Percutaneous Management of High-Risk Pulmonary Embolism

Abstract

Acute pulmonary embolism (PE) leads to an abrupt increase in pulmonary vascular resistance and right ventricular afterload, and when significant enough, can result in hemodynamic instability. High-risk PE is a dire cardiovascular emergency and portends a poor prognosis. Traditional therapeutic options to rapidly reduce thrombus burden like systemic thrombolysis and surgical pulmonary endarterectomy have limitations, both with regards to appropriate candidates and efficacy, and have limited data demonstrating their benefit in high-risk PE. There are growing percutaneous treatment options for acute PE that include both localized thrombolysis and mechanical embolectomy. Data for such therapies with high-risk PE are currently limited. However, given the limitations, there is an opportunity to improve outcomes, with percutaneous treatments options offering new mechanisms for clot reduction with a possible improved safety profile compared with systemic thrombolysis. Additionally, mechanical circulatory support options allow for complementary treatment for patients with persistent instability, allowing for a bridge to more definitive treatment options. As more data develop, a shift toward a percutaneous approach with mechanical circulatory support may become a preferred option for the management of high-risk PE at tertiary care centers.

High-risk (also known as massive) acute pulmonary embolism (PE) is among the direst of cardiovascular emergencies with poor outcomes and few data guiding optimal management. The incidence of PE is increasing with the aging population, increased medical complexity, and more-sensitive imaging modalities. The proportion of patients presenting with high-risk PE is also increasing.¹ Acute PE causes an abrupt increase in pulmonary vascular resistance due to thrombotic obstruction, hypoxemic vasoconstriction, and the release of pulmonary artery (PA) vasoconstrictors.² The increase in right ventricular (RV) afterload leads to increased RV wall tension and RV ischemia with subsequent RV systolic dysfunction and dilation. If this process is severe, bowing of the interventricular septum leads to decreased left ventricular preload and reduced cardiac output.³ The obstructive shock can precipitate hypotension and cardiac arrest. The goal of acute PE management is to decrease RV afterload through clot reduction—the rapidity at which this is clinically required depends on the degree of RV compromise and resultant hemodynamic consequences. Early risk stratification utilizing clinical, imaging, and serologic markers of RV strain are essential following PE diagnosis. The overwhelming majority of patients presenting with PE are low risk to intermediate-risk and have good short-term outcomes with anticoagulation alone.^4–7 However, patients that present with intermediate-risk PE and concerning features (eg, abnormal vital signs, significant right ventricular strain, elevated cardiac biomarkers), and particularly those with hemodynamically unstable, high-risk PE continue to have poor outcomes despite advances in therapies. Prompt evaluation and treatment is necessary as death occurs early in those with high-risk PE, as high as 27% within 24 hours of presentation and up to 49% by 72 hours.⁸ Notably, there are minimal high-quality data to guide strategies for this patient population given the relative infrequent incidence at individual centers and the difficulty in studying critically ill patients. Due to the poor outcomes in this group with traditional treatment options, there is growing interest in additional therapies, particularly catheter-based therapy and use of mechanical circulatory support (MCS), to help improve survival.

In this review, we discuss the traditional treatment options for high-risk PE and their limitations, as well as the potential for a paradigm shift to upfront percutaneous treatment of high-risk PE and the role of MSC.

HIGH-RISK PULMONARY EMBOLISM

High-risk PE as defined by the European Society of Cardiology Guidelines includes patients with: (1) cardiac arrest, (2) systolic blood pressure <90 mm Hg; or vasopressors to maintain a systolic blood pressure ≥90 mm Hg; and end-organ hypoperfusion (altered mental status, cold, oliguria/anuria, elevated lactate), or (3) systolic blood pressure <90 mm Hg; or drop in systolic blood pressure ≥40 mm Hg; lasting longer than 15 minutes without another clear explanation (eg, arrhythmia, sepsis, hypovolemia).⁹ Syncope is more common in patients with high-risk PE and more frequently associated with RV strain, but syncope alone does not meet criteria for high-risk PE. It can be a concerning feature on presentation; however, outcomes appear to be more associated with the concomitant hemodynamic profile of the patient rather than syncope itself.¹⁰ Incidence of high-risk PE varies based on the cohort evaluated, with estimates of 3% to 7% in multicenter studies and up to 12% in larger single institutional series.^1,4,5,11 Mortality in patients with high-risk PE is significantly higher than those with low- or intermediate-risk PE, with ranges of inpatient mortality reported from 15% to 52%.^5,8,11–13

The mainstay of immediate treatment in patients with confirmed or highly suspected PE is parenteral anticoagulation.⁹ Anticoagulation allows the body’s intrinsic fibrinolytic system to breakdown the thrombus, but this occurs over days to weeks and thus is often insufficient alone for patients with high-risk PE. As such, more active clot reduction therapies to rapidly reduce RV afterload are required. Advanced therapeutic options include systemic thrombolysis, catheter-directed thrombolysis, catheter-directed embolectomy, or surgical embolectomy. Additionally, MCS can be employed for patients with refractory shock to unload the RV as a bridge to recovery with or without concomitant advanced therapy.

NONCATHETER-BASED ADVANCED THERAPEUTIC OPTIONS

Currently, there is a dearth of evidence to support any specific advanced therapies for high-risk PE. As such, systemic thrombolysis remains the guideline recommended therapy for most patients with high-risk PE (Table 1).^9,14–18 Randomized data evaluating systemic thrombolysis in high-risk PE patients is extremely limited (ie, an 8-patient study in 1990s).¹⁹ A meta-analysis of studies comparing systemic thrombolysis to anticoagulation demonstrated a reduction in all-cause mortality among studies that included patients with high-risk PE (OR, 0.48 [CI, 0.20–1.15]), with a stronger association when evaluating PE-related mortality (OR, 0.15 [CI, 0.03–0.78]).²⁰ Systemic thrombolysis is widely available and easily administered without the need for specific equipment or training. However, there is often hesitancy to give systemic thrombolysis even in high-risk PE patients, because relative and absolute contraindications to systemic thrombolysis are common among high-risk PE patients.¹⁴ Older patients and those with comorbidities account for a large proportion of patients with high-risk PE and thus are much less likely to receive systemic thrombolysis, decreasing chances of survival.^1,21 Major bleeding is increased approximately 3-fold and intracranial hemorrhage nearly 8-fold in patients who receive systemic thrombolysis compared with anticoagulation alone.²² Observational data report major bleeding rates of over 20% and intracranial hemorrhage rates over 2% in high-risk PE patients.^23,24 As such, it is unsurprising that, in one nationwide study, systemic thrombolysis was administered to only 16.1% of unstable PE patients.¹³ Furthermore, systemic thrombolysis is often unsuccessful, and shock may persist despite administration. MCS is a potential option for such patients, but there is significant risk with large bore arterial cannulation required for veno-arterial extracorporeal membrane oxygenation (V-A ECMO) soon after systemic thrombolysis.²⁵ As such, it is unsurprising that national estimates of mortality with high-risk PE following systemic thrombolysis remain in the 40% to 50% range.¹³

Table 1.

Society Guidelines and Statements for Management of High-Risk PE

Society	Recommendation for high-risk PE management	Grade, level of evidence
ESC (2019)9	Systemic thrombolysis	I,B
	Surgical embolectomy (when systemic thrombolysis failed or contraindicated)	I,C
	Catheter-based therapy (when systemic thrombolysis failed or contraindicated)	IIa, C
	ECMO	IIb, C
AHA	Systemic thrombolysis	IIa, B
Scientific Statement (2011)14	Surgical embolectomy (when systemic thrombolysis failed or contraindicated)	IIb, C
Consensus Statement (2019)15	Catheter-based therapy (when systemic thrombolysis failed or contraindicated)	IIb, C
CHEST 2021 Update16	Systemic thrombolysis	Weak, low-certainty
	Catheter-based therapy (when systemic thrombolysis failed or contraindicated)	Weak, low-certainty
PERT Consortium Consensus Practice (2019)17	Systemic thrombolysis	NA
	Catheter-based therapy (when systemic thrombolysis failed or contraindicated)
	Surgical embolectomy (when systemic thrombolysis failed or contraindicated)
	MSC (with refractory shock or cardiac arrest)
Society of Interventional Radiology Position Statement (2018)18	Catheter-based therapy in highly selected patients	NA

AHA indicates American Heart Association; CHEST, The American College of Chest Physicians; ECMO, extracorporeal membrane oxygenation; ESC, European Society of Cardiology; MSC, mechanical circulatory support; NA, not applicable; PE, pulmonary embolism; and PERT, Pulmonary Embolism Response Team.

Surgical embolectomy is an additional option for management of high-risk PE. It is limited to institutions with cardiac surgery available and traditionally utilized in patients who have a contraindication to systemic thrombolysis; as a bail out procedure in those with persistent instability following thrombolysis; and for those with clot-in-transit. It requires cardiopulmonary bypass so patients are still exposed to high doses of heparin; as such patients at very high bleeding risk may not be candidates. In a national estimate, surgical embolectomy was performed in 4.3% of patients with high-risk PE, and 0.2% of all PEs.^13,26 Surgical embolectomy does provide rapid reduction in RV afterload and includes the option to bridge to MCS in the rare instances of persistent shock.²⁷ Utilization and experience differ widely, even among centers that have cardiac surgery available. In a nationwide cohort from 1998 to 2008, mortality remained 30% to 40% following surgical embolectomy in high-risk PE.²⁸ Additional data regarding outcomes are mostly limited to single-center observational series but demonstrate good outcomes in high-volume centers.^27,29 As such, surgical embolectomy may have an important role in dedicated centers but is unlikely to become a universal or practical primary treatment modality.

PERCUTANEOUS TREATMENT OPTIONS

The development of novel catheter-based therapies for PE offer alternatives to systemic thrombolysis for high-risk PE and have altered the risk-benefit ratio for invasive therapies in many patients. Despite limited data in high-risk PE, utilization of catheter-based therapies has increased in recent years. In a nationwide cohort study from 2016 to 2019, catheter-based therapy was performed in 6.6% of high-risk PE patients.²⁶ Percutaneous treatment options for acute PE consist of 2 main categories—with or without thrombolytic agents, as well as a hybrid approach (Table 2). In recent years, there has been rapid growth of mechanical thrombectomy that does not require thrombolysis. Mechanisms of mechanical thrombectomy include suction/aspiration, rheolytic therapy, extraction, and disruption. There are currently no randomized trials that compared catheter-based treatment to systemic thrombolysis or anticoagulation alone in the setting of high-risk PE. In the following sections, contemporary devices will be reviewed in the context of high-risk PE. A more extensive list and supporting data for intermediate- and high-risk PE is presented in Table S1.

Table 2.

Catheter-Based Therapy Options for Acute PE

Category	Device	Size cannula	Mechanism	Pros	Cons
Mechanical Thrombectomy
	Pigtail catheter, peripheral balloon catheters	5–7 Fr	Fragmentation of thrombus by spinning pigtail catheter and inflation of balloon catheters (6–16 mm)	Generally atraumatic, widely available	Risk for distal embolization. Often used in combination with thrombolysis or aspiration
	Aspirex	8 – 11 Fr	Spiral tip rotates at 40K RPM to fragment thrombus. Thrombus is then aspirated through adjacent port as device is advanced and withdrawn through thrombus	Both fragmentation and suction component. Rotational speed can be adjusted to prevent excessive suction and collapse of vessel	Primarily used in DVT, limited clinical experience in PE
	Amplatz thrombectomy device	7 – 8 Fr	Fragmentation, rheolytic. Air driven, high-pressure, high-speed impeller creates a vacuum which pulls clot into distal tip of catheter, fragments the thrombus, and then expels the particles	Both fragmentation and suction component	Difficult to maneuver especially into lobar arteries
	Angiojet	6 – 8 Fr	Rheolytic. Saline jet at the distal end of the catheter is used to fragment thrombus. Fragmented clot is subsequently aspirated via side ports using Bernoulli principle. Optional thrombolysis	Both fragmentation and suction component. Option for local thrombolysis	Systemic complications due to hemolysis and release of adenosine which can cause bradycardia, now with Black Box FDA warning (however brief treatment runs seems to mitigate). Renal insufficiency related to hemoglobinuria has also been seen
Suction or aspiration thrombectomy
	Angiovac	26 Fr outflow, 16 – 20 Fr inflow	Suction catheter combined with veno-venous recirculation system. Suction catheter with balloon actuated expandable funnel at distal tip.	Veno-venous blood return limits blood loss. Able to remove large volume of thrombus quickly due to large size	Requires perfusionist. Manipulation and steering can be difficult. Large bore access required.
	FlowTriever	20 – 22 Fr access, varying catheter sizes (6–10, 11–14, 15–18 mm)	Suction catheter connected to a retraction aspirator which provides vacuum for clot aspiration	Able to remove large volumes of thrombus quickly. Aspirated blood can be filtered and reinfused.	Size and rigidity makes access to PA branches more difficult. Large bore access required.
	Penumbra Indigo	8 Fr, 12 Fr	Flexible aspiration catheter connected to continuous suction or vacuum aspiration system. Wire separator in catheter lumen facilitates aspiration	Flexibility for placement in segmental branches	No mechanism for recirculation. Luminal diameter limits volume of thrombus aspirated
Catheter-directed thrombolysis
	Unifuse	4 – 5 Fr	Multihole catheter (end and sides). Introduced over guidewire. Occluding ball then advanced and obstructs end hole, forces infusate out of side holes	Pressure-response technology should provide even distribution of lytic	Typically requires 12–24 h for thrombolytic infusion
	Cragg-McNamara	4 – 5 Fr	Valved tip single lumen catheter with multihole infusion segment. Radiopaque markers located at proximal and distal ends of infusion segment	No guidewire needed to occlude (as in Unifuse), theoretically allowing for larger volume to be infused	Typically requires 12–24 h for thrombolytic infusion
	Ekosonic	5 Fr	Ultrasound-assisted infusion generates acoustic field and disperses fibrinolytic agent into the clot	Ultrasound may break up fibrin strands and allow for better penetration of thrombolytic allowing for lower doses	Benefit derived from ultrasound remains unclear, more expensive than standard CDT
	Bashir	7 Fr	Pharmaco-mechanical; expandable basket with 6 reinforced infusion limbs which can be opened within the thrombus	Mechanical disruption of clot may create a wide channel through the thrombus to improve delivery of thrombolytics	Limited clinical experience

CDT indicates catheter-directed thrombolysis; DVT, deep vein thrombosis; FDA, Food and Drug Administration; Fr, French; PA, pulmonary artery; PE, pulmonary embolism; and RPM, revolutions per minute.

CATHETER-DIRECTED THROMBOLYSIS

Catheter-directed thrombolysis delivers a thrombolytic agent directly into the pulmonary arteries with a lower total dose than is administrated intravenously and can be performed with or without ultrasound-facilitation. Ultrasound-facilitated, catheter-directed thrombolysis with the EKOSonic endovascular system (Boston Scientific, Marlborough, MA; Figure 1) is currently the only FDA-cleared local thrombolytic delivery system for the treatment of acute PE. With the EKOSonic system, small bore catheters are placed via any available venous access route into the pulmonary arteries and ultrasound waves are used to improve binding of tPA (tissue-type plasminogen activator) to fibrin crosslinks and help facilitate thrombolysis.³⁰

Figure 1.

Chest x-ray of Ekosonic catheter in pulmonary arteries.

In a small randomized trial of intermediate-risk PE, ultrasound-facilitated, catheter-directed thrombolysis more rapidly reduced PA pressures and markers of RV strain on imaging compared to anticoagulation alone.³¹ A variety of treatment regimens have been studied, with 8 to 24 mg of tPA administered over 4 to 24 hours in various studies.^32,33 The data for ultrasound-facilitated, catheter-directed thrombolysis is limited in high-risk PE and represents a small proportion of prior reports.

Bleeding may be reduced with catheter-directed thrombolysis compared to systemic thrombolysis based on observational data. In the single-armed SEATTLE II trial of 119 intermediate- and 31 high-risk PE patients, the overall rate of bleeding was 10%, with high-risk PE representing a significant risk factor bleeding (adjusted OR, 3.6 [95% CI, 1.01–12.9]; P=0.049).^32,34 With increasing experience and lower doses of thrombolytics, bleeding rates have decreased in more recent reports.³³ In a nationwide administrative claims database analysis, the bleeding rate was nearly double in patients who received systemic thrombolysis compared with catheter-directed thrombolysis (15.0% versus 8.7%; P<0.001), with triple the rate of intracranial hemorrhage (1.4% versus 0.5%; P 0.003).³⁵ A meta-analysis of prospective studies performed in 2019 reported major bleeding rates of 4.3% and intracranial hemorrhage rates of 0.7%.¹⁵ A recent registry study of ultrasound-facilitated, catheter-directed thrombolysis of 489 patients with intermediate-risk PE reported ISTH major bleeding risk of 2.5% with no intracranial hemorrhage.³⁶

There are limited comparisons of catheter-directed thrombolysis with and without ultrasound-facilitation (eg, Craig-McNamara catheter), particularly in high-risk PE patients. The SUNSET PE trial randomized 82 patients with intermediate-risk PE to catheter-directed thrombolysis with and without ultrasound-facilitation and found no difference in clot reduction.³⁷ Observational trials, which included a small percentage of patients with high-risk PE (7%–16%) and claims-based data in intermediate- and high-risk PE patients, also did not demonstrate a difference between the 2 approaches.^24,38,39 Ultrasound-facilitated, catheter-directed thrombolysis is notably approximately 10 times more expensive but is also associated with higher hospital reimbursement.³⁷

The more recently studied Bashir endovascular catheter (Thrombolex, Inc, New Britain, PA) is a 7 French (Fr) pharmaco-mechanical infusion device consisting of an expandable basket of 6 nitinol reinforced infusion limbs. The basket can be expanded and collapsed to create fissures for possible greater thrombolytic exposure to the thrombus. In a study of 109 patients with intermediate-risk PE, 7 mg of tPA were delivered per PA over 5 hours.⁴⁰ There was a 33.3% reduction in RV/LV ratio and 35.9% reduction in thrombus burden at 48 hours, with only 2 serious adverse events occurring in one patient.

A concern of catheter-directed thrombolysis in the high-risk PE patient is the rate at which RV afterload reduction occurs. There is a clear improvement after several hours, and it may be sufficient in patients that are relatively stable on vasopressor support or with low cardiac index without progressive hypotension; however, in patients with significant instability, a more rapid reduction in RV afterload is often required to avoid further decompensation. Additionally, the low, but non-negligible, risk of bleeding can be prohibitive in some patients and may also limit adjunctive treatment options like mechanical circulatory support.

MECHANICAL THROMBECTOMY

Over the past decade, there has been rapid growth in mechanical thrombectomy devices to extract clot from the PA, and many more are in development. Early catheter-based attempts at clot reduction primarily utilized thrombus fragmentation. Studies were generally small and effectiveness poorly described. Concomitant thrombolytic administration was also often employed making it difficult to isolate the benefit of mechanical therapy alone. In a meta-analysis reviewing mechanical therapy for high-risk PE published from 1990 to 2008, only 33% of the 594 patients received mechanical thrombectomy alone.⁴¹ Rotational thrombectomy via a rotating pigtail was most frequently utilized (70%). The primary outcome of clinical success, defined as stabilization of hemodynamics, resolution of hypoxia, and survival from PE, occurred at a rate of 86.5%. The pooled success rate was higher in studies that utilized concomitant thrombolytics (91%). Major procedural complications occurred in 2.4%.

Since this time, there has been development of multiple purpose-built PE mechanical thrombectomy devices—the most frequently used include the Flow-Triever System (Inari Medical, Irvine, CA), the Indigo Thrombectomy System (Penumbra, Inc, Alameda, CA) and the Angiovac/AlphaVac (Angiodynamics, Inc, Latham, NY). One of the first aspiration devices on the market was the Angiovac, which is a large bore (26 Fr outflow and 16–20 Fr inflow) veno-venous recirculation system that filters aspirated thrombus. The distal portion includes a balloon actuated, expandable funnel-shaped tip for aspiration. This was originally designed to facilitate removal of peripheral and intracardiac thrombi. However, given the large size and need for a perfusionist, it was not considered an optimal device for PE. In a study of 182 patients (81 with vegetation, 101 with thrombosis), procedural success rate was higher for vegetation removal (74.5%) and RA/caval thrombosis (80.5%) compared to PE (32.4%).⁴² Operative mortality was 14.6% with vegetation removal, 14.8% with RA/caval thrombus and 32.3% for PE patients. More recently, the AlphaVac (Angiodynamics) was developed using a newer, more deliverable cannula and a manual aspiration system independent of circulatory support. An ongoing study is examining its use for PE extraction (APEX-AV study; NCT05318092).

The FlowTriever System comes in various sizes and shapes, with the most frequently utilized device sized at 20 to 24 Fr. Access sites are typically either the internal jugular or femoral vein. The catheter is attached to a retraction aspirator to provide a vacuum for clot aspiration. There is an external blood return system to strain thrombus and return blood called FlowSaver. There are also add-on devices, including self-expanding nitinol disks and cages, which can aid in macerating thrombus and retrieving clot. These are more commonly utilized for perceived chronic or wall-adherent thrombus.

The FlowTriever System (Figure 2A and 2B) was initially evaluated in the single-armed FLARE trial in 106 intermediate-risk PE patients.⁴³ The primary outcome was change in RV/LV ratio, which was reduced by 25.1% at 48-hour follow-up. In those patients with an elevated mean PA pressure at baseline, there was a drop of 3.2-mm Hg; post-procedure (34.7±7.1 to 31.5±7.7 mm Hg; P<0.0001). Only 2 patients received adjunctive thrombolytics. Four patients with major procedure related clinical deteriorate, 1 of which was major bleeding due to pulmonary vascular injury. The recently reported FLASH registry evaluated the FlowTriever in 800 patients, 7.9% of which were high-risk PE.⁴⁴ Specific data for high-risk PE are not available, but mortality for the total cohort was 0.8% at 30 days with no device-related deaths. As an attempt to address the paucity of data for patients with massive PE, Inari Medical is also sponsoring the FLAME Trial (NCT04795167), which is examining 100 to 150 patients with high-risk PE. In parallel, Inari’s randomized PEERLESS trial comparing FlowTriever to catheter-directed thrombolysis in intermediate-high risk patients is ongoing (https://www.clinicaltrials.gov; Unique identifier: NCT05111613).

Figure 2. A patient with significant bilateral proximal pulmonary embolism (PE), hypotension, and hypoxemia requiring intubation was taken emergently to the cardiac catheterization lab.

FlowTriever catheter on fluoroscopy (A) and thrombus removed following procedure (B). Patient subsequently underwent placement of ProtekDuo right ventricular support with oxygenator given persistent shock and respiratory failure (C).

There have been a variety of retrospective, single-center case series utilizing FlowTriever in elevated risk PE with generally good outcomes, technical success, and low adverse events. The largest of these includes a report of 58 patients with intermediate-high or high-risk PE—28 received FlowTriever compared with 30 with routine care (the majority of which was anticoagulation alone).⁴⁵ Although confounded by treatment selection bias, in-hospital mortality was lower (3.6% versus 23.3%; P<0.05) and ICU length of stay was shorter (2.1±1.2 versus 6.1±8.6 days; P<0.05) in the FlowTriever group.

One limitation of the FlowTriever is the size and rigidity of the catheter, which can make it challenging to deliver or advance into more distal vessels. In addition, there is a learning curve, and experienced operators are required for optimal results compared with the relatively simple placement of an infusion catheter.

The Indigo system is an additional FDA-approved mechanical aspiration PE device with a unique design. It has a smaller profile with an 8 and 12 Fr catheter available and is a flexible aspiration catheter that is connected to a continuous suction vacuum system. There is also a wire separator in the catheter lumen to facilitate aspiration. Given the smaller caliber catheter, more distal clot can be accessed compared to the larger profile of the FlowTriever. There are limited data for utilization of the device in high-risk PE, but prospective data for intermediate risk PE. The EXTRACT-PE prospective, multicenter single-arm trial enrolled 119 intermediate-risk PE patients for treatment with the 8 Fr system. The primary outcome was RV/LV ratio at 48 hours post-procedure, which was reduced by 0.43 (95% CI, 0.38–0.47) compared to baseline of 1.47. Two major adverse events were reported including 1 patient with distal vessel perforation who died. Major bleeding occurred in 1.7%.^46,47 The ongoing STRIKE-PE Study (NCT04798261) will examine longer term safety and effectiveness of the Indigo system. A limitation of this system is the small Fr-size catheter, which may be useful for navigating deeper into the pulmonary tree, but may be less successful at aspirating large proximal or organized thrombus.

MECHANICAL CIRCULATORY SUPPORT

MCS can be complementary to percutaneous treatment options in the management of high-risk PE. MCS options include isolated right ventricular assist devices, with or without an oxygenator, or biventricular support, specifically with V-A ECMO. Currently marketed right ventricular assist devices include Impella RP and ProtekDuo. The Impella RP (Abiomed, Danvers, MA) is in axial flow pump placed through the RV and across the pulmonic valve in the left PA. It is inserted via a femoral vein with a 23 Fr cannula and can provide over 4 L/min of output. Theoretically, the Impella RP would be best utilized in patients with persistent RV failure that have already had clot reduction either by systemic thrombolysis or percutaneous therapy; otherwise, the pump may deliver blood up against any persistent left PA obstruction. Impella RP has not been specifically studied for acute PE with data limited to small case series but with reports of improved hemodynamics.⁴⁸ Given venous only access, it is a viable MCS option in patients that have recently received systemic thrombolysis.

The ProtekDuo (Tandem [Liva Nova], London, United Kingdom; Figure 2C) is a 29 or 31 Fr dual lumen cannula placed via the internal jugular vein with inflow cannula in the right atrium and outflow cannula in the main PA. It is attached to an extracorporeal pump with the option to splice in an oxygenator if additional oxygenation is needed. Similar to the Impella RP, more significant RV unloading can occur if clot burden is successfully reduced prior to placement. There are only a few small reports currently available regarding its role in acute PE, but it is a viable option in patients with significant hypoxemia and persistent shock.⁴⁹

The most reported MCS utilized in high-risk PE is V-A ECMO, though rates of use are only 0.2% of all PEs.^26,50 V-A ECMO offers biventricular support via inflow cannula from the femoral vein or IJ and outflow cannula via a peripheral artery. V-A ECMO has been reported in patients who have already received systemic fibrinolysis, surgical embolectomy, or used adjunctively with catheter-based therapies. However, V-A ECMO completely offloads the RV and bypasses the pulmonary circulation; thus, a clot reduction strategy is not required prior to initiation and may not be needed if recovery is observed with anticoagulation alone. A limitation compared to percutaneous right ventricular assist device support is the need for a large bore arterial cannula, which comes with access site and limb related complications. In a report from a large registry evaluating outcomes in patients with a variety of indications for VA-ECMO, cannulation site bleeding rates were 15% and approximately 7% developed limb ischemia.⁵¹ Reported bleeding rates following thrombolysis are as high as 61% to 100%.^25,52 If thrombolysis is given, the minimal time for safe cannulation is not well-defined. However, in patients with progressive shock or ongoing cardiac arrest, V-A ECMO is often the only option for potential survival.

The data are limited regarding the benefit of V-A ECMO. A pre/post V-A ECMO availability study at a single institution demonstrated 30-day survival for high-risk PE increased from 17.2% in the pre-ECMO era to 41.4% in the post-ECMO era, although unmeasured confounding likely contributed to this large survival difference.⁵³ In a large European report of V-A ECMO in PE, overall 30-day survival was 51.7%, and was the highest in those that received concomitant surgical embolectomy (85.7%), whereas no patient underwent concomitant catheter-based therapy.⁵⁴ In a high-volume center with robust V-A ECMO capabilities, early initiation in selected high-risk PE patients resulted in a mortality of 14.9%.⁵⁵

A PARADIGM SHIFT?: A PROPOSED NUANCED APPROACH TO HIGH-RISK PE

The current state of data is too limited to make definitive recommendations of utilization of catheter-based therapy and MCS in high-risk PE. Given the rapid evolution of percutaneous and MCS options for PE, current guidelines offer little input regarding the most appropriate role for these treatment options in high-risk PE (Table 1).

Though there has been a clear increase in the utilization of catheter-based therapies for acute PE, there has not yet been a true paradigm shift towards a more interventional approach for high-risk PE. However, as more data arises, safety profiles of these percutaneous interventions improve and V-A ECMO becomes more widely available, there may be a shift towards a primary percutaneous approach similar to what occurred with ST-elevation myocardial infarction and acute ischemic stroke. Given the significant shortcomings of systemic thrombolysis for PE, coupled with dismal outcomes and a growing armamentarium for PE treatment, it is time to consider a more comprehensive approach to high-risk PE in an attempt to improve outcomes.

At institutions with the full breadth of advanced therapies available, a more nuanced risk stratification approach for high-risk PE is warranted, weighing the patient’s risk and benefits of the potential interventions. In our centers, we differentiate high-risk PE from “catastrophic” PE. Catastrophic PE includes progressive shock despite multiple vasopressors/inotropes, impending/active cardiac arrest, or persistent shock despite thrombolysis (Figure 3A through 3D), whereas high-risk PE are able to remain clinically stable on a single vasopressor alone without rapidly escalating doses. For patients with “stable” high-risk PE, we favor catheter-based thrombectomy with backup MCS if needed post-procedure. We try to avoid thrombolysis, either systemically or via catheter, given the risk of bleeding that can delay other life-saving therapies like MCS.

Figure 3. Approaches to management of high-risk pulmonary embolism.

CBT indicates catheter-based thrombectomy; ECMO, extracorporeal membrane oxygenation; MCS, mechanical circulatory support; PE, pulmonary embolism; RV, right ventricular; and V-A, veno-arterial.

For patients with catastrophic PE, we favor initiation of mechanical circulatory support with V-A ECMO and then consider PE specific therapies once the patient is stabilized. For a subset of these patients, if V-A ECMO cannot be mobilized in a sufficiently rapid fashion, bolus systemic thrombolysis is utilized.⁵⁶ For patients with persistent shock early following catheter-based intervention, we pursue right-sided MCS as it often affords adequate support and the option for oxygenation while only requiring venous access. If patients are 6 to 12 hours post-thrombolysis, we will consider V-A ECMO if needed. Such decisions are made in discussion with the advanced heart failure/shock service to determine the candidacy of the patient and to select the MCS platform.

A more nuanced approach to high-risk PE requires a multidisciplinary team with institutional algorithms to optimize efficient delivery of complex care, as delays can have dire consequences. Defined treatment pathways integrating emergency departments (both at the tertiary care center and referring institutions), critical care providers, interventional cardiology/radiology, cardiac surgery, and cardiovascular medicine are vital to expedite care. PE response teams are well positioned to take on this role. Improved outcomes have been demonstrated at high-volume centers and expedited transfers should occur for high-risk PE patients.^57–59 The development of PE treatment networks can be important to assure adequate risk stratification and facilitate efficient and appropriate transfer.

CONCLUSIONS

Improving survival among patients presenting with high-risk PE is a priority among those invested in the evolving landscape of PE management. Greater device-based treatment options for managing acute PE highlight the importance of an endovascular approach even for high-risk PE. Given the limitations of systemic thrombolysis, there may be a paradigm shift toward a greater role for catheter-based therapies and MCS for treating high-risk PE in the coming years but will require parallel investments in generating high-quality evidence.

Supplementary Material

Table S1

References 60–107

Supplemental Material is available at https://www.ahajournals.org/doi/suppl/10.1161/CIRCINTERVENTIONS.122.012166.

Nonstandard Abbreviations and Acronyms

MCS

mechanical circulatory support

PA

pulmonary artery

PE

pulmonary embolism

RV

right ventricular

V-A ECMO

veno-arterial extracorporeal membrane oxygenation