152 AJTCCM VOL. 29 NO. 4 2023
ORIGINAL RESEARCH: ARTICLES
Background. Extracorporeal membrane oxygenation (ECMO) is an advanced, resource-intensive technology used in a limited capacity in
South Africa (SA). Minimal data on the use of ECMO in SA are available.
Objectives. To describe the indications, early outcome and comorbidities of patients placed on ECMO in the highest-volume ECMO
centre in SA.
Methods. We performed a single-centre retrospective review of all adult patients supported with any form of ECMO from August 2016 to
December 2018. Operative and clinical records were reviewed. e primary objective of this study was to review the outcome of patients
placed on ECMO in the form of survival to hospital discharge. e secondary objectives were to identify population-specic comorbidities
and indications for ECMO that could be associated with non-survival and to compare outcome with known risk scores in the form of the
Respiratory ECMO Survival Prediction (RESP) and Survival Aer Venoarterial ECMO (SAVE) scores.
Results. One hundred and seven patients were identied. e primary indication for ECMO was respiratory support in 78 patients and
cardiac support in 29 patients. Forty-seven patients were discharged from hospital, with a 44.0% overall survival rate. Gender (p=0.039),
age (p=0.019) and hypertension (p=0.022) were associated with death in univariate logistic regression analysis. However, aer adjusting
for potential confounding in multivariate logistic regression analysis, the association was no longer signicant. In the all respiratory
support group, patients in risk class IV had better than predicted survival according to the RESP score, while risk classes I, II and III
had worse than predicted survival. In the circulatory support group, all risk classes had worse than predicted survival according to the
SAVE score.
Conclusion. We report ECMO outcomes in SA for the rst time. We identied very high mortality rates for patients transferred on
ECMO from other facilities and for patients converted from venovenous ECMO to venoarterial ECMO. Although our outcomes were
comparable in some of the risk classes, further external validation of the SAVE and RESP scores will be needed to compare our outcomes
with these scores.
Keywords. Extracorporeal membrane oxygenation, ECMO, venovenous ECMO, venoarterial ECMO, ECMO indications, ECMO outcomes,
lung transplant, circulatory support.
Afr J Thoracic Crit Care Med 2023;29(4):e211. https://doi.org/10.7196/AJTCCM.2023.v29i4.211
Extracorporeal membrane oxygenation (ECMO), also known as
extracorporeal life support (ECLS), is a mechanical support system
used in patients with inadequate oxygenation, ventilation or perfusion.
ECMO may be used to provide physiological support until recovery of
the failing organ or as a bridge to denitive treatment.
ere are two major types of ECMO: venovenous (VV-ECMO)
and venoarterial (VA-ECMO). In VV-ECMO, only venous access is
needed and mainly respiratory support is provided, while both venous
and arterial access are required in VA-ECMO in order to facilitate
both gas exchange and mechanical cardiac output support.[1]
ECMO technology has advanced considerably in recent times, with
circuits becoming smaller and safer. With these advances, ECMO use
worldwide has seen a marked increase.[2,3] While ECMO is commonly
used in most developed countries, its use in the developing world is
Extracorporeal membrane oxygenation in South Africa:
Experience from a single centre in the private sector
N L F van Zijl,1 MB ChB, MMed (or Surg), FC Cardio (SA); J T Janson,¹ MB ChB, MMed (or Surg), FC Cardio (SA), PhD;
M Sussman,² MB ChB, FC Cardio (SA); A Geldenhuys,² MB ChB, FC Cardio (SA)
1 Division of Cardiothoracic Surgery, Department of Surgery, Faculty of Medicine and Health Sciences, Stellenbosch University, Cape Town, South Africa
2 Netcare Milpark Hospital, Johannesburg, South Africa
Corresponding author: N L F van Zijl (nicholasvanzijl@gmail.com)
Study synopsis
What the study adds. We report on extracorporeal membrane oxygenation (ECMO) outcomes in South Africa for the rst time.
Weidentied a high mortality rate in patients transferred on ECMO from other facilities, and in patients converted from venovenous
ECMO to venoarterial ECMO.
Implications of the ndings. Transferred patients had a high mortality rate. e reason for this should be further investigated and may
highlight the need for possible protocols to assist with appropriate timing of patient transfers and possible earlier intervention or transfer.
AJTCCM VOL. 29 NO. 4 2023 153
ORIGINAL RESEARCH: ARTICLES
extremely limited owing to its high cost and a scarcity of dedicated
ECMO intensive care units (ICUs) with the highly specialised and
skilled sta needed to care for the patients.
South Africa (SA) currently only has three Extracorporeal Life
Support Organization (ELSO)-registered centres that provide ECMO
for adults. Only one of the centres provides care to the public sector.
With advances in ECMO technology, the indications for ECMO are
also expanding. Some of the indications are hypoxaemic respiratory
failure with a mortality risk >50%, hypercapnic respiratory failure
with a pH <7.2, bridge to transplantation, refractory cardiogenic
shock, massive pulmonary embolism, and failure to wean from
cardiopulmonary bypass aer cardiac surgery.[3-6]
Although ECMO is certainly lifesaving in some patients, it is an
invasive treatment option with a signicant potential for complications.
e need to predict which patients would benet from this resource-
intensive technology has given rise to multiple risk prediction scores
in the ELSO literature. e Respiratory ECMO Survival Prediction
(RESP) and Survival Aer Venoarterial ECMO (SAVE) scores are two
of the most frequently used.[7,8] Risk models use multiple variables, are
sometimes diagnosis specic, and need to be externally validated to
assess their accuracy in dierent cohorts.[9,10]
We present the experience of an ELSO-registered ECMO centre
in SA.
Methods
A retrospective review was conducted of all patients who underwent
any form of ECMO at Netcare Milpark Hospital in Johannesburg
from August 2016 to December 2018. e study was approved by
the Research Operations Committee of Stellenbosch University
(ref.no. UNIV-2019-0004). Informed consent for individual patients
was waived. e primary objective of the study was to review our
outcomes with ECMO in the form of survival to hospital discharge.
The secondary objectives were to identify population-specific
comorbidities and indications for ECMO that could be related to
mortality and to compare our outcomes with known risk scores in
the form of the RESP and SAVE scores.
All adult patients (≥18 years of age) who underwent ECMO,
had hospital les available for review and were discharged or died
before 31 December 2018 were included. e primary outcome was
dened as death or discharge from hospital before 31 December 2018.
e duration of ECMO treatment was calculated from the time of
ECMO cannulation until decannulation or death. ese data were
also compared with data captured by the Milpark Cardiothoracic
Centre. Other data collected included age, sex, days in the ICU, days
on ventilator prior to ECMO placement, type of ECMO (VV or VA),
indication for ECMO (respiratory failure or cardiogenic shock), injury
type, surgery prior to ECMO placement, ECMO circuit changed or
replaced, site of cannulation, whether tracheostomy was performed,
whether patients were transferred on ECMO from another facility, and
risk assessment in the form of a risk score.
To analyse our primary outcomes, the patients with respiratory
failure were divided into an all respiratory support group and a
respiratory support without lung transplant group. Patients with
cardiogenic shock were divided into a circulatory support group and
a circulatory support without cardiac transplant group.
To assess our secondary outcomes, patients were grouped into three
categories according to the type of ECMO support received, namely
VV-ECMO, VA-ECMO, or VV+VA-ECMO if changed from one type
to another.
In patients with cardiogenic shock, the SAVE score was used, and for
primarily adult respiratory failure, the RESP score was calculated.[7,8]
Statistical analysis
Descriptive data were presented as either means with standard
deviations (SDs) for normal distribution or medians with
interquartile ranges (IQRs) for skewed distribution. Data were
analysed using Pearsons χ2 test and Fishers exact test. We tested
associations between clinical characteristics using logistic regression
analysis. We controlled for potential confounding using multivariate
logistic regression analysis. We report the odds ratio as measures of
association with the corresponding condence intervals. Statistical
signicance was set at p<0.05.
Results
There were 107 patients in the study. Forty-three patients were
transferred from other hospitals prior to ECMO placement at Milpark
Hospital, and 6 patients were placed on ECMO at the referring hospital
and retrieved on ECMO. The mean (SD) age of patients placed
on ECMO was 47 (13.3) years. e mean age of ECMO survivors
was lower than that of non-survivors (44 (13.8) v. 50 (12.4) years,
respectively; p=0.019). e study population comprised slightly more
male than female patients (53.3% male; n=57).
In the total group, the primary indication for ECMO placement
was respiratory support in 78 patients and cardiac support in 29. VV-
ECMO was initiated in 58 patients, 40 patients received VA-ECMO,
and 9 patients received VV+VA-ECMO. Forty-seven patients were
discharged from hospital, with a 44.0% overall survival rate. e
patients on VA-ECMO and VV-ECMO had similar survival rates of
45.8% and 46.5%, respectively, while the VV+VA-ECMO group had a
22.2% survival rate. Four patients were changed from VV-ECMO to
VA-ECMO, all of whom died (p=0.13).
A total of 60 patients died in hospital, 45 while still on ECMO. e
median (IQR) duration of ECMO treatment was 8 (4 - 18) days, the
median time in hospital was 33 (14 - 60) days, and the median time on
ECMO aer lung transplant was 4 (0 - 11) days. e longest duration
of ECMO recorded was 83 days; unfortunately, this patient died. e
longest duration of ECMO aer which the patient was successfully
discharged was 48 days, and the longest hospital stay with successful
discharge was 197 days.
Sixty-three patients had surgery performed prior to ECMO placement.
Bilateral sequential lung transplant and single-lung transplant were the
most common procedures performed prior to ECMO placement, and
accounted for 22 of the participants. Afurther breakdown of procedures
performed is presented in Table1. Indications for lung transplant were
interstitial lung disease (n=7), cystic brosis (n=3), chronic obstructive
pulmonary disease (COPD) (n=4), pulmonary arterial hypertension
(n=3) and miscellaneous (n=5).
VV-ECMO was mostly established with a dual-lumen cannula. e
most frequent site of cannulation in the VV-ECMO group was the
right jugular vein (81.0% of cases), followed by the le jugular vein
(13.7%) and the femoral vein (5.2%). Tracheostomy was performed
in 42 of the patients.
154 AJTCCM VOL. 29 NO. 4 2023
ORIGINAL RESEARCH: ARTICLES
A total of 24 different cardiopulmonary injury types leading to
ECMO placement, or indications for placement, were identied in
our study (Table2). e main indications for ECMO support were
aer bilateral lung transplant (n=21), followed by bacterial and viral
pneumonia (n=18 and 12, respectively). One patient was treated with
extracorporeal cardiopulmonary resuscitation (ECPR), but died.
Age and sex were compared between survivors and non-survivors,
and many comorbidities were identied and compared (Table3).
Non-survivors were older (49 years v. 43 years, respectively;
p=0.019), and more males than females did not survive (64.9% v.
46.0%; p=0.039). Acute kidney injury was diagnosed in 34 patients at
the time of ECMO placement, ranging from an Acute Kidney Injury
Network (AKIN) classication of 1 (n=13) to AKIN 2 (n=11) and
AKIN 3 (n=10). Only 4 patients were known to have asthma prior
to ECMO placement, and all survived (p=0.035). Asthma was only a
concomitant comorbidity and not the primary indication for ECMO
placement in any of the patients. For the patients known to have
asthma, the indications for ECMO were bacterial pneumonia (n=1),
fungal pneumonia (n=1) and post bilateral lung transplant (n=2).
e ECMO outcome prediction scores for the likelihood of survival
in the form of a SAVE score for patients with adult cardiogenic shock
and a RESP score for adult respiratory failure were compared with our
data. Two patients in the all respiratory support group did not have all
the variables needed to generate an accurate RESP score.
In the all respiratory support group (n=76), 46.1% survived. e
respiratory support without lung transplant group (n=54) had a
survival rate of 42.5%. In the circulatory support group (n=29), the
survival rate was 34.4%. In the circulatory support group without
transplant (n=20), 35.0% survived.
e predicted survival likelihood compared with actual survival
is shown in Table4. In the all respiratory support group, patients
in risk class IV had better than predicted survival according to the
RESP score, while risk classes I, II and III had worse than predicted
survival. In the circulatory support group, all risk classes had worse
than predicted survival.
Overall survival for the transferred patients and the patients
retrieved on ECMO was 44.2% (n=19/43) and 16.7% (n=1/6),
respectively.
Discussion
ECMO is an expensive, specialised technology with a high mortality
rate, but it can be lifesaving for certain patients with reversible
cardiopulmonary failure. ECMO circuits are continuously improving
and are becoming smaller, safer to manage, and hopefully more
accessible. Identifying which patients would benet from ECMO
support, and the development of risk scores to assist with this process,
are therefore of the utmost importance.
To our knowledge this is the first study to report on ECMO
outcomes in SA.
Milpark Hospital is the only ELSO-registered ECMO referral centre
for adults in the northern part of SA.[3] Although improved outcomes
have been documented at specialist referral centres, delays in transfer
could lead to a delay in ECMO initiation and possibly an increase
in mortality.[3] The outcomes in the patients retrieved on ECMO
could possibly reect the increased mortality in critically ill patients
who are far from specialised facilities, and also the increased risk of
transporting them. e protocols and indications for the retrieval of
these patients will need to be reviewed to improve outcomes in this
specic group.
In our cohort, there was 100% survival in the patients known to
have asthma as a comorbidity. However, we had no patients placed on
ECMO for status asthmaticus in our cohort. ECMO has been shown
Table1. Procedures performed prior to ECMO placement
n (% of total
procedures)
Bilateral sequential lung transplant and single-
lung transplant 22 (34.9)
Coronary artery bypass graing 3 (4.8)
Valve replacement and coronary artery bypass
graing 2 (3.2)
Valve replacement or repair 6 (9.5)
Heart transplant 9 (14.2)
Coronary artery stenting 1 (1.6)
General surgical 8 (12.7)
oracic surgery 7 (11.1)
Bilateral pulmonary endarterectomy 4 (6.3)
Redo cardiac surgery 1 (1.6)
Total procedures 63
ECMO = extracorporeal membrane oxygenation.
Table2. Injury types/indications for ECMO placement
n (% of total
patients)
Post lung transplant 22 (20.6)
Bacterial pneumonia 18 (16.8)
Viral pneumonia 12 (11.2)
Post cardiotomy 9 (8.4)
Burns 2 (1.9)
Myocardial infarction 2 (1.9)
Bridge to transplant 2 (1.9)
Post cardiac transplant 8 (7.4)
Aspiration pneumonia 3 (2.8)
Pulmonary embolism 4 (3.7)
Haemorrhagic shock 1 (0.9)
Cardiac shock 3 (2.8)
Trauma 1 (0.9)
Gra failure 1 (0.9)
Transfusion-related lung injury 2 (1.9)
Cardiomyopathy 2 (1.9)
Le ventricular assist device placement 1 (0.9)
Fungal pneumonia 1 (0.9)
Tracheal surgery 2 (1.9)
Reperfusion injury post pulmonary
endarterectomy 4 (3.7)
Pulmonary contusion 2 (1.9)
Post lung resection 3 (2.8)
Chronic gra rejection 1 (0.9)
ECPR 1 (0.9)
Total 107
ECMO = extracorporeal membrane oxygenation;
ECPR = extracorporeal cardiopulmonary resuscitation.
AJTCCM VOL. 29 NO. 4 2023 155
ORIGINAL RESEARCH: ARTICLES
to improve survival in patients with status asthmaticus.[11] Asthma
was also found to be a protective factor during the development of
the RESP score. Pre-ECMO asthma diagnosis was associated with
17.7 increased odds of survival to hospital discharge (p<0.0001) in
the Predicting Survival aer ECMO for Severe Acute Respiratory
Failure study, and carries the highest weighting of 11 in the RESP
score.[7] It is interesting that in our study, patients with asthma as a
comorbidity showed statistically signicant survival. However, this
was in the setting of a retrospective review, and a further prospective
study will be necessary to evaluate this nding.
ECPR refers to the rapid deployment of VA-ECMO during eorts
to resuscitate a patient during cardiac arrest. ECPR is the most rapidly
growing indication for ECMO use, with 8 558 runs during 2020.[2]
ECPR has also been associated with the worst outcomes, with only
29% of patients surviving to discharge or transfer according to the
ECLS registry report of 2020. Only one patient in our study received
ECPR, and unfortunately died. e increased mortality rate of 83.3%
for patients transferred on ECMO and poor survival for patients
placed on ECMO during ECPR should be considered in non-ELSO-
registered hospitals in SA.
e role of ECMO in lung transplantation has expanded considerably.
ECMO can be used as a bridge to transplant, for intraoperative
support during transplantation, and for postoperative support. e
main indications for lung transplantation are idiopathic interstitial
pneumonia, COPD, cystic brosis, pulmonary arterial hypertension
and retransplant.[12,13]ese indications correlated with our cohort.
Pulmonary donor graft dysfunction continues to play a large
part in postoperative morbidity of transplant patients.[14] e use of
intraoperative ECMO support during lung transplantation with the
option to extend support postoperatively has multiple benets and
should continue to rise in view of excellent results achieved by some
groups.[15,16] It is therefore not surprising that the most common
postoperative indication for ECMO use in the present study was in
the setting of lung transplantation.
Although the primary aim of this study was not to externally
validate the RESP and SAVE scores, it did provide us with information
regarding the predicted survival likelihoods for our cohort. In the
all respiratory support group, risk class IV had better than predicted
survival according to the RESP score, while risk classes I, II and III
had worse than predicted survival. In the circulatory support group,
all risk classes had worse than predicted survival. However, these
groups were small and inadequately powered to make denitive
conclusions. A possible reason for the worse than predicted survival
for the circulatory support group is that with the initial development
of the SAVE score, patients who received VA-ECMO during CPR
and patients on their second run of ECMO were excluded,[8]
while these patients were included in our cohort. Further external
validation of the risk scores will be needed to assess their accuracy
in our population group.
Conversion from VV-ECMO to VA-ECMO is associated with
decreased survival.[17] Our patient cohort included 4 patients who
were placed on VA-ECMO after initial VV-ECMO. Although all
4patients died, this did not reach statistical signicance owing to the
small number of patients in this group.
is retrospective study has several limitations. We were unable to
collect accurate data on complications and causes of death. Patients
Table3. Demographics and comorbidities of all ECMO patients*
All ECMO (N=107),
n (%)
Survivors (n=47),
n (%)
Non-survivors
(n=60), n (%)p-value
Age (years), mean (SD) 47 (13.2) 43 (13.8) 49 (12.4) 0.019
Sex male 57 (53.3) 20 (42.6) 37 (61.7) 0.039
Hypertension 37 (34.5) 11 (23.4) 26 (43.3) 0.022
COPD 11 (10.2) 5 (10.6) 6 (10.0) 0.884
Diabetes 21 (19.6) 8 (17.0) 13 (21.7) 0.585
Asthma 4 (3.7) 4 (8.5) 0 0.035
Chronic pulmonary hypertension 27 (25.2) 12 (25.5) 15 (25.0) 0.899
Sarcoidosis 1 (0.9) 1 (2.1) 0 0.256
Epilepsy 2 (1.8) 0 2 (3.3) 0.206
Fungal pneumonia 5 (4.7) 2 (4.2) 3 (5.0) 0.462
Trauma/burn 7 (6.5) 3 (6.4) 4 (6.7) 0.619
Interstitial lung disease 17 (15.9) 10 (21.3) 7 (11.7) 0.640
Tracheostomy 42 (39.2) 15 (31.9) 27 (45.0) 0.373
Cystic brosis 4 (3.7) 2 (4.2) 2 (3.3) 0.802
Chronic thromboembolic pulmonary hypertension 4 (3.7) 3 (6.4) 1 (1.7) 0.201
Cancer 2 (1.8) 0 2 (3.3) 0.502
Acute kidney injury upon initiation 34 (31.7) 10 (21.3) 24 (40.0) 0.368
AKIN 1 13 (12.1) 4 (8.5) 9 (15.0) 0.185
AKIN 2 11 (10.2) 3 (6.4) 8 (13.3) 0.147
AKIN 3 10 (9.3) 3 (6.4) 7 (11.7) 0.219
ECMO = extracorporeal membrane oxygenation; COPD = chronic obstructive pulmonary disease; AKIN = Acute Kidney Injury Network classication.
*Some patients had more than one comorbidity.
Except where otherwise indicated.
156 AJTCCM VOL. 29 NO. 4 2023
ORIGINAL RESEARCH: ARTICLES
who undergo ECMO patients have prolonged ICU stays with complex
diagnoses and multiple factors impacting on their outcomes. Owing to
the wide variation in ECMO indications in our cohort, the subgroups
became too small to make deductions regarding specic indications
and mortality.
Conclusion
We report ECMO outcomes in SA for the rst time. We identied very
high mortality rates for patients transferred on ECMO from other
facilities and for patients converted from VV-ECMO to VA-ECMO.
Although our outcomes were comparable in some of the risk classes,
further external validation of the SAVE and RESP scores will be
needed to compare our outcomes with these scores.
Declaration. e research for this study was done in partial fullment of
the requirements for NLFvZs MMed (or Surg) degree at Stellenbosch
University.
Acknowledgements. We would like to thank Dr Paul G Williams for his
contribution.
Author contributions. NLFvZ, JTJ, AG and MS contributed to the
write-up and review of the article. All authors read and approved the
nal version.
Funding.None.
Conicts of interest.None.
1. Brogan TV, Lequier L, Lorusso R, MacLaren G, Peek G, eds. Extracorporal Life Support:
e ELSO Red Book. https://www.elso.org/ecmo-resources/6theditionredbook.aspx
2. Extracorporeal Life Support Organization. ECLS registry report & international summary
of statisitcs. 2023. https://www.elso.org/registry/internationalsummaryandreports.aspx
(accessed 6 November 2023).
3. Extracorporeal Life Support Organization (ELSO). Registry of active ELSO centers
using ECMO. https://www.elso.org/Registry.aspx (accessed 23 July 2021).
4. Braune S, Sieweke A, Brettner F, etal. e feasibility and safety of extracorporeal
carbon dioxide removal to avoid intubation in patients with COPD unresponsive to
noninvasive ventilation for acute hypercapnic respiratory failure (ECLAIR study):
Multicentre case-control study. Intensive Care Med 2016;42(9):1437-1444. https://
doi.org/10.1007/s00134-016-4452-y
Table4. Predicted survival compared with actual survival
NGrade Predicted survival, % Mortality, nSurvival per grade, %
All respiratory support
RESP score
≤6 4 I 92 1 75.0
3 - 5 26 II 76 12 53.9
–1 - 2 33 III 57 19 42.5
–2 - –5 8 IV 33 5 37.5
≥–6 5 V 18 4 20.0
Total 76 41 46.1
Respiratory support without lung transplant
RESP score
≤6 4 I 92 1 75.0
3 - 5 16 II 76 8 50.0
–1 - 2 23 III 57 14 29.1
–2 - –5 7 IV 33 5 28.6
≥–6 4 V 18 3 25.0
Total 54 31 42.5
All circulatory support
SAVE score
>5 3 I 75 1 66.6
1 - 5 8 II 58 4 50.0
–4 - 0 10 III 42 8 20.0
–9 - –5 8 IV 30 6 25.0
≤–10 0 V 18 - -
Total 29 19 34.4
Circulatory support without cardiac transplant
SAVE score
>5 2 I 75 1 50.0
1 - 5 5 II 58 3 40.0
–4 - 0 6 III 42 4 33.3
–9 - –5 7 IV 30 5 28.6
≤–10 0 V 18 - -
Total 20 13 35.0
RESP = Respiratory ECMO Survival Prediction; SAVE = Survival Aer Venoarterial ECMO.
AJTCCM VOL. 29 NO. 4 2023 157
ORIGINAL RESEARCH: ARTICLES
5. Ouweneel DM, Schotborgh JV, Limpens J, etal. Extracorporeal life support during
cardiac arrest and cardiogenic shock: A systematic review and meta-analysis. Intensive
Care Med 2016;42(12):1922-1934. https://doi.org/10.1007/s00134-016-4536-8
6. Hakim AH, Ahmad U, McCurry KR, etal. Contemporary outcomes of extracorporeal
membrane oxygenation used as bridge to lung transplantation. Ann orac Surg
2018;106(1):192-198. https://doi.org/10.1016/J.ATHORACSUR.2018.02.036
7. Schmidt M, Bailey M, Sheldrake J, etal. Predicting survival after extracorporeal
membrane oxygenation for severe acute respiratory failure: The Respiratory
Extracorporeal Membrane Oxygenation Survival Prediction (RESP) score. Am J Respir
Crit Care Med 2014;189(11):1374-1382. https://doi.org/10.1164/rccm.201311-2023OC
8. Schmidt M, Burrell A, Roberts L, etal. Predicting survival aer ECMO for refractory
cardiogenic shock: e survival aer veno-arterial-ECMO (SAVE)-score. Eur Heart J
2015;36(33):2246-2256. https://doi.org/10.1093/eurheartj/ehv194
9. Ahuja A, Shekar K. Patient selection for VV ECMO: Have we found the crystal ball? J
orac Dis 2018;10(Suppl 17):S1979-S1981. https://doi.org/10.21037/JTD.2018.05.83
10. Hilder M, Herbstreit F, Adamzik M, etal. Comparison of mortality prediction
models in acute respiratory distress syndrome undergoing extracorporeal membrane
oxygenation and development of a novel prediction score: e PREdiction of Survival
on ECMO erapy-Score (PRESET-Score). Crit Care 2017;21(1):301. https://doi.
org/10.1186/s13054-017-1888-6
11. Di Lascio G, Prii E, Messai E, etal. Extracorporeal membrane oxygenation support
for life-threatening acute severe status asthmaticus. Perfusion 2017;32(2):157-163.
https://doi.org/10.1177/0267659116670481
12. Bos S, Vos R, van Raemdonck DE, Verleden GM. Survival in adult lung transplantation:
Where are we in 2020? Curr Opin Organ Transplant 2020;25(3):268-273. https://doi.
org/10.1097/MOT.0000000000000753
13. Chambers DC, Perch M, Zuckermann A, etal. e International oracic Organ
Transplant Registry of the International Society for Heart and Lung Transplantation:
Thirty-eighth adult lung transplantation report – 2021; Focus on recipient
characteristics. J Heart Lung Transplant 2021;40(10):1060-1072. https://doi.
org/10.1016/j.healun.2021.07.021
14. Snell GI, Yusen RD, Weill D, etal. Report of the ISHLT Working Group on Primary
Lung Gra Dysfunction, part I: Denition and grading – a 2016 Consensus Group
statement of the International Society for Heart and Lung Transplantation. J Heart Lung
Transplant 2017;36(10):1097-1103. https://doi.org/10.1016/J.HEALUN.2017.07.021
15. Hoetzenecker K, Benazzo A, Stork T, et al. Bilateral lung transplantation on
intraoperative extracorporeal membrane oxygenator: An observational study.
J Thorac Cardiovasc Surg 2020;160(1):320-327.e1. https://doi.org/10.1016/J.
JTCVS.2019.10.155
16. Hoetzenecker K, Schwarz S, Muckenhuber M, etal. Intraoperative extracorporeal
membrane oxygenation and the possibility of postoperative prolongation improve
survival in bilateral lung transplantation. J orac Cardiovasc Surg 2018;155(5):2193-
2206.e3. https://doi.org/10.1016/J.JTCVS.2017.10.144
17. Falk L, Fletcher-Sandersjöö A, Hultman J, Broman LM. Conversion from venovenous
to venoarterial extracorporeal membrane oxygenation in adults. Membranes (Basel)
2021;11(3):188. https://doi.org/10.3390/membranes11030188
Submitted 7 February 2022. Accepted 10 September 2023. Published 27 November 2023.