ADDENDA Aug. 5, 2004

6. A Consultant Hematologist in Victoria, Australia agrees with sentiments expressed about lack of evidence and utility of automatic protocols. However, she wishes to comment on item #4 on the previous page, about the role of cryoprecipitate. She manages obstetric hemorrhage at a large hospital where the hematologist is part of the 'team' approach to management of this condition, and has a role in co-ordination of blood product support, interpretation of lab results and assisting the laboratory in work priorities during an emergency. She finds that cryoprecipitate has a definite role in some obstetric hemorrhage and that without aggressive and early replacement of fibrinogen in some clinical circumstances, things can "get away from you".

In her opinion, any protocol for massive transfusion should assess the likelihood of DIC. In obstetric hemorrhage DIC should be considered if: (i) the patient has had abruption and amniotic fluid embolism, or primary post partum hemorrhage. and (ii) clinical assessment reveals bleeding from everywhere (lines, endotracheal tube, mucous membranes, etc) versus bleeding from the uterus/surgical site only. With abruption and 'bleeding from everywhere' the Australian colleague recommends that cryo be used early. The same can apply to trauma management where cryo is more likely to be medically indicated for patients with obvious clinical coagulopathy or major head injury, severe acidosis, hypothermia etc.

ADDENDA Aug. 6, 2004

7. The transfusion medicine physician in Colorado who initiated this discussion has read colleagues' comments with interest, and has read the Canadian Journal of Anesthesia article, too. He also canvassed level I and II trauma centers in Colorado (there are only 12 in addition to his own) and found only two with "automatic" protocols. One thaws 2 units of FFP after the first 6 RBC's have been issued, and then informs the surgeon that the FFP is available, if they need it. The other just instituted a new protocol where 2 units of FFP are thawed and sent after the first 4 RBC's are issued. After that, 2 units of FFP are thawed and sent for every 4 RBC's thereafter. Neither center has an automatic protocol for issuance of cryo or platelets. Both are Level I centers. After reviewing the Colorado data and literature, his Trauma Committee decided to give some "guidelines" including thawing an initial two units of FFP, but not doing any additional products without specific orders.

ADDENDA Aug. 11, 2004

8. The director of a transfusion service at a large New York City trauma hospital (who submitted the information for posting #4) wishes to continue the discussion with the physician in Canada who submitted the information in posting #5 above. He acknowledges that the cited paper (Can J Anesth 2004;51:293-310) is an excellent review of the massive transfusion literature. However he partially disagrees with the Canadian's characterization that "a deficit in coagulation factors (specially fibrinogen) is more likely to be the primary cause of defective hemostasis", if the emphasis on fibrinogen is an attempt to justify the automatic use of cryoprecipitate. First of all, the review specifically recommends (p. 304-305) that coagulation factor deficiencies INCLUDING decreased levels of fibrinogen, be treated with FFP, and that only "if the fibrinogen concentration remains below 1 g/L despite therapy with FFP, the administration of cryoprecipitate ... may be indicated". Second, the review cites only a single paper (Anes Analg 1995;81:360-5) that averages blood loss greater than 1.42 blood volumes reduces fibrinogen below 1 g/L, while 2.0 blood volumes on average is needed to reduce other factors below 20%. This paper is based on elective surgery patients, not trauma patients (as pointed out by the author), and the conclusion is derived from an extrapolated regression line that is based on only TWO patients with blood loss over 1.5 volumes. Moreover, two papers are cited (Anesth 1988;69:839-845 and Br J Haem 1987:67:365-368) that conclude factors V and VIII are reduced below 20% before fibrinogen falls below 1 g/L.

The New York colleague would also like to ask the Consultant Hematologist from Australia (posting #6): What is the evidence that "cryoprecipitate has a definite role in some obstetrical hemorrhage" requiring "aggressive and early replacement of fibrinogen" (implying use of cryo before levels approach 1 g/L)?

ADDENDA Aug. 18, 2004

9. The consultant Hematologist in Victoria (#6 above) replies that in response to the New York colleague (posting #8) and her own earlier comments (posting #6) about use of cryoprecipitate in obstetric hemorhage, she did not intend to imply that they use cryoprecipitate before fibrinogen levels approach 1g/L (100 mg/dL). She states " My 'evidence' is purely anecdotal." In her experience, in major obstetric DIC often the first set of coagulation results shows a fibrinogen either undetectable or in the range 0.2-0.3 (200-300 mg/dL). In these patients, if they use FFP alone, they appear to take a very long time to catch up and see an improvement in clinical bleeding or a fibrinogen >1.0g/L. In these patients she reports that they use cryo early in the massive transfusion phase. Sometimes these patients have predominantly DIC and do not need major red cell transfusion (so technically not a massive transfusion form a red cell point of view, but massive in their need for coagulation support).

ADDENDA Sept. 22, 2004

10. A colleague in British Columbia suggests that others might be interested in reviewing the article by Stainsby D et al in the British Journal of Anaesthesia 85: 487-91 2000 entitled "Management of massive blood loss: a template guideline". Included in the article is a template for a massive transfusion protocol.

ADDENDA Oct. 31, 2004

11. Kay Elliott, MT (ASCP) SBB, the colleague from the Pacific Northwest who submitted ADDENDUM #3 to this discussion on July 26, 2004, has graciously agreed to share a presentation describing her institution's approach to massive transfusion. (PDF file- Note: file size approx. 6 MB)

ADDENDA Nov. 9, 2004

12. A transfusion service medical director in Virginia reports that his hospital does not have a strict protocol for massive transfusion cases, but if the transfusion service has dispensed more than 10 units of RBCs within 24 hours for one patient, they call and ask if platelet and/or plasma transfusion is contemplated. They feel strongly that if there is difficulty in controlling bleeding at this point, or if fresh microvascular bleeding is observed, that it is far preferable to transfuse platelets and plasma while laboratory results are pending, than to jeopardize the patient’s outcome by waiting for these results. In his opinion, clinical judgment must be the foremost consideration.

ADDENDA Dec. 5, 2004

13. David Orchard, M.T. (ASCP) of St. Vincent Hospital Blood Bank in Green Bay, Wisconsin (attribution used with permission) has graciously provided a document in use at his facility (MS Word file) that addresses their massive transfusion protocol.

ADDENDA Dec. 18, 2004

14. A transfusion medicine physician in Oregon reports that his institution's massive transfusion protocol is very similar to that used by Kay Elliott, as described in posting #11 above. They allow empiric use of "white" products when testing is unavailable, but prefer aggressive factor replacement based on discrete laboratory testing (platelets, PT-INR and fibrinogen). The main advantage of this approach is that by aggressively correcting abnormal coagulation parameters they believe that they are in a better position to separate "medical bleeding" from "surgical bleeding." In his experience, persistent bleeding in the face of a normal patient body temperature and acceptable coagulation studies is more often due to a condition that is amenable to surgical intervention. This is in contrast to coagulopathic patients whose blood loss is more often due to diffuse oozing with no surgically correctable lesion. He feels that this approach makes for the most efficient use of blood products and minimizes unnecessary delays in correcting the underlying pathology.

ADDENDA June 26, 2008

15. A physician in Ohio asks if any institution administers group AB plasma in the resuscitation of massively bleeding trauma patients whose ABO group is unknown, in order to follow a protocol based on a formulaic ratio of one plasma unit for each red cell unit transfused? In addition, he wants to know if due to inventory constraints the use of AB plasma is not an option, would it be reasonable to use group A plasma in the resuscitation of the patient until their true ABO group is known?

ADDENDA July 3, 2008

16. Dr. Stephen Apfelroth, Medical Director of the Blood Bank at the Jacobi Medical Center of the Albert Einstein College of Medicine (attribution used with permission), writes that he has been getting requests for more FFP earlier in trauma rescusication. He cites J. Trauma 2006;60:S91-96 as stating that Univ. Maryland Shock-Trauma Center at Baltimore is issuing 10 units FFP together with 10 units PRBC for trauma patients. He points out that "the surgeons are claiming that FFP is kept thawed in advance there, which would seem hard to manage as a selection of thawed FFP types would have be maintained (or else all type AB used)". He also cites Surgery News Dec 2007 (Vol 3 No 12), reporting that the blood bank at Vanderbilt delivers a box of 10 type O PRBC, 4 thawed type AB FFP, and two doses of apheresis platelets on demand (and can follow with additional boxes of 6 red cells, 4 FFP, 2 apheresis platelets).

He writes "Particularly problematic are the aspects that FFP is supposedly kept thawed at all times and that type AB FFP may be freely used together with type O red cells before typing is available. Does anyone else have issues or experience with requests of this nature?"

The Editors also wonder: for those providing thawed FFP, are they providing AB? For those providing group O RBCs, are they providing O neg?

ADDENDA July 16, 2008

17. Neil Blumberg MD, Professor of Pathology & Laboratory Medicine and Director of Transfusion Medicine at University of Rochester Medical Center (attribution used with permission) writes:

I have a number of concerns about the calls for immediate replacement of FFP and platelets prior to massive transfusion occurring, and in particular, with the use of ABO mismatched plasma (e.g., AB for everyone) and platelets (what type do you use--there is no "universal donor" ABO type for platelets).

1. Firstly, we should admit that the evidence for the practices that are being advocated (and for our current practices, to be fair) is far from strong. It is observational, empiric and potentially wrong. The new approach is based upon experience with the treatment of healthy young adults in a war zone, which are not the patients in which most massive transfusion protocols will be employed in a civilian hospital. The presence of multi-organ disease and older immune systems, hearts, lungs, etc. may render practices that were reasonably efficacious and safe in 25 year olds decidedly unsafe in 55 or 70 year olds. That said, some sort of protocol is clearly indicated for patients receiving greater than 1 blood volume or so in a few hours. There is no evidence I am aware of that implementing FFP and platelet transfusion with the first red cell unit is better than waiting for the fifth, eighth or tenth red cell unit. There is good reason to be believe FFP and platelets are dangerous products for many patients, predisposing to thrombosis, acute lung injury and multi-organ failure at the very least.
2. The proposal to use AB FFP and who-knows-what ABO group platelets before knowing the patient's type is, in my view, a likely dangerous, possibly ineffective and unproven measure. Data from our own center suggests the mortality of cardiac surgery patients transfused with ABO non-identical platelets increases 3-4 fold and red cell utilization increases by 50% compared with patients receiving similar numbers of ABO identical platelets (Transfusion 41:790, 2001). Agneta Shanwell and colleagues have reported in abstract form that patients receiving ABO compatible but non-identical FFP (primarily AB) have 7% increased mortality overall (Vox Sanguinis 91[Supplement 3]: 15, 2006). These data should give one pause before recommending liberal use of AB plasma and non-identical platelet transfusions for patients who may well need neither urgently. Use of ABO mismatched transfusions may actually impair hemostasis rather than improve it in some patients.
3. We recently reviewed our experience with about 700+ surgical patients receiving at least one platelet transfusion. 50% of them ultimately received less than 10 units of red cells in their entire hospital stay, suggesting that most of these patients did not need platelet transfusions at all. We are a level I trauma center, our surgeons do liver and heart transplants and VAD placements and all the usual stuff any 750 bed city hospital does. Thus if we implemented immediate FFP and platelet replacement on demand, many patients would probably suffer morbidity and mortality from transfusion who do not do so currently.
4. Considering all these issues, we've come up with an alternate approach that we will implement. The main reason for giving AB FFP rather than ABO type specific is that there is not timely receipt of a blood specimen from the ER or OR. We will beat on people until the absolute first specimen that goes anywhere, goes to the blood bank so we can get an ABO type done in a few minutes. The first four units of red cells will go to the patient's bedside in a cooler as O neg (or O Pos sometimes). The next four units of red cells are type specific if possible. After 8 units of red cells in less than 2 hours, we will start 1:1:1 automatic shipments of red cells/FFP/platelets for all future transfusion orders. This will actually work out to one pool of platelets with every 4-5 red cells ordered. The attending physician can initiate this protocol earlier than the 9th red cell unit, but we will discourage this protocol's use when we don't have a sample for ABO type determination. On balance, I feel this approach has a fighting chance of providing maximal benefit to those patients who are already experiencing severe bleeding and minimize harm to patients who may not ultimately need FFP and platelets, or who would have to receive large volumes of ABO mismatched plasma. I make no claims it's a proven approach, but the same is true of giving AB plasma to everyone, and there is early data that this latter approach may well increase mortality.

Please submit comments to the e-Network Forum.

Ira A. Shulman, MD
CBBS e-Network Forum Editor & Moderator

W. Tait Stevens, MD
CBBS e-Network Forum Assistant Editor & Moderator