Prone position in intubated, mechanically ventilated patients with COVID-19: a multi-centric study of more than 1000 patients.

BACKGROUND: Limited data are available on the use of prone position in intubated, invasively ventilated patients with Coronavirus disease-19 (COVID-19). Aim of this study is to investigate the use and effect of prone position in this population during the first 2020 pandemic wave. METHODS: Retrospective, multicentre, national cohort study conducted between February 24 and June 14, 2020, in 24 Italian Intensive Care Units (ICU) on adult patients needing invasive mechanical ventilation for respiratory failure caused by COVID-19. Clinical data were collected on the day of ICU admission. Information regarding the use of prone position was collected daily. Follow-up for patient outcomes was performed on July 15, 2020. The respiratory effects of the first prone position were studied in a subset of 78 patients. Patients were classified as Oxygen Responders if the PaO2/FiO2 ratio increased ≥ 20 mmHg during prone position and as Carbon Dioxide Responders if the ventilatory ratio was reduced during prone position. RESULTS: Of 1057 included patients, mild, moderate and severe ARDS was present in 15, 50 and 35% of patients, respectively, and had a resulting mortality of 25, 33 and 41%. Prone position was applied in 61% of the patients. Patients placed prone had a more severe disease and died significantly more (45% vs. 33%, p < 0.001). Overall, prone position induced a significant increase in PaO2/FiO2 ratio, while no change in respiratory system compliance or ventilatory ratio was observed. Seventy-eight % of the subset of 78 patients were Oxygen Responders. Non-Responders had a more severe respiratory failure and died more often in the ICU (65% vs. 38%, p = 0.047). Forty-seven % of patients were defined as Carbon Dioxide Responders. These patients were older and had more comorbidities; however, no difference in terms of ICU mortality was observed (51% vs. 37%, p = 0.189 for Carbon Dioxide Responders and Non-Responders, respectively). CONCLUSIONS: During the COVID-19 pandemic, prone position has been widely adopted to treat mechanically ventilated patients with respiratory failure. The majority of patients improved their oxygenation during prone position, most likely due to a better ventilation perfusion matching. TRIAL REGISTRATION: clinicaltrials.gov number: NCT04388670.

Development of machine learning models to predict RT-PCR results for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in patients with influenza-like symptoms using only basic clinical data.

BACKGROUND: Reverse Transcription-Polymerase Chain Reaction (RT-PCR) for Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-COV-2) diagnosis currently requires quite a long time span. A quicker and more efficient diagnostic tool in emergency departments could improve management during this global crisis. Our main goal was assessing the accuracy of artificial intelligence in predicting the results of RT-PCR for SARS-COV-2, using basic information at hand in all emergency departments. METHODS: This is a retrospective study carried out between February 22, 2020 and March 16, 2020 in one of the main hospitals in Milan, Italy. We screened for eligibility all patients admitted with influenza-like symptoms tested for SARS-COV-2. Patients under 12 years old and patients in whom the leukocyte formula was not performed in the ED were excluded. Input data through artificial intelligence were made up of a combination of clinical, radiological and routine laboratory data upon hospital admission. Different Machine Learning algorithms available on WEKA data mining software and on Semeion Research Centre depository were trained using both the Training and Testing and the K-fold cross-validation protocol. RESULTS: Among 199 patients subject to study (median [interquartile range] age 65 [46-78] years; 127 [63.8%] men), 124 [62.3%] resulted positive to SARS-COV-2. The best Machine Learning System reached an accuracy of 91.4% with 94.1% sensitivity and 88.7% specificity. CONCLUSION: Our study suggests that properly trained artificial intelligence algorithms may be able to predict correct results in RT-PCR for SARS-COV-2, using basic clinical data. If confirmed, on a larger-scale study, this approach could have important clinical and organizational implications.