Association Between Hospital Type and Resilience During COVID-19 Caseload Stress : A Retrospective Cohort Study.

BACKGROUND: Imbalances between hospital caseload and care resources that strained U.S. hospitals during the pandemic have persisted after the pandemic amid ongoing staff shortages. Understanding which hospital types were more resilient to pandemic overcrowding-related excess deaths may prioritize patient safety during future crises. OBJECTIVE: To determine whether hospital type classified by capabilities and resources (that is, extracorporeal membrane oxygenation [ECMO] capability, multiplicity of intensive care unit [ICU] types, and large or small hospital) influenced COVID-19 volume-outcome relationships during Delta wave surges. DESIGN: Retrospective cohort study. SETTING: 620 U.S. hospitals in the PINC AI Healthcare Database. PARTICIPANTS: Adult inpatients with COVID-19 admitted July to November 2021. MEASUREMENTS: Hospital-months were ranked by previously validated surge index (severity-weighted COVID-19 inpatient caseload relative to hospital bed capacity) percentiles. Hierarchical models were used to evaluate the effect of log-transformed surge index on the marginally adjusted probability of in-hospital mortality or discharge to hospice. Effect modification was assessed for by 4 mutually exclusive hospital types. RESULTS: Among 620 hospitals recording 223 380 inpatients with COVID-19 during the Delta wave, there were 208 ECMO-capable, 216 multi-ICU, 36 large (≥200 beds) single-ICU, and 160 small (<200 beds) single-ICU hospitals. Overall, 50 752 (23%) patients required admission to the ICU, and 34 274 (15.3%) died. The marginally adjusted probability for mortality was 5.51% (95% CI, 4.53% to 6.50%) per unit increase in the log surge index (strain attributable mortality = 7375 [CI, 5936 to 8813] or 1 in 5 COVID-19 deaths). The test for interaction showed no difference (P = 0.32) in log surge index-mortality relationship across 4 hospital types. Results were consistent after excluding transferred patients, restricting to patients with acute respiratory failure and mechanical ventilation, and using alternative strain metrics. LIMITATION: Residual confounding. CONCLUSION: Comparably detrimental relationships between COVID-19 caseload and survival were seen across all hospital types, including highly advanced centers, and well beyond the pandemic's learning curve. These lessons from the pandemic heighten the need to minimize caseload surges and their effects across all hospital types during public health and staffing crises. PRIMARY FUNDING SOURCE: Intramural Research Program of the National Institutes of Health Clinical Center.

Statistical Power and Performance of Strategies to Analyze Composites of Survival and Duration of Ventilation in Clinical Trials.

BACKGROUND: Patients with acute hypoxemic respiratory failure are at high risk of death and prolonged time on the ventilator. Interventions often aim to reduce both mortality and time on the ventilator. Many methods have been proposed for analyzing these endpoints as a single composite outcome (days alive and free of ventilation), but it is unclear which analytical method provides the best performance. Thus, we aimed to determine the analysis method with the highest statistical power for use in clinical trials. METHODS: Using statistical simulation, we compared multiple methods for analyzing days alive and free of ventilation: the t, Wilcoxon rank-sum, and Kryger Jensen and Lange tests, as well as the proportional odds, hurdle-Poisson, and competing risk models. We compared 14 scenarios relating to: 1) varying baseline distributions of mortality and duration of ventilation, which were based on data from a registry of patients with acute hypoxemic respiratory failure and 2) the varying effects of treatment on mortality and duration of ventilation. RESULTS AND CONCLUSIONS: All methods have good control of type 1 error rates (i.e., avoid false positive findings). When data are simulated using a proportional odds model, the t test and ordinal models have the highest relative power (92% and 90%, respectively), followed by competing risk models. When the data are simulated using survival models, the competing risk models have the highest power (100% and 92%), followed by the t test and a ten-category ordinal model. All models struggled to detect the effect of the intervention when the treatment only affected one of mortality and duration of ventilation. Overall, the best performing analytical strategy depends on the respective effects of treatment on survival and duration of ventilation and the underlying distribution of the outcomes. The evaluated models each provide a different interpretation for the treatment effect, which must be considered alongside the statistical power when selecting analysis models.

Association Between Inability to Stand at ICU Discharge and Readmission: A Historical Cohort Study.

OBJECTIVES: The aim of this study was to determine if being unable to stand at ICU discharge was associated with an increased probability of ICU readmission. DESIGN: A multicenter retrospective cohort study was conducted using the Toronto Intensive Care Observational Registry (iCORE) project. SETTING: Nine tertiary academic ICUs in Toronto, Canada, affiliated with the University of Toronto. PATIENTS: All patients admitted to ICUs participating in iCORE from September 2014 to January 2020 were included. Patients had to be mechanically ventilated for more than 4 hours to be included in iCORE. Exclusion criteria were death during the initial ICU stay, transfer to another institution not included in iCORE at ICU discharge, and a short ICU stay defined as less than 2 days. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: The main exposure in this study was the inability of the patient to stand at ICU discharge, documented daily in the database within the ICU Mobility Scale. The primary outcome of this study was readmission to the ICU. After adjusting for potential confounders, being unable to stand at ICU discharge was associated with increased odds of readmission (odds ratio, 1.85; 95% CI, 1.31-2.62; p < 0.001). CONCLUSIONS: In patients with an ICU stay of 2 days or more, being unable to stand at ICU discharge is associated with increased odds of readmission to the ICU.

Understanding ventilator-induced lung injury: The role of mechanical power.

Mechanical ventilation stands as a life-saving intervention in the management of respiratory failure. However, it carries the risk of ventilator-induced lung injury. Despite the adoption of lung-protective ventilation strategies, including lower tidal volumes and pressure limitations, mortality rates remain high, leaving room for innovative approaches. The concept of mechanical power has emerged as a comprehensive metric encompassing key ventilator parameters associated with the genesis of ventilator-induced lung injury, including volume, pressure, flow, resistance, and respiratory rate. While numerous animal and human studies have linked mechanical power and ventilator-induced lung injury, its practical implementation at the bedside is hindered by calculation challenges, lack of equation consensus, and the absence of an optimal threshold. To overcome the constraints of measuring static respiratory parameters, dynamic mechanical power is proposed for all patients, regardless of their ventilation mode. However, establishing a causal relationship is crucial for its potential implementation, and requires further research. The objective of this review is to explore the role of mechanical power in ventilator-induced lung injury, its association with patient outcomes, and the challenges and potential benefits of implementing a ventilation strategy based on mechanical power.

Epidemiology, ventilation management and outcomes of COVID-19 ARDS patients versus patients with ARDS due to pneumonia in the Pre-COVID era.

BACKGROUND: Ventilation management may differ between COVID-19 ARDS (COVID-ARDS) patients and patients with pre-COVID ARDS (CLASSIC-ARDS); it is uncertain whether associations of ventilation management with outcomes for CLASSIC-ARDS also exist in COVID-ARDS. METHODS: Individual patient data analysis of COVID-ARDS and CLASSIC-ARDS patients in six observational studies of ventilation, four in the COVID-19 pandemic and two pre-pandemic. Descriptive statistics were used to compare epidemiology and ventilation characteristics. The primary endpoint were key ventilation parameters; other outcomes included mortality and ventilator-free days and alive (VFD-60) at day 60. RESULTS: This analysis included 6702 COVID-ARDS patients and 1415 CLASSIC-ARDS patients. COVID-ARDS patients received lower median VT (6.6 [6.0 to 7.4] vs 7.3 [6.4 to 8.5] ml/kg PBW; p < 0.001) and higher median PEEP (12.0 [10.0 to 14.0] vs 8.0 [6.0 to 10.0] cm H2O; p < 0.001), at lower median ΔP (13.0 [10.0 to 15.0] vs 16.0 [IQR 12.0 to 20.0] cm H2O; p < 0.001) and higher median Crs (33.5 [26.6 to 42.1] vs 28.1 [21.6 to 38.4] mL/cm H2O; p < 0.001). Following multivariable adjustment, higher ΔP had an independent association with higher 60-day mortality and less VFD-60 in both groups. Higher PEEP had an association with less VFD-60, but only in COVID-ARDS patients. CONCLUSIONS: Our findings show important differences in key ventilation parameters and associations thereof with outcomes between COVID-ARDS and CLASSIC-ARDS. TRIAL REGISTRATION: Clinicaltrials.gov (identifier NCT05650957), December 14, 2022.

Echocardiographic Findings in Critically Ill COVID-19 Patients Treated With and Without Extracorporeal Membrane Oxygenation.

OBJECTIVES: To describe echocardiographic findings among mechanically ventilated patients with COVID-19 acute respiratory distress syndrome, comparing those with and without venovenous extracorporeal membrane oxygenation (VV ECMO) support. DESIGN: Single-center, retrospective cohort study. SETTING: Intensive care unit (ICU) of a quaternary academic center. PARTICIPANTS: Patients with COVID-19 admitted between March 2020 and June 2021 receiving mechanical ventilation, with an echocardiogram within 72 hours of admission. INTERVENTIONS: Admission and follow-up echocardiograms during ICU stay. MEASUREMENTS: Patient characteristics and echocardiographic findings were analyzed. Mortality odds ratio (OR) for right ventricular (RV) systolic dysfunction and acute cor pulmonale (ACP) was calculated. MAIN RESULTS: Among 242 patients, 145 (60%) received VV ECMO. Median (IQR) PaO2/FiO2 was 76 (65-95) and 98 (85-140) in ECMO and non-ECMO patients, respectively (p ≤ 0.001). Initial echocardiograms showed no significant differences in left ventricular systolic dysfunction (10% v 15 %, p = 0.31) and RV systolic dysfunction (38% v. 27%, p = 0.27) between ECMO and non-ECMO patients. ACP was more frequent in the ECMO group at baseline (41% v. 26 %, p = 0.02). During the ICU stay, patients on ECMO exhibited a higher prevalence of RV systolic dysfunction (55% v 34%, p = 0.001) and ACP (51% v 26%, p = 0.002). RV systolic dysfunction (OR 1.99; 95% CI 1.09-3.63) and ACP (OR 2.95; 95% CI 1.55-5.62) on the follow-up echocardiograms were associated with higher odds of ICU mortality. CONCLUSIONS: The prevalence of echocardiographic abnormalities, in particular RV dysfunction, was frequent among patients with COVID-19 receiving VV ECMO support and was associated with worse clinical outcomes.

Patient education and extracorporeal membrane oxygenation preferences of patients and providers in COVID care.

BACKGROUND: Extracorporeal membrane oxygenation (ECMO) represents an important but limited treatment for patients with severe COVID-19. We assessed the effects of an educational intervention on a person's ECMO care preference and examined whether patients and providers had similar ECMO preferences. METHODS: In the Video+Survey group, patients watched an educational video about ECMO's purpose, benefits, and risks followed by an assessment of ECMO knowledge and care preferences in seven scenarios varying by hypothetical patient age, function, and comorbidities. Patients in the Survey Only group and providers didn't watch the video. Logistic regression was used to estimate the probability of agreement for each ECMO scenario between the two patient groups and then between all patients and providers. RESULTS: Video+Survey patients were more likely (64% vs. 17%; p = 0.02) to correctly answer all ECMO knowledge questions than Survey Only patients. Patients in both groups agreed that ECMO should be considered across all hypothetical scenarios, with predicted agreement above 65%. In adjusted analyses, patients and providers had similar predicted agreement for ECMO consideration across six of the seven scenarios, but patients showed greater preference (84% vs. 41%, p = 0.003) for the scenario of a functionally dependent 65-year-old with comorbidities than providers. DISCUSSION AND CONCLUSIONS: An educational video increased a person's ECMO knowledge but did not change their ECMO preferences. Clinicians were less likely than patients to recommend ECMO for older adults, so advanced care planning discussion between patients and providers about treatment options in critically ill patients with COVID-19 is critical.

Resource Utilization and Costs Associated With Cardiogenic Shock Complicating Myocardial Infarction: A Population-Based Cohort Study.

Background: Cardiogenic shock due to acute myocardial infarction (AMI-CS) is associated with significant short- and long-term morbidity and mortality. Despite this, little is known about associated cost. Objectives: The purpose of this study was to evaluate the health care costs and resource use associated with AMI-CS using administrative data from the province of Ontario, Canada. Methods: This was a retrospective cohort study of adult patients with AMI-CS from April 2009 to March 2019. One-year costs following index admission were reported at an individual level. We used generalized linear models to identify factors associated with increased cost. We stratified patients by revascularization strategy to compare cost in each group and examined total cost at a patient level per individual fiscal year. Results: We included 9,789 consecutive patients with AMI-CS across 135 centers in Ontario (mean age 70.5 years; 67.7% male). Mortality in-hospital was 30.2%, and mortality at 2 years was 45.9%. The median inpatient cost per patient was $23,912 (IQR: $12,234-$41,833) with a median total 1-year cost of $37,913 (IQR: $20,113-$66,582). The median 1-year cost was $17,730 (IQR: $9,323-$38,379) for those who died in hospital, and $45,713 (IQR: $29,688-$77,683) for those surviving to discharge, with $12,719 (IQR: $4,262-$35,275) occurring after discharge. Patients who received coronary artery bypass grafting incurred the highest cost among revascularization groups. No significant differences were observed in cost per fiscal year from 2009 to 2019. Conclusions: AMI-CS is associated with significant health care costs, both during the index hospitalization and following discharge. To optimize cost-effectiveness, future therapies should aim to reduce disability in addition to improving mortality.

Comparison of venovenous extracorporeal membrane oxygenation, prone position and supine mechanical ventilation for severely hypoxemic acute respiratory distress syndrome: a network meta-analysis.

PURPOSE: Severe acute respiratory distress syndrome (ARDS) with PaO2/FiO2 < 80 mmHg is a life-threatening condition. The optimal management strategy is unclear. The aim of this meta-analysis was to compare the effects of low tidal volumes (Vt), moderate Vt, prone ventilation, and venovenous extracorporeal membrane oxygenation (VV-ECMO) on mortality in severe ARDS. METHODS: We performed a frequentist network meta-analysis of randomised controlled trials (RCTs) with participants who had severe ARDS and met eligibility criteria for VV-ECMO or had PaO2/FiO2 < 80 mmHg. We applied the Grades of Recommendation, Assessment, Development, and Evaluation (GRADE) methodology to discern the relative effect of interventions on mortality and the certainty of the evidence. RESULTS: Ten RCTs including 812 participants with severe ARDS were eligible. VV-ECMO reduces mortality compared to low Vt (risk ratio [RR] 0.77, 95% confidence interval [CI] 0.59-0.99, moderate certainty) and compared to moderate Vt (RR 0.75, 95% CI 0.57-0.98, low certainty). Prone ventilation reduces mortality compared to moderate Vt (RR 0.78, 95% CI 0.66-0.93, high certainty) and compared to low Vt (RR 0.81, 95% CI 0.63-1.02, moderate certainty). We found no difference in the network comparison of VV-ECMO compared to prone ventilation (RR 0.95, 95% CI 0.72-1.26), but inferences were based solely on indirect comparisons with very low certainty due to very wide confidence intervals. CONCLUSIONS: In adults with ARDS and severe hypoxia, both VV-ECMO (low to moderate certainty evidence) and prone ventilation (moderate to high certainty evidence) improve mortality relative to low and moderate Vt strategies. The impact of VV-ECMO versus prone ventilation remains uncertain.

Extended Versus Standard Proning Duration for COVID-19 Associated Acute Respiratory Distress Syndrome: A Target Trial Emulation Study.

RATIONALE: Prone positioning for > 16 hours in moderate-to-severe acute respiratory distress syndrome (ARDS) improves survival. However, the optimal duration of proning is unknown. OBJECTIVE: To estimate the effect of extended versus standard proning duration on patients with moderate-to-severe COVID-19 ARDS. METHODS: Data were extracted from a 5-hospital electronic medical record registry. Patients who were proned within 72 hours of mechanical ventilation were categorized as receiving extended (> 24 hours) versus standard (16-24 hours) proning based on the first proning session length. We used a target trial emulation design to estimate the effect of extended vs. standard proning on the primary outcome of 90-day mortality, and secondary outcomes of ventilator liberation and ICU discharge. Analytically we used inverse probability of treatment weighted (IPTW) Cox, or Fine and Gray regression models. RESULTS: 314 patients were included, 234 who received extended proning, and 80 who received standard duration. Extended proning patients were older, had greater comorbidity, were more often at an academic hospital, and had shorter time from admission to mechanical ventilation. After IPTW, characteristics were well balanced. Unadjusted 90-day mortality in the extended vs. standard proning groups was 39% vs 58%. In doubly-robust IPTW analyses, we found no significant effects of extended vs. standard proning duration on mortality (hazard ratio [95% CI] 0.95 [0.51-1.77]), ventilator liberation (sub-distribution hazard [sHR] 1.60, [0.97-2.64], or ICU discharge (sHR 1.31 [0.82-2.10]). CONCLUSION: Using target trial emulation, we found no significant effect of extended vs. standard proning duration on mortality, ventilator liberation, or ICU discharge. However, given the imprecision of estimates, further study is justified. This article is open access and distributed under the terms of the Creative Commons Attribution Non-Commercial No Derivatives License 4.0 (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Mental health sequelae in survivors of cardiogenic shock complicating myocardial infarction. A population-based cohort study.

PURPOSE: Cardiogenic shock secondary to acute myocardial infarction (AMI-CS) is associated with substantial short- and long-term morbidity and mortality. However, there are limited data on mental health sequelae that survivors experience following discharge. METHODS: We conducted a retrospective, population-based cohort study in Ontario, Canada of critically ill adult (≥ 18 years) survivors of AMI-CS, admitted to hospital between April 1, 2009 and March 31, 2019. We compared these patients to AMI survivors without shock. We captured outcome data using linked health administrative databases. The primary outcome was a new mental health diagnosis (a composite of mood, anxiety, or related disorders; schizophrenia/psychotic disorders; and other mental health disorders) following hospital discharge. We secondarily evaluated incidence of deliberate self-harm and death by suicide. We compared patients using overlap propensity score-weighted, cause-specific proportional hazard models. RESULTS: We included 7812 consecutive survivors of AMI-CS, from 135 centers. Mean age was 68.4 (standard deviation (SD) 12.2) years, and 70.3% were male. Median follow-up time was 767 days (interquartile range (IQR) 225-1682). Incidence of new mental health diagnosis among AMI-CS survivors was 109.6 per 1,000 person-years (95% confidence interval (CI) 105.4-113.9), compared with 103.8 per 1000 person-years (95% CI 102.5-105.2) among AMI survivors without shock. After propensity score adjustment, there was no difference in the risk of new mental health diagnoses following discharge [hazard ratio (HR) 0.99 (95% CI 0.94-1.03)]. Factors associated with new mental health diagnoses following AMI-CS included female sex, pre-existing mental health diagnoses, and discharge to a long-term hospital or rehabilitation institute. CONCLUSION: Survivors of AMI-CS experience substantial mental health morbidity following discharge. Risk of new mental health diagnoses was comparable between survivors of AMI with and without shock. Future research on interventions to mitigate psychiatric sequelae after AMI-CS is warranted.

Physical Rehabilitation and Mobilization in Patients Receiving Extracorporeal Life Support: A Systematic Review.

OBJECTIVES: We planned to synthesize evidence examining the potential efficacy and safety of performing physical rehabilitation and/or mobilization (PR&M) in adult patients receiving extracorporeal life support (ECLS). DATA SOURCES: We included any study that compared PR&M to no PR&M or among different PR&M strategies in adult patients receiving any ECLS for any indication and any cannulation. We searched seven electronic databases with no language limitations. STUDY SELECTION AND DATA EXTRACTION: Two reviewers, independently and in duplicate, screened all citations for eligibility. We used the Cochrane Risk of Bias 2 and Cochrane Risk Of Bias In Non-randomized Studies of Interventions tools to assess individual study risk of bias. Although we had planned for meta-analysis, this was not possible due to insufficient data, so we used narrative and tabular data summaries for presenting results. We assessed the overall certainty of the evidence for each outcome using the Grading of Recommendations Assessment, Development, and Evaluation framework. DATA SYNTHESIS: We included 17 studies that enrolled 996 patients. Most studies examined venovenous extracorporeal membrane oxygenation (ECMO) and/or venoarterial ECMO as a bridge to recovery in the ICU. We found an uncertain effect of high-intensity/active PR&M on mortality, duration of mechanical ventilation, ICU length of stay, hospital length of stay, or quality of life compared with low-intensity/passive PR&M in patients receiving ECLS (very low certainty due to very serious imprecision). There was similarly an uncertain effect on safety events including clinically important bleeding, spontaneous intracerebral hemorrhage, limb ischemia, accidental decannulation, or ECLS circuit dysfunction (very low certainty due to very serious risk of bias and imprecision). CONCLUSIONS: Based on the currently available summary of evidence, there is an uncertain effect of high-intensity/active PR&M on patient important outcomes or safety in patients receiving ECLS. Despite indirect data from other populations suggesting potential benefit of high-intensity PR&M in the ICU; further high-quality randomized trials evaluating the benefits and risks of physical therapy and/or mobilization in this population are needed.

Designing a Bayesian adaptive clinical trial to evaluate novel mechanical ventilation strategies in acute respiratory failure using integrated nested Laplace approximations.

BACKGROUND: Adaptive trials usually require simulations to determine values for design parameters, demonstrate error rates, and establish the sample size. We designed a Bayesian adaptive trial comparing ventilation strategies for patients with acute hypoxemic respiratory failure using simulations. The complexity of the analysis would usually require computationally expensive Markov Chain Monte Carlo methods but this barrier to simulation was overcome using the Integrated Nested Laplace Approximations (INLA) algorithm to provide fast, approximate Bayesian inference. METHODS: We simulated two-arm Bayesian adaptive trials with equal randomization that stratified participants into two disease severity states. The analysis used a proportional odds model, fit using INLA. Trials were stopped based on pre-specified posterior probability thresholds for superiority or futility, separately for each state. We calculated the type I error and power across 64 scenarios that varied the probability thresholds and the initial minimum sample size before commencing adaptive analyses. Two designs that maintained a type I error below 5%, a power above 80%, and a feasible mean sample size were evaluated further to determine the optimal design. RESULTS: Power generally increased as the initial sample size and the futility threshold increased. The chosen design had an initial recruitment of 500 and a superiority threshold of 0.9925, and futility threshold of 0.95. It maintained high power and was likely to reach a conclusion before exceeding a feasible sample size. CONCLUSIONS: We designed a Bayesian adaptive trial to evaluate novel strategies for ventilation using the INLA algorithm to efficiently evaluate a wide range of designs through simulation.

Head-to-toe bedside ultrasound for adult patients on extracorporeal membrane oxygenation.

Bedside ultrasound represents a well-suited diagnostic and monitoring tool for patients on extracorporeal membrane oxygenation (ECMO) who may be too unstable for transport to other hospital areas for diagnostic tests. The role of ultrasound, however, starts even before ECMO initiation. Every patient considered for ECMO should have a thorough ultrasonographic assessment of cardiac and valvular function, as well as vascular anatomy without delaying ECMO cannulation. The role of pre-ECMO ultrasound is to confirm the indication for ECMO, identify clinical situations for which ECMO is not indicated, rule out contraindications, and inform the choice of ECMO configuration. During ECMO cannulation, the use of vascular and cardiac ultrasound reduces the risk of complications and ensures adequate cannula positioning. Ultrasound remains key for monitoring during ECMO support and troubleshooting ECMO complications. For instance, ultrasound is helpful in the assessment of drainage insufficiency, hemodynamic instability, biventricular function, persistent hypoxemia, and recirculation on venovenous (VV) ECMO. Lung ultrasound can be used to monitor signs of recovery on VV ECMO. Brain ultrasound provides valuable diagnostic and prognostic information on ECMO. Echocardiography is essential in the assessment of readiness for liberation from venoarterial (VA) ECMO. Lastly, post decannulation ultrasound mainly aims at identifying post decannulation thrombosis and vascular complications. This review will cover the role of head-to-toe ultrasound for the management of adult ECMO patients from decision to initiate ECMO to the post decannulation phase.

Venovenous Extracorporeal Membrane Oxygenation Candidacy Decision-Making: Lessons and Hypotheses From a Single-Center Observational Analysis.

BACKGROUND: Use of venovenous extracorporeal membrane oxygenation (ECMO) is increasing, but candidacy selection processes are variable and subject to bias. RESEARCH QUESTION: What are the reasons behind venovenous ECMO candidacy decisions, and are decisions made consistently across patients? STUDY DESIGN AND METHODS: Prospective observational study of all patients, admitted or outside hospital referrals, considered for venovenous ECMO at a tertiary referral center. Relevant clinical data and reasons for candidacy determination were cross-referenced with other noncandidates and candidates and were assessed qualitatively. RESULTS: Eighty-one consultations resulted in 44 noncandidates (54%), 29 candidates (36%; nine of whom subsequently underwent cannulation), and eight deferred decisions (10%). Fifteen unique contraindications were identified, variably present across all patients. Five contraindications were invoked as the sole reason to deny ECMO to a patient. In patients with three or more contraindications, additional contraindications were cited even if the severity was relatively minor. All but four contraindications invoked to deny ECMO to a patient were nonprohibitive for at least one other candidate. Contraindications documented in noncandidates were present but not mentioned in 21 other noncandidates (47%). Twenty-six candidates (90%) had at least one contraindication that was prohibitive in a noncandidate, including a contraindication that was the sole reason to deny ECMO. Contraindications were proposed as informing three prognostic domains, through which patterns of inconsistency could be understood better: (1) irreversible underlying pulmonary process, (2) unsurvivable critical illness, and (3) clinical condition too compromised for meaningful recovery. INTERPRETATION: ECMO candidacy decisions are inconsistent. We identified four patterns of inconsistency in our center and propose a three-domain model for understanding and categorizing contraindications, yielding five lessons that may improve candidacy decision processes until further research can guide practice more definitively.

Age and associated outcomes among patients receiving venovenous extracorporeal membrane oxygenation for acute respiratory failure: analysis of the Extracorporeal Life Support Organization registry.

PURPOSE: Venovenous extracorporeal membrane oxygenation (VV-ECMO) can be used to support patients with refractory acute respiratory failure, though guidance on patient selection is lacking. While age is commonly utilized as a factor in establishing the potential VV-ECMO candidacy of these patients, little is known regarding its association with outcome. We studied the association between increasing patient age and outcomes among patients with acute respiratory failure receiving VV-ECMO. METHODS: In this registry-based cohort study, we used individual patient data from 144 centres. We included adult patients (≥ 18 years of age) receiving VV-ECMO from 2017 to 2022. The primary outcome was hospital mortality. Secondary outcomes included a composite of complications following initiation of VV-ECMO. We conducted Bayesian analyses to estimate the association between chronological age and outcomes. RESULTS: We included 27,811 patients receiving VV-ECMO. Of these, 11,533 (41.5%) died in hospital. For the analysis conducted using weakly informed priors, and as compared to the reference category of age 18-29, the age brackets of 30-39 (odds ratio [OR] 1.17, 95% credible interval [CrI] 1.06-1.31), 40-49 (OR 1.65, 95% CrI 1.49-1.82), 50-59 (OR 2.39, 95% CrI 2.16-2.61), 60-69 (OR 3.29, 95% CrI 2.97-3.67), 70-79 (OR 4.57, 95% CrI 3.90-5.37), and ≥ 80 (OR 8.08, 95% CrI 4.85-13.74) were independently associated with increasing hospital mortality. Similar results were found between increasing age and post-ECMO complications. CONCLUSIONS: Among patients receiving VV-ECMO for acute respiratory failure, increasing age is significantly associated with poorer outcomes, and this association emerges as early as 30 years of age.