COVID-19 is a thromboinflammatory disease [1]. Anticoagulant prophylaxis is currently widely used to attenuate thrombotic complications, and aspirin has also been recently proposed as a potential therapeutic option [2, 3]. However, there is limited evidence of anticoagulants’ pharmacodynamic efficacy, and no study of the pharmacodynamic efficacy of aspirin is available. No published experience with bedside assays to assess anticoagulant or antiplatelet response in COVID-19 is available.

Here, we report a sub-analysis of the evaluation of hemostasis in hospitalized COVID-19 patients (TARGET-COVID) study (URL: https://www.clinicaltrials.gov; Unique identifier: NCT04493307). The study was performed in accordance with standard ethical principles and approved by the local institutional review board. All patients provided written consent. We enrolled hospitalized patients diagnosed with COVID-19 by reverse transcription-polymerase chain reaction assay (n = 120) and patients without COVID-19 but with pneumonia and an elevated D-dimer (n = 13). Using a bedside thromboelastography assay (TEG-6s, Haemonetics Corporation, Braintree, MA, USA), we measured the reaction time (R) as an indicator of anticoagulant response and platelet function with the platelet mapping assay at the time of hospital administration [4, 5]. Aspirin effect was further determined by urinary 11-dehydro-thromboxane B2 (u11-dh TxB2) with enzyme-linked immune assay (Inflammatory Markers Laboratory, Wichita, KS, USA), total thrombus formation analysis system (T-TAS) with platelet chip (Fujimori Kogyo Co, Tokyo, Japan) and whole blood aggregometry (Chronolog Corporation, Havertown, PA, USA).

The majority of our COVID-19 positive patients were African Americans (67%). The difference in R (ΔR) ≥ 1 min between the kaolin and kaolin plus heparinase channels in the TEG6s indicated an anticoagulant effect [5]. Aspirin response was assessed in patients admitted on chronic aspirin therapy (≥ 14 days) and anticoagulant therapy was assessed ≥ 24 h after administration.

Compared to patients on enoxaparin prophylaxis (subcutaneous enoxaparin < 80 mg BID) (n = 50), patients on heparin prophylaxis (subcutaneous unfractionated heparin ≤ 7500 units TID) (n = 21) and therapeutic anticoagulation (intravenous unfractionated heparin or subcutaneous enoxaparin ≥ 80 mg BID) (n = 17) exhibited a significantly greater ΔR (0.31 ± 0.8 versus 1.2 ± 1.7 and 1.5 ± 1.4 min, respectively, p < 0.004 for both comparisons). There was a higher incidence of poor anticoagulant response (ΔR < 1 min) with enoxaparin prophylaxis compared to heparin prophylaxis and intravenous heparin (84% versus 62% and 53%, respectively, p < 0.05 for both comparisons) (Fig. 1a).

Fig. 1
figure 1

a Anticoagulant response indicated by the change in reaction time by thromboelastography in patients with COVID-19. R reaction time, CK-R reaction time with kaolin channel, CKH-R reaction time in kaolin plus heparinase channel. *p < 0.004 as compared to enoxaparin prophylaxis. +p < 0.05 as compared to enoxaparin prophylaxis. Urinary 11-dehydro thromboxane B2 in COVID-19 positive and negative patients with and without Aspirin. TI thromboinflammation, TR therapeutic response [6]

In total, 29% of patients were on aspirin therapy. With TEG-6s, 94% of COVID-19 patients demonstrated > 50% platelet aggregation induced by 1 mmol/L arachidonic acid (AA) (aspirin nonresponsiveness) (data are not shown). In patients on aspirin, u11-dh TxB2 was similar in COVID-19 positive and negative patients (3760 ± 2295 versus 3051 ± 1488 pg/mg creatinine). However, the number of non-COVID-19 patients were small to draw a definitive conclusion. In COVID-19 patients, u11-dh TxB2 was lower in aspirin-treated patients than patients not on aspirin (3760 ± 2295 versus 13,125 ± 11,474 pg/mg creatinine, p = 0.003). An inadequate therapeutic aspirin response was observed in 91% of COVID-19 patients on 81 mg daily aspirin and 50% of patients on ≥ 162 mg daily aspirin (Fig. 1b). The frequency of thromboinflammation was highest in COVID-19 patients not on aspirin (81%) and lower on 81 mg daily (55%) and lowest on ≥ 162 mg daily aspirin (25%). Compared to COVID-19 patients not on aspirin, patients on 81 mg daily aspirin exhibited a trend for lower whole blood aggregation in response to 5 µg/mL collagen (7.6 ± 4.1% versus 9.4 ± 3.3%, p = 0.08) and 50 umol/L epinephrine (5.0 ± 4.3% versus 6.7 ± 4.1%, p = 0.1). With T-TAS platelet chip, COVID-19 patients on 81 mg daily aspirin exhibited lower area under the curve compared to patients not on aspirin (226 ± 174 versus 305 ± 135, p = 0.01). No significant correlation between arachidonic acid- or collagen-induced platelet aggregation and u11-dh TxB2 was observed.

This is the first report of a point-of-care care assay demonstrating an inadequate pharmacodynamic response to anticoagulants and aspirin in a high percentage of COVID-19 patients, most of whom were African Americans. An insufficient pharmacodynamic effect of 81 mg daily aspirin indicated by high levels of u11-dh TxB2 and area under the curve with platelet chip with T-TAS suggests that low dose aspirin is insufficient to provide a meaningful clinical effect in the presence of elevated systemic inflammation (cytokine storm) and hypercoagulability. Furthermore, most of the COVID-19 positive patients on anticoagulant therapy, particularly enoxaparin prophylaxis, may not achieve the desired pharmacodynamic effect attributed to high fibrinogen levels. Our results should be considered hypothesis-generating for personalizing antithrombotic therapy to improve outcomes in COVID-19. Extrapolation of our findings to other races and ethnicities will require further study. These findings are relevant to ongoing studies of antithrombotic therapy in COVID-19.