Extracorporeal membrane oxygenation aggravates platelet glycoprotein V shedding and δ-granule deficiency in COVID-19-associated acute respiratory distress syndrome.

BACKGROUND: Extracorporeal membrane oxygenation (ECMO) is a lifesaving therapy in patients with acute respiratory distress syndrome (ARDS). Hemostatic complications are frequently observed in patients on ECMO and limit the success of this therapy. Platelets are key mediators of hemostasis enabling activation, aggregation, and thrombus formation by coming in contact with exposed matrix proteins via their surface receptors such as glycoprotein (GP) VI or GPIb/V/IX. Recent research has elucidated a regulatory role of the GPV subunit. The cleaved soluble GPV (sGPV) ectodomain was identified to spatiotemporally control fibrin formation through complex formation with thrombin. OBJECTIVES: We aimed to decipher the impact of ECMO on platelet phenotype and function, including the role of GPV and plasmatic sGPV. METHODS: We recruited 36 patients with ARDS in the wake of COVID-19 pneumonia and performed a longitudinal comparison of platelet phenotype and function in non-ECMO (n = 23) vs ECMO (n = 13) compared with those of healthy controls. Patients were assessed at up to 3 time points (t1 = days 1-3; t2 = days 4-6; and t3 = days 7-14 after cannulation/study inclusion). RESULTS: Agonist-induced platelet activation was assessed by flow cytometry and revealed decreased GPIIb/IIIa activation and α-granule release in all ARDS patients. During ECMO treatment, agonist-induced δ-granule release continuously decreased, which was independently confirmed by electron microscopy and was associated with a prolonged in vitro bleeding time. GPV expression on the platelet surface markedly decreased in ECMO patients compared with that in non-ECMO patients. Plasma sGPV levels were increased in ECMO patients and were associated with poor outcome. CONCLUSION: Our data demonstrate an ECMO-intrinsic platelet δ-granule deficiency and hemostatic dysfunction beyond the underlying ARDS.

Uncoupling of platelet granule release and integrin activation suggests GPIIb/IIIa as a therapeutic target in COVID-19.

Thromboembolic events are frequent and life-threating complications of COVID-19 but are also observed in patients with sepsis. Disseminated thrombosis can occur despite anticoagulation, suggesting that platelets play a direct but incompletely understood role. Several studies demonstrated altered platelet function in COVID-19 with some controversial findings, while underlying disease-specific mechanisms remain ill defined. We performed a comprehensive cohort study with 111 patients, comprising 37 with COVID-19, 46 with sepsis, and 28 with infection, compared with control participants. Platelet phenotype and function were assessed under static and flow conditions, revealing unexpected disease-specific differences. From hospital admission onward, platelets in COVID-19 failed to activate the integrin glycoprotein IIb/IIa (GPIIb/IIIa) in response to multiple agonists. Dense granule release was markedly impaired due to virtually missing granules, also demonstrated by whole-mount electron microscopy. By contrast, α-granule marker CD62P exposure was only mildly affected, revealing a subpopulation of PAC-1-/CD62P+ platelets, independently confirmed by automated clustering. This uncoupling of α-granule release was not observed in patients with sepsis, despite a similar disease severity. We found overall unaltered thrombus formation in COVID-19 and sepsis samples under venous shear rates, which was dependent on the presence of tissue factor. Unexpectedly, under arterial shear rates, thrombus formation was virtually abrogated in sepsis, whereas we detected overall normal-sized and stable thrombi in blood from patients with COVID-19. These thrombi were susceptible to subthreshold levels of GPIIb/IIIa blockers, eptifibatide, or tirofiban that had only a minor effect in control participants' blood. We provide evidence that low-dose GPIIb/IIIa blockade could be a therapeutic approach in COVID-19.