How much ECMO is enough? Center experience, learning curves, run durations, and outcomes in venovenous ECMO: An ELSO registry analysis

How much ECMO is enough? Center experience, learning curves, run durations, and outcomes in venovenous ECMO: An ELSO registry analysis | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article How much ECMO is enough? Center experience, learning curves, run durations, and outcomes in venovenous ECMO: An ELSO registry analysis Adam Green, Abhimanyu Chandel, Clifford Chang, Nitin Puri, Christopher Noel, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8770342/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Purpose Expanding availability of venovenous ECMO (VV ECMO) and increasing frequency of prolonged ECMO runs challenge standard definitions of center experience based solely on case counts. We examined the relationship between center experience and patient outcome using annual VV ECMO-day-volume (total days of VV ECMO support per center-year), a metric that integrates both procedural frequency and exposure to prolonged support and may better reflect institutional expertise. Methods We examined adult patients in the Extracorporeal Life Support Organization (ELSO) Registry supported with VV ECMO between 2014 and 2024. The association between center experience, in-hospital mortality, and ECMO-related complications were examined using mixed-effect models adjusted for illness severity, calendar year, and clustering within centers. Results Among 54,036 patients, center experience was independently associated with lower in-hospital mortality and fewer ECMO-related complications. Each doubling of annual VV ECMO-days was associated with a 4% reduction in the odds of death (OR 0.96; 95% CI, 0.93–0.98). The association was most pronounced among centers with longer average ECMO run duration (> 18 days; OR 0.95; 95% CI, 0.91–0.99), whereas no significant volume-mortality association was observed in centers with shorter average run duration. Center learning was observed to increase quickly and plateau as experience is gained, allowing for newly established centers to rapidly achieve expertise. Conclusion Higher VV ECMO-day-volume is associated with lower in-hospital mortality and fewer ECMO-related complications. This metric incorporates frequency and duration of support and provides a more comprehensive measure of center experience than case counts alone. Extracorporeal Membrane Oxygenation Extracorporeal Life Support Organization Registry Volume-Outcome Relationship Center Experience Figures Figure 1 Figure 2 Figure 3 Introduction Venovenous extracorporeal membrane oxygenation (VV ECMO) is an established and expanding supportive tool for patients with severe, potentially reversible respiratory failure. As VV ECMO utilization has expanded, understanding institutional factors associated with improved outcomes has become increasingly important. A positive correlation between center volume and outcomes has been observed in other areas, including post-cardiac arrest care, postoperative recovery, and mechanical ventilation. 1 , 2 In the current literature, ECMO center volume is most often characterized by the number of separate ECMO cases per year, regardless of the run duration. Analyses of the ELSO Registry and national datasets have shown lower risk-adjusted mortality at centers performing a greater number of ECMO cases annually. 3 – 8 The association between center volume and improved outcomes is likely multifactorial and includes patient selection, timing of ECMO initiation, and institutional capabilities. 4 , 5 While the number of cases per year (ECMO-case-volume) is an intuitive and reproducible metric, it does not account for substantial heterogeneity in ECMO run duration between centers. Contemporary adult VV ECMO increasingly involves prolonged support, reflecting changes in patient selection, improvements in supportive care, and greater use of ECMO as a bridge to recovery or transplantation. 9 Previous ELSO-based analyses have demonstrated that prolonged ECMO support is not independently associated with mortality, suggesting that experience with sustained ECMO care can be accrued without proportional increases in risk. 9 Therefore, cumulative exposure to ECMO management, reflected by the total number of days of support per center-year (ECMO-day-volume), may better capture institutional experience. Annual ECMO-day-volume incorporates the frequency and duration of care, reflects systems-level expertise exercised daily during ECMO support, and avoids misclassifying centers with fewer but longer runs as inexperienced centers. Prior studies have not evaluated annual ECMO-day-volume as a center-level exposure. Accordingly, we evaluated the association between center-level VV ECMO experience, defined by annual ECMO-day-volume, and patient outcomes in adult patients supported by VV ECMO. We hypothesized that higher annual ECMO-day-volume would be associated with lower in-hospital mortality and fewer ECMO-related complications, independent of patient-level severity of illness and secular trends. By reframing ECMO experience in terms of cumulative exposure rather than case counts alone, this study aims to refine the understanding of volume–outcome relationships in adult VV ECMO. Methods Study Design and Population We conducted a retrospective cohort study using data from the ELSO Registry. Established in 1989, the ELSO Registry is the world’s largest, comprehensive database of patient data on patients receiving ECMO with 780 active centers and data on 270,960 cases ( https://www.elso.org/registry.aspx ). Adult patients (≥ 18 years) supported with VV ECMO between 2014 and 2024 were included. ECMO support was analyzed at the run level, with patients clustered within ECMO centers. The ECMO run was the unit of analysis for all primary models. Exposure: Center experience was quantified using annual center-level VV ECMO-day-volume, defined as the total number of days of VV ECMO support delivered by a center within a given calendar year. To facilitate interpretability and account for right-skewness, annual ECMO-day-volume was log-transformed and parameterized per doubling of exposure. As a sensitivity analysis, center experience was alternatively defined using annual ECMO-case-volume, modeled using the same framework. Outcomes The primary outcome was in-hospital mortality. Secondary outcomes included the overall ECMO-related complication rate and subtype-specific complication rates. Complications were identified using ELSO-defined complication codes and were considered present only if recorded as active during the ECMO run. Covariates Patient-level severity of illness was adjusted using the Respiratory Extracorporeal Membrane Oxygenation Survival Prediction (RESP) score (a composite score of disease severity comprising prognostically significant components including age, primary diagnosis, pre-ECLS management, and comorbidities), constructed from available ELSO variables according to published methodology. 10 Calendar year of ECMO initiation was included to account for secular trends in ECMO practice and outcomes. COVID-19 status was included as a binary variable and evaluated as a potential effect modifier of the volume–outcome relationship. Statistical Analysis The association between center VV ECMO experience and in-hospital mortality was evaluated using hierarchical logistic regression models with a random intercept for center to account for clustering of runs within centers. The primary model included annual ECMO-day-volume (per doubling), RESP score, calendar year, COVID-19 status, and an interaction term between center volume and COVID-19 status. Adjusted odds ratios (ORs) with 95% confidence intervals (CIs) were reported. ECMO-related complications were modeled as rates per ECMO-day using mixed-effects negative binomial regression, with a random intercept for center and an offset for log-transformed ECMO duration to account for differences in time at risk. Incidence rate ratios (IRRs) with 95% CIs were reported. Exploratory and Sensitivity Analyses Exploratory analyses examined whether experience with longer ECMO runs modified the volume–outcome relationship by incorporating center-level average ECMO run duration, calculated annually as total ECMO-days divided by run count. Additional exploratory analyses evaluated within-center learning by modeling cumulative prior ECMO-day-volume among centers initiating VV ECMO after 2014. Sensitivity analyses assessed alternative exposure definitions using ECMO-case-volume, evaluated nonlinearity using restricted cubic splines, and addressed missing RESP scores using indicator-based adjustment and inverse probability weighting. Additional details regarding the above methodology, handling of missing data, and model diagnostics are provided in the online data supplement. Ethical Considerations The ELSO Registry contains de-identified data. This study was conducted in accordance with institutional policies governing secondary analyses of de-identified data and was considered exempt from human subject review. Results Study population Between 2014 and 2024, a total of 54,036 adult patients supported with VV ECMO were recorded in the ELSO Registry across 697 contributing centers and were included for analysis. Baseline patient characteristics stratified by center ECMO-day-volume (quartile [Q1-Q4] of annual ECMO-days) are included in Table 1 . Overall, patients treated at lower- and higher-volume centers, defined based on annual ECMO-day-volume, were broadly similar with respect to age, gender, and body mass index. However, several differences in case composition were observed across volume strata. Patients supported at centers with higher day-volume were less likely to have markers of advanced physiologic vulnerability at the time of VV ECMO initiation, including a lower prevalence of pre-ECMO cardiac arrest (Q4 vs. Q1: 6.2% vs. 10%), but were more likely to have incorporated advanced non-ECMO acute respiratory distress syndrome (ARDS) therapies prior to cannulation. These centers had higher rates of prone positioning (32% vs. 17%), neuromuscular blockade (56% vs. 48%), and inhaled pulmonary vasodilator use (24% vs. 20%), as well as longer durations of mechanical ventilation prior to VV ECMO initiation (49 vs. 31 hours). Despite a longer pre-ECMO course, patients cannulated at centers with higher day-volume were less likely to require vasopressors (54% vs. 61%) or bicarbonate (7.4% vs. 13%). Finally, the proportion of patients with COVID-19–associated respiratory failure was substantially higher among centers with higher annual ECMO-day-volume (53% vs. 12%). These trends were similar when baseline characteristics were stratified by annual ECMO cases per year (eTable 1). Among all patients, 21,949/54,036 (40.6%) died (Q4 vs. Q1: 40.7% vs. 39.3%). A comparison of observed mortality by quartile of annual ECMO-day-volume stratified by COVID-19 status is found in Fig. 1 and eTable 2. Unadjusted survival was significantly higher in high day-volume centers (p < 0.001) in both strata. Lung transplantation was performed in 3,302 (6.1%) (Q4: N = 866 [6.4%] vs. Q1: N = 627 [4.6%]) patients. Survivors were most commonly discharged home (11,867/31,106; 38.2%) or to long-term acute care or rehabilitation facilities (9,334/31,106; 30.0%) (eTable3). Center VV ECMO-day-volume and in-hospital mortality In hierarchical logistic regression models adjusted for RESP score, calendar year, and center-level clustering, higher annual VV ECMO-day-volume was independently associated with lower in-hospital mortality (Table 2 ). Among patients without COVID-19, each doubling of annual VV ECMO-days was associated with a 4% reduction in the odds of death (OR 0.96; 95% CI, 0.93–0.98; p = 0.001). The protective effect of center day-volume was found to be stronger in patients with COVID-19 than in patients supported with VV ECMO for other reasons (eFigure 1). Among patients with COVID-19, each doubling of annual VV ECMO-day-volume was associated with an approximately 11% (OR volume [0.96] * OR interaction [0.93] = 0.89) reduction in mortality odds. Nonlinearity in the association between center day-volume and mortality was assessed using restricted cubic splines. Spline-based models did not demonstrate a statistically significant departure from linearity (likelihood ratio test p > 0.05). ECMO-related complications ECMO-related complications stratified by ECMO-day-volume are displayed in eTable 4. Total complications were more common in centers with higher day-volume. However, when rate of complications was considered per ECMO-day, higher center VV ECMO-day-volume was associated with lower rates of ECMO-related complications. In mixed-effects negative binomial models with an offset for ECMO duration, each doubling of annual VV ECMO-day-volume was associated with a lower overall complication rate (across all ECMO complications) per ECMO-day (Fig. 2). Average VV ECMO run duration and the volume–outcome relationship The median run duration in the cohort was 15 days (IQR: 10–21 days), and centers with higher VV ECMO-day-volume had higher median run duration (Q4 vs. Q1: 25 vs. 10 days; p < 0.001). To explore whether experience managing longer VV ECMO runs modified the volume–outcome relationship, centers were stratified by annual average VV ECMO run duration (short, intermediate, and long [by tertile of run duration]) (eTable 5). In stratified analyses, the association between center VV ECMO-day-volume and mortality was most pronounced in centers with longer average run durations (Fig. 3). In centers with long average VV ECMO runs, each doubling of annual VV ECMO-day-volume was associated with a significant reduction in mortality (OR 0.95; 95% CI, 0.91–0.99) (eFigure 2), whereas no significant association was observed in centers with shorter or intermediate run durations. These findings suggest that prolonged VV ECMO support may be a modifier of the volume–outcome relationship. Within-center learning and cumulative experience To evaluate whether outcomes improve as an individual center gains experience over time, we performed a within-center learning analysis restricted to centers that began recording VV ECMO cases after 2014. After adjustment for contemporaneous annual volume, RESP score, and calendar year; each doubling of cumulative prior VV ECMO-day-volume at a given center was associated with a 3% lower odds of in-hospital mortality (OR 0.97; 95% CI, 0.95–0.99), consistent with a center-level learning effect over time that increases quickly while plateauing (after approximately 100 cumulative VV ECMO-days) as experience is gained (eFigure 3). Considerable heterogeneity in learning trajectories was observed across centers, indicating that, although outcomes generally improved with experience, the rate at which centers accrued benefits varied. Sensitivity analyses Results were robust across multiple sensitivity analyses. Defining center volume by ECMO-case-volume yielded qualitatively similar associations with mortality (eTable 6). However, the model using annual VV ECMO-day-volume demonstrated superior predictive performance compared with the model using annual case-volume, with a higher cross-validated log-likelihood and a lower mean Brier score (0.221 vs. 0.229), indicating improved out-of-sample fit and calibration when considering annual ECMO-day-volume. Likewise, when considering center experience based on ECMO-case-volume, the complication rate was significantly lower for higher case-volume centers (eFigure 4). To address missing RESP scores, both indicator-based adjustment and inverse probability weighting produced effect estimates consistent with the primary analysis (eTables 7 and 8). Together, these findings support the robustness of the observed associations between ECMO-day-volume, mortality, and complications. Discussion In this large, contemporary analysis of the international ELSO Registry, we found that higher center VV ECMO experience, quantified by annual ECMO-day-volume, was independently associated with lower in-hospital mortality and fewer ECMO-related complications. This association persisted after adjustment for patient severity of illness, calendar year, and center-level clustering. A previous ELSO registry analysis, defining experience as annual case-volume, likewise demonstrated lower mortality at high-volume centers. 5 However, this analysis was conducted more than a decade ago, combined venoarterial (VA) and VV ECMO, and preceded the COVID-19 pandemic. Since this study, ECMO practices have evolved, with an increase in prolonged VV ECMO runs. 9 Accordingly, we re-examined the volume–outcome relationship using annual ECMO-day-volume as the measure of center experience, a metric that more fully captures cumulative exposure to ECMO management over time. By restricting the analysis to VV ECMO, we aimed to create a more homogeneous population. We propose two main reasons for these findings: optimal patient selection and the acquisition of clinical expertise. Although baseline demographic characteristics and total RESP score were similar across volume-strata, differences in case composition and distribution of RESP components were evident. Centers with higher day-volume cannulated fewer patients with immunocompromised states, central nervous system dysfunction, and pre-ECMO cardiac arrest, suggesting a focus on patient selection. 11 , 12 Likewise, high day-volume centers appear to prioritize ARDS care prior to cannulation, as evidenced by a higher percentage of patients receiving neuromuscular blockers and prone ventilation. Additionally, a longer interval between intubation and ECMO cannulation was observed at high day-volume centers, likely an attempt to optimize medical management before resorting to ECMO. 13 Despite this, fewer patients required vasopressors or bicarbonate at the time of cannulation at the high day-volume centers. Collectively, these findings suggest experienced centers employ more selective and nuanced patient-selection strategies and may be more adept at identifying patients likely to benefit from ECMO. In addition to patient selection, evidence of clinical expertise at high ECMO-day-volume centers was also observed. First, higher day-volume centers experienced fewer ECMO-related complications. There were fewer complications observed per ECMO day in all complication types as day-volume increases. When all complications are considered, there was a 14% reduction in complications per doubling of ECMO day-volume (IRR 0.86; 95% CI, 0.85–0.88). This is likely due to a variety of reasons: provider experience, established care protocols, ECMO dedicated resources, and the availability of ancillary services in this specific population - all of these factors are influenced by increased ECMO volume. The consistency of these findings across multiple complication types suggests that improved technical and clinical management, rather than patient selection alone contributes to better outcomes. Further supporting a component of clinical expertise, the volume-outcome relationship was particularly strong in patients supported with VV ECMO for COVID-19. In these patients, each doubling of annual ECMO-day-volume was linked to approximately a 11% decrease in in-hospital mortality. This reflects the capacity of high-volume hospitals to provide robust ECMO services even during periods of significant resource strain. Centers with longer ECMO runs showed a stronger link between volume and lower mortality, unlike centers with shorter runs. Experience from managing prolonged ECMO, not just higher case numbers, seems key to better outcomes. Longer runs expose teams to more complications and learning opportunities, making ECMO-day-volume a better measure than annual case volume. Importantly, in at least one year during the period analyzed, 5.7% (40/697) of centers would have been classified in the lower half of experience based on annual ECMO-case-volume but would be placed in the top third of centers based on annual ECMO-day-volume. This discordance highlights that a small but meaningful number of centers may be misclassified based on case count alone. We also observed evidence of a learning curve within centers. In analysis restricted to programs that did not report cases in 2014, increasing cumulative ECMO experience over time was associated with progressively lower mortality. Each doubling of prior cumulative ECMO-day-volume within a center was associated with a 3% reduction in mortality. Centers can improve significantly with experience, quickly increasing and then plateauing. This is encouraging for regions without ECMO centers. Once established and volume builds, outcomes can match longstanding ECMO programs. These findings impact VV ECMO program development, benchmarking, and regionalization. Relying solely on case count thresholds is insufficient, as they don't account for experience from long runs. ECMO-day-volume provides a more nuanced measure aligned with current practice. Policymakers and societies could include ECMO-day-volume in quality metrics, accreditation, and referrals. The within-center learning association highlights the importance of early experience, protocols, and mentorship to fast-track expertise in new programs. Prior investigations of the volume-outcome relationship have examined a wide range of interventions. 1 , 2 , 5 , 14 In nearly all such studies, institutional experience has been measured using annual case-volume. For many interventions, beyond what we demonstrated for VV ECMO, outcomes are likely influenced not only by the frequency of initiation but also by the duration of ongoing management. The duration of management is expected to determine cumulative exposure to complications, resource utilization, staffing demands, and systems-level expertise. This may be particularly relevant for outcomes related to mechanical ventilation, renal replacement therapy, post-transplant critical care, and VA ECMO where management extends over days to weeks and requires sustained multidisciplinary coordination. Our findings suggest that applying duration-based metrics to these other interventions may be of value and future study is needed to apply this metric across critical care interventions. Several limitations merit consideration. First, as an observational registry study, residual confounding by unmeasured patient- and center-level factors cannot be excluded, and observed associations should not be interpreted as strictly causal. Centers with greater annual ECMO-day-volume may differ systematically from lower-volume centers in ways not fully captured by available variables. This limitation prevents examination of granular data such as staffing models, referral patterns, institutional resources, cannulation strategies, post-ECMO care, and discharge location. Second, the use of ECMO-day-volume as an exposure introduces the potential for survival bias, as longer ECMO runs inherently contribute more days to a center’s volume metric. Although our primary analysis focuses on center-level annual experience rather than patient-level duration, centers with better outcomes may accumulate more ECMO-day-volume because patients survive longer. To address this concern, we included a sensitivity analysis examining outcomes based on ECMO-case-volume. Given that ECMO-case-volume is not directly influenced by run duration or survival, these concordant results suggest that survivor bias alone is unlikely to fully explain the observed volume-outcome relationship. Third, complications were modeled as rates per ECMO-day, which assumes a constant risk over time and does not capture potential time-varying hazards across the ECMO course. Further, complication ascertainment in registry data may be subject to misclassification or differential reporting by center volume. Conclusion Higher VV ECMO experience, measured by annual ECMO-day-volume, links to lower in-hospital mortality and fewer complications, especially in centers with longer ECMO runs. Better outcomes at high-volume centers likely result from improved patient selection and clinical expertise. These factors are important when starting new VV ECMO programs and for quality initiatives in existing ones. Further research should explore how day-volume impacts outcomes in other critical care interventions. Declarations Statements and Disclosure: Dr. Rackley received funding from Inspira and Roche. Dr. King received funding from Merck and United Therapeutics. Dr. King received funding from Merck and United Therapeutics. Dr. Green received consulting fees from Ceribell Inc. The remaining authors have disclosed that they have no potential conflicts of interest. Take Home Message : This large ELSO registry analysis found an inverse relationship with total days of VV ECMO support per center-year and in-hospital mortality. Fewer ECMO-related complications per day occurred in centers with higher VV ECMO-day-volume. Center learning was observed to increase quickly, and plateau as experience was gained. These findings support the establishment of new ECMO centers, assuming the volume to support expertise development exists. References Nguyen YL, Wallace DJ, Yordanov Y et al (2015) The Volume-Outcome Relationship in Critical Care: A Systematic Review and Meta-analysis. Chest Jul 148(1):79–92. 10.1378/chest.14-2195 Kahn JM, Goss CH, Heagerty PJ, Kramer AA, O'Brien CR, Rubenfeld GD (2006) Hospital volume and the outcomes of mechanical ventilation. N Engl J Med Jul 6(1):41–50. 10.1056/NEJMsa053993 Karamlou T, Vafaeezadeh M, Parrish AM et al (2013) Increased extracorporeal membrane oxygenation center case volume is associated with improved extracorporeal membrane oxygenation survival among pediatric patients. J Thorac Cardiovasc Surg Feb 145(2):470–475. 10.1016/j.jtcvs.2012.11.037 Freeman CL, Bennett TD, Casper TC et al (Mar 2014) Pediatric and neonatal extracorporeal membrane oxygenation: does center volume impact mortality?*. Crit Care Med 42(3):512–519. 10.1097/01.ccm.0000435674.83682.96 Barbaro RP, Odetola FO, Kidwell KM et al (2015) Association of hospital-level volume of extracorporeal membrane oxygenation cases and mortality. Analysis of the extracorporeal life support organization registry. Am J Respir Crit Care Med Apr 15(8):894–901. 10.1164/rccm.201409-1634OC Hayes D Jr., Tobias JD, Tumin D (2016) Center Volume and Extracorporeal Membrane Oxygenation Support at Lung Transplantation in the Lung Allocation Score Era. Am J Respir Crit Care Med Aug 1(3):317–326. 10.1164/rccm.201511-2222OC McCarthy FH, McDermott KM, Spragan D et al (2016) Unconventional Volume-Outcome Associations in Adult Extracorporeal Membrane Oxygenation in the United States. Ann Thorac Surg Aug 102(2):489–495. 10.1016/j.athoracsur.2016.02.009 Muguruma K, Kunisawa S, Fushimi K, Imanaka Y (2020) Epidemiology and volume-outcome relationship of extracorporeal membrane oxygenation for respiratory failure in Japan: A retrospective observational study using a national administrative database. Acute Med Surg Jan-Dec 7(1):e486. 10.1002/ams2.486 Chandel A, Fabyan KD, Mendelsohn S et al (2024) Prevalence and Survival of Prolonged Venovenous Extracorporeal Membrane Oxygenation for Acute Respiratory Distress Syndrome: An Analysis of the Extracorporeal Life Support Organization Registry. Crit Care Med Jun 1(6):869–877. 10.1097/CCM.0000000000006200 Schmidt M, Bailey M, Sheldrake J et al (2014) Predicting survival after extracorporeal membrane oxygenation for severe acute respiratory failure. The Respiratory Extracorporeal Membrane Oxygenation Survival Prediction (RESP) score. Am J Respir Crit Care Med Jun 1(11):1374–1382. 10.1164/rccm.201311-2023OC Nasr VG, Raman L, Barbaro RP et al (2019) Highlights from the Extracorporeal Life Support Organization Registry: 2006–2017. ASAIO J Aug 65(6):537–544. 10.1097/MAT.0000000000000863 Whiteside HL, Hillerson D, Abdel-Latif A, Gupta VA (2023) Prognostic Implication of Pre-Cannulation Cardiac Arrest in Patients Undergoing Extracorporeal Membrane Oxygenation for the Management of Cardiogenic Shock. J Intensive Care Med Feb 38(2):202–207. 10.1177/08850666221115606 Wu MY, Huang CC, Wu TI, Chang YS, Wang CL, Lin PJ (2017) Is There a Preinterventional Mechanical Ventilation Time Limit for Candidates of Adult Respiratory Extracorporeal Membrane Oxygenation. ASAIO J Sep/Oct 63(5):650–658. 10.1097/MAT.0000000000000577 Deitz RL, Emerel L, Chan EG et al (2023) Waitlist Mortality and Extracorporeal Membrane Oxygenation Bridge to Lung Transplant. Ann Thorac Surg Jul 116(1):156–162. 10.1016/j.athoracsur.2023.02.062 Tables Table 1 Baseline characteristics by annual ECMO-day-volume (categorized by quartile [Q1-Q4]) Parameter Overall N = 54,036 1 < 174 (Q1) N = 13,509 1 174–417 (Q2) N = 13,509 1 418–866 (Q3) N = 13,509 1 ≥ 867 (Q4) N = 13,509 1 Baseline characteristics Age, years 48.6 (36.2, 59.2) 48.5 (34.7, 59.9) 49.7 (36.6, 60.5) 48.6 (36.4, 59.0) 47.9 (36.9, 57.4) Gender Female 19,418 (36%) 4,901 (36%) 4,838 (36%) 4,899 (37%) 4,780 (36%) Missing 350 64 61 115 110 Body mass index, kg/m² 29.6 (25.1, 35.4) 28.8 (24.4, 34.8) 29.3 (24.8, 35.1) 29.8 (25.4, 35.5) 30.5 (26.0, 36.2) Missing 9,700 2,636 2,441 2,644 1,979 COVID-19 14,327 (30%) 1,369 (12%) 2,491 (21%) 3,599 (30%) 6,868 (53%) Missing 5,486 1,626 1,681 1,535 644 Diagnosis Category Aspiration pneumonitis 1,106 (2.2%) 381 (2.9%) 335 (2.6%) 239 (1.9%) 151 (1.2%) Asthma 2,879 (5.6%) 812 (6.2%) 706 (5.4%) 748 (5.9%) 613 (4.9%) Bacterial pneumonia 2,706 (5.3%) 986 (7.6%) 870 (6.7%) 584 (4.6%) 266 (2.1%) Nonrespiratory or chronic respiratory diagnosis 3,814 (7.5%) 1,244 (9.6%) 1,013 (7.8%) 918 (7.3%) 639 (5.1%) Other acute respiratory diagnosis 18,853 (37%) 5,534 (43%) 5,186 (40%) 4,575 (36%) 3,558 (28%) Trauma or burn 5,971 (12%) 1,633 (13%) 1,760 (14%) 1,553 (12%) 1,025 (8.2%) Viral pneumonia 15,828 (31%) 2,407 (19%) 3,158 (24%) 4,014 (32%) 6,249 (50%) Missing 2,879 512 481 878 1,008 Immunocompromised 3,778 (7.4%) 808 (6.2%) 1,128 (8.7%) 1,053 (8.3%) 789 (6.3%) Missing 2,879 512 481 878 1,008 History of CNS dysfunction 2,174 (4.2%) 606 (4.7%) 630 (4.8%) 567 (4.5%) 371 (3.0%) Missing 2,879 512 481 878 1,008 Pre-ECMO Disease Severity PaCO₂, mmHg 58.0 (47.0, 74.0) 58.0 (46.5, 75.0) 58.0 (47.0, 74.0) 57.0 (46.0, 73.0) 59.0 (48.0, 74.0) Missing 12,464 2,478 3,085 3,241 3,660 Peak inspiratory pressure, cm H₂O 33.0 (29.0, 38.0) 34.0 (29.0, 39.0) 33.0 (29.0, 38.0) 33.0 (29.0, 38.0) 33.0 (29.0, 37.0) Missing 23,137 5,249 5,688 5,998 6,202 Mean airway pressure, cm H₂O 21.0 (17.0, 25.0) 21.0 (17.0, 25.0) 21.0 (17.0, 25.0) 20.0 (17.0, 24.0) 21.0 (17.0, 24.0) Missing 30,906 7,197 7,797 7,699 8,213 Serum lactate, mmol/L 1.9 (1.2, 3.5) 2.1 (1.3, 4.4) 1.9 (1.2, 3.7) 1.9 (1.2, 3.4) 1.7 (1.1, 2.8) Missing 29,421 7,026 7,437 7,695 7,263 Cardiac arrest 4,421 (8.2%) 1,412 (10%) 1,209 (8.9%) 964 (7.1%) 836 (6.2%) Neuromuscular blockade 27,415 (51%) 6,464 (48%) 6,480 (48%) 6,915 (51%) 7,556 (56%) Pulmonary vasodilator 11,324 (21%) 2,636 (20%) 2,443 (18%) 2,949 (22%) 3,296 (24%) Bicarbonate infusion 5,534 (10%) 1,821 (13%) 1,516 (11%) 1,195 (8.8%) 1,002 (7.4%) Prone positioning 12,122 (22%) 2,238 (17%) 2,607 (19%) 2,962 (22%) 4,315 (32%) Vasopressor use 30,604 (57%) 8,211 (61%) 7,753 (57%) 7,358 (54%) 7,282 (54%) Time from intubation to ECMO, hours 37.0 (9.0, 120.0) 31.0 (8.0, 106.0) 34.0 (8.0, 116.0) 40.0 (9.0, 126.0) 49.0 (11.0, 127.0) Missing 10,576 2,224 2,595 2,660 3,097 RESP score total 2.0 (0.0, 4.0) 2.0 (0.0, 4.0) 2.0 (0.0, 4.0) 2.0 (0.0, 4.0) 2.0 (0.0, 4.0) Missing 29,691 6,846 7,186 7,635 8,024 1 Median (Q1, Q3); n (%) Table 2 Adjusted hierarchical logistic regression of center annual VV ECMO-day-volume and in-hospital mortality Parameter OR 95% CI p-value Center VV ECMO volume (per doubling of ECMO-day-volume) 0.96 0.93, 0.98 0.001 COVID-19 (yes vs no) No — — Yes 4.19 2.87, 6.11 < 0.001 RESP score (per point increase) 0.86 0.85, 0.87 < 0.001 Calendar year (per year) 0.98 0.97, 0.99 0.013 Volume × COVID-19 interaction Center VV ECMO volume (per doubling of ECMO-day-volume) * Yes 0.93 0.90, 0.97 < 0.001 Abbreviations: CI = Confidence Interval, OR = Odds Ratio Supplementary Files eFigure1.png eFigure 1: Adjusted predicted probability of in-hospital mortality across annual center VV ECMO-day-volume Predictions are derived from a hierarchical logistic regression model adjusted for RESP score and calendar year with random intercepts for center. Shaded areas represent 95% confidence intervals. Solid lines indicate COVID-19 status. efigure2.png eFigure 2: Adjusted association between average center VV ECMO-day-volume and in-hospital mortality eFigure3.png eFigure 3: Adjusted in-hospital mortality as a function of within-center cumulative VV ECMO experience Adjusted population-average predicted probabilities of in-hospital mortality derived from a mixed-effects logistic regression model, plotted against cumulative center VV ECMO-day-volume (linear x-axis) accrued prior to the index year. Shaded bands represent 95% confidence intervals. The vertical dashed line allows for visualization of a potential inflection point in the learning curve. efigure4.png eFigure 4: Forest plot of adjusted incidence rate ratios (IRR) for ECMO related complications per doubling of annual ECMO-case-volume. Estimates are derived from mixed-effects negative binomial models with an offset for ECMO duration (to estimate complication rate per ECMO-day rather than the total number of complications) and random intercepts for center, adjusted for RESP score and calendar year. Error bars represent 95% confidence intervals. STROBEchecklistcohort.docx supplementfinal.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8770342","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":587810438,"identity":"c29e6912-27d8-4ea2-b164-7ea99e2b1275","order_by":0,"name":"Adam Green","email":"data:image/png;base64,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","orcid":"https://orcid.org/0000-0003-2562-746X","institution":"Cooper University Health Care","correspondingAuthor":true,"prefix":"","firstName":"Adam","middleName":"","lastName":"Green","suffix":""},{"id":587810439,"identity":"eac9f74d-964c-4d96-a2bd-4981e3618104","order_by":1,"name":"Abhimanyu Chandel","email":"","orcid":"","institution":"Uniformed Services University: Uniformed Services University of the Health Sciences","correspondingAuthor":false,"prefix":"","firstName":"Abhimanyu","middleName":"","lastName":"Chandel","suffix":""},{"id":587810440,"identity":"618fd5a5-409c-4941-8802-ef7a36179fcf","order_by":2,"name":"Clifford Chang","email":"","orcid":"","institution":"Cooper Hospital University Medical Center: Cooper University Health Care","correspondingAuthor":false,"prefix":"","firstName":"Clifford","middleName":"","lastName":"Chang","suffix":""},{"id":587810441,"identity":"6a25bdf3-995d-4517-9331-efa59cee1fbf","order_by":3,"name":"Nitin Puri","email":"","orcid":"","institution":"Cooper Hospital University Medical Center: Cooper University Health Care","correspondingAuthor":false,"prefix":"","firstName":"Nitin","middleName":"","lastName":"Puri","suffix":""},{"id":587810442,"identity":"ddee0b21-835b-4402-ad83-7cad5a53a497","order_by":4,"name":"Christopher Noel","email":"","orcid":"","institution":"Cooper Hospital University Medical Center: Cooper University Health Care","correspondingAuthor":false,"prefix":"","firstName":"Christopher","middleName":"","lastName":"Noel","suffix":""},{"id":587810443,"identity":"7d056147-16c8-4689-9d4a-4eda19a2e8ce","order_by":5,"name":"Taylor S. Conrad","email":"","orcid":"","institution":"Louisiana State University Medical Center: University Health Shreveport","correspondingAuthor":false,"prefix":"","firstName":"Taylor","middleName":"S.","lastName":"Conrad","suffix":""},{"id":587810444,"identity":"661a4465-aeeb-4e34-8c74-de18bf48bc87","order_by":6,"name":"Emily Damuth","email":"","orcid":"","institution":"Cooper Hospital University Medical Center: Cooper University Health Care","correspondingAuthor":false,"prefix":"","firstName":"Emily","middleName":"","lastName":"Damuth","suffix":""},{"id":587810445,"identity":"53405b9e-cc78-4905-abaf-6de95b2ceb2e","order_by":7,"name":"Craig R. Rackley","email":"","orcid":"","institution":"Duke University Health System","correspondingAuthor":false,"prefix":"","firstName":"Craig","middleName":"R.","lastName":"Rackley","suffix":""},{"id":587810446,"identity":"6257a7f5-38b5-4b7e-a899-d2522e345b89","order_by":8,"name":"Christopher S. King","email":"","orcid":"","institution":"Inova Fairfax Medical Campus: Inova Fairfax Hospital","correspondingAuthor":false,"prefix":"","firstName":"Christopher","middleName":"S.","lastName":"King","suffix":""},{"id":587810447,"identity":"2578cd47-385d-4f8c-8076-9ce5b281a2d0","order_by":9,"name":"Steven A. Conrad","email":"","orcid":"","institution":"Louisiana State University Medical Center: University Health Shreveport","correspondingAuthor":false,"prefix":"","firstName":"Steven","middleName":"A.","lastName":"Conrad","suffix":""}],"badges":[],"createdAt":"2026-02-03 03:13:24","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8770342/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8770342/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":102515956,"identity":"0a511e6c-dcf2-439e-838d-8dd737dd17a1","added_by":"auto","created_at":"2026-02-12 13:46:05","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":379725,"visible":true,"origin":"","legend":"\u003cp\u003eIn-hospital mortality by center VV ECMO-day-volume (divided into quartiles) and COVID-19 status\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-8770342/v1/725f8c83b96b0bb4d84862f6.png"},{"id":102515963,"identity":"2c7e32f9-eaeb-47c7-a5ff-51534eb7b656","added_by":"auto","created_at":"2026-02-12 13:46:05","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":485377,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eForest plot of adjusted incidence rate ratios (IRR) for ECMO related complications per doubling of annual VV ECMO-day-volume.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEstimates are derived from mixed-effects negative binomial models with an offset for ECMO duration (to estimate complication rate per ECMO-day rather than the total number of complications) and random intercepts for center, adjusted for RESP score and calendar year. Error bars represent 95% confidence intervals. Limb and neurologic complications were observed too infrequent to accurately estimate confidence intervals.\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-8770342/v1/06b0a4916ca8867329084ed1.png"},{"id":102515957,"identity":"7988236e-a3e3-4e37-bbb7-98840f74d3c3","added_by":"auto","created_at":"2026-02-12 13:46:05","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":454697,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAdjusted in-hospital mortality by center VV ECMO volume quartile, stratified by average VV ECMO run length\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAdjusted probabilities of in-hospital mortality are shown across quartiles of annual center VV ECMO-day-volume, stratified by tertiles of average VV ECMO run length (short ≤12 days, intermediate 12–18 days, long \u0026gt;18 days). Estimates were derived from hierarchical logistic regression models adjusted for RESP score and calendar year, with a random intercept for center. Points represent adjusted predicted mortality, with error bars indicating 95% confidence intervals.\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-8770342/v1/7d6639e46816bc2d52284d73.png"},{"id":107106078,"identity":"97a38058-a25d-4992-8cf1-15a8733016ca","added_by":"auto","created_at":"2026-04-16 21:09:09","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1467383,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8770342/v1/10342044-0f54-48e8-b2f7-8aa86a280918.pdf"},{"id":102746940,"identity":"29dd6d43-21e3-4654-8a31-53009cdef4a9","added_by":"auto","created_at":"2026-02-16 09:03:02","extension":"png","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":468074,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eeFigure 1: Adjusted predicted probability of in-hospital mortality across annual center VV ECMO-day-volume\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePredictions are derived from a hierarchical logistic regression model adjusted for RESP score and calendar year with random intercepts for center. Shaded areas represent 95% confidence intervals. Solid lines indicate COVID-19 status.\u003c/p\u003e","description":"","filename":"eFigure1.png","url":"https://assets-eu.researchsquare.com/files/rs-8770342/v1/3935cc66f43cb5ecd018ec61.png"},{"id":102515960,"identity":"5e92af64-6f89-4602-b79d-6a1df59198b0","added_by":"auto","created_at":"2026-02-12 13:46:05","extension":"png","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":435967,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eeFigure 2: Adjusted association between average center VV ECMO-day-volume and in-hospital mortality\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"efigure2.png","url":"https://assets-eu.researchsquare.com/files/rs-8770342/v1/c87bb0ca5bf83f56515e2529.png"},{"id":102746280,"identity":"bf9f5012-e2be-41e9-8701-9767765aa6e9","added_by":"auto","created_at":"2026-02-16 08:56:25","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":436275,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eeFigure 3: Adjusted in-hospital mortality as a function of within-center cumulative VV ECMO experience\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAdjusted population-average predicted probabilities of in-hospital mortality derived from a mixed-effects logistic regression model, plotted against cumulative center VV ECMO-day-volume (linear x-axis) accrued prior to the index year. Shaded bands represent 95% confidence intervals. The vertical dashed line allows for visualization of a potential inflection point in the learning curve.\u003c/p\u003e","description":"","filename":"eFigure3.png","url":"https://assets-eu.researchsquare.com/files/rs-8770342/v1/033f37d1a18a9a717f114942.png"},{"id":102515962,"identity":"c9467a11-2614-4ac8-8ef3-6d8890c83bb7","added_by":"auto","created_at":"2026-02-12 13:46:05","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":493392,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eeFigure 4:\u003c/strong\u003e \u003cstrong\u003eForest plot of adjusted incidence rate ratios (IRR) for ECMO related complications per doubling of annual ECMO-case-volume.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEstimates are derived from mixed-effects negative binomial models with an offset for ECMO duration (to estimate complication rate per ECMO-day rather than the total number of complications) and random intercepts for center, adjusted for RESP score and calendar year. Error bars represent 95% confidence intervals.\u003c/p\u003e","description":"","filename":"efigure4.png","url":"https://assets-eu.researchsquare.com/files/rs-8770342/v1/6c3ec028a31b0ff61becb958.png"},{"id":102515958,"identity":"6a53e7de-b874-44f3-beb9-447215831878","added_by":"auto","created_at":"2026-02-12 13:46:05","extension":"docx","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":33771,"visible":true,"origin":"","legend":"","description":"","filename":"STROBEchecklistcohort.docx","url":"https://assets-eu.researchsquare.com/files/rs-8770342/v1/fd8dc96d27dd3f6a658b3397.docx"},{"id":102515964,"identity":"20668269-2711-4c62-a15f-46092655ed33","added_by":"auto","created_at":"2026-02-12 13:46:05","extension":"docx","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":67881,"visible":true,"origin":"","legend":"","description":"","filename":"supplementfinal.docx","url":"https://assets-eu.researchsquare.com/files/rs-8770342/v1/9512d0ffec26c94e90430b26.docx"}],"financialInterests":"","formattedTitle":"How much ECMO is enough? Center experience, learning curves, run durations, and outcomes in venovenous ECMO: An ELSO registry analysis","fulltext":[{"header":"Introduction","content":"\u003cp\u003eVenovenous extracorporeal membrane oxygenation (VV ECMO) is an established and expanding supportive tool for patients with severe, potentially reversible respiratory failure. As VV ECMO utilization has expanded, understanding institutional factors associated with improved outcomes has become increasingly important.\u003c/p\u003e \u003cp\u003eA positive correlation between center volume and outcomes has been observed in other areas, including post-cardiac arrest care, postoperative recovery, and mechanical ventilation.\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e In the current literature, ECMO center volume is most often characterized by the number of separate ECMO cases per year, regardless of the run duration. Analyses of the ELSO Registry and national datasets have shown lower risk-adjusted mortality at centers performing a greater number of ECMO cases annually.\u003csup\u003e\u003cspan additionalcitationids=\"CR4 CR5 CR6 CR7\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eThe association between center volume and improved outcomes is likely multifactorial and includes patient selection, timing of ECMO initiation, and institutional capabilities.\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e While the number of cases per year (ECMO-case-volume) is an intuitive and reproducible metric, it does not account for substantial heterogeneity in ECMO run duration between centers. Contemporary adult VV ECMO increasingly involves prolonged support, reflecting changes in patient selection, improvements in supportive care, and greater use of ECMO as a bridge to recovery or transplantation.\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e Previous ELSO-based analyses have demonstrated that prolonged ECMO support is not independently associated with mortality, suggesting that experience with sustained ECMO care can be accrued without proportional increases in risk.\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e Therefore, cumulative exposure to ECMO management, reflected by the total number of days of support per center-year (ECMO-day-volume), may better capture institutional experience. Annual ECMO-day-volume incorporates the frequency and duration of care, reflects systems-level expertise exercised daily during ECMO support, and avoids misclassifying centers with fewer but longer runs as inexperienced centers. Prior studies have not evaluated annual ECMO-day-volume as a center-level exposure.\u003c/p\u003e \u003cp\u003eAccordingly, we evaluated the association between center-level VV ECMO experience, defined by annual ECMO-day-volume, and patient outcomes in adult patients supported by VV ECMO. We hypothesized that higher annual ECMO-day-volume would be associated with lower in-hospital mortality and fewer ECMO-related complications, independent of patient-level severity of illness and secular trends. By reframing ECMO experience in terms of cumulative exposure rather than case counts alone, this study aims to refine the understanding of volume\u0026ndash;outcome relationships in adult VV ECMO.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy Design and Population\u003c/h2\u003e \u003cp\u003eWe conducted a retrospective cohort study using data from the ELSO Registry. Established in 1989, the ELSO Registry is the world\u0026rsquo;s largest, comprehensive database of patient data on patients receiving ECMO with 780 active centers and data on 270,960 cases (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.elso.org/registry.aspx\u003c/span\u003e\u003cspan address=\"https://www.elso.org/registry.aspx\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Adult patients (\u0026ge;\u0026thinsp;18 years) supported with VV ECMO between 2014 and 2024 were included. ECMO support was analyzed at the run level, with patients clustered within ECMO centers. The ECMO run was the unit of analysis for all primary models.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eExposure:\u003c/h3\u003e\n\u003cp\u003eCenter experience was quantified using annual center-level VV ECMO-day-volume, defined as the total number of days of VV ECMO support delivered by a center within a given calendar year. To facilitate interpretability and account for right-skewness, annual ECMO-day-volume was log-transformed and parameterized per doubling of exposure. As a sensitivity analysis, center experience was alternatively defined using annual ECMO-case-volume, modeled using the same framework.\u003c/p\u003e\n\u003ch3\u003eOutcomes\u003c/h3\u003e\n\u003cp\u003eThe primary outcome was in-hospital mortality. Secondary outcomes included the overall ECMO-related complication rate and subtype-specific complication rates. Complications were identified using ELSO-defined complication codes and were considered present only if recorded as active during the ECMO run.\u003c/p\u003e\n\u003ch3\u003eCovariates\u003c/h3\u003e\n\u003cp\u003ePatient-level severity of illness was adjusted using the Respiratory Extracorporeal Membrane Oxygenation Survival Prediction (RESP) score (a composite score of disease severity comprising prognostically significant components including age, primary diagnosis, pre-ECLS management, and comorbidities), constructed from available ELSO variables according to published methodology.\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e Calendar year of ECMO initiation was included to account for secular trends in ECMO practice and outcomes. COVID-19 status was included as a binary variable and evaluated as a potential effect modifier of the volume\u0026ndash;outcome relationship.\u003c/p\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eStatistical Analysis\u003c/h2\u003e \u003cp\u003eThe association between center VV ECMO experience and in-hospital mortality was evaluated using hierarchical logistic regression models with a random intercept for center to account for clustering of runs within centers. The primary model included annual ECMO-day-volume (per doubling), RESP score, calendar year, COVID-19 status, and an interaction term between center volume and COVID-19 status. Adjusted odds ratios (ORs) with 95% confidence intervals (CIs) were reported.\u003c/p\u003e \u003cp\u003eECMO-related complications were modeled as rates per ECMO-day using mixed-effects negative binomial regression, with a random intercept for center and an offset for log-transformed ECMO duration to account for differences in time at risk. Incidence rate ratios (IRRs) with 95% CIs were reported.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eExploratory and Sensitivity Analyses\u003c/h2\u003e \u003cp\u003eExploratory analyses examined whether experience with longer ECMO runs modified the volume\u0026ndash;outcome relationship by incorporating center-level average ECMO run duration, calculated annually as total ECMO-days divided by run count. Additional exploratory analyses evaluated within-center learning by modeling cumulative prior ECMO-day-volume among centers initiating VV ECMO after 2014.\u003c/p\u003e \u003cp\u003eSensitivity analyses assessed alternative exposure definitions using ECMO-case-volume, evaluated nonlinearity using restricted cubic splines, and addressed missing RESP scores using indicator-based adjustment and inverse probability weighting.\u003c/p\u003e \u003cp\u003eAdditional details regarding the above methodology, handling of missing data, and model diagnostics are provided in the online data supplement.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eEthical Considerations\u003c/h3\u003e\n\u003cp\u003eThe ELSO Registry contains de-identified data. This study was conducted in accordance with institutional policies governing secondary analyses of de-identified data and was considered exempt from human subject review.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eStudy population\u003c/h2\u003e \u003cp\u003eBetween 2014 and 2024, a total of 54,036 adult patients supported with VV ECMO were recorded in the ELSO Registry across 697 contributing centers and were included for analysis. Baseline patient characteristics stratified by center ECMO-day-volume (quartile [Q1-Q4] of annual ECMO-days) are included in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Overall, patients treated at lower- and higher-volume centers, defined based on annual ECMO-day-volume, were broadly similar with respect to age, gender, and body mass index. However, several differences in case composition were observed across volume strata. Patients supported at centers with higher day-volume were less likely to have markers of advanced physiologic vulnerability at the time of VV ECMO initiation, including a lower prevalence of pre-ECMO cardiac arrest (Q4 vs. Q1: 6.2% vs. 10%), but were more likely to have incorporated advanced non-ECMO acute respiratory distress syndrome (ARDS) therapies prior to cannulation. These centers had higher rates of prone positioning (32% vs. 17%), neuromuscular blockade (56% vs. 48%), and inhaled pulmonary vasodilator use (24% vs. 20%), as well as longer durations of mechanical ventilation prior to VV ECMO initiation (49 vs. 31 hours). Despite a longer pre-ECMO course, patients cannulated at centers with higher day-volume were less likely to require vasopressors (54% vs. 61%) or bicarbonate (7.4% vs. 13%). Finally, the proportion of patients with COVID-19\u0026ndash;associated respiratory failure was substantially higher among centers with higher annual ECMO-day-volume (53% vs. 12%). These trends were similar when baseline characteristics were stratified by annual ECMO cases per year (eTable 1).\u003c/p\u003e \u003cp\u003eAmong all patients, 21,949/54,036 (40.6%) died (Q4 vs. Q1: 40.7% vs. 39.3%). A comparison of observed mortality by quartile of annual ECMO-day-volume stratified by COVID-19 status is found in Fig.\u0026nbsp;1 and eTable 2. Unadjusted survival was significantly higher in high day-volume centers (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) in both strata. Lung transplantation was performed in 3,302 (6.1%) (Q4: N\u0026thinsp;=\u0026thinsp;866 [6.4%] vs. Q1: N\u0026thinsp;=\u0026thinsp;627 [4.6%]) patients. Survivors were most commonly discharged home (11,867/31,106; 38.2%) or to long-term acute care or rehabilitation facilities (9,334/31,106; 30.0%) (eTable3).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eCenter VV ECMO-day-volume and in-hospital mortality\u003c/h2\u003e \u003cp\u003eIn hierarchical logistic regression models adjusted for RESP score, calendar year, and center-level clustering, higher annual VV ECMO-day-volume was independently associated with lower in-hospital mortality (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Among patients without COVID-19, each doubling of annual VV ECMO-days was associated with a 4% reduction in the odds of death (OR 0.96; 95% CI, 0.93\u0026ndash;0.98; p\u0026thinsp;=\u0026thinsp;0.001). The protective effect of center day-volume was found to be stronger in patients with COVID-19 than in patients supported with VV ECMO for other reasons (eFigure 1). Among patients with COVID-19, each doubling of annual VV ECMO-day-volume was associated with an approximately 11% (OR volume [0.96] * OR interaction [0.93]\u0026thinsp;=\u0026thinsp;0.89) reduction in mortality odds. Nonlinearity in the association between center day-volume and mortality was assessed using restricted cubic splines. Spline-based models did not demonstrate a statistically significant departure from linearity (likelihood ratio test p\u0026thinsp;\u0026gt;\u0026thinsp;0.05).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eECMO-related complications\u003c/h2\u003e \u003cp\u003eECMO-related complications stratified by ECMO-day-volume are displayed in eTable 4. Total complications were more common in centers with higher day-volume. However, when rate of complications was considered per ECMO-day, higher center VV ECMO-day-volume was associated with lower rates of ECMO-related complications. In mixed-effects negative binomial models with an offset for ECMO duration, each doubling of annual VV ECMO-day-volume was associated with a lower overall complication rate (across all ECMO complications) per ECMO-day (Fig.\u0026nbsp;2).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eAverage VV ECMO run duration and the volume\u0026ndash;outcome relationship\u003c/h2\u003e \u003cp\u003eThe median run duration in the cohort was 15 days (IQR: 10\u0026ndash;21 days), and centers with higher VV ECMO-day-volume had higher median run duration (Q4 vs. Q1: 25 vs. 10 days; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). To explore whether experience managing longer VV ECMO runs modified the volume\u0026ndash;outcome relationship, centers were stratified by annual average VV ECMO run duration (short, intermediate, and long [by tertile of run duration]) (eTable 5). In stratified analyses, the association between center VV ECMO-day-volume and mortality was most pronounced in centers with longer average run durations (Fig.\u0026nbsp;3). In centers with long average VV ECMO runs, each doubling of annual VV ECMO-day-volume was associated with a significant reduction in mortality (OR 0.95; 95% CI, 0.91\u0026ndash;0.99) (eFigure 2), whereas no significant association was observed in centers with shorter or intermediate run durations. These findings suggest that prolonged VV ECMO support may be a modifier of the volume\u0026ndash;outcome relationship.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eWithin-center learning and cumulative experience\u003c/h2\u003e \u003cp\u003eTo evaluate whether outcomes improve as an individual center gains experience over time, we performed a within-center learning analysis restricted to centers that began recording VV ECMO cases after 2014. After adjustment for contemporaneous annual volume, RESP score, and calendar year; each doubling of cumulative prior VV ECMO-day-volume at a given center was associated with a 3% lower odds of in-hospital mortality (OR 0.97; 95% CI, 0.95\u0026ndash;0.99), consistent with a center-level learning effect over time that increases quickly while plateauing (after approximately 100 cumulative VV ECMO-days) as experience is gained (eFigure 3). Considerable heterogeneity in learning trajectories was observed across centers, indicating that, although outcomes generally improved with experience, the rate at which centers accrued benefits varied.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eSensitivity analyses\u003c/h2\u003e \u003cp\u003eResults were robust across multiple sensitivity analyses. Defining center volume by ECMO-case-volume yielded qualitatively similar associations with mortality (eTable 6). However, the model using annual VV ECMO-day-volume demonstrated superior predictive performance compared with the model using annual case-volume, with a higher cross-validated log-likelihood and a lower mean Brier score (0.221 vs. 0.229), indicating improved out-of-sample fit and calibration when considering annual ECMO-day-volume. Likewise, when considering center experience based on ECMO-case-volume, the complication rate was significantly lower for higher case-volume centers (eFigure 4).\u003c/p\u003e \u003cp\u003eTo address missing RESP scores, both indicator-based adjustment and inverse probability weighting produced effect estimates consistent with the primary analysis (eTables 7 and 8). Together, these findings support the robustness of the observed associations between ECMO-day-volume, mortality, and complications.\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this large, contemporary analysis of the international ELSO Registry, we found that higher center VV ECMO experience, quantified by annual ECMO-day-volume, was independently associated with lower in-hospital mortality and fewer ECMO-related complications. This association persisted after adjustment for patient severity of illness, calendar year, and center-level clustering. A previous ELSO registry analysis, defining experience as annual case-volume, likewise demonstrated lower mortality at high-volume centers.\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e However, this analysis was conducted more than a decade ago, combined venoarterial (VA) and VV ECMO, and preceded the COVID-19 pandemic. Since this study, ECMO practices have evolved, with an increase in prolonged VV ECMO runs.\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e Accordingly, we re-examined the volume\u0026ndash;outcome relationship using annual ECMO-day-volume as the measure of center experience, a metric that more fully captures cumulative exposure to ECMO management over time. By restricting the analysis to VV ECMO, we aimed to create a more homogeneous population.\u003c/p\u003e \u003cp\u003eWe propose two main reasons for these findings: optimal patient selection and the acquisition of clinical expertise. Although baseline demographic characteristics and total RESP score were similar across volume-strata, differences in case composition and distribution of RESP components were evident. Centers with higher day-volume cannulated fewer patients with immunocompromised states, central nervous system dysfunction, and pre-ECMO cardiac arrest, suggesting a focus on patient selection.\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e Likewise, high day-volume centers appear to prioritize ARDS care prior to cannulation, as evidenced by a higher percentage of patients receiving neuromuscular blockers and prone ventilation. Additionally, a longer interval between intubation and ECMO cannulation was observed at high day-volume centers, likely an attempt to optimize medical management before resorting to ECMO.\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e Despite this, fewer patients required vasopressors or bicarbonate at the time of cannulation at the high day-volume centers. Collectively, these findings suggest experienced centers employ more selective and nuanced patient-selection strategies and may be more adept at identifying patients likely to benefit from ECMO.\u003c/p\u003e \u003cp\u003eIn addition to patient selection, evidence of clinical expertise at high ECMO-day-volume centers was also observed. First, higher day-volume centers experienced fewer ECMO-related complications. There were fewer complications observed per ECMO day in all complication types as day-volume increases. When all complications are considered, there was a 14% reduction in complications per doubling of ECMO day-volume (IRR 0.86; 95% CI, 0.85\u0026ndash;0.88). This is likely due to a variety of reasons: provider experience, established care protocols, ECMO dedicated resources, and the availability of ancillary services in this specific population - all of these factors are influenced by increased ECMO volume. The consistency of these findings across multiple complication types suggests that improved technical and clinical management, rather than patient selection alone contributes to better outcomes.\u003c/p\u003e \u003cp\u003eFurther supporting a component of clinical expertise, the volume-outcome relationship was particularly strong in patients supported with VV ECMO for COVID-19. In these patients, each doubling of annual ECMO-day-volume was linked to approximately a 11% decrease in in-hospital mortality. This reflects the capacity of high-volume hospitals to provide robust ECMO services even during periods of significant resource strain.\u003c/p\u003e \u003cp\u003eCenters with longer ECMO runs showed a stronger link between volume and lower mortality, unlike centers with shorter runs. Experience from managing prolonged ECMO, not just higher case numbers, seems key to better outcomes. Longer runs expose teams to more complications and learning opportunities, making ECMO-day-volume a better measure than annual case volume. Importantly, in at least one year during the period analyzed, 5.7% (40/697) of centers would have been classified in the lower half of experience based on annual ECMO-case-volume but would be placed in the top third of centers based on annual ECMO-day-volume. This discordance highlights that a small but meaningful number of centers may be misclassified based on case count alone.\u003c/p\u003e \u003cp\u003eWe also observed evidence of a learning curve within centers. In analysis restricted to programs that did not report cases in 2014, increasing cumulative ECMO experience over time was associated with progressively lower mortality. Each doubling of prior cumulative ECMO-day-volume within a center was associated with a 3% reduction in mortality. Centers can improve significantly with experience, quickly increasing and then plateauing. This is encouraging for regions without ECMO centers. Once established and volume builds, outcomes can match longstanding ECMO programs.\u003c/p\u003e \u003cp\u003eThese findings impact VV ECMO program development, benchmarking, and regionalization. Relying solely on case count thresholds is insufficient, as they don't account for experience from long runs. ECMO-day-volume provides a more nuanced measure aligned with current practice. Policymakers and societies could include ECMO-day-volume in quality metrics, accreditation, and referrals. The within-center learning association highlights the importance of early experience, protocols, and mentorship to fast-track expertise in new programs.\u003c/p\u003e \u003cp\u003ePrior investigations of the volume-outcome relationship have examined a wide range of interventions.\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e,\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e In nearly all such studies, institutional experience has been measured using annual case-volume. For many interventions, beyond what we demonstrated for VV ECMO, outcomes are likely influenced not only by the frequency of initiation but also by the duration of ongoing management. The duration of management is expected to determine cumulative exposure to complications, resource utilization, staffing demands, and systems-level expertise. This may be particularly relevant for outcomes related to mechanical ventilation, renal replacement therapy, post-transplant critical care, and VA ECMO where management extends over days to weeks and requires sustained multidisciplinary coordination. Our findings suggest that applying duration-based metrics to these other interventions may be of value and future study is needed to apply this metric across critical care interventions.\u003c/p\u003e \u003cp\u003eSeveral limitations merit consideration. First, as an observational registry study, residual confounding by unmeasured patient- and center-level factors cannot be excluded, and observed associations should not be interpreted as strictly causal. Centers with greater annual ECMO-day-volume may differ systematically from lower-volume centers in ways not fully captured by available variables. This limitation prevents examination of granular data such as staffing models, referral patterns, institutional resources, cannulation strategies, post-ECMO care, and discharge location. Second, the use of ECMO-day-volume as an exposure introduces the potential for survival bias, as longer ECMO runs inherently contribute more days to a center\u0026rsquo;s volume metric. Although our primary analysis focuses on center-level annual experience rather than patient-level duration, centers with better outcomes may accumulate more ECMO-day-volume because patients survive longer. To address this concern, we included a sensitivity analysis examining outcomes based on ECMO-case-volume. Given that ECMO-case-volume is not directly influenced by run duration or survival, these concordant results suggest that survivor bias alone is unlikely to fully explain the observed volume-outcome relationship. Third, complications were modeled as rates per ECMO-day, which assumes a constant risk over time and does not capture potential time-varying hazards across the ECMO course. Further, complication ascertainment in registry data may be subject to misclassification or differential reporting by center volume.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eHigher VV ECMO experience, measured by annual ECMO-day-volume, links to lower in-hospital mortality and fewer complications, especially in centers with longer ECMO runs. Better outcomes at high-volume centers likely result from improved patient selection and clinical expertise. These factors are important when starting new VV ECMO programs and for quality initiatives in existing ones. Further research should explore how day-volume impacts outcomes in other critical care interventions.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eStatements and Disclosure:\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDr. Rackley received funding from Inspira and Roche. Dr. King received funding from Merck and United Therapeutics. Dr. King received funding from Merck and United Therapeutics. Dr. Green received consulting fees from Ceribell Inc. The remaining authors have disclosed that they have no potential conflicts of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTake Home Message\u003c/strong\u003e: This large ELSO registry analysis found an inverse relationship with total days of VV ECMO support per center-year and in-hospital mortality. Fewer ECMO-related complications per day occurred in centers with higher VV ECMO-day-volume. Center learning was observed to increase quickly, and plateau as experience was gained. These findings support the establishment of new ECMO centers, assuming the volume to support expertise development exists.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eNguyen YL, Wallace DJ, Yordanov Y et al (2015) The Volume-Outcome Relationship in Critical Care: A Systematic Review and Meta-analysis. Chest Jul 148(1):79\u0026ndash;92. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1378/chest.14-2195\u003c/span\u003e\u003cspan address=\"10.1378/chest.14-2195\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKahn JM, Goss CH, Heagerty PJ, Kramer AA, O'Brien CR, Rubenfeld GD (2006) Hospital volume and the outcomes of mechanical ventilation. N Engl J Med Jul 6(1):41\u0026ndash;50. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1056/NEJMsa053993\u003c/span\u003e\u003cspan address=\"10.1056/NEJMsa053993\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKaramlou T, Vafaeezadeh M, Parrish AM et al (2013) Increased extracorporeal membrane oxygenation center case volume is associated with improved extracorporeal membrane oxygenation survival among pediatric patients. J Thorac Cardiovasc Surg Feb 145(2):470\u0026ndash;475. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jtcvs.2012.11.037\u003c/span\u003e\u003cspan address=\"10.1016/j.jtcvs.2012.11.037\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFreeman CL, Bennett TD, Casper TC et al (Mar 2014) Pediatric and neonatal extracorporeal membrane oxygenation: does center volume impact mortality?*. Crit Care Med 42(3):512\u0026ndash;519. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/01.ccm.0000435674.83682.96\u003c/span\u003e\u003cspan address=\"10.1097/01.ccm.0000435674.83682.96\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarbaro RP, Odetola FO, Kidwell KM et al (2015) Association of hospital-level volume of extracorporeal membrane oxygenation cases and mortality. Analysis of the extracorporeal life support organization registry. Am J Respir Crit Care Med Apr 15(8):894\u0026ndash;901. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1164/rccm.201409-1634OC\u003c/span\u003e\u003cspan address=\"10.1164/rccm.201409-1634OC\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHayes D Jr., Tobias JD, Tumin D (2016) Center Volume and Extracorporeal Membrane Oxygenation Support at Lung Transplantation in the Lung Allocation Score Era. Am J Respir Crit Care Med Aug 1(3):317\u0026ndash;326. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1164/rccm.201511-2222OC\u003c/span\u003e\u003cspan address=\"10.1164/rccm.201511-2222OC\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcCarthy FH, McDermott KM, Spragan D et al (2016) Unconventional Volume-Outcome Associations in Adult Extracorporeal Membrane Oxygenation in the United States. Ann Thorac Surg Aug 102(2):489\u0026ndash;495. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.athoracsur.2016.02.009\u003c/span\u003e\u003cspan address=\"10.1016/j.athoracsur.2016.02.009\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMuguruma K, Kunisawa S, Fushimi K, Imanaka Y (2020) Epidemiology and volume-outcome relationship of extracorporeal membrane oxygenation for respiratory failure in Japan: A retrospective observational study using a national administrative database. Acute Med Surg Jan-Dec 7(1):e486. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/ams2.486\u003c/span\u003e\u003cspan address=\"10.1002/ams2.486\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChandel A, Fabyan KD, Mendelsohn S et al (2024) Prevalence and Survival of Prolonged Venovenous Extracorporeal Membrane Oxygenation for Acute Respiratory Distress Syndrome: An Analysis of the Extracorporeal Life Support Organization Registry. Crit Care Med Jun 1(6):869\u0026ndash;877. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/CCM.0000000000006200\u003c/span\u003e\u003cspan address=\"10.1097/CCM.0000000000006200\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchmidt M, Bailey M, Sheldrake J et al (2014) Predicting survival after extracorporeal membrane oxygenation for severe acute respiratory failure. The Respiratory Extracorporeal Membrane Oxygenation Survival Prediction (RESP) score. Am J Respir Crit Care Med Jun 1(11):1374\u0026ndash;1382. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1164/rccm.201311-2023OC\u003c/span\u003e\u003cspan address=\"10.1164/rccm.201311-2023OC\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNasr VG, Raman L, Barbaro RP et al (2019) Highlights from the Extracorporeal Life Support Organization Registry: 2006\u0026ndash;2017. ASAIO J Aug 65(6):537\u0026ndash;544. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/MAT.0000000000000863\u003c/span\u003e\u003cspan address=\"10.1097/MAT.0000000000000863\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWhiteside HL, Hillerson D, Abdel-Latif A, Gupta VA (2023) Prognostic Implication of Pre-Cannulation Cardiac Arrest in Patients Undergoing Extracorporeal Membrane Oxygenation for the Management of Cardiogenic Shock. J Intensive Care Med Feb 38(2):202\u0026ndash;207. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1177/08850666221115606\u003c/span\u003e\u003cspan address=\"10.1177/08850666221115606\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu MY, Huang CC, Wu TI, Chang YS, Wang CL, Lin PJ (2017) Is There a Preinterventional Mechanical Ventilation Time Limit for Candidates of Adult Respiratory Extracorporeal Membrane Oxygenation. ASAIO J Sep/Oct 63(5):650\u0026ndash;658. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/MAT.0000000000000577\u003c/span\u003e\u003cspan address=\"10.1097/MAT.0000000000000577\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDeitz RL, Emerel L, Chan EG et al (2023) Waitlist Mortality and Extracorporeal Membrane Oxygenation Bridge to Lung Transplant. Ann Thorac Surg Jul 116(1):156\u0026ndash;162. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.athoracsur.2023.02.062\u003c/span\u003e\u003cspan address=\"10.1016/j.athoracsur.2023.02.062\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Tables","content":" \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eBaseline characteristics by annual ECMO-day-volume (categorized by quartile [Q1-Q4])\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eParameter\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOverall N\u0026thinsp;=\u0026thinsp;54,036\u003csup\u003e\u003cem\u003e1\u003c/em\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;174 (Q1) N\u0026thinsp;=\u0026thinsp;13,509\u003csup\u003e\u003cem\u003e1\u003c/em\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e174\u0026ndash;417 (Q2) N\u0026thinsp;=\u0026thinsp;13,509\u003csup\u003e\u003cem\u003e1\u003c/em\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e418\u0026ndash;866 (Q3) N\u0026thinsp;=\u0026thinsp;13,509\u003csup\u003e\u003cem\u003e1\u003c/em\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026ge;\u0026thinsp;867 (Q4) N\u0026thinsp;=\u0026thinsp;13,509\u003csup\u003e\u003cem\u003e1\u003c/em\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"6\" nameend=\"c6\" namest=\"c1\"\u003e \u003cp\u003eBaseline characteristics\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAge, years\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e48.6 (36.2, 59.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e48.5 (34.7, 59.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e49.7 (36.6, 60.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e48.6 (36.4, 59.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e47.9 (36.9, 57.4)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eGender\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFemale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e19,418 (36%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4,901 (36%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4,838 (36%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4,899 (37%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e4,780 (36%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMissing\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e350\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e64\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e61\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e115\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e110\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eBody mass index, kg/m\u0026sup2;\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e29.6 (25.1, 35.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e28.8 (24.4, 34.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e29.3 (24.8, 35.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e29.8 (25.4, 35.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e30.5 (26.0, 36.2)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMissing\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e9,700\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2,636\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2,441\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2,644\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1,979\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eCOVID-19\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e14,327 (30%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1,369 (12%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2,491 (21%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3,599 (30%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e6,868 (53%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMissing\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5,486\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1,626\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1,681\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1,535\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e644\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eDiagnosis Category\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAspiration pneumonitis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1,106 (2.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e381 (2.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e335 (2.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e239 (1.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e151 (1.2%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAsthma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2,879 (5.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e812 (6.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e706 (5.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e748 (5.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e613 (4.9%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBacterial pneumonia\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2,706 (5.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e986 (7.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e870 (6.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e584 (4.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e266 (2.1%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNonrespiratory or chronic respiratory diagnosis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3,814 (7.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1,244 (9.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1,013 (7.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e918 (7.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e639 (5.1%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOther acute respiratory diagnosis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e18,853 (37%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5,534 (43%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5,186 (40%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4,575 (36%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e3,558 (28%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTrauma or burn\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5,971 (12%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1,633 (13%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1,760 (14%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1,553 (12%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1,025 (8.2%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eViral pneumonia\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e15,828 (31%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2,407 (19%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3,158 (24%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4,014 (32%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e6,249 (50%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMissing\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2,879\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e512\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e481\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e878\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1,008\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eImmunocompromised\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3,778 (7.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e808 (6.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1,128 (8.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1,053 (8.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e789 (6.3%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMissing\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2,879\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e512\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e481\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e878\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1,008\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eHistory of CNS dysfunction\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2,174 (4.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e606 (4.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e630 (4.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e567 (4.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e371 (3.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMissing\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2,879\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e512\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e481\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e878\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1,008\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c6\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003ePre-ECMO Disease Severity\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003ePaCO₂, mmHg\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e58.0 (47.0, 74.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e58.0 (46.5, 75.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e58.0 (47.0, 74.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e57.0 (46.0, 73.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e59.0 (48.0, 74.0)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMissing\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e12,464\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2,478\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3,085\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3,241\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e3,660\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003ePeak inspiratory pressure, cm H₂O\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e33.0 (29.0, 38.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e34.0 (29.0, 39.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e33.0 (29.0, 38.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e33.0 (29.0, 38.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e33.0 (29.0, 37.0)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMissing\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e23,137\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5,249\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5,688\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e5,998\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e6,202\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eMean airway pressure, cm H₂O\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e21.0 (17.0, 25.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e21.0 (17.0, 25.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e21.0 (17.0, 25.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e20.0 (17.0, 24.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e21.0 (17.0, 24.0)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMissing\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e30,906\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7,197\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7,797\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e7,699\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e8,213\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSerum lactate, mmol/L\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.9 (1.2, 3.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.1 (1.3, 4.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.9 (1.2, 3.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.9 (1.2, 3.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.7 (1.1, 2.8)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMissing\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e29,421\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7,026\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7,437\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e7,695\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e7,263\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eCardiac arrest\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4,421 (8.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1,412 (10%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1,209 (8.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e964 (7.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e836 (6.2%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNeuromuscular blockade\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e27,415 (51%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6,464 (48%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6,480 (48%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e6,915 (51%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e7,556 (56%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003ePulmonary vasodilator\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e11,324 (21%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2,636 (20%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2,443 (18%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2,949 (22%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e3,296 (24%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eBicarbonate infusion\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5,534 (10%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1,821 (13%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1,516 (11%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1,195 (8.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1,002 (7.4%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eProne positioning\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e12,122 (22%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2,238 (17%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2,607 (19%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2,962 (22%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e4,315 (32%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eVasopressor use\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e30,604 (57%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e8,211 (61%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7,753 (57%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e7,358 (54%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e7,282 (54%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eTime from intubation to ECMO, hours\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e37.0 (9.0, 120.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e31.0 (8.0, 106.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e34.0 (8.0, 116.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e40.0 (9.0, 126.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e49.0 (11.0, 127.0)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMissing\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e10,576\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2,224\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2,595\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2,660\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e3,097\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eRESP score total\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.0 (0.0, 4.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.0 (0.0, 4.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.0 (0.0, 4.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2.0 (0.0, 4.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e2.0 (0.0, 4.0)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMissing\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e29,691\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6,846\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7,186\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e7,635\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e8,024\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"6\"\u003e\u003csup\u003e\u003cem\u003e1\u003c/em\u003e\u003c/sup\u003eMedian (Q1, Q3); n (%)\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eAdjusted hierarchical logistic regression of center annual VV ECMO-day-volume and in-hospital mortality\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eParameter\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e95% CI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eCenter VV ECMO volume (per doubling of ECMO-day-volume)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.96\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.93, 0.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eCOVID-19 (yes vs no)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4.19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.87, 6.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eRESP score (per point increase)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.86\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.85, 0.87\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eCalendar year (per year)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.97, 0.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eVolume \u0026times; COVID-19 interaction\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCenter VV ECMO volume (per doubling of ECMO-day-volume) * Yes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.93\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.90, 0.97\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"4\"\u003eAbbreviations: CI\u0026thinsp;=\u0026thinsp;Confidence Interval, OR\u0026thinsp;=\u0026thinsp;Odds Ratio\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e "}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":true,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Extracorporeal Membrane Oxygenation, Extracorporeal Life Support Organization Registry, Volume-Outcome Relationship, Center Experience","lastPublishedDoi":"10.21203/rs.3.rs-8770342/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8770342/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003ePurpose\u003c/h2\u003e \u003cp\u003eExpanding availability of venovenous ECMO (VV ECMO) and increasing frequency of prolonged ECMO runs challenge standard definitions of center experience based solely on case counts. We examined the relationship between center experience and patient outcome using annual VV ECMO-day-volume (total days of VV ECMO support per center-year), a metric that integrates both procedural frequency and exposure to prolonged support and may better reflect institutional expertise.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eWe examined adult patients in the Extracorporeal Life Support Organization (ELSO) Registry supported with VV ECMO between 2014 and 2024. The association between center experience, in-hospital mortality, and ECMO-related complications were examined using mixed-effect models adjusted for illness severity, calendar year, and clustering within centers.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eAmong 54,036 patients, center experience was independently associated with lower in-hospital mortality and fewer ECMO-related complications. Each doubling of annual VV ECMO-days was associated with a 4% reduction in the odds of death (OR 0.96; 95% CI, 0.93\u0026ndash;0.98). The association was most pronounced among centers with longer average ECMO run duration (\u0026gt;\u0026thinsp;18 days; OR 0.95; 95% CI, 0.91\u0026ndash;0.99), whereas no significant volume-mortality association was observed in centers with shorter average run duration. Center learning was observed to increase quickly and plateau as experience is gained, allowing for newly established centers to rapidly achieve expertise.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eHigher VV ECMO-day-volume is associated with lower in-hospital mortality and fewer ECMO-related complications. This metric incorporates frequency and duration of support and provides a more comprehensive measure of center experience than case counts alone.\u003c/p\u003e","manuscriptTitle":"How much ECMO is enough? Center experience, learning curves, run durations, and outcomes in venovenous ECMO: An ELSO registry analysis","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-12 13:45:54","doi":"10.21203/rs.3.rs-8770342/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"fdd8791e-e28b-4976-8d40-47d25f1375e5","owner":[],"postedDate":"February 12th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-04-16T21:08:02+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-12 13:45:54","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8770342","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8770342","identity":"rs-8770342","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}