A Machine Learning Prediction Model of Respiratory Failure Within 48 Hours of Patient Admission for COVID-19: Model Development and Validation.

BACKGROUND: Predicting early respiratory failure due to COVID-19 can help triage patients to higher levels of care, allocate scarce resources, and reduce morbidity and mortality by appropriately monitoring and treating the patients at greatest risk for deterioration. Given the complexity of COVID-19, machine learning approaches may support clinical decision making for patients with this disease. OBJECTIVE: Our objective is to derive a machine learning model that predicts respiratory failure within 48 hours of admission based on data from the emergency department. METHODS: Data were collected from patients with COVID-19 who were admitted to Northwell Health acute care hospitals and were discharged, died, or spent a minimum of 48 hours in the hospital between March 1 and May 11, 2020. Of 11,525 patients, 933 (8.1%) were placed on invasive mechanical ventilation within 48 hours of admission. Variables used by the models included clinical and laboratory data commonly collected in the emergency department. We trained and validated three predictive models (two based on XGBoost and one that used logistic regression) using cross-hospital validation. We compared model performance among all three models as well as an established early warning score (Modified Early Warning Score) using receiver operating characteristic curves, precision-recall curves, and other metrics. RESULTS: The XGBoost model had the highest mean accuracy (0.919; area under the curve=0.77), outperforming the other two models as well as the Modified Early Warning Score. Important predictor variables included the type of oxygen delivery used in the emergency department, patient age, Emergency Severity Index level, respiratory rate, serum lactate, and demographic characteristics. CONCLUSIONS: The XGBoost model had high predictive accuracy, outperforming other early warning scores. The clinical plausibility and predictive ability of XGBoost suggest that the model could be used to predict 48-hour respiratory failure in admitted patients with COVID-19.

The interplay of DAMPs, TLR4, and proinflammatory cytokines in pulmonary fibrosis.

Pulmonary fibrosis is a chronic debilitating condition characterized by progressive deposition of connective tissue, leading to a steady restriction of lung elasticity, a decline in lung function, and a median survival of 4.5 years. The leading causes of pulmonary fibrosis are inhalation of foreign particles (such as silicosis and pneumoconiosis), infections (such as post COVID-19), autoimmune diseases (such as systemic autoimmune diseases of the connective tissue), and idiopathic pulmonary fibrosis. The therapeutics currently available for pulmonary fibrosis only modestly slow the progression of the disease. This review is centered on the interplay of damage-associated molecular pattern (DAMP) molecules, Toll-like receptor 4 (TLR4), and inflammatory cytokines (such as TNF-α, IL-1ß, and IL-17) as they contribute to the pathogenesis of pulmonary fibrosis, and the possible avenues to develop effective therapeutics that disrupt this interplay.

Therapeutic Potential of B-1a Cells in COVID-19.

Coronavirus disease 2019 (COVID-19) is a life-threatening respiratory illness caused by the novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Its clinical presentation can vary from the asymptomatic state to acute respiratory distress syndrome (ARDS) and multi-organ dysfunction. Due to our insufficient understanding of its pathophysiology and lack of effective treatment, the morbidity and mortality of severe COVID-19 patients are high. Patients with COVID-19 develop ARDS fueled by exaggerated neutrophil influx into the lungs and cytokine storm. B-1a cells represent a unique subpopulation of B lymphocytes critical for circulating natural antibodies, innate immunity, and immunoregulation. These cells spontaneously produce natural IgM, interleukin (IL)-10, and granulocyte-monocyte colony stimulating factor (GM-CSF). Natural IgM neutralizes viruses and opsonizes bacteria, IL-10 attenuates the cytokine storm, and GM-CSF induces IgM production by B-1a cells in an autocrine manner. Indeed, B-1a cells have been shown to ameliorate influenza virus infection, sepsis, and pneumonia, all of which are similar to COVID-19. The recent discovery of B-1a cells in humans further reinforces their potentially critical role in the immune response against SARS-CoV-2 and their anticipated translational applications against viral and microbial infections. Given that B-1a cells protect against ARDS via immunoglobulin production and the anti-COVID-19 effects of convalescent plasma treatment, we recommend that studies be conducted to further examine the role of B-1a cells in the pathogenesis of COVID-19 and explore their therapeutic potential to treat COVID-19 patients.