Objective:

Describe the incidence of unintentional therapeutic anticoagulation and associated patient-specific factors in neurocritical care patients receiving pharmacologic venous thromboembolism (VTE) prophylaxis.

Background:

Neurocritically ill patients have high risk for VTE and central nervous system (CNS) bleeding complications. Optimal pharmacologic VTE prophylaxis regimens remain undetermined.

Design/Methods:

This retrospective cohort study included adult patients admitted to one neurocritical care unit from January 1, 2023 to December 31, 2024, with peak, steady state anti-factor Xa (aFXa) levels drawn while receiving subcutaneous unfractionated heparin (UFH) or enoxaparin for VTE prophylaxis. Primary outcome was incidence of therapeutic anticoagulation, defined as aFXa >0.3 IU/mL and >0.6 IU/mL for UFH and enoxaparin, respectively. Secondary outcomes included aFXa level, dose adjustments based on aFXa, and factors associated with therapeutic aFXa using univariate and multivariable analyses. Safety outcomes included VTE and CNS bleeding.

Results:

155 patients with 203 aFXa levels were included. Mean age, weight, and BMI were 65 years, 83 kg, and 29.1, respectively. 54% were female. Primary diagnoses included intracerebral hemorrhage (26%), acute ischemic stroke (21%), subarachnoid hemorrhage (17%), and traumatic brain injury (14%). 183 UFH and 20 enoxaparin aFXa levels were obtained. Therapeutic anticoagulation occurred in eight patients (5.2%), seven receiving UFH and one enoxaparin. 54% of UFH aFXa were <0.1. Of remaining UFH levels, median aFXa was 0.14 (IQR 0.11-0.19). Median enoxaparin aFXa was 0.24 (IQR 0.19-0.37). 44 dose adjustments occurred utilizing aFXa (21.7%). On univariate analysis, female gender and number of doses before aFXa were associated with therapeutic aFXa, while weight and admission creatinine clearance trended towards significance. Number of doses before aFXa remained significant in the multivariable analysis (OR 1.15, CI 1.03-1.27, p=0.012). VTE occurred in 16 patients (10%) and CNS bleeding in 7 (5%). 

Conclusions:

Unintentional therapeutic anticoagulation can occur with pharmacologic VTE prophylaxis in neurocritically ill patients. Monitoring aFXa might be useful in this population.