Disseminated Intravascular Coagulation: A Predictor for Arterial and Venous Thrombosis in Patients with Acute Myeloid Leukemia

About 10 percent of adult patients with acute myeloid leukemia (AML) treated with intensive chemotherapy developed venous or arterial thrombosis, according to results from a prospective study published in Blood. Study authors, led by Eduard J. Libourel, MD, from the Department of Hematology/Oncology at the Sint Franciscus Gasthuis in Rotterdam, the Netherlands, also found that disseminated intravascular coagulation (DIC) at AML diagnosis could identify patients likely to develop thrombosis.

“Thrombosis is a well-known complication in patients with malignancy; however, it is not well known how often this occurs in AML patients,” co-author Frank W. G. Leebeek, MD, PhD, from the Erasmus University Medical Centre in Rotterdam, the Netherlands, told ASH Clinical News.

Although DIC is associated with venous thromboembolism (VTE) and bleeding risk in certain hematologic malignancies, including acute promyelocytic leukemia (APL) and acute lymphocytic leukemia, this relationship has not been demonstrated in other types of AML.

Dr. Leebeek and authors tested their hypothesis that DIC contributed to thrombosis risk in AML patients in a large, observational study of 272 younger adult patients (mean age = 47 years; range = 18-65 years) with newly diagnosed AML or higher-risk myelodysplasia (MDS). Patients with APL were excluded from the study. Results from this cohort were validated in a cohort of 132 older patients with AML (mean age = 68 years; range = 61-77 years).

DIC parameters (including fibrinogen levels, D-dimer levels, platelet count, prothrombin time [PT], anti-thrombin levels, and alpha-2 anti-plasmin levels) were measured at AML or MDS diagnosis. No patients received anticoagulant prophylaxis and imaging studies were performed to confirm the diagnosis of thrombotic events.

The prevalence of thrombosis was 8.7 percent in the younger cohort: 4.7 percent venous thrombosis and 4 percent arterial thrombosis.

Among the AML patients:

  • 2.9% had pulmonary embolism (PE)
  • 1.4% had venous thrombosis of the leg
  • 0.4% had thrombosis of the upper extremity

During a median follow-up of 478 days (range = 0-109 months), 18 patients developed a symptomatic venous thrombosis (n=9) or arterial thrombotic event (ATE; n=9).

One patient had a thrombotic event four days prior to starting chemotherapy, but most thrombotic events occurred before the start of the second course of chemotherapy, with 12 patients (67%) developing thrombosis a median of eight days after initiating chemotherapy (range = 2-60 days). The remaining five patients (28%) developed thrombosis after the start of the second course of chemotherapy (median = 32 days; range = 6-47 days).

Rates of thrombosis were higher in the validation cohort of older patients, with a prevalence of 10.4 percent (4.4% venous thrombosis and 5.9% arterial thrombosis.

Among the AML patients:

  • 1.5% had PE
  • 2.2% had venous thrombosis of the leg
  • 0.7% had thrombosis of the upper extremity

“Hematologists should be aware that thrombosis – both venous and arterial – occur frequently in patients with AML,” Dr. Leebeek told ASH Clinical News. “If patients present with symptoms suggestive of deep vein thrombosis or PE, adequate diagnostics should be performed.”

DIC scores (per the International Society on Thrombosis and Hemostasis DIC scores) were available for 99 percent of patients in the younger cohort (n=270) and 95 percent in the older cohort (n=126).

DIC incidence (defined as an ISTH DIC score ≥5) was comparable between the younger and older groups: 8.5 percent and 6.3 percent. “There was no apparent difference regarding AML cytogenetic features for patients with and without DIC (p=0.25),” the authors reported. However, in patients with DIC, white blood cell count (p=0.005 in the younger and 0.04 in the older groups) and blast count in bone marrow (p<0.001 and 0.009) were significantly higher than in those without DIC.

“[Among younger patients] the ISTH scoring system for DIC was a strong predictor for the occurrence of thrombosis, with a hazard ratio (HR) of 4.79 (95% CI 1.71-13.45) for a DIC score ≥5,” the authors observed. This association was confirmed in the validation cohort, where the HR for any thrombotic event was 11.08 (95% CI 3.23-38.06) for patients with a DIC score ≥5.

Eight percent of younger patients (n=22) developed major bleeding, though this did not occur more often in those with a thrombotic event and was not associated with any of the DIC parameters at diagnosis. The same was true in the validation cohort: 18 patients developed major bleeding, but none were associated with a thrombotic event or DIC parameters.

As seen in the TABLE, of the individual DIC parameters, only D-dimer level was significantly associated with thrombosis in both cohorts. “We found that a [highly] increased D-dimer level (>4 mg/L) is the best predictor for a thrombotic event,” the authors wrote. Mean D-dimer levels, they added, were similar between both cohorts: 1.2 mg/L (95% CI 0-18.5) in younger patients and 1.2 mg/L (95% CI 0.1-21.8) in older patients.

“So far, no standardized tests exist for the determination of DIC in patients with acute leukemia. Two out of four parameters of the most commonly used ISTH DIC score (low platelet counts and reduced fibrinogen levels) seem of little value in AML patients,” Dr. Libourel and colleagues explained.

“The high risk of thrombosis may indicate that AML patients may benefit from thromboprophylaxis,” Dr. Leebeek said. “There is no consensus on the use of thromboprophylaxis in these patients and is therefore not routinely given to AML patients, because of their risk for thrombocytopenia and bleeding. Therefore, it is important to identify which patients have the highest risk of thrombosis.”

This study, he added, “can be a start for future studies to investigate whether the subset of AML patients with high D-dimer levels could benefit from thromboprophylaxis.”

The authors noted, however, that acquired and inherited thrombophilic factors other than DIC could have contributed to the increased thrombotic risk among AML patients. The use of anticoagulant therapy in 22 study participants during follow-up also introduced a possible bias.

Limitations of the study include its observational design and lack of a formal comparator arm.


Libourel EJ, Klerk CPW, van Norden Y. Disseminated intravascular coagulation at diagnosis is a strong predictor for both arterial and venous thrombosis in newly diagnosed acute myeloid leukemia. Blood. 2016 June 29. [Epub ahead of print]

TABLE. DIC Parameters Associated with Thrombosis Risk
Younger Cohort Validation Cohort
Hazard ratios (95% CI) p Value Hazard ratios (95% CI) p Value
Prothrombin time (versus <16.3 seconds)
≤19.3 seconds 3.03 (0.86-10.64) 0.020 4.39 (1.13-16.98 0.092
>19.3 seconds 11.71 (2.63-52.17) 5.10 (0.62-41.63)
D-dimer level (versus ≤0.5 mg/L)
0.5-4.0 mg/L 2.79 (0.84-9.28) <0.001 0.73 (0.16-3.24) 0.01
>4.0 mg/L 12.03 (3.39-42.64) 7.82 (1.95-31.38)
Antithrombin (versus >0.8 IU/ml)
≤0.8 IU/ml 3.45 (1.27-9.32) 0.025 0.74 (0.16-3.43) 0.7
Alpha-2 anti-plasmin (versus >0.8 IU/ml)
≤0.8 IU/ml 4.52 (1.32-15.44) p=0.029 N/A N/A