ABSTRACT
Pandemic diseases have inflicted significant morbidity and mortality on humankind throughout history, often from acute respiratory failure due to pneumonia and the acute respiratory distress syndrome (ARDS). Extracorporeal membrane oxygenation (ECMO) has been used for the management of patients with ARDS during pandemics of the 21st century, including both the influenza A (H1N1) and coronavirus disease 2019 pandemics. While ECMO has been shown to improve outcomes in ARDS, it is resource intensive. During pandemic conditions, demand for ECMO has often outnumbered access and availability. In light of such limitation, healthcare systems, medical societies, and national governments have, at times, standardized approaches to and coordinated care of patients requiring ECMO in attempts to optimize patient outcomes and access to ECMO. © 2023 Elsevier Inc. All rights reserved.
ABSTRACT
Background: Revised guidelines clarify indications for extracorporeal membrane oxygenation (ECMO) support in patients with COVID-19-related acute respiratory distress syndrome (ARDS). Commercially available ECMO analytics software records granular perfusion data continuously throughout the run. To date, electronic-medical record (EMR) clinical data has not been integrated with ECMO perfusion data and analyzed with machine learning-based algorithms to improve patient care. Methods: Retrospective chart review was performed on all SARS-CoV2 positive patients cannulated to veno-venous ECMO at an urban highvolume regional referral center from March 1st, 2020, through December 31st, 2021. Categorical data including patient demographics, clinical outcomes, and laboratory data (complete blood count, basic metabolic panel, arterial blood gas, lactate, anticoagulation assays) and vital signs (pulse, arterial line blood pressure, oxygen saturation) were collected for the entirety of the ECMO run. Time-series perfusion data (arterial flow normalized to body surface area (BSA), sweep gas, delta pressures normalized to arterial flow) were captured every 60-120 seconds. We constructed a predictive long-short term memory (LSTM) predictive model that integrated clinical and time-series data using an extended machine learning (ML) framework with neural network. Primary outcome was successful ECMO decannulation. Data were truncated to discrete and relative timepoints (7, 14, 21 days, or percent of the run). Receiver operating characteristic (ROC) curves show the model's diagnostic accuracy. Results: 42 patients were included in the analysis (30 male, 12 female). Mean age was 43.9 (SD=11.5) years old, and mean duration of ECMO run was 36.2 (SD=30.1) days. 24 patients were successfully decannulated and 4 are currently supported on ECMO. When provided the complete data, the LSTM model showed an area under the ROC curve >0.95, demonstrating strong diagnostic accuracy in predicting successful ECMO decannulation (Figure 1A). When data were truncated to only the first two weeks of the ECMO run, the area under the ROC curve was 0.93 (Fig. 1B). Patterns of arterial flow normalized to BSA and sweep gas normalized to flow also appear different in patients with divergent clinical outcomes (Fig 2). Conclusion: Characterizing key determinants of ECMO support may offer intensive care unit healthcare teams potentially lifesaving information in real-time. Our machine-learning model successfully integrates clinical and perfusion data from the mind's eye of a clinician managing the care of a patient supported with ECMO. We have identified critical variables with the most meaningful impact on the mechanics of ECMO support. Our model may also help predict patient outcomes into and offer clinicians opportunities for interventions to improve care. (Figure Presented).
ABSTRACT
Background/aims: A specialist palliative care service evaluation in an acute hospital during the first wave of COVID-19 showed that those from ethnic minority backgrounds, especially women, were referred later. Improvements in treatments, and operational and system-level changes to the palliative care service which were introduced to address this disparity, may have improved access for those from ethnic minorities. Aim: To assess the effectiveness of operational and system-level changes to the hospital specialist palliative care service, by examining care patterns and trends for those with COVID-19. Methods: Retrospective service evaluation comparing patients referred to an acute hospital palliative care service with confirmed COVID-19 infection either at the peak of the first (Mar-Apr 2020, W1) or second (Dec 2020-Feb 2021, W2) wave of the pandemic. Demographic, clinical characteristics, and outcomes data were collected and compared using statistical tests;generalised linear mixed models for modelling of elapsed time from admission to referral;and survival analysis for each wave. Results: Data from 165 patients (W1 = 60, W2 =105) were included. Overall, patients in W1 were referred earlier to palliative care than in W2, particularly in the first 8 days from admission. Receiving dexamethasone, anticoagulants and absence of dementia, hypertension, and fever were associated with longer time to referral. The delay in referral from W1 of Black and Asian patients of 2-4 days, accounting for 22%-44% of the overall time from admission to death, was no longer observable in W2. The Australian-modified Karnofsky Performance Status (HR < 0.92, upper CI < 0.97) and phase of illness (HR > 3, lower CI >2) were good predictors of survival in both waves. Conclusions: The delayed referrals for ethnic minorities were not seen in W2. Actions to integrate palliative care within organisational COVID-19 planning, engaging with minority ethnic groups, and educating the workforce on culturally sensitive approaches to care may have had a positive impact on access to palliative care.