LMWH to UH administered three times a day. “We also couldn’t measure interruptions in dosing due to surgery or patient refusal,” he said. “However, if it the case that UH is more likely to be refused based on the need for more frequent dosing, perhaps that is another feather in the cap of low molecular weight heparin-based thromboprophylaxis strategies. Larger prospective studies are needed, that take into account prophylaxis type and dosing, and are powered to detect a difference with respect to PE.”
Dr. Byrne reported having no financial disclosures.