10.1161/CIRCULATIONAHA.118.038269

1

Unloading the Left Ventricle Before Reperfusion in Patients with Anterior

ST-Segment Elevation Myocardial Infarction:

A Pilot Study Using the Impella CP®

Running Title: Kapur et al.; LV Unloading with Impella before PCI in STEMI

Navin K. Kapur et al.

The full author list is available on page 17.

Address for Correspondence:

Navin K. Kapur, MD

The CardioVascular Center

and The Molecular Cardiology Research Institute

Tufts Medical Center,

800 Washington Street, Box # 80,

Boston, MA 02111

Tel: 617-636-9371

Fax: 617-636-1444

Email: Nkapur@tuftsmedicalcenter.org

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

10.1161/CIRCULATIONAHA.118.038269

2

Abstract

Background: In ST-elevation myocardial infarction (STEMI), infarct size correlates directly

with heart failure and mortality. Preclinical testing has shown that compared with reperfusion

alone, mechanically unloading the left ventricle (LV) before reperfusion reduces infarct size and

in addition, that 30 minutes of unloading activates a cardioprotective program that limits

reperfusion injury. The Door-To-Unload in STEMI Pilot Trial represents the first exploratory

study testing whether LV unloading and delayed reperfusion in patients with STEMI without

cardiogenic shock is safe and feasible.

Methods: In a multi-center, prospective, randomized exploratory safety and feasibility trial, we

assigned 50 patients with anterior STEMI to LV unloading using the Impella CP followed by

immediate reperfusion (U-IR) versus delayed reperfusion after 30 minutes of unloading (U-DR).

The primary safety outcome was a composite of major adverse cardiovascular and

cerebrovascular events (MACCE) at 30 days. Efficacy parameters included assessment of infarct

size using cardiac magnetic resonance (CMR) imaging.

Results: All patients completed the U-IR (n=25) or U-DR (n=25) protocols with respective mean

door-to-balloon times of 72 versus 97 minutes. MACCE rates were not statistically different

between the U-IR versus U-DR groups (8% vs. 12%, respectively, p=0.99). Compared with the

U-IR group, delaying reperfusion in the U-DR group did not affect 30-day mean infarct size

measured as a percent of LV mass (15±12% vs. 13±11%, U-IR vs. U-DR, p=0.53).

Conclusions: We report that LV unloading using the Impella CP device with a 30 minute delay

before reperfusion is feasible within a relatively short time period in anterior STEMI. The DTU-

STEMI pilot trial did not identify prohibitive safety signals that would preclude proceeding to a

larger pivotal study of LV unloading before reperfusion. An appropriately powered pivotal trial

comparing LV unloading before reperfusion to the current standard of care is required.

Clinical Trial Registration: URL: https://www.clinicaltrials.gov. Unique identifier:

NCT03000270

Key Words: acute myocardial infarction, heart failure, infarct size, left ventricular unloading,

mechanical circulatory support, ischemia reperfusion injury

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

10.1161/CIRCULATIONAHA.118.038269

3

Clinical Perspective

What is new?

• Preclinical testing has shown that compared with reperfusion alone, mechanically

unloading the left ventricle (LV) for 30 minutes before reperfusion activates a

cardioprotective program that limits reperfusion injury.

• The Door-To-Unload in STEMI Pilot Trial represents the first exploratory study testing

the safety and feasibility of LV unloading and delayed reperfusion as a method to reduce

infarct size in patients with STEMI without cardiogenic shock.

• Findings from this study suggest for the first time that LV unloading using the Impella

CP device with a 30-minute delay before reperfusion is feasible within a relatively short

Door to Balloon Time.

What are the clinical implications?

• The DTU-STEMI pilot trial did not identify any prohibitive safety signals that would

preclude proceeding to a larger pivotal study of LV unloading before reperfusion.

• Based on results from this pilot trial, an appropriately powered pivotal trial comparing

LV unloading and delayed reperfusion with the current standard of care is currently in

development.

• The finding that LV unloading and delaying reperfusion is feasible potentially opens

many new doors for further exploration by suggesting that other adjunct interventions

(i.e. pharmacologic) may now be considered during that ‘window-period’ where the

coronary artery is occluded, but the LV is mechanically unloaded in a STEMI.

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

10.1161/CIRCULATIONAHA.118.038269

4

Introduction

Acute myocardial infarction (AMI) is a leading cause of heart failure (HF), with an annual

incidence of over 700,000 in the United States alone.1 Contemporary in-hospital management of

ST-segment elevation AMI (STEMI) focuses on reducing door-to-balloon (DTB) time to reduce

infarct size. However, despite intense resource allocation to achieve DTB times under 90

minutes, the incidence of post-AMI heart failure remains high, with every 5% increase in

myocardial infarct size associated with a 20% increase in one-year hospitalization for HF and

one-year mortality.2-4 New approaches are needed to reduce infarct size and to decrease both the

incidence of HF and late-term mortality after STEMI.

Multiple preclinical models over the past 20 years have shown that compared with

reperfusion alone, mechanically unloading the left ventricle before, not after, coronary

reperfusion reduces ischemia-reperfusion injury and myocardial infarct size in AMI.5-7 More

recently, preclinical testing identified that unloading the LV and delaying reperfusion by 30

minutes activates a cardioprotective signaling program that reduces myocardial damage after

reperfusion and promotes myocardial recovery 30 days after AMI.8,9 These findings suggest that

LV unloading before reperfusion may limit reperfusion injury in the treatment of AMI.

The Impella CP (Abiomed, Danvers, MA) is a percutaneously delivered hemodynamic

support device that reduces LV myocardial oxygen demand. The Impella CP is clinically

indicated for high-risk coronary intervention and cardiogenic shock10-12, but is not indicated for

STEMI without cardiogenic shock. Furthermore, in STEMI, any delay to reperfusion is not

recommended. To inform the development of an appropriately powered randomized controlled

study comparing the efficacy of a LV unloading strategy with the current standard of care for

STEMI, we wished to test the feasibility of activating an unloading device with or without a

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

10.1161/CIRCULATIONAHA.118.038269

5

delay to coronary reperfusion and to begin exploring whether in the setting of unloading,

delaying reperfusion improves myocardial salvage. The Door-To-Unload in STEMI Pilot Trial is

the first exploratory study testing the feasibility and safety of LV unloading before reperfusion in

STEMI without cardiogenic shock.

Methods

All data and supporting materials of this study will not be made available to other researchers for

purposes of reproducing the results or replicating the procedure.

Trial Design

The DTU-STEMI study was a prospective, multicenter, randomized pilot trial involving 14

centers in the United States to explore the feasibility and safety and potential benefit of

mechanical unloading prior to coronary reperfusion in patients presenting with anterior STEMI.

Based on the preclinical work described above, all patients received acute mechanical unloading

with the Impella CP system (Abiomed Inc., Danvers, MA) and were then randomized to one of

two arms: LV unloading followed by immediate reperfusion (U-IR) or LV unloading with a 30-

minute delay to reperfusion (U-DR) (Figure 1). This comparison was specifically designed to test

whether unloading the myocardium for 30 minutes before reperfusion was feasible and safe (i.e.

based on the preclinical findings) and if there was any evidence that delaying reperfusion

increased infarct size.

Institutional review boards at each site approved the trial and each enrolled patient

provided written informed consent. The study was designed by the principal investigators and

Abiomed. The study steering committee members and an independent Data and Safety Monitor

oversaw the safety and feasibility of the trial. A clinical events committee, blinded to

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

10.1161/CIRCULATIONAHA.118.038269

6

randomization group, independently adjudicated clinical safety endpoints in the study. The

sponsor was responsible for the study conduct, reporting, monitoring and managed the database.

Angiographic, cardiovascular magnetic resonance (CMR), echocardiography and ECG studies

were analyzed by core laboratories blinded to the treatment allocation. The study design was

approved by the Food and Drug Administration and registered accordingly (NCT03000270).

Study Population

Patients 21-80 years of age presenting between 1 to 6 hours from chest pain onset and with ST-

segment elevation of ≥2 mm in 2 or more contiguous anterior leads or ≥ 4 mm total ST-segment

deviation sum in the anterior leads were eligible for enrollment. Patients with prior MI or

coronary artery bypass grafting surgery, out of hospital cardiac arrest requiring cardiopulmonary

resuscitation, cardiogenic shock, inability to undergo Impella CP insertion, fibrinolysis within 72

hours of presentation, or contraindications to CMR imaging were excluded.

Study Procedures

Patients were randomized to either U-IR or U-DR arms immediately after femoral vascular

access was obtained. The Impella CP was placed prior to diagnostic coronary angiography and

operators were instructed to perform percutaneous coronary intervention (PCI) using second-

generation drug eluting stents and to follow guideline-directed post-AMI care. Timing of

coronary angiography after LV unloading was not specified in the protocol. In the U-DR group,

operators were allowed to shorten the time between unloading and reperfusion if deemed

clinically necessary. After PCI, the Impella CP was explanted after a minimum of 3 hours of LV

support. The protocol allowed for an extended duration of hemodynamic support as clinically

indicated.

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

10.1161/CIRCULATIONAHA.118.038269

7

Study Endpoints

The primary safety outcome was a composite of major adverse cardiovascular and

cerebrovascular events (MACCE) including cardiovascular mortality, reinfarction, stroke, or

major vascular events at 30 days. Supplemental Table 1 contains definitions used to adjudicate

each component of MACCE. Additional safety parameters included all-cause mortality,

hemolysis, acute renal dysfunction, hospitalization for heart failure, ventricular arrhythmias, LV

thrombus, bleeding and minor vascular events. Bleeding in Academic Research Consortium

(BARC) bleeding definitions are provided as Supplemental Methods. The primary efficacy

endpoint was an assessment of infarct size as percent of total LV mass at 30 days using CMR.

Secondary efficacy endpoints included infarct size by CMR at 3-5 days and 30 days. Exploratory

endpoints included a comparison of infarct size normalized to area at risk at 3-5 days between

groups. Qualifying 12-lead electrocardiograms were evaluated to quantify ST Segment Elevation

Sum (∑STE), an established clinical marker of area at risk in STEMI.13 Specifically, ∑STE was

quantified by measuring the magnitude of ST-segment elevation 0.08 seconds after the J-point

across precordial leads as compared to the isoelectric segment of the P-R interval in an

independent core lab, blinded to the study group allocation.

Cardiac Magnetic Resonance

Patients underwent cardiac magnetic resonance (CMR) imaging on days 3 to 5 and again on day

30 (±7 days). Standard protocols were used as detailed previously.14,15 In brief, cine-CMR was

performed using a steady-state free-precession sequence and analyzed for LV ejection fraction,

volumes, and mass. Delayed-enhancement imaging was performed using a 2-D segmented

inversion-recovery sequence, 10 minutes after administration of routine extracellular gadolinium

contrast (0.15 mmol/kg of body weight), and analyzed for the presence and extent of infarction

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

10.1161/CIRCULATIONAHA.118.038269

8

and microvascular obstruction. Infarct size was expressed as a percentage of total LV mass as

well as a percentage of the area-at-risk. Determination of the area-at-risk was based on the

physiological principle that the lateral boundaries of the area-at-risk equal the lateral boundaries

of the infarction starting relatively early in the process of evolving myocardial infarction.16

Hence, the area-at-risk (as percentage of LV mass) was calculated as the endocardial surface area

of the infarct (as depicted by the delayed-enhancement images) divided by the total LV

endocardial surface area.17 If an infarct was not visualized, infarct size as a percentage of the

area-at-risk was considered to be zero. A central core laboratory (Duke Cardiovascular Magnetic

Resonance Center, Durham, NC) qualified participating sites, performed quality assessment on

the images during the conduct of the study, and manually performed assessment of CMR

parameters on de-identified images without knowledge or access to treatment assignment or

clinical outcomes.

Statistical Analysis

Baseline demographic and clinical variables were summarized for the two treatment groups. The

study was powered to detect a large difference in infarct size assuming a large standard deviation

(SD) that may be expected in a small STEMI study. Specifically, a power of 0.88 and an alpha

of 0.05 was used to detect an absolute difference in infarct size of 10% with an assumed SD of

10%. Given the exploratory nature of the trial and the intent to inform subsequent trial design, no

corrections were made for the multiplicity of testing. All continuous variables were summarized

as mean±SD or as median (interquartile range, IQR), and compared between treatment groups

using the appropriate parametric or non-parametric tests. Categorical variables were summarized

as frequencies and percentages and compared between treatment groups using Pearson’s χ2 test

for contingency tables or Fisher Exact test, as appropriate. All statistical tests and/or confidence

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

10.1161/CIRCULATIONAHA.118.038269

9

intervals, as appropriate, were performed at α=0.05 (2-sided). The comparability among

treatment groups was evaluated with respect to all clinically relevant demographic and baseline

characteristic variables. As this study was a pilot study, the primary analysis did not involve

imputation. Thus, treatments were compared based upon the primary endpoint using only

subjects with primary endpoint data (evaluable CMR data at 30 days).

Results

A total of 50 patients with anterior STEMI were enrolled and randomized to either the U-IR or

U-DR arms (n=25/group) between April 2017 and May 2018. Baseline characteristics were not

statistically different between the groups (Table 1, Supplemental Table 2). Mean age of trial

participants was 59.7 years and 38 patients (76%) were male. On average, patients were

hypertensive on presentation with time from chest pain onset to LV unloading not statistically

different between the groups (176.2±73.4 minutes vs. 200.2±151.8 minutes, U-DR vs. U-IR,

p=0.48). ∑STE was >4 in 90% (45/50) of patients. Prior to Impella CP placement, LV end-

diastolic pressure was elevated in both groups (25.0±9.6mmHg vs. 24.0±8.1mmHg, U-DR vs. U-

IR, p=0.73). Baseline LVEF was obtained using left ventriculography prior to randomization in

90% (45/50) of patients using the required PCI arterial vascular access (either femoral or radial

per the discretion of the operator). Baseline LVEF was lower in the U-DR group (41.9±12.3%

vs. 32.7±12.7%, U-IR vs. U-DR, p=0.02). The Impella CP was successfully implanted in all 50

patients with a mean power (P-level) of 7.6±1.0 and mean device flow of 2.8±0.4 L/min during

the 3 hours of support required by the study protocol, indicating successful unloading of the LV.

Mean time from the start of the procedure to Impella CP implantation and activation was

15.4±8.4 minutes for the total population. The time from unloading (Impella CP insertion and

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

10.1161/CIRCULATIONAHA.118.038269

10

activation) to the first coronary angiogram was significantly longer in the U-DR group

(13.8±10.3 minutes vs. 1.7±6.9 minutes, U-DR vs.. U-IR, p<0.001). All timing elements are

reported in Table 2. Radial artery access was used for PCI in 60% (30/50) of patients. The use of

a vascular closure device was at the discretion of the operators. In 58% (29/50) of the patients, a

femoral artery closure device was used (56% (14/25) vs. 60% (15/25), U-DR vs. U-IR, p=0.99).

The left anterior descending artery was identified as the culprit coronary artery and treated with

stenting in 98% (49/50) of patients (Supplemental Table 3). One patient randomized to the U-DR

arm did not have any coronary lesions requiring PCI. All patients undergoing PCI received a

P2Y12 inhibitor prior to PCI; 8% of patients received bivalirudin and 94% received

unfractionated heparin. Among these, one patient received both bivalirudin and unfractionated

heparin. A glycoprotein IIb/IIIa receptor inhibitor was used in 8% of patients in addition to dual

antiplatelet therapy prior to PCI. Coronary angiography was performed after LV unloading was

initiated. Thrombolysis in Myocardial Infarction (TIMI) 0 to 1 flow was observed in 40%

(10/25) of U-DR vs. 64% (16/25) of U-IR patients before PCI (p=0.16). Post-PCI TIMI 3 flow

was observed in 98% (48/49) of patients undergoing PCI.

All patients assigned to the U-DR arm completed 30 minutes of LV Unloading prior to

reperfusion without need for bailout PCI in any patient (Figure 2). Timing elements including

device implantation to balloon reperfusion are shown in Table 2. Mean DTB time was longer in

the U-DR arm (96.7±26 minutes vs. 72.6±24 minutes, U-DR vs. U-IR, p=0.002), driven by a

prolonged unloading to balloon time in the U-DR group (34.1±3 minutes vs. 10.5±7 minutes, U-

DR vs. U-IR, p<0.001).

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

10.1161/CIRCULATIONAHA.118.038269

11

Primary Efficacy Endpoint and Additional Efficacy Parameters

CMR was performed in 84% (21/25) and 80% (20/25) of patients between days 3 to 5 in the U-

DR and U-IR groups respectively, and in 84% (21/25) and 76% (19/25) of patients at 30 days of

follow up in the U-DR and U-IR groups respectively. The primary efficacy endpoint of infarct

size normalized to total LV mass at 30 days was not statistically different between the two study

arms (13.1±11.3% vs. 15.3±11.5%, U-DR vs. U-IR, p=0.53). Among the secondary endpoints, at

3-5 days, mean infarct size normalized to total LV mass was not statistically different between

the groups (16.7±13.3% vs. 19.1±14.0% U-DR vs. U-IR, p=0.58). Infarct size normalized to the

area at risk was 44.2±18.9% versus 51.6±23.6% for the U-DR and U-IR groups respectively

(p=0.28). Mean microvascular obstruction was 1.3% versus 2.7% for the U-DR and U-IR groups

respectively (p=0.22). LV ejection fraction and LV volumes were not statistically different

between the groups at 3-5 days and 30 days. TIMI flow did not correlate with infarct size in the

U-IR and U-DR groups. An analysis of infarct size normalized to the area at risk by CMR and

stratified by ∑STE was performed (Supplemental Table 4 and Supplemental Figure 1).

CMR was performed in 82% (41/50) of patients between days 3 to 5 and in 80% (40/50)

at 30 days of follow up. The primary efficacy endpoint of infarct size normalized to total LV

mass at 30 days was 14.1% (40/50) for the total group. No difference was observed between

groups (13.1±11.3% vs. 15.3±11.5%, U-DR vs. U-IR, p=0.53). Among the secondary and

exploratory endpoints, at 3-5 days, mean infarct size normalized to total LV mass of 17.9±13.5%

and infarct size normalized to the area at risk of 47.9±21.4% were observed for the total group

(n=40; Table 3). Infarct size normalized to the area at risk was not statistically different between

the groups (44.2±18.9% vs. 51.6±23.6%, U-DR vs. U-IR, p=0.28). Mean microvascular

obstruction was 1.3% versus 2.7% for the U-DR and U-IR groups, respectively (p=0.22). LV

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

10.1161/CIRCULATIONAHA.118.038269

12

ejection fraction and LV volumes were not statistically different between the groups at 3-5 days

and 30 days. TIMI flow did not correlate with infarct size in the U-IR and U-DR groups. An

analysis of infarct size normalized to the area at risk by CMR and stratified by ∑STE was

performed (Supplemental Table 4 and Supplemental Figure 1).

Primary Safety Endpoint and Additional Safety Parameters

The incidence of the composite 30-day MACCE outcome was 12% (3/25) and 8% (2/25) for the

U-DR and U-IR groups, respectively (p=0.99) (Table 4). Cardiovascular mortality was 4% (2/50)

in the overall cohort, with 1 death per group. No non-cardiovascular mortality was observed.

One patient had a stroke one day after enrollment (2%) and two patients had major vascular

events (4%) related to flow limiting dissections of the femoral artery at device removal.

BARC ≥2 bleeding was observed in 16% (n=4/25) versus 12% (3/25) for the U-DR and U-IR

groups respectively (p=0.99). No BARC 3C (intracranial), 4 (CABG-related) or 5 (fatal) events

were observed. Blood transfusions were administered to 3 patients (2/25 in the U-DR and 1/25 in

the U-IR), with each patient requiring a single unit of packed red blood cells only. LV thrombus

was observed in 12% (3/25) versus 16% (4/25) for the U-DR and U-IR groups respectively

(p=0.99), and detected by CMR only, and not by pre-discharge echocardiography, in 4 out of 7

patients (2 in each arm). Ventricular tachycardia or fibrillation (VT/VF) was observed in 20%

(5/25) versus 16% (n=4/25) for the U-DR and U-IR groups respectively (p=0.99). Supplemental

Tables 5 and 6 provides details of all additional clinical events.

Discussion

The Impella CP device is currently indicated for both elective and emergent high-risk PCI in the

setting of LV dysfunction. The device is not indicated for STEMI without cardiogenic shock.

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

10.1161/CIRCULATIONAHA.118.038269

13

Furthermore, intentionally delaying reperfusion time in STEMI is not recommended. The

mechanistic preclinical research leading up to the DTU-STEMI pilot study challenge the existing

paradigm of STEMI management by suggesting that LV unloading and delaying reperfusion may

reduce myocardial damage and promote myocardial salvage in anterior STEMI. Before

proceeding with a large randomized controlled trial with 2 or 3 experimental arms involving LV

unloading with or without a delay before reperfusion, preliminary clinical data suggesting that

LV unloading with a planned delay of the DTB time by 30 minutes (U-DR) is feasible and free

from prohibitive adverse events and outcomes was required. The DTU-STEMI pilot study

represents the first human experience of mechanically unloading the LV and intentionally

delaying coronary reperfusion (primary unloading) in anterior STEMI. These findings suggest

for the first time that it is feasible to study whether focusing initially on reducing myocardial

oxygen consumption (unloading) and then restoring coronary reperfusion in anterior STEMI has

clinical benefit. Specifically we report that: 1) LV unloading using the Impella CP device with a

30 minute delay before reperfusion is feasible within a relatively short time period and 2) infarct

size with LV unloading and immediate reperfusion was not different from LV unloading with

delayed reperfusion. The DTU-STEMI pilot trial did not identify any prohibitive safety signals

that would preclude proceeding to a larger pivotal study of LV unloading before reperfusion. An

appropriately powered pivotal trial comparing LV unloading before reperfusion with the current

standard of care is required.

Over the past 30 years, mechanical circulatory support devices have evolved from

surgically implanted pneumatic systems to percutaneously delivered micro-axial flow catheters

that can rapidly reduce ventricular wall stress and myocardial oxygen consumption, increasing

forward flow and thereby decreasing ventricular pressure and volume.18-20 In addition to

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

10.1161/CIRCULATIONAHA.118.038269

14

favorably altering myocardial supply-demand mismatch during an AMI, more recent preclinical

testing suggests that LV unloading and delaying coronary reperfusion by 30 minutes is necessary

to promote myocardial recovery by enhancing activity of kinases known to limit ischemia

reperfusion injury, preserving mitochondrial integrity, increasing activity of cardioprotective

cytokines (ie stromal derived factor one-alpha), and enhancing microcirculatory coronary flow.21-

23 While promising, major limitations to translating these preclinical findings include concerns

regarding the technical safety and feasibility of pump implantation during a STEMI and that

delaying the DTB time may increase myocardial infarct size.24

Multiple attempts to limit infarct size have been tested, however no prior clinical trial has

intentionally extended the delay to reperfusion after initiating a cardioprotective treatment

strategy.25-27 Given the disruptive concept of first unloading the LV and delaying reperfusion,

30-day MACCE was selected as the primary safety endpoint to provide a rigorous and sensitive

analysis of any potential risk associated with the DTU-STEMI strategy. Overall MACCE rates

were not significantly different in the U-IR and U-DR arms, with no events of reinfarction or

prohibitive safety signals. Among the individual MACCE elements, CV mortality was observed

in one patient for each arm of the study and approaches national benchmarks for 30-day STEMI

mortality rates.28 One patient was newly diagnosed with chronic pulmonary fibrosis after

treatment for STEMI. On post-procedure day 3, the patient developed progressive oxygen

dependency requiring pulse steroid therapy and cyclophosphamide without improvement. The

patient eventually developed multi organ failure and expired 24 days after revascularization for

STEMI. The second mortality was an elderly patient who presented in cardiogenic shock that

was detected only after enrollment. The incidence of major vascular events in the DTU-STEMI

study was comparable to the pump arm of the Intra-aortic Balloon Counterpulsation and Infarct

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

10.1161/CIRCULATIONAHA.118.038269

15

Size in Patients with Acute Anterior Myocardial Infarction Without Shock (CRISP-AMI)

study.27 Overall, BARC bleeding >2 in the DTU-STEMI was lower than reported in a recent

analysis of bleeding events involving percutaneous ventricular assist devices (27%)29, and higher

than those reported in other STEMI trials involving drug therapy or intra-aortic balloon pumps

(3%).27 Further study is required to better understand the impact of Impella CP implantation on

bleeding risk that should be assessed in a future pivotal trial. LV thrombus was observed in

several patients with a high rate of detection by CMR only, which is consistent with prior reports

for anterior STEMI.30-32 VT/VF was also observed in several patients and after a review of

optimal insertion techniques and best practices with all investigators, we subsequently noted a

decrease in VT/VF events for the remainder of the study.

A key aspect of testing the feasibility concept was gaining a better understanding of the

time required to establish LV unloading prior to PCI and its impact on door-to-balloon and

overall ischemic time. From the start of the procedure to insertion and activation of the Impella

CP required 15.4±8.4 minutes on average for the total 50 patient study (Table 2). This time

includes prepping, draping, vascular access, left ventriculography and insertion of the Impella

CP device. Furthermore, despite being allowed to provide PCI before completion of the 30

minute delay to reperfusion, all operators complied with the 30 minute delay to reperfusion in the

U-DR arm without a single bailout PCI (Figure 2). These observations highlight key insights

gained from this pilot study including: 1) it is feasible to implant and activate this unloading

device in a timely manner during an anterior STEMI; 2) despite this inherent delay, operators

were able to achieve average door to balloon times of 84.4±27.6 minutes across all 50 patients;

and 3) despite this inherent delay, overall infarct sizes were low relative to recent reports

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

10.1161/CIRCULATIONAHA.118.038269

16

including CRISP-AMI27, 33, 34 and did not correlate with DTB times. These findings support that

the DTU-STEMI strategy may be tested in a larger pivotal trial.

Several limitations to the study include the small number of patients studied, a low

absolute number of adverse events and the lack of a standard-of-care study arm. However, the

objective of this study was to demonstrate feasibility of LV unloading before reperfusion to

support the development of a pivotal trial. Complications noted in the study require further

exploration. A 30-day endpoint may not be sufficient to capture the clinical impact of these

complications, and CMR data were not available for 20% of patients, which limited the analysis.

However, this attrition rate is not unexpected and is similar to the CRISP-AMI study. Clinical

characteristics of patients with and without CMR data are presented in the supplemental material.

A pivotal study will require a larger number of patients, a randomization scheme that includes

standard of care for STEMI, and a longer period of follow up to thoroughly assess the safety and

potential efficacy of LV unloading before reperfusion. Given the small number of patients in this

pilot study, all results should be interpreted as hypothesis generating.

In conclusion, despite the clear importance of reducing infarct size, no therapies

specifically targeting ischemia-reperfusion injury exist. The DTU-STEMI pilot study overcomes

a critical barrier to progress in the field of cardioprotection and myocardial recovery by

suggesting for the first time that it is possible to delay coronary reperfusion, thereby allowing

enough time for LV unloading to precondition the myocardium and reduce ischemia-reperfusion

injury and overall myocardial damage in AMI. We have shown that infarct size did not differ for

patients with unloading who had subsequent immediate versus delayed reperfusion. Furthermore,

we observed that LV unloading and delaying reperfusion is feasible in a relatively short time

interval based on a small patient sample. With these observations in mind, an appropriately

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

10.1161/CIRCULATIONAHA.118.038269

17

powered pivotal trial comparing LV unloading before reperfusion to the current standard of care

in anterior STEMI is in development. Furthermore, the finding that LV unloading and delaying

reperfusion is feasible potentially opens many new doors for further exploration by suggesting

that other adjunct interventions (i.e. pharmacologic) may now be considered during that

‘window-period’ where the coronary artery is occluded, but the LV is unloaded.

Authors

Navin K. Kapur, MD1; Mohamad Alkhouli, MD2; Tony DeMartini, MD3; Haroon Faraz, MD4;

Zachary George, MD5; Mark Goodwin, MD3; Jaime Hernandez-Montfort, MD6;

Vijay Iyer, MD7; Noam Josephy, MD8,9; Sanjog Kalra, MD10; Amir Kaki, MD11;

Richard Karas, MD, PhD1; Carey Kimmelstiel, MD1; Gerald Koeing, MD, PhD12;

Evan Lau, MD6; Kapil Lotun, MD13; Ryan Madder, MD14; Salvatore Mannino, MD15;

Perwaiz Meraj, MD16; Jason Moreland, MD2; Jeffrey Moses, MD17; Raymond Kim, MD18;

Theodore Schreiber, MD11; James Udelson, MD1; Christian Witzke, MD10; David Wohns, MD14;

William O’Neill, MD12

Affiliations

1The CardioVascular Center, Tufts Medical Center, Boston, MA; 2West Virginia University

Heart and Vascular Institute, Morgantown, WV; 3Advocate Heart Institute, Naperville, IL;

4Hackensack University Medical Center, Hackensack, NJ; 5Greenville Memorial Hospital,

Greenville, SC; 6Baystate Medical Center, Springfield, MA; 7University of Buffalo, Buffalo,

NY; 8Abiomed Inc, Danvers, MA; 9Massachusetts Institute of Technology, Cambridge, MA;

10Einstein Medical Center, Philadelphia, PA; 11Detroit Medical Center, Detroit, MI;

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

10.1161/CIRCULATIONAHA.118.038269

18

12Henry Ford Medical Center, Detroit, MI; 13University of Arizona, Phoenix, AZ;

14Spectrum Health, Grand Rapids, MI; 15WellStar Kennestone Hospital, Marietta, GA;

16Northwell Health; Long Island, NY; 17Columbia University, New York, NY;

18Duke University Medical Center, Durham, NC

Sources of Funding

Funding for this study was provided by Abiomed Inc (Danvers, Massachusetts)

Disclosures

Dr. Kapur has received research funding from Abiomed Inc (Danvers, Massachusetts), Cardiac

Assist Inc (Pittsburgh, Pennsylvania), and Maquet Cardiovascular Inc, (Rastatt, Germany).

References

1. Benjamin EJ, Blaha MJ, Chiuve SE, Cushman M, Das SR, Deo R, de Ferranti SD, Floyd

J, Fornage M, Gillespie C, Isasi CR, Jiménez MC, Jordan LC, Judd SE, Lackland

D, Lichtman JH, Lisabeth L, Liu S, Longenecker CT, Mackey RH, Matsushita

K, Mozaffarian D, Mussolino ME, Nasir K, Neumar RW, Palaniappan L, Pandey

DK, Thiagarajan RR, Reeves MJ, Ritchey M, Rodriguez CJ, Roth GA, Rosamond

WD, Sasson C, Towfighi A, Tsao CW, Turner MB, Virani SS, Voeks JH, Willey

JZ, Wilkins JT, Wu JH, Alger HM, Wong SS, Muntner P; American Heart Association

statistics committee and stroke statistics subcommittee. Heart disease and stroke

statistics-2017 update: a report from the American Heart Association. Circulation. 2017;

135:e146-e603.

2. Menees DS, Peterson ED, Wang Y, Curtis JP, Messenger JC, Rumsfeld JS, Gurm HS.

Door-to-balloon time and mortality among patients undergoing primary PCI. N Engl J

Med. 2013; 369:901-909.

3. Stone GW, Selker HP, Thiele H, Patel MR, Udelson JE, Ohman EM, Maehara A, Eitel I,

Granger CB, Jenkins PL, Nichols M, Ben-Yehuda O. Relationship between infarct size

and outcomes following primary PCI: patient-level analysis from 10 randomized trials. J

Am Coll Cardiol 2016; 67:1674-1683.

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

10.1161/CIRCULATIONAHA.118.038269

19

4. Ezekowitz JA, Kaul P, Bakal JA, Armstrong PW, Welsh RC, McAlister FA. Declining

in-hospital mortality and increasing heart failure incidence in elderly patients with first

myocardial infarction. J Am Coll Cardiol. 2009; 53:13-20.

5. Achour H, Boccalandro F, Felli P, Amirian J, Uthman M, Buja M, Smalling RW.

Mechanical left ventricular unloading prior to reperfusion reduces infarct size in a canine

infarction model. Catheter Cardiovasc Interv. 2005; 64:182-192.

6. Meyns B, Stolinski J, Leunens V, Verbeken E, Flameng W. Left ventricular support by

catheter-mounted axial flow pump reduces infarct size. J Am Coll Cardiol. 2003;

41:1087-1095.

7. Saku K, Kakino T, Arimura T, Sunagawa G, Nishikawa T, Sakamoto T, Kishi T, Tsutsui

H, Sunagawa K. Left ventricular mechanical unloading by total support of Impella in

myocardial infarction reduces infarct size, preserves left ventricular function, and

prevents subsequent heart failure in dogs. Circ Heart Fail. 2018; 11:e004397.

8. Kapur NK, Paruchuri V, Urbano-Morales JA, Mackey EE, Daly GH, Qiao X, Pandian N,

Perides G, Karas RH. Mechanically unloading the left ventricle before coronary

reperfusion reduces left ventricular wall stress and myocardial infarct size. Circulation.

2013; 128:328-336.

9. Esposito ML, Zhang Y, Qiao X, Reyelt L, Paruchuri V, Schnitzler GR, Morine KJ,

Annamalai SK, Bogins C, Natov PS, Pedicini R, Breton C, Mullin A, Mackey EE, Patel

A, Rowin E, Jaffe IZ, Karas RH, Kapur NK. Left ventricular unloading

before reperfusion promotes functional recovery after acute myocardial infarction. J Am

Coll Cardiol. 2018; 72:501-514.

10. O'Neill WW, Kleiman NS, Moses J, Henriques JP, Dixon S, Massaro J, Palacios I, Maini

B, Mulukutla S, Dzavík V, Popma J, Douglas PS, Ohman M. A prospective, randomized

clinical trial of hemodynamic support with Impella 2.5 versus intra-aortic balloon pump

in patients undergoing high-risk percutaneous coronary intervention: the PROTECT II

study. Circulation. 2012; 126:1717-1727.

11. O'Neill WW, Schreiber T, Wohns DH, Rihal C, Naidu SS, Civitello AB, Dixon SR,

Massaro JM, Maini B, Ohman EM. The current use of Impella 2.5 in acute myocardial

infarction complicated by cardiogenic shock: results from the USpella Registry. J Interv

Cardiol. 2014; 27:1-11.

12. Flaherty MP, Khan AR, O'Neill WW. Early initiation of Impella in acute myocardial

infarction complicated by cardiogenic shock improves survival: a meta-analysis. JACC

Cardiovasc Interv. 2017; 10:1805-1806.

13. Rodríguez-Palomares JF, Figueras-Bellot J, Descalzo M, Moral S, Otaegui I, Pineda V,

Del Blanco BG, González-Alujas MT, Evangelista Masip A, García-Dorado D. Relation

of ST-segment elevation before and after percutaneous transluminal coronary angioplasty

to left ventricular area at risk, myocardial infarct size and systolic function. Am J Cardiol.

2014; 113:593-600.

14. Patel MR, Worthley SG, Stebbins A, Dill T, Rademakers FE, Valeti US, Barsness GW,

Van de Werf F, Hamm CW, Armstrong PW, Granger CB, Kim RJ. Pexelizumab and

infarct size in patients with acute myocardial infarction undergoing primary PCI: a

delayed enhancement cardiac magnetic resonance substudy from the APEX-AMI trial.

JACC Cardiovasc Imaging. 2010; 3:52-60.

15. Melloni C, Rao SV, Povsic TJ, Melton L, Kim RJ, Kilaru R, Patel MR, Talan M, Ferrucci

L, Longo DL, Lakatta EG, Najjar SS, Harrington RA. Design and rationale of the

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

10.1161/CIRCULATIONAHA.118.038269

20

Reduction of Infarct Expansion and Ventricular Remodeling with Erythropoietin After

Large Myocardial Infarction (REVEAL) trial. Am Heart J. 2010; 160:795-803.

16. Reimer KA, Jennings RB. The “wavefront phenomenon” of myocardial ischemic cell

death. II. Transmural progression of necrosis within the framework of ischemic bed size

(myocardium at risk) and collateral flow. Lab Invest 1979; 40:633–644.

17. Versteylen MO, Bekkers SC, Smulders MW, Winkens B, Mihl C, Winkens MH, Leiner

T, Waltenberger JL, Kim RJ, Gorgels AP. Performance of angiographic,

electrocardiographic, and MRI methods to assess the area at risk in acute myocardial

infarction. Heart. 2012; 98:109-115.

18. Burkhoff D, Sayer G, Doshi D, Uriel N. Hemodynamics of mechanical circulatory

support. J Am Coll Cardiol. 2015; 66:2663-2674.

19. Uriel N, Sayer G, Annamalai S, Kapur NK, Burkhoff D. Mechanical unloading in heart

failure. J Am Coll Cardiol. 2018; 72:569-580.

20. Kapur NK, Paruchuri V, Pham DT, Reyelt L, Murphy B, Beale C, Bogins C, Wiener D,

Nilson J, Esposito M, Perkins S, Perides G, Karas RH. Hemodynamic effects of left atrial

or left ventricular cannulation for acute circulatory support in a bovine model of left heart

injury. ASAIO J. 2015; 61:301-306.

21. Kapur NK, Qiao X, Paruchuri V, Morine KJ, Syed W, Dow S, Shah N, Pandian N, Karas

RH. Mechanical pre-conditioning with acute circulatory support before reperfusion limits

infarct size in acute myocardial infarction. JACC Heart Fail. 2015; 3:873-882.

22. Hausenloy DJ, Yellon DM. Myocardial ischemia-reperfusion injury: a neglected

therapeutic target. J Clin Invest. 2013; 123:92-100.

23. Watanabe S, Fish K, Kovacic JC, Bikou O, Leonardson L, Nomoto K, Aguero J, Kapur

NK, Hajjar RJ, Ishikawa K. Left ventricular unloading using an Impella CP improves

coronary flow and infarct zone perfusion in ischemic heart failure. J Am Heart Assoc.

2018; 7:e006462.

24. Bates ER. Limiting infarct size in ST-Segment myocardial infarction: the holy grail of

reperfusion therapy. JACC Heart Fail. 2015; 3:883-885.

25. Hausenloy, D. J. & Yellon, D. M. Targeting myocardial reperfusion injury--the search

continues. N Engl J Med. 2015; 373:1073-1075.

26. Cung TT, Morel O, Cayla G, Rioufol G, Garcia-Dorado D, Angoulvant D, Bonnefoy-

Cudraz E, Guérin P, Elbaz M, Delarche N, Coste P, Vanzetto G, Metge M, Aupetit JF,

Jouve B, Motreff P, Tron C, Labeque JN, Steg PG, Cottin Y, Range G, Clerc J, Claeys

MJ, Coussement P, Prunier F, Moulin F, Roth O, Belle L, Dubois P, Barragan P, Gilard

M, Piot C, Colin P, De Poli F, Morice MC, Ider O, Dubois-Randé JL, Unterseeh T, Le

Breton H, Béard T, Blanchard D, Grollier G, Malquarti V, Staat P, Sudre A, Elmer E,

Hansson MJ, Bergerot C, Boussaha I, Jossan C, Derumeaux G, Mewton N, Ovize M..

Cyclosprine before PCI in patients with acute myocardial infarction. N Engl J Med. 2015;

373:1021-1031.

27. Patel MR, Smalling RW, Thiele H, Barnhart HX, Zhou Y, Chandra P, Chew D, Cohen

M, French J, Perera D, Ohman EM. Intra-aortic balloon counterpulsation and infarct size

in patients with acute anterior myocardial infarction without shock: the CRISP AMI

randomized trial. JAMA. 2011; 306:1329-1337.

28. Peterson ED, Dai D, DeLong ER, Brennan JM, Singh M, Rao SV, Shaw RE, Roe MT,

Ho KK, Klein LW, Krone RJ, Weintraub WS, Brindis RG, Rumsfeld JS, Spertus

JA; NCDR Registry Participants. Contemporary mortality risk prediction for

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

10.1161/CIRCULATIONAHA.118.038269

21

percutaneous coronary intervention: results from 588,398 procedures in the National

Cardiovascular Data Registry. J Am Coll Cardiol. 2010; 55:1923-1932.

29. Redfors B, Watson BM, McAndrew T, Palisaitis E, Francese DP, Razavi M, Safirstein J,

Mehran R, Kirtane AJ, Généreux P. Mortality, length of stay, and cost implications of

procedural bleeding after percutaneous interventions using large-bore catheters. JAMA

Cardiol. 2017; 2: 798-802.

30. McCarthy CP, Vaduganathan M, McCarthy KJ, Januzzi JL Jr, Bhatt DL, McEvoy JW.

Left ventricular thrombus after acute myocardial infarction: Screening, prevention, and

treatment. JAMA Cardiol. 2018; 3: 642-649.

31. Solheim S, Seljeflot I, Lunde K, Bjørnerheim R, Aakhus S, Forfang K, Arnesen H.

Frequency of left ventricular thrombus in patients with anterior wall acute myocardial

infarction treated with percutaneous coronary intervention and dual antiplatelet therapy.

Am J Cardiol. 2010; 106:1197-1200.

32. Delewi R, Nijveldt R, Hirsch A, Marcu CB, Robbers L, Hassell ME, de Bruin RH,

Vleugels J, van der Laan AM, Bouma BJ, Tio RA, Tijssen JG, van Rossum AC, Zijlstra

F, Piek JJ. Left ventricular thrombus formation after acute myocardial infarction as

assessed by cardiovascular magnetic resonance imaging. Eur J Radiol. 2012; 81: 3900-

3904.

33. Ibanez B, Macaya C, Sánchez-Brunete V, Pizarro G, Fernández-Friera L, Mateos A,

Fernández-Ortiz A, García-Ruiz JM, García-Álvarez A, Iñiguez A, Jiménez-Borreguero

J, López-Romero P, Fernández-Jiménez R, Goicolea J, Ruiz-Mateos B, Bastante T, Arias

M, Iglesias-Vázquez JA, Rodriguez MD, Escalera N, Acebal C, Cabrera JA, Valenciano

J, Pérez de Prado A, Fernández-Campos MJ, Casado I, García-Rubira JC, García-Prieto J,

Sanz-Rosa D, Cuellas C, Hernández-Antolín R, Albarrán A, Fernández-Vázquez F, de la

Torre-Hernández JM, Pocock S, Sanz G, Fuster V. Effect of early metoprolol on infarct

size in ST-segment-elevation myocardial infarction patients undergoing primary

percutaneous coronary intervention: the Effect of Metoprolol in Cardioprotection During

an Acute Myocardial Infarction (METOCARD-CNIC) trial. Circulation. 2013; 128:

1495-1503.

34. Er F, Dahlem KM, Nia AM, Erdmann E, Waltenberger J, Hellmich M, Kuhr K, Le MT,

Herrfurth T, Taghiyev Z, Biesenbach E, Yüksel D, Eran-Ergöknil A, Vanezi M, Caglayan

E, Gassanov N. Randomized control of sympathetic drive with continuous intravenous

esmolol in patients with acute ST-Segment elevation myocardial infarction: The BEtA-

Blocker Therapy in Acute Myocardial Infarction (BEAT-AMI) Trial. JACC Cardiovasc

Interv. 2016; 9: 231-240.

D

ownloaded from

http://ahajournals.org by on Novem

ber 24, 2018

22

Table 1. Baseline Characteristics

U-DR U-IR

n=25 n=25

Age, mean (±SD), years 60.6 (10.7) 58.8 (11.4)

Male, n (%) 21 (84.0) 17 (68.0)

Race, n (%)

American Indian/Alaskan Native 1 (4.0) 0 (0.0)

Asian 2 (8.0) 5 (20.0)

Black or African American 4 (16.0) 4 (16.0)

White or Caucasian 18 (72.0) 16 (64.0)

Other 0 (0.0) 0 (0.0)

BMI, mean (±SD), kg/m2 30.0 (6.0) 29.6 (9.8)

Height, mean (±SD), cm 175.4 (8.5) 169.8 (15.3)

Weight, mean (±SD), kg 92.8 (21.1) 83.7 (19.6)

Medical history, n (%)

Hypertension (receiving drug therapy) 14 (56.0) 12 (48.0)

Stroke 0 (0.0) 1 (4.0)

Transient ischemic attack 1 (4.0) 1 (4.0)

Current nicotine use 8 (32.0) 5 (20.0)

Dyslipidemia (receiving drug therapy) 9 (36.0) 14 (56.0)

Renal insufficiency 0 (0.0) 0 (0.0)

Diabetes mellitus 6 (24.0) 4 (28.6)

Prior peripheral arterial disease 0 (0.0) 0 (0.0)

At presentation

Blood pressure, mean (±SD), mmHg

Systolic 149 (34) 157 (26)

Diastolic 88 (15) 95 (19)

MAP 108 (20) 116 (20)

Heart rate, mean (±SD), beats/min 89 (22) 87 (16)

Pre-Impella LVEDP n=22 n=23

LVEDP, mean (±SD), mmHg 25.0 (9.6) 24.0 (8.1)

Heart rate, mean (±SD), bpm 87.9 (19.6) 76.5 (13.6)

Anterior ST Elevation Sum, mm, n (%), n=25 n=25

0-<2 0 (0. 0) 2 (8.0)

2-<4 1 (4.0) 2 (8.0)

4-<6 3 (12.0) 2 (8.00)

≥6 21 (84.0) 19 (76.0)

Baseline LVEF n=22 n=23

LVEF, mean (±SD), % 32.7 (12.7) 41.9 (12.3)

U-DR, delayed reperfusion after 30 minutes of LV unloading by the Impella CP; U-IR, immediate

reperfusion after placement of the Impella CP; CP, Cardiac Power; SD, standard deviation; BMI, body

mass index; MAP, mean arterial pressure; LV, left ventricle; LVEDP, left ventricular end diastolic

pressure; LVEF, left ventricular ejection fraction.

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

10.1161/CIRCULATIONAHA.118.038269

23

Table 2. Timing Elements

U-DR U-IR P value

Total Door to Balloon n=24* n=25

DTB, mean (±SD), minutes 96.7 (26.1) 72.6 (24.0) 0.002

DTB, median (IQR), minutes 98.0 (76.0-112.5) 68.0 (55.0-87.0)

Door to Unload n=25 n=25

Door to Unload, mean (±SD), minutes 64.0 (25.5) 62.1 (24.2) 0.79

Door to Unload, median (IQR), minutes 63.0 (43.0-83.5) 56.0 (42.0-79.0)

Symptom to Unload n=25 n=25

Symptom to Unload, mean (±SD), minutes 176.2 (73.4) 200.2 (151.8) 0.48

Symptom to Unload, median (IQR), minutes 153.0 (119.0-196.0) 174.0 (124.0-223.0)

Procedure Start to Impella Insertion n=25 n=25

Insertion Time, mean (±SD), minutes 15.1 (7.9) 15.8 (9.0) 0.78

Insertion Time, median (IQR), minutes 15.0 (8.0-20.0) 15.0 (12.0-20.0)

Unload to Coronary Balloon n=24* n=25

Unload to Balloon, mean (±SD), minutes 34.1 (2.6) 10.5 (6.7) <0.001

Unload to Balloon, median (IQR), minutes 34.0 (32.0-35.3) 10.0 (5.0-12.0)

Duration of Support n=25 n=25

Duration of Support, mean (±SD), hours 8.2 (7.9) 5.2 (3.9) 0.10

Duration of Support, median (IQR), hours 3.9 (3.4-11.9) 4.2 (3.4-4.9)

*One patient in the U-DR arm did not have a PCI. U-DR, delayed reperfusion after 30 minutes of LV

unloading by the Impella CP; U-IR, immediate reperfusion after placement of the Impella CP; DTB,

door-to-balloon; IQR, interquartile range; SD, standard deviation; CP, cardiac power; LV, left ventricle;

PCI, percutaneous coronary intervention.

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

10.1161/CIRCULATIONAHA.118.038269

24

Table 3. CMR Results

30 Days

U-DR U-IR P value

Infarct Size, Number assessed, n (%) 21 (84.0) 19 (76.0)

mean (±SD), % 13.1 (11.3) 15.3 (11.5) 0.53

median (IQR),% 10.4 (5.0-26.1) 13.0 (3.8-22.9)

LVEF, Number assessed, n (%) 21 (84.0) 19 (76.0)

mean (±SD), % 49.2 (12.9) 48.5 (13.4) 0.87

median (IQR), % 47.4 (39.4-61.0) 47.2 (35.9-59.9)

LVESV, Number assessed, n (%) 20 (80.0) 19 (76.0)

mean (±SD), ml 76.0 (43.9) 78.3 (34.4) 0.86

median (IQR), ml 69.0 (38.7-100.6) 65.5 (50.2-106.0)

LVEDV, Number assessed, n (%) 20 (80.0) 19 (76.0)

mean (±SD), ml 140.9 (50.8) 147.9 (37.1) 0.63

median (IQR), ml 147.8 (97.8-171.6) 149.6 (119.2-171.0)

3 to 5 Days

U-DR* U-IR P value

Infarct Size, Number assessed, n (%) 20 (80.0) 20 (80.0)

mean (±SD), % 16.7 (13.3) 19.1 (14.0) 0.58

median (IQR), % 15.2 (6.7-23.9) 15.3 (7.4-30.5)

Infarct/AAR, Number assessed, n (%) 20 (80.0) 20 (80.0)

mean (±SD), % 44.2 (18.9) 51.6 (23.6) 0.28

median (IQR), % 47.3 (28.8-59.5) 57.1 (38.4-71.8)

MVO, Number assessed, n (%) 20 (80.0) 20 (80.0)

mean (±SD), % 1.3 (2.7) 2.7 (4.4) 0.22

median (IQR), % 0.0 (0.0-1.7) 0.7 (0.0-3.0)

Salvage Index, Number assessed, n (%) 20 (80.0) 20 (80.0)

mean (±SD), % 55.8 (18.9) 48.4 (23.6) 0.28

median (IQR), % 52.7 (41.3-71.0) 43.0 (28.2-59.8)

LVEF, Number assessed, n (%) 21 (84.0) 20 (80.0)

mean (±SD), % 44.7 (9.2) 46.2 (14.3) 0.69

median (IQR), % 45.0 (37.9-52.2) 47.3 (32.6-54.9)

LVESV, n (%) 20 (80.0) 20 (80.0)

mean (±SD), ml 82.7 (39.4) 78.3 (30.1) 0.69

median (IQR), ml 79.6 (60.6-93.8) 80.8 (57.8-102.9)

LVEDV, n (%) 20 (80.0) 20 (80.0)

mean (±SD), ml 145.1 (47.8) 143.1 (32.5) 0.88

median (IQR), ml 143.9 (118.0-167.4) 143.6 (123.7-167.4)

*A patient in the U-DR arm had a test without contrast, the core laboratory could only read the LVEF.

CMR, cardiac magnetic resonance; U-DR, delayed reperfusion after 30 minutes of LV unloading by the

Impella CP; U-IR, immediate reperfusion after placement of the Impella CP; SD, standard deviation;

IQR, interquartile range; LVEF, left ventricular ejection fraction; LVESV, left ventricular end systolic

volume; LVEDV, left ventricular end diastolic volume; AAR, area at risk; MVO, microvascular

obstruction; Salvage Index = [1-(Infarct Size/AAR)]*100; CP, cardiac power; LV, left ventricle.

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

10.1161/CIRCULATIONAHA.118.038269

25

Table 4. MACCE at 30 Days U-DR

95% CI U-IR

95% CI P Value n=25 n=25

MACCE, n (%) 3 (12%) [2.55%,31.22%] 2 (8%) [0.98%,26.03%] 0.99

CV Mortality, n (%) 1 (4%) [0.10%,20.35%] 1 (4%) [0.10%,20.35%] 0.99

Reinfarction, n (%) 0 (0%) [0.00%,13.72%] 0 (0%) [0.00%,13.72%] -

Stroke/TIA, n (%) 0 (0%) [0.00%,13.72%] 1 (4%) [0.10%,20.35%] 0.99

Major Vascular Events, n (%) 2 (8%) [0.98%,26.03%] 0 (0%) [0.00%,13.72%] 0.49

MACCE, major adverse cardiac and cerebrovascular events; U-DR, delayed reperfusion after 30 minutes

of LV unloading by the Impella CP; U-IR, immediate reperfusion after placement of the Impella CP; CV,

cardiovascular; TIA, transient ischemic attack; CI, confidence interval; LV, left ventricle; CP, cardiac

power.

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

10.1161/CIRCULATIONAHA.118.038269

26

Figure Legends

Figure 1. Study Design and Flowchart

*One patient had an incomplete scan. U-DR, delayed reperfusion after 30 minutes of LV

unloading by the Impella CP; U-IR, immediate reperfusion after placement of the Impella CP;

CP, cardiac power; LV, left ventricle; CMR, cardiac magnetic resonance; IUD, intrauterine

device; PCI, percutaneous coronary intervention; MACCE, major adverse cardiac and

cerebrovascular events.

Figure 2. Scatter Plot of LV Unloading to Balloon Times. The time from Impella CP

activation to balloon inflation across the 2 arms of the study shows full compliance with the

protocolized delay to balloon time of 30 minutes and no bailout PCI in the U-DR group. LV, left

ventricle; CP, cardiac power; PCI, percutaneous coronary intervention; U-DR, delayed

reperfusion after 30 minutes of LV unloading by the Impella CP; U-IR, immediate reperfusion

after placement of the Impella CP.

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018

Dow

nloaded from http://ahajournals.org by on N

ovember 24, 2018