Mechanical Circulatory

Support in Acute
Decompensated Heart Failure
and Shock
Nishtha Sodhi, MD, John M. Lasala, MD, PhD*

KEYWORDS
 Interventional management  Cardiogenic shock  Decompensated heart failure
 Intraaortic balloon pump  TandemHeart  Extracorporeal membrane support

KEY POINTS
 In the current era, failure of maximal medical therapy is no longer a justifiable endpoint given
the array of available advanced mechanical options.
 Deciding which mechanical device is most suitable depends largely on the degree of support
needed.
 These temporary support devices, if implemented in a timely fashion, can often bridge
patients to decision, recovery, long-term support devices (ventricular assist devices [VADs]
and total artificial heart), and/or heart transplant.

INTRODUCTION INTRA-AORTIC BALLOON PUMP

Introduction and Components
Approximately 6 million adults in the United The IABP has grown to be the most widely used
States have congestive heart failure. Many of hemodynamic support device since its introduc-
these patients, at some point in their care and tion in the 1960s.4 This device uses the counter-
evaluation, pass through a cardiac catheteriza- pulsation of a balloon in the descending aorta to
tion laboratory.1 The emerging subspecialty of improve cardiac output and increase coronary
“interventional heart failure” has arisen due to perfusion. The system comprises a dual-lumen
this expanding patient population who require 7.5F to 8.0F catheter with a polyethylene
expertise in not only the pathophysiology but balloon and the control console. The inner cath-
also the practical application and implementa- eter lumen accepts the guide wire during place-
tion of mechanical therapeutics to improve ment and transduces aortic pressure for
such hemodynamics, particularly once refractory monitoring. The gas lumen serves as the conduit
to optimal medical management (Fig. 1).2,3 An for the rapid exchange of helium in and out of
array of interventional therapeutics is available the balloon. Helium has low viscosity and is
in the modern era, with uses depending on absorbed rapidly in blood if the balloon inadver-
acute or chronic situations (Fig. 2). This article tently ruptures.
focuses on support in acute decompensated
heart failure and cardiogenic shock, including Hemodynamic Effects
intra-aortic balloon pumps (IABPs), continuous The hemodynamic consequences of counterpul-
aortic flow augmentation, and extracorporeal sation can be organized into those that occur
membrane oxygenation (ECMO). during inflation and those during deflation

Disclosure Statement: None.

Cardiovascular Division, Washington University, 660 South Euclid Avenue, St Louis, MO 63110, USA
* Corresponding author.
E-mail address: [email protected]

Intervent Cardiol Clin 6 (2017) 387–405

http://dx.doi.org/10.1016/j.iccl.2017.03.008
2211-7458/17/ª 2017 Elsevier Inc. All rights reserved.
388 Sodhi & Lasala

Fig. 1. The emerging subspecialty of “interventional heart failure” has arisen due to the expanding patient pop-
ulation who requires expertise in not only the pathophysiology but the practical application and implementation of
mechanical therapeutics to improve such hemodynamics, particularly once refractory to optimal medical manage-
ment, which forebodes incredible mortality. (Adopted from Samuels LE, Kaufman MS, Thomas MP, et al. Pharma-
cological criteria for ventricular assist device insertion following postcardiotomy shock: experience with the
Abiomed BVS system. J Card Surg 1999;14(4):288–93.)

(Fig. 3). Inflation occurs at the onset of diastole improvement in coronary perfusion. Balloon
and causes a displacement of blood that in- deflation is timed to occur immediately prior to
creases the diastolic pressure in the aorta. There systole, leading to an abrupt drop in aortic
is a resulting increase in systemic mean arterial pressure just prior to ventricular ejection. This
pressure and cardiac output as well as an reduces the ventricular afterload and leads to

Fig. 2. The array of mechanical support options for LV failure and RV failure, respectively. (Data from Kapur NK,
Esposito ML. Door to unload: a new paradigm for the management of cardiogenic shock. Curr Cardiovasc Risk
Rep 2016;10:41.)
 Acute Mechanical Circulatory Support 389

Fig. 3. Effects of inflation and deflation of IABP. (Data from Krishna M, Zacharowski K. Principles of intra-aortic
balloon pump counterpulsation. Contin Educ Anaesth Crit Care Pain 2009;9(1): 24–28.)

decreased workload and improved cardiac peripheral vascular disease. The IABP may also
output. be inserted without a sheath; however, this tech-
nique is contraindicated if there is significant
Technical Considerations and scar tissue or if the patient is obese.
Contraindications
Proper balloon volume sizing can maximize the Management
hemodynamic benefit for patients. Increases in One of the most important functions of the con-
balloon volume (up to 50 mL) are accomplished sole is to provide real-time information
by adding length to the balloon. Thus, sizing regarding the timing of balloon inflation and
charts are based on patient height to avoid deflation. There are several modes to set
balloon obstruction of left subclavian and bilat- balloon trigger, including ECG and pressure.
eral renal perfusion (Fig. 4). Balloon position ECG triggering is the most commonly used;
should be performed under fluoroscopy to avoid however, it is susceptible to arrhythmias.
this potential complication. If the placement Recognition of mistimed balloon inflation
cannot be performed under live fluoroscopy, and/or deflation is vital to ensuring proper he-
then immediate verification via a chest plain modynamic support. The correct IABP waveform
film is warranted. The proper position of the consists of inflation in diastole at the dicrotic
distal tip is at 1 cm to 2 cm distal to left subcla- notch, causing an augmentation of the diastolic
vian artery (second to third intercostal space), pressure above the unassisted pressure
although one study suggests that placing the (Fig. 5). Deflation should occur just before sys-
tip 2 cm above the carina may be a more reliable tole and cause a decrease in the end-diastolic
landmark.5 Additionally, serial monitoring of the pressure and peak systolic pressure. There are
left radial pulse and urine output can signal 4 scenarios to recognize: early inflation, late
malposition or movement of the IABP. inflation, early deflation, and late deflation.
Contraindications to IABP placement include When early inflation is occurring, the balloon is
severe aortic insufficiency, aortic aneurysm, and expanding prior to the dicrotic notch and can
390 Sodhi & Lasala

Fig. 4. Proper placement of IABP.

The tip of the catheter should be
below the subclavian, usually
achieved by placing this at the
tracheal bifurcation. Placement
should ensure that there is no
obstruction of left subclavian and
bilateral renal perfusion. (Data
from Ragosta M. Textbook of clin-
ical hemodynamics. Philadelphia:
Saunders/Elsevier; 2008.)

encroach onto the previous systolic phase in aortic pressure during the diastolic phase and
(Fig. 6). The detrimental effects can include pre- limits pressure augmentation and coronary flow
mature closure of the aortic valve with resulting (Fig. 7). It can also lead to poor afterload reduc-
increased filling pressures and afterload that re- tion and increased myocardial oxygen demand.
sults in increased myocardial oxygen demand. Finally, late deflation causes the balloon to
Late inflation occurs after the dicrotic notch impede on the subsequent systolic phase and
and can lead to poor diastolic augmentation the work of the heart against an inflated balloon
and with it poor coronary perfusion. In addition, leads to increased afterload and increased
there is ineffective afterload reduction and myocardial oxygen demand.
increased myocardial oxygen consumption de- Potential complications of IABP use include
mand. Early deflation leads to a sharp drop right limb ischemia, bleeding, thrombocytopenia,

Fig. 5. Proper timing of inflation

and deflation of an IABP. The
balloon inflates after the dicrotic
notch on the aortic pressure wave-
form (arrow). Deflation should be
completed before aortic ejection
begins on the next cardiac cycle,
which is evident by a lower end-
diastolic pressure of the augmented
beat compared with the unaug-
mented beat. (Data from Ragosta
M. Textbook of clinical hemody-
namics. Philadelphia: Saunders/
Elsevier; 2008.)
 Acute Mechanical Circulatory Support 391

Clinical Data and Guidelines for Intra-aortic

Balloon Pump
In the Intra-aortic balloon counterpulsation in pa-
tients with acute myocardial infarction compli-
cated by cardiogenic shock (IABP-SHOCK) I trial,
there was no benefit in hemodynamics with IABP
use in acute myocardial infarction (AMI) cardio-
genic shock, likely due to poor native hemody-
namics (Fig. 8A).7 IABP-SHOCK II concluded no
mortality benefit of IABP compared with medical
therapy in the setting of AMI complicated by
cardiogenic shock (Fig. 8B).8 At 30 days, 39.7%
of the patients in the IABP group and 41.3% of
the patients in the control group had died. At
12-month follow-up of these patients, there was
no survival benefit observed between the IABP
Fig. 6. Early inflation and late deflation. Balloon infla-
arm and control arm.8 In a meta-analysis, the
tion is occurring prior to the dicrotic notch (arrows)
and the end-diastolic pressure of the augmented IABP was found to increase the risk of bleeding
beat is higher than that of the unaugmented beat and stroke in AMI cardiogenic shock patients.9
consistent with late deflation. (Data from Ragosta M. Subsequently, the European Society of Cardiol-
Textbook of clinical hemodynamics. Philadelphia: ogy downgraded the guidelines for the IABP to
Saunders/Elsevier; 2008.) class III (harm), advising that the IABP should not
be used routinely in cardiogenic shock patients.10
The US population study by Stretch and col-
infection, and aortic dissection. Rupture of the leagues analyzed the contemporary use of MCS
balloon may also occur and is signaled by the devices from 2004 to 2011 and determined that
presence of blood within the tubing or with a IABP use prior to MCS was a predictor of mortality
gas loss alarm. In this case, the device should and increased costs (Table 1).11,12 Thus, although
be stopped immediately and removed. it is readily available and used frequently, there is
Long-term IABP support can be safely accom- controversy regarding clinical benefits of IABP.
plished via a subclavian approach that minimizes Currently, IABP has received a class IIa indication
infection and allows patients to ambulate.6 This for use during ST elevation myocardial infarction
approach can be considered for hemodynamic (STEMI) complicated by cardiogenic shock in the
support for patients undergoing stabilization 2013 American College of Cardiology Foundation
and subsequent work-up for transplant or left (ACCF)/American Heart Association (AHA) guide-
ventricular assist device (LVAD). lines. The current ACCF/AHA and most recent
Society for Cardiovascular Angiography and Inter-
ventions (SCAI) expert consensus document on
percutaneous coronary intervention (PCI) without
on-site cardiac surgery agree that the ability to
provide IABP support during transport of unstable
patients is a requirement for such centers.13

LEFT ATRIAL TO AORTA ASSIST DEVICE:

TANDEM HEART
Introduction and Components
The TandemHeart (CardiacAssist, Pittsburgh,
PA) is a support device that delivers blood
from the left atrium (LA) to the arterial system
(femoral artery) using an extracorporeal centrifu-
gal pump. The TandemHeart system is
Fig. 7. Early deflation along with late inflation. Early composed of (1) an inflow cannula (21F) with
deflation leads to a sharp drop in aortic pressure dur- transseptal placement into the LA, (2) an outflow
ing diastole (arrow), thereby limiting coronary perfu- cannula (15F or 17F) placed into the femoral ar-
sion. (Courtesy of Ragosta M. Textbook of clinical tery, (3) centrifugal pump, and (4) control con-
hemodynamics. Philadelphia: Saunders/Elsevier; 2008.) sole (Fig. 9).
392 Sodhi & Lasala

Fig. 8. (A) IABP-SHOCK I and (B) IABP-SHOCK II demonstrated no hemodynamic or mortality benefit with IABP
use. (Data from Prondzinsky R, Unverzagt S, Russ M, et al. Hemodynamic effects of intra-aortic balloon counterpul-
sation in patients with acute myocardial infarction complicated by cardiogenic shock: the prospective, randomized
IABP SHOCK trial. Shock 2012;37(4):378–84; and Thiele H, Zeymer U, Neumann FJ, et al. Intra-aortic balloon coun-
terpulsation in acute myocardial infarction complicated by cardiogenic shock (IABP-SHOCK II): final 12 month re-
sults of a randomised, open-label trial. Lancet 2013;382(9905):1638–45.)

Inflow and outflow cannulae as self-diagnostic algorithms and alarms. The

The inflow cannula is placed from the femoral console provides power to the system and has
vein into the LA via a transseptal puncture. The a battery backup that can provide up to 1 hour
cannula is 21F and comes in either 62-cm or of operation.
72-cm lengths. Blood inflow comes through 14
side holes at the distal tip. The arterial outflow Hemodynamic Effects
cannula is 17 cm in length and comes in either The TandemHeart works in parallel, or in tan-
15F or 17F. Both cannulae contain 3 radiopaque dem, with the left ventricle (LV). The redirection
markings at the distal end, allowing accurate of blood from the LA to the peripheral arterial
fluoroscopic placement. system unloads the LV, thereby reducing work-
load, wall stress, and myocardial oxygen de-
Centrifugal pump mand. Systemic circulation is perfused by both
The centrifugal pump is hydrodynamic and uses parallel pumps, generally with a greater contri-
a fluid-bearing system created by constant saline bution from the TandemHeart due to LV
infusion into the lower housing that minimizes response to the pump, which may virtually cease
friction. The pump itself has a maximum flow of due to changing loading conditions (decreased
4 L/min to 5 L/min; however, flow is also depen- preload and increased afterload). The 19F arte-
dent on inflow and outflow cannula size, with the rial cannula allows up to 5 L/min of flow whereas
variable in this system the outflow arterial the 15F cannula allows up to 3.5 L/min.
cannula.
Technical Considerations and
Console Contraindications
The controller is designed for ease of use with As with other types of hemodynamic support
onscreen step-by-step set-up guidance as well devices, the size of the peripheral vasculature

Table 1
Predictors of mortality in acute myocardial infarction cardiogenic shock
Odds Ratio Lower 95% CI Upper 95% CI P Value
CPR administration 3.50 2.20 5.57 <.001
IABP use 2.00 1.58 2.52 <.001
Intubation 1.71 1.27 2.30 <.001
Vasopressor use 1.39 .75 2.58 .30
Performed or administered up to 7 days before percutaneous ventricular assist device use. Data were collected from the
Nationwide Inpatient Sample from the Healthcare Cost and Utilization Project between 2004 and 2011, and determined
that IABP use prior to MCS was a predictor of mortality and increased costs.
Data from Stretch R, Sauer CM, Yuh DD, et al. National trends in the utilization of short-term mechanical circulatory sup-
port: incidence, outcomes, and cost analysis. J Am Coll Cardiol 2014;64(14):1407-15.
 Acute Mechanical Circulatory Support 393

Fig. 9. TandemHeart. The Tandem-

Heart system is composed of (1) an
inflow cannula (21F) with transseptal
placement into the LA, (2) an outflow
cannula (15F or 17F) placed into the
femoral artery, (3) centrifugal pump,
and (4) control console. (Data from
Naidu SS. Novel percutaneous
cardiac assist devices. Circulation
2011;123:533–43.)

can limit the use of the TandemHeart. The Adequate right ventricular (RV) function is
venous cannula is 21F whereas the arterial can- required to maintain LA volume. Contraindica-
nula can be 15F or 17F. Serial assessment must tions for the placement of the TandemHeart
be made of the access sites to assess for any include any condition that prohibits anticoagula-
possible complication or compromise of distal tion. In addition, the presence of a ventricular
perfusion. Although frequent examination of septal defect or aortic insufficiency precludes
limb color and temperature is needed, the distal placement.13,14
pulse may not be palpable due to the nonpulsa-
tile flow from the device. Placement of a sheath Management
to provide antegrade flow to the distal limb may The flow provided by the TandemHeart is
help prevent ischemia. dependent on several factors, including the sys-
The placement of the TandemHeart venous temic and pulmonary resistance, cannula size
cannula requires a transseptal puncture. For an and position, and fluid balance. Support is initi-
experienced operator, a transseptal puncture is ated at 5500 revolutions per minute (rpm) and
a relatively safe procedure. Approximately 1% subsequently raised by 250 rpm to 500 rpm until
of cases, however, develop complications 45, there is no long an increase in flow. This is the
including cardiac wall perforation, aortic root maximum flow for that particular set of parame-
puncture, pericardial effusion or tamponade, ters. Should more flow be desired, parameters,
stroke, and death 46. such as fluid balance, vascular resistance, and
For those operators with less experience, RV function, should be assessed and addressed.
methods to help delineate the atrial septal anat- This highlights the need for continued hemody-
omy, such as intracardiac echo or transesopha- namic monitoring as well as the utility of a
geal echo, may be more time consuming and Swan-Ganz catheter to help assess total cardiac
costly but may minimize complications. The output and filling pressures.
placement of a pigtail catheter to the aortic The TandemHeart console provides alarms
annulus marks the structure and helps avoid categorized into 3 settings: low priority, medium
inadvertent puncture of the aortic root. priority, and high priority. The alarms are
394 Sodhi & Lasala

accompanied by the triggering conditions and to IABP and/or pharmacotherapy who received
associated causes that can guide troubleshooting. the TandemHeart and found significant improve-
Vibration in the system’s tubing may signal ments in cardiac index, systolic blood pressure,
inadequate filling of the LA and should trigger and urine output over an average implant of
an evaluation for the root cause, including hypo- 5.8 days.15 Similar hemodynamics were found
volemia, pulmonary hypertension, cardiac tam- by Burkhoff and colleagues16 in 33 patients ran-
ponade, bleeding, RV failure, or arrhythmias. domized to TandemHeart or IABP (Fig. 10).
Kinks in the tubing, cannula migration, and Although no randomized trial of high-risk (HR)-
thrombus in the circuit should also be assessed. PCI with TandemHeart exists, in a small series
The importance of adequate anticoagulation of 54 patients who were deemed high risk for
has been discussed previously. Although proto- surgery, 97% procedural success and 87%
cols may vary among centers, activated clotting 6-month survival were found in those patients
time (ACT) should be maintained between 180 who underwent complex PCI with TandemHeart
seconds and 220 seconds. Additional heparin support.17 The 2015 SCAI/ACC/Heart Failure
administration should be considered if the flow Society of American (HFSA)/Society of Thoracic
drops below 1 L/min or if the system is stopped Surgeons (STS) Expert Consensus suggest that
for more than 5 minutes (not recommended). TandemHeart may be considered in (1) severe
Complications for this device are discussed LV dysfunction (ejection fraction [EF] <35%) or
previously but are summarized and include recent decompensated heart failure with associ-
bleeding, limb ischemia, arrhythmias, tampo- ated technically challenging or prolonged PCI
nade, atrial perforation, and residual atrial septal (depending on vascular anatomy, local exper-
defect.13,14 tise, and availability) and (2) continued deteriora-
Weaning from the TandemHeart should be tion of cardiogenic shock patients despite IABP
performed when deemed clinically appropriate and/or Impella 2.5 or Impella CP (Abiomed,
by the managing team. Flow rate is decreased Danvers, MA).13
by 50% every hour but is not to be reduced
below 1 L/min. If a patient remains stable, then LEFT VENTRICULAR TO AORTA ASSIST
the device may be turned off and the arterial DEVICES: IMPELLA
cannula immediately clamped followed by the Introduction and Components
venous cannula. The device is then retracted The Impella is a percutaneous ventricular sup-
from the LA into the right atrium and the heparin port device that uses a microaxial pump to
is stopped. Once the ACT reaches appropriate move blood continuously from the LV to the
levels (per hospital protocol) the femoral ascending aorta. There are 3 classes available
cannulae may be removed. that provide increasing levels of LV support:
Impella 2.5 (2.5 L/min; 12F system), Impella CP
Clinical Data and Guidelines for (approximately 3.5 L/min; 14F system), and the
TandemHeart Impella 5.0 (5.0 L/min; 21F system). All 3 have
The Texas Heart experience analyzed 117 pa- been approved in the United States to provide
tients with severe cardiogenic shock refractory hemodynamic support for up to 6 hours.

Fig. 10. Compared with the IABP, the patients who received the TandemHeart had a greater increase in cardiac index
and mean arterial pressure (A, B) and decrease in pulmonary capillary wedge pressure (C) but no difference in severe
adverse events of 30-day mortality. (Data from Burkhoff D, Cohen H, Brunckhorst C, et al. A randomized multicenter clin-
ical study to evaluate the safety and efficacy of the TandemHeart percutaneous ventricular assist device versus conven-
tional therapy with intraaortic balloon pumping for treatment of cardiogenic shock. Am Heart J 2006;152:469.e1–8.)
 Acute Mechanical Circulatory Support 395

Additionally, based on data from trials, such as the device’s reliance on a functional RV in
PROTECT I and PROTECT II,9,18 the Impella 2.5 providing LV filling (unless there is concurrent
and, more recently, the Impella CP have been RV support). And although transient arrhythmias
approved for use in hemodynamically stable pa- can be tolerated, more sustained tachycardias
tients undergoing elective or urgent HR-PCI. or asystole compromises the hemodynamic
The Impella support system is comprised of 3 support.
major components: (1) catheter, (2) purge sys-
tem, and (3) automated controller (Fig. 11). An Technical Considerations and
impeller and its adjacent motor are located Contraindications
near the outlet area in the ascending aorta. As A common limitation to use of Impella support is
it rotates, negative pressure draws ventricular the inadequacy of the peripheral vasculature in ac-
blood into the inlet area and through the can- commodating the large bore catheters. Although
nula. The flow through the cannula is dependent the Impella 2.5 (12F) and Impella CP (14F) can
on the rotation speed of the impeller for which often be placed percutaneously via the femoral ar-
there are 9 settings: P0 through P8. To protect tery, the Impella 5.0 device requires a surgical cut-
the motor, the purge fluid (5% dextrose with down of the femoral, axillary, or subclavian artery.
heparin) forms a hydraulic pressure shield that There is a need for anticoagulation (goal
prevents blood from migrating proximally past ACT >250 s at placement, 160–180 s for mainte-
the impeller and into the motor housing. nance), so the presence of a coagulopathy or
recent hemorrhage may prohibit its use. The pres-
Hemodynamic Effects ence of a mechanical aortic valve or a LV mural
There are 2 primary effects that the Impella im- thrombus precludes Impella use, as does signifi-
parts: (1) an unloading of the ventricle (lower cant aortic valve stenosis (valve area 0.6 cm2 or
end-diastolic volume and pressure) and (2) an in- less, except when performed with valvuloplasty)
crease in forward flow (higher mean arterial or insufficiency (21 or greater by echo). These clin-
pressure). There is a reduction in LV end- ical parameters do not prohibit TandemHeart or
diastolic pressure and volume that translates ECMO utilization.
into decreased wall tension and myocardial oxy-
gen demand.13,14 Furthermore, there is evi- Management
dence of improved coronary perfusion pressure Appropriate postprocedural management of the
and hyperemic flow velocity with a decrease in Impella system is paramount. The Impella
microvascular resistance with Impella 2.5.13,14 controller is an important tool that allows contin-
An important concept to recognize with uous automated oversight of device parameters
Impella, as with other forms of LV support, is and function. Although the device representative

Fig. 11. Impella. The Impella sup-

port system comprises 3 major
components: (1) catheter, (2)
purge system, and (3) automated
controller. (Data from Cardiogenic
Shock Therapy with Impella Clin-
ical Dossier.)
396 Sodhi & Lasala

is an invaluable resource for help with trouble- presence of thrombus. If correct position has
shooting, there are several scenarios with which been confirmed, then the catheter may need to
operators should be familiar. be removed and fully inspected while assuring
that an appropriate ACT exists (>250 s). The
Position monitoring presence of suction may lead to hemolysis.
The automated Impella controller has 2 main dis-
plays that are useful in monitoring catheter posi- Hemolysis
tion and function. The first is the home screen, Patients must be monitored for signs of possible
which contains an alarm window that can display hemolysis, such as new or worsening anemia and
up to 3 alarms simultaneously with trouble- the presence of dark-colored urine. Laboratory
shooting suggestions. In addition, there is a cen- testing may help confirm the presence of hemo-
tral display area that displays an illustration of lysis, including bilirubin, lactate dehydrogenase,
the determined Impella position with corre- haptoglobin, and plasma-free hemoglobin. He-
sponding position indicator message. The place- molysis usually indicates improper position but
ment screen displays 2 waveforms: placement may be due to 3 things: (1) inlet obstruction (po-
signal waveform and motor current waveform. sition or volume), (2) obstruction within the can-
These waveforms signals can be used to deter- nula (eg, clot), and (3) outflow obstruction (AV or
mine the catheter position. If the catheter needs AO wall close to outlet). These scenarios can be
to be repositioned, then echocardiographic or visualized with transthoracic echo and Doppler
fluoroscopic guidance is recommended. assessment.
A severely dysfunctional LV may be unable to
Heparin-induced thrombocytopenia
generate a significant pressure difference across
If heparin-induced thrombocytopenia is sus-
the aortic valve. This leads to dampening of the
pected, then all heparin products should be
placement signal and motor current waveforms
stopped and confirmatory testing performed.
that limits their utility in determining catheter
The team must then make a risk-benefit decision
position. In this case, the operator must rely on
as to continue without heparin in the system.
patient hemodynamic parameter and imaging
Direct thrombin inhibitors may be used for
to monitor position.
systemic anticoagulation or in the purge
During the initial placement of the Impella
solution (50).
catheter, there is tendency for the catheter to
dive forward into the ventricle. This can be Weaning
avoided by taking precautionary steps, including Once a decision has been made to wean the pa-
placing the inlet area at approximately 3.5 cm tient from support, then the flow should be
below the aortic valve, removing the slack from decreased 2 P-levels at a time as hemodynami-
the catheter over the aortic arch prior to starting cally tolerated until reaching P-2. If a patient
the pump, tightening the Tuohy-Borst valve, and tolerates P-2, the catheter may be pulled
placing a leg immobilizer. completely into the aorta and systemic anticoa-
gulation held. Once the ACT is less than 150 sec-
Suction alarm onds, the flow is reduced to P-0 and the catheter
Suction may occur with the Impella due to and introducer may be removed. In general, the
improper positioning or inadequate LV volume. longer the Impella has been in place, the longer
The Impella position should always be confirmed the weaning process will take. If used to support
with imaging and adjustments made to space HR-PCI, weaning should take 5 minutes to 10 mi-
the inlet from the ventricular wall. Inadequate nutes; however, if it has been left in for several
LV volume may be secondary to overall volume days, then it may take hours.
depletion but can be seen when there is poor
RV function leading to poor filling of the left Clinical Data and Guidelines: Impella
side. Echocardiographically or Swan-guided he- The PROTECT II trial randomized 452 symptom-
modynamics can help assess the root cause of atic patients with complex multivessel disease or
a suction alarm. unprotected left main disease and severely
When encountered with suction repositioning depressed LV function with primary endpoint
the catheter, decreasing the P-level, addressing of a 30-day composite of 11 adverse events.
disturbances in intravascular volume, and mini- Impella provided superior hemodynamic sup-
mizing movement with the use of a leg immobi- port in comparison with IABP, and at 90 days a
lizer can be attempted. It is paramount to trend toward decreased events was observed
recognize that a suction alarm at the initial in the intent-to-treat population. Larger differ-
placement of the catheter may signal the ences favoring Impella were seen in the per
 Acute Mechanical Circulatory Support 397

protocol population. A subsequent analysis transplant, or open-heart surgery. Larger pro-

redefining myocardial infarction as the develop- spective study is ongoing.
ment of new Q waves or creatine kinase–MB
more than 8 times the upper limit of normal EXTRACORPOREAL MEMBRANE
demonstrated lower rates of events, major car- OXYGENATION
diac adverse events, and cerebrovascular events Introduction and Components
in patients treated with Impella.13,19–21 ECMO is a form of mechanical cardiopulmonary
The Catheter-based Ventricular Assist Device support that can be provided for a prolonged
(cVAD) registry is an observational, multicenter, period. There are 2 general modes of ECMO
retrospective registry of patients supported support: (1) venovenous (VV) and (2) venoarterial
with Impella 2.5, Impella CP, Impella 5.0, Impella (VA) (Fig. 13). In VV-ECMO, blood is taken from
LD, or Impella RP and reflects real-world use and the right atrium, then oxygenated, and CO2
suggests greater survival with pre-PCI Impella removed prior to being returned to the right
insertion compared with pre-PCI IABP and/or atrium. This form of ECMO provides only respi-
pharmacotherapy alone (Fig. 12). ratory support; patients are still dependent on
Current guidelines for Impella use are summa- their native hemodynamic condition. In contrast,
rized in Table 2. VA-ECMO provides both respiratory and hemo-
dynamic support. In this case, blood is pulled
Impella RP from the venous system (right atrium or inferior
The RECOVER RIGHT study was conducted in vena cava) and, after gas exchange, is delivered
2014 to evaluate the safety and efficacy of the back to the arterial system at either central or
Impella RP. Findings included successful implan- peripheral cannulation sites.
tation in 90% of patients suffering from right There are a variety of configurations for an
heart failure and 73% successful survival to either ECMO system; however, there are several
30 days or to hospital discharge.22–24 The device fundamental components, including (1) drainage
is implanted into the femoral vein for inflow and perfusion cannulae, (2) centrifugal (most
through the inferior vena cava to the outlet common) or roller pump, (3) membrane oxygen-
area in the pulmonary artery. Delivery is via a ator and heat exchanger.
0.08 wire preferentially placed in the left lower This discussion focuses on VA-ECMO and its
pulmonary artery. role in hemodynamic support for patients with
The Impella RP is currently Food and Drug cardiogenic shock or requiring salvage during
Administration approved for humanitarian device cardiac arrest. There several indications for VA-
exemption for patients who develop acute right ECMO, all with the underlying notion that ulti-
heart failure or decompensation after LVAD mately the patient is thought to have a reversible
implantation, myocardial infarction, heart course of cardiac failure. This includes its

Fig. 12. The cVAD registry is an

observational, multicenter, retro-
spective registry of patients sup-
ported with Impella 2.5, Impella
CP, Impella 5.0, Impella LD, or
Impella RP and reflects real-world
use and suggests greater survival
with pre-PCI Impella insertion
compared with pre-PCI IABP and/
or pharmacotherapy alone. (Data
from O’Neill WW, Kleiman NS,
Moses J, et al. A prospective, ran-
domized clinical trial of hemody-
namic support with Impella 2.5
versus intraaortic balloon pump in
patients undergoing high-risk
percutaneous coronary interven-
tion: The PROTECT II study. Circu-
lation 2012;126:1717–27.)
398 Sodhi & Lasala

Table 2
Summary of guideline recommendations for Impella use
Clinical Society Guideline Impella Food and Drug
Populationsa Class Latest Update Administration Approval
PCI in cardiogenic shock I 2013 2016
Multiorgan failure, cardiogenic I 2013 2016
shock
PCI in low EF, complex CAD IIb 2011 2015
Bridge to recovery or decision, IIa 2013 2016
cardiogenic shock
STEMI and cardiogenic shock IIb 2013 2016
STEMI and urgent CABG IIa 2013 2016
Acutely decompensated heart IIa 2012 To be determined
failure
Consensus document on N/A 2015 2015/16
hemodynamic support
a
ACCF, HFSA, Heart Rhythm Society, International Society for Heart & Lung Transplantation, SCAI, STS.
Courtesy of Cardiogenic Shock Therapy with Impella Clinical Dossier.

Fig. 13. ECMO: there are a variety of configurations for an ECMO system; however, there are several fundamental
components, including (1) drainage and perfusion cannulae, (2) centrifugal (most common) or roller pump, (3) mem-
brane oxygenator and heat exchanger. Central cannulae (A), peripheral cannulae (B), and single cannulae for res-
piratory support alone (C) are depicted. (Data from Martinez G, Vuylsteke A. Extracorporeal membrane
oxygenation in adults. Contin Educ Anaesth Crit Care Pain 2012;12(2):57–61.)
 Acute Mechanical Circulatory Support 399

common usage as a bridge to advanced heart Cannulae for VA ECMO can be placed either
failure therapies, such as LVAD and/or trans- centrally or peripherally. Central access usually
plant.25–27 Patients who fail to wean from cardio- consists of the drainage cannula placed the in
pulmonary bypass (CPB) after cardiac surgery the right atrium and the perfusion cannula in
can also be transitioned to VA-ECMO.25–27 the ascending aorta. This approach requires a
sternotomy or thoracotomy and is most often
an option for those surgical patients who fail to
Hemodynamic Effects
wean from CPB. Peripheral cannulation may
There are 2 competing effects of ECMO on the
require a cutdown but can often be performed
LV. As blood is pulled from the venous system,
percutaneously. The drainage cannula may ac-
there is a decrease in preload and consequently
cess the venous circulation at the jugular or
the end-diastolic volume and end-diastolic pres-
femoral vein. The perfusion cannula is most
sure in the LV, which reduces wall tension and
often placed in the femoral artery but may also
work. In contrast, as blood returns to the arterial
use the axillary, subclavian, or in rare circum-
system there is an increase in afterload and
stances the carotid artery (although this site is
work.28
more commonly used in infants). In femoral ar-
tery cannulation, especially in patients with pe-
Technical Considerations and ripheral vascular disease, there may be a need
Contraindications to place an additional cannula into the femoral
Contraindications to ECMO include the pres- artery directing blood flow anterograde
ence of an irreversible process, multiorgan fail- down the ipsilateral limb to prevent ischemia
ure, prolonged cardiopulmonary resuscitation (Fig. 14).29
(CPR) (>60 mins), aortic dissection, and severe
aortic regurgitation. In addition, patients with Management
active bleeding or with contraindication to anti- The active management of ECMO can become
coagulation are generally not candidates for complex and its discussion is outside of the
ECMO support. Another factor to consider in- scope of this article. This article aims to address
cludes the presence of significant peripheral several of the more practical aspects to manage-
arterial disease. ment. A perfusionist is required to manage the

Fig. 14. In femoral artery cannulation, especially in patients with peripheral vascular disease, there may be a need
to place an additional cannula into the femoral artery directing blood flow anterograde down the ipsilateral limb to
prevent ischemia. (Data from Frischer JS, Stolar CJ. Extracorproeal membrane oxygenation. In: Holcomb GW,
Murphy JP, Ostlie DJ. Ashcraft’s Pediatric Surgery. 5th edition. Philadelphia: Elsevier; 2010. p. 74–86.)
400 Sodhi & Lasala

ECMO circuits and should be present during any CLINICAL DATA AND GUIDELINES:
procedure in the cardiac catheterization labora- EXTRACORPOREAL MEMBRANE
tory. Anticoagulation is key in preventing the OXYGENATION
formation of thrombus in the system. Heparin
is most commonly used with goal ACT of 180 There are no large randomized controlled trials
to 250. on use of ECMO. The Extracorporeal Life Sup-
Common complications include bleeding, port Organization registry, however, demon-
thrombosis, infection, and limb ischemia (58). strated a 27% survival to hospital
Hemorrhage is often seen and most commonly discharge.13,30 More recently, a 49% survival
involves the access site but may occur in any or- was demonstrated with the use of either me-
gan and can be devastating when the central chanical support or ECMO in cardiogenic
nervous system is involved.27,28 The risk of shock.13,31–33 Current expert guidelines recom-
thrombosis can range from 8% to 17%.27,28 mend the use of ECMO when concomitant hyp-
Managing the balance between the risks for oxemia and RV failure are present.
hemorrhage and thrombosis requires frequent
monitoring. The risk of infection of any PERCUTANEOUS HEART PUMP
indwelling line must be recognized and appro- The SHIELD II trial is a prospective, randomized
priate aseptic procedures followed. Limb multicenter trial comparing HeartMate PHP
ischemia can be secondary to a direct effect of (Thoratec, Pleasanton, CA) to the Impella 2.5 in
the cannula interrupting distal flow or from an patients undergoing HR-PCI. The PHP device is
embolic phenomenon. Supplemental antegrade capable of generating 4 L/min to 5 L/min flow
flow via a cannula placed in the femoral artery, with the advantage of percutaneous insertion
as discussed previously, may be help circumvent via a 14F introducer sheath. Once positioned
this problem. Other options include a cannula across the aortic valve, the outer sheath of the
placed in the dorsalis pedis or posterior tibial PHP is retracted, thereby fully expanding the
arteries.27,28 cannula to 24F (Fig. 15). The study is currently

Fig. 15. The HeartMate PHP (Thoratec) is capable of generating 4 L/min to 5 L/min flow with the advantage of
percutaneous insertion via a 14F introducer sheath. Once positioned across the aortic valve, the outer sheath of
the PHP is retracted, thereby fully expanding the cannula to 24F. (Courtesy of Thoratec, Pleasanton, CA.)
 Table 3
Options for mechanical device support in acute decompensated heart failure and cardiogenic shock
Device Pump Type Vascular Access Trigger Maximum Support
IABP Volume displacement Femoral artery for 7F or 8F sheath ECG or Pressure Dependent on native circulation;
pump 0.5 L/min
Impella Axial continuous flow Percutaneous—Impella 2.5 (12F) Continuous flow Impella 2.5 (2.5 L/min) and Impella
pump and Impella CP (14F) via femoral CP (approximately 3.5 L/min)
artery Impella 5.0 (5.0 L/min)
Surgical cutdown of the femoral,
axillary, or subclavian artery—
Impella 5.0 (21F)
TandemHeart Centrifugal continuous 21F inflow cannula into femoral Continuous flow 15F–4.0 L/min
flow pump vein with transseptal placement 17F–5.0 L/min
into the LA
15F or 17F outflow cannula into
femoral artery
VV-ECMO Respiratory support alone RA to oxygenator back to RA Continuous oxygenation Dependent on native circulation
dependent on native circulation

Acute Mechanical Circulatory Support

VA-ECMO Respiratory and Sternotomy or thoracotomy– Continuous oxygenation and Complete CPB
hemodynamic support central access—drainage hemodynamic flow and support
cannula placed the in the right
atrium and the perfusion
cannula in the ascending aorta
Percutaneous–peripheral access—
venous drainage cannula may
access (jugular or femoral vein)
and the arterial perfusion
cannula (femoral, axillary,
subclavian, carotid arteries)
It is best to match device with severity of cardiogenic shock right from the initial clinical assessment rather than proceeding with a fixed predetermined algorithm of escalation of device
therapy.

401
402 Sodhi & Lasala

Table 4
Hemodynamic impacts of the available pharmacologic and mechanical support options
Intra- Surgical
aortic Extracorporeal Ventricular
Balloon Membrane Assist
Inotropes Pump Oxygenation TandemHeart Impella Device
Advantages
Flow (L/min) <0.5 0.5 4 3.5 2.5–5.0 6.0
Coronary [ [[ [[ [[
perfusion
LV end-diastolic [ Y [[[[ YY YYY YYYY
pressure
Limitations
Arrhythmia 111
Stoke 11 11 1 1 111
Limb ischemia NA 1 111 11 1 NA
Bleeding NA 11 1111 111 1/11 1111
Cost $ $ $$$ $$$ $$$ $$$$$

enrolling and primary endpoint includes 90-day support requiring inotropic/pressor medications
composite analysis of cardiac death, myocardial to restore hemodynamics).34
infarction, stroke, repeat revascularization (PCI
or coronary artery bypass graft), bleeding
SUMMARY
(Bleeding Academic Research Consortium type
3 or 5), aortic insufficiency 12 over baseline, In the current era, failure of maximal medical
and severe hypotension (defined as systolic therapy is no longer a justifiable endpoint given
blood pressure <90 mm Hg while on a device the array of available advanced mechanical

Fig. 16. Optimal patient selection

for AMCS pumps. AMCS, acute
mechanical circulatory support;
EDP, end diastolic pressure; ESP,
end systolic pressure; MAP, mean
arterial pressure; MR, mitral regur-
gitation; MV-CAD, multivessel cor-
onary artery disease. (Data from
Kapur NK, Esposito ML. Door to un-
load: a new paradigm for the man-
agement of cardiogenic shock. Curr
Cardiovasc Risk Rep 2016;10:41.)
 Acute Mechanical Circulatory Support 403

options.1,2,12,35–37 Deciding which device is most These temporary support devices, if imple-
suitable depends largely on the degree of sup- mented in a timely fashion, can often bridge
port needed (Tables 3 and 4). It is best to match patients to decision, recovery, long-term sup-
device with severity of cardiogenic shock right port devices (VADs and total artificial heart),
from the initial clinical assessment rather than and/or heart transplant. They can even be
proceeding with a fixed predetermined algo- sued as adjunct support for HR procedures tar-
rithm of escalation of device therapy (Fig. 16). geting the underlying cardiogenic shock etiol-
Certain hemodynamic parameters, such as car- ogy (eg, HR-PCI) (see Fig. 17). Underlying
diac index, pulmonary artery pulsatility index, successful device selection and ultimate
pulmonary artery systolic pressure, and pulmo- outcome requires a multidisciplinary heart
nary capillary wedge pressure, can also be team approach, including heart failure special-
used to match mechanical support needed for ists, interventional cardiologists, and cardiotho-
the degree of LV and/or RV failure (Figs. 17 racic surgeons, along with patient preferences
and 18). executed in a timely fashion. Door to support

Fig. 17. Algorithm for percutaneous MCS device selection in patients with cardiogenic shock, cardiac arrest, and
HR-PCI. 3VD, 3-vessel coronary artery disease; AS, aortic stenosis; BiV, biventricular; CI, cardiac index; CPO, car-
diac power; HR, heart rate; LVEDP, LV end-diastolic pressure; MCS, mechanical circulatory support; MR, mitral
regurgitation; PCWP, pulmonary capillary wedge pressure; ROSC, return of spontaneous circulation; RVAD, right
VAD; SBP, systolic blood pressure; UPLMN, unprotected left main artery. (Data from Atkinson T, Ohman EM,
O’Neill WW, et al. A practical approach to mechanical circulatory support in patients undergoing percutaneous cor-
onary intervention. Interventional Scientific Council of the American College of Cardiology. JACC Cardiovasc Interv
2016;9(9):871–83.)
404 Sodhi & Lasala

Fig. 18. Proposed algorithm for patient presenting in cardiogenic shock. AMCS, acute mechanical circulatory sup-
port; BiV, biventricular; PA, pulmonary artery; PAPi, pulmonary artery pulsatility index; PCWP, pulmonary capillary
wedge pressure; TH, tandem heart; Vent, LV vent; VF, ventricular fibrillation; VT, ventricular tachycardia. (Data from
Kapur NK, Esposito ML. Door to unload: a new paradigm for the management of cardiogenic shock. Curr Cardi-
ovasc Risk Rep 2016;10:41.)

time is one of the most influential parameters in 7. Prondzinsky R, Unverzagt S, Russ M, et al. Hemody-
successful outcomes. namic effects of intra-aortic balloon counterpulsa-
tion in patients with acute myocardial infarction
REFERENCES complicated by cardiogenic shock: the prospec-
tive, randomized IABP SHOCK trial. Shock 2012;
1. Kapur NK, Dimas V, Sorajja P, et al. The Interven- 37(4):378–84.
tional Heart Failure Initiative: a mission statement 8. Thiele H, Zeymer U, Neumann FJ, et al. Intra-aortic
for the next generation of invasive cardiologists. balloon counterpulsation in acute myocardial infarc-
Catheter Cardiovasc Interv 2015;86(2):353–5. tion complicated by cardiogenic shock (IABP-
2. Kapur NK. Interventional Heart Failure. Cardiac In- SHOCK II): final 12 month results of a randomised,
terventions Today 2013. open-label trial. Lancet 2013;382(9905):1638–45.
3. Samuels LE, Kaufman MS, Thomas MP, et al. Phar- 9. Sjauw KD, Engström AE, Vis MM, et al.
macological criteria for ventricular assist device A systematic review and meta-analysis of intra-
insertion following postcardiotomy shock: experi- aortic balloon pump therapy in ST-elevation
ence with the Abiomed BVS system. J Card Surg myocardial infarction: should we change the guide-
1999;14(4):288–93. lines? Eur Heart J 2009;30(4):459–68.
4. Thompson KA, Philip KJ, Simsir S, et al. The new 10. Kolh P, Windecker S, Alfonso F, et al. 2014 ESC/
concept of ’interventional heart failure therapy’: EACTS Guidelines on myocardial revascularization:
part 2–inotropes, valvular disease, pumps, and the Task Force on Myocardial Revascularization of
transplantation. J Cardiovasc Pharmacol Ther the European Society of Cardiology (ESC) and the
2010;15(3):231–43. European Association for Cardio-Thoracic Surgery
5. Kim JT, Lee JR, Kim JK, et al. The carina as a useful (EACTS). Eur J Cardiothorac Surg 2014;46(4):517–92.
radiographic landmark for positioning the intraaortic 11. Stretch R, Sauer CM, Yuh DD, et al. National trends
balloon pump. Anesth Analg 2007;105(3):735–8. in the utilization of short-term mechanical circula-
6. Estep JD, Cordero-Reyes AM, Bhimaraj A, et al. Percu- tory support: incidence, outcomes, and cost anal-
taneous Placement of an Intra-Aortic Balloon Pump in ysis. J Am Coll Cardiol 2014;64(14):1407–15.
the Left Axillary/Subclavian Position Provides Safe, 12. Atkinson T, Ohman EM, O’Neill WW, et al.
Ambulatory Long-Term Support as Bridge to Heart A practical approach to mechanical circulatory sup-
Transplantation. JACC Heart Fail 2013;1(5):382–8. port in patients undergoing percutaneous coronary
 Acute Mechanical Circulatory Support 405

intervention. Interventional Scientific Council of the 23. Ragosta M. Textbook of clinical hemodynamics.
American College of Cardiology. JACC Cardiovasc Philadelphia: Saunders/Elsevier; 2008.
Interv 2016;9(9):871–83. 24. Martinez G, Vuylsteke A. Extracorporeal membrane
13. Rihal CS, Naidu SS, Givertz MM, et al. 2015 SCAI/ oxygenation in adults. Contin Educ Anaesth Crit
ACC/HFSA/STS Clinical Expert Consensus State- Care Pain 2012;12(2):57–61.
ment on the Use of Percutaneous Mechanical Cir- 25. Rousse N, Juthier F, Pinçon C, et al. ECMO as a
culatory Support Devices in Cardiovascular Care. bridge to decision: Recovery, VAD, or heart trans-
J Am Coll Cardiol 2015;65(19):2140–1. plantation? Int J Cardiol 2015;187:620–7.
14. Naidu SS. Novel percutaneous cardiac assist de- 26. Chung JC, Tsai PR, Chou NK, et al. Extracorporeal
vices. Circulation 2011;123:533–43. originally pub- membrane oxygenation bridge to adult heart
lished February 7, 2011. transplantation. Clin Transplant 2010;24(3):375–80.
15. Kar B, Gregoric ID, Basra SS, et al. The percutaneous 27. Abrams D. Extracorporeal membrane oxygenation
ventricular assist device in severe refractory cardio- in cardiopulmonary disease in adults. J Am Coll
genic shock. J Am Coll Cardiol 2011;57:688–96. Cardiol 2014;63(25 Pt A):2769–78.
16. Burkhoff D, Cohen H, Brunckhorst C, et al. 28. Burkhoff D. Hemodynamics of mechanical circula-
A randomized multicenter clinical study to evaluate tory support. J Am Coll Cardiol 2015;66(23):2663–74.
the safety and efficacy of the TandemHeart percu- 29. Gaffney AM, Wildhirt SM, Griffin MJ, et al. Extracor-
taneous ventricular assist device versus conven- poreal life support. BMJ 2010;341:c5317.
tional therapy with intraaortic balloon pumping 30. Thiagarajan RR, Brogan TV, Scheurer MA, et al.
for treatment of cardiogenic shock. Am Heart J Extracorporeal membrane oxygenation to support
2006;152:469.e1-8. cardiopulmonary resuscitation in adults. Ann
17. Alli OO, Singh IM, Holmes DR Jr, et al. Percuta- Thorac Surg 2009;87:778–85.
neous left ventricular assist device with Tandem- 31. Takayama H, Truby L, Koekort M, et al. Clinical
Heart for high-risk percutaneous coronary outcome of mechanical circulatory support for re-
intervention: The Mayo Clinic experience. Catheter fractory cardiogenic shock in the current era.
Cardiovasc Interv 2012;80:728–34. J Heart Lung Transplant 2013;32:106–11.
18. Krishna M, Zacharowski K. Principles of intra-aortic 32. Frischer JS, Stolar CJ. Extracorproeal membrane
balloon pump counterpulsation. Contin Educ oxygenation. In: Holcomb GW, Murphy JP,
Anaesth Crit Care Pain 2009;9(1):24–8. Ostlie DJ, editors. Ashcraft’s Pediatric Surgery.
19. O’Neill WW, Kleiman NS, Moses J, et al. 5th edition. Philadelphia: Elsevier; 2010. p. 74–86.
A prospective, randomized clinical trial of hemody- 33. Kapur NK, Esposito ML. Door to unload: a new
namic support with Impella 2.5 versus intraaortic paradigm for the management of cardiogenic
balloon pump in patients undergoing high-risk shock. Curr Cardiovasc Risk Rep 2016;10:41.
percutaneous coronary intervention: The PROTECT http://dx.doi.org/10.1007/s12170-016-0524-3.
II study. Circulation 2012;126:1717–27. 34. Supporting Patients Undergoing HIgh-Risk PCI
20. Dangas GD, Kini AS, Sharma SK, et al. Impact of he- Using a High-Flow PErcutaneous Left Ventricular
modynamic support with Impella 2.5 versus intra- Support Device (SHIELD II) (SHIELD II). Available
aortic balloon pump on prognostically important clin- at: https://clinicaltrials.gov/show/NCT02468778.
ical outcomes in patients undergoing high-risk percu- Accessed September 1, 2016.
taneous coronary intervention (from the PROTECT II 35. Shekar K, Gregory SD, Fraser JF, et al. Mechanical
randomized trial). Am J Cardiol 2014;113:222–8. circulatory support in the new era: an overview. Crit
21. Cohen MG, Ghatak A, Kleiman NS, et al. Opti- Care 2016;20:66.
mizing rotational atherectomy in high-risk percuta- 36. Burkhoff D, Naidu SS. The science behind percuta-
neous coronary interventions: Insights from the neous hemodynamic support: a review and compar-
PROTECT IotaIota study. Catheter Cardiovasc ison of support strategies. Catheter Cardiovasc
Interv 2014;83:1057–64. Interv 2012;80(5):816–29.
22. Anderson MB, Goldstein J, Milano C, et al. Benefits 37. Morley D, Litwak K, Ferber P, et al. Hemodynamic
of a novel percutaneous ventricular assist device for effects of partial ventricular support in chronic
right heart failure: The prospective RECOVER heart failure: results of simulation validated with
RIGHT study of the Impella RP device. J Heart in vivo data. J Thorac Cardiovasc Surg 2007;
Lung Transplant 2015;34(12):1549–60. 133(1):21–8.