Extra-Corporeal Cardiopulmonary Resuscitation (ECPR) for Cardiac Arrest Patients-3-Year Experience in the Era of Left Ventricular Decompression

Abstract

Background
Cardiac arrest has been a great threat, regardless of its occurrence in or out of the hospital (OHCA and IHCA). ECMO has been regarded as the last resort to save these endangering lives. This prospective cohort intended to clarify how ECMO-assisted CPR (ECPR) benefited patients of cardiac arrests, both OHCA and IHCA, and addressed the survival benefit of left ventricular decompression.

Methods
Consecutive patients of OHCA and IHCA, refractory to resuscitation and without certain exclusion criteria for mechanical support, were enrolled since 2018. Primary endpoint was in-hospital mortality.

Results
From 2018 Jan. to 2021 Mar., 147 patients of both OHCA and IHCA were put on ECPR. Acute coronary syndrome took 63.3%. Rate of weaning-off from ECMO/modified LVAD was 45.6%, and the in-hospital mortality rate was 67.3%. Peak serum lactate (10.9 ± 5.5 vs. 16.4 ± 7.7 mmol/L, p < 0.001) and left ventricular ejection fraction (35.8 ± 20.6% vs. 26.7 ± 21.7%, p = 0.029) during mechanical support were of prognostic significance. The low-flow time was not correlated with these two prognostic parameters and in-hospital mortality in our series. Forty-five patients with poorer left ventricular function were put on modified LVAD, which rendered survival benefit over VA-ECMO by 3 days: 1st day mortality (VA-ECMO vs modified LVAD 18.6% vs. 2.2%, p = 0.008); 3rd day (35.3% vs 17.8%, p = 0.034). The Kaplan-Meyer analysis depicted the survival benefit of modified LVAD (log-rank p = 0.024)

Conclusion
Our prospective cohort, enrolling both OHCA and IHCA patients put on ECPR, implied that fair outcomes could be achieved in both groups, as long as high quality CPR could be persistently applied, regardless of the low-flow time. Peak serum lactate and LVEF were proved again to be closely correlated to outcomes. Moreover, the procedure of LV decompression could render further survival benefit to those with poorer LV function.

Full text

Page 1/17
Extra-Corporeal Cardiopulmonary Resuscitation
(ECPR) for Cardiac Arrest Patients-3-Year
Experience in the Era of Left Ventricular
Decompression
Ho-Tsung Hsin (  hsinht@gmail.com )
Far Eastern Memorial Hospital https://orcid.org/0000-0001-7802-3243
Kuan-Ming Chiu
Far Eastern Memorial Hospital
Ying-Chiu Wu
Far Eastern Memorial Hospital
Fu-Chien Hsieh
Far Eastern Memorial Hospital
Jer-Shen Chen
Far Eastern Memorial Hospital
Jih-Hsin Huang
Far Eastern Memorial Hospital
Mei-Ling Wu
Far Eastern Memorial Hospital
Chieh-Fu Chen
Far Eastern Memorial Hospital
Kei-Ip Cheong
Far Eastern Memorial Hospital
Cheng-Hung How
Far Eastern Memorial Hospital
Research Article
Keywords: cardiac arrest, OHCA, IHCA, ECMO, LV decompression
Posted Date: September 28th, 2021
DOI: https://doi.org/10.21203/rs.3.rs-919268/v1
License:   This work is licensed under a Creative Commons Attribution 4.0 International License. 
Read Full License

Page 2/17

Page 3/17
Abstract
Background
Cardiac arrest has been a great threat, regardless of its occurrence in or out of the hospital (OHCA and
IHCA). ECMO has been regarded as the last resort to save these endangering lives. This prospective
cohort intended to clarify how ECMO-assisted CPR (ECPR) beneted patients of cardiac arrests, both
OHCA and IHCA, and addressed the survival benet of left ventricular decompression.
Methods
Consecutive patients of OHCA and IHCA, refractory to resuscitation and without certain exclusion criteria
for mechanical support, were enrolled since 2018. Primary endpoint was in-hospital mortality.
Results
From 2018 Jan. to 2021 Mar., 147 patients of both OHCA and IHCA were put on ECPR. Acute coronary
syndrome took 63.3%. Rate of weaning-off from ECMO/modied LVAD was 45.6%, and the in-hospital
mortality rate was 67.3%. Peak serum lactate (10.9 ± 5.5 vs. 16.4 ± 7.7 mmol/L, p < 0.001) and left
ventricular ejection fraction (35.8 ± 20.6% vs. 26.7 ± 21.7%, p = 0.029) during mechanical support were of
prognostic signicance. The low-ow time was not correlated with these two prognostic parameters and
in-hospital mortality in our series. Forty-ve patients with poorer left ventricular function were put on
modied LVAD, which rendered survival benet over VA-ECMO by 3 days: 1st day mortality (VA-ECMO vs
modied LVAD 18.6% vs. 2.2%, p = 0.008); 3rd day (35.3% vs 17.8%, p = 0.034). The Kaplan-Meyer
analysis depicted the survival benet of modied LVAD (log-rank p = 0.024)
Conclusion
Our prospective cohort, enrolling both OHCA and IHCA patients put on ECPR, implied that fair outcomes
could be achieved in both groups, as long as high quality CPR could be persistently applied, regardless of
the low-ow time. Peak serum lactate and LVEF were proved again to be closely correlated to outcomes.
Moreover, the procedure of LV decompression could render further survival benet to those with poorer LV
function.
Background
Cardiac arrests, both out-of-hospital (OHCA) and in-hospital (IHCA), threatened hundreds of thousands of
lives each year. Extra-corporeal membranous oxygenation (ECMO), composed of a blood pump, gas
exchange device, conduits with vascular access and heat exchanger, has been advocated as the last
resort to rescue those patients refractory to our resuscitation efforts[1, 2]. The incorporation of ECMO in

Page 4/17
our resuscitation efforts has been referred to as extra-corporeal cardiopulmonary resuscitation (ECPR).
Despite the ushering of the concept of ECPR, the American Heart Association (AHA) still reported
insucient evidence to recommend the routine use of ECPR for patients with cardiac arrests in 2019. It
only suggested the application of ECPR when conventional CPR failed[3]. In addition, left ventricular (LV)
decompression, either by active apical venting or passive upstream left atrial drainage, has been
introduced to avoid overload of the failing left ventricle[4]. The combination of ECMO and LV
decompression in cardiac arrest has not been explored extensively.
We conducted this prospective cohort since 2018, during which our protocol of LV decompression was
well established. The study aims to clarify the ecacy of ECPR with or without LV decompression in
rescuing patients of cardiac arrest.
Methods
Setting
This study was conducted in a tertiary referring medical center in Northern Taiwan, where at least 450
open-heart surgeries were performed each year. Patients of OHCA were put on CPR by either bystander or
emergent medical technician (EMT), and followed by efforts made by the medical team in emergency
service in our hospital. Resuscitation of IHCA were provided by dedicated medical personnel, including
senior nurse practitioner, medical resident, senior nurse, respiratory therapist, and duty attending
physician. All members providing CPR are certied by authorized board of advanced cardiac life-support
(ACLS) in Taiwan. The ECMO task force consisted of a duty cardiovascular surgeon, a surgical resident, a
perfusionist, a nurse practitioner, and staff of Operation Theater if needed. A designated perfusionist was
assigned to record all relevant data. All patients put on ECMO were all taken care in the cardiovascular
intensive care unit (CVICU), in which around 6,900 patient-days was accommodated annually. The cause
of cardiac arrest was delineated by a committee organized by the intensivist of CVICU. All patients
without meaningful responses after the resuscitation would be put on target temperature management
(TTM) for 24 hours, unless there was contra-indications, such as uncontrolled bleeding diathesis (gastro-
intestinal or intracranial hemorrhage), and refractory hypothermia related cardiac arrhythmias. We
adhered to contemporary clinical guidelines to apply adjunctive medical treatment for specic disease
entity, such as myocardial infarction, cardiogenic shock, heart failure, pulmonary embolism, etc[5–7].
Patients and procedures
Cardiac arrest of any cause, which mandated CPR for more than 10 minutes, would activate the ECMO
task force by the CPR team. The task force was responsible for the decision to proceed or not, after
assessment and discussion with the CPR team and the aicted family. We excluded patients, who were
elder than 80-years-old, presented with active massive bleeding, bed-ridden status, severe irreversible
brain damage, with known terminal disease with less than 6 months of life expectancy, and those who
already requested “Do Not Resuscitate” (DNR). The study protocol was approved by the institutional
review board. Informed consent for any necessary invasive procedure and data collection were obtained

Page 5/17
either from the patient's legitimate family. The whole study has been approved and scrutinized by the
Institutional Review Board of Far-Eastern Memorial Hospital, New Taipei City, Taiwan.
Procedure
In general, the interval between activation and arrival of ECMO task force in our hospital was around 10
min 24/7, as there was always a duty cardiovascular surgeon. In addition, we have always made one set
of heparinized saline (2 U/ml)-primed circuit and centrifugal pump available. Indwelling an antegrade
reperfusion catheter in the supercial femoral artery has been our routine to avoid any distal limb
ischemia. The ECMO circuit was connected to a temperature console, immediately after the system was
established. As the setup of ECMO system was accomplished, a brain computed tomography would be
performed, in order to exclude any intra-cranial pathology. Echocardiographic surveillance was performed
on daily basis, in order to clarify myocardial status, focusing on left ventricular ejection fraction (LVEF),
right ventricular function and other relevant complications. Only the LVEF taken in 72 hours of
resuscitation was enlisted, in order to avoid the inuence of myocardial stunning.
Under one of the following conditions, we would apply procedures of LV decompression, such as direct
surgical LV apical venting, upstream LV drainage via one of the upper pulmonary veins either by surgical
sternotomy or percutaneous trans-septal approach. The indications included: 1. Refractory lung edema, 2.
Echocardiographic evidence of marked LV distension, 3. Refractory ventricular tachycardia/brillation, 4.
Elevated LV end-diastolic pressure (LVEDP), either illustrated by pulmonary arterial occlusive pressure
(PAOP) by Swan-Ganz catheter (PAOP > 25 mmHg), or by direct measurement during LV catheterization
(LVEDP > 25 mmHg) in the cath lab[8]. The drainage tube of LV decompression was bridged to the ECMO
venous tube, connected to the inlet of the centrifugal pump together. The arterial access returning to the
patient would be left as the original ECMO arterial access, or implanted to the ascending aorta directing
antegradely, if the patient underwent any surgical intervention, such as coronary artery bypass graft
surgery (CABG), or the aforementioned surgical LV decompression. If the systolic function of right
ventricle (RV) recovered (dened by echocardiographic evidence: direct 2D visual or tricuspid annular
plane systolic excursion (TAPSE))[9], the femoral venous access of ECMO would be de-cannulated, and
the drainage tube of LV decompression would be connected to the inlet of centrifugal pump directly. As
the patient’s oxygenation became fair, we would taper the support of our ECMO’s membranous
oxygenator, which would be removed after the patient became independent of it. In order to unify the
terminology, the whole system, consist of a tube draining the LV, the centrifugal pump, and the arterial
access returning to the patient, would be referred to as “modied left ventricular assist device (modied
LVAD)”. In the context, the terms of “LV decompression” and “modied LVAD” were synonymous and
interchangeable.
Success of weaning-off of ECMO or modied LVAD is dened as no deterioration of hemodynamics,
necessitating repeated application of any mechanical support or mortality in 72 hours after
discontinuation. Bridge to implantable ventricular assist device, such as Heartware™, or heart transplant
was also regarded as a successful weaning-off.

Page 6/17
The primary endpoint was in-hospital all-cause mortality. Successful weaning-off of ECMO/modied
LVAD is regarded as 2nd endpoint.
Statistical analysis
Continuous variables were expressed as mean ± standard deviation or median, depending on the
normality test results. Categorical variables are expressed as number and percentage of patients, and
were compared by using χ2 and Fischer’s exact tests. Group means for continuous variables were
compared by Student’s t-test or the Wilconxin test as appropriate. The low-ow-time was dened as the
beginning of witnessed CPR to the establishment of ECMO ow. The mortalities were evaluated at the
intervals of 1, 3, 7, 14, 21, 28-days and as a whole by overall in-hospital mortality, which was also
compared b yχ2 and Fischer’s exact tests. Kaplan-Meyer survival curve was adopted to illustrate the
difference between ECMO with/without LV decompression. The correlations were analysed by a
correlation coecient, χ2 test for categorical variables and Pearson’s correlation for continuous variables.
All statistical tests were two tailed, and a p value < 0.05 was deemed statistically signicant. The analysis
was done with Statistical Package for Social Sciences software (SPSS ver.12.0 for Windows).
Results
From 2018 Jan. to 2021 Mar., there were 147 consecutive patients of cardiac arrests put on ECPR. The
enrollment of the study was withheld in Apr. 2021, because the COVID-19 pandemic in northern Taiwan
interrupted the normal conduct of resuscitation and activation of ECMO task force, especially for those
patients presenting with OHCA, who might pose unknown infectious threat to our medical staff. Our
enrollee were 58.8 ± 12.3 years-old, and a majority of them was male (86.4%). IHCA patients comprised
57.8% of our study group. Acute coronary syndrome accounted for 63.3% of the underlying cause of
cardiac arrest. It took less than 30mins for the ECMO task force to establish the system after activation,
both for OHCA and IHCA patients (25 ± 12 vs. 27 ± 14 min, p = 0.355). The overall success of weaning of
ECMO/modied LVAD was 45.6%, and the primary endpoint of in-hospital all-cause mortality was
achieved by 67.3% of the enrollee, in other words, 32.7% of our patients of cardiac arrests survived to
discharge. Our patients were put on mechanical support for 8.7 ± 15.5 days, and they would stay in the
hospital for 22.0 ± 24.0 days. (Table1)

Page 7/17
Table 1
Demographics of all patients, categorized by OHCA and IHCA
OHCA IHCA Overall p value
Case No. 62 85 147 
Age (yr) 57.3 ± 11.2 60.1 ± 13.0 58.8 0.085
Sex    
Male 57(91.9%) 70(82.4%) 127(86.4%) 0.143
Female 5(8.1%) 15(17.6%) 20(15.6% 
Hypertension 29(46.8%) 51(60.0%) 80(54.4%) 0.132
Diabetes 17(27.4%) 32(37.6%) 49(33.3%) 0.218
Smoking 30(48.4%) 37(43.5%) 67(45.6%) 0.616
Low ow time# (min) 80.7 ± 44.5 45.6 ± 31.3  < 0.001
ECMO esta. Time* (min) 25 ± 12 27 ± 14  0.355
Peak lactate (mmol/L) 14.8 ± 6.5 14.2 ± 8.1  0.657
LVEF+ (%) 26.1 ± 21.1 33.5 ± 21.7  0.079
Diagnosis    0.317
CAD/MI@43(69.4%) 50(58.8%) 93(63.3%) 
pulmonary embolism 2(3.2%) 3(3.5%) 5(3.4%) 
myocarditis 1(1.6%) 2(2.4%) 3(2.0%) 
peri-open heart shock&0(0%) 4(4.7%) 4(2.7%) 
other shock 15(24.2%) 20(23.5%) 35(23.8%) 
Aortic dissection 1(1.6%) 6(7.1%) 7(4.8%) 
ECMO status    
# Time interval between witnessed CPR and the establishment of ECMO ow
* ECMO establish time: interval between the activation of ECMO task force and the establishment of
ECMO ow
+ Left ventricular ejection fraction
@ Coronary artery disease/myocardial infarction
& Shock during open heart surgery

Page 8/17
OHCA IHCA Overall p value
VA ECMO 43(69.4%) 59(69.4%) 102(69.4%) 1.0
Modied LVAD 19(30.6%) 26(30.6%) 45(30.6%) 1.0
Wean off 27(43.5%) 40(47.1% 67(45.6%) 0.738
ECMO/VAD days 7.1 ± 10.0 9.9 ± 18.5 8.7 ± 15.5 0.280
Hospital stay (day) 17.9 ± 21.9 24.9 ± 25.1 22.0 ± 24.0 0.080
Mortality 42(67.7%) 57(67.1%) 99(67.3%) 1.0
# Time interval between witnessed CPR and the establishment of ECMO ow
* ECMO establish time: interval between the activation of ECMO task force and the establishment of
ECMO ow
+ Left ventricular ejection fraction
@ Coronary artery disease/myocardial infarction
& Shock during open heart surgery
Regarding the primary endpoint of in-hospital all-cause mortality, both survivors and those who
succumbed to were similar in demographics, including age, co-morbidities (diabetes, hypertension,
dyslipidemia and current smoking). There was no difference in the rapidity of setting up the ECMO
system (survivor vs. mortality: 23.8 ± 11.2 vs. 27.3 ± 14.0 mins, p = 0.129). The proportion of OHCA in
both groups was not different, either (41.7% vs. 42.4%, p = 0.93). Male predominance was more
signicant in the mortality group (75.0% vs. 91.9%, p = 0.005). Duration of ECMO or modied LVAD did
not vary signicantly (survivor vs. mortality: 12.2 ± 20.9 vs. 7.1 ± 11.9 days, p = 0.065). Wean-off rate was
much higher in the survivor group (93.8% vs 22.2%, p < 0.001). Longer hospital stay (40.0 ± 25.6 vs. 13.2
± 17.5 days, p < 0.001, 95% conf int 19.6, 33.9) was also expected among survivors. Peak serum lactate
level (survivor vs. mortality 10.9 ± 5.5 vs. 16.4 ± 7.7 mmol/L, p < 0.001, 95% conf int. -8.1, -3.1) and LVEF
during mechanical support (35.8 ± 20.6 vs. 26.7 ± 21.7%, p = 0.029, 95% conf int.1.0, 17.4) were
signicantly correlated with prognosis. (Table2)

Page 9/17
Table 2
Demographics categorized by survival and mortality
Survival Mortality p value
Case No. 48 99 
Age (yr) 58.5 ± 10.2 59.2 ± 13.2 0.768
Sex   0.005
Male 36(75.0%) 91(91.9%) 
Female 12(25.0%) 8(8.1%) 
Hypertension 21(43.8%) 59(59.6%) 0.07
Diabetes 16(33.3%) 33(33.3%) 1.0
Smoking 27(56.3%) 40(40.4%) 0.07
OHCA 20(41.7%) 42(42.4%) 0.93
IHCA 28(58.3%) 57(57.6%) 0.93
Low ow time# (min) 60.0 ± 44.1 60.4 ± 39.8 0.98
ECMO/VAD days 12.2 ± 20.9 7.1 ± 11.9 0.065
Hospital stay (day) 40 ± 25.6 13.2 ± 17.5 < 0.001
ECMO esta. Time* (min) 23.8 ± 11.2 27.3 ± 14.0 0.129
Wean off 45(93.8%) 22(22.2%) < 0.001
Peak lactate (mmol/L) 10.9 ± 5.5 16.4 ± 7.7 < 0.001
LVEF+ (%) 35.8 ± 20.6 26.7 ± 21.2 0.029
LDL (mg/dL) 91.9 ± 39.4 80.2 ± 40.0 0.13
#Time interval between witnessed CPR and the establishment of ECMO ow
*ECMO establish time: interval between the activation of ECMO task force and the establishment of
ECMO ow
+Left ventricular ejection fraction
Forty-ve patients (30.6% of our registry) underwent the procedure of LV decompression (surgical LA
drain/surgical apical venting/trans-septal LA drain: 28/4/13 cases), with similar distribution in
demographics and ECMO establish time, except that more male patients were in the group of modied
LVAD (VA-ECMO vs. Modied LVAD: 82.4% vs 95.6%, p = 0.036). Poorer LVEF rendered more LV
decompression (39.4 ± 21.3 vs. 16.2 ± 12.9%, p < 0.001, 95% conf int. 16.0, 30.5). Those patients put on
modied LVAD got longer dependence on the mechanical support (3.7 ± 3.7 vs. 20.3 ± 24.1 days, p <

Page 10/17
0.001, 95% conf int. -21.4, -11.7) and longer hospital stay (17.8 ± 19.1 vs 31.4 ± 30.7 days, p = 0.001, 95%
conf int. -21.8, -5.4). Wean-off rate in both modalities of mechanical support was nearly identical (46.1%
vs. 44.4%, p = 1.0). The mortality rate by interval signicantly differed on the 1st day (18.6% vs. 2.2%, p =
0.008) and the 3rd day (35.3% vs 17.8%, p = 0.034) (Table3). The Kaplan-Meyer analysis further depicted
the survival benet of modied LVAD (log-rank p = 0.024). (Fig.1)

Page 11/17
Table 3
Demographics categorized by VA-ECMO and modied LVAD
VA-ECMO Modied LVAD p value
Case No. 102 45 
Sex   0.036
Male 84(82.4%) 43(95.6%) 
Female 18(17.6%) 2(4.4%) 
Age (yr) 59.8 ± 12.9 57 ± 10.7 0.202
Hypertension 56(54.9%) 24(53.3%) 0.86
DM 39(38.2%) 10(22.2%) 0.061
Smoking 45(44.1%) 22(48.9%) 0.596
Low ow time# (min) 64.5 ± 41.4 50.7 ± 39.3 0.068
ECMO est. time*(min) 26.8 ± 13.0 24.6 ± 13.6 0.355
IHCA 53(52.0%) 24(53.3%) 1.0
OHCA 43(48.0%) 19(46.7%) 1.0
ECMO/VAD days 3.7 ± 3.7 20.3 ± 24.1 < 0.001
Hospital stay(days) 17.8 ± 19.1 31.4 ± 30.7 0.001
Wean off 47(46.1%) 20(44.4%) 1.0
Peak lactate (mmol/L) 14.4 ± 7.4 14.3 ± 7.7 0.944
LVEF+ (%) 39.4 ± 21.3 16.2 ± 12.9 < 0.001
LDL (mg/dL) 84.8 ± 40.8 84.8 ± 39.1 0.997
Mortality   
1 day 19(18.6%) 1(2.2%) 0.008
3 day 36(35.3%) 8(17.8%) 0.034
7 day 45(44.1%) 14(31.1%) 0.149
14 day 54 (52.9%) 17(37.8%) 0.108
#Time interval between witnessed CPR and the establishment of ECMO ow
*ECMO establish time: interval between the activation of ECMO task force and the establishment of
ECMO ow
+Left ventricular ejection fraction

Page 12/17
VA-ECMO Modied LVAD p value
21 day 56(54.9%) 20(44.4%) 0.284
28 day 60(58.8%) 24(53.3%) 0.589
In-hospital 67(65.7%) 32(71.1%) 0.571
#Time interval between witnessed CPR and the establishment of ECMO ow
*ECMO establish time: interval between the activation of ECMO task force and the establishment of
ECMO ow
+Left ventricular ejection fraction
Discussion
Our study has illustrated current practice of ECPR in a referring center in northern Taiwan. The protocol of
LV decompression/modied LVAD has been established at the end of 2017, and hence the study began in
2018. First of all, coronary artery disease contributed to 63.3% of our registry of cardiac arrest. Secondly,
peak serum lactate and LVEF predicted mortality in our cohort. Most intriguing, modied LVAD, mandated
by poorer LVEF, seemed to benet those who were expected to come with worse outcome.
OHCA vs. IHCA
Our cohort enrolled patients of both OHCA and IHCA of any causes. The composition of diagnosis was
compatible with known literature, with the consensus that nearly two-thirds of cardiac arrest events were
attributed to coronary artery disease [10–12]. Previously, the survival of OHCA patients was expected to
be inferior to IHCA ones, as IHCA patients might merit more ecient CPR efforts. The overall survival to
discharge of ECPR-treated IHCA patients was 37.9%, while that of ECPR-treated OHCA patients was
heterogeneously between 6.9%and 56.0%, averagely around 20% [13, 14]. OHCA patients were inevitably
aicted with longer low-ow time, which was believed to be detrimental in prognosis. In our series, OHCA
patients got longer low-ow time by 35.2 mins, in comparison with their IHCA counterparts. (80.7 ± 44.5
vs 45.6 ± 31.3 min, p < 0.001, 95% conf Int. 22.5, 47.8). But this drawback did not necessarily mean poorer
systemic perfusion, which could be vividly reected by peak serum lactate level after resuscitation [15].
The length of low-ow time did not correlate with the peak serum lactate after resuscitation in our study
(R2 = 0.001). (Fig. 2a)
Therefore, the peak serum lactate of our OHCA and IHCA patients were not distinct, implying similar
suciency of resuscitation was achieved. As a result, our OHCA and IHCA patients put on ECPR shared
similar rate of weaning-off from mechanical support, and nearly identical rate of survival to discharge.
(32.3% vs. 32.9%, p = 1.0). In our opinion, if high quality CPR could be applied consistently and
persistently, patients of cardiac arrest could share the same outcome, in spite of the difference of low-
ow time.

Page 13/17
The prognostic factors: peak serum lactate and LVEF
The peak serum lactate and LVEF cast signicant predictability. Those with higher lactate level or poorer
LV function suffered higher mortality rate in our cohort. These two factors have been well reported to tell
the prognosis in the literature [16]. In our study, the low-ow time did not differ between those who
survived and those who succumbed to (60.0 ± 44.1 vs. 60.4 ± 39.8 mins, p = 0.948). Neither were the peak
serum lactate (R2 = 0.001) (Fig.2) nor the LVEF (R2 = 0.002) correlated with the low-ow time. (Fig.2b).
Furthermore, the peak serum lactate and LVEF were not interactive and no correlation existed (p = 0.956).
In addition, the duration of dependence on mechanical support did not inuence the outcome (survival vs.
mortality 12.2 ± 20.9 vs. 7.1 ± 11.9 days, p = 0.065), but it was the wean-off rate that mattered (93.8 vs.
22.2%, p < 0.001). In other words, as long as the patient could get rid of the indwelled machinery, he or she
may be able to survive to discharge, regardless of how long the patient was put on the machine.
The LV decompression/modied LVAD
So far, our study demonstrated comparable and compatible characters with currently available literature,
such as diagnosis composition, mortality of ECPR-treated IHCA, prognostic predictability of peak serum
lactate and LVEF. This legitimized further extrapolation of our cohort, to explore the ecacy of LV
decompression in the care of cardiac arrest patients. In addition to equal distribution of demographics in
the two groups of VA-ECMO and modied LVAD, the highly predictive parameter, peak serum lactate, was
not distinct among them. Even the low-ow time was similar. LVEF was the only different parameter. But
in this category, it should be regarded as the driving factor that rendered urgent indication for LV
decompression, rather than the contributor to the outcome. The ration could be, the lower the LVEF, the
more the need for LV decompression, and resultantly, the better the survival. Based on the cross-sectional
mortality, the modied LVAD offered survival benet over VA-ECMO by 3 days in our ECPR patients, which
was further illustrated by Kaplan-Meyer analysis.
In the study conducted by Chen YS et al., the authors found that ECPR did benet IHCA patients, but
OHCA patients with longer low-ow-time were not enrolled [17]. There was no further delineation of the
role of LV decompression then. To our knowledge, there were only 3 individual series that comprised more
than 20 cases of LV decompression: Centofanti P et al. 24 cases, Eastaugh LJ et al. 44 cases, Hacking
DF et al. 49 cases [18–20]. The latter two were studying pediatric patients, and only Centofanti P et al.
addressed the survival benet of trans-apical LV venting in adults with cardiogenic shock put on VA-
ECMO. Moreover, none of the above mentioned cardiac arrest. Our cohort enrolled 45 cases of LV
decompression, by direct surgical LV venting, upstream LA drainage either by sternotomy, or peripheral
trans-septal approach. In addition, these 45 patients were even more complicated, as they suffered from
cardiac arrests and ensuing resuscitation. Our patients of both OHCA and IHCA could have more survival
probability, if modied LVAD were appropriately indicated.
Study limitations

Page 14/17
This is a prospective cohort study, and inheriting the drawback of selection bias was not inevitable.
However, nearly all items of demographic were evenly distributed, which should offset the inuences
derived from shortcomings of a non-randomized design.
Conclusions
Our prospective cohort, enrolling both OHCA and IHCA patients put on ECPR, implied that fair outcomes
could be achieved, as long as high quality CPR could be persistently applied, regardless of how long the
low-ow time was. Peak serum lactate and LVEF were proved again to be closely correlated to outcomes.
Moreover, the procedure of LV decompression could render further survival benet to those with poorer LV
function.
List Of Abbreviations
ECPR
extra-corporeal cardiopulmonary resuscitation
OHCA
out-of-hospital cardiac arrest
IHCA
in-hospital cardiac arrest
ECMO
extra-corporeal membranous oxygenation
LVAD
left ventricular assist device
LVEF
left ventricular ejection fraction
Declarations
Ethics approval and consent to participate:All procedures performed in studies involving human
participants were in accordance with the ethical standards of the institutional and/or national research
committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical
standards.This prospective cohort was approved and scrutinized by theInstitutional Review Board of
Far-Eastern Memorial Hospital, New Taipei City, Taiwan.
The concept of patient involvement was incorporated in the study design, especially in the process of
enrollment. Informed consent was the basic requirement. The pros and cons of the intended treatment
were well discussed with the patient or the patient’s legitimate delegates, using laypersons’ language to
avoid the gap of information. We then revisited our aims and revised the research plans to align them
better with things that patients cared about the most. The patient or his/her legitimate delegates were all
well informed and allowed to express their thoughts during the treatment and study process.

Page 15/17
Consent for publication:All authors have provided consent for publication of the manuscript.
Availability of data and materials: all information as available by contacting the correspondent author Dr.
Hsin HT. by email
Competing interests: none to declare
Funding: none
Authors' contributions:H-TH, conducted the study, analysis andnished the manuscript drafting. K-MC
organized the team and designed the study protocol, Y-CW nished the data collection, others performed
the procedures of ECMO/modied LVAD, dedicated care and offered draft suggestions.
Acknowledgement:not applicable
References
1. Hadaya J, Dobaria V, Aguayo E, et al. National trends in utilization and outcomes of extracorporeal
support for in- and out-of-hospital cardiac arrest. Resuscitation. 2020;151:181–8.
2. Extracorporeal Life Support Organization. ELSO Guidelines for Cardiopulmonary Extracorporeal Life
Support. Version 1.4.. Extracorporeal Life Support Organization (Ann Arbor, MI); 2017.
3. Panchal A, Berg K, Hirsch K, et al. 2019 American heart association focused update on advanced
cardiovascular life support: use of advanced airways, vasopressors, and extracorporeal
cardiopulmonary resuscitation during cardiac arrest: an update to the American Heart Association
guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation.
2019;140:e881 –94.
4. Meani P, Gelsomino S, Natour E, et al. Modalities and Effects of Left Ventricle Unloading on
Extracorporeal Life support: a Review of the Current Literature. Eur J Heart Fail. 2017;19:84–91.
5. Thiele H, Ohman EM, Waha-Thiele S, Zeymer W, Desch S. Management of cardiogenic shock
complicating myocardial infarction: an update 2019. Eu Heart J. 2019;40:2671–83.
. Konstantinides SV, Meyer G, Becattini C, et al. The Task Force for the diagnosis and management of
acute pulmonary embolism of the European Society of Cardiology (ESC). 2019 ESC Guidelines for
the diagnosis and management of acute pulmonary embolism developed in collaboration with the
European Respiratory Society (ERS): The Task Force for the diagnosis and management of acute
pulmonary embolism of the European Society of Cardiology (ESC).Eur Respir J. 2019;54:1901647.
doi: 10.1183/13993003.01647-2019.
7. Wang CC, Wu CK, Tsai ML, et al. 2019 Focused Update of the Guidelines of the Taiwan Society of
Cardiology for the Diagnosis and Treatment of Heart Failure. Acta Cardiol Sin. 2019;35:244–83.
. Judd W. Landsberg MD. Heart failure for the pulmonary critical care physician. Manual for
Pulmonary and Critical Care Medicine; 2018.

Page 16/17
9. Bleeker GB, Steendijk P, Holman ER, et al. Assessing right ventricular function: the role of
echocardiography and complementary technologies. Heart. 2006;92(Suppl 1):i19–26.
10. . Stecker EC, Reinier K, Marijon E, et al. Public health burden of sudden cardiac death in the United
States. Circ Arrhythm Electrophysiol. 2014;7:212–7.
11. Thomas AC, Knapman PA, Krikler DM, Davies MJ. Community study of the causes of “natural”
sudden death. BMJ. 1988;297:1453–6.
12. Leach IH, Blundell JW, Rowley JM, Turner DR. Acute ischaemic lesions in death due to ischaemic
heart disease: an autopsy study of 333 cases of out-of-hospital death. Eur Heart J. 1995;16:1181–5.
13. Yan S, Gan Y, Jiang N, et al. The global survival rate among adult out-of-hospital cardiac arrest
patients who received cardiopulmonary resuscitation: a systematic review and meta-analysis. Crit
Care. 2020;24:61. doi:10.1186/s13054-020-2773-2.
14. D’Arrigo S, Cacciola S, Dennis M, et al. Predictors of favourable outcome after in-hospital cardiac
arrest treated with extracorporeal cardiopulmonary resuscitation: a systematic review and meta-
analysis. Resuscitation. 2017;121:62–70.
15. Casserly B, Phillips GS, Schorr C, et al. Lactate measurements in sepsis-induced tissue
hypoperfusion: results from the Surviving Sepsis Campaign database. Crit Care Med. 2015;43:567–
73.
1. Nolan JP, Sandroni C, Böttiger BW, et al. European Resuscitation Council and European Society of
Intensive Care Medicine guidelines 2021: post-resuscitation care. Intensive Care Med. 2021;25:1–53.
17. Chen YS, Lin JW, Yu HY, et al. Cardiopulmonary resuscitation with assisted extracorporeal life
support versus conventional cardiopulmonary resuscitation in adults with in-hospital cardiac arrest:
an observational study and propensity analysis. Lancet. 2008;372:554–61.
1. Centofanti P, Attisani M, La Torre M, et al. Left Ventricular Unloading during Peripheral Extracorporeal
Membrane Oxygenator Support: A Bridge To Life In Profound Cardiogenic Shock. J Extra Corpor
Technol. 2017;49:201–5.
19. Eastaugh LJ, Thiagarajan RR, Darst JR, et al. Percutaneous left atrial decompression in patients
supported with extracorporeal membrane oxygenation for cardiac disease. Pediatr Crit Care Med.
2015;16:59–65.
20. Hacking DF, Best D, d'Udekem Y, et al. Elective decompression of the left ventricle in pediatric patients
may reduce the duration of venoarterial extracorporeal membrane oxygenation. Artif Organs.
2015;39:319–26.
Figures

Page 17/17
Figure 1
The Kaplan-Meyer survival curve of VA-ECMO vs. modied LVAD showed signicant survival benet of
modied LVAD.
Figure 2
2a The low-ow time was not correlated with peak serum lactate 2b The low-ow time was not correlated
with left ventricular ejection fraction during mechanical support