Tuesday, January 8, 2013

Over Transfused

A new paper from NEJM indicates that perhaps we need to re-calibrate our blood transfusion strategies in patients with severe upper gastrointestinal bleeds.  Patients were randomly assigned to liberal (transfuse for HgB under 9) vs restrictive (transfuse only for HgB under 7) transfusion treatment protocols and outcomes were measured (survival, bleeding, portal venous gradients, etc). 
The probability of survival was slightly higher with the restrictive strategy than with the liberal strategy in the subgroup of patients who had bleeding associated with a peptic ulcer (hazard ratio, 0.70; 95% CI, 0.26 to 1.25) and was significantly higher in the subgroup of patients with cirrhosis and Child–Pugh class A or B disease (hazard ratio, 0.30; 95% CI, 0.11 to 0.85), but not in those with cirrhosis and Child–Pugh class C disease (hazard ratio, 1.04; 95% CI, 0.45 to 2.37).

Historically, patients who come in with massive GI bleeds due to ulcer disease are aggressively resuscitated with blood products and saline until they achieve hemodynamic stability (and if not, it's off to the OR).  The decision to transfuse blood typically was a clinical decision, not one driven primarily by arbitrary hemoglobin counts.

The real life applicability of these findings is somewhat dubious.  In an emergency situation, when a patient is hemodynamically labile in the ER and you aren't certain of the past medical history and blood is spewing out in torrents like the hallways in the Shining, it's tough to sit around and wait for the lab to run a CBC before deciding whether to thaw some blood or not. 

From recent wartime experiences, we know from the trauma literature that outcomes are superior in those patients who are aggressively resuscitated with blood products early and often.  It just doesn't make a whole lot of sense to delay treatment of a potentially exsanguinating patient based on an arbitrary laboratory result cut-off.  Certainly, the game changes once a patient has stabilized.  There's no reason to transfuse a HgB of 8.7 for the mere fact that 8.7 is less than 9. 

I'll stick with the old standby of "clinical judgment", thank you very much.    


Anonymous said...

If you are going to draw from Trauma what about permissive hypertension?

Paracelsus said...

I wholly agree with you, although I admit I did not take the time to read the entire article.

Symptomatic anaemia is (or should be) another extremely important factor in deciding to transfuse. A patient who gets dizziness, SoB and chest pains at Hb 8 is going to get two units from me, even if they don't have any KNOWN heart history. Therefore, indeed, an Hb threshold established more or less arbitrary, might not mean much.

For the specific problem of upper GI bleed (but not only), the rate of the blood loss is, I think, just as important as any instantaneous Hb value. If I get a 9 on admission I won't be going to sleep reassured. If I get another 9 in two hours, yes, maybe.

This issue should probably looked at the same way as trauma, I agree.

This sort of study reminds me of the one comparing outcomes in appendicitis with or without surgery, with all its inherent patient selection problems, etc.. It proved, of course, that surgery is superior. This type of enterprise illustrates, in my opinion, a good way to abuse evidence-based medicine. Statistics do not absolve the scientist from starting with a decent hypothesis AND a plausible mechanism behind what is being researched.