|Values are valid only on day of printing.|
Click CC to turn on closed captioning.
Published: March 2014Print Record of Viewing
In an ideal world, all trauma patients needing transfusion would receive ABO-identical plasma. However, in reality, many patients received ABO-compatible plasma, which results in an increase in overall complications. In this presentation, Dr. Stubbs reviews the current knowledge about transfusion complications related to the emergency use of plasma and the patient care benefits of using group A thawed plasma.
Presenter: James Stubbs, MD
Welcome to Mayo Medical Laboratories Hot Topics. These presentations provide short discussion of current topics and may be helpful to you in your practice.
Our speaker for this program is Dr. James Stubbs, Chair of the Division of Transfusion Medicine, Department of Laboratory Medicine and Pathology at Mayo Clinic. Dr. Stubbs discusses the Mayo Clinic experience in the use of group A thawed plasma in emergency situations.
Thank you for that introduction and it’s a pleasure to be here today. I have nothing to disclose.
The Mayo Clinic practice trauma center in Rochester was established in 1994 and treats over 2400 patients annually. It utilizes a multidisciplinary approach to maximize the quality of patient care. Our level 1 trauma facility serves a large area, including portions of western Wisconsin and northern Iowa
This is some of the area we serve, and the icons show the different transport options that are utilized. When we first looked at using thawed plasma in our level 1 trauma center, the area we served was 18,600 square miles and encompassed 30 referring hospitals. No facility that we served had on-site thawed plasma available, and there were only 3 hospitals that actually stocked plasma on site. This created what we termed a “geographic plasma deficit” that was magnified by a rather prolonged mean transport time of 40 minutes to our level 1 facility.
Now, let’s talk a little bit about trauma and the ability to clot following injury. In severely injured patients, approximately 25% actually have a hemostatic problem, or what we call a coagulopathy, when they arrive at the trauma center. As a result, the standard approach has evolved to trying to correct trauma-induced coagulopathy with aggressive transfusion therapy. Massive transfusion protocols that utilize early and aggressive coagulation factor replacement with plasma have become the norm, and using thawed plasma as the replacement plasma has become an integral part of these protocols.
Let’s define ABO-identical and ABO-compatible plasma. ABO-identical plasma comes from donors who have the identical ABO group as the recipients. By contrast ABO-compatible plasma is defined as plasma that would not cause hemolysis because of isoagglutinins present in the plasma product, but it is not of an identical ABO group as the recipient. Now, in an ideal world, all patients would receive ABO-identical plasma. In fact, some investigators found that the administration of ABO-compatible plasma has resulted in an observed increase in overall patient complications, in particular, acute respiratory distress syndrome and sepsis.
Unfortunately, we live in a non-ideal world, and it is not feasible or possible to transfuse ABO-identical plasma in all settings. The concept of damage control resuscitation using early and aggressive transfusion therapy to combat trauma-induced coagulopathy, often requires the administration of plasma prior to the determination of the patient’s blood type. Traditionally, it has been understood that Group AB plasma should be utilized as the “universal donor plasma.” Group AB plasma is considered a universal donor because it does not have any anti-A, anti-B, or anti-A,B isoagglutinins present and, therefore, that obviates the risk of passive isoagglutinin-mediated hemolysis.
But AB plasma also has risks, and one of the major risks that we’ve noted is transfusion-related acute lung injury or TRALI. The Red Cross recently reviewed the effects of using predominately male plasma as a mitigation procedure to reduce the risk of TRALI, which was initiated in 2006. Prior to implementation, the Red Cross found that there were 18.6 TRALI cases per million plasma units distributed by the Red Cross. In 2008 through 2011, which is a period of time post-TRALI mitigation with predominantly male plasma, the observed number of cases dropped to 4.2 TRALI cases per million plasma units distributed, which is a statistically significant decrease.
Interestingly, when you look at just AB plasma, pre and post TRALI-risk mitigation strategy, AB plasma cases were unchanged, there were 26.3 cases of TRALI per million AB plasma units distributed. One of the main reasons for this is that only 60% of the AB plasma came from male donors, while 40% was from females. Compare postmitigation TRALI risk with A, B, and O plasma, and it dropped to 1.8 per million units distributed, and that is directly related to the fact that >99% of the donors of A, B, and O plasma were male. So the odds ratio of TRALI risk associated with AB plasma compared to A, B, and O plasma in this study was 14.5.
Well, what about the use of Group A plasma vs. Group AB plasma for emergent transfusions? Should we worry about the fact that Group A plasma has anti-B in it? At this time, our practice for emergent transfusions is to use Group A thawed plasma from male donors only, which is a TRALI-mitigation strategy. We use Group A plasma as a replacement for Group AB plasma in emergent settings because of greater availability and better TRALI mitigation. Our clinical experience with the use of Group A plasma in emergent settings has been very, very positive in that we have had no clinical reports of hemolysis from Group A thawed plasma in any patient since 2008 when we first implemented the Group A emergent plasma transfusion protocol. So, clinically, it appears to be safe.
From a laboratory perspective, we decided to look at the extent of anti-B titers in our male A plasma donors to determine the potential risk of hemolysis in Group B or Group AB recipients if, in fact, they were to receive A plasma with a significant amount of anti-B in it. In the laboratory we performed anti-B titers on 120 male Group A whole blood donors using serial dilutions of serum and saline that were reacted with commercially prepared 3% Group A1 red cells. The tubes were incubated for 30 minutes at 37°C, washed and read for agglutination both macroscopically and at the antihuman globulin phase. We considered a significant titer to be the highest dilution that reacted 1+ with our agglutination scoring system.
Our results are shown here. The median anti-B titer was 16, no donor had a titer of more than 512, and close to 92% of our donors had a titer of 64 or less. So anti-B titers in our male Group A plasma donors appear to be very low, and that’s consistent with what we’ve seen clinically, with no evidence of hemolysis in any patient who has received emergent Group A plasma since incorporation of this emergent transfusion strategy in 2008.
What about the practical realities of the use of AB plasma as opposed Group A plasma? Group AB, as a blood type, is present in 4% of the US population. With increasing adoption of damage control resuscitation for severely injured trauma patients, if Group AB thawed plasma was used for all emergent transfusion episodes, it could pose a significant supply challenge. If, in fact, group-identical are preferable to group-compatible plasma transfusions, sufficient supply must be available to meet group-identical AB plasma transfusion demands in both emergent and nonemergent settings. In this situation, the emergent demands could potentially compromise the residual supply of Group AB plasma for transfusions in nonemergent settings.
So we evaluated our ability to supply Group AB plasma to our trauma services with the available donor pool at Mayo Clinic. What we found was that in 1 year we collected 872 AB whole blood units; 353 of those AB plasma units were required to manufacture cryoprecipitate. That leaves 519 AB plasma units available for use as thawed plasma. Over a period of 1 year, it required 1349 thawed plasma units to appropriately stock our Emergency Department and Helipad refrigerators. Therefore, we would have a shortfall of 830 units if we relied on AB thawed plasma. So, even if we wanted to, currently we could not supply those refrigerators solely with AB thawed plasma; we simply don’t have the supply.
So for Group AB vs Group A emergent plasma transfusions, the reality is that the AB plasma supply must rely on a significant percentage of female donors in order to meet demands. This results in a significantly increased risk of TRALI related to the female donor plasma pool. The choice we had to make, above and beyond the shortfall from a supply standpoint, was a remote risk of hemolysis in association with the transfusion of A plasma to Group B or AB recipients vs a genuine risk of TRALI with Group AB plasma with our current donor pool. Remember, the American Red Cross recently estimated the risk to be 14.5 times the risk of Group A, B, or O plasma. From a patient care perspective, therefore, we chose to go with Group A plasma as our component of choice for emergent plasma transfusions.
In 2008, we implemented Group A plasma for emergent transfusions in our trauma patient population. We did a few small initial studies just to find out what the impact would be on supply and outdating, and reported our results in “Stocking Thawed Plasma in a Level 1 Trauma Emergency Department.” Because thawed plasma has a 5-day storage period in the refrigerated environment, we established a rotation out of the Emergency Department on the third day, restocking with younger units, and utilizing the older thawed plasma in the hospital patient population as needed. The Transfusion Laboratory was immediately notified of the use of the thawed plasma in the ED, and we would restock the ED refrigerator with additional Group A thawed plasma units.
When we reviewed this process and compared standard plasma, with a 24-hour use period with thawed plasma having a 5-day use period, our inventory loss decreased by 74% using thawed plasma. We were able to recycle the majority of the thawed plasma that cycled out of the ED with other Mayo patients. Interestingly, what we found in emergent transfusions was that 1.26% of those transfusions were ABO-incompatible, that is, they went to Group B patients. When we looked at percentage of patients as opposed to the percentage of transfusions, 3.03% of patients actually got ABO-incompatible plasma transfusions without any adverse consequences reported to us.
Another study looked at the effect of thawed plasma on discard rates for 6-month periods pre- and post-thawed plasma implementation, which occurred in July of 2008. This compares the 24-hour out-date of thawed fresh frozen plasma to the 5-day out-date of thawed plasma. Preimplementation, a total of 427 plasma units were discarded over the 6-month period, of which 46.4% of them outdated; in other words, they reached the limit of their acceptable storage and had to be discarded. Postimplementation, we discarded 394 plasma units, but only 13.7% of those actually outdated. So the utilization of thawed plasma vs fresh frozen plasma stored in the refrigerator resulted in a large decrease in out-date rates.
In a much larger study we looked at our clinical experience in trauma patients. It was a retrospective review of all trauma patients transfused with at least 1 emergency release Group A plasma units from 2008 through 2011.
We collected information on a large number of patients, the breakdown is shown here. Our review showed that Group A plasma, ABO-compatible plasma was transfused to 219 patients or 86% of the plasma recipient population, and the remainder, therefore, were ABO-incompatible.
As outlined here, you can see that there were no differences in the ABO-compatible and ABO-incompatible plasma transfusion groups with regard to their demographics, injury severity scores, predicted survival based on the trauma and injury severity score, the percentage of patients transferred from the scene, the time from injury to the admission to our trauma bay, the amount of time in our trauma bay, or the time spent at our referring hospitals. So basically, we are comparing apples to apples with regard to the patients who received ABO-incompatible vs ABO-compatible plasma transfusions.
During this study, 1647 plasma units were transfused; 734 were actually emergency release plasma transfusions, so almost 50%.122 plasma units or 7% of all the transfusions were ABO-incompatible plasma transfusions. If you consider that 734 plasma units were emergency release plasma transfusions, and you add to that the 25 Group AB plasmas that were transfused overall, if we had used Group AB plasma as opposed to Group A plasma for emergent transfusions, we would have performed 759 Group AB plasma transfusions when, in fact, the number of Group AB plasma transfusions was 25. We had an overall a 97% reduction in the number of potential Group AB plasma transfuse ons as a result of using Group A plasma for emergent plasma transfusions.
This looks at the ABO-incompatible and ABO-compatible recipients from the standpoint of outcomes. You can see that the overall and specific rates of complications were similar. In particular, there were no differences between the groups with regard to the rates of acute lung injury, the rate of TRALI, or acute respiratory distress syndrome, acute renal failure, sepsis, or death.
This table describes the blood product use for recipients of ABO-compatible and ABO-incompatible plasma transfusions. There is 1 statistically significant difference disclosed in this table, and that’s where only the ABO-incompatible group received ABO-incompatible units, which is by design, and it does achieve statistical significance. But all other transfusion parameters between the ABO-incompatible and -compatible groups are comparable and not statistically significant.
This table compares data on patients who received massive transfusions and whether they got a high plasma-to-red-cell ratio of transfusions vs a low plasma-to-red-cell transfusion ratio in their initial resuscitations. As you can see, there were no statistically significant differences in outcomes between the high and low ratio groups.
What is interesting here is that the mortality rate of massively transfused patients who received at least 1 ABO-incompatible plasma unit was 8%, while the mortality rate of the massively transfused patients who received all ABO-compatible plasma units was 40%, which achieved statistical significance. We’re not sure exactly what this means regarding the true meaning of incompatible vs compatible plasma; it may just be a byproduct of having low numbers of massively transfused patients. Suffice it to say, the ABO-incompatible plasma recipients did as well as the ABO-compatible recipients from the standpoint of mortality.
In total, 272 apheresis platelets were transfused to 103 patients. Both groups, ABO-compatible and ABO-incompatible, received an average of 1.1 apheresis platelet units. No plasma incompatible platelet transfusions were administered to either group, so the only incompatible plasma that was administered in these 2 groups came via plasma transfusions.
More importantly, there were no episodes of hemolysis associated with the emergent transfusion of Group A plasma, and there were no differences in adverse events in patients who received ABO-incompatible vs ABO-compatible plasma.
We have subsequently extended our emergent Group A plasma transfusions into the prehospital environment of our air ambulances and are able to administer such components out in the field. This is one of the publications that describe our experience with prehospital plasma resuscitation. To be able to administer early resuscitation and combat trauma-induced coagulopathy in severely injured patients, we implemented a “Pre-Hospital Thawed Plasma First Transfusion Protocol,” which can occur out in the field and in our air ambulances. The protocol calls for the delivery of thawed plasma by our highly trained personnel in the air ambulance services.
As you can see, we also have developed a Pre-Hospital Massive Transfusion Protocol. This clearly describes which patients are candidates to receive such transfusions. Based on this criteria, patients with 2 or more of the conditions listed would trigger our prehospital massive transfusion protocol.
In further study of our remote damage control resuscitation Pre-Hospital Plasma Protocol for warfarin reversal and traumatic brain injury, we have recently published our experience in this particular publication.
This is an aerial picture of the Saint Marys Hospital in Rochester, which is the home of our level 1 trauma center. The red circle marks our helipad where our air ambulances take off and land.
So what has been the evolution of our massive transfusion protocol at Mayo Clinic Rochester? In 2006, we provided groups of blood products on demand. The breakdown is listed here. In 2008, we changed from thawed fresh frozen plasma with a 24-hour out-date to thawed Group A plasma as part of the massive transfusion protocol.
In 2009, we were stocking 4 Group A thawed plasma in our ED refrigerator and 2 Group A thawed plasma in the Helipad cooler. With the more aggressive use of plasma in the damage control resuscitation of trauma patients, we evolved to: in the air ambulances, 3 thawed plasma and 3 red blood cells to allow for a 1:1 plasma to red cell ratio for resuscitation out in the field; and we evolved to stocking 6 red cells and 6 thawed plasma in our ED refrigerator for use in the trauma room; and our hospital massive transfusion pack is now 6 red cells, 6 thawed plasma, and 1 apheresis platelet for delivery to the bedside for massive transfusion protocols. We no longer provide cryoprecipitate as part of that package.
I hope you enjoyed this presentation and I thank you very much for your attention.