Incidence of Bleeding Events in Acute Myeloid Leukemia (AML) Patients Receiving Anticoagulation Therapy for Treatment of Deep Vein Thrombosis (DVT) or Pulmonary Embolism (PE) Who Are Also Undergoing Induction or Re-Induction Therapy

Background/Rationale:Venous thromboembolic events (VTE) remain a prevalent complication in cancer patients. Despite this increased risk, the decision on whether or not to treat and/or use prophylactic therapy remains highly debatable among clinicians. Induction and re-induction therapy, in combination with the pathophysiology of the disease can cause profound thrombocytopenia. Additionally, leukemic cells can cause an inappropriate release of procoagulants directly into the bloodstream, initiating the clotting cascade and causing simultaneous clot formation and hemorrhage leading to disseminated intravascular coagulation (DIC). Another major factor to consider in this patient population is central venous access. Although the National Comprehensive Cancer Network (NCCN) guidelines recommend ≥3 months of therapeutic anticoagulation for catheter-associated VTEs, clinicians use anticoagulation cautiously in the AML population due to severe thrombocytopenia and the increased risk of bleeding. Data that can be used to optimize treatment strategies for catheter-associated VTE in patients with acute leukemia are needed as the risk versus benefit for anticoagulation is currently unclear. A recent article by Rajasekhar and colleagues investigated this risk and noted that among patients with severe thrombocytopenia (platelets <50,000/μL) who received therapeutic anticoagulation, 27% experienced a significant bleeding event compared to only 2% of patients who did not receive therapeutic anticoagulation. In order to establish the bleeding complications associated with patients receiving induction or re-induction therapy for AML, we conducted a retrospective chart review evaluating the incidence of bleeding in patients receiving therapeutic anticoagulation for treatment of VTE.

Objective:The primary outcome of this study is incidence of bleeding events in AML patients receiving anticoagulation therapy for treatment of DVT or PE who are also undergoing induction or re-induction therapy. Secondary outcomes include incidence of chemical and non-chemical DVT prophylaxis, as well as the difference in incidence of bleeding events among agents.

Methods:This is a retrospective chart review evaluating the incidence of bleeding in AML patients receiving anticoagulation for treatment of VTE during induction or re-induction therapy at WFBH from October 2012 to October 2016. AML patients who were >18 years of age who had undergone induction or re-induction chemotherapy and developed a VTE were included in the study. Clinical diagnosis of VTE was confirmed by objective imaging procedures. The following induction or re-induction regimens were excluded: azacitidine or decitabine. Descriptive statistics were used for data analysis.

Results: Out of 523 patients screened, 24 met eligibility criteria for a total VTE incidence of 4.6%. The age range for the patients included was 22 to 77, with 14 patients (58.3%) being female and 10 patients (41.7%) being male. Of the patients who developed a VTE, 18 (75%) received non-chemical DVT prophylaxis. Twenty (83.3%) of the 24 VTEs were catheter-associated DVTs. Twenty-two of 24 patients received therapeutic anticoagulation with either heparin (n=21) or enoxaparin (n=1). The 2 patients who did not receive anticoagulation did not have a bleeding event. Patients who developed a bleed were more likely to have supratherapeutic PTTs (defined as PTTs ≥ 90 seconds); however, supratherapeutic levels were isolated and not associated with the time of the bleed. Of the patients who received therapeutic anticoagulation, 8 (36.4%) had a major bleed and 9 (40.9%) had a minor bleed, for an overall bleeding rate of 77.3%. There were no deaths as a result of bleeding events. Of the 20 patients with catheter-associated VTEs, 7 (35%) had a major bleed; 4 of these patients required anticoagulation reversal with protamine.

Conclusions:AML patients who received induction or re-induction therapy had a low incidence of VTE overall. However, when patients received anticoagulation for VTE treatment there was a high incidence of bleeding. The results found in this retrospective chart review showed that for patients with catheter-associated VTEs, the risk of bleeding outweighed the risk of VTE. We recommend a risk-adapted strategy where patients do not receive therapeutic anticoagulation until platelet recovery.