Alex K. Pearce

Background/Objectives: The COVID-19 pandemic highlighted the limitations of pulse oximetry in detecting occult hypoxemia. The superiority of the alveolar gas monitor (AGM) compared to pulse oximetry (SpO₂) in predicting respiratory deterioration among COVID-19-positive individuals has previously been demonstrated. Here, we combine COVID-19 and non-COVID-19 individuals as a combined cohort of participants to determine if the AGM has similar utility across a larger, more generalizable cohort. Methods: Adult patients (n = 75) at risk of respiratory deterioration in the emergency department (ED) underwent prospective assessments of their oxygen deficit (OD) and SpO₂, simultaneously measured during quiet breathing on the AGM. The OD and SpO₂ were then compared for their ability to predict the dichotomous outcome of the need for supplemental oxygen. The administration of supplemental oxygen was ordered by the clinical care team with no knowledge of the patients’ enrollment in this study. Results: In the logistic regression analysis, both SpO₂ and OD significantly predicted the need for supplemental oxygen among COVID-19-negative individuals. However, in the multivariable regression, only OD (p < 0.001) significantly predicted the need for supplemental oxygen, while SpO₂ (p = 0.05) did not in the combined cohort of COVID-19-negative and -positive individuals. Receiver operating characteristic (ROC) curve analysis demonstrated the superior discriminative ability of OD (area under ROC curve = 0.937) relative to SpO₂ (area under ROC curve = 0.888) to predict the need for supplemental oxygen. Conclusions: The noninvasive AGM, which combines the measurement of exhaled partial pressures of gas with SpO₂, outperforms SpO₂ alone in predicting the need for supplemental oxygen among individuals in the ED at risk of respiratory deterioration regardless of the etiology for their symptoms (COVID-19-positive or -negative).

Background: Mortality in patients with acute respiratory failure remains high. Predicting progression of acute respiratory failure may be critical to improving patient outcomes. Machine learning, a subset of artificial intelligence is a rapidly expanding area, which is being integrated into several areas of clinical medicine. This manuscript will address the knowledge gap in predicting the onset and progression of respiratory failure, provide a review of existing prognostic strategies, and provide a clinical perspective on the implementation and future integration of machine learning into clinical care. Main body: Existing strategies for predicting respiratory failure, such as prediction scores and biomarkers, offer both strengths and limitations. While these tools provide some prognostic value, machine learning presents a promising, data-driven approach to prognostication in the intensive care unit. Machine learning has already shown success in various areas of clinical medicine, although relatively few algorithms target respiratory failure prediction specifically. As machine learning grows in the context of respiratory failure, outcomes such as the need for invasive mechanical ventilation and escalation of respiratory support (e.g. non-invasive ventilation) have been identified as key targets. However, the development and implementation of machine learning models in clinical care involves complex challenges. Future success will depend on rigorous model validation, clinician collaboration, thoughtful trial design, and the application of implementation science to ensure integration into clinical care. Conclusion: Machine learning holds promise for optimizing treatment strategies and potentially improving outcomes in respiratory failure. However, further research and development are necessary to fully realize its potential in clinical practice.

In some patients with acute respiratory distress syndrome (ARDS), a paradoxical improvement in respiratory system compliance (C_RS) has been observed when assuming a supine (head of bed [HOB] 0°) compared with semirecumbent (HOB 35–40°) posture. We sought to test the hypothesis that mechanically ventilated patients with ARDS would have improved C_RS, due to changes in ventilation distribution, when moving from the semirecumbent to supine position. We conducted a prospective, observational ICU study including 14 mechanically ventilated patients with ARDS. For each patient, ventilation distribution (assessed by electrical impedance tomography) and pulmonary mechanics were compared in supine versus semirecumbent postures. Compared with semirecumbent, in the supine posture C_RS increased (33 ± 21 vs. 26 ± 14 mL/cm H₂O, p = 0.005), driving pressure was reduced (14 ± 6 vs. 17 ± 7 cm H₂O, p < 0.001), and dorsal fraction of ventilation was decreased (48.5 ± 14.1% vs. 54.5 ± 12.0%, p = 0.003). Posture change from semirecumbent to supine resulted in a favorable physiologic response in terms of improved C_RS and reduced driving pressure—with a corresponding increase in ventral ventilation, possibly related to reduced ventral overdistension.

Secondary infections can occur during or after the treatment of an initial infection. Glucocorticoids may decrease mortality in patients with severe COVID-19; however, risk of secondary infection is not well described. Our primary objective was to investigate the risk of secondary infection among critically ill patients with COVID-19 treated with glucocorticoids. We examined patients with COVID-19 being treated in the intensive care unit at two academic medical centers from 1 to 7/2020. One hundred-seven patients were included. Of these, 31 received steroids and 76 patients did not. Analysis of the larger cohort was performed followed by a matched pairs analysis of 22 steroid and 22 non-steroid patients. Secondary infection was seen in 14 patients (45.2%) receiving steroids compared to 35(46.1%) not receiving steroids (p = 0.968). Secondary infections were most frequently encountered in the respiratory tract. Escherichia coli and Staphylococcus aureus were the most frequently identified organisms. Mortality was 16.1% in the steroid-treated group compared to 23.7% in the control group (p = 0.388). After performing matched pairs analysis and multivariable logistic regression there was no significant difference between secondary infection or mortality and steroid receipt. Secondary infections were common among critically ill patients with COVID-19, but the incidence of secondary infection was not significantly impacted by steroid treatment.

Background and Objective: The COVID-19 pandemic magnified the importance of gas exchange abnormalities in early respiratory failure. Pulse oximetry (SpO₂) has not been universally effective for clinical decision-making, possibly because of limitations. The alveolar gas monitor (AGM100) adds exhaled gas tensions to SpO₂ to calculate the oxygen deficit (OD). The OD parallels the alveolar-to-arterial oxygen difference (AaDO₂) in outpatients with cardiopulmonary disease. We hypothesized that the OD would discriminate between COVID-19 patients who require hospital admission and those who are discharged home, as well as predict need for supplemental oxygen during the index hospitalization. Methods: Patients presenting with dyspnea and COVID-19 were enrolled with informed consent and had OD measured using the AGM100. The OD was then compared between admitted and discharged patients and between patients who required supplemental oxygen and those who did not. The OD was also compared to SpO₂ for each of these outcomes using receiver operating characteristic (ROC) curves. Results: Thirty patients were COVID-19 positive and had complete AGM100 data. The mean OD was significantly (p = 0.025) higher among those admitted 50.0 ± 20.6 (mean ± SD) vs. discharged 27.0 ± 14.3 (mean ± SD). The OD was also significantly (p < 0.0001) higher among those requiring supplemental oxygen 60.1 ± 12.9 (mean ± SD) vs. those remaining on room air 25.2 ± 11.9 (mean ± SD). ROC curves for the OD demonstrated very good and excellent sensitivity for predicting hospital admission and supplemental oxygen administration, respectively. The OD performed better than an SpO₂ threshold of <94%. Conclusions: The AGM100 is a novel, noninvasive way of measuring impaired gas exchange for clinically important endpoints in COVID-19.