A locally optimised machine learning approach to early prognostication of long-term neurological outcomes after out-of-hospital cardiac arrest.

Background: Out-of-hospital cardiac arrest (OHCA) represents a major burden for society and health care, with an average incidence in adults of 67 to 170 cases per 100,000 person-years in Europe and in-hospital survival rates of less than 10%. Patients and practitioners would benefit from a prognostication tool for long-term good neurological outcomes. Objective: We aim to develop a machine learning (ML) pipeline on a local database to classify patients according to their neurological outcomes and identify prognostic features. Methods: We collected clinical and biological data consecutively from 595 patients who presented OHCA and were routed to a single regional cardiac arrest centre in the south of France. We applied recursive feature elimination and ML analyses to identify the main features associated with a good neurological outcome, defined as a Cerebral Performance Category score less than or equal to 2 at six months post-OHCA. Results: We identified 12 variables 24 h after admission, capable of predicting a six-month good neurological outcome. The best model (extreme gradient boosting) achieved an AUC of 0.96 and an accuracy of 0.92 in the test cohort. Conclusion: We demonstrated that it is possible to build accurate, locally optimised prediction and prognostication scores using datasets of limited size and breadth. We proposed and shared a generic machine-learning pipeline which allows external teams to replicate the approach locally.

Platelet activation and partial desensitization are associated with viral xenophagy in patients with severe COVID-19.

Mild thrombocytopenia, changes in platelet gene expression, enhanced platelet functionality, and presence of platelet-rich thrombi in the lung have been associated with thromboinflammatory complications of patients with COVID-19. However, whether severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) gets internalized by platelets and directly alters their behavior and function in infected patients remains elusive. Here, we investigated platelet parameters and the presence of viral material in platelets from a prospective cohort of 29 patients with severe COVID-19 admitted to an intensive care unit. A combination of specific assays, tandem mass spectrometry, and flow cytometry indicated high levels of protein and lipid platelet activation markers in the plasma from patients with severe COVID-19 associated with an increase of proinflammatory cytokines and leukocyte-platelets interactions. Platelets were partly desensitized, as shown by a significant reduction of αIIbß3 activation and granule secretion in response to stimulation and a decrease of surface GPVI, whereas plasma from patients with severe COVID-19 potentiated washed healthy platelet aggregation response. Transmission electron microscopy indicated the presence of SARS-CoV-2 particles in a significant fraction of platelets as confirmed by immunogold labeling and immunofluorescence imaging of Spike and nucleocapsid proteins. Compared with platelets from healthy donors or patients with bacterial sepsis, platelets from patients with severe COVID-19 exhibited enlarged intracellular vesicles and autophagolysosomes. They had large LC3-positive structures and increased levels of LC3II with a co-localization of LC3 and Spike, suggesting that platelets can digest SARS-CoV-2 material by xenophagy in critically ill patients. Altogether, these data show that during severe COVID-19, platelets get activated, become partly desensitized, and develop a selective autophagy response.

Platelet Versus Megakaryocyte: Who Is the Real Bandleader of Thromboinflammation in Sepsis?

Platelets are mainly known for their key role in hemostasis and thrombosis. However, studies over the last two decades have shown their strong implication in mechanisms associated with inflammation, thrombosis, and the immune system in various neoplastic, inflammatory, autoimmune, and infectious diseases. During sepsis, platelets amplify the recruitment and activation of innate immune cells at the site of infection and contribute to the elimination of pathogens. In certain conditions, these mechanisms can lead to thromboinflammation resulting in severe organ dysfunction. Here, we discuss the interactions of platelets with leukocytes, neutrophil extracellular traps (NETs), and endothelial cells during sepsis. The intrinsic properties of platelets that generate an inflammatory signal through the NOD-like receptor family, pyrin domain-containing 3 (NLRP3) inflammasome are discussed. As an example of immunothrombosis, the implication of platelets in vaccine-induced immune thrombotic thrombocytopenia is documented. Finally, we discuss the role of megakaryocytes (MKs) in thromboinflammation and their adaptive responses.

Evolution of platelet activation parameters during septic shock in intensive care unit.

During severe sepsis, platelet activation may induce disseminate microvascular thrombosis, which play a key role in critical organ failure. Crucially, most of the studies in this field have explored platelet-leukocyte interactions in animal models, or explored platelets under the spectrum of thrombocytopenia or disseminated intravascular coagulation and have not taken into account the complex interplay that might exist between platelets and leukocytes during human septic shock nor the kinetics of platelet activation. Here, we assessed platelet activation parameters at the admission of patients with sepsis to the intensive care unit (ICU) and 48 hours later. Twenty-two patients were enrolled in the study, thirteen (59.1%) of whom were thrombocytopenic. The control group was composed of twelve infection-free patients admitted during the study period. The activation parameters studied included platelet-leukocyte interactions, assessed by flow cytometry in whole blood, as well as membrane surface and soluble platelet activation markers measured by flow cytometry and dedicated ELISA kits. We also investigated platelet aggregation and secretion responses of patients with sepsis following stimulation, compared to controls. At admission, the level of circulating monocyte-platelet and neutrophil-platelet heterotypic aggregates was significantly higher in sepsis patients compared to controls and returned to a level comparable to controls or even below 48 hours later. Basal levels of CD62P and CD63 platelet membrane exposure at admission and 48 hours later were low and similar to controls. In contrast, plasma level of soluble GPVI and soluble CD40 ligand was significantly increased in septic patients, at the two times of analysis, reflecting previous platelet activation. Platelet aggregation and secretion responses induced by specific agonists were significantly decreased in septic conditions, particularly 48 hours after admission. Hence, we have observed for the first time that critically ill septic patients compared to controls have both an early and durable platelet activation while their circulating platelets are less responsive to different agonists.

Thromboembolic events and Covid-19.

The novel Corona virus infection (Covid-19) first identified in China in December 2019 has rapidly progressed in pandemic leading to significant mortality and unprecedented challenge for healthcare systems. Although the clinical spectrum of Covid-19 is variable, acute respiratory failure and systemic coagulopathy are common in severe Covid-19 patients. Lung is an important target of the SARS-CoV-2 virus causing eventually acute respiratory distress syndrome associated to a thromboinflammatory state. The cytokinic storm, thromboinflammation and pulmonary tropism are the bedrock of tissue lesions responsible for acute respiratory failure and for prolonged infection that may lead to multiple organ failure and death. The thrombogenicity of this infectious disease is illustrated by the high frequency of thromboembolic events observed even in Covid-19 patients treated with anticoagulation. Increased D-Dimers, a biomarker reflecting activation of hemostasis and fibrinolysis, and low platelet count (thrombocytopenia) are associated with higher mortality in Covid-19 patients. In this review, we will summarize our current knowledge on the thromboembolic manifestations, the disturbed hemostatic parameters, and the thromboinflammatory conditions associated to Covid-19 and we will discuss the modalities of anticoagulant treatment or other potential antithrombotic options.