Characteristics and outcomes of ECMO cannula-related infections: a European multicenter retrospective study

Characteristics and outcomes of ECMO cannula-related infections: a European multicenter retrospective study | 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 Characteristics and outcomes of ECMO cannula-related infections: a European multicenter retrospective study Sofia Ortuno, Nicolas Massart, Charles Vidal, Etienne de Montmollin, and 10 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3940633/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 Only few data regarding epidemiology and management of ECMO cannula-related infections (CRIs) exist. The aim of our study was to describe their epidemiology and prognosis, and to evaluate factors associated with outcome. Methods We performed a multicenter retrospective study in 12 European ICUs, including patients with CRI, defined as a clinical suspicion plus a positive bacterial sample of ECMO-cannulation site. Primary objective was to describe CRI characteristics and outcomes. Secondary objectives were to evaluate the rates of infection recurrence, their risk factors, and to evaluate the impact of antimicrobial treatment duration on outcome. Results During the study period, 124 patients with CRI (78 having concomitant positive blood culture with the same pathogen) were included. Pathogens responsible for infections were predominantly Enterobacteriaceae, coagulase-negative Staphylococcus and Enterococcus spp., and 40% of episodes were polymicrobial. Rates of infection recurrence was 24% and ICU-mortality rate was 50%. Whereas veno-venous ECMO (as compared to veno-arterial ECMO), and therefore ECMO duration was associated with infection recurrence, characteristics of CRI and its management (and in particular duration of antimicrobial treatment) were not associated with recurrence. Patients with antibiotic course ≤8 days had similar infection recurrence rate and outcomes (including mortality) than patients with prolonged (>8 days) antibiotic course. Conclusion CRIs are frequently associated with BSI and frequently polymicrobial. Main risk factor of infection recurrence is ECMO duration. Duration of antimicrobial treatment for CRI ≤8 days is not associated with an increased risk of recurrence or death, as compared to longer treatment. ECMO cannula-related infection hospital-acquired infection bloodstream infection Figures Figure 1 Take Home Message ECMO cannula-related infections are frequently polymicrobial and associated with bloodstream infection, and are mainly due to Enterobacteriaceae, coagulase-negative Staphylococus and Enterococcus spp.. Recurrence is associated with ECMO duration, and an antimicrobial duration ≤8 days is neither associated with infection recurrence nor outcome. Introduction Extracorporeal membrane oxygenation (ECMO) complications include hemorrhage, thrombosis, hemolysis, neurological events and infections [ 1 ]. Infections are one of the main complications, occurring in 42 to 64% of ECMO patients [ 2 – 6 ]. Ventilator-associated pneumonia (VAP) is the most frequent one, but other ECMO-related infections (cannula-related infection (CRI), associated or not with bloodstream infection (BSI)) are frequent and may lead to increased morbidity/mortality. However, data regarding CRI are scarce. In a retrospective large cohort of nosocomial infections in ECMO-patients, Schmidt et al. found that CRI represents 10% of them [ 2 ], whereas Grasselli et al. found a 3% rate of CRI [ 5 ]. Discriminating cannula colonization from infection is sometimes challenging, since traditional clinical and biological signs of infection are frequently absent in ECMO patients. Additionally, little is known about bacterial epidemiology of CRI, since published studies mostly focus on pediatric patients [ 7 , 8 ]. In addition, no consensus on diagnostic criteria or management of CRI exists, and little is known about their impact on prognosis. Data showed that long ECMO run (up then 10 days) were associated with increased rate of CRI, and that these infections were associated with prolonged hospital stay [ 8 , 9 ]. Practices regarding treatment of these infections are heterogeneous [ 10 ], and optimal duration of antimicrobial treatment is unknown [ 11 , 12 ]. We therefore design this study to describe the epidemiology, management and prognosis of CRI, and to determine factors associated with outcomes, with a particular focus on the duration of antimicrobial treatment. Methods Patients Patients requiring a veno-arterial- (VA) or veno-venous (VV)-ECMO for respectively refractory cardiogenic shock or refractory acute respiratory distress syndrome (ARDS) for more than two days, and having developed an ECMO-related infection, as defined below, between January 2012 and December 2022 in one of the 12 participating ICUs, were included. Investigators from the 12 participating ICUs retrospectively collected data from their most recent consecutive patients with CRI. Investigators were invited to include, if possible, at least 10 patients per center, with the possibility to include more patients. Study objectives The primary objective was to describe the characteristics of CRI, as defined hereafter, and its outcomes. Secondary objectives were to evaluate the rate of infection recurrence (see definition below), and to evaluate the impact of antimicrobial treatment duration on the outcome. Definitions CRI was defined as clinically suspected infection at one of the ECMO-cannulation site, confirmed by a positive bacteriological cannulation-site sampling. CRI was suspected in case of inflammatory signs at the site of cannulation (i.e., redness, swelling), with or without purulent drainage, with or without general signs of infection (fever, hyperleukocytosis); in case of isolated purulent drainage from the cannulation site; in case of clinical suspicion of infection (fever, hyperleukocytosis) with inflammation of the cannulation site; or in case of hemodynamic worsening (introduction of vasopressors or increase of more than 30% of their dose) suspected of being due to an infection, with inflammation of the cannulation site. In all these cases, physicians performed bacteriological sampling of cannulation site using swab or collection drainage using a syringe. CRI was confirmed when cannulation site sampling retrieved pathogens (bacteria or fungus). CRI was considered complicated by BSI when same pathogen (same species, same resistance profile) was retrieved from blood culture and cannula sampling at the same time (within a 48-hrs time window). Patients with infection of the cannulation site occurring after ECMO removal were not included. Patients with cannula-colonization (positive bacterial sampling of cannula without local or general signs of infection) were not included. Patients with primary BSI (BSI without portal of entry and without bacteriologically-proven CRI) were not included in the present study. Although ECMO cannula may be the hidden portal of entry of such infection, we choose not to include them since we wanted to focus on CRI and its management. Recurrence of CRI was defined as a new infection episode in a patient having developed a first episode, whatever the pathogen responsible for the second episode. Second episodes occurring after ECMO removal were also considered as recurrences. In these cases, CRI recurrence was considered if the patient had local signs of CRI plus a pathogen retrieved at cannulation site. Patients without second episode of CRI were considered as cured from infection. Duration of antimicrobial treatment for the first episode was defined as the total duration of antimicrobial treatment effective against the pathogen(s) responsible for ECMO-related infection, including time on antibiotics given for another infection than ECMO-related infection. Antimicrobial treatment was considered appropriate when the incriminated microorganism was susceptible to the antibiotic administered on antimicrobial susceptibility testing. As an example, if a patient received amoxicillin for a CRI due to wild-type E. coli had to be switched at day 7 of amoxicillin to a 7-day course of piperacillin-tazobactam for Pseudomonas aeruginosa VAP episode, he was considered to have received 14 days of effective antibiotic treatment for the initial CRI. Multi drug resistant organism (MDRO) was considered according to the definition of the European Society of Clinical Microbiology and Infectious Disease, as pathogens resistant to more than one antimicrobial agent [ 13 ]. We also recorded local cannula site complications after decannulation (hemorrhage, infection and thrombosis), mortality rate before ECMO weaning, duration of ECMO support and ventilatory support, length of ICU stays and ICU mortality rate. We then determined factors associated with recurrence. Statistical analysis Categorical variables are expressed as percentages and continuous variables as median ant interquartile range (IQR). Chi-Square test and Fisher exact test were used to compare categorical variables and Mann-Whitney U test or Wilcoxon test for continuous variables. To determine the risk factors of infection recurrence, we performed uni-and multivariable analysis of factors associated with infection recurrence, taking into account death and ECMO decannulation as competing events, using a Fine and Gray analysis. Values were expressed as subdistribution hazard ratio (sdHR). In a second analysis, factors associated with recurrence were compared using a Poisson regression model. Values were expressed in incidence rate ratio (IRR), taking in account the time of exposition to risk, i.e., ECMO duration in our study. To explore factors associated with ICU-mortality, uni- and multivariable analysis of factors associated with ICU death were performed using a Cox proportional model. Variables with a p-value < 0.2 in univariable analysis were entered into the model. To further explore the impact of antimicrobial treatment duration of the first episode on outcome, we selected the 105 patients who survived at least 8 days after infection onset, and split them into 2 categories: short duration of antimicrobial treatment, i.e. ≤8 days, and long duration of treatment, i.e. >8 days. Uni- and multivariable analysis of factors associated with death was performed in this population. Analyses were performed with an alpha risk of 5%. Statistical analysis was performed with the statistical software R 4.1.1. Ethics In accordance with current French law, informed written consent for demographic, physiologic and hospital-outcome data analyses was not obtained because this observational study did not modify existing diagnostic or therapeutic strategies. Nonetheless, patients and/or relatives were informed about the anonymous data collection and told that they could decline inclusion. The protocol was approved by the ethics committee of the Société de Réanimation de Langue Française (registration no. CE-SRLF-22-074) and the database is registered by the Commission Nationale de l’Informatique et des Libertés (CNIL, registration no 2228484v 0). Results Between January 2012 and December 2022, 124 ECMO patients with CRI were included. Baseline patients’ characteristics are presented in Table 1 . Patients were predominantly male (67%), young (median (IQR) age 52 (40–61 years), and 26 (21%) were immunosuppressed. We included higher number of patients with VA-ECMO (n=86, 79%) than patients with VV-ECMO (n=38, 31%). CRI characteristics CRI characteristics (clinical presentation, pathogens responsible for infection and management of infection) are displayed in Table 2 . Among the 124 episodes of CRI, 78 (63%) had blood culture positive for the same pathogen that the one retrieved at the cannula site. Although the majority (59%) had inflammatory signs at cannula insertion site, 56 had no local sign of infection: among them, 15 had CRI and 41 had CRI with concomitant BSI. Pathogens responsible for ECMO-related infection are shown on Table 2: Enterobacteriaceae were the most common pathogen responsible for these infections, followed by coagulase-negative Staphylococcus and Enterococcus spp.. Interestingly, 50 (40%) of episodes were polymicrobial. Details on treatment are reported in Table 2 : median (IQR) duration of antimicrobial treatment was 10 (7–15) days, similar in patients with and without concomitant BSI (median (IQR) durations of 10 (7–15) and 10 (7–16) days, respectively, p = 0.6); 29 patients (23%) were treated using combination therapy during all the duration of antimicrobial treatment, 37 (30%) had surgery for their CRI, and 25 (20%) had change of cannulation site for the infection. Outcomes Among the 124 patients, 29 had a second episode of infection that occurred after a median (IQR) of 6 (0–11) days after the end of antimicrobial treatment of the first episode. Details regarding the outcomes and pathogens responsible for the second episode of infection are reported in Table 3 . Cumulative incidences of infection recurrence, death and ECMO decannulation are shown on Figure 1 . Uni and multivariable analyses of factors associated with mortality are shown in eTable 1 : chronic cardiac failure and septic shock complicating CRI were associated with ICU-mortality, as was coagulase-negative Staphylococcus as the pathogen responsible for CRI. Factors associated with infection recurrence Tables 1 and 2 display the baseline and infection characteristics of CRI in patients with and without infection recurrence. Baseline characteristics of patients were similar in the 2 groups, as well as most infection characteristics: timing of infection, type of infection (CRI with or without concomitant BSI). However, coagulase-negative Staphylococcus was less frequently responsible for infection in patients who will experience infection recurrence. Among the 29 second episodes, only 7 (25%) were due to the same pathogen. Interestingly, characteristics of treatment (duration, use of combination therapy, cannulation site change and need for surgery) were similar in both groups. To explore the risk factors of infection recurrence, we performed uni-and multivariable analysis of factors associated with recurrence of infection, taking into account death and decannulation as competing events (see eTable 2 in the online supplement). The only factors associated with recurrence of infection were veno-venous ECMO (sdHR 14.7, 95% CI 1.58–137), whereas the femoro-femoral site of cannulation (as compared to femoro-jugular and other cannulation site) was a protective factor for infection recurrence (sdHR 0.05, 95% CI 0.01–0.4). Neither type of pathogen responsible for infection nor infection characteristics (associated BSI, inflammatory signs at cannula insertion site, septic shock) and infection management (appropriate empiric antimicrobial treatment, prolonged duration of antimicrobial treatment beyond 8 days, combination therapy, change of cannula) were associated with recurrence of infection ( eTable 2 , online supplement). Similar results were obtained when using a Poisson regression model to explore factors associated with recurrence: infection management was not associated with infection recurrence ( eTable 3 , online supplement). Impact of antimicrobial treatment duration To explore specifically the impact of the duration of antimicrobial treatment of the first episode on outcomes, patients were grouped, according to the duration of treatment (short, i.e. ≤8 days or long, i.e. >8 days) and compared. For this analysis, patients who died before day 8 were excluded, thus 105 patients were included in this analysis. Results are presented in the online supplement: 35 patients had a short (≤8 days) duration of antimicrobial treatment and 70 had a long (>8 days) duration of antimicrobial treatment; with median (IQR) durations of treatment of 7 (6–8) and 15 (11–16) days, respectively. Characteristics of patients were similar between patients with short or long duration of treatment (see eTable 4 , online supplement). Data regarding infection characteristics, microbiological data and management of infection according to duration of antimicrobial treatment are presented in eTable 5 (online supplement). Patients with long duration of treatment had more frequently surgery on cannulation site than patients with short duration; however the outcomes were similar between the 2 groups; particularly the recurrence rates and ICU mortality rates were similar. In a multivariable cox model analysis evaluating factors associated with death in this specific population (patients surviving at least 8 days after ECMO-related infection onset), prolonged (>8 days), duration of antimicrobial treatment and combination therapy had no protective effect on death in either univariate or multivariable analysis, with respective HR (95% CI) in multivariable analysis of 1.00 (0.49–2.07), and 0.78 (0.34–1.80). Discussion In this multicenter study of 124 patients with ECMO-related infection, we found that 2/3 of patients with cannula-related infection had associated BSI with the same pathogen. Main pathogens responsible for ECMO-related infection were Enterobacteriaceae , coagulase-negative Staphylococcus and Enterococcus spp., and 40% of these infections involved more than one pathogen. Outcomes of ECMO-related infection were globally poor, since 1/4 experienced infection recurrence. While veno-venous ECMO (as compared to veno-arterial), and therefore ECMO duration was associated with infection recurrence, characteristics of CRI and its management (duration of antimicrobial treatment, use of combination, therapy) were not associated with infection recurrence. More specifically, patients with a short duration of antimicrobial treatment (≤ 8 days) had similar outcomes (including infection recurrence) than patients with long duration of antimicrobial treatment (> 8 days). Last, factors associated with death were cardiac comorbidity, CRI complicated by septic shock, and coagulase-negative Staphylococcus as the pathogen responsible for infection. This is the largest multicenter cohort regarding CRI in adults, and the first that describes its epidemiology and outcomes. Among the previous observational studies evaluating infection in ECMO patients, small case series showed that Enterobacteriaceae and non-fermenting Gram negative bacilli were the most frequent pathogen involved in ECMO-associated infection [ 8 , 11 , 14 ], while other highlighted the central role of Enterococcus spp. in nosocomial infections [ 2 , 5 ]. However, most infections recorded in these studies were not specifically ECMO-related, i.e., included VAP or urinary tract infection [ 2 , 5 ]. Among them, only few CRIs were documented [ 2 , 5 , 14 , 15 ]. Nevertheless, our findings are in agreement with these results, since the main pathogens responsible for CRI were Enterococcus spp. and Enterobacteriaceae. Of note, coagulase-negative Staphylococcus represented 30% of CRI in our study, whereas a previous study that included 21 CRI episodes reported a 19% rate of coagulase-negative Staphylococcus [ 2 ]. Last, pathogens responsible for infections in our study were less frequently MDRO than in others, [ 5 ], probably related to the relatively low rate of MDRO in French ICUs, as compared to other European countries. Few data regarding outcomes of patients with CRI are available. A small single center cohort study showed that ECMO patients who developed infection had significantly higher mortality rate than patients without infection (40.4% vs. 20.0%, respectively, p = 0.037). However this study reported only 3 CRIs and 5 BSIs [ 5 ]. Moreover, these results were not confirmed in others [ 2 , 11 , 14 ]; Schmidt et al., in their cohort study of 220 ECMO patients, found that infected and non-infected patients had similar mortality rate, 51% vs 62%, p = 0.15, respectively. Although our study was not designed to compare patients with and without CRI, ICU-mortality rate of our patients was similar to that observed in the Schmidt study [ 2 ]. One of the originalities of our work was to focus on other outcomes, and in particular infection recurrence. Interestingly, none of the characteristics of infection was associated with an increased risk of recurrence: neither severity of infection (septic shock), nor positive blood cultures, nor inflammatory signs at ECMO cannulation site. Only coagulase-negative Staphylococcus as pathogen responsible for infection was associated with a decreased risk of infection recurrence. Management of CRI may be challenging, with specific issues regarding antimicrobial treatment duration and need for cannula change to cure infection. To date, no data focusing on these issues exist in the literature. We showed here that neither antimicrobial treatment duration, nor combination therapy, nor cannula change were associated with an increased risk of infection recurrence. Specifically, patients with short duration of antimicrobial treatment (≤ 8 days) had similar outcomes than patients with prolonged (> 8 days) duration of antimicrobial treatment. This observation is in line with recent data on infections in ICU patients, which showed that short durations of antimicrobial treatment, even for severe infections, are not associated with worse outcomes [ 16 ]. The only factors associated with infection recurrence were VV-ECMO (as compared to VA-ECMO), whereas femoro-femoral cannulation (mostly used for VA-ECMO in our patients) was a protective factor. These observations underlined the role of ECMO duration on infection outcome: patients with VV-ECMO had a longer duration of ECMO support than VA-ECMO patients, thus are more exposed to the risk of recurrence. Our study has several limitations that should be underlined. Firstly, due to its retrospective design, we cannot be sure that all parameters recorded in the present study were reported in the patient chart. Moreover, the retrospective design limits the interpretation of outcomes. Secondly, there is no clear definition of CRI and no consensus on the diagnosis process (which sample, which type of sampling…). Therefore, definition may vary from one center to another, and we cannot be sure that some patients included had cannula colonization rather than infection. However, since we predefined CRI as clinical suspicion of infection plus positive bacterial sample, we think that the population of patients included here is homogeneous. Thirdly, we excluded patients with primary BSI and therefore could have miss some patients with CRI: indeed, some patients with primary BSI could have in fact CRI (not diagnosed for several reasons). However, we preferred to include only patients with bacteriologically proven CRI. Fourthly, although we performed 2 different analyses exploring factors associated with infection recurrence, we could have missed confounding factors. However, the use of Fine and Gray analysis lowers this bias, and allowed us to avoid the 2 major cofounders for infection recurrences, namely death and decannulation. Finally, even if this is a multicenter cohort, our study mostly performed in French ICUs. We cannot be sure that similar results would be observed in ICUs with different bacterial ecologies. Conclusion CRI is frequently polymicrobial, mainly due to Enterobacteriaceae, coagulase-negative Staphylococci and Enterococcus spp, and frequently associated with BSI. Risk of infection recurrence is mainly driven by ECMO duration, whereas infection management, and more specifically duration of antimicrobial treatment does not influence infection recurrence and mortality. Abbreviations BSI bloodstream infection CRI cannula-related infection ECMO extracorporeal membrane oxygenation ICU intensive care unit IQR interquartile range IRR incidence rate ratio sdHR subdistribution hazard ratio VAP ventilated-associated pneumonia VA veno-arterial VV veno-venous Declarations Ethical approval In accordance with current French law, informed written consent for demographic, physiologic and hospital-outcome data analyses was not obtained because this observational study did not modify existing diagnostic or therapeutic strategies. Nonetheless, patients and/or relatives were informed about the anonymous data collection and told that they could decline inclusion. The protocol was approved by the ethics committee of the Société de Réanimation de Langue Française (registration no. CE-SRLF-22-074) and the database is registered by the Commission Nationale de l’Informatique et des Libertés (CNIL, registration no 2228484v 0). Consent for publication Not applicable Availability of data and materials The datasets generated during the current study are available from the corresponding author on reasonable request. Competing interest . C. - E. L.received lecture fees from Merck, Aerogen and AdvanzPharma, and grant from Eumedica and Merck, all outside the submitted work. The other authors have no conflicts of interest to declare in relationship to this manuscript. Funding . None. Authors’ contribution SO and CEL designed the study, collected, compiled, analyzed and interpreted the data and wrote the manuscript. SO, NM, CV, EdM, NM, SH, AB, PM, FB, SH, BA, HR and CEL collected data. NM performed statistical analysis, analyzed and interpreted the data. All authors approved the final version of the manuscript. Acknowledgments Not applicable Contributors Groupe Hospitalier Pitié-Salpêtrière, Paris, France : Juliette Chommeloux, Guillaume Hékimian, Alain Combes, Nima Djavidi, Aymeric Lancelot, Pauline Dureau, Guillaume Lebreton. CH de Saint-Brieuc, France : Pierre Fillâtre. Centre Hospitalier Universitaire de La Réunion, Saint Denis, France : Laura Rivoire, Nicolas Allou, Radj Cally . Hôpital Bichat-Claude Bernard, Paris, France : Hermann Do Rego, Mariem Dlela, Marc Doman. Centre Hospitalier universitaire de Rennes : Christophe Camus, Erwan Flecher. Hôpital Louis Pradel, Lyon, France : Jean-Luc Fellahi, Matteo Pozzi, Matthias Jacquet-Lagreze. Hôpitaux Universitaires de Genève, Genève, Switzerland : Raphaël Giraud, Karim Bendjelid. Hôpital Henri Mondor, Créteil, France : Keyvan Razazi, Armand Mekontso-Dessap. Hôpital universitaire de Lille, Lille, France : Anahita Rouzé, Saad Nseir, Thibault Duburcq. Hôpital Nord, Marseille, France : Christophe Guervilly, Jean-Marie Forel. Hôpital Haut Leveque, Pessac, France : Eline Bonnardel. Hôpital Universitaire de Bruxelles, Brussels, Belgium : Fabio S Taccone. References Pineton de Chambrun M, Bréchot N, Combes A (2018) Mechanical circulatory devices in acute heart failure. Curr Opin Crit Care 24:286–291. https://doi.org/10.1097/MCC.0000000000000520 Schmidt M, Bréchot N, Hariri S et al (2012) Nosocomial infections in adult cardiogenic shock patients supported by venoarterial extracorporeal membrane oxygenation. Clin Infect Dis 55:1633–1641. https://doi.org/10.1093/cid/cis783 Danial P, Hajage D, Nguyen LS et al (2018) Percutaneous versus surgical femoro-femoral veno-arterial ECMO: a propensity score matched study. Intensive Care Med 44:2153–2161. https://doi.org/10.1007/s00134-018-5442-z Aubron C, Cheng AC, Pilcher D et al (2013) Infections acquired by adults who receive extracorporeal membrane oxygenation: risk factors and outcome. Infect Control Hosp Epidemiol 34:24–30. https://doi.org/10.1086/668439 Grasselli G, Scaravilli V, Di Bella S et al (2017) Nosocomial Infections During Extracorporeal Membrane Oxygenation: Incidence, Etiology, and Impact on Patients’ Outcome. Crit Care Med 45:1726–1733. https://doi.org/10.1097/CCM.0000000000002652 Winiszewski H, Boyadjian C, Besch G et al (2022) Extracorporeal Membrane Oxygenation Cannula-Related Infections: Epidemiology and Risk Factors. ASAIO J 68:571–576. https://doi.org/10.1097/MAT.0000000000001505 Bizzarro MJ, Conrad SA, Kaufman DA et al (2011) Infections acquired during extracorporeal membrane oxygenation in neonates, children, and adults. Pediatr Crit Care Med 12:277–281. https://doi.org/10.1097/PCC.0b013e3181e28894 Vogel AM, Lew DF, Kao LS, Lally KP (2011) Defining risk for infectious complications on extracorporeal life support. J Pediatr Surg 46:2260–2264. https://doi.org/10.1016/j.jpedsurg.2011.09.013 Hsu M-S, Chiu K-M, Huang Y-T et al (2009) Risk factors for nosocomial infection during extracorporeal membrane oxygenation. J Hosp Infect 73:210–216. https://doi.org/10.1016/j.jhin.2009.07.016 Glater-Welt LB, Schneider JB, Zinger MM et al (2016) Nosocomial Bloodstream Infections in Patients Receiving Extracorporeal Life Support: Variability in Prevention Practices: A Survey of the Extracorporeal Life Support Organization Members. J Intensive Care Med 31:654–669. https://doi.org/10.1177/0885066615571540 Burket JS, Bartlett RH, Vander Hyde K, Chenoweth CE (1999) Nosocomial infections in adult patients undergoing extracorporeal membrane oxygenation. Clin Infect Dis 28:828–833. https://doi.org/10.1086/515200 Sun W, Wang S, Chen Z et al (2020) Impact of high mechanical shear stress and oxygenator membrane surface on blood damage relevant to thrombosis and bleeding in a pediatric ECMO circuit. Artif Organs 44:717–726. https://doi.org/10.1111/aor.13646 Magiorakos A-P, Srinivasan A, Carey RB et al (2012) Multidrug-resistant, extensively drug-resistant and pandrug-resistant bacteria: an international expert proposal for interim standard definitions for acquired resistance. Clin Microbiol Infect 18:268–281. https://doi.org/10.1111/j.1469-0691.2011.03570.x Sun H-Y, Ko W-J, Tsai P-R et al (2010) Infections occurring during extracorporeal membrane oxygenation use in adult patients. J Thorac Cardiovasc Surg 140:1125–1132e2. https://doi.org/10.1016/j.jtcvs.2010.07.017 Kim HS, Park S, Ko HH et al (2021) Different characteristics of bloodstream infection during venoarterial and venovenous extracorporeal membrane oxygenation in adult patients. Sci Rep 11:9498. https://doi.org/10.1038/s41598-021-89108-4 Mokrani D, Chommeloux J, Pineton de Chambrun M et al (2023) Antibiotic stewardship in the ICU: time to shift into overdrive. Ann Intensive Care 13:39. https://doi.org/10.1186/s13613-023-01134-9 Tables Table 1. Baseline characteristics according to the recurrence of infection or not Overall population n = 124 No infection recurrence n = 95 Infection recurrence n = 29 p Baseline characteristics Male sex 83 (67) 62 (65) 21 (72) 0.6 Age, years 52 (40–61) 54 (41–63) 50 (40–58) 0.3 BMI, kg/m² a 28 (25–35) 28 (24–33) 33 (26–40) 0.06 Pre-existing conditions Chronic cardiac disease Chronic renal disease Chronic respiratory disease Diabetes mellitus Peripheral vascular disease Immunocompromised b Ongoing pregnancy 45 (36) 17 (14) 19 (15) 38 (31) 21 (17) 26 (21) 1 (1) 36 (38) 15 (16) 17 (18) 34 (36) 15 (16) 20 (21) 0 9 (31) 2 (7) 2 (7) 4 (14) 6 (21) 6 (21) 1 (3) 0.7 0.4 0.3 0.04 0.7 1 0.2 Admission SAPS2 Admission SOFA score 42 (30–57) 8 (4–12) 41 (31–56) 9 (4–12) 45 (27–61) 8 (5–12) 0.9 0.8 ECMO characteristics Veno-venous ECMO 38 (31) 23 (24) 15 (52) 0.01 Percutaneous cannulation 69/104 (66) 53/82 (65) 16/22 (73) 0.6 Ongoing antibiotics at ECMO start 52 (42) 37 (39) 15 (51) 0.3 SARS-Cov2 infection 27 (22) 16 (17) 11 (38) 0.03 Indication of ECMO Cardiogenic shock ARDS Post-cardiotomy Septic shock E-CPR Other 64 (52) 38 (31) 12 (10) 3 (2) 5 (4) 2(2) 52 (55) 25 (26) 10 (11) 2 (2) 4 (4) 2 (2) 12 (41) 13 (45) 2 (7) 1 (3) 1 (3) 0 0.5 Cannulation by mobile unit 27/119 (23) 20 (22) 7 (27) 0.8 Data are expressed as median (IQR) or n (%). Abbreviations: BMI : Body Mass Index ; SAPS2 : Simplified acute physiology score 2; SOFA : Sepsis-related organ failure assessment ; ECMO : Extra-corporeal membrane oxygenation; E-CPR: Extracorporeal cardiopulmonary resuscitation. a data available for 114/124 patients b Solid organ transplant recipients, active hematological malignancy or receiving immunosuppressant drug (including corticosteroids at a dose≥0.5 mg/kg/d for≥1 month) Table 2: First ECMO-related infection episode characteristics Overall population n = 124 No infection recurrence n = 95 Infection recurrence n = 29 p Time between ICU admission and first episode, days 9 (6–16) 8 (6–14) 13 (6–19) 0.08 Concomitant BSI 78 (63) 61 (64) 17 (59) 0.745 Septic shock 69/121 (58) 56/93 (60) 13/27 (48) 0.4 Inflammatory sign at cannula insertion site 68/116 (59) 50/88 (57) 18/28 (64) 0.6 Pathogen Staphylococcus aureus Methicillin resistant Coagulase-negative Staphylococcus Methicillin resistant Streptococcus spp Enterococcus spp Vancomycin resistant Enterobacteriaceae ESBL-producer Carbapenem resistant Non fermenting Gram-negative bacilli Multiresistant NF-GNB Anaerobe MDRO Polymicrobial 6 (5) 1 (1) 37 (30) 6 (5) 4 (3) 38 (31) 0 64 (52) 15 (12) 4 (3) 23 (18) 5 (4) 7 (6) 21 (17) 50 (40) 5 (5) 1 (1) 33 (35) 4 (4) 3 (3) 28 (30) 0 28 (30) 8 (8) 1 (1) 18 (19) 4 (4) 6 (6) 13 (14) 40 (42) 1 (3) 0 4 (14) 2 (7) 1 (3) 10 (35) 0 16 (62) 7 (24) 3 (10) 5 (17) 1 (3) 1 (3) 8 (28) 10 (35) 1 1 0.04 0.9 1 0.8 0.3 0.05 0.06 1 1 1 0.1 0.6 Time from diagnostic to empiric treatment, days 0 (0–0) 0 (0–0) 0 (0–0) 0.7 Appropriate empiric antimicrobial treatment 77/109 (71) 58 (69) 19 (76) 0.7 Time from infection onset to appropriate antimicrobial treatment, days 2 (0–2) 1 (0–2) 2 (1–3) 0.3 Duration of antimicrobial treatment of first ECMO-related infection, days a 10 (7–15) 10 (7–15) 13 (8–18) 0.06 Combination antimicrobial treatment 29 (23) 25 (26.3) 4 (13.8) 0.3 Oxygenator change during infection 29 (23) 19 (20.0) 10 (34.5) 0.2 Cannula site change for infection 25 (20) 19 (20.0) 6 (20.7) 1.0 Surgery needed for cannula site infection 37 (30) 28 (29.5) 9 (31.0) 1.0 Time to new infection, days b 6 (0–11) – 6 (0–11) Treatment failure is defined as recurrence of ECMO-related infection or death during ICU stay. Data are expressed as median [IQR] or n (%). ECMO, Extra-corporeal membrane oxygenation. ICU, intensive care unit. CRI, Cannula related infection. ESBL, extended spectrum beta-lactamase. NF-GNB, Non fermenting Gram-negative bacilli. MDRO, multidrug resistant organism a data available for 121/124 patients b time from end of antimicrobial treatment of first episode to second episode onset. Table 3: Outcomes Overall population n = 124 Recurrence of infection 29 (23) Time to new infection, days a 6 (0–10) Pathogen responsible for second episode Staphylococcus aureus Coagulase-negative Staphylococcus Streptococcus spp Enterococcus spp. Enterobacteriaceae Non fermenting Gram-negative bacilli MDRO Polymicrobial 0 6 (21) 1 (3) 10 (34) 11 (38) 4 (14) 14 (48) 14 (48) Complications on cannula site after decannulation Cannula site infection Arterial thrombosis Hemorrhage 31 (26) 20 (16) 11 (9) 7 (6) Duration of ECMO support, days 15 (9–26) Duration of invasive mechanical ventilation, days 20 (11–34) ICU length of stay, days 26 (19–45) Death on ECMO 40 (32) ICU mortality rate 62 (50) Data are expressed as median (IQR) or n (%). ECMO, Extra-corporeal membrane oxygenation. ICU, intensive care unit a time from end of antimicrobial treatment of first episode to second episode onset. Supplementary Files CRIFonlinesuplementICM.docx STROBEchecklistOrtuno.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-3940633","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":272213051,"identity":"f78755bb-bc9b-4c6c-8892-d9d6331d76dc","order_by":0,"name":"Sofia Ortuno","email":"","orcid":"","institution":"Sorbonne Université: Sorbonne Universite","correspondingAuthor":false,"prefix":"","firstName":"Sofia","middleName":"","lastName":"Ortuno","suffix":""},{"id":272213052,"identity":"cd703ac0-c65f-4456-972b-408ed8e563f4","order_by":1,"name":"Nicolas Massart","email":"","orcid":"","institution":"CH Saint Brieuc: Centre Hospitalier de Saint-Brieuc","correspondingAuthor":false,"prefix":"","firstName":"Nicolas","middleName":"","lastName":"Massart","suffix":""},{"id":272213053,"identity":"da838d9f-cf5b-4d76-bd1b-589761129cd2","order_by":2,"name":"Charles Vidal","email":"","orcid":"","institution":"CHU de La Réunion: Centre Hospitalier Universitaire de la Reunion","correspondingAuthor":false,"prefix":"","firstName":"Charles","middleName":"","lastName":"Vidal","suffix":""},{"id":272213054,"identity":"f140b568-ad6d-4b03-bffd-d3d93e24bd1b","order_by":3,"name":"Etienne de Montmollin","email":"","orcid":"","institution":"Université Paris Cité Faculté de Santé: Universite Paris Cite Faculte de Sante","correspondingAuthor":false,"prefix":"","firstName":"Etienne","middleName":"","lastName":"de Montmollin","suffix":""},{"id":272213055,"identity":"c8d5f05c-8650-40ed-8aea-b4d288810a39","order_by":4,"name":"Adrien Bouglé","email":"","orcid":"","institution":"Sorbonne Université: Sorbonne Universite","correspondingAuthor":false,"prefix":"","firstName":"Adrien","middleName":"","lastName":"Bouglé","suffix":""},{"id":272213056,"identity":"76cec068-f697-41af-8fd8-4acfc7079be8","order_by":5,"name":"Nicolas Nesseler","email":"","orcid":"","institution":"CHU Rennes: Centre Hospitalier Universitaire de Rennes","correspondingAuthor":false,"prefix":"","firstName":"Nicolas","middleName":"","lastName":"Nesseler","suffix":""},{"id":272213057,"identity":"bdd9ee5d-c279-458a-b081-c581677dda0a","order_by":6,"name":"Frank Bidar","email":"","orcid":"","institution":"Hospices Civils de Lyon","correspondingAuthor":false,"prefix":"","firstName":"Frank","middleName":"","lastName":"Bidar","suffix":""},{"id":272213058,"identity":"5971e21c-46ca-4048-b62b-bf3382e788f4","order_by":7,"name":"Benjamin Assouline","email":"","orcid":"","institution":"Geneva University Hospitals: Hopitaux Universitaires Geneve","correspondingAuthor":false,"prefix":"","firstName":"Benjamin","middleName":"","lastName":"Assouline","suffix":""},{"id":272213059,"identity":"af845aec-044a-4599-a4fd-4ef3d4e8e490","order_by":8,"name":"Paul Masi","email":"","orcid":"","institution":"Hôpital Henri Mondor: Hopital Henri Mondor","correspondingAuthor":false,"prefix":"","firstName":"Paul","middleName":"","lastName":"Masi","suffix":""},{"id":272213060,"identity":"cd328705-7cb5-4264-948c-fc11b8224800","order_by":9,"name":"Samuel Henri","email":"","orcid":"","institution":"CHRU de Lille: Centre Hospitalier Universitaire de Lille","correspondingAuthor":false,"prefix":"","firstName":"Samuel","middleName":"","lastName":"Henri","suffix":""},{"id":272213061,"identity":"37e3a656-37b3-405b-a463-f7d4b762b984","order_by":10,"name":"Sami Hraiech","email":"","orcid":"","institution":"Assistance Publique Hopitaux de Marseille","correspondingAuthor":false,"prefix":"","firstName":"Sami","middleName":"","lastName":"Hraiech","suffix":""},{"id":272213062,"identity":"79b259ad-ffdd-46fc-850d-302d95229dbd","order_by":11,"name":"Hadrien Rozé","email":"","orcid":"","institution":"CHU de Bordeaux: Centre Hospitalier Universitaire de Bordeaux","correspondingAuthor":false,"prefix":"","firstName":"Hadrien","middleName":"","lastName":"Rozé","suffix":""},{"id":272213063,"identity":"5bbc0f68-635a-446c-bd3c-e960b383ce2a","order_by":12,"name":"Francesca Manicone","email":"","orcid":"","institution":"Université Libre de Bruxelles: Universite Libre de Bruxelles","correspondingAuthor":false,"prefix":"","firstName":"Francesca","middleName":"","lastName":"Manicone","suffix":""},{"id":272213064,"identity":"83a15b8c-9ab3-408c-aeaf-d1ca94bdeedb","order_by":13,"name":"Charles-Edouard Luyt","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABb0lEQVRIie3RP0vDQBgG8KccJA6nWa9Y+hlOCi2liF8lpZAs8R8uHSStFOJS94hfouDSweFKoFOLa4UOFSFTkYhLi0W8pEmtFZ0F88Ald5Af73tvgDRp/mKYXJNop4cPgp3wFQC55TGOAPLrRF8nCpC5cAG6SQq/khbdJPgk2k3LD/S6DU2t9dn8rgKFWaWn/a5NoQ4FOamPUVKHj70Z+HFcZNwvuPrAQ7btG3KZIdlrHQ48CnqkE3fgo9w2uTzycjMinOkFVB3R6IysIqPCi4kjZGNW6W3b8RodYcADFhwxMV9RfbdxIEl2kZCyIxvTppxIAn7vI2xsRSxZpUnCTXF3VSXjEApmxWRkQBZNCBtZZ9D7HtjAr1VywqQK9U+vr+RdFPasExrexfXhUZ4QzTVvEZzLiV3Weg9TUcnL0XWCedfOa9qhIDScmGaQl1k9Id8ib73FkZHDUZa/RmD56U8gijqJCL6SNGnSpPm/+QAax4H5ijs+7AAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0001-7424-2705","institution":"Sorbonne University: Sorbonne Universite","correspondingAuthor":true,"prefix":"","firstName":"Charles-Edouard","middleName":"","lastName":"Luyt","suffix":""}],"badges":[],"createdAt":"2024-02-08 17:06:36","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3940633/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3940633/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":51081653,"identity":"0e909dbc-c36f-4409-b4d0-951f9cbb8115","added_by":"auto","created_at":"2024-02-13 19:17:55","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":12938,"visible":true,"origin":"","legend":"\u003cp\u003eLegend not included with this version\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-3940633/v1/3c8ed5f8968fd8adbf4656b4.png"},{"id":52666291,"identity":"09072478-8895-44a5-a0eb-0e9f795d81af","added_by":"auto","created_at":"2024-03-14 08:58:24","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":335882,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3940633/v1/3299096f-4be9-4581-b49c-4b9d0b95234a.pdf"},{"id":51081651,"identity":"c0b57856-cb47-41a9-a36b-bfa36244d555","added_by":"auto","created_at":"2024-02-13 19:17:55","extension":"docx","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":54735,"visible":true,"origin":"","legend":"","description":"","filename":"CRIFonlinesuplementICM.docx","url":"https://assets-eu.researchsquare.com/files/rs-3940633/v1/e4ac7fa9fe6551ac58865fdb.docx"},{"id":51081650,"identity":"85d4d6bd-4abd-46c0-a34d-93da9f7cab92","added_by":"auto","created_at":"2024-02-13 19:17:55","extension":"docx","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":30636,"visible":true,"origin":"","legend":"","description":"","filename":"STROBEchecklistOrtuno.docx","url":"https://assets-eu.researchsquare.com/files/rs-3940633/v1/191ff0e4b77b56e87c7bf359.docx"}],"financialInterests":"","formattedTitle":"Characteristics and outcomes of ECMO cannula-related infections: a European multicenter retrospective study","fulltext":[{"header":"Take Home Message","content":"\u003cp\u003eECMO cannula-related infections are frequently polymicrobial and associated with bloodstream infection, and are mainly due to Enterobacteriaceae, coagulase-negative Staphylococus and Enterococcus spp.. Recurrence is associated with ECMO duration, and an antimicrobial duration \u0026le;8 days is neither associated with infection recurrence nor outcome.\u003c/p\u003e"},{"header":"Introduction","content":"\u003cp\u003eExtracorporeal membrane oxygenation (ECMO) complications include hemorrhage, thrombosis, hemolysis, neurological events and infections [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Infections are one of the main complications, occurring in 42 to 64% of ECMO patients [\u003cspan additionalcitationids=\"CR3 CR4 CR5\" citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Ventilator-associated pneumonia (VAP) is the most frequent one, but other ECMO-related infections (cannula-related infection (CRI), associated or not with bloodstream infection (BSI)) are frequent and may lead to increased morbidity/mortality. However, data regarding CRI are scarce. In a retrospective large cohort of nosocomial infections in ECMO-patients, Schmidt et al. found that CRI represents 10% of them [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e], whereas Grasselli et al. found a 3% rate of CRI [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Discriminating cannula colonization from infection is sometimes challenging, since traditional clinical and biological signs of infection are frequently absent in ECMO patients.\u003c/p\u003e \u003cp\u003eAdditionally, little is known about bacterial epidemiology of CRI, since published studies mostly focus on pediatric patients [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. In addition, no consensus on diagnostic criteria or management of CRI exists, and little is known about their impact on prognosis. Data showed that long ECMO run (up then 10 days) were associated with increased rate of CRI, and that these infections were associated with prolonged hospital stay [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. Practices regarding treatment of these infections are heterogeneous [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e], and optimal duration of antimicrobial treatment is unknown [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eWe therefore design this study to describe the epidemiology, management and prognosis of CRI, and to determine factors associated with outcomes, with a particular focus on the duration of antimicrobial treatment.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003ePatients\u003c/h2\u003e \u003cp\u003ePatients requiring a veno-arterial- (VA) or veno-venous (VV)-ECMO for respectively refractory cardiogenic shock or refractory acute respiratory distress syndrome (ARDS) for more than two days, and having developed an ECMO-related infection, as defined below, between January 2012 and December 2022 in one of the 12 participating ICUs, were included.\u003c/p\u003e \u003cp\u003eInvestigators from the 12 participating ICUs retrospectively collected data from their most recent consecutive patients with CRI. Investigators were invited to include, if possible, at least 10 patients per center, with the possibility to include more patients.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eStudy objectives\u003c/h2\u003e \u003cp\u003eThe primary objective was to describe the characteristics of CRI, as defined hereafter, and its outcomes. Secondary objectives were to evaluate the rate of infection recurrence (see definition below), and to evaluate the impact of antimicrobial treatment duration on the outcome.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eDefinitions\u003c/h2\u003e \u003cp\u003eCRI was defined as clinically suspected infection at one of the ECMO-cannulation site, confirmed by a positive bacteriological cannulation-site sampling. CRI was suspected in case of inflammatory signs at the site of cannulation (i.e., redness, swelling), with or without purulent drainage, with or without general signs of infection (fever, hyperleukocytosis); in case of isolated purulent drainage from the cannulation site; in case of clinical suspicion of infection (fever, hyperleukocytosis) with inflammation of the cannulation site; or in case of hemodynamic worsening (introduction of vasopressors or increase of more than 30% of their dose) suspected of being due to an infection, with inflammation of the cannulation site. In all these cases, physicians performed bacteriological sampling of cannulation site using swab or collection drainage using a syringe. CRI was confirmed when cannulation site sampling retrieved pathogens (bacteria or fungus). CRI was considered complicated by BSI when same pathogen (same species, same resistance profile) was retrieved from blood culture and cannula sampling at the same time (within a 48-hrs time window). Patients with infection of the cannulation site occurring after ECMO removal were not included. Patients with cannula-colonization (positive bacterial sampling of cannula without local or general signs of infection) were not included. Patients with primary BSI (BSI without portal of entry and without bacteriologically-proven CRI) were not included in the present study. Although ECMO cannula may be the hidden portal of entry of such infection, we choose not to include them since we wanted to focus on CRI and its management.\u003c/p\u003e \u003cp\u003eRecurrence of CRI was defined as a new infection episode in a patient having developed a first episode, whatever the pathogen responsible for the second episode. Second episodes occurring after ECMO removal were also considered as recurrences. In these cases, CRI recurrence was considered if the patient had local signs of CRI plus a pathogen retrieved at cannulation site. Patients without second episode of CRI were considered as cured from infection.\u003c/p\u003e \u003cp\u003eDuration of antimicrobial treatment for the first episode was defined as the total duration of antimicrobial treatment effective against the pathogen(s) responsible for ECMO-related infection, including time on antibiotics given for another infection than ECMO-related infection. Antimicrobial treatment was considered appropriate when the incriminated microorganism was susceptible to the antibiotic administered on antimicrobial susceptibility testing. As an example, if a patient received amoxicillin for a CRI due to wild-type \u003cem\u003eE. coli\u003c/em\u003e had to be switched at day 7 of amoxicillin to a 7-day course of piperacillin-tazobactam for \u003cem\u003ePseudomonas aeruginosa\u003c/em\u003e VAP episode, he was considered to have received 14 days of effective antibiotic treatment for the initial CRI.\u003c/p\u003e \u003cp\u003eMulti drug resistant organism (MDRO) was considered according to the definition of the European Society of Clinical Microbiology and Infectious Disease, as pathogens resistant to more than one antimicrobial agent [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eWe also recorded local cannula site complications after decannulation (hemorrhage, infection and thrombosis), mortality rate before ECMO weaning, duration of ECMO support and ventilatory support, length of ICU stays and ICU mortality rate. We then determined factors associated with recurrence.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eCategorical variables are expressed as percentages and continuous variables as median ant interquartile range (IQR). Chi-Square test and Fisher exact test were used to compare categorical variables and Mann-Whitney U test or Wilcoxon test for continuous variables.\u003c/p\u003e \u003cp\u003eTo determine the risk factors of infection recurrence, we performed uni-and multivariable analysis of factors associated with infection recurrence, taking into account death and ECMO decannulation as competing events, using a Fine and Gray analysis. Values were expressed as subdistribution hazard ratio (sdHR). In a second analysis, factors associated with recurrence were compared using a Poisson regression model. Values were expressed in incidence rate ratio (IRR), taking in account the time of exposition to risk, i.e., ECMO duration in our study. To explore factors associated with ICU-mortality, uni- and multivariable analysis of factors associated with ICU death were performed using a Cox proportional model. Variables with a p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.2 in univariable analysis were entered into the model.\u003c/p\u003e \u003cp\u003eTo further explore the impact of antimicrobial treatment duration of the first episode on outcome, we selected the 105 patients who survived at least 8 days after infection onset, and split them into 2 categories: short duration of antimicrobial treatment, i.e. \u0026le;8 days, and long duration of treatment, i.e. \u0026gt;8 days. Uni- and multivariable analysis of factors associated with death was performed in this population. Analyses were performed with an alpha risk of 5%. Statistical analysis was performed with the statistical software R 4.1.1.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eEthics\u003c/h2\u003e \u003cp\u003eIn accordance with current French law, informed written consent for demographic, physiologic and hospital-outcome data analyses was not obtained because this observational study did not modify existing diagnostic or therapeutic strategies. Nonetheless, patients and/or relatives were informed about the anonymous data collection and told that they could decline inclusion. The protocol was approved by the ethics committee of the Soci\u0026eacute;t\u0026eacute; de R\u0026eacute;animation de Langue Fran\u0026ccedil;aise (registration no. CE-SRLF-22-074) and the database is registered by the Commission Nationale de l\u0026rsquo;Informatique et des Libert\u0026eacute;s (CNIL, registration no 2228484v 0).\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003eBetween January 2012 and December 2022, 124 ECMO patients with CRI were included. Baseline patients\u0026rsquo; characteristics are presented in \u003cstrong\u003eTable 1\u003c/strong\u003e. Patients were predominantly male (67%), young (median (IQR) age 52 (40\u0026ndash;61 years), and 26 (21%) were immunosuppressed. We included higher number of patients with VA-ECMO (n=86, 79%) than patients with VV-ECMO (n=38, 31%).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eCRI characteristics\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eCRI characteristics (clinical presentation, pathogens responsible for infection and management of infection) are displayed in \u003cstrong\u003eTable 2\u003c/strong\u003e. Among the 124 episodes of CRI, 78 (63%) had blood culture positive for the same pathogen that the one retrieved at the cannula site. Although the majority (59%) had inflammatory signs at cannula insertion site, 56 had no local sign of infection: among them, 15 had CRI and 41 had CRI with concomitant BSI. Pathogens responsible for ECMO-related infection are shown on Table 2: \u003cem\u003eEnterobacteriaceae\u003c/em\u003e were the most common pathogen responsible for these infections, followed by coagulase-negative \u003cem\u003eStaphylococcus\u003c/em\u003e and \u003cem\u003eEnterococcus\u003c/em\u003e spp.. Interestingly, 50 (40%) of episodes were polymicrobial.\u003c/p\u003e\n\u003cp\u003eDetails on treatment are reported in \u003cstrong\u003eTable 2\u003c/strong\u003e: median (IQR) duration of antimicrobial treatment was 10 (7\u0026ndash;15) days, similar in patients with and without concomitant BSI (median (IQR) durations of 10 (7\u0026ndash;15) and 10 (7\u0026ndash;16) days, respectively, p = 0.6); 29 patients (23%) were treated using combination therapy during all the duration of antimicrobial treatment, 37 (30%) had surgery for their CRI, and 25 (20%) had change of cannulation site for the infection.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eOutcomes\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eAmong the 124 patients, 29 had a second episode of infection that occurred after a median (IQR) of 6 (0\u0026ndash;11) days after the end of antimicrobial treatment of the first episode. Details regarding the outcomes and pathogens responsible for the second episode of infection are reported in \u003cstrong\u003eTable 3\u003c/strong\u003e. Cumulative incidences of infection recurrence, death and ECMO decannulation are shown on \u003cstrong\u003eFigure 1\u003c/strong\u003e. Uni and multivariable analyses of factors associated with mortality are shown in \u003cstrong\u003eeTable 1\u003c/strong\u003e: chronic cardiac failure and septic shock complicating CRI were associated with ICU-mortality, as was coagulase-negative \u003cem\u003eStaphylococcus\u003c/em\u003e as the pathogen responsible for CRI.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eFactors associated with infection recurrence\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTables 1 and 2\u003c/strong\u003e display the baseline and infection characteristics of CRI in patients with and without infection recurrence. Baseline characteristics of patients were similar in the 2 groups, as well as most infection characteristics: timing of infection, type of infection (CRI with or without concomitant BSI). However, coagulase-negative \u003cem\u003eStaphylococcus\u003c/em\u003e was less frequently responsible for infection in patients who will experience infection recurrence. Among the 29 second episodes, only 7 (25%) were due to the same pathogen. Interestingly, characteristics of treatment (duration, use of combination therapy, cannulation site change and need for surgery) were similar in both groups.\u003c/p\u003e\n\u003cp\u003eTo explore the risk factors of infection recurrence, we performed uni-and multivariable analysis of factors associated with recurrence of infection, taking into account death and decannulation as competing events (see \u003cstrong\u003eeTable 2\u003c/strong\u003e in the online supplement). The only factors associated with recurrence of infection were veno-venous ECMO (sdHR 14.7, 95% CI 1.58\u0026ndash;137), whereas the femoro-femoral site of cannulation (as compared to femoro-jugular and other cannulation site) was a protective factor for infection recurrence (sdHR 0.05, 95% CI 0.01\u0026ndash;0.4). Neither type of pathogen responsible for infection nor infection characteristics (associated BSI, inflammatory signs at cannula insertion site, septic shock) and infection management (appropriate empiric antimicrobial treatment, prolonged duration of antimicrobial treatment beyond 8 days, combination therapy, change of cannula) were associated with recurrence of infection (\u003cstrong\u003eeTable 2\u003c/strong\u003e, online supplement). Similar results were obtained when using a Poisson regression model to explore factors associated with recurrence: infection management was not associated with infection recurrence (\u003cstrong\u003eeTable 3\u003c/strong\u003e, online supplement).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eImpact of antimicrobial treatment duration\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTo explore specifically the impact of the duration of antimicrobial treatment of the first episode on outcomes, patients were grouped, according to the duration of treatment (short, i.e. \u0026le;8 days or long, i.e. \u0026nbsp;\u0026gt;8 days) and compared. For this analysis, patients who died before day 8 were excluded, thus 105 patients were included in this analysis. Results are presented in the online supplement: 35 patients had a short (\u0026le;8 days) duration of antimicrobial treatment and 70 had a long (\u0026gt;8 days) duration of antimicrobial treatment; with median (IQR) durations of treatment of 7 (6\u0026ndash;8) and 15 (11\u0026ndash;16) days, respectively. Characteristics of patients were similar between patients with short or long duration of treatment (see \u003cstrong\u003eeTable 4\u003c/strong\u003e, online supplement). Data regarding infection characteristics, microbiological data and management of infection according to duration of antimicrobial treatment are presented in \u003cstrong\u003eeTable 5\u003c/strong\u003e (online supplement). Patients with long duration of treatment had more frequently surgery on cannulation site than patients with short duration; however the outcomes were similar between the 2 groups; particularly the recurrence rates and ICU mortality rates were similar.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn a multivariable cox model analysis evaluating factors associated with death in this specific population (patients surviving at least 8 days after ECMO-related infection onset), prolonged (\u0026gt;8 days), duration of antimicrobial treatment and combination therapy had no protective effect on death in either univariate or multivariable analysis, with respective HR (95% CI) \u0026nbsp;in multivariable analysis of 1.00 (0.49\u0026ndash;2.07), and 0.78 (0.34\u0026ndash;1.80).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this multicenter study of 124 patients with ECMO-related infection, we found that 2/3 of patients with cannula-related infection had associated BSI with the same pathogen. Main pathogens responsible for ECMO-related infection were \u003cem\u003eEnterobacteriaceae\u003c/em\u003e, coagulase-negative \u003cem\u003eStaphylococcus\u003c/em\u003e and \u003cem\u003eEnterococcus\u003c/em\u003e spp., and 40% of these infections involved more than one pathogen. Outcomes of ECMO-related infection were globally poor, since 1/4 experienced infection recurrence. While veno-venous ECMO (as compared to veno-arterial), and therefore ECMO duration was associated with infection recurrence, characteristics of CRI and its management (duration of antimicrobial treatment, use of combination, therapy) were not associated with infection recurrence. More specifically, patients with a short duration of antimicrobial treatment (\u0026le;\u0026thinsp;8 days) had similar outcomes (including infection recurrence) than patients with long duration of antimicrobial treatment (\u0026gt;\u0026thinsp;8 days). Last, factors associated with death were cardiac comorbidity, CRI complicated by septic shock, and coagulase-negative \u003cem\u003eStaphylococcus\u003c/em\u003e as the pathogen responsible for infection.\u003c/p\u003e \u003cp\u003eThis is the largest multicenter cohort regarding CRI in adults, and the first that describes its epidemiology and outcomes. Among the previous observational studies evaluating infection in ECMO patients, small case series showed that \u003cem\u003eEnterobacteriaceae\u003c/em\u003e and non-fermenting Gram negative bacilli were the most frequent pathogen involved in ECMO-associated infection [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e], while other highlighted the central role of \u003cem\u003eEnterococcus\u003c/em\u003e spp. in nosocomial infections [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. However, most infections recorded in these studies were not specifically ECMO-related, i.e., included VAP or urinary tract infection [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Among them, only few CRIs were documented [\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, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Nevertheless, our findings are in agreement with these results, since the main pathogens responsible for CRI were \u003cem\u003eEnterococcus\u003c/em\u003e spp. and \u003cem\u003eEnterobacteriaceae.\u003c/em\u003e Of note, coagulase-negative \u003cem\u003eStaphylococcus\u003c/em\u003e represented 30% of CRI in our study, whereas a previous study that included 21 CRI episodes reported a 19% rate of coagulase-negative \u003cem\u003eStaphylococcus\u003c/em\u003e [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Last, pathogens responsible for infections in our study were less frequently MDRO than in others, [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e], probably related to the relatively low rate of MDRO in French ICUs, as compared to other European countries.\u003c/p\u003e \u003cp\u003eFew data regarding outcomes of patients with CRI are available. A small single center cohort study showed that ECMO patients who developed infection had significantly higher mortality rate than patients without infection (40.4% vs. 20.0%, respectively, p\u0026thinsp;=\u0026thinsp;0.037). However this study reported only 3 CRIs and 5 BSIs [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Moreover, these results were not confirmed in others [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]; Schmidt et al., in their cohort study of 220 ECMO patients, found that infected and non-infected patients had similar mortality rate, 51% vs 62%, p\u0026thinsp;=\u0026thinsp;0.15, respectively. Although our study was not designed to compare patients with and without CRI, ICU-mortality rate of our patients was similar to that observed in the Schmidt study [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. One of the originalities of our work was to focus on other outcomes, and in particular infection recurrence. Interestingly, none of the characteristics of infection was associated with an increased risk of recurrence: neither severity of infection (septic shock), nor positive blood cultures, nor inflammatory signs at ECMO cannulation site. Only coagulase-negative \u003cem\u003eStaphylococcus\u003c/em\u003e as pathogen responsible for infection was associated with a decreased risk of infection recurrence.\u003c/p\u003e \u003cp\u003eManagement of CRI may be challenging, with specific issues regarding antimicrobial treatment duration and need for cannula change to cure infection. To date, no data focusing on these issues exist in the literature. We showed here that neither antimicrobial treatment duration, nor combination therapy, nor cannula change were associated with an increased risk of infection recurrence. Specifically, patients with short duration of antimicrobial treatment (\u0026le;\u0026thinsp;8 days) had similar outcomes than patients with prolonged (\u0026gt;\u0026thinsp;8 days) duration of antimicrobial treatment. This observation is in line with recent data on infections in ICU patients, which showed that short durations of antimicrobial treatment, even for severe infections, are not associated with worse outcomes [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. The only factors associated with infection recurrence were VV-ECMO (as compared to VA-ECMO), whereas femoro-femoral cannulation (mostly used for VA-ECMO in our patients) was a protective factor. These observations underlined the role of ECMO duration on infection outcome: patients with VV-ECMO had a longer duration of ECMO support than VA-ECMO patients, thus are more exposed to the risk of recurrence.\u003c/p\u003e \u003cp\u003eOur study has several limitations that should be underlined. Firstly, due to its retrospective design, we cannot be sure that all parameters recorded in the present study were reported in the patient chart. Moreover, the retrospective design limits the interpretation of outcomes. Secondly, there is no clear definition of CRI and no consensus on the diagnosis process (which sample, which type of sampling\u0026hellip;). Therefore, definition may vary from one center to another, and we cannot be sure that some patients included had cannula colonization rather than infection. However, since we predefined CRI as clinical suspicion of infection plus positive bacterial sample, we think that the population of patients included here is homogeneous. Thirdly, we excluded patients with primary BSI and therefore could have miss some patients with CRI: indeed, some patients with primary BSI could have in fact CRI (not diagnosed for several reasons). However, we preferred to include only patients with bacteriologically proven CRI. Fourthly, although we performed 2 different analyses exploring factors associated with infection recurrence, we could have missed confounding factors. However, the use of Fine and Gray analysis lowers this bias, and allowed us to avoid the 2 major cofounders for infection recurrences, namely death and decannulation. Finally, even if this is a multicenter cohort, our study mostly performed in French ICUs. We cannot be sure that similar results would be observed in ICUs with different bacterial ecologies.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eCRI is frequently polymicrobial, mainly due to Enterobacteriaceae, coagulase-negative \u003cem\u003eStaphylococci\u003c/em\u003e and \u003cem\u003eEnterococcus\u003c/em\u003e spp, and frequently associated with BSI. Risk of infection recurrence is mainly driven by ECMO duration, whereas infection management, and more specifically duration of antimicrobial treatment does not influence infection recurrence and mortality.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eBSI bloodstream infection\u003c/p\u003e\n\u003cp\u003eCRI cannula-related infection\u003c/p\u003e\n\u003cp\u003eECMO extracorporeal membrane oxygenation\u003c/p\u003e\n\u003cp\u003eICU intensive care unit\u003c/p\u003e\n\u003cp\u003eIQR interquartile range\u003c/p\u003e\n\u003cp\u003eIRR incidence rate ratio\u003c/p\u003e\n\u003cp\u003esdHR subdistribution hazard ratio\u003c/p\u003e\n\u003cp\u003eVAP ventilated-associated pneumonia\u003c/p\u003e\n\u003cp\u003eVA veno-arterial\u003c/p\u003e\n\u003cp\u003eVV veno-venous\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthical approval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn accordance with current French law, informed written consent for demographic, physiologic and hospital-outcome data analyses was not obtained because this observational study did not modify existing diagnostic or therapeutic strategies. Nonetheless, patients and/or relatives were informed about the anonymous data collection and told that they could decline inclusion. The protocol was approved by the ethics committee of the Société de Réanimation de Langue Française (registration no. CE-SRLF-22-074) and the database is registered by the Commission Nationale de l’Informatique et des Libertés (CNIL, registration no 2228484v 0).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets generated during the current study are available from the corresponding author on reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interest\u003c/strong\u003e. C.\u003cstrong\u003e-\u003c/strong\u003eE. L.received lecture fees from Merck, Aerogen and AdvanzPharma, and grant from Eumedica and Merck, all outside the submitted work. The other authors have no conflicts of interest to declare in relationship to this manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e. None.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors’ contribution\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSO and CEL designed the study, collected, compiled, analyzed and interpreted the data and wrote the manuscript. SO, NM, CV, EdM, NM, SH, AB, PM, FB, SH, BA, HR and CEL collected data. NM performed statistical analysis, analyzed and interpreted the data. All authors approved the final version of the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eContributors\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eGroupe Hospitalier Pitié-Salpêtrière, Paris, France\u003c/em\u003e: Juliette Chommeloux, Guillaume Hékimian, Alain Combes, Nima Djavidi, Aymeric Lancelot, Pauline Dureau, Guillaume Lebreton. \u003cem\u003eCH de Saint-Brieuc, France\u003c/em\u003e: Pierre Fillâtre. \u003cem\u003eCentre Hospitalier Universitaire de La Réunion, Saint Denis, France\u003c/em\u003e: Laura Rivoire, Nicolas Allou, Radj Cally .\u0026nbsp;\u003cem\u003eHôpital Bichat-Claude Bernard, Paris, France\u003c/em\u003e: Hermann Do Rego, Mariem Dlela, Marc Doman. \u003cem\u003eCentre Hospitalier universitaire de Rennes\u003c/em\u003e : Christophe Camus, Erwan Flecher.\u0026nbsp;\u003cem\u003eHôpital Louis Pradel, Lyon, France\u003c/em\u003e: Jean-Luc Fellahi, Matteo Pozzi, Matthias Jacquet-Lagreze.\u0026nbsp;\u003cem\u003eHôpitaux Universitaires de Genève, Genève, Switzerland\u003c/em\u003e : Raphaël Giraud, Karim Bendjelid.\u0026nbsp;\u003cem\u003eHôpital Henri Mondor, Créteil, France\u003c/em\u003e: Keyvan Razazi, Armand Mekontso-Dessap. \u003cem\u003eHôpital universitaire de Lille, Lille, France\u003c/em\u003e: Anahita Rouzé, Saad Nseir, Thibault Duburcq.\u0026nbsp;\u003cem\u003eHôpital Nord, Marseille, France\u003c/em\u003e: Christophe Guervilly, Jean-Marie Forel.\u0026nbsp;\u003cem\u003eHôpital Haut Leveque, Pessac, France\u003c/em\u003e : Eline Bonnardel. Hôpital Universitaire de Bruxelles, Brussels, Belgium : Fabio S Taccone.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003ePineton de Chambrun M, Br\u0026eacute;chot N, Combes A (2018) Mechanical circulatory devices in acute heart failure. Curr Opin Crit Care 24:286\u0026ndash;291. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1097/MCC.0000000000000520\u003c/span\u003e\u003cspan address=\"10.1097/MCC.0000000000000520\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchmidt M, Br\u0026eacute;chot N, Hariri S et al (2012) Nosocomial infections in adult cardiogenic shock patients supported by venoarterial extracorporeal membrane oxygenation. Clin Infect Dis 55:1633\u0026ndash;1641. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/cid/cis783\u003c/span\u003e\u003cspan address=\"10.1093/cid/cis783\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDanial P, Hajage D, Nguyen LS et al (2018) Percutaneous versus surgical femoro-femoral veno-arterial ECMO: a propensity score matched study. Intensive Care Med 44:2153\u0026ndash;2161. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00134-018-5442-z\u003c/span\u003e\u003cspan address=\"10.1007/s00134-018-5442-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAubron C, Cheng AC, Pilcher D et al (2013) Infections acquired by adults who receive extracorporeal membrane oxygenation: risk factors and outcome. Infect Control Hosp Epidemiol 34:24\u0026ndash;30. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1086/668439\u003c/span\u003e\u003cspan address=\"10.1086/668439\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrasselli G, Scaravilli V, Di Bella S et al (2017) Nosocomial Infections During Extracorporeal Membrane Oxygenation: Incidence, Etiology, and Impact on Patients\u0026rsquo; Outcome. Crit Care Med 45:1726\u0026ndash;1733. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1097/CCM.0000000000002652\u003c/span\u003e\u003cspan address=\"10.1097/CCM.0000000000002652\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWiniszewski H, Boyadjian C, Besch G et al (2022) Extracorporeal Membrane Oxygenation Cannula-Related Infections: Epidemiology and Risk Factors. ASAIO J 68:571\u0026ndash;576. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1097/MAT.0000000000001505\u003c/span\u003e\u003cspan address=\"10.1097/MAT.0000000000001505\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBizzarro MJ, Conrad SA, Kaufman DA et al (2011) Infections acquired during extracorporeal membrane oxygenation in neonates, children, and adults. Pediatr Crit Care Med 12:277\u0026ndash;281. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1097/PCC.0b013e3181e28894\u003c/span\u003e\u003cspan address=\"10.1097/PCC.0b013e3181e28894\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVogel AM, Lew DF, Kao LS, Lally KP (2011) Defining risk for infectious complications on extracorporeal life support. J Pediatr Surg 46:2260\u0026ndash;2264. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jpedsurg.2011.09.013\u003c/span\u003e\u003cspan address=\"10.1016/j.jpedsurg.2011.09.013\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHsu M-S, Chiu K-M, Huang Y-T et al (2009) Risk factors for nosocomial infection during extracorporeal membrane oxygenation. J Hosp Infect 73:210\u0026ndash;216. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jhin.2009.07.016\u003c/span\u003e\u003cspan address=\"10.1016/j.jhin.2009.07.016\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGlater-Welt LB, Schneider JB, Zinger MM et al (2016) Nosocomial Bloodstream Infections in Patients Receiving Extracorporeal Life Support: Variability in Prevention Practices: A Survey of the Extracorporeal Life Support Organization Members. J Intensive Care Med 31:654\u0026ndash;669. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/0885066615571540\u003c/span\u003e\u003cspan address=\"10.1177/0885066615571540\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBurket JS, Bartlett RH, Vander Hyde K, Chenoweth CE (1999) Nosocomial infections in adult patients undergoing extracorporeal membrane oxygenation. Clin Infect Dis 28:828\u0026ndash;833. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1086/515200\u003c/span\u003e\u003cspan address=\"10.1086/515200\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun W, Wang S, Chen Z et al (2020) Impact of high mechanical shear stress and oxygenator membrane surface on blood damage relevant to thrombosis and bleeding in a pediatric ECMO circuit. Artif Organs 44:717\u0026ndash;726. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/aor.13646\u003c/span\u003e\u003cspan address=\"10.1111/aor.13646\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMagiorakos A-P, Srinivasan A, Carey RB et al (2012) Multidrug-resistant, extensively drug-resistant and pandrug-resistant bacteria: an international expert proposal for interim standard definitions for acquired resistance. Clin Microbiol Infect 18:268\u0026ndash;281. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1469-0691.2011.03570.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1469-0691.2011.03570.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun H-Y, Ko W-J, Tsai P-R et al (2010) Infections occurring during extracorporeal membrane oxygenation use in adult patients. J Thorac Cardiovasc Surg 140:1125\u0026ndash;1132e2. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jtcvs.2010.07.017\u003c/span\u003e\u003cspan address=\"10.1016/j.jtcvs.2010.07.017\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim HS, Park S, Ko HH et al (2021) Different characteristics of bloodstream infection during venoarterial and venovenous extracorporeal membrane oxygenation in adult patients. Sci Rep 11:9498. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-021-89108-4\u003c/span\u003e\u003cspan address=\"10.1038/s41598-021-89108-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMokrani D, Chommeloux J, Pineton de Chambrun M et al (2023) Antibiotic stewardship in the ICU: time to shift into overdrive. Ann Intensive Care 13:39. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s13613-023-01134-9\u003c/span\u003e\u003cspan address=\"10.1186/s13613-023-01134-9\" 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\u003cstrong\u003eTable 1.\u0026nbsp;\u003c/strong\u003eBaseline characteristics according to the recurrence of infection or not\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"662\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"36.858006042296076%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.580060422960724%\" valign=\"top\"\u003e\n \u003cp\u003eOverall population\u003c/p\u003e\n \u003cp\u003en = 124\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.069486404833835%\" valign=\"top\"\u003e\n \u003cp\u003eNo infection recurrence\u003c/p\u003e\n \u003cp\u003en = 95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.882175226586103%\" valign=\"top\"\u003e\n \u003cp\u003eInfection recurrence\u003c/p\u003e\n \u003cp\u003en = 29\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.610271903323262%\" valign=\"top\"\u003e\n \u003cp\u003ep\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"36.858006042296076%\" valign=\"top\"\u003e\n \u003cp\u003eBaseline characteristics\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.580060422960724%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.069486404833835%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.882175226586103%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.610271903323262%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"36.858006042296076%\" valign=\"top\"\u003e\n \u003cp\u003eMale sex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.580060422960724%\" valign=\"top\"\u003e\n \u003cp\u003e83 (67)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.069486404833835%\" valign=\"top\"\u003e\n \u003cp\u003e62 (65)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.882175226586103%\" valign=\"top\"\u003e\n \u003cp\u003e21 (72)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.610271903323262%\" valign=\"top\"\u003e\n \u003cp\u003e0.6\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"36.858006042296076%\" valign=\"top\"\u003e\n \u003cp\u003eAge, years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.580060422960724%\" valign=\"top\"\u003e\n \u003cp\u003e52 (40\u0026ndash;61)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.069486404833835%\" valign=\"top\"\u003e\n \u003cp\u003e54 (41\u0026ndash;63)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.882175226586103%\" valign=\"top\"\u003e\n \u003cp\u003e50 (40\u0026ndash;58)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.610271903323262%\" valign=\"top\"\u003e\n \u003cp\u003e0.3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"36.858006042296076%\" valign=\"top\"\u003e\n \u003cp\u003eBMI, kg/m\u0026sup2; \u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.580060422960724%\" valign=\"top\"\u003e\n \u003cp\u003e28 (25\u0026ndash;35)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.069486404833835%\" valign=\"top\"\u003e\n \u003cp\u003e28 (24\u0026ndash;33)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.882175226586103%\" valign=\"top\"\u003e\n \u003cp\u003e33 (26\u0026ndash;40)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.610271903323262%\" valign=\"top\"\u003e\n \u003cp\u003e0.06\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"36.858006042296076%\" valign=\"top\"\u003e\n \u003cp\u003ePre-existing conditions\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Chronic cardiac disease\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Chronic renal disease\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Chronic respiratory disease\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Diabetes mellitus\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Peripheral vascular disease\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Immunocompromised\u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e\n \u003cp\u003eOngoing pregnancy\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.580060422960724%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e45 (36)\u003c/p\u003e\n \u003cp\u003e17 (14)\u003c/p\u003e\n \u003cp\u003e19 (15)\u003c/p\u003e\n \u003cp\u003e38 (31)\u003c/p\u003e\n \u003cp\u003e21 (17)\u003c/p\u003e\n \u003cp\u003e26 (21)\u003c/p\u003e\n \u003cp\u003e1 (1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.069486404833835%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e36 (38)\u003c/p\u003e\n \u003cp\u003e15 (16)\u003c/p\u003e\n \u003cp\u003e17 (18)\u003c/p\u003e\n \u003cp\u003e34 (36)\u003c/p\u003e\n \u003cp\u003e15 (16)\u003c/p\u003e\n \u003cp\u003e20 (21)\u003c/p\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.882175226586103%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e9 (31)\u003c/p\u003e\n \u003cp\u003e2 (7)\u003c/p\u003e\n \u003cp\u003e2 (7)\u003c/p\u003e\n \u003cp\u003e4 (14)\u003c/p\u003e\n \u003cp\u003e6 (21)\u003c/p\u003e\n \u003cp\u003e6 (21)\u003c/p\u003e\n \u003cp\u003e1 (3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.610271903323262%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e0.7\u003c/p\u003e\n \u003cp\u003e0.4\u003c/p\u003e\n \u003cp\u003e0.3\u003c/p\u003e\n \u003cp\u003e0.04\u003c/p\u003e\n \u003cp\u003e0.7\u003c/p\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003cp\u003e0.2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"36.858006042296076%\" valign=\"top\"\u003e\n \u003cp\u003eAdmission SAPS2\u003c/p\u003e\n \u003cp\u003eAdmission SOFA score\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.580060422960724%\" valign=\"top\"\u003e\n \u003cp\u003e42 (30\u0026ndash;57)\u003c/p\u003e\n \u003cp\u003e8 (4\u0026ndash;12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.069486404833835%\" valign=\"top\"\u003e\n \u003cp\u003e41 (31\u0026ndash;56)\u003c/p\u003e\n \u003cp\u003e9 (4\u0026ndash;12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.882175226586103%\" valign=\"top\"\u003e\n \u003cp\u003e45 (27\u0026ndash;61)\u003c/p\u003e\n \u003cp\u003e8 (5\u0026ndash;12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.610271903323262%\" valign=\"top\"\u003e\n \u003cp\u003e0.9\u003c/p\u003e\n \u003cp\u003e0.8\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"36.858006042296076%\" valign=\"top\"\u003e\n \u003cp\u003eECMO characteristics\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.580060422960724%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.069486404833835%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.882175226586103%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.610271903323262%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"36.858006042296076%\" valign=\"top\"\u003e\n \u003cp\u003eVeno-venous ECMO\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.580060422960724%\" valign=\"top\"\u003e\n \u003cp\u003e38 (31)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.069486404833835%\" valign=\"top\"\u003e\n \u003cp\u003e23 (24)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.882175226586103%\" valign=\"top\"\u003e\n \u003cp\u003e15 (52)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.610271903323262%\" valign=\"top\"\u003e\n \u003cp\u003e0.01\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"36.858006042296076%\" valign=\"top\"\u003e\n \u003cp\u003ePercutaneous cannulation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.580060422960724%\" valign=\"top\"\u003e\n \u003cp\u003e69/104 (66)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.069486404833835%\" valign=\"top\"\u003e\n \u003cp\u003e53/82 (65)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.882175226586103%\" valign=\"top\"\u003e\n \u003cp\u003e16/22 (73)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.610271903323262%\" valign=\"top\"\u003e\n \u003cp\u003e0.6\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"36.858006042296076%\" valign=\"top\"\u003e\n \u003cp\u003eOngoing antibiotics at ECMO start\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.580060422960724%\" valign=\"top\"\u003e\n \u003cp\u003e52 (42)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.069486404833835%\" valign=\"top\"\u003e\n \u003cp\u003e37 (39)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.882175226586103%\" valign=\"top\"\u003e\n \u003cp\u003e15 (51)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.610271903323262%\" valign=\"top\"\u003e\n \u003cp\u003e0.3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"36.858006042296076%\" valign=\"top\"\u003e\n \u003cp\u003eSARS-Cov2 infection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.580060422960724%\" valign=\"top\"\u003e\n \u003cp\u003e27 (22)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.069486404833835%\" valign=\"top\"\u003e\n \u003cp\u003e16 (17)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.882175226586103%\" valign=\"top\"\u003e\n \u003cp\u003e11 (38)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.610271903323262%\" valign=\"top\"\u003e\n \u003cp\u003e0.03\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"36.858006042296076%\" valign=\"top\"\u003e\n \u003cp\u003eIndication of ECMO\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp;Cardiogenic shock\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp;ARDS\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp;Post-cardiotomy\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp;Septic shock\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp;E-CPR\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp;Other\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.580060422960724%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e64 (52)\u003c/p\u003e\n \u003cp\u003e38 (31)\u003c/p\u003e\n \u003cp\u003e12 (10)\u003c/p\u003e\n \u003cp\u003e3 (2)\u003c/p\u003e\n \u003cp\u003e5 (4)\u003c/p\u003e\n \u003cp\u003e2(2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.069486404833835%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e52 (55)\u003c/p\u003e\n \u003cp\u003e25 (26)\u003c/p\u003e\n \u003cp\u003e10 (11)\u003c/p\u003e\n \u003cp\u003e2 (2)\u003c/p\u003e\n \u003cp\u003e4 (4)\u003c/p\u003e\n \u003cp\u003e2 (2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.882175226586103%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e12 (41)\u003c/p\u003e\n \u003cp\u003e13 (45)\u003c/p\u003e\n \u003cp\u003e2 (7)\u003c/p\u003e\n \u003cp\u003e1 (3)\u003c/p\u003e\n \u003cp\u003e1 (3)\u003c/p\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.610271903323262%\" valign=\"top\"\u003e\n \u003cp\u003e0.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"36.858006042296076%\" valign=\"top\"\u003e\n \u003cp\u003eCannulation by mobile unit\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.580060422960724%\" valign=\"top\"\u003e\n \u003cp\u003e27/119 (23)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.069486404833835%\" valign=\"top\"\u003e\n \u003cp\u003e20 (22)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.882175226586103%\" valign=\"top\"\u003e\n \u003cp\u003e7 (27)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.610271903323262%\" valign=\"top\"\u003e\n \u003cp\u003e0.8\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eData are expressed as median (IQR) or n (%).\u003c/p\u003e\n\u003cp\u003eAbbreviations: BMI\u0026nbsp;: Body Mass Index\u0026nbsp;; SAPS2\u0026nbsp;:\u0026nbsp;Simplified acute physiology score 2; SOFA\u0026nbsp;: Sepsis-related organ failure assessment\u0026nbsp;; ECMO\u0026nbsp;: Extra-corporeal membrane oxygenation; E-CPR: Extracorporeal cardiopulmonary resuscitation.\u003c/p\u003e\n\u003cp\u003e\u003csup\u003ea\u003c/sup\u003e data available for 114/124 patients\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003csup\u003eb\u003c/sup\u003e Solid organ transplant recipients, active hematological malignancy or receiving immunosuppressant drug (including corticosteroids at a dose\u0026ge;0.5 mg/kg/d for\u0026ge;1 month)\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2:\u0026nbsp;\u003c/strong\u003eFirst ECMO-related infection episode characteristics\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"858\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"39.62703962703963%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.5990675990676%\" valign=\"top\"\u003e\n \u003cp\u003eOverall population\u003c/p\u003e\n \u003cp\u003en = 124\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.365967365967364%\" valign=\"top\"\u003e\n \u003cp\u003eNo infection recurrence\u003c/p\u003e\n \u003cp\u003en = 95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.55011655011655%\" valign=\"top\"\u003e\n \u003cp\u003eInfection recurrence\u003c/p\u003e\n \u003cp\u003en = 29\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.857808857808857%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ep\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"39.62703962703963%\" valign=\"top\"\u003e\n \u003cp\u003eTime between ICU admission and first episode, days\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.5990675990676%\" valign=\"top\"\u003e\n \u003cp\u003e9 (6\u0026ndash;16)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.365967365967364%\" valign=\"top\"\u003e\n \u003cp\u003e8 (6\u0026ndash;14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.55011655011655%\" valign=\"top\"\u003e\n \u003cp\u003e13 (6\u0026ndash;19)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.857808857808857%\" valign=\"top\"\u003e\n \u003cp\u003e0.08\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"39.62703962703963%\" valign=\"top\"\u003e\n \u003cp\u003eConcomitant BSI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.5990675990676%\" valign=\"top\"\u003e\n \u003cp\u003e78 (63)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.365967365967364%\" valign=\"top\"\u003e\n \u003cp\u003e61 (64)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.55011655011655%\" valign=\"top\"\u003e\n \u003cp\u003e17 (59)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.857808857808857%\" valign=\"top\"\u003e\n \u003cp\u003e0.745\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"39.62703962703963%\" valign=\"top\"\u003e\n \u003cp\u003eSeptic shock\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.5990675990676%\" valign=\"top\"\u003e\n \u003cp\u003e69/121 (58)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.365967365967364%\" valign=\"top\"\u003e\n \u003cp\u003e56/93 (60)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.55011655011655%\" valign=\"top\"\u003e\n \u003cp\u003e13/27 (48)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.857808857808857%\" valign=\"top\"\u003e\n \u003cp\u003e0.4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"39.62703962703963%\" valign=\"top\"\u003e\n \u003cp\u003eInflammatory sign at cannula insertion site\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.5990675990676%\" valign=\"top\"\u003e\n \u003cp\u003e68/116 (59)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.365967365967364%\" valign=\"top\"\u003e\n \u003cp\u003e50/88 (57)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.55011655011655%\" valign=\"top\"\u003e\n \u003cp\u003e18/28 (64)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.857808857808857%\" valign=\"top\"\u003e\n \u003cp\u003e0.6\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"39.62703962703963%\" valign=\"top\"\u003e\n \u003cp\u003ePathogen\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp;Staphylococcus aureus\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003eMethicillin resistant\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp;Coagulase-negative Staphylococcus\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003eMethicillin resistant\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp;Streptococcus\u003c/em\u003e spp\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp;Enterococcus\u003c/em\u003e spp\u003c/p\u003e\n \u003cp\u003eVancomycin resistant\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp;Enterobacteriaceae\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; ESBL-producer\u003c/p\u003e\n \u003cp\u003eCarbapenem resistant\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp;Non fermenting Gram-negative bacilli\u003c/p\u003e\n \u003cp\u003eMultiresistant NF-GNB\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp;Anaerobe\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp;MDRO\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp;Polymicrobial\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.5990675990676%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e6 (5)\u003c/p\u003e\n \u003cp\u003e1 (1)\u003c/p\u003e\n \u003cp\u003e37 (30)\u003c/p\u003e\n \u003cp\u003e6 (5)\u003c/p\u003e\n \u003cp\u003e4 (3)\u003c/p\u003e\n \u003cp\u003e38 (31)\u003c/p\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003cp\u003e64 (52)\u003c/p\u003e\n \u003cp\u003e15 (12)\u003c/p\u003e\n \u003cp\u003e4 (3)\u003c/p\u003e\n \u003cp\u003e23 (18)\u003c/p\u003e\n \u003cp\u003e5 (4)\u003c/p\u003e\n \u003cp\u003e7 (6)\u003c/p\u003e\n \u003cp\u003e21 (17)\u003c/p\u003e\n \u003cp\u003e50 (40)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.365967365967364%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e5 (5)\u003c/p\u003e\n \u003cp\u003e1 (1)\u003c/p\u003e\n \u003cp\u003e33 (35)\u003c/p\u003e\n \u003cp\u003e4 (4)\u003c/p\u003e\n \u003cp\u003e3 (3)\u003c/p\u003e\n \u003cp\u003e28 (30)\u003c/p\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003cp\u003e28 (30)\u003c/p\u003e\n \u003cp\u003e8 (8)\u003c/p\u003e\n \u003cp\u003e1 (1)\u003c/p\u003e\n \u003cp\u003e18 (19)\u003c/p\u003e\n \u003cp\u003e4 (4)\u003c/p\u003e\n \u003cp\u003e6 (6)\u003c/p\u003e\n \u003cp\u003e13 (14)\u003c/p\u003e\n \u003cp\u003e40 (42)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.55011655011655%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e1 (3)\u003c/p\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003cp\u003e4 (14)\u003c/p\u003e\n \u003cp\u003e2 (7)\u003c/p\u003e\n \u003cp\u003e1 (3)\u003c/p\u003e\n \u003cp\u003e10 (35)\u003c/p\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003cp\u003e16 (62)\u003c/p\u003e\n \u003cp\u003e7 (24)\u003c/p\u003e\n \u003cp\u003e3 (10)\u003c/p\u003e\n \u003cp\u003e5 (17)\u003c/p\u003e\n \u003cp\u003e1 (3)\u003c/p\u003e\n \u003cp\u003e1 (3)\u003c/p\u003e\n \u003cp\u003e8 (28)\u003c/p\u003e\n \u003cp\u003e10 (35)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.857808857808857%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003cp\u003e0.04\u003c/p\u003e\n \u003cp\u003e0.9\u003c/p\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003cp\u003e0.8\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e0.3\u003c/p\u003e\n \u003cp\u003e0.05\u003c/p\u003e\n \u003cp\u003e0.06\u003c/p\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003cp\u003e0.1\u003c/p\u003e\n \u003cp\u003e0.6\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"39.62703962703963%\" valign=\"top\"\u003e\n \u003cp\u003eTime from diagnostic to empiric treatment, days\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.5990675990676%\" valign=\"top\"\u003e\n \u003cp\u003e0 (0\u0026ndash;0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.365967365967364%\" valign=\"top\"\u003e\n \u003cp\u003e0 (0\u0026ndash;0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.55011655011655%\" valign=\"top\"\u003e\n \u003cp\u003e0 (0\u0026ndash;0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.857808857808857%\" valign=\"top\"\u003e\n \u003cp\u003e0.7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"39.62703962703963%\" valign=\"top\"\u003e\n \u003cp\u003eAppropriate empiric antimicrobial treatment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.5990675990676%\" valign=\"top\"\u003e\n \u003cp\u003e77/109 (71)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.365967365967364%\" valign=\"top\"\u003e\n \u003cp\u003e58 (69)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.55011655011655%\" valign=\"top\"\u003e\n \u003cp\u003e19 (76)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.857808857808857%\" valign=\"top\"\u003e\n \u003cp\u003e0.7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"39.62703962703963%\" valign=\"top\"\u003e\n \u003cp\u003eTime from infection onset to appropriate antimicrobial treatment, days\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.5990675990676%\" valign=\"top\"\u003e\n \u003cp\u003e2 (0\u0026ndash;2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.365967365967364%\" valign=\"top\"\u003e\n \u003cp\u003e1 (0\u0026ndash;2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.55011655011655%\" valign=\"top\"\u003e\n \u003cp\u003e2 (1\u0026ndash;3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.857808857808857%\" valign=\"top\"\u003e\n \u003cp\u003e0.3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"39.62703962703963%\" valign=\"top\"\u003e\n \u003cp\u003eDuration of antimicrobial treatment of first ECMO-related infection, days\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.5990675990676%\" valign=\"top\"\u003e\n \u003cp\u003e10 (7\u0026ndash;15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.365967365967364%\" valign=\"top\"\u003e\n \u003cp\u003e10 (7\u0026ndash;15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.55011655011655%\" valign=\"top\"\u003e\n \u003cp\u003e13 (8\u0026ndash;18)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.857808857808857%\" valign=\"top\"\u003e\n \u003cp\u003e0.06\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"39.62703962703963%\" valign=\"top\"\u003e\n \u003cp\u003eCombination antimicrobial treatment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.5990675990676%\" valign=\"top\"\u003e\n \u003cp\u003e29 (23)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.365967365967364%\" valign=\"top\"\u003e\n \u003cp\u003e25 (26.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.55011655011655%\" valign=\"top\"\u003e\n \u003cp\u003e4 (13.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.857808857808857%\" valign=\"top\"\u003e\n \u003cp\u003e0.3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"39.62703962703963%\" valign=\"top\"\u003e\n \u003cp\u003eOxygenator change during infection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.5990675990676%\" valign=\"top\"\u003e\n \u003cp\u003e29 (23)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.365967365967364%\" valign=\"top\"\u003e\n \u003cp\u003e19 (20.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.55011655011655%\" valign=\"top\"\u003e\n \u003cp\u003e10 (34.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.857808857808857%\" valign=\"top\"\u003e\n \u003cp\u003e0.2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"39.62703962703963%\" valign=\"top\"\u003e\n \u003cp\u003eCannula site change for infection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.5990675990676%\" valign=\"top\"\u003e\n \u003cp\u003e25 (20)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.365967365967364%\" valign=\"top\"\u003e\n \u003cp\u003e19 (20.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.55011655011655%\" valign=\"top\"\u003e\n \u003cp\u003e6 (20.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.857808857808857%\" valign=\"top\"\u003e\n \u003cp\u003e1.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"39.62703962703963%\" valign=\"top\"\u003e\n \u003cp\u003eSurgery needed for cannula site infection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.5990675990676%\" valign=\"top\"\u003e\n \u003cp\u003e37 (30)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.365967365967364%\" valign=\"top\"\u003e\n \u003cp\u003e28 (29.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.55011655011655%\" valign=\"top\"\u003e\n \u003cp\u003e9 (31.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.857808857808857%\" valign=\"top\"\u003e\n \u003cp\u003e1.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"39.62703962703963%\" valign=\"top\"\u003e\n \u003cp\u003eTime to new infection, days\u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.5990675990676%\" valign=\"top\"\u003e\n \u003cp\u003e6 (0\u0026ndash;11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.365967365967364%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026ndash;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.55011655011655%\" valign=\"top\"\u003e\n \u003cp\u003e6 (0\u0026ndash;11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.857808857808857%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eTreatment failure is defined as recurrence of ECMO-related infection or death during ICU stay.\u003c/p\u003e\n\u003cp\u003eData are expressed as median [IQR] or n (%).\u003c/p\u003e\n\u003cp\u003eECMO, Extra-corporeal membrane oxygenation. ICU, intensive care unit. CRI, Cannula related infection. ESBL, extended spectrum beta-lactamase. NF-GNB, Non fermenting Gram-negative bacilli. MDRO, multidrug resistant organism\u003c/p\u003e\n\u003cp\u003e\u003csup\u003ea\u003c/sup\u003e data available for 121/124 patients\u003c/p\u003e\n\u003cp\u003e\u003csup\u003eb\u003c/sup\u003e time from end of antimicrobial treatment of first episode to second episode onset.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 3: Outcomes\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"558\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"61.04129263913824%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"38.95870736086176%\" valign=\"top\"\u003e\n \u003cp\u003eOverall population\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003en = 124\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"61.04129263913824%\" valign=\"top\"\u003e\n \u003cp\u003eRecurrence of infection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"38.95870736086176%\" valign=\"top\"\u003e\n \u003cp\u003e29 (23)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"61.04129263913824%\" valign=\"top\"\u003e\n \u003cp\u003eTime to new infection, days\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"38.95870736086176%\" valign=\"top\"\u003e\n \u003cp\u003e6 (0\u0026ndash;10)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"61.04129263913824%\" valign=\"top\"\u003e\n \u003cp\u003ePathogen responsible for second episode\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u003cem\u003eStaphylococcus aureus\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Coagulase-negative \u003cem\u003eStaphylococcus\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Streptococcus\u003c/em\u003e spp\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Enterococcus\u003c/em\u003e spp.\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Enterobacteriaceae\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Non fermenting Gram-negative bacilli\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; MDRO\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Polymicrobial\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"38.95870736086176%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003cp\u003e6 (21)\u003c/p\u003e\n \u003cp\u003e1 (3)\u003c/p\u003e\n \u003cp\u003e10 (34)\u003c/p\u003e\n \u003cp\u003e11 (38)\u003c/p\u003e\n \u003cp\u003e4 (14)\u003c/p\u003e\n \u003cp\u003e14 (48)\u003c/p\u003e\n \u003cp\u003e14 (48)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"61.04129263913824%\" valign=\"top\"\u003e\n \u003cp\u003eComplications on cannula site after decannulation\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Cannula site infection\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Arterial thrombosis\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Hemorrhage\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"38.95870736086176%\" valign=\"top\"\u003e\n \u003cp\u003e31 (26)\u003c/p\u003e\n \u003cp\u003e20 (16)\u003c/p\u003e\n \u003cp\u003e11 (9)\u003c/p\u003e\n \u003cp\u003e7 (6)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"61.04129263913824%\" valign=\"top\"\u003e\n \u003cp\u003eDuration of ECMO support, days\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"38.95870736086176%\" valign=\"top\"\u003e\n \u003cp\u003e15 (9\u0026ndash;26)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"61.04129263913824%\" valign=\"top\"\u003e\n \u003cp\u003eDuration of invasive mechanical ventilation, days\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"38.95870736086176%\" valign=\"top\"\u003e\n \u003cp\u003e20 (11\u0026ndash;34)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"61.04129263913824%\" valign=\"top\"\u003e\n \u003cp\u003eICU length of stay, days\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"38.95870736086176%\" valign=\"top\"\u003e\n \u003cp\u003e26 (19\u0026ndash;45)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"61.04129263913824%\" valign=\"top\"\u003e\n \u003cp\u003eDeath on ECMO\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"38.95870736086176%\" valign=\"top\"\u003e\n \u003cp\u003e40 (32)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"61.04129263913824%\" valign=\"top\"\u003e\n \u003cp\u003eICU mortality rate\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"38.95870736086176%\" valign=\"top\"\u003e\n \u003cp\u003e62 (50)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eData are expressed as median (IQR) or n (%).\u003c/p\u003e\n\u003cp\u003eECMO, Extra-corporeal membrane oxygenation. ICU, intensive care unit\u003c/p\u003e\n\u003cp\u003e\u003csup\u003ea\u003c/sup\u003e time from end of antimicrobial treatment of first episode to second episode onset.\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":false,"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":"ECMO, cannula-related infection, hospital-acquired infection, bloodstream infection","lastPublishedDoi":"10.21203/rs.3.rs-3940633/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3940633/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003ePurpose\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOnly few data regarding epidemiology and management of ECMO cannula-related infections (CRIs) exist. The aim of our study was to describe their epidemiology and prognosis, and to evaluate factors associated with outcome.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe performed a multicenter retrospective study in 12 European ICUs, including patients with CRI, defined as a clinical suspicion plus a positive bacterial sample of ECMO-cannulation site. Primary objective was to describe CRI characteristics and outcomes. Secondary objectives were to evaluate the rates of infection recurrence, their risk factors, and to evaluate the impact of antimicrobial treatment duration on outcome.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDuring the study period, 124 patients with CRI (78 having concomitant positive blood culture with the same pathogen) were included. Pathogens responsible for infections were predominantly Enterobacteriaceae, coagulase-negative \u003cem\u003eStaphylococcus\u003c/em\u003e and \u003cem\u003eEnterococcus\u003c/em\u003espp., and 40% of episodes were polymicrobial. Rates of infection recurrence was 24% and ICU-mortality rate was 50%. Whereas veno-venous ECMO (as compared to veno-arterial ECMO), and therefore ECMO duration was associated with infection recurrence, characteristics of CRI and its management (and in particular duration of antimicrobial treatment) were not associated with recurrence. Patients with antibiotic course ≤8 days had similar infection recurrence rate and outcomes (including mortality) than patients with prolonged (\u0026gt;8 days) antibiotic course.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCRIs are frequently associated with BSI and frequently polymicrobial. Main risk factor of infection recurrence is ECMO duration. Duration of antimicrobial treatment for CRI ≤8 days is not associated with an increased risk of recurrence or death, as compared to longer treatment.\u003c/p\u003e","manuscriptTitle":"Characteristics and outcomes of ECMO cannula-related infections: a European multicenter retrospective study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-02-13 19:17:50","doi":"10.21203/rs.3.rs-3940633/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":"c4c239d5-b78f-47f6-bc7f-ecc4e5acc268","owner":[],"postedDate":"February 13th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-03-14T08:50:17+00:00","versionOfRecord":[],"versionCreatedAt":"2024-02-13 19:17:50","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-3940633","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3940633","identity":"rs-3940633","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}