JounalReview_WholeBloodResuscitation

Hey everyone, welcome to another episode of Behind the Knife's Journal Review Series. We are the Miami Trauma Team, back with another episode from Jackson Memorial Hospital's Reiter Trauma Center. In this episode, we're going to be talking about the use of whole blood in the resuscitation of trauma patients.

This is something that's really seen a resurgence in the last five to 10 years or so, and there's a lot of different data out there, so we're gonna try and get you the information you need to know. It's funny how we so often see the pendulum swing back and forth over the years in medicine. All of us in trauma have seen the evolution of blood product resuscitation of the last several decades, and while the use of whole blood was common during early global conflicts.

The discovery of blood fractionation techniques, which allow blood bankers to deconstruct whole blood into its individual components resulted in a shift to the component paced transfusion strategy for the latter part of the 20th century. And so the fractionated blood is just what we had and used for a while.

Now there's been a resurgent interest in the use of whole blood for trauma resuscitation as a result of promising data from

recent world conflicts in Iraq and Afghanistan. But how do we translate this into practice? That's exactly what we're going to focus on today. We're going to take two recent papers on whole blood resuscitation that sucked them and figure out how to translate these findings to clinical practice.

The first paper is a multi centered perspective observational study by Dr. Hazleton and co authors that was published in Annals of Surgery in 2022. And the second is another multi centered perspective observational study that was published in 2022. But this one's by Dr. Jason Sperry and the SWAT Study Group, published in the Journal of the American College of Surgeons in 2023.

I'd also be remiss if I didn't point out that one of the, our co hosts today, Dr. Mezzoso, is the first author of a recent e systematic review, meta analysis, and practice management guidelines on whole blood that was published last year. So we should have a really good discussion today. But before we get any further down the rabbit hole, let's get some introductions out of the way.

I'm Mike Kobler Lichter, PGY 4 in General Surgery, and my second, I have two years of dedicated research with our trauma faculty here in Miami. My name's Eva Oretziga, I'm a vascular

surgery fellow at the University of Pennsylvania, but I graduated from general surgery residency at UM last year. I'm Eugenia Kwan, trauma and surgical critical care attending at Loma Linda University, having graduated from fellowship at the Rider Trauma Center.

My name is Jonathan Maisoso, I'm an attending trauma surgeon at the Rider Trauma Center and assistant professor of surgery at the University of Miami. So let's just get into the first paper. This first one is titled The Use of Cold Stored Whole Blood is Associated with Improved Mortality and Hemostatic Resuscitation of Major Bleeding.

As mentioned, Dr. Joshua Hazleton is the first author on this, and it was a multi center perspective observational study that included 14 different centers. They included any trauma patient who presented to one of these 14 centers and received any blood products as part of their initial resuscitation, which was defined as from hospital arrival to trauma bay or emergency room discharge.

They defined the whole blood group as patients who received any volume of whole blood as part of their

initial resuscitation and the component therapy group as patients who received only component therapy, no whole blood. for the entirety of their initial resuscitation. And just to be clear, this was low titer, cold stored, type O whole blood, not the warm, fresh whole blood which has been used in published works from the military previously.

Of note, the authors state that there was no standardized protocol for determining which patients got whole blood, and that each center basically had their own criteria for this. The primary outcome was in hospital mortality, and secondary outcomes included AKI, DVT, PE, pulmonary complications, TRALI, and bleeding complications.

The authors then did a mixed effects multiple regression to control for center level effects. age, sex, pre hospital blood administration, mechanism, ISS, and then developed Kaplan Meier curves to compare survival between patients in the whole blood group and the component therapy group. So I know we usually have the trainees answering most of the questions in these

episodes, but since you are the real whole blood expert in the room today, we're going to flip this a little bit.

Dr. Mezzo. So. So what are some important considerations in the design of this study that we need to be aware of when determining the effect of whole blood on resuscitation outcomes and trauma? So I think the first thing to be aware of, uh, actually understanding what blood product we're using and talking about.

A lot of the initial work with whole blood was done with data from the military setting where they're using warm whole blood, often from walking blood banks, which is really not the case in the civilian setting. Cold stored whole blood probably behaves somewhat similarly to warm whole blood and likely has some of the same benefits, but it's not the same product and we need to be aware of that when we're comparing different studies on whole blood, especially between military and civilian populations.

The second point is one the authors themselves point out, which is there a significant variation in the decision of who gets whole blood as part of their initial resuscitation, both between centers and even between different providers at the same center. Not only is there variation in who gets whole blood, but what

proportion of the total resuscitation is given in whole blood.

I think we as a field are moving towards giving more patients more whole blood, but there's still a lot of variation, and the data's kind of all over the place. Oftentimes, sicker, more unstable patients will be preferentially given whole blood. And this will certainly have an effect on the results of observational trials.

Authors do a good job here of attempting to control for that, but it's still a potential bias to be aware of and highlights the importance of doing prospective randomized work to assess the potential benefits of whole blood and trauma. Lastly, this study used in hospital mortality as its endpoint.

There is a lot of heterogeneity, as well, between different studies in looking at mortality at specific time points, which also makes the whole field a little bit more difficult to interpret. Um, and when you look at a lot of the work that's been done, In hemorrhage control, there's actually a paper now probably five or so years old by John Holcomb and Angela Sawaya and a bunch of other authors in this topic in hemorrhage control where they

really emphasize.

And I agree with them that we should be looking at earlier time points for mortality when we're looking at hemorrhage control interventions, because that's, that's really where this particular intervention may make a difference, whether that's, you know, three hours, six hours. 12 hour or 24 hour mortality.

If you're implementing a hemorrhage control intervention, that's what you wanted to do is to improve that early survival to get you to the next step of being able to take care of the patient. But that's enough talking for me. Mike, why don't you tell us what this study found? So they were able to get data on 1, 623 trauma patients over five years from 2016 to 2021.

As you would expect, the data was from primarily young males, 83 percent male with a median age of 35. Perhaps somewhat surprisingly, though, was this population was made up of 53 percent penetrating trauma, and 73 percent of the population actually fell into the whole blood group, with only 27 percent in the component therapy group.

The whole blood group tended to be a little bit older,

36 versus 32, more likely male, 87 percent versus 71%, more commonly with comorbid conditions, 33 percent versus 29%, With blunt trauma, 49 percent versus 40%, higher presenting Glasgow Coma Scale, 14 as opposed to 9, and a higher shock index on representation of 0.

98 as compared to 0. 83. Groups were similar in terms of pre existing anticoagulant or antiplatelet use. as well as use of TXA. In terms of lab values, the whole blood group tended to have higher initial and 24 hour hemoglobin values, but lower four hour platelet values. But these are pretty similar. There was no difference in massive transfusion or ultra massive transfusion rate between the two groups around 32 percent for massive transfusion around 9 percent for ultra massive transfusion for transfusion requirement.

The component therapy group got more packed red blood cells in the first 24 and 4 hours as expected, but no difference in the total amount of product transfused at 24 hours. The

whole blood group was 9 percent less likely to experience a bleeding complication. But there were no differences in the rate of AKI, DVT, PE, pulmonary complications, or length of stay between the two groups.

On multivariate analysis, when controlling for age, sex, pre hospital blood product administration, mechanism, and injury severity score, whole blood exposure was associated with a 48 percent decrease in the likelihood of mortality, compared to the component therapy group. And 9 percent decreased likelihood of bleeding complications.

Okay, so to summarize, the whole blood group was probably a little sicker at baseline. There was no difference in transfusion requirement at 4 or 24 hours. But on multivariate regression, whole blood exposure was associated with improved mortality and lower bleeding complication rates. What about any limitations to the study, Mike?

So, the authors do note that they are unable to control for all potential confounding, which is the real crux of the issue in trying to determine the benefit of whole blood through

observational studies. Even with the fancy statistics, parenthesis score matching, multiple regression, et cetera, surgeons are choosing to administer whole blood selectively, and the group that receives whole blood versus only component therapy are not the same, and they cannot be made to be the same when looking at retrospective observational data.

The other important thing that they make note of is the definition of the whole blood group. There is no method of accounting for how much whole blood a person got. If a person received an ultramassive transfusion and got 40 units of blood, a person who got 1 unit of whole blood and 39 units of packed red blood cells would be in the same group as someone who got 40 units of whole blood and no packed red blood cells.

And those are two clearly very different scenarios. Dr. Mezoso, what is your practice in regards to selecting which patients to give early whole blood to and what is the role of whole blood in your mind and trauma currently? Yeah, so I think that's a good question. You know, you made a good point in the, you know, the scenario between between a person who gets one unit of whole blood and then 39 units of PAX cells.

I think that really is going to

make a difference in how we interpret all of the data that's out there right now. Because you know, I don't know that there's, I don't know that there is much of a difference in giving somebody one or two units of whole blood when the rest of their resuscitation is going to be made up primarily of components.

Although, maybe, maybe it will, you know. Dr. Hazleton's study maybe suggests that it does, you know, once they controlled for a bunch of things, they did find that there was still some benefit to getting whole blood. But I think we still need more data to really understand, you know, that kind of discretionary one or two units when someone is just, you know, a touch hypotensive, but don't look severely injured, you know, does that really benefit those patients?

And I think, you know, one important point, since we are talking about whole blood is there is a Prospective multicenter study that is being conducted in a randomized fashion out of University of Alabama called the troop study, where they're looking at a exclusively whole blood

resuscitation in the initial phase versus an exclusively component based resuscitation in the initial phase to see if there is actually a difference.

Between those groups. And I think the results of that study are really going to be telling for what we're going to be doing with whole blood in the next five years or so. I think the other thing that's really important to consider and it's something that we did include in the East practice management guidelines that you mentioned is that.

A lot of this is going to also come down to availability. What I am going to have available to me to be able to give to patients at an urban level one trauma center with all the bells and whistles and all of the resources is very different than what somebody else is going to be able to resuscitate their patients with in a smaller level two or level three hospital or just an emergency room where they may get a trauma patient before they can get them transferred out.

And so those recitations are also not going to look the same. Agreed. I think we hit all the important points of the first article. So let's get to our second and final one,

whole blood resuscitation and association with survival and injured patients with an elevated probability of mortality by Jason Sperry et al and the SWAT group.

Take it away, Ava. What's this paper all about? Yes. So this was again, a multicenter perspective observational study over three and a half years. That took place at seven level one trauma centers. It included injured patients at risk of massive transfusion, which was defined as an assessment of blood consumption or ABC criteria score of two or higher.

And that means they had to have at least two of the following satisfied. So hypotension with a systolic of less than 90. the penetrating mechanism of injury, a positive FAST, and a heart rate of greater than 120. And, importantly, the patients had to have required both a blood transfusion as well as hemorrhage control procedures, either in the operating room or in the IR suite, within 60 minutes of arrival.

Of note, pre hospital vitals and transfusions were included for ABC score calculation. They also note that

deferred FAST for purposes of immediate transport to the OR did count as having a positive FAST for the purposes of calculating that score. Exclusion criteria were children under 15, penetrating brain injury, more than five minutes of CPR, death before initiation of transfer to OR or IR suite, known prisoners, and known pregnancy.

So just looking at these criteria, you can see that the inclusion criteria was pretty different. The first study basically included any patients that were getting blood, whereas in this study patients were at risk of large volume transfusion, so basically patients who needed an intervention for it, not just those who got a single unit.

This study also wanted to examine a subgroup of TBI patients, and for their TBI criteria, they used positive CT brain imaging. Once again though, the whole blood group was defined as at least one unit of low, titer, O, whole blood, either pre hospital or in hospital phases of care, while the component therapy group received no whole blood at all.

Of note, there were, again, no standardized inter facility guidelines for

who would receive whole blood versus component therapy, and again, this was left up to the protocols of each facility or the treating physician. In fact, four of the seven centers did not actually have whole blood resuscitation programs at the time of the study initiation.

The primary outcome was four hour mortality, while secondary outcomes were 24 hour mortality, 28 day mortality, achievement of hemostasis, adjudicated death from exsanguination or hemorrhage, and the incidence of unexpected survivors based on a pre hospital mortality of greater than 50 percent at 28 days.

The authors also compared 4 and 24 hour transfusion requirements, lab values, and incidence of various in hospital complications. Of note, the definition for massive transfusion in this study was 3 units or more of any red blood cell containing product within a 60 minute time period during the first 4 hours of arrival.

So, really critical administration threshold, not just the traditional definition of MTP, compared to the traditional 10 units and 24

hours definition that was used in the other study. And this helps to eliminate some of the bias of those who didn't survive long enough to reach those thresholds. The effect of whole blood exposure on the primary outcome of four hour mortality was then evaluated with an inverse probability weighting and regression analysis to weigh observations by the calculated likelihood of receiving whole blood initially and then subsequently controlled for age, sex mechanism, ISS, pre hospital systolic blood pressure, and the need for pre hospital blood.

Okay, so you touched on some important points that highlight the difficulties in synthesizing all the literature on this topic. There really is a ton of heterogeneity between papers on this topic, from inclusion criteria to local practice patterns and even chosen outcomes, and it sometimes makes comparing studies difficult and complicates how we should be applying the results to the patients that we're seeing ourselves.

What did this study show, though? So this study eventually enrolled 1, 051 patients. As you can imagine, these were again, very injured patients,

median ISS of 22, greater than 60 percent penetrating trauma, over 70 percent of patients requiring their definition of MTP, four hour mortality of 8 percent and 28 day mortality of 17%.

Just under two thirds of patients at whole blood sites received whole blood compared to 75 percent of patients at whole blood sites for the TBI subgroup. So this is actually a lot fewer patients getting whole blood than the first study. This is only two third of patients at the whole blood centers compared to almost three quarters of all patients in the first study.

Again, highlighting how different local practice patterns influence these observational studies. The whole blood patients were more likely to be males with lower SPPs, lower GCS. We're more likely to have TBI and we're more likely to get massive transfusion at four hours. So 75 percent for 65 percent of the component group.

Unadjusted mortality was similar between cohorts, 8. 2 versus 7. 5%. After propensity adjustment, there were again, no differences in four

hour, 24 hour, 28 day mortality in the overall cohort or in the TBI subgroup. There were no difference in progression of TBI, rates of achieving chemostasis at four hours, rates of death from exsanguination, or transfusion requirements in units at four or 24 hours, though the whole blood group received more volume at both four and 24 hours.

median of 3. 2 liters versus 2. 7 liters at 4 hours and 4. 1 versus 3 liters at 24 hours. In terms of hospital complications, there were no differences between rates of VTE, multiple organ dysfunction, or infectious complication, but the whole blood group did have a slightly lower median of vent free and ICU free days.

Coagulation parameters on labs were also very similar except for a very slightly lower LY30 value in the whole blood group at 24 hours. They also looked at the rate of unexpected survivors between the two groups, which is I think where this gets pretty interesting.

They started with a pre hospital model to predict mortality based on vitals and variables available in the pre hospital setting prior to arrival.

And they had pretty good discrimination in this model with an area under the receiver operator curve of 0. 89. They then selected all patients with estimated mortality of greater than 50 percent to look at the actual mortality rates in this group. 73 percent of the component therapy patients with predicted mortality of greater than 50 percent died.

Compared to only 40 percent of the whole blood patients, and in fact, on multiple regression to control for other confounding variables, the whole blood group had a 35 percent lower risk of 28 day mortality. Further analysis showed 48 percent lower risk of mortality in patients with a pre hospital predicted mortality of 5 percent or greater, 33 percent lower risk for predicted mortality of 10 percent greater, and 30 percent lower risk for predicted mortality of 20 percent or greater.

In other words, there is a trend that whole blood offers a survival benefit in those patients who are at a relatively increased

risk of death from their injury. Lastly, they examined the potential for a dose response relationship by including the ratio of total whole blood transfused related to the total component product received in 24 hours, and found that the risk of mortality decreases as the ratio of whole blood to component therapy increases.

So I think this is starting to get at one of the other current hot topics in whole blood resuscitation, which the authors themselves do address. We need to better explicitly characterize the population that actually benefits from whole blood. I think when you look at certain populations, there's a lot of, there's probably a lot more unanswered questions, which is, what is the benefit of whole blood?

in patients with TBI, for example, or polytrauma with TBI. What's the effect on pediatric patients? Is this a population that this could potentially be another therapeutic target for? The other thing I think to consider is that just like, uh, if you think about other interventions that we've studied in the past, right?

So if you think

about, for example, plasma, early plasma, And we talk about the pamper trial and the combat trial, right? The combat trial didn't find a difference between patients who've got FFP in the prehospital setting and those who didn't, but that difference was likely related to the very short prehospital transport time compared to when you look at the pamper trial, which looked at essentially the same question, um, and found that there was an improvement in patients who received FFP in the prehospital setting.

But again, they had much longer prehospital transport times. Or if you look at TXA, for example, you know, there's probably, uh, not something that is beneficial for everyone, but there probably is a population that remains to be better defined. The intervention could provide the most benefit. So I think whole blood is in that same category where, you know, we know that it has its benefits for sure.

We know that it's hemostatic capacity is superior, but we don't know what that actually means or translates into

clinically. And that all needs to be studied a little bit further. Eva, why don't you talk a little bit about the limitations of this study? Sure. So the officers do point out the typical issues with observational studies of this nature that we've already touched on.

How there's inherent differences in populations based on exposure, the whole blood cohort presented with more severe injuries, lower presenting systolic blood pressure, lower GCS, and higher rates of massive transfusion and transfusion requirement. Additionally, the touch on the potential for inter center variability in practice of who gets whole blood, as well as changing practices regarding early whole blood administration over time.

One other thing to consider, though, that the authors did not comment so much on is related to their pre hospital model. In their methods, they state that the only patients in the component therapy group were used to fit the model that predicted pre hospital probability of mortality, but that this model was subsequently applied across the entire cohort.

Although this model demonstrated pretty good discrimination with an area under the curve of 0. 89, This is a little bit misleading. Since

this model was only fitted based on patients in the component therapy group, it'll undoubtedly be more accurate at predicting the mortality in that group than the whole blood group.

Since there aren't any reported metrics regarding the model's performance in the whole blood versus the component group, or any metrics related to the model calibration, we don't know how much better this model is at predicting mortality in one group compared to the other, and we don't know how precise the predictions actually are to begin with.

If there is a large difference, or if the calibration isn't great, it makes comparisons in observed to expected mortality difficult between the two groups, as the predicted mortalities in the whole blood group are going to be less accurate. Similarly, we don't actually know the distribution of the predicted mortalities estimated by the model.

If all the patients in the component therapy group had a predicted mortality close to 100%, but the whole blood group were in the 55 range, they would all be included in this analysis. But clearly the actual expected mortality should be higher in the component group in that case, and it wouldn't necessarily be an effect of the treatment.

Regardless, it still seems like there certainly is a

signal that the whole blood exposure likely does improve mortality in this population. Okay, that's a pretty extensive summary of these two articles. Drs. Kwan and Mezzo, so maybe we can finish up by having you two talk a little bit about how you currently view and use whole blood in your practice.

the patient populations you think it makes the biggest differences in and necessary future directions for research in this topic. I think Dr. Mezzo mentioned this earlier, but the role of whole blood and resuscitation is really going to depend on what's available at the institution and what protocols are in place.

You know, at Ryder and Loma Linda, whole blood's available in the recess bay. And I think Loma Linda, we're actually in the works of getting approved for the use of whole blood in all trauma patients, including pediatrics. Bye. There's data out there that's showing that, is it safe for all of these populations?

Sure, but I think there's probably a need for further research, you know, whether there's always these questions about, um, like RH factor or pediatrics, like who specifically is going to benefit the most from getting whole blood versus component therapy. And, and I

think that's, that's something that still needs to be looked at.

further clarified? I don't know. What do you think, Dr. Meza? So, yeah, um, no, I think those are good points. I, you know, in my opinion, I think that the patients that likely benefit the most are those that are in hemorrhagic shock, right? Those are kind of the, I think that's what gets to the meat of this is what we're really should be talking about are patients who are in shock, who require, you know, a massive transfusion.

And those patients, I think almost for sure, you know, until we have a A definitive randomized trial. We won't be able to say for sure, but almost for sure. I think those patients will benefit from getting more whole blood than component therapy. Outside of that, I think that, like we mentioned already, I think the data are all over the place and you can really make an argument, I think, either way.

I don't think anyone would fault you for using, uh, whole blood if you have that, uh, commodity available if, if you do need to give the patient any transfusion, but I, I don't think

that that's something that we can say across the board is what should happen as Dr. Kwon mentioned. Safety wise, I think it's just as safe, uh, to use whole blood as it is to use component therapy.

And in fact, maybe even safer since you're just talking about one donor. So I think those are all things to consider. You know, I think the thing about the RH, uh, status is, is an interesting one. Um, there are, uh, more and more studies coming out that say that it probably does not make much of a difference with the low titer group of whole blood that we're giving.

And there's actually some interesting data out there as well that was done in, in women of childbearing age, um, and when asked if they, would consider getting whole blood as part of their resuscitation if they were bleeding to death, knowing that there was a very, very, very small chance of some kind of issue with a fetus in the future.

Um, and the overwhelming majority said that they would consent to receiving whole blood. So I think that's an

interesting thought and an interesting finding. That further supports that, you know, once, once we decide, yes, this is what we should be doing as a standard practice. It shouldn't shy us away from giving it to women of childbearing age as well.

And actually most large trauma centers are already doing that. I know we are all right. Great work team. That was an excellent discussion of whole blood resuscitation. And I think we've finally gotten to that point where it's time for some quick hits. Number one, there is significant heterogeneity in study design across whole blood resuscitation studies, complicating comparison.

Number two, there is likely a mortality benefit to whole blood resuscitation and trauma. However, this is likely dependent on the population. And number three, future research directions should focus on prospective randomized work to try and better quantify the exact benefit of whole blood and determine in which populations this benefit is actually realized.

Thanks for listening and remember dominate the day.