IMPORTANCEThe National COVID Cohort Collaborative (N3C) is a centralized, harmonized, highgranularity electronic health record repository that is the largest, most representative COVID-19 cohort to date. This multicenter data set can support robust evidence-based development of predictive and diagnostic tools and inform clinical care and policy.OBJECTIVES To evaluate COVID-19 severity and risk factors over time and assess the use of machine learning to predict clinical severity. DESIGN, SETTING, AND PARTICIPANTSIn a retrospective cohort study of 1 926 526 US adults with SARS-CoV-2 infection (polymerase chain reaction >99% or antigen <1%) and adult patients without SARS-CoV-2 infection who served as controls from 34 medical centers nationwide between January 1, 2020, and December 7, 2020, patients were stratified using a World Health Organization COVID-19 severity scale and demographic characteristics. Differences between groups over time were evaluated using multivariable logistic regression. Random forest and XGBoost models were used to predict severe clinical course (death, discharge to hospice, invasive ventilatory support, or extracorporeal membrane oxygenation). MAIN OUTCOMES AND MEASURESPatient demographic characteristics and COVID-19 severity using the World Health Organization COVID-19 severity scale and differences between groups over time using multivariable logistic regression. RESULTSThe cohort included 174 568 adults who tested positive for SARS-CoV-2 (mean [SD] age, 44.4 [18.6] years; 53.2% female) and 1 133 848 adult controls who tested negative for SARS-CoV-2 (mean [SD] age, 49.5 [19.2] years; 57.1% female). Of the 174 568 adults with SARS-CoV-2, 32 472(18.6%) were hospitalized, and 6565 (20.2%) of those had a severe clinical course (invasive ventilatory support, extracorporeal membrane oxygenation, death, or discharge to hospice). Of the hospitalized patients, mortality was 11.6% overall and decreased from 16.4% in March to April 2020 to 8.6% in September to October 2020 (P = .002 for monthly trend). Using 64 inputs available on the first hospital day, this study predicted a severe clinical course using random forest and XGBoost models (area under the receiver operating curve = 0.87 for both) that were stable over time. The factor most strongly associated with clinical severity was pH; this result was consistent across machine learning methods. In a separate multivariable logistic regression model built for inference, (continued) Key Points Question In a US data resource large enough to adjust for multiple confounders, what risk factors are associated with COVID-19 severity and severity trajectory over time, and can machine learning models predict clinical severity? Findings In this cohort study of 174 568 adults with SARS-CoV-2, 32 472 (18.6%) were hospitalized and 6565 (20.2%) were severely ill, and first-day machine learning models accurately predicted clinical severity. Mortality was 11.6%

The erythrocyte membrane is a newly appreciated platform for thiol-based circulatory signaling, and it requires robust free thiol maintenance. We sought to define physiological constraints on erythrocyte antioxidant defense. Hemoglobin (Hb) conformation gates glycolytic flux through the hexose monophosphate pathway (HMP), the sole source of nicotinamide adenine dinucleotide phosphate (NADPH) in erythrocytes. We hypothesized elevated intraerythrocytic deoxyHb would limit resilience to oxidative stress. Human erythrocytes were subjected to controlled oxidant (superoxide) loading following independent manipulation of oxygen tension, Hb conformation, and glycolytic pathway dominance. Sufficiency of antioxidant defense was determined by serial quantification of GSH, NADPH, NADH redox couples. Hypoxic erythrocytes demonstrated greater loss of reduction potential [Delta GSH E(hc) (mV): 123.4+/-9.7 vs. 57.2+/-11.1] and reduced membrane thiol (47.7+/-5.7 vs. 20.1+/-4.3%) (hypoxia vs. normoxia, respectively; P<0.01), a finding mimicked in normoxic erythrocytes after HMP blockade. Rebalancing HMP flux during hypoxia restored resilience to oxidative stress at all stages of the system. Cell-free studies assured oxidative loading was not altered by oxygen tension, heme ligation, or the inhibitors employed. These data indicate that Hb conformation controls coupled glucose and thiol metabolism in erythrocytes, and implicate hypoxemia in the pathobiology of erythrocyte-based vascular signaling.

There is a paucity of data regarding the use of direct thrombin inhibitors such as bivalirudin for children on extracorporeal life support (ECLS). We sought to compare the outcomes of children on ECLS anticoagulated with bivalirudin versus heparin. Patients transitioned from heparin to bivalirudin were treated as a separate group. A single‐institution, retrospective review of all consecutive children (neonate to 18 years) placed on ECLS in the cardiac or pediatric intensive care units was performed (June 2018‐December 2019). Data collected included demographics, anticoagulation strategy, number of circuit interventions, blood product use on ECLS, survival to decannulation, and survival to discharge. Fifty‐four children were placed on ECLS for a total of 56 runs. Demographics and venovenous versus venoarterial ECLS were similar. The bivalirudin group had longer median duration of support compared to the heparin group––11.0 days [IQR 6.2, 23.1] versus 3.3 days [2.1, 6.2], P < .001. Patients switched from heparin to bivalirudin had a similar duration of support (10.3 days [8.3, 18.3]) as those on bilvalirudin alone. However, there was no difference in red blood cell, fresh frozen plasma, or platelet transfusions. There was no difference in the number of circuit interventions, survival to decannulation or discharge. The freedom to first circuit intervention was longer with bivalirudin compared to heparin. Our data suggest that even with longer pediatric ECLS runs on bivalirudin, there were no differences in the outcomes between the heparin and bivalirudin groups, with longer freedom from first circuit intervention with bivalirudin. While this is the largest reported series comparing children on ECLS anticoagulated with heparin versus bivalirudin, larger studies are needed to determine the optimal anticoagulation strategy for this diverse and complicated group of children.

Here, we review current data elucidating the role of red blood cell derived microparticles (RMPs) in normal vascular physiology and disease progression. Microparticles (MPs) are submicron-size, membrane-encapsulated vesicles derived from various parent cell types. MPs are produced in response to numerous stimuli that promote a sequence of cytoskeletal and membrane phospholipid changes and resulting MP genesis. MPs were originally considered as potential biomarkers for multiple disease processes and more recently are recognized to have pleiotropic biological effects, most notably in: promotion of coagulation, production and handling of reactive oxygen species, immune modulation, angiogenesis, and in initiating apoptosis. RMPs, specifically, form normally during RBC maturation in response to injury during circulation, and are copiously produced during processing and storage for transfusion. Notably, several factors during RBC storage are known to trigger RMP production, including: increased intracellular calcium, increased potassium leakage, and energy failure with ATP depletion. Of note, RMP composition differs markedly from that of intact RBCs and the nature/composition of RMP components are affected by the specific circumstances of RMP genesis. Described RMP bioactivities include: promotion of coagulation, immune modulation, and promotion of endothelial adhesion as well as influence upon vasoregulation via influence upon nitric oxide (NO) bioavailability. Of particular relevance, RMPs scavenge NO more avidly than do intact RBCs; this physiology has been proposed to contribute to the impaired oxygen delivery homeostasis that may be observed following transfusion. In summary, RMPs are submicron particles released from RBCs, with demonstrated vasoactive properties that appear to disturb oxygen delivery homeostasis. The clinical impact of RMPs in normal and patho-physiology and in transfusion recipients is an area of continued investigation.