Hey everyone, I'm Brandon Odo.
And I'm Brian Bulling.
And this is Critical Care Scenarios, the podcast where we use clinical cases, narrative storytelling,
and expert guests to impact how critical care is practiced in the real world.
Alright everyone, welcome back.
It is another turbo with Brandon Odo.
Bear with me, I have a cold as I more or less have ever since I had my kid.
However, I would like to chat about a topic today which I think is interesting, although
I'm not sure everyone agrees, but it's one of those areas that is kind of one step removed
from the core of our clinical work in the ICU.
And yet we make use of it in the interface with it all the time.
And I think those are the kinds of areas that it's really important to have as deep of an
understanding of as possible.
The number of kind of black boxes that we deal with should be as few as possible because
you really don't want to be offering care to patients or ordering things that you just
don't understand.
You don't want to click a button and say this should happen and then you wave your hands,
smoke appears, magic occurs, and then hopefully it worked.
You should understand what is specifically going to happen because then you understand
how to prepare for it when it's appropriate, how to troubleshoot it.
When all you understand is the most superficial level, you can't integrate it with what you're
doing in any other way.
So the topic I'm hinting at here is transfusion medicine, also known as blood banking.
So we give patients blood, how does that work, what's it all about?
This is pulled shamelessly from my blog, critical concepts.
So if you're more of a reader or you want a reference, I'll put those links in the show
notes, but for you joggers and commuters, let's get audio.
I want to kind of follow blood all the way from donation to when it gets to the patient
so we understand what's happening with it and the nuances.
So most of the blood that we give is donated.
Once you've donated blood in the US here, most of it is through the American Red Cross.
They do certain screening to ensure that your blood is unlikely to have transmissible diseases.
And then you go and they pull off generally one unit of whole blood.
Whole blood of course is just the stuff that's in your veins.
What is a unit?
It's around 500 CCs, half a liter, which is about a pint for you non-metric people.
It is just what you had in your blood.
However, whole blood is not generally what is stored in the bank and given to patients.
And the main reason, well, there's two reasons.
One is that by splitting it into components, you can give it to multiple patients.
And since most disease states only need one of those components, your one donation can
benefit two or three other people instead of just one.
The other practical reason is shelf life.
Placements can survive for a lot longer than whole blood.
A very large reason is because components like red cells can be refrigerated.
Components like plasma can be frozen.
Components like platelets cannot have either of those things done.
So you can't really store them that way.
So we have to kind of split them up and store them appropriately.
So they take it and they centrifuge it into components.
And the three main ones are red blood cells, which they store as packed red blood cells
and plasma and then platelets.
Now the red blood cells end up as what we call packed red blood cells, PRBCs.
And if you're wondering how much you get out, if you're a person who likes to look at hematocrits
versus hemoglobin of normal blood, of course the hematocrit is just the volume of red cells
in the blood.
So if you know a normal hematocrit is a little less than 50%, that's what you're getting
out of a unit of whole blood.
So you'll have maybe three to four hundred CCs probably of packed red blood cells.
Some of that is going to be, there is a little plasma in there, but then also some saline
in just preservatives.
Usually there's citrate added to keep it from clotting.
Citrate of course binds up calcium, which is a cofactor for coagulation.
So that keeps it liquid in the bag.
This gets refrigerated.
This is pretty much the remainder of that volume, which has all of those proteins and proteins
like clotting factors and so on in there.
This has to be frozen pretty quickly actually because a lot of those factors are labile.
They don't last very long on the shelf.
So they freeze it pretty promptly and they call it fresh frozen plasma.
And then when it actually needs to be used, it gets thawed.
They put it in a little warm water bath.
It circulates around.
Takes about 20 minutes.
And some of the most busy, usually trauma centers, they may keep some liquid plasma on
hand either thawed or actually liquid plasma that was never frozen.
So you don't have to wait for it to thaw.
And then there's platelets.
Now platelets are a little confusing because you can pull these out of the whole blood.
These are called whole blood derived platelets.
However, the blood donation process that is most common, at least here in the United
States does not make use of these.
Those whole blood derived platelets are not pulled out.
They just are actually left in usually the plasma.
So in a way you're giving them, but they're probably not active once they've been frozen
in thawed.
The platelets we actually give are usually from a feresis process.
So you donate these separately.
They go and take your blood.
They spin off platelets, return it to you.
It's a longer donation process.
So they take a larger pool of it out.
It usually ends up being the equivalent of about maybe five or six units of whole blood
derived platelets.
So these feresis or single donor platelet packs come in these kind of larger bags, whereas
a unit of red cells or of plasma is equivalent to what came out of one unit of whole blood.
So a unit is not a fixed volume.
It starts out as a unit of whole blood and then a unit of components is what came out
of one unit of whole blood.
And then the confusing part is the platelets.
I actually recommend to people not to say a unit of platelets because it becomes unclear.
Do you mean that unit equivalent of a whole blood derived platelet unit or one bag, which
is most common because most places here in the States, that's all you can give.
That's what they have.
Well, I just call it a pack of platelets usually.
And each just know is it the equivalent in your area of five units worth or six units
worth or whatever.
Now there's a belief that storing platelets at anything other than room temperature can
be harmful to them.
Temperature streams, extremes, either cooling or actually warming affects their function.
Now there's, I know some controversy on this in the blood banking world and it's not
academic because this ends up meeting that platelets can only be stored at room temperature
so they don't last very long.
And this is why you've probably run out of platelets at times, even in larger centers.
Platelets are always in shortage.
So if we could say refrigerate them, they would last a lot longer.
But this is also why if you're doing like a massive transfusion, you're not really supposed
to put platelets through the warmer.
You're just supposed to run them to gravity at room temperature.
And it's also why platelets are considered the highest risk product for transfusion related
bacterial infections because they're just sitting there incubating on the shelf.
There's one other component which is not always produced called cryoprecipitat.
Now cryo is a collection of specific proteins and factors from the plasma.
Now normally these we left in the FFP, but if you want to produce cryo, you kind of partially
thaw FFP and centrifuge it and they sort of skim out this highly concentrated sludge,
which is specific factors that have been concentrated down.
It ends up being generally fibrinogen, which is what we usually give it for, as well as
factor eight.
And then there's some von Willebrand factor, factor nine and fibrinectin, which who knows
what that is.
The remaining plasma turns into this weird product called cryo pore plasma that is not
used for a whole lot, sometimes given to TTP patients and some other specific uses, but
otherwise gets tossed.
Kind of like platelets, because that little bit of cryo that comes out of one unit is
so small, we don't generally give one unit of cryo, they pool it either at your center
or at the blood bank, and usually the pool dose is between five and ten units, but again,
be aware of kind of what a unit means for you.
So those are the components.
And now bear in mind that if I take the packed blood cells, the FFP, and even if we had the
whole blood derived platelets and I mix them back together, I've kind of reconstituted
the whole blood that was donated, but only kind of.
It's not the same.
It's been stored and that does affect it.
It affects the red cells, their pliability, their two, three DPG, and their ability to,
their affinity for oxygen, their pH is affected, but also we've added stuff, right?
We've added diluent and preservatives and anticoagulants, so we've actually diluted
all this stuff.
And what we've added can have physiologic effects.
For instance, giving a great deal of the citrate in red blood cells can bind up the patient's
calcium.
So just bear in mind that banked blood is not the same as fresh blood that's in your
patient's body, even if you do your best to try to recreate that.
Okay, let's talk about the process for matching these two patients.
First of all, again, we screen donors to try to minimize any risks on their end.
Then they do test the blood for a number of infectious things.
Really packed red blood cells, and this is fairly ubiquitous in the US here now, are
lukeo-reduced.
They filter or wash out most of the white blood cells.
You don't need these, and they probably contribute to inflammatory responses in the recipient.
Kind of a mild graft versus host disease.
A lot of the mild, febrile transfusion reactions we see are probably from white cells.
So most of the blood you're giving these days for us is already lukeo-reduced.
Another case is you can lukeo-filter it out while it's being given.
A more extreme version of this is irradiation, where you actually bombard the blood with ionizing
radiation to totally deactivate any of the lukeocytes that are there.
This is really only for specific uses, like to prevent real graft versus host disease
in patients who are really at risk, which is heavy amino suppression, immunologic malignancy.
And some chemotherapy and that sort of thing.
Now when you get to the patient that you want to give the blood to, the most important thing
is to avoid a mismatch in their ABO types.
So you know you have A and B antigens on your blood cells.
O would mean you have neither one.
AB means you have both.
We do not want to mismatch that because ABO incompatibility causes the most immediate
and severe transfusion reactions.
Now it's pretty straightforward and reliable to do AB typing as well as RH typing, the positive
or negative.
However, mistakes still get made and most of the mistakes are logistical ones, clerical
errors.
Blood just gets mixed up between patients.
So the current standards here in the states are that before you give blood, you actually
have to type them twice.
You have to do their ABO typing and then actually repeat it at least once.
Now one of them can be historical, like a prior admission if you have that in your system.
But they have to have at least two to make absolutely sure you know what blood type the
patient is.
Okay.
The second step here, so that's the type in a type and screen.
What is the screen?
The screening is looking for non-AB antibodies in the recipient's blood.
So this includes against the RA chainogen, the positive or negative.
But also there's a million other little antigens that can be found on red blood cells.
And they come in these various systems with have various components, KEL, Duffy, Lewis,
lots of names and they all have kind of letter abbreviations, capital or lowercase.
They're just proteins or carbohydrates that can appear on red blood cells and which can
stimulate an immune response.
Now you have probably some of these markers and others you don't.
If you're exposed to ones you don't have non-self antigens, then you can stimulate
an immune response, which means the next time you see an antigen, you'll have a reaction
to it.
Again, unlike A and B, these are all learned responses.
You're not going to have a reaction that first time you have to be sensitized by some
kind of exposure.
You have transfusion, although also a mother can be sensitized by fetal blood crossing
the placenta in appropriate circumstances.
That's why they give things like a roganm.
If there's maternal fetal mixing of blood to try to avoid sensitization in the mother
to anti-RH antibody formation.
Remember A and B, you already have those antibodies.
If those are non-self, who knows from where I think the theory I heard was maybe you get
exposed by bugs or something.
But by the time you are an adult, you're going to have those antibodies to whatever
antigens you don't have.
The point is we want to know if you have antibodies to any of these things.
They'll screen you.
There's a panel for all these non-ABO antibodies.
They're just going to run the recipients of blood through it.
It's an all or nothing test.
It's either positive or negative.
It's negative.
You don't have any of these antibodies.
These are the ones they consider more or less clinically significant.
There's probably a million others that are even more mild.
The ones that are here can vary in significance.
Some can cause quite a significant reaction if you had a response.
Others less so.
But you want to know, if you pop up positive, they're going to go and find out which antibodies
you do have.
That's a lengthier process of identifying the specific antibodies.
You'll see somebody pops up with a positive antibody screen.
It will eventually result saying they have anti-K or whatever.
If somebody does have an antibody, that means the blood that they can receive is going to
need to lack that antigen.
If they have anti-K, they can only get blood that does not have the K antigen.
You can see if you get a lot of transfusions, you can start to accumulate antibodies that
makes it harder and harder to find blood that doesn't have those antigens.
You might think it would make sense to just match the blood ahead of time so that whatever
antigens you don't have, the blood you get also doesn't have those.
My understanding is that's not totally practical.
That would mean it's hard to find anyone blood ever.
We go this way where we just give you blood and then if you start to develop antibodies,
it starts to narrow down the blood you can get.
That is the type in screen process.
What about cross matching?
You want to cross match unit blood so you can donate that.
This is confirming that the blood you're going to give is okay.
You've already done the screening and stuff.
They pulled out a unit from the shelf that should lack any antibodies you happen to have
or if you have no antibodies then just as long as it's AB compatible as well as the RH.
A big part of the cross matching is actually just another confirmation of ABO and compatibility.
You need to really make sure no one has screwed this up.
You can do this different ways.
The simplest is called an electronic cross match which basically means if you have a
kind of specific well validated computer system, you're just letting the computer check that
the ABO and RH type you're giving matches the patient.
Otherwise you need to actually do something in the lab.
The simplest is immediate spin cross match.
They just mix a little of the patient's blood with a little bit of the donor blood, give
it a quick centrifuge and look for any homolysis or agglutination.
It happens really quickly when there's incompatibility so it just takes minutes.
In more complicated situations such as somebody who had a positive antibody screen, they have
to do a lengthier cross match which means a more full mixing study and longer waiting.
Suffice to say the weirdest gets, the longer it's going to take you to get blood.
What about the other products?
Plasma is simpler.
You still need to have ABO compatibility but remember it is flipped.
The issue here is giving antibodies to attack the recipient's red cells not the other way
around.
For instance, remember that O negatives considered universal donor red cells because it has no
antigens on it.
For plasma, AB is universal because it has no antibodies in it.
That being said, beyond ABO and RH compatibility, this is pretty simple.
They don't generally try to match it for all of these other antigens because banked plasma
has no other antibodies.
They screen this when it's donated and if you do have, let's say, that anti-K antibody
in there because the donor had it, they don't make plasma out of that blood.
So this kind of simplifies that process.
What about platelets?
Even easier.
There are A and B antigens on platelets but they're very lightly expressed and all of
these other antigens are not present at all.
You don't even need to cross match it for A and B and you don't need to screen for other
antibodies.
Maybe in an ideal world they would do some of this.
Like maybe if they did use type specific platelets, they would have a little bit better
survival and so on.
But again, they're always in shortage and it is safe to do it this way.
That is the common standard practice.
Now there are some other antigens on platelets.
There are HLA antigens, for instance, as there are in most of our cells and there are also
some other antigens called anti-human platelet antigens.
So you can develop immunization, some of these things and patients who have proceeded
a lot of platelets do start to have a diminishing response which is probably from some of these
factors.
So platelet refractoryness, as they say, is a whole specific transfusion medicine area
of interest.
You can try AB type specific platelets if you're having issues.
You can start to try to match HLA types either kind of loosely or true matching like you
would do for bone marrow donation which usually means you have to call in the donor to give
that blood.
It's not going to be in the bank or some other things.
But this is pretty specific stuff that blood bankers can help you with if you really think
your patient is refractory and not just say consuming platelets.
All right, that might be enough for us to kind of touch on these general topics of how
the blood banking system works.
There's a lot more that can be said such as when we should transfuse thresholds and so
on.
The only other thing I want to touch on today and maybe we'll do another episode though
is that we remain in a real shortage of blood here in the States.
You'd probably call it a crisis.
Now there have always been fluctuating availability and shortages of blood but since COVID in
particular which really affected numbers of donations it has gotten only worse and of
course all the more so for some of these products like platelets but really for everything.
So please bear in mind, patients do not benefit as much or as often from transfusion as you
probably think.
The more you study some of these things the more and more refined that we can lower thresholds
and be more sparing and transfusion and by and large in general patients seem to do just
as well.
So really consider next time you're going to give a transfusion for little to no reason
just to make yourself feel better or to make a number look a little bit less red in your
system whether it's necessary because it very well may mean that another patient who truly
does need it is not going to be able to.
See you next time.
♪♪♪♪