JOURNAL OF THE AMERICAN COLLEGE OF CARDIOLOGY VOL. 76, NO.

18, 2020

PUBLISHED BY ELSEVIER ON BEHALF OF THE AMERICAN

COLLEGE OF CARDIOLOGY FOUNDATION

JACC FOCUS SEMINAR: VENOUS THROMBOEMBOLISM

JACC FOCUS SEMINAR

Advanced Management of Intermediate-

and High-Risk Pulmonary Embolism
JACC Focus Seminar

Gregory Piazza, MD, MS

ABSTRACT

Intermediate-risk (submassive) pulmonary embolism (PE) describes normotensive patients with evidence of right
ventricular compromise, whereas high-risk (massive) PE comprises those who have experienced hemodynamic decom-
pensation with hypotension, cardiogenic shock, or cardiac arrest. Together, these 2 syndromes represent the most
clinically challenging manifestations of the PE spectrum. Prompt therapeutic anticoagulation remains the cornerstone of
therapy for both intermediate- and high-risk PE. Patients with intermediate-risk PE who subsequently deteriorate despite
anticoagulation and those with high-risk PE require additional advanced therapies, typically focused on pulmonary artery
reperfusion. Strategies for reperfusion therapy include systemic fibrinolysis, surgical pulmonary embolectomy, and a
growing number of options for catheter-based therapy. Multidisciplinary PE response teams can aid in selection of
appropriate management strategies, especially where gaps in evidence exist and guideline recommendations are sparse.
(J Am Coll Cardiol 2020;76:2117–27) Published by Elsevier on behalf of the American College of Cardiology Foundation.

H eterogeneity of clinical
limited randomized controlled trial data,
and a burgeoning number of advanced
treatment options have established pulmonary embo-
presentation, (2,3). In-hospital mortality approaches 7% for all pa-
tients with PE and 33% for those presenting with he-
modynamic instability (4). Death due to PE is largely
from progressive right heart failure and occurs most
lism (PE) as one of the most challenging cardiovascu- commonly in patients with signs of right ventricular
lar disorders in clinical medicine. The incidence of PE (RV) dysfunction (intermediate-risk PE) or hemody-
in the United States is estimated to be 121 per 100,000 namic instability (high-risk PE) (5).
population (1). Greater sensitivity of diagnostic imag-
ing, an aging population, and increasing prevalence PATHOPHYSIOLOGY
of venous thromboembolism risk factors, such as
obesity and cancer, can be expected to continue to HEMODYNAMICS. Acute PE results in an abrupt in-
drive PE incidence. Although case fatality rates crease in pulmonary vascular resistance and RV
appear to be decreasing, PE-related mortality in afterload through direct physical obstruction, hyp-
the United States continues to be high, with oxemic vasoconstriction, and release of pulmonary
estimates ranging from 19.4 to 32.3 per 100,000 artery vasoconstrictors (Figure 1). Acute increases in

Listen to this manuscript’s

audio summary by
Editor-in-Chief
Dr. Valentin Fuster on From the Division of Cardiovascular Medicine, Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School,
JACC.org. Boston, Massachusetts. Dr. Piazza has received significant research grant support from BTG International, Bristol-Myers Squibb,
Daiichi-Sankyo, Bayer, Portola, and Janssen; and has received modest consulting fees from Pfizer and Thrombolex.
The author attests he is in compliance with human studies committees and animal welfare regulations of the authors’ institutions
and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC
author instructions page.

Manuscript received March 23, 2020; revised manuscript received May 4, 2020, accepted May 5, 2020.

ISSN 0735-1097/$36.00 https://doi.org/10.1016/j.jacc.2020.05.028

2118 Piazza JACC VOL. 76, NO. 18, 2020

Advanced Management of PE NOVEMBER 3, 2020:2117–27

ABBREVIATIONS afterload lead to RV dilation and hypo-

AND ACRONYMS HIGHLIGHTS
kinesis, tricuspid regurgitation, and ulti-
mately acute RV failure. Patients with RV  Patients with intermediate- and high-risk
CT = computed tomography
failure may abruptly decompensate, with PE represent the populations at highest
CTEPH = chronic
systemic arterial hypotension, cardiogenic risk for early mortality.
thromboembolic pulmonary
hypertension shock, and cardiac arrest.
 Although immediate anticoagulation is
DOAC = direct oral RV pressure overload may also result in
the cornerstone of management, pa-
anticoagulant interventricular septal deviation toward the
tients with intermediate- to high-risk
DVT = deep vein thrombosis left ventricle (LV), thereby limiting LV dias-
PE who deteriorate despite anticoagu-
ECMO = extracorporeal tole. Abnormal LV filling can be detected
lant therapy and those with high-risk
membrane oxygenation echocardiographically by transmitral Doppler
PE should be considered for advanced
FDA = U.S. Food and Drug with left atrial contraction, represented by
Administration therapies.
the A-wave, making a paradoxically greater
IVC = inferior vena cava contribution to diastole than passive filling,  Options for pulmonary reperfusion
LV = left ventricular represented by the E-wave (Figure 1). RV include systemic fibrinolysis, surgical
PE = pulmonary embolism pressure overload produces increased wall embolectomy, and a growing number of
PESI = Pulmonary Embolism stress and resultant ischemia by increasing catheter-based therapies.
Severity Index myocardial oxygen demand while simulta-
RV = right ventricular
 Clinical outcome-driven, randomized
neously limiting supply.
sPESI = simplified PESI
controlled trials are needed to define the
GAS EXCHANGE. Ventilation-to-perfusion
place of these advanced therapies in
t-PA = tissue-plasminogen mismatch, increases in total dead space, and
activator management pathways for intermediate-
right-to-left shunting contribute to pertur-
and high-risk PE.
bations of gas exchange in patients with acute PE.

F I G U R E 1 The Spectrum of RV Pathophysiological Changes in Acute PE as Demonstrated by Transthoracic Echocardiography

Increased RV Dilation,
Decreased LV
Pulmonary Hypokinesis,
Cardiac Output RV Ischemia,
Vascular and Abnormal LV
and Systemic Infarction, and
Resistance and Interventricular Filling
Arterial Failure
RV Pressure Septal
Hypotension
Overload Deviation

Increased pulmonary vascular resistance results in RV dilation and hypokinesis, interventricular septal deviation, and abnormal LV filling. Decreased left-sided cardiac
output results in systemic arterial hypotension, coronary artery hypoperfusion, and RV ischemia. LV ¼ left ventricular; PE ¼ pulmonary embolism; RV ¼ right
ventricular.
JACC VOL. 76, NO. 18, 2020 Piazza 2119
NOVEMBER 3, 2020:2117–27 Advanced Management of PE

The 2 most common gas exchange abnormalities are present with risk factors for adverse outcomes (2).
hypoxemia and increased alveolar-arterial oxygen Patients with high-risk (also termed massive) PE
gradient. Some patients with acute PE may hyper- present with syncope, systemic arterial hypotension,
ventilate, leading to hypocapnia and respiratory cardiogenic shock, or cardiac arrest. Patients with
alkalosis. Hypercapnia may accompany high-risk PE catastrophic, or “super-massive,” PE demonstrate
due to impaired minute ventilation and increased refractory shock or require ongoing cardiopulmonary
anatomic and physiological dead space. resuscitation and may require mechanical circulatory
support such as extracorporeal membrane oxygena-
LONG-TERM SEQUELAE
tion (ECMO). Intermediate-risk (also called sub-
massive) PE comprises a heterogeneous population
CHRONIC THROMBOEMBOLIC PULMONARY HYPERTENSION.
characterized by initially normal hemodynamics and
Chronic thromboembolic pulmonary hypertension
evidence of RV dysfunction. Intermediate-risk PE
(CTEPH) is characterized by persistent pulmonary
may be particularly challenging because a subset may
arterial obstruction, pulmonary vasoconstriction, and
suddenly, often without warning, develop systemic
a secondary small-vessel arteriopathy resulting in
arterial hypotension, cardiogenic shock, and sudden
chronic dyspnea and functional limitation and occurs
death, despite prompt therapeutic anticoagulation.
in 2% to 4% of patients after PE (6). Pulmonary
Intermediate-risk PE demonstrates considerable het-
thromboendarterectomy is the most effective and
erogeneity with regard to degree of RV dysfunction
durable therapy for CTEPH. Pulmonary vasodilators,
and prognosis, and may be further subcategorized
such as riociguat, offer the potential for improved
(Figure 2). Intermediate–high-risk patients with evi-
symptoms and functional capacity to patients with
dence of RV dysfunction on imaging and positive
CTEPH who are inoperable and or have post-
cardiac biomarkers are more likely to clinically dete-
thromboendarterectomy pulmonary hypertension
riorate than intermediate–low-risk patients who have
(7). A growing experience in balloon pulmonary an-
RV dysfunction on imaging, positive cardiac bio-
gioplasty provides an additional option for patients
markers, or neither.
who are not candidates for surgery (8). Advanced
therapy for PE, in particular systemic fibrinolysis, has RISK STRATIFICATION. Bedside scoring systems for
not been proven to prevent CTEPH (9). Because the prognostication of the risk of adverse outcomes and
evaluation and treatment are complex and evolving, therapeutic decision making, such as the PESI (Pul-
patients with CTEPH should be referred to special- monary Embolism Severity Index) and simplified PESI
ized centers. (sPESI), are validated tools for stratification of
POST-PE SYNDROME. The post-PE syndrome is patients with PE based on clinical parameters. High
characterized by persistent symptoms, including PESI and sPESI scores define a subset of patients with
chest pain and dyspnea, functional limitation, and increased 30-day mortality.
exercise intolerance in the absence of pulmonary Elevations in cardiac biomarkers, in particular
hypertension and is more common than CTEPH (10). cardiac troponin and brain-type natriuretic peptide,
The relative contribution of deconditioning versus correspond with RV pressure overload and
persistent cardiopulmonary limitations due to PE on result from RV microinfarction and increased shear
such symptoms is unclear (10). The impact of stress, respectively. Increased cardiac troponin and
advanced therapies such as systemic fibrinolysis or brain-type natriuretic peptide are associated with
catheter-based intervention on the frequency of the increased short-term mortality and adverse outcomes
post-PE syndrome remains undefined. However, in in patients with acute PE. Among normotensive pa-
long-term follow-up from the PEITHO (Pulmonary tients, cardiac biomarkers distinguish intermediate-
Embolism International Thrombolysis) trial, systemic risk from low-risk PE.
fibrinolysis did not reduce symptom burden or func- Detection of RV enlargement by contrast-enhanced
tional limitation in patients with intermediate-risk chest computed tomography (CT) has become an
PE (9). especially convenient risk stratification tool because
it uses data acquired from the initial diagnostic scan.
PE SYNDROMES AND THE APPROACH TO Based on measurements from an axial CT view, RV
RISK STRATIFICATION enlargement, defined as an RV diameter–to–left
ventricular (LV) diameter (RV-to-LV) ratio of >0.90,
SPECTRUM OF PE SYNDROMES. Although most pa- is an independent predictor of 30-day PE mortality. A
tients with PE have normal blood pressure, preserved systematic approach to echocardiography provides
RV function, and normal cardiac biomarkers, a subset accurate assessment of RV dysfunction while also
2120 Piazza JACC VOL. 76, NO. 18, 2020

Advanced Management of PE NOVEMBER 3, 2020:2117–27

F I G U R E 2 Risk Categories for Patients Presenting With Acute PE

Guidelines Category Hemodynamic PE Severity Index Evidence of RV Dysfunction

Status (PESI)
(or Simplified PESI)

American Heart Typically Abnormal RV on

Association (AHA, Massive Unstable High Imaging, Elevated
2011) Troponin, OR Both
May Have Abnormal RV on
Submassive Stable High Imaging OR Elevated
Troponin OR Both

Low Risk Stable Typically Low None

Typically Abnormal RV on
European Society of
High Risk Unstable High Imaging, Elevated
Cardiology (ESC,
Troponin, OR Both
2019)
Intermediate- Abnormal RV on Imaging,
Stable High
High Risk AND Elevated Troponin

May Have Abnormal RV on

Intermediate-Low
Stable High Imaging OR Elevated
Risk
Troponin But Not Both

Low Risk Stable Low None

Although terminology may differ, risk stratification strategies are consistent in considering hemodynamics, clinical presentation, and evidence of RV dysfunction.
PESI ¼ Pulmonary Embolism Severity Index; RV ¼ right ventricular.

evaluating for pulmonary hypertension in patients THE ROLE OF MULTIDISCIPLINARY

diagnosed with PE (2). Echocardiographic evidence RESPONSE TEAMS
of RV dysfunction defines intermediate-risk PE
and detects patients with an increased risk of Multidisciplinary PE response teams have emerged in
systemic arterial hypotension, cardiogenic shock, response to the clinical challenges of limited high-
and death. Echocardiography should be performed quality comparative data regarding advanced thera-
in patients with acute PE and clinical evidence of pies, rapidly advancing technology for device ther-
RV failure, elevated cardiac biomarkers, suspected apy, and a number of evidence-based clinical practice
pulmonary arterial hypertension, or clinical guidelines with varying and sometimes conflicting
deterioration. recommendations for patients with intermediate- or
Risk stratification algorithms should integrate high-risk PE (12). Modeled after established “heart
clinical prognostic indicators, cardiac biomarkers, team” approaches to myocardial infarction, stroke,
and evidence of RV dysfunction as detected by either and acute aortic syndromes, this rapid response
echocardiography or contrast-enhanced chest CT strategy harnesses multidisciplinary expertise,
(Central Illustration) (2). Patients with high-risk PE including cardiovascular medicine, pulmonology,
should be considered for advanced therapies for hematology, radiology, cardiac surgery, and inter-
reperfusion because of a high mortality with anti- ventionalists, to individualize PE care. Widespread
coagulation alone. A subset of patients with inter- adoption of a multidisciplinary response team
mediate–high-risk PE may be considered for rescue concept has been predicated on the promise of
reperfusion on a case-by-case basis based on clinical reduced heterogeneity of PE management both
deterioration or failure to improve despite therapeu- within individual medical centers and across health
tic anticoagulation (11). care systems, improved access to advanced therapies
JACC VOL. 76, NO. 18, 2020 Piazza 2121
NOVEMBER 3, 2020:2117–27 Advanced Management of PE

C ENTR AL I LL U STRA T I O N Approach to Risk Stratification and Risk-Based Management of Acute Pulmonary
Embolism Syndromes

A Patient with Acute Pulmonary Embolism

STEP 1: Is the patient hemodynamically stable?

YES NO

STEP 2: Does the patient have high-risk clinical features

(increased PESI or sPESI)?

STEP 3: Does the patient have CT or echocardiographic evidence of RV

dysfunction OR elevated troponin OR both?

Low-risk clinical High-risk clinical High-risk clinical

features, normal RV, features, abnormal RV features, abnormal RV,
AND normal troponin OR elevated troponin AND elevated troponin
OR neither

LOW-RISK PE INTERMEDIATE INTERMEDIATE HIGH-RISK PE

-LOW-RISK PE -HIGH-RISK PE

B Patient with Acute Pulmonary Embolism

STEP 1: Can the patient be anticoagulated?

YES NO

STEP 2: Start immediate therapeutic anticoagulation Consider IVC filter

LOW-RISK PE INTERMEDIATE INTERMEDIATE HIGH-RISK PE

-LOW-RISK PE -HIGH-RISK PE

STEP 3:
STEP 3: Continue anticoagulation STEP 3: Continue anticoagulation,
Continue anticoagulation, monitor in critical care setting,
monitor, and consider support hemodynamics,
reperfusion if deterioration and consider reperfusion

Piazza, G. J Am Coll Cardiol. 2020;76(18):2117–27.

Risk stratification incorporates assessment of hemodynamics and clinical features, cardiac biomarker determination, and detection of RV dysfunction on imaging.
Selection of advanced therapy should consider the risk of hemodynamic decompensation and early mortality with interventional complications and bleeding. (A)
Approach to risk stratification of patients with acute PE. (B) Risk-based management of acute PE syndromes. CT ¼ computed tomogram; IVC ¼ inferior vena cava;
PE ¼ pulmonary embolism; PESI ¼ Pulmonary Embolism Severity Index; RV ¼ right ventricular; sPESI ¼ simplified Pulmonary Embolism Severity Index.
2122 Piazza JACC VOL. 76, NO. 18, 2020

Advanced Management of PE NOVEMBER 3, 2020:2117–27

for those patients at increased risk of adverse out- SYSTEMIC FIBRINOLYSIS. The rationale for systemic
comes, more appropriate use of interventional ther- fibrinolysis in intermediate-risk PE is to avert
apies, improved clinical outcomes, and reduced impending hemodynamic collapse and death from
length of stay and health care costs (13). Despite the progressive right-sided heart failure. In patients with
rapid proliferation of this multidisciplinary approach high-risk PE, systemic fibrinolytic therapy is admin-
internationally, there have been limited reported istered to rapidly reverse hemodynamic compromise,
data regarding its real-world impact. The 2019 Euro- RV dysfunction, and gas exchange abnormalities.
pean Society of Cardiology Guidelines encourage a Systemic fibrinolysis is considered a lifesaving ther-
multidisciplinary response team approach in manage- apy in patients presenting with intermediate- and
ment of patients with intermediate- and high-risk PE (2). high-risk PE (15,16). The Europe-based PEITHO is the
largest randomized controlled trial of systemic fibri-
ANTICOAGULATION IN INTERMEDIATE-
nolysis in PE to date, enrolling 1,006 patients with
AND HIGH-RISK PE
intermediate-risk PE (17). The study evaluated the
impact of systemic fibrinolysis with tenecteplase fol-
Regardless of whether patients receive advanced lowed by anticoagulation with heparin versus heparin
therapy, prompt therapeutic-intensity anticoagulation alone on the primary outcome of all-cause mortality
comprises the cornerstone of treatment for interme- or hemodynamic collapse within 7 days of randomi-
diate- and high-risk PE (Figure 3). Anticoagulant zation. Systemic fibrinolysis reduced the frequency of
strategies for immediate treatment of acute PE the primary outcome (2.6% vs. 5.6%, p ¼ 0.015) with
include intravenous unfractionated heparin or most of the benefit due to a reduction in hemody-
injectable therapy with low-molecular weight heparin namic collapse within 7 days (1.6% vs. 5%, p ¼ 0.002).
or fondaparinux with transition to a direct oral anti- However, the benefit of fibrinolysis came at the cost
coagulant (DOAC) or vitamin K antagonist, or of increased major hemorrhage (6.3% vs. 1.5%,
completely oral monotherapy with a DOAC, such as p < 0.001), with approximately 2% of the
apixaban or rivaroxaban. Because it can be dis- tenecteplase-treated patients suffering intracra-
continued and rapidly reversed, unfractionated hep- nial hemorrhage.
arin has been the preferred anticoagulant for patients The U.S. Food and Drug Administration (FDA) has
undergoing advanced therapy with fibrinolysis, approved 100 mg tissue-plasminogen activator
catheter-based intervention, or surgery for PE. How- (t-PA) as a continuous infusion via a peripheral vein
ever, greater awareness regarding the danger of sub- over 2 h for the fibrinolysis of acute PE. Patients
therapeutic anticoagulation associated with being considered for fibrinolysis should be meticu-
unfractionated heparin has driven a growing prefer- lously assessed for contraindications. Concern over
ence for the consistent antithrombotic effect of low- the risk of intracranial hemorrhage, which ap-
molecular weight heparin in patients with PE with proaches 3% to 5% outside of clinical trials has
increased risk of adverse outcomes. DOACs represent dampened clinician enthusiasm for full-dose sys-
a considerable advancement in anticoagulation for temic fibrinolysis and has sparked development of
acute PE with comparable efficacy to vitamin K an- alternative fibrinolytic strategies with lower
tagonists but substantial reductions in bleeding bleeding risk.
complications and greater convenience (14). One alternative strategy has focused on half-dose
Evidence-based clinical practice guidelines recom- systemic fibrinolysis. Initial enthusiasm for this
mend DOACs as first line for oral anticoagulation in strategy was based on limited international and
patients with acute PE (2,11). single-center experiences (18,19). However, a more
recent propensity score–matched study comparing
ADVANCED THERAPIES FOR
outcomes in 3,768 patients receiving 50 mg versus
INTERMEDIATE- AND HIGH-RISK PE
full-dose 100 mg of alteplase for PE demonstrated

Advanced therapies for acute PE include systemic that half-dose fibrinolysis was associated with

fibrinolysis, catheter-based intervention, surgical increased frequency of treatment escalation (53.8%

pulmonary embolectomy, and mechanical circulatory vs. 41.4%; p < 0.01), driven largely by secondary

support (Table 1). Choosing a particular advanced fibrinolysis (25.9% vs. 7.3%; p < 0.01) and catheter-

therapy depends on the individual patient’s risk for directed therapy (14.2% vs. 3.8%; p < 0.01) (20).

adverse outcomes due to PE and for major bleeding, Furthermore, hospital mortality (13% vs. 15%;

most importantly intracranial hemorrhage (Central p ¼ 0.3), intracranial hemorrhage (0.5% vs. 0.4%;

Illustration). p ¼ 0.67), gastrointestinal bleeding (1.6% vs. 1.6%;

JACC VOL. 76, NO. 18, 2020 Piazza 2123
NOVEMBER 3, 2020:2117–27 Advanced Management of PE

F I G U R E 3 Advanced Management of Intermediate- and High-Risk Pulmonary Embolism

STEP 1: Administer
Therapeutic Anticoagulation

STEP 2: Risk Stratify to

Identify Intermediate-High
and High-Risk PE

STEP 3: Consider Advanced

Therapy for High-Risk and
Decompensated Intermediate-High
Pulmonary Embolism and High-Risk PE

A step-wise approach to management of pulmonary embolism (PE) begins with prompt initiation of therapeutic anticoagulation. Next, risk
stratification identifies patients with normal blood pressure but evidence of right ventricular dysfunction (intermediate-risk PE) and those
with hemodynamic instability (high-risk PE). Patients with intermediate–high-risk PE are monitored closely and considered for rescue
reperfusion if they develop subsequent clinical deterioration. Patients with high-risk PE should undergo expeditious reperfusion therapy, or if
in refractory cardiogenic shock, mechanical circulatory support.

p ¼ 0.99), and acute blood loss anemia (6.9% vs. the U.S.-based single-arm, multicenter SEATTLE
4.6%; p ¼ 0.11) were similar. (Prospective, Single-Arm Multi-Center Trial of
CATHETER-BASED THERAPY. Catheter-based ther- EkoSonic Endovascular System and Activase for
apy for treatment of acute PE includes pharmaco- Treatment of Acute Pulmonary Embolism) II trial, the
mechanical therapy, catheter-directed fibrinolysis, safety and efficacy of ultrasound-facilitated, catheter-
and mechanical embolectomy. Catheter-based ther- directed fibrinolysis (24 mg t-PA) was assessed in 150
apy combining local fibrinolysis with mechanical patients with high- (n ¼ 31) or intermediate-risk
thrombectomy offers the potential advantage of (n ¼ 119) PE (23). Mean RV-to-LV ratio decreased by
increased efficacy of thrombus dissolution due to the 25% (1.55 vs. 1.13; mean difference, 0.42; p < 0.0001),
synergistic effects of higher local fibrinolytic drug mean pulmonary artery systolic pressure decreased by
concentrations and mechanical disruption with 30% (51.4 mm Hg vs. 36.9 mm Hg; mean
greater exposed thrombus surface area. Because difference, 14.4 mm Hg; p < 0.0001), and mean
higher local drug concentration is achieved with modified Miller angiographic obstruction index
lower overall dose of fibrinolytic agent, catheter- diminished by 30% (22.5 vs. 15.8; mean
based therapy may offer the advantage of decreased difference, 6.6; p < 0.0001) from pre-procedure to
hemorrhagic complications. The evidence base for the 48 h post-procedure. Major bleeding occurred in 10%
efficacy and safety of these various catheter-based of patients, with none experiencing intracranial
techniques varies and is lacking the randomized hemorrhage. On May 21, 2014, the FDA cleared
controlled trials powered to evaluate clinical out- ultrasound-facilitated, catheter-directed fibrinolysis
comes (21). with the EkoSonic Endovascular System for treatment
The most extensively studied percutaneous of PE. In a subsequent dose-ranging trial, 4
technique for treatment of acute PE is ultrasound- accelerated-dosing regimens (8 min/2 h, 8 min/4 h,
facilitated, catheter-directed fibrinolysis (Boston 12 min/6 h, and 24 min/6 h) for ultrasound-facilitated,
Scientific Corporation, Marlborough, Massachusetts) catheter-directed fibrinolysis were evaluated in 101
(22–24). In a European randomized controlled trial of patients with intermediate-risk PE (24). All 4 regimens
59 patients with intermediate-risk PE, ultrasound- improved RV function comparable to 24 mg of t-PA
facilitated, catheter-directed fibrinolysis with low- administered over 12 to 24 h, based on the CT-
dose t-PA (20 mg total) plus anticoagulation reduced calculated RV-to-LV ratio from baseline to 48 h. Ma-
a surrogate endpoint, RV-to-LV ratio, from baseline to jor bleeding was observed in 4% of patients, with
24 h to a greater extent than anticoagulation (22). In intracranial hemorrhage in 1 patient who received an
2124 Piazza JACC VOL. 76, NO. 18, 2020

Advanced Management of PE NOVEMBER 3, 2020:2117–27

T A B L E 1 Options for Advanced Therapy in Acute PE

Option Indications Advantages Disadvantages

Systemic fibrinolysis High- and intermediate–high-  Rapid administration  2%–5% risk of ICH
risk PE  Decreases mortality
 Prevents hemodynamic collapse
 Expedites RV recovery and symptom relief
Catheter-directed therapy High- and intermediate–high-  Expedites RV recovery and symptom relief  Limited long-term and
risk PE  Reduced risk of ICH comparative data
 Option for mechanical embolectomy with  May take time to mobilize
some devices
Surgical embolectomy High- and intermediate–high-  Expedites RV recovery and symptom relief  Limited long-term and
risk PE  Reduced risk of ICH comparative data
 Avoids need for fibrinolysis  May take time to mobilize
 Limited to more centrally
located PE
ECMO Refractory cardiogenic shock  Supports hemodynamics and oxygenation in  Limited long-term and
patients with refractory shock or hypoxemia comparative data
 May take time to mobilize

ECMO ¼ extracorporeal membrane oxygenation; ICH ¼ intracranial hemorrhage; PE ¼ pulmonary embolism; RV ¼ right ventricular.

additional 50 mg of t-PA intravenously and another 2018. The Indigo Thrombectomy System (Penumbra,
with baseline pancytopenia and previously unknown Inc., Alameda, California) is a smaller-bore aspiration
arteriovenous malformation. In 1-year follow-up, catheter that does not require fibrinolytic adminis-
these accelerated lower-dose t-PA regimens for tration and that was evaluated in a single-arm study
ultrasound-facilitated, catheter-directed fibrinolysis of 119 patients with intermediate-risk PE (EXTRACT-
resulted in sustained RV recovery as assessed by serial PE [Evaluating the Safety and Efficacy of the Indigo
echocardiography and continued improvements in Aspiration System in Acute Pulmonary Embolism];
functional status and quality of life (25). A study using NCT03218566). Treatment with the Indigo device
a novel technique for 3-dimensional reconstruction of resulted in a 27% reduction in the mean CT-measured
the pulmonary vasculature from chest CT data ob- RV-to-LV diameter ratio and was associated with
tained in the SEATTLE II trial demonstrated that 3 major adverse events in 2 patients (28). The
reduction in RV volume correlated with increased AngioVac system (AngioDynamics, Inc., Latham, New
blood volume through the distal, rather than proximal, York) is a veno-veno bypass system that includes a
pulmonary arteries (26). These data suggest that 22-F suction thrombectomy catheter. Evidence for the
ultrasound-facilitated, catheter-directed fibrinolysis use of the AngioVac system to treat PE has been
may function to relieve RV pressure overload through limited (29).
distal pulmonary artery reperfusion. Other devices for catheter-based therapy in acute
Purely mechanical catheter embolectomy tech- PE are in various stages of development and study
niques may be advantageous in patients with PE with (21). Catheter-direct therapy, using local fibrinolysis
contraindications to fibrinolytic therapy. The Flow- without targeted mechanical thrombus disruption,
Triever system (Inari Medical, Irvine, California) is a has undergone limited prospective evaluation and
large-bore device that mechanically engages may be a consideration for patients with high-risk or
thrombus via 3 self-expanding nitinol disks and then intermediate-high-risk PE (30). Important knowledge
aspirates trapped thrombus. In typical practice, the gaps, such as the impact on clinical outcomes
device is used as a simple large-bore aspiration compared with anticoagulation alone and implica-
catheter without deployment of the nitinol disks. In a tions of time to catheter placement, procedure time,
U.S.-based single-arm, multicenter study of 106 pa- operator learning curve, procedural volumes, and
tients with intermediate-risk PE, embolectomy with cost, hinder the integration of catheter-based thera-
the FlowTriever system resulted in a 25% reduction in pies into clinical practice pathways for treatment of
CT-measured RV-to-LV ratio (mean difference, 0.38; PE (2,21). A number of guidance documents have
p < 0.0001) and 10% decrease in mean modified highlighted the lack of mortality data and the need for
Miller index (mean difference, 1.90; p < 0.001) (27). randomized controlled trials to elucidate the clinical
In the study, 6 major adverse events occurred in benefits versus risks of catheter-based therapy in
4 patients within 48 h of the procedure, including intermediate–high-risk and high-risk PE (2,11,21,31).
1 major bleeding event. The FlowTriever device Current evidence-based clinical practice guidelines
received FDA clearance for treatment of PE in May reflect the limitations in the data in their positions on
JACC VOL. 76, NO. 18, 2020 Piazza 2125
NOVEMBER 3, 2020:2117–27 Advanced Management of PE

catheter-based therapy. The 2019 European Society of maintaining systemic arterial perfusion pressure.
Cardiology Guidelines offer catheter-based therapy as Norepinephrine, epinephrine, and dopamine have
an alternative to surgical embolectomy for patients dual mechanisms of action as both inotropes and
with high-risk PE in whom systemic fibrinolysis has vasopressors and therefore may be preferred in the
failed or is contraindicated (Class IIa; Level of Evi- initial support of patients with high-risk PE. Inotropes
dence: C) and as an alternative to systemic fibrino- such as dobutamine may be required to augment
lysis in other patients with PE who have experienced cardiac output but may also cause systemic arterial
hemodynamic deterioration despite anticoagulation hypotension. In these cases, the addition of a vaso-
(Class IIa; Level of Evidence: C) (2). In patients with pressor may be necessary to support end-organ
acute PE associated with hypotension and have a high perfusion while administering inotropes. In other
risk of bleeding, have failed systemic fibrinolysis, or patients with high-risk PE and tachycardia, a primary
have shock that is likely to result in death before vasopressor such as vasopressin or phenylephrine
systemic fibrinolysis can take effect, the 2016 Amer- may be most appropriate to avoid accelerating the
ican College of Chest Physicians Guidelines suggest heart rate further. Although pulmonary vasodilators
catheter-directed therapy over no intervention, if the have the theoretical benefit of reducing pulmonary
expertise and resources are available (Grade 2C) (11). vascular resistance and improving RV function, a
multicenter randomized placebo-controlled trial of
SURGICAL EMBOLECTOMY. Surgical pulmonary em-
inhaled nitric oxide in patients with intermediate-risk
bolectomy is considered in patients with intermedi-
PE demonstrated no benefit on a primary endpoint of
ate–high- or high-risk PE in whom fibrinolysis has
complete RV recovery and normalization of cardiac
failed or is contraindicated (32). Rescue surgical pul-
troponin and a secondary endpoint of normalization
monary embolectomy after failed fibrinolysis is
of brain-type natriuretic peptide and Borg dyspnea
preferred over repeat fibrinolytic administration.
score <3 (34).
Other indications include paradoxical embolism,
ECMO is indicated for hemodynamic and ventila-
“clot-in-transit,” and hemodynamic collapse or res-
tory support in patients with severe RV failure and
piratory failure requiring cardiopulmonary resuscita-
refractory cardiogenic shock due to acute PE. Analysis
tion. Surgical pulmonary embolectomy is most
of the U.S. National Inpatient Sample has shown in
effective in patients with large centrally located PE.
upward trend in utilization of ECMO for patients with
In experienced centers, surgical pulmonary embo-
high-risk PE over the time period from 2005 to 2013
lectomy has been shown to be safe and effective (33).
(35). In-hospital mortality for high-risk PE patients
Optimal results are achieved when the patient is
receiving ECMO during this time period remained
referred before the development of pressor-
high at 61.6%. Predictors of increased mortality with
dependent hypotension or cardiogenic shock.
the use of ECMO for high-risk PE included increasing
HEMODYNAMIC SUPPORT FOR HIGH-RISK PE age, female sex, obesity, heart failure, and chronic
lung disease. Although ECMO has been traditionally
Although the initial strategy to manage hemodynamic used as a temporizing measure until advanced ther-
instability is often to augment RV preload with bolus apy, such as surgical embolectomy, can be instituted,
administration of intravenous fluids, excessive vol- more recent data suggest that most patients who
ume resuscitation may exacerbate RV failure by present with high-risk PE and are supported with
overdistending the RV, increasing wall stress, wors- ECMO will recover with anticoagulation alone (36).
ening RV ischemia, decreasing contractility, and Recent cannulation for ECMO is often viewed as a
causing further interventricular septal shift toward contraindication to systemic fibrinolysis.
the LV, thereby limiting LV filling and systemic car-
diac output. An initial trial of intravenous volume is INFERIOR VENA CAVA FILTERS
most likely to be successful in patients without signs
of increased right-sided preload, such as those with Inferior vena cava (IVC) filter insertion is considered in
central venous pressures of <15 mm Hg. In patients patients with acute PE with contraindications to anti-
with central venous pressures of >15 mm Hg, volume coagulation or with recurrent PE despite therapeutic
loading should be avoided, and administration of anticoagulation (2,11). IVC filter insertion had also
vasopressors and inotropes should be the initial step been considered on an individual basis for patients
in hemodynamic support. with intermediate- or high-risk PE who were receiving
The optimal agent for the hemodynamic support of therapeutic anticoagulation but who had limited car-
patients with high-risk PE should augment RV func- diopulmonary reserve, such that a subsequent PE
tion through positive inotropic effects while also would likely be fatal. This indication was the focus of
2126 Piazza JACC VOL. 76, NO. 18, 2020

Advanced Management of PE NOVEMBER 3, 2020:2117–27

the PREPIC2 (Prevention of Recurrent Pulmonary FUTURE DIRECTIONS IN ADVANCED CARE

Embolism by Vena Cava Interruption) trial, which FOR INTERMEDIATE- AND HIGH-RISK PE
randomly assigned 399 normotensive patients with
acute PE, concomitant lower extremity deep vein Although the past decade was marked by remark-
thrombosis (DVT), and at least 1 risk factor for adverse able growth in PE-related clinical investigation,
outcomes to retrievable IVC filter implantation plus several critical research needs persist. More precise
anticoagulation versus anticoagulation alone (37). risk stratification tools to pre-emptively identify
Adjunctive insertion of a retrievable IVC filter, patients with intermediate-risk PE with the highest
compared with anticoagulation alone, did not reduce risk of clinical deterioration will be necessary for
the risk of symptomatic recurrent PE or mortality at 3 selection of those who would benefit from
or 6 months. Based on these findings, IVC filters should advanced therapies. The mortality rate for high-risk
not be routinely inserted in patients with intermedi- PE remains unacceptably high. Strategies for
ate- and high-risk PE who can be treated with anti- selection of the optimal advanced therapy and he-
coagulation. In a meta-analysis summarizing data modynamic support in patients with high-risk PE
from randomized controlled trials and prospective are sorely needed. Although gaining widespread
controlled observational studies, IVC filters appear to acceptance and demonstrating great potential,
reduce the short-term risk of subsequent PE, increase multidisciplinary PE response teams warrant simi-
the long-term risk for DVT, and have no impact on larly rigorous clinical evaluation as have been
overall mortality (38). demanded from medical and device therapies to
After an FDA advisory regarding IVC filter utiliza- better understand their benefits and costs. Finally,
tion in 2010 and updates to societal guidelines, the burgeoning area of device therapies for PE calls
annual IVC filter use has declined in the United States for appropriately powered, clinical endpoint-driven,
(39). Despite data demonstrating the safety and ease randomized controlled trials to define their place in
with which retrievable IVC filters can be removed, up clinical pathways for intermediate- and high-risk PE
to 50% remain permanently indwelling (40,41). management.
Device-related complications include strut fracture,
filter migration, strut embolization, device tilt, IVC
penetration, perforation of surrounding structures, ADDRESS FOR CORRESPONDENCE: Dr. Gregory
PE, DVT, and IVC thrombosis. To avoid such compli- Piazza, Division of Cardiovascular Medicine, Brigham
cations, IVC filters should be retrieved as soon as no and Women’s Hospital, 75 Francis Street, Boston,
longer necessary and anticoagulation has been safely Massachusetts 02115. E-mail: gpiazza@partners.org.
initiated. Twitter: @BrighamWomens.

REFERENCES

1. Centers for Disease Control and Prevention. 7. Ghofrani HA, D’Armini AM, Grimminger F, et al. 12. Dudzinski DM, Piazza G. Multidisciplinary pul-
Venous thromboembolism in adult hospitaliza- Riociguat for the treatment of chronic thrombo- monary embolism response teams. Circulation
tions - United States, 2007–2009. MMWR Morb embolic pulmonary hypertension. N Engl J Med 2016;133:98–103.
Mortal Wkly Rep 2012;61:401–4. 2013;369:319–29.
13. Giri JS, Piazza G. A midterm report card for
2. Konstantinides SV, Meyer G, Becattini C, et al. 8. Kataoka M, Inami T, Kawakami T, Fukuda K, pulmonary embolism response teams. Vasc Med
2019 ESC Guidelines for the diagnosis and Satoh T. Balloon pulmonary angioplasty (percu- 2018;23:72–4.
management of acute pulmonary embolism taneous transluminal pulmonary angioplasty) for
14. van Es N, Coppens M, Schulman S,
developed in collaboration with the European chronic thromboembolic pulmonary hypertension:
Middeldorp S, Buller HR. Direct oral anticoagu-
Respiratory Society (ERS). Eur Heart J 2019;41: a Japanese perspective. J Am Coll Cardiol Intv
lants compared with vitamin K antagonists for
543–603. 2019;12:1382–8.
acute venous thromboembolism: evidence from
3. Wendelboe AM, Raskob GE. Global burden of 9. Konstantinides SV, Vicaut E, Danays T, et al. phase 3 trials. Blood 2014;124:1968–75.
thrombosis: epidemiologic aspects. Circ Res 2016; Impact of thrombolytic therapy on the long-term
15. Chatterjee S, Chakraborty A, Weinberg I, et al.
118:1340–7. outcome of intermediate-risk pulmonary embo-
Thrombolysis for pulmonary embolism and risk of
lism. J Am Coll Cardiol 2017;69:1536–44.
4. Casazza F, Becattini C, Bongarzoni A, et al. all-cause mortality, major bleeding, and intracra-
Clinical features and short term outcomes of pa- 10. Kahn SR, Akaberi A, Granton JT, et al. Quality nial hemorrhage: a meta-analysis. JAMA 2014;311:
tients with acute pulmonary embolism. The Italian of life, dyspnea, and functional exercise capacity 2414–21.
Pulmonary Embolism Registry (IPER). Thromb Res following a first episode of pulmonary embolism: 16. Marti C, John G, Konstantinides S, et al. Sys-
2012;130:847–52. results of the ELOPE Cohort Study. Am J Med temic thrombolytic therapy for acute pulmonary
5. Piazza G. Submassive pulmonary embolism. 2017;130:990.e9–21. embolism: a systematic review and meta-analysis.
JAMA 2013;309:171–80. Eur Heart J 2015;36:605–14.
11. Kearon C, Akl EA, Ornelas J, et al. Antith-
6. Piazza G, Goldhaber SZ. Chronic thromboem- rombotic therapy for VTE Disease: CHEST guide- 17. Meyer G, Vicaut E, Danays T, et al. Fibrinolysis
bolic pulmonary hypertension. N Engl J Med 2011; line and expert panel report. Chest 2016;149: for patients with intermediate-risk pulmonary
364:351–60. 315–52. embolism. N Engl J Med 2014;370:1402–11.
JACC VOL. 76, NO. 18, 2020 Piazza 2127
NOVEMBER 3, 2020:2117–27 Advanced Management of PE

18. Sharifi M, Bay C, Skrocki L, Rahimi F, TLE-3D): a substudy of SEATTLE-2. Circ Car- 36. Ghoreishi M, DiChiacchio L, Pasrija C,
Mehdipour M, “MOPETT” Investigators. Moderate diovasc Imaging 2019;12:e009903. et al. Predictors of recovery in patients
pulmonary embolism treated with thrombolysis supported with VA-ECMO for acute massive
27. Tu T, Toma C, Tapson VF, et al. A prospective,
(from the “MOPETT” Trial). The Am J Cardiol 2013; pulmonary embolism. Ann Thorac Surg 2020;
single-arm, multicenter trial of catheter-directed
111:273–7. 110:70–5.
mechanical thrombectomy for intermediate-risk
19. Wang C, Zhai Z, Yang Y, et al. Efficacy and acute pulmonary embolism: The FLARE Study. J 37. Mismetti P, Laporte S, Pellerin O, et al.
safety of low dose recombinant tissue-type plas- Am Coll Cardiol Intv 2019;12:859–69. Effect of a retrievable inferior vena cava filter
minogen activator for the treatment of acute plus anticoagulation vs anticoagulation alone
28. VIVA 2019: Penumbra Indigo Aspiration sys-
pulmonary thromboembolism: a randomized, on risk of recurrent pulmonary embolism: a
tem IDE trial for acute PE meets primary safety
multicenter, controlled trial. Chest 2010;137: randomized clinical trial. JAMA 2015;313:
and efficacy endpoints. 2019. Available at: https://
254–62. 1627–35.
vascularnews.com/viva19-penumbra-indigo-aspi-
20. Kiser TH, Burnham EL, Clark B, et al. Half-dose ration-system-ide-trial-for-acute-pe-meets-pri- 38. Bikdeli B, Chatterjee S, Desai NR, et al. Inferior
versus full-dose alteplase for treatment of pul- mary-safety-and-efficacy-endpoints/. Accessed vena cava filters to prevent pulmonary embolism:
monary embolism. Crit Care Med 2018;46: March 21, 2020. systematic review and meta-analysis. J Am Coll
1617–25. Cardiol 2017;70:1587–97.
29. Donaldson CW, Baker JN, Narayan RL, et al.
21. Giri J, Sista AK, Weinberg I, et al. Interventional Thrombectomy using suction filtration and veno- 39. Ahmed O, Patel K, Patel MV, et al.
therapies for acute pulmonary embolism: current venous bypass: single center experience with a Declining national annual IVC filter utilization:
status and principles for the development of novel novel device. Catheter Cardiovasc Interv 2015;86: an analysis on the impact of societal and
evidence: a scientific statement from the American E81–7. governmental communications. Chest 2017;151:
Heart Association. Circulation 2019;140:e774–801. 1402–4.
30. Kuo WT, Banerjee A, Kim PS, et al. Pulmonary
22. Kucher N, Boekstegers P, Muller O, et al. Embolism Response to Fragmentation, Embolec- 40. Rottenstreich A, Spectre G, Roth B,
Randomized controlled trial of ultrasound- tomy, and Catheter Thrombolysis (PERFECT): Bloom AI, Kalish Y. Patterns of use and
assisted catheter-directed thrombolysis for acute initial results from a prospective multicenter reg- outcome of inferior vena cava filters in a
intermediate-risk pulmonary embolism. Circula- istry. Chest 2015;148:667–73. tertiary care setting. Eur J Haematol 2015 Mar
tion 2014;129:479–86. 6 [E-pub ahead of print].
31. Jaff MR, McMurtry MS, Archer SL, et al. Man-
41. Sutphin PD, Reis SP, McKune A,
23. Piazza G, Hohlfelder B, Jaff MR, et al. agement of massive and submassive pulmonary
Ravanzo M, Kalva SP, Pillai AK. Improving
A prospective, single-arm, multicenter trial of embolism, iliofemoral deep vein thrombosis, and
inferior vena cava filter retrieval rates with the
ultrasound-facilitated, catheter-directed, low- chronic thromboembolic pulmonary hypertension:
define, measure, analyze, improve, control
dose fibrinolysis for acute massive and sub- a scientific statement from the American Heart
methodology. J Vasc Interv Radiol 2015;26:
massive pulmonary embolism: The SEATTLE II Association. Circulation 2011;123:1788–830.
491–8.e1.
Study. J Am Coll Cardiol Intv 2015;8:1382–92. 32. Poterucha TJ, Bergmark B, Aranki S, Kaneko T,
24. Tapson VF, Sterling K, Jones N, et al. Piazza G. Surgical pulmonary embolectomy. Cir-
A randomized trial of the optimum duration of culation 2015;132:1146–51.
acoustic pulse thrombolysis procedure in acute 33. Kolkailah AA, Hirji S, Piazza G, et al. Surgical KEY WORDS anticoagulation, catheter-
intermediate-risk pulmonary embolism: The OPTA- pulmonary embolectomy and catheter-directed directed therapy, catheter thrombolysis,
LYSE PE Trial. J Am Coll Cardiol Intv 2018;11:1401–10. thrombolysis for treatment of submassive pul- fibrinolysis, IVC filter, mechanical circulatory
25. Piazza G, Sterling KM, Tapson VF, et al. One-year monary embolism. J Card Surg 2018;33:252–9. support, pulmonary embolism, right
echocardiographic, functional, and quality of life 34. Kline JA, Puskarich MA, Jones AE, et al. ventricular failure, surgical embolectomy,
outcomes after ultrasound-facilitated catheter- Inhaled nitric oxide to treat intermediate risk thrombolysis
based fibrinolysis for pulmonary embolism. Circ pulmonary embolism: a multicenter randomized
Cardiovasc Interv 2020 Aug 6 [E-pub ahead of print]. controlled trial. Nitric Oxide 2019;84:60–8.
Go to http://www.acc.org/
26. Rahaghi FN, San José Estépar R, Goldhaber SZ, 35. Elbadawi A, Mentias A, Elgendy IY, et al.
jacc-journals-cme to take
et al. Quantification and significance of the National trends and outcomes for extra-
the CME/MOC/ECME quiz
pulmonary vascular response to ultrasound- corporeal membrane oxygenation use in high-
for this article.
facilitated, catheter-directed fibrinolysis in risk pulmonary embolism. Vasc Med 2019;24:
patients with acute pulmonary embolism (SEAT- 230–3.