… but shouldn’t we start transfusing him?

I was initially going to do something on chest pain but that will be pushed back for a quick discussion on when to transfuse patients in the ED.  While most of this information can be generalized to any patient actively bleeding I will focus on the patient with GI hemorrhage with some literature hot off the press.

Transfusing patients with blood products is not a benign therapy despite the fact that we often do not have to deal with the consequences of such in the ED.  One of the most feared complications is known as Transfusion-related Acute Lung Injury (TRALI) as it is the leading cause of transfusion-related mortality. (Crit Care Med 2005 33(4):721-6.)

A little over a decade ago Hebert et al. conducted a study in all patients admitted to the intensive care unit that after establishing euvolemia with a hemoglobin concentration of less than 9 g/dL.  Patients were randomized to a restrictive strategy (goal 7-9 g/dL) or a liberal strategy (10-12 g/dL).  While the lower morality among the restrictive group at 30 days did not reach statistical significance (19 vs 23%) the in-hospital morality did reach statistical significance (22 vs 28%).  The restrictive group also noted lower multi-organ dysfunction scores and ARDS.  Given the additional costs and risks of transfusing blood products is probably of minimal to no benefit for most patients.  The only subgroup that may benefit from a more liberal strategy is patients with evidence of active cardiac ischemia.  In short, a lower threshold for transfusion is probably better for our patients.  It will not only save on cost but save on downstream injuries. (N Engl J Med 1999 340(6):409-417)

To the news hot off the press (it’s 02JAN and this paper is listed as publication date of 03JAN – does that mean I’m actually reading into the future?)…

Villanueva et al randomized 921 patients admitted to their hospital for GI hemorrhage to a restrictive strategy (threshold 7 with target 7-9 g/dL) versus a liberal strategy (threshold 9 with target 9-11 g/dL).  Patients were excluded if they showed no evidence of on-going bleeding and were unlikely to need blood products at all or if they were actively exsanguinating.  So what did they find…
-Further bleeding occurred more often in the liberal group (16% vs 10%)
-Adverse event rates were higher in the liberal group (48 vs 40%)
-Patients with cirrhosis had significant increases portal-pressures versus no change in the restrictive group
-Overall mortality was higher in the liberal group (9 vs 5%)
-Cirrhotic patient mortality was higher in the liberal group (18 vs 11%)
-Child-Pugh Class A/B mortality was higher in the liberal group (12 vs 4%)
-Morality rates were also lower in the Child-Pugh class C group, variceal bleeds, and peptic ulcer groups however these differences did not reach statistical significance
-All adverse events, serious and minor, were higher in the liberal group

Moral of the story: Less is more.  Aim for a target transfusion goal of 7-9 g/dL. (N Engl J Med 2013 368(1):11-21)

About these ads
This entry was posted in Medical. Bookmark the permalink.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s