Quantifying Oxygen Demand by Patients Hospitalized with COVID-19 at a Large Safety-Net Hospital Using Multiple Methodologies (preprint)

Background: During the COVID-19 pandemic, many facilities worldwide struggled to forecast oxygen demand, which often exceeded oxygen supply to the detriment of patient care. Accurate estimates of oxygen demand by patients with COVID-19 are scarce, and proposed estimation methods have not been fully evaluated or implemented. To address this knowledge gap, oxygen demand by COVID-19 patients was calculated at a large safety-net hospital in the United States using patient consumption (demand) data, oxygen procurement (supply) data, and modeled data with a novel calculator tool. Methods: Data were extracted from electronic medical records of patients admitted with COVID-19 to Zuckerberg San Francisco General Hospital (ZSFG) from March 2020 to March 2022, including every recorded peripheral oxygen saturation (SpO2) measurement as well as oxygen delivery device(s) and settings. Total patient oxygen consumption was calculated as the sum of oxygen delivery amounts for each recorded time interval during hospitalization. Oxygen delivery amounts were calculated using delivery device-specific formulas. Patient and treatment-specific factors which may impact oxygen demand were also reported. For comparison, oxygen procurement logs from the study period were reviewed to estimate supply consumed, and the Oxygencalculator.com tool was used to model oxygen demand using an experimental patient population of the same size. Results: In total, 282,095 time points from 1,076 patients were analyzed. Two-thirds of patients received oxygen, of which 24.3% received high-flow oxygen (HFO) therapy and 16.0% received invasive mechanical ventilation (IMV) at some point. In-hospital mortality was 7.5% overall, 10.8% for patients who received oxygen, and 28.3% for patients who received IMV. The median (IQR) duration of oxygen therapy was 3.1 (0.8-8.9) days, mean (SD) oxygen flow was 5.6 (5.0) liters per minute (LPM), and mean (SD) total volume of oxygen delivered was 180,115 (510,330) liters (L) per hospitalization. Both the supply- and model-based methods overestimated oxygen consumption compared to demand estimated from patient data. Conclusions: This study represents one of the largest cohorts of patients with COVID-19 for which oxygen demand has been calculated, including patient clinical characteristics which may help explain variations in oxygen demand. Moreover, oxygen demand was quantified using a methodology that could be applied in any setting.

AnmO2l: An open-source pulse-dose oxygen conserving device for the COVID-19 crisis (preprint)

The surge of COVID-19 cases in India and around the world has resulted in acute oxygen shortage. Oxygen therapy through nasal cannula with flow rates of 1-6 L/min is an effective treatment for many COVID-19 patients in non-critical conditions and may help prevent disease progression. The same treatment is also used in post-acute care for recovering COVID-19 patients. Here we describe a simple, open source, and rapidly manufacturable oxygen conservation device for use with dual-port nasal cannula that can extend the life of current oxygen supply by almost two to three times, which we hope will help towards coping with the ongoing crisis. We estimate the bill of materials cost to be under $50 for manufacturing in bulk.

Management and Outcomes of Critically-Ill Patients with COVID-19 Pneumonia at a Safety-net Hospital in San Francisco, a Region with Early Public Health Interventions: A Case Series (preprint)

Background: Following early implementation of public health measures, San Francisco has experienced a slow rise and a low peak level of coronavirus disease 2019 (COVID-19) cases and deaths. Methods and Findings: We included all patients with COVID-19 pneumonia admitted to the intensive care unit (ICU) at the safety net hospital for San Francisco through April 8, 2020. Each patient had [≥]15 days of follow-up. Among 26 patients, the median age was 54 years (interquartile range, 43 to 62), 65% were men, and 77% were Latinx. Mechanical ventilation was initiated for 11 (42%) patients within 24 hours of ICU admission and 20 patients (77%) overall. The median duration of mechanical ventilation was 13.5 days (interquartile range, 5 to 20). Patients were managed with lung protective ventilation (tidal volume <8 ml/kg of ideal body weight and plateau pressure [≤]30 cmH2O on 98% and 78% of ventilator days, respectively). Prone positioning was used for 13 of 20 (65%) ventilated patients for a median of 5 days (interquartile range, 2 to 10). Seventeen (65%) patients were discharged home, 1 (4%) was discharged to nursing home, 3 (12%) were discharged from the ICU, and 2 (8%) remain intubated in the ICU at the time of this report. Three (12%) patients have died. Conclusions: Good outcomes were achieved in critically ill patients with COVID-19 by using standard therapies for acute respiratory distress syndrome (ARDS) such as lung protective ventilation and prone positioning. Ensuring hospitals can deliver sustained high-quality and evidence-based critical care to patients with ARDS should remain a priority.