Proactive Use of Plasma and Platelets in Massive Transfusion in Trauma

Some 3 decades ago, a small number of frontline clinicians started calling for the establishment of a massive transfusion protocol (MTP), consisting of a proactive and more aggressive use of fresh frozen plasma (FFP) in exsanguinating trauma patients.1,2 MTP was to be applied only to patients presenting with shock, acidosis, hypothermia, major abdominal bleeding, uncontrolled hemorrhage, and/or pelvic fracture.3 Traditionally, hypovolemic shock in general had been resuscitated first with crystalloid solution and packed red blood cells (PRBCs), with FFP given after ≥10 units of PRBCs had been infused, a high INR had been reported, or diffuse bleeding was evident.2,3 Most clinicians, ourselves included, had found the turnaround time for laboratory coagulation studies and the wait for the FFP to be delivered and thawed unacceptably long in severe trauma. Because many of the most seriously injured patients were already coagulopathic, acidotic, and hypothermic on presentation, it was felt that early use of plasma might prevent further deterioration of the coagulopathy, or even reverse it. Intuitively, the amount of plasma given should be such that when combined with PRBCs, the resulting mixture should also keep the hematocrit within an acceptable range. Thus, starting with a plasma:PRBC ratio not far from 1:1 until blood results are available and/or clinical stability emerges seemed appropriate. In the trauma literature, because of the similar number of units of plasma and PRBCs, MTP is also commonly known as “1:1” or “formula-based protocol.” Much of the earlier work had focused on plasma as maintaining adequate coagulation is more challenging than maintaining an adequate platelet count in an exsanguinating patient. When platelet is included, MTP is sometimes also referred to as “1:1:1.” In the ensuing years, there was much resistance to adopting this MTP in resuscitating exsanguinating patients. Without exception, one main reason put forward against MTP was the lack of randomized controlled trials (RCTs) validating it. As recently as June 2011, the Canadian National Advisory Committee on Blood and Blood Products determined that a transfusion ratio for plasma:platelet:PRBC of 1:1:1 cannot be recommended as the standard of care for massive transfusion during trauma resuscitation in Canada because of the lack of RCTs.4 Advocates of MTP might even have been considered reckless for not insisting on practicing evidence-based medicine (EBM). It was not until the wars in Afghanistan and Iraq that the concept of earlier and more aggressive hemostatic resuscitation began to gain more traction. The apparent need for more proactive and aggressive use of plasma or plasma and platelet in severe trauma spawned many retrospective studies comparing patients who had received blood component ratio approaching 1:1:1 against those who had much less plasma/platelet.5–8 These studies showed a strong association between cumulative plasma/platelet use approaching 1:1:1 with PRBCs and 24-h survival. However, a true validation of MTP (early plasma ± platelets and at equivalent number of units as PRBCs) was not possible because the only patients available for inclusion in these studies were transfused the traditional way (plasma/platelet started late). Most patients actually did not receive “1:1” or “1:1:1”; those who got high doses of plasma ± platelet received them late, like everyone else, and had to survive long enough to be “caught up.”6–8 A survivorship bias was created when those early deaths were categorized in the low plasma/platelet group. Although many subsequent studies attempted with various degrees of success to eliminate this bias,6 this flaw was often used to paint with the same brush all studies meant to validate MTP. Indeed, the Canadian National Advisory Committee on Blood and Blood Products stated in June 2011 that future retrospective studies would likely contain survivorship bias and would not contribute to the determination of the most effective ratio.4 In fact, both sides of this debate were guilty of confirmation bias, that is, citing mainly studies, many of which are equally flawed, that bolster their points of view. During (and after) resuscitation, some of us had received inquiries from the blood bank and colleagues: “Why FFP so early and so much?”; “What are (were) the INR and platelet count?”; “Where is the evidence for MTP?”; “What about survivor bias?” This “scorn” went on for years. To date, no RCT comparing the traditional approach with MTP approach has yet been done. Nonetheless, based on a small 2011 survey, 70% of academic medical centers in the United States had abandoned the traditional approach and, instead, adopted a protocol that calls for the early release of MTP packs consisting of a similar number of units of plasma, platelets, and PRBCs.9 The European multidisciplinary Task Force for Advanced Bleeding Care in Trauma’s 2016 guideline on management of major bleeding and coagulopathy after trauma offers “strong recommendation in support of a plasma:PRBC ratio of ≥1:2” to be applied “to most patients in most circumstances without reservation.”10 Recognition of the importance of fibrinogen is transforming the approach further to include fibrinogen or cryoprecipitate in some centers.11 How has this come about? What happened to EBM? Most importantly, what has finally silenced the critics and brought about the acceptance? WHAT REALLY IS THIS MTP ANYWAY? In the face of exsanguination, MTP is a crude attempt to replace what is being lost—whole blood—real time. The optimum ratios between these blood components are likely variable because trauma is a heterogeneous condition. The best plasma:platelet:PRBC ratio, if there was one, could be, say, 1:1:1–3. For example, Kashuk et al12 showed in their retrospective study that mortality associated with a plasma:PRBC ratio of 1:2 may be lower than that associated with a ratio of 1:1 and 1:3. Likewise, Teixeira et al13 retrospectively determined that the best ratio appears to be 1:2–3. In contrast, Maegele et al14 found in their observational study that a mean plasma:PRBC ratio of 1:1.9, 1:1, and 1.5:1 was associated with highest, lower, and lowest 6-h, 24-h, 30-day, and in-hospital mortality rates, respectively. The PROMMTT (PRospective, Observational, Multicenter, Major Trauma Transfusion) study found that in the initial 6-h time interval, patients in the moderate (both plasma:PRBC and platelet:PRBC >1:2 and <1:1) or high ratio group (both plasma:PRBC and platelet:PRBC ≥1:1) had lower mortality rates than the low ratio group (both plasma:PRBC and platelet:PRBC ≤1:2).15 Importantly, plasma and platelet ratios were not associated with 24-h or 30-day mortality in that study, suggesting no overall harm due to early use of these blood products.15 The authors also found that early transfusion of plasma was associated with reduced 24-h and 30-day mortality compared with patients who received lower plasma:PRBC ratios, or who did not receive early plasma but were “caught up” to ratios approaching 1:1 by 24 h.15 The PROPPR (Pragmatic, Randomized Optimal Platelet and Plasma Ratios) trial randomly assigned patients to transfusions of plasma, platelets, and PRBCs in ratios of either 1:1:1 or 1:1:2.16 Similar to the PROMMTT study, death from hemorrhage was significantly less common in the 1:1:1 cohort at 6 h after injury; however, no significant difference was seen at the 24-h and 30-day time points. No differences in the rates of ARDS, multiorgan failure, thromboembolism, and sepsis were seen between groups.16 Therefore, although the “1:1:1” term is sometimes used synonymously as MTP, it is important to note that the best ratio, if there was one, has been very much the subject of investigation. That said, evidence so far suggests that a plasma:platelet:PRBC close to 1:1:1 is a good start. Ratios significantly lower than 1:1:2 would fall outside the “1:1:1” paradigm. Of note is that mixing a unit of plasma, platelets, and PRBCs does not reconstitute a unit of fresh whole blood. Because of degradation and the presence of diluents, the 1:1:1 mixture has a coagulation factor activity of 40%–50%, platelet count of 90×109/L, and hematocrit of ~26%.17 As we strive for homeostasis (under close clinical observation and laboratory/point-of-care monitoring), a ratio significantly different from this borderline anemic, coagulopathic, and thrombocytopenic mixture may make it difficult to restore homeostatic balance later. MTP is not achieving equivalent numbers of units of plasma, platelets, and PRBCs late (after 6 or 24 h) during the resuscitation, as might be the case with the traditional way of transfusing a continuously hemorrhagic patient. Although Group O PRBC is immediately available for resuscitation, thawed plasma is typically not. As such, 1:1 for plasma during the first 2 to 3 h of resuscitation is rarely achievable. Without MTP and multidisciplinary (emergency medicine, anesthesiology, surgery, blood bank, nursing, etc.) commitment, a 1:1 ratio in the early hours is even more elusive. Stocking refrigerated thawed plasma (factors V and VIII at ~50% of original levels at the end of their 5-day shelf-life) in the emergency department will help achieve “1:1” but has major logistic and resource implications. Even with great effort, a more realistic ratio is a unit of plasma for every 1–3 units of PRBCs during the early hours. That is already a great improvement from the traditional way of transfusing in which the first unit of FFP usually starts late and a near-unity ratio occurs (if ever) much later. In our opinion, one of the objectives of having MTP is to steer clinicians away from the mindset of waiting for the laboratory results (eg, elevated INR) or until 10 units of PRBCs have been given, or microvascular bleeding is obvious, before requesting for FFP to be thawed. In other words, the main difference between MTP and the traditional approaches is not only how much plasma and platelets are given in relationship to PRBCs, but also the timing of when to initiate plasma and platelet transfusion. In the end, both approaches may wind up giving similar blood product ratios as clinicians using the traditional approach play catch up with the clotting factors and platelets. This point is highlighted by Riskin et al,18 who reported in their before-and-after study that after implementation of MTP, the mean times to first units of FFP and platelets were 169 and 241 min, significantly shorter than the 254 and 416 min, respectively, before MTP implementation when the traditional transfusion approach was used. At 24 h, FFP:PRBC was 1:1.8 in both the before-MTP and after-MTP cohorts. Importantly, after MTP implementation, the amount of blood products used and mortality were both significantly reduced.18 Nowadays, many trauma centers have made prethawed plasma immediately available, further shrinking the time gap between the first units of PRBC and FFP. WHERE IS THE EVIDENCE? There have been many observational studies on the early use of FFP and platelets, most, but not all, of them supportive. Unfortunately, no experimental direct comparison between MTP and the traditional method has been done. In a randomized trial attempting to look at MTP versus the traditional laboratory approach, Nascimento et al19 wound up comparing a FFP:platelet:PRBC ratio of 1:1:1 versus 1:0.8:1.7, respectively, at 12 h after admission. Notably, both ratios were within the vaguely defined “1:1:1” paradigm, that is, ~1 unit of FFP/platelets for each ~1–2 units of PRBCs. However, MTP is not just ratios. As mentioned earlier, it requires that the time gap between the first PRBC and FFP/platelet units be kept to a minimum. In this study, because there was no prethawed plasma constantly available, the first unit of FFP was given on average 60 (“1:1:1”) and 78 (laboratory-driven) min after the first unit of PRBCs. For the first hour of admission when hemorrhagic death occurs at a high rate, no plasma or platelets were given in either group. That said, the time gaps between the first PRBC and FFP units in both cohorts were vastly improved from previous studies (eg, Riskin et al18). Our take from their study is that a heightened vigilance for coagulopathy pushes the traditional approach toward MTP. The PROPPR trial mentioned earlier attempted to define the best ratio within the MTP paradigm.16 Blood component transfusions were stopped once hemostasis was achieved and subsequent transfusions were guided in part by laboratory results. The results showed that a starting transfusion ratio of 1:1:1 resulted in significantly reduced mortality from exsanguination within the first 6 h, with similar complication rates to the 1:1:2 protocol.16 However, the study was underpowered to show a significant difference in overall 24-h or 30-day mortality. The Early Whole Blood in Patients Requiring Transfusion After Major Trauma trial compared whole blood + platelets versus PRBC + FFP + platelets based on close monitoring, including of coagulation parameters.20 Clearly both approaches represent different forms of “1:1:1.” This pilot study found that, in patients without head injury, use of stored whole blood reduced transfusion volume.20 THE POWER OF OBSERVATION AND PHYSIOLOGIC SENSE Our own clinical experience suggests that striving for almost as many units of plasma and platelets as PRBCs as early as possible reduces coagulopathy in trauma with massive bleeding but without RCTs, we may never know whether the observed reduction in coagulopathy and apparent improved outcomes in observational studies actually occur. The problem with the traditional approach is that waiting for laboratory/clinical confirmation of coagulopathy in an exsanguinating patient means that the battle may already have been lost. Replacing blood loss with erythrocyte alone until one blood volume has been transfused may worsen the preexisting coagulopathy. The traditional approach evolved from decades of experience from using whole blood for massive transfusion and was brilliant. However, since the late 1970s/early 1980s, with blood fractionation, the use of PRBCs means that dilution of clotting factors appears early. The need for early and aggressive use of plasma is supported by pharmacokinetic models showing that reversing coagulopathy with a fluid that contains 1 unit/mL of clotting factors (namely plasma) takes hours.21 Somehow, the traditional approach to massive transfusion did not get changed along with this change in blood bank practice. THE ADVENT OF EBM Serendipitously, EBM became mainstream in the 1980s, roughly around the time when blood fractionation became standard. RCTs have since become the gold standard and so embarking on a new approach with major resource and logistic implications such as MTP without robust evidence became, in some peoples’ minds, borderline sacrilegious. The mere statement “There is no RCT!” is enough to silence anyone. The irony of this debate between the traditional approach versus MTP is that the former has never been subjected to any trial at all! Thus, we had the situation in which well-intentioned and intelligent people opposed MTP (which makes more physiologic sense and is supported by many observational studies and observations by frontline clinicians), on the basis of lack of robust proof, while defending an approach that had essentially no proof and which makes less physiologic sense.22 The positions on either side of this debate do have legitimate concerns. While anesthesiologists’ and surgeons’ immediate priority is homeostasis, blood bankers must remain good stewards of a valuable resource. All must worry about outcomes such as multiorgan failure, sepsis, fluid overload, transfusion-related acute lung injury, and mortality. Hence, for any investigation or clinical endeavor related to massive transfusion to succeed, it is paramount that all concerned specialties are on the same page. THE TRADITIONAL APPROACH AND MTP HAVE MORPHED INTO ONE There are numerous reasons (cost, logistics, etc.) for why an RCT comparing the traditional approach and MTP has not been done, one being that the 2 approaches have morphed into one and the equipoise is disappearing.23 Indeed, there is good evidence, again gained through military experience, to suggest whole blood and warm fresh whole blood, the ultimate “1:1:1”s, are superior for massive transfusion.20,24,25 As mentioned earlier, the main problem with basing FFP transfusion on laboratory results (part of the traditional approach) is that traditional coagulation tests take too long and FFP takes time to thaw. As technology progresses, the delays in test results is becoming less of a problem. When results are derived from fast point-of-care devices and prethawed plasma is available, the 2 Achilles heels of the traditional approach are no longer, and the relative advantages/disadvantages of the so-called laboratory-based and formula-based approaches start to vanish and indeed look, in our opinion, favorably at the former. For example, use of rapid thrombelastography to analyze coagulation profiles and guide blood component therapy shows great promise of reducing unnecessary plasma use while improving as does the use of factor such as factors and This more and approach to hemostatic is considered by to be more than the However, most trauma centers in do not have and a of coagulation factors is not when a patient is hypovolemic and is bleeding there is sometimes but to restore with a and more physiologic plasma in to The to plasma, such as and all have some have may contain a balance of with other factors and that are important in resuscitation and of most clinicians start with a massive transfusion consisting of similar number of units of plasma, platelets, and PRBCs when there is a high of massive while for blood thrombelastography if INR if blood and of microvascular This is by therapy based on point-of-care and clinical There is a best blood product ratio for each case as it Unfortunately, this initial ratio is not known with the of the blood product almost even within the same case as time In our and based on evidence so striving for a near-unity ratio when exsanguinating patients is a start. some patients will from the equivalent of whole blood from the start. For many if plasma and platelets are given no than after the first units of PRBCs, and are given with equivalent number of units as PRBCs until hemorrhage is we that the of coagulopathy we had seen in the with the traditional way of transfusing could often be As resuscitation clinical and test will guide further Our should be on how to that are to massive and how we and our patients and blood may also be gained by comparing trauma resuscitation results between use of thrombelastography and of factors is more and MTP consisting of “1:1:1” use of blood components is more although differences in case and other of care may the results. EBM AND This is not meant to be a on exsanguination There are already on this from the of and This is a that EBM is a that evidence is in the of RCTs, our should be guided by Therefore, we should and to validate our with In the of RCTs, however, clinical should be based on experience, and less robust evidence derived from observations and studies, while being of their associated For example, given the more approach of “1:1:1” and the of to the of to plasma, and experience, we strive to be proactive in plasma in all clinical as massive in which hemorrhage is immediately and showing no of other words, until we would the same as when whether to use a when of a should keep in that the of (part of all the during the initial in a and coagulopathic, and has also never been subjected to a randomized Indeed, have erythrocyte transfusion and in are not against EBM but against a on it. For example, a that there is no evidence that has improved in Even the will remain the first put on and the to if at all, from patients in the there is that EBM has improved patient and the that EBM is is at the of the EBM no more than may be the for clinical it is only as good as the evidence and the people the an in clinical medicine will through such as RCTs, by some by and often by a of all of must be constantly to whether are were be as in the case of massive transfusion in the of blood fractionation, are no That is how is EBM is a clinical is when no RCT is available or This the This the This the This was