Bioprosthetic Valve Thrombosis · 2017-05-30 · tiny within the context of ongoing evaluations of...

Listen to this manuscript’s

audio summary by

JACC Editor-in-Chief

Dr. Valentin Fuster.

J O U R N A L O F T H E AM E R I C A N C O L L E G E O F C A R D I O L O G Y V O L . 6 9 , N O . 1 7 , 2 0 1 7

ª 2 0 1 7 B Y T H E AM E R I C A N C O L L E G E O F C A R D I O L O G Y F O U N D A T I O N

P U B L I S H E D B Y E L S E V I E R

I S S N 0 7 3 5 - 1 0 9 7 / $ 3 6 . 0 0

h t t p : / / d x . d o i . o r g / 1 0 . 1 0 1 6 / j . j a c c . 2 0 1 7 . 0 2 . 0 5 1

REVIEW TOPIC OF THE WEEK

Bioprosthetic Valve Thrombosis
Rishi Puri, MBBS, PHD,a,b,c Vincent Auffret, MD, MSC,a,d Josep Rodés-Cabau, MDa
ABSTRACT

Fro

Ce

Au

La

Me

an

rep

thi

Ma

Bioprosthetic valve (BPV) thrombosis is considered a relatively rare clinical entity. Yet a more recent analysis involving a

more systematic echocardiographic follow-up, the advent of transcatheter heart valve (THV) technologies coupledwith the

highly sensitive nature of 4-dimensional computed tomographic imaging for detecting subclinical thrombi upon both

surgically implanted andTHVs, has generated enormous interest in thisfield, casting new light on both its true incidence and

clinical relevance. Debate continues among clinicians as to both the clinical relevance of subclinical BPV thrombosis and the

value of empirical oral anticoagulation following BPV implantation. Furthermore, currently no systematic, prospective data

exist regarding the optimal treatment approach in THV recipients. The authors provide an overview of the clinical and

subclinical spectrum of BPV thrombosis of surgical and THVs, outline its diagnostic challenges, summarize its pathophys-

iological basis, and discuss various therapeutic options that are emerging, particularly within the rapidly expanding field of

THV implantation. (J Am Coll Cardiol 2017;69:2193–211) © 2017 by the American College of Cardiology Foundation.

O f the >110,000 valve replacementsundertaken annually within the UnitedStates alone (1), there has been a gradual

shift from mechanical to bioprosthetic valve (BPV)implantations; bioprostheses now account fornearly 80% of all surgical aortic valve replacements(SAVRs) within the United States (2). This stemslargely from the changing demographics of valve re-cipients, who are now older, with higher surgicalrisk profiles, and who also harbor greater bleedingrisk because of the need for lifelong oral anticoagula-tion (OAC) following valve replacement. AlthoughBPVs are less thrombogenic than their mechanicalcounterparts, clinically apparent BPV thrombosis(BPVT) is a rare yet important clinical entity, knownto occur at all 4 valve locations. Its incidence and clin-ical relevance have recently come under intense scru-tiny within the context of ongoing evaluations of theperformance and durability of transcatheter heart

m the aQuébec Heart and Lung Institute, Laval University, Québec Cit

nter for Clinical Research, Cleveland, Ohio; cDepartment of Medicine,

stralia; and the dDepartment of Cardiology, Rennes University Hospita

boratory, Rennes 1 University, Rennes, France. Dr. Rodés-Cabau has re

dtronic, and Heart Leaflet Technologies. Dr. Auffret has received fellowsh

d research grants from Abbott, Edwards Lifesciences, Medtronic, Biose

orted that he has no relationships relevant to the contents of this paper to

s paper.

nuscript received November 25, 2016; revised manuscript received Febru

valves (THVs) and the advent of 4-dimensional (4D)computed tomography (CT) as a novel means ofvalve surveillance. This, coupled with the continuingdebate regarding optimal antithrombotic regimens inboth surgical and THV recipients, has reignited thefocus on BPVT per se. Moreover, its true incidenceis probably greater than previously thought, andits clinical relevance is likely to increase in conjunc-tion with the rapidly expanding field of THVtechnologies. In this review, we summarize BPVTas a clinical and subclinical entity, outline its diag-nostic challenges, provide an overview of its patho-physiological basis, and discuss various therapeuticoptions.

VALVE BIOPROSTHESES

SURGICAL BIOPROSTHESES. BPVs are derivedfrom either human or animal tissue (3–5) (Figure 1).

y, Québec, Canada; bCleveland Clinic Coordinating

University of Adelaide, Adelaide, South Australia,

l, and Department of Signal and Image Processing

ceived research grants from Edwards Lifesciences,

ip support from Fédération Française de Cardiologie

nsors, Terumo, and Boston Scientific. Dr. Puri has

disclose. Raj Makkar, MD, served as Guest Editor for

ary 10, 2017, accepted February 16, 2017.

https://s3.amazonaws.com/ADFJACC/JACC6917/JACC6917_fustersummary_08




http://crossmark.crossref.org/dialog/?doi=10.1016/j.jacc.2017.02.051&domain=pdf

http://dx.doi.org/10.1016/j.jacc.2017.02.051

ABBR EV I A T I ON S

AND ACRONYMS

4D = 4-dimensional

ACT = activated clotting time

AF = atrial fibrillation

BPV = bioprosthetic valve

BPVT = bioprosthetic valve

thrombosis

CT = computed tomography

DAPT = dual-antiplatelet

therapy

HALT = hypoattenuated leaflet

thickening

NOAC = novel oral

anticoagulant agent

OAC = oral anticoagulation

PPM = prosthesis-patient

mismatch

SAVR = surgical aortic valve

replacement

TAVR = transcatheter aortic

valve replacement

TEE = transesophageal

echocardiography

THV = transcatheter heart

valve

TTE = transthoracic

echocardiography

Puri et al. J A C C V O L . 6 9 , N O . 1 7 , 2 0 1 7

Bioprosthetic Valve Thrombosis M A Y 2 , 2 0 1 7 : 2 1 9 3 – 2 1 1

2194

Human tissue valves can be divided into 2categories: allografts and autografts. Allo-grafts are cadaveric valves that are cry-opreserved in liquid nitrogen until use andimplanted without a stent. Autografts arecomposed of the patient’s own valve. Themost common autograft is the Ross proced-ure, during which the patient’s pulmonaryvalve is transplanted to the aortic position,and an allograft valve is placed in the pul-monary position (3). Clinical indications forhuman tissue valves are currently limited(i.e., women of childbearing age, endocardi-tis). On the contrary, animal tissue valves,known as xenografts, are the most widelyused type of BPV, falling into 2 categories:stented and stentless valves, and can beeither porcine or bovine. Valve leaflets arecrosslinked with glutaraldehyde and moun-ted on a metallic or polymer supportingstent. Pericardial valves consist of sheets ofbovine pericardium mounted inside oroutside a supporting stent (4). Stentless BPVstend to be hemodynamically superiorcompared with stented BPVs (5). Suturelessvalves have recently emerged as an option toreduce cross-clamp and cardiopulmonarybypass duration, thereby improving surgical

outcomes and facilitating minimally invasive ap-proaches. These valves seem also especially suitablefor patients with small aortic annuli, to optimize he-modynamic results (6).

THVs. Most THVs are stent-caged BPVs implantedusing a catheter-based delivery system (Figure 1).First-generation devices have been the most widelyused THVs in clinical practice to date. The SAPIENvalve (Edwards Lifesciences, Irvine, California) and itscurrent iterations (SAPIEN XT and SAPIEN 3) consistof bovine pericardial leaflets sewed within a balloon-expandable cobalt-chromium stent, with the SAPIEN3 THV being covered by a polymer skirt at the bottomof the stent frame aimed to reduce paravalvular leaks.The CoreValve, CoreValve Evolut R and CoreValveEvolut PRO (Medtronic, Minneapolis, Minnesota) aremade of a self-expanding nitinol stent with porcinepericardial leaflets (7). Several manufacturers havedeveloped newer generation devices, which aremainly nitinol-based self-expandable THVs, usingeither bovine or porcine pericardium (Figure 1). Ofnote, THV leaflets are thinner than surgical BPV leaf-lets (w0.25 mm vs. w0.40 mm, respectively) (8).

More recently, several THV platforms haveemerged for treating mitral regurgitation (9). Most of

these valves consist of a stent frame made of nitinoland 3 leaflets of bovine or porcine pericardium su-tured inside the stent frame. The most commonmechanism of valve fixation consists of capturingthe native mitral leaflets, in addition to radial force.

DIFFERENCES BETWEEN SURGICAL

BPVs AND THVs

Fundamental differences between BPVs and THVs liein their respective implantation techniques. Althoughmanipulation of BPVs is avoided during SAVR,crimping the valve upon the delivery catheter isrequired for THV implantation, and this may translateinto disrupted collagen fiber orientation on the valvesurface (10,11). Although the clinical relevance ofthese findings remains unclear, over the medium tolong term, these subtle surface alterations could serveas a nidus for calcification, structural valve deterio-ration, and subsequent thrombus formation. More-over, the mounting of the valves within a rigid stentincreases the amount of mechanical stress subjectedupon the THV leaflets, compared with surgical BPVs,which have greater flexibility within the stent. Thepresence of native valve calcification may also leadto suboptimal deployment, and paravalvular leaks,which may also increase the risk for valve failure andthrombosis (5,12,13). However, the superior hemo-dynamic results currently demonstrated by THVscompared with surgical BPVs, with a much lowerincidence of prosthesis-patient mismatch (PPM) (14),may nevertheless contribute to improved durability,but data on long-term THV outcomes are currentlylimited (5).

DEFINITION AND DIAGNOSIS OF BPVT

The diagnosis of BPVT remains challenging in clinicalpractice, in part because of a lack of general aware-ness of this condition, coupled with the lack of auniversal definition, and thus is often confused orthe term is used interchangeably with “valvedeterioration.” In published surgical reports, valvethrombosis is considered any thrombosis unrelated toinfection, attached to or within close proximity of thevalve per se, thus occluding the path of blood flow,interfering with valve function, or sufficiently largeso as to warrant treatment (15). In contrast, structuralvalve deterioration includes dysfunction or deterio-ration of the prosthetic valve, exclusive of infectionor thrombosis, as determined by reoperation,autopsy, or clinical investigation. This usuallyinvolves wear and tear, fracture, calcification, leaflettear, stent creep, suture line disruption of valvecomponents, and so forth. Nonstructural dysfunction

FIGURE 1 Bioprosthetic Valves

Stented porcine valves

Stented pericardial valves

Stentless porcine valves

Stentless pericardial valves

MedtronicHancock II

MedtronicMosaic

St. Jude MedicalBiocorTM

St. Jude MedicalTrifectaTM

St. Jude MedicalPortico TM

SymetisACURATE neo TM

Boston ScientificLotusTM

Direct FlowMedical

Carpentier-EdwardsMagna EaseTM

Sorin groupMitroflowTM

Sorin groupPericarbon FreedomTM

St. Jude MedicalToronto SPVTM ·

· ·

· MedtronicFreestyle ·

Medtronic3f Enable ·

Transcatheter heart valves

EdwardsSAPIEN 3

MedtronicCoreValve Evolut R

Bioprosthetic heartvalves

Xenografts

Stented Stentless

Transcatheter heart valves

First generationiterations

Edwards SAPIENEdwards SAPIEN XTMedtronic CoreValve

PorcineEdwards Prima PlusMedtronic Freestyle

St. Jude Medical Toronto SPV

PericardialCE PERIMOUNT lineEdwards Intuity Elite

St. Jude Medical TrifectaSorin MitroflowSorin Crown PRTSorin PericarbonSorin Perceval

PorcineCE SAV

Medtronic Hancock IIMedtronic Mosaic

St. Jude Medical BiocorSt. Jude Medical Epic

PericardialSorin Pericarbon Freedom

Sorin Solo SmartMedtronic ATS 3f

Newer generationDevices

Edwards SAPIEN 3Medtronic CoreValveEvolut R, Evolut PRO

Boston Scientific LotusDirect Flow Medical

St. Jude Medical PorticoSymetis ACURATE

JenaValve

Human tissue valves

AllograftsCadaveric aortic valve

AutograftsRoss procedure

A

B

(A) The general classification of bioprosthetic valves. (B) The various types of surgical and transcatheter heart valves.

J A C C V O L . 6 9 , N O . 1 7 , 2 0 1 7 Puri et al.M A Y 2 , 2 0 1 7 : 2 1 9 3 – 2 1 1 Bioprosthetic Valve Thrombosis

2195

TABLE 1 Clinical and Imaging Characteristics and Predictive Models of Surgical

Bioprosthetic Valve Thrombosis

Clinical and Imaging Characteristics

Clinical features � Occurrence within 1–2 yrs of BPV implantation� Presence of risk factors: paroxysmal AF, subtherapeutic INR,

recent withdrawal of OAC, history of TE event, depressedLVEF, known hypercoagulable condition.

� New-onset acute HF, progressively worsening dyspnea, newTE event

� Regression of HF symptoms with OAC

Echocardiographic features � Direct visualization of valve thrombosis in rare cases� 50% mean gradient increase compared with the initial

post-operative evaluation� Increased cusp thickness (>2 mm), especially on the downstream

aspect of the BPV (ventricular side for mitral and tricuspidprostheses, arterial side for aortic and pulmonary prostheses)

� Abnormal/reduced leaflet mobility� Regression of BPV abnormalities with OAC, usually within

1–3 months of OAC initiation

CT features � Reduced leaflet motion on 4D CT� HALT

Predictive Models of Surgical BPVT as Proposed by Egbe et al. (16)

Variables Total Score Sensitivity (%) Specificity (%) PPV (%) NPV (%)

A. 50%meangradient increase 1 45 89 68 77

B. Increased cusp thickness 1 74 69 55 84

C. Abnormal cusp mobility 1 63 70 51 81

D. Paroxysmal AF 1 63 73 54 80

E. Subtherapeutic INR 1 30 92 67 73

Combination of variables

A and B 2 86 57 50 89

A, B, and C 3 72 90 78 87

A, B, C, and D 4 70 94 87 86

A, B, C, D, and E 5 66 93 85 89

AF ¼ atrial fibrillation; BPV ¼ bioprosthetic valve; BPVT ¼ bioprosthetic valve thrombosis; CT ¼ computedtomography; 4D¼ 4-dimensional; HALT¼ hypoattenuated leaflet thickening; HF¼ heart failure; INR¼ internationalnormalized ratio; LVEF ¼ left ventricular ejection fraction; OAC ¼ oral anticoagulation; TE ¼ thromboembolic.



2196

relates to problems (excluding thrombus and infec-tion) indirectly involving valve components, whilestill resulting in a dysfunctional valve (i.e., pannus,paravalvular leak, inappropriate sizing or positioning)(15). In clinical practice, frequent confusion of theseterms and definitions has invariably led to under-reporting of valve thrombosis. Furthermore, the ma-jority of BPVTs have been reported in the aorticposition, thus presenting a major limitation in trulyunderstanding data related to BPVT in other posi-tions, including those involving THVs.

In their series, Egbe et al. (16) demonstrated thatthe possibility of surgical BPVT was mentioned inonly 2 of 42 (5%) transthoracic echocardiograms andin 6 of 45 (13%) transesophageal echocardiograms,although echocardiography reports described theseBPVs as being “abnormal” in all instances. However,Daniel et al. (17) reported that if transthoracic echo-cardiography were to remain the first-line screeningexamination for specifically detecting BPV abnor-malities, it would identify BPVT in only 13% of cases.

Therefore, transesophageal echocardiography(TEE) is strongly recommended when the clinicalsuspicion for BPV abnormalities remains high,particularly for BPVs within the mitral or tricuspidposition (17,18).

Summarized in Table 1, Egbe et al. (16) recentlyproposed a diagnostic model combining 3 echocar-diographic predictors: 1) a 50% increase in the trans-valvular gradient compared with baseline within5 years of surgery, in the absence of a high cardiacoutput state; 2) increased cusp thickness (>2 mm),especially on the downstream aspect of the valve; and3) abnormal cusp mobility. Applying this simplemodel to the 138 patients included in their studyyielded rates of 72%, 90%, 78%, and 87% for sensi-tivity, specificity, and positive and negative predictedvalues for diagnosing BPVT, respectively (16).Although the limited number of events (n ¼ 46) in thestudy makes the multivariate model prone to over-fitting, to date, this model remains the best availableoption to simply and accurately diagnose BPVT.Therefore, it is our opinion that this model should beexternally validated in future studies and subse-quently implemented in clinical practice for identi-fying BPVT of both surgical BPVs and THVs.Finally, because most patients with BPVT underwentsurgery within the first 5 years post-implantation,consistent with the work of Egbe et al. early andyearly echocardiographic surveillance of BPV re-cipients (although not currently supported by clinicalguidelines) (19) may be warranted. Figure 2 highlightssome echocardiographic examples of BPVT.

THV THROMBOSIS. Recent systematic retrospectiveanalyses have attempted to describe and categorizeTHV thrombosis (20,21). Although limited by reportingand publication biases, the incidence of THV throm-bosis is likely to be at least 0.6%, possibly more. Mostcases occurred within the first year post–transcatheteraortic valve replacement (TAVR), with a median onsetof 6 months post-TAVR. The majority of patients withTHV thrombosis (65%) presented with gradual onset ofdyspnea, with no reported or published cases to dateof overt systemic embolism. Nearly one-third ofpatients were asymptomatic. Direct echocardio-graphic visualization of thrombus was possible in onlya minority of cases, with almost all patients demon-strating a rising transaortic gradient from baseline(>90% presented with a mean transaortic gradient>20 mm Hg). Morphological features demonstrable onechocardiography related to leaflet thickening andreduced leaflet mobility. Furthermore, a preponder-ance of reported cases occurred following balloon-expanding valve implantation, with a large majority

FIGURE 2 Echocardiographic Images of Bioprosthetic Valve Thrombosis

(A) A transesophageal echocardiographic (TEE) frame of mitral bioprosthesis thrombosis with leaflets thickened by echodense material

(arrow) on both sides of the valve in this case. Reprinted with permission from Egbe et al. (16). (B) A TEE frame of a large thrombus (arrow)

on the aortic side of a bioprosthetic valve. Reprinted with permission from Cremer et al. (35). (C) A TEE frame of a 29-mm SAPIEN XT

transcatheter heart valve (THV) with thrombus (arrow) on the downstream side of the leaflets. Reprinted with permission from Latib et al.

(21). (D) A transthoracic echocardiographic frame of a mobile pedunculated thrombus (arrow) attached on the leaflet of a 26-mm SAPIEN

XT THV and floating in the left ventricular outflow tract. Reprinted with permission of Professor Donal, Cardiology Department, Rennes

University Hospital, France.


2197

of THV thrombosis cases occurring despite the use ofdual-antiplatelet therapy (DAPT). However, theprompt and empirical commencement of OAC wasnearly universal in restoring transaortic gradients toclose to their baseline values, following a medianperiod of 40 days post-OAC commencement.

De Marchena et al. (22) provided unique patho-logical insights on the basis of autopsy specimens of3 THV recipients who presented with BPVT (involvingboth balloon and self-expanding THVs). DespiteTHV leaflets being thinner (with the assumption ofgreater fragility) than leaflets of surgically implantedvalves (8), pathological examination of these 3 au-topsy cases failed to reveal the presence of micro-injury or leaflet degeneration that could have arisen

at the time of implantation, illustrating that BPVTmay not necessarily arise from periprocedural leafletinjury. Moreover, leaflet thrombosis was associatedwith macrophage layering beneath and inflammatoryinfiltrate within the thrombus. The absence of alymphocytic infiltrate in these cases also points to anunderlying nonautoimmune process.

Data in the transcatheter mitral valve replacementfield are limited to a few cases in the context of first-in-human or early feasibility studies (9). However,cases of clinically relevant transcatheter mitralvalve thrombosis have already been reported (23),emphasizing the importance of vigilance about thispotential complication in the transcatheter mitralspace.



2198

MECHANISMS AND RISK FACTORS RELATED

TO VALVE THROMBOSIS

Although we currently lack a precise understandingof mechanisms leading to BPVT, the underlyingprinciples invariably relate to perturbations in bloodflow and activation of various hemostatic factorsinvolving mechanisms common to medical device-induced thrombosis.

HEMOSTATIC ACTIVATION. Whilst an intact endo-thelium effectively regulates vessel tone and anantithrombotic state, artificial surface contact withblood promotes clot formation and involves multiplemechanisms, including protein adsorption, as well asplatelet, leukocyte, and red cell adhesion, thrombingeneration, and complement activation. The initi-ating event underlying device-induced thrombosis isthought to involve plasma protein adsorption, withthe protein layer modulating further reactions.Hydrophilicity is a key determinant of proteinadsorption, with more proteins adsorbing to hydro-phobic compared with hydrophilic surfaces. Subse-quent deposition of fibrinogen, fibronectin, and vonWillebrand factor mediates platelet adhesion andactivation, the release of thromboxane A2, adenosinediphosphate, and a host of other platelet agonists,which serve to amplify this overall process, ultimatelyresulting in thrombosis and eventual platelet con-sumption following prolonged exposure to theforeign material. Red cell adhesion also occurs butfollows a passive process, with adherent red cellsalso releasing adenosine diphosphate, further aug-menting platelet activation. Additional leukocyteadhesion (involving the interaction between neutro-phils and P-selectins), along with complement acti-vation, serve to intensify the milieu of local plateletactivation (24).

PERTURBATIONS IN BLOOD FLOW. Flow conditionsaround valve prostheses are associated with endo-thelial damage, vascular remodeling, and thrombosis.Studies evaluating wall shear stress following valveimplantation suggest that wall shear stress valuesincrease from the base of the valve leaflets toward theleaflet tips, where maximum flow activation occurs.High wall shear stress levels promote platelet acti-vation, and prolonged exposure (blood residencetime) to these shear stresses, combined with flowrecirculation, stimulate localized thrombogenesis,especially within the prosthetic valve region, wherethere is a complex interplay between flow patternsand blood stasis (25). Blood residence time wasrecently found to be significantly greater on THVscompared with surgically implanted valves (26).

Moderate or severe PPM occurs when the effectiveaortic valve area of a normal functioning prostheticvalve is small in relation to patient body surface areaand is present in up to 44% of SAVR recipients, influ-encing post-SAVR mortality (27). Wall shear rate, anindirect measure of shear stress, was significantlyhigher in those TAVR recipients demonstratingPPM compared with those without PPM. Although adirect link between PPM and BPVT has yet to beelucidated, its various rheological perturbations areconducive to valve thrombosis. Elevated transvalvulargradients and subsequent alterations in shearstress during PPM significantly alter the configurationof circulating von Willebrand factor, while concur-rently activating the coagulation cascade, thus pre-disposing to bleeding and thrombosis (28,29). Alteredflow patterns result in areas of relative blood stasis,particularly behind BPV leaflets, promoting plateletadhesion and thrombosis (30). Post-SAVR, PPM hasbeen shown to directly associate with the stenosis typeof structural valve deterioration with pannus over-growth (31), which may increase the risk of throm-bosis. Future studies are therefore warranted toexplore the possible relationship between PPM andBPVT.

PATIENT-RELATED RISK FACTORS FOR BPVT. Certainpatient-related comorbidities may further predisposeBPV recipients to valve thrombosis by promoting ahypercoagulable state. These include the presenceof renal insufficiency, obesity, diabetes mellitus,concomitant smoking, and anemia. Additionally,periprocedural trauma activates coagulation path-ways, with subsequent exposure of tissue factorwithin the endothelial layer, a potent prothromboticstimulus. Furthermore, low cardiac output states alsopromote thrombosis. BPVT affecting the tricuspid andpulmonary valves (regions of relatively slower flowcompared with the aortic and mitral valves) are farmore frequent than in left-sided heart valves (32),with the propensity for valve thrombosis beinghigher in the tricuspid compared with pulmonaryposition. Thus far, the majority of reported THVthromboses were in those who received balloon-expandable aortic valves (20,21). These findingsshould, however, be considered preliminary,requiring ongoing investigation and more extensiveclinical reporting, as we currently lack a detailedunderstanding of the possible mechanisms underly-ing such observations that are specific to each valveand patient phenotype. Moreover, a degree ofreporting bias, is likely to exist, as both clinicallysignificant and subclinical valve thrombosis havebeen reported in a range of THV systems.


2199

A recent retrospective analysis of valve hemody-namic deterioration (defined as a $10 mm Hg increasein transprosthetic mean gradient during follow-up)in more than 1,500 TAVR recipients identified anumber of clinical criteria associated with thisphenomenon over a mean period of 20 months (33).The absence of OAC post-TAVR, valve-in-valve pro-cedures (TAVR within a previously implanted surgicalvalve), smaller sized THVs, and greater body massindexes are each independently associated with valvehemodynamic deterioration. The incidence of valvehemodynamic deterioration was <2% in those pre-scribed OAC post-TAVR compared with 6% in thosenot prescribed OAC. These results suggest that anunderlying thrombotic process post-TAVR may be animportant, more prevalent phenomenon thancurrently considered.

CLINICAL INSIGHTS FROM BPVT

CLINICAL PRESENTATION. The spectrum of BPVTcan range from an incidental finding in asymptomaticpatients to overt syncope, acute heart failure, and, inits most severe form, cardiogenic shock due to me-chanical obstruction (34,35). Progressive dyspnea isperhaps the most common clinical presentation(20,21). Moreover, in some cases, patients may pre-sent with systemic or pulmonary embolism fromfriable valve-associated thrombi owing to a process ofdelayed thrombus organization (36).

PREVALENCE. The precise incidence of BPVT isdifficult to ascertain, as most large prospective reg-istries failed to systematically report BPVTs, insteadreferring to the more general notion of thromboem-bolism. Table 2 provides a summary of pertinentpublished reports regarding surgical BPVT. In a recentmeta-analysis involving SAVR recipients, early andlate BPVT were only reported in 3 and 13 of the54 included studies, with rates of 0.34% and 0.04%per year, respectively (37). Puvimanasinghe et al. (38)also reported an extremely low 0.03% per year pre-dicted rate of BPVT using microsimulation models.However, this result was derived from only 3 eventsreported across 4 studies encompassing 9,925 pa-tients. Other investigators reported up to a 6% inci-dence of BPVT in retrospective analyses of mitralvalve recipients (39,40), occurring at a mean follow-up of 12 months. Butnaru et al. (39) described 9cases (6%) of BPVT by retrospectively reviewing TEEperformed in 149 patients with suspected prosthesisdysfunction on transthoracic echocardiography.BPVT was confirmed by pathology following reoper-ation in 3 of these cases and following resolutionof prosthesis dysfunction with OAC in the other

6 patients. Similarly, Oliver et al. (40) reported10 cases (6.2%) of BPVT, diagnosed at a mean follow-up of 83 months post-implantation, among 161patients evaluated by TEE for mitral BPV dysfunction.Pathology confirmed BPVT in 3 patients who under-went reoperation, whereas the other 7 patients weresuccessfully treated with OAC.

More recently, Egbe et al. (16) reported on thelargest series of BPVT to date. Using a Mayo Clinicpathology database, they identified 46 cases (11.6%)of histologically proven BPVT among 397 patientswho underwent BPV explantation between 1997 and2013. Interestingly, the prevalence of BPVT waslargely comparable at all 4 valve locations, with ratesof 10.9%, 12.7%, 12.1%, and 11.6% within the aortic,mitral, tricuspid, and pulmonary positions, respec-tively. On the basis of the total number of patientswith BPV implantations and at least 1 follow-upechocardiographic examination during the studyperiod (n ¼ 3,161), the investigators estimated a BPVTincidence of 1.46%. However, in a previous reportfrom the same group during the same time period(41), 17 patients with suspected BPVT were success-fully managed with either OAC or thrombolytic ther-apy and thus did not undergo BPV explantation.Therefore, the estimated incidence of BPVT couldconceivably be 1.99% or more within this institutionaldatabase.

A crucial point highlighted in a recent case series isthe common misconception that the risk for BPVT islimited to within the first 3 months post-implantation(16,39,41–43). Pislaru et al. (41) reported a peak inci-dence of BPVT at 13 to 24 months, whereas only24% of cases (n ¼ 8 of 34 patients) occurred within3 months post-surgery in a recent review of publishedreports from Dohi et al. (43). In their report, Egbeet al. (16) also demonstrated a median time toexplantation for BPVT of 24 months, with 15% ofcases occurring more than 5 years post-implantation.However, this delay was still significantly lower thanthe time to explantation of age-, sex-, and prosthesisposition–matched control subjects with structuralvalve failure (median 108 months) (16). Figure 3shows 2 examples of the morphology of explantedBPVs.

CLINICALLY ASSOCIATED FACTORS. Table 1 summarizesthe clinical and imaging characteristics associatedwith BPVT. Specific risk factors that might help di-agnose BPVT remain elusive and largely empirical, asthe scarcity of this clinical issue has hindered theability to identify robust predictors. Current guide-lines state that BPV recipients with hypercoagulableconditions, histories of thromboembolic events,

TABLE 2 Overview of Studies in the Setting of Surgical Bioprosthetic Valve Thrombosis

First Author (Ref. #) Design Number of Patients Main Findings

Oliver et al. (40) Single-center retrospectiveanalysis, 1990–1995

161 10 of 161 (6%, mean age 63 yrs) mitral bioprosthesisrecipients, underwent TEE for clinical or echocardiographiccriteria of prosthetic malfunction, presented with BPVT

Mean time from implantation 83 months8 patients with CHF, 2 with TIANo case was diagnosed by TTE2 patients underwent reoperation, 8 remaining patients

underwent OAC; partial or complete resolution of valvedysfunction demonstrated on TEE a mean of 86 d afterinitiation of OAC

Puvimanasinghe et al. (38) Meta-analysis and microsimulation 5,837 Annual rate of BPVT: 0.03%/yr predicted from 4 reports with9,925 patient-yrs of follow-up

Brown et al. (42) Single-center retrospectiveanalysis, 1993–2009

4,568 8 patients (0.18%, mean age 77 yrs) underwent reoperationfor BPVT a median of 398 days from implantation; allpatients received stented porcine bioprostheses

Incidence of BPVT ranged from 0.37% to 1.26%, depending onvalve type

All patients received aspirin early after surgery, but only3 received warfarin for the first 3 months post-operatively

Butnaru et al. (39) Single-center retrospectiveanalysis, 2002–2011

149 9 of 149 patients (6%) who underwent bioprosthetic mitralvalve replacement showed evidence of BPVT; all patientshad their native valves preserved

Mean time from implantation was 16 months5 patients presented with CHF (56% of BPVT cases) among

whom diagnosis was made at reoperation in 2 and aftersuccessful trial of OAC in 3

Echocardiographic controls performed after 9–75 days showedcomplete resolution of abnormalities in the 6 patientstreated with OAC

Dohi et al. (43) Case report and literature review — 26 reports (n ¼ 34, mean age 69 yrs) of aortic BPVT since1990, excluding patients on OAC or thrombolytic therapy

BPVT occurred >13 months post-operatively in 50% of cases,whereas only 24% of BPVT occurred within 3 months ofsurgery

31 of 34 valves (91%) were porcine22 of 30 cases (73%) with known antithrombotic regimen

occurred among patients (n ¼ 18) under aspirin or notreatment (n ¼ 4)

Egbe et al. (16) Single-center retrospective case-matched analysis, 1997–2013

138 (46 cases of BPVTmatched with 92 cases ofstructural deterioration)

BPVT represented 11.6% of indications of bioprostheticexplantations

BPVT prevalence was similar for the 4 positions, ranging from10.9% to 12.7%

Median time to explantation was 24 months for BPVT vs.108 months for structural deterioration (p < 0.001)

Possible BPVT was mentioned in only 2 of 42 TTEs (5%) and in6 of 45 TEEs (13%)

Paroxysmal AF, subtherapeutic INR on OAC, 50% increase ingradient within 5 yrs, increased cusp thickness (>2 mm,commonly on the downstream aspect of the valve), andabnormal cusp mobility were independent predictors ofBPVT

Jander et al. (44) Single-center retrospectiveanalysis, 2007–2012

1,751 17 BPVT cases among 749 porcine bioprosthesis recipients(2.3%) vs. 0 cases among 1,002 bovine bioprosthesisrecipients (p < 0.001)

Echocardiographic parameters returned to post-operativevalues after OAC initiation in 15 stable patients

2 patients underwent reoperation because of instability orsuspicion of endocarditis

Pislaru et al. (41) Single-center retrospectiveanalysis, 1997–2013

32 Comparison of 15 BPVT patients managed initially with OAC,with 17 BPVT cases managed by surgery/thrombolysis

Low prevalence of traditional risk factors for BPVTPeak incidence of BPVT: 13–24 monthsOnly 19% of cases were identified on initial TTEEchocardiographic improvement (normalization in 13 of

15 patients) without death or stroke in all stable patientswho underwent OAC, regardless of BPVT position

Huygens et al. (37) Meta-analysis of 54 studiespublished 2000–2015

55,712 Early BPVT rate: 0.3%/yr among 3 studiesLate BPVT rate: 0.04%/yr among 13 studiesReintervention for BPVT: 0.03%/yr among 25 studies

CHF ¼ congestive heart failure; TEE ¼ transesophageal echocardiography; TIA ¼ transient ischemic attack; TTE ¼ transthoracic echocardiography; other abbreviations as in Table 1.



2200

FIGURE 3 Pathological Specimen of an Explanted Bioprosthesis

(A) An example of bioprosthetic valve thrombosis of a 25-mm Bicor prosthesis (St. Jude Medical, St. Paul, Minnesota) with thrombi on

the downstream side of the cusps. Reprinted with permission from Cremer et al. (35). (B) An example of an explanted Sorin Mitraflow

bioprosthetic valve with thrombus on the downstream side of the valve leaflets. Reprinted with permission from Dr. E. Flecher, Cardiothoracic

Surgery Department, Rennes University Hospital, France.


2201

depressed left ventricular function, and ongoingatrial fibrillation (AF) are at a greater thrombotic risk(19). In their multivariate analysis, Egbe et al. (16) alsoidentified suboptimal anticoagulation in patientswith OAC therapy and paroxysmal AF as clinicalpredictors of surgical BPVT. Two recent studies(42,44) suggested a higher risk for BPVT with porcinecompared with pericardial valves, especially beyondthe first 3 post-operative months, which may berelated to a specific design feature of porcine valvesthat could promote blood stasis within leaflets (42).

SUBCLINICAL BPVT

Advances in computed tomographic technology havepermitted highly sensitive volume-rendered 4D valveimaging, yielding intriguing insights into thebehavior (in particular leaflet motion) of BPVs rela-tively soon post-implantation (Figure 4). These dataare summarized in Table 3. Pache et al. (45) used CT asa means of evaluating 156 consecutive SAPIEN S3 THVrecipients, with patients imaged at a mean time of5 days post-TAVR. Hypoattenuated leaflet thickening(HALT) was formally assessed (without 4D imageacquisition) and discovered in 16 patients (10%). Thisphenomenon seemed to primarily involve a single-leaflet cusp in most cases, with associated leafletrigidity. These findings were corroborated by Leet-maa et al. (46), who used CT between 1 and 3 monthspost-procedure in 140 TAVR recipients. HALT wasnoted in 5 patients (4%), with 1 patient (with pre-

existing congestive heart failure) presenting withfurther heart failure symptoms coupled with re-ductions in valve effective orifice area. HALT onCT was confirmed with TEE, which revealedrestricted cusp motion, with 3 months of OAC effec-tively restoring THV function. In a consecutive series(n ¼ 405) of patients undergoing post-TAVR CT(in addition to TEE), Hansson et al. (47) reported theincidence of overt THV thrombosis (evidenced byHALT) to be 7%, with the majority of cases (23 of 28patients) being subclinical in nature. The absence ofOAC and larger sized (29 mm) THVs were indepen-dently associated with THV thrombosis. Warfarinadministration reverted THV thrombosis in 85% ofcases, as determined by follow-up CT and TEE.

More recently, investigators at separate sites re-ported similar rates of THV leaflet thickening post-TAVR using CT. Seventy recipients of the Lotus THVsystem (Boston Scientific, Marlborough, Massachu-setts) underwent serial CT at multiple time pointspost-TAVR out to 3 years (48). The incidence of HALTby 1 year post-TAVR was 15.7%, with a variable timeof onset. None of the patients who developed HALTwere receiving OAC at the time of detection. Althoughtransvalvular gradients appeared to rise significantlyhigher in the HALT group compared with the no-HALT group, there were no clinical events attribut-able to HALT in this series. A separate cohort of70 TAVR recipients revealed a similar 1-year post-TAVR HALT cumulative incidence of 14.2% (49),again without significant clinical sequelae.

FIGURE 4 Four-Dimensional Computed Tomographic Images of Subclinical Bioprosthetic Valve Thrombosis

Two-dimensional computed tomographic (CT) images (grayscale images, A and D) and volume-rendered CT images (color) for a CoreValve

(A to C) and a Portico valve (D to F). Arrows depict hypoattenuating opacities during diastole and systole. Leaflets with reduced motion are

depicted by wedge-shaped or semilunar opacities during the cardiac cycle. Adapted and reprinted with permission from Makkar et al. (50).



2202

Makkar et al. (50) recently detailed the 4Dcomputed tomographic data in both surgical BPVsand THVs that have perhaps gained the most notori-ety. These data constitute a pooled analysis of a

clinical trial of patients undergoing TAVR and 2ongoing registries involving patients undergoingeither TAVR or SAVR, whereby those patients withevaluable computed tomographic data were

TABLE 3 Summary of Clinical and Subclinical Findings of Patients With Suspected Bioprosthetic Valve Thrombosis Post-Surgical Aortic Valve Replacement and

Post-Transcatheter Aortic Valve Replacement in Studies That Performed Systematic Computed Tomographic Evaluations

First Author (Ref. #)

TAVR SAVRMedian Time BetweenTAVR/SAVR and CT,

days

MTG(m) or PTG(p),mm Hg

Post-TAVR/SAVR DAPT,Suspected BPVT

Resolution With OAC,Suspected BPVT TIA/StrokeHALT RLM HALT RLM Discharge Follow-up

Leetma et al. (46) 4/140 (3) 5/140 (4) - - 91 (66-92) 14.2 (p) 19.2 (p) 3/5 (60) 4/5 (90) 0 (0)

Makkar et al. (50) 22/55 (40)* 37/177 (23) - 2/27 (7) PORTICO IDE: 32 (28-37)Pooled registries: 30�10

9.1 (m) 9.6 (m) 21/39 (54) 11/11 (100) 5 (13)

Pache et al. (45) 16/156 (10) 8 /156 (5) - - 5 (5-6) 8.2 (m) 12.8 (m) 10/16 (63) 4/4 (100) 0 (0)

Hansson et al. (47) 28/405 (7) - - 42 (25-59) 10 (m) 10 (m) 19/28 (68) - 2/17 (12)

Yanagisawaet al. (49)

10/70 (14)† - - - - z10 (m) z10 (m) 3/10 (30) - 0/10 (0)

Gooley et al. (48) 11/70 (16) 11/70 (16) - - - 12 (m) 21 (m) - - 0/11 (0)

Chakravartyet al. (51)

101/752 (13) 101/752 (13) 5/138 (4) 5/138 (4) 83 (33-281) 9.8 (m) 13.8 (m) 31/208 (15) 36/36 (100) 4/4 (4)

Values are n/N (%) or N (%), unless otherwise indicated. *Data from the PORTICO IDE trial in which all patients with RLM also had HALT (n ¼ 22). HALT was not reported for the 15 TAVR patients and 2 SAVRpatients with RLM in the pooled registries. †Incidence of HALT at 1 year post-TAVR. At 6 months post-TAVR, 7 patients (10%) had HALT.

DAPT ¼ dual-antiplatelet therapy; MTG ¼ mean transaortic gradient; PORTICO IDE ¼ Portico Re-Sheathable Transcatheter Aortic Valve System US IDE; PTG ¼ peak transaortic gradient; RLM ¼ reducedleaflet motion; SAVR ¼ surgical aortic valve replacement; TAVR ¼ transcatheter aortic valve replacement; other abbreviations as in Tables 1 and 2.


2203

described (n ¼ 55 within the clinical trial [mediantime from TAVR to CT 32 days]), n ¼ 132 within the 2registries [median time from TAVR or SAVR to CT86 days]). Reduced leaflet motion was observed in22 of 55 patients (40%) within the clinical trial and in 17of 132 patients (13%) within the 2 registries, coincidingwith HALT in each of these patients. Transaortic gra-dients across this population did not significantlydiffer compared with those patients demonstratingnormal BPV motion. Interestingly, reduced leafletmotion was demonstrable in both THV (14%) and sur-gical BPV (7%) recipients. In patients already receivingOAC, the incidence of reduced leaflet motion wassignificantly reduced. Furthermore, the incidence ofstroke or transient ischemic attack was significantlyhigher in those with reduced leaflet motion comparedwith those with normal leaflet motion (18% vs. 1%, p ¼0.01), although the absolute number of events was low(4 of 132 patients), rendering the clinical relevance ofthese findings somewhat inconclusive. In those pa-tients reevaluated with serial computed tomographicBPV imaging, OAC restored leaflet motion in all pa-tients (11 of 11), compared with 1 of 10 patients notreceiving OAC (p < 0.001).

Chakravarty et al. (51) recently reported novel in-sights gleaned from the CT findings from 890 patients(TAVR: n ¼ 752; SAVR: n ¼ 138) included in theRESOLVE (Assessment of Transcatheter and SurgicalAortic Bioprosthetic Valve Thrombosis and its Treat-ment with Anticoagulation; NCT02318342) and SA-VORY (Subclinical Aortic Valve BioprosthesisThrombosis Assessed by 4D CT; NCT02426307) reg-istries. The CT examinations were performed at amedian of 83 days post-valve implantation. Subclin-ical valve thrombosis was diagnosed in 12% of

patients, being more frequent following TAVRcompared with SAVR (13% vs. 4%; p ¼ 0.001), and lessfrequent in those patients receiving anticoagulantscompared with those receiving antiplatelet therapy(4% vs. 15%; p < 0.001). No differences in HALT orRLM were observed in those prescribed single versusdual antiplatelet therapy. All cases of HALT and RLMwere hemodynamically subclinical (mean gradients inall cases being <20 mm Hg). Although no differencesin stroke rates were observed, the likelihood of apost-procedural TIA was significantly greater in thosewith HALT/RLM compared with patients withoutthese imaging findings. In all cases of HALT/RLM,systemic anticoagulation with warfarin (or a NOAC)resolved these abnormalities, with no apparent dif-ferences in efficacy between warfarin and NOACs.These findings will require further prospectiveconfirmation in ongoing trials evaluating newer gen-eration THV platforms that encompass embedded 4DCT imaging, prior to impacting clinical guidelines.However, one could speculate on the need for dual-antiplatelet therapy post-TAVR given the apparentlack of differences in valve motion abnormalitiescompared with single antiplatelet therapy.

Collectively, these data highlight the exquisitelysensitive nature of 4D CT in assessing BPV leafletmotion (and possibly early subclinical leaflet throm-bosis) much earlier than any potential apparenthemodynamic changes detectable with transtho-racic echocardiography. However, the anatomic-functional-clinical correlations of these observationsremain uncertain, especially over time. It is never-theless conceivable, following confirmation in largerprospective datasets, that early detection of subclini-cal nonobstructive thrombosis could prove useful in

https://clinicaltrials.gov/ct2/show/NCT02318342




2204

enabling preventive therapy (i.e., a short course ofOAC therapy). However, routine CT for BPV screeningassessment cannot presently be justified. Questionsalso remain regarding the optimal timing of post-valveimplantation imaging, as well as the issue of dealingwith the not so infrequent occurrence of image arti-facts (up to 30% incidence) (52). In light of theseemerging data regarding subclinical thrombosis, andin light of the controversy regarding the optimal defi-nition of THV degeneration compared with the tradi-tional definitions used in published surgical reports(which mandate the need for reoperation or patho-logical examination for an official diagnosis of valvethrombosis), we propose that clinicians adopt a prag-matic approach for assessing THV thrombosis anddegeneration. This would include heightened clinicaland echocardiographic surveillance in those patientsdeemed to harbor risk factors for BPVT and THVdegeneration (i.e., PPM, large sized THVs, valve-in-valve procedures), qualification with CT when thesuspicion arises of an underlying thrombotic process(especially when echocardiography fails to provideclarity regarding diagnosis), and prompt initiation ofanticoagulation when echocardiographic and CTcharacteristics suggestive of BPVT arise.

TREATMENT OF BPVT

OBSTRUCTIVE AND NONOBSTRUCTIVE RIGHT- AND

LEFT-SIDED BPVT. In general, the treatment strategyfor BPVT depends on its mode of clinical presentation;the patient’s hemodynamic status, presenceor absence of BPV obstruction, and valve location.Conventional treatment options include surgery,fibrinolysis, and anticoagulation (32). For non-obstructive left-sided BPVTs that are large (>5 mm),mobile, and pedunculated thrombi, surgery could beconsidered if intravenous heparin fails to resolve thesefeatures. For small (<5 mm) thrombi, medical therapywith OAC is usually the preferred option. In cases ofleft-sided obstructive thrombi, surgery or fibrinolysiscould each be considered. Surgical mortality hastraditionally related directly to the patient’s functionalstatus: 4% for patients in New York Heart Associationclass I versus 18% in those with class IV symptoms (32).Fibrinolysis appears efficacious in about 80% of occa-sions, yet carries an approximately 10% to 15% mor-tality ratewith a 12% rate of systemic embolism (53,54).As such, fibrinolysis is nowadays considered a second-line therapy reserved for patients with surgical con-traindications. However, fibrinolysis is usuallyconsidered acceptable for obstructed tricuspid or pul-monary BPVs, with acceptable complication rates, andsurgery is reserved for fibrinolytic failure (55).

THV thrombosis has been usually managed withOAC therapy, with significant clinical and hemody-namic improvement and normalization of trans-valvular gradients following 2 months of OAC therapyin the vast majority of cases (20,21). The duration ofOAC following BPVT remains unclear and should bedetermined on an individual basis, taking intoconsideration the risks for bleeding events versusrecurrent valve thrombosis. Long-term OAC therapymay be the preferred strategy, but in the event of OACcessation, antiplatelet therapy and frequent echo-cardiographic surveillance should probably berecommended.

Although beyond the scope of this review to pre-sent in detail, in the current era of THV therapies, ingeneral, transcatheter valve-in-valve procedures nowrepresent a viable option for treating patients withobstructive BPVs across all native valve positions.Figure 5 provides a proposed treatment algorithmfor patients presenting with various manifestationsof BPVT.

PREVENTING BPVT

POST-PROCEDURAL THROMBOPROPHYLAXIS FOLLOWING

SAVR. Table 4 summarizes current societal antith-rombotic guideline recommendations following sur-gical or transcatheter BPV implantation (19,56–59).Following surgical BPV implantation, 3 months ofOAC is generally recommended (irrespective of valveposition), during which time appropriate endotheli-alization of the BPV surface generally occurs.Following the initial 3-month period post-SAVR,guidelines tend to recommend the indefinite contin-uation of at least aspirin. However, with respectto aortic BPVs, treatment guidelines in generalremain inconclusive. This stems from the lack of anappropriately designed randomized controlled trialevaluating the nature, intensity, and duration ofantithrombotic therapies following aortic BPV im-plantation. Within the large Society of Thoracic Sur-geons Adult Cardiac Surgery Database, from 2004 to2006, in more than 25,600 patients $65 years of ageacross nearly 800 U.S. hospitals, Brennan et al. (60)uncovered lower adjusted death and embolic eventsin those prescribed both aspirin and warfarin withinthe first 3 months post-SAVR (with BPVs) comparedwith aspirin only, with the cost of an added bleedingrisk. Interestingly, an aspirin-only versus warfarin-only strategy was equally efficacious in terms ofreducing mortality and thromboembolic complica-tions, outlining the synergistic benefit of inhibitingplatelets, as well as the coagulation cascade. A similarbenefit of warfarin was demonstrated within the first

FIGURE 5 Treatment Algorithm for Patients Presenting With Bioprosthetic Valve Thrombosis

Bioprosthetic Valve Thrombosis (BPVT)

Right-sided BPVT Left-sided BPVT

Clinically unstable(NYHA III-IV or cardiogenic shock)

Clinically unstable(NYHA III-IV or cardiogenic shock)

Recent pulmonary embolism Recent systemic embolism

Clinically stable(NYHA I-II)

Clinically stable(NYHA I-II)

Fibrinolysis

FibrinolysisOAC in therapeutic range andno subtherapeutic INR within

the past weeks/months

OAC in therapeutic range andno subtherapeutic INR within

the past weeks/months

Thrombus <5-10 mm or 0.8 cm2

Absence of left atrial thrombus

IV Heparin +Start or intensify OAC

IV Heparin +Start or intensify OAC

Consider transcathetervalve-in-valve therapy

Continue OAC + Follow-up Continue OAC + Follow-up

Success Success Success Success Success Success

Success

Success

SurgeryConsider transcathetervalve-in-valve therapy

(if thrombus <5 mm or 0.8 cm2;consider using EPD)

Surgery

High surgicalrisk

High surgicalrisk

Failure

Failure

FailureFailure

No

No

No

YesNo

No

No

No

Yes

Yes

Yes

Yes

Yes

Yes

A proposed management approach for patients with both right- and left-sided bioprosthetic valve thrombosis (BPVT). EPD ¼ embolic protection device;

IV ¼ intravenous; NYHA ¼ New York Heart Association; OAC ¼ oral anticoagulation.


2205

3 months post–surgical BPV implantation within aDanish national registry (61). Those patients whodiscontinued warfarin within 6 months post-SAVRdemonstrated higher cardiovascular death rates.However, this analysis was limited by the lack ofinformation on concurrent aspirin use and by signif-icant confounding factors that were unable to beassessed to better understand why patients notreceiving warfarin demonstrated bleeding rates morethan twice those of patients receiving warfarin.

As such, the requirement for and duration of OACfollowing surgical BPV implantation currently re-mains contentious and highly variable (62). In theirrecent meta-analysis, Riaz et al. (63) attempted toclarify this issue by evaluating the safety and use ofOAC (vs. either aspirin or no OAC or antiplatelettherapy) following bioprosthetic SAVR. The in-vestigators included 13 studies comprising 6,431 cases

versus 18,210 control subjects. Most data stemmedfrom observational cohorts, across a variety of set-tings with differing operative management and clin-ical follow-up, and with significant heterogeneitynoted among included studies. The investigatorsfound that warfarin use post-SAVR using a BPVassociated with a significantly elevated the risk forbleeding (odds ratio: 1.96; 95% confidence interval:1.25 to 3.08; p <0.0001) compared with aspirin orplacebo. With regard to composite primary outcomevariables (risk for venous thromboembolism, stroke,or transient ischemic attack) at 3 months, no signifi-cant difference was observed with warfarin (oddsratio: 1.13; 95% confidence interval: 0.82 to 1.56;p ¼ 0.67), and OAC use was not shown to improveclinical outcomes beyond 3 months post-SAVR(odds ratio: 1.12; 95% confidence interval: 0.80 to1.58; p ¼ 0.79). These data are contrary to current

TABLE 4 Current Antithrombotic Therapeutic Recommendations for Bioprosthetic Valve Implantation

ACC/AHA Guidelines (19)ESC/EACTS

Guidelines (57) ACCP Guidelines (56)ACCF/AATS/SCAI/STSExpert Consensus (59)

CCS PositionStatement (58)

SAVRMVR or repair

ASA 75–100 mg/d lifelong(IIa B) þ VKA (INR of 2.5)for the first 3 months (IIa Cfor MVR or repair; IIb B forSAVR)

Low-dose ASA (IIa C) or VKA(IIb C) for the first3 months post-SAVR

VKA for the first 3 monthsafter MVR or repair (IIa C)

ASA (50–100 mg/day) overVKA in the first 3 monthspost-SAVR (grade 2C)

VKA (INR of 2.5) for the first3 months post-MVR

ASA over no therapy after3 months in all cases(grade 2C)

TAVR ASA 75–100 mg/daylifelong þ clopidogrel75 mg/day for the first6 months post-TAVR(IIb C)

DAPT (duration unspecified)In setting of OAC, avoid triple

therapy and use warfarinwith either ASA orclopidogrel

DAPT over VKA therapy andover no antiplatelettherapy in the first 3months (grade 2C)

IV heparin with an ACT goalof 300 s during theprocedure

DAPT for 3–6 months, thenASA 81 mg indefinitely

In setting of OAC, continueASA, but notclopidogrel

DAPT for 1–3 months,then ASA 81 mgindefinitely

In setting of OAC, avoidtriple therapy

AATS ¼ American Association for Thoracic Surgery; ACC ¼ American College of Cardiology; ACCP ¼ American College of Chest Physicians; ACT ¼ activated clotting time; AHA ¼ American Heart Association;ASA ¼ aspirin; CCS ¼ Canadian Cardiovascular Society; EACTS ¼ European Association for Cardio-Thoracic Surgery; ESC ¼ European Society of Cardiology; IV ¼ intravenous; MVR ¼ mitral valve replacement;SCAI ¼ Society for Cardiovascular Angiography and Interventions; STS ¼ Society of Thoracic Surgeons; VKA ¼ vitamin K antagonist; other abbreviations as in Tables 1 and 3.



2206

clinical guideline recommendations, highlighting theurgent need for prospective studies.

PERI- AND POST-PROCEDURAL THROMBOPROPHYLAXIS

FOLLOWING TAVR. Although now a well-establishedtreatment option in intermediate- and high-risksurgical patients with severe aortic stenosis, adjunc-tive pharmacological strategies for optimizing TAVRoutcomes and minimizing thromboembolic complica-tions remain under investigation. Periprocedurally,intravenous heparin anticoagulation 100 units/kgfollowed by additional boluses to maintain an acti-vated clotting time (ACT) of $250 s has been the mostcommon regime. A single-center retrospective anal-ysis of more than 360 patients undergoing TAVRdemonstrated significantly lower 30-day major andlife-threatening bleeding rates according to baselineachieved ACT-guided therapy compared withbody weight–adjusted periprocedural heparin-guidedtherapy (a non-ACT-guided approach) (64). However,lower ACT levels (<250 s) have been associated withthe near systematic presence of thromboembolicmaterial captured within an embolic protection device(65). Therefore, future studies will need to focus onidentifying the optimal periprocedural anti-coagulation strategy that associates with lowerthromboembolic rates, while minimizing acutebleeding risk. In the BRAVO-3 (Effect of Bivalirudin onAortic Valve Intervention Outcomes 3) trial random-izing more than 800 TAVR recipients to either biva-lirudin (a direct thrombin inhibitor) or heparin, nosignificant differences were observed in in-hospitalmajor bleeding rates (6.9% vs. 9.0%, p ¼ 0.27, respec-tively) or 30-day net adverse cardiovascular events(14.4% vs. 16.1%, p ¼ 0.35) (66). Unlike during primary

percutaneous coronary intervention, these data likelysuggest no added benefit of direct thrombin inhibitionover heparin for lowering bleeding rates during TAVR.

Current post-TAVR antithrombotic recommenda-tions have essentially been empirically extrapolatedfrom the strategies adopted for percutaneouscoronary intervention. Nevertheless, current U.S.guidelines recommend indefinite aspirin therapy and3 to 6 months of DAPT post-TAVR with aspirin andclopidogrel (59), whereas Canadian guidelines differslightly by recommending post-TAVR DAPT for 1 to3 months (Table 4) (58). Table 5 summarizes clinicaltrials currently under way investigating a variety ofantithrombotic regimens post-TAVR. The ongoingARTE (Aspirin Versus Aspirin þ Clopidogrel FollowingTranscatheter Aortic Valve Implantation) pilot trial istesting the hypothesis of single antiplatelet therapyversus DAPT post-TAVR in patients not requiringanticoagulation. Small-scale analyses have evaluatedthe efficacy of DAPT versus monoplatelet therapypost-TAVR. Although the focus of these trials may notspecifically relate to BPVT, collectively these resultshave demonstrated that post-TAVR DAPT fails toincrementally reduce stroke rates, while increasingbleeding rates (67–69), findings that were recentlycorroborated in meta-analyses (70,71). The REAC-TAVI (Platelet Reactivity After TAVI) trial is a small-scale mechanistic study to evaluate the utility of3 months of ticagrelor versus DAPT in those withelevated baseline on-treatment platelet reactivity.Follow-up platelet reactivity will be measured at3 months post-randomization. The POPular-TAVI(Antiplatelet Therapy for Patients Undergoing Trans-catheter Aortic Valve Implantation) trial will assessfreedom from bleeding across 2 patient subsets (those

TABLE 5 Currently Registered Ongoing Clinical Trials Testing Antithrombotic Therapies During Bioprosthetic Valve Implantation

Trial Design n Intervention Therapies Tested Primary Endpoints

ARTE (NCT01559298) Randomized (pre-TAVR) 300 TAVR ASA (80 mg/day) for 6 months vs.DAPT for 3 months, followed byongoing ASA (80 mg/day)

MACE: all-cause death, MI, stroke/TIA at 30 days and 1 yr

Safety: life-threatening or majorbleeding at 30 days and 1 yr

ATLANTIS (NCT02664649) 2 strata, 1:1 randomizationper stratum

1,509 TAVR Stratum 1 (indication for OAC): standardof care vs. apixaban 5 mg bid for6 months

Stratum 2 (no indication for OAC):standard of care—DAPT/SAPT vs.apixaban 5 mg bid for 6 months

MACE: all-cause death, MI, stroke/TIA/systemic embolism,intracardiac or bioprostheticthrombus, DVT/PE

Safety: major bleeding

AVATAR (NCT02735902) Randomized (post-TAVR) 170 TAVR VKA (INR of 2–3) for 12 months vs.VKA (INR of 2–3) þ ASA (75–100mg/day) for 12 months

Composite outcome: death from anycause, MI, stroke, valvethrombosis, and hemorrhage $2as defined by the VARC-2 scale

GALILEO (NCT02556203) PROBE1:1 randomization

1,520 TAVR Rivaroxaban 10 mg/dayþ ASA 75–100mg/day for 3 months, thenrivaroxaban 10 mg/day for 12–24months vs. DAPT for 3 months, thenASA 75–100mg/day for 12–24months

MACE: all-cause death; stroke; MI;valve thrombosis; PE; DVT;systemic embolism

Safety: life-threatening, disabling, ormajor bleeding

POPular-TAVI (NCT02247128) RandomizedCohort A: TAVR patients with

no indication for OACCohort B: TAVR patients with

an indication for OAC

1,000 TAVR Cohort A: ASA (<100 mg/day) vs. DAPTfor 3 months, then continue ASA(<100 mg/day) until 12-monthperiod

Cohort B: warfarin (target INR of 2) vs.clopidogrel 75 mg/day for3 months þ warfarin (target INRof 2), then continue warfarin alonethrough 12-month period

Safety: freedom from all bleedingcomplications

Coprimary endpoint: freedom fromnon-procedure-related bleedingcomplications (defined accordingto BARC and VARC)

REAC-TAVI (NCT02224066) Randomized on the basis ofon-treatment PRU

60 TAVR PRU 208: ticagrelor 90 mg bid vs.DAPT for 3 months

PRU <208: DAPT for 3 months

Suppression of residual plateletreactivity (assessed byVerifyNow P2Y12 assay)measured at 3 months

Frequency of ReducedLeaflet Motion AfterSAVR and TAVR(NCT02696226)

Pilot, randomized(post-induction ofanesthesia)

100 TAVR/SAVR TAVR patients: warfarin (INR 2–3) þclopidogrel 75 mg/day for 3 monthsfollowed by DAPT for 3 months andASA 81 mg/day indefinitelythereafter vs. DAPT for 6 monthsfollowed by ASA 81 mg/dayindefinitely

SAVR patients: warfarin (INR of 2–3) for3months followedbyASA 81mg/dayindefinitely vs. ASA indefinitely

Impact of periproceduralanticoagulation on the frequencyof reduced leaflet motion on 4DCT after SAVR and TAVR at 4–6weeks post-procedure

Inflammation andThrombosis in PatientsWith Severe AorticStenosis After TAVR(NCT02486367)

Randomized, open-label 60 TAVR Clopidogrel 600-mg loading dosefollowed by 75 mg/day vs.ticagrelor 180-mg loading dosefollowed by 90 mg bid for 1 month

Primary endpoint: platelet reactivity,expressed as PRU using theVerifyNow system

Key secondary endpoint: percentageof inflammatory (CD14þ CD16þ)monocytes as a proportion oftotal monocytes measured usingflow cytometry

ARTE ¼ Aspirin Versus Aspirin þ ClopidogRel Following Transcatheter Aortic Valve Implantation; ATLANTIS ¼ Anti-Thrombotic Strategy After Trans-Aortic Valve Implantation for Aortic Stenosis;AVATAR ¼ Anticoagulation Alone Versus Anticoagulation and Aspirin Following Transcatheter Aortic Valve Interventions (1:1); BARC ¼ Bleeding Academic Research Consortium; DVT ¼ deep vein thrombosis;GALILEO ¼ Global Study Comparing a Rivaroxaban-Based Antithrombotic Strategy to an Antiplatelet-Based Strategy After Transcatheter Aortic Valve Replacement to Optimize Clinical Outcomes;MACE ¼ major adverse cardiovascular event; MI ¼ myocardial infarction; PE ¼ pulmonary embolism; POPular-TAVI ¼ Antiplatelet Therapy for Patients Undergoing Transcatheter Aortic Valve Implantation;PROBE¼ prospective, randomized, open-label, blinded endpoint evaluation; PRU ¼ platelet reactivity units; REAC-TAVI¼ Platelet Reactivity After TAVI: A Multicenter Pilot Study; SAPT ¼ single-antiplatelettherapy; VARC ¼ Valve Academic Research Consortium; other abbreviations as in Tables 1 to 4.


2207

either with or without an indication for OAC) byevaluating shorter periods (<3 months post-TAVR) ofDAPT versus aspirin, followed by longer term aspirinpost-TAVR, or in those with an indication for OAC,the addition of clopidogrel (vs. placebo) for 3 monthspost-TAVR, followed by indefinite warfarin therapy.

Several studies are currently evaluating the use ofNOACs in TAVR recipients for thromboembolic pro-phylaxis. Apixaban (an oral direct thrombin/factor Xa

inhibitor) compared with warfarin reduced car-dioembolic events, while reducing bleeding rates inpatients with nonvalvular AF (72). These properties,coupled with its similar safety profile to aspirin inpatients with AF who are unsuitable for warfarin, hasled to ATLANTIS (Anti-Thrombotic Strategy AfterTrans-Aortic Valve Implantation for Aortic Stenosis),whereby apixaban versus DAPT is being tested in1,500 patients. Furthermore, GALILEO (Global Study





https://clinicaltrials.gov/ct2/show/NCT02247128?term=NCT02247128&rank=1




CENTRAL ILLUSTRATION Bioprosthetic Valve Thrombosis: Prevalence, Diagnostic Criteria, Current TreatmentOptions, and Ongoing Controversies

Clinical Presentation Diagnostic Criteria Treatment Options

Broad clinical spectrumof Bioprosthetic ValveThrombosis (BPVT):• Incidental or asymptomatic

• Systemic or pulmonaryembolism

• Cardiogenic shock

• Peak incidence 1–2 yearspost valve implantation

• Incidence - obstructive/symptomatic: ≤1%- subclinical BPVT: up to 14% at 1 year

3 echocardiographic features:

50% increase intransvalvular gradient

Increased cusp thickness(>2 mm)

Abnormal cusp mobility

SurgeryFor patients with:• Obstructed left-sided BPVT • Hemodynamic instability• Decompensated heart failure

ThrombolysisFor • High risk surgical candidates• Right-sided obstructive BPVT

Transcatheter valve-in-valveFor non-surgical patients, consider use of embolic protection device if large thrombotic burden

Systemic oral anticoagulationFor patients with minimallyobstructive and/or small(< 5mm or < 0.8cm2) thrombi

on routine valve imaging

Puri, R. et al. J Am Coll Cardiol. 2017;69(17):2193–211.

These are depicted with accompanying pathological, echocardiographic, and 4-dimensional (4D) computed tomographic images of bioprosthetic valve (BPV)

thrombosis. BPVT ¼ bioprosthetic valve thrombosis; CT ¼ computed tomography; NOAC ¼ novel oral anticoagulant agent; SAVR ¼ surgical aortic valve replacement;

TAVR ¼ transcatheter aortic valve replacement.



2208

Comparing a Rivaroxaban-Based AntithromboticStrategy to an Antiplatelet-Based Strategy AfterTranscatheter Aortic Valve Replacement to OptimizeClinical Outcomes) is evaluating the efficacy of low-dose rivaroxaban versus DAPT in TAVR patientswithout AF.

CONCLUSIONS

The advent of TAVR, and recent findings ofsubclinical leaflet thrombosis in both transcatheterand surgically implanted BPVs, has spurred intenseinterest in the thrombogenic profiles of BPVs per se,particularly with regard to defining optimal thera-peutic strategies for preventing thromboemboliccomplications. Meanwhile, the lack of prospective,randomized data following surgical BPV implanta-tion has also led some to question the role, efficacy,and duration of empirical systemic OAC followingsurgical BPV implantation, particularly withinthe aortic position. Although symptomatic BPVT

is considered a relatively rare phenomenon, accu-mulating evidence suggests that this phenomenonmay be more prevalent than previously thought,particularly with THVs. This perhaps relates tothe current lack of a harmonized definition ofBPVT. Clinicians should have a heightened clinicalawareness of this phenomenon, especially in BPVrecipients who present with dyspnea and/orrising transprosthetic gradients at any time post-implantation (Central Illustration). This shouldprompt echocardiographic (and possibly computedtomographic) imaging to evaluate bioprostheticleaflet morphology and motion (Central Illustration).In the absence of clinically overt BPV obstruction,OAC therapy is recommended, with follow-up echo-cardiography scheduled 1 to 2 months followingtreatment initiation (Central Illustration). Althoughexquisitely sensitive in detecting leaflet anomalies,the role of routine 4D CT post-BPV insertion remainsuncertain until further research identifies clinicalutility. However, CT could play an important role in


2209

the diagnosis of BPVT in symptomatic patients orthose presenting with rising transprosthetic gradi-ents. It also remains to be seen whether subclinicalBPVT detected with 4D CT will serve as surrogateendpoint for clinical trials testing novel antith-rombotic strategies, given the unknown relationshipbetween subclinical BPVT and clinical events. Therole of routine OAC following surgical BPV implan-tation within the aortic position remains unclear.Nevertheless, until prospective studies are under-taken, clinical guidelines suggesting a limited period(1 to 3 months) of OAC with or without aspirin shouldprobably be adhered to. Following TAVR, a period ofDAPT is also advisable. Thereafter, the precise long-term antithrombotic regimen should be dictated bypatient-specific comorbidities relating to boththromboembolic and bleeding risk. Future studies

are required to better elucidate THV durability andidentify novel risk factors for BPVT, such as PPM,smaller-sized THVs, newer iteration transcatheterprocedures (i.e., valve-in-valve procedures), andTHVs implanted in nonaortic positions. A range ofclinical trials are currently under way that will bepivotal in informing us of the role of novel OACsfollowing TAVR.

ACKNOWLEDGMENT The authors thank MélanieCôté for her invaluable assistance in preparing thefigures.

ADDRESS FOR CORRESPONDENCE: Dr. Josep Rodés-Cabau, Quebec Heart and Lung Institute, Laval Uni-versity, 2725 Chemin Ste-Foy, Québec City QC G1V4G5, Canada. E-mail: [email protected].

RE F E RENCE S

1. Osnabrugge RLJ, Mylotte D, Head SJ, et al.Aortic stenosis in the elderly: disease prevalenceand number of candidates for transcatheteraortic valve replacement: a meta-analysis andmodeling study. J Am Coll Cardiol 2013;62:1002–12.

2. Brown JM, O’Brien SM, Wu C, Sikora JA,Griffith BP, Gammie JS. Isolated aortic valvereplacement in North America comprising 108,687patients in 10 years: changes in risks, valve types,and outcomes in the Society of Thoracic SurgeonsNational Database. J Thorac Cardiovasc Surg2009;137:82–90.

3. Siddiqui RF, Abraham JR, Butany J. Bio-prosthetic heart valves: modes of failure. Histo-pathology 2009;55:135–44.

4. Pibarot P, Dumesnil JG. Prosthetic heartvalves: selection of the optimal prosthesis andlong-term management. Circulation 2009;119:1034–48.

5. Arsalan M, Walther T. Durability of prosthesesfor transcatheter aortic valve implantation. NatRev Cardiol 2016;13:360–7.

6. Phan K, Tsai YC, Niranjan N, et al. Suturelessaortic valve replacement: a systematic review andmeta-analysis. Ann Cardiothorac Surg 2015;4:100–11.

7. Kheradvar A, Groves EM, Goergen CJ, et al.Emerging trends in heart valve engineering: PartII. Novel and standard technologies for aorticvalve replacement. Ann Biomed Eng 2015;43:844–57.

8. Martin C, Sun W. Comparison of transcatheteraortic valve and surgical bioprosthetic valvedurability: a fatigue simulation study. J Biomech2015;48:3026–34.

9. Regueiro A, Granada J, Dagenais F, Rodés-Cabau J. Transcatheter mitral valve replace-ment: insights from early clinical experienceand future challenges. J Am Coll Cardiol 2017[E-pub ahead of print].

10. Alavi SH, Groves EM, Kheradvar A. The effectsof transcatheter valve crimping on pericardialleaflets. Ann Thorac Surg 2014;97:1260–6.

11. Thubrikar MJ, Deck JD, Aouad J, Nolan SP. Roleof mechanical stress in calcification of aortic bio-prosthetic valves. J Thorac Cardiovasc Surg 1983;86:115–25.

12. Arsalan M, Mack MJ. Durability of devices:long-term results and clinical outcomes. Euro-Intervention 2015;11 Suppl W:W119–22.

13. Prandoni P, Pengo V, Boetto P, Zambon G,Menozzi L. Do malfunctioning bioprosthetic heartvalves represent a potential thrombogenic focus?Haemostasis 1985;15:337–44.

14. Pibarot P, Weissman NJ, Stewart WJ, et al.Incidence and sequelae of prosthesis-patientmismatch in transcatheter versus surgical valvereplacement in high-risk patients with severeaortic stenosis: a PARTNER trial cohort–A analysis.J Am Coll Cardiol 2014;64:1323–34.

15. Akins CW, Miller DC, Turina MI, et al. Guide-lines for reporting mortality and morbidity aftercardiac valve interventions. J Thorac CardiovascSurg 2008;135:732–8.

16. Egbe AC, Pislaru SV, Pellikka PA, et al. Bio-prosthetic valve thrombosis versus structuralfailure: clinical and echocardiographic predictors.J Am Coll Cardiol 2015;66:2285–94.

17. Daniel WG, Mügge A, Grote J, et al. Com-parison of transthoracic and transesophagealechocardiography for detection of abnormalitiesof prosthetic and bioprosthetic valves in themitral and aortic positions. Am J Cardiol 1993;71:210–5.

18. Pislaru SV, Pellikka PA, Schaff HV,Connolly HM. Bioprosthetic valve thrombosis:the eyes will not see what the mind does notknow. J Thorac Cardiovasc Surg 2015;149:e86–7.

19. Nishimura RA, Otto CM, Bonow RO, et al. 2014AHA/ACC guideline for the management of

patients with valvular heart disease: a report ofthe American College of Cardiology/AmericanHeart Association Task Force on Practice Guide-lines [published correction appears in J Am CollCardiol 2014;63:2489]. J Am Coll Cardiol 2014;63:e57–185.

20. Córdoba-Soriano JG, Puri R, Amat-Santos I,et al. Valve thrombosis following transcatheteraortic valve implantation: a systematic review. RevEsp Cardiol (Engl Ed) 2015;68:198–204.

21. Latib A, Naganuma T, Abdel-Wahab M, et al.Treatment and clinical outcomes of transcatheterheart valve thrombosis. Circ Cardiovasc Interv2015;8:e001779.

22. De Marchena E, Mesa J, Pomenti S, et al.Thrombus formation following transcatheteraortic valve replacement. J Am Coll Cardiol Intv2015;8:728–39.

23. Rodés-Cabau J. Transcatheter mitral valvereplacement with the Fortis valve: procedural andmidterm results of the first-in-man compassionateclinical use experience. Presented at: EuroPCR;May 20, 2015; Paris, France.

24. Jaffer IH, Fredenburgh JC, Hirsh J, Weitz JI.Medical device-induced thrombosis: what causes itand how can we prevent it? J Thromb Haemost2015;13 Suppl 1:S72–81.

25. Kopanidis A, Pantos I, Alexopoulos N,Theodorakakos A, Efstathopoulos E, Katritsis D.Aortic flow patterns after simulated implantationof transcatheter aortic valves. Hellenic J Cardiol2015;56:418–28.

26. Vahidkhah K, Barakat M, Abbasi M, et al. Valvethrombosis following transcatheter aortic valvereplacement: significance of blood stasis on theleaflets. Eur J Cardiothorac Surg. In press.

27. Head SJ, Mokhles MM, Osnabrugge RL, et al.The impact of prosthesis-patient mismatch onlong-term survival after aortic valve replacement:a systematic review and meta-analysis of 34observational studies comprising 27,186 patients

mailto:[email protected]

http://refhub.elsevier.com/S0735-1097(17)36016-3/sref1




















































































































2210

with 133,141 patient-years. Eur Heart J 2012;33:1518–29.

28. Vincentelli A, Susen S, Le Tourneau T, et al.Acquired von Willebrand syndrome in aortic ste-nosis. N Engl J Med 2003;349:343–9.

29. Yoshida K, Tobe S, Kawata M, Yamaguchi M.Acquired and reversible von Willebrand diseasewith high shear stress aortic valve stenosis. AnnThorac Surg 2006;81:490–4.

30. Ducci A, Tzamtzis S, Mullen MJ, Burriesci G.Hemodynamics in the Valsalva sinuses aftertranscatheter aortic valve implantation (TAVI).J Heart Valve Dis 2013;22:688–96.

31. Flameng W, Herregods MC, Vercalsteren M,Herijgers P, Bogaerts K, Meuris B. Prosthesis-patient mismatch predicts structural valvedegeneration in bioprosthetic heart valves. Circu-lation 2010;121:2123–9.

32. Roudaut R, Serri K, Lafitte S. Thrombosis ofprosthetic heart valves: diagnosis and therapeuticconsiderations. Heart 2007;93:137–42.

33. Del Trigo M, Muñoz-Garcia AJ,Wijeysundera HC, et al. Incidence, timing, andpredictors of valve hemodynamic deteriorationafter transcatheter aortic valve replacement:multicenter registry. J Am Coll Cardiol 2016;67:644–55.

34. Cáceres-Lóriga FM, Pérez-López H,Santos-Gracia J, Morlans-Hernandez K. Pros-thetic heart valve thrombosis: pathogenesis,diagnosis and management. Int J Cardiol2006;110:1–6.

35. Cremer PC, Rodriguez LL, Griffin BP, et al.Early bioprosthetic valve failure: mechanistic in-sights via correlation between echocardiographicand operative findings. J Am Soc Echocardiog2015;28:1131–48.

36. Schoen FJ, Gotlieb AI. Heart valve health,disease, replacement, and repair: a 25-year car-diovascular pathology perspective. CardiovascPath 2016;25:341–52.

37. Huygens SA, Mokhles MM, Hanif M, et al.Contemporary outcomes after surgical aorticvalve replacement with bioprostheses andallografts: a systematic review and meta-analysis. Euro J Cardiothorac Surg 2016;50:605–16.

38. Puvimanasinghe JP, Steyerberg EW,Takkenberg JJ, et al. Prognosis after aortic valvereplacement with a bioprosthesis: predictionsbased on meta-analysis and microsimulation. Cir-culation 2001;103:1535–41.

39. Butnaru A, Shaheen J, Tzivoni D, Tauber R,Bitran D, Silberman S. Diagnosis and treatment ofearly bioprosthetic malfunction in the mitral valveposition due to thrombus formation. Am J Cardiol2013;112:1439–44.

40. Oliver JM, Gallego P, Gonzalez A,Dominguez FJ, Gamallo C, Mesa JM. Bioprostheticmitral valve thrombosis: clinical profile, trans-esophageal echocardiographic features, andfollow-up after anticoagulant therapy. J Am SocEchocardiog 1996;9:691–9.

41. Pislaru SV, Hussain I, Pellikka PA, et al. Mis-conceptions, diagnostic challenges and treatment

opportunities in bioprosthetic valve thrombosis:lessons from a case series. Eur J Cardiothorac Surg2015;47:725–32.

42. Brown ML, Park SJ, Sundt TM, Schaff HV. Earlythrombosis risk in patients with biologic valves inthe aortic position. J Thorac Cardiovasc Surg 2012;144:108–11.

43. Dohi M, Doi K, Yaku H. Early stenosis of anaortic porcine bioprosthesis due to thrombosis:case report and literature review. J Thorac Car-diovasc Surg 2015;149:e83–6.

44. Jander N, Sommer H, Pingpoh C, et al. Theporcine valve type predicts obstructive thrombosisbeyond the first three postoperative months inbioprostheses in the aortic position. Int J Cardiol2015;199:90–5.

45. Pache G, Schoechlin S, Blanke P, et al. Earlyhypo-attenuated leaflet thickening in balloon-expandable transcatheter aortic heart valves. EurHeart J 2016;37:2263–71.

46. Leetmaa T, Hansson NC, Leipsic J, et al. Earlyaortic transcatheter heart valve thrombosis:diagnostic value of contrast-enhanced multi-detector computed tomography. Circ CardiovascInterv 2015;8:e001596.

47. Hansson NC, Grove EL, Andersen HR, et al.Transcatheter aortic valve thrombosis: incidence,predisposing factors, and clinical implications.J Am Coll Cardiol 2016;68:2059–69.

48. Gooley R. Subclinical leaflet thickening in anovel mechanical expanded TAVI device. Pre-sented at: PCR London Valves; September 20,2016; London, United Kingdom.

49. Yanagisawa R, Hayashida K, Yamada Y, et al.Incidence, predictors, and mid-term outcomes ofpossible leaflet thrombosis after TAVR. J Am CollCardiol Img 2017;10:1–11.

50. Makkar RR, Fontana G, Jilaihawi H, et al.Possible subclinical leaflet thrombosis in bio-prosthetic aortic valves. N Engl J Med 2015;373:2015–24.

51. Chakravarty T, Sondergaard L, Friedman J, et al.Subclinical leaflet thrombosis in surgical and trans-catheter bioprosthetic aortic valves: an observa-tional study. Lancet 2017 [E-pub ahead of print].

52. Bax JJ, Delgado V, Prendergast B. Doescomputed tomography detect bioprosthetic aorticvalve thrombosis? New findings, new questions?Eur Heart J 2016;37:2272–5.

53. Lengyel M, Fuster V, Keltai M, et al.Guidelines for management of left-sidedprosthetic valve thrombosis: a role forthrombolytic therapy. Consensus Conferenceon Prosthetic Valve Thrombosis. J Am CollCardiol 1997;30:1521–6.

54. Özkan M, Gündüz S, Biteker M, et al. Com-parison of different TEE-guided thrombolyticregimens for prosthetic valve thrombosis: theTROIA trial. J Am Coll Cardiol Img 2013;6:206–16.

55. Shapira Y, Sagie A, Jortner R, Adler Y,Hirsch R. Thrombosis of bileaflet tricuspid valveprosthesis: clinical spectrum and the role ofnonsurgical treatment. Am Heart J 1999;137:721–5.

56. Whitlock RP, Sun JC, Fremes SE, Rubens FD,Teoh KH. Antithrombotic and thrombolytic ther-apy for valvular disease: antithrombotic therapyand prevention of thrombosis, 9th ed: AmericanCollege of Chest Physicians evidence-based clin-ical practice guidelines. Chest 2012;141:e576S–600S.

57. Authors/Task Force Members, Vahanian A,Alfieri O, et al. Guidelines on the managementof valvular heart disease (version 2012): JointTask Force on the Management of ValvularHeart Disease of the European Society of Car-diology and the European Association forCardio-Thoracic Surgery (EACTS). Eur Heart J2012;33:2451–96.

58. Webb J, Rodés-Cabau J, Fremes S, et al.Transcatheter aortic valve implantation: a Cana-dian Cardiovascular Society position statement.Can J Cardiol 2012;28:520–8.

59. Holmes DR Jr., Mack MJ, Kaul S, et al. 2012ACCF/AATS/SCAI/STS expert consensus documenton transcatheter aortic valve replacement. J AmColl Cardiol 2012;59:1200–54.

60. Brennan JM, Edwards FH, Zhao Y, et al. Earlyanticoagulation of bioprosthetic aortic valves inolder patients: results from the Society of ThoracicSurgeons Adult Cardiac Surgery National Data-base. J Am Coll Cardiol 2012;60:971–7.

61. Mérie C, Køber L, Skov Olsen P, et al. Associ-ation of warfarin therapy duration after bio-prosthetic aortic valve replacement with risk ofmortality, thromboembolic complications, andbleeding. JAMA 2012;308:2118–25.

62. Brennan JM, Alexander KP, Wallace A, et al.Patterns of anticoagulation following bio-prosthetic valve implantation: observations fromANSWER. J Heart Valve Dis 2012;21:78–87.

63. Riaz H, Alansari SAR, Khan MS, et al. Safetyand use of anticoagulation after aortic valvereplacement with bioprostheses: a meta-anal-ysis. Circ Cardiovasc Qual Outcomes 2016;9:294–302.

64. Bernelli C, Chieffo A, Montorfano M, et al.Usefulness of baseline activated clotting time-guided heparin administration in reducingbleeding events during transfemoral transcatheteraortic valve implantation. J Am Coll Cardiol Intv2014;7:140–51.

65. Van Mieghem NM, El Faquir N, Rahhab Z, et al.Incidence and predictors of debris embolizing tothe brain during transcatheter aortic valve im-plantation. J Am Coll Cardiol Intv 2015;8:718–24.

66. Dangas GD, Lefèvre T, Kupatt C, et al., for theBRAVO-3 Investigators. Bivalirudin versus heparinanticoagulation in transcatheter aortic valvereplacement: the randomized BRAVO-3 trial. J AmColl Cardiol 2015;66:2860–8.

67. Ussia GP, Scarabelli M, Mulè M, et al. Dualantiplatelet therapy versus aspirin alone inpatients undergoing transcatheter aortic valveimplantation. Am J Cardiol 2011;108:1772–6.

68. Stabile E, Pucciarelli A, Cota L, et al. SAT-TAVI (single antiplatelet therapy for TAVI) study:a pilot randomized study comparing double tosingle antiplatelet therapy for transcatheter





























































































































































































2211

aortic valve implantation. Int J Cardiol 2014;174:624–7.

69. Durand E, Blanchard D, Chassaing S, et al.Comparison of two antiplatelet therapy strategiesin patients undergoing transcatheter aortic valveimplantation. Am J Cardiol 2014;113:355–60.

70. Hassell ME, Hildick-Smith D, Durand E, et al.Antiplatelet therapy following transcatheter aorticvalve implantation. Heart 2015;101:1118–25.

71. Gandhi S, Schwalm JD, Velianou JL,Natarajan MK, Farkouh ME. Comparison of dual-antiplatelet therapy to mono-antiplatelet therapyafter transcatheter aortic valve implantation: sys-tematic review and meta-analysis. Can J Cardiol2015;31:775–84.

72. Connolly SJ, Eikelboom J, Joyner C, et al.,for the AVERROES Steering Committee andInvestigators. Apixaban in patients with

atrial fibrillation. N Engl J Med 2011;364:806–17.

KEY WORDS 4-dimensional computedtomography, heart valve prosthesis, surgicalaortic valve replacement, thromboembolism,thrombolytic therapy, transcatheter aorticvalve replacement





















Bioprosthetic Valve Thrombosis · 2017-05-30 · tiny within the context of ongoing evaluations of...

Documents

Transcript of Bioprosthetic Valve Thrombosis · 2017-05-30 · tiny within the context of ongoing evaluations of...