Thromboelastography MightBe More Applicable to Guide Anticoagulant Therapy than Fibrinolytic Therapy in Critically Ill Patients with COVID-19

We read with great interest the recent report by Wright and colleagues¹ “Fibrinolysis Shutdown Correlation with Thromboembolic Events in Severe COVID-19 Infection.” With the aim of developing a screening tool to predict which SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2)-positive patients are at risk of thromboembolic events, these authors used coagulation laboratory values and thromboelastography (TEG) results to develop a receiver operating characteristic curve to predict micro- and macrothrombotic events in 44 critically ill patients with COVID-19. Wright and colleagues conclude that “fibrinolysis shutdown” (d-dimer levels > 2,600 ng/mL and 0% clot lysis at 30 minutes on TEG) is predictive of thromboembolic events and need for hemodialysis. Their results also suggest that renal failure (a microthrombotic event) precedes macrothrombotic events in critically ill patients with COVID-19.

To date, at least 3 other published studies and unpublished data from the American Society of Hematology have reported the results of viscoelastic hemostasis assays (TEG and rotational thromboelastometry) are consistent with hypercoagulability in critically ill patients with COVID-19.2, 3, 4, 5 We found that 48% of 21 critically ill patients with COVID-19 had a significantly elevated maximum amplitude on TEG, and that this metric had a 100% sensitivity and 100% negative predictive value for thrombotic events.² Based on these results, we implemented a TEG-based anticoagulation protocol for critically ill patients with COVID-19 who are 18 years or older with no contraindications to anticoagulation. Briefly, if a patient has an elevated maximum amplitude on TEG (>67 mm), a heparin drip is initiated at 5 U/kg/h with a partial thromboplastin time goal of 40 to 60 seconds. In our cohort of 21 critically ill SARS-CoV-2-positive patients, 46 thrombotic events developed in 13 patients (62%; range 1 to 8 events per patient) before initiation of our anticoagulation protocol²; strikingly, only 1 thrombotic event occurred in this cohort of 21 patients after initiation of this protocol (data not published).

Although the discussion highlights the impact of hypercoagulability and anticoagulation in critically ill patients with COVID-19, more data are required to describe the significance of “fibrinolysis shutdown” reported by Wright and colleagues.¹ We also observed this phenomenon in our cohort of critically ill patients with COVID-19. Specifically, in patients with 2 or more compared with 0 to 1 thrombotic events, clot lysis at 30 minutes on TEG was 0.5% vs 1.3%. However, the mean d-dimer level was 6,800 ng/mL vs 2,900 ng/mL in these groups, respectively, suggesting that clot lysis was not impaired in patients with higher thrombosis rates.² Given these data, we agree with Wright and colleagues that investigating whether thrombolysis is occurring at sites of fibrin clots (producing elevated d-dimer), but not systemically (resulting in low clot lysis on TEG) will likely be needed to guide the use of fibrinolytic therapy in these patients.

Finally, although Wright and colleagues¹ use their data to predict renal failure, future studies using this model to predict other thromboembolic complications (stroke, acute limb ischemia, or right heart failure—all of which have also been associated with this pandemic) will be needed to support incorporation of this model into our clinical care guidelines.⁶