Immunomodulators in patients receiving extracorporeal membrane oxygenation for COVID-19: a propensity-score weighted analysis of the ELSO registry

Background: Mortality for patients receiving extracorporeal membrane oxygenation (ECMO) for COVID-19 has increased over time. We investigated the association between immunomodulators and mortality for patients receiving ECMO for COVID-19. Method(s): We analysed the Extracorporeal Life Support Organisation Registry from Jan 1, 2020 through Dec 31, 2021, comparing in-hospital mortality and the overall survival since ECMO initiation of patients who did and did not receive immunomodulators (selective interleukin blockers, corticosteroids, janus-kinase inhibitors, convalescent plasma, and intravenous immunoglobulins) before or during ECMO, by using logistic regression and Cox regression. We calculated the propensity scores, and applied the overlap-weightage method to account for confounding factors. We conducted sensitivity analyses including regression models using inverse propensity score weightage method, models adjusted by propensity score, and unadjusted models to ensure robustness of results. A subgroup analysis on patients receiving corticosteroids was conducted. Result(s): 7180 patients were included in the final analysis. Immunomodulators were associated with increased mortality (OR: 1.168, 95%CI: 1.059-1.289, p=0.0020) and shorter survival since ECMO (HR: 1.05, 95%- CI: 1.078, 95%CI: 1.007-1.154, p=0.031). Similarly, corticosteroids were associated with a significant increase in mortality (OR: 1.302, 95%-CI: 1.180-1.437, p<0.0001) and shorter survival (HR: 1.221, 95%-CI: 1.139- 1.308, p<0.0001). Sensitivity analyses did not significantly change the overall results. Conclusion(s): Immunomodulators and corticosteroids, in particular, were associated with a significant increase in mortality amongst patients receiving ECMO for COVID-19, even after adjusting for potential confounding variables. Further studies are required to evaluate the timing of immunomodulators and understand the possible mechanisms behind this association.

COVID-19-related ARDS Supported with Extracorporeal Membrane Oxygenation: Using Machine Learning Models to Improve Care

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).

Invasive mechanical ventilation in patients with acute respiratory distress syndrome receiving extracorporeal support: a narrative review of strategies to mitigate lung injury.

Veno-venous extracorporeal membrane oxygenation is indicated in patients with acute respiratory distress syndrome and severely impaired gas exchange despite evidence-based lung protective ventilation, prone positioning and other parts of the standard algorithm for treating such patients. Extracorporeal support can facilitate ultra-lung-protective ventilation, meaning even lower volumes and pressures than standard lung-protective ventilation, by directly removing carbon dioxide in patients needing injurious ventilator settings to maintain sufficient gas exchange. Injurious ventilation results in ventilator-induced lung injury, which is one of the main determinants of mortality in acute respiratory distress syndrome. Marked reductions in the intensity of ventilation to the lowest tolerable levels under extracorporeal support may be achieved and could thereby potentially mitigate ventilator-induced lung injury and theoretically patient self-inflicted lung injury in spontaneously breathing patients with high respiratory drive. However, the benefits of this strategy may be counterbalanced by the use of continuous deep sedation and even neuromuscular blocking drugs, which may impair physical rehabilitation and impact long-term outcomes. There are currently a lack of large-scale prospective data to inform optimal invasive ventilation practices and how to best apply a holistic approach to patients receiving veno-venous extracorporeal membrane oxygenation, while minimising ventilator-induced and patient self-inflicted lung injury. We aimed to review the literature relating to invasive ventilation strategies in patients with acute respiratory distress syndrome receiving extracorporeal support and discuss personalised ventilation approaches and the potential role of adjunctive therapies in facilitating lung protection.

An appraisal of respiratory system compliance in mechanically ventilated COVID-19 patients

Rationale Heterogeneous respiratory system static compliance (CRS) values and levels of hypoxemia in patients with novel coronavirus disease (COVID-19) requiring mechanical ventilation have been reported in previous smallcase series or studies conducted at a national level.Methods We designed a retrospective observational cohort study with rapid data gathering from the international COVID-19 Critical Care Consortium study to comprehensively describe the impact of CRS on the ventilatory management and outcomes of COVID-19 patients on mechanical ventilation (MV), admitted to intensive care units (ICU) worldwide.Results We enrolled 318 COVID-19 patients enrolled into the study from January 14th through September 31th, 2020 in 19 countries and stratified into two CRS groups. CRS was calculated as: tidal volume/[airway plateau pressure-positive endexpiratory pressure (PEEP)] and available within 48h from commencement of MV in 318 patients. Patients were mean±SD of 58.0±12.2, predominantly from Europe (54%) and males (68%). Median CRS (IQR) was 34.1 mL/cmH2O (26.5-45.5) and PaO2/FiO2 was 119 mmHg (87.1-164) and was not correlated with CRS. Female sex presented lower CRS than in males (95% CI:-13.8 to-8.5 P<0.001) and higher body mass index (34.7±10.9 vs 29.1±6.0, p<0.001). Median (IQR) PEEP was 12 cmH2O (10-15), throughout the range of CRS, while median (IQR) driving pressure was 12.3 (10-15) cmH2O and significantly decreased as CRS improved (p<0.001). No differences were found in comorbidities and clinical management between CRS strata. In addition, 28-day ICU mortality and hospital mortality did not differ between CRSgroups.Conclusions This multicentre report provides a comprehensive account of CRS in COVID-19 patients on MV-predominantly males or overweight females, in their late 50s-admitted to ICU during the first international outbreaks. Phenotypes associated with different CRS upon commencement of MV could not be identified.

Interleukin-6 and pulmonary compliance in COVID-19-related acute respiratory distress syndrome

Rationale: Higher levels of circulating interleukin-6 (IL-6) and lower respiratory system compliance have each been associated with increased mortality in severe coronavirus 2019 (COVID-19). IL-6 levels are associated with disease severity and mortality in non-COVID-19-related acute respiratory distress syndrome (ARDS). The purpose of this study was to examine the relationship between IL-6 and respiratory mechanics in COVID-19-related ARDS. Methods: This retrospective cohort study took place at two Columbia University Irving Medical Center hospitals. We identified patients age >18 years with laboratory confirmed COVID-19, who were intubated from March 1st through April 30th, 2020, and met the Berlin definition of ARDS. Electronic medical records were reviewed for clinical data. Outcomes were censored at 90 days after intubation. For patients without IL-6 levels recorded on the initial day of intubation, serum samples were obtained from the Columbia University Biobank and tested using the Quantikine Human IL-6 Immunoassay. IL-6 values were log-transformed. The primary outcome was respiratory system compliance. Secondary outcomes were calculated ventilatory ratio, PaO2:FiO2 ratio, and mortality. Linear regression and logistic regression were used for statistical analyses. Results: During the study period, 483 patients had COVID-19-associated ARDS. Median time of follow up was 37 days (IQR 11-90). At 90 days, 260 (53.8%) patients were deceased, 206 (42.7%) had been discharged, and 17 (3.5%) were still admitted. Two hundred sixteen (44.7%) patients had available data on respiratory system compliance and serum IL-6 levels from the initial day of mechanical ventilation. The median IL-6 value was 204.1 pg/ml (IQR 110-469.7). Median compliance was 25.5 ml/cmH2O (IQR 21.4-33.3), median ventilatory ratio was 1.96 (IQR 1.51-2.57), and median PaO2:FiO2 ratio was 134 (IQR 87-196). In unadjusted linear regression, higher IL-6 was associated with lower respiratory system compliance (log [IL-6] coefficient-1.80, p = 0.001) (Figure 1). This relationship remained significant when adjusting for age, sex, body mass index, race, ethnicity, and Sequential Organ Failure Assessment (SOFA) score (coefficient-2.43, p<0.001). There was no significant association between IL-6 and ventilatory ratio (0.76 p=0.08) or PaO2:FiO2 ratio (-6.15 p=0.06). Higher IL-6 was associated with higher odds of death at 90 days (OR 1.35 per unit increase in log [IL-6], p-value 0.022) when adjusting for age, sex, body mass index, race, ethnicity, and SOFA score. Conclusion: In COVID-19-associated ARDS, higher levels of IL-6 were associated with lower respiratory system compliance even adjusting for measured confounders. Higher IL-6 was also associated with higher mortality.

Factors associated with mortality in patients with COVID-19 requiring mechanical ventilation: An international cohort study from 139 intensive care unit across 6 continents

Rationale: Patients with COVID-19 commonly develop severe hypoxemic respiratory failure and require invasive mechanical ventilation (MV). The disease burden and predictors of mortality in this population remain uncertain. Methods: Prospective observational cohort study from 139 intensive care units of the international COVID-19 Critical Care Consortium. Patients enrolled from January 14th through November 31st 2020 were included in the analysis. Patient's characteristics and clinical data were assessed. Multivariable Cox proportional hazards analysis was conducted to identify indipendent predictors of mortality within 28 days from commencement of MV. Results: 1578 patients on MV were included into the analysis. Mean±SD age was 59 years±13 and patients were predominantly males (66%). 542 Patients (34.4%) died within 28 days from commencement of MV. Nonsurvivors were slightly older (mean age±SD 62±13 vs. 59±13) and presented more frequently hypertension, chronic cardiac disease and diabetes. Median (IQR) PaO2/FiO2 upon commencement of MV was 96 (68-135) and 111 (81-173) in patients who did not survive vs. survivors, respectively (p=0.04). ECMO (13% vs 25%, p<0.01), inhaled nitric oxide (11% vs 15%, p=0.02) and recruitment manoeauvres (26% vs 31%, p<0.01) were used less frequently in patients who did not survive. Independent risk factors associated with 28-day mortality included age older than 70 years (hazard ratio [HR], 2.83;95% CI, 1.32-6.07), higher creatinine levels upon ICU admission (HR, 1.20;95% CI, 1.03-1.40), and lower pH within 24h from commencement of MV (HR, 0.12;95% CI, 0.02-0.62), while a shorter period (day) from early symptoms to hospitalisation reduced mortality risks (HR, 0.96;95% CI, 0.93-0.99). Conclusions: Our findings from a large international cohort of critically-ill COVID-19 patients on mechanical ventilation emphasises that elderly patients, not promptly admitted to the hospital, and who present higher creatinine levels and acidosis are at higher risk of mortality.

Impact of immunomodulatory drugs on infectious complications in COVID-19 patients on extracorporeal membrane oxygenation: Analysis of the ELSO registry

Objective: To identify the incidence of infections in those receiving immunomodulatory drugs for COVID- 19 during ECMO and the risk factors for infection. Methods: Deidentified data on all patients who had ECMO for COVID-19 till July 2020 were analysed from the ELSO Registry. A comparison cohort of patients who did not acquire superinfections during ECMO was used to identify risk factors for infection. Our primary outcome measure was incidence of infections pre- or on ECMO in patients receiving immunomodulatory drugs. Univariate analysis assessed potential associations between survival and various pre-ECMO/ECMOrelated factors. Variables (p< 0.1) entered a logistic regression model which identified predictors of infections in this cohort. Results: Of the 1237 patients who required ECMO for COVID-19 related complications, 911 patients (73.6%) received immunomodulatory drugs. 47% of these patients had superinfections, predominantly with gram negative bacteria (56%). Pre-ECMO factors associated with a higher odds of infection included immunodeficiency and treatment with selective cytokine blockers. ECMO complications (mechanical, renal, pulmonary, infectious and metabolic) increased the odds of infection. (Table 1) Patients who developed an infection preor on ECMO had significantly longer ECMO runs than those who did not (491.1±308.9 hours vs 293.4± 240.6hours, p< 0.001) with no mortality difference (45.7% Vs 43.4%, p = 0.45). Conclusions: Of the three quarter of patients who received immunomodulatory drugs for COVID-19 during ECMO, 47% had superinfections. Immunodeficiency and use of selective cytokine blockers were risk factors for infections pre or on ECMO in addition to ECMO related complications.

Tracheostomy practices in patients with COVID-19 supported by VV-ECMO: A review of the ELSO registry

Objective: Extracorporeal membrane oxygenation (ECMO) use in COVID-19 has been supported by major health organizations and studies, but optimal management strategies need further research. We characterize and describe the tracheostomy practice in ECMO-supported patients with acute respiratory failure related to viral pneumonia without SARS-CoV-2 in 2019 and to COVID-19 in 2020. Methods: Analysis of the Extracorporeal Life Support Organization (ELSO) Registry including patients receiving respiratory ECMO support for COVID-19 in 2020 or other viral-induced respiratory failure in 2019. We compared tracheostomy practices between the 2019 and 2020 cohorts and reported outcomes of those that receive a tracheostomy on ECMO. Results: We identified 1960 patients who received VV-ECMO support for acute respiratory failure related to COVID-19. 57% of patients received a tracheostomy during the hospitalization, and 25% had a tracheostomy placed while on ECMO. The proportion of patients receiving a tracheostomy was similar in 2019 viral disease, but tracheostomies were performed 4 days earlier in 2019 compared to 2020 (median 6.7 days [IQR 3.0- 12.0 days] vs. 11.2 days [IQR 5.8-17.0 days], p< 0.01). More patients were mobilized in 2019 than 2020, but patients who received a tracheostomy on ECMO were more likely to be mobilized in both cohorts. 8% of patients were reported as having surgical site bleeding after a tracheostomy placement on ECMO in 2020. The median ECMO duration in those who received a tracheostomy on ECMO was 23.5 days [IQR 15.8-35.3]. Hospital mortality was 45% for patients with COVID- 19 (46% for patients who received a tracheostomy on ECMO). Conclusions: Tracheostomies are commonly performed in COVID-19 patients on ECMO. The number of tracheostomies performed on ECMO in the SARS-CoV-2 pandemic is similar to other viral diseases in 2019, however, they are performed later in the ECMO run. Patients who receive a tracheostomy are more likely to achieve mobilization.

Transition from VV to VA or hybrid ECMO: An ELSO Overview

Objective: The use of veno-venous extracorporeal membrane oxygenation (V-V ECMO) has increased significantly in the last years, and in particular in the recent pandemic COVID-19, representing a valuable therapy to treat severe respiratory failure. However, few patients remain hypoxemic and become hemodynamically instable during V-V ECMO support requiring a change in ECMO configuration. Methods: The Extracorporeal Life Support Organization Registry was reviewed for all cases of adult ARDS in patients undergoing ECMO requiring a change in configuration during the support run. (2017 to 2019). All main aspects, if available, characterizing the patients undergoing ECMO support modifications were collected and analyzed in terms of incidence, causes, patterns, complications and outcomes. Results: Of 9936 V-V ECMO runs, there were 354 ECMO that requiring a in changing configurations (3,5%). Over the study period the proportion of ECMO changing configuration was 1,3% (136) for VA, 1,6% (166) for V-VA and 0,8(85%) for Other configurations with an increasing in configuration changing during the years. The mortality rate in the shifting population was 55%. The new configurations were associated with major bleeding, stroke and renal failure was similar. Main determinants of poor outcome were the severity of the underlying illness, the delay in recognizing evolution of the disease, the complications associated with the conversion itself, and the more complex management of the hybrid setting. Conclusions: Our review of ARDS patients showed the variation in configuration VA ECMO or Hybrid ECMO was not associated with worse survival but with a high rates of complications rates compared with VV ECMO. These data suggest that in very selected patients it may reasonable to initially instituted VV ECMO support, reserving VA or Hybrid ECMO forms for conversion for refractory cardiopulmonary failure and cardiogenic shock that represent the major reasons of VV ECMO failure.

Evaluation of an online social learning environment in type 1 diabetes self-management education

Aims: Self-management education is an essential component of type 1 diabetes care which improves health outcomes. Online mediums of education delivery are growing in popularity, offering a flexible alternative to traditional group education. Interest in online education has increased as healthcare restructuring secondary to covid-19 has restricted face-to- face offerings. We delivered and evaluated a free open online course in type 1 diabetes self-management education. Methods: Understanding type 1 diabetes was a free open online course accessible via registration on an online platform. The course was delivered over a 2-day period in December 2020 and focused on key self-management concepts featuring videos, quizzes and moderated comment sections accompanied by a daily live Q&A session. The course was evaluated via a post-course questionnaire to establish user feedback and experience. Results: 216 users started the course, with 106 users completing >75% of the course steps (completion rate 49.1%). Among the 103 users who completed the optional post-course questionnaire, 69 had type 1 diabetes and 21 were healthcare professionals. Most users agreed that the course was easy to follow (95.1%), improved their knowledge of diabetes (76.2%), motivated them to manage their diabetes (79.7%) and improved their self-management confidence (79.4%). Thematic analysis of free-text responses identified appreciation of the peer-learning environment provided with minimal technical difficulties experienced. Conclusions: Open online courses offer a remotely accessible, engaging and scalable method of self-management education delivery that motivated and improved the self-management confidence of their user base.

Massive open online course for type 2 diabetes self-management: Adapting education in the COVID-19 era

Introduction: Type 2 diabetes self-management education is an essential component of type 2 diabetes care that is traditionally delivered in a face-to-face setting. In response to the recent COVID-19 pandemic, innovative solutions are urgently needed, allowing provision of self-management education that can be delivered in compliance with social distancing policies. Innovations that are self-service and can deliver education efficiently at low cost are particularly appealing to healthcare providers and commissioners. Methods: We aimed to evaluate user uptake, dropout, acceptability, satisfaction, perceived short-term knowledge gain and health benefits/behaviour changes in relation to a free massive open online course (MOOC) in diabetes self-management education, created and delivered during the COVID-19 pandemic. This course, focusing on addressing knowledge and self-management needs for people with type 2 diabetes, made use of online interactive content including expert and patient videos, quizzes, moderated discussion boards and live social media that encouraged personal reflection and goal setting. User expectations and experiences were explored via survey-based methods. Here, we present our experience of developing the course and describe users' experiences. Results: 1991 users registered interest in the course over a 2-week period, with 976 users starting the course and 640 (65.6%) users completing the course in full. Users engaged well, finding the course educational, user-friendly and motivating, demonstrating high completion rates and user satisfaction. A statistically significant (p<0.001) increase in self-reported self-management ability and health knowledge was observed among participants with type 2 diabetes. Discussion: MOOCs in type 2 diabetes self-management education have great potential for delivering education efficiently at scale and low cost. Although engagement can be limited by digital literacy, benefits include flexible and remote access to up-to-date, evidence-based education delivered by a multidisciplinary team of healthcare professionals. © 2020 Author(s) (or their employer(s)).