Possibilities of information systems for prediction of outcomes of new coronavirus infection COVID-19

The pandemic of coronavirus infection COVID-19 (Coronavirus Disease 2019), caused by a new strain of coronavirus SARSCoV- 2 (severe acute respiratory syndrome coronavirus 2), has caused high mortality worldwide. The clinical manifestations of COVID-19 are nonspecific. Diagnostics includes clinical, laboratory and radiological data. The importance of introducing information systems into medical practice in order to improve the quality of medical care is noted. It is stated that the development of medical artificial intelligence is associated with the development of artificial intelligence programs designed to help the clinic in making a diagnosis, prescribing treatment, as well as predicting the outcome of the disease. Such systems include artificial neural networks, fuzzy expert systems, and hybrid intelligent systems. The article analyzes data from a number of studies on the use of artificial intelligence for diagnosing COVID-19, predicting the risk of mortality and studying risk factors for severe course and lethal outcome in various groups. Using clusters of predictors, models have been developed to predict mortality and understand the relationship of various characteristics and diseases with mortality from COVID-19. The article also summarizes the key factors that worsen the prognosis for COVID-19. Scales for detecting or predicting the development of COVID-19-induced “cytokine storm” are marked as a separate item. © 2022, Remedium Group Ltd. All rights reserved.

Results of mechanical ventilation in patients with covid-19 complicated by acute respiratory distress syndrome

Objective. To analyze the results of mechanical ventilation in patients with COVID-19 pneumonia complicated by acute respiratory distress syndrome (ARDS). Material and methods. A single-center observational study included 60 patients aged 63 [54;74] years with severe and extremely severe forms of COVID-19 complicated by ARDS. All patients were divided into 2 groups: The 1st group-14 patients with a favorable outcome, the 2nd group-46 patients with an unfavorable outcome. All patients underwent complex therapy including respiratory support in accordance with the Russian guidelines for the treatment of coronavirus infection and acute respiratory distress syndrome. Clinical state was assessed using the SAPS II scale, severity of organ dysfunction-SOFA scale, ARDS severity-LIS scale. Mechanical properties of lungs were analyzed considering F, Vt, MV, PIP, PEEP, CPAP. Gas exchange was analyzed via continuous monitoring of SpO2, PetCO2, plethysmography and blood gas tests. These data were recorded in mechanical ventilation card for subsequent calculation of some parameters (FiO2, PaO2, PaCO2, pH, PaO2/FiO2). Laboratory tests included leukocyte, lymphocyte, platelet count and C-reactive protein. Examinations were carried out at the following stages of respiratory support: 1 day (I stage), 3 days (II stage), 5 days (III stage), 10 days (IV stage), 14 days (stage V) and in preterminal period in patients with unfavorable outcome (stage VI). We analyzed duration of respiratory support, ICU-stay, hospital-stay, morbidity and mortality rates (within 28 days). Results. Patients with acute hypoxemic respiratory failure following severe and extremely severe COVID-19 pneumonia had severe lung damage (ARDS LIS score>2.5) regardless of the outcome (favorable or unfavorable). There was only a tendency of less LIS score of lung injury in patients with favorable outcome. Patients with subsequent unfavorable outcome required higher PEEP for adequate ventilation while other parameters of respiratory support (Vt, F, MV, PIP) were comparable. Severe and extremely severe COVID-19 pneumonia followed by unfavorable outcome was characterized by significantly more severe disorders of gas exchange (PaO2/FiO2) which could not be corrected despite intensive care including mechanical ventilation. LIS score increased up to 3.8 in preterminal period compared to 2.8 at the beginning of ventilation. This fact indicated ARDS progression. Oxygen status disorders were extremely severe (PaO2/FiO2 92.1 mm Hg) but did not reach critical hypoxemia (PaO2/FiO2<75 mm Hg). Conclusion. Mechanical ventilation was effective in 23.3% of patients with severe and extremely severe forms of new coronavirus infection. However, this study has certain limitations associated with single-center design, no pathogenetic (effective antiviral therapy) treatment and different versions of the COVID-19 treatment guidelines. © A.I. GRITSAN, N.V. AVDEEV, I.V. DEMKO, V.V. ISHUTIN, E.E. KORCHAGIN.