Prone Positioning During Extracorporeal Membrane Oxygenation in Patients With Severe ARDS: The PRONECMO Randomized Clinical Trial

Matthieu Schmidt; David Hajage; Guillaume Lebreton; Martin Dres; Christophe Guervilly; Jean Christophe Richard; Romain Sonneville; Hadrien Winiszewski; Gregoire Muller; Gaëtan Beduneau; Emmanuelle Mercier; Hadrien Roze; Mathieu Lesouhaitier; Nicolas Terzi; Arnaud W Thille; Isaura Laurent; Antoine Kimmoun; Alain Combes

doi:10.1001/jama.2023.24491

. 2023 Dec 1;330(24):2343–2353. doi: 10.1001/jama.2023.24491

Prone Positioning During Extracorporeal Membrane Oxygenation in Patients With Severe ARDS

The PRONECMO Randomized Clinical Trial

Matthieu Schmidt ^1,^2,^3,^✉, David Hajage ⁴, Guillaume Lebreton ^1,⁵, Martin Dres ^3,⁶, Christophe Guervilly ⁷, Jean Christophe Richard ⁸, Romain Sonneville ⁹, Hadrien Winiszewski ¹⁰, Gregoire Muller ¹¹, Gaëtan Beduneau ¹², Emmanuelle Mercier ¹³, Hadrien Roze ¹⁴, Mathieu Lesouhaitier ¹⁵, Nicolas Terzi ¹⁶, Arnaud W Thille ¹⁷, Isaura Laurent ⁴, Antoine Kimmoun ¹⁸, Alain Combes ^1,^2,³, for the PRONECMO Investigators, the REVA Network, and the International ECMO Network (ECMONet)

¹Sorbonne Université, INSERM, UMRS_1166-ICAN, Institute of Cardiometabolism and Nutrition, Paris, France

²Assistance Publique–Hôpitaux de Paris, Service de Médecine Intensive-Réanimation, Institut de Cardiologie, Hôpital Pitié-Salpêtrière, Paris, France

³GRC 30 RESPIRE, Sorbonne Université, Paris, France

⁴Sorbonne Université, INSERM, Institut Pierre Louis d’Epidémiologie et de Santé Publique, Assistance Publique–Hôpitaux de Paris, Hôpitaux Universitaires Pitié Salpêtrière–Charles Foix, Département de Santé Publique, Centre de Pharmacoépidémiologie (Cephepi), CIC-1421, Paris, France

⁵Assistance Publique–Hôpitaux de Paris, Thoracic and Cardiovascular Department, Hôpital Pitié-Salpêtrière, Paris France

⁶Assistance Publique–Hôpitaux de Paris, Groupe Hospitalier Universitaire APHP-Sorbonne Université, Site Pitié-Salpêtrière, Service de Médecine Intensive et Réanimation (Département R3S), et Sorbonne Université, INSERM, UMRS-1158 Neurophysiologie Respiratoire Expérimentale et Clinique, Paris, France

⁷Médecine Intensive Réanimation, Hôpital Nord, AP-HM, Centre d’Etudes et de Recherches sur les Services de Santé et Qualité de Vie EA 3279, Marseille, France

⁸Médecine Intensive Réanimation, Hospices Civils de Lyon, Hôpital de la Croix Rousse, Lyon, France

⁹Université Paris Cité, INSERM U1137, F-75018 Paris, APHP Nord, Service de Médecine Intensive Réanimation, Hôpital Bichat–Claude Bernard, Paris, France

¹⁰Service de Médecine Intensive Réanimation, CHU Besançon, Besançon, France

¹¹Service de Médecine Intensive-Réanimation, Centre Hospitalier Régional d’Orléans, Orléans, France

¹²Université Rouen Normandie, Normandie Université, GRHVN UR 3830, Medical Intensive Care Unit, Rouen University Hospital, Rouen, France

¹³Service de Médecine Intensive Réanimation, CHU de Tours, CRICS-TRIGGERSEP Network, Tours, France

¹⁴Service d’Anesthésie Réanimation Thoraco-Abdominale, CMC Magellan, Hôpital, Haut Leveque, CHU de Bordeaux, INSERM 1045: Centre de Recherche Cardio-Thoracique, Université de Bordeaux, Pessac, France

¹⁵CHU Rennes, Service de Maladies Infectieuses et Réanimation Médicale, Rennes, France

¹⁶Service de Médecine Intensive Réanimation, Université de Grenoble-Alpes, Inserm U1042, Grenoble, France

¹⁷CHU de Poitiers, Médecine Intensive Réanimation, Poitiers, France

¹⁸Université de Lorraine, CHRU de Nancy, Institut Lorrain du Cœur et des Vaisseaux, Service de Médecine Intensive-Réanimation, U1116, FCRIN-INICRCT, Nancy, France

^✉

Corresponding Author: Matthieu Schmidt, MD, Service de Medecine Intensive Reanimation, iCAN, Institute of Cardiometabolism and Nutrition, Hôpital de la Pitié-Salpêtrière, 47 Boulevard de l’Hôpital, 75651 Paris CEDEX 13, France (matthieu.schmidt@aphp.fr).

Accepted for Publication: November 7, 2023.

Published Online: December 1, 2023. doi:10.1001/jama.2023.24491

Author Contributions: Drs Schmidt, Hajage, and Combes had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Schmidt, Hajage, Lebreton, Combes.

Acquisition, analysis, or interpretation of data: Schmidt, Hajage, Dres, Guervilly, Richard, Sonneville, Winiszewski, Muller, Beduneau, Mercier, Roze, Lesouhaitier, Terzi, Thille, Laurent, Kimmoun, Combes.

Drafting of the manuscript: Schmidt, Laurent, Combes.

Critical review of the manuscript for important intellectual content: Schmidt, Hajage, Lebreton, Dres, Guervilly, Richard, Sonneville, Winiszewski, Muller, Beduneau, Mercier, Roze, Lesouhaitier, Terzi, Thille, Kimmoun, Combes.

Statistical analysis: Hajage, Laurent, Combes.

Obtained funding: Schmidt, Hajage, Combes.

Administrative, technical, or material support: Schmidt, Hajage, Sonneville, Muller, Terzi, Combes.

Supervision: Schmidt, Lebreton, Dres, Muller, Lesouhaitier, Combes.

Conflict of Interest Disclosures: Dr Schmidt reported receipt of lecture fees from Getinge, Dräger, Baxter, and Fresenius Medical Care. Dr Dres reported receipt of personal fees from Lungpacer Medical Inc. Dr Guervilly reported receipt of personal fees from Xenios FMC. Dr Sonneville reported receipt of grants from the French Ministry of Health and LFB. Dr Thille reported receipt of grants from the French Ministry of Health and receipt of personal fees (payment for lectures and travel/accommodation to attend scientific meetings) and nonfinancial support from Fisher & Paykel and GE Healthcare. Dr Combes reported receipt of grants from Getinge and personal fees from Getinge, Baxter, and Xenios. No other disclosures were reported.

Funding/Support: This study was funded by grant PHRC APHP-180607 from the Programme Hospitalier de Recherche Clinique–PHRC 2018 (French Ministry of Health).

Role of the Funder/Sponsor: The study funder provided support for research coordination, statistical support, and monitoring but had no role in the design of the study; data collection, study management, and interpretation of the data; preparation, review, or approval of the manuscript; or decision to submit the manuscript for publication.

Group Information: The PRONECMO Investigators are listed in Supplement 3.

Data Sharing Statement: See Supplement 4.

Additional Contributions: We thank all those who worked on or participated to make the PRONECMO trial a success, including the patients enrolled in the trial, all site staff who helped enroll participants, and all health care professionals who cared for the patients throughout the protocol duration. In particular, we thank Armand Mekontso Dessap, MD, PhD (Department of Intensive Care, Hôpital Henri Mondor, Assistance Publique–Hôpitaux de Paris, Creteil, France), Fabio Taccone, MD, PhD (Department of Intensive Care, Hôpital Universitaire de Bruxelles, Université Libre de Bruxelles, Brussels, Belgium), Audrey De-Jong, MD, PhD (Department of Anaesthesia and Intensive Care Unit, Regional University Hospital of Montpellier, St-Eloi Hospital, University of Montpellier, Montpellier, France), and Daniel Mathieu, MD, PhD (CHU Lille, Pôle de Médecine Intensive–Réanimation, Hôpital Roger Salengro, Lille, France) for joining the data and safety monitoring board. We thank Jessica Palmyre, MSc, and Shirmone Botha, MSc (Unité de Recherche Clinique, Centre Hospitalier Universitaire Pitié-Salpêtrière, Assistance Publique–Hôpitaux de Paris, Créteil, France) for their help with study coordination and data collection. These contributors were not compensated beyond their regular salaries.

^✉

Corresponding author.

PMCID: PMC10692949 PMID: 38038395

Key Points

Question

Prone positioning may improve the outcome of patients with severe acute respiratory distress syndrome (ARDS), but is prone position superior to supine position among patients receiving venovenous extracorporeal membrane oxygenation (VV-ECMO) for severe ARDS?

Findings

In this randomized clinical trial that included 170 patients primarily with COVID-19 who were undergoing VV-ECMO, successful ECMO weaning at day 60 occurred in 38 patients (44%) in the prone ECMO group compared with 37 patients (44%) in the supine ECMO group, a nonsignificant difference (subdistribution hazard ratio, 1.11).

Meaning

Among patients with severe ARDS undergoing VV-ECMO, prone positioning did not reduce time to successful ECMO weaning compared with supine position.

Abstract

Importance

Prone positioning may improve outcomes in patients with severe acute respiratory distress syndrome (ARDS), but it is unknown whether prone positioning improves clinical outcomes among patients with ARDS who are undergoing venovenous extracorporeal membrane oxygenation (VV-ECMO) compared with supine positioning.

Objective

To test whether prone positioning vs supine positioning decreases the time to successful ECMO weaning in patients with severe ARDS supported by VV-ECMO.

Design, Setting, and Participants

Randomized clinical trial of patients with severe ARDS undergoing VV-ECMO for less than 48 hours at 14 intensive care units (ICUs) in France between March 3, 2021, and December 7, 2021.

Interventions

Patients were randomized 1:1 to prone positioning (at least 4 sessions of 16 hours) (n = 86) or to supine positioning (n = 84).

Main Outcomes and Measures

The primary outcome was time to successful ECMO weaning within 60 days following randomization. Secondary outcomes included ECMO and mechanical ventilation–free days, ICU and hospital length of stay, skin pressure injury, serious adverse events, and all-cause mortality at 90-day follow-up.

Results

Among 170 randomized patients (median age, 51 [IQR, 43-59] years; n = 60 women [35%]), median respiratory system compliance was 15.0 (IQR, 10.7-20.6) mL/cm H₂O; 159 patients (94%) had COVID-19–related ARDS; and 164 (96%) were in prone position before ECMO initiation. Within 60 days of enrollment, 38 of 86 patients (44%) had successful ECMO weaning in the prone ECMO group compared with 37 of 84 (44%) in the supine ECMO group (risk difference, 0.1% [95% CI, −14.9% to 15.2%]; subdistribution hazard ratio, 1.11 [95% CI, 0.71-1.75]; P = .64). Within 90 days, no significant difference was observed in ECMO duration (28 vs 32 days; difference, −4.9 [95% CI, −11.2 to 1.5] days; P = .13), ICU length of stay, or 90-day mortality (51% vs 48%; risk difference, 2.4% [95% CI, −13.9% to 18.6%]; P = .62). No serious adverse events were reported during the prone position procedure.

Conclusions and Relevance

Among patients with severe ARDS supported by VV-ECMO, prone positioning compared with supine positioning did not significantly reduce time to successful weaning of ECMO.

Trial Registration

ClinicalTrials.gov Identifier: NCT04607551

This randomized trial assesses whether prone vs supine positioning decreases the time to successful weaning of extracorporeal membrane oxygenation (ECMO) in patients with severe acute respiratory distress syndrome (ARDS).

Introduction

Among patients with acute respiratory distress syndrome (ARDS), prone ventilation provided for at least 16 hours per day significantly reduced 90-day mortality.¹ Prone position promotes better overall ventilation/perfusion matching through a more homogeneous distribution of gas-tissue ratios along the dependent-nondependent axis in addition to decreased levels of lung stress and strain.^2,3,4 There is reduced overdistension in nondependent lung regions and less cyclical airspace opening and closing in dependent regions during prone position, perhaps reducing ventilator-induced lung injury.^5,6 Prone positioning may also improve hemodynamics.⁷ Considering these benefits, the use of prone positioning during venovenous extracorporeal membrane oxygenation (VV-ECMO) has increased in many ECMO centers, particularly during the COVID-19 pandemic.⁸ However, evidence supporting prone positioning during ECMO remains low, mostly from retrospective observational studies and meta-analyses suggesting accelerated ECMO weaning and lower mortality when prone positioning was applied during ECMO.^9,10,11,12

To address this knowledge gap, the PRONECMO multicenter randomized clinical trial was designed to determine the effect of early prone positioning during VV-ECMO vs supine positioning on time to successful ECMO weaning in patients with severe ARDS.

Methods

Trial Design

The PRONECMO trial was an investigator-initiated, open-label, parallel-group, multicenter randomized clinical trial conducted at 14 sites in France (eTable 1 in Supplement 1). Eligible intensive care units (ICUs) were medical and surgical centers in France that had clinicians who were experienced in treating adult patients undergoing VV-ECMO and had the capability to perform prone positioning during ECMO. The trial protocol (Supplement 2) was designed by the trial scientific committee and approved by an institutional review board (Comite de Protection des Personnes Ouest II–Angers 2020/59). The study followed CONSORT reporting guidelines.¹³ The trial was financed by a grant from the French Ministry of Health and was sponsored by the Direction de la Recherche Clinique et du Développement, Assistance Publique–Hôpitaux de Paris, which coordinated operational processes, conducted data monitoring and quality checks, and performed statistical analyses. Patient safety was regularly monitored by an independent data and safety monitoring board, who analyzed adverse events in a blinded manner.

Patients

The decision to initiate VV-ECMO was at the discretion of the treating team, although the ECMO to Rescue Lung Injury in Severe ARDS (EOLIA) trial inclusion criteria were recommended.¹⁴ Adult patients were enrolled if they were undergoing invasive mechanical ventilation and were supported by VV-ECMO for less than 48 hours. Exclusion criteria were being younger than 18 years or older than 75 years, pregnancy and breastfeeding, initiation of VV-ECMO more than 48 hours prior, end-stage chronic lung disease, ARDS secondary to abdominal surgery, a moribund condition or a Simplified Acute Physiology Score (SAPS II) value of more than 90 (on a scale from 0 to 163, with higher scores indicating greater severity of illness¹⁵) on the day of randomization, irreversible ARDS with no expectation of lung function recovery, lung transplant, and contraindications for prone positioning (eTable 2 in Supplement 1). As patients were unable to give written consent at inclusion, written informed consent was obtained from a close relative or surrogate. According to the specifications of emergency consent, randomization without close relative or surrogate consent could be performed. The family member or close relative/legal representative and the patient were informed of the inclusion as soon as possible and consented in writing to their continuation in the trial.

Randomization

Eligible patients were randomly assigned in a 1:1 ratio to either prone positioning while undergoing ECMO (prone ECMO group) or supine positioning while undergoing ECMO (supine ECMO group) in permuted blocks of size 4 or 6. Randomization was performed through a secure web-based system using the minimization method: the first 20 patients were randomly assigned (1:1). The remaining patients had an 80% probability of being assigned to the group that minimizes imbalances on the following characteristics: center, prone positioning before ECMO (yes vs no), respiratory system compliance (≤20 or >20 mL/cm H₂0) at randomization, and COVID-19 status (positive vs negative).

Trial Intervention

For patients assigned to the prone ECMO group, the proning procedure was applied according to a written protocol (see video at https://we.tl/t-1K5mLgITRi and eTable 3 in Supplement 1). At least 4 prone position sessions of 16 hours were performed during the first 4 days unless a patient met the following predefined criteria for early stopping of the prone intervention: oxygenation and respiratory system compliance improvement leading to achievement of ECMO weaning criteria, Pao₂ deterioration after 16 hours of prone positioning by more than 20% relative to supine, or severe proning-related complications leading to immediate interruption of the session (eTable 3 in Supplement 1). Use of continuous neuromuscular blockade during prone positioning or prone positioning while undergoing ECMO after 4 sessions was recommended but left to the physician’s discretion. For patients assigned to the supine ECMO group, prone positioning while undergoing ECMO was not allowed before day 60. Physicians could place a patient in prone position as rescue therapy in cases of refractory hypoxemia during ECMO (eTable 3 in Supplement 1).

Cointerventions

In both groups, patients received similar sedation, anticoagulation, ECMO, and circuit management. Ultraprotective lung ventilation targeting a low tidal volume (<4 mL/kg), a positive end-expiratory pressure greater than 10 cm H₂0, low plateau pressure (≤24 cm H₂O), and low respiratory rate (≤20/min) was applied in both groups and maintained during prone position sessions (eTable 4 in Supplement 1). Protocolized management was applied to both groups regarding weaning from ECMO. The protocol was undertaken daily as soon as improvement was noted in respiratory system compliance, arterial oxygenation, or chest radiography. Successful weaning criteria allowing ECMO withdrawal are detailed in eTable 5 in Supplement 1.¹⁴

Data Collection

Patients’ characteristics, severity of illness, pre-ECMO arterial blood gas measures, mechanical ventilation settings (see Box for definitions of mechanical ventilation– and ECMO-related measures), and adjunctive therapies were documented at enrollment. Intervention and cointervention data during ICU stay were recorded up to 60 days after randomization, including mechanical ventilation settings, Sequential Organ Failure Assessment (SOFA) scores, and vasopressor administration. Data were specifically documented before and after 16 hours of prone positioning during the first 4 days of ECMO. Data were collected on the number of patients for whom prone positioning was prematurely stopped.

Box. Definitions of Mechanical Ventilation and Extracorporeal Membrane Oxygenation (ECMO) Measures.

Mechanical Ventilation

Positive end-expiratory pressure (PEEP): Measured in centimeters of H₂O, PEEP is the positive pressure that remains in the airways at the end of the respiratory cycle (end of exhalation) that is greater than the atmospheric pressure in patients receiving mechanical ventilation.
Tidal volume: Measured in milliliters per kilogram of predicted body weight, tidal volume is the amount of air that moves in or out of the lungs with each respiratory cycle.
Driving pressure: Measured in centimeters of H₂O, driving pressure is plateau airway pressure minus PEEP. It can also be expressed as the ratio of tidal volume to respiratory system compliance, indicating the decreased functional size of the lung.
Respiratory rate: Measured in breaths per minute, respiratory rate is the rate at which breathing occurs; it can be set and controlled by the ventilator.
Respiratory system compliance: Measured in millimeters per centimeter of H₂O, respiratory system compliance measures lung ability to expand and accommodate changes in volume (ie, elasticity) in response to changes in pressure. It is expressed as tidal volume/(plateau pressure − PEEP).

ECMO Configuration/Settings

Femorojugular: Configuration of the venovenous ECMO setting. The drainage cannula is inserted into the femoral vein and the return cannula into the internal jugular vein.
ECMO flow: Measured in liters per minute, ECMO flow is the amount of blood that is pumped continuously out of the body and sent through a membrane that adds oxygen and removes carbon dioxide.
Fmo₂: Measured in percentage, Fmo₂ is the fraction of oxygen of the gas flowing through the membrane oxygenator.
Sweep gas flow: Measured in liters per minute, sweep gas flow is gas flow through the membrane oxygenator. An increase in sweep gas flow increases CO₂ elimination.

Outcomes

The primary outcome was time to successful ECMO weaning within 60 days following randomization. ECMO weaning (ie, ECMO separation) was considered successful only if a patient survived without ECMO or lung transplant for 30 days after ECMO discontinuation. Two competing events were therefore considered: (1) weaning failure, defined as need for a second ECMO run or lung transplant or death within 30 days after ECMO separation, and (2) death while undergoing ECMO. Patients still alive while undergoing ECMO 60 days after randomization without any competing events were censored. In patients for whom several types of events (successful weaning, weaning failure, death) occurred during follow-up, only the first event was considered in the analysis of the primary outcome. Thus, the qualification for successful ECMO weaning required 30 days of follow-up after ECMO discontinuation (ie, up to a maximum of 90 days after randomization if ECMO was removed on day 60).

The predefined secondary outcomes were all-cause mortality at 90-day follow-up, time to achieve a respiratory system compliance greater than 30 mL/cm H₂O within 7 days, number of days alive without cardiovascular or kidney failure (defined by a SOFA cardiovascular and kidney score of 0-1), incidence of ventilator-associated pneumonia, ECMO- and mechanical ventilation–free days (death was considered 0 days), ICU and hospital lengths of stay, and need for venoarterial ECMO at 60-day and 90-day follow-up.

Predefined adverse events included skin pressure injuries evaluated daily during the first 7 days after randomization with the Revised Pressure Injury Staging System.¹⁶ This scale grades pressure ulcers in 12 sections of the body from a grade 1 pressure injury (ie, soft tissue injury without ulceration) to a grade 4 pressure injury (ie, full-thickness skin loss and tissue loss). The score ranges from 0 to 48, with higher scores indicating more severe skin pressure injuries. Predefined serious adverse events (unintentional decannulation, nonscheduled extubation during the procedure, hemoptysis, cardiac arrest) were also assessed at 60-day follow-up.

Sample Size Calculation

Based on data from a randomized clinical trial conducted in patients with severe ARDS undergoing VV-ECMO in France,¹⁴ we estimated that the cumulative incidence of successful ECMO weaning in the control group would be at least 60% in the presence of competing causes (death). According to Latouche et al,¹⁷ 170 patients in total (85 per group) were required to have 80% power to detect a subdistribution hazard ratio of 1.75 with an α = .05 (bilateral formulation).

Statistical Analysis

Baseline characteristics are reported as counts with percentages for categorical variables and as means with SDs or medians with IQRs for continuous variables. Primary and secondary analyses were conducted according to as-randomized principles. The primary analysis was adjusted for minimization stratification factors. The primary end point was time to successful ECMO weaning within the 60 days following randomization, in the presence of the competing risks of death and weaning failure. The cumulative incidence curves for these 3 competing events were drawn for each randomization group. The cumulative incidence of successful ECMO weaning was compared between groups using the Gray test. The subdistribution hazard ratios were estimated (with their 95% CIs) for the 3 events using a Fine and Gray competing risk regression. The estimated subdistribution hazard ratio associated with successful weaning represents an estimation of the total effect of the randomized intervention for the primary event of interest. We conducted analyses of the primary end point in predefined and post hoc subgroups of interest. We postulated that patients with respiratory system compliance of less than or equal to 20 mL/cm H₂O or with a higher body mass index (ie, exceeding the median body mass index of our study population) would have been more likely to experience a successful ECMO weaning with prone positioning. A sensitivity analysis was also performed by estimating cause-specific hazard ratios and their 95% CIs with a cause-specific Cox regression model. The cause-specific hazard ratio associated with successful weaning represents an estimation of the direct effect of the randomized intervention on the primary event of interest in a counterfactual world in which competing events are eliminated.

Categorical outcomes were compared with the χ² test or Fisher exact test and continuous outcomes with the t test or Wilcoxon signed rank test, as appropriate. Censored outcomes were described over time using the Kaplan-Meier method and compared using log-rank tests. All analyses were conducted at a 2-sided α = .05 and were performed using R software, version 4.1.3 (R Foundation for Statistical Computing).

Results

Study Sites and Patients

From March 3, 2021, to December 7, 2021, 250 patients undergoing ECMO for severe ARDS were assessed for eligibility, of whom 170 were randomized (86 randomized to prone positioning and 84 to supine positioning) (Figure 1) and completed follow-up. As shown in Table 1, the median age was 51 (IQR, 43-59) years and 60 patients (35%) were women. The leading cause of ARDS was COVID-19 pneumonia (n = 159 [94%]). The pre-ECMO median ratio of Pao₂ to fraction of inspired oxygen was 66 mm Hg (IQR, 56-80 mm Hg), and patients underwent cannulation after a median of 4 days (IQR, 1-6 days) after intubation. Prior to the initiation of ECMO, 85 (99% in the prone ECMO group) and 79 (94% in the supine ECMO group) of the patients had been in prone position, respectively. The median SOFA score was 9 (IQR, 8-13), the median tidal volume during ECMO was 3.0 (IQR, 2.0-4.3) mL/kg of predicted body weight, and the median respiratory system compliance was 14.0 (IQR, 10.0-23.8) mL/cm H₂O and 15.0 (IQR, 11.7-19.4) mL/cm H₂O in the prone and supine ECMO groups, respectively.

Figure 1. — ARDS indicates acute respiratory distress syndrome; SAPS II, Simplified Acute Physiology Score; and VV-ECMO, venovenous extracorporeal membrane oxygenation.

Table 1. Baseline Participant Characteristics^a.

Characteristics	Prone ECMO (n = 86)	Supine ECMO (n = 84)
Age, median (IQR), y	52 (44-60)	50 (41-58)
Sex, No. (%)
Female	24 (37.9)	36 (32.9)
Male	62 (72.1)	48 (57.1)
Body mass index, median (IQR)^b	32.7 (28.4-38.1)	33.1 (28.8-36.8)
SAPS II score, median (IQR)^c	51 (36-60)	50 (33-56)
Comorbidities, No. (%)
Diabetes	18 (20.9)	17 (20.2)
Chronic respiratory disease^d	7 (8.1)	10 (11.9)
Ischemic cardiomyopathy	6 (7.0)	0
Immunocompromised^e	4 (5)	3 (4)
Time from ICU admission to intubation, median (IQR), d	1 (0-4)	1 (0-4)
Time from intubation to ECMO, median (IQR), d	3 (1-6)	5 (2-7)
Time from ECMO initiation to randomization, median (IQR), d	1 (0-1)	1 (0-1)
ARDS etiology, No. (%)
COVID-19 pneumonia	80 (93.0)	79 (92.9)
Bacterial pneumonia	6 (7.0)	3 (3.6)
Other	0	2 (2.4)^f
Pre-ECMO parameters
Ventilation, median (IQR)
Fio₂	100 (100-100)	100 (100-100)
Positive end-expiratory pressure, cm H₂O	12 (10-15)	12 (10-14)
Tidal volume, mL/kg of predicted body weight	5.9 (5.3-6.3)	6.0 (5.2-6.4)
Respiratory rate, /min	30 (25-33)	30 (25-32)
Plateau pressure, cm H₂O	30 (29-33)	31 (29-34)
Respiratory system compliance, mL/cm H₂O	22.0 (16.0-29.5)	20.5 (14.7-27.0)
Blood gas measurements
pH, median (IQR)	7.31 (7.26-7.40)	7.30 (7.20-7.40)
Pao₂/Fio₂, median (IQR), mm Hg	66 (55-77)	67 (59-80)
≤80, No. (%)	67 (83.7)	59 (76.6)
≤50, No. (%)	9 (11.2)	9 (11.7)
Paco₂, median (IQR), mm Hg	54 (48-64)	59 (51-70)
Paco₂ ≥60 mm Hg and pH ≤7.25, No. (%)	14 (17.5)	20 (25.6)
Arterial lactate, median (IQR), mmol/L	1.8 (1.0-2.0)	1.8 (1.2-2.0)
Adjunctive therapies
Prone positioning, No. (%)	85 (98.8)	79 (94.1)
No. of sessions, median (IQR)	2 (1-3)	3 (2-4)
Continuous neuromuscular blockade, No. (%)	78 (94.0)	65 (95.6)
Inhaled nitric oxide, No. (%)	44 (53.7)	35 (52.2)
High-dose corticosteroids, No. (%)^g	22 (26.5)	20 (29.4)
Recruitment maneuvers, No. (%)	4 (5.2)	6 (9.7)
Pneumothorax, No. (%)	7 (8.2)	7 (8.4)
Parameters at randomization
SOFA score, median (IQR)^h	9 (8-13)	9 (8-12)
Ventilation parameters while undergoing ECMO
Positive end-expiratory pressure, median (IQR), cm H₂O	12 (10-14)	12 (12-14)
Tidal volume, median (IQR), mL/kg of predicted body weight	3.0 (2.0-4.3)	3.1 (2.3-3.9)
Driving pressure, median (IQR), cm H₂O	14 (11-15)	14 (11-14)
Respiratory rate, median (IQR), /min	20 (12-20)	20 (12-20)
Respiratory system compliance, median (IQR), mL/cm H₂O	14.0 (10.0-23.8)	15.0 (11.7-19.4)
Respiratory system compliance ≤20 mL/cm H₂O, No. (%)	40 (46.5)	42 (50)
ECMO settings
Femorojugular, No. (%)	85 (100)	82 (98.8)
ECMO flow, median (IQR), L/min	4.8 (4.2-5.5)	4.8 (4.1-5.3)
Fmo₂, median (IQR), %	100 (90-100)	100 (100-100)
Sweep gas flow, median (IQR), L/min	3 (2-5)	4 (2-5)
Receiving vasopressor or inotropes, No. (%)	50 (58.1)	46 (54.8)

Open in a new tab

Abbreviations: ARDS, acute respiratory distress syndrome; ECMO, extracorporeal membrane oxygenation; Fio₂, fraction of inspired oxygen; Fmo₂, fraction of oxygen in the ECMO membrane; ICU, intensive care unit.

^{^a}

See the Box for definitions of mechanical ventilation and ECMO measures.

^{^b}

Calculated as weight in kilograms divided by height in meters squared.

^{^c}

The Simplified Acute Physiology Score (SAPS II) measures severity of illness 24 hours after admission to the ICU. It is based on 12 physiological variables and 3 disease-related variables. The score ranges from 0 to 163, with higher scores indicating more severe disease and higher risk of death.

^{^d}

Defined as chronic obstructive or restrictive pulmonary disease or asthma.

^{^e}

Defined as hematological malignancy, active treatment for solid tumor, solid organ transplant, AIDS, or long-term treatment with corticosteroids or immunosuppressants.

^{^f}

One instance each of pancreatitis and abdominal-related septic shock.

^{^g}

High dose defined as more than 1 mg/kg per day of equivalent prednisone.

^{^h}

The Sequential Organ Failure Assessment (SOFA) score ranges from 0 to 24, with higher scores indicating a greater degree of organ dysfunction.

Intervention

All patients in the interventional group were placed in prone position immediately after randomization, and 69 patients (80%) received at least 4 prone position sessions (median, 4 [IQR, 4-5] sessions). In the prone ECMO group, 17 patients (19.8%) discontinued the prone intervention before achievement of 4 prone position sessions (eTable 7 in Supplement 1), most often due to improvement in oxygenation and respiratory system compliance. No significant differences in respiratory system compliance were present between groups during the first 7 days of ECMO (P = .08) (eFigure 1 in Supplement 1). Similarly, respiratory system compliance, Pao₂, and Paco₂ did not significantly change before and after the first prone position session (eTable 8 in Supplement 1). Among patients randomized to the supine ECMO group, 2 (2.4%) were placed in prone position for refractory hypoxemia while undergoing ECMO (eTable 3 in Supplement 1).

Primary Outcome

Successful ECMO weaning within 60 days of randomization occurred in 38 of 86 patients (44.2%) and 37 of 84 patients (44.0%) in the prone and supine ECMO groups, respectively (risk difference, 0.1% [95% CI, −14.9% to 15.2%]; subdistribution hazard ratio, 1.11 [95% CI, 0.71-1.75]; P = .64) (Table 2 and Figure 2). ECMO weaning failure and death, the 2 other competing components of the primary outcome, were not significantly different between groups (Table 2). Among patients with ECMO weaning failure, 10 were weaned from ECMO but died (8 patients in the prone ECMO group; 2 patients in the supine ECMO group), and 2 patients needed a second ECMO run within 30 days after ECMO separation. None had a lung transplant (eFigure 2 in Supplement 1). At day 60, 7 patients in the prone ECMO group and 13 patients in the supine ECMO group were still undergoing ECMO and were censored. Results of the cause-specific analysis were similar (eTable 9 in Supplement 1).

Table 2. Primary and Secondary End Points and Adverse Events^a.

Outcomes/events	Prone ECMO (n = 86)	Supine ECMO (n = 84)	Mean, median, or risk difference, (95% CI)	Relative difference (95% CI)	P value
Primary outcome
Successful ECMO weaning by day 60, No. (%)	38 (44.2)	37 (44.0)	0.1 (−14.9 to 15.2)	sHR, 1.11 (0.71-1.75)	.64
Competing events
Death before ECMO weaning, No. (%)	32 (37.2)	31 (36.9)	0.3 (−14.5 to 14.1)	sHR, 3.76 (0.71-21.1)	.12
ECMO weaning failure, No. (%)^b	9 (10.5)	3 (3.6)	6.9 (−1.9 to 15.7)	sHR, 0.94 (0.58-1.53)	.80
Secondary outcomes
Respiratory system compliance ≥30 mL/cm H₂O, No. (%)^c
On day 2	24 (27.9)	17 (20.2)	7.7 (−6.3 to 21.6)	1.38 (0.80-2.38)	.24
On day 7	33 (38.4)	26 (30.9)	7.4 (−8 to 22.9)	1.24 (0.82-1.88)	.31
Days alive and free from kidney failure within 7 days, median (IQR)^d	7 (6-7)	7 (6-7)	0 (0 to 0)		.86
Days alive and free from cardiovascular failure within 7 days, median (IQR)^d	5 (3-7)	5 (1-7)	0 (−2.5 to 1)		.32
Pneumothorax by day 60, No. (%)	14 (16)	17 (21)	−4 (−16.7 to 8.8)	0.80 (0.42-1.53)	.46
≥1 Ventilatory-associated pneumonia episode, No. (%)	73 (85)	75 (89)	−4.4 (−15.6 to 6.8)	0.95 (0.85-1.07)	.49
All-cause day 60 mortality, No. (%)	40 (47)	35 (42)	4.8 (−11.2 to 20.9)	1.18 (0.75-1.87)	.48
All-cause day 90 mortality, No. (%)	44 (51)	40 (48)	2.4 (−13.9 to 18.6)	1.1 (0.72-1.69)	.62
Days free from ECMO by day 90, median (IQR)	0 (0-73)	0 (0-64)	0 (−51.5 to 37)		.60
Days alive and free from mechanical ventilation by day 90, median (IQR)	0 (0-51)	0 (0-50)	0 (−4 to 21.9)		.84
Days receiving ECMO during first 90 days, mean (SD)	27.51 (20.39)	32.19 (23.95)	−4.9 (−11.2 to 1.5)		.13
Days receiving mechanical ventilation during first 90 days, mean (SD)	49.22 (30.06)	52.21 (28.78)	−3.0 (−10.9 to 4.8)		.62
Days in intensive care unit during first 90 days, mean (SD)	42.47 (25.44)	46.26 (26.88)	−3.8 (−10.6 to 4.3)		.43
Days in hospital during first 90 days, mean (SD)	59.79 (28.86)	59.36 (28.15)	0.4 (−8.0 to 8.9)		.97
Adverse events by day 60
Serious adverse events, No. (%)
≥1 Cardiac arrest	3 (3.5)	11 (13.1)	−9.6 (−19 to −0.2)	0.27 (0.08-0.92)	.05
Bleeding event requiring packed red blood cell transfusion	24 (27.9)	32 (38.1)	−3.0 (−17.3 to 11.4)	0.89 (0.54-1.48)	.79
Hemorrhagic stroke	2 (2.3)	1 (1.2)	1.1 (−3.9 to 6.2)	1.95 (0.18-21.14)	>.99
Unintentional ECMO decannulation	0	0
Nonscheduled extubation, No. (%)	0	0
Severe hemoptysis, No. (%)	0	0
Maximum Revised Pressure Injury Staging System score, median (IQR)^e	8 (4-11)	6 (2-10)	2 (−1 to 6)		.14

Open in a new tab

Abbreviations: ECMO, extracorporeal membrane oxygenation; sHR, subdistribution hazard ratio.

^{^a}

No patients were lost to follow-up.

^{^b}

ECMO weaning failure was defined as the need for a second ECMO run or lung transplant or death within 30 days after ECMO discontinuation.

^{^c}

Respiratory system compliance measures the ability of the lungs to expand and accommodate changes in volume (ie, elasticity) in response to changes in pressure. It is expressed as tidal volume/(plateau pressure − positive end-expiratory pressure).

^{^d}

Being free from kidney failure or cardiovascular failure was defined as a kidney or cardiovascular component score of less than 2 on the Sequential Organ Failure Assessment, respectively.

^{^e}

This scale grades pressure ulcers in 12 sections of the body, from a grade 1 pressure injury (ie, soft tissue injury without ulceration) to a grade 4 pressure injury (ie, full-thickness skin loss and tissue loss). The score ranges from 0 to 48, with higher scores indicating more severe skin pressure injuries. It was evaluated daily during the first 7 days.

Figure 2. — Weaning failure was defined as need for a second extracorporeal membrane oxygenation (ECMO) run or lung transplant or death within 30 days after ECMO discontinuation. Among patients with ECMO weaning failure, 8 in the prone ECMO group and 2 in the supine ECMO group were weaned from ECMO but died within 30 days after ECMO discontinuation, 1 patient in each group needed a second ECMO run within 30 days after ECMO discontinuation, and none had a lung transplant.

^aEvents refer to death, successful weaning, or weaning failure.

Secondary Outcomes

No significant differences were present in any of the prespecified secondary end points (Table 2). At day 90, 44 patients (51%) in the prone ECMO group and 40 (48%) in the supine ECMO group had died (absolute risk difference, 2.4% [95% CI, −13.9% to 18.6%]; hazard ratio, 1.12 [95% CI, 0.72-1.72]; P = .62) (eFigure 3 in Supplement 1). ECMO and mechanical ventilation durations, ICU length of stay, and hospital length of stay were not significantly different between groups (eFigures 4-7 in Supplement 1). By 60 days, 73 of 86 patients (85%) had at least 1 ventilator-associated episode of pneumonia in the prone ECMO group compared with 75 of 84 patients (89%) in the supine ECMO group (absolute risk difference, −4.4% [95% CI, −15.6% to 6.8%]; relative risk, 0.95 [95% CI, 0.85-1.07]; P = .49). Incomplete data on the occurrence of acute core pulmonale and involved lung quadrants on chest radiography limited those analyses. Two patients were switched from VV-ECMO to venoarterial ECMO.

Adverse Events

The rate of cardiac arrest was significantly greater in patients in the supine ECMO group compared with those in the prone ECMO group (11 [11.1%] vs 3 [3.5%]; absolute risk difference, −9.6% [95% CI, −19.0% to −0.2%]; relative risk, 0.27 [95% CI, 0.08-0.92]; P = .05). The rates of bleeding events leading to packed red blood cell transfusion and hemorrhagic stroke were similar in the 2 groups. No unintentional ECMO decannulation, nonscheduled extubation, or severe hemoptysis occurred during prone positioning procedures. The Revised Pressure Ulcer Injury Staging System score was not significantly higher in the prone ECMO group (median, 8 [IQR, 4-11] vs 6 [IQR, 2-10]; relative risk difference, 2% [95% CI, −1% to 6%]; P = .14) (Table 2).

Subgroup Analyses

There were no significant treatment effect differences in prespecified subgroups (Figure 3). The low rates of non-COVID ARDS (6.5%) and patients not placed in prone position before ECMO (3.6%) precluded performing prespecified analyses in these subgroups. A post hoc subgroup analysis by median body mass index and center ECMO experience (defined as more than 30 patients included in the study) did not show statistically significant heterogeneity of treatment effect between the 2 study groups.

Discussion

In this randomized clinical trial among patients with severe ARDS receiving VV-ECMO, the early application of prone position during ECMO was not associated with a shorter time to successful ECMO weaning. Secondary outcomes including survival at days 60 and 90, ECMO- and mechanical ventilation–free days, and ICU and hospital lengths of stay were not different between groups.

Recent retrospective cohort studies of patients receiving VV-ECMO showed that prone positioning during ECMO was significantly associated with a greater probability of being alive and weaned from ECMO.^9,10 In recent systematic review and meta-analysis of 1836 patients from 13 observational studies, prone positioning during ECMO was associated with a significant improvement in ventilator-free days and ICU survival (relative risk, 1.31; 95% CI, 1.21-1.41), present in both COVID-19 and non–COVID-19 patients receiving ECMO.¹⁸ Indeed, a shorter duration of ECMO support may decrease ECMO-related complications such as severe hemorrhage,^19,20 neurological events,²¹ nosocomial infections,²² and decreased resource utilization. As such, the primary outcome in the current trial was pragmatic and combined both mortality at day 60 alone and ECMO weaning status (ie, success or failure).

Despite promising findings in observational studies, prone positioning of patients undergoing ECMO failed to reduce ECMO duration or mortality in this randomized trial. Several factors may explain this finding. First, the majority of enrolled patients were placed in prone position before ECMO, compared with widely variable proportions of patients in previous cohorts.^{10,11,18,23,24} Second, the majority of patients had COVID-19–related ARDS, perhaps with a greater severity of pulmonary injury. Prior international cohorts^25,26,27 suggested that patients with COVID-19 rescued by ECMO had longer ECMO runs and ICU lengths of stay than that reported in the EOLIA trial¹⁴ or other large, retrospective series of ECMO for non–COVID-19–related ARDS.^28,29,30 Thus, it remains uncertain whether the current trial findings are generalizable to patients with severe non–COVID-19–related ARDS.

The neutral result also raises the prospect that the etiology or severity of ARDS and subsequent biology may exhibit differential responses to prone positioning. The 60-day mortality rate of 44% was greater than the rates reported in cohorts of patients with COVID-19 undergoing ECMO treated in experienced ECMO centers in Europe.⁸ The COVID-19 variant may also contribute to prone positioning response, as less favorable pandemic outcomes were observed in late 2020^8,27,31 and the Delta variant was predominant in France during part of the trial.⁸ This contrasts with early reports of prone positioning benefit in COVID-19 patients receiving ECMO during predominance of the wild-type SARS-CoV-2 strain.^{12,14,18,24,25,26}

Limitations

This study has several limitations. First, the sample size may have been insufficient to detect a clinically meaningful treatment effect. However, the rate of use of ECMO in patients with ARDS is still low—only 7% of patients with severe ARDS included in the Large Observational Study to Understand the Global Impact of Severe Acute Respiratory Failure (LUNG SAFE), which was performed before the COVID-19 pandemic.³² This rate may have been even lower if the EOLIA criteria for ECMO initiation had been strictly applied,¹⁴ especially with the systematic use of prone positioning before considering ECMO. Second, 69 of 170 patients were enrolled from a single very experienced center, which could limit the generalizability of our results. However, it should also be noted that 170 of 250 patients screened during the study period were included in the trial, suggesting that our inclusion criteria are broadly applicable to this patient population. Third, the trial was unblinded due to the nature of the intervention, and this design may introduce bias.

Conclusions

Among patients with severe ARDS supported by VV-ECMO, prone positioning during ECMO did not significantly reduce time to successful weaning of ECMO.

Section Editor: Christopher Seymour, MD, Associate Editor, JAMA (christopher.seymour@jamanetwork.org).

Supplement 1.

eTable 1. Characteristics of participating sites

eTable 2. Inclusion and exclusion criteria

eTable 3. Prone positioning procedure on ECMO

eTable 4. ECMO management protocol

eTable 5. ECMO weaning

eTable 6. Patient enrollment by participating sites

eTable 7. Reasons for discontinuing PP or having < 4 PP sessions for patients randomized in the prone ECMO arm

eTable 8. Evolution of respiratory system compliance, PaO2, and PaCO2 before and after 16 hours of prone positioning on ECMO (first prone positioning session).

eTable 9. Cause-specific analysis of the primary outcome

eFigure 1. Median respiratory system compliance (interquartile range) and its 95% confidence interval from Day 1 to Day 7 post-randomization according to prone or supine position

eFigure 2. Cumulate incidence function for the events of ECMO successful weaning, weaning failure due to death after ECMO separation, weaning failure due to the need for a second ECMO run, and death, by position group

eFigure 3. Kaplan-Meier time-to-event curves for survival in the prone and supine ECMO groups

eFigure 4. Kaplan-Meier time-to-event curves for the probability of being on ECMO in the prone and supine ECMO groups

eFigure 5. Kaplan-Meier time-to-event curves for the probability of being on mechanical ventilation in the prone and supine ECMO groups

eFigure 6. Kaplan-Meier time-to-event curves for the probability of remaining in the ICU in the prone and supine ECMO groups

eFigure 7. Kaplan-Meier time-to-event curves for the probability of remaining in the hospital in the prone and supine ECMO groups

Click here for additional data file.^{(846.5KB, pdf)}

Supplement 2.

Trial Protocol

Click here for additional data file.^{(1.9MB, pdf)}

Supplement 3.

Nonauthor Collaborators. PRONECMO Investigators

Click here for additional data file.^{(192.9KB, pdf)}

Supplement 4.

Data Sharing Statement

Click here for additional data file.^{(35.3KB, pdf)}

References

1.Guérin C, Reignier J, Richard JC, et al. ; PROSEVA Study Group . Prone positioning in severe acute respiratory distress syndrome. N Engl J Med. 2013;368(23):2159-2168. doi: 10.1056/NEJMoa1214103 [DOI] [PubMed] [Google Scholar]
2.Gattinoni L, Taccone P, Carlesso E, Marini JJ. Prone position in acute respiratory distress syndrome: rationale, indications, and limits. Am J Respir Crit Care Med. 2013;188(11):1286-1293. doi: 10.1164/rccm.201308-1532CI [DOI] [PubMed] [Google Scholar]
3.Gattinoni L, Pesenti A, Carlesso E. Body position changes redistribute lung computed-tomographic density in patients with acute respiratory failure: impact and clinical fallout through the following 20 years. Intensive Care Med. 2013;39(11):1909-1915. doi: 10.1007/s00134-013-3066-x [DOI] [PubMed] [Google Scholar]
4.Mure M, Domino KB, Lindahl SG, Hlastala MP, Altemeier WA, Glenny RW. Regional ventilation-perfusion distribution is more uniform in the prone position. J Appl Physiol (1985). 2000;88(3):1076-1083. doi: 10.1152/jappl.2000.88.3.1076 [DOI] [PubMed] [Google Scholar]
5.Broccard AF, Shapiro RS, Schmitz LL, Ravenscraft SA, Marini JJ. Influence of prone position on the extent and distribution of lung injury in a high tidal volume oleic acid model of acute respiratory distress syndrome. Crit Care Med. 1997;25(1):16-27. doi: 10.1097/00003246-199701000-00007 [DOI] [PubMed] [Google Scholar]
6.Broccard A, Shapiro RS, Schmitz LL, Adams AB, Nahum A, Marini JJ. Prone positioning attenuates and redistributes ventilator-induced lung injury in dogs. Crit Care Med. 2000;28(2):295-303. doi: 10.1097/00003246-200002000-00001 [DOI] [PubMed] [Google Scholar]
7.Jozwiak M, Teboul JL, Anguel N, et al. Beneficial hemodynamic effects of prone positioning in patients with acute respiratory distress syndrome. Am J Respir Crit Care Med. 2013;188(12):1428-1433. doi: 10.1164/rccm.201303-0593OC [DOI] [PubMed] [Google Scholar]
8.Schmidt M, Hajage D, Landoll M, et al. ; ECMO-SURGES Investigators . Comparative outcomes of extracorporeal membrane oxygenation for COVID-19 delivered in experienced European centres during successive SARS-CoV-2 variant outbreaks (ECMO-SURGES): an international, multicentre, retrospective cohort study. Lancet Respir Med. 2023;11(2):163-175. doi: 10.1016/S2213-2600(22)00438-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Guervilly C, Prud’homme E, Pauly V, et al. Prone positioning and extracorporeal membrane oxygenation for severe acute respiratory distress syndrome: time for a randomized trial? Intensive Care Med. 2019;45(7):1040-1042. doi: 10.1007/s00134-019-05570-9 [DOI] [PubMed] [Google Scholar]
10.Petit M, Fetita C, Gaudemer A, et al. Prone-positioning for severe acute respiratory distress syndrome requiring extracorporeal membrane oxygenation. Crit Care Med. 2022;50(2):264-274. doi: 10.1097/CCM.0000000000005145 [DOI] [PubMed] [Google Scholar]
11.Giani M, Rezoagli E, Guervilly C, et al. ; European Prone Positioning During Extracorporeal Membrane Oxygenation (EuroPronECMO) Investigators . Timing of prone positioning during venovenous extracorporeal membrane oxygenation for acute respiratory distress syndrome. Crit Care Med. 2023;51(1):25-35. doi: 10.1097/CCM.0000000000005705 [DOI] [PubMed] [Google Scholar]
12.Zaaqoq AM, Barnett AG, Griffee MJ, et al. ; COVID-19 Critical Care Consortium . Beneficial effect of prone positioning during venovenous extracorporeal membrane oxygenation for coronavirus disease 2019. Crit Care Med. 2022;50(2):275-285. doi: 10.1097/CCM.0000000000005296 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Schulz KF, Altman DG, Moher D; CONSORT Group . CONSORT 2010 statement: updated guidelines for reporting parallel group randomised trials. BMJ. 2010;340:c332. doi: 10.1136/bmj.c332 [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Combes A, Hajage D, Capellier G, et al. ; EOLIA Trial Group, REVA, and ECMONet . Extracorporeal membrane oxygenation for severe acute respiratory distress syndrome. N Engl J Med. 2018;378(21):1965-1975. doi: 10.1056/NEJMoa1800385 [DOI] [PubMed] [Google Scholar]
15.Le Gall JR, Lemeshow S, Saulnier F. A new Simplified Acute Physiology Score (SAPS II) based on a European/North American multicenter study. JAMA. 1993;270(24):2957-2963. doi: 10.1001/jama.1993.03510240069035 [DOI] [PubMed] [Google Scholar]
16.Edsberg LE, Black JM, Goldberg M, McNichol L, Moore L, Sieggreen M. Revised National Pressure Ulcer Advisory Panel Pressure Injury Staging System: Revised Pressure Injury Staging System. J Wound Ostomy Continence Nurs. 2016;43(6):585-597. doi: 10.1097/WON.0000000000000281 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Latouche A, Porcher R. Sample size calculations in the presence of competing risks. Stat Med. 2007;26(30):5370-5380. doi: 10.1002/sim.3114 [DOI] [PubMed] [Google Scholar]
18.Papazian L, Schmidt M, Hajage D, et al. Effect of prone positioning on survival in adult patients receiving venovenous extracorporeal membrane oxygenation for acute respiratory distress syndrome: a systematic review and meta-analysis. Intensive Care Med. 2022;48(3):270-280. doi: 10.1007/s00134-021-06604-x [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Mansour A, Flecher E, Schmidt M, et al. ; ECMOSARS Investigators . Bleeding and thrombotic events in patients with severe COVID-19 supported with extracorporeal membrane oxygenation: a nationwide cohort study. Intensive Care Med. 2022;48(8):1039-1052. doi: 10.1007/s00134-022-06794-y [DOI] [PubMed] [Google Scholar]
20.Martucci G, Schmidt M, Agerstrand C, et al. ; International ECMO Network (ECMONet) . Transfusion practice in patients receiving VV ECMO (PROTECMO): a prospective, multicentre, observational study. Lancet Respir Med. 2023;11(3):245-255. doi: 10.1016/S2213-2600(22)00353-8 [DOI] [PubMed] [Google Scholar]
21.Luyt CE, Bréchot N, Demondion P, et al. Brain injury during venovenous extracorporeal membrane oxygenation. Intensive Care Med. 2016;42(5):897-907. doi: 10.1007/s00134-016-4318-3 [DOI] [PubMed] [Google Scholar]
22.Abrams D, Grasselli G, Schmidt M, Mueller T, Brodie D. ECLS-associated infections in adults: what we know and what we don’t yet know. Intensive Care Med. 2020;46(2):182-191. doi: 10.1007/s00134-019-05847-z [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Giani M, Martucci G, Madotto F, et al. Prone positioning during venovenous extracorporeal membrane oxygenation in acute respiratory distress syndrome: a multicenter cohort study and propensity-matched analysis. Ann Am Thorac Soc. 2021;18(3):495-501. doi: 10.1513/AnnalsATS.202006-625OC [DOI] [PubMed] [Google Scholar]
24.Massart N, Guervilly C, Mansour A, et al. ; Extracorporeal Membrane Oxygenation for Respiratory Failure and/or Heart Failure Related to Severe Acute Respiratory Syndrome Coronavirus 2 (ECMOSARS) Investigators . Impact of prone position in COVID-19 patients on extracorporeal membrane oxygenation. Crit Care Med. 2023;51(1):36-46. doi: 10.1097/CCM.0000000000005714 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Schmidt M, Hajage D, Lebreton G, et al. ; Groupe de Recherche Clinique en Reanimation et Soins Intensifs du Patient en Insuffisance Respiratoire Aigue (GRC-RESPIRE) Sorbonne Université; Paris-Sorbonne ECMO-COVID Investigators . Extracorporeal membrane oxygenation for severe acute respiratory distress syndrome associated with COVID-19: a retrospective cohort study. Lancet Respir Med. 2020;8(11):1121-1131. doi: 10.1016/S2213-2600(20)30328-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Barbaro RP, MacLaren G, Boonstra PS, et al. ; Extracorporeal Life Support Organization . Extracorporeal membrane oxygenation support in COVID-19: an international cohort study of the Extracorporeal Life Support Organization Registry. Lancet. 2020;396(10257):1071-1078. doi: 10.1016/S0140-6736(20)32008-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Barbaro RP, MacLaren G, Boonstra PS, et al. ; Extracorporeal Life Support Organization . Extracorporeal membrane oxygenation for COVID-19: evolving outcomes from the international Extracorporeal Life Support Organization Registry. Lancet. 2021;398(10307):1230-1238. doi: 10.1016/S0140-6736(21)01960-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Davies A, Jones D, Bailey M, et al. ; Australia and New Zealand Extracorporeal Membrane Oxygenation (ANZ ECMO) Influenza Investigators . Extracorporeal membrane oxygenation for 2009 influenza A(H1N1) acute respiratory distress syndrome. JAMA. 2009;302(17):1888-1895. doi: 10.1001/jama.2009.1535 [DOI] [PubMed] [Google Scholar]
29.Pham T, Combes A, Rozé H, et al. ; REVA Research Network . Extracorporeal membrane oxygenation for pandemic influenza A(H1N1)-induced acute respiratory distress syndrome: a cohort study and propensity-matched analysis. Am J Respir Crit Care Med. 2013;187(3):276-285. doi: 10.1164/rccm.201205-0815OC [DOI] [PubMed] [Google Scholar]
30.Schmidt M, Pham T, Arcadipane A, et al. Mechanical ventilation management during extracorporeal membrane oxygenation for acute respiratory distress syndrome: an international multicenter prospective cohort. Am J Respir Crit Care Med. 2019;200(8):1002-1012. doi: 10.1164/rccm.201806-1094OC [DOI] [PubMed] [Google Scholar]
31.Riera J, Roncon-Albuquerque R Jr, Fuset MP, Alcántara S, Blanco-Schweizer P; ECMOVIBER Study Group . Increased mortality in patients with COVID-19 receiving extracorporeal respiratory support during the second wave of the pandemic. Intensive Care Med. 2021;47(12):1490-1493. doi: 10.1007/s00134-021-06517-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Bellani G, Laffey JG, Pham T, et al. ; LUNG SAFE Investigators; ESICM Trials Group . Epidemiology, patterns of care, and mortality for patients with acute respiratory distress syndrome in intensive care units in 50 countries. JAMA. 2016;315(8):788-800. doi: 10.1001/jama.2016.0291 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials