Introduction: Balancing the safety and efficacy of anticoagulation (AC) for venous thromboembolism (VTE) among patients with thrombocytopenia (TCP) is challenging as there are limited data on outcomes. As a general practice, patients with VTE and platelet count (PC) > 50,000, a full dose of AC is considered safe. In 2018 International Society of Thrombosis and Hemostasis (ISTH) Standardization Committee updated guidance statement on the management of AC in Cancer-associated Thrombosis and TCP. ISTH recommended full dose low molecular weight heparin (LMWH)/unfractionated heparin with transfusion support or dose-reduced LMWH based on risk of thrombus progression for patients with PC between 25,000 and 50,000. Despite increasing clinical use of DOAC (direct oral anticoagulants) in patients with VTE, including patients with TCP, the safety of DOAC in TCP is unclear, and additional data is needed to support clinical decision making.

Methods: In this retrospective cohort study conducted at Emory Healthcare from 2017 to 2022, patients ≥18 years with the following ICD-10 diagnosis codes: I82(other venous embolism and thrombosis), I26(Pulmonary embolism), I81(portal vein thrombosis), and PC <100,000 were identified from electronic health records(EHR).The primary outcome of interest is a composite of either major bleeding (MB), clinically-relevant non-major bleeding (CRNMB), minor bleeding, recurrent VTE, or VTE progression. MB and CRNMB were defined by ISTH criteria. Patients were followed from their index date (defined as the first simultaneous occurrence of VTE, PC <100,000, and AC use either with DOAC or LMWH) until their PC improved to at least 100,000 for 7 or more days; they stopped AC; they experienced one of the composite outcome events, or until the study end (6 months after index date). This study was approved by the Institutional Review Board. Descriptive statistics were calculated for each demographic and clinical covariate by treatment group. Systematic differences between treatment groups were examined with Pearson-chi square test, Fisher's exact test, two-sample t-tests and Kruskal-Wallis tests.

Results: Three hundred EHR were reviewed, 69 patients met inclusion criteria, and 231 were excluded. DOAC was prescribed to 47 patients (68.1%) and LMWH to 22 patients (31.9%). The mean age was 60 years (SD=12.8). There were 31 males (46.3%) and 36 females (53.7%), 36 participants were White (53.7%), 26 Black (38.8%), and 4 Asian (6.0%). Though the number of comorbidities was slightly higher among patients on DOAC (mean=2.2, SD=1.1) compared to LMWH (mean=1.3, SD=0.6), none of the differences met the statistical significance threshold. LMWH was used more commonly among cancer patients (86.4% vs 63.8%) and DOAC among hypertensive patients (46.8% vs. 22.7%). Among patients with an adverse outcome, no significant differences were observed in the prevalence of either comorbidity occurrence or by treatment type.

A composite outcome was observed in 17 (25%) patients. While the proportion of events was higher among patients receiving DOAC compared to LMWH (14, 29.8% vs. 3, 13.6%), the difference was not statistically significant (p 0.154). Composite outcomes in the DOAC cohort were 8 MB (57.1%), 3 CRNMB (21.4%), 1 minor bleeding (7.1%), 1 VTE progression and 1 multiple events (MB and VTE). The only reported outcome in LMWH cohort was CRNMB (3, 100%). In the DOAC cohort, 11 patients received Apixaban and 3 received Rivaroxaban. Median PC on the day of bleeding in the DOAC cohort was 80,000. Median PC in the LMWH cohort on the outcome days couldn't be accurately determined because of missing data, but PC was <100,000. There was no anti-platelet use in any of the patients with outcomes, except one patient with both MB and VTE. There was no difference in bleeding outcomes based on DOAC doses.

Conclusion: Based on our preliminary data from a single-institutional cohort study, composite outcomes were higher among patients on DOAC compared to LMWH (29.8% vs. 13.6%), but the difference did not meet statistical significance. LMWH cohort notably only had 3 CRNMB without any MB or new VTE. Despite increasing clinical use of DOAC, our study provides a signal for a better safety profile for LMWH in patients with TCP. We will continue to collect data to increase the sample size and evaluate whether any subgroup of patients with TCP and VTE will have a safety profile at lower doses of DOAC comparable to LMWH.

This content is only available as a PDF.
2025
Sign in via your Institution