VA-ECMO is a type of mechanical circulatory support that provides systemic flow of 4-6 L/min in cases of severe cardiorespiratory failure. However, its retrograde flow configuration through femoral cannulation can lead to LV overload, causing distension, blood stasis, thrombosis, and pulmonary congestion. Although LV unloading modalities are used to prevent these complications, their effectiveness and timing of implementation are still unclear, as they may add additional morbidity to an already critically ill patient.

This review is divided into three parts to address different questions. The first part focuses on identifying patients who could benefit from LV unloading, using reliable clinical, hemodynamic, and echocardiographic predictors. The second part analyzes the various unloading methods available, offering practical recommendations based on the latest scientific evidence. Pharmacological and non-invasive options, the insertion of decompression catheters, and a wide variety of unloading devices are examined. The third part discusses the optimal timing for deploying an unloading device, comparing preventive versus reactive unloading according to certain criteria. Finally, goals are presented to follow once the unloading method is implemented, providing guidelines for managing the patient in various circumstances. The purpose of this article in our blog is to simplify and summarize the most relevant points of this comprehensive review.

INDICATIONS FOR LV UNLOADING

Established complications caused by increased afterload, such as pulmonary edema or heart failure due to lack of aortic valve opening, clearly indicate the need for LV unloading. However, the timing for primary unloading in the absence of obvious complications to facilitate LV recovery remains unclear. The use of unloading devices is not without cost and carries potential risks of vascular and hemorrhagic complications, and also adds practical complexity to device setup, anticoagulation management, monitoring of vascular access sites, and difficulty in mobilizing the patient.

To identify patients at higher risk of complications due to increased afterload, several predictors must be considered. Hemodynamic predictors include reduced arterial pulsatility and elevated pulmonary capillary wedge pressure (PCWP), which reflects high LV filling pressure. Echocardiographic predictors include increased LV dimensions, blood stasis and thrombi in the LV, absence of aortic valve opening, or a left ventricular outflow tract (LVOT) velocity-time integral (VTI) <10 cm. Clinical predictors include the development of pulmonary edema or refractory ventricular arrhythmias.

LV unloading has been shown to improve survival in patients with cardiogenic shock (CS) secondary to myocardial infarction, with an absolute risk reduction of 6.6%, due to improved subendocardial perfusion. Prophylactic LV unloading has also shown clear benefit in patients with chronic heart failure and elevated PCWP, who are especially vulnerable to increased afterload.

UNLOADING STRATEGIES

Pharmacological and Non-Invasive Approaches

Before considering invasive approaches, it is important to reduce afterload in patients. To achieve this, the goal is to optimize VA-ECMO flow to provide adequate systemic perfusion with the lowest possible afterload. Reduction/withdrawal of vasoconstrictor support can be an initial step to maintain mean arterial pressures of 60 mmHg and systolic pressures of 80-90 mmHg. Low flow reduction, below 2.2 L/min/m², may be sufficient to maintain adequate perfusion and decrease ventricular distension. However, very low flows below 1.5 L/min/m² increase the risk of thromboembolic complications in the ECMO circuit.

Preload correction may be another strategy. Fluid therapy optimization is an option, but it is not suitable for all patients due to frequent increases in capillary permeability and third spacing. Increasing diuresis with diuretics or hemofiltration may be attempted, especially in those patients who have not yet developed complications related to high afterload.

Inotropes can be used to improve ventricular contractility. This strategy is employed as an initial measure to counteract increased afterload or as a bridge to more invasive unloading approaches in patients who have developed or are at risk of developing complications. However, it is important to apply it temporarily since inotropes increase myocardial oxygen demand, and increased mortality has been observed in observational studies.

LV Decompression by Catheters

Catheter insertion into the LV, left atrium (LA), or pulmonary artery is performed, and these are connected to the venous cannula of the VA-ECMO circuit. Some groups use transesophageal echocardiography and 7Fr Pigtail catheters for this insertion. Others have achieved similar results with even smaller 5F and 6F catheters. However, due to limited flow and a higher risk of hemolysis, routine use of these smaller catheters is not recommended.

Percutaneous Atrial Septostomy

This technique is performed using a balloon guided by fluoroscopy or echocardiography. Although common in pediatric populations, there is limited evidence available for its use in adults. There is a risk of developing thrombi in the LV or aortic root, as well as the risk of stroke after removing the venous cannula (paradoxical embolism). Another possibility is left-sided VA-ECMO with transseptal puncture, placing a multi-perforated venous cannula in the LA. This allows for simultaneous biatrial drainage. This approach is particularly useful in patients with LV thrombi or unilateral peripheral vascular disease, where implanting a large-caliber femoral cannula is contraindicated. It may also be an alternative in patients in whom a transaortic device cannot be used (e.g., aortic stenosis or mechanical prosthesis). The ultimate evolution of this modification is the conversion to TandemHeart®, a centrifugal pump that uses a 21F transseptal cannula located in the LA for drainage and maintains the same output as the original ECMO via arterial cannulation. An oxygenator can be added to the TandemHeart® system, or, as mentioned before, the LA cannula can be connected to the VA-ECMO drainage cannula via a Y-connector, allowing for biatrial drainage and unloading of both ventricles.

Pulmonary Artery Drainage

Pulmonary artery drainage is a technique used in VA-ECMO to reduce LV pressure and volume by inserting a single- or multi-perforated cannula. Among the available cannulas, the Protek Duo® stands out. This bilumen cannula was initially conceived to create a percutaneous right ventricular support system, with drainage from atrial perforations and infusion into the distal pulmonary artery, limiting recirculation present in other systems like the Avalon® cannula. When used for unloading, both lumens would be connected to the ECMO inflow. It is available in sizes of 29F and 31F and can be inserted via the jugular vein with good navigability, allowing flows of up to 4.5 L. The results are promising and it is particularly useful when converting the system to right or biventricular support.

Intra-Aortic Balloon Pump (IABP)

This is the most widely used device for LV unloading, as it reduces afterload and improves coronary flow. Although it has demonstrated benefit in patients with CS secondary to AMI, there are no randomized studies justifying its use as an LV unloading method. The results of observational studies evaluating its efficacy are heterogeneous and inconclusive.

Percutaneous Left Ventricular Assist Devices (pLVAD)

The combination of pLVAD and VA-ECMO is increasingly used to treat CS. pLVADs are microaxial flow pumps that continuously move blood from the LV to the aortic root, reducing workload, pressure-volume area, and myocardial oxygen consumption. The Impella CP® is the most commonly used device for LV unloading in this configuration, often referred to as “ECMELLA” or “ECPELLA.” The pLVAD is generally inserted percutaneously through the femoral artery contralateral to the ECMO perfusion cannula, although it can also be surgically implanted via the axillary artery. Contraindications for its use include having a mechanical prosthesis in the aortic position, severe aortic insufficiency, LV thrombus, and peripheral arterial disease.

The combination of pLVAD and VA-ECMO has been shown to reduce PCWP, improve pulmonary flow, and reduce LV size. However, as with IABP, there are no randomized studies to support its use for LV unloading. Recent registries suggest that “ECPELLA” reduces mortality compared to VA-ECMO alone, although this benefit may be offset by an increased risk of bleeding and ischemic complications in the lower extremities.

A critical aspect in managing “ECPELLA” is to adequately balance the flows of both devices. This must be dynamic, with a higher VA-ECMO flow prioritized initially to ensure adequate systemic perfusion. Then, a gradual transition is made to pLVAD support to promote cardiac recovery, and in a final phase, VA-ECMO weaning is considered. To avoid Harlequin syndrome in patients with hypoxemic respiratory failure, a lower pLVAD flow is recommended until pulmonary function improves, after which flows can be adjusted again, although conversion of the system to V-VA ECMO alongside pLVAD may still be necessary.

The initial flow of the Impella® should generally be lower than the maximum achieved by these devices. The Impella 5.5® can provide anterograde flows of up to 6 L/min, which is usually sufficient for adequate systemic perfusion in most cases. Its use as an unloading method has been associated with earlier VA-ECMO weaning and lower VA-ECMO flow to improve RV function and oxygenation before decannulation. Axillary insertion of the Impella® requires a surgical approach, but allows for high flows over significantly longer periods than the Impella CP® (cases of up to 83 days have been reported) and greater positional stability to ensure proper functioning. Although its license in Europe is for 30 days and in the United States for 14 days, it presents potential advantages such as the possibility of patient mobilization and a lower incidence of hemolysis and thrombosis, making it an ideal option as a bridge to transplantation or long-term LVAD implantation after VA-ECMO decannulation. This strategy may be beneficial in cases of decompensated heart failure, patients with poor recovery prospects, or when prolonged mechanical circulatory support is anticipated.

Surgical Approaches

Central VA-ECMO is used in postcardiotomy shock or graft failure following heart transplantation, where a sternotomy has already been performed. Although cannulation of the ascending aorta avoids the retrograde flow characteristic of peripheral VA-ECMO, ventricular distension can still occur when the LV is compromised, leading many centers to routinely perform LV unloading. This can be achieved by surgical insertion of a 16-20F cannula through the LV apex, pulmonary vein, or pulmonary artery, connected to the VA-ECMO drainage system via a Y-connector. Some minimally invasive approaches, such as subxiphoid or anterolateral thoracotomy, have also been described for cannula insertion with the same function in situations with an intact chest or when the sternotomy has already been closed. In this regard, inserting an apical unloading cannula allows conversion of the ECMO system to a paracorporeal left ventricular assist system (Levitronix®).

Selection of the Strategy

The chosen strategy largely depends on the degree of increased afterload. IABP is suitable for slightly increased afterload, while for a distended LV with little contractility and high filling pressures, active unloading with pLVAD or a surgical decompression cannula is more appropriate. We are awaiting the results of the HERACLES study, the first clinical trial comparing these unloading strategies.

TIMING OF UNLOADING

Unloading devices can be implanted before, during, or after the initiation of VA-ECMO, either prophylactically or in response to clinical, echocardiographic, or hemodynamic manifestations of increased afterload. Observational studies suggest that prophylactic unloading, performed within 2 hours of VA-ECMO initiation, may improve mortality compared to reactive unloading. However, randomized trials are needed to confirm these findings, and we await the results of the EARLY-UNLOAD and REVERSE trials, which will evaluate the benefits of early unloading with LA septostomy and Impella CP®, respectively. The ECLS-SHOCK trial evaluates the benefit of ECMO in patients with refractory shock secondary to AMI, and the results from the subgroup using unloading methods will provide relevant information.

GOALS OF UNLOADING

Once unloading is initiated, continuous monitoring of ECMO flow is essential, as well as optimizing device parameters to achieve an adequate mean arterial pressure that ensures optimal systemic perfusion and PCWP < 15 mmHg. Aortic valve opening should also be assessed using echocardiography. Echocardiography plays a crucial role in evaluating RV function and LV contraction recovery, which will be reflected in an improvement in the arterial pulse wave. As hemodynamics improve, efforts should be made to gradually reduce inotropic drug doses. If evidence of cardiac recovery is observed, such as a pulse pressure > 10 mmHg, mean arterial pressure > 60 mmHg with low doses of inotropes, and EF > 30%, VA-ECMO weaning may be considered.

If an additional device, such as pLVAD or IABP, is used, VA-ECMO weaning should be attempted first, if feasible, to reduce afterload and myocardial oxygen consumption, potentially improving cardiac recovery. This is particularly relevant if an improvement in RV function and oxygenation is observed, allowing a transition to exclusive LV support. If cardiac recovery is deemed possible but VA-ECMO weaning is not yet feasible, another long-term support device, such as the Impella 5.5® or a long-term LVAD, may be considered. In cases where cardiac recovery seems unlikely, VA-ECMO can be maintained as a bridge to transplantation. Finally, in patients who do not improve despite LV unloading or develop multiorgan failure, palliative measures may be the only option.

In conclusion, the choice of unloading method largely depends on local expertise, although there is still a lack of data to guide selection among different strategies. The algorithm presented in this review provides guidance in this regard. In any case, ongoing and future randomized clinical trials will help determine when, for whom, and how LV unloading should be performed in patients supported with VA-ECMO.

REFERENCE:

Ezad SM, Ryan M, Donker DW, Pappalardo F, Barrett N, Camporota L, et al. Unloading the left ventricle in venoarterial ECMO: in whom, when, and how? Circulation. 2023 Apr 18;147(16):1237-1250. doi: 10.1161/CIRCULATIONAHA.122.062371.