Retrospective review of hemoglobin and/or hematocrit levels with occurrence of thrombosis in cancer patients treated with erythropoiesis stimulating agents.

BACKGROUND

No data exists that directly compares hemoglobin and hematocrit levels between cancer patients with and without occurrence of thrombosis during treatment with erythropoiesis stimulating agents (ESAs).

OBJECTIVE

To determine the association of hemoglobin and hematocrit levels with the occurrence of thrombosis in cancer patients treated with ESAs.

METHODS

A retrospective case-control study approved by the Institutional Review Board was conducted on cancer patients billed for epoetin or darbepoetin between 1 July 2002 and 30 June 2007. Cases were defined as patients billed for thrombosis while controls were defined as patients not billed for thrombosis.

RESULTS

Sixteen patients had an occurrence of thrombosis (cases) and were matched to 16 patients that did not have an occurrence of thrombosis (controls) based on age, sex, and cancer type. The mean peak hemoglobin levels for cases and controls were 12.6 +/- 1.2 g/dL versus 12.6 +/- 1.4 g/dL (p = 0.9). The mean peak hematocrit levels for cases and controls were 37.3 +/- 3.8% versus 37.9 +/- 4.3% (p = 0.8). For the 16/586 (2.7%) patients with thrombosis, the mean hemoglobin and hematocrit at time of thrombosis were 9.6 +/- 1.0 g/dL and 28.9 +/- 3.1%. A significant identifiable risk factor for thrombosis between the cases and controls was history of thrombosis 31.3% versus 0% (p = 0.04).

CONCLUSION

There was no statistical difference in peak hemoglobin and hematocrit levels between patients with thrombosis and those without thrombosis. Further study is warranted to determine if these levels are true risk factors for thrombosis.