Impact of Mechanical

C i rc u l a t o r y Su p p o r t on
Posttransplant Outcomes
Todd F. Dardas, MD, MS

KEYWORDS
 Heart transplant  Mechanical circulatory support  Temporary circulatory support  Artificial heart
 Survival  Risk factors  Health utility

KEY POINTS
 Mechanical circulatory support is not associated with reduced posttransplant survival in most
cases.
 Significant reductions in transplant survival may occur following left ventricle assist device (LVAD)
support complicated by infection, total artificial heart, and extracorporeal life support.
 Continuous-flow LVAD support is associated with an increased risk of posttransplant vasoplegia
syndrome.

INTRODUCTION article focuses on posttransplant outcomes spe-

cifically associated with selected MCS devices.
Mechanical circulatory support (MCS) devices
allow for myocardial recovery, serve to stabilize
end-organ function before definitive therapy, act DURABLE DEVICES
as permanent support, provide a bridge to trans- Left Ventricular Assist Devices
plant (BTT), and dramatically increase the likeli-
Each left VAD (LVAD) generation has achieved
hood of surviving cardiogenic shock. When used
lower mortality than prior generations, and high-
as part of a BTT strategy, MCS may extend life
quality prospective trials have established the
to the median survival of heart transplant, which
efficacy of durable LVADs as BTT.3–5 Declining mor-
is currently more than 12 years. However, MCS
tality among LVAD-supported candidates (43%
devices are associated with complications that
mortality in 2005–2006 decreasing to 8% in 2015–
may prevent transplant or reduce posttransplant
2016) and increasing prevalence of MCS at the
survival. For those truly in need of MCS, the poten-
time of listing (increasing from 10% to 35% preva-
tial gain in survival of MCS far outweighs the alter-
lence between 2005–2016) suggest increasing
native of waiting for a transplant unsupported.
experience and improved efficacy.1 Mortality
Accordingly, ventricular assist device (VAD) sup-
for status 1A registrants declined dramatically
port at listing is present among 25% to 40% of
over the same time, and this decline is partly attrib-
candidates in recent series.1,2 Experience with
utable to LVADs used as rescue therapy. Thus,
MCS before a transplant is growing rapidly; as
the use of LVADs as BTT is well accepted; the
more experience accumulates, the advanced
advanced heart failure community is responsible
heart failure community will reduce the risk of mor-
for identifying and managing risk factors contrib-
cardiology.theclinics.com

tality and morbidity attributable to MCS. This

uting to higher rates of posttransplant outcomes.

Disclosure: Research grant (International Society for Heart and Lung Transplantation/Medtronic).
Department of Medicine, University of Washington, 1959 Northeast Pacific Street, Box 356422, Seattle, WA
98195, USA
E-mail address: tdardas@uw.edu

Cardiol Clin 36 (2018) 551–560

https://doi.org/10.1016/j.ccl.2018.06.009
0733-8651/18/Ó 2018 Elsevier Inc. All rights reserved.
552 Dardas

Posttransplant mortality LVADs (85%) and status 1A VAD-supported pa-

Early experience with pulsatile-flow LVADs (PF- tients with complications (78%, P 5 .01). Don-
VADs) as BTT suggested no association between neyong and colleagues15 used a time-dependent
LVADs and posttransplant mortality.6–9 In the PF- Cox model with propensity matching to evaluate
VAD era, mortality was similar between VAD- and the effect of HeartMate II support before transplant
non–VAD-supported recipients after adjusting for compared with no support. When adjusted for
severity of illness, donor-recipient mismatches, donor, recipient, and propensity to receive LVAD
and cold ischemic time.6–9 Consistent with high support, the investigators found that 30-day (HR
rates of infectious complications encountered 1.23; 95% CI 0.79–1.95) and 30- to 365-day (HR
with pulsatile devices available at the time, these 1.31; 95% CI 0.85–2.01) mortality rates were
reports describe a higher risk of first infection higher among the LVAD-supported group, though
and dying of infection after transplant among PF- statistical significance was not met. Higher risk
VAD–supported recipients.8,9 Early comparisons among LVAD-supported recipients was not seen
between pulsatile and continuous-flow (CF) de- in the large, pooled International Society of Heart
vices suggested equality between the two modal- Lung Transplant (ISHLT) adult heart transplant
ities with regard to posttransplant survival.10 Registry analysis (sampled from 2004–2008),
CF devices have proven more reliable with fewer which demonstrated a relative risk (RR) of 1.16
complications and improved survival compared (95% CI RR 0.82–1.65) for CF-VADs compared
with pulsatile devices and, accordingly, have with no inotropes at transplant and an RR of 1.19
replaced pulsatile devices as the mainstay of dura- (95% CI RR 0.84–1.69) when CF-VADs were
ble MCS. In the CF era, mortality has decreased compared with those without inotropes or
for registrants in general and among persons MCS.16 Posttransplant mortality was not signifi-
initially supported with LVADs.1,11 However, cantly increased in a modern ISHLT Registry anal-
increasing survival among registrants obligatorily ysis (2005–2015), which demonstrated only 2%
increases the number of prevalent registrants absolute risk reduction for inotropes when
and exacerbates the supply-demand mismatch compared with pretransplant CF-VAD support.2
of donor hearts, resulting in scrutiny of even small Although there are differing mortality signals
differences in mortality as programs attempt to among LVAD-bridged patients, most data suggest
optimize the utility of each donor. Given the obli- no decrement in survival among recipients trans-
gate need for repeat surgery and high burden of planted from durable LVAD support (Fig. 1).
complications, posttransplant outcomes with du- The exception is LVAD-supported patients with
rable LVADs will remain of great interest. complications. Quader and colleagues17 reported
Several reports address posttransplant mortality 14% posttransplant mortality at 1 year among
among LVAD-supported transplant recipients those recipients with device complications and
when compared with unsupported recipients in 10% mortality among recipients with CF-VADs
the CF-VAD era and found no clinically significant without complications. Although these complica-
difference in posttransplant mortality (Table 1). tions were not further delineated in this investiga-
Weiss and colleagues12 reported no significant in- tion, most reports have found a higher risk of
crease in posttransplant mortality for persons sup- posttransplant mortality among patients with de-
ported with the HeartMate II (Abbott Corp, Abbott vice infections. Concern for recrudescent infec-
Park IL) (multivariable hazard ratio [HR] 1.22; 95% tions among device-supported recipients is
confidence interval [CI] 0.87–1.72). Dardas and justified given the obligate need for posttransplant
colleagues13 reported posttransplant survival for immunosuppression and the high frequency of in-
MCS and non-MCS registrants from the Organ fections encountered during VAD support.
Procurement and Transplantation Network Although pump-related infections may resolve
(OPTN) registry. This report suggests an increase with explant at transplant, resistant organisms
in posttransplant morality among persons with and/or deeply seated infections at relatively
implantable LVADs using elective time (HR 1.2, impenetrable sites may not resolve or recur in
95% CI 0.88–1.3) and among LVADs with compli- the presence of immunosuppression. Reports
cations (HR 1.2, 95% CI 0.99–1.4), though neither from the CF-VAD era indicate variable signals to-
were statistically significant when compared with a ward increased posttransplant mortality following
reference group of recipients supported with dual LVADs complicated by infection. John and col-
inotropes and intra-aortic balloon pumps (IABPs). leagues18 reported a nonsignificant decrease in
Trivedi and colleagues14 noted clinically nonsignif- 1-year survival posttransplant survival between
icant differences in 3-year survival among regis- those persons without a percutaneous lead infec-
trants transplanted from status 1B with an LVAD tion (89%) when compared with those with a
using elective status 1A time (84%), status 1B percutaneous lead infection (75%, P 5 .07). In
 Table 1
Posttransplant survival among continuous-flow ventricular assist devices and comparator groups

Sample Time Follow-up

Author Source Frame Group of Interest Reference Group Time HR 95% CI KM P Value
16
Nativi et al, ISHLT Registry 2004–2008 CF devices Inotropes, no VAD 4y 1.2 0.82–1.65 — .41
2011 No inotropes, no VAD 4y 1.2 0.84–1.69 — .32
Donneyong UNOS 2004–2010 HMII All other recipients 0–30 d 1.2 0.79–1.95 — .36
et al,15 2014 HMII All other recipients 30–365 d 1.3 0.85–2.01 — .22
HMII All other recipients >365 d 0.4 0.36–0.77 — .01
Dardas et al,13 UNOS 2005–2010 CF-VAD listed status Status 1A by dual All available 1.2 0.88–1.3 — .44
2012 1A using elective time inotropes/Swan or IABP

Impact of MCS on Posttransplant Outcomes

CF-VAD listed status Status 1A by dual All available 1.2 0.99–1.4 — .07
1A with complication inotropes/Swan or IABP
Lund et al,2 ISHLT Registry 2005–2015 CF-VAD All other recipients 1y — — 90 vs 88% —
2017
Trivedi et al,14 UNOS 2006–2013 CF-VAD listed status CF-VAD listed status 1B 3y — — 84 vs 85% .5
2016 1A using elective time
CF-VAD listed status CF-VAD listed status 1B 3y — — 84 vs 78% —
1A with complication
Weiss et al,12 UNOS 2009–2010 HMII All other recipients 1y 1.2 0.87–1.72 — .25
2011 CF-VAD All other recipients 1y 2.0 1.06–3.97 — .03

Abbreviations: CI, confidence interval; HMII, HeartMate II LVAD; HR, hazard ratio; IABP, intra-aortic balloon pump; ISHLT, International Society of Heart Lung Transplant; KM, Kaplan-
Meier product limit estimate; UNOS, United Network of Organ Sharing.

553
554 Dardas

outcomes but that reducing the prevalence of in-

fections among VAD-supported registrants may
result in incremental improvement in posttrans-
plant survival. Although posttransplant mortality
may be nominally higher among registrants with
infections, this association does not disqualify
these patients for transplant but certainly high-
lights the need to minimize complications and their
effects.
Studies of transplant timing from VAD support
have not provided clarity on whether an optimal
time for transplant exists. The first report of trans-
plant timing among persons with CF devices was
reported by John and colleagues18 who noted 1-
Fig. 1. Five-year posttransplant survival from studies year posttransplant survival of 87% among per-
of mechanically supported recipients. TAH, total artifi-
sons supported for a median of 151 days with
cial heart. (Data from Trivedi JR, Rajagopal K, Schumer
EM, et al. Differences in status 1A heart transplanta-
the HeartMate II device. When examining the dura-
tion survival in the continuous flow left ventricular tion of support and effect on posttransplant out-
assist device era. Ann Thorac Surg 2016;102(5):1512– comes, the investigators noted no significant
6; and Copeland JG, Copeland H, Gustafson M, et al. difference between persons receiving a transplant
Experience with more than 100 total artificial heart at less than 30 days, 30 to 89 days, 90 to 179 days,
implants. J Thorac Cardiovasc Surg 2012;143(3):727– and greater than 180 days at 30 days after trans-
34; and Unpublished data from the author, status 1A plant (P 5 .28) and 1-year posttransplant mortality
transplants without mechanical circulatory support (P 5 .18). A more modern series of patients re-
group.) ported by Kamdar and colleagues20 suggested
no significant difference (P 5 .47) between pa-
their series, the occurrence of any type of infection
tients supported for less than 180 days (1-year sur-
was not associated with worse outcomes
vival 89%) and those supported for more than
(P 5 .38). Healy and colleagues19 evaluated a
180 days (1-year survival 93%). The investigators
larger series of recipients and described the effect
reported an effect opposite the larger series report
of various status 1A complication groups on post-
by John and colleagues,18 which suggests
transplant survival (see Fig. 1, Table 2). These in-
declining survival with longer support (1-year sur-
vestigators noted that justifications for infection
vival 84% for transplants 90–179 days vs 79%
increased the risk of posttransplant morality
for transplants more than 365 days after implant).
when compared with justification for elective
Colvin and colleagues21 and Dardas and col-
30 days of status 1A time (1-year survival 79%
leagues22 evaluated transplant timing via simula-
vs 84%, P 5 .012).19 These studies suggest that
tion of the transplant system. Colvin and
carefully selected persons with device-related in-
colleagues21 found no marked change in post-
fections can have acceptable posttransplant
transplant survival when increasing the number
of elective status 1A days from 30 days to
90 days. Dardas and colleagues22 reported fewer
Table 2
transplants (53% vs 51% at 5 years), higher moral-
Hazard ratios for selected left ventricular assist
device complications at 10-year follow-up
ity (17% vs 21% at 5 years), and a higher fraction
death among transplants occurring after LVAD
Group HR 95% CI complications (7% vs 9%). Conclusions from
these studies are not definitive, as urgency for
Status 1A LVAD using Reference
transplant, therapeutic bias, and survivor bias
elective time
figure strongly into posttransplant outcomes. Ran-
Thromboembolism 1.12 0.81, 1.56 domized controlled trials in the area of timing are
Device infection 1.3 1.09, 1.56 infeasible, and the current practice of favoring
Device malfunction 0.81 0.62, 1.06 rapid transplant remains the appropriate strategy
Arrhythmia 1.24 0.76, 2.00 for most BTT scenarios.
Other 0.95 0.68, 1.32 The improved survival among persons in need of
VAD support justifies their use for BTT. The aggre-
Data from Healy AH, Baird BC, Drakos SG. Impact of ven-
tricular assist device complications on posttransplant sur-
gate data suggest a nonsignificant decrease in
vival: an analysis of the united network of organ sharing posttransplant survival among VAD-supported re-
database. Ann Thorac Surg 2013;95(3):870–75. cipients without complications. Those persons
 Impact of MCS on Posttransplant Outcomes 555

with LVAD complications have decreased survival, heart transplant recipients, VS portends a higher
but there is sufficient survival to justify transplant. risk of mortality with reports of case-fatality rates
The prospect of extending life for a decade or of 17% to 25%.24,25 Vasopressin deficiency,
more with the provision of transplant via BTT increased nitric oxide (NO) synthesis, and cytokine
LVAD strategies is a more valuable goal than maxi- release contribute to initiation and maintenance of
mizing the utility of every donor heart. refractory vasodilation (Fig. 2).23–28 Data also sug-
gest that increases in pulsatility (generated exper-
Selected posttransplant morbidities imentally by lower VAD speed) correlate with a
Vasoplegia Hypotension refractory to numerous paradoxic decrease in muscle sympathetic nerve
high-dose pressors with adequate cardiac output activity (a surrogate for sympathetic tone).29 This
characterizes the vasoplegia syndrome (VS). mechanism may explain the strong association
Awareness of this syndrome and evidence for a between CF-VADS and VS at the time of trans-
strong association with MCS are increasing. The plant. Given the refractory nature of this condition,
incidence of VS varies by report but may be as identifying both causation and risk factors remains
high as 10% to 27% after cardiac surgery.23 VS of paramount interest to avoid excess morbidity
may be more common in the CF-VAD era. Among and mortality.

Fig. 2. Pathophysiology and mediators of VS. ACE, angiotensin-converting enzyme; Ca11, calcium ions; CI, car-
diac index; CPB, cardiopulmonary bypass; IL, interleukin; KATP, ATP-sensitive potassium channel; NF, nuclear factor;
SIRS, systemic inflammatory response syndrome; SVR, systemic vascular resistance; TNF, tumor-necrotizing factor.
(From Liu H, Yu L, Yang L, et al. Vasoplegic syndrome: an update on perioperative considerations. J Clin Anesth
2017;40:66; with permission.)
556 Dardas

Risk factors for VS vary between studies. How- patients. At baseline, 33% exhibited greater than
ever, CF-VADs as BTT are a consistent risk factor. 10% of panel reactive antibodies (PRAs). Sensi-
In a recent series of 138 transplants from a single tized patients increased to 50% of the sample after
institution, Truby and colleagues23 identified male LVAD implant, and sensitization occurred against
sex (odds ratio [OR] 10.64, P 5 .023), body mass both class I and class II antigens. Risk factors for
index (OR 1.17, P 5 .001), BTT LVAD (OR 3.29, sensitization included younger age, female sex,
P 5 .041), and preoperative inotrope score (OR and pre-VAD sensitization. No differences were
4.57, P 5 .002) as risk factors for VS. The strong detected in posttransplant survival in this series.
association with CF-VADs is consistent with the Similarly, Ko and colleagues36 reported sensitiza-
findings of Patarroyo and colleagues,25 who re- tion (increase to >10% PRA among persons with
ported an OR of 2.8 (95% CI 1.1–7.4) for BTT <10% PRA before LVAD implant) among 23% of
CF-VADs among a series of 348 heart transplants. CF-LVAD implants (Fig. 3). Importantly, the in-
This investigation also notes an increased risk of creases in preformed antibody resolved in 67%
VS attributable to thyroid disease (OR 2.7; 95% of patients. The investigators also note a higher
CI 1.0–7.0, P 5 .04), though this factor was weakly risk of acute cellular rejection and antibody-
significant. Many studies associated VS with the mediated rejection during follow-up among newly
severity of illness before transplant, which must sensitized LVAD patients but not among newly
be considered with caution, as hypotension is sensitized patients without LVADs. Despite the
part of the syndrome’s definition. Older data sug- higher rates of rejection, medium-term survival
gest a strong association between angiotensin- was similar between the newly sensitized group
converting enzyme (ACE) use for vasoplegia with LVADs and those without MCS. Among
following cardiac surgery.30 Given this effect, LVAD patients with very high levels of sensitization,
many centers are attempting to minimize ACE- mortality was 40%. Currently, there are no effective
inhibitor (ACE-I) use before surgery. However, strategies for minimizing bleeding without poten-
given the uncertainty of transplant timing and tially increasing the thrombosis risk in CF-
known benefits of ACE-I in heart failure treatment, VADs.37,38 Thus, rapid treatment of bleeding sour-
randomized controlled trials are necessary to ces, a high transfusion threshold, and careful man-
establish the harm of ACE-I cessation at arbitrary agement of both antiplatelet and anticoagulant
intervals before transplant.31 agents are critical to reduce allosensitization
Treatment of VS is largely supportive with some events, as is the development of more biocompat-
sources advocating for the use of methylene blue, ible materials. Ultimately, short-term and long-term
which inhibits NO synthase and the cyclic GMP survival after transplant are similar for supported
messenger system. Although no randomized trial and unsupported recipients suggesting that allo-
of incident VS treatment exists, Ozal and col- sensitization does not play a prominent role in
leagues32 randomized coronary artery bypass sur- posttransplant survival.
gery patients at high risk of VS (defined as use of
ACE-I, calcium channel blockers, or heparin) to
methylene blue or no methylene blue. The treatment
group had no VS, whereas the untreated group had
a 26% incidence of VS (P<.001).32–34 As more LVAD
patients proceed to transplant, VS has the potential
to become a greater issue than the other aforemen-
tioned risk factors for severe morbidity and mortality
after BTT; further studies with methylene blue in this
population are warranted.

Allosensitization Durable LVAD support induces

allosensitization via hemolysis and frequent trans-
fusions encountered with gastrointestinal bleeding.
Allosensitization decreases the available donor
Fig. 3. Calculated PRA (cPRA) values before and after
pool and increases the risk of posttransplant mor-
LVAD implant (red horizontal line). Values increased
tality. Given the need to provide transplant to
at the time of LVAD implant and then reduced, on
VAD-supported patients quickly after complica- average, thereafter (black line). (From Ko BS, Drakos
tions develop, allosensitization is equivalent to a S, Kfoury AG, et al. Immunologic effects of
risk factor for mortality among this population. continuous-flow left ventricular assist devices before
Alba and colleagues35 investigated HLA antibody and after heart transplant. J Heart Lung Transplant
formation among 143 CF-LVAD–supported 2016;35(8):1027; with permission.)
 Impact of MCS on Posttransplant Outcomes 557

Total Artificial Heart exceeds 8 years.39 As discussed for durable

LVADs, any decrease in posttransplant survival
Biventricular support is increasing in the United
attributable to TAH support or BiVAD support
States, while remaining a small segment of the du-
must be considered against the very high mortality
rable device experience. However, persons with
without support as long as meaningful posttrans-
biventricular and isolated RV failure are among
plant survival is achieved. However, TAH survival
the most challenging patients to successfully sup-
is far less than the current median survival of
port to durable devices and transplant as the result
12 years; whether these transplants represent
of concomitant liver and kidney dysfunction. If du-
reasonable use of donor hearts is debatable.
rable support is feasible, options for durable de-
vices include the SynCardia Total Artificial Heart
(SynCardia Systems, LLC, Tucson, AZ) and either TEMPORARY CIRCULATORY SUPPORT
paracorporeal or biventricular CF-VADs. The development of the TandemHeart (CardiacAs-
A recent query of the OPTN registry identified sist, Inc, Pittsburgh, PA), Impella (Abiomed, Dan-
only 212 total artificial heart (TAH)–supported re- vers, MA), and miniaturized extracorporeal life
cipients.39 TAH support shares many of the risk support (ECLS) devices allows for temporary cir-
factors described with CF support. The introduc- culatory support of left ventricular (LV) and right
tion of a wearable, ambulatory driver for the TAH ventricular (RV) failure via more direct ventricular
allows for improved physical rehabilitation whether unloading. These devices are commonly used to
in or out of the hospital, though not all patients evaluate the trajectory of end-organ function to
meet the criteria for Freedom driver (SynCardia assess the feasibility of durable support or trans-
LLC) use and other morbidities, such as stroke plant. Therefore, many patients supported with
and depression, may impinge on rehabilitation ef- these devices had recent and severe end-organ
forts. Renal failure and liver dysfunction can occur dysfunction and are at an increased risk for end-
among TAH patients as the result of venous hyper- organ decompensation after transplant. However,
tension, despite adequate cardiac output deliv- if end-organ function is improved and perfusion
ered by the device. Intravascular volume and maintained, these devices offer additional perfu-
device settings must be assessed and meticu- sion when compared with optimal medical therapy
lously adjusted to avoid worsening renal and liver alone.
function that may lead to delisting or a higher risk
of organ failure at the time of transplant. Cheng TandemHeart
and colleagues39 reported a 24% prevalence of
renal failure or dialysis after TAH transplants. The TandemHeart offers LV and RV unloading via
Posttransplant survival among TAH patients is inflow proximal to the LV or RV and outflow above
decreased when compared with persons without the semilunar valves. Unloading at the level of the
MCS and to those supported with CF-LVADs left atrium or inferior vena cava/superior vena cava
(see Fig. 1). The force of mortality is greatest in offers substantial ventricular unloading. These de-
the perioperative phase with 80% survival at vices are complicated by the need for relative
1 year as compared with 88% to 93% 1-year sur- immobility to maintain inflow cannula position
vival among other transplant recipients.40 Graft (left-sided support) or outflow cannula position
failure and sepsis early after transplant are com- (right-sided support) and require heparinization.
mon and mirror the cause of death among persons Hemolysis is rare with the TandemHeart.42,43 A
with ongoing TAH support suggesting that end- query of the OPTN registry revealed 35 patients
organ function may be more difficult to restore with TandemHeart at the time of transplant with
with transplant alone and that infections may be 90-day survival of 88% (95% CI 71%–95%) and
harder to identify and treat during TAH support.41 1-year survival of 84% (95% CI 66%–93%, Dardas
In modern series, median posttransplant survival TF, unpublished data, 2018). Thus, limited data
is 6 to 7 years following TAH support, which is suggest that TandemHeart before transplant re-
among the lowest posttransplant survival of sults in no incremental decrease in posttransplant
devices discussed in this article.39,40 When survival among carefully selected patients.
compared with biventricular VADs (paracorporeal,
pulsatile, and CF–biventricular assist device Impella
[BiVAD] configurations), posttransplant survival
The Impella family of devices offers a rapidly
was greater for the BiVAD group compared with
implantable temporary MCS option. The intraven-
the TAH group empirically; but no differences
tricular inflow and outflow above the aortic or pul-
were found after adjusting for confounders (TAH
monic valve may allow for greater ventricular
HR 1.29, P 5 .1).39 Median survival with BiVADs
unloading than IABPs, extracorporeal membrane
558 Dardas

oxygenation (ECMO), or TandemHeart. These de- raised during the development of the new heart
vices require relatively large sheaths or surgical con- allocation system as to whether ECLS-supported
duits for delivery, heparinization, and are subject to candidates should have access to the highest ur-
hemolysis and, less commonly, mechanical failure. gency tier of the new system.48 The probability of
Placement of Impella devices via an axillary or sub- death after 30 days of ELCS support is 24%. Post-
clavian approach is feasible and may allow for transplant survival after ECLS is 76% at 1 year, and
meaningful physical therapy before transplant.44,45 posttransplant dialysis occurred in 24% of persons
Reports of BTT success after long periods of not on dialysis before transplant (unpublished
Impella support exist.46 A query of the OPTN reg- data). These values may underrepresent post-
istry, identified 55 transplants with previous transplant mortality in this population. Regardless,
Impella with 90-day survival of 91% (95% CI ECLS-supported candidates were afforded tier 1
79%–96%) and 1-year survival of 88% (95% CI urgency for 2 weeks in the new allocation system.
76%–95%, unpublished data). Placement of the After this time, urgency is subject to Regional
Impella in the axillary or subclavian artery may Review Board approval. For programs, ECLS use
allow for ambulation and physical therapy. As will remain a resource-intensive mode of circula-
randomized experiences with larger Impella de- tory support that must be applied early in
vices grows, more data will become available to the course of cardiogenic shock and maintained
assess whether Impella devices are superior to by expert teams of providers. As time on ECLS ac-
IABP and other forms of temporary circulatory cumulates, teams will need to carefully assess
support. whether transplant remains beneficial.

Extracorporeal Life Support SUMMARY

ECMO or ECLS can support critical cardiogenic MCS continues to develop and serves a critical
shock as a bridge to transplant, durable MCS, or role in the treatment armamentarium for persons
recovery. ECLS can provide support for left- and with cardiogenic shock. Posttransplant survival
right-sided failure but requires large cannulae, among persons supported with durable LVADs is
which predispose patients to significant bleeding, equivalent to patients receiving a transplant
infections, and embolic complications. Although without MCS, whereas temporary MCS and biven-
attractive for its ability to offer full circulatory sup- tricular MCS have lower but acceptable posttrans-
port and oxygenation, ultimately, the high burden plant survival when compared with waiting
of complications limit the use of ECLS as a reliable unsupported. Caution is appropriate when consid-
bridge to transplant. A recent meta-analysis of ering transplant of MCS patients supported with
ECMO complications documents the high inci- CF-LVADs complicated by infections and those
dence of complications that may preclude trans- with either TAH or ECLS support. However, for
plant (Fig. 4).47 Consistent with these highly most patients supported with MCS, transplant re-
morbid complications, numerous concerns were mains a positive outcome and a means to end the
burden of MCS. Case-by-case evaluation of the
prospect for reasonable posttransplant survival is
necessary among these high-risk populations.

REFERENCES
1. Colvin M, Smith JM, Hadley N, et al. OPTN/SRTR
2016 annual data report: heart. Am J Transplant
2018;18(Suppl 1):291–362.
2. Lund LH, Khush KK, Cherikh WS, et al. The registry
of the international society for heart and lung trans-
plantation: thirty-fourth adult heart transplantation
report-2017; focus theme: allograft ischemic time.
J Heart Lung Transplant 2017;36(10):1037–46.
3. Pagani FD, Miller LW, Russell SD, et al. Extended
Fig. 4. Major complications during ECLS. LE, lower ex-
tremity. (Adapted from Cheng R, Hachamovitch R, Kit- mechanical circulatory support with a continuous-
tleson M. Complications of extracorporeal membrane flow rotary left ventricular assist device. J Am Coll
oxygenation for treatment of cardiogenic shock and Cardiol 2009;54(4):312–21.
cardiac arrest: a meta-analysis of 1866 adult patients. 4. Aaronson KD, Slaughter MS, Miller LW, et al. Use of
Ann Thorac Surg 2014;97(2):614; with permission.) an intrapericardial, continuous-flow, centrifugal
 Impact of MCS on Posttransplant Outcomes 559

pump in patients awaiting heart transplantation. Cir- 17. Quader MA, Wolfe LG, Kasirajan V. Heart transplan-
culation 2012;125(25):3191–200. tation outcomes in patients with continuous-flow left
5. Mehra MR, Naka Y, Uriel N, et al. A fully magnetically ventricular assist device-related complications.
levitated circulatory pump for advanced heart fail- J Heart Lung Transplant 2015;34(1):75–81.
ure. N Engl J Med 2017;376(5):440–50. 18. John R, Pagani FD, Naka Y, et al. Post-cardiac trans-
6. Frazier OH, Rose EA, McCarthy P, et al. Improved plant survival after support with a continuous-flow left
mortality and rehabilitation of transplant candidates ventricular assist device: impact of duration of left
treated with a long-term implantable left ventricular ventricular assist device support and other variables.
assist system. Ann Surg 1995;222(3):327–36 [dis- J Thorac Cardiovasc Surg 2010;140(1):174–81.
cussion: 336–8]. 19. Healy AH, Baird BC, Drakos SG, et al. Impact of
7. Massad MG, McCarthy PM, Smedira NG, et al. Does ventricular assist device complications on posttrans-
successful bridging with the implantable left ventric- plant survival: an analysis of the United network of
ular assist device affect cardiac transplantation organ sharing database. Ann Thorac Surg 2013;
outcome? J Thorac Cardiovasc Surg 1996;112(5): 95(3):870–5.
1275–81 [discussion: 1282–3]. 20. Kamdar F, John R, Eckman P, et al. Postcardiac
8. Patlolla V, Patten RD, Denofrio D, et al. The effect of ven- transplant survival in the current era in patients
tricular assist devices on post-transplant mortality an receiving continuous-flow left ventricular assist
analysis of the United network for organ sharing devices. J Thorac Cardiovasc Surg 2013;145(2):
thoracic registry. J Am Coll Cardiol 2009;53(3):264–71. 575–81.
9. Jaski BE, Kim JC, Naftel DC, et al. Cardiac trans- 21. Colvin M, Miranda-Herrera D, Gustafson SK, et al.
plant outcome of patients supported on left ventric- Impact of increased time at the highest urgency
ular assist device vs. intravenous inotropic therapy. category on heart transplant outcomes for candi-
J Heart Lung Transplant 2001;20(4):449–56. dates with ventricular assist devices. J Heart Lung
10. Klotz S, Stypmann J, Welp H, et al. Does contin- Transplant 2016;35(3):326–34.
uous flow left ventricular assist device technology 22. Dardas TF, Cheng RK, Mahr CM, et al. Adverse ef-
have a positive impact on outcome pretransplant fects of delayed transplant listing among patients
and posttransplant? Ann Thorac Surg 2006;82(5): with implantable left ventricular assist devices.
1774–8. J Card Fail 2018. https://doi.org/10.1016/j.cardfail.
11. Schulze PC, Kitada S, Clerkin K, et al. Regional differ- 2018.01.003.
ences in recipient waitlist time and pre- and post- 23. Truby LK, Takeda K, Farr M, et al. Incidence and
transplant mortality after the 2006 United Network for impact of on-cardiopulmonary bypass vasoplegia
organ sharing policy changes in the donor heart alloca- during heart transplantation. ASAIO J 2018;64(1):
tion algorithm. JACC Heart Fail 2014;2(2):166–77. 43–51.
12. Weiss ES, Allen JG, Arnaoutakis GJ, et al. Creation of 24. Byrne JG, Leacche M, Paul S, et al. Risk factors and
a quantitative recipient risk index for mortality predic- outcomes for ’vasoplegia syndrome’ following car-
tion after cardiac transplantation (IMPACT). Ann diac transplantation. Eur J Cardiothorac Surg
Thorac Surg 2011;92(3):914–21 [discussion: 921–2]. 2004;25(3):327–32.
13. Dardas T, Mokadam NA, Pagani F, et al. Transplant 25. Patarroyo M, Simbaqueba C, Shrestha K, et al. Pre-
registrants with implanted left ventricular assist de- operative risk factors and clinical outcomes associ-
vices have insufficient risk to justify elective organ ated with vasoplegia in recipients of orthotopic heart
procurement and transplantation network status 1A transplantation in the contemporary era. J Heart
time. J Am Coll Cardiol 2012;60(1):36–43. Lung Transplant 2012;31(3):282–7.
14. Trivedi JR, Rajagopal K, Schumer EM, et al. Differ- 26. Fischer GW, Levin MA. Vasoplegia during cardiac
ences in status 1A heart transplantation survival in surgery: current concepts and management. Semin
the continuous flow left ventricular assist device Thorac Cardiovasc Surg 2010;22(2):140–4.
era. Ann Thorac Surg 2016;102(5):1512–6. 27. Levin MA, Lin H-M, Castillo JG, et al. Early on–
15. Donneyong M, Cheng A, Trivedi JR, et al. The asso- cardiopulmonary bypass hypotension and other fac-
ciation of pretransplant HeartMate II left ventricular tors associated with vasoplegic syndrome. Circula-
assist device placement and heart transplantation tion 2009;120(17):1664–71.
mortality. ASAIO J 2014;60(3):294–9. 28. Liu H, Yu L, Yang L, et al. Vasoplegic syndrome: an
16. Nativi JN, Drakos SG, Kucheryavaya AY, et al. update on perioperative considerations. J Clin
Changing outcomes in patients bridged to heart Anesth 2017;40:63–71.
transplantation with continuous- versus pulsatile- 29. Cornwell WK 3rd, Tarumi T, Stickford A, et al. Resto-
flow ventricular assist devices: an analysis of the ration of pulsatile flow reduces sympathetic nerve
registry of the International Society for Heart and activity among individuals with continuous-flow left
Lung Transplantation. J Heart Lung Transplant ventricular assist devices. Circulation 2015;
2011;30(8):854–61. 132(24):2316–22.
560 Dardas

30. Boeken U, Feindt P, Mohan E, et al. Post-perfusion device support as bridge-to-transplantation. J Card
syndrome and disturbed microcirculation after car- Surg 2016;31(10):648–53.
diac surgery: the role of angiotensin-converting- 40. Copeland JG, Copeland H, Gustafson M, et al.
enzyme inhibitors. Thorac Cardiovasc Surg 1999; Experience with more than 100 total artificial heart
47(6):347–51. implants. J Thorac Cardiovasc Surg 2012;143(3):
31. Raja SG, Fida N. Should angiotensin converting 727–34.
enzyme inhibitors/angiotensin II receptor antago- 41. Copeland JG, Smith RG, Arabia FA, et al. Total arti-
nists be omitted before cardiac surgery to avoid ficial heart bridge to transplantation: a 9-year expe-
postoperative vasodilation? Interact Cardiovasc rience with 62 patients. J Heart Lung Transplant
Thorac Surg 2008;7(3):470–5. 2004;23(7):823–31.
32. Ozal E, Kuralay E, Yildirim V, et al. Preoperative 42. Bruckner BA, Jacob LP, Gregoric ID, et al. Clin-
methylene blue administration in patients at high ical experience with the TandemHeart percuta-
risk for vasoplegic syndrome during cardiac sur- neous ventricular assist device as a bridge to
gery. Ann Thorac Surg 2005;79(5):1615–9. cardiac transplantation. Tex Heart Inst J 2008;
33. Mehaffey JH, Johnston LE, Hawkins RB, et al. Meth- 35(4):447–50.
ylene blue for vasoplegic syndrome after cardiac 43. Tempelhof MW, Klein L, Cotts WG, et al. Clinical
operation: early administration improves survival. experience and patient outcomes associated with
Ann Thorac Surg 2017;104(1):36–41. the TandemHeart percutaneous transseptal assist
device among a heterogeneous patient population.
34. Levin RL, Degrange MA, Bruno GF, et al. Methylene
ASAIO J 2011;57(4):254–61.
blue reduces mortality and morbidity in vasoplegic
44. Bansal A, Bhama JK, Patel R, et al. Using the mini-
patients after cardiac surgery. Ann Thorac Surg
mally invasive impella 5.0 via the right subclavian ar-
2004;77(2):496–9.
tery cutdown for acute on chronic decompensated
35. Alba AC, Tinckam K, Foroutan F, et al. Factors asso-
heart failure as a bridge to decision. Ochsner J
ciated with anti-human leukocyte antigen antibodies
2016;16(3):210–6.
in patients supported with continuous-flow devices
45. Lotun K, Shetty R, Patel M, et al. Percutaneous left
and effect on probability of transplant and post-
axillary artery approach for Impella 2.5 liter circula-
transplant outcomes. J Heart Lung Transplant
tory support for patients with severe aortoiliac arte-
2015;34(5):685–92.
rial disease undergoing high-risk percutaneous
36. Ko BS, Drakos S, Kfoury AG, et al. Immunologic ef- coronary intervention. J Interv Cardiol 2012;25(2):
fects of continuous-flow left ventricular assist de- 210–3.
vices before and after heart transplant. J Heart 46. Lima B, Kale P, Gonzalez-Stawinski GV, et al. Effec-
Lung Transplant 2016;35(8):1024–30. tiveness and safety of the impella 5.0 as a bridge to
37. Kirklin JK, Naftel DC, Kormos RL, et al. Interagency cardiac transplantation or durable left ventricular
registry for mechanically assisted circulatory sup- assist device. Am J Cardiol 2016;117(10):1622–8.
port (INTERMACS) analysis of pump thrombosis in 47. Cheng R, Hachamovitch R, Kittleson M, et al. Com-
the HeartMate II left ventricular assist device. plications of extracorporeal membrane oxygenation
J Heart Lung Transplant 2014;33(1):12–22. for treatment of cardiogenic shock and cardiac ar-
38. Starling RC, Moazami N, Silvestry SC, et al. Unex- rest: a meta-analysis of 1,866 adult patients. Ann
pected abrupt increase in left ventricular assist de- Thorac Surg 2014;97(2):610–6.
vice thrombosis. N Engl J Med 2014;370(1):33–40. 48. Kittleson MM, Kobashigawa JA. Cardiac transplan-
39. Cheng A, Trivedi JR, Van Berkel VH, et al. Compari- tation: current outcomes and contemporary contro-
son of total artificial heart and biventricular assist versies. JACC Heart Fail 2017;5(12):857–68.