# Mechanical Circulatory Support: IABP, Impella, ECMO

| Device | Mechanism | Max Flow Support | Cannulation | LV Effect | Key Indication | Landmark Trial |
|---|---|---|---|---|---|---|
| IABP | Counterpulsation (diastolic augmentation) | +0.5–1.0 L/min | Percutaneous femoral artery | Modest afterload reduction | Bridge to surgery (VSD, MR); LV vent on ECMO | IABP-SHOCK II (no benefit) |
| Impella 2.5 | Microaxial flow pump (LV→aorta) | 2.5 L/min | Percutaneous femoral artery | Direct LV unloading | Low-flow support, high-risk PCI | — |
| Impella CP | Microaxial flow pump (LV→aorta) | 3.7–4.0 L/min | Percutaneous femoral artery | Direct LV unloading | AMI-cardiogenic shock (SCAI C-D) | DanGer Shock (mortality benefit, NNT 8) |
| Impella 5.0/5.5 | Microaxial flow pump (LV→aorta) | 5.0–6.0 L/min | Surgical cutdown (axillary/femoral) | Direct LV unloading | Bridge to recovery/decision | — |
| Impella RP | Microaxial flow pump (RA→PA) | 4.0 L/min | Percutaneous femoral vein | RV unloading | RV failure | — |
| VA-ECMO | Centrifugal pump + oxygenator | 4–6 L/min | Femoral vein→femoral artery (or central) | Increases LV afterload (requires venting) | Biventricular failure, ECPR, post-cardiotomy | ECLS-SHOCK (no benefit; more complications) |
| VV-ECMO | Centrifugal pump + oxygenator | 3–6 L/min | Femoral vein→IJ vein (or bicaval DL) | No hemodynamic support | Severe refractory ARDS | EOLIA (trend toward benefit) |

## Intra-Aortic Balloon Pump (IABP)

### Mechanism

The intra-aortic balloon pump operates on the principle of counterpulsation, using a helium-filled balloon of 30 to 50 mL positioned in the descending thoracic aorta. During diastole, the balloon inflates, augmenting diastolic aortic pressure and thereby increasing coronary perfusion pressure and myocardial oxygen supply. During systole, the balloon rapidly deflates just before ventricular ejection, reducing the aortic pressure at the onset of systole and effectively decreasing left ventricular afterload. The net hemodynamic effect of this counterpulsation cycle is a modest increase in cardiac output of approximately 0.5 to 1.0 L/min, a decrease in myocardial oxygen demand through afterload reduction, and improved coronary blood flow through diastolic augmentation.

### Placement and Management

The IABP is inserted percutaneously via the femoral artery using a 7 to 8 French catheter, with sheathless techniques available that reduce the incidence of limb ischemia. The balloon tip should be positioned 1 to 2 cm distal to the origin of the left subclavian artery, confirmed by chest radiography at the level of the carina or second intercostal space. Timing is set at a 1:1 ratio, augmenting every cardiac cycle, with adjustments guided by the arterial waveform. Optimal timing is confirmed when the diastolic augmentation exceeds the unassisted systolic pressure and the assisted end-diastolic pressure falls below the unassisted end-diastolic pressure. Anticoagulation with heparin infusion targeting an aPTT of 50 to 70 seconds is required with most protocols. Weaning proceeds by reducing the augmentation ratio from 1:1 to 1:2 to 1:3 as hemodynamics improve, followed by removal.

### Evidence

The IABP-SHOCK II trial of 2012 demonstrated no 30-day or 12-month mortality benefit for the IABP in AMI-related cardiogenic shock, fundamentally altering clinical practice. Current guidelines classify routine IABP use in AMI-cardiogenic shock as Class III, indicating no benefit. The remaining indications for IABP include bridge to surgery for mechanical complications such as ventricular septal defect and acute mitral regurgitation, refractory angina, and LV venting during VA-ECMO support. Contraindications include moderate-to-severe aortic insufficiency, which is worsened by diastolic augmentation, aortic dissection, aortic aneurysm, and severe peripheral vascular disease.

## Impella Devices

### Device Options

The Impella family comprises several devices with escalating support capacity. The Impella 2.5 is a catheter-mounted axial flow pump providing up to 2.5 L/min of flow, placed percutaneously via the femoral artery. The Impella CP delivers up to 3.7 to 4.0 L/min and is the most commonly used device in cardiogenic shock, also placed percutaneously. The Impella 5.0 provides up to 5.0 L/min but requires surgical cutdown via the axillary or femoral artery. The Impella 5.5 delivers up to 6.0 L/min with surgical placement and offers repositionability, serving as a bridge to recovery or decision. The Impella RP provides right ventricular support, positioned from the femoral vein to the pulmonary artery with flows up to 4.0 L/min. The Impella ECP is the newest design, featuring a foldable mechanism that allows fully percutaneous deployment.

### Mechanism

The Impella functions as a microaxial flow pump positioned across the aortic valve with its inlet in the left ventricle and outlet in the ascending aorta. This positioning directly unloads the left ventricle by reducing end-diastolic pressure, volume, and wall stress while simultaneously increasing forward cardiac output and aortic root pressure to improve coronary and systemic perfusion. A critical distinction from VA-ECMO is that the Impella reduces LV afterload, providing true ventricular unloading rather than the afterload increase that characterizes VA-ECMO.

### DanGer Shock Trial (2024)

The DanGer Shock trial represents a watershed moment in mechanical circulatory support. This multicenter randomized controlled trial enrolled 360 patients with AMI-related cardiogenic shock classified as SCAI stage C or D and compared Impella CP against standard care consisting of IABP, vasopressors, and inotropes. The primary outcome of 180-day all-cause mortality was 46 percent in the Impella group versus 59 percent in the standard care group, yielding a hazard ratio of 0.74 with a p-value of 0.04 and a number needed to treat of 8. This was the first randomized controlled trial to demonstrate a mortality benefit for any mechanical circulatory support device in cardiogenic shock, establishing a new paradigm for management. The Impella group did experience higher rates of hemolysis, bleeding, renal replacement therapy, and limb ischemia, underscoring the need for experienced teams and meticulous complication monitoring.

### Complications

Hemolysis from mechanical shear stress requires monitoring of plasma-free hemoglobin, LDH, and haptoglobin, with pump speed reduction if hemolysis is significant. Limb ischemia occurs in 5 to 15 percent of patients and should be monitored with near-infrared spectroscopy and pulse oximetry on the ipsilateral foot. Bleeding, both at the access site and gastrointestinal, requires careful anticoagulation management. Device migration or malposition necessitates continuous waveform monitoring and repositioning under echocardiographic or fluoroscopic guidance. Aortic valve injury is rare but has been reported. Device thrombosis is prevented by maintaining purge solution flow and appropriate anticoagulation.

<image>Comparative illustration of three MCS devices side by side in anatomical position. Left panel: IABP showing balloon in descending aorta with inflation/deflation cycle annotated, arterial waveform showing diastolic augmentation. Center panel: Impella CP showing catheter crossing aortic valve with inlet in LV and outlet in ascending aorta, with axial flow pump mechanism detail and motor housing annotated. Right panel: VA-ECMO showing drainage cannula in right atrium via femoral vein and return cannula in femoral artery with extracorporeal circuit (centrifugal pump, oxygenator, heat exchanger). Include hemodynamic effect annotations for each: IABP (CO +0.5-1 L/min, afterload reduction), Impella (CO +3-5 L/min, true LV unloading), VA-ECMO (CO +4-6 L/min, increases LV afterload). Below each device: key trial names and outcomes.</image>

## Veno-Arterial ECMO (VA-ECMO)

### Indications

VA-ECMO is indicated for cardiogenic shock refractory to pharmacological support and other mechanical circulatory support, cardiac arrest as a bridge to decision through extracorporeal CPR, post-cardiotomy shock, massive pulmonary embolism with hemodynamic collapse, and as a bridge to recovery, decision, transplant, or durable ventricular assist device. Its primary advantage over the Impella is the ability to support both ventricles, making it the modality of choice for biventricular failure.

### Circuit Components

The VA-ECMO circuit consists of a venous drainage cannula of 21 to 25 French placed in the femoral vein with the tip in the right atrium, or via direct central cannulation, and an arterial return cannula of 15 to 19 French in the femoral artery or axillary artery. A centrifugal pump generates continuous non-pulsatile flow at speeds of 2000 to 5000 RPM. A polymethylpentene hollow-fiber membrane oxygenator provides extracorporeal gas exchange with adjustable sweep gas FiO2 and flow rate. A heat exchanger enables temperature regulation. Target flow rates are 3 to 6 L/min, aiming for a cardiac index of 2.2 to 2.5 L/min/m2.

### Hemodynamic Effects

VA-ECMO increases systemic oxygen delivery by providing extracorporeal gas exchange and circulatory support. However, its critical limitation is that it increases LV afterload through retrograde aortic flow, which elevates aortic pressure and prevents the left ventricle from ejecting against the increased afterload. This leads to progressive LV distension, which in turn causes worsening pulmonary edema, myocardial ischemia from increased wall stress, and impaired myocardial recovery. This fundamental hemodynamic limitation distinguishes VA-ECMO from the Impella, which directly unloads the ventricle.

### LV Venting Strategies

The LV distension problem necessitates an LV venting strategy whenever VA-ECMO is initiated. The IABP can reduce afterload and provide some unloading and is the most commonly used approach. The Impella, used in combination with VA-ECMO in the ECMELLA configuration, provides direct LV unloading and represents the most effective venting strategy. Atrial septostomy creates a left-to-right shunt that decompresses the left atrium. Surgical LV venting through the apex or pulmonary vein is available in post-surgical settings. Monitoring for adequate venting includes assessment of left atrial pressure, pulmonary capillary wedge pressure, LV distension on echocardiography, and verification that the aortic valve continues to open, as loss of aortic valve opening creates blood stasis that promotes intracardiac thrombus formation.

### ECLS-SHOCK Trial (2023)

The ECLS-SHOCK trial enrolled 420 patients with AMI-related cardiogenic shock planned for revascularization and compared VA-ECMO against standard care. The 30-day mortality was 47.7 percent versus 49.0 percent, demonstrating no benefit from VA-ECMO. The ECMO group experienced significantly more complications including moderate-to-severe bleeding at 23 versus 10 percent and limb ischemia at 11 versus 4 percent. Notably, not all patients received LV venting, which may have contributed to the negative results. The conclusion is that VA-ECMO should not be used routinely in AMI-cardiogenic shock and should be reserved for biventricular failure or refractory shock situations.

### Complications of VA-ECMO

Limb ischemia on the cannulated side is prevented by mandatory placement of a distal perfusion cannula in the superficial femoral artery. Bleeding from access sites and gastrointestinal sources requires careful anticoagulation balance. Hemolysis results from shear stress through the pump and oxygenator. Harlequin syndrome, also known as differential hypoxemia, occurs when the upper body receives poorly oxygenated blood from native cardiac output while the lower body receives well-oxygenated ECMO blood. This is monitored by right radial SpO2 or arterial blood gas, and if upper body hypoxemia is detected, intervention is required by increasing ECMO flow, adding a VV-ECMO component, or converting to central cannulation. Infection risk increases with duration. Circuit thrombosis, stroke, and limb thromboembolism require maintenance of ACT at 180 to 220 seconds.

## Veno-Venous ECMO (VV-ECMO)

### Indications

VV-ECMO is indicated for severe ARDS refractory to conventional lung-protective ventilation, prone positioning, and rescue therapies. The EOLIA trial criteria provide a framework for referral: a PaO2/FiO2 below 50 on FiO2 1.0 for more than 3 hours, a PaO2/FiO2 below 80 on FiO2 1.0 for more than 6 hours, or a pH below 7.25 with a PaCO2 of 60 or greater despite a respiratory rate of 35 and optimal ventilator settings. VV-ECMO also serves as a bridge to lung transplantation and is used in severe air leak syndromes. Critically, VV-ECMO provides gas exchange only and does not provide any hemodynamic support.

### Cannulation Strategies

The most common configuration uses femoral-jugular cannulation, with drainage from the femoral vein positioned in the inferior vena cava and return through the right internal jugular vein into the superior vena cava. The bicaval dual-lumen cannula, such as the Avalon catheter, is a single catheter inserted through the right internal jugular vein with drainage ports in both the SVC and IVC and a return port directed at the tricuspid valve toward the right ventricle and pulmonary artery. Its advantages include a single cannulation site and facilitated patient mobilization, though the risks of malposition and perforation require echocardiographic or fluoroscopic guidance for placement. Recirculation, in which oxygenated return blood is immediately drawn back into the drainage cannula, is the principal source of circuit inefficiency and is minimized by maintaining adequate distance between cannula tips of at least 10 cm and optimizing cannula positioning.

### Management

Blood flow is typically set at 50 to 80 mL/kg/min, corresponding to 3 to 6 L/min, and determines the oxygen delivery capacity of the circuit. Sweep gas flow determines CO2 removal and is initially set at a 1:1 ratio with blood flow and titrated to the PaCO2 target. Sweep gas FiO2 is typically set at 1.0 through the oxygenator. Ventilator settings during ECMO should be reduced to lung rest settings: FiO2 0.3 to 0.4, PEEP 10 to 15 cmH2O, tidal volume 1 to 4 mL/kg ideal body weight, and respiratory rate 10 to 15, with the goal of allowing lung recovery while avoiding ongoing ventilator-induced lung injury. Anticoagulation with unfractionated heparin targets an anti-Xa of 0.3 to 0.7 IU/mL or ACT of 180 to 220 seconds.

### EOLIA Trial (2018)

The EOLIA trial randomized 249 patients with severe ARDS to early VV-ECMO versus conventional treatment with crossover to ECMO as rescue. The primary outcome of 60-day mortality was 35 percent versus 46 percent, which did not reach statistical significance with a p-value of 0.09. However, the 28 percent crossover rate from the control arm substantially diluted the treatment effect, and Bayesian post-hoc analysis demonstrated a greater than 95 percent probability that ECMO reduces mortality. The practical interpretation is that early VV-ECMO for severe ARDS is likely beneficial, and early referral to an ECMO center is essential because delays in transfer are consistently associated with worse outcomes.

<image>VA-ECMO and VV-ECMO configurations illustrated side by side. VA-ECMO panel: shows femoral venous drainage cannula (25 Fr) with tip in right atrium, femoral arterial return cannula (17 Fr) in common femoral artery with distal perfusion cannula in SFA, extracorporeal circuit (centrifugal pump → oxygenator → arterial return), and arrows showing blood flow path. Annotation showing LV distension problem with arrow indicating increased afterload. Harlequin syndrome illustration showing upper body cyanosis with lower body pink. VV-ECMO panel: shows femoral venous drainage (25 Fr) in IVC and right IJ return (19 Fr) in SVC, with circuit in between. Alternative bicaval dual-lumen catheter shown in inset. Arrows demonstrating recirculation concept. Key differences table between VA and VV-ECMO: hemodynamic support, cannulation sites, complications, indications.</image>

## Durable Mechanical Circulatory Support

### Left Ventricular Assist Devices (LVADs)

The current standard for durable left ventricular assist devices is the continuous-flow centrifugal pump, exemplified by the HeartMate 3. This device features a fully magnetically levitated impeller that reduces shear stress, resulting in less hemolysis and pump thrombosis compared to earlier-generation devices. The MOMENTUM 3 trial demonstrated the superiority of the HeartMate 3 over the HeartMate II, with improved survival free of disabling stroke or reoperation for device malfunction at 77 versus 63 percent at 2 years. The indications for LVAD implantation include bridge to transplant, destination therapy for patients not eligible for transplant, and bridge to candidacy or decision. Complications of long-term LVAD support include driveline infection at 15 to 20 percent at 1 year, gastrointestinal bleeding at 15 to 30 percent from acquired von Willebrand disease, stroke at 10 to 15 percent, and right ventricular failure post-implant in 20 to 30 percent of patients.

### Total Artificial Heart (TAH)

The SynCardia total artificial heart provides orthotopic biventricular replacement and serves as a bridge to transplant in patients with biventricular failure who are not suitable for LVAD implantation. It requires a sufficient body habitus with a body surface area greater than 1.7 m2 due to device size constraints. The AESON total artificial heart represents a newer bioprosthetic design that has received CE marking.

## Key Clinical Pearls

- IABP provides minimal hemodynamic support (0.5-1 L/min) and has no mortality benefit in AMI-cardiogenic shock (IABP-SHOCK II)
- Impella CP is the first MCS device to demonstrate mortality benefit in AMI-cardiogenic shock in an RCT (DanGer Shock, NNT = 8)
- VA-ECMO increases LV afterload — always plan for LV venting (IABP, Impella, or surgical vent) when initiating VA-ECMO
- ECLS-SHOCK showed no benefit of VA-ECMO in AMI-cardiogenic shock — do not use routinely; reserve for biventricular failure or ECPR
- Monitor for Harlequin syndrome on VA-ECMO: right radial SpO2/ABG is essential; upper body hypoxemia requires intervention
- Distal perfusion cannula is mandatory for femoral VA-ECMO to prevent limb ischemia
- VV-ECMO provides gas exchange only — it does NOT provide hemodynamic support
- Refer patients with severe ARDS to an ECMO center early — delays worsen outcomes; the EOLIA trial suggests likely mortality benefit

## References

1. Thiele H, Zeymer U, Neumann FJ, et al. Intraaortic balloon support for myocardial infarction with cardiogenic shock. N Engl J Med. 2012;367(14):1287-1296.
2. Moller JE, Engstrom T, Jensen LO, et al. Microaxial flow pump or standard care in infarct-related cardiogenic shock. N Engl J Med. 2024;390(15):1382-1393.
3. Thiele H, Zeymer U, Akin I, et al. Extracorporeal life support in infarct-related cardiogenic shock. N Engl J Med. 2023;389(14):1286-1297.
4. Combes A, Hajage D, Capellier G, et al. Extracorporeal membrane oxygenation for severe acute respiratory distress syndrome. N Engl J Med. 2018;378(21):1965-1975.
5. Mehra MR, Uriel N, Naka Y, et al. A fully magnetically levitated left ventricular assist device — final report. N Engl J Med. 2019;380(17):1618-1627.