Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
© Annals of Cardiothoracic Surgery. All rights reserved. Ann Cardiothorac Surg 2021;10(3):353-363 | http://dx.doi.org/10.21037/acs-2020-cfmcs-251
Impact of extra-corporeal life support (ECLS) cannulation strategy
on outcome after durable mechanical circulation support system
implantation on behalf of durable MCS after ECLS Study Group
Diyar Saeed1,2, Evgenij Potapov3,4, Antonio Loforte5, Michiel Morshuis6, David Schibilsky7,
Daniel Zimpfer8, Julia Riebandt8, Federico Pappalardo9, Matteo Attisani10, Mauro Rinaldi10,
Davide Pacini5, Assad Haneya11, Faiz Ramjankhan12, Dirk W. Donker12, Ulrich P. Jorde13, Wolfgang Otto1,
Julia Stein3, Dmytro Tsyganenko3, Ameen Al-Naamani1, Radi Wieloch2, Rafael Ayala7, Jochen Cremer11,
Michael Borger1, Artur Lichtenberg2, Jan Gummert6
1Department of Cardiac surgery, Leipzig Heart Center, Leipzig, Germany; 2Department for Cardiac Surgery, Duesseldorf University Hospital,
Duesseldorf, Germany; 3Department of Cardiac Surgery, German Heart Center Berlin, Berlin, Germany; 4DZHK (German Center for
Cardiovascular Research), Partner Site Berlin, Berlin, Germany; 5Division of Cardiac Surgery, IRCCS Azienda Ospedaliero-Universitaria di Bologna,
Bologna, Italy; 6Department of Cardiovascular and Thoracic Surgery, Heart and diabetes Center NRW, Bad Oeynhausen, Germany; 7Department
of Cardiac and Vascular Surgery, Freiburg University, Freiburg, Germany; 8Department of Cardiac Surgery, Medical University Vienna, Vienna,
Austria; 9Advanced Heart Failure and Mechanical Circulatory Support Program, San Raffaele Hospital, Vita Salute University, Milan, Italy;
10Department of Cardiac Surgery, University of Turin, Turin, Italy; 11Department of Cardiac Surgery, University Hospital Schleswig Holstein,
Campus Kiel, Kiel, Germany; 12Department of Cardiothoracic Surgery, University Medical Center Utrecht, Utrecht, The Netherland; 13Department
of Medicine, Monteore Medical Center, Bronx, NY, USA
Correspondence to: Diyar Saeed, MD, PhD. University Department for Cardiac Surgery, Leipzig Heart Center, Strümpellstr. 39, 04289 Leipzig,
Germany. Email: diyar.saeed@helios-gesundheit.de.
Background: The literature on outcomes of patients requiring durable mechanical circulatory support
(MCS) after extra-corporeal life support (ECLS) is limited. The aim of this study was to investigate the
impact of preoperative ECLS cannulation on postoperative outcome after durable MCS implantation.
Methods: The durable MCS after ECLS registry is a multicenter retrospective study that gathered data
on consecutive patients who underwent durable MCS implantation after ECLS between January 2010 and
August 2018 in eleven high volume European centers. Patients who underwent the implantation of total
articial heart, pulsatile pumps, or rst-generation pumps after ECLS were excluded from the analysis. The
remaining patients were divided into two groups; central ECLS group (cECLS) and peripheral ECLS group
(pECLS). A 1:1 propensity score analysis was performed to identify two matched groups. The outcome of
these two groups was compared.
Results: A total of 531 durable MCS after ECLS were implanted during this period. The ECLS
cannulation site was peripheral in 87% (n=462) and central in 13% (n=69) of the patients. After excluding
pulsatile pumps and total articial heart patients, a total of 494 patients remained (pECLS =434 patients,
cECLS =60 patients). A 1:1 propensity score analysis resulted in 2 matched groups (each 55 patients)
with median age of 54 years (48–60 years) in cECLS group and 54 years (43–60 years) in pECLS group.
HeartWare HVAD (Medtronic, Minneapolis, MN) was implanted in the majority of the patients (cECLS
=71% vs. pECLS =76%, P=0.67). All postoperative morbidities were comparable between the groups. The
thirty-day, one year and long-term survival was comparable between the groups (P=0.73).
Conclusions: The cannulation strategy of ECLS appears to have no impact on the post-operative outcome
after durable MCS implantation.
Keywords: Mechanical circulatory support (MCS); extra-corporeal life support (ECLS); cannulation strategy;
outcome; ventricular assist device
Featured Article
363
354 Saeed et al. ECLS cannulation strategy and outcome after VAD support
© Annals of Cardiothoracic Surgery. All rights reserved. Ann Cardiothorac Surg 2021;10(3):353-363 | http://dx.doi.org/10.21037/acs-2020-cfmcs-251
Introduction
Mechanical circulatory support (MCS) systems, in
particular, extra-corporeal life support (ECLS), have become
a widely accepted therapy option for patients in cardiogenic
shock (1). The application of ECLS for immediate
resuscitation with the potential of end organ function
recovery is a useful strategy to improve survival of this
otherwise extremely sick patient population. After ECLS
implantation, the primary aim remains patient stabilization
and weaning from ECLS. However, some patients take a
different path and either expire on ECLS due to multiple
organ failure and/or neurological deficit, or need to be
considered for a durable MCS. The decision to implant a
durable MCS in a patient with ECLS is very challenging
and the implantation threshold may vary between
institutions (2,3). Our group recently published the largest
series on durable MCS patients bridged with ECLS (4).
We were able to identify several survival predictors. An app
“Durable MCS after ECLS calculator” was introduced that
aids future patient selection and helps to avoid unnecessary
resource utilization (4).
The implantation of ECLS is usually on an emergency
basis for cardiogenic shock patients. The implantation
may be performed using either a central approach with
the oxygenated blood from ECLS returning to the
aorta/subclavian artery or a peripheral approach with
oxygenated blood returned through the femoral artery.
There are several studies investigating the impact of ECLS
cannulation approach on various post implant morbidities
and outcome after ECLS implantation (5,6) However, we
are not aware of any study that specically investigates the
impact of preoperative (prior to durable MCS implantation)
ECLS cannulation strategy on postoperative morbidities
and outcome after durable MCS implantation. Therefore,
the primary objective of this study was to investigate
whether preoperative ECLS cannulation strategy has any
impact on the outcome after durable MCS implantation
using data from the durable MCS after ECLS registry. We
hypothesize that the more physiological form of support,
central ECLS, may be associated with lower postoperative
morbidities after durable MCS implantation.
Methods
Patient population
The durable MCS after ECLS registry is a multicenter
retrospective study that gathered data on consecutive
patients who underwent durable MCS implantation directly
after ECLS between January 2010 and August 2018 in
eleven high volume European centers. The primary aim
after ECLS implantation was to wean the patient off
mechanical support. Patients who did not meet the weaning
criteria were considered for durable MCS after adequate
neurological evaluation. There was no specific protocol
when and how to proceed with durable MCS therapy. All
perioperative data and postoperative complications were
considered. The Interagency Registry for Mechanically
Assisted Circulatory Support (INTERMACS) definitions
were used for postoperative complications except for right
ventricular failure, which was considered only if mechanical
support of the right ventricle was necessary.
For this study, all patients who underwent the
implantation of total artificial heart, pulsatile pumps, or
earlier generation pumps after ECLS were excluded from
the analysis. Therefore, only patients who were supported
with HeartWare HVAD (Medtronic, Minneapolis, MN),
HeartMate II (Abbott, Abbott Park, Ill) or HeartMate III
(Abbott, Abbott Park, Ill) were included. The patients were
divided into two groups; central ECLS group (cECLS)
and peripheral ECLS group (pECLS). A 1:1 propensity
score analysis was then performed to identify two matched
groups. The outcome of these two groups was analyzed
and compared. The study protocol was approved by the
Individual Health Research Ethics Boards.
Surgical techniques
ECLS implantation
The ECLS implantation at each institution was performed
on an emergency basis in cardiogenic shock patients
for various reasons. Peripheral ECLS approach was
predominantly performed through femoral vein and
arteries. Distal leg perfusion cannula was exclusively used in
Submitted Dec 21, 2020. Accepted for publication Mar 29, 2021.
doi: 10.21037/acs-2020-cfmcs-251
View this article at: http://dx.doi.org/10.21037/acs-2020-cfmcs-251
355
Annals of cardiothoracic surgery, Vol 10, No 3 May 2021
© Annals of Cardiothoracic Surgery. All rights reserved. Ann Cardiothorac Surg 2021;10(3):353-363 | http://dx.doi.org/10.21037/acs-2020-cfmcs-251
pECLS patients. The cannulation was either percutaneous
or through surgical cut down. The arterial cannulation
strategy for central ECLS approach was either through
direct cannulation of the aorta, graft anastomosis to the
aorta, or graft anastomosis to the subclavian artery. The
venous cannulation of central ECLS was predominately
percutaneous using the femoral vein.
Durable MCS implantation
The durable MCS was implanted either in a standard
fashion through a median sternotomy or using less
invasive techniques. The outflow graft was connected to
the ascending aorta in all cases. Some of the patients were
operated on ECLS. However, the cardiopulmonary bypass
machine was used at the time of implant for other patients,
either due to the necessity of concomitant procedures or
surgeon preference.
Statistical analysis
Continuous study variables were evaluated for both normal
distribution and outlier activity. They were reported as mean
with standard deviation if normally distributed, or as median
with the interquartile range otherwise. For categorical data,
the frequencies are given. Statistical tests were performed
according to type, normality, and scedasticity of data with
Welch two-sample t-test, Wilcoxon signed-rank test, or χ2
test. A 5-fold multiple imputation was applied to address
missing values in risk factors. Patients in the cECLS group
were compared to patients in the pECLS group. As these
two groups were not randomized, a propensity score
analysis was computed with univariate analysis assessing a
predened set of preoperative risk factors. The 1:1 nearest
neighbor propensity score matching with a caliper of 0.2
was applied onto relevant baseline variables stratifying
55 patients into each group (Figure 1). Kaplan-Meier
estimates were calculated to describe overall survival in
the matched pECLS and cECLS groups with the date of
MCS implantation as the starting point. Comparison of
the matched cECLS versus pECLS groups was performed
using log-rank test. For statistical calculations and graphics,
we used R software, Version 4.0.3 (R Core Team 2018. R:
A Language and Environment for Statistical Computing. R
Foundation for Statistical Computing, Vienna, Austria) with
the packages mice, MatchIt and tidyverse.
Results
A total of 531 durable MCS on VA-ECMO with average age
of 53±12 years were implanted between January 2010 and
August 2018. Ischemic cardiomyopathy was the dominant
reason for cardiogenic shock with 300 patients (57%). The
ECLS cannulation was peripheral in 462 (87%) of patients
and central in 69 patients (13%). Up to 173 (33%) patients
had history of cardiopulmonary resuscitation (CPR) prior to
ECLS implantation. After excluding pulsatile pumps, first
generation pumps and total articial heart patients, a total
of 494 patients remained (pECLS =434 patients, cECLS
=60 patients). Table 1 shows the preoperative characteristics
of the patients. The preoperative characteristics were
comparable except for statistically higher body mass index
(BMI) in the cECLS group (BMI ≥30: 35% in cECLS vs.
19% in pECLS, P=0.0079), higher rate of previous cardiac
surgery in the cECLS group (48% vs. 22% in pECLS,
P=0.0001), as well as higher C-reactive protein (CRP) in
the cECLS group {18 [12–30] vs. 13 [6.7–24], P=0.013}.
Further, norepinephrine was more in use in the cECLS
group (58% vs. 44%, P=0.04) and the cardiopulmonary
bypass time was significantly longer in the cECLS group
{120 min [78–180] vs. 99 min [60–130], P=0.0014}.
To adjust for differences between the groups, a 1:1
propensity score analysis was computed with univariate
analysis assessing a predefined set of preoperative risk
factors. The matching resulted in two matched groups
(each 55 patients) based on the baseline characteristics of
the patients (Figure 1). Table 2 shows the characteristics of
the matched groups. The groups were well matched with
no statistically signicant differences between the matched
groups. The cECLS group had median age of 54 years
[48–60 years] vs. 54 years [43–60 years] in the pECLS
group (P=0.67). The median duration of ECLS support
was ve days [two-seven days] in the cECLS group vs. ve
days [three-ten days] in the pECLS group (P=0.39). The
HeartWare HVAD pump was predominantly implanted
in both groups (71% cECLS vs. 76% in pECLS, P=0.67).
Table 3 shows postoperative (after durable MCS
implantation) outcome in the matched group of patients
over the entire duration of follow up. Postoperative
mechanical right ventricular support was necessary in 53%
of the cECLS patients vs. 44% of the pECLS patients
(P=0.45). The amount of chest tube output in the first
twenty-four hours and rate of re-exploration for bleeding
356 Saeed et al. ECLS cannulation strategy and outcome after VAD support
© Annals of Cardiothoracic Surgery. All rights reserved. Ann Cardiothorac Surg 2021;10(3):353-363 | http://dx.doi.org/10.21037/acs-2020-cfmcs-251
were comparable between the groups. All other post
durable MCS related complications were comparable
between the groups. There was a slightly higher rate of
pump thrombosis after durable MCS implantation in the
cECLS group (0.15 events per patient year vs. 0.06 events
per patient year). However, looking at the pump thrombosis
events in the first year of follow up, no statistically
signicant difference was observed (Log-rank test, P=0.82).
Meanwhile, the rate of postoperative stroke was comparable
between the groups (27% in cECLS vs. 25% in pECLS,
P=1.00). The median duration of the durable MCS support
was 320 days [26–970 days] in cECLS group vs. 240 days
Figure 1 Bias reduction plotted as standardized mean differences of baseline variables between the central ECLS and peripheral ECLS
groups. Baseline differences before propensity score matching (blue dots) were effectively balanced after matching (red dots). ECLS, extra-
corporeal life support.
Standardized mean differences
−0.25 0.00 0.25 0.50 0.75
Distance
AST
Epinephrine use
Hb value
Previous cardiac surgery
Corotrop use
International normalized ratio
C-reactive protein
ALT
Bilirubin value
Ischemic cardiomyopathy
Age at implantation
Norepinephrine use
Base excess
Diabetes mellitus
Society of thoracic surgeons (STS) score
Creatinine
Atrial fibrillation
Dialyse
Cardiopulmonary resuscitation prior to ECLS
Model of end stage liver disease (MELD) score
Blood urea nitrogen
Ph
Peripheral vascular disease
ECLS support duration
Model of end stage liver disease (MELD) XI score
Gender: male
WBC count
PreOP IABP
Body mass index ≥30
Platelet count
Lactate level
Covariate balance
357
Annals of cardiothoracic surgery, Vol 10, No 3 May 2021
© Annals of Cardiothoracic Surgery. All rights reserved. Ann Cardiothorac Surg 2021;10(3):353-363 | http://dx.doi.org/10.21037/acs-2020-cfmcs-251
Table 1 Pre- and intraoperative characteristics of the overall population
Parameter Central ECLS (n=60), median [IQR] Peripheral ECLS (n=434), median [IQR] P value
Age (years) 55 [48–60] 54 [47–60] 0.7
Female gender 10 (17%) 82 (19%) 0.81
BMI ≥30 21 (35%) 83 (19%) 0.0079
Diagnosis (ICM) 39 (65%) 245 (56%) 0.26
Atrial fibrillation 19 (32%) 131 (30%) 0.93
Diabetes mellitus 17 (28%) 105 (24%) 0.59
Peripheral vascular disease 5 (8%) 25 (6%) 0.62
Hx of previous cardiac surgery 29 (48%) 94 (22%) 0.0001
STS score 5% [4–7] 5% [4–7] 0.5
CPR prior to ECLS 17 (28%) 140 (32%) 0.64
ECMO support duration (days) 5 [2–8] 5 [3–8] 0.87
Renal replacement therapy on ECMO 17 (28%) 133 (31%) 0.83
IABP + ECMO 19 (32%) 91 (21%) 0.089
Laboratory parameters
Creatinine (mg/dL) 1.3 [0.77–1.7] 1.2 [0.81–1.8] 0.7
Blood urea nitrogen (mg/dL) 61 [36–100] 58 [34–88] 0.47
AST (U/L) 110 [48–180] 93 [48–250] 0.93
ALT (U/L) 86 [41–230] 96 [39–360] 0.39
Serum bilirubin (mg/dL) 1.7 [1.2–4.3] 1.7 [0.99–3.7] 0.64
INR 1.3 [1.1–1.6] 1.3 [1.1–1.6] 0.8
MELD score 16 [13–25] 18 [11–25] 0.96
MELD score XI 18 [13–27] 19 [12–28] 0.95
Hb value (mg/dL) 9.5 [8.8–10] 9.5 [8.8–10] 0.79
WBC count (103/µL) 12 [9.7–15] 11 [8.7–15] 0.27
Platelets count (103/µL) 78 [54–120] 86 [59–120] 0.46
CRP (mg/dL) 18 [12–30] 13 [6.7–24] 0.013
Lactate value (mg/dL) 1.1 [0.79–1.6] 1.1 [0.7–1.6] 0.79
PH value 7.4 [7.3–7.4] 7.4 [7.3–7.5] 0.21
BE (mmol/L) 1.9 [−1.2 to 3.2] 1.4 [−1.9 to 4.1] 0.62
Catecholamine use on ECLS
Norepinephrine use 35 (58%) 189 (44%) 0.04
Epinephrine use 31 (52%) 186 (43%) 0.25
Milrinone use 13 (22%) 64 (15%) 0.23
Table 1 (continued)
358 Saeed et al. ECLS cannulation strategy and outcome after VAD support
© Annals of Cardiothoracic Surgery. All rights reserved. Ann Cardiothorac Surg 2021;10(3):353-363 | http://dx.doi.org/10.21037/acs-2020-cfmcs-251
Table 1 (continued)
Parameter Central ECLS (n=60), median [IQR] Peripheral ECLS (n=434), median [IQR] P value
Intraoperative parameters
CPB used for VAD implantation 41 (68%) 247 (57%) 0.12
Less Invasive VAD implantation 2 (3%) 40 (9%) 0.2
Total surgery time (min) 250 [190–340] 240 [190–320] 0.7
CPB time (min) 120 [78–180] 99 [60–130] 0.001
Concomitant procedures + VAD 13 (22%) 81 (19%) 0.7
VAD type: HeartMate II 10 (17%) 69 (16%) 1.00
VAD type: HeartMate III 6 (10%) 39 (9%) 0.94
VAD type: HeartWare HVAD 44 (73%) 326 (75%) 0.92
BMI, body mass index; ICM, ischemic cardiomyopathy; CPR, cardiopulmonary resuscitation; ECLS, extra-corporeal life support; VAD,
ventricular assist device; IABP, intra-aortic balloon pump; INR, international normalized ratio; MELD, model of end stage liver disease;
WBC, white blood cell; CRP, C-reactive protein; BE, base excess; STS score, Society of Thoracic Surgery score.
[24–840 days] in the pECLS group (P=0.83). During the
follow up, heart transplantation was performed in 20% of
the cECLS group vs. 24% in the pECLS group (P=0.82).
Further, up to 60% of the cECLS patients expired while
on durable MCS vs. 55% of the pECLS patients (P=0.7).
Figure 2 shows the Kaplan-Meier survival curve of the
matched groups. There were no statistically significant
differences in the short- and long-term outcomes between
the groups (Log rank test, P=0.73). The thirty-day and
one-year mortality of the matched groups were 29% in
cECLS group vs. 27% in pECLS group (P=1.00) and
49% in cECLS group vs. 49% in pECLS group (P=1.00)
respectively. No major differences in the outcome between
the three types of pumps in the matched groups were
observed. The thirty-day and one-year survival rates in
the 39 matched patients in cECLS group supported with
HeartWare HVAD were 72% and 49% respectively. This
was comparable with the thirty-day and one-year survival
rates of 70% and 60% in the 10 matched cECLS group
supported with HeartMate II as well as the thirty-day and
one-year survival rates of 67% and 67% in the 6 matched
cECLS patients supported with HeartMate III pump.
Discussion
The recent years have witnessed a widespread use of
ECLS systems for patients in cardiogenic shock (7,8). In
this multicentre study, we performed a sub analysis of the
Durable MCS after ECLS registry data to specifically
investigate whether the preoperative ECLS implantation
strategy has any impact on the outcome following durable
MCS implantation. The ndings of this study show that the
outcome is comparable after durable MCS implantation in
matched groups of patients regardless of the preoperative
ECLS cannulation strategy. The postoperative MCS related
morbidities of the matched groups were also comparable.
Durable MCS implantation and/or heart transplantation
remains the only therapy option in otherwise viable
patients on ECLS who fail to show adequate recovery
of the ventricular function in absence of neurological
deficits. Even though the new US allocation system now
prioritizes patients waiting on ECLS, outcome after heart
transplantation on ECLS remains poor (9,10). Furthermore,
considering issues of organ shortage and resulting extended
waiting times in Europe for heart transplantation, timely
transplantation may not be a realistic therapy option for
patients on ECLS. Therefore, at least in Europe, durable
MCS therapy remains the main treatment option for this
patient population. In an effort to determine “the point of
no return” in this patient population on ECLS, our group
established the Durable MCS after ECLS registry, which
includes data from 531 patients who underwent durable
MCS implantation after ECLS support (4). In the first
analysis of the registry data, we were able to show that the
overall survival is very limited in this patient population
and not comparable to the outcome in “traditional” MCS
359
Annals of cardiothoracic surgery, Vol 10, No 3 May 2021
© Annals of Cardiothoracic Surgery. All rights reserved. Ann Cardiothorac Surg 2021;10(3):353-363 | http://dx.doi.org/10.21037/acs-2020-cfmcs-251
Table 2 Pre- and intraoperative characteristics of the patients after 1:1 propensity score matching
Parameter Central ECLS (n=55), median [IQR] Peripheral ECLS (n=55), median [IQR] P value
Age (years) 54 [48–60] 54 [43–60] 0.67
Female gender 10 (18%) 8 (15%) 0.80
BMI ≥30 17 (31%) 22 (40%) 0.43
Diagnosis (ICM) 35 (64%) 33 (60%) 0.84
Atrial fibrillation 15 (27%) 16 (29%) 1.00
Diabetes mellitus 15 (27%) 15 (27%) 1.00
Peripheral vascular disease 5 (9%) 6 (11%) 1.00
Hx of previous cardiac surgery 24 (44%) 20 (36%) 0.56
STS score 5% [4–7] 7% [4–8] 0.77
CPR prior to ECLS 16 (29%) 17 (31%) 1.00
ECMO support duration (days) 5 [2–7] 5 [3–10] 0.39
Renal replacement therapy on ECLS 16 (29%) 17 (31%) 1.00
IABP + ECLS 16 (29%) 20 (36%) 0.54
Laboratory parameters
Creatinine (mg/dL) 1.4 [0.75–1.8] 1.2 [0.86–1.7] 0.73
Blood urea nitrogen (mg/dL) 63 [37–96] 58 [36–100] 0.88
AST (U/L) 110 [48–180] 97 [52–180] 0.73
ALT (U/L) 83 [40–220] 90 [38–170] 0.96
Serum bilirubin (mg/dL) 1.7 [1.1–4.3] 1.7 [1–3.7] 0.49
INR 1.3 [1.1–1.6] 1.3 [1.2–1.5] 0.59
MELD score 17 [12–24] 18 [12–25] 0.77
MELD score XI 18 [13–28] 20 [14–28] 0.72
Hb value (mg/dL) 9.5 [8.8–10] 9.3 [8.6–10] 0.37
WBC count (103/µL) 12 [9.6–15] 12 [10–16] 0.70
Platelets count (103/µL) 79 [56–120] 94 [60–140] 0.28
CRP (mg/dL) 16 [9.4–31] 15 [8.4–24] 0.34
Lactate value (mg/dL) 1.1 [0.79–1.6] 1.1 [0.68–1.6] 0.96
PH value 7.4 [7.3–7.4] 7.4 [7.3–7.4] 0.86
BE (mmol/L) 1.9 [−1.2 to 3.1] 0.4 [−2.4 to 3.6] 0.33
Catecholamine use
Norepinephrine use 31 (56%) 30 (55%) 1.00
Epinephrine use 28 (51%) 23 (42%) 0.44
Milrinone use 12 (22%) 9 (16%) 0.63
Table 2 (continued)
360 Saeed et al. ECLS cannulation strategy and outcome after VAD support
© Annals of Cardiothoracic Surgery. All rights reserved. Ann Cardiothorac Surg 2021;10(3):353-363 | http://dx.doi.org/10.21037/acs-2020-cfmcs-251
candidates. The postoperative morbidities, manifested
as increased right ventricular failure and higher rate of
postoperative complications, were also higher than the
morbidities following durable MCS implantation in patients
without ECLS support. Furthermore, we were able to
nd several predictors of one-year mortality in this patient
Table 2 (continued)
Parameter Central ECLS (n=55), median [IQR] Peripheral ECLS (n=55), median [IQR] P value
Intraoperative parameters
CPB used for VAD implantation 38 (69%) 39 (71%) 1.00
Less invasive VAD implantation 2 (4%) 3 (5%) 1.00
Total surgery time (min) 250 [190–340] 250 [210–330] 0.67
CPB time (min) 120 [78–180] 100 [85–140] 0.13
Concomitant procedures + VAD 13 (24%) 18 (33%) 0.40
VAD type: HeartMate II 10 (18%) 10 (18%) 1.00
VAD type: HeartMate III 6 (11%) 3 (5%) 0.49
VAD type: HeartWare HVAD 39 (71%) 42 (76%) 0.67
BMI, body mass index; ICM, ischemic cardiomyopathy; CPR, cardiopulmonary resuscitation; ECLS, extra-corporeal life support; VAD,
ventricular assist device; IABP, intra-aortic balloon pump; INR, international normalized ratio; MELD, model of end stage liver disease;
WBC, white blood cell; CRP, C-reactive protein; BE, base excess; STS score, Society of Thoracic Surgery score.
Table 3 Postoperative outcome of matched groups
Parameter Central ECLS (n=55), median [IQR] Peripheral ECLS (n=55), median [IQR] P value
Chest tube output in 24 h (mL) 1,000 [600–1,700] 1,000 [780–1,400] 0.96
No. of RBC units 7 [4–12] 8 [4.5–10] 0.73
No. of FFP units 6 [3.5–10] 6 [2.5–10] 0.97
No. of platelet units 4 [2–5.5] 3 [2–4] 0.29
Re-exploration rate for bleeding 25 (45%) 23 (42%) 0.85
Mechanical RV support for RVF 29 (53%) 24 (44%) 0.45
Postoperative reparatory failure 37 (67%) 33 (60%) 0.55
Postoperative liver failure 25 (45%) 24 (44%) 1.00
Postoperative renal failure 32 (58%) 30 (55%) 0.85
Postoperative stroke 15 (27%) 14 (25%) 1.00
Pump thrombosis 13 (24%) 5 (9%) 0.07
GI bleeding 7 (13%) 11 (20%) 0.44
Driveline infection 14 (25%) 8 (15%) 0.23
Durable MCS support duration (days) 320 [26–970] 240 [24–840] 0.83
RBC, red blood cells; FFP, fresh frozen plasma; RV, right ventricle; RVF, right ventricular failure; GI, gastro-intestinal; MCS, mechanical
circulatory support.
361
Annals of cardiothoracic surgery, Vol 10, No 3 May 2021
© Annals of Cardiothoracic Surgery. All rights reserved. Ann Cardiothorac Surg 2021;10(3):353-363 | http://dx.doi.org/10.21037/acs-2020-cfmcs-251
population including; age, female gender, lactate value,
high MELD XI score, history of atrial brillation, history
of previous cardiac surgery and BMI ≥30. On the basis
of this data, a formula to estimate one-year survival after
durable MCS implantation and an app (durable MCS after
ECLS calculator) were created to facilitate future candidate
selection (4). The app may be downloaded for free from the
Apple Store or the Google Play Store.
Patients requiring ECLS support may be cannulated
using a femoro-femoral approach; the so called peripheral
ECLS technique, or using aorta /subclavian artery for
returning the oxygenated blood from ECLS circuit; the
so called central ECLS approach. In patients with cECLS
and aortic cannulation, direct cannulation of the aorta may
be performed. Alternatively, a graft is anastomosed to the
aorta and tunnelled allowing sternum closure (6). Each of
these two cannulation strategies has its own advantages and
disadvantages. Some centers prefer the central cannulation
approach with the advantage of having direct flow stream
from the outflow cannula into the aorta and the arch vessels,
coronaries and the rest of the body. Other centers prefer
the peripheral cannulation technique as it is faster, can be
performed at the bed side, and is less invasive compared
to the central cannulation technique. However, higher
peripheral vascular complications have been reported in
pECLS patients (11). There are many studies investigating
the differences in outcome and post implantation
complications in patients undergoing central or peripheral
ECLS implantation. In a different study, the outcome
of these two cannulation techniques was investigated in
37 patients (12). In that study, no particular oxygenation/
ventilation, hemodynamic, or end organ function advantage
was observed with either cannulation technique. However,
more bleeding and resternotomy complications were
observed in cECLS patients. In another multicentre study of
postcardiotomy patients supported with ECLS, Mariscalco
et al. analysed the outcome of 781 patients receiving ECLS
for postcardiotomy cardiogenic shock (5). The ndings of
that study showed higher in-hospital mortality in cECLS
patients as well as higher reoperation rate for bleeding/
tamponade and blood transfusion requirements. All of the
published studies so far questioned the impact of ECLS
cannulation strategy on the outcome (5,11,12). However,
none of the published studies specifically investigate the
Figure 2 The Kaplan-Meier survival curves after durable MCS implantation in patients on ECLS. The thirty-day and one-year mortality
of the matched groups were 29% in central ECLS group vs. 27% in peripheral ECLS group, and 49% in central ECLS group vs. 49% in
peripheral ECLS group respectively (P=1.00 each). The short and long-term survival rates were comparable between the matched groups (Log
rank test, P=0.73). ECLS, extra-corporeal life support; MCS, mechanical circulatory support.
0 1 2 3 4 5
Central 55 (100) 28 (51) 22 (40) 15 (27) 9 (16) 7 (13)
Peripheral 55 (100) 28 (51) 23 (42) 18 (33) 13 (24) 11 (20)
0 1 2 3 4 5
Time [Years]
Time [Years]
Number at risk: n (%)
Strata
(log-rank test: P=0.73)
Strata Central peripheral
100%
75%
50%
25%
0%
Survival rate
362 Saeed et al. ECLS cannulation strategy and outcome after VAD support
© Annals of Cardiothoracic Surgery. All rights reserved. Ann Cardiothorac Surg 2021;10(3):353-363 | http://dx.doi.org/10.21037/acs-2020-cfmcs-251
impact of the preoperative ECLS cannulation strategy
on the outcome following durable MCS implantation in
patients on ECLS. The ndings of this study show that the
preoperative ECLS cannulation strategy has no impact on
the postoperative outcome after durable MCS implantation.
In contrary to our hypothesis, the more physiological form
(cECLS strategy) showed no advantage with regard to
postoperative morbidity and mortality.
The main limitation of this study is its retrospective
nature. However, data in many centers were prospectively
collected and entered in a corresponding data bank.
Additionally, this was not a randomized study and despite
matching our ECLS groups using several covariates, there
may still be confounding by indication. Nevertheless, as
shown in Figure 1, the groups were well matched and we
therefore believe that the study outcome is representative.
However, as the peripheral cannulation strategy dominated
the study population, it may potentially introduce a hidden
exclusion bias.
In conclusion, this study showed that the outcome after
durable MCS implantation in matched groups of patients
bridged with ECLS is similar regardless of the preoperative
ECLS cannulation strategy. The implications of this
study may discourage some clinicians from changing a
well-functioning peripheral or central ECLS cannulation
approach based on the incorrect notion that one approach
may be superior to the other. Future research may need to
focus on determining whether left ventricular pressures and
risk of lung oedema in ECLS vary based on cannulation
strategy.
Acknowledgments
Funding: None.
Footnote
Conicts of Interest: The authors have no conflicts of interest
to declare.
Open Access Statement: This is an Open Access article
distributed in accordance with the Creative Commons
Attribution-NonCommercial-NoDerivs 4.0 International
License (CC BY-NC-ND 4.0), which permits the non-
commercial replication and distribution of the article with
the strict proviso that no changes or edits are made and the
original work is properly cited (including links to both the
formal publication through the relevant DOI and the license).
See: https://creativecommons.org/licenses/by-nc-nd/4.0/.
References
1. Marasco SF, Lukas G, McDonald M, et al. Review of
ECMO (extra corporeal membrane oxygenation) support
in critically ill adult patients. Heart Lung Circ 2008;17
Suppl 4:S41-7.
2. Durinka JB, Bogar LJ, Hirose H, et al. End-organ recovery
is key to success for extracorporeal membrane oxygenation
as a bridge to implantable left ventricular assist device.
ASAIO J 2014;60:189-92.
3. Maxhera B, Albert A, Ansari E, et al. Survival predictors in
ventricular assist device patients with prior extracorporeal
life support: selecting appropriate candidates. Artif Organs
2014;38:727-32.
4. Saeed D, Potapov E, Loforte A, et al. Transition From
Temporary to Durable Circulatory Support Systems. J Am
Coll Cardiol 2020;76:2956-64.
5. Mariscalco G, Salsano A, Fiore A, et al. Peripheral
versus central extracorporeal membrane oxygenation for
postcardiotomy shock: Multicenter registry, systematic
review, and meta-analysis. J Thorac Cardiovasc Surg
2020;160:1207-1216.e44.
6. Saeed D, Maxhera B, Westenfeld R, et al. An Alternative
Approach for Perioperative Extracorporeal Life Support
Implantation. Artif Organs 2015;39:719-23.
7. Aubin H, Petrov G, Dalyanoglu H, et al. Four-year
experience of providing mobile extracorporeal life support
to out-of-center patients within a suprainstitutional
network-Outcome of 160 consecutively treated patients.
Resuscitation 2017;121:151-7.
8. Aubin H, Petrov G, Dalyanoglu H, et al. A
Suprainstitutional Network for Remote Extracorporeal
Life Support: A Retrospective Cohort Study. JACC Heart
Fail 2016;4:698-708.
9. Barge-Caballero E, Almenar-Bonet L, Gonzalez-
Vilchez F, et al. Clinical outcomes of temporary
mechanical circulatory support as a direct bridge to heart
transplantation: a nationwide Spanish registry. Eur J Heart
Fail 2018;20:178-86.
10. Fukuhara S, Takeda K, Kurlansky PA, et al. Extracorporeal
membrane oxygenation as a direct bridge to heart
transplantation in adults. J Thorac Cardiovasc Surg
363
Annals of cardiothoracic surgery, Vol 10, No 3 May 2021
© Annals of Cardiothoracic Surgery. All rights reserved. Ann Cardiothorac Surg 2021;10(3):353-363 | http://dx.doi.org/10.21037/acs-2020-cfmcs-251
Cite this article as: Saeed D, Potapov E, Loforte A,
Morshuis M, Schibilsky D, Zimpfer D, Riebandt J, Pappalardo F,
Attisani M, Rinaldi M, Pacini D, Haneya A, Ramjankhan F,
Donker DW, Jorde UP, Otto W, Stein J, Tsyganenko D,
Al-Naamani A, Wieloch R, Ayala R, Cremer J, Borger M,
Lichtenberg A, Gummert J. Impact of extra-corporeal life
support (ECLS) cannulation strategy on outcome after durable
mechanical circulation support system implantation on behalf
of durable MCS after ECLS Study Group. Ann Cardiothorac
Surg 2021;10(3):353-363. doi: 10.21037/acs-2020-cfmcs-251
2018;155:1607-1618.e6.
11. Ranney DN, Benrashid E, Meza JM, et al. Central
Cannulation as a Viable Alternative to Peripheral
Cannulation in Extracorporeal Membrane Oxygenation.
Semin Thorac Cardiovasc Surg 2017;29:188-95.
12. Saeed D, Stosik H, Islamovic M, et al. Femoro-femoral
versus atrio-aortic extracorporeal membrane oxygenation:
selecting the ideal cannulation technique. Artif Organs
2014;38:549-55.
