Extracorporeal membrane oxygenation in covid 19 related acute respiratory distress syndrome a euro elso international survey

Extracorporeal Membrane Oxygenation in COVID-19-related Acute

2

Respiratory Distress Syndrome – a EuroELSO international survey

3
4

Sebastian Mang1,2, Armin Kalenka3, Lars Mikael Broman4, Alexander Supady5,

5

Justyna Swol6, Guy Danziger1,2, André Becker1,2, Sabrina I. Hörsch1,7, Thilo

6

Mertke1,7, Ralf Kaiser1,2, Hendrik Bracht8, Viviane Zotzmann5, Frederik Seiler1,2,

7

Robert Bals1,2, Fabio Silvio Taccone9, Onnen Moerer10, Roberto Lorusso11, Jan

8

Bělohlávek12, Ralf M. Muellenbach13, and Philipp M. Lepper1,2 for the COVEC-Study

9

Group*

pe
er
re
v

iew
ed

1

10

Interdisciplinary COVID-19-Center, University Medical Centre, Saarland University,
Homburg/Saar, Germany
2 Department of Internal Medicine V - Pneumology, Allergology and Critical Care Medicine,
University Medical Centre, Saarland University, Homburg/Saar, Germany
3 Department of Anaesthesiology and Intensive Care Medicine, District Hospital Bergstrasse,
University Hospital Heidelberg, Heppenheim, Germany
4 ECMO Centre Karolinska, Department of Pediatric Perioperative Medicine and Intensive
Care, Karolinska University Hospital, Stockholm, Sweden
5 Department of Medicine III (Interdisciplinary Medical Intensive Care), Medical Center –
University of Freiburg, Faculty of Medicine, University of Freiburg, Germany
6 Department of Pneumology, Allergology and Sleep Medicine, and Intensive Care Medicine,
Paracelsus Medical University, General Hospital Nuremberg, Nuremberg, Germany
7 Department of Anaesthesiology, Critical Care Medicine and Pain Medicine, University
Medical Centre, Saarland University, Homburg/Saar, Germany
8 Department. of Anaesthesiology and Critical Care Medicine, University Hospital of Ulm,
Ulm, Germany
9 Department of Intensive Care, Erasme University Hospital, Université Libre de Bruxelles,
Brussels, Belgium

10 Department of Anaesthesiology, University Hospital of Göttingen, Göttingen, Germany
11Cardio-Thoracic Surgery Department - Heart & Vascular Centre - Maastricht University
Medical Centre, Maastricht, Netherlands
12 2nd Department of Internal Cardiovascular Medicine, General University Hospital, Prague,
Czech Republic
13 Department of Anaesthesiology and Critical Care Medicine, Campus Kassel of the
University of Southampton, Kassel, Germany

rin

tn

ot

1

ep

* Members of the study group are listed in the appendix

Word count: 2,544
Keywords:
COVID-19, SARS-CoV-2, ECMO, survey, CARDS

Pr

11
12
13
14

15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44

45
46
47
48

Running title:
ECMO for COVID-19 induced ARDS – EURO-ELSO-Survey

This preprint research paper has not been peer reviewed. Electronic copy available at: />

ECMO for COVID-19 induced ARDS – EURO-ELSO Survey

Correspondence to:

50
51
52
53
54
55
56
57
58
59

Philipp M. Lepper, MD
Department of Internal Medicine V – Pneumology, Allergology and Intensive Care
Medicine and ECLS Center Saar
University Hospital of Saarland
Kirrberger Str. 100

66421 Homburg
Germany
Phone:
+49-6841 16 15000
Fax:
+49-6841 16 15208
Email:


Pr

ep

rin

tn

ot

pe
er
re
v

iew
ed

49

2

This preprint research paper has not been peer reviewed. Electronic copy available at: />

Mang et al.

Background: Extracorporeal membrane oxygenation (ECMO) is a means to

61

support patients with acute respiratory failure. Initially, recommendations to treat

62

severe cases of pandemic Coronavirus Disease 2019 (COVID-19) with ECMO have

63

been restrained. In the meantime, ECMO has been shown to produce similar

64

outcomes in patients with severe COVID-19 compared to existing data on ARDS

65

mortality.

iew
ed

60

Objective: We performed an international email survey to assess how ECMO

67

providers worldwide have previously used ECMO during the treatment of critically ill

68

patients with COVID-19.

pe
er
re
v

66

69

Methods: A questionnaire with 45 questions (covering e.g. indication,

70

technical aspects, benefit and reasons for treatment discontinuation), mostly

71

multiple-choice, was distributed by email to ECMO centers. The survey was

72

approved by the European branch of the Extracorporeal Life Support Organization

73

(ELSO).

Results: 276 centers worldwide responded that they employed ECMO for

75

very severe COVID-19 cases, mostly in veno-venous configuration (87%). The most

76

common reason to establish ECMO was isolated hypoxemic respiratory failure

77

(50%), followed by a combination of hypoxemia and hypercapnia (39%). Only a

78

small fraction of patients required veno-arterial cannulation due to heart failure (3%).

79

Time on ECMO varied between less than two and more than four weeks. The main

80

reason to discontinue ECMO treatment prior to patient’s recovery was lack of clinical

81

improvement (53%), followed by major bleeding, mostly intracranially (13%). Only

82

4% of respondents reported that triage situations, lack of staff or lack of oxygenators

83

were responsible for discontinuation of ECMO support. Most ECMO physicians (66%

84

 26%) agreed that patients with COVID-19 induced ARDS (CARDS) benefitted from

85

ECMO. Overall mortality of COVID-19 patients on ECMO was estimated to be about

86

55%, scoring higher than what has previously been reported for Influenza patients on

87

ECMO (29 – 36%).

89

tn

rin

Conclusion: ECMO has been utilized successfully during the COVID-19

pandemic to stabilize CARDS patients in hypoxemic or hypercapnic lung failure. Age
and multimorbidity limited the use of ECMO. Triage situations were rarely a concern.

Pr

90

ep

88

ot

74

91

ECMO providers stated that patients with severe COVID-19 benefitted from ECMO.

3

This preprint research paper has not been peer reviewed. Electronic copy available at: />

ECMO for COVID-19 induced ARDS – EURO-ELSO Survey

An increasing use in patients with respiratory failure in a future stage of the

93

pandemic may be expected.

Pr

ep

rin

tn

ot

pe
er
re
v

iew
ed

92

4

This preprint research paper has not been peer reviewed. Electronic copy available at: />

Mang et al.

94

Introduction

iew
ed

95
96

Early in 2020, countries worldwide have been facing a surge of patients with

97

acute respiratory distress syndrome (ARDS) due to pandemic Severe Acute

98

Respiratory Syndrome Coronavirus-2 (SARS-CoV-2) disease 2019 (COVID-19).

99

Survival of those most severely affected by COVID-19-related ARDS (CARDS) might

100

depend on extracorporeal membrane oxygenation (ECMO) as bridge to recovery1.

In this global pandemic, hospitals and healthcare systems have been pushed

102

to the verge of collapse. During the first phase of the COVID-19 pandemic, the

103

number of critically ill patients requiring invasive ventilation often exceeded ventilator

104

capacities, creating a need for ICU triage2. In this scenario, it was highly unlikely that

105

ECMO would be broadly recommended to critical care providers to treat COVID-19,

106

given its high demands on personnel and resources3. In its initial guidance

107

document, ELSO considered to offer ECMO only to specific patients not responding

108

to maximal conventional therapy, including proning and neuromuscular blockade4.

109

Additionally, early reports suggested mortality rates could be higher than 90% in

110

COVID-19 patients supported with ECMO5.

pe
er
re
v

101

A recent trial reported that ECMO reduced 60-day mortality in non-COVID-19

112

related ARDS to 35% in the ECMO group versus 46% in the conventional

113

management group (relative risk 0.76, 95% CI 0.55–1.04; p=0.09)6. The study

114

highlighted that ECMO can facilitate protective ventilation of ARDS patients with

115

reduced tidal volumes, plateau and driving pressures, mostly due to effective

116

extracorporeal CO2 removal. CARDS might not differ as much from non-COVID

117

ARDS as was previously expected7. Physiological considerations make it thus

118

reasonable to think about ECMO as a bail-out strategy in critically ill patients with

119

CARDS. A recently published retrospective data suggested that mortality of patients

120

with CARDS receiving ECMO might be comparable to past ARDS cohorts8. Given

121

that COVID-19 pathophysiology is still poorly understood, little is currently known

123

tn

rin

about how to tailor ECMO treatment to meet COVID-19 specific challenges, e.g.
hypercoagulable state9, or how long ECMO should be continued when patients fail to
improve.

Pr

124

ep

122

ot

111

125

We therefore designed an online survey to elicit how ECMO providers

126

worldwide have previously employed ECMO to treat critically ill COVID-19 patients.

5

This preprint research paper has not been peer reviewed. Electronic copy available at: />

ECMO for COVID-19 induced ARDS – EURO-ELSO Survey

Our survey was approved by the European branch of the Extracorporeal Life Support

128

Organization (EuroELSO).

Pr

ep

rin

tn

ot

pe
er
re
v

iew
ed

127

6

This preprint research paper has not been peer reviewed. Electronic copy available at: />

Mang et al.

129

Methods

iew
ed

130
131

We created a questionnaire consisting of 45 questions and distributed it to

132

4,193 physicians that had published on an ECMO-related topic since the year 2000

133

in a PubMed-listed journal with an available E-Mail using a commercially available

134

internet survey platform (SurveyMonkey Inc., San Mateo, California).

135
136

The ethical committee (Ärztekammer des Saarlandes) waived the need for a
formal approval since the questionnaire did not retrieve actual patient data.

138

pe
er
re
v

137

Questionnaire

139

The questionnaire was composed of two sections: the first dealt with general

141

questions regarding contact information, details on hospital and ICU capacity as well

142

as years of ECMO experience. The second part was designed to elicit most common

143

indications for ECMO use in COVID-19, details about ECMO circuit configuration as

144

well as complications and reasons for possible treatment discontinuation. We did not

145

ask for any patient-specific data. For conformity reasons and to facilitate participation

146

in the survey, most of the questions were multiple choice with two to nine possible

147

answers per question. The last eight items requested the participant to express his

148

extent of agreement with a specific statement about ECMO therapy in the context of

149

COVID-19 on a visual analogue scale. The survey is partly available in the

150

Supplementary Material.

151

The survey questions and multiple-choice responses with their respective

152

organization in the different sections were circulated and consented between a group

153

of 23 very experienced physicians in this field. When consensus of all questions and

154

answers was reached, the survey was transferred to an online platform

155

(SurveyMonkey Inc., San Mateo, California). Automatic data retrieval and descriptive

156

statistics were retrieved through this platform. More than one answer from centers

158

tn

rin

were possible. This was allowed, as many centers comprise several departments
with physicians from different backgrounds (e.g. anesthesiology and surgery).
Results from multiple-choice questions are expressed in median, participants’

Pr

159

ep

157

ot

140

160

extent of agreement or disagreement in mean and standard deviation in percent. The

161

survey was launched on 8th, deadline for return was 20th of June, 2020. Final

7

This preprint research paper has not been peer reviewed. Electronic copy available at: />

ECMO for COVID-19 induced ARDS – EURO-ELSO Survey

analysis of results was performed using an extrapolation tool provided by

163

SurveyMonkey as well as SPSS.

iew
ed

162

164

Participants were given the opportunity to be listed as collaborators. Those

165

participants who did not supply hospital or contact information or who did not

166

complete the survey could not be included in the list of collaborators.

167

168

Results

169

General data on ECMO centers and treatment capacities

pe
er
re
v

170

276 ECMO professionals from 98 centers in 30 different countries on four

172

continents (North America, South America, Europe, Asia) responded to the survey,

173

yielding a response rate of 6.6%. Sixty-four percent of responding centers were

174

ELSO members. As it was possible to skip questions, sometimes the denominator is

175

less than 276. In this case, the number of respondents is given in brackets.

176

Centers’ median number of years with ECMO experience was 14, mainly in ECMO

177

treatment of adults or adult and pediatric patients (85%). Only 1.3% of participants

178

were exclusively specialized in neonatal ECMO. Most common numbers of patients

179

supported with ECMO per year prior to COVID-19 in the participants’ centers ranged

180

from 21 to 50, 13% of centers having even supported more the 100 patients on

181

veno-venous ECMO (VV-ECMO) per year prior to the pandemic.

ot

171

182

Numbers of patients with COVID-19 with or without ECMO

tn

183

The majority of ECMO providers (30%) stated that 2 to 6 patients with COVID-

185

19 had received ECMO in their center, 85% of all centers having supported a

186

maximum of 15 patients on ECMO by survey deadline. ECMO treatment had mostly

187

been initiated in the participants’ hospitals (63%), only a minority of patients was

188

retrieved on ECMO by mobile ECMO retrieval teams from other hospitals.

190
191

Indication for ECMO and circuit configuration
The most common reason to initiate ECMO for COVID-19 was isolated

hypoxemia (50%), followed by a combination of hypoxemia and hypercapnia (39%).

Pr

192

ep

189

rin

184

193

Isolated hypercapnia was rarely a reason to cannulate a patient (3%). Only 6%

194

stated that ECMO was started to facilitate lung-protective ventilation (n = 105). The

195

majority of ECMO cannulations (88%) were performed in veno-venous configuration.
8

This preprint research paper has not been peer reviewed. Electronic copy available at: />

Mang et al.

Eight percent of centers used veno-arterial configuration (VA-ECMO) in one or more

197

patients and 3% had to extend to a V-AV-circuit in at least one case (one venous

198

draining cannula, one arterial returning cannula, one venous returning cannula). In

199

those cases, where an arterial cannulation was required, the indications were

200

specified as biventricular failure (n = 2) and, in one case, right heart failure due to

201

pulmonary embolism (n = 1). See also Figure 1, panel A and B.

iew
ed

196

202
203

Anticoagulation management

Targeting anticoagulation therapy in patients with COVID-19 on ECMO, 60%

205

of participants (n = 110) stated that they did not change their standard

206

anticoagulation strategy compared to cases of ARDS due to other causes. Forty

207

percent used higher doses of anticoagulants than usual, monitored by higher

208

prothrombin time or higher activated clotting time. Only one of 110 ECMO providers

209

stated that they deliberately used lower doses of anticoagulants than usual for

210

ECMO in COVID-19. Antiplatelet therapy was also rarely used (1%) to prevent

211

clotting. The details of the anticoagulants or antiplatelet agents administered were

212

not part of the survey. See also Figure 1 C.

213
214

Reasons to abstain from ECMO

pe
er
re
v

204

The two main reasons to refrain from ECMO initiation were patient age (74%)

216

and comorbidities (85%, not further specified). 28% of participants stated that ECMO

217

was withdrawn due to a patient’s known or suspected wishes. Nine percent decided

218

against ECMO because it was not actively recommended for COVID-19 induced

219

ARDS by responsible scientific societies at that time. Seven percent reported that a

220

surge of COVID-19 patients and overwhelming workload made ECMO impracticable.

221

Only 5% of participants reported that they had to abstain from ECMO initiation due to

222

a shortage of oxygenators, machines or ECMO cannulas.

Pr

ep

rin

tn

ot

215

9

This preprint research paper has not been peer reviewed. Electronic copy available at: />

ECMO for COVID-19 induced ARDS – EURO-ELSO Survey

223

Duration of ECMO support
Most patients were supported with ECMO for less than two weeks. However,

225

50% of all participants stated that they had also treated patients with ECMO for more

226

than four weeks (Figure 1 D).

iew
ed

224

227
228

Reasons for ECMO discontinuation

Seventy-two percent of participants confirmed that their center would

230

withdraw ECMO if there was no perspective for a COVID-19 patient to recover. If

231

ECMO treatment was discontinued prior to recovery, futility was mostly stated as the

232

reason (n = 50 from 94 responses, 53%). ECMO-related complications were the

233

second most important reason for treatment discontinuation. Fourteen percent of

234

ECMO providers stated that they had terminated ECMO due to major bleeding (n =

235

15), mainly intracranial hemorrhage (n = 13) and, less frequently, extracranially. In

236

1% of cases, not further specified technical issues led to ECMO withdrawal. The

237

question also offered ‘lack of staff’ as a possible answer, which was not chosen.

238

However, 2% of participants (n = 2, Germany and France) stated that a triage

239

situation forced physicians to discontinue ECMO prior to the patient’s possible

240

recovery. Two percent of respondents named lack of ECMO oxygenators, ECMO

241

machines or consumables as the reason for ECMO discontinuation. See also Figure

242

1 E.

ot

pe

er
re
v

229

243

Estimation of patients’ outcome

tn

244

When asked to estimate the percentage of patients who died while on ECMO

246

due to COVID-19, average mortality was estimated to be 55%, meaning that 45% of

247

patients had survived on ECMO at least until the end of the survey.

249
250
251
252

The last eight questions were designed to investigate a participant’s opinion

on certain statements about ECMO and COVID-19, measured in percentage of
agreement. Participants agreed to 63%  24% on average that patients were longer
on ECMO due to COVID-19 compared to other causes of ARDS. The claim that

Pr

253

ECMO providers’ opinions on COVID-19 and ECMO

ep

248

rin

245

254

CARDS patients on ECMO required more sweep gas flow than what the individual

255

ECMO physician was used to was accepted by 58%  25%. The statement that
10

This preprint research paper has not been peer reviewed. Electronic copy available at: />

Mang et al.

oxygenator change was more frequently required in CARDS patients on ECMO had

257

an acceptance rate of 54% on average  29%. The claim that disturbed coagulation

258

in COVID-19 patients would make ECMO impossible was mostly rejected (24%

259

agreement  26%). The assumption that ECMO offers patients with COVID-19

260

induced ARDS a chance to recover found relatively strong acceptance (agreement

261

extent of 76%  25%), with 66%  26% agreement with the claim that CARDS

262

patients benefitted from ECMO therapy. Box plots are displayed in Figure 2.

263

pe
er
re
v

For results not outlined here, see Supplementary Material.

Pr

ep

rin

tn

ot

264

iew
ed

256

11

This preprint research paper has not been peer reviewed. Electronic copy available at: />

ECMO for COVID-19 induced ARDS – EURO-ELSO Survey

Discussion

265

iew
ed

266

Survey results prove that critical care providers worldwide have repeatedly

268

successfully used extracorporeal membrane oxygenation to support patients in

269

COVID-19 related respiratory failure despite official recommendations were limited to

270

carefully singled-out patients. At the time the survey was launched, data regarding

271

the outcome of COVID-19 patients treated with ECMO was limited to case reports

272

and small case series. Recent retrospective cohort studies on COVID-19 patients

273

receiving ECMO have produced encouraging results8 but were unpublished at that

274

time. Critical care providers hence had to rely on what was already known about

275

ARDS from different causes. Both clinical experience and pathophysiology of

276

CARDS made ECMO appealing to treat COVID-19, even when data on mortality or

277

complication rate was missing.

pe
er
re
v

267

Several groups have reported evidence that COVID-19 might be associated

279

with a hypercoagulable state, resulting in an increased risk of thromboembolic

280

complications11. Exposing hypercoagulable blood to the artificial surface of an ECMO

281

circuit could therefore come with a considerable risk of thromboembolism.

282

Interestingly, when asked about ECMO-related complications in COVID-19, 13% of

283

ECMO providers stated that ECMO had to be terminated due to major bleeding, but

284

only one case of relevant pulmonary embolism on VV-ECMO was reported. It should

285

be noted though that minor thrombosis or thromboembolism could have remained

286

undetected during ECMO therapy, hence a reliable incidence of all thrombotic events

287

cannot be provided or was underestimated by the ECMO providers. Nevertheless,

288

given that 50% of all participants had treated patients on ECMO for more than four

289

weeks, the overall occurrence of lethal thromboembolism on ECMO in COVID-19

290

was surprisingly low. Only a minority of ECMO physicians said that they increased

291

the dosage of anticoagulants to prevent clotting on ECMO. There was also no

292

general agreement that oxygenator change was required more frequently compared

294

tn

rin

to patients with ARDS from other causes. Although increased rate of oxygenator
pump

head

thrombosis

in

COVID-19

has

been

reported12,

suspected

hypercoagulation in COVID-19 did not seem to translate to a higher incidence of life-

Pr

295

ep

293

ot

278

296

threatening pulmonary or systemic thromboembolism in patients with COVID-19 on

297

ECMO. Prospective studies should address how much anticoagulation for ECMO in

12

This preprint research paper has not been peer reviewed. Electronic copy available at: />

Mang et al.

COVID-19 is needed so that both the risk of thrombosis and hemorrhage can be

299

minimized.

iew
ed

298

Although predominantly affecting the lungs, multiple organ involvement of

301

COVID-19 has been reported. Cardiac complications in COVID-19 have recently

302

gathered broad attention. Acute myocardial injury in the absence of macroscopic

303

coronary artery disease is suspected to occur in up to 10% of patients especially in

304

the critically ill13,14,15,16. While 11% of our participants reported that they employed

305

also veno-arterial ECMO for COVID-19, only 3% of participants specified that arterial

306

cannulation was required because of acute heart or circulatory failure, in one case

307

caused by fulminant pulmonary embolism. So far, our survey results do not support

308

the hypothesis that severe cardiac involvement in COVID-19 translated to an

309

increased need for cardiac support (e.g. with VA-ECMO) in patients with CARDS.

pe
er
re
v

300

A recently published study investigating outcomes of all adult patients with

311

CARDS treated with ECMO using a EuroELSO registry estimated overall 90-day

312

mortality to be less than 40%8. When asked about the mortality of COVID-19

313

patients on ECMO, the mean estimation in our survey was 55%, meaning that

314

roughly 45% could be weaned off extracorporeal support at survey deadline. If this

315

data translates to real-life mortality, survival of COVID-19 patients on ECMO could

316

be comparable to non-COVID-19 induced ARDS. At the time of report, no

317

prospective data on the outcome nor complication rate of ECMO therapy in COVID-

318

19 is available. For the time being, our survey suggests that ECMO for COVID-19 is

319

practicable, effective and does not lead to higher complication rates in COVID-19 if it

320

is utilized in experienced ECMO centers. Most participants agreed that their patients

321

benefitted from ECMO therapy and stated that, in a hypothetical scenario of a future

322

respiratory pandemic, they would use it more readily to treat acute respiratory failure.

tn

rin

323

ot

310

Our survey has several limitations. Firstly, while it reflects opinions of a fairly

325

large cohort of physicians providing ECMO services, we nevertheless did not ask for

326
327

patient-specific data outcome. Hence, we cannot provide statistical evidence on
ECMO-related end points, e.g. overall survival, 28-day mortality or discharge from
ICU. Secondly, the survey was addressed mainly to tertiary care centers who have

Pr

328

ep

324

329

repeatedly published on ECMO, leading to a skewness of representation, possibly to

330

the disadvantage of primary and secondary care hospitals with less ECMO

331

experience.
13

This preprint research paper has not been peer reviewed. Electronic copy available at: />

ECMO for COVID-19 induced ARDS – EURO-ELSO Survey

332

Critical care providers need to know not only whether to use ECMO, but also

334

how to manage ECMO therapy in a disease as poorly understood as COVID-19.

335

Currently, this knowledge is expanding8. Across Europe, more than 2,100 patients in

336

180 centers were already treated on ECMO for respiratory and/or circulatory failure

337

due to COVID-19 ( last accessed

338

on November 23rd). The data from this survey (theoretical considerations early in the

339

pandemia) and the EuroELSO (real world data with weekly updates) data are in line,

340

suggesting that ECMO is a useful adjunctive tool in COVID-19-related respiratory

341

failure.

Pr

ep

rin

tn

ot

pe
er
re
v

iew
ed

333

14

This preprint research paper has not been peer reviewed. Electronic copy available at: />

Mang et al.

342

Declarations

344

iew
ed

343

Conflict of interest disclosure

345

Robert Bals declares funding from AstraZeneca, Boehringer Ingelheim,

347

GlaxoSmithKline, Grifols, Novartis, CLS Behring, the German Federal Ministry of

348

Education and Research (BMBF) Competence Network Asthma and COPD

349

(ASCONET), Sander-Stiftung, Schwiete-Stiftung, Krebshilfe and Mukoviszidose eV.

350

Conflicts that the editors consider relevant to the content of the manuscript have

351

been disclosed. All other authors: No potential conflicts of interest

352
353

Financial funding

354
355
356

COVID-19 research was funded by the Federal state of Saarland, Saarland
University and Dr. Rolf M. Schwiete Foundation.

357
358

pe
er
re
v

346

Acknowledgements

359

The authors are grateful to all colleagues worldwide for their generous

361

participation in our survey. We thank the EuroELSO steering committee for

362

supporting the survey.

364

tn

363

Consent to be listed as collaborator

365

368
369
370
371

were included as collaborators.
Authorship Contribution
S.M., A.K., A.B., G.D., S.H., T.M. and P.L. drafted the survey. L.M.B, J.S.,

H.B., F.S.T., O.M., R.L., J.B. and R.M. revised the survey for important content. P.L.

Pr

372

rin

367

All colleagues who participated in the survey and agreed to be named publicly

ep

366

ot

360

373

launched the survey and oversaw collection of data. R.K. provided IT-tools for survey

374

preparation. S.M. drafted manuscript. A.K., L.M.B., A.S., J.S., G.D., S.H., F.S., R.B.,

15

This preprint research paper has not been peer reviewed. Electronic copy available at: />

ECMO for COVID-19 induced ARDS – EURO-ELSO Survey

F.T., O.M., R.L, J.B., R.M. and P.L. revised the manuscript for important intellectual

376

content. All authors have seen and approved the final version of the manuscript.

Pr

ep

rin

tn

ot

pe
er
re
v

iew
ed

375

16

This preprint research paper has not been peer reviewed. Electronic copy available at: />

Mang et al.

*The COVID-19 ECMO (COVEC) study group

378

iew
ed

377

379

Nicholas Barrett, Guy’s and St Thomas’ NHS Foundation Trust, London, UK;

380

Daniel Duerschmied, University Hospital of Freiburg, Freiburg, Germany;

381

Eddy Fan, Toronto General Hospital, Toronto, Canada;

382

Falk Fichtner, Universitätsklinikum Leipzig, Leipzig, Germany;

383

Hendrik Haake, Krankenhaus Maria Hilf, Mönchengladbach, Germany;

384

Frank Langer, University Medical Centre, Saarland University, Homburg/Saar,
Germany;

pe
er
re
v

385
386

Haitham Mutlak, Sana Klinikum Offenbach, Offenbach, Germany;

387

Markus Kredel, University Hospital of Würzburg, Würzburg, Germany;

388

Thomas Müller, University Hospital of Regensburg, Regensburg, Germany;

389

Alessandro Protti, Humanitas Research and Clinical Centre-IRCCS, Milan, Italy

390

Alexander Raddatz, University Medical Centre, Saarland University, Homburg/Saar,

391

Germany

392

Tobias Spangenberg, Marienkrankenhaus Hamburg, Hamburg, Germany;

393

Dawid Staudacher, University Hospital of Freiburg, Freiburg, Germany;

394

Holger Wehrfritz, University Medical Centre, Saarland University, Homburg/Saar,

395

Germany

Tobias Wengenmayer, University Hospital of Freiburg, Freiburg, Germany;

397

Arne Westheider, University Medical Centre, Saarland University, Homburg/Saar,
Germany;

tn

398

ot

396

Pr

ep

rin

399

17

This preprint research paper has not been peer reviewed. Electronic copy available at: />

ECMO for COVID-19 induced ARDS – EURO-ELSO Survey

Survey Collaborators:

401
402

France

403

iew
ed

400

Simon Dang Van, University Hospital of Angers, France;

405

Cedric Daubin, Hospital Center University of Caen Normandie, France;

406

Philippe Gaudard, CHU Montpellier, University Montpellier, France;

407

Thomas Godet, University Hospital of Clermont-Ferrand, France;

408

Pierre-Grégoire Guinot, Département d’Anesthésie-Réanimation, Centre Hospitalier

409

pe
er
re
v

404

Universitaire de Dijon, France;

410

Lọc Le Guennec, Hơpital Universitaire Pitié Salpêtrière, Paris, France;

411

Bruno Megarbane, Lariboisière Hospital, University of Paris, France;

412

Alain Mercat, Centre Hospitalier Universitaire d'Angers, France;

413

Romain Sonneville, Bichat-Claude Bernard University Hospital, France;

414

Elie Zogheib, Medical Intensive care - CHU Amiens, France;

415

Tai Pham, Medecine Intensive-Réanimation, Hôpital Bicêtre, APHP, Le Kremlin

416

417
418

Bicêtre, France

Hadrien Winiszewski, Centre Hospitalier Rộgional Universitaire de Besanỗon,
France;

420

ot

419

Austria

tn

421

Peter Schellongowski, Dept. of Medicine I, Medical University of Vienna, Austria;

423

Thomas Staudinger, Dept. of Medicine I, Medical University of Vienna, Austria;

424

Dominik Wiedemann, Department of Cardiac Surgery, Medical University of Vienna,

425

rin

422

Austria;

Corinna Velik-Salchner, Medical University Innsbruck, Austria;

427

Michael Joannidis, Medical University Innsbruck, Dept. Internal Medicine, Division of

428
429

Intensive Care and Emergency Medicine, Austria;

Pr

430

ep

426

18

This preprint research paper has not been peer reviewed. Electronic copy available at: />

Mang et al.

Germany

432

iew
ed

431

433

Marc Bodenstein, University Medical Center Mainz, Germany;

434

Heinrich Volker Groesdonk, Dept. of Intensive Care Medicine and Intermediate Care;
Helios Clinic Erfurt, Germany;

435

Stefan Guth, Kerckhoff-Klinik, Bad Nauheim, Germany;

437

Matthias Hecker, University Hospital Giessen, Medical Clinic II, Germany;

438

Faeq Husain-Syed, University Hospital Giessen and Marburg, Giessen, Germany;

439

Christian Jung, Heinrich-Heine-University Düsseldorf, Germany;

440

L. Christian Napp, Hannover Medical School, Germany;

441

Ruslan Natanov, Hannover Medical School, Germany;

442

Georg Trummer, Universitaets-Herzzentrum Freiburg Bad Krotzingen, Germany;

443

Sascha Treskatsch, Charité – University Hospital Berlin, Department of
Anesthesiology and Intensive Care Medicine, Berlin, Germany;

444
445

pe
er
re

v

436

Henryk Welp, University Hospital Münster, Germany;

446
447

Italy

448

Leonello Avalli, Ospedale San Gerardo, Monza, Italy;

450

Lorenzo Ball, Università degli studi di Genova, Genova, Italy;

451

Mirko Belliato, Foundation IRCCS Policlinico San Matteo, Pavia, Italy;

452

Manuela Bonizzoli, Intensive Care Unit, Careggi Hospital, Florence, Italy;

453

Emma Borrelli, University of Siena, Siena, Italy;

454

Giacomo Cavallaro, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico,

tn

Milano, Italy;

rin

455

ot

449

Andrea Franci, Azienda Odpedaliero Universitaria Careggi, Florence, Italy;

457

Silvia Gramaticopolo, Ospedale San Bortolo Vicenza, Italy;

458

Mauro Panigada, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico

459
460

Milano, Italy;

Luigi Tritapepe, San Camillo-Forlanini Hospital, Rome, Italy;

Pr

461

ep

456

462

Saudi Arabia

463
464

Salman Abdulaziz, King Saud Medical City, Saudi Arabia;
19

This preprint research paper has not been peer reviewed. Electronic copy available at: />

ECMO for COVID-19 induced ARDS – EURO-ELSO Survey

Canada

466

iew
ed

465

467

David Bracco, McGill University Health Center, Montreal, Canada;

468

Yiorgos Alexandros, Cavayas Hôpital du Sacré-Coeur de Montréal, Montréal,

469

Canada;
Ari Joffe, Stollery Children's Hospital, Alberta, Canada;

471

A. Dave Nagpal, London Health Sciences Centre, London, Canada;

472

Ying Sia, University Institute of Cardiology and Respirology of Quebec, Quebec,

473

Canada;

474
475

United Kingdom

476

pe
er
re
v

470

477

Georg Auzinger, King's College Hospital London, United Kingdom;

478

Vasileios Zochios, University Hospitals of Leicester NHS Trust, Glenfield Hospital

479

ECMO Unit, Lecester, United Kingdom;

480
481

United States of America

482

Alejandro Garcia, Johns Hopkins University, Baltimore, Maryland, United States;

484

Katja Gist, University of Colorado, Boulder, Colorado, United States;

485

Dana Lustbader, ProHEALTH an OPTUM Company, New York, New York, United
States;

tn

486

ot

483

Demetris Yannopoulos, University of Minnesota, Minneapolis, Minnesota, United

488

States;

489

R. Scott Stephens, Johns Hopkins University, Baltimore, Maryland, United States;

490

Joseph Tonna, University of Utah, Salt Lake City, Utah, United States;

491

Linda Paxton, St. Mary's Hospital, Richmond, Virginia, United States;

492

Hitoshi Hirose, Thomas Jefferson University Hospital, Philadelphia, Pennsylvania,

494

United States;

Bo Kim, Johns Hopkins University, Baltimore, Maryland, United States;

Pr

495

ep

493

rin

487

496

Sweden

497
498

Magnus Dalén, Karolinska University Hospital and Karolinska Institutet, Sweden;
20

This preprint research paper has not been peer reviewed. Electronic copy available at: />

Mang et al.

499

Czech Republik

501

Martin Balik, Complex Cardiac Center, General University Hospital, Prague, Czech
Republic;

502
503

iew
ed

500

David Janak, Charles University of Prague, Department of Cardiovascular Surgery,
Prague, Czech Republic;

504
505
506

Chile

pe
er
re
v

507
508

Luis Castillo, Hospital Barros Luco, Santiago, Chile;

509

Alejandro Bruhn; Pontificia Universidad Catolica de Chile, Santiago, Chile;

510
511

Colombia

512
513

Jorge Luis Alvarado Socarras, Fundacion Cardiovascular de Colombia,
Bucaramanga, Santander, Colombia;

514
515
516

South Korea

517

Taeyun Kim, Seongnam Citizens Medical Center, Seongnam, South Korea;

519

Hyoung Soo Kim, Hallym University Sacred Heart Hospital, Seoul, South Korea;

520

Joung Hun Byun, Gyeongsang National University College of Medicine, Changwon,

tn

ot

518

South Korea;

521

523
524

Brazil

rin

522

Guilherme Mainardi, Hospital São Camilo, Brazil;

526

Pedro Mendes, Hospital das Clinicas de São Paulo HC-FMUSP, São Paulo, Brazil;

527
528

Switzerland

Pr

529

ep

525

530

Raphaël Giraud, Geneva University Hospitals, Geneva, Switzerland;

531
532

21

This preprint research paper has not been peer reviewed. Electronic copy available at: />

ECMO for COVID-19 induced ARDS – EURO-ELSO Survey

533

Portugal

iew
ed

534
535

Philip Fortuna, Centro Hospitalar Universitário de Lisboa Central, Lisbon, Portugal;

536

Japan

537
538

Tatsuma Fukuda, University of the Ryukyus, Nakagami District, Okinawa, Japan;

539
540

The Netherlands

542

pe
er
re
v

541

Jacinta Maas, Leiden University Medical Center, Leiden, The Netherlands;

543
544

Poland

545
546

Dariusz Maciejewski, Regional Teaching Hospital, Bielsko-Biała, Poland;

547
548

India

549
550

Deblal Pandit, Medica Superspecialty Hospital Kolkata, Calcutta, India;

551

Israel

ot

552
553

Yosv Psz, General Intensive Care Unit, Tel Aviv Sourasky Medical Center, Tel Aviv,
Israel;

555
556

Slovenia

rin

557
558
559

Peter Radsel, University Medical Center Ljubljana, Slovenia;

562

China

Gangfeng Yan, Children's Hospital of Fudan University, China;

Pr

563

ep

560
561

tn

554

22

This preprint research paper has not been peer reviewed. Electronic copy available at: />

Mang et al.

References

iew
ed

564

1. Savarimuthu, Sugeevan; BinSaeid, Jalal; Harky, Amer (2020) The role of
ECMO in COVID-19: Can it provide rescue therapy in those who are critically
ill? Journal of cardiac surgery. 35:6;1298–1301.

569
570
571

2. Phua, Jason; Weng, Li; Ling, Lowell, et al (2020) Intensive care management
of coronavirus disease 2019 (COVID-19): challenges and recommendations.
Lancet Respir Med. 2020 May; 8(5): 506–517

572
573
574
575

3. MacLaren, Graeme; Fisher, Dale; Brodie, Daniel et al (2020) Preparing for the
Most Critically Ill Patients With COVID-19: The Potential Role of
Extracorporeal Membrane Oxygenation. JAMA. DOI:

10.1001/jama.2020.2342.

576
577
578

4. Bartlett, Robert H.; Ogino, Mark T.; Brodie, Daniel (2020) Initial ELSO
Guidance Document: ECMO for COVID-19 Patients with Severe
Cardiopulmonary Failure. ASAIO Journal. 66(5):472-474

579
580
581
582
583

5. Henry, Brandon Michael; Lippi, Giuseppe (2020) Poor survival with
extracorporeal membrane oxygenation in acute respiratory distress syndrome
(ARDS) due to coronavirus disease 2019 (COVID-19): Pooled analysis of
early reports. Journal of critical care, 58:27–28. DOI:
10.1016/j.jcrc.2020.03.011.

584
585
586
587

6. Combes, Alain; Hajage, David; Capellier, Gilles et al. (2018) Extracorporeal
Membrane Oxygenation for Severe Acute Respiratory Distress Syndrome.
The New England journal of medicine, 378:21;1965–1975. DOI:

10.1056/NEJMoa1800385.

588
589
590
591

7. Ferrando, Carlos; Suarez-Sipmann, Fernando; Mellado-Artigas, et al. (2020)
Clinical features, ventilatory management, and outcome of ARDS caused by
COVID-19 are similar to other causes of ARDS. Intensive care medicine. DOI:
10.1007/s00134-020-06192-2

592
593
594
595

8. Barbaro, Ryan P.; MacLaren, Graeme; Boonstra, Philip S. et al. (2020)
Extracorporeal membrane oxygenation support in COVID-19: an international
cohort study of the Extracorporeal Life Support Organization registry. The
Lancet. DOI: 10.1016/S0140-6736(20)32008-0.

596
597
598

9. Iba, Toshiaki; Levy, Jerrold H.; Levi, Marcel; Connors, Jean Marie; Thachil,
Jecko (2020) Coagulopathy of Coronavirus Disease 2019. Critical care
medicine. DOI: 10.1097/CCM.0000000000004458.

ot

tn

rin

ep

10. Amato, Marcelo B P; O Meade, Maureen; Slutsky, Arthur S et al (2015)
Driving pressure and survival in the acute respiratory distress syndrome. The
New England Journal of Medicine, 19;372(8):747-55

Pr

599
600
601

pe
er
re
v

565
566
567
568

602
603

604
605

11. Ranucci, Marco; Ballotta, Andrea; Di Dedda, Umberto et al. (2020) The
procoagulant pattern of patients with COVID-19 acute respiratory distress
syndrome. Journal of thrombosis and haemostasis,18:7;1747–1751. DOI:
10.1111/jth.14854
23

This preprint research paper has not been peer reviewed. Electronic copy available at: />

ECMO for COVID-19 induced ARDS – EURO-ELSO Survey

12. Bemtgen, Xavier; Zotzmann, Viviane; Benk, Christoph; et al. (2020)
Thrombotic circuit complications during venovenous extracorporeal
membrane oxygenation in COVID-19. Journal of thrombosis and
thrombolysis. DOI: 10.1007/s11239-020-02217-1.

610
611
612

13. Wang, Dawei; Hu, Bo; Hu, Chang et al (2020) Clinical Characteristics of 138
Hospitalized Patients With 2019 Novel Coronavirus-Infected Pneumonia in
Wuhan, China. JAMA. DOI: 10.1001/jama.2020.1585

613
614
615
616

14. Chen, C.; Yan, J. T.; Zhou, N.; Zhao, J. P et al (2020) Analysis of myocardial
injury in patients with COVID-19 and association between concomitant
cardiovascular diseases and severity of COVID-19. Zhonghua xin xue guan
bing za zhi, 48:0, E008. DOI: 10.3760/cma.j.cn112148-20200225-00123.

617
618
619
620

15. Lippi, Giuseppe; Lavie, Carl J.; Sanchis-Gomar, Fabian (2020) Cardiac
troponin I in patients with coronavirus disease 2019 (COVID-19): Evidence
from a meta-analysis. Progress in cardiovascular diseases. DOI:
10.1016/j.pcad.2020.03.001.

621
622
623
624

16. Esposito, Antonio; Palmisano, Anna; Natale, Luigi; Ligabue, Guido; Peretto,
Giovanni; Lovato, Luigi et al. (2020) Cardiac Magnetic Resonance
Characterization of Myocarditis-Like Acute Cardiac Syndrome in COVID-19.
In : JACC. Cardiovascular imaging. DOI: 10.1016/j.jcmg.2020.06.003.

Pr

ep

rin

tn

ot

pe
er
re
v

iew
ed

606
607
608
609

24

This preprint research paper has not been peer reviewed. Electronic copy available at: />

Mang et al.

628
629

Sebastian Mang1,2, Armin Kalenka3, Lars Mikael Broman4, Alexander Supady5,

630

Justyna Swol6, Guy Danziger1,2, André Becker1,2, Sabrina I. Hörsch1,7, Thilo

631

Mertke1,7, Ralf Kaiser1,2, Hendrik Bracht8, Viviane Zotzmann5, Frederik Seiler1,2,

632

Robert Bals1,2, Fabio Silvio Taccone9, Onnen Moerer10, Roberto Lorusso11, Jan

633

Bělohlávek12, Ralf M. Muellenbach13, and Philipp M. Lepper1,2 for the COVEC-Study

634

Group*

pe
er
re
v

iew
ed

627

Extracorporeal Membrane Oxygenation in COVID-19-related Acute
Respiratory Distress Syndrome – a EuroELSO international survey

626

635
636

Supplementary Data

637
638

Results – Details on participating ECMO centers

639

Countries with most replies were Germany (32), the United States of America

641

(32), Italy (16), France (15) and Canada (11) (Supplementary Figure 1).

642

Participants were mostly ECMO providers at larger hospitals and tertiary care

643

centers, the majority having a capacity of over 1,000 beds in total, 60% having at

644

least 21 ICU beds on their ward. ICUs employing ECMO were mostly

645

interdisciplinary (66%) and, to a lesser extent, surgical (18%) or non-surgical (16%).

646

ECMO wards were mainly run by critical care physicians (59%), followed by Cardiac

647

surgeons (19%) and Anesthesiologists (9%). The vast majority of ECMO providers

648

(70%) stated that their center had previously increased ECMO capacity in the face of

649

COVID-19, in median by 25 – 50%. 19% managed to double their ECMO capacity

650

(Supplementary Table 1 and 2).

653

654
655

tn

Results – Numbers of critically-ill patients with or without ECMO
By survey deadline, most centers had treated a median of 21-40 patients with
COVID-19 (both intensive and normal care units), 73 of 149 had treated more than
100 patients. Those centers experiencing a high emergence of COVID-19 patients

Pr

656

rin

652

ep

651

ot

640

657

were mostly localized in countries which were most severely affected by the

658

pandemic at that time (France 18%, Italy 17%, USA 14% and Germany 11%). 41%
25

This preprint research paper has not been peer reviewed. Electronic copy available at: />