Goal-directed transfusion algorithm for trauma patients with severe hemorrhage using TEG 6S: Results of a Delphi consensus survey and expert panel recommendations

Abstract

The leading cause of early preventable death following injury remains uncontrolled hemorrhage.1,2 Trauma-induced coagulopathy (TIC), resulting from the complex interaction between severe tissue injury and hypoperfusion, leads to adaptive and maladaptive inflammatory, immune, and coagulation disruptions. Trauma-induced coagulopathy is observed in upward of 25% of critically injured patients and is significantly associated with poor outcomes, including death.3–5 Viscoelastography is a highly useful tool for measuring alterations to clot formation, strength, and breakdown central to TIC, identifying patients at high risk of requiring massive transfusion, and directing both blood product and prohemostatic therapy as a part of a massive transfusion protocol.6–12 Many trauma-related studies are based on thromboelastography (TEG) and, specifically, use of the TEG 5000 (Haemonetics Corp., Boston, MA) platform, which has been used to develop specific thresholds to predict the need for massive transfusion and to optimize goal-directed transfusion therapy.10,13,14 The TEG 5000 analyzer is a rotational pin-in-cup platform, which requires frequent calibration and is sensitive to movement and vibration. This platform is being replaced with the TEG 6S, which is a cartridge based analyzer that obviates calibration and motion sensitivity concerns. Consequently, many centers have transitioned to instituting the TEG 6S platform for viscoelastic testing. One study compared results using the TEG 5000 versus the TEG 6S platforms and found excellent concordance in detecting coagulopathy.15 However, there are no studies describing the appropriate thresholds for the various TEG 6S parameters to identify injured patients at risk for massive hemorrhage, to guide transfusion of blood components, or to inform administration of prohemostatic agents as part of a goal-directed massive transfusion strategy. Currently, threshold values for transfusion are based on published TEG 6S manufacturer limits defining coagulopathy and therefore may not be applicable to severely injured patients with TIC. Given that it is not practical to repeat the previous studies, which were done using the TEG 5000 platform, an international panel was convened to provide evidence-informed expert consensus of thresholds for transfusion using the TEG 6S platform. An algorithm for goal-directed transfusion using the TEG 6S platform in the setting of severe trauma-related hemorrhage is presented herein. PATIENTS AND METHODS A modified Delphi process was used (Fig. 1). A Delphi method is a group-based method, which is based on the premise that a group of acknowledged experts who asked the same question(s) in an iterative process can develop either an expert opinion–based or evidence-based consensus to a problem.16 As previously noted, because it is not practical to repeat the studies previously done to identify transfusion thresholds using the TEG 5000 (Haemonetics Corp.) platform, an expert panel using a Delphi process was convened to make evidence-informed recommendations with the TEG 6S system. The process was facilitated by the Bioscript Group using funding from Haemonetics Corp., but the sponsor had no role in the formulation of questions or involvement in the Delphi process. They also did not have access to the data. The group consisted of 13 members of an advisory board that had previously been convened by Haemonetics Corp. to discuss the current state of evidence regarding use of TEG to guide therapy in massive transfusion for trauma and opportunities to inform the trauma surgery community on optimal use of the TEG 6S in this setting (Table 1). All participants except for J.C., who is a hematologist, are actively practicing trauma surgeons. Following this, two cochairs (B.S. and L.Z.K.) were chosen to formulate the specific questions and scenarios for the group to consider. All scenarios were based on a severely bleeding trauma patient. There were no changes to the expert panel during the course of this study. This project was not registered with searchable research platforms. Panelists did not receive any compensation for participation. The project was carried out from April to July 2024. An Equator Accurate Consensus Reporting Document (ACCORD)checklist is included (Supplemental Digital Content, Supplementary Data 1, https://links.lww.com/TA/E371).Figure 1: Delphi consensus process. TABLE 1 - Members of the Panel Convened for This Project Faculty Member Name Faculty Member Institution Brian C. Thurston Spartanburg Regional Healthcare System, Spartanburg, SC Noelle Saillant Massachusetts General Hospital and Harvard Medical School, Boston, MA Philip C. Spinella University of Pittsburgh, Pittsburgh, PA Matthew D. Neal University of Pittsburgh, Pittsburgh, PA Christine Leeper University of Pittsburgh, Pittsburgh, PA Jonathan P. Meizoso University of Miami; Ryder Trauma Center, Jackson Memorial Hospital, Miami, FL Bryan Cotton McGovern Medical School at UTHealth, Houston, TX Babak Sarani Center for Trauma and Critical Care, George Washington University, Washington, DC Robert Winfield University of Kansas, Kansas City, KS Jeannie Callum Queen's University and Kingston Health Sciences Centre, Kingston, ON. Canada Lucy Z. Kornblith University of California, San Francisco; Zuckerberg San Francisco General Hospital Ernest E. Moore Ernest E Moore Shock Trauma Center at Denver Health and University of Colorado, Denver, CO Andrew Beckett University of Toronto and St. Michael's Hospital, Toronto, ON, Canada An iterative questionnaire process was conducted via email using Survey Monkey. Consensus was defined as 80% or higher agreement. Two rounds of questionnaires followed by a final consensus virtual meeting were used to arrive at the current recommendations. A pilot survey was not used. Response rate for each step was 100%. Whereas the responses using Survey Monkey were anonymous, the final consensus conference was live and virtual. A 7-point Likert scale was used to assess agreement with each question, and participants were able to comment on each of their responses. Tabulated responses were sent to the panelists after each round of questions. No representative of Haemonetics Corp. was allowed in the conference calls or had access to the questionnaire results. The massive transfusion protocol of Zuckerberg San Francisco General Hospital, an American College of Surgeons–verified Level I trauma center, was used as a starting point for discussions regarding TEG 6S thresholds for transfusion of various blood components and delivery of prohemostatic agents. RESULTS First Round of Questions In the first round of questions, we assessed recommendations regarding blood product transfusion based on abnormalities of the various TEG 6S parameters (Table 2). All members agreed with the statement that an algorithm incorporating TEG 6S parameters should help guide decision-making in hemorrhaging adult trauma patients. All but two panelists felt that there was no need to have a separate algorithm for patients receiving low-titer group O whole blood, as compared with ratio-driven component-based resuscitation. TABLE 2 - Responses to the First Round of Questions Questions Response, n (%) SA A SWA N SWD D SD Q1. An algorithm incorporating TEG 6S parameters would help guide treatment decision-making in bleeding adult trauma patients. 6 (46.2) 6 (46.2) 1 (7.7) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Q2. There is no clinical need for the TEG 6S algorithm threshold values and associated treatments to differ according to whether a 1:1:1 RBC/FFP/PLT resuscitation protocol or LTOWB resuscitation protocol is used. 4 (30.8) 4 (30.8) 2 (15.4) 1 (7.7) 0 (0.0) 1 (7.7) 1 (7.7) Q3. Information from both the manufacturer's instructions and the literature should guide development of the algorithm. 6 (46.2) 4 (30.8) 3 (23.1) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Q4. In a coagulopathic patient, a prolonged CK-R time value suggests that a factor deficiency is contributing to the cause of coagulopathy. 5 (38.5) 4 (30.8) 3 (23.1) 1 (7.7) 0 (0.0) 0 (0.0) 0 (0.0) Q5. When a prolonged CK-R time value is observed, intervention with FFP should be considered. 4 (30.8) 8 (61.5) 1 (7.7) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Q6. In a coagulopathic patient, a decreased CFF-MA value suggests a deficiency in the contribution of fibrin to clot formation 5 (38.5) 5 (38.5) 3 (23.1) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Q7. When a decreased CFF-MA value is observed, intervention with cryoprecipitate or fibrinogen should be considered. 5 (38.5) 7 (53.9) 1 (7.7) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Q8. In a coagulopathic patient, a decreased CRT-MA value suggests a deficiency in the contribution of platelets and/or fibrin to clot formation. 3 (23.1) 8 (61.5) 2 (15.4) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Q9. When a decreased CRT-MA value is observed with no abnormality observed for CFF-MA, intervention with platelets should be considered. 2 (15.4) 9 (62.9) 1 (7.7) 1 (7.7) 0 (0.0) 0 (0.0) 0 (0.0) Q10. In a coagulopathic patient, an increased LY30 value suggests an increase in fibrinolysis. 6 (46.2) 6 (46.2) 1 (7.7) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0 Q11. When an increased LY30 value is observed, intervention with intravenous TXA should be considered. 5 (38.5) 5 (38.5) 1 (7.7) 1 (7.7) 0 (0.0) 0 (0.0) 1 (7.7) Q12. Following intervention to treat the identified deficiency/deficiencies, if bleeding persists, TEG should be repeated at regular time intervals until hemorrhage is controlled. 9 (62.9) 3 (23.1) 1 (7.7) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) A, agree; D, disagree; FFP, fresh frozen plasma; LTOWB, low-titer group O whole blood; N, neutral; PLT, platelet; RBC, red blood cell; SA, strongly agree; SD, strongly disagree; SWA, somewhat agree; SWD, somewhat disagree. Citrated Kaolin Reaction Time All but one panelist agreed that a prolonged citrated kaolin reaction time (CK-R time) value suggests that factor deficiency is contributing to coagulopathy and ongoing bleeding. The dissenting expert commented that the increased CK-R time value may not be due to factor deficiency but rather elaboration of an anticoagulant factor. However, there was unanimous agreement that an increased CK-R time value should be treated with plasma transfusion unless the patient is known to be on an anticoagulant. Citrated Functional Fibrinogen Maximum Amplitude There was unanimous agreement that a decreased citrated functional fibrinogen maximum amplitude (CFF-MA) value suggests a deficiency in the contribution of fibrinogen to the clot, and the patient should receive cryoprecipitate or fibrinogen concentrate. Citrated Rapid TEG Maximum Amplitude Similarly, there was unanimous agreement that a decreased citrated rapid TEG maximum amplitude (CRT-MA) value suggests a deficiency in the contribution of platelets and/or fibrinogen to the clot. All but one panelist agreed that platelet transfusion is indicated in the setting of a decreased CRT-MA with a normal CFF-MA since a normal CFF-MA excludes fibrinogen deficiency as a contributing factor to this finding. The dissenting expert remained neutral on this question and commented that the question warranted a more nuanced understanding of the contribution of fibrinogen versus platelet activity to the maximum amplitude value. Percent of Clot Lysis at 30 Minutes All panelists agreed that an increased clot lysis at 30 minutes (LY30) is suggestive of increased fibrinolysis, but one panelist strongly disagreed and another was neutral with the statement that an elevated LY30 value should be treated with tranexamic acid (TXA). The person who strongly disagreed with this recommendation stated that TXA should be given to all bleeding trauma patients within 1 hour of injury regardless of the LY30 value because the studies supporting administration of TXA for bleeding trauma patients are based on clinical signs of severe hemorrhage rather than TEG. Repeat Testing All agreed that TEG 6S should be repeated at regular time intervals until hemorrhage is controlled. However, there was no agreement regarding the time should with panelists a time as and after a of blood have been as after of red blood or one to two rounds of a massive transfusion Round of Questions In round we assessed recommendations regarding specific threshold values for each as for transfusion of blood or delivery of prohemostatic agents to a severely bleeding trauma patient (Table The threshold values used were based on the manufacturer's recommendation of the normal limits for each TABLE 3 - Responses to the Round of Questions Delphi Consensus Response, n (%) SA A SWA N SWD D SD Q1. In a coagulopathic patient, a CK-R time value of than minutes of suggests that a coagulation factor is contributing to the cause of coagulopathy. 4 (30.8) 7 (53.9) 2 (15.4) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Q2. When a CK-R time value of than minutes is observed, intervention with 4 FFP plasma should be considered. 2 (15.4) 5 (38.5) 2 (15.4) 1 (7.7) 2 (15.4) 1 (7.7) 0 (0.0) Q3. When a CK-R time value of between and minutes is observed, intervention with 2 FFP plasma should be considered. 1 (7.7) 6 (46.2) 3 (23.1) 0 (0.0) 1 (7.7) 1 (7.7) 1 (7.7) Q4. In a coagulopathic patient, a CFF-MA value of than of suggests a deficiency in fibrin contribution to clot formation. 4 (30.8) 7 (53.9) 2 (15.4) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Q5. When a CFF-MA value of than is observed, intervention with cryoprecipitate or fibrinogen should be considered. 3 (23.1) 9 (62.9) 0 (0.0) 0 (0.0) 1 (7.7) 0 (0.0) 0 (0.0) Q6. In a coagulopathic patient, a CRT-MA value of than of suggests a deficiency in platelet and/or fibrin contribution to clot formation. 5 (38.5) 6 (46.2) 1 (7.7) 0 (0.0) 1 (7.7) 0 (0.0) 0 (0.0) Q7. When a CRT-MA value of is observed, with no abnormality observed for CFF-MA, intervention with platelets should be considered. 2 (15.4) 8 (61.5) 2 (15.4) 0 (0.0) 1 (7.7) 0 (0.0) 0 (0.0) Q8. When a CRT-MA value of than is observed with no abnormality observed for CFF-MA, intervention with platelets should be considered. 1 (7.7) 6 (46.2) 3 (23.1) 0 (0.0) 1 (7.7) 0 (0.0) 2 (15.4) Q9. In a coagulopathic patient, a value of than of suggests an increase in fibrinolysis. 5 (38.5) 7 (53.9) 0 (0.0) 0 (0.0) 0 (0.0) 1 (7.7) 0 (0.0) Q10. When a value than is observed, intervention with TXA should be considered. 4 (30.8) 5 (38.5) 2 (15.4) 0 (0.0) 0 (0.0) 1 (7.7) 1 (7.7) Q11. Following intervention to treat the identified deficiency/deficiencies, if bleeding persists, the TEG should be repeated 2 (15.4) 8 (61.5) 2 (15.4) 1 (7.7) 0 (0.0) 0 (0.0) 0 (0.0) Q12. Following intervention to treat the identified deficiency/deficiencies, if bleeding persists, the TEG should be repeated with round of 5 (38.5) 5 (38.5) 1 (7.7) 1 (7.7) 0 (0.0) 0 (0.0) 1 (7.7) A, agree; D, disagree; FFP, fresh frozen plasma; LTOWB, low-titer group O whole blood; N, neutral; PLT, platelet; RBC, red blood cell; SA, strongly agree; SD, strongly disagree; SWA, somewhat agree; SWD, somewhat disagree. Citrated Kaolin Reaction Time The first question asked if a CK-R time value than minutes suggests that a coagulation factor is There was unanimous agreement with this However, there was more in to the question, which asked if 4 of plasma should be the CK-R time value of agreed with this but and one person was The on the recommendation for a 4 of plasma to be commented that should be based and should also for the of in the CK-R In because of plasma product plasma is in and not by of the Responses to the question, which asked if 2 of plasma should be the CK-R time is between and were with panelists regarding of plasma by 2 In panelists plasma transfusion the is within normal that one to the of hemorrhage in the to Citrated Functional Fibrinogen Maximum Amplitude There was agreement that a CFF-MA value than a deficiency in the contribution of fibrinogen to clot formation. In all but one panelist agreed that transfusion of 4 to 6 of fibrinogen is indicated in this The dissenting panelist stated that a CFF-MA value of 8 to is for clot formation and therefore felt that a threshold of for transfusion was Citrated Rapid TEG Maximum Amplitude The question as to whether a CRT-MA value than suggests a deficiency in platelet activity and/or fibrinogen all but one panelist felt that this was the on this question that a threshold of was high and that transfusion should be until a threshold is as to In to the panelists were asked if a CRT-MA threshold of to should be treated with one adult for platelet transfusion the CFF-MA is All but one panelist agreed with this, but the dissenting panelist stated that the threshold for transfusion should be to panelists disagreed with a question if 2 of platelets should be indicated the CRT-MA is than commented that there are no studies a for platelet transfusion, transfusion of 1 adult for at a time with with TEG or clinical and one person that 2 should be at a CRT-MA as than to Percent of LY30 The panelists were asked threshold values for All except two panelists agreed that a LY30 value than should be treated with One panelist that all bleeding trauma patients should receive as The panelist that the threshold for TXA is and a higher threshold should be used. Repeat Testing in to repeated with TEG 6S, the panelists agreed that the should be repeated One person strongly disagreed that TEG 6S should be repeated with each round of a massive transfusion because of the time for transfusion of Consensus Following round a consensus virtual meeting was of 13 panelists of the consensus meeting were sent to the panelists who not and that agreed with the consensus (Table The final algorithm is in TABLE 4 - Consensus Responses and A CK-R time suggests that a coagulation factor may be contributing to treatment with plasma at was agreed to use plasma by to make the algorithm more to centers of the a of to assess clinical need and if was agreed that is the that would be of clinical value in an actively bleeding trauma patient. A CFF-MA value of suggests a deficiency in fibrin contribution to clot formation, and treatment with a of fibrinogen is a CFF-MA fibrinogen as either fibrinogen or was felt that the of fibrinogen should be with to the fibrinogen of their cryoprecipitate if used as the of The American to a that cryoprecipitate is to 3 fibrinogen and to 6 fibrinogen concentrate. A from blood to an of A CRT-MA value of suggests a deficiency in platelet and/or fibrin contribution to clot formation. the CRT-MA value is but the CFF-MA is treatment with a of platelets should be CRT-MA between and 1 adult to 1 a may 1 or 2 adult to the was chosen because of clinical with using the TEG values between and the to to whether 1 or 2 of platelets would be as there may be a of the appropriate and there is no published A LY30 value of than suggests that an increase in and treatment with TXA should be if not and within 3 of the In patients with with by TEG after TXA administration may be at There was the appropriate LY30 for an actively bleeding patient, was following manufacturer of normal in of a of clinical in trauma with the TEG As patients may have been TXA at the of or more in the it was to the that TXA should be given if not In the regarding 3 from the injury was given the increased risk associated with TXA administration in However, it was to that it is to TXA of this in of with fibrinolysis. bleeding persists, TEG should be repeated or with round of A of TEG was felt to be useful and practical in a clinical with a of 30 to was also felt to the for after a of or after a This was as round of and in as an the FFP, fresh frozen plasma; LTOWB, low-titer group O whole blood; massive transfusion PLT, platelet; RBC, red blood algorithm on the use of the TEG 6S to massive Kaolin Reaction Time There was agreement that plasma should be at to 30 the CK-R time value All felt that is the in an adult patient who is bleeding and Citrated Functional Fibrinogen Maximum Amplitude In to CFF-MA, there was agreement that 4 to 6 of fibrinogen should be either as fibrinogen or cryoprecipitate the CFF-MA value is than Citrated Rapid TEG Maximum Amplitude there was agreement the CRT-MA is to and the CFF-MA is a bleeding adult patient should receive 1 one adult of and, the CRT-MA is than the should 1 or 2 of The threshold for platelet transfusion was from the manufacturer to based on expert and clinical Percent of LY30 There was agreement that an LY30 value than should be treated with all but one person agreed that TXA should be to patients in who more than 3 following injury if the LY30 value is than The dissenting expert felt that may increase TXA is given more than 3 from the time of injury but also stated that this requires more study. Repeat Testing In of of repeat with there was agreement that the should be repeated 30 to minutes or with each round of massive transfusion until bleeding is controlled. is a useful tool for the and specific of coagulopathy in trauma patients and for a goal-directed transfusion during a massive transfusion is able to but the specific threshold values in the literature are based on the TEG 5000 platform, which is being replaced by the TEG 6S platform. This group of experts was able to agreement regarding specific TEG 6S thresholds that should for transfusion or delivery of prohemostatic agents and for the or in severely coagulopathic adult trauma patients. is not practical to repeat the studies that have been published using the TEG 5000 platform in trauma patients with the TEG 6S platform. As one has to use clinical and of the literature to using the TEG 6S platform. were able to a group of acknowledged experts in and to a practical algorithm to blood transfusion in bleeding adult patients. recommendations also provide a starting point for which to study use of TEG 6S in this patient One of the of was the of There is in the of plasma to critically and severely injured panel agreed that a of at of plasma is to make a to coagulation and coagulation Similarly, there are studies the of platelets that should be given to a bleeding patient. suggests that the platelet can be to by for each of platelet However, there not to be a between platelet transfusion, platelet or platelet and clinical in injured in part because of ongoing The panel commented that platelet should be in to platelet and it is in the setting of injury to identify changes in platelet a normal platelet In to TXA there are studies, including associated with administration of TXA following There are also studies that there is no between TXA administration and the Denver group has published studies that a in patients with As we the of TXA in trauma patients unless there is the to a TEG than 1 hour from In TEG can be used to identify patients with As group as a whole felt that TXA should be given regardless of time from injury if the LY30 value is elevated However, this has been and expert recommendations As ongoing research is on the of early TXA administration on the Repeat TEG until hemorrhage is also the to identify early administration of the of the panelists who a of is that it is based on expert The panelists which were on the to by the responses for the first two rounds of questionnaires anonymous, but the final consensus conference was virtual and This may have in to responses. the of the CRT-MA the threshold values for transfusion are based on published TEG 6S manufacturer limits defining and the of transfusion are based on previously published studies as as on expert recommendations for thresholds to transfusion or to prohemostatic as as the of that should be massive transfusion for the hemorrhaging patient, including the use of and factor were not included in the that the questions as part of the Delphi process. As the of literature in this may need to be in the the TEG 6S thresholds for of for goal-directed massive transfusion in this The thresholds provide a starting point for the of and studies in injured patients.

Topics & Concepts