A Single Institutional Experience Comparing Cryoprecipitate to Plasma in Patients with Disseminated Intravascular Coagulopathy (DIC) from Acute Myeloid Leukemia (AML)

Background: Hemorrhagic complications from disseminated intravascular coagulopathy (DIC) constitute a leading cause of morbidity and early mortality in acute leukemia, and particularly in acute promyelocytic leukemia (APL). In addition to timely institution of anti-leukemic therapy, supportive transfusions are essential in early management. Both cryoprecipitate (cryo) and plasma transfusions (plasma) are widely used in this setting to replenish consumed coagulation factors, including fibrinogen. Data comparing the safety and efficacy of cryo to plasma in this setting are lacking.

Objectives: To compare outcomes of bleeding, transfusion reactions, and volume overload amongst AML patients managed primarily with cryo to those managed with plasma from the onset of DIC.

Methods: In this retrospective study, all adult patients treated for AML at Weill Cornell Medicine from 2010-2015 that had a fibrinogen <150 mg/dL were identified by query of electronic medical records (EMR). Cases with evidence of infection, shock, and/or organ failure were excluded. Data on demographics, the number and type of transfused blood products, lab values, volume status, bleeding events and transfusion reactions were recorded for all patients from the first detection of hypofibrinogenemia until the occurrence of an outcome or a 48-hour endpoint (whichever occurred first). Outcomes included new or ongoing bleeding events, transfusion reactions, and volume overload. Logistic regression analysis was utilized to correlate variables with outcome.

Results: A total of 96 patients were identified. The median age was 66 years (17-90). Fifty-two (55%) were male and 44 (45%) were female. Seventy patients (73%) were newly diagnosed cases of AML, inclusive of 29 (30%) with APL, while the remaining 26 (27%) had relapsed and/or refractory disease. Fifty-eight patients (60%) had DIC on admission, and the remaining 38 (40%) acquired it during hospitalization. Thirty-seven patients (39%) presented with bleeding, of which 7 were major. A total of 86 units of plasma were given to 20 patients (median 3.5 units per patient, range 1-11, volume per unit ~250mL) and a total of 225 pools of cryo to 61 patients (median 3 pools per patient, range 1-11. One pool is 5 donor units with volume of ~110mL). Of those patients, 13 received both cryo and plasma (14%). The remaining 28 (29%) did not receive cryo or plasma.

By the end of follow-up, 18 patients had new or ongoing bleeding (19%) with 5 major bleeds and 1 mortality. Accounting for age, sex, baseline fibrinogen, PT/INR, Platelets, WBC and bleeding at baseline, there was no significant difference in the risk of new or ongoing bleed for patients treated with plasma alone compared to cryo alone, cryo+plasma, or neither (0, 23%, 23%, 14% respectively) or in the resolution of a pre-existing bleed (100%, 65%, 78%, and 50% respectively). New or worsening volume overload occurred in 6 patients: 2 in the plasma alone group (29%); 2 in the cryo alone group (4.2%); and 2 in the cryo+plasma (15%). There was no significant difference identified between groups. Transfusion reactions were uncommon with only 2 minor reactions reported for patients receiving plasma (2.3%, febrile non-hemolytic and allergic reactions) and 1 minor reaction for patients receiving cryo (0.4%, allergic reaction).

Conclusion: In the management of DIC with hypofibrinogenemia in AML patients, we identified no differences in clinical outcomes of patients treated with cryoprecipitate compared to plasma in terms of bleeding risk, volume overload, or transfusion reactions. Additional factors that should influence the choice of product include donor pool exposure (median 3.5 donors for plasma vs median 15 donors for cryo) and healthcare costs. Prospective studies are needed to compare the safety and efficacy of routine administration of cryo to that of plasma in this critical setting.