When Should VTE Prophylaxis Be Initiated in Patients With Thrombocytopenia?

Among critically ill patients with thrombocytopenia, use of venous thromboembolism (VTE) prophylaxis decreased in patients with a platelet count less than 100 x 10⁹/L, and the risk of bleeding associated with VTE prophylaxis was the highest at a platelet count of less than 50 x 10⁹/L, according to a research letter published in the Annals of the American Thoracic Society.

Little evidence exists to guide clinicians in decisions regarding VTE prophylaxis in patients with thrombocytopenia. Researchers therefore sought to quantify the risks of bleeding and VTE associated with early VTE prophylaxis in critically ill patients with thrombocytopenia in a range of platelet counts.

The study enrolled adults admitted to an intensive care unit (ICU) on hospital day 1 who had thrombocytopenia (initial platelet count ≤150 x 10⁹/L). The participants were a subset of the Premier Healthcare Database (from 2016 to 2020).

The exposures of interest were early VTE prophylaxis on hospital day 1 and initial platelet count. The outcomes were red blood cell (RBC) transfusion on days 2 or 3 of hospital admission and a diagnosis of new VTE that was not present at admission.

A total of 137,579 patients were admitted to an ICU with a platelet count on day 1; of those, 40,009 were included in the analysis: 12,048 (30.1%) of these patients (median age, 62 years; 42% female) received VTE prophylaxis at hospital admission, and 27,961 had no VTE prophylaxis (median age, 61 years; 41% female).

The adjusted probability for early VTE prophylaxis initiation increased as the platelet count increased from 0 to 100 x 10⁹/L. At a level higher than 100 x 10⁹/L, the probability of early VTE prophylaxis was stable.

A significant interaction between platelet count and early VTE prophylaxis (P <.001) was found in the model for RBC transfusion, with the effect estimate direction indicating that the RBC transfusion risk was greater when VTE prophylaxis was initiated at a lower platelet count. In the model with natural cubic splines, the interaction between early VTE prophylaxis initiation, platelet count, and transfusion was nonlinear, with a threshold effect less than a count of 50 x10⁹/L.

No interaction was found between early VTE prophylaxis initiation and platelet count for new VTE (P =.83), including the sensitivity analysis, which used a more sensitive definition of VTE (P =.88). The risk of new VTE in patients with thrombocytopenia was low for all platelet counts.

Study limitations include the low sensitivity of the algorithm used to identify VTE and a lack of clarity in determining whether VTE prophylaxis doses decreased at lower platelet counts. In addition, minor bleeding that did not require transfusion and was undetectable could have affected clinicians’ decisions to use VTE prophylaxis.

“These results can help inform the decision to initiate VTE prophylaxis in critically ill patients with thrombocytopenia,” the study authors commented.