Hemoglobin, blood transfusions, and tanking pt's

[winterallsummer]

Hello,

 

I would post this in the specialty form but wasn't sure where to post it - surgery, internal, EM - since it applies to several fields.  Actually it would be nice to hear from a few specialists and generalists opinions.  

 

Patient presents with trauma and an active bleed.  Stabilized, etc in surgery and then is on the floor.  Patient's hemoglobin drops from 10 post-op to 8, 7, 6.  Gets the "industry standard", 2 units of blood.  H&H are taken later and it is around 8.5, next day 8.3 or 8.4.

 

Patient's BP is 130/70 and very slightly tachy (100-105).  Otherwise, no problems.  Hospital "management" pushes for discharge.  No fever, cleared by everyone else, etc.  What is the best option here?

 

Is he safe to d/c?  Should he have gotten more than 2 units, tanked up so to speak, to begin with?  What if he was at risk for the side effects of viscosity (eg hx of CAD or CVA), how much would the strategy change?  What if it was PUD in a non-compliant patient instead of trauma?

 

What are your general guidelines for blood?  Do you wait for 7 or if you see a falling trend, will you transfuse before you get there?  

 

Do you ever give more than 2 units, excluding massive blood loss (eg the pt who obviously needs 4-6+ units)?  If so, how much lasix and fluids, if any, between doses?

 

Thanks as always.

[Acebecker]

I don't recall the specific study, but there is good evidence that conservative blood resuscitation yields better results than aggressive reuscitation. IIRC Hb of 8-9 produced less mortality than a goal of 10-11 or so. In your case, do need to make sure it's stable or 12-24 hours prior to DC.

[Guest JMPA]

you need to discern active bleeding from dilution and/or poor phlebotomy technique. i would not discharge a patient unless stable q 24 hour as noted above. there are instances where blood is drawn down stream from an active IV.

[rcdavis]

The answer is: depends.

 

Depends on what kind of trauma (high speed unrestrained driver or falling from 30 feet? GSw?, assault?

Head? Chest? Abd? Pelvic.

 

Depends on surgery.. Fractured spleen capsule packing

or splenectomy? Fractured liver? Esophageal varicella bleed? Gastric ulcer ? Perforation?

 

Femoral fx?

 

Retro peritoneal involvement?

 

Depends on fluid resuscitation... How much crystalloid was given before blood?

 

Is hgb true or dilutional?

 

2 units prbc ain't anything in true trauma. And 2 units at hgb of 8 is a little aggressive unless here was evidence on o2 delivery compromise. As stated above, most trauma guys now lean towards letting hgb fall down to 6-7 mg% before replacement ( excluding cops and myocardio insufficiency pts). We seem to get better quicker with anemia than when being tanked up.

 

Any data as to what patient's usual hgb is? Eg, does he live at 8-9, but came in dry?

 

How old is patient? Old speaks to stays in hospital, hound favors d/c if all else equal.

 

In your assessment of the patient, to what do you attribute the anemia? Fluid shift or active blood loss?

 

If active blood loss, the patient needs to stay ... To check coags, post trauma syndromes, 3rd compartment spacing, etc.

 

If " physiologic" , and young patient, then d/c with close24 hr follow up.

 

The other questions, involving diuretics, iron overload, calcium, fibrinogen levels, are Germaine in much higher transfusion levels than 2 units.

 

davis

[coloradopa]

Hello,

 

I would post this in the specialty form but wasn't sure where to post it - surgery, internal, EM - since it applies to several fields.  Actually it would be nice to hear from a few specialists and generalists opinions.  

 

Patient presents with trauma and an active bleed.  Stabilized, etc in surgery and then is on the floor.  Patient's hemoglobin drops from 10 post-op to 8, 7, 6.  Gets the "industry standard", 2 units of blood.  H&H are taken later and it is around 8.5, next day 8.3 or 8.4.

 

Patient's BP is 130/70 and very slightly tachy (100-105).  Otherwise, no problems.  Hospital "management" pushes for discharge.  No fever, cleared by everyone else, etc.  What is the best option here?

 

Is he safe to d/c?  Should he have gotten more than 2 units, tanked up so to speak, to begin with?  What if he was at risk for the side effects of viscosity (eg hx of CAD or CVA), how much would the strategy change?  What if it was PUD in a non-compliant patient instead of trauma?

 

What are your general guidelines for blood?  Do you wait for 7 or if you see a falling trend, will you transfuse before you get there?  

 

Do you ever give more than 2 units, excluding massive blood loss (eg the pt who obviously needs 4-6+ units)?  If so, how much lasix and fluids, if any, between doses?

 

Thanks as always.

We wouldn't have transfused the patient in the first place absent bleeding. If you look at critical care there's good evidence that you aren't helping and probably hurting patients by transfusing patients that don't need it. Increased infection and increased mortality as well as longer ICU stays are just part of the picture. The TRICC trial is generally what is referred to here. It looked at a threshold of 7 vs 10 for transfusion with increased mortality in the liberal (10) group. 

http://www.nejm.org/doi/full/10.1056/NEJM199902113400601

 

In reality because of the way the study was run it was really 8 vs 10. In our ICU, in a non bleeding patient we will transfuse below 7 and between 7 and 8 on request of the surgical services. If the patient is actively bleeding then we will push the hemoglobin up to 9 or more (since it takes time to equilibrate). 

 

Here is a pretty good article from Annals that looks at some of the issues:

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3207872/

[UGoLong]

Transfusing the two units when they did makes sense to me. I would have waited to d/c the patient to see if the hemoglobin holds for a day to be sure there is no further unexplained blood loss. I don't see anyone get a transfusion over Hgb of 7 or so without symptoms of uncompensated shock. A heart rate in the low 100s isn't really a concern.

 

The other question is where they got d/c-ed to. If it was to an LTAC, they can follow his or her hemoglobin and deal with it there. 

[Guest Paula]

Whatever you do don't send the patient back to their PCP if they live in a rural area and the Hgb is not stable.  I hate getting patients like this in follow up.  We try to send them back when they continue to worsen and then the ER/Hospital refuses to re-admit because it hasn't been three days.  

 

Stupid Medicare rules compromise patient care from my perspective out here in the boonies. 

[Acebecker]

So Paula - where would you send that patient to follow-up? Not being flip or a wise-guy - I'm on the receiving end of those hospital DCs, too and I send them back to the ER. But I'm always trying to figure out why. Where is the right place for them? If still bleeding after a 2 day med-Surg stay, they likely need a procedure of some kind dep on where the bleeding is. If the Hb loss is dilutional, perhaps they need another unit or two - can't we use outpatient hospital departments to give them their blood? Or perhaps their Hb loss is hemolytic - autoimmune or a valve issue. Thoughts?

[delco714]

In my hospital we have a dedicated heme team with pretty strict guidelines. In no setting of cardiac problems or shock, under 6 will get you 1-2 packs. Before that you will get iv iron (venofer). For your patient I'm assuming there was a post op blood loss mixed with hemodilution. I would never dc a pt <24hr after xfuse. Must show stable h/h over 24hrs at the minimum. Then they will be dc with cbc the next day or two. I also agree you must try your best to find their baseline hh. Check for dz or conditions causing chronic anemia. They could of walked in iron deficient. Long story short, medicine is an art, each pt unique. But no, I would of fought a dc until hemodyn stable >24hr and pulse closer to nl.

[GetMeOuttaThisMess]

winterallsummer said that this was a trauma patient with an active bleed prior to going to surgery.  The first Hb mentioned is post-op at 10 gm/dL, followed by subsequent decreases which then stabilize, which I take as being consistent with the true Hb being measured following the acute bleed and time itself adjusting the actual intravascular hemoglobin concentration/blood products (it isn't stated how long the time interval is between onset of bleeding and the obtaining of the Hb of 10 mg/dL.).  Winter, do you know the pre-op Hb obtained in the ED for reference purposes?  I bet it was > 10 gm/dL.  As an aside, I used to have a cardiac pt. that we knew when his Hb. dropped below 7 gm/dL because he'd come into the office with angina.  Put him in the hospital, fill the tank, clean the windshield and off he'd go.  Who needs a lab?

[Guest Paula]

So Paula - where would you send that patient to follow-up? Not being flip or a wise-guy - I'm on the receiving end of those hospital DCs, too and I send them back to the ER. But I'm always trying to figure out why. Where is the right place for them? If still bleeding after a 2 day med-Surg stay, they likely need a procedure of some kind dep on where the bleeding is. If the Hb loss is dilutional, perhaps they need another unit or two - can't we use outpatient hospital departments to give them their blood? Or perhaps their Hb loss is hemolytic - autoimmune or a valve issue. Thoughts?

There is no other place to send for follow-up for our patients.  We are it.  But it just gripes me when patients are d/c'd too early.  I would appreciate a phone call from the hospital or PA/NP/MD/DO first and discuss the case with us so we are prepared.  Often times the patient just shows up and we had no idea they had been in the hospital.  D/C notes are not always sent to us.  So when a patient comes in on a Friday at the end of the day and we scramble to figure it all out, the hgb is low (maybe not quite low enough for another unit, but is 2 down from discharge and on the edge, what do we do?  Our patients tend to not understand why they were in the hospital so getting a history is a challenge.  

 

I suppose I could give the patient an order to travel to the hospital lab on Sat for repeat H & H but then there is no way for us to manage the result.  Lab won't give results to the patient.  I'm not on call.  Neither is the doc.  WAIT!   I just had a brilliant idea..  I will put the doc's cell phone number on the order and to have the lab call if there is a critical value and he can take care of it.  That is what he is paid the big bucks for.    Problem solved. Why didn't I think of this before?  

 

It would make sense for the hospital outpatient department (or ER) to coordinate with the PCP to provide the appropriate care.

[SocialMedicine]

it would be great if ERs had a better way to communicate pt information . Obtaining that information from many of the ERs in New York City is an absolute joke.

[winterallsummer]

To follow up... Pt was 30 y/o no PMH and had GSW with femur fx. Dc"d home, cleared by pt etc. no past history. Sent to follow up at a local FM practice. This was not my pt. thx all for the discussion. Will read the articles when I get a minute.

 

Interested in getting iv iron without blood. Can you elaborate?

[True Anomaly]

it would be great if ERs had a better way to communicate pt information . Obtaining that information from many of the ERs in New York City is an absolute joke.

 

As an aside to the discussion in regards to this, it is my fervent hope that with EMR's becoming more ubiquitous across the country that at some point some genius IT guy will find a way for the different systems to talk to each other.  I would absolutely love it if just the medical centers in a single large metro area all shared the same data.  Or hell, even just share the same imaging studies on the same system. 

[whoRyou]

As an aside to the discussion in regards to this, it is my fervent hope that with EMR's becoming more ubiquitous across the country that at some point some genius IT guy will find a way for the different systems to talk to each other.  I would absolutely love it if just the medical centers in a single large metro area all shared the same data.  Or hell, even just share the same imaging studies on the same system. 

Yep, that would make life better in the fast lane        pardon the pun LOL

[medic25]

As an aside to the discussion in regards to this, it is my fervent hope that with EMR's becoming more ubiquitous across the country that at some point some genius IT guy will find a way for the different systems to talk to each other. I would absolutely love it if just the medical centers in a single large metro area all shared the same data. Or hell, even just share the same imaging studies on the same system.

We have this to some degree since transitioning to EPIC. Our whole health system is now on the same EMR, including outpatient clinics and PCP offices. It's great being able to read the primary's notes, outpatient labs, etc. on most of our ED patients.

[delco714]

The state of Maine has a resource that collects most the pts medical data. Pretty great. And we can share images from hospital to hospital easily. Additionally my hospital is the flagship of a system that makes up 2/3 the state. We're all on the same records.

[True Anomaly]

We have this to some degree since transitioning to EPIC. Our whole health system is now on the same EMR, including outpatient clinics and PCP offices. It's great being able to read the primary's notes, outpatient labs, etc. on most of our ED patients.

 

Ours is getting that way too- I still wish inpatient consults were done on the computer so I don't have to decipher hieroglyphics.  But I know that you would LOVE to be able to look up patients who bounce from one ER system to the next looking for "second opinions" on their chronic abdominal/back/chest pain.

[medic25]

But I know that you would LOVE to be able to look up patients who bounce from one ER system to the next looking for "second opinions" on their chronic abdominal/back/chest pain.

That's the advantage to a small city; we took over the only other hospital in town a couple of years ago, so unless they are traveling around the state we can see everything. We also have a good state narcotic database which helps to screen for shoppers.