Understanding Coagulation and Inflammation in Burns

Trauma injury is the leading cause of death in people 1-44 years old in the US. Burn injury is a particularly debilitating form of trauma. In the U.S. the incidence of burn injury is estimated to be greater than 2 million cases per year. Burns account for ~ 300,000 deaths worldwide every year. In 1996, the cost of caring for burn patients in the U.S. was estimated at $573 million per year.

Notably, 20-40% of trauma deaths that occur after hospital admission involve massive bleeding. Hemorrhage is the second most common cause of early in-hospital mortality accounting for a large portion of trauma deaths that occur within the first 24 hours. Resuscitation has dramatically changed over the last decade or so. Nowadays, we are substituting the normal saline and the other resuscitating fluids with blood products. This early administration of blood products during resuscitation is referred to as damage control resuscitation (DCR). The goal of DCR is to prevent and immediately correct trauma associated coagulopathy. DCR was initially practiced in the military where a balanced ratio of FFP:platelets:RBCs of 1:1:1 was employed. DCR has resulted in improved outcomes in both the military as well as civilian trauma setting as compared to previous resuscitation patterns. Despite the early use of a 1:1:1 ratio, a 1:1:2 ratio also became commonly used. Thus, the Pragmatic, Randomized Optimal Platelet and Plasma Ratios (PROPPR) study was done to investigate the best ratio of products to use during resuscitation. The PROPPR study found no differences in mortality between the 1:1:1 and the 1:1:2 groups at 24 hours or at 30 days. Additionally, there was no difference in complications between the two groups. Notably, exsanguination was significantly decreased in the 1:1:1 group and more patients in the 1:1:1 group reached hemostasis. The Prospective Observational Multicenter Major Trauma Transfusion (PROMMTT) study showed that early transfusion (within minutes of arriving at hospital) was associated with improved 6-hour survival. Furthermore, patients with increased plasma to RBC ratios (>1:2) were found to have improved 30-day survival as compared to patients who received lower plasma to RBC ratios (<1:2). Notably, 1-day and 30-day survival were found to be increased when patients received higher ratios of platelets to RBCs. Delayed but balanced transfusion ratios did not have the same protective effect as receiving plasma early.

To date there have been few studies on the effects of hemorrhage during burn and soft tissue excision and optimal blood product resuscitation. It is unclear whether DCR is optimal in burn patients. Moreover, it is unknown exactly what is happening in burn patients with respect to coagulation, platelet function, and microparticles (MPs). Patients with burns and soft tissue injuries typically have considerable bleeding during surgeries. Intraoperative blood loss is estimated at 9.2% of blood volume for every 1% of total body surface area (TBSA) burn excised in adults. In children undergoing burn excision, blood loss is estimated as 2% of blood volume for every 1% TBSA excised in extremities and trunk and 5% of blood volume for each 1% TBSA excised for the face. Moreover, early complete excision and grafting of major burns has been shown to decrease transfusions, infections, and mortality in children. Interestingly, many burn surgeons anecdotally report that their burn patients develop microvascular bleeding during the surgery. This phenomenon may be indicative of development of coagulopathy.

There have been several prospective studies looking at perioperative coagulation status in burn patients. These studies show that the coagulation proteins (FV, FVIII, FIX, and fibrinogen) decrease during surgery. Notably, in most cases, the factor levels were still within "normal" reference ranges. FVIII and fibrinogen are acute phase reactants and were elevated in the burn patients preoperatively. These studies were performed before the adoption of DCR. Additionally, several studies have shown that antithrombin, protein S, and protein C (the natural anticoagulants) levels decrease in burn injuries. A recent study by Palmieri et al. looked at compared a transfusion ratio of 1:1 RBCs:FFP to 4:1 RBCs:FFP in children with >20% TBSA burns. In their study of 16 children, they found a trend toward increased length of stay (LOS), peak PELOD score (measure of organ dysfunction), increased time to wound healing, and increased infection rates in the 4:1 group; however, the differences were not statistically significant. Notably, they found that the 1: 1 ratio was safe in the burn patients and that the 1:1 ratio was also less expensive ($26,635 versus $34,485). A recent study by Pidcoke et al. showed that current blood product resuscitation during burn and soft tissue excision is not hemostatic. Thus, the effect of using a balanced resuscitation ratio of blood products including platelets in burn patients has not been well studied.

In general, platelet dysfunction is believed to play a role in the development of trauma associated coagulopathy (TAC). The effect of burns on patients' platelets count and activity is unknown. Upon hospital admission, burn patients typically have normal platelet counts. By days 3-5, the patients' platelet counts usually drop, especially in burns of large TBSA. In severe burn patients, platelet mediators, such as platelet factor 4 (PF4) and thromboxane B2 (TxB2), have been found to be elevated. This is thought to be attributable to platelet activation and consumption. Thus, further investigation of the effect of burns on platelets will offer further insight into the transfusion needs of burn patients.

In a study done by Lu et al. on the development of TAC in burn patients, TAC was defined as an INR ≥ 1.3, aPTT ≥ 1.5 times the mean normal limit, and normal platelet counts, no patient presented with TAC upon admission. Only few changes in INR and aPTT values were observed over time. Further studies are needed; however, the authors may have underdiagnosed TAC in the burn patients based upon their definition of TAC. It is likely that platelets and MPs play a significant role in the development of TAC as well as the bleeding seen in burn patients during excision and grafting.

MPs are small vesicles (< 1 µm) that resemble their parent cell in terms of similar surface proteins and membrane lipids. Interestingly, all blood cells can release MPs. The most abundant MPs in the blood are platelet MPs (PMPs). PMPs have been found to have 50- to 100-fold higher procoagulant activity than activated whole platelets. PMPs are formed via five mechanisms: 1) platelet activation and subsequent shedding of membrane fragments, 2) complement mediated membrane attack, 3) high shear forces, 4) senescence and apoptosis of platelets and megakaryocytes, and 5) platelet cytoskeletal abnormalities. In the PROMMTT study, patient were found to have increased levels of MPs derived from endothelial cells, RBCs, and leukocytes as well as increased levels of tissue factor bearing MPs (TF-MPs). Notably, coagulopathic trauma patients were found to have much lower levels of PMPs, TF-MPs, and thrombin generation in addition to more bleeding and increased mortality.

To further understand the role of PMPs, Matijevic et al. examined PMPs in plasma. They compared thawed plasma at day 5 to freshly thawed fresh frozen plasma (FFP) at day 0. They found that the majority of MPs were indeed PMPs and that day 5 plasma had a 50% reduction in MPs and a 29% decrease in procoagulant activity. A recent prospective observational study of trauma patients found that patients with low levels of phosphatidylserine (PS) positive PMPs had impaired clot formation and were more likely to receive more RBC transfusions during the first 24 hours following injury. These studies suggest that PMPs may play an important previously overlooked role in TAC. The role of MPs in burn patients is unknown. It is likely that MPs play a significant and overlooked role in the coagulation and development of coagulopathies in burn patients. Further studies are needed to further evaluate the role of MPs particularly the number, identity, size, and contribution to coagulation in burn patients.

Following surgery with burn excision and grafting, patients remain hospitalized for various lengths of time. This often depends upon the TBSA burned, smoke inhalation, and infectious complications, and other complicating medical problems. Based upon previous studies, burn patients are thought to become hypercoagulable. This hypercoagulable state has been attributed to increased levels of some clotting factors (FV, FVIII), platelets, and fibrinogen, and decreased levels of antithrombin, protein S, and protein C. Notably, the incidence of deep vein thrombosis (DVT) and pulmonary embolism (PE) in burn patients is low. A study using the National Burn Repository database found the rate of venous thromboembolism (VTE) to be 0.61%. Another study found the rate to be 0.25% for DVT and 0.05% for PE. Yet another study which used duplex ultrasonography to screen burn patients for DVT on admission and at discharge found a prevalence of 6.1%. Thus, the rate of VTE in burn patients is thought to be much higher than previously expected due to asymptomatic DVTs as well as the difficulty in diagnosing DVTs in this population due to overlapping symptoms between DVTs and burns (ex. extremity swelling).

There have been several published reports of VTE in burn patients. Recently, Van Haren et al. studied the coagulation status of 24 burn patients over the course of their hospital stay using thromboelastography (TEG) and standard laboratory coagulation tests. Notably, repeat samples taken one week after admission were found to be hypercoagulable (decreased R and K times, increased α angle and increased MA) and levels of fibrinogen, protein S, protein C, and antithrombin were found to be increased. Two patients that were hypercoagulable (TEG showing decreased R time) at admission were found to later develop VTE despite anticoagulation therapy. Surprisingly, prophylactic use of anticoagulation is not overall common in burn patients. A survey of burn centers showed that about one quarter of USA centers did not use any form of VTE prophylaxis. Thus, additional studies are needed to further explore the coagulation status and various coagulation and anticoagulation factor levels of burn patients throughout their hospital course. Only by understanding what is physiologically taking place in these patients can we offer the best therapeutic treatment (no anticoagulation versus anticoagulation, and if so, which one?).

The goal of the proposed study is to further understand the physiologic changes that occur initially in burn patients and how the coagulation status of the burn patient changes over time. In addition to changes in coagulation and anticoagulation proteins, this study will focus on the effect of burns on platelets and MPs. We will also examine the MP populations and characterize the identity, size, PS content of the MPs and elucidate their effect on coagulation parameters as well as inflammation.