Mechanical Circulatory Assist Devices: Available Modalities and Review of Literature.

Despite advancements in the field interventional cardiology, the prognosis in patients who suffer from cardiogenic shock is poor. Over the years, the use of percutaneous mechanical circulatory support (MCS) devices has increased with the aim to improve short- and long-term outcomes. In this article, we aim to review the different modalities available for MCS devices and current literature comparing their uses. Copyright:

INTRODUCTION

Almost 6 million Americans suffer from heart failure, and this is expected to reach >8 million by 2030.[1] Acute coronary syndrome is a common cause of heart failure and is commonly complicated by acute myocardial infarction cardiogenic shock (AMI-CS), which is seen in approximately 5%–8% of patients presenting with acute ST-elevation myocardial infarction (STEMI).[2] Studies have shown that AMI-CS patients have in-hospital mortality of over 60% and long-term mortality of 80%, despite early revascularization.[3] Even though acute ischemic changes and cardiopulmonary arrest (CA-CS) are the most common cause of CS, nonischemic causes such as left ventricular (LV) and right ventricular (RV) dysfunction, myopericarditis, severe diseases of the aortic and mitral valves, and Takotsubo cardiomyopathy should also be on the differential.

The exact definition of CS varies among literature, but these patients commonly present with hypotension, tachycardia, oliguria, altered mental status, and cool, clammy, and cyanotic extremities with evidence of end organ damage. The goal in treatment of CS is to perfuse end organs such as kidneys, brain, and liver. In addition to vasoactive agents, short-term mechanical circulatory support (MCS) devices are designed to help with tissue and end-organ perfusion, reduce intracardiac filling pressure, LV volume, wall stress, myocardial oxygen demand, and potentially infarct size, and increase coronary perfusion and improve the overall hemodynamic status of the patient. The hemodynamic condition of LV follows the pressure–volume loop, and it can provide information such as contractile function, stroke volume, and myocardial oxygen demand.[456] The pressure–volume loop does not provide any information regarding RV function or extracardiac functions. The MCS will alter the patient's innate physiology, compensation, and hemodynamics.

The most common short-term MCS devices are intra-aortic balloon pumps (IABP), extracorporeal membrane oxygenation (ECMO), and peripheral ventricular assist devices (pVADs) such as Impella and TandemHeart. Some of the percutaneous MCS devices can be inserted at bedside or catheterization laboratory without surgical cutdown. These devices are used for a wide range of clinical conditions from prophylactic insertion for high-risk percutaneous coronary intervention (PCI) to intraoperatively for both cardiac and even noncardiac surgeries in high-risk patients.

The use of MCS devices has increased over the past few decades, and an in-depth understanding of the working principles, indications, and their hemodynamic effect is needed for the best patient management.

INTRA-AORTIC BALLOON PUMP

The IABP counter pulsation system consists of a flexible double lumen catheter that allows for flushing or pressure monitor and delivery of helium gas, and a mobile console that contains the digital inflation and deflation cycle control. The catheter is usually between 30 and 50 cc and most frequently inserted peripherally through femoral access under fluoroscopic guidance or at bedside with postinsertion X-ray.[7] The proximal tip of the balloon should be positioned in the proximal ascending aorta and distal tip not obstructing the renal arteries. The balloon inflates during diastole to augment coronary perfusion with counterpulsation and deflates during systole, effectively reducing afterload and increase cardiac output. Electrocardiogram (ECG) or arterial pressure waveform timing is used to calibrate the inflation and deflation cycles. Newer technologies in IABP design have allowed the incorporation of aortic valve closure and utilization of digital algorithms to respond to arrhythmias.[2]

Hemodynamically, the IABP increases cardiac output by 0.5 L/min but can also increase cerebral and peripheral perfusion pressure and as a result, reduces myocardial stress and oxygen demand.[89] The exact effect may be highly variable depending on the position and volume of the balloon, heart rate and rhythm, and systemic vascular resistance.[1011] Increased arterial compliance and elasticity is associated with the greater hemodynamic enhancement from IABP. IABP's effect on total coronary blood flow is highly debated, with certain studies suggesting significantly increased flow while other studies found no changes.[12131415] However, the measurable increase in diastolic filling pressure should lead to increased flow into stenotic territories.

Clinically, the IABP is often considered for patients with the following conditions.

Prophylaxis for high-risk or complicated PCI

AMI-CS

CA-CS

Other causes of CS

Intractable angina

Refractory heart failure bridging

Ventricular arrythmia bridging

Severe/critical aortic stenosis

Intraoperatively for cardiac and noncardiac surgeries in select patients.

Prophylactic placement of IABP before CABG has been done in the past,[16] and the decision should be jointly made by cardiac surgeon, interventional cardiologist, and cardiothoracic anesthesiologist evaluating the risk of decompensation during induction of anesthesia and intraoperatively. Contraindications for IABP insertion include significant aortic regurgitation, aortic dissection, aortic aneurysm, uncontrolled sepsis, and severe peripheral arterial disease (PAD) that is untreated.

IABP insertion-associated complications can be classified as vascular and nonvascular. Vascular complications are most feared including acute limb ischemia, major hemorrhage, and vessel laceration.[1718] Nonvascular complications include cholesterol embolization, balloon rupture, sepsis, and peripheral neuropathy.[171920] Overall, these complications occur in about 7% of the patients, and 2.6% are classified as major complications (acute limb ischemia, balloon leak, and death).[21] PAD, female gender, diabetes mellitus, hypertension, prolonged support, obesity, catheter size >9.5F, and cardiac index <2.2 L/min/m2 are risk factors associated with a higher probability of IABP complications.[2122232425]

While IABP has been in widespread clinical use for decades, the data from clinical trials have been mixed. The Should We Emergently Revascularize Occluded Coronaries for CS (SHOCK) trial published in 1999 that showed a benefit for early revascularization in the setting of AMI-CS retrospectively observed that those treated with IABPs and fibrinolytics had improved 1-year mortality compared to fibrinolytics alone.[2526] The IABP in CS II (IABP-SHOCK II) trial randomly assigned 600 patients with AMI-CS to IABP group (n = 301) and no IABP group (n = 299).[27] The primary end point was 30-day mortality. There was no significant difference with regard to primary end point (IABP vs. no IABP, 39.7% vs. 41.3%; P = 0.69 for intention to treat, 37.5% vs. 41.4%; P = 0.35 per-protocol). There was no difference in secondary outcomes such as intensive care unit length of stay, duration of catecholamines, time to hemodynamic stabilization, LV assist device (LVAD) placement, hospital re-infarction, stroke, bleeding, and sepsis. However, there was a 10% cross over from control to IABP group, mostly due to protocol violations.

The timing of IABP varied, as 86.6% was inserted post-PCI. The sample size was relatively small, and long-term results are pending. As a result of this trial, 2012 ESC STEMI guidelines downgraded IABP use in STEMI patients to Grade 2B recommendation from 1C.[28] The American College of Cardiology and the American Heart Association (ACC/AHA) 2013 guidelines classified the use of IABP for the management of STEMI who do not stabilize quickly with medical therapy as class 2A recommendation.[29]

The Counterpulsation to Reduce Infarct Size Pre-PCI AMI (CRISP AMI) randomly assigned 337 patients with acute anterior STEMI without CS to receive IABP before PCI (n = 161) or PCI alone (n = 176).[30] The primary end point was infarction size. There was no significant difference in infarct size between the two groups (IABP plus PCI vs. PCI alone, 42.1% vs. 37.5%, difference of 4.6%; P = 0.06, imputed difference of 4.5%; P = 0.07). At 30 days, there was no significant difference observed in these two groups in terms of major vascular complications, major bleeding, or transfusion. One may suggest that the benefit of IABP was offset by the time spent inserting the device. However, the median time from first medical contact to the first device related to infarct artery was only 9 min longer for IABP plus PCI group (77 min vs. 68 min; P = 0.04).

Another possibility is that potential benefit of IABP may not be observed since the patients in this trial are very late in the course, with median symptom onset to first device time of 196 min. Studies have shown that myocardial salvage is minimal if ischemic time is >120 min where patients in this trial had an ischemic time of longer than 3 h on average.[3132]

This trial's primary end point was infarct size on cardiac magnetic resonance imaging (MRI), but this is a physiologic measurement, and it may not correlate exactly to the true or long-term myocardial damage or remodeling. Fifteen or 8.5% of patients in PCI alone arm crossed over to rescue IABP arm, mostly due to sustained hypotension or development of CS. This may have led to small, but no significant increase in number of patients that died before they could receive cardiac MRI in the PCI group. This trial supports IABP as a salvage strategy for high-risk PCI in STEMI patients.

IMPELLA

There are currently three Impella devices: Impella 2.5, Impella CP, and Impella 5.0. These devices are able to create axial flow and actively draw blood from LV and pump it into the proximal ascending aorta. As a result, it directly increases cardiac output and mean arterial pressure and reduces myocardial oxygen demand and LV end-diastolic pressure. Due to the innate physiology of the device, LV may collapse on itself from Impella flow. To prevent such events, central venous pressure should be monitored closely and between 8 and 12 mmHg. These devices are generally inserted under fluoroscopic guidance. The positioning can be easily adjusted through transthoracic echocardiography. Since the Impella device provides continuous flow, ECG monitoring for arrhythmia is not required.

Impella 2.5 is capable of providing up to 2.5 L/min of cardiac output augmentation and can be inserted percutaneously through a 13F catheter into the femoral artery.[33] The catheter is placed across the aortic valve and is connected to an external console device for pump speed monitoring and adjustment and may act as invasive blood pressure monitor. The rotational speed can vary from 2000 to 50,000 rpm, and 2.5 L/min augmentation happens at the maximum rpm. The patient must be heparinized to maintain a partial thromboplastin time of 50–56 s.[34] Impella CP provides 3.0–4.0 L/min of cardiac output augmentation, while Impella 5.0 provides 5.0 L/min cardiac output increase. The Impella 2.5 and Impella CP can be inserted percutaneously, while Impella 5.0 requires surgical cutdown for access since it uses 21F sheath. Contraindications for Impella device include mechanical valve, LV thrombus, and ventricular septal defect.

The efficacy study of LVAD to treat patients with CS (ISAR-SHOCK) trial randomized 26 patients with AMI-CS into receiving IABP (n = 13) and Impella 2.5 (n = 12).[3536] One patient died before device could be placed. The primary end point was a change in cardiac index (ΔCI) 30 min after implantation from baseline. Authors noted a significant greater ΔCI in Impella group versus the IABP group (Impella vs. IABP, 0.49 ± 0.46 L/min/m2 vs. 0.11 ± 0.31 L/min/m2; P = 0.02). Mean arterial pressure was significantly increased in Impella group as well (Impella vs. IABP, 9.0 ± 14.0 mmHg versus −1.2 ± 16.2 mmHg; P = 0.09). However, the 30-day mortality was 46% in both groups. The small number of patients enrolled in this study may not have allowed for a full extrapolation of mortality difference. The primary end point of ΔCI was chosen to avoid early mortality, but retrospectively, may have hindered further extrapolation from Impella's ability to improve hemodynamics.

A Prospective Randomized Clinical Trial of Hemodynamic Support With Impella 2.5 versus IABP in Patients Undergoing High-Risk PCI (PROTECT II) trial randomized 452 symptomatic patients with a complex triple vessel or unprotected left main disease and severely depressed LV function with ejection fraction of ≤35% to receive IABP (n = 226) or Impella 2.5 (n = 226) during high-risk PCI.

The primary end point was 30-day incidence of major adverse events. There was no significant difference in the primary end point between the two (IABP vs. Impella, 40.1% vs. 35.1%; P = 0.227 for intention to treat population and 42.2% vs. 34.3%, P = 0.092 for per protocol population). There were fewer major events such as stroke, MI, death, and repeat revascularization in the Impella group versus IABP group (9.8% vs. 18.6, P = 0.009) after hospital discharge. However, at 90-day follow-up, the Impella group showed a statistically insignificant trend toward lower major adverse events compared to IABP group (40.6% vs. 49.3%; P = 0.066).

In terms of secondary outcome, maximal decrease in cardiac output from baseline was lower in Impella compared to IABP (−0.04 ± 0.24 W in comparison with −0.14 ± 0.27 W; P = 0.001). Creatinine clearance, in-hospital composite major adverse events, or any of its components were all similar between the two groups. The biggest limitation was early termination of the study due to been deemed futile, and as a result, only 69% of the planned enrollment occurred. However, authors believed that there was a significant learning curve with Impella and the patients in the latter half of the trial did much better compared to the first half. In addition, operators were unable to be blinded to treatment assignment due to distinct radiographic appearance of the two devices.

As we can see, there is a paucity of data on Impella devices in the current literature. Despite repeated attempts at organizing randomized controlled trials with Impella device, enrollment rate is extremely slow. In addition, patients in active CS are not readily enrolled in clinical trials since time is of the essence, and they often have high mortality and have high procedural risks due to their multiple comorbidities.[37] In addition, Impella 2.5 was used in all the available clinical trials and not the Impella 5.0. As a result, physicians often have to evaluate each patient's physiology and use their clinical judgment based on individual cases.

TANDEMHEART

TandemHeart is a pVAD that utilizes an extracorporeal axial flow pump that is capable of providing up to 5 L/min of cardiac output augmentation. This device uses a 21F inflow cannula and is inserted into femoral vein (FV), going up the inferior vena cava/right atrium where a transseptal puncture is performed to pass the cannula into the left atrium. The attached pump will aspirate the oxygenated blood and is attached to the outflow cannula. The outflow cannula is usually between 15F and 19F and is inserted into femoral artery with a centrifugal pump. This leads to a reduction LV end-diastolic pressure, but an increase in afterload due to backflow of blood into the arterial system. In addition, this configuration requires adequate RV function as well as pulmonary oxygenation and vascular compliance. Once the device is in place, the patient must be heparinized to maintain Activated Clotting Time (ACT) of 200 s. It is normally used as a bridging device to long-term VAD placement or heart transplant.[38]

TandemHeart can be removed at bed side or at the time of surgery. Although the atrial septal defect created with the transseptal puncture does not need to be repaired unless there is evidence of left to right shunting, it is typically repaired if the patient undergoes open surgery.[3940] There have been multiple reported cases where TandemHeart was used to bridge advanced heart failure and even cardiopulmonary resuscitation (CPR) patients to heart transplants.[4142]

There are several advantages of TandemHeart. It can be used in patients with LV thrombus since it completely bypasses that chamber. It can also be used in aortic stenosis patients since it does not require anterograde blood flow through the aortic valve. The transseptal puncture itself is a procedure that requires technical expertise of an experienced interventional cardiologist.

Complications associated with the transseptal puncture include perforation of adjacent structures such as aortic root, atrial wall, and coronary sinus. The patient will be immobilized due to the femoral insertion site. Since a large cannula is required, bleeding and ischemic limb risk is higher compared to IABP. Ventricular septal defect, RV failure, aortic regurgitation, aortic dissection, and severe PAD are contraindications to TandemHeart placement.[3343]

Data have shown that the use of pVADs does not have a significant impact on mortality in CS patients.[44] A subsequent study was designed to randomize 41 patients with AMI-CS to TandemHeart (n = 21) and IABP (n = 20) for hemodynamic support.[45] The primary end point was cardiac power index within 2 h after device implantation. The study showed that TandemHeart improved CPI more compared to IABP (TandemHeart vs. IABP, 0.22–0.37 W/m2 vs. 0.22–0.28 W/m2; P = 0.004). The renal function measured in urine output was better in TandemHeart group as well (TandemHeart vs. IABP, 30–80 ml/h vs. 28–30 ml/h; P = 0.04).

However, severe bleeding and limb ischemia were significantly higher in TandemHeart group, whereas 30-day mortality was similar between the two (TandemHeart vs. IABP, 43% vs. 45%; P = 0.86). Another observational study looked at outcomes of 117 patients with severe CS, of whom 56 patients (47.9%) underwent active CPR immediately before or at the time of device placement. The cardiac index improved from 0.52 to 3.0, P < 0.001.[46] The pulmonary capillary wedge pressure, serum lactic acid level, and creatinine all had statistically significant decrease. As a result, pVAD can rapidly and significantly improve patient's hemodynamics at risk of bleeding and limb ischemia.

EXTRACORPORAL MEMBRANE OXYGENATION

ECMO is different from VADs since it is used for patients with concomitant cardiac and pulmonary failure. It is able to pump blood throughout the body as well as oxygenating the blood with artificial membrane lung. ECMO can be implemented into venovenous (VV) or venoarterial (VA) configuration. Peripherally, inflow extracts blood from internal jugular (IJ)/FV for both VV and VA and are pumped through a filter membrane that removes carbon dioxide and adds oxygen. Then, the fully oxygenated blood is pumped back into the circuit through the outflow cannula. The VV uses IJ/right atrium for outflow, while VA uses femoral artery for outflow.

ECMO can be inserted centrally as well, but it must be done in the operating room. VV is for oxygenation only and VA can provide full cardiopulmonary support to the patient. ECMO can provide up to 6 L/min of increase in cardiac output. Similar to TandemHeart, LV afterload may be increased in VA ECMO due to the flow into arterial system. Occasionally, IABP or Impella may be inserted on top of ECMO for additional hemodynamic support. Even though peripheral ECMO can be placed at bedside, surgical cutdown of the femoral artery may still be required due to the size of the catheter.

ECMO catheter size varies, but 31F is usually the largest and most commonly used for adult males. Since limb ischemia is common with femoral access in VA ECMO configuration, additional arterial cannula can be inserted distal to femoral artery cannula to perfuse the limb distal to the femoral artery cannula itself. If femoral vessels are not suitable due to severe PAD, common carotid or subclavian artery can be used instead.[47]

ECMO as short-term MCS device can be used in the following conditions: acute CS and CA with severe pulmonary congestion, prophylaxis, and rescue for high-risk PCI, ARDS, massive pulmonary embolism, and failure to wean from cardiopulmonary bypass postcardiotomy. It is contraindicated in significant aortic regurgitation, severe PAD, recent stroke or traumatic brain injury, and uncontrolled sepsis.

Surgical and cannula site bleed, renal failure, limb ischemia, and infection are the most common complications with ECMO when used with cardiac indication.[48] The patient must be heparinized with partial thrombin time maintained between 50 and 70 s. While ECMO has been in use for decades, there are limited data and randomized controlled trials to assess ECMO's role for short-term MCS device. Since there is no recommendation available on ACC/AHA guidelines, physicians must use their own clinical judgment on initiating ECMO treatment based on the individual patient.

CONCLUSION

Although the use of MCS devices theoretically seems promising, current literature provides conflicting results. To optimize outcomes, a multidisciplinary approach and meticulous patient and device selection remains key in ensuring better short- and long-term outcomes.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.