Incidence of Venous Thromboembolism in Patients with Acute Leukemia.

Introduction: Although the association between solid tumors and venous thromboembolism (VTE: lower extremity deep venous thrombosis and pulmonary embolism) is well known, less is known about the risk of thrombosis and its clinical impact in acute leukemia. Recent reports suggest that VTE in hematological malignancies is not rare.

Materials and Methods: The California Cancer Registry was linked to both the California Patient Discharge Data Set and the master death registry to determine the incidence of VTE (exclusive of upper extremity DVT) in patients with acute myelogenous leukemia (AML) and acute lymphocytic leukemia (ALL) over a 6-year period and overall survival. Cox proportional hazards modeling was performed to analyze the effect of potential risk factors on the incidence of VTE and death.

Results: There were 5394 cases with AML (45.4% women, mean age 60.4 ± 21 years). The cumulative incidence and person-time rate of VTE are shown in the table. Significant predictors of VTE were female sex (hazard ratio (HR)=1.4, CI: 1.0–1.9), 2 chronic co-morbidities (HR= 1.8, CI: 1.2–2.6), the presence of a catheter (HR=1.6, CI: 1.1–2.3), and advancing age (HR ≥4 for each decade starting at 16 years), but not FAB category. Diagnosis of VTE was not associated with reduced survival. Significant predictors of death within 2 years were an increasing number of chronic co-morbidities (HR ≥1.3 for increasing co-morbidities), and advancing age (referent age < 16) (HR ≥1.6 for each decade). FAB M3 was associated with improved survival. There were 2482 cases of ALL (42.7% women, mean age 25.3 ± 25 years). The cumulative incidence and person-time incidence rate of VTE are shown in the table. Three or more co-morbidities (HR=2.5, CI: 1.2–5.0), increasing age (HR ≥5.4 for each decade), and the presence of a catheter (HR=2.3, CI: 1.3–4.3) were risks factors for VTE, but not FAB classification. Predictors of death in patients with ALL were ≥3 co-morbidities (HR=1.5, CI: 1.2–1.9), increasing age (HR ≥3.6 for each decade), and FAB L3, but not VTE.

Conclusions: VTE in patients with AML/ALL was not rare. The two-year cumulative incidence of VTE in patients with either AML or ALL was about 3.6%, with over half the cases diagnosed in the first 3 months. This incidence and time course are similar to the findings reported among patients with solid tumors. Presence of chronic medical co-morbidities was associated with development of VTE, as was the presence of a catheter. The higher risk of VTE in females with AML is unexplained. In contrast to patients with solid tumors, VTE was not a predictor for shortened survival in acute leukemia patients. Thromboprophylaxis should be considered in patients hospitalized with acute leukemia.