A Computational Approach in the Diagnostic Process of COVID-19: The Missing Link between the Laboratory and Emergency Department.

BACKGROUND: The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is responsible for the COVID-19 pandemic and so it is crucial the right evaluation of viral infection. According to the Centers for Disease Control and Prevention (CDC), the Real-Time Reverse Transcription PCR (RT-PCR) in respiratory samples is the gold standard for confirming the disease. However, it has practical limitations as time-consuming procedures and a high rate of false-negative results. We aim to assess the accuracy of COVID-19 classifiers based on Arificial Intelligence (AI) and statistical classification methods adapted on blood tests and other information routinely collected at the Emergency Departments (EDs). METHODS: Patients admitted to the ED of Careggi Hospital from April 7th-30th 2020 with pre-specified features of suspected COVID-19 were enrolled. Physicians prospectively dichotomized them as COVID-19 likely/unlikely case, based on clinical features and bedside imaging support. Considering the limits of each method to identify a case of COVID-19, further evaluation was performed after an independent clinical review of 30-day follow-up data. Using this as a gold standard, several classifiers were implemented: Logistic Regression (LR), Quadratic Discriminant Analysis (QDA), Random Forest (RF), Support Vector Machine (SVM), Neural Networks (NN), K-nearest neighbor (K-NN), Naive Bayes (NB). RESULTS: Most of the classifiers show a ROC >0.80 on both internal and external validation samples but the best results are obtained applying RF, LR and NN. The performance from the external validation sustains the proof of concept to use such mathematical models fast, robust and efficient for a first identification of COVID-19 positive patients. These tools may constitute both a bedside support while waiting for RT-PCR results, and a tool to point to a deeper investigation, by identifying which patients are more likely to develop into positive cases within 7 days. CONCLUSIONS: Considering the obtained results and with a rapidly changing virus, we believe that data processing automated procedures may provide a valid support to the physicians facing the decision to classify a patient as a COVID-19 case or not.

Diagnostic accuracy of physician's gestalt in suspected COVID-19: Prospective bicentric study.

OBJECTIVES: Physicians' gestalt is central in the diagnostic pipeline of suspected COVID-19, due to the absence of a single tool allowing conclusive rule in or rule out. The aim of this study was to estimate the diagnostic test characteristics of physician's gestalt for COVID-19 in the emergency department (ED), based on clinical findings or on a combination of clinical findings and bedside imaging results. METHODS: From April 1 to April 30, 2020, patients with suspected COVID-19 were prospectively enrolled in two EDs. Physicians prospectively dichotomized patients in COVID-19 likely or unlikely twice: after medical evaluation of clinical features (clinical gestalt [CG]) and after evaluation of clinical features and results of lung ultrasound or chest x-ray (clinical and bedside imaging-integrated gestalt [CBIIG]). The final diagnosis was adjudicated after independent review of 30-day follow-up data. RESULTS: Among 838 ED enrolled patients, 193 (23%) were finally diagnosed with COVID-19. The area under the curve (AUC), sensitivity, and specificity of CG and CBIIG for COVID-19 were 80.8% and 91.6% (p < 0.01), 82.9% and 91.4% (p = 0.01), and 78.6% and 91.8% (p < 0.01), respectively. CBIIG had similar AUC and sensitivity to reverse transcription-polymerase chain reaction (RT-PCR) for SARS-CoV-2 on the first nasopharyngeal swab per se (93.5%, p = 0.24; and 87%, p = 0.17, respectively). CBIIG plus RT-PCR had a sensitivity of 98.4% for COVID-19 (p < 0.01 vs. RT-PCR alone) compared to 95.9% for CG plus RT-PCR (p = 0.05). CONCLUSIONS: In suspected COVID-19, CG and CBIIG have fair diagnostic accuracy, in line with physicians' gestalt for other acute conditions. Negative RT-PCR plus low probability based on CBIIG can rule out COVID-19 with a relatively low number of false-negative cases.

Can non-invasive ventilation modify central venous pressure? Comparison between invasive measurement and ultrasonographic evaluation.

Central venous pressure (CVP) is primarily measured to assess intravascular volume status and heart preload. In clinical practice, the measuring device most commonly used in emergency departments and intensive care units, is an electronic transducer that interconnects a central venous catheter (CVC) with a monitoring system. Non-invasive ventilation (NIV) consists in a breathing support that supplies a positive pressure in airways through a mask or a cask though not using an endotracheal prosthesis. In emergency settings, non-invasive ultrasonography evaluation of CVP, and hence of intravascular volume status entail the measurement by a subxiphoid approach of inferior vena cava diameter and its variations in relation to respiratory activity. In the literature, there are many studies analyzing the ability to estimate CVP through ultrasonography, rating inspiratory and expiratory vena cava diameters and their ratio, defined as inferior vena cava collapsibility index (IVC-CI). At the same time, the effects of invasive mechanical ventilation on blood volume and the correlation during ventilation between hemodynamic invasive measurement of CVP and inferior vena cava diameters have already been demonstrated. Nevertheless, there are no available data regarding the hemodynamic effects of NIV and the potential correlations during this kind of ventilation between invasive and non-invasive CVP measurements. Therefore, this study aims to understand whether there exists or not an interrelationship between the values of CVP assessed invasively through a CVC and non-invasively through the IVC-CI in patients with severe respiratory distress, and above all to evaluate if these means of assessment can be influenced using NIV.