ACUTE CORONARY SYNDROMES Left ventricular thrombus ... · ACUTE CORONARY SYNDROMES Left ventricular thrombus formation after acute myocardial infarction Ronak Delewi,1 Felix Zijlstra,2

ACUTE CORONARY SYNDROMES

Left ventricular thrombus formationafter acute myocardial infarctionRonak Delewi,1 Felix Zijlstra,2 Jan J Piek1

Cardiovascular disease remains the leading cause ofdeath in western society. Mortality from acutemyocardial infarction (AMI) has decreased since theintroduction of primary percutaneous coronaryintervention (PCI), which has proved to be superiorto thrombolytic therapy by demonstrating lowermortality rates and reduced clinical adverse events.Nevertheless, postinfarct complications still lead tomorbidity and mortality in a large number ofpatients.One of the most feared complications is the

occurrence of thromboembolic events (mostlycerebrovascular accidents) due to left ventricular(LV) thrombus formation. The risk of LV thrombusformation is highest during the first 3 monthsfollowing acute myocardial infarction, but thepotential for cerebral emboli persists in the largepopulation of patients with chronic LV dysfunc-tion. Since these thromboembolic events areusually unheralded by warning signs of transientcerebral ischaemia, the only truly satisfactorymedical approach is adequate management of thesehigh risk groups. This article discusses the inci-dence, diagnosis and management of LV thrombusformation after an AMI.

PATHOGENESIS OF LV THROMBUSThe combination of blood stasis, endothelial injuryand hypercoagulability, often referred to asVirchow’s triad, is a prerequisite for in vivothrombus formation. In the presence of LVthrombus formation after AMI, the three compo-nents of this triad can also be recognised (figure 1).LV regional wall akinesia and dyskinesia result inblood stasis, often recognised on two dimensionalechocardiography by the occurrence of spontaneousLV contrast. Prolonged ischaemia leads to suben-docardial tissue injury with inflammatory changes.Finally, patients with an acute coronary syndromedisplay a hypercoagulable state with, for example,increased concentrations of prothrombin, fibrino-peptide A, and von Willebrand factor, and decreasedconcentrations of the enzyme responsible forcleaving von Willebrand factor (ADAMTS13).w1 w2

This triad can result in the formation of LVthrombus composed of fibrin, red blood cells, andplatelets.LV thrombus can occur within 24 h after AMI.

One study performing serial echocardiographicstudies showed that about 90% of thrombi areformed at a maximum of 2 weeks after the indexevent.w3 However, some patients develop a new LV

thrombus after discharge, often in association withworsening LV systolic function. Spontaneous oranticoagulant induced resolution is relativelycommon in LV thrombus formation after AMI.Thrombus seems to disappear more often inpatients with apical akinesia than those with apicalaneurysm or dyskinesia.1

It has been speculated that LV thrombus playsa positive role in the acutely infarcted myocardium,by offering mechanical support to the infarctedmyocardium and therefore protecting against LVrupture.w4 The thrombus becomes firmly attachedto its site of origin, enhancing the underlyingmyocardial scar, limiting potential infarct expan-sion, and partially restoring the thickness of themyocardial wall. As a consequence, bulging isreduced, resulting in a more effective myocardialcontraction. Often, however, expansion of theinfarct zone occurs very early after infarction,before the thrombus has time to organise and isable to prevent formation of LV aneurysm andmyocardial rupture.

INCIDENCEEarly data from the prethrombolytic and throm-bolytic eras suggest that in the setting of AMI, LVthrombus was present in 7e46% of patients, mostfrequently in acute anterior or apical myocardialinfarction.2e4 w3ew5 Differences in diagnostictechniques, timing of examination and use ofantithrombotic treatment cause substantial varia-tion in the reported frequency of thrombus fromdifferent series. In addition, it should be noted thatthe incidence as reported in autopsy studies isconsistently higher compared with clinical studies,probably due to better accuracy but also due topatient selection.Nowadays the reported incidence is lower. This is

probably due to (1) more aggressive anticoagulationtherapies in the acute phase (eg, the use of heparin,bivalirudin), (2) smaller infarctions, and (3)improved LV remodelling. Although the use of ACEinhibitors is also thought to be associated withimproved LV remodelling, the GISSI-3 study foundno difference in LV thrombus rates betweenpatients who did and did not receive lisinopril.5

There are limited data on the exact frequency ofLV thrombus in PCI treated AMI patients. Twostudies found LV thrombus formation in 5.4% and7.1% of patients with acute anterior wall myocar-dial infarctions.w6 w7 However, these studies wereretrospective, non-serial and only assessed LV

< Additional references arepublished online only. To viewthese references please visit thejournal online (http://dx.doi.org/10.1136/heartjnl-2012-301962).1Department of Cardiology,Academic Medical Center,University of Amsterdam,Amsterdam, the Netherlands2Department of Cardiology,Erasmus Medical Center,Rotterdam, the Netherlands

Correspondence toProfessor Dr Jan J Piek,Department of Cardiology,Academic Medical Center,University of Amsterdam, POBox 22660, 1100 DDAmsterdam, The Netherlands;[email protected]

Education in Heart

Heart 2012;98:1743–1749. doi:10.1136/heartjnl-2012-301962 1743

on 20 June 2019 by guest. Protected by copyright.

http://heart.bmj.com

/H

eart: first published as 10.1136/heartjnl-2012-301962 on 14 Novem

ber 2012. Dow

nloaded from

http://heart.bmj.com/

thrombus formation at a single point in time andduring the early phase of recovery after myocardialinfarction.In the latter study a follow-up echocardiography

was performed at 1e3 months, showing LVthrombus in an additional 8% of the patients.w7

Solheim et al reported a similar incidence of 15% inthe first 3 months in a selected group of AMIpatients treated by primary PCI.6 So, the timing ofLV thrombus assessment is crucial, as assessmenttoo soon after the onset of myocardial infarctionwill probably lead to failure to detect the thrombusin a significant percentage of patients.

CLINICAL FACTORS CONTRIBUTING TO LVTHROMBUS FORMATIONRisk factors for the development of LV thrombusare consistently irrespective of infarct treatmentand include large infarct size, severe apical asynergy(ie, akinesis or dyskinesis), LV aneurysm, andanterior MI.2 5e8 w6 This is consistent with anincreased contribution of at least two of the threecomponents of Virchow’s triad, namely a largerarea of blood stasis as well as an increased area ofinjured subendocardium.In a study of more than 8000 patients with ST

elevation myocardial infarction (STEMI), LVthrombus was found in 427 patients (5.1%). Thisincidence is relatively low compared to otherstudies, probably because of the exclusion of highrisk patients with severe LV dysfunction. Patientswith anterior AMI had a higher incidence of LVthrombus compared to patients with AMI at otherregions (11.5% vs 2.3%, p<0.0001). The incidenceof LV thrombus was also higher in patients with anejection fraction #40% (10.5% vs 4%, p<0.0001).In patients with an anterior AMI and an ejectionfraction #40% this percentage was as high as17.8%.5

Thrombus formation is not exclusively locatedapically; approximately 11% occurs at the septalwall and 3% at the inferoposterior wall.4 Theprevalence of thrombus in non-anterior myocardialinfarction increases when inferior necrosis extends

towards the posterolateral wall. In such cases theprevalence is similar to that observed in anteriorwall AMIs of comparable extension.w5 Thrombican also be found in small apical infarcts, with goodglobal systolic function.w3

The presence of thrombi is significantly relatedto the region of most severe functional impairmentand/or the region with myocardial enhancement(ie, infarction or scarring).7 LV thrombus appearsearlier in the course of the disease when initialejection fraction #40%, in the presence of multi-vessel coronary artery disease, or a high peak crea-tine kinase value.w8

There is conflicting evidence with respect to theinfluence of b-blockers. Several studies havereported a higher frequency of thrombus develop-ment in patients treated with b-blockers whichcould be related to the negative inotropic action ofthese drugs and thus increased blood stasis. Inparticular, in a randomised study, Johannessen et alreported an increased occurrence of thrombus inpatients with anterior AMI after oral b-blockertherapy.w9 Turpie et al reported similar results aftertreatment with b-blockers in a large population ofpatients with AMI.9 The GISSI-2 study, however,observed the same rate of LV thrombi in patientswith or without atenolol.10

It has been demonstrated that mitral regurgita-tion prevents thrombus formation in patientswith dilated cardiomyopathy.w10 The protectiveeffect of mitral regurgitation may be the conse-quence of augmented early diastolic flow velocitiesat the mitral annulus level, as well through theentire length of the left ventricle, protecting theLV cavity from a stagnant, thrombogenic bloodflow pattern. In addition, studies suggest abnormalflow profiles are associated with the presence of anLV thrombus.11 w11 However, to date no studieshave demonstrated the same association in patientswith AMI.There have been few studies on the use of

biomarkers in the setting of LV thrombus forma-tion. It could be postulated that factors involved inthe coagulation cascade could serve as biomarkers toidentify patients at increased risk for LV thrombusdevelopment. Data presented at the EuropeanSociety of Cardiology in 2011 demonstrated highersoluble tissue factor and d-dimer concentrations inpatients with LV thrombus formation.w12 Anotherstudy observed mildly elevated anticardiolipinantibody levels in patients with LV thrombusformation after AMI.w13 Whether these factors areindeed capable of predicting LV thrombus formationneeds to be evaluated.

DIAGNOSTIC MODALITIES TO DETECT LVTHROMBUSRadionuclide based techniquesIn 39 series using radionuclide ventriculography,a so-called ‘square left ventricle’ was reported to beassociated with LV thrombus.w14 The use ofindium-111 labelled platelets is much better docu-mented. It provides excellent specificity (95%) in

Figure 1 The three components of the Virchow’s triad inleft ventricular thrombus formation. ACS, acute coronarysyndrome; LV, left ventricular.

Education in Heart

1744 Heart 2012;98:1743–1749. doi:10.1136/heartjnl-2012-301962



/H


ber 2012. Dow

nloaded from


identifying LV thrombus, and its sensitivity wasreported to be 70% compared with transthoracicechocardiography (TTE).w15 It is not appliedwidely though because it is time consuming andexpensive, not universally available, and involvesradiation exposure. Furthermore, this scintigraphictechnique is ineffective in identifying relativelysmall thrombi, and it has good specificity andsensitivity only if there is active platelet aggrega-tion on the surface of the LV mural thrombus at thetime of imaging. In patients with an elevated lefthemidiaphragm, indium-111 activity in the spleenmay be confused with that from the LV apex.Finally, in patients with a large LVaneurysm but noLV thrombus, a large amount of relatively staticblood within the LV aneurysm may increaseindium-111 activity.w16

EchocardiographyTwo dimensional TTE is the technique used mostoften for assessing the presence, shape and size ofan LVmural thrombus. When the thoracic anatomyof the patient allows sufficient visualisation of theheart, two dimensional echocardiography providesexcellent specificity (85e90%) and sensitivity (95%)in detecting LV thrombus.12 w17 w18 LV thrombuson echocardiography is defined as a discrete echo-dense mass in the left ventricle with definedmargins that are distinct from the endocardium andseen throughout systole and diastole. It should belocated adjacent to an area of the LV wall which ishypokinetic or akinetic and seen from at least twoviews (usually apical and short axis). Care must betaken to exclude false tendons and trabeculae and torule out artefacts (reverberations, side lobe or nearfield artefacts), which constitute the most commoncause for a false diagnosis of a thrombus.13 14

Another source of false-positive studies result fromtangentially-cut LV wall. Varying gain settings anddepth of field, as well as using transducers withdifferent carrier frequencies in multiple positionsand orientations, are helpful to minimise suchfalse-positive studies.w17

In addition, often the LV apex cannot be clearlydefined and the presence or absence of a thrombusmay be very difficult to establish, leading to anestimated 10e46% of echocardiograms that areinconclusive.w20 w21 Intravenous echo contrastduring TTE may improve the diagnostic assess-ment of LV thrombus.12 w22 However, in Europethe use of most compounds is contraindicated bythe European Medicines Agency in cardiac patientswith acute coronary syndromes, recent PCI, acuteor chronic severe heart failure or severe cardiacarrhythmias. Also non-protruding and small muralLV thrombi may still go undetected.14

Transoesophageal echocardiography (TOE) haslittle to offer in the detection of LV thrombus.Although it is the technique of choice for detectingatrial masses and thrombi in the left atrialappendage, its value for diagnosing LV thrombus islimited because the apex is most often not wellvisualised.12 w23 Nevertheless, some data suggest

that TOE is superior to TTE in providing optimalvisualisation of small LV apical thrombi.w24

Computed tomographyCTscanning provides about the same specificity andsensitivity as two dimensional TTE in the identifi-cation of LV thrombus.w25 This technique is notused in daily practice since it requires the intra-venous injection of radiographic contrast materialand exposes the patient to ionising radiation.

Magnetic resonance imagingCardiac magnetic resonance imaging (CMR) withcontrast (delayed enhancement (DE)) has signifi-cantly better accuracy than TTE and TOE for thediagnosis of LV thrombus7 12 w26 w27 (table 1 andfigure 2). A study by Srichai et al compared CMRand late gadolinium enhancement with echocardio-graphy in a cohort of patients undergoing LVreconstruction surgery in whom surgical and/orpostmortem verification of thrombus wasperformed.12 This study reported that the sensi-tivity of TTE was 40%, compared with 88% forCMR. Another study reported an echo sensitivityand specificity of 33% and 91%, respectively, ina heterogeneous population of patients with LVsystolic dysfunction.7 These studies reported lowersensitivity for detection of LV thrombus thanpreviously described, probably due to exclusion ofsuboptimal echocardiographic examinations in thepreviously mentioned studies. Also, echocardio-graphic examinations were often reinterpreted withemphasis on LV thrombus detection and led todifferent findings when the presence or absence ofLV thrombus was based on routine clinical echo-cardiographic reading as part of the patient’s eval-uation.DE-CMR allows for a relatively rapid assessment

of thrombus presence, size, and location and isnowadays considered the gold standard. Theintravenous administration of gadolinium chelatesgreatly enhances the ability to detect and charac-terise LV thrombi. Immediately after contrastadministration, the homogeneous, strongenhancement of the LV cavity allows easy detec-tion of abnormal intraventricular structures (dark),which frequently occur adjacent to scarredmyocardium (bright hyperenhanced).Cine-CMR (without a contrast agent such as

gadolinium) seems to be less suitable for LV

Table 1 Sensitivities and specificities of differentdiagnostic modalities for the detection of left ventricularthrombus formation

Sensitivity Specificity

TOE 35% 90%

Routine clinical TTE 35e40% 90%

TTE (indication suspect LV thrombus) 60% 90%

CT Comparable with TTE

Cine CMR 60% 90%

DE-CMR 88% 99%

CMR, cardiac magnetic resonance imaging; CT, computed tomography;DE, delayed enhancement; LV, left ventricular; TOE, transoesophagealechocardiography; TTE, transthoracic echocardiography.

Education in Heart




/H


ber 2012. Dow

nloaded from


thrombus detection. Thrombus was missed in44e50% of the cases as detected by DE-CMR.7 8

The ability of DE-CMR to identify thrombus basedon tissue characteristics rather than anatomicalappearance alone may explain why it providesimproved thrombus imaging compared with cine-CMR. It should be mentioned that the criteria todifferentiate no-reflow zones from mural thrombiare not definite, and thus differentiation may notalways be straightforward. Also, further researchand histopathological correlation is needed toevaluate the role of DE-CMR in differentiatingsubacute from organised clots.

Embolic complicationsIn the prethrombolytic era, embolic complicationswere reported in approximately 10% ofcases,15 w28 w29 whereas in the thrombolytic era,embolic complications occurred in 2e3% ofpatients. There are poor data regarding emboliccomplications in LV thrombus patients treated byprimary PCI. Also, exact percentages regarding thesite of embolisation are not available.Several studies have suggested that LV thrombi

that protrude into the ventricular cavity or thatexhibit independent mobility are associated witha higher rate of embolisation than thrombi withoutthese features16 17 w30 (figure 3). A thrombus isconsidered as protruding when it projects predom-inantly into the LV cavity and as mural when itappears flat and parallel to the endocardial surface.Echocardiographic studies analysing mainly retro-spective and non-serial data have indicated a posi-tive relationship between the embolic potential ofLV thrombi and their protruding shape and/orintracavitary motion.15 w30 However, spontaneoustime-course variation in the morphologic aspects,such as shape and mobility pattern, are common.By performing serial echocardiography on 59untreated patients, Domenicucci et al found thatthese morphological features demonstratedpronounced spontaneous variability in the firstseveral months after acute infarction, and therefore

suggested that the assessment of these features wasnot helpful. They noted that 41% of 59 thrombihad significant changes in shape and 29% hadchanges in mobility.18 Also, it has been reportedthat up to 40% of embolism episodes occur inpatients whose thrombi are neither protuberant normobile.Other thrombus characteristics, such as

thrombus size,16 central echolucency17 or hyper-kinesia of the myocardial segments adjacent to thethrombus,4 were found in some studies to beassociated with an increased risk of embolism, butwere not confirmed by others.Other conditions that increase the risk of

systemic embolisation are: (1) severe congestiveheart failure, (2) diffuse LV dilatation and systolicdysfunction, (3) previous embolisation, (4) atrialfibrillation, and (5) advanced patient age. It hasbeen suggested that the risk of embolisation islower in patients with LV aneurysm, since theabsence of LV contraction near the site of thethrombus makes dislodgement unlikely.w31

PHARMACOLOGICAL MANAGEMENTIf indeed systemic embolisation is the highest riskof LV thrombus, the central question arises as tohow these patients should be treated to preventembolisation. In the past, if recurrent systemicemboli developed despite anticoagulant therapy,surgical removal of the thrombus was considerednecessary.17 w31 w32 Nowadays antithrombotictherapy is thought to prevent embolic complica-tions of LV thrombus.

ThrombolysisVaitkus and Barnathan pooled the data from sixstudies comprising a total of 390 patients andassessed the incidence of LV thrombus formation inthose patients treated with thrombolysis versusthose without thrombolytic therapy. They werenot able to demonstrate a statistical difference inthe incidence of LV thrombus formation, onlya trend in favour of thrombolysis.19 These studies

Figure 2 Left ventricular (LV) thrombus formation on delayed gadolinium contrast cardiac MRI and transthoracic echocardiography. Transthoracicechocardiographic appearance of a thrombus (asterisk) in the apex of the left ventricle (A); cine cardiovascular magnetic resonance of the same patientalso delineates the apical thrombus (B); late gadolinium enhancement imaging clearly confirms the avascular non-enhancing thrombus (asterisk, dark)close to the transmural infarcted myocardium (bright hyperenhanced, black arrowheads) with areas of microvascular obstruction (black, whitearrowheads) (C). Courtesy of Dr A C van Rossum, Dr R Nijveldt, Department of Cardiology, VU University Medical Center, Amsterdam, the Netherlands,and Dr B J Bouma, Department of Cardiology, Academic Medical Center, Amsterdam, the Netherlands.

Education in Heart




/H


ber 2012. Dow

nloaded from


were not randomised but often utilised patientsseen 3 h after symptom onset as a control group.Data from the GISSI-3 database, including morethan 8000 patients, showed no reduced incidence ofthrombus formation in patients who receivedeither thrombolytic therapy or heparin.5

Intravenous thrombolysis has also been used fortreatment of documented LV thrombus. In a reportof 16 patients with LV thrombus on echocardio-graphy, urokinase was infused intravenously ata rate of 60 000 U/h for 2e8 days in combinationwith intravenous heparin (200 units/kg312 h).LV thrombi were successfully lysed in 10 of 16patients. None of the patients suffered from clinicalembolism, and therapy had to be discontinued inonly one patient due to haematuria.w33 In a laterstudy, four patients with mobile LV thrombus weretreated with intravenous urokinase or strepto-kinase. In the first two cases, lysis of thrombus wasachieved without complication. In the latter twocases, however, systemic embolism occurred, withtransient diplopia in one and stroke followed bydeath in the other.1 It was concluded that fibrino-lytic agents are capable of lysing ventricularthrombi but that the risks of this therapy aretoo high.

HeparinData regarding the benefit of heparin treatment inpatients with documented LV thrombus on echo-cardiography during the first 2 weeks are somewhatconflicting, leading us to believe that there may bea benefit, at least in the short term. In a randomisedcontrolled trial, AMI survivors who were treatedwith high dose heparin (12 500 units subcutane-

ously every 12 h) showed a lower incidence ofLV thrombus formation than those administereda low dose (5000 units subcutaneously every 12 h)(11% vs 32%, p<0.001) during a 10 day period.9

Results from the SCATI study showed a similarreduction in LV thrombus formation for the groupthat was treated with calciumeheparin comparedto the control group in patients undergoingthrombolysis.w34 In the GISSI-2-connected study,however, high dose heparin did not preventthrombus formation (27% vs 30%, p¼NS).10 Ina study with 23 consecutive patients with mobileand protruding thrombi, high dose heparin wasgiven intravenously over a period of 14e22 days(mean 1464). In all 23 patients LV thrombidecreased in size, with disappearance of the highrisk features. No embolic events were detectedduring treatment, and the only complication wasan upper gastrointestinal haemorrhage.w35 Dalte-parin, a low molecular weight heparin, reduced theincidence of LV mural thrombus formation but hadno influence on the risk of systemic embolisation,and its use was associated with an increased risk ofhaemorrhage.w36

Vitamin K antagonistObservational studies conducted in the pre-thrombolytic and thrombolytic eras have providedsupport for the hypothesis that anticoagulationreduces the risk of embolisation.1 2 4 w3 w37ew39

A 1993 meta-analysis included 11 studies of 856patients who had an anterior myocardial infarction;the odds ratio (OR) for an embolic event was 5.5(95% CI 3.0 to 9.8).19 The meta-analysis includedseven studies with 270 patients that included dataon the relationship between anticoagulation for6 months and embolisation. Although all sevenstudies presented data suggesting that systemicanticoagulation reduces embolic complications, thistrend reached significance only in three trials. Whenpooling the data, anticoagulation compared withno anticoagulation was associated with a reductionin the rate of embolisation (OR 0.14, 95% CI 0.04to 0.52).Based on these data, both current European

Society of Cardiology and American College ofCardiology/American Heart Association guidelinesrecommend vitamin K antagonist therapy inpatients with an LV thrombus after myocardialinfarction.w40 w41

However, vitamin K antagonists do not appear toaffect the likelihood of resolution of the throm-busw3 and, unfortunately, no large randomised trialshave been performed to evaluate the efficacy of longterm anticoagulation to prevent embolisation inpatients with LV thrombus. Therefore the effects oflong term anticoagulants on the risk of embolisationare the subject of debate. Among the many ques-tions left unanswered is when to withdraw anti-coagulant medication when thrombus is identifiedsince the risk of embolisation decreases over time,likely as a result of organisation of thrombus whichincludes thrombus neovascularisation. However,retrospective studies documented ongoing embolic

Figure 3 Transthoracic echocardiographic appearance of a mobile, protruding leftventricular thrombus. Courtesy of J Vleugels and Rianne H A de Bruin, Department ofCardiology, Department of Cardiology, Academic Medical Center, Amsterdam, theNetherlands.

Education in Heart




/H


ber 2012. Dow

nloaded from


risk in LV thrombus patients.w42 In indium-111platelet imaging studies most thrombi, regardless ofage, have been observed to have externally detect-able ongoing platelet accumulation, indicatingcontinued surface activity.20 The European guide-lines recommend vitamin K antagonist for at least3e6 months, while the American guidelinesrecommend indefinite treatment in patientswithout increased risk of bleeding.Although there are limited data regarding the

appropriate follow-up and timing of cessation ofvitamin K antagonists in these patients, thefollowing approach seems appropriate for mostpatients:< Assess LV thrombus within the first month

after AMI, preferably with CMR in high riskpatients, and start vitamin K antagonist whenLV thrombus is present and no contraindicationexists

< Re-evaluate LV thrombus formation after6 months since data show that LV thrombusresolution in the initial months is very common,also in patients treated with vitamin K antag-onistsw43

< When LV thrombus is not present and there isno other indication for vitamin K antagonist,assess bleeding risk and consider stoppingtherapy.Newer anticoagulants are presently being

developed and some of them are alreadyregistered.w44ew46 It can be envisioned that in thelonger term these new anticoagulants will replacevitamin K antagonists. However, at presentvitamin K antagonist therapy is still the standard ofcare for the treatment of LV thrombus. Moreimportantly, the newer anticoagulants also havethe risk of fatal and non-fatal bleedings and theirrole in LV thrombus patients should be furtherassessed.

Antiplatelet therapy and triple therapy inthe PCI eraAnother issue is that nowadays STEMI patients aretreated by primary PCI and receive long term dualantiplatelet therapy (including aspirin and a P2Y12

inhibitor). Consequently, patients with LVthrombus or at increased risk of LV thrombus aftera myocardial infarction are frequently being treatedwith vitamin K antagonist in addition to dualantiplatelet therapy (triple antithrombotic therapy)and therefore are subjected to an increased bleedingrisk. It is unclear, however, if long term anti-coagulation is still necessary in STEMI patientstreated by primary PCI and subsequent dualantiplatelet therapy.Large prospective studies show a yearly incidence

of bleeding of approximately 3.7% for dual anti-platelet therapy and 12% for triple antithrombotictherapy.w47 The most common site of bleeding isthe gastrointestinal tract (30e40%) and cerebrum(9e10%), with 25% of episodes in the latter siteproving fatal. Furthermore, non-fatal bleedings arean important predictor of mortality post-PCI atfollow-up.w48 Also, in regard to hospitalisation

Left ventricular thrombus formation after myocardial infarction: keypoints

< Left ventricular (LV) regional wall akinesia and dyskinesia resulting in bloodstasis, prolonged ischaemia leading to subendocardial tissue injury withinflammatory changes and a hypercoagulable state, are consistent withVirchow’s triad, resulting in LV thrombus formation.

< Risk factors for the development of LV thrombus include:– large infarct sizes– severe apical asynergy– LV aneurysm– anterior myocardial infarction

< There is reported controversy regarding the negative influence of b-blockersand the protective effect of mitral regurgitation.

< Early data from the prethrombolytic and thrombolytic era suggest that in thesetting of acute myocardial infarction, LV thrombus was present in 7e46% ofpatients.

< Nowadays the reported incidence is lower, probably due to (1) moreaggressive anticoagulation therapies in the acute phase (eg, use of heparin,bivalirudin), (2) smaller infarctions, and (3) improved LV remodelling.

< Timing of LV thrombus assessment is crucial, as assessment too soon afterthe onset of myocardial infarction will miss LV thrombus formation.

< Transthoracic echocardiography is most often used for assessing LVthrombus. However, it is estimated that 10e46% of echocardiograms areinconclusive.

< Delay enhancement cardiac magnetic resonance imaging (CMR) is nowadaysconsidered the gold standard.

< Cine-CMR, transoesophageal echocardiography, radionuclide angiography,and CT seem less appropriate for LV thrombus detection.

< Conditions that increase the risk of systemic embolisation in patients with LVthrombus are: (1) severe congestive heart failure, (2) diffuse LV dilatation andsystolic dysfunction, (3) previous embolisation, (4) advanced age, and (5)presence of LV protruding or mobile thrombi.

< Observational studies conducted in the prethrombolytic and thrombolytic erashave provided support for the hypothesis that warfarin reduces the risk ofembolisation.

You can get CPD/CME credits for Education in Heart

Education in Heart articles are accredited by both the UK Royal College ofPhysicians (London) and the European Board for Accreditation in Cardiologydyou need to answer the accompanying multiple choice questions (MCQs). Toaccess the questions, click on BMJ Learning: Take this module on BMJLearning from the content box at the top right and bottom left of the onlinearticle. For more information please go to: http://heart.bmj.com/misc/education.dtl< RCP credits: Log your activity in your CPD diary online (http://www.

rcplondon.ac.uk/members/CPDdiary/index.asp)dpass mark is 80%.< EBAC credits: Print out and retain the BMJ Learning certificate once you

have completed the MCQsdpass mark is 60%. EBAC/ EACCME Credits cannow be converted to AMA PRA Category 1 CME Credits and are recognisedby all National Accreditation Authorities in Europe (http://www.ebac-cme.org/newsite/?hit¼men02).

Please note: The MCQs are hosted on BMJ Learningdthe best availablelearning website for medical professionals from the BMJ Group. If prompted,subscribers must sign into Heart with their journal’s username and password. Allusers must also complete a one-time registration on BMJ Learning and subse-quently log in (with a BMJ Learning username and password) on every visit.

Education in Heart




/H


ber 2012. Dow

nloaded from


after emergency department visits in the USA foradverse drug events in patients above 65 years,33.3% of the 99 628 hospitalisations concernedwarfarin.w49 Moreover, in the general STEMIpopulation treated with primary PCI and dualantiplatelet therapy but no anticoagulationtherapy, symptomatic cerebral infarction is rare,occurring in 0.75e1.2% of all STEMI patients.w50

Thus, the potential benefit of vitamin K antagonisttreatment on top of dual antiplatelet therapy maynot outweigh the increased bleeding risk. This callsfor a randomised trial to be conducted to determinewhether anticoagulation treatment preventsembolic complications in AMI patients treatedwith primary PCI.

Acknowledgements We gratefully acknowledge the valuablecontribution of ME Hassell, MD.

Contributors RD: drafting the manuscript; FZ: critical revision; JP:interpretation of the data.

Funding This work was supported by a grant from the Dutch HeartFoundation and National Health Insurance Board/ZON MW, theNetherlands to RD. Grant number 2011 T022 + 40-00703-98-11629.

Competing interests In compliance with EBAC/EACCME guidelines,all authors participating in Education in Heart have disclosed potentialconflicts of interest that might cause a bias in the article. Theauthors have no competing interests.

Provenance and peer review Commissioned; internally peerreviewed.

REFERENCES1. Keren A, Goldberg S, Gottlieb S, et al. Natural history of left

ventricular thrombi: their appearance and resolution in theposthospitalization period of acute myocardial infarction. J Am CollCardiol 1990;15:790e800.

< Clinical study assessing appearance and resolution of LVthrombi on echocardiographic follow-up.

2. Asinger RW, Mikell FL, Elsperger J, et al. Incidence of left-ventricular thrombosis after acute transmural myocardial infarction.Serial evaluation by two-dimensional echocardiography. N Engl JMed 1981;305:297e302.

< Eminent clinical study on the incidence of LV thrombusafter AMI.

3. Visser CA, Kan G, Lie KI, et al. Left ventricular thrombus followingacute myocardial infarction: a prospective serial echocardiographicstudy of 96 patients. Eur Heart J 1983;4:333e7.

< Clinical study evaluating the incidence of LV thrombusformation after AMI.

4. Jugdutt BI, Sivaram CA. Prospective two-dimensionalechocardiographic evaluation of left ventricular thrombus andembolism after acute myocardial infarction. J Am Coll Cardiol1989;13:554e64.

< Clinical study evaluating LV thrombi and systemicembolism after AMI.

5. Chiarella F, Santoro E, Domenicucci S, et al. Predischargetwo-dimensional echocardiographic evaluation of left ventricularthrombosis after acute myocardial infarction in the GISSI-3 study.Am J Cardiol 1998;81:822e7.

< Substudy of the GISSI-3 study providing one of the largestdatasets to date on the incidence and risk factors of LVthrombus formation.

6. Solheim S, Seljeflot I, Lunde K, et al. Frequency of leftventricular thrombus in patients with anterior wall acutemyocardial infarction treated with percutaneous coronaryintervention and dual antiplatelet therapy. Am J Cardiol2010;106:1197e200.

< Clinical study assessing the prevalence of LV thrombusformation in STEMI patients treated with primary PCI.

7. Mollet NR, Dymarkowski S, Volders W, et al. Visualization ofventricular thrombi with contrast-enhanced magnetic resonance

imaging in patients with ischemic heart disease. Circulation2002;106:2873e6.

< Clinical study describing the unique position of contrastenhanced MRI in LV thrombus formation.

8. Weinsaft JW, Kim HW, Crowley AL, et al. LV thrombus detectionby routine echocardiography: insights into performancecharacteristics using delayed enhancement CMR. J Am CollCardiol img 2011;4:702e12.

< Clinical study providing more insights on contrast enhancedCMR.

9. Turpie AG, Robinson JG, Doyle DJ, et al. Comparison of high-dosewith low-dose subcutaneous heparin to prevent left ventricularmural thrombosis in patients with acute transmural anteriormyocardial infarction. N Engl J Med 1989;320:352e7.

< Randomised controlled trial comparing high dose with lowdose heparin in patients with LV thrombus formation.

10. Vecchio C, Chiarella F, Lupi G, et al. Left ventricular thrombus inanterior acute myocardial infarction after thrombolysis. A GISSI-2connected study. Circulation 1991;84:512e19.

< Echocardiographic substudy of GISSI-2 assessing the effectof two different thrombolytic agents and heparin on theincidence and features of LV thrombi.

11. Delemarre BJ, Visser CA, Bot H, et al. Prediction of apicalthrombus formation in acute myocardial infarction based on leftventricular spatial flow pattern. J Am Coll Cardiol1990;15:355e60.

< Clinical study assessing the predictive value of LV spatialflow pattern for thrombus formation.

12. Srichai MB, Junor C, Rodriguez LL, et al. Clinical, imaging, andpathological characteristics of left ventricular thrombus:a comparison of contrast-enhanced magnetic resonance imaging,transthoracic echocardiography, and transesophagealechocardiography with surgical or pathological validation. Am HeartJ 2006;152:75e84.

< Clinical study comparing TTE, TOE and contrast enhancedMRI for the detection of LV thrombi.

13. van Dantzig JM, Delemarre BJ, Bot H, et al. Left ventricularthrombus in acute myocardial infarction. Eur Heart J1996;17:1640e5.

< Eminent review on diagnosis and management of LVthrombus patients.

14. Weinsaft JW, Kim RJ, Ross M, et al. Contrast-enhancedanatomic imaging as compared to contrast-enhanced tissuecharacterization for detection of left ventricular thrombus. J AmColl Cardiol img 2009;2:969e79.

< Clinical study providing important information on contrastenhanced anatomic imaging compared to contrastenhanced tissue characterisation of LV thrombi.

15. Meltzer RS, Visser CA, Fuster V. Intracardiac thrombi andsystemic embolization. Ann Intern Med 1986;104:689e98.

< Excellent review on intracardiac thrombi and anticoagulanttreatment.

16. Johannessen KA, Nordrehaug JE, von der LG, et al. Risk factorsfor embolisation in patients with left ventricular thrombi and acutemyocardial infarction. Br Heart J 1988;60:104e10.

< Clinical study assessing risk factors for systemicembolisation in LV thrombus patients.

17. Haugland JM, Asinger RW, Mikell FL, et al. Embolic potential ofleft ventricular thrombi detected by two-dimensionalechocardiography. Circulation 1984;70:588e98.

< Clinical study seeking an association betweenechocardiographic appearances of LV thrombi andsystemic embolisation.

18. Domenicucci S, Bellotti P, Chiarella F, et al. Spontaneousmorphologic changes in left ventricular thrombi: a prospectivetwo-dimensional echocardiographic study. Circulation1987;75:737e43.

< Clinical study with serial echocardiographic studiesdemonstrating spontaneous morphologic changes inLV thrombi.

19. Vaitkus PT, Barnathan ES. Embolic potential, prevention andmanagement of mural thrombus complicating anterior myocardialinfarction: a meta-analysis. J Am Coll Cardiol 1993;22:1004e9.

< Relevant meta-analysis regarding the embolic potential andmanagement of LV thrombus.

20. Stratton JR, Ritchie JL. 111In platelet imaging of left ventricularthrombi. Predictive value for systemic emboli. Circulation1990;81:1182e9.

< Clinical study with indium-111 platelet imaging of LVthrombi.

Education in Heart




/H


ber 2012. Dow

nloaded from


Documents

ACUTE CORONARY SYNDROMES Left ventricular thrombus ... · ACUTE CORONARY SYNDROMES Left ventricular thrombus formation after acute myocardial infarction Ronak Delewi,1 Felix Zijlstra,2