JOURNAL OF THE AMERICAN COLLEGE OF CARDIOLOGY VOL. 75, NO.

14, 2020

ª 2020 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION

PUBLISHED BY ELSEVIER

Antithrombotic Therapy for Patients With

Left Ventricular Mural Thrombus
Benoit Lattuca, MD,a,* Nesrine Bouziri, MD,a,* Mathieu Kerneis, MD,a Jean-Jacques Portal, PHD,b
Jiannong Zhou, BSC,c Marie Hauguel-Moreau, MD,a Amel Mameri, MD,a Michel Zeitouni, MD,a Paul Guedeney, MD,a
Nadjib Hammoudi, MD, PHD,a Richard Isnard, MD, PHD,a Françoise Pousset, MD,a Jean-Philippe Collet, MD, PHD,a
Eric Vicaut, MD, PHD,b Gilles Montalescot, MD, PHD,a Johanne Silvain, MD, PHD,a for the ACTION Study Group

ABSTRACT

BACKGROUND Contemporary data are lacking regarding the prognosis and management of left ventricular thrombus
(LVT).

OBJECTIVES The purpose of this study was to quantify the effect of anticoagulation therapy on LVT evolution using
sequential imaging and to determine the impact of LVT regression on the incidence of thromboembolism, bleeding, and
mortality.

METHODS From January 2011 to January 2018, a comprehensive computerized search of LVT was conducted using
90,065 consecutive echocardiogram reports. Only patients with a confirmed LVT were included after imaging review by 2
independent experts. Major adverse cardiovascular events (MACE), which included death, stroke, myocardial infarction, or
acute peripheral artery emboli, were determined as well as major bleeding events (BARC $3) and all-cause mortality rates.

RESULTS There were 159 patients with a confirmed LVT. Patients were treated with vitamin K antagonists (48.4%),
parenteral heparins (27.7%), and direct oral anticoagulants (22.6%). Antiplatelet therapy was used in 67.9% of the
population. A reduction of the LVT area from baseline was observed in 121 patients (76.1%), and total LVT regression
occurred in 99 patients (62.3%) within a median time of 103 days (interquartile range: 32 to 392 days). The independent
correlates of LVT regression were a nonischemic cardiomyopathy (hazard ratio [HR]: 2.74; 95% confidence interval [CI]:
1.43 to 5.26; p ¼ 0.002) and a smaller baseline thrombus area (HR: 0.66; 95% CI: 0.45 to 0.96; p ¼ 0.031). The fre-
quency of MACE was 37.1%; mortality 18.9%; stroke 13.3%; and major bleeding 13.2% during a median follow-up of
632 days (interquartile range: 187 to 1,126 days). MACE occurred in 35.4% and 40.0% of patients with total LVT
regression and those with persistent LVT (p ¼ 0.203). A reduced risk of mortality was observed among patients with total
LVT regression (HR: 0.48; 95% CI: 0.23 to 0.98; p ¼ 0.039), whereas an increased major bleeding risk was observed
among patients with persistent LVT (9.1% vs. 12%; HR 0.34; 95% CI: 0.14 to 0.82; p ¼ 0.011). A left ventricular ejection
fraction $35% (HR: 0.46; 95% CI: 0.23 to 0.93; p ¼ 0.029) and anticoagulation therapy >3 months (HR: 0.42; 95% CI:
0.20 to 0.88; p ¼ 0.021) were independently associated with less MACE.

CONCLUSIONS The presence of LVT was associated with a very high risk of MACE and mortality. Total LVT regression,
obtained with different anticoagulant regimens, was associated with reduced mortality.
(J Am Coll Cardiol 2020;75:1676–85) © 2020 by the American College of Cardiology Foundation.

Listen to this manuscript’s

audio summary by
T he advent of reperfusion therapy and the
widespread use of primary percutaneous cor-
onary intervention (PCI) have markedly
reduced the incidence of post–myocardial infarction
(MI) left ventricular thrombus (LVT)
last decades (1–3). Nevertheless, contemporary epide-
miological studies suggest that the incidence of LVT
may remain as high as 15% to 25% in patients with
over the

Editor-in-Chief
Dr. Valentin Fuster on
From the aSorbonne Université, ACTION Study Group, INSERM UMR_S 1166, Institut de Cardiologie, Pitié-Salpêtrière Hospital
JACC.org.
(AP-HP), Paris, France; bUnité de Recherche Clinique, Lariboisière Hospital (AP-HP), ACTION Study Group, Paris, France; and the
c
Information system department, Pitié-Salpêtrière Hospital (AP-HP), Paris, France. *Drs. Lattuca and Bouziri contributed equally
to this work. This work was supported by Allies in Cardiovascular Trials Initiatives and Organized Networks ACTION Group. Dr.
Lattuca has received research grants from Biotronik, Boston Scientific, Daiichi-Sankyo, Fédération Française de Cardiologie, and
Institute of CardioMetabolism and Nutrition; has received consultant fees from Daiichi-Sankyo and Eli Lilly; and has received
lecture fees from AstraZeneca and Novartis. Dr. Kerneis has received research grants from Fédération Française de Cardiologie,

ISSN 0735-1097/$36.00 https://doi.org/10.1016/j.jacc.2020.01.057

JACC VOL. 75, NO. 14, 2020 Lattuca et al. 1677
APRIL 14, 2020:1676–85 Left Ventricular Thrombus and Cardiovascular Events

ST-segment elevation myocardial infarction (STEMI) contrast transthoracic echocardiography, ABBREVIATIONS

(4–6) and up to 36% in the setting of dilated cardio- cardiac computed tomography scan, or car- AND ACRONYMS

myopathy (7,8) when detected with optimal imaging diac magnetic resonance imaging) and a third
BARC = Bleeding Academic
modalities. expert (N.H.) was requested to confirm the Research Consortium
Despite adequate anticoagulation therapy, LVT diagnosis. Patients with right ventricular
DOAC = direct oral
remains a severe complication associated with a high thrombus and atrial thrombus were excluded anticoagulant
risk of cerebral and peripheral arterial embolism and from the study. LVEF = left ventricular
subsequent mortality (9–11). American and European THROMBUS EVALUATION. LVT was defined
ejection fraction

Guidelines (12,13) recommend vitamin K antagonist as an echodense mass adjacent to a hypo- LVT = left ventricular

(VKA) for at least 3 to 6 months. The duration must be thrombus

kinetic or akinetic myocardial segment. To be
individualized according to bleeding risk. The need MACE = major adverse
distinguishable from the underlying myocar-
cardiovascular events
for concomitant antiplatelet therapy may be consid- dium, a clear thrombus–blood interface was
MI = myocardial infarction
ered. There is limited supportive evidence (14,15). required and the LVT had to be visible on at
The effect of anticoagulation on LVT regression as STEMI = ST-segment elevation
least 2 views during all of the cardiac cycle. myocardial infarction
well as the optimal anticoagulation therapy duration LVT diameters (mm), area (cm 2), and volume
VKA = vitamin-K antagonist
and their potential impact on clinical outcomes are (cm 3) were assessed for each transthoracic
lacking. Moreover, no study has, to date, evaluated echocardiogram considering the mean of 3 measures
the efficacy and the safety of direct oral anticoagu- to study LVT evolution over time (Supplemental
lants (DOACs) in this clinical setting. Figure 1). Mobility and location of LVT were also
SEE PAGE 1686 defined. LVT was considered mobile if thrombus had
a pedunculated nonmural component. A calcified LVT
The purposes of this study were: first, to describe was defined as a persistent left ventricular mural
the effect of anticoagulation therapy on LVT evolu- thrombus encapsulated by thickened and calcified
tion; and second, to identify the independent pre- endocardium. Left ventricular characteristics
dictors of LVT regression and clinical outcomes. including left ventricular ejection fraction (LVEF),
METHODS left ventricular volume, wall motion, cardiac output,
and potential mechanical complications were also
STUDY DESIGN AND POPULATION. From January collected. All echocardiograms were reviewed using
2011 to December 2017, all reports of consecutive the dedicated TOMTEC imaging system tools (TOM-
echocardiogram performed at the Institute of Cardi- TEC, Unterschleissheim, Germany).
ology, Pitié-Salpêtrière hospital (Paris, France), were BASELINE AND DATA COLLECTION. All patient
analyzed using a computerized case-sensitive search. baseline characteristics including cardiovascular risk
All patients with a reported LVT, regardless of the factors, past medical history of major bleeding or
underlying disease, were screened, and only patients ischemic event, underlying disease, and long-term
with a confirmed LVT by 2 independent experts (N.B. use of anticoagulation therapy before the diagnosis
and A.M.), were included. In case of discordance, an of LVT were collected through medical reports. Spe-
additional imaging modality was performed (i.e., cial attention was given to the type (VKA, DOACs,

European Society of Cardiology, and Servier; and has received consultant fees from AstraZeneca, Bayer, Daiichi-Sankyo, Eli Lilly,
and Sanofi. Dr. Zeitouni has received research grants from Fédération Française de Cardiologie and Servier. Dr. Hammoudi has
received research grants to the institution or consulting/lecture fees from Philips, Bayer, Laboratoires Servier, Novartis Pharma,
AstraZeneca, Bristol-Myers Squibb, Merck Sharp & Dohme, Fédération Française de Cardiologie, and ICAN. Dr. Isnard has served
on the Advisory Board of and led lectures for Novartis, Servier, and Bayer; and has led lectures for Amgen and Pfizer. Dr. Collet has
received research grants from AstraZeneca, Bayer, Bristol-Myers Squibb, Boston Scientific, Daiichi-Sankyo, Eli Lilly, Fédération
Française de Cardiologie, Lead-Up, Medtronic, Merck Sharp & Dohme, Sanofi, and WebMD. Dr. Vicaut has received research grants
from Boehringer and Sanofi; and has received consultant fees from Abbott, Eli Lilly, Hexacath, Fresenius, LFB, Medtronic,
Novartis, Pfizer, and Sanofi. Dr. Montalescot has received research grants from Abbott, Amgen, Actelion, American College of
Cardiology Foundation, AstraZeneca, Axis-Santé, Bayer, Boston Scientific, Boehringer Ingelheim, Bristol-Myers Squibb, Beth Israel
Deaconess Medical, Brigham Women’s Hospital, China Heart House, Daiichi-Sankyo, Idorsia, Elsevier, Europa, Fédération Fran-
çaise de Cardiologie, ICAN, Lead-Up, Medtronic, Menarini, Merck Sharp & Dohme, NovoNordisk, Partners, Pfizer, Quantum Ge-
nomics, Sanofi, Servier, and WebMD. Dr. Silvain has received research grants from Algorythm and Fondation de France; and has
received lecture fees from Amed, Amgen, AstraZeneca, Bayer, CSL Behring, Daiichi-Sankyo, Eli Lilly, Gilead Science, Iroko Cardio,
Sanofi, and St. Jude Medical. All other authors have reported that they have no relationships relevant to the contents of this paper
to disclose. Michael D. Ezekowitz MBChB, DPhil, served as Guest Associate Editor for this paper.

Manuscript received June 28, 2019; revised manuscript received January 28, 2020, accepted January 30, 2020.
1678 Lattuca et al. JACC VOL. 75, NO. 14, 2020

Left Ventricular Thrombus and Cardiovascular Events APRIL 14, 2020:1676–85

F I G U R E 1 Study Flow Chart

Case-sensitive computerized echocardiography search

January 2011 to January 2018
90,065 reports

Excluded:
- Atrial thrombus
- Right ventricular thrombus

Left ventricular thrombus on echocardiography report

n = 174 patients

Thrombus reviewed by 2 independent experts

Excluded by experts by echocardiography (n = 1)

Excluded by a concomitant imaging (n = 14)

- by cardiac magnetic resonance imaging (n = 10)
- by contrast echocardiography (n = 2)
- by cardiac CT-scan (n = 2)

Confirmed left ventricular thrombus

n = 159 patients

The figure represents the patients screened and included in the final analysis. CT ¼ computed tomography.

low-molecular-weight heparin, unfractionated hepa- thrombus dimension, a stable thrombus, or a partial

rin) of anticoagulant used, duration of anti- thrombus regression at the last available follow-up.
coagulation therapy, and the association with We collected the following clinical outcomes dur-
antiplatelet therapy. Treatment duration was ob- ing the period of observation: major adverse cardio-
tained through medical reviews of each hospital visit vascular events (MACE) defined as the composite of
and by attending physician or patient interviews. all-cause death, ischemic stroke or transient
STUDY OBJECTIVES. The objectives of this study ischemic attack, MI, or acute peripheral artery
were to evaluate LVT regression under anti- emboli; any embolic complications defined as the
coagulation therapy, the time needed for total LVT composite of ischemic stroke or transient ischemic
regression using sequential noninvasive imaging, and attack, acute coronary emboli, or acute peripheral
the independent correlates associated with total LVT artery emboli (limb, renal, or digestive arteries); all-
regression. The impact of LVT regression on throm- cause death; and bleeding events defined as clini-
boembolism, bleeding, and mortality and factors cally relevant bleeding events (Bleeding Academic
associated with major cardiovascular adverse events Research Consortium [BARC] $2) and major bleeding
were determined. events (BARC$3) (16).

ENDPOINT DEFINITIONS. Total LVT regression was STATISTICAL ANALYSIS. All data were shown as
defined by a complete disappearance of LVT on all mean  SD or median (interquartile range) for
echocardiography views at the last available follow- continuous variables and as number (%) of patients
up. LVT persistence was classified as an increased for categorical variables. Non-Gaussian variables
JACC VOL. 75, NO. 14, 2020 Lattuca et al. 1679
APRIL 14, 2020:1676–85 Left Ventricular Thrombus and Cardiovascular Events

alone), baseline thrombus area (defined as a contin-

T A B L E 1 Baseline Clinical Characteristics
uous variable), history of atrial fibrillation, atrial
Patients with LVT 159
fibrillation complication (occurring after thrombus
Age, yrs 58  13
diagnosis), and major bleeding (BARC $3). A second
Female 28 (17.6)
Body mass index, kg/m2 25.0 (22.6–27.4) multivariate Cox model, including the aforemen-
Active smoking 73 (45.9) tioned variables with antithrombotic treatment
Hypertension 57 (35.8) duration and total LVT regression as additional vari-
Family history of cardiovascular disease 19 (11.9) ables, was performed to identify independent corre-
Diabetes mellitus 38 (23.9) lates of MACE, all-cause death, and all embolic
Dyslipidemia 65 (40.9)
complications in this specific population of LVT pa-
Creatinine clearance <60 ml/min 33 (20.8)
tients. All tests had a 2-sided significance level of 5%
Prior coronary stenting 93 (58.5)
Prior coronary artery bypass graft 8 (5.0)
and were performed with the use of SAS software,
Peripheral artery disease 18 (11.3) version 9.2 (SAS Institute, Cary, North Carolina).
Prior stroke or transient ischemic attack 20 (12.6)
Prior atrial fibrillation 18 (11.3) RESULTS
Prior major bleeding 8 (5.0)
Previous anticoagulant therapy 20 (12.6) CHARACTERISTICS OF THE PATIENTS AND
Vitamin K antagonist 14 (8.8) THROMBUS DESCRIPTION. Of the 90,065 echocar-
Direct oral anticoagulants 6 (3.8) diograms performed during the study period, a total
Underlying disease
of 174 patients had a reported LVT and 159 patients
Coronary artery disease 125 (78.6)
were definitively included in this analysis after
NSTEMI 18 (11.3)
STEMI 56 (35.2)
imaging review by 2 independent experts (Figure 1).
Successful revascularization 45 (80.3) The baseline characteristics of the patients with
STEMI >24 h 12 (21.4) confirmed LVT are described in Table 1. The popu-
Dilated cardiomyopathy 23 (14.5) lation was relatively young (mean age 58  13 years)
Hypertrophic cardiomyopathy 2 (1.3) with a high prevalence of cardiovascular risk factors
Myocarditis 6 (3.8)
leading to ischemic cardiomyopathy (78.6%,
Takotsubo cardiomyopathy 2 (1.3)
n ¼ 125) including STEMI in 1 of 3 patients (35.2%;
ARVD with associated LV impairment 1 (0.6)
n ¼ 56). Severe left ventricular dysfunction was
Values are n, mean  SD, n (%), or median (interquartile range). common (mean LVEF 31.9  12.5%) with more than
ARVD ¼ arrhythmogenic right ventricular dysplasia; LVT ¼ left ventricular one-half of the population, which had a mean car-
thrombus; NSTEMI ¼ non–ST-segment elevation myocardial infarction;
STEMI ¼ ST-segment elevation myocardial infarction. diac output index of <2.6 l/min. The most frequent
left ventricular wall motion abnormality was aki-
nesis of the apical segments where LVT were pre-
were compared with the use of the Mann-Whitney dominantly located (98.1%). Left ventricular
test. Chi-square testing was used for frequency com- aneurysms were found in 15.1% (n ¼ 24) of patients.
parisons. The 95% confidence interval (CI) for the LVT were classified as mobile in 34.6% (n ¼ 55) of
hazard ratio (HR) is presented. The population was patients and calcified LVT were observed in 1.3%
evaluated according to total LVT regression or (n ¼ 2) of patients. The largest diameter, surface
incomplete regression during follow-up. Time to first area, and thrombus volume as well as the detailed
clinical endpoints was estimated using the Kaplan- baseline echocardiographic measurements are dis-
Meier method, and differences between total LVT played in Table 2.
regression and persistent LVT were assessed with the ANTITHROMBOTIC THERAPY. Only 2 patients (1.2%)
log-rank test. Univariate and multivariate Cox were treated with antiplatelet therapy alone due to
regression analyses were used to identify indepen- high bleeding risk. The vast majority of the study
dent correlates for total LVT regression. The population (98.8%) was exposed to anticoagulation
following variables related to clinical outcomes in therapy including VKA (48.4%; n ¼ 77) mostly
univariate analysis (p < 0.10) were included in the fluindione (46.5%; n ¼ 74), DOACs (22.6%; n ¼ 36),
first multivariate model: age, underlying disease low molecular weight heparin (23.3%; n ¼ 37), or
(ischemic vs. nonischemic cardiomyopathy), LVEF, unfractionated heparin (4.4%; n ¼ 7). Concomitant
anticoagulation by DOACs or heparin (considering antiplatelet therapy was prescribed in 67.9% (n ¼ 108)
VKA treatment as reference), combined antith- of patients (Table 3). Median duration of anti-
rombotic treatment by anticoagulant and antiplatelet coagulation therapy was 508 days (interquartile
therapy (in comparison to anticoagulation therapy range: 115 to 986 days) without significant difference
1680 Lattuca et al. JACC VOL. 75, NO. 14, 2020

Left Ventricular Thrombus and Cardiovascular Events APRIL 14, 2020:1676–85

T A B L E 2 Baseline Echocardiographic Evaluation T A B L E 3 Antithrombotic Strategy Following Left Ventricular

Patients with LVT 159 Thrombus Diagnosis

Left ventricular ejection fraction, % 31.9  12.5 Patients with LVT 159
Cardiac output, l/min 4.1 (3.0–4.9) Antiplatelet therapy only 2 (1.2)
Left ventricular wall motion Anticoagulation only 45 (28.3)
Akinesis in apical segments 130 (81.8) Anticoagulation þ antiplatelet therapy
Hypokinesis in apical segments 6 (3.8) Aspirin with anticoagulant 52 (32.7)
Global hypokinesis 23 (14.5) Clopidogrel with anticoagulant 3 (1.9)
Left ventricular thrombus Ticagrelor with anticoagulant 1 (0.6)
Largest diameter, mm 19 (13–24) Anticoagulant þ dual antiplatelet therapy 56 (35.2)
Area, cm2 1.34 (0.8–2.57) Anticoagulation type
Volume, cm3 1.06 (0.49–2.40) Vitamin K antagonist 77 (48.4)
Mobile thrombus 55 (34.6) Coumadin 3 (1.9)
Apical thrombus 156 (98.1) Fluindione 74 (46.5)
Left ventricular aneurysm 24 (15.1) Direct oral anticoagulant 36 (22.6)
Other mechanical complication 1 (0.6) Apixaban 2.5 mg  2 9 (5.7)
Apixaban 5 mg  2 9 (5.7)
Values are n, mean  SD, median (interquartile range), or n (%). Dabigatran 110 mg  2 5 (3.1)
LVT ¼ left ventricular thrombus.
Dabigatran 150 mg  2 0 (0.0)
Rivaroxaban 15 mg 10 (6.3)
Rivaroxaban 20 mg 3 (1.9)
regarding the anticoagulant drug administered Low–molecular-weight heparin 37 (23.3)
Unfractionated heparin 7 (4.4)
(Supplemental Table 1).
THROMBUS REGRESSION. A reduction of the Values are n or n (%).

thrombus area from the baseline echocardiography to LVT ¼ left ventricular thrombus.

the final echocardiography (median delay of 338 days

(interquartile range: 38 to 907 days) was observed in
76.1% (n ¼ 121) of patients. The time-dependent cu- Reduced mortality was observed in patients with
mulative total regression of LVT is represented in total LVT regression (15.2% vs. 25.0%; HR: 0.48;
Figure 2 showing a progressive decrease of thrombus 95% CI: 0.23 to 0.98; p ¼ 0.039) whereas an
area. Total LVT regression was achieved in 62.3% increased major bleeding risk was observed among
(n ¼ 99) of patients within a median time of 103 days patients with persistent LVT (9.1% vs. 12%; HR:
(interquartile range: 32 to 392 days). 0.34; 95% CI: 0.14 to 0.82; p ¼ 0.011) (Table 4,
After multivariate adjustment, the presence of a Figure 3). MACE occurred more frequently in pa-
nonischemic cardiomyopathy (HR: 2.74; 95% CI: 1.43 tients treated with a dual antithrombotic strategy
to 5.26; p ¼ 0.002) and a smaller baseline thrombus (54%; n ¼ 30 for the association of single anti-
area (HR: 0.66; 95% CI: 0.45 to 0.96; p ¼ 0.031) were platelet therapy and anticoagulant) than those
independently associated with total LVT regression. A treated with a triple antithrombotic strategy (29%;
thrombus recurrence or an increase of thrombus area n ¼ 16 for the association of dual antiplatelet ther-
was observed in 14.5% (n ¼ 23) of patients. A review of apy and anticoagulant) with similar rate of major
medical records reported that recurrent or increased bleeding: 14%, n ¼ 8 and 12.5%, n ¼ 7, respectively.
size of LVT was associated with poor treatment After multivariate adjustment, an LVEF $35% (HR:
adherence; prothrombotic conditions such as active 0.46; 95% CI: 0.23 to 0.93; p ¼ 0.029) and a pro-
cancer or inflammatory or hematological diseases; or longed anticoagulation therapy over 3 months (HR:
chronic renal failure (Supplemental Table 2). 0.42; 95% CI: 0.20 to 0.88; p ¼ 0.021) were inde-
pendently associated with a reduced risk of MACE
CARDIOVASCULAR EVENTS. MACE and embolic (Figure 4). Moreover, a numerically lower rate of
complications occurred in 37.1% (n ¼ 59) and 22.2% embolic complications was observed in patients
(n ¼ 35) of patients, respectively, within a median treated by a prolonged anticoagulation therapy over
follow-up period of 632 days (interquartile range: 3 months (HR: 0.46; 95% CI: 0.18 to 1.14; p ¼ 0.093)
187 to 1,126 days). All-cause mortality was 18.9% (data not shown). It has to be noted that the type of
(n ¼ 30). Clinically relevant (BARC $2) and major anticoagulant therapy was not independently asso-
(BARC $3) bleeding events occurred in 17% (n ¼ 27) ciated with cardiovascular events in our study (data
and 13.2% (n ¼ 21) of patients, respectively. not shown).
JACC VOL. 75, NO. 14, 2020 Lattuca et al. 1681
APRIL 14, 2020:1676–85 Left Ventricular Thrombus and Cardiovascular Events

F I G U R E 2 Time-Dependent Cumulative Total Left Ventricular Thrombus Regression

10

8
Thrombus Area (cm2)

0
Echocardiographic Index 2 3 4 5
Images
Median time between
echocardiograms 26 days 52 days 69 days 311 days

Cumulative total 43.4% 50.5% 61.0% 62.3%

thrombus regression (n = 69) (n = 93) (n = 97) (n = 99)

Each point represents thrombus area of each patient at the different echocardiographic evaluation times. The red error bars represent the
median interquartile ranges.

DISCUSSION associated with poor outcomes. This analysis pro-

vides new insights from a large and well-
Although the incidence of LVT has decreased in characterized contemporary cohort of patients with
modern times, it continues to complicate both LVT (Central Illustration). For patients with LVT, total
ischemic and nonischemic cardiomyopathies and is regression was only achieved in two-thirds of

T A B L E 4 Cardiovascular Events in the Whole Population and According to the Occurrence of a Total Thrombus Regression

All Population Total Thrombus Regression Persistent Thrombus Log-Rank

(n ¼ 159) (n ¼ 99) (n ¼ 60) Test p Value

Major adverse cardiovascular events: composite of 59 (37.1) 35 (35.4) 24 (40.0) 0.203

all-cause death, ischemic stroke/TIA, MI, or
acute peripheral artery emboli
Composite of cardiovascular death, ischemic stroke/ 46 (28.9) 29 (29.3) 17 (28.3) 0.524
TIA, MI, or acute peripheral artery emboli
Cardiovascular death 14 (8.8) 8 (8.1) 6 (10.0) 0.434
All-cause death 30 (18.9) 15 (15.2) 15 (25.0) 0.039
Stroke/TIA 21 (13.3) 14 (14.1) 7 (11.9) 0.871
Acute peripheral artery emboli 20 (12.7) 14 (14.1) 6 (10.2) 0.962
All embolic complications 35 (22.2) 22 (22.2) 13 (22.0) 0.474
BARC $2 bleeding 27 (17.0) 12 (12.1) 15 (25.0) 0.002
BARC $3 bleeding 21 (13.2) 9 (9.1) 12 (20.0) 0.011

Values are n (%).

BARC ¼ Bleeding Academic Research Consortium; MI ¼ myocardial infarction; TIA ¼ transient ischemic attack.
1682 Lattuca et al. JACC VOL. 75, NO. 14, 2020

Left Ventricular Thrombus and Cardiovascular Events APRIL 14, 2020:1676–85

F I G U R E 3 Time-to-Event Curves According to Left Ventricular Thrombus Evolution on Treatment

Major Adverse
A Cardiovascular Events
B All-Cause Death

1.00 1.00
Event-Free Proportion (%)

0.75 0.75

Survival Rate (%)

0.50 0.50

0.25 0.25

0.00 0.00
HR: 0.71; 95% CI: 0.42-1.21; p = 0.203 HR: 0.48; 95% CI: 0.23-0.98; p = 0.039

0 500 1,000 1,500 0 500 1,000 1,500

Days Since LVT Diagnosis Days Since LVT Diagnosis
No 60 27 13 4 No 60 29 15 7
Yes 96 55 27 11 Yes 99 68 34 15

C All Embolic Complications D Major Bleeding Events

1.00 1.00
Event-Free Proportion (%)

Event-Free Proportion (%)

0.75 0.75

0.50 0.50

0.25 0.25

0.00 0.00
HR: 0.77; 95% CI: 0.38-1.57; p = 0.474 HR: 0.34; 95% CI: 0.14-0.82; p = 0.011

0 500 1,000 1,500 0 500 1,000 1,500

Days Since LVT Diagnosis Days Since LVT Diagnosis
No 59 26 13 5 No 60 26 13 6
Yes 96 55 27 11 Yes 98 60 31 14
Total Thrombus Regression Persistent Thrombus

(A) Major adverse cardiovascular events, (B) all-cause death, (C) embolic complications, and (D) major bleeding events according to left
ventricular thrombus (LVT) evolution on treatment. The blue curve represents patients who reached total thrombus regression on treatment
during follow-up and the red curve represents patients with thrombus persistence at the end of follow-up. CI ¼ confidence interval;
HR ¼ hazard ratio.

our population within 3 months for one-half of them nonischemic cardiomyopathy and a smaller
using full anticoagulation. The use of antiplatelet baseline thrombus area. LVT patients had a very
agents was common. The independent correlates high risk of MACE, embolic or major bleeding
of total LVT regression were an underlying complications, and mortality. Most importantly,
JACC VOL. 75, NO. 14, 2020 Lattuca et al. 1683
APRIL 14, 2020:1676–85 Left Ventricular Thrombus and Cardiovascular Events

F I G U R E 4 Independent Correlates of Major Adverse Cardiovascular Events

Adjusted HR
[95% CI] p value

Baseline LVEF ≥35% 0.46 [0.23-0.93] 0.029

Baseline thrombus area* 1.11 [0.94-1.31] 0.206

Anticoagulation by heparin 2.11 [0.97-4.60] 0.059

Anticoagulation therapy >3 months 0.42 [0.20-0.88] 0.021

Atrial fibrillation† 1.38 [0.58-3.29] 0.472

0.1 1 10

More Frequent MACE

*The baseline thrombus area is the thrombus area measured on the first echocardiography and the variable is presented as a continuous growing variable (unit ¼ 1 cm2).
†Atrial fibrillation was defined by the presence of permanent or paroxysmal atrial fibrillation. LVEF ¼ left ventricular ejection fraction; MACE ¼ major adverse
cardiovascular events; other abbreviations as in Figure 3.

total LVT regression was associated with a LEFT VENTRICULAR THROMBUS AND CLINICAL
reduced mortality, which was obtained with a full OUTCOMES. The main result of the present study is
anticoagulation therapy using either heparin, VKA, the high rate of cardiovascular events, as 4 of 10 pa-
or DOACs. tients had a MACE and 1 of 5 patients died during the

THROMBUS REGRESSION AND SUBOPTIMAL

follow-up period, a dramatically higher case mortality

ANTICOAGULATION THERAPY. This study highlights

than reported in STEMI patients without LVT (17,18).

that the current antithrombotic regimen needs to be These findings are again difficult to compare with the

improved, because one-third of patients did not few prior observational studies published so far

achieve total LVT regression and remained exposed (5,14,15,19,20) but are consistent with 2 recent studies

to a high risk of clinical complications even when (9,11) and highlight the severity of this population

combining with antiplatelet agents. Consistent with and the need for more efficient therapeutic strategy

our findings, a recent study screened more than to reverse such poor prognosis. Several approaches to

140,000 echocardiograms and found that LVT was a improve the prognosis of LVT patients may be

rare complication, as 128 LVT patients were identi- considered. One would be to refine the antith-

fied. Total LVT regression was reached in 71% of pa- rombotic regimen to improve and accelerate LVT

tients treated by anticoagulation therapy, but, regression, a factor associated with lower mortality in

unfortunately, echocardiographic follow-up and data our study. It could be argued that LVT regression

concerning LVT regression were not available for all could at least partially be the consequence of

of the patients (11). In a single cohort of 92 post-MI thrombus embolization. However, although we

patients treated with VKA, the authors found that cannot exclude asymptomatic embolization, we did

LVT regression was dependent on time in therapeutic not find a significant difference in embolic compli-

range. Unlike our study, there was no experience cations between patients with or without LVT

using DOACs (9). regression. These embolic complications could occur

1684 Lattuca et al. JACC VOL. 75, NO. 14, 2020

Left Ventricular Thrombus and Cardiovascular Events APRIL 14, 2020:1676–85

C E NT R AL IL L U STR AT IO N Clinical Outcomes Associated With Left Ventricular Thrombus and Impact of Total
Thrombus Regression on Prognosis

From 2011 to 2017

90,065 echocardiograms

1.00
Confirmed
left ventricular thrombus
n = 159
0.75

Survival (%)
0.50
Median follow-up 1.7 years

0.25

Reduction of mortality 0.00 Adjusted HR: 0.48; 95% CI: 0.23-0.98;

Total thrombus regression
among patients with p = 0.039
obtained by anticoagulation
total thrombus regression
0 500 1,000 1,500
Only 62.3% of the patients Days
Median time: 103 days No 60 29 15 7
Yes 99 68 34 15
Similarly with vitamin K antagonist or Total Thrombus Regression
direct oral anticoagulant or heparin
Persistent Thrombus

Lattuca, B. et al. J Am Coll Cardiol. 2020;75(14):1676–85.

Major bleeding events were defined by Bleeding Academic Research Consortium types 3 or 5. CI ¼ confidence interval; HR ¼ hazard ratio.

at any time under antithrombotic therapy, as a complications that are already reported in more than
consequence of thrombus fragmentation, quick 10% of the patients in our study. Although not
thrombus regression, or on the contrary, persistence designed and powered for a direct comparison be-
of the thrombus. The achievement of total LVT tween DOACs versus VKA, our study suggests a similar
regression seems to be a marker of morbidity and rate of total LVT regression, MACE, and bleeding
mortality of the patient more than the associated complications with both anticoagulant strategies.
embolic risk. It should not be a predominant factor to
shorten the antithrombotic treatment. Moreover, STUDY LIMITATIONS. This was a retrospective study
after adjustment for confounding factors, atrial from a single center, albeit the largest volume center
fibrillation was not associated with a higher rate of in the Paris area. We also acknowledge that the
clinical outcomes and our data support that thrombus diagnosis of LVT was mainly based on echocardiog-
regression is mostly due to the duration of antith- raphy, which may have a lower sensitivity and spec-
rombotic therapy and the size of the LVT. Beyond the ificity for detection compared with other imaging
efficacy of each anticoagulant drug, the optimal modalities, such as cardiac magnetic resonance im-
treatment duration is a major clinical issue. In our aging (4,19,21). We also acknowledge that the com-
study, we observed a lower occurrence of MACE parison between the different antithrombotic drugs
among patients who were treated by an anti- was limited by the small sample size and that our
coagulation therapy for a longer duration than 3 results should be considered exploratory rather than
months. We can hypothesize that a longer antith- definitive. An individualized risk stratification based
rombotic treatment could lead to a better prevention on the patient characteristics, the underlying disease,
of embolic complications or MI and prevent LVT and the LVT evolution under antithrombotic treat-
recurrence. Such findings need to be confirmed by ment would be ideal to guide physician decision-
adequately sized randomized trials. However, any making and to refine the management of LVT with
intensification of the antithrombotic treatment may the main objective of reducing the risk of clin-
be compromised by more frequent bleeding ical outcomes.
JACC VOL. 75, NO. 14, 2020 Lattuca et al. 1685
APRIL 14, 2020:1676–85 Left Ventricular Thrombus and Cardiovascular Events

CONCLUSIONS PERSPECTIVES

We conclude that the clinical prognosis of patients

COMPETENCY IN PATIENT CARE AND PROCEDURAL
with LVT is poor with a very high risk of major car-
SKILLS: Left ventricular thrombus is an uncommon
diovascular events and mortality. Refinement in the
complication of ischemic and nonischemic cardiomyopathies
antithrombotic management is needed to improve
associated with a high risk of adverse events and mortality.
clinical outcomes of these relatively young patients.
Regression of thrombus is associated with lower mortality,
Total LVT regression can be obtained with different
but in about one-third of cases, thrombus does not
anticoagulation regimens including DOACs and
resolve despite anticoagulation with or without antiplatelet
thereby reduce mortality.
therapy.
ADDRESS FOR CORRESPONDENCE: Dr. Gilles
TRANSLATIONAL OUTLOOK: Randomized studies should
Montalescot, ACTION Study Group, Institut de
assess the safety and efficacy of target-specific oral anticoagu-
Cardiologie, Centre Hospitalier Universitaire Pitié-
lants, clarify the optimum duration of treatment, and identify
Salpêtrière, 47 Boulevard de l’Hôpital, 75013 Paris,
predictors of recurrence after thrombus resolution.
France. E-mail: gilles.montalescot@aphp.fr. Twitter:
@ActionCoeur, @docjohanne.

REFERENCES

1. Habash F, Vallurupalli S. Challenges in man- patients with left ventricular thrombus after first Academic Research Consortium. Circulation 2011;
agement of left ventricular thrombus. Ther Adv acute myocardial infarction. Eur Heart J 2018;39: 123:2736–47.
Cardiovasc Dis 2017;11:203–13. 201–8.
17. Jernberg T, Johanson P, Held C, et al. Associ-
2. Gianstefani S, Douiri A, Delithanasis I, et al. Inci- 10. Johannessen KA, Nordrehaug JE, von der ation between adoption of evidence-based treat-
dence and predictors of early left ventricular Lippe G, Vollset SE. Risk factors for embolisation ment and survival for patients with ST-elevation
thrombus after ST-elevation myocardial infarction in in patients with left ventricular thrombi and acute myocardial infarction. JAMA 2011;305:1677–84.
the contemporary era of primary percutaneous cor- myocardial infarction. Br Heart J 1988;60:104–10. 18. Rosamond WD, Chambless LE, Heiss G, et al.
onary intervention. Am J Cardiol 2014;113:1111–6.
11. McCarthy CP, Murphy S, Venkateswaran RV, Twenty-two-year trends in incidence of myocar-
3. Mao TF, Bajwa A, Muskula P, et al. Incidence of et al. Left ventricular thrombus: contemporary dial infarction, coronary heart disease mortality,
left ventricular thrombus in patients with acute st- etiologies, treatment strategies, and outcomes. and case fatality in 4 US communities, 1987-2008.
segment elevation myocardial infarction treated J Am Coll Cardiol 2019;73:2007–9. Circulation 2012;125:1848–57.
with percutaneous coronary intervention. Am J
12. O’Gara PT, Kushner FG, Ascheim DD, et al. 19. Weinsaft JW, Kim J, Medicherla CB, et al.
Cardiol 2018;121:27–31.
2013 ACCF/AHA guideline for the management of Echocardiographic algorithm for post-myocardial
4. McCarthy CP, Vaduganathan M, McCarthy KJ, ST-elevation myocardial infarction: a report of the infarction LV thrombus: a gatekeeper for
Januzzi JL, Bhatt DL, McEvoy JW. Left ventricular American College of Cardiology Foundation/ thrombus evaluation by delayed enhancement
thrombus after acute myocardial infarction: American Heart Association Task Force on Practice CMR. J Am Coll Cardiol Img 2016;9:505–15.
screening, prevention, and treatment. JAMA Car- Guidelines. J Am Coll Cardiol 2013;61:e78–140.
20. Visser CA, Kan G, Meltzer RS, Dunning AJ,
diol 2018;3:642.
13. Ibanez B, James S, Agewall S, et al. 2017 ESC Roelandt J. Embolic potential of left ventricular
5. Delewi R, Zijlstra F, Piek JJ. Left ventricular guidelines for the management of acute myocar- thrombus after myocardial infarction: a two-
thrombus formation after acute myocardial dial infarction in patients presenting with dimensional echocardiographic study of 119 pa-
infarction. Heart 2012;98:1743–9. ST-segment elevation: The Task Force for the tients. J Am Coll Cardiol 1985;5:1276–80.

6. Robinson AA, Jain A, Gentry M, McNamara RL. management of acute myocardial infarction in 21. Srichai MB, Junor C, Rodriguez LL, et al. Clinical,
Left ventricular thrombi after STEMI in the primary patients presenting with ST-segment elevation of imaging, and pathological characteristics of left
PCI era: A systematic review and meta-analysis. Int the European Society of Cardiology (ESC). Eur ventricular thrombus: a comparison of contrast-
J Cardiol 2016;221:554–9. Heart J 2018;39:119–77. enhanced magnetic resonance imaging, trans-
14. Vaitkus PT, Barnathan ES. Embolic potential, thoracic echocardiography, and transesophageal
7. Gottdiener JS, Gay JA, VanVoorhees L,
prevention and management of mural thrombus echocardiography with surgical or pathological
DiBianco R, Fletcher RD. Frequency and embolic
complicating anterior myocardial infarction: a validation. Am Heart J 2006;152:75–84.
potential of left ventricular thrombus in dilated
cardiomyopathy: assessment by 2-dimensional meta-analysis. J Am Coll Cardiol 1993;22:1004–9.
echocardiography. Am J Cardiol 1983;52:1281–5. 15. Weinreich DJ, Burke JF, Pauletto FJ. Left KEY WORDS antithrombotic, direct oral
8. Ezekowitz MD, Wilson DA, Smith EO, et al. ventricular mural thrombi complicating acute anticoagulant, left ventricular thrombus,
Comparison of indium-111 platelet scintigraphy myocardial infarction. Long-term follow-up with mortality, myocardial infarction, thrombus
and two-dimensional echocardiography in the serial echocardiography. Ann Intern Med 1984; regression
diagnosis of left ventricular thrombi. N Engl J Med 100:789–94.
1982;306:1509–13.
16. Mehran R, Rao SV, Bhatt DL, et al. Standard- A PP END IX For a supplemental figure and
9. Maniwa N, Fujino M, Nakai M, et al. Anti- ized bleeding definitions for cardiovascular clinical tables, please see the online version of this
coagulation combined with antiplatelet therapy in trials: a consensus report from the Bleeding paper.