Doc, I Think I'm A Quart Low
October 8, 2006 12:15 PM   Subscribe

When someone needs a blood transfusion, how do they know how much blood to put in?

So my partner's sister has been hospitalized several times in the last few months as part of her leukemia treatment. From time to time she needs blood transfusions, platelets, chemo, and so on.

On the most recent occasion, he was on the phone with her and they were getting ready to give her a transfusion so he had to hang up. I asked him how much they were going to give her, and he didn't know. So we started speculating: Is there a test? A standard rule of thumb? < humor> A dipstick of some sort? < /humor> Do they just keep going until they think they're done...?

She is a former nurse, and the next day we asked her this question, and she didn't know the answer either. So I Googled it and came up with nothing. So now I'm passing it on to you. Anybody know?
posted by Robert Angelo to Health & Fitness (13 answers total) 2 users marked this as a favorite
And while we're at it why doesn't pumping stuff in cause a rise in blood pressure?
posted by phrontist at 12:22 PM on October 8, 2006

Not sure, but I think it's a reactive thing - that is, if your blood pressure is too low (or other signs that you don't have enough of X type blood - perhaps certain types of anemia or whatever) then they transfuse you. If there's still a problem, then transfuse again.

In the case of something like leukemia, my guess would be that she has blood, so her BP isn't the problem, but the red blood cells are damaged? In which case, perhaps there's just a standard replacement dosage.
posted by spaceman_spiff at 12:23 PM on October 8, 2006

Wouldn't a low pressure valve on the IV just mean that your vein would suck in just enough blood to keep your body at a certain blood pressure (or range)?

I mean, I know in horrible accidents where someone is injured and has lost blood they just drop it in by the unit, but as far as I understand IV's are set to drip either by rate or pressure. But I'm not a doctor, so I could be way off base.

I know from giving blood a lot (and reading the information/sales pitch) that blood products are divided at least four different ways, plasma, platelets, etc. Maybe (as the previous poster said) for leukemia she's not getting whole blood, but just the red blood cells she's lost due to the treatment.
posted by tiamat at 12:33 PM on October 8, 2006

I asked my mom who is an RN:

Every unit or two of blood (she cant remember the exact amount) is known to raise the hematocrit by so much. So you start by looking at the base hematocrit, and then transfuse units based on what their normal hematocit value (which is different for men and women) should be. If the person is asymptomatic, and their hematocrit is in the 20s, the doctor will often start with 2 units of blood.

Blood pressure is not raised if the patient is given whole blood (as there is an equalibrium between solids and liquids.), which apparently doesnt happen very often any more. But most often patients are given packed red blood cells (PRBCs), and this (being only a solid, and thus not keeping equilibrium between liquid and solid) *does* raise blood pressure as well as temperature. PRBCs are preferred because there's less danger of fluid volume overload. She said that at her hopsital they get around that to some degree by also giving the patient a diuretic, which eliminates some of the bodies fluid.
posted by supercrayon at 1:08 PM on October 8, 2006

IANAD, or a medical guy, but I've hung around hospitals a lot more than I'd ever have liked in the last 5 years. Typical blood volume per unit weight in humans is a fairly standard value, but the human body adjusts quickly to accidental rapid blood loss by several mechanisms, including the shock response, which is not, unfortunately, always optimally adapted for survival. In situations of external blood loss, volume estimation from observed blood at the injury scene can be a useful guide.
"unit [3] (of blood)
a unit of volume for human blood and various blood components or products. A unit of whole blood is 450 milliliters, which is about 0.9510 U.S. pint. For components of blood, one unit is the amount of that substance that would normally be found in one unit of whole blood. The adult human body contains roughly 12 units of whole blood."
In cases of longer term blood loss, due to internal bleeding, or failure to replace blood volume components (as in leukemia), additional blood products may be added, by estimation of blood volume from body weight ratio, and information obtained from diagnostic blood counts taken both before and after the procedure. Since the volume of blood products added in such cases, is typically a small percentage of total blood volume, the circulatory system absorbs the addition readily, due to the elastic nature of the system, and the kidneys relatively large ability to return the body quickly to fluid stasis.
posted by paulsc at 1:18 PM on October 8, 2006

The standard teaching is that each unit of PRBCs (packed red blood cells) will raise your hemoglobin 1 point or your hematocrit 3-4 points. You get a CBC which tells you the patient's hemoglobin/hematocrit, and then decide how many points you want to bump them (depends on the patient and other medical conditions/risk factors). Doesn't always have to be up to a "normal" level.

Pumping "stuff" (saline, blood products, etc.) increases the person's "volume status" (they have more fluid in their blood vessels and body) and will raise the BP if it's very low, but otherwise it's pretty tightly controlled by the heart and nervous system.
posted by gramcracker at 2:16 PM on October 8, 2006

There is a condition called "hypovolemia" in which one's blood pressure is low simply because there isn't enough blood in one's veins. If I remember right, giving blood to someone who's hypovolemic will help raise their blood pressure. But that's a special case.
posted by nebulawindphone at 2:53 PM on October 8, 2006

Thank you! This helped us understand it much better. In her case, she has received both whole blood and platelets recently, as needed.
posted by Robert Angelo at 3:23 PM on October 8, 2006

Re: hypovolemia, blood isn't used as a fluid volume replacer. A patient who's only mildly hypovolemic (ie. someone whose only sign is a low pressure and who hasn't been bleeding) wouldn't get a blood transfusion; the donated blood is too valuable and the risks are too great to warrant it. Also, transfusing them would actually raise their hematocrit to abnormal levels because their hypovolemia doesn't have anything to do with a loss of red blood cells. Instead, they'll get lactated Ringer's or normal saline to expand their fluid volume and raise their pressure.

A massive blood loss from trauma or hemorrhage, OTOH, where fluid volume deficit is only part of the problem, requires both blood transfusions and regular fluid replacement. People are frequently transfused intraoperatively to counteract blood loss from surgery, and trauma victims often need blood to counteract the blood loss from their injuries.

Most otherwise-healthy people can lose a decent amount of blood and tolerate it pretty well. During a c-section, it's not uncommon to lose a liter of blood, and we rarely have to transfuse women afterwards.
posted by jesourie at 7:49 PM on October 8, 2006

Blood products are occasionally used as a volume replacer in the setting of severe acute trauma, but in general it's far too costly and dangerous to use it that way. Every unit of typed, crossmatched blood carries about a 1 in 250 risk of a minor transfusion reaction, a 1 in 5000 risk of a life threatening reaction, and variable risks of transmitting infectious disease. So it's not given casually or lightly.

Whole blood is never given in the United States - blood is always depleted of its white cells first, because these tend to attack the person they're transfused into. Such a blood product is referred to as leukocyte-depleted or leukocyte-poor.

Before transfusing blood, often a "complete blood count" will be done. A C.B.C. measures white blood cells, red blood cells and platelets per unit volume. Red blood cells are often expressed as hematocrit, which is the percentage of total blood volume that is occuped by red blood cells. Normal is 39-45% or thereabouts, depending on if you're a man or woman, young or old, sick or well, etc.

On average, a single unit of packed red cells, about 450cc total volume, raises the hematocrit about 3 points, after all the fluid shifts and equilibrations occur.

Fresh frozen plasma, which is blood minus all its cellular parts, is often given to folks who have clotting problems. Platelets are, unsurprisingly, given to folks who are low on platelets and at risk of bleeding. FFP comes in units and platelets come in 12-packs; various lab tests are used to quantify the risks of bleeding and the amount of blood product to use is decided accordingly.
posted by ikkyu2 at 11:14 PM on October 8, 2006

Most of the answers here are right on the money; some that I'd like to highlight:
  • Truly, giving whole blood is incredibly rare, for the reasons ikkyu2 highlights.
  • In a setting as described by the OP, you'd need a measurement of the hemoglobin and/or hematocrit -- part of the CBC -- in order to decide how much blood to give. Then, you'd use the rules as others have described them (one unit raises the hemoglobin roughly 1 g/dl).
  • The "1 unit = 1 g/dl of hemoglobin" rule only applies to adults, since a unit of packed red blood cells is a standard size -- 450 cc or so -- no matter what the age of the recipient. So 450 cc of PRBCs going into an average adult will raise the hemoglobin a predictable amount, but going into a smaller-sized child will obviously raise it much more. So in peds, we use the general rule of giving around 10 cc per kilogram of the child's weight; that raises the the hemoglobin somewhere in the 1 to 2 g/dl range, in general.
  • Blood is used as a volume repletion agent a lot more often than people here acknowledge. In the trauma setting, it's actually critical (since if a patient's lost 50%+ of his or her blood volume, the loss of the actual blood cells is as much a part of the problem is the loss of volume); in the intensive care unit setting (especially when talking about oncology patients), blood has a moderately important role as a volume expander.

posted by delfuego at 7:30 AM on October 9, 2006

Just out of curiosity, what would happen in a situation where someone's normal hematocrit is quite low? I have a genetic anemia, and my hematocrit is generally around 23-25. One one hand, I can imagine doctors wanting to raise this, albeit temporarily. On the other hand, why waste blood products?
posted by nekton at 2:06 PM on October 10, 2006

Nekton, a physician should only try to return any one person to his or her own norm, not the generic physiologic normal. A person'ss body becomes very, very accustomed to the hematocrit that results from whatever genetic melange affects it; people with sickle cell anemia function fine at hematocrits that would make "normal" people faint, and that sort of thing. There's nothing beneficial to take someone whose norm is 23-25 and raise it to 35-36; in fact, it can be very harmful, for a slew of reasons (stroke risk, sequestration, etc.).
posted by delfuego at 8:02 PM on October 10, 2006

« Older Helping the elephants   |   Justified anger, or plain paranoia? Newer »
This thread is closed to new comments.