Jump to content

Leaderboard

Popular Content

Showing content with the highest reputation on 04/23/2008 in all areas

  1. JCruz

    Age of blood

    Although publication in the NEJM gives this article the veneer of credibility, there are a number of problems with this study. First, it is a retrospective study, so it does not have the power that a prospective randomized controlled trial would have (despite the authors' attempt to argue that their retrospective data was "prospectively" by clinicians carrying out patient care. Although the authors argue that the populations were matched, they simply were not. Here are the differences between the populations that by the authors own work are statistically significant: Statistically Significant Differences in the New Blood vs. Old Blood Populations Blood group (# of units/total # of units - %)53.1% of new blood units were type O, only 31.1% of old blood were type O56% of old blood were type A, 37.9% new blood type ABlood group (# of patients/total # of patients - %)50.9% new blood patients type O, only 30.4 percent old blood type O49.4 % old blood type A, 34.7% new blood type AHigher rate of abnormal left ventricular function in the old blood group (63.1 vs 57.9%Higher rate of mitral regurgitation for old blood (67.3 vs. 64.1%)Peripheral vascular disease higher in old blood patients (58.5 vs. 54.4%)Leukocyte reduction: significantly higher number of old blood patients receiving both leukoreduced and nonleukoreduced products (11.4 vs. 3.9)Larger body surface area in older blood patients #'s 3, 4 and 5 indicate a higher risk population than the "new blood" folks. The authors talk about ventilator time, but have made no adjustment for pre-existing lung function - these are elderly people - want to bet how many of those folks with peripheral vascular disease were/are smokers with less than optimal lung function? Take a look at Figure 1A. This is where "demonstrate" the two populations are the same with regard to # of units transfused. The scale on the Y axis is 0-30. Now look at figure 3 - the Kaplan Meier curve they use to show their "significant difference" between old and new blood folks. Looks like a big difference, right? Until you check out the scale, where the entire graph is a 15% spread (from 85-100%). So they've magnified a really small difference to make it seem big. Oh, and if you pull that same little trick on figure 1a, it shows the same "huge difference" in the two populations as they are claiming for fig 3. NEJM did a big disservice in printing this article in this form. The most this article does is support the need for a randomized controlled prospective trial. We are all aware the storage lesion exists, but none of us know its true impact on patient outcome in cardiac patients or any other. That said, it unfortunately was well publicized on a slow news day. "Old Blood Can Kill You" in our local rag - it almost made me bleed into my brain. Although some of you are struggling with your surgeons and clinicians, my greatest concern is what we will be hearing from patients and their families every time we transfuse them and they question how old the blood is. And should there be a bad outcome for any reason, well - welcome to litigation land, because surely it was the "bad old blood" that killed the 85 year-old 3 pack a day smoker with CHF, diabetes and metastatic cancer, right?
    1 point
  2. I would like to respond since I am a senior tech at the IRL who did the ID on the patient in question. Of course, we follow the AABB IRL guidelines when identifying a specificity. The patient in question also has anti-E, -S, -K and -Fya in addition to the –Cw, -V and –Dia mentioned previously. This patient has made all of the commonly encountered allo antibodies that we would expect with his phenotype. He has a history of GI bleeding and has had multiple episodes at various local hospitals dating back to 2005. We often fill his requirements from our Ro phenotyped inventory due to his antibody history. As can be assumed, most of our Ro donors are either African American or Hispanic and therefore it is not unlikely that the patient has been exposed to the V and Dia antigens. Since the patient does not require C negative units, exposure to the Cw antigen is also not unexpected. As we test this patient in the future, we will likely identify other antibodies to low incidence antigens as these happen to be on selected cell panels. I would imagine that other reference lab staff would agree with me that we do not identify these specificities as a whim. Once identified, we are faced with having to decide how to handle them every time we get new requests for red cells. Julie
    1 point
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.