Abstract
Overview and Mission of the A-CURE Working Group
The first annual symposium of the Acute Cardiac Unloading and Recovery (A-CURE) meeting was held in Rome, Italy, on August 26, 2016. The mission of A-CURE is to advance the science and mechanistic understanding of acute cardiac unloading and to support the translation of basic and clinical research into therapies aimed at heart muscle recovery. Heart failure is a global cause of morbidity and mortality. Despite major advances in pharmacologic therapy, the number of individuals diagnosed with heart failure is growing exponentially and is expected to reach >8 million individuals (1 in every 33) with an estimated medical cost of US$53 billion in the United States alone by 2030. Over the past 3 decades, major advances in mechanical circulatory support technology have led to a revolution in the use of these devices for acute myocardial infarction, cardiogenic shock, and advanced heart failure. The primary objective of these support pumps is to mechanically unload the heart. Scientists from around the globe have embraced these devices as a potentially important platform for myocardial recovery. The plenary lectures by Dr. Eugene Braunwald and Dr. Joseph Hill identified the problem of myocardial recovery and opened the door for more aggressive research in this arena. The abstracts and lectures presented during this inaugural meeting of the A-CURE include fundamental basic, translational, and clinical science focused on acute cardiac unloading and myocardial recovery. We anticipate that future meetings will continue to support the development of novel therapeutic approaches to address the critical problem of heart failure around the world.
Navin Kapur, MD, Tufts Medical Center
Daniel Burkhoff, MD, PhD, Columbia University
Mark Anderson, MD, Einstein Medical Center
Co-Chairs, The A-CURE Working Group, www.A-CURE.org
From Door to Balloon to Door to Unload: A Transcriptomic Analysis Identifies That Primary Unloading Globally Shifts Gene Expression and Preserves Mitochondrial Integrity Within the Infarct Zone During the Acute Phase of AMI
Michele L. Esposito, Yali Zhang, Gavin Schnitzler, Vikram Paruchuri, Xiaoying Qiao, Shiva Annamalai, Emily Mackey, Kevin Morine, Andrew Mullin, Robert Pedicini, Catalina Breton, Richard Karas, and Navin K. Kapur
Tufts University Medical Center
We recently reported that compared to primary reperfusion (PR), first reducing myocardial oxygen demand by activating an acute mechanical circulatory support (AMCS) device while delaying coronary reperfusion (primary unloading [PU]) reduces myocardial damage in models of acute myocardial infarction (AMI). We now employed a transcriptomics-directed approach to identify the underlying mechanisms by which PU reduces myocardial infarct size. Acute myocardial infarction was induced by occlusion of the left anterior descending artery (LAD) for 90 minutes in male swine (n = 4/group). In the PR group, the LAD was reperfused for 120 minutes. In the PU group, after 90 minutes of ischemia, an AMCS device was activated and the LAD occluded for an additional 30 minutes, followed by 120 minutes of reperfusion. Myocardial infarct size was quantified by 2,3,5-Triphenyltetrazolium chloride (TTC) staining. Whole-transcript expression analysis was performed on RNA from the infarct zone using Porcine 1.0 ST microarrays and ConsensusPathDB pathway/ontology programs. Quantitative polymerase chain reaction confirmed expression of select genes from regulated pathways. Scanning electron microscopy (SEM) evaluated mitochondrial integrity within infarct zones. Sham-operated LV samples served as controls. Compared to PR, PU reduced myocardial infarct size (65% vs 34%, infarct size/area at risk; P < .01). Principle component analysis of the transcriptomes showed that infarct tissue from the PU group clustered separately from the PR and Sham groups. Compared to PR, pathway analysis indicated that PU triggered reduced inflammation, reduced fibrosis, and increased metabolic enzyme expression and cellular respiration. The SEM showed a higher ratio of intact mitochondria/cardiomyocyte area within the infarct zone after PU, not PR. We identified for the first time that primary left ventricular unloading triggers a global shift in gene expression and preserves mitochondrial integrity within the infarct zone during the acute phase of AMI. Further studies exploring the clinical utility of PU are required.
Neuromechanical Unloading for Acute Myocardial Ischemia Synergistically, Strikingly Reduces Infarct Size, and Prevents Heart Failure in the Long Term
Takahiro Arimura, Keita Saku, Takamori Kakino, Takafumi Sakamoto, Takuya Akashi, Yoshinori Murayama, Takuya Nishikawa, Genya Sunagawa, Takeshi Tohyama, Tomomi Ide, and Kenji Sunagawa
Kyushu University
Despite the widespread use of early coronary reperfusion therapy, acute myocardial infarction (AMI) remains a leading cause of chronic heart failure. Therefore, there remains a clear unmet need to prevent the post-MI heart failure. We have recently shown in AMI that left ventricular assist device (LVAD) markedly reduced the pressure–volume area and myocardial oxygen consumption and resulted in marked reduction in infarct size. Recent investigations also indicated that vagal nerve stimulation (VNS) has powerful anti-infarct, anti-heart failure effect through complex mechanisms. We hypothesized that the combination of mechanical unloading with VNS, that is, neuromechanical unloading, would have synergistic, powerful anti-infarct effects and thereby prevent heart failure in the long term. The purpose of this investigation is to examine the impact of neuromechanical unloading of AMI on the cardiac function and infarct size in the long term. In 33 dogs, we ligated the left anterior descending coronary artery for 180 minutes and then reperfused. We performed partial LVAD support (pLVAD, LV output equals LVAD output) or total LVAD support (tLVAD, no LV output) with percutaneous transvascular LVAD (Impella). For VNS, we transvenously stimulated the right vagus nerve in the superior vena cava (−Δ heart rate = 31 ± 7 bpm). We started the neuromechanical unloading 60 minutes after the onset of ischemia to 60 minutes after reperfusion. Four weeks after AMI, we compared cardiac function and infarct size among 6 groups, ischemia/reperfusion (I/R; no treatment, n = 6), pLVAD (n = 6), tLVAD (n = 6), VNS (n = 5), pLVAD + VNS (n = 5), and tLVAD + VNS (n = 5). The tLVAD + VNS nearly normalized LV ejection fraction (2D echo) (I/R: 48% ± 4%; pLVAD + VNS: 62% ± 3%; tLVAD + VNS: 64% ± 3%; P < .05) and end-diastolic pressure (I/R: 13.6 ± 0.6; pLVAD + VNS: 3.7 ± 1.2; tLVAD + VNS: 5.5 ± 2.3 mm Hg; P < .05). Histological examination indicated that tLVAD + VNS markedly reduced infarct size (I/R: 14.8% ± 4.4%; pLVAD + VNS: 2.9% ± 1.3%; tLVAD + VNS: 1.7% ± 0.5%; P < .05). Neuromechanical unloading synergistically and strikingly reduces the infarct size and prevents the deterioration of cardiac function and heart failure in the long term.
Impella Unloading Combined With Bradycardic Agent Synergistically Suppresses Myocardial Oxygen Consumption and Strikingly Reduces Infarct Size in Acute Myocardial Infarction
Keita Saku, Genya Sunagawa, Takahiro Arimura, Takamori Kakino, Takuya Akashi, Yoshinori Murayama, Takafumi Sakamoto, Takuya Kishi, and Kenji Sunagawa
Kyushu University
Reducing myocardial oxygen consumption (MVO2) in acute myocardial infarction (AMI) lessens ischemia and reduces infarct size. Since left ventricular (LV) pressure–volume area (PVA) and heart rate (HR) are the major determinant of MVO2, we hypothesized that the combination of transvascular LV assist device (Impella), which reduces LV-PVA, and bradycardic agent, Ivabradine (IVA, selective if channel inhibitor), reduces infarct size via the powerful synergistic suppression of MVO2 in AMI. In 4 mongrel dogs, we occluded the left circumflex coronary artery for 1 hour, measured PVA and MVO2 (Fick’s rule), and compared them among 3 conditions; Control, Impella unloading, and the combination of Impella and IVA (1 mg/kg, intravenous). Impella significantly reduced PVA (−25.0% ± 36.4%) and MVO2 (−34.0% ± 27.3%, vs Control). Concomitant administration of IVA markedly slowed HR (−43% ± 8%, vs Control) and further reduced MVO2 (−62.0% ± 15.5%, vs Control) without compromising hemodynamics. In 14 mongrel dogs, we ligated coronary arteries for 180 minutes and then reperfused. We started Impella and IVA 1 hour after the onset of ischemia. We allocated animals into 3 groups; Control (n = 4, no Impella or IVA treatment), Impella (n = 5), and Impella + IVA (n = 5), and compared the infarct size (relative to the area at risk) 3 hours after reperfusion. Although Impella significantly reduced infarct size, the impact was by far larger in the combination therapy than in Impella (Control: 55.4% ± 5.6%; Impella: 40.3% ± 14.7%; Impella + IVA: 23.9% ± 10.6%; P < .01). The combination therapy of Impella and IVA, which synergistically decreases MVO2 in the acute phase of AMI, markedly reduces infarct size without compromising hemodynamics. Since both therapeutic modalities have been clinically approved in many countries, and thus they are readily available in clinical, the combination therapy may serve as a powerful therapeutic option to reduce infract size and in turn to prevent the future development of heart failure.
Early Unloading With Impella Support in Acute MI With Cardiogenic Shock Linked to Decreased Infarct Size, In-Hospital Mortality, and Hospital Length of Stay
Shailendra Singh, Yonathan Litwok, Alexander Lee, and Evelina Grayver
Hofstra Northwell School of Medicine
Acute MI (AMI) complicated by cardiogenic shock (CS) continues to yield a devastating challenge with mortality rates approaching 50%. Hemodynamic stabilization before and/or after early revascularization remains the primary goal in these patients. Guidelines discourage the first-line use of mechanical support in patients with CS and emphasize refractory shock as the indication. We hypothesize that by decreasing the time to Impella in AMI with CS, it will reduce myocardial oxygen demand and support systemic hemodynamics earlier, which will subsequently decrease the prolonged need for ionotropic/vasopressor support, improve stability during angioplasty/intervention, decrease length of stay (LOS), and decrease in-hospital mortality. We performed a retrospective, multicenter observational study of 71 consecutive patients who presented with AMI from January 2013 to May 2016 who were implanted with an Impella device. All patients were taken to the cardiac catheterization laboratory for primary percutaneous coronary intervention and received appropriate guideline-based medical therapies. The study consisted of 71 patients presenting to our institutions with AMI with CS. Patients who received hemodynamic support with intra-aortic balloon pump (IABP), Impella, or both mechanical support devices were implanted at varied times within their hospitalization. Patients who had Impella support placed earlier in their presentation were found to have quicker hemodynamic improvement, decreased infarct size, and decreased LOS. Patients who were initially supported with only IABP and were transitioned to Impella benefitted from increased cardiac power and improved outcomes. The present study is the only study to date to demonstrate the fact that earlier implantation of Impella support in AMI complicated by CS improved cardiovascular outcomes and length of stay. Delayed or lack of judicious use of mechanical hemodynamic support in the setting AMI with CS continues to yield unfavorable outcomes. More research with larger enrollments on the benefits of early unloading in the setting of AMI with CS need to be completed.
Improving Left Ventricular Unloading Following Prolonged Cardiac Arrest Using a Minimally Invasive Left Ventricular Assist Device: A Prospective Animal Study in Pigs
A Ebeling, R Zayat, M Fries, and M Derwall
University Hospital RWTH Aachen
Cardiopulmonary resuscitation (CPR) is an emergency procedure deployed when a patient suffers cardiac arrest (CA). Compared to conventional CPR, improved survival is observed after CA when CPR is conducted in the presence of a minimally invasive left ventricular assist device, Impella 2.5 (iCPR). However, data on myocardial function during and following iCPR are lacking. The aim is to assess cardiac functional parameters during and following iCPR. Five 55.2 ± 2.4 kg pigs were anesthetized, intubated, and implanted with an Impella 2.5. Ventricular fibrillation was electrically induced and left untreated for 9 minutes before defibrillation was attempted following 6 minutes of iCPR. During iCPR, the Impella device was set to the maximally achievable flow. One hour following return of spontaneous circulation (ROSC), mild therapeutic hypothermia was induced for 16 hours using a total of 4 liters of 6°C cold saline infusions and ice bags. To assess myocardial recovery, we used 2-dimensional echocardiography, tissue Doppler (TDI), and speckle tracking. All animals received transesophageal echocardiography at baseline, during untreated CA, at the initiation of iCPR, 30 minutes, and 5 hours following ROSC. Left ventricular (LV) systolic parameters returned to baseline values 5 hours after ROSC (global longitudinal strain: −25% ± 4.3% vs −20% ± 2.7%; P = .388; ejection fraction [%]: 64 ± 8.8 vs 61.32 ± 10.3; P = .971; stroke volume index (mL/m2): 28.32 ± 8.9 vs 24.71 ± 12.86; P = .545). The LV volume unloading was also observed over the same time period. The LV end-diastolic volume was 55.38 ± 2.8 mL at baseline, peaked after CA at 64.7 ± 9.9 mL, fell to 45.69 ± 7.4 mL 30 minutes after initiation of iCPR, and was maintained at 49.46 ± 13.9 mL 5 hours after ROSC. Recovery of the RV systolic parameters was not observed during the first 5 hours following ROSC (baseline vs 5 hours after ROSC: TDI-derived tricuspid annular systolic velocity (TASV) (cm/s): 11.6 ± 1 vs 8.5 ± 1; P = .005, right ventricular (RV) fractional area change (%): 42 ± 6.2 vs 33 ± 6.9; P = .006). The iCPR is able to achieve a full recovery of LV systolic parameters and provides sufficient LV volume unloading. The observed RV distension is likely attributable to extensive volume loading. Further studies are needed to analyze long term LV and RV function following CA and iCPR support.
Relationship Between Therapeutic Effects on Infarct Size in Acute MI and 1-Year Outcomes: Patient-Level Analysis of Randomized Clinical Trials
James Udelson, Harry P. Selker, Robin Ruthazer, Holger Thiele, Manesh R. Patel, E. Magnus Ohman, Akiko Maehara, Paul L. Jenkins, Melissa Nichols, Ori Ben-Yehuda, and Gregg W. Stone
Tufts University Medical Center
Infarct size (IS) after myocardial infarction (MI) is related to long-term outcomes. Whether a change in IS from an intervention is related to the intervention’s effect on outcomes is unknown. A therapy-induced change in IS is related in direction and/or magnitude to the outcome effect of that therapy. We combined patient-level data from 10 randomized clinical trials of therapies for ST-segment elevation myocardial infarction (STEMI). The IS was assessed by sestamibi imaging or cardiac magnetic resonance with analysis in core labs. Each patient within a trial was assigned a variable to represent a treatment’s mean effect on IS. Cox proportional hazards models estimated the association of treatment-related IS to 1-year adjudicated clinical outcomes of hospitalization for heart failure (HF) and all-cause mortality. The 10 trials included 2458 patients, 24% women. The IS was measured at median 5 days post-MI. Mean trial IS in the control groups in the 10 trials ranged from 12% to 35% of the LV and from 12% to 40% among treatment groups. There was a significant relation of treatment effect on IS to treatment effect on 1-year HF hospitalization (heart rate [HR]: 0.83; 95% confidence interval [CI]: 0.75-0.93; P < .001). There was no significant relation between treatment effect on IS to treatment effect on 1-year mortality (HR: 1.04; 95% CI: 0.94-1.15). The relation to HF hospitalization was stable in sensitivity analyses adjusting for time from MI to IS assessment, and for considering HF as the main outcome and death as a competing risk. This patient-level analysis of randomized placebo-controlled trials of multiple therapeutics for STEMI suggests that a treatment-induced effect on IS is related in direction and quantifiable magnitude to a treatment effect on HF hospitalizations. The data enable the consideration of incorporating IS assessment into novel trial analytic approaches as a surrogate end point to assess new therapeutics.
LV Unloading Using an Impella CP Reduces Wall Stress and Improves Coronary Flow and Perfusion in Infarcted Myocardium
Kiyotake Ishikawa, Shin Watanabe, Lauren Leonardson, Kenneth Fish, and Roger J Hajjar
Icahn School of Medicine at Mt. Sinai
Left ventricular (LV) unloading may improve coronary perfusion by increasing cardiac output and reducing LV wall stress. Whether sustained LV unloading improves myocardial perfusion in a postmyocardial infarction (MI) setting remains uncertain. Unloading a post-MI LV with an Impella CP increases coronary flow and myocardial perfusion by reducing LV wall stress and increasing cardiac output. To mimic patients who need LV support, large anterior transmural MI was induced by occluding the proximal left anterior descending artery (LAD) for 90 minutes in Yorkshire pigs (n = 5, 40-50 kg) followed by a thrombus injection through the balloon lumen to induce total occlusion of the LAD. Two-weeks after the MI, animals underwent LV unloading with an Impella CP for 120 minutes. Epicardial coronary flow was assessed by coronary flow wire before, 5 minutes, and 120 minutes after LV unloading. Myocardial perfusion was assessed using fluorescent microspheres before and 120 minutes after LV unloading. Coronary angiograms revealed TIMI 2 flow in 4 animals and TIMI 3 flow in 1 animal. The LV unloading with maximal pump support (P8) for 2 hours resulted in an increase in total cardiac output (3.08-3.93 L/min; P = .07). Impella support significantly reduced end-diastolic volume (109 ± 17-85 ± 12 mL; P = .02) and end-diastolic pressure (29.3 ± 5.6 to 19.2 ± 6.9 mm Hg) resulting in a significant decrease in LV end-diastolic wall stress (infarct: 34.4 ± 5.5 to 20.2 ± 5.2 kdynes/cm2; P = .03; remote: 32.4 ± 8.6 to 19.3 ± 6.0 kdynes/cm2; P = .03). Coronary flow increased acutely (LAD: 6.7 ± 1.8 to 10.2 ± 1.5 cm/s, P = .03; LCx: 8.4 ± 2.7 to 10.4 ± 3.6 cm/s, P = .31) and remained elevated at 120 minutes (LAD: 9.8 ± 1.3 cm/s, P = .058; LCx: 12.4 ± 4.2 cm/s, P = .058). Compared to baseline, myocardial perfusion as measured by fluorescent microspheres within the infarct zone was significantly increased (87% ± 67%; P = .02), whereas perfusion of the remote nonischemic myocardium was similar compared to the baseline (−2% ± 15%; P = .89), likely due to autoregulation in the noninjured myocardium. Sustained LV unloading using an Impella CP increases coronary flow and perfusion of infarcted myocardium for at least 2 hours.
ECG-Synchronized Pulsatile Extracorporeal Life Support Preserves Left Ventricular Function and Coronary Flow in a Porcine Model of Cardiogenic Shock
Petr Ostadal, Mikulas Mlcek, Holger Gorhan, Ivo Simundic, Svitlana Strunina, Matej Hrachovina, Andreas Kruger, Dagmar Vondrakova, Marek Janotka, Pavel Hala, Martin Mates, Otomar Kittnar, and Petr Neuzil
Na Homolce Hospital
Venoarterial extracorporeal life support (ECLS) is increasingly being used to treat rapidly progressing or severe cardiogenic shock. However, it has been repeatedly shown that increased afterload associated with ECLS significantly diminishes left ventricular (LV) performance. A new electrocardiogram (ECG)-synchronized pulsatile cardiac assist system, which offers full circulatory support with increased diastolic and decreased systolic extracorporeal flow, has recently been introduced. The ECG-synchronized pulsatile ECLS flow improves variables of LV function and coronary flow in comparison with standard continuous-flow ECLS support. Sixteen female swine (mean body weight: 45 kg) underwent ECLS implantation under general anesthesia and artificial ventilation. Subsequently, acute cardiogenic shock, with documented signs of tissue hypoperfusion, was induced by initiating global myocardial hypoxia. Hemodynamic cardiac performance variables and coronary flow were then measured at different rates of continuous or pulsatile ECLS flow (ranging from 1L/min to 4L/min) using arterial and venous catheters, a pulmonary artery catheter, an LV pressure–volume loop catheter, and a Doppler coronary guide wire. Myocardial hypoxia resulted in declines in mean cardiac output (CO) to 2.3 ± 1.2 L/min, systolic blood pressure to 61 ± 7 mm Hg, and LV ejection fraction (LVEF) to 21% ± 7%. Synchronized pulsatile flow was associated with a significant reduction in LV end-systolic volume by 6.2 mL (6.7%), an increase in LV stroke volume by 5.0 mL (17.4%), higher LVEF by 4.5% (18.8% relative), CO by 0.37 L/min (17.1%), and mean arterial pressure by 3.0 mm Hg (5.5%), when compared with continuous ECLS flow at all ECLS flow rates (P < .05). At selected ECLS flow rates, pulsatile flow also reduced LV end-diastolic pressure, end-diastolic volume, and systolic pressure. The ECG-synchronized pulsatile flow was also associated with significantly increased (7%-22%) coronary flow at all ECLS flow rates. The ECG-synchronized pulsatile ECLS flow preserved LV function and coronary flow compared with standard continuous-flow ECLS in a porcine model of cardiogenic shock.
LV Unloading Using Impella CP Improves LA Pressure, Function, and Stiffness
Shin Watanabe, Kenneth Fish, Lauren Leonardson, Roger J Hajjar, and Kiyotake Ishikawa
Icahn School of Medicine at Mt. Sinai
The impact of partial left ventricular assist device on left anterior (LA) function and stiffness in heart failure (HF) remains unclear. Hypothesis: Unloading the LV with an Impella CP improves transmitral pressure gradient, leading to a reduction of LA pressure and improved LA function and stiffness. The HF is induced by percutaneously occluding the proximal left anterior descending artery for 90 minutes in Yorkshire pigs (n = 4; 40-50 kg). Two weeks after the myocardial infarction (MI), animals underwent left ventricular (LV) unloading with an Impella CP for 150 minutes. The LA pressure was directly measured by a transseptal approach, and LA volumes were assessed by 3-dimensional echocardiography. The LA stiffness was approximated as the slope of minimal and maximal LA pressure–volume coordinates as described previously (Circ Heart Fail. 2015;8:295-303). Two weeks after the MI, the animals presented with impaired LV ejection fraction (LVEF; 69.7% ± 10.5% to 38.6% ± 7.0 %; P = .03) and a dilated LV (LV end-systolic volume: 24.55 ± 10.7 to 65.3 ± 16.3 mL; P = .04) without significant mitral regurgitation. The LV unloading with maximal pump support (P8) resulted in an increase in total cardiac output (2.80 ± 0.18 to 3.27 ± 0.22 L/min; P = .03) and reduced LV end-diastolic pressure (27.5 ± 10.1 to 17.9 ± 4.5 mm Hg; P = .06). These changes were accompanied by a significant reduction in mean LA pressure (17.3 ± 2.6 to 10.8 ± 2.1 mm Hg; P = .001). The LA volumes assessed by 3-dimensional echocardiography were also reduced (LA maximal volume: 40.3 ± 4.6 to 29.0 ± 2.3 mL, P = .006; LA minimal volume: 20.3 ± 2.7 to 15.0 ± 2.3 mL, P = .002). Although the total LAEF was not altered from the baseline (49.3 ± 6.4 to 48.5 ± 6.5%; P = .85), passive Left atrial ejection fraction (LAEF) was significantly increased (17.7% ± 1.9% to 39.4% ± 5.6%; P = .008), suggesting an improved transmitral suction effect. Additionally, LA stiffness assessed by pressure–volume coordinates was improved with Impella support (1.41 ± 0.52 to 0.30 ± 0.16 mm Hg/mL; P = .03). The LV unloading using an Impella CP improves passive LA function and reduces mean LA pressure in a recent MI setting. Along with improved LA stiffness, these data implicate a beneficial impact of LV unloading on relieving HF symptoms.
Unloading Left Ventricle via Impella CP in Cardiogenic Shock Improves Microcirculation and Hemodynamic State
Jan-Thorben Sieweke, Dominik Berliner, Florian Zauner, Christian L Napp, Ulrike Flierl, Jörn Tongers, Johann Bauersachs, and Andreas Schäfer
Department of Cardiology and Angiology, Hannover Medical School
Despite advances in understanding and management of cardiogenic shock (CS), morbidity and mortality particularly in patients’ refractory to standard therapy remain unacceptably high. The evolving field of percutaneous mechanical support, especially left ventricular unloading, has revolutionized treatment strategies in CS. In our single-center registry, 93 patients with CS were treated with Impella CP between February 2013 and December 2015 (76% men, age 58.5 [interquartile range (IQR): 50-68] years). Cardiogenic shock resulted from ST-segment elevation myocardial infarction (STEMI) (43%), non–STEMI (26%), and cardiomyopathy (24%). Patients were severely ill during the intensive care unit course: 89% mechanical ventilation, 43% dialysis, 40% extra-hospital resuscitation, and 28% intrahospital resuscitation. Impella was inserted and removed percutaneously via femoral access (duration of support: 96 hours [IQR: 30; 166 hours]). On mechanical support with Impella CP, the hemodynamic state improved. Reflecting the immediate improvement in microcirculation, lactate levels normalized within 6 hours, and lactate levels were reduced at every time point. In this population, patients with lactate levels above 10.2 hours prior to Impella implantation had a mortality of 80%. Patients with lactate levels above 5.5 4 hours after Impella CP implantation had a survival below 10%. Despite being negatively selected due to shock state, intrahospital survival in this highly fragile population was 40%. Thirty-nine patients died on mechanical support for various reasons including intentional termination of therapy (n = 2 due to patient’s provision; n = 11 due to severe anoxic brain damage). Although 43% were bridged to recovery, another 14% were bridged to left ventricular assist device implantation. The safety profile was reasonable (7% disseminated intravascular coagulation, 4% compartment syndrome, 8% leg ischemia, and 3% stroke). Minor vascular access site problems occurred in only 8 patients. Our results suggest that the concept of mechanical support using Impella microaxial pump for left ventricular unloading maintaining end-organ perfusion is efficacious in rescuing patients in CS with a favorable relationship to adverse side effects.
Evidence that Mesenchymal to Endothelial Transition Occurs Secondary to Cardiac Unloading Through LVAD Support
Ana Sofia Cruz Solbes, Cesar Uribe, Keith A Youker, Javier Amione-Guerra, Bashaar Hanawi, Jerry D Estep, Barry H Trachtenberg, Guha Ashrith, Guillermo Torre-Amione, Erick Suarez, and Arvind Bhimaraj
Methodist Hospital
Endothelial to mesenchymal transition (EndMT) contributes to fibrosis in heart failure. The role of the reverse phenomenon of mesenchymal to endothelial transition (MET) in recovery of organ pathology is increasingly being recognized. Chronic unloading with left ventricular assist devices (LVADs) provides an opportunity to study the mechanisms facilitated by unloading. We have demonstrated reduction in transforming growth factor β (a known inducer of EndMT) post-LVAD. The current study assesses changes in EndMT/MET after unloading of the LV. The MET contributes to decreased fibrosis after LVAD implantation secondary to cardiac unloading. We performed reverse transaiption-polymerase chain reaction on 13 paired samples of human myocardial LV obtained at LVAD implant and at time of heart transplant. The SNAI1 was studied as surrogate marker of EndMT/MET. Fibrosis was quantified using Masson trichrome staining while fibroblast was counted with immunohistochemistry (FSP-1. Spearman correlation was performed. Echocardiographic data were compared to assess effective unloading. Mean age was 54.8, 78% male, 50% African American, 42% ICM, and all LVADs were HeartMateII. Median days of support was 314 days (100-1426 days). Echo data suggest significant unloading (LV ejection fraction improved, P = 0.03; left anterior volume decreased ∼40%, P < .0001; LVAD decreased, P = .03). Significant decrease in fibrosis post-LVAD occurred in 86% of the patients (P = .008) with a decrease in absolute number of fibroblasts (P = .03). The SNAI1 increased significantly (P = .03) after LVAD support, compared with pre-LVAD. Extent of fibrosis pre-LVAD had a significant correlation (r = 0.75; P = .003) to SNAI1 gene expression post-LVAD. The EndMT/MET plays a role in fibrotic/antifibrotic mechanisms in human myocardium. The SNAI1 expression had been proposed to play a role in the initial transition of MET. Our data supports this hypothesis: Higher SNAI1 signal was associated with decreased fibrosis and fibroblast count post-LVAD. Correlation between post-LVAD SNAI1 with increased fibrosis pre-LVAD suggests more MET occurring in patients with a higher number of fibroblasts at baseline. Further studies of these pathways during acute unloading might be relevant.
Inhibition of the Inflammatory Injury Following Myocardial Ischemia Reperfusion in the Mouse
Antonio Abbate and Stephano Toldo
Virginia Commonwealth University
Reperfusion is effective in reducing ischemic injury in acute myocardial infarction (AMI). However, ischemia triggers a secondary injury, known as reperfusion injury, contributing to the overall infarct size. Multiple mechanisms are being explored to favorably modify the effects of reperfusion injury. We hypothesize that inhibition of the Nod-like Receptor Protein-3 (NLRP3) inflammasome limits infarct size following myocardial ischemia/reperfusion, by inhibiting the inflammatory component of the reperfusion injury. We induced AMI in adult mice by transient ligation of the left anterior descending coronary artery for 30 or 75 minutes. We tested 3 different strategies to inhibit the NLRP3 inflammasome: a newly designed small molecule specifically inhibiting the inflammasome (NLRP3inh), plasma-derived alpha-1 antitrypsin (AAT) shown to inhibit the NLRP3 inflammasome, and a synthetic oligopeptide (SP16) designed to reproduce the C-terminal peptide of AAT. Infarct size was measured at 1, 3, and 24 hours and expressed as percentage of area at risk. Infarct size increased with duration of ischemia from 43% ± 4% with 30 minutes to 65% ± 3% with 75 minutes if ischemia (P < .001) showing a wavefront of ischemic injury. After 30 minutes of ischemia, however, infarct size progressively increased from 1 to 24 hours after reperfusion (11% ± 2% at 1 hour, 30% ± 5% at 3 hours, and 43% ± 4% at 24 hours) showing a wavefront of reperfusion injury. Administration of the NLRP3inh, AAT, or SP16 given immediately at reperfusion or within 30 minutes after reperfusion following 30 or 75 minutes of ischemia significantly reduced infarct size at 24 hours (−56%, −44%, −55%, respectively, vs vehicle, all P < .01). Pharmacological inhibition of the NLRP3 inflammasome within 1 hour of reperfusion limits the secondary inflammatory injury and infarct size following myocardial ischemia reperfusion in the mouse. Pharmacological interventions alone or in conjunction with other interventions show promise to significantly further improve outcome postmyocardial infarction.
To Have Clinical Value, Future Cardioprotective Interventions Must Provide Additive Protection to That From a P2Y12 Receptor Antagonist
Michael Cohen and James M. Downey
University of South Alabama College of Medicine
Patients with acute myocardial infarction (AMI) are treated with P2Y12 receptor antagonists to suppress platelet aggregation. Although P2Y12 antagonists are used for their anticoagulant effect, our studies reveal that they are also potent postconditioning mimetics. The latter properties are influencing preclinical and clinical searches for adjunctive cardioprotective interventions in unexpected ways. In situ monkey, rabbit, and rat hearts were subjected to ischemia/reperfusion after which hearts were removed, areas at risk determined, and infarct size measured by triphenyltetrazolium chloride staining. Animals were given cangrelor, an intravenous P2Y12 antagonist, starting 10 minutes before reperfusion to simulate treatment of patients with AMI prior to angioplasty. Infarction was approximately half that seen in untreated animals. This cardioprotection was abrogated by inhibitors of signaling used by postconditioning including wortmannin, LY294002, PD98059, 5-hydroxydecanoate, 8-sulfophenyltheophylline, MRS1754, and 2-mercaptopropionylglycine, suggesting that cangrelor protects by a similar mechanism. None restored platelet reactivity indicating that protection was independent of anticoagulation. Our hypothesis was further strengthened by failure of ischemic pre- or postconditioning to potentiate cangrelor’s cardioprotection. Similar results were seen with clopidogrel and ticagrelor, other P2Y12 antagonists, indicating a class effect. We propose that much of the ability of P2Y12 inhibitors to improve outcomes in AMI can be attributed to direct infarct size reduction from conditioning rather than prevention of stent thrombosis. Because patients treated with P2Y12 antagonists prior to revascularization are already postconditioned, adding another conditioning intervention will be futile. This would explain why recent large trials of ischemic postconditioning or the postconditioning mimetic cyclosporine in patients with AMI were unsuccessful in spite of strong preclinical results. Additional protection in today’s patients will require an intervention not dependent on the conditioning mechanism. Any such intervention must first demonstrate additive protection in animals that have additionally been treated with a P2Y12 antagonist before clinical trials are considered.
Unloading Left Ventricle Using Impella pLVAD During Acute MI Reveals Survival Benefit
Perwaiz Meraj, Rajkumar Doshi, Avneets Singh, and Perwaiz Meraj
Northwell Health
The Impella percutaneous left ventricular assist device (pLVAD) has been increasingly used in patients with severe heart failure, cardiogenic shock, and high risk percutaneous intervention. However, use of Impella in patients with severe coronary artery disease (CAD) and left ventricular end-diastolic pressure (LVEDP) above 35 mm Hg is underreported. The LVEDP is a useful indicator of hemodynamic status in patients with low ejection fraction. Our goal is to report our experience in using Impella LVAD in patients with CAD presenting with LVEDP above 35 mm Hg. We identified all the patients who underwent cardiac intervention using percutaneous LVAD from January 2011 to December 2015. A total of 173 patients with an LVEDP above 35 mm Hg were included in our study, divided into 2 groups: those treated with Impella pLVAD and those with intra-aortic balloon pump (IABP). The primary end point evaluated major adverse events (death, myocardial infarction, and cardiogenic shock), mortality at 30 days, 1 year, and 2 years. Baseline demographics were obtained. Unadjusted and adjusted analysis were performed using SPSS version 22.0. Among the patients treated with Impella, the patients had an overwhelmingly lower ejection fraction (22.05% vs 33.59%; P < .000). In unadjusted and adjusted analysis, both bleeding (2.6% vs 19.4%; P < .000) and transfusion requirements (0% vs 32.1%; P < .010) were significantly better in the Impella-treated group. Mortality did not differ. The data for the use of Impella pLVAD in near shock patients with elevated LVEDP does not demonstrate a clear benefit; however, in clinical practice, a benefit has been anecdotally seen. These data demonstrate that there is a clear benefit in adverse outcomes except mortality in those patients with an LVEDP > 35 mm Hg treated with the Impella pLVAD over an IABP. Most practitioners would utilize an IABP in this population; however, this strategy may demonstrate harm, just as seen in the cardiogenic shock population. The LV unloading with an Impella pLVAD appears to be the device of choice at this time.
Transcatheter Renal Venous Decongestion in Acute Decompensated Heart Failure
Ehud Schwammenthal, Yosi Tuval, Daniel Glozman, Shany Levinkron, Dvir Afek, and Shira Rosenberg-Tzur
Mangenta Medical Ltd
The main reason for hospitalization in acute decompensated heart failure (ADHF) is congestion rather than low cardiac output, and the encapsulated kidneys are particularly vulnerable to congestion: Increased renal venous pressure (RVP) has been shown to impair renal blood flow and function and worsen diuretic responsiveness. Renal dysfunction in ADHF increases in-hospital death rate 7-fold and length of stay 3-fold. We selectively reduce RVP in order to mechanically unload the kidneys, improve renal perfusion and function, and promote fluid and sodium removal. To this end, we developed a transcatheter renal venous decongestion (TRVD) system, including self-expandable flow pumps that are easily deployed in the renal veins through 7F introducer sheaths. The system was tested in a controlled porcine model in 6 experimental interventions where caval venous pressure was raised using an inflatable balloon, comparing urine output and sodium excretion in 1 treated kidney to that of the untreated kidney (internal control). The system was then introduced for clinical testing in a first-in-man study in 7 patients with ADHF. Animal experiments: Urine output at baseline was similar; 22 ± 1 mL/10 min and 21 ± 4 mL/10 min, respectively (NS). During increased venous pressure, urine output was 29 ± 10 mL/10 min in treated kidneys but fell to 10 ± 4 mL/10 min in untreated kidneys (P < .02 to baseline; P < .002 to treatment). Human data: Plasma creatinine (under identical medication, n = 5) ranged from 0.84 to 1.6 mg/dL before and 0.7 to 1.22 mg/dL after therapy; sodium excretion (n = 4) increased from 43 ± 7 mmol/h to 82 ± 26 mmol/h. Inferior vena cava diameter decreased from 2.85 ± 0.3 cm to 1.72 ± 0.38 cm at discharge (P < .02). The TRVD shows favorable effects on urine output, sodium excretion, and diuretic responsiveness (clinically) and, therefore, has the potential to become an important interventional tool in the treatment of patients with ADHF.
LV Unloading Using Impella Reduces LA Pressure and Volume Overload in Chronic Mitral Regurgitation
Kiyotake Ishikawa, Shin Watanabe, Lauren Leonardson, Kenneth Fish, and Roger J. Hajjar
Icahn School of Medicine at Mt. Sinai
Mitral regurgitation is a common presentation in patients administered due to the decompensated chronic heart failure. Whether a left ventricular (LV) to aorta percutaneous left ventricular assist device (pLVAD) can be effective in relieving left anterior (LA) overload remains unclear. The LV unloading using an Impella CP reduces LA pressure and volume by actively pumping the blood toward forward direction. Chronic heart failure with mitral regurgitation was induced in Yorkshire pigs (n = 3; 20 kg) by percutaneously severing chordae tendineae of the mitral apparatus with a biopsy catheter. Three months later (body weight: 43.5 ± 5.0 kg), the animals underwent LV unloading with an Impella CP with a maximal flow support (p8). Hemodynamics before and during the LV unloading were assessed by Swan-Ganz catheter and pressure volume loop catheter (Millar catheter) in both LA and LV. Additionally, LA volumes were assessed by 3-dimensional (3D) echocardiography before and during the Impella support. At 3 months, animals presented with moderate mitral regurgitation (regurgitant fraction: 38% ± 10%) with dilated LV (LV end-diastolic volume: 45.5 ± 1.7 mL to 89.7 ± 18.0, P = .04; LV end-systolic volume: 12.5 ± 1.6 mL to 31.2 ± 10.9, P = .10; Day 0 to 3 month, respectively). The LV unloading resulted in a significant reduction of LV end-diastolic pressure (13.6 ± 2.6 to 4.0 ± 4.0 mm Hg; P = .029). Although the visual assessment of magnetic resonance (MR) degree by color Doppler echocardiography did not change by LV unloading, mean LA pressure decreased significantly (12.3 ± 7.1 to 9.3 ± 6.1 mm Hg; P = .035). The LA v-wave, which is accentuated in the mitral regurgitation due to the regurgitant flow, also reduced significantly, indicating a reduction of quantitative MR (17.3 ± 11.2 to 12.3 ± 9.5 mm Hg; P = .038). Furthermore, maximum LA volume assessed by 3D echocardiography was significantly decreased (46.6 ± 13.4 to 29.7 ± 15.9 mL; P = .043). The LV to aorta pLVAD can alleviate LA pressure and volume overload in a heart failure due to mitral regurgitation.
Ventricular Changes Evaluated by Echocardiography During Impella Support: Single Pediatric Institution Experience
Ricardo H. Pignatelli, Dhaval Parekh, Athar Qureshi, Iki Adachi, Douglas Feagin, Sebastian Tume, Ricardo A. Najera, Yunfei Wang, John Colquitt, Henri Justino, and Aamir Jeewa
Baylor College of Medicine, Texas Children’s Hospital
Impella (IMP) is a percutaneous microaxial flow ventricular assist device (VAD) device. It is one the latest options available for circulatory support in critically ill patients in cardiogenic shock. The ability to unload the left ventricle and support the ventricular function during and shortly after explantation has not been shown in this age cohort. The aim is to review the changes in hemodynamics and ventricular function evaluated by echocardiography in a pediatric cohort supported with the IMP. Retrospective study of 10 consecutive patients on IMP supports since 2014. Demographic data, hemodynamics (pulmonary capillary wedge pressure [PCWP]), left ventricular (LV) end-diastolic Z score (LVDz), LV ejection fraction (EF)%, and right ventricular fractional area change % (RVFAC%) were collected pre- and postimplant at 24 hours, 72 hours, and 120 hours as well first echo postexplant (EPex). The PCWP P value was calculated by Wilcoxon rank test and the remaining parameters by a mixed model analysis. Ten patients were included in our study with a median age 15 (6-22) years, and Body surface area (BSA) of 1.7 (0.9-2) m2. Six patients had graft rejection after heart transplant, and 2 had myocarditis. The PCWP decreased in 6/6 patients (100%; P = .03) at 24 hours. Trend toward decreased LVDz 4/10 patients (40%; P = .23) at EPex, LVEF% increased in 5/10 patients (50%; P = .01) at EPex, and RVFAC% improved in 6/10 patients (60%; P = .01) at EPex. Of all patients who were supported with IMP, 6 had recovery of ventricular function, 2 were supported to transplant, and 2 bridged to long-term VAD. The IMP support showed ability to unload the LV, decrease PCWP, and improve biventricular function by echocardiographic parameters in our cohort. The IMP successfully assists in myocardial recovery and can be considered as a bridge to transplant/long-term VAD in this age group.
Calcium Homeostasis and Cardiac Arrhythmias After Acute Cardiac Unloading in Rodents
Alexander P. Schwoerer and Heimo Ehmke
University Medical Center Hamburg-Eppendorf
Mechanical unloading of failing human hearts with ventricular assist devices (VADs) greatly increases the risk for ventricular arrhythmias. Data obtained in patients after VAD implantation have shown alterations in cellular electrophysiology, but the interpretation of these studies is difficult because of the confounding effects of the underlying heart disease. We therefore investigated the impact of mechanical unloading on cardiac function in the nonfailing heart using the animal model of heterotopic heart transplantation. We could previously demonstrate that mechanical unloading causes a marked prolongation of ventricular action potentials and an upregulation of the L-type Ca2+ current in this experimental model. A detailed analysis of intracellular Ca2+ cycling revealed a pronounced suppression of sarcoplasmic Ca2+ content and Ca2+-induced Ca2+ release, which is fully compensated by an increased Ca2+ influx through L-type Ca2+ channels. Accordingly, the systolic Ca2+ transient remained normal in the unloaded myocardium, however, at the expense of larger changes in Ca2+ concentrations at the cell membrane. To investigate whether this translates to an impaired repolarization in vivo and thus may induce arrhythmias, we performed continuous long-term telemetric electrocardiogram recordings for 56 days after heterotopic heart transplantation. Although heart rate regulation normalized within the first 4 days after surgery and remained stable over the remaining weeks, ventricular repolarization was severely impaired in the unloaded hearts. Following the perioperative phase, the QT interval progressively increased by ∼65% within the first week and remained at this level. In parallel to this QT prolongation, the occurrence of ventricular arrhythmias over 56 days was ∼100-fold higher in the unloaded than in the control hearts. These data support the concept that mechanical unloading, independent from heart disease, impairs myocardial repolarization and Ca2+ homeostasis leading to ventricular arrhythmias. This implicates that a reduction in left ventricular workload may be causally linked with the occurrence of ventricular arrhythmias in patients with VAD support.
Hydrostatic Pressure Gradient Ultrafiltration Device: A Novel Approach for Extracellular Fluids Removal
Yair Feld, Nitai Hanani, and Giorgi Shtenberg
Rambam Healthcare Campus
This study explored the concept of a novel intra peritoneal absorption chamber for fluids removal through the peritoneal membranes. Fluid overload is a common and challenging clinical problem in patients with acute decompensated heart failure (ADHF). Normalization of fluid status in ADHF is associated with improved long-term prognosis. Diuretic therapy is limited by kidney function and perfusion pressure, whereas dialysis and ultrafiltration are associated with significant hemodynamic and electrolyte imbalances when performed in the acute settings. We suggest a novel approach in which a permeable absorption chamber is implanted in the peritoneum. A negative hydrostatic pressure in the absorption chamber is induced by a pump, prompting fluids ultrafiltration through the peritoneal membranes into the chamber. The accumulated extracellular fluids are drained to an ex vivo collection system or into the urinary system. To examine the feasibility of this concept, we implanted an absorption chamber in the peritoneum of rats and drained fluids through the transplanted chamber. An absorption chamber was prepared from a stainless steel coil with a diameter of 0.4 cm and a length of 3 to 6 cm covered by a collagen membrane (Permacol; Medtronic). The absorption chamber was implanted in the peritoneum cavity of 4 Sprague-Dawley rats. Two-week postimplantation, the rats were anesthetized, and a peritoneal needle was inserted to the chamber. Extracellular fluid was drained from the chamber at an average rate of 16 ± 6 cc/kg/day during 3 hours of negative hydrostatic pressure induction. The fluid electrolytes and proteins were comparable to the serum content. Implantable absorption chamber enables extracellular fluids removal through the peritoneal membranes. This study suggests that an implantable absorption chamber may be used in fluid overload clinical conditions and serve as a possible novel heart failure therapy in acute and potentially chronic settings.
Potential of Regenerative Therapy With Medical Assist Devices
Timothy Henry
Cedars-Sinai Medical Center
Heart failure continues to be the leading cause of morbidity and mortality not only in the United States but throughout the world. Despite advances in medical and device therapy, many patients continue to have significant symptoms leaving heart transplantation, left ventricular assist devices (LVADs), or palliative care as the only options. This has stimulated interest in regenerative therapies including stem-cell therapy. Based on positive preclinical results, the initial clinical trials used predominantly autologous bone marrow mononuclear stem cells (BMC). Trials with BMC demonstrated excellent safety but only modest efficacy most likely due to the significant variability with autologous BMC, which is related to the decline in the number and potency of stem cells with age and cardiac risk factors. This has stimulated the next generation of cell therapies, which include allogeneic cells, cardiac derived cells, and enhanced cultured autologous cells. A recent large double-blind, placebo-controlled phase 2 trial using enhanced BMC cells (IxCell-DCM) demonstrated a significant reduction in mortality and cardiovascular hospitalizations in cell-treated patients. A second large phase 2 trial using enhanced mesenchymal stem cells (MSCs) will be presented at embryonic stem cells, and a large phase 3 trial with allogeneic MSCs is underway. An even more attractive idea is to combine the benefits of novel, mechanical left ventricular support devices with regenerative therapy. To date, there have been a total of 67 patients randomized in 11 published clinical trials with the combination of cell therapy and LVAD trials. The largest of these (N = 30) was an NIH-sponsored trial using allogeneic MSCs, which demonstrated a potential benefit in LVAD weaning and a suggestion of mortality benefit. In summary, despite optimal medical and device therapy, the number of patients with advanced HF continues to grow. Both novel mechanical assist devices and regenerative therapy represent potential solutions. The combination of these 2 unique therapies may be a particularly attractive solution.