Browse the corpus

We appreciate the issues related to normal age-related D-dimer concentrations raised by Marco Zuin and colleagues. The ACTION study1 is a randomised trial designed pragmatically to answer the question of whether patients hospitalised with COVID-19 and elevated D-dimer concentration, without another indication for anticoagulation, should routinely receive therapeutic anticoagulation with the factor Xa inhibitor, rivaroxaban. We found that rivaroxaban 20 mg once daily for 30 days had no benefit and significantly increased bleeding compared with in-hospital prophylactic heparin. Early in the COVID-19 pandemic, observational data suggested that thrombotic events were high in patients with COVID-19 and even higher among patients with an elevated D-dimer concentration.2 In light of the available information at the time, coupled with the fact that the D-dimer concentration of many patients with COVID-19 was measured when they were hospitalised, we included an elevated D-dimer concentration, defined as above the assay (not age-adjusted) upper limit of normal in each site, as an inclusion criterion for the ACTION trial. D-dimer concentrations are not routinely measured to guide therapeutic anticoagulation decision making in patients without COVID-19,3 and the purpose of enrolling patients with an elevated D-dimer concentration in ACTION was to increase the trial population's risk of thrombotic events and not to establish D-dimer concentrations as a diagnostic tool to guide therapeutic anticoagulation in patients hospitalised with COVID-19.

atients without COVID-19,3 and the purpose of enrolling patients with an elevated D-dimer concentration in ACTION was to increase the trial population's risk of thrombotic events and not to establish D-dimer concentrations as a diagnostic tool to guide therapeutic anticoagulation in patients hospitalised with COVID-19. A prespecified subgroup analysis of ACTION showed that the main results were consistent, irrespective of D-dimer concentrations.1 Similar results have been shown in other randomised trials investigating anticoagulation in patients with COVID-19, in which D-dimer concentrations at presentation (elevated vs normal vs not collected) did not influence the main results.4, 5 Furthermore, when we used age-adjusted D-dimer upper limits of normal, as proposed by Zuin and colleagues, we found that most (>90%) patients over the age of 50 years still had an elevated age-adjusted D-dimer concentration at study entry. Not surprisingly, when we excluded the fewer than 10% of patients who did not have an elevated age-adjusted D-dimer concentration, our main results remained consistent with those in the overall population (win ratio 0·87 [95% CI 0·59–1·26]). Therefore, the results from the ACTION trial are relevant, robust, and provide high-quality evidence to avoid the routine use of therapeutic rivaroxaban—in the absence of another evidence-based indication for oral anticoagulation—in patients hospitalised with COVID-19, irrespective of D-dimer concentration.