Journal of the American College of Cardiology Vol. 62, No. 1, 2013� 2013 by the American College of Cardiology Foundation ISSN 0735-1097/$36.00Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jacc.2013.02.088

Evaluation and Management ofParavalvular Aortic Regurgitation AfterTranscatheter Aortic Valve Replacement

Jan-Malte Sinning, MD,* Mariuca Vasa-Nicotera, MD,* Derek Chin, MD,yChristoph Hammerstingl, MD,* Alexander Ghanem, MD,* Johan Bence, MD,y Jan Kovac, MD,yEberhard Grube, MD,* Georg Nickenig, MD,* Nikos Werner, MD*

Bonn, Germany; and Leicester, United Kingdom

P

From the *Medizin

Germany; and th

Leicester, United

and Efficacy Study

and Efficacy study

Medtronic. Dr. Gr

they have no relati

Manuscript rece

2013, accepted Fe

aravalvular aortic regurgitation (PAR) negatively affects the prognosis after transcatheter aortic valve replacement(TAVR) with dramatically increased morbidity and mortality in patients with more than mild PAR. Becausetranscatheter heart valves are implanted in a sutureless fashion using oversizing to anchor the prosthesis stent frameat the level of the virtual aortic annulus, stent frame underexpansion due to heavily calcified cusps, suboptimalplacement of the prosthesis, and/or annulus-prosthesis-sizemismatch due tomalsizing can contribute to paravalvularleakage. In contrast to open heart surgery, TAVR does not offer the opportunity to measure the aortic annulus underdirect vision during the procedure. Therefore, the dilemma before each TAVR procedure is the appropriate sizing of thedimensions of the aortic annulus and to choose not only the size but also the transcatheter heart valve type(self-expanding vs. balloon-expandable) that fits the given anatomy best. Because precise echocardiographicquantification of PAR in patients with TAVR remains challenging especially in the acute implantation situation,a multimodal approach for the evaluation of PAR with the use of hemodynamic measurements and imagingmodalities is imperative to precisely quantify the severity of aortic regurgitation immediately after valve implantationand to identify patients who will benefit from corrective measures such as post-dilation or valve-in-valve implantation.Every measure has to be taken to prevent or reduce PAR to provide a satisfying long-term clinical outcome.(J Am Coll Cardiol 2013;62:11–20) ª 2013 by the American College of Cardiology Foundation

Transcatheter aortic valve replacement (TAVR) has becomean alternative to surgical aortic valve replacement for inop-erable or surgical high-risk patients with severe aorticstenosis and prohibitive surgical risk (1,2). Transcatheterheart valves (THVs) are implanted in a sutureless fashionusing oversizing to anchor the prosthesis stent frame at thelevel of the aortic annulus. Therefore, incomplete circum-ferential apposition of the prosthesis with the annulus mightlead to paravalvular aortic regurgitation (PAR). BecausePAR negatively affects the prognosis after TAVR withdramatically increased morbidity and mortality in patientswith more than mild PAR (3–16), this procedure-relatedissue has to be addressed to further improve the outcomeof patients after TAVR (3–5).

ische Klinik und Poliklinik II, Universitätsklinikum Bonn, Bonn,

e yDepartment of Cardiology, Glenfield University Hospital,

Kingdom. Dr. Chin is a co-investigator for the CoreValve Safety

, Medtronic CoreValve Advance study, and St. Jude Portico Safety

. Dr. Kovac is a proctor for Edwards Lifesciences Corporation and

ube is a proctor for Medtronic. All other authors have reported that

onships relevant to the contents of this paper to disclose.

ived December 26, 2012; revised manuscript received January 23,

bruary 5, 2013.

This paper reviews the evaluation of significant PAR inpatients with TAVR with a focus on precise quantificationand therapeutic options to manage PAR after TAVR.

Background

Anatomy of the aortic valve. Precise assessment of thedimensions of the aortic valvular complex is fundamentalbefore each TAVR procedure. Although the concept of anannulus is ingrained in the surgical vocabulary, the aorticroot is not constructed on the basis of a ring-like structuresupporting the leaflets of the aortic valve. The true histo-logical ventricular–aortic junction, which is circular butcrossed by the semilunar attachments of the leaflets of theaortic valve in a crown-like fashion (the “surgical” annulus)within the cylindrical aortic root, cannot be seen on multi-slice computed tomography or other noninvasive imagingmodalities (17,18). Therefore, the so-called aortic annulus isdefined as a virtual ring that has three anchor points at thenadir of each of the attachments of the aortic cusps (Fig. 1).This virtual aortic annulus is mostly noncircular and oftenmay have an oval or elliptical shape (17–22).

Abbreviationsand Acronyms

AR = aortic regurgitation

ARI = aortic regurgitation

index

LV = left ventricle

LVEDP = left ventricular

end-diastolic pressure

LVOT = left ventricular

outflow tract

PAR = paravalvular aortic

regurgitation

TAVR = transcatheter aortic

valve replacement

TEE = transesophageal

echocardiography

THV = transcatheter heart

valve

Sinning et al. JACC Vol. 62, No. 1, 2013Paravalvular Aortic Regurgitation After TAVR July 2, 2013:11–20

12

Paravalvular AorticRegurgitation

PAR results from incompletecircumferential apposition of theprosthesis with the annulus. Dis-cordance in the orthogonal dia-meters of oval-shaped annuli withthe circular prosthesis leads toprosthetic undersizing or over-sizing, a cause of significant PARor annular rupture (21). Degen-erative calcification of the nativeaortic valve (e.g., severely calcifiedcusp) causes PAR through mal-apposition of the skirt with theannulus or displacement of theprosthesis into a high or low po-sition within the aortic root

Figure 1 Anatomy of the Aortic Root

The aortic root with the aortic valve, which is suspended in a crown-like fashion

within the root, contains 3 circular rings (A): the sinotubular junction, the

“surgical” annulus (anatomic ventriculo-arterial junction), and the virtual aortic

annulus, which is formed by joining the nadir of each aortic valvular cusp (B).

VA ¼ virtual aortic.

(6,22). It may hamper proper positioning of the THVand contribute to the occurrence of significant PAR afterTAVR (23).

Another risk factor for PAR is a functional bicuspid aorticvalve, which is predominantly caused by fusion of thecommissures as a consequence of heavy calcification that maypreclude the expansion of the prosthesis and full apposition ofthe THV with the annulus. On a compassionate off-labelbasis, early results of TAVR for bicuspid aortic valvestenosis seem promising, and most cases of PAR are mild(24–26). However, the long-term impact of noncircularprosthesis expansion on hemodynamics and durability of theTHVhas to be elucidated before TAVR can be recommendedin patients with bicuspid aortic stenosis.

Clinically significant acute AR is characterized bya considerable diastolic runoff into the left ventricle (LV)leading to acute volume overload, which has a disastrouseffect, especially in patients with severely impaired leftventricular function: The noncompliant, hypertrophic LV,which has been adapted to chronic pressure overload, isunable to increase the end-diastolic volume, so that the leftventricular end-diastolic pressure (LVEDP) increasesrapidly with consecutive decrease of cardiac output leadingto hemodynamic deterioration (27,28).

Several studies have shown that up to 70% of all patientswith TAVR have PAR after the procedure, graded asmoderate or severe in approximately 15% of the patients(Table 1). Meanwhile, it is well recognized that the occur-rence of more than mild PAR has a significant impact onprognosis after TAVR with a two- to four-fold increased1-year mortality risk compared with patients without clini-cally significant PAR (6–16). In 3,195 patients of theprospective FRANCE-2 (French Aortic National Core-Valve and Edwards Registry 2) TAVR registry, the inves-tigators observed that patients with a more than mild PARhad a 2.5-fold increased mortality risk compared withpatients without PAR or with only mild PAR (11). The 2-

year results of the PARTNER (Placement of AorticTranscatheter Valves) Trial cohort A (10) even suggestedthat a lesser extent of PAR may be harmful for patients withTAVR. Therefore, the precise quantification of the severityof AR immediately after valve implantation (within thecatheterization laboratory) is paramount to identify patientswith clinically significant PAR.

Imaging Before and After TAVR

In contrast to open heart surgery, TAVR does not offer theopportunity to measure the aortic annulus under directvision during the procedure. Therefore, the dilemma beforeeach TAVR procedure is the appropriate sizing of thedimensions of the aortic annulus and to choose not only thesize but also the THV type (self-expanding vs. balloon-expandable) that fits the given anatomy best.

Table1

Incide

nceof

Mod

erate/

Sev

erePARAfter

TAVR

FirstAutho

r(R

ef.#)

Patients

EuroS

CORE

Acc

essRou

teTH

VTy

pePARRate

Asses

sedby

MortalityforMore

Than

Mild

PAR

Abd

el-W

ahab

etal.(8)

690

20.4

�13.1

92.4%

TF,3.5%

TA84.3%

CV(M

edtron

icInc.,M

inne

apolis,

Minne

sota),15.7%

ES(Edw

ards

Lifesciences

Corpo

ratio

n,Irvine

,Califo

rnia)

17.2%

Ang

iograp

hyNA

Leon

etal.(1)

179

26.4

�17.2

100%

TF100%

ES15.2%

Echo

cardiograp

hyNA

Tambu

rinoet

al.(9)

663

23.0

�13.7

90.3%

TF,9.7%

TS100%

CV

21.0%

Echo

cardiograp

hyNA

Smith

etal.(2)

348

29.3

�16.5

70.1%

TF,29.9%

TA100%

ES13.1%

Echo

cardiograp

hyNA

Moa

tet

al.(12)

870

18.5

(11.7–27.9)

68.9%

TF,26.4%

TA52.0%

CV,

48.0%

ES13.6%

Ang

iograp

hyNA

Sinn

inget

al.(6)

146

30.2

�18.0

91.8%

TF,8.2%

TS100%

CV

15.1%

Echo

cardiograp

hy,a

ngiograp

hy,

hemod

ynam

ics

30-day:2

2.7%,1-yr:63.6%

Gila

rdet

al.(11)

1,915*

21.9

�14.3

73.9%

TF,17.7%

TA66.9%

ES,33.1%

CV

16.5%

Echo

cardiograp

hyNA

Gotzm

annet

al.(16)

198

22.0

�16.0

97.5%

TF,2.5%

TS100%

CV

14.1%

Echo

cardiograp

hy,h

emod

ynam

ics

30-day:2

1.0%,1-yr:57.0%

Vasa-Nicoteraet

al.(34)

122

22.4

�13.0

97.5%

TF,1.7%

TA79.5%

CV,

20.5%

ES16.4%

Echo

cardiograp

hy,a

ngiograp

hy,

hemod

ynam

ics

30-day:3

0.0%,1-yr:60.0%

Hayashida

etal.(15)

400

22.3

(17.1–30.3)

NA

86.8%

ES,13.2%

CV

3.0%

Echo

cardiograp

hy30-day:1

6.7%,1-yr:60.0%

*In1,915/3

,195pa

tients,

PARwas

assessed

aftertheTA

VRproced

ure.

euroSC

ORE¼

Europe

anSystem

forCardiac

Ope

rativeRiskEvalua

tion;

CV¼

CoreV

alve;ES

¼Ed

wards

SAPIEN;N

A¼

notavailable;

PAR¼

paravalvular

aortic

regu

rgita

tion;

TA¼

tran

sapical;TF

¼tran

sfem

oral;TH

V¼

tran

scathe

terhe

artvalve;

TS¼

tran

s-subclavian

.

JACC Vol. 62, No. 1, 2013 Sinning et al.July 2, 2013:11–20 Paravalvular Aortic Regurgitation After TAVR

13

Two-dimensional transesophageal echocardiography(TEE) is not sufficient for accurate THV sizing and tends tounderestimate the real dimension of the aortic annulus,because the basal attachments of the aortic leaflets are cutbut the maximum diameter of the aortic annulus is nottransected (18,21). Therefore, pre-procedural 3-dimensionalimaging techniques for detailed and accurate assessment ofthe annulus are essential to reduce the incidence of para-valvular leakage after TAVR (21,29–33). For TAVRplanning, multislice computed tomography is the goldstandard because it permits precise measurement of the aorticroot at any desired level or plane, and thus may provide moreaccurate measurement. Because the aortic annulus is oftenoval shaped, it is recommended to measure both themaximum and the minimum diametersdor even morepragmatic, the perimeter or area of the annulusdfor precisesizing of the THV. With respect to annulus area or perim-eter, an oversizing by more than 10% seems to be associatedwith a lesser extent of PAR (6,16,21,32,34).

Angiographic Assessment of PAR

The qualitative angiographic grading of PAR is an easy-to-use method, but regurgitant flow within each angio-graphic grade varies widely, and a considerable overlap fromone grade to another has been reported (35,36). However, inrecent TAVR studies, the angiographically assessed (quali-tative) degree of PAR correlated well with echocardiographyin patients with TAVR (6,34). PAR can be classifiedaccording to the visually estimated density of opacification ofthe LV into three degrees adapted to the Valve AcademicResearch Consortium 2 criteria: mild (reflow of contrast inthe outflow tract and middle portion of the LV but clearingwith each beat), moderate (reflow of contrast in the wholeleft ventricular cavity with incomplete washout in a singlebeat and faint opacification of the entire LV over severalcardiac cycles), and severe (opacification of the entire LVwith the same intensity as in the aorta and persistence of thecontrast after a single beat). Because many patients withTAVR undergo the procedure in conscious sedation andtherefore the implantation of the valve prosthesis ispredominantly angiographically guided, the amount ofcontrast dye should be kept as low as possible, especially inpatients with chronic renal failure and/or at high risk for theoccurrence of acute kidney injury (3,7).

Echocardiographic Assessment

Despite the recently updated Valve Academic ResearchConsortium 2 criteria, which refer to parameters recom-mended for surgical prosthetic heart valves that have yet tobe validated in THVs, the echocardiographic quantificationof PAR after TAVR remains challenging (37,38). Even theuse of an integrative echocardiographic approach for gradingaortic regurgitation (AR) remains imprecise in practice,especially in the implantation situation, for the following

Sinning et al. JACC Vol. 62, No. 1, 2013Paravalvular Aortic Regurgitation After TAVR July 2, 2013:11–20

14

reasons: 1) the acute hemodynamic changes that occurduring TAVR affect Doppler and color flow assessment;2) semiquantitative parameters of AR severity, such as jetwidth, vena contracta, or pressure half-time, are best appliedin central jets and thus are not ideal for quantification of theeccentric, circumferential PAR jets seen in patients withTAVR; and 3) acoustic shadowing by the prosthesis andnative calcification also may obscure PAR jets. Given thesequantitative limitations, TEE screening criteria include a jetdepth extending beyond the left ventricular outflow tract(LVOT), multiple PAR jets, and holodiastolic flow reversalin the descending aorta (37). Prominent holodiastolic flowreversal in the descending aorta identifies patients with morethan mild PAR with a sensitivity of up to 86% and rules outsignificant PAR with a specificity of up to 92% (p < 0.001)(6,13). In addition, it is desirable to measure the circum-ferential extent of the jet in the short-axis view in patientswith clinically significant PAR (<10%: mild, 10% to 29%:moderate, and �30%: severe PAR) (37). However, it isimportant to realize that at this time, the body of evidencesupporting these criteria for the assessment of paravalvularAR may be limited and requires further validation inpatients with THV as our experience continues to expand.

Nonetheless, TEE has an essential role in defining theorigin and mechanism of PAR. The jets of transvalvular AR

Figure 2 Origin and Mechanism of Paravalvular Aortic Regurgitation

Transesophageal echocardiography (TEE) long-axis view of Edwards SAPIEN (Edwards Life

and PAR (A, Online Video 1). PAR after deployment of an Edwards SAPIEN XT 26-mm pros

the root (B, Online Video 2). TEE short-axis view of underexpanded oval-shaped waist of a M

TEE long-axis view of an Edwards SAPIEN XT 26-mm prosthesis with diameters of 24 to 26

TEE of malposition PAR after low implantation of a Medtronic CoreValve prosthesis. The P

(“supra-skirt” PAR) into the paravalvular space and LVOT (E, Online Video 3). TEE of malpos

posterior PAR jet passes from the aortic sinus below the tissue skirt (“infra-skirt” PAR) int

Video 4). AR ¼ aortic regurgitation.

are easy to differentiate from PAR, lying within thecircumference of the prosthetic stent frame (Fig. 2A,Online Video 1). The valve leaflets should be assessed todetermine whether there is structural damage causingtransleaflet AR, intact leaflets with insufficient diastolicpressure to close them, or central leaflet malappositionrelated to the prosthetic design or tissue. Malappositionparavalvular AR jets occur outside the circumference of theprosthetic stent frame. These jets are caused by incompleteapposition between the prosthesis and the annulus because ofheavy calcification or underexpansion of the prosthesis(Fig. 2B, Online Video 2). TEE assessment of prostheticgeometry (circular or noncircular) (Fig. 2C) and in vivodiameter may be required to guide treatment (Fig. 2D).

Furthermore, malposition PAR jets may occur across thestent frame, when the prosthesis is implanted in too low ortoo high a position relative to the native annulus. In the lowimplant (“too ventricular”), the prosthesis is deployed ata depth that exceeds the height of its tissue skirt; the PARjet passes above the skirt (“supra-skirt” PAR), from withinthe aortic portion of the stent frame into the paravalvularspace and the LVOT (Fig. 2E, Online Video 3). In the highimplant (“too aortic”), the prosthesis is deployed partiallyabove the native annulus; the PAR jet passes from theparavalvular space across the irregular inflow edge of the

(PAR)

sciences Corporation, Irvine, California) 23-mm prosthesis with both transvalvular AR

thesis because of malapposition with the left annulus. The device was not co-axial to

edtronic CoreValve 29-mm prosthesis (Medtronic Inc., Minneapolis, Minnesota) (C).

mm indicating reasonable expansion in the presence of PAR (D). Color 3-dimensional

AR jet passes from within the aortic portion of the stent frame above the tissue skirt

ition PAR after the high implantation of a Medtronic CoreValve 29-mm prosthesis. The

o the LVOT where the irregular inflow edge rises above the native annulus (F, Online

JACC Vol. 62, No. 1, 2013 Sinning et al.July 2, 2013:11–20 Paravalvular Aortic Regurgitation After TAVR

15

prosthesis into the LVOT (“infra-skirt” PAR) (Fig. 2F,Online Video 4).

Hemodynamic Assessment During TAVR

Because an objective and quantitative parameter for thedirect assessment of the severity of PAR during the proce-dure would be invaluable in patients with TAVR to applyeffective measures to reduce the severity of PAR, the use ofhemodynamic parameters for a quantitative evaluation ofPAR seems to be useful. The so-called aortic regurgitationindex (ARI) is the ratio of the transvalvular gradient betweendiastolic blood pressure (RRdia) in the aorta and LVEDPto systolic blood pressure (RRsys) in the aorta: [(RRdia –

LVEDP)/RRsys] � 100 (Figs. 3A and 3B) (6). The ARIwas developed in a cohort of TAVR patients from Bonn,Germany, and was validated in an independent TAVRcohort from Leicester, United Kingdom (6,13). The ARIshows an inverse correlation to the severity of PAR; differ-entiates between patients with mild, moderate, or severePAR; and independently predicts the associated 1-yearmortality risk. In both cohorts, an ARI cutoff value of 25was calculated by receiver operating curve analysis as theoptimal predictor of the 1-year mortality risk after TAVR.

Figure 3 ARI in Patients With Moderate and Trivial PAR

After adjustment for the systolic blood pressure level, the transvalvular gradient

results in an ARI of 16.7 in a patient with moderate PAR (A) and an ARI of 30.8 in

a patient with trivial PAR (B). LVEDP ¼ left-ventricular end-diastolic blood pressure;

RRdia ¼ end-diastolic blood pressure in the aorta; RRsys ¼ systolic blood pressure

in the aorta.

In addition, this ARI cutoff value had a negative predictivevalue of 95% to 100% for the occurrence of more than mildPAR, when used complementary to the angiographicallyor echocardiographically assessed PAR severity (6,13).Although the ARI does not consider the severity of pre-procedural AR and concomitant mitral regurgitation, it isa helpful tool to identify patients after TAVR with the needto take effective corrective measures to decrease the severityof PAR. Another study on the impact of PAR on long-termoutcome recently confirmed that the ARI significantlydiffered between patients with moderate/severe PAR com-pared to patients with no or mild PAR (16). However,the ARI still has to be validated in a larger and controlledstudy population.

When evaluating hemodynamics after TAVR with thedimensionless ARI, it is recommended to perform themeasurements approximately 10 min after valve deploymentto prevent confounding by an increased LVEDP due tomyocardial ischemia and/or diastolic dysfunction after rapidpacing and balloon valvuloplasty. The determination of theARI should be performed as mean value over several cardiaccycles (especially in patients with atrial fibrillation) witha heart rate of 60 to 80 beats/min and without extrasystolicbeats, because with increasing heart rate and shortenedduration of the diastole, the diastolic pressure in the aorta alsoincreases and therebymight lead to a false-negativeARI abovethe cutoff of 25.

Treatment Options to Reduce PAR

Several corrective measures have been proposed to overcomesignificant residual PAR after TAVR. However, data onthese measures predominantly originate from small series orcase reports, and the impact of corrective measures for thereduction of PAR on long-term outcome and especiallyvalve durability still has to be clarified in future studies(39–45).

A standardized algorithm with a multimodal approachmight be helpful for the evaluation of PAR after TAVR to:1) quantify the severity of AR immediately after valveimplantation (within the catheterization laboratory); 2)evaluate the hemodynamic tolerability of PAR with regardto the prognosis after TAVR; and 3) identify patients whowill benefit from corrective measures, such as post-dilationor valve-in-valve implantation to reduce PAR and improveoutcome (Fig. 4). After valve deployment, the degree ofPAR can be assessed by aortic root angiography or echo-cardiography. If no PAR is present, no measures have to betaken. In all other cases with mild to severe PAR, thedetermination of the ARI is helpful to more preciselyquantify the extent of PAR and to have a point of referencebefore corrective measures are taken. In patients with more-than-mild PAR and/or an ARI <25, the evaluation of PARby echocardiography, preferably TEE, is recommended toelucidate the cause of PAR and its mechanism. Whencorrective measures have been taken in patients with

Figure 4 Treatment Algorithm for PAR After TAVR

Treatment algorithm using a multimodal approach with the use of hemodynamic measurements and imaging modalities to quantify the severity of PAR during the TAVR procedure

and to identify patients who will benefit from corrective measures. AR ¼ aortic regurgitation; TEE ¼ transesophageal echocardiography; TTE ¼ transthoracic echocardiography.

Figure 5 Balloon Post-Dilation

Underexpansion of a CoreValve 29-mm prosthesis frame caused by a severely calcified cusp (A) resulted in moderate PAR with an eccentric jet near the left coronary cusp in

aortic root angiography (B, arrows show eccentric AR jet) and an ARI of 20.8 (C). Post-dilation with a straight 28-mm valvuloplasty balloon (D) led to a satisfying procedural result

with only mild PAR (E) and increased the ARI to 31.8 (F). AR ¼ aortic regurgitation.

Sinning et al. JACC Vol. 62, No. 1, 2013Paravalvular Aortic Regurgitation After TAVR July 2, 2013:11–20

16

Figure 6 Valve-in-Valve Implantation

Deep (“too ventricular”) implantation of a CoreValve 31-mm prosthesis (A) led to severe PAR in aortic root angiography (B, arrows show supra-skirt PAR) and an ARI of 14.3,

reflecting unfavorable hemodynamics (C). After implantation of a second CoreValve 31-mm prosthesis in valve-in-valve-technique, which was delivered 5 mm higher than the first

CoreValve under TEE control (D), only trivial PAR was left in angiography (E), and the ARI increased from 14.3 to 45.8, indicating a good procedural result (F). Abbreviation as in

Figure 5.

Figure 7 Snare Technique

Deep implantation of a CoreValve prosthesis (left coronary cusp: 13.2 mm; noncoronary cusp: 12.3 mm) resulted in severe “supra-skirt” PAR (A, B) with beginning equalization

between the diastolic blood pressure in the aorta and the left ventricular end-diastolic pressure (LVEDP), resulting in an ARI of 16.0 (C). TEE showed the circumferential extent of

massive paravalvular leakage (D). Via left-sided transbrachial access, the CoreValve prosthesis was pulled into a more favorable position (left coronary cusp: 4.2 mm; non-

coronary cusp: 2.6 mm) with use of an Amplatz (AGA Medical Corp., Plymouth, Minnesota) gooseneck snare catheter (E), resulting in an ARI of 32.1 (F) without evidence of

residual PAR (E). Abbreviation as in Figure 5.

JACC Vol. 62, No. 1, 2013 Sinning et al.July 2, 2013:11–20 Paravalvular Aortic Regurgitation After TAVR

17

Sinning et al. JACC Vol. 62, No. 1, 2013Paravalvular Aortic Regurgitation After TAVR July 2, 2013:11–20

18

clinically significant PAR, the severity of PAR can be re-evaluated by imaging modalities and the AR index.

Balloon Post-Dilation

The presence of severely calcified cusps of the native aorticvalve might prevent complete apposition of the prosthesiswith the annulus leading to a typical eccentric AR jet(Fig. 5). Several studies identified a higher degree of nativeaortic valve calcification as a predictor of more than mildPAR (22,46). Balloon post-dilation is an option to reducethe degree of PAR by obtaining a better expansion of theprosthesis stent frame and a better sealing of the paravalvularspace if the THV has been deployed at the correctimplantation depth (6,26,40–43,45,47). Post-dilation is alsothe treatment option of choice for patients with frameunderexpansion as the reason for severe PAR that can occurin rare casesddespite predilation of the native aorticvalvedwith the use of self-expanding THVs (48).

The size of the balloon for post-dilation should conform tothe aortic annulus dimension and not exceed the maximumdiameter of the native aortic valve. For the Medtronic Cor-eValve prosthesis (Medtronic Inc.,Minneapolis,Minnesota),a straight valvuloplasty balloon with a maximum diameter of22, 25, 28, and 29 mm is recommended for the 23-, 26-, 29-,and 31-mm CoreValve, respectively (6). For the Edwards

Figure 8 Interventional Closure of Paravalvular Leakage

Despite balloon post-dilation and oversizing with a straight 22-mm valvuloplasty balloon,

23-mm prosthesis with the annulus caused by a severely calcified cusp (A, B) remained 2

depth and showed an unfavorable ARI of 21.7 (C). TEE was used to identify the PAR pat

paravalvular defect was obtained with use of a straight hydrophilic wire over a diagnostic

Edwards SAPIEN prosthesis. Before insertion of the 9F delivery sheath, the THV was secu

placed in the LV to control the Edwards SAPIEN prosthesis in case of dislodgement or em

real-time 3-dimensional TEE and successful leak closure with only trace PAR left in angio

SAPIEN prosthesis (Edwards Lifesciences Corporation,Irvine, California), balloon post-dilation should be performedwith the same balloon as used for delivery of the valve pros-thesis stepwise adding 1 ml saline to the total volume toincrease its diameter (43). In a recent study, THV oversizingafter deployment by post-dilation did not lead to leafletmalapposition with consecutive central AR (15,43).

Valve-in-Valve Implantation

Accurate positioning of the THV with respect to the nativeaortic annulus is critical to ensure a successful procedure,whereas suboptimal deployment can result in incompleteapposition of the valve and annulus or even worse inincomplete sealing by the pericardial skirt of the stent frameallowing a considerable diastolic backflow into the LV. Formalpositioned THVs with too shallow (“too aortic”) or toodeep (“too ventricular”) implantation of the prosthesis, valve-in-valve implantation is a viable treatment strategy to reducesignificant PAR and to prevent bailout cardiac surgery(Fig. 6). The second valve can be deployed in a way that thesealing pericardial skirts of both valves overlap and that thesecond valve ensures sealing with the native valve annulus(44). Thus, initial procedural failure can be converted intoprocedural success in up to 90% of the attempts, and valve-in-valve implantation results in comparable hemodynamic

moderate AR due to incomplete circumferential apposition of an Edwards SAPIEN

months after implantation of this accurately sized prosthesis with proper implantation

homechanism and identified the localized paravalvular leak (D). Passage of the

4F Judkins right coronary catheter, avoiding crossing through the stent struts of the

red by transvalvular insertion of an Amplatz Super Stiff wire via a 5F pigtail catheter

bolization. After deployment of an Amplatz Vascular Plug III under guidance with

graphy (E), the ARI increased to 30.8 (F) (51,52). Abbreviation as in Figure 5.

JACC Vol. 62, No. 1, 2013 Sinning et al.July 2, 2013:11–20 Paravalvular Aortic Regurgitation After TAVR

19

results with satisfactory short-term and midterm outcomes(42,44,49). In addition, the valve-in-valve technique isa viable treatment option for significant transvalvular ARdue to severe prosthetic leaflet dysfunction and has impor-tant implications for patients who develop late failure ofTHVs, also in case of restenosis as a “re-do” procedure (50).However, one major drawback of THV-in-THV implan-tation with the Medtronic CoreValve is the restricted accessto the coronary ostia. In the near future, repositionabilityand retrievability of THVs will be addressed by the next-generation devices and help to reduce the occurrence ofmalpositioning (4).

Snare Technique

The Snare technique represents an option worth consideringfor a too deeply implanted Medtronic CoreValve (too“ventricular” placement of the prosthesis) (Fig. 7). In thiscase, correction of the device position may be achieved byengaging one of the anchoring hooks and pulling witha snare catheter (39). To increase the leverage effect, thesnaring maneuver can be performed via transbrachial access.However, this corrective maneuver is not without hazards(e.g., aortic dissection) and bears the potential risk of THVembolization into the ascending aorta (6,40,45). If theattempt to reposition a Medtronic CoreValve with the snaretechnique fails, valve-in-valve implantation can be consid-ered to prevent conversion to emergency open heart surgery.

Interventional Closure

Interventional closure of paravalvular leaks after TAVR hasbeen described for the Edwards SAPIEN prosthesis (51,52)(Fig. 8). If the implantation depth of a THV is appropriateand the THV is not undersized, balloon post-dilation can bethefirst step to obtain a better expansion of the prosthesis stentframe with improved sealing of the paravalvular space.However, if significant PAR remains because of heavy calci-fications of the native aortic valve and a localizedAR jet can beidentified, transcatheter device closure with use of theAmplatzer Vascular Plug III (AVP III, AGAMedical Corp.,Plymouth, Minnesota) can be attempted analogous to para-valvular leak closure in surgical heart valves (53). However,potential risks associated with transcatheter device closure ofparavalvular leaks after TAVR include stroke, THVdislodgment, and embolization of the closure device (52).

Future Developments

The reduction in the incidence and severity of PAR representsan obvious target for technical improvements in the designof upcoming “next-generation” THVs and of implantationtechniques (4): sealing mechanisms at the lower part of theprosthesis skirt that help to conform to irregular surfaces ofthe native anatomy or full repositionability and retrievabilityto achieve an optimal procedural result with proper placementof the prosthesis and a lesser extent of PAR.

Conclusions

For the evaluation of PAR after TAVR, a multimodalapproach with the use of hemodynamic measurements andimaging modalities is imperative to precisely quantify theseverity of AR immediately after valve implantation and toidentify patients who will benefit from corrective measures,such as post-dilation or valve-in-valve implantation. Everymeasure has to be taken to prevent or reduce PAR to providea satisfying long-term clinical outcome.

Reprint requests and correspondence: Dr. Nikos Werner,Medizinische Klinik und Poliklinik II, Universitätsklinikum Bonn,Sigmund-Freud-Str. 25, 53105 Bonn, Germany. E-mail: nikos.werner@ukb.uni-bonn.de.

REFERENCES

1. Leon MB, Smith CR, Mack M, et al. Transcatheter aortic-valveimplantation for aortic stenosis in patients who cannot undergosurgery. N Engl J Med 2010;363:1597–607.

2. Smith CR, Leon MB, Mack MJ, et al. Transcatheter versus surgicalaortic-valve replacement in high-risk patients. N Engl J Med 2011;364:2187–98.

3. Rodés-Cabau J. Transcatheter aortic valve implantation: current andfuture approaches. Nature Pub Group 2011:1–15.

4. Sinning JM, Werner N, Nickenig G, Grube E. Transcatheter aorticvalve implantation: upcoming new devices. Interv Cardiol Clin 2012;1:37–43.

5. Webb JG, Wood DA. Current status of transcatheter aortic valvereplacement. J Am Coll Cardiol 2012;60:483–92.

6. Sinning JM, Hammerstingl C, Vasa-Nicotera M, et al. Aortic regur-gitation index defines severity of peri-prosthetic regurgitation andpredicts outcome in patients after transcatheter aortic valve implanta-tion. J Am Coll Cardiol 2012;59:1134–41.

7. Sinning JM, Ghanem A, Steinhäuser H, et al. Renal function aspredictor of mortality in patients after percutaneous transcatheter aorticvalve implantation. J Am Coll Cardiol Intv 2010;3:1141–9.

8. Abdel-Wahab M, Zahn R, Horack M, et al. Aortic regurgitation aftertranscatheter aortic valve implantation: incidence and early outcome.Results from the German transcatheter aortic valve interventionsregistry. Heart 2011;97:899–906.

9. Tamburino C, Capodanno D, Ramondo A, et al. Incidence andpredictors of early and late mortality after transcatheter aortic valveimplantation in 663 patients with severe aortic stenosis. Circulation2011;123:299–308.

10. Kodali SK, Williams MR, Smith CR, et al. Two-year outcomes aftertranscatheter or surgical aortic-valve replacement. N Engl J Med 2012;366:1686–95.

11. Gilard M, Eltchaninoff H, Iung B, et al. Registry of transcatheteraortic-valve implantation in high-risk patients. N Engl J Med 2012;366:1705–15.

12. Moat NE, Ludman P, de Belder MA, et al. Long-term outcomes aftertranscatheter aortic valve implantation in high-risk patients with severeaortic stenosis: the U.K. TAVI (United Kingdom Transcatheter AorticValve Implantation) Registry. J Am Coll Cardiol 2011;58:2130–8.

13. Athappan G, Patvardhan E, Tuzcu EM, et al. Incidence, Predictorsand Outcomes of Aortic Regurgitation after Transcatheter Aortic ValveReplacement: Meta-analysis and Systematic Review of Literature. J AmColl Cardiol 2013;61:1585–95.

14. Hayashida K, Morice MC, Chevalier B, et al. Sex-related differences inclinical presentation and outcome of transcatheter aortic valve implan-tation for severe aortic stenosis. J Am Coll Cardiol 2012;59:566–71.

15. Hayashida K, Lefèvre T, Chevalier B, et al. Impact of post-proceduralaortic regurgitation on mortality after transcatheter aortic valveimplantation. J Am Coll Cardiol Intv 2012;5:1247–56.

16. Gotzmann M, Korten M, Bojara W, et al. Long-term outcome ofpatients with moderate and severe prosthetic aortic valve regurgitation

Sinning et al. JACC Vol. 62, No. 1, 2013Paravalvular Aortic Regurgitation After TAVR July 2, 2013:11–20

20

after transcatheter aortic valve implantation. Am J Cardiol 2012;110:1500–6.

17. Anderson RH, Devine WA, Ho SY, Smith A, McKay R. The myth ofthe aortic annulus: The anatomy of the subaortic outflow tract. AnnThorac Surg 1991;52:640–6.

18. Piazza N, De Jaegere P, Schultz C, Becker AE, Serruys PW,Anderson RH. Anatomy of the aortic valvar complex and its implica-tions for transcatheter implantation of the aortic valve. Circ CardiovascInterv 2008;1:74–81.

19. Schultz CJ, Moelker A, Piazza N, et al. Three dimensional evaluationof the aortic annulus using multislice computer tomography: aremanufacturer’s guidelines for sizing for percutaneous aortic valvereplacement helpful? Eur Heart J 2010;31:849–56.

20. Ho SY. Structure and anatomy of the aortic root. Eur J Echocardiogr2009;10:i3–10.

21. Jilaihawi H, Kashif M, Fontana G, et al. Cross-sectional computedtomographic assessment improves accuracy of aortic annular sizing fortranscatheter aortic valve replacement and reduces the incidence ofparavalvular aortic regurgitation. J Am Coll Cardiol 2012;59:1275–86.

22. Buellesfeld L, Grube E. A permanent solution for a temporaryproblem. J Am Coll Cardiol Intv 2012;5:578–81.

23. John D, Buellesfeld L, Yuecel S, et al. Correlation of device landingzone calcification and acute procedural success in patients undergoingtranscatheter aortic valve implantations with the self-expanding Cor-eValve prosthesis. J Am Coll Cardiol Intv 2010;3:233–43.

24. Roberts WC, Janning KG, Ko JM, Filardo G, Matter GJ. Frequency ofcongenitally bicuspid aortic valves in patients �80 years of ageundergoing aortic valve replacement for aortic stenosis (with or withoutaortic regurgitation) and implications for transcatheter aortic valveimplantation. Am J Cardiol 2012;109:1632–6.

25. Wijesinghe N, Ye J, Rodés-Cabau J, et al. Transcatheter aortic valveimplantation in patients with bicuspid aortic valve stenosis. J Am CollCardiol Intv 2010;3:1122–5.

26. Himbert D, Pontnau F, Messika-Zeitoun D, et al. Feasibility andoutcomes of transcatheter aortic valve implantation in high-risk patientswith stenotic bicuspid aortic valves. Am J Cardiol 2012;110:877–83.

27. Gotzmann M, Lindstaedt M, Mügge A. From pressure overload tovolume overload: aortic regurgitation after TAVI. Am Heart J 2012;163:903–11.

28. Azadani AN, Jaussaud N, Matthews PB, et al. Energy loss due toparavalvular leak with transcatheter aortic valve implantation. AnnThorac Surg 2009;88:1857–63.

29. Jabbour A, Ismail TF, Moat N, et al. Multimodality imaging intranscatheter aortic valve implantation and post-procedural aorticregurgitation. J Am Coll Cardiol 2011;58:2165–73.

30. Husser O, Rauch S, Endemann DH, et al. Impact of three-dimensional transesophageal echocardiography on prosthesis sizingfor transcatheter aortic valve implantation. Catheter Cardiovasc Interv2012;80:956–63.

31. Tops LF, Wood DA, Delgado V, et al. Noninvasive evaluation of theaortic root with multislice computed tomography. J Am Coll CardiolImg 2008;1:321–30.

32. Willson AB, Webb JG, LaBounty TM, et al. 3-dimensional aorticannular assessment by multidetector computed tomography predictsmoderate or severe paravalvular regurgitation after transcatheter aorticvalve replacement. J Am Coll Cardiol 2012;59:1287–94.

33. Schueler R, Sinning JM, Momcilovic D, et al. Three-dimensionalspeckle-tracking analysis of left ventricular function after transcatheteraortic valve implantation. J Am Soc Echocardiogr 2012;25:827–34.

34. Vasa-Nicotera M, Sinning JM, Chin D, et al. Impact of paravalvularleakage on outcome in patients after transcatheter aortic valveimplantation. J Am Coll Cardiol Intv 2012;5:858–65.

35. Sellers RD, Levy MJ, Amplatz K, Lillehei CW. Left retrograde car-dioangiography in acquired cardiac disease: technic, indications, andinterpretations in 700 cases. Am J Cardiol 1964;14:437–47.

36. Michel PL, Vahanian A, Besnainou F, Acar J. Value of qualitativeangiographic grading in aortic regurgitation. Eur Heart J 1987;8Suppl C:11–4.

37. Kappetein AP, Head SJ, Généreux P, et al. Updated standardizedendpoint definitions for transcatheter aortic valve implantation. TheValve Academic Research Consortium-2 Consensus Document. J AmColl Cardiol 2012;60:1438–54.

38. Zoghbi WA, Chambers JB, Dumesnil JG, et al. Recommendations forevaluation of prosthetic valves with echocardiography and Dopplerultrasound. J Am Soc Echocardiogr 2009;22:975–1014.

39. Ussia GP, Barbanti M, Immè S, et al. Management of implant failureduring transcatheter aortic valve implantation. Catheter CardiovascInterv 2010;76:440–9.

40. Takagi K, Latib A, Al-Lamee R, et al. Predictors of moderate-to-severe paravalvular aortic regurgitation immediately after CoreValveimplantation and the impact of postdilatation. Catheter CardiovascInterv 2011;78:432–43.

41. Schultz C, Rossi A, van Mieghem N, et al. Aortic annulus dimensionsand leaflet calcification from contrast MSCT predict the need forballoon post-dilatation after TAVI with the Medtronic CoreValveprosthesis. EuroIntervention 2011;7:564–72.

42. Ussia GP, Barbanti M, Ramondo A, et al. The valve-in-valve tech-nique for treatment of aortic bioprosthesis malposition. J Am CollCardiol 2011;57:1062–8.

43. Nombela-Franco L, Rodés-Cabau J, DeLarochellière R, et al.Predictive factors, efficacy, and safety of balloon post-dilation aftertranscatheter aortic valve implantation with a balloon-expandable valve.J Am Coll Cardiol Intv 2012;5:499–512.

44. Toggweiler S, Wood DA, Rodés-Cabau J, et al. Transcatheter valve-in-valve implantation for failed balloon-expandable transcatheteraortic valves. J Am Coll Cardiol Intv 2012;5:571–7.

45. Eggebrecht H, Doss M, Schmermund A, Nowak B, Krissel J,Voigtländer T. Interventional options for severe aortic regurgitationafter transcatheter aortic valve implantation: balloons, snares, valve-in-valve. Clin Res Cardiol 2012;101:503–7.

46. Koos R, Mahnken AH, Dohmen G, et al. Association of aortic valvecalcification severity with the degree of aortic regurgitation aftertranscatheter aortic valve implantation. Int J Cardiol 2011;150:142–5.

47. Webb JG, Binder RK. Post-dilating transcatheter heart valves. J AmColl Cardiol Intv 2012;5:513–4.

48. Sinning JM, Vasa-Nicotera M, Ghanem A, et al. An Exceptional caseof frame underexpansion of a self-expandable transcatheter heart valvedespite predilation. J Am Coll Cardiol Intv 2012;5:1288–9.

49. Gerckens U, Latsios G, Mueller R, et al. Procedural and mid-termresults in patients with aortic stenosis treated with implantation of 2(in-series) CoreValve prostheses in 1 procedure. J Am Coll Cardiol Intv2010;3:244–50.

50. Hammerstingl C, Nickenig G, Grube E. Treatment of a degenerativestenosed CoreValve� aortic bioprosthesis by transcatheter valve-in-valve insertion. Catheter Cardiovasc Interv 2011;79:748–55.

51. Sinning JM, Vasa-Nicotera M, Werner N, Nickenig G,Hammerstingl C. Interventional closure of paravalvular leakage aftertranscatheter aortic valve implantation. Eur Heart J 2012;33:2498.

52. Whisenant B, Jones K, Horton KD, Horton S. Device closure ofparavalvular defects following transcatheter aortic valve replacementwith the Edwards Sapien valve. Catheter Cardiovasc Interv 2013;81:901–5.

53. Rihal CS, Sorajja P, Booker JD, Hagler DJ, Cabalka AK. Principles ofpercutaneous paravalvular leak closure. J Am Coll Cardiol Intv 2012;5:121–30.

Key Words: ARI - paravalvular aortic regurgitation - periprostheticregurgitation - TAVI - TAVR.

APPENDIX

For supplemental videos and their legends, please see the online version ofthis article.