We studied the predictive value of the PaO2/FiO2 ratio for classifying COVID-19-positive patients who will develop severe clinical outcomes. One hundred fifty patients were recruited and categorized into two distinct populations (“A” and “B”), according to the indications given by the World Health Organization. Patients belonging the population “A” presented with mild disease not requiring oxygen support, whereas population “B” presented with a severe disease needing oxygen support. The AUC curve of PaO2/FiO2 in the discovery cohort was 0.838 (95% CI 0.771–0.908). The optimal cut-off value for distinguishing population “A” from the “B” one, calculated by Youden’s index, with sensitivity of 71.79% and specificity 85.25%, LR+4.866, LR−0.339, was < 274 mmHg. The AUC in the validation cohort of 170 patients overlapped the previous one, i.e., 0.826 (95% CI 0.760–0.891). PaO2/FiO2 ratio < 274 mmHg was a good predictive index test to forecast the development of a severe respiratory failure in SARS-CoV-2-infected patients. Moreover, our work highlights that PaO2/FiO2 ratio, compared to inflammatory scores (hs-CRP, NLR, PLR and LDH) indicated to be useful in clinical managements, results to be the most reliable parameter to identify patients who require closer respiratory monitoring and more aggressive supportive therapies. Clinical trial registration: Prognostic Score in COVID-19, prot. NCT04780373 https://clinicaltrials.gov/ct2/show/NCT04780373 (retrospectively registered).

No abstract

To support physicians in clinical decision process on patients affected by Coronavirus Disease 2019 (COVID-19) in areas with a low vaccination rate, we devised and evaluated the performances of several machine learning (ML) classifiers fed with readily available clinical and laboratory data. Our observational retrospective study collected data from a cohort of 779 COVID-19 patients presenting to three hospitals of the Lazio-Abruzzo area (Italy). Based on a different selection of clinical and respiratory (ROX index and PaO2/FiO2 ratio) variables, we devised an AI-driven tool to predict safe discharge from ED, disease severity and mortality during hospitalization. To predict safe discharge our best classifier is an RF integrated with ROX index that reached AUC of 0.96. To predict disease severity the best classifier was an RF integrated with ROX index that reached an AUC of 0.91. For mortality prediction the best classifier was an RF integrated with ROX index, that reached an AUC of 0.91. The results obtained thanks to our algorithms are consistent with the scientific literature an accomplish significant performances to forecast safe discharge from ED and severe clinical course of COVID-19.