Anti-Fibrinolytics (lysine analogues) for the Prevention of Bleeding in Patients with Hematological Disorders: A Systematic Review

Patients with hematological disorders are frequently at risk of severe or life-threatening bleeding as a result of thrombocytopenia. This is despite the routine use of prophylactic platelet transfusions (PlTx) to prevent bleeding once the platelet count falls below a certain threshold. PlTx are not without risk and adverse events may be life threatening. A possible adjunct to prophylactic PlTxs is the use of anti-fibrinolytics, specifically the lysine analogues tranexamic acid (TXA) and epsilon aminocaproic acid (EACA). The objective of this review was to establish the current evidence for the safety and efficacy of these agents in thrombocytopenic patients with hematological disorders who would routinely receive PlTx.

The protocol was pre-specified and published in the Cochrane Database of Systematic Reviews (CD009733). We searched (in full) MEDLINE (1948–2011); EMBASE (1980–2011); CENTRAL (The Cochrane Library Issue 4, 2011); CINAHL (1982–2011); LILACS; Transfusion Evidence Library: WHO ICTRP; ISRCTN ClinicalTrials.gov; EU Clinical Trials Register; and 4 other electronic databases up to 31^st October 2011 as well as additional records from hand-searching articles. There were no restrictions on language or publication period. Eligible studies were randomized-controlled trials (RCTs) involving patients (of all ages) with a hematological disorder who were severely thrombocytopenic due to bone marrow failure, who required PlTxs, and who received TXA or EACA (any dose, via any route). Trials of patients with immune thrombocytopenia were excluded. Two authors extracted data independently. Study and participant characteristics, details of the intervention and comparator, and key outcomes were recorded. Primary outcomes were bleeding and thromboembolism. Secondary outcomes included mortality, laboratory measures of fibrinolysis, number of platelet (plt) and red blood cell (RBC) transfusions (Tx), and adverse events of anti-fibrinolytic agents or transfusions. Risk of study bias was assessed using the Cochrane Collaboration criteria.

Of 446 initially identified, 415 articles were excluded on the basis of the title and abstract. Thirty-one full text articles were reviewed from which, 4 studies reported in 5 articles were eligible for inclusion. One TXA study (8 patients (pts)) was excluded from the qualitative analysis due to poor study design. Three studies (2 TXA (12 to 56pts), 1 EACA (18pts)) reported in 4 articles (published 1983 to 1995) were included in the qualitative analysis. No quantitative analysis was performed due to differences in the way outcomes were reported and the paucity of data available.

All studies reported bleeding, but it was reported in different ways. All 3 studies suggested anti-fibrinolytics reduced the risk of bleeding. The first study showed a difference in average bleeding score of 42 in placebo (P) vs. 3 (TXA). The second study only showed a difference in bleeding episodes during consolidation chemotherapy, mean 2.6 episodes/pt (SD 2.2) (P) vs. mean 1.1 episodes/pt (SD 1.4) (TXA). The final study reported bleeding on 50% of days at risk (P) vs. bleeding on 31% of days at risk (EACA). Two studies reported thromboembolism and no events occurred. All 3 studies reported a reduction in PlTx usage. The first study reported a difference of 222 plt units (P) vs. 69 plt units (TXA). The second study only showed a difference in total plt usage during consolidation chemotherapy, mean 9.3 units (SD 3.3) (P) vs 3.7 (SD 4.1) (TXA). The final study reported 1 PlTx every 10.5 days at risk (P) vs. 1 PlTx every 13.3 days at risk (EACA). One of the 2 studies that reported RBCTx usage found a reduction in use. None of the studies reported overall mortality. One study reported death due to bleeding, and none occurred. Only 1 study reported adverse events of TXA and none occurred.

Our results indicate that the evidence base for the use of anti-fibrinolytics in this patient group is poor. TXA and EACA may be useful adjuncts to PlTx in order to minimize their use and any associated complications because all of the studies showed the same direction of effect. They appear to be well tolerated although the data are sparse. Larger RCTs are required to evaluate the use of anti-fibrinolytics before they can be widely adopted in clinical hematology practice.

No relevant conflicts of interest to declare.