There are over 1.7 million cases per year of traumatic brain injury (TBI) in the United States, with most injuries occurring secondary to blunt trauma (1). Despite improvements in medical triage and tertiary care, TBI encompasses a significant healthcare burden with significant patient morbidity and mortality. The sequelae of TBI, including coagulopathy, hypercoagulability, and a pro-inflammatory response, all contribute to secondary brain injury and can be as detrimental to patient outcomes as the initial injury itself (2, 3).
Initially, platelet dysfunction and coagulopathy predominate the acute response to TBI (4). In patients with severe TBI, platelet dysfunction manifests as an increase in platelet adenosine diphosphate (ADP) receptor inhibition, with the degree of inhibition strongly correlating to TBI severity and mortality (5, 6). Following the acute phase of coagulopathy, hypercoagulability dominates the subacute phase, which in humans can range from the first 6 h to several days from the initial TBI insult (7, 8). In humans, hypercoagulability may present in the form of early disseminated intravascular coagulation syndrome with consumption of clotting factors, or as microvascular thrombi, venous thromboembolism (VTE), or cerebral ischemia (9, 10). While these changes are clinically well documented in severe TBI, the changes in coagulation following mild to moderate TBI are not as well understood.
Defining the temporal relationship of platelet dysfunction, subsequent hyperaggregation, and the relationship to coagulopathy following TBI has remained elusive and complex. In this study, we sought to investigate the coagulation changes observed following head injury utilizing an established murine model of moderate TBI and to define the time course between changes in platelet function and overall coagulation. We hypothesized that changes in platelet-induced coagulation are dependent on the type of platelet agonist and time following TBI, overall coagulopathy mirrors changes in platelet-induced coagulation, and the changes in platelet-induced coagulation correlate with altered pro-inflammatory mediators during the subacute phase.
MATERIALS AND METHODS
Closed weight drop TBI model
All the murine experiments were approved by the Institutional Animal Care and Use Committee of the University of Cincinnati. Male C57BL6/J mice between ages 8 and 10 weeks were purchased from Jackson Laboratory (Bar Harbor, Maine). The animals were acclimated for at least 1 week prior to experiments and housed in controlled conditions with a 12 h light–dark cycle. The closed weight drop model was used, as previously described, to induce a moderate TBI (11, 12). Briefly, mice were anesthetized with 2% inhaled isoflurane for 2 min in 100% oxygen at 1 L/min. The animals were placed in a prone position with the head centered under a 400 g cylindrical weight. The weight was dropped from a height of 1.5 cm above the table surface and centered in both rostral/caudal and lateral directions. Moderate TBI was defined and internally validated by the righting reflex response time to turn three times from a supine to prone position.
As a blunt non-TBI injury control, a separate group of mice underwent tail injury. Mice were anesthetized as above, placed in a prone position on the surface of our mild traumatic brain injury apparatus and the same 400 g steel rod was released from the same 1.5 cm height onto an area 2 cm distal to the base of the tail (13). The righting reflex response was performed immediately following TBI to determine appropriate injury severity.
Sham mice were anesthetized, handled, and positioned below the impactor weight but did not sustain TBI or tail trauma. Mice were sacrificed after 10 min (acute), 6 (subacute), or 24 h post TBI, anesthetized with 0.1 mg/g of pentobarbital administered by intraperitoneal injection, and whole blood samples were collected via cardiac puncture.
Whole blood samples were anticoagulated with 10% hirudin. The Multiplate impedance aggregometry software (Roche Diagnostics, Rotkreuz, Switzerland) was used to measure platelet aggregation induced by arachidonic acid (AA), ADP, collagen, and protease activated receptor (PAR)-4. Results are reported as area under the curve.
Platelet count determination
Whole blood samples were anticoagulated with heparin, and platelet count was obtained using the Coulter AcT 10 Hematology Analyzer (Beckman Coulter, Brea, Calif).
Coagulant factor and cytokine measurements by ELISA
Serum was obtained from whole blood samples using serum separator tubes (BD Bioscience, San Diego, Calif) that were centrifuged at 1,000 g for 10 min then frozen until analysis. Platelet poor plasma (PPP) was obtained by centrifugation of citrated blood (10%) at 260 g for 5 min to generate platelet-rich plasma. This plasma was then centrifuged again at 10,000 g for 10 min with the resulting supernatant saved for analysis as PPP.
Serum and PPP levels of fibrinogen, interleukin-6 (IL-6), keratinocyte chemoattractant (KC), tissue factor (TF), von Willebrand factor (vWF), and P-selectin were measured by ELISA according to the manufacturer's protocol (R&D Systems, Minneapolis, Minn).
Whole blood samples were anticoagulated with 10% citrate. Changes in coagulation parameters after TBI were determined using rotation thromboelastometry (ROTEM, TEM Systems Inc, Durham, NC) analyses, as previously described (14, 15). Analyses were performed within 10 min of sample collection. Overall coagulation (NATEM), fibrin contribution to clot (FIBTEM), and the extrinsic pathway coagulation (EXTEM) were determined. Viscoelastic coagulation testing included analysis of reaction (R) time, clot formation time (CFT), rate of clot formation (α-angle), and maximum clot firmness (MCF). Per manufacturer's instructions, 20 μL of thromboplastin was added to 300 μL of citrated whole blood to initiate clot formation in the EXTEM test. In addition to thromboplastin, cytochalasin D was added to FIBTEM samples to negate the contribution of platelets during clot formation. Percent of platelet contribution (%MCF-Platelet) was calculated by the equation: (EXTEMMCF − FIBTEMMCF)/EXTEMMCF, as previously described (16).
Tail bleed times
Mice were anesthetized with 0.1 mg/kg pentobarbital and remained warmed on a heating pad for the duration of the experiment. The distal 1 mm of the tail was removed with a tail guillotine. The remaining proximal tail was inserted into warmed 0.9% saline. The time until bleeding cessation was measured and defined as no blood flow for 1 min.
Two-tailed Student t tests were used to make comparisons between two mice groups, and Analysis of Variance with Tukey post-test was used to make comparisons between three or more groups. Means and standard errors of the mean were calculated in each experiment. A P value of less than 0.05 was considered significant. All data analyses were performed with Prism 6 (GraphPad Software, La Jolla, Calif).
Immediately after TBI, platelet aggregation is blunted, but subsequently increases 6 h after TBI
There was a significant decrease in ADP-induced platelet activation in mice subjected to TBI (n = 16) compared with their sham (n = 19) counterparts at 10 min (Fig. 1A, 13.6 ± 2.3 units TBI versus 29.5 ± 5.2 units sham, P < 0.05). At 10 min a similar difference in ADP-induced platelet activation was observed between blunt tail injury (n = 5) and sham mice (Fig. 1A, 13.5 ± 4.3 units blunt tail injury vs. 29.5 ± 5.2 units sham, P < 0.05). However, at 6 h after injury, there was a significant increase in ADP-induced platelet activation in TBI mice (n = 6) compared with their sham (n = 8) counterparts (Fig. 1B, 82.3 ± 13.9 units TBI vs. 59.0 ± 5.5 sham, P < 0.01), as well as the blunt tail injury (n = 5) counterparts (Fig. 1B, 82.3 ± 13.9 units TBI vs. 49.6 ± 22.1 blunt injury P < 0.05.) Notably, this relationship was similar in both AA (TBI n = 6, sham n = 6) and collagen-induced (TBI n = 6, sham n = 9) platelet aggregation at 6 h after injury (Fig. 1C and D). There was no difference in PAR4-induced platelet activation at any time point (TBI n = 6, sham n = 6) (Fig. 1E).
Platelet counts and fibrinogen levels are unaltered after TBI injury
Given our findings of increased platelet activation 6 h post TBI, we measured circulating platelet counts and fibrinogen levels. Whole blood platelet counts (Fig. 2A) were no different between TBI (n = 6) and sham mice (n = 8). Similarly, no differences existed in serum fibrinogen levels (Fig. 2B) between TBI (n = 4) and sham mice (n = 8).
Hypercoagulability occurs 6 h after TBI despite unchanged platelet count and fibrinogen levels
We measured coagulation parameters at 6 h postinjury to assess the contribution of platelet activation to whole blood hypercoagulability. There was a significant increase in overall clot strength in TBI (n = 8) compared with sham mice (n = 8) (Fig. 3A, NATEM MCF, 73.9 ± 1.9 mm TBI vs. 68.9 ± 1.1 mm Sham, P = 0.04). This difference persisted when examining fibrin contribution to clot strength between TBI (n = 6) and sham mice (n = 8) (Fig. 3B, FIBTEM MCF, 29.2 ± 3.3 mm TBI vs. 21.2 ± 1.7 mm Sham, P = 0.04). In addition, there was a significantly faster CFT within the extrinsic pathway in TBI (n = 6) compared with sham (n = 8) mice, as demonstrated by decreased CFT (Fig. 3C, EXTEM CFT, 32 ± 3.5 s TBI vs. 63.9 ± 5.6 s Sham, P < 0.001) and increased α-angle (Fig. 3D, EXTEM α-angle, 83.7 ± 0.7° TBI vs. 76.9 ± 1.2° Sham, P < 0.001).
Tail bleeding times were also investigated at 10 min post-TBI and 6 h post-TBI (Fig. 4). Tail bleeding times were unchanged at 10 min after TBI compared with baseline uninjured mice. TBI mice (n = 5) displayed significantly shorter bleeding times at 6 h compared with sham mice (n = 16) (56.5 ± 3.6 s TBI vs. 83.5 ± 11.0 s sham, P < 0.001). Interestingly, neither the fibrinogen concentration nor the %MCF-platelet (data not shown) were independently increased at 6 h after TBI, suggesting that platelet changes occur as part of a more global hypercoagulability.
Increase in platelet activation is paralleled by an increase in pro-inflammatory response after TBI
To evaluate the systemic inflammatory response after TBI, serum levels of pro-inflammatory cytokines IL-6, KC, and coagulation and platelet function markers P-selectin, vWF, and TF were measured 6 h following TBI. Compared with sham mice (n = 5), TBI mice (n = 5) showed a significant increase in IL-6 (Fig. 5A, 123.9 ± 35.5 pg/mL TBI vs. 26.8 ± 9.1 pg/mL sham, P = 0.029) and KC (Fig. 5B, 287.8 ± 20.2 pg/mL TBI vs. 101.7 ± 21.2 pg/mL sham, P = 0.015). TBI mice (n = 8) also showed an increase in P-selectin (Fig. 5C, 194.5 ± 21.4 ng/mL TBI vs. 119.1 ± 8.3 ng/mL sham, P = 0.005) compared with sham mice (n = 8). There was no difference in the level of TF (Fig. 5D, 1.86 ± 4.2 pg/mL TBI vs. 1.49 ± 2.1 pg/mL sham, P = 0.9) nor in the level of vWF (Fig. 5E, 0.04 ± 0.04 ng/mL TBI vs. 0.13 ± 0.11 ng/mL sham, P = 0.2) at 6 h postinjury. The post-TBI changes in P-selectin, vWF, and TF levels were similar when measured in PPP at 6 h post-TBI.
In this study, we demonstrate an initial ADP-dependent platelet dysfunction in a murine TBI model, followed by a rebound hypercoagulability and platelet hyperaggregation 6 h postinjury. Contributing factors to subacute hypercoagulability were increased platelet and fibrin clotting strength (MA) and speed to CFT, without concomitant changes in platelet counts and fibrinogen levels. This hypercoagulability is associated with a marked increase in select systemic pro-inflammatory cytokines post-TBI.
Our findings of acute hypocoagulability are consistent with previous human and murine TBI studies (4–6, 17). Although these studies employed a severe TBI model to demonstrate platelet dysfunction, we observed similar acute alterations in platelet responsiveness to the agonists ADP, AA, and collagen in our moderate murine TBI model. Here, we show both moderate TBI and blunt tail trauma resulted in acute ADP-induced platelet dysfunction. These findings are consistent with the observations reported by Wohlauer et al. (17), in which blunt trauma of any kind was associated with ADP-induced coagulopathy in humans as measured by thromboelastography. However, the hypercoagulability demonstrated at 6 h after TBI was unique to this injury compared with blunt tail injury. Rather than conventional measures of serum coagulation profiles (plasma prothrombin time and activated partial thromboplastin time), thromboelastometry was utilized in our analysis, as conventional tests of coagulation have not been shown to be a sensitive measure of acute trauma-induced coagulopathy (4, 8, 17). By contrast, viscoelastic coagulation testing is an established method of detecting changes in coagulability in trauma patients that has demonstrated greater accuracy than plasma prothrombin time and activated partial thromboplastin time in predicting thromboembolic events (18–20).
We are the first group to demonstrate post-TBI platelet hyperaggregation and coagulation changes with impedance aggregometry and thromboelastometry in a murine TBI model. These findings will allow for future investigations into the development and consequences of these platelet-related changes, such as in the role of VTE following TBI. A significant increase in MCF and α-angle, and a significant decrease in CFT in TBI mice compared with sham at 6 h post-TBI was observed. This not only represents increased platelet activity, but also increased platelet–fibrin interaction. These observations are consistent with previous studies that have established the concept of delayed thrombosis in animal models, while hypercoagulability has also been demonstrated following TBI in humans by thromboelastometry (2, 21–25). Specifically, our findings mirror those of Massaro et al. (8), who reported a progressive and delayed hypercoagulable state in their study of TBI patients admitted to their Neurocritical Care Unit. These findings of hypercoagulability following TBI are especially important as the incidence of deep vein thrombosis and pulmonary embolism continue to rise despite aggressive initiation of mechanical and chemoprophylaxis regimens in the trauma patient population (26). Recently, Brill et al. (27) observed a significantly higher rate of (VTE) in trauma patients with hypercoagulable admission thrombelastography (TEG) compared with those patients with either normal or hypocoagulable admission TEG testing. It should be noted, however, that while isolated TBI has been associated with a higher incidence of VTE, no study has yet examined the rate of VTE in isolated TBI patients presenting with a hypercoagulable TEG (28).
In the aftermath of the initial TBI insult, vascular thrombotic events remain a grave concern, as these events may exacerbate secondary brain injury (29). Due to concern for ongoing cerebral hemorrhagic complications, no consensus exists with regard to prophylactic anticoagulation in these patients, despite the elevated risk for morbidity and mortality associated with VTE following TBI (30). Given the transition from coagulopathy to hypercoagulability after TBI, additional concerns arise regarding the appropriate timing of anticoagulation. Reiff et al. (31) reported from their single-center experience that TBI was an independent risk factor for the development of deep venous thromboses, despite receiving prophylactic anticoagulation. In addition, the elevated risk of VTE persists after TBI patients’ index hospitalisation (32). Identifying the phases of the inflammatory milieu and stage of hypercoagulability may facilitate development of more effective and tailored VTE prophylaxis to decrease the rates of VTEs and capillary microthrombi in vulnerable TBI patients.
There are several limitations of our study. First, though we were able to demonstrate a hypercoagulable state via thromboelastometry, the development of thrombotic and microthrombotic events was not directly examined. Second, this study does not include an examination of varying TBI severity or the contribution of additional blunt polytrauma on coagulopathy. Both the variables should be addressed in future studies. Third, though we identified a concomitant rise in the levels of several pro-inflammatory mediators during the subacute hypercoagulable state following TBI, we have not yet identified a cause-and-effect relationship. Fourth, for this study we have determined the severity of the murine TBI with the righting reflex response time, thus results may differ if injury is titrated to a different marker of severity, such as the neurological severity score (11). Fifth, the age of the mice used in the study was between 8 and 10 weeks, which may not be generalizable to older or younger populations. Finally, future comparisons of isolated TBI to other models of blunt injury and polytrauma may need additional refinement to standardize the host response to injury and further elucidate post-traumatic coagulation and platelet aggregation responses.
In conclusion, the acute hypocagulability that occurs initially after TBI transitions to a subacute hypercoagulability phase that is established by 6 h postinjury in a murine model. This is temporally associated with a significant pro-inflammatory response. Further studies to evaluate the crosstalk between the inflammatory response, platelet activation, and enzymatic clotting activity are needed, which may lead to new therapeutic targets in preventing micro- and macrovascular thrombotic insults after traumatic brain injury.
1. Summers CR, Ivins B, Schwab KA. Traumatic brain injury
in the United States: an epidemiologic overview. Mt Sinai J Med
76 2:105–110, 2009.
2. Stein SC, Chen XH, Sinson GP, Smith DH. Intravascular coagulation: a major secondary insult in nonfatal traumatic brain injury
. J Neurosurg
97 6:1373–1377, 2002.
3. Stein SC, Smith DH. Coagulopathy in traumatic brain injury
. Neurocrit Care
1 4:479–488, 2004.
4. Ploplis VA, Donahue DL, Sandoval-Cooper MJ, MorenoCaffaro M, Sheets P, Thomas SG, Walsh M, Castellino FJ. Systemic platelet dysfunction is the result of local dysregulated coagulation and platelet activation in the brain in a rat model of isolated traumatic brain injury
. J Neurotrauma
31 19:1672–1675, 2014.
5. Davis PK, Musunuru H, Walsh M, Cassady R, Yount R, Losiniecki A, Moore EE, Wohlauer MV, Howard J, Ploplis VA, et al. Platelet dysfunction is an early marker for traumatic brain injury
-induced coagulopathy. Neurocrit Care
18 2:201–208, 2013.
6. Castellino FJ, Chapman MP, Donahue DL, Thomas S, Moore EE, Wohlauer MV, Fritz B, Yount R, Ploplis V, Davis P, et al. Traumatic brain injury
causes platelet adenosine diphosphate and arachidonic acid receptor inhibition independent of hemorrhagic shock in humans and rats. J Trauma Acute Care Surg
76 5:1169–1176, 2014.
7. Sun Y, Wang J, Wu X, Xi C, Gai Y, Liu H, Yuan Q, Wang E, Gao L, Hu J, et al. Validating the incidence of coagulopathy and disseminated intravascular coagulation in patients with traumatic brain injury
—analysis of 242 cases. Br J Neurosurg
25 3:363–368, 2011.
8. Massaro AM, Doerfler S, Nawalinski K, Michel B, Driscoll N, Ju C, Patel H, Quattrone F, Frangos S, Maloney-Wilensky E, et al. Thromboelastography defines late hypercoagulability
after TBI: a pilot study. Neurocrit Care
22 1:45–51, 2015.
9. Hulka F, Mullins RJ, Frank EH. Blunt brain injury activates the coagulation process. Arch Surg
131 9:923–927, 1996.
10. Stein SC, Graham DI, Chen XH, Smith DH. Association between intravascular microthrombosis and cerebral ischemia in traumatic brain injury
54 3:687–691, 2004.
11. Flierl MA, Stahel PF, Beauchamp KM, Morgan SJ, Smith WR, Shohami E. Mouse closed head injury model induced by a weight-drop device. Nat Protoc
4 9:1328–1337, 2009.
12. Yang SH, Gustafson J, Gangidine BS, Stepien D, Schuster R, Pritts TA, Goodman MD, Remick DG, Lentsch AB. A murine model of mild traumatic brain injury
exhibiting cognitive and motor deficits. J Surg Res
184 2:981–988, 2013.
13. Hsieh T, Vaickus MH, Stein TD, Lussier BL, Kim J, Stepien DM, Duffy ER, Chiswick EL, Remick DG. The role of substance P in pulmonary clearance of bacteria in comparative injury models. Am J Pathol
186 12:3236–3245, 2016.
14. Luddington RJ. Thrombelastography/thromboelastometry
. Clin Lab Haematol
27 2:81–90, 2005.
15. Midura EF, Kuethe JW, Rice TC, Veile R, England LG, Friend LA, Caldwell CC, Goodman MD. Impact of platelets and platelet-derived microparticles on hypercoagulability
following burn injury. Shock
45 1:82–87, 2016.
16. Midura EF, Jernigan PL, Kuethe JW, Friend LA, Veile R, Makley AT, Caldwell CC, Goodman MD. Microparticles impact coagulation after traumatic brain injury
. J Surg Res
197 1:25–31, 2015.
17. Wohlauer MV, Moore EE, Thomas S, Sauaia A, Evans E, Harr J, Silliman CC, Ploplis V, Castellino FJ, Walsh M. Early platelet dysfunction: an unrecognized role in the acute coagulopathy of trauma. J Am Coll Surg
214 5:739–746, 2012.
18. Park MS, Martini WZ, Dubick MA, Salinas J, Butenas S, Kheirabadi BS, Pusateri AE, Vos JA, Guymon CH, Wolf SE, et al. Thromboelastography as a better indicator of hypercoagulable state after injury than prothrombin time or activated partial thromboplastin time. J Trauma
67 2:266–275, 2009.
19. Kashuk JL, Moore EE, Sabel A, Barnett C, Haenel J, Le T, Pezold M, Lawrence J, Biffi WL, Cothren CC, et al. Rapid thrombelastography (r-TEG) identifies hypercoagulability
and predicts thromboembolic events in surgical patients. Surgery
146 4:764–772, 2009.
20. Schreiber MA, Differding J, Thorborg P, Mayberry JC, Mullins RJ. Hypercoagulability
is most prevalent early after injury and in female patients. J Trauma
58 3:475–480, 2005.
21. Tikk A, Noormaa U. The significance of cerebral and systemic disseminated intravascular coagulation in early prognosis of brain injury. Acta Neurochir Suppl (Wien)
28 1:96–97, 1979.
22. Lu D, Mahmood A, Goussev A, Qu C, Zhang ZG, Chopp M. Delayed thrombosis after traumatic brain injury
in rats. J Neurotrauma
21 12:1756–1766, 2004.
23. Kumar MA. Coagulopathy associated with traumatic brain injury
. Curr Neurol Neurosci Rep
13 11:391, 2013.
24. Nekludov M, Bellander BM, Blomback M, Wallen HN. Platelet dysfunction in patients with severe traumatic brain injury
. J Neurotrauma
24 11:1699–1706, 2007.
25. Branco BC, Inaba K, Ives C, Okoye O, Shulman I, David JS, Schochl H, Rhee P, Demetriades D. Thromboelastogram evaluation of the impact of hypercoagulability
in trauma patients. Shock
41 3:200–207, 2014.
26. Cotton BA, Minei KM, Radwan ZA, Matijevic N, Pivalizza E, Podbielski J, Wade CE, Kozar RA, Holcomb JB. Admission rapid thrombelastography predicts development of pulmonary embolism in trauma patients. J Trauma Acute Care Surg
72 6:1470–1475, 2012.
27. Brill JB, Badiee J, Zander AL, Wallace JD, Lewis PR, Sise MJ, Bansal V, Shackford SR. The rate of deep vein thrombosis doubles in trauma patients with hypercoagulable thromboelastography. J Trauma Acute Care Surg
83 3:413–419, 2017.
28. Van Gent JM, Bandle J, Calvo RY, Zander AL, Olson EJ, Shackford SR, Peck KA, Sise CB, Sise MJ. Isolated traumatic brain injury
and venous thromboembolism. J Trauma Acute Care Surg
77 2:238–242, 2014.
29. Maegele M. Coagulopathy after traumatic brain injury
: incidence, pathogenesis, and treatment options. Transfusion
53 (suppl 1):28S–37S, 2013.
30. Phelan HA. Pharmacologic venous thromboembolism prophylaxis after traumatic brain injury
: a critical literature review. J Neurotrauma
29 10:1821–1828, 2012.
31. Reiff DA, Haricharan RN, Bullington NM, Griffin RL, McGwin G Jr, Rue LW 3rd. Traumatic brain injury
is associated with the development of deep vein thrombosis independent of pharmacological prophylaxis. J Trauma
66 5:1436–1440, 2009.
32. Olufajo OA, Yorkgitis BK, Cooper Z, Rios-Diaz A, Metcalfe D, Havens JM, Kelly E, Haider AH, Gates JD, Salim A. How long should we fear? Long-term risk of venous thromboembolism in patients with traumatic brain injury
. J Trauma Acute Care Surg
81 1:71–78, 2016.