Transcatheter Valve-in-Valve Implantation for Failed ...pulmonic, and tricuspid positions. Despite some core similarities to transcatheter therapy of native valve dis-ease, V-in-V

Journal of the American College of Cardiology Vol. 58, No. 21, 2011© 2011 by the American College of Cardiology Foundation ISSN 0735-1097/$36.00

FOCUS ISSUE: STRUCTURAL HEART DISEASE Prosthetic Valve Failure

State-of-the-Art Paper

Transcatheter Valve-in-Valve Implantationfor Failed Surgical Bioprosthetic Valves

Ronen Gurvitch, MBBS,*† Anson Cheung, MD,* Jian Ye, MD,* David A. Wood, MD,*Alexander B. Willson, MBBS,* Stefan Toggweiler, MD,* Ronald Binder, MD,* John G. Webb, MD*

Vancouver, British Columbia, Canada; and Melbourne, Victoria, Australia

When bioprosthetic cardiac valves fail, reoperative valve replacement carries a higher risk of morbidity and mor-tality compared with initial valve replacement. Transcatheter heart valve implantation may be a viable alterna-tive to surgical aortic valve replacement for high-risk patients with native aortic stenosis, and valve-in-valve (V-in-V) implantation has been successfully performed for failed surgical bioprostheses in the aortic, mitral,pulmonic, and tricuspid positions. Despite some core similarities to transcatheter therapy of native valve dis-ease, V-in-V therapy poses unique clinical and anatomic challenges. In this paper, we review the challenges, se-lection criteria, techniques, and outcomes of V-in-V implantation. (J Am Coll Cardiol 2011;58:2196–209)© 2011 by the American College of Cardiology Foundation

Published by Elsevier Inc. doi:10.1016/j.jacc.2011.09.009

Despite the shorter durability of bioprosthetic cardiac valvescompared with mechanical prostheses, the former are oftenused to reduce thromboembolic risk and to avoid anticoag-ulation and the associated increased risk of bleeding (1,2).When bioprosthetic valves degenerate, reoperative valvereplacement is the current standard of care (3–5). However,patients requiring redo valvular surgery are generally at anincreased risk of adverse perioperative events. Factors thathave been shown to portend a higher reoperative riskinclude advanced age, renal failure, pulmonary disease,cognitive impairment, reduced ejection fraction, higherNew York Heart Association functional class, need forconcomitant bypass surgery, need for more than 1 reopera-tion, and reoperation for failed mitral valves (higher riskthan aortic) (3,5–12). Operative mortality for redo valvesurgery may be higher than 20% in such high-risk patients,whereas in low-risk patients, it may be �4% (8,12).

Transcatheter aortic valve replacement for native aorticstenosis has evolved as a viable alternative to open heartsurgery in patients with high or prohibitive surgical risk(13–20). As transcatheter aortic valve replacement evolvedand clinical outcomes improved, transcatheter treatment of

From the *Department of Cardiology and Cardiothoracic Surgery, St. Paul’s Hospital,University of British Columbia, Vancouver, British Columbia, Canada; and the†Department of Cardiology, The Royal Melbourne Hospital, Melbourne, Victoria,Australia. Drs. Cheung, Ye, and Webb are consultants to Edwards Lifesciences Inc.Dr. Wood is a consultant to St. Jude Medical and Edwards Lifesciences. All otherauthors have reported that they have no relationships relevant to the contents of this
paper to disclose.
Manuscript received September 12, 2011; accepted September 13, 2011.

degenerated bioprosthetic valves became possible usingvalve-in-valve (V-in-V) implantation, in which the newtranscatheter valve is inserted inside the degenerated bio-prosthesis. V-in-V implantation has been successfully per-formed in degenerated aortic, mitral, pulmonic, and tricus-pid bioprostheses as well as in pulmonary conduits. Avariety of percutaneous and minimally invasive surgicaltechniques have been used. This article reviews the chal-lenges, selection criteria, techniques, and what is known ofthe outcomes of this evolving therapy.Structural elements of bioprosthetic valves. Contempo-rary surgical bioprostheses generally incorporate leafletsderived from porcine valve leaflets or bovine pericardium(heterograft/xenograft). These are generally preserved inglutaraldehyde to attenuate calcific degeneration by cross-linking collagen fibers, reducing antigenicity, and prevent-ing remodeling of extracellular matrix (21,22). Humantissue (homograft) is less commonly used. For the lesscommon pulmonic position, valved conduits of various typesare often used.

Bioprosthetic valves may be stented or stentless. Stentedvalves are usually constructed of whole porcine aortic valvesor bovine pericardium, suspended from a support structuresuch as a stent or frame (Fig. 1). The support structures ofcurrent valves are composed of various alloys or plastics,designed to absorb forces on the leaflets during valvefunction and maintain structural integrity. The supportstructure is attached to a basal ring, which is covered by afabric sewing cuff. The basal ring may be circular or scallop
shaped. Newer valve designs often incorporate low-profile

2197JACC Vol. 58, No. 21, 2011 Gurvitch et al.November 15, 2011:2196–209 Valve-in-Valve Therapy for Failed Surgical Valves

or supra-aortic design features intended to optimize hemody-namic functioning (e.g., Mitroflow or Soprano valve, Sorin,Canada).

Stentless valves are sutured to the root in the position ofa native valve and eliminate the support stent/frame toimprove hemodynamic performance and durability (23,24).These may be of autograft, heterograft, or homograft origin.Although stentless valves tend to have better laminar flowand lower post-operative gradients, clinical outcomes havenot been superior to stented valves (25,26). Some stentlessvalves have been associated with exuberant calcification ofthe aortic root, posing a particular challenge for reoperation.Stentless valves pose unique challenges for V-in-V therapygiven the lack of a frame to anchor the new transcatheterheart valve (THV), and the absence of radiopaque markersto aid with positioning.Causes of bioprosthetic valve failure. Failure of biopros-thetic valves may occur due leaflet failure or nonleafletfailure or both. Leaflet failures are the result of leafletdegeneration (wear and tear or calcification) or leafletdestruction (due to endocarditis). Nonleaflet failures aretypically the result of pannus, thrombus, or paravalvularleaks. The outcomes of these failures may be stenosis,regurgitation, or a combination of both. In general, failuredue to calcification, pannus formation, or thrombus resultsin valvular stenosis, and failure due to leaflet destruction(endocarditis) or paravalvular leak results in regurgitation.In failure due to wear and tear, which is frequently associ-ated with calcification and abnormal leaflet coaptation,mixed stenosis/regurgitation is common. With current bio-

Figure 1 Schematic Diagram of Stented Bioprosthetic Valves

Valve leaflet (A), stent frame (B), and external sewing ring (C). The internal, oute

prosthetic valves, freedom fromfailure at 10 years is between 70%and 90% and at 15 years between40% and 70% (22,27–31). Com-mon risk factors for bioprostheticfailure include mitral valve posi-tion, younger age at implanta-tion, renal failure, and hyperparathyroidism (22,29–31).Bioprosthetic valve calcification is markedly accelerated inyounger patients, with failure rates as high as 40% at 4 yearsin patients younger than 30 years old (29). Not surprisingly,the lifetime risk of needing reoperation decreases with theincreasing age of the patient (29,32).

When bioprosthetic valves degenerate, calcific depositsare often located within the leaflet tissue, with particularpredisposition to areas of high leaflet stress (e.g., commis-sural and attachment points). Glutaraldehyde fixation treat-ments creating stable crosslinks between collagen aim torender the valve material inert, although residual glutaral-dehyde may also serve as calcium binding sites (22). Variousanticalcification treatments aim to reduce residual glutaral-dehyde or phospholipids (e.g., Medtronic AOS [MedtronicInc., Minneapolis, Minnesota] or Edwards ThermaFix[Edwards Lifesciences, Irvine, California]). Frequently,leaflet calcification and subsequent tears also result in somedegree of valvular regurgitation. Although calcification isresponsible for the majority of bioprosthetic valve failures,progressive collagen damage related to a number of immuneand atherosclerotic processes may also be contributory(33–35). Pannus formation results from growth of host

external diameters all represent different dimensions of surgical bioprostheses.

Abbreviationsand Acronyms

THV � transcatheter heartvalve

V-in-V � valve-in-valve

r, and

2198 Gurvitch et al. JACC Vol. 58, No. 21, 2011Valve-in-Valve Therapy for Failed Surgical Valves November 15, 2011:2196–209

tissue and is to some extent part of the normal healingreaction to prosthesis implantation. More than 60% ofexplanted valves show evidence of mild pannus at thetissue-valve interface. In those with severe pannus forma-tion, leaflet dysfunction can occur largely due to leafletstiffness and restricted mobility. An important distinctionmust be made between pannus and thrombus, with quali-tative and quantitative ultrasound intensity on transesoph-ageal echocardiography helping differentiate them (36).

In patients with bioprosthetic valve regurgitation, clini-cians must clarify whether it is transvalvular or paravalvularregurgitation. Although the former may be successfullytreated using V-in-V therapy, the latter is not suitable forsuch techniques. Patients with bioprosthetic paravalvular

Figure 2 The 3 Types of Widely Available Transcatheter System

Edwards SAPIEN valve (A) and Edwards Novaflex delivery system (with Edwards SAMedtronic CoreValve delivery system (D), Medtronic Melody valve (E), and Medtro

regurgitation who are deemed unsuitable for reoperationmay, however, benefit from other percutaneous techniquesto occlude the paravalvular leak (37).

Technical Considerations

Transcatheter systems for V-in-V implantation. To date,3 devices have been described in the setting V-in-V implan-tation. 1) The Edwards SAPIEN valve (Edwards Life-sciences) is balloon expandable, composed of a metal stentframe with bovine pericardial leaflets crimped onto a bal-loon catheter (15,16) (Fig. 2). Initial models used stainlesssteel frames, whereas recent versions use a lower profilecobalt chromium frame (38). The current SAPIEN XT

XT valve) (B), Medtronic CoreValve (C),lody delivery system (F).

s

PIENnic Me


valve is manufactured with expanded external diameters of20, 23, 26, and 29 mm, allowing treatment of failedbioprostheses in all positions. 2) The CoreValve system(Medtronic Inc.) is composed of a self-expandable nitinolmultilevel frame with porcine pericardial leaflets (Fig. 2).This valve is available in external diameters of 26 and 29mm, with additional sizes anticipated. The device is de-ployed by retraction of a sheath. The valve can only bemounted in 1 direction within the restraining sheath; hence,the current device can only be delivered in a retrogradeorientation via a transfemoral or subclavian/axillary/transaortic approach. The long frame limits its utility forV-in-V implantation for failed aortic bioprostheses. 3) TheMelody transcatheter valve (Medtronic Inc.) is composed ofa bovine jugular venous valve attached to a platinum iridiumstent scaffold delivered using a balloon-in-balloon system tofacilitate positioning during expansion (Fig. 2). The valve isdesigned for use in the pulmonary circulation, mostly totreat dysfunctional right ventricular outflow tract conduitsor other pulmonary bioprostheses in patients with congen-ital heart disease.Manufacturer sizing and labeling of surgical bioprostheticheart valves. The methodologies for labeling valve sizes arenot standardized and vary by the different manufacturers.

Valve Dimensions for Selected 18- to 23-mm Stper Manufacturer Product InformationTable 1 Valve Dimensions for Selected 18-per Manufacturer Product Informati

ValveLabel Size Valve Type/Model (Manufacturer)

18 Soprano (Sorin Biomedica)

19 Magna (Edwards Lifesciences)

Perimount (Edwards Lifesciences)

Mosaic (Medtronic)

Hancock Ultra (Medtronic)

Hancock II (Medtronic)

Mitroflow (Sorin Biomedica)

Trifecta (St. Jude Medical)

Epic/Biocor (St. Jude Medical)

Epic Supra/Biocor Supra (St. Jude Medical)




Mosaic/Hancock II (Medtronic)

Hancock/Hancock Ultra (Medtronic)









Hancock/Hancock Ultra (Medtronic)





N/A � not available.

Internal diameters vary markedly for a given labeled size(39,40). The labeling of valve sizes may refer to internal orouter diameters of the valve (in the case of stented valves,Fig. 1) or the external diameter for most stentless valves. Inmost cases, the labeled size of a stented valve refers to thestent outer diameter (apart from the Soprano valve in whichthe label reflects the internal diameter). Different biopros-theses with the same label size frequently have differentinternal diameters and external sewing ring diameters. Theexternal sewing ring diameter generally reflects the originalnative annulus diameter as measured by the surgeon at thetime of the original valve implantation.

The exact internal dimensions of a surgical bioprosthesisare those that are most relevant for V-in-V therapy, and inthe case of most stented valves, these diameters are signif-icantly smaller than the labeled valve size. Most stentedbioprostheses with labeled diameters of 19 to 21 mm mayhave internal diameters that are too small to achieveacceptable hemodynamics with V-in-V therapy. Given theheterogeneity of valve types and sizing nomenclature, oper-ators contemplating V-in-V therapy must familiarize them-selves with the structural elements and dimensions of thespecific bioprosthesis that they are treating. Obtaining adetailed operative report is critical. Tables 1, 2, and 3

Surgical Bioprostheses,-mm Stented Surgical Bioprostheses,

wing Ring ExternalDiameter, mm

Stent OuterDiameter, mm

Stent InternalDiameter, mm

26 21 17.8

24 19 18

26 19 18

25 19 17.5

24 19 17.5

N/A N/A N/A

21 18.6 15.4

24 19 N/a

N/A N/A N/A

N/A N/A N/A

28 23 19.8

26 21 20

29 21 20

27 21 18.5

26 21 18.5

23 20.7 17.3

26 21 N/A

N/A 21 19

N/A 21 21

30 25 21.7

28 23 22

31 23 22

30 23 20.5

28 23 22

26 22.7 19

28 23 N/A

N/A 23 21

N/A 23 23

entedto 23on

Se


provide the relevant dimensions for a range of commonlyimplanted surgical bioprostheses.

In general, a V-in-V implant is chosen with a nominalexternal diameter matching or exceeding the reported inter-nal diameter of the failed surgical valve to allow for securefixation and sealing. It is important to consider that excessV-in-V oversizing may be associated with significant un-derexpansion the newly implanted valve, resulting in com-promised hemodynamics and durability. The acceptableboundaries of underexpansion remain undefined in thiscontext, although clinical experience suggests that a range of10% to 15% may be acceptable.

The mechanism of failure of the initial valve deservesconsideration. Regurgitant valves with torn leaflets mayhave relatively larger internal diameters, whereas valves withprominent pannus or calcification may have reduced internaldimensions. Sizing consideration needs to take into accountthe reported internal diameter, bulkiness of the degeneratedvalve, nature of the valve failure, and location of calcificationor pannus. Pre-procedural screening with both computedtomography and transesophageal echocardiography mayhelp quantify discrepancies between in vitro and in vivobioprosthetic internal diameters; provide information re-garding valve bulkiness, calcification, and pannus; and

Valve Dimensions for Selected 24- to 29-mm Stper Manufacturer Product InformationTable 2 Valve Dimensions for Selected 24-per Manufacturer Product Informati

ValveLabel Size Valve Type/Model (Manufacturer)





Mosaic Ultra/Hancock I Ultra (Medtronic)




Epic Supra (St. Jude Medical)





Mosaic Ultra/Hancock II Ultra (Medtronic)









Mosaic Ultra/Hancock II Ultra (Medtronic)






define the internal diameters when the specific valve modelor size is unknown.Access route. Depending on the valve location and thedelivery system, a number of approaches may be considered.Aortic V-in-V implantation may be performed usingtransapical access, which may facilitate crossing a stenoticvalve and coaxial positioning, or various transarterial ap-proaches that may avoid the need for a thoracotomy and ageneral anesthetic. Mitral V-in-V implantation appearsideally suited, at least anatomically, for the transapicalapproach, although the more circuitous transvenous-transseptal approach has been successfully used and avoidsthe need for a thoracotomy and possibly the need for generalanesthesia (41–46). Pulmonary conduits are ideally suitedfor percutaneous femoral, subclavian, or internal jugularvenous access depending on the characteristics of the par-ticular delivery system (47–49). Tricuspid V-in-V implan-tation has been achieved using percutaneous transvenousaccess from the superior vena cava and the femoral vein, aswell as using an open transatrial approach with direct rightatrial puncture (44,48,50,51).The role of pre-dilation. Balloon pre-dilation may beperformed to facilitate crossing or positioning during THVimplantation. However, degenerated bioprostheses are often

Surgical Bioprostheses,-mm Stented Surgical Bioprostheses,

ing Ring ExternalDiameter, mm

Stent OuterDiameter, mm

Stent InternalDiameter, mm

32 27 23.7

28 23 22

31 23 22

30 23 20.5

30 25 22.5

29 25.1 21

31 25 N/A

N/A 25 23

N/A 25 25

35 29 25.6

32 27 26

35 27 26

36 27 24

32 27 24

31 27.3 22.9

33 27 N/A

N/A 27 27

N/A 27 27

38 31 27.6

34 29 28

37 29 28

39 29 26

34 29 26

33 29.5 24.7

35 29 N/A

N/A 29 27

N/A N/A N/A

entedto 29on

Sew

Cpbc

ttpt

hm

wmmV


bulky and friable. Debris embolization is of particularconcern for left-sided bioprostheses, and leaflet tears mayresult in acute severe regurgitation, which may be problem-atic if there is any delay in implanting the THV. Pre-dilation may be necessary when a retrograde technique isused for a severely stenosed and calcified valve (e.g., tran-sarterial crossing of an aortic bioprosthesis). Predilation isgenerally not necessary with antegrade crossing of a stenoticbioprosthesis as the funneled inflow and leaflets facilitateTHV crossing. Similarly, predilation is not generally nec-essary with severely regurgitant bioprostheses, regardless ofthe access mode or the type of THV chosen.Device selection for V-in-V implantation. No doubtmany of the valves currently undergoing evaluation or underdevelopment will be used for future V-in-V implantations.To date only 3 types of THVs have been used in this setting.

AORTIC. The Edwards SAPIEN/XT and the MedtronicoreValve devices have been widely used in the aorticosition. The Medtronic Melody valve has also been used,ut the durability of this venous valve in the systemicirculation is doubtful.

MITRAL. Only the SAPIEN/XT valves have been used inhe mitral position to our knowledge (Fig. 3). The design ofhe current CoreValve system is not suited to mitral im-lantation and the Melody device would not be anticipatedo be sufficiently durable in the systemic circulation.

PULMONARY. The Melody and SAPIEN/XT (Fig. 4) valvesave been used in this position. Pulmonary valve disease is

Valve Dimensions for Selected Stentless Valvesper Manufacturer Product InformationTable 3 Valve Dimensions for Selected Stenper Manufacturer Product Informati

ValveLabel Size Bioprosthesis Ma

19 Freestyle Medtro

Prima Plus Edward

3F Therapeutics ATS Me

Toronto SPV St. Jude

Pericarbon Freedom Sorin B

21 Freestyle Medtro

Prima Plus Edward



Pericarbon freedom Sorin B

23 Freestyle Medtro

Prima Plus Edward




25 Freestyle Medtro

Prima Plus Edward





ostly part of more complex congenital structural disease,

ith pulmonary conduits of relatively low pressure andostly tubular, often in patients who have undergoneultiple previous procedures and hence ideally suited to-in-V therapy.

TRICUSPID. The Edwards SAPIEN/XT (Fig. 5) and Mel-ody valves have been used in this position.Positioning principles and deployment. For secure de-ployment without embolization, the THV is ideallypositioned to overlap the surgical valve annular sewingring (Fig. 6). To achieve this, the exact type of biopros-thesis and its characteristics must be known, includingwhether it is stented or stentless, dimensions, and radio-logic appearance (Fig. 7). Some valves may have ra-diopaque markers or rings at the basal portion close to theanatomic sewing ring (Fig. 8), whereas in others, themarkers may be close to the tip of the stent posts or theremay be no radiopaque markers at all.

Accurate placement requires coaxial positioning withinthe valve, which in turn requires determining the appropri-ate C-arm angulation representing perpendicularity to thevalve plane (52). Fluoroscopically visible components ofsurgical valves can be used effectively to achieve perpendic-ularity. If the radiographic marker is a basal ring, rotatingthe C arm until the ring is side on (appearing as a straight“line”) suggests perpendicularity to the valve plane and anappropriate implant view (Fig. 9). The typical aortic valve isdirected slightly to the right and posterior, such that leftanterior oblique-cranial views are often helpful. The typicalmitral valve is directed to the left and anterior, such that

Valves,

urerOuter

Diameter, mmInternal

Diameter, mm

19 16

ciences 19 16

19 17

cal N/A N/A

ica 19 17

21 18

ciences 21 18

21 19

cal 21 18

ica 21 19

23 20

ciences 23 20

23 21

cal 23 20

ica 23 21

25 21

ciences 25 21

25 23

cal 25 21

ica 25 23

,tlesson

nufact

nic

s Lifes

dical

Medi

iomed

nic

s Lifes

dical

Medi

iomed

nic

s Lifes

dical

Medi

iomed

nic

s Lifes

dical

Medi

iomed

right anterior oblique views are often helpful. Further


manipulation of the delivery system or guidewire may thenbe required to achieve coaxial alignment. For radiolucentvalves (e.g., stentless) angiography with a pigtail catheter inone of the valve cusps and transesophageal echocardio-graphic guidance may be helpful.

When treating failed pulmonic homografts/conduits,pre-stenting with a bare metal stent may provide additionalradial strength and reduce the risk of stent fracture with theMelody valve. Similarly, pre-stenting may provide a longer“landing zone” with a greater margin of safety duringpositioning with the relatively short Edwards SAPIEN/XTvalve (less necessary when treating failed stented biopros-theses that may provide an adequate support structure).Pre-stenting with a covered stent may also reduce the riskassociated with conduit rupture when this is a concern.Pacing. The Edwards valves have generally been deployedwith rapid ventricular pacing to aid accurate positioningof the relatively short valve. This may not be required inthe lower pressure venous system; although, if cardiacmotion is problematic, we have used either rapid pacingor induced ventricular fibrillation. The Medtronic Mel-ody valve is generally deployed without rapid pacing in

Figure 3 Mitral Valve-in-Valve

(A) Fluoroscopic image of a regurgitant Carpentier Edwards valve in the mitral posimage intensifier. (B) Same valve is shown “down the barrel.” This would be a poprojection after implantation of an Edwards SAPIEN valve. (D) “Down the barrel” pvalve (white arrow).

the more commonly treated pulmonary and tricuspid

positions (50), but with pacing in the aortic position.Rapid pacing is generally not used when using theCoreValve device, although this may be of value duringdifficult deployment, particularly when treating a severelyregurgitant bioprosthesis.

Results

Early pre-clinical studies. Boudjemline et al. (53) evalu-ated the concept of mitral V-in-V therapy in a sheep model.A bovine jugular valve was mounted onto a stent andimplanted off-pump using a transatrial approach. Walther etal. (54) evaluated the use of a more contemporary transcath-eter system in both the aortic and mitral positions in pigs.Edwards THVs were implanted into Carpentier Edwardsxenografts transapically in the beating heart model. Azadaniet al. (55) evaluated the hemodynamic performance of23-mm THVs within degenerated surgical bioprostheses.They demonstrated in an in vitro model that incompletestent expansion resulted in leaflet distortion and centralregurgitation when implanted in 19- and 21-mm biopros-

ith the stent frame radiopaque (white arrow). The valve is perpendicular to theiographic projection in which to position a transcatheter valve. (C) Perpendicularon after implantation of an Edwards SAPIEN valve (black arrow) inside the old

ition, wor angrojecti

theses. In a subsequent report, the same group used a


custom-designed supravalvular THV (56), achieving morefavorable hemodynamics.Valve hemodynamics post–V-in-V. THV expansion dur-ing V-in-V therapy is constrained by the internal dimen-

Figure 4 Pulmonic Valve-in-Valve

(A) A Carpentier Edwards Perimount 27-mm valve (a left pulmonary artery stent isseverely regurgitant with moderate to severe stenosis. (B) An Edwards SAPIEN 26ing no residual pulmonary regurgitation. (D) Final image post-implantation.

Figure 5 Tricuspid Valve-in-Valve

(A) A stenosed and regurgitant tricuspid Mitroflow Synergy 27-mm valve is seen pis also seen above it). (B) A sheath is introduced via the right atrium through a mimage after implantation.

sions of the bioprosthetic valve, especially by the relativelyundilatable sewing ring (Fig. 8F). Although this may reducethe chance of annular rupture, heart block, and coronaryocclusion compared with therapy with native aortic valve

een) in a 21-year-old man with repaired tetralogy of Fallot. The valve wasalve is being positioned. (C) Post-deployment pulmonary angiogram demonstrat-

icular to the image intensifier (a bileaflet mechanical valve in the mitral positiont thoracotomy. An Edwards SAPIEN 26-mm valve is being positioned. (C) Final

also s-mm v

erpendini righ


stenosis, valve underexpansion may contribute to hemody-namically significant residual gradients after V-in-V im-plantation. Table 4 provides a summary of the hemody-namic outcomes associated with V-in-V implantation(57–66).

Valve underexpansion may be expected to affect transval-vular gradients, effective orifice areas, and leaflet and stentdurability. Post-procedural transaortic mean gradients are inthe range of 10 to 22 mm Hg in most series (range: 4 to 30mm Hg). In comparison, mean gradients after transcathetertherapy for native aortic stenosis are generally around 10mm Hg (67,68). Although elevated gradients may beacceptable in some patients who cannot undergo open-heartsurgery, they may be inadequate in others with longerexpected survival.

There are currently insufficient long-term outcome datato understand the clinical sequelae of such residual gradi-

Figure 6In Vitro Demonstration of aTranscatheter Valve (Edwards SAPIEN)Implanted Within a Carpentier Edwards Valve

(A) Incorrect positioning: the transcatheter valve is implanted too high withinthe outflow tract of the surgical valve. This may result in splaying of the surgi-cal valve posts and transcatheter valve embolization. (B) Correct valve posi-tioning: the transcatheter valve (arrow) is implanted such that it overlaps thesurgical sewing ring, allowing better anchoring and a more secure position.Reprinted, with permission, from Webb et al. (46).

ents. From our own experience, we have yet to see patients

fail to improve or return for further intervention because ofelevated gradients. For example, 2 patients from our cohorthad transaortic gradients of 24 and 27 mm Hg, and at afollow-up of 610 and 540 days, respectively; both remainimproved and relatively asymptomatic. However, it is likelythat some patients may fail to obtain adequate symptomaticimprovement if sufficiently low transvalvular gradients arenot obtained. V-in-V designs that minimize residual gradi-ents will become increasingly important as younger and

Figure 7 Fluoroscopic Appearance ofVarious Stented Surgical Bioprostheses

Carpentier Edwards (CE) Perimount Standard (A), CE Porcine SupraAnnular (B),CE pericardial (the radiopaque base ring has 3 “holes”) (C), Hancock II (D),CE Porcine (E), Mosaic (radiopaque rings at the top of each stent post only)(F), Mitroflow (G), Epic (very faint base ring only) (H).

lower risk patients are treated in the future.


Regurgitation post–V-in-V. Although paravalvular leaksare common after transcatheter aortic valve replacement fornative aortic stenosis, regurgitation appears to be absent ormild in most published V-in-V reports thus far. The

Figure 8 Fluoroscopic Positioning for V-in-V Implantation for Bi

Importance of knowing the radiological appearance of the surgical valve treated. (although the rigid sewing ring below this required for V-in-V fixation is radiolucent iportion (arrow indicates lowest portion of the SAPIEN valve). (C) Sorin Mitroflow vaMedtronic Mosaic valve: the radiopaque markers are near the top of the surgical sers (white arrows). (F) Deployed Edwards SAPIEN valve. The “waist” at the lower(white arrow). The SAPIEN valve remained slightly underexpanded, and the residu

Figure 9 Perpendicular Alignment for Implantation

Flouroscopic images of a Carpentier Edwards pericardial valve in the mitral positiointensifier, leading to foreshortening. (B) In the 25o right anterior oblique projectioaccurate positioning of a transcatheter valve.

circular sewing ring of surgical bioprostheses appears tofacilitate intervalvular sealing. We suggest reporting 3sources of regurgitation in the context of V-in-V therapy(Fig. 10):

thetic Aortic Valve Failure

pentier Edwards pericardial valve (the wire frame within the valve posts is visible,model). (B) Positioning of the Edwards SAPIEN just below the lowest radiopaque) Positioning the Edwards SAPIEN just below the lowest radiopaque portion. (E)

osts (black arrows), hence the valve is positioned completely below these mark-f the implanted valve demonstrates the narrowest location of the surgical valven gradient was 30 mm Hg.

In the posterior-anterior projection, the valve is not perpendicular to the imagevalve is now perpendicular/orthogonal to the image intensifier, allowing more

opros

A) Carn thislve. (Dtent ppart oal mea

n. (A)n, the

ChTtbtenVGmabtme


1. Paravalvular: arising between the native valve and theoriginal failed bioprosthetic valve. THV implantationis unlikely to ameliorate this, and paravalvular leakclosure with an occlusive plug may be an option.

2. Intervalvular: arising from the area between the old“outer” bioprosthetic valve and the new “inner” trans-catheter valve. This may occur from incomplete THVexpansion (undersizing, severe calcification of biopros-thetic valve leaflets). This has not been a major issue inpublished series thus far.

3. Transvalvular: arising through the newly implantedTHV. This may occur from incomplete leaflet coap-tation due to an oversized and deformed THV or aTHV leaflet “stuck” in an open position. This has notbeen a major issue in published series thus far.

oronary concerns. Most stented surgical bioprosthesesave leaflet tissue mounted internal to the stent frame (Fig. 11).o some extent, this ensures that degenerated bioprosthetic

issue will not be displaced external to the valve posts. Whenioprosthetic tissue is mounted external to the valve frame orhere is no stent frame, there is no such assurance. Forxample, acute coronary ostial occlusion by displaced exter-ally mounted pericardial tissue has been reported after-in-V implantation within the Mitroflow valve (Sorinroup, Burnaby, British Columbia, Canada) (69), althoughany successful and uncomplicated such implantations have

lso been performed. Coronary obstruction has similarlyeen observed with stentless prostheses. It is reasonable tohink that post–V-in-V implantation coronary compromiseight also occur with other failed bioprostheses that have

Clinical and Hemodynamic Outcomes of V-in-V ImplantationTable 4 Clinical and Hemodynamic Outcomes of V-in-V Implant

Valve Position

First Author (Ref. #) No. of Valves Reported THV Im

Aortic

Khawaja et al. (42) 4 CoreValve

Maroto et al. (57) 2 Edwards SAPIEN

Pasic et al. (58) 14 Edwards SAPIEN

Seiffert et al. (59) 4 Edwards SAPIEN

Kempfer et al. (60) 11 Edwards SAPIEN

Gotzmann et al. (61) 5 CoreValve

Webb et al. (46) 10 Edwards SAPIEN

Mitral

Cheung et al. (43) 11 Edwards SAPIEN

Seiffert et al. (59) 1 Edwards SAPIEN

De Weger et al. (62) 1 Edwards SAPIEN (into a faile

Cerillo et al. (44) 3 Edwards SAPIEN

Tricuspid

Roberts et al. (48) 15 Melody

Webb et al. (46) 1 Edwards SAPIEN

Cerillo et al. (44) 1 Edwards SAPIEN

Van Garsee et al. (63) 1 Edwards SAPIEN

Pulmonic

Zahn et al. (64) 34 Melody

Khambadkone et al. (65) 59 Early version of the Melody

Boone et al. (66) 7 Edwards SAPIEN

McElhinney et al. (47) 124 Melody

MG � mean gradient; PG � peak gradient; THV � transcatheter heart valve.

xternally mounted leaflets, are nonstented, implanted in a

supra-annular position, or have unusually bulky diseasedleaflets, particularly in patients with low-lying coronary ostiaand shallow sinuses. When considering aortic V-in-V im-

Figure 10 Possible Sources of Regurgitation AfterV-in-V Implantation

“Down the barrel/end on” image of an Edwards SAPIEN valve (white arrow)implanted within a Medtronic Mosaic valve (black arrows indicate theradiopaque markers). The large surrounding circle represents surroundingnative tissue. Regurgitation may arise from 3 areas depicted schematically bythe 3 black circles: 1, paravalvular, arising between the native tissue and theoriginal failed bioprosthetic valve; 2, intervalvular, arising from the areabetween the old “outer” surgical valve and the new “inner” transcatheter valve;and 3, transvalvular, arising from within the transcatheter valve.

Residual Transvalvular Gradient, mm Hg 30-Day Survival

PG 30–39 100%

PG 13 and 17 Not reported

MG 3–23 (mean MG 13.1 � 6.4) 86%

MG 12–30 (mean MG 19.0 � 12.4) 75%

MG 11 � 4 100%

Mean MG 16.4 � 3.6 100%

Mean MG 20.2 � 6.7 100%

Mean MG 7 (range 5–8) 100%

MG 3 0%

al annuloplasty ring) MG 4 100%

MG 4–9 75%

MG 3.9 93%

MG 4 100%

MG 9 100%

MG 2 Not reported

Mean MG 17.3 � 7.3 100%

Mean MG 19.5 � 15.3 100%

MG 14.9 � 6.9 100%

Median peak 12 99%

ation

planted

d mitr

valve


plants, the relationship of the failed valve to the coronaryostia must be carefully evaluated.

Conclusions

Initial results with V-in-V therapy are very encouraging.However, in the absence of rigorous evaluation and long-term follow-up, V-in-V therapy is probably best consideredonly for patients who present with a prohibitive reoperativerisk. Therapy of small (e.g., �21-mm aortic valves) shouldbe approached with caution as significant residual gradientsmay remain with currently available valves. Operators shouldbe encouraged to share their experience, whether favorable orunfavorable. Future technologic advances may continue toimprove both hemodynamic and clinical outcomes.

Reprint requests and correspondence: Dr. John G. Webb, St.Paul’s Hospital, 1081 Burrard Street, Vancouver, British Colum-bia V6Z 1Y6, Canada. E-mail: [email protected].

REFERENCES

1. Gummert JF, Funkat A, Beckmann A, et al. Cardiac surgery inGermany during 2008. A report on behalf of the German Society forThoracic and Cardiovascular Surgery. Thorac Cardiovasc Surg 2009;57:315–23.

2. Ruel M, Chan V, Bedard P, et al. Very long-term survival implicationsof heart valve replacement with tissue versus mechanical prostheses in

Figure 11 Selected Stented Surgical Bioprosthetic Valves

Carpentier Edwards Perimount (A), Medtronic Mosaic (B), St. Jude Medical EPIC (In A to C, the leaflets are inside the stent frame. In D, the leaflets are outside of

adults �60 years of age. Circulation 2007;116:I294–300.

3. Maganti M, Rao V, Armstrong S, Feindel CM, Scully HE, David TE.Redo valvular surgery in elderly patients. Ann Thorac Surg 2009;87:521–5.

4. Balsam LB, Grossi EA, Greenhouse DG, et al. Reoperative valvesurgery in the elderly: predictors of risk and long-term survival. AnnThorac Surg 2010;90:1195–200, discussion 1201.

5. Jones JM, O’kane H, Gladstone DJ, et al. Repeat heart valve surgery:risk factors for operative mortality. J Thorac Cardiovasc Surg 2001;122:913–8.

6. Akins CW, Buckley MJ, Daggett WM, et al. Risk of reoperative valvereplacement for failed mitral and aortic bioprostheses. Ann ThoracSurg 1998;65:1545–51, discussion 1551–2.

7. Albeyoglu SC, Filizcan U, Sargin M, et al. Determinants of hospitalmortality after repeat mitral valve surgery for rheumatic mitral valvedisease. Thorac Cardiovasc Surg 2006;54:244–9.

8. Leontyev S, Borger MA, Davierwala P, et al. Redo aortic valve surgery:early and late outcomes. Ann Thorac Surg 2011;91:1120–6.

9. Beurtheret S, Gariboldi V, Feier H, et al. Short-term results of repeatvalve replacement: a predictive factor analysis. J Heart Valve Dis2010;19:326–32.

10. Lytle BW, Cosgrove DM, Taylor PC, et al. Reoperations for valvesurgery: perioperative mortality and determinants of risk for 1,000patients, 1958-1984. Ann Thorac Surg 1986;42:632–43.

11. Piehler JM, Blackstone EH, Bailey KR, et al. Reoperation onprosthetic heart values. Patient-specific estimates of in-hospital events.J Thorac Cardiovasc Surg 1995;109:30–48.

12. Kirsch M, Nakashima K, Kubota S, Houel R, Hillion ML, LoisanceD. The risk of reoperative heart valve procedures in Octogenarianpatients. J Heart Valve Dis 2004;13:991–6, discussion 996.

13. Cribier A, Eltchaninoff H, Bash A, et al. Percutaneous transcatheterimplantation of an aortic valve prosthesis for calcific aortic stenosis:first human case description. Circulation 2002;106:3006–8.

14. Webb JG, Altwegg L, Masson JB, Al Bugami S, Al Ali A, Boone RA.A new transcatheter aortic valve and percutaneous valve delivery

d Sorin Medical Mitroflow (D) valves.ent frame (the stent frame is not visible).

C), anthe st

system. J Am Coll Cardiol 2009;53:1855–8.

mailto:[email protected]


15. Webb JG, Chandavimol M, Thompson CR, et al. Percutaneous aorticvalve implantation retrograde from the femoral artery. Circulation2006;113:842–50.

16. Lichtenstein SV, Cheung A, Ye J, et al. Transapical transcatheteraortic valve implantation in humans: initial clinical experience. Circu-lation 2006;114:591–6.

17. Walther T, Dewey T, Borger MA, et al. Transapical aortic valveimplantation: step by step. Ann Thorac Surg 2009;87:276–83.

18. Webb JG. Percutaneous aortic valve replacement will become acommon treatment for aortic valve disease. J Am Coll Cardiol Intv2008;1:122–6.

19. Piazza N, Grube E, Gerckens U, et al. Procedural and 30-dayoutcomes following transcatheter aortic valve implantation using thethird generation (18 Fr) corevalve revalving system: results from themulticentre, expanded evaluation registry 1-year following CE markapproval. EuroIntervention 2008;4:242–9.

20. Tamburino C, Capodanno D, Mule M, et al. Procedural success and30-day clinical outcomes after percutaneous aortic valve replacementusing current third-generation self-expanding CoreValve prosthesis.J Invasive Cardiol 2009;21:93–8.

21. Hilbert SL, Ferrans VJ. Porcine aortic valve bioprostheses: morpho-logic and functional considerations. J Long Term Eff Med Implants1992;2:99–112.

22. Schoen FJ, Levy RJ. Calcification of tissue heart valve substitutes:progress toward understanding and prevention. Ann Thorac Surg2005;79:1072–80.

23. Kallikourdis A, Jacob S. Is a stentless aortic valve superior toconventional bioprosthetic valves for aortic valve replacement? InteractCardiovasc Thorac Surg 2007;6:665–72.

24. Dunning J, Graham RJ, Thambyrajah J, Stewart MJ, Kendall SW,Hunter S. Stentless vs. stented aortic valve bioprostheses: a prospectiverandomized controlled trial. Eur Heart J 2007;28:2369–74.

25. Perez de Arenaza D, Lees B, Flather M, et al. Randomized compar-ison of stentless versus stented valves for aortic stenosis: effects on leftventricular mass. Circulation 2005;112:2696–702.

26. Chambers JB, Rimington HM, Hodson F, Rajani R, Blauth CI. Thesubcoronary Toronto stentless versus supra-annular Perimount stentedreplacement aortic valve: early clinical and hemodynamic results of arandomized comparison in 160 patients. J Thorac Cardiovasc Surg2006;131:878–2.

27. Vesey JM, Otto CM. Complications of prosthetic heart valves. CurrCardiol Rep 2004;6:106–11.

28. Jamieson WR, Munro AI, Miyagishima RT, Allen P, Burr LH, TyersGF. Carpentier-Edwards standard porcine bioprosthesis: clinical per-formance to seventeen years. Ann Thorac Surg 1995;60:999–1006,discussion 1007.

29. Jamieson WR, Rosado LJ, Munro AI, et al. Carpentier-Edwardsstandard porcine bioprosthesis: primary tissue failure (structural valvedeterioration) by age groups. Ann Thorac Surg 1988;46:155–62.

30. Ruel M, Kulik A, Rubens FD, et al. Late incidence and determinantsof reoperation in patients with prosthetic heart valves. Eur J Cardio-thorac Surg 2004;25:364–70.

31. Jamieson WR, Miyagishima RT, Burr LH, Lichtenstein SV, FradetGJ, Janusz MT. Carpentier-Edwards porcine bioprostheses: clinicalperformance assessed by actual analysis. J Heart Valve Dis 2000;9:530–5.

32. van Geldorp MW, Eric Jamieson WR, Kappetein AP, et al. Patientoutcome after aortic valve replacement with a mechanical or biologicalprosthesis: weighing lifetime anticoagulant-related event risk againstreoperation risk. J Thorac Cardiovasc Surg 2009;137:881–6, 886e1-5.

33. Sacks MS, Schoen FJ. Collagen fiber disruption occurs independent ofcalcification in clinically explanted bioprosthetic heart valves. J BiomedMater Res 2002;62:359–71.

34. Vesely I, Barber JE, Ratliff NB. Tissue damage and calcification maybe independent mechanisms of bioprosthetic heart valve failure.J Heart Valve Dis 2001;10:471–7.

35. Mohammadi S, Baillot R, Voisine P, Mathieu P, Dagenais F.Structural deterioration of the Freestyle aortic valve: mode of presen-tation and mechanisms. J Thorac Cardiovasc Surg 2006;132:401–6.

36. Barbetseas J, Nagueh SF, Pitsavos C, Toutouzas PK, Quinones MA,Zoghbi WA. Differentiating thrombus from pannus formation inobstructed mechanical prosthetic valves: an evaluation of clinical,transthoracic and transesophageal echocardiographic parameters. J Am
Coll Cardiol 1998;32:1410–7.
37. Nietlispach F, Johnson M, Moss RR, et al. Transcatheter closure ofparavalvular defects using a purpose-specific occluder. J Am CollCardiol Intv 2010;3:759–65.

38. Nietlispach F, Wijesinghe N, Wood D, Carere RG, Webb JG.Current balloon-expandable transcatheter heart valve and deliverysystems. Catheter Cardiovasc Interv 2010;75:295–300.

39. Christakis GT, Buth KJ, Goldman BS, et al. Inaccurate and mislead-ing valve sizing: a proposed standard for valve size nomenclature. AnnThorac Surg 1998;66:1198–203.

40. Shibata T, Inoue K, Ikuta T, Bito Y, Yoshioka Y, Mizoguchi H.Which valve and which size should we use in the valve-on-valvetechnique for re-do mitral valve surgery? Interact Cardiovasc ThoracSurg 2009;8:206–10.

41. Sharp AS, Michev I, Colombo A. First trans-axillary implantation ofEdwards Sapien valve to treat an incompetent aortic bioprosthesis.Catheter Cardiovasc Interv 2010;75:507–10.

42. Khawaja MZ, Haworth P, Ghuran A, et al. Transcatheter aortic valveimplantation for stenosed and regurgitant aortic valve bioprosthesesCoreValve for failed bioprosthetic aortic valve replacements. J Am CollCardiol 2010;55:97–101.

43. Cheung AW, Gurvitch R, Ye J, et al. Transcatheter transapical mitralvalve-in-valve implantations for a failed bioprosthesis: a case series.J Thorac Cardiovasc Surg 2011;141:711–5.

44. Cerillo AG, Chiaramonti F, Murzi M, et al. Transcatheter valve invalve implantation for failed mitral and tricuspid bioprosthesis. Cath-eter Cardiovasc Interv 2011 Apr 28 [E-pub ahead of print]; doi:10.1002/ccd.23112.

45. Montorfano M, Latib A, Chieffo A, et al. Successful percutaneousanterograde transcatheter valve-in-valve implantation in the mitralposition. J Am Coll Cardiol Intv 2011. In press.

46. Webb JG, Wood DA, Ye J, et al. Transcatheter valve-in-valveimplantation for failed bioprosthetic heart valves. Circulation 2010;121:1848–57.

47. McElhinney DB, Hellenbrand WE, Zahn EM, et al. Short- andmedium-term outcomes after transcatheter pulmonary valve placementin the expanded multicenter US melody valve trial. Circulation2010;122:507–16.

48. Roberts PA, Boudjemline Y, Cheatham JP, et al. Percutaneoustricuspid valve replacement in congenital and acquired heart disease.J Am Coll Cardiol 2011;58:117–22.

49. Huber CH, Hurni M, Tsang V, von Segesser LK. Valved stents fortransapical pulmonary valve replacement. J Thorac Cardiovasc Surg2009;137:914–8.

50. Roberts P, Spina R, Vallely M, Wilson M, Bailey B, Celermajer DS.Percutaneous tricuspid valve replacement for a stenosed bioprosthesis.Circ Cardiovasc Interv 2010;3:e14–5.

51. Kenny D, Hijazi ZM, Walsh KP. Transcatheter tricuspid valvereplacement with the Edwards SAPIEN valve. Catheter CardiovascInterv 2011;78:267–70.

52. Gurvitch R, Wood DA, Leipsic J, et al. Multislice computed tomog-raphy for prediction of optimal angiographic deployment projectionsduring transcatheter aortic valve implantation. J Am Coll Cardiol Intv2010;3:1157–65.

53. Boudjemline Y, Pineau E, Borenstein N, Behr L, Bonhoeffer P. Newinsights in minimally invasive valve replacement: description of acooperative approach for the off-pump replacement of mitral valves.Eur Heart J 2005;26:2013–7.

54. Walther T, Falk V, Dewey T, et al. Valve-in-a-valve concept fortranscatheter minimally invasive repeat xenograft implantation. J AmColl Cardiol. 2007;50:56–60.

55. Azadani AN, Jaussaud N, Matthews PB, Ge L, Chuter TA, TsengEE. Transcatheter aortic valves inadequately relieve stenosis in smalldegenerated bioprostheses. Interact Cardiovasc Thorac Surg 2010;11:70–7.

56. Azadani AN, Jaussaud N, Matthews PB, et al. Valve-in-valve implan-tation using a novel supravalvular transcatheter aortic valve: proof ofconcept. Ann Thorac Surg 2009;88:1864–9.

57. Maroto LC, Rodriguez JE, Cobiella J, Marcos P. Transapical off-pump aortic valve-in-a-valve implantation in two elderly patients witha degenerated porcine bioprosthesis. Eur J Cardiothorac Surg 2010;37:738–40.

58. Pasic M, Unbehaun A, Dreysse S, et al. Transapical aortic valve implan-tation after previous aortic valve replacement: clinical proof of the
”valve-in-valve” concept. J Thorac Cardiovasc Surg 2011;142:270–7.


59. Seiffert M, Franzen O, Conradi L, et al. Series of transcathetervalve-in-valve implantations in high-risk patients with degeneratedbioprostheses in aortic and mitral position. Catheter Cardiovasc Interv2010;76:608–15.

60. Kempfert J, Van Linden A, Linke A, et al. Transapical off-pumpvalve-in-valve implantation in patients with degenerated aortic xeno-grafts. Ann Thorac Surg 2010;89:1934–41.

61. Gotzmann M, Mugge A, Bojara W. Transcatheter aortic valve implan-tation for treatment of patients with degenerated aortic bioprostheses—valve-in-valve technique. Catheter Cardiovasc Interv 2010;76:1000–6.

62. de Weger A, Tavilla G, Ng AC, et al. Successful transapical trans-catheter valve implantation within a dysfunctional mitral bioprosthesis.J Am Coll Cardiol Img 2010;3:222–3.

63. Van Garsse LA, Ter Bekke RM, van Ommen VG. Percutaneoustranscatheter valve-in-valve implantation in stenosed tricuspid valvebioprosthesis. Circulation 2011;123:e219–21.

64. Zahn EM, Hellenbrand WE, Lock JE, McElhinney DB. Implanta-tion of the melody transcatheter pulmonary valve in patients with a
dysfunctional right ventricular outflow tract conduit early results fromthe U.S. clinical trial. J Am Coll Cardiol 2009;54:1722–9.
65. Khambadkone S, Coats L, Taylor A, et al. Percutaneous pulmonaryvalve implantation in humans: results in 59 consecutive patients.Circulation 2005;112:1189–97.

66. Boone RH, Webb JG, Horlick E, et al. Transcatheter pulmonary valveimplantation using the Edwards SAPIEN transcatheter heart valve.Catheter Cardiovasc Interv 2010;75:286–94.

67. Grube E, Buellesfeld L, Mueller R, et al. Progress and current statusof percutaneous aortic valve replacement: results of three devicegenerations of the CoreValve Revalving system. Circ Cardiovasc Interv2008;1:167–75.

68. Webb JG, Altwegg L, Boone RH, et al. Transcatheter aortic valveimplantation: impact on clinical and valve-related outcomes. Circula-tion 2009;119:3009–16.

69. Gurvitch R, Cheung A, Bedogni F, Webb JG. Coronary obstructionfollowing transcatheter aortic valve-in-valve implantation for failedsurgical bioprostheses. Catheter Cardiovasc Interv 2011;77:439–44.

Key Words: failed surgical bioprosthesis y transcatheter aortic valveimplantation y valve-in-valve.

Documents

Transcatheter Valve-in-Valve Implantation for Failed ...pulmonic, and tricuspid positions. Despite some core similarities to transcatheter therapy of native valve dis-ease, V-in-V