Balancing Bleeding and Thrombosis Risk With Anticoagulation in Patients With Hematologic Malignancies and Thrombocytopenia

Two studies presented at the 2016 ASH Annual Meeting examined the safety of anticoagulation therapy in patients with hematologic malignancies – a population with an increased risk of venous thromboembolism (VTE), despite periods of thrombocytopenia in subsets. While the first study supports the safety of anticoagulation, the second study questions the use of a “standard” approach to anticoagulation in these patients, after determining that bleeding and VTE risk varies during the disease course.

Anticoagulation Safe for Certain Patients With Thrombocytopenia

The management of thrombosis in patients receiving curative-intent chemotherapy or hematopoietic cell transplantation (HCT) is difficult, but research presented by Bethany T. Samuelson, MD, from the Department of Medicine at the University of Washington in Seattle, Washington, and authors from the Seattle Cancer Care Alliance and Fred Hutchinson Cancer Research Center, suggests that patients with low platelet counts can safely receive anticoagulation to prevent thrombosis without increasing the risk of major bleeding.1 These patients also were able to safely receive platelet transfusion to maintain a platelet count of ≥50 x 109.

Dr. Samuelson and authors performed a retrospective chart review of cases of VTE diagnosed at the three participating centers over a four-year period, identifying 79 adult patients being treated for hematologic malignancies who had experienced a period of five or more days of moderate to severe treatment-related thrombocytopenia (platelets <50 x 109) within 30 days of VTE diagnosis.

Eighty-three events occurred among the 79 patients identified; 66 patients (80%) were started on anticoagulation at the time of VTE diagnosis.

Thirty-six patients did not experience bleeding, while 23 patients experienced World Health Organization (WHO) grade 2 bleeding events, and 10 experienced WHO grade 3/4 bleeding events (TABLE). “While patients who experienced thrombocytopenia for five days or longer had higher rates of minor (grade 2) bleeding, increased rates of clinically significant (grade 3/4 bleeding) were not observed,” Dr. Samuelson noted.

The majority of anticoagulated patients (n=61; 92%) were set to receive platelet transfusions when their platelets dropped below ≥50 x 109 for the majority of anticoagulated patients, which Dr. Samuelson said was common “in the absence of evidence to guide anticoagulation and thrombosis risk management” in this setting. However, “patients experienced a number of adverse effects potentially related to transfusions,” including eight patients who experienced transfusion reactions (7 cases of hives and 1 febrile non-hemolytic transfusion reaction), and 25 who experienced volume overload that may have been associated with or worsened by platelet transfusions. Ten patients discontinued anticoagulation because the platelet count could not be maintained above the 50 x 109 threshold.

The optimal platelet transfusion threshold for patients who require anticoagulation should be the subject of future investigations, the authors concluded.

Preventing Recurrent VTE and Bleeding Events Associated With Severe Thrombocytopenia

In the second study, Damon E. Houghton, MD, from the Department of Hematology/Oncology at the University of North Carolina in Chapel Hill, and authors found that clotting and bleeding risks are not static across disease, and the anticoagulation strategy should be modified according to the degree and duration of thrombocytopenia.

Dr. Houghton and colleagues conducted a single-center retrospective chart review of 78 adult patients (median age = 53 years; range = 19-81 years) with active hematologic malignancies who were enrolled between April 2004 and December 2015. Patients had either acute or chronic VTE and were on anticoagulation before experiencing a thrombocytopenic episode (defined as a platelet count dropping below 50,000/µL) or had acute VTE during a thrombocytopenic episode.2

The median duration of severe thrombocytopenia was 14 days (range = 6-35 days). When platelets dropped below 50,000/µL, patients were initially treated with:

  • a therapeutic dose of anticoagulation with supportive platelet transfusions to maintain a count of >40,000/µL to 50,000/µL (n=43; 55%)
  • no anticoagulation (n=33; 42%)
  • a reduced or prophylactic dose of anticoagulation (n=2; 3%)

The primary outcome of this study was a composite of the duration of the thrombocytopenic episode, recurrent VTE, clinically significant bleeding, and the combined incidence of bleeding and/or recurrent VTE.

The risk of recurrent VTE and clinically significant bleeding changed over time, Dr. Houghton and authors noted, with bleeding events tending to occur early during the thrombocytopenic episode, and VTE events tending to occur later.

During the 11.5 years of follow-up, 13 patients experienced clinically significant bleeding after a median of eight days (range = 5-22 days); the majority of these bleeding events occurred on or before day 30 of thrombocytopenia onset (n=11; 85%). Two patients experienced a life-threatening bleed. Six patients experienced recurrent VTE after a median of 62 days of thrombocytopenia (range = 40-91 days). Most VTE events occurred after 40 days of thrombocytopenia onset (n=5; 83%). None of these events resulted in patient death, the researchers reported.

The composite outcome occurred in 29 percent (n=13/45) of patients in the anticoagulation group, compared with 18 percent (n=6/33) of patients whose anticoagulation was held at the onset of thrombocytopenia (incidence rate ratio [IRR] = 3.2; 95% CI 1.13-10.26; p value not reported).

Within 100 days after the onset of severe thrombocytopenia, anticoagulation was associated with a higher rate of clinically significant bleeding compared with patients who had anticoagulation held (27% [n=12/45] vs. 3% [n=1/33]; IRR=17.7 [95% CI 2.62-756.98]; p value not reported). The risk of recurrent VTE during this period, however, was non-significantly higher in patients who had their anticoagulation held (2% [n=1/45] vs. 15% [n=5/33]; IRR=0.3 (95% CI 0.01-2.64]; p value not reported).

The risks of major bleeding and composite outcome were highest in patients who received the therapeutic-dose anticoagulation with supportive platelet transfusions during episodes of severe thrombocytopenia, which “raises concerns over this treatment as the primary management strategy for VTE [in this patient population],” Dr. Houghton said.

“The fluctuating risk of bleeding and recurrent VTE over time … illustrates that anticoagulation strategies should vary in aggressiveness depending on degree and duration of thrombocytopenia,” the authors concluded, adding the the study is limited by its retrospective and single-center design. “Larger safety and efficacy trials incorporating dynamic risk-adapted anticoagulation strategies are urgently needed to determine the best treatment strategy for VTE occurring in patients with hematologic malignancies and severe thrombocytopenia.”


  1. Samuelson BT, Walter RB, Gernsheimer T, et al. A platelet count <50 x109 was not associated with increased rates of major bleeding among anticoagulated patients. Abstract #164. Presented at the 2016 ASH Annual Meeting, December 3, 2016; San Diego, CA.
  2. Houghton DE, Laux JP, Key NS, et al. Treatment of venous thromboembolism in patients with hematological malignancies and severe thrombocytopenia: A retrospective cohort analysis. Abstract #531. Presented at the 2016 ASH Annual Meeting, December 4, 2016; San Diego, CA.

TABLE. Outcomes Related to Bleeding Events1
  No Bleeding


WHO Grade 2


WHO Grade 3/4


p Value
Mean platelet count 91 81 98 0.547*



Median mean platelet count 76 60 74
Mean platelet count at time of bleed 85 159 0.23
Median platelet count at time of bleed 55 70
Mean days on anticoagulation 21.2 22.6 18.7 0.340*



Mean days <50 x 109 6.7 11.5 5 0.192*



Median days <50 x 109 5 8 4
Percentage of days on anticoagulation with platelets <50 x 109 24.8 56.5 35.9 0.975*



Percentage of events occurring at <50 x 109 40% 20%
*Grade 3/4 bleeding vs. no bleeding

**No bleeding vs. grade 2 bleeding

***Grade 2 vs. grade 3/4 bleeding

WHO = World Health Organization