Journal of the American College of Cardiology
`© 2013 by the American College of Cardiology Foundation
`Published by Elsevier Inc.
`
`STATE-OF-THE-ART PAPER
`
`Vol. 61, No. 11, 2013
`ISSN 0735-1097/$36.00
`http://dx.doi.org/10.1016/j.jacc.2012.08.1039
`
`Paravalvular Leak After
`Transcatheter Aortic Valve Replacement
`The New Achilles’ Heel? A Comprehensive Review of the Literature
`
`Philippe Généreux, MD,*†‡ Stuart J. Head, MSC,§ Rebecca Hahn, MD,*† Benoit Daneault, MD,*†
`Susheel Kodali, MD,*† Mathew R. Williams, MD,*† Nicolas M. van Mieghem, MD,储
`Maria C. Alu, MM,* Patrick W. Serruys, MD, PHD,储 A. Pieter Kappetein, MD, PHD,§
`Martin B. Leon, MD*†
`New York, New York; Montréal, Québec, Canada; and Rotterdam, the Netherlands
`
`Paravalvular leak (PVL) is a frequent complication of transcatheter aortic valve replacement (TAVR) and is seen
`at a much higher rate after TAVR than after conventional surgical aortic valve replacement. Recent reports indi-
`cating that PVL may be correlated with increased late mortality have raised concerns. However, the heterogene-
`ity of methods for assessing and quantifying PVL, and lack of consistency in the timing of such assessments, is
`a hindrance to understanding its true prevalence, severity, and effect. This literature review is an effort to consol-
`idate current knowledge in this area to better understand the prevalence, progression, and impact of post-TAVR
`PVL and to help direct future efforts regarding the assessment, prevention, and treatment of this troublesome
`complication.
`(J Am Coll Cardiol 2013;61:1125–36) © 2013 by the American College of Cardiology Foundation
`
`Transcatheter aortic valve replacement (TAVR) has become
`the treatment of choice for inoperable patients with severe
`aortic stenosis (1) and is comparable to surgical aortic valve
`replacement (SAVR) for patients at high risk (2). However,
`paravalvular leak (PVL) is more frequently seen after TAVR
`than after SAVR, and its potential association with mortal-
`ity has raised concerns (3–6). Moreover, recent reports have
`suggested that PVL could negatively impact mid- and
`long-term prognosis following TAVR (7,8). Although con-
`cerning, the lack of standardized quantitative and qualitative
`methods to assess and categorize PVL and the heterogene-
`ity in the timing of post-procedural assessment of PVL
`warrant caution in interpretation of these data. Therefore,
`we sought to perform a systematic review of the current
`literature to better define the rate, progression over time,
`
`From the *Columbia University Medical Center/New York Presbyterian Hospital,
`New York, New York; †Cardiovascular Research Foundation, New York, New York;
`‡Hôpital du Sacré-Coeur de Montréal, Montréal, Québec, Canada; §Department of
`Cardiothoracic Surgery, Erasmus University Medical Center, Rotterdam, the Neth-
`erlands; and the 储Department of Cardiology, Erasmus University Medical Center,
`Rotterdam, the Netherlands. Dr. Généreux has received speaker honoraria, consulting
`fees, and research grants from Edwards Lifesciences. Dr. Kodali has received
`consulting fees from Edwards Lifesciences and St. Jude Medical. Dr. Kappetein is
`member of a steering committee of the SURTAVI (Surgical Replacement and
`Transcatheter Aortic Valve Implantation) trial sponsored by Medtronic. Dr. Leon is
`a nonpaid member of the scientific advisory board of Edwards Lifesciences. All other
`authors have reported that they have no relationships relevant to the contents of this
`paper to disclose. Drs. Généreux and Head are joint first authors.
`Manuscript received June 27, 2012; revised manuscript received August 7, 2012,
`accepted August 21, 2012.
`
`predictors, and consequences of PVL after TAVR. Further-
`more, recommendations for measuring PVL are provided to
`improve consistency throughout the literature.
`
`Rate of PVL
`
`Multiple studies have reported the frequency and severity of
`PVL after TAVR (9). There is, however, significant heter-
`ogeneity that is caused by differences in: 1) imaging modal-
`ities (transthoracic echocardiography, transesophageal echo-
`cardiography, angiography); 2)
`timing of assessment
`(immediately after implantation, before discharge, at 30
`days); 3) transcatheter heart valve (THV) system; 4) grading
`scale; and 5) adjudication of events. When PVL was
`evaluated before hospital discharge and without central core
`laboratory analysis, its absence was reported in 6% to 59% of
`patients, whereas moderate or severe PVL was seen in 0% to
`24% (1–5,10–16) (Table 1).
`Thus far, only the PARTNER (Placement of Aortic
`Transcatheter Valve) trial has used a central echocardiogra-
`phy core laboratory to evaluate PVL (1,2). PVL was graded
`in accordance with the American Society of Echocardiog-
`raphy recommendations for native valves (17) because there
`were no recommendations for prosthetic valve assessment
`at the start of the trial. In addition, because of the
`inevitable eccentric nature of the jet and the frequent
`“spray” of the jet contour in the outflow tract, the color
`Doppler in the available parasternal short-axis view(s)
`was weighted in a subjective fashion more heavily than
`
`Downloaded From: http://content.onlinejacc.org/ on 08/03/2016
`
`Page 01 of 12
`
`

`

`1126
`
`Généreux etal.
`Paravalvular Leak After TAVR
`
`JACC Vol. 61, No. 11, 2013
`March 19, 2013:1125–36
`
`AV ⴝ atrioventricular
`
`LV ⴝ left ventricle/
`ventricular
`
`PVL ⴝ paravalvular leak
`
`SAVR ⴝ surgical aortic
`valve replacement
`
`TAVR ⴝ transcatheter
`aortic valve replacement
`
`THV ⴝ transcatheter heart
`valve
`
`Abbreviations
`and Acronyms
`
`AR ⴝ aortic regurgitation
`
`other signals in providing an
`integrated assessment. The fol-
`lowing definition was applied:
`no PVL (no regurgitant color
`flow),
`trace (pinpoint
`jet
`in
`atrioventricular [AV] short-axis
`view), mild (jet arc length ⬍10%
`of the AV annulus short-axis view
`circumference), moderate (jet arc
`length 10% to 30% of the AV
`annulus short-axis view circumfer-
`ence), and severe (jet arc length
`⬎30% of the AV annulus short-
`axis view circumference). In the
`PARTNER trial, trace/mild PVL
`was found in 66% of patients and moderate/severe in
`12% (1,2).
`Thus far, no prospective direct comparison of the rate of
`PVL after TAVR has been published between the 2 most
`frequently used THV systems (balloon-expandable THV,
`Edwards Lifesciences, Irvine, California; self-expandable
`CoreValve THV, Medtronic, Minneapolis, Minnesota).
`However, moderate to severe post-procedural PVL seems to
`be slightly higher with the CoreValve (9% to 21%) (
`)
`than the Edwards (6% to 13.9%) (1–3,5,18,21,22) device.
`Recent 1-year data presented from the FRANCE 2 (French
`Aortic National CoreValve and Edwards 2) Registry seemed to
`confirm this finding—the use of self-expandable prosthesis was
`identified as one of the major determinants of significant PVL
`after TAVR. At patient discharge, self-expandable prosthesis
`was associated with a moderate to severe PVL rate of 19.8%,
`compared with 12.2% for balloon-expandable prosthesis
`(p value not available) (23).
`
`Progression Over Time
`
`One of the initial concerns about PVL was potential
`worsening during extended follow-up. Because a large
`percentage of patients are discharged with trace or mild
`PVL, worsening of PVL could have important conse-
`quences on the volume load imposed on the left ventricle
`(LV), ultimately resulting in significant heart failure. In
`addition,
`if many cases progress to clinically significant
`leakage, hemolysis requiring repeated transfusions or reop-
`eration may further complicate the course of patients.
`Despite the lack of “common language” among previous
`reports in assessment of PVL severity, several studies have
`reported comparable findings with respect to time trends of
`PVL severity. Webb et al. (24) reported the evolution of
`PVL over time in a cohort of 168 patients and found
`that PVL was generally mild and remained stable between
`30-day and 1-year follow-ups, a result
`that has been
`confirmed by other studies (Table 2). A recent report by
`Ussia et al. (16) showed that rates of mild (53%) and
`moderate (15%) post-procedural PVL had been reduced to
`47% and 10%, respectively, at a follow-up of 3 years. Some
`
`attrition of the “sickest” patients might have been due to
`patients with worsening PVL dying, but there were no cases
`of worsening from mild to moderate/severe regurgitation in
`individual patient progression of PVL.
`Data from the PARTNER trial suggested, however, that
`PVL at 2 years had increased by ⱖ1 grade in 22.4% of
`patients, whereas it remained unchanged in 46.2% and
`improved by ⱖ1 grade in 31.5% of patients (Fig. 1) (8). So
`far, no studies have explored the mechanisms behind im-
`provement or worsening of PVL in individual patients, and
`measurement methods may explain, at least in part, these
`changes.
`
`Impact on Clinical Outcomes
`
`After SAVR, moderate to severe residual aortic regurgita-
`tion (AR) occurs infrequently in approximately 4% of
`patients (25). A recent study showed that AR after SAVR
`was an independent predictor of long-term mortality with a
`hazard ratio of 1.7 (95% CI: 1.2 to 2.3). The TAVR
`community has focused extensively on the effect of AR on
`survival because its prevalence is much higher after TAVR
`than after SAVR (8). A number of studies have identified
`AR ⱖ2⫹ to be an independent predictor of short- and
`long-term mortality (Table 3) (3). Furthermore, patients
`with AR ⱖ2⫹ were 10 times more likely to be nonre-
`sponders to therapy at 6 months’ follow-up; nonresponsive-
`ness was defined as either death or New York Heart
`Association classification ⱖ2.
`Few studies have devoted analyses specifically to PVL.
`This is not surprising because the low post-operative rate of
`PVL in surgical series makes statistical analysis not mean-
`ingful. However, even in TAVR after which post-
`procedural AR is largely paravalvular, there have been only
`a few large registries and randomized trials focused on PVL.
`Data on 663 patients from the Italian registry found that
`PVL grade ⱖ2⫹ was not associated with early 30-day
`mortality, but multivariate analysis did find a hazard ratio of
`3.79 for patients with PVL ⱖ2⫹ for late mortality beyond
`30 days (6). More disturbingly, although it was generally
`believed that only moderate or severe regurgitation would
`impact long-term outcomes (26), the recently published
`2-year results from the PARTNER trial showed that even
`mild PVL was associated with significant mortality (Fig. 2)
`(8). Multivariable analyses did not identify AR or PVL as
`independent predictors of mortality in this trial, but, inter-
`estingly, there is a trend toward improved survival
`in
`patients undergoing TAVR compared with SAVR if PVL
`was negligible (70% vs. 65%).
`Importantly, based on the current literature, the direct
`causal relationship between PVL and mortality (vs. PVL
`being a marker for other factors) still needs to be deter-
`mined. Careful analyses of baseline patient characteristics,
`the repercussion of all degrees of PVL on LV geometry and
`remodeling, and the determination of the precise cause of
`death (cardiovascular vs. noncardiovascular) are needed to
`
`Downloaded From: http://content.onlinejacc.org/ on 08/03/2016
`
`Page 02 of 12
`
`

`

`JACC Vol. 61, No. 11, 2013
`March 19, 2013:1125–36
`
`Généreux etal.
`Paravalvular Leak After TAVR
`
`1127
`
`confirm the strength and the nature of this relationship. At
`this point, any previous observations linking PVL (espe-
`cially mild) with mortality should be considered hypothesis
`generating.
`
`Predictors of PVL
`
`Significant PVL most commonly results from: 1) incom-
`plete prosthesis apposition to the native annulus due to
`patterns or extent of calcification (11,27–30) or annular
`eccentricity (26); 2) undersizing of the device (10,31,32);
`and/or 3) malpositioning of the valve (33). These observa-
`tions seem to be true for both balloon-expandable and
`self-expandable THVs.
`Valve sizing has been shown to be one of the strongest
`predictors of PVL. A low cover index reflecting a lower
`degree of oversizing of the prosthesis based on transthoracic
`echocardiography annulus measurement predicts significant
`PVL (10). More recently, studies have evaluated the use of
`multidetector computed tomography (MDCT) for THV
`sizing, and MDCT showed good predictability and reduced
`rates of significant PVL (34–37). Furthermore, larger and
`eccentric annuli were identified as predictors of PVL in
`multiple studies and most likely reflect inadequate sizing of
`the THV (3,15,26). A smaller aortic valve area was found to
`predict PVL in one study, but this was likely because the
`smaller area indicates a greater degree of calcification (3).
`The extent of calcification and asymmetrical distribution, as
`well as the location of calcium on the aortic wall, valve
`commissure, or THV landing zone, as a predictor for PVL
`has been confirmed in several studies (11,26,29,37,38).
`In studies specifically evaluating the CoreValve (Medtronic), a
`lower depth of implantation and a greater angle between the
`aorta and LV outflow tract were found to predict PVL
`(14,15).
`
`Assessment of Paravalvular Regurgitation
`
`Angiographic and hemodynamic assessment. Aortic root
`angiography is an established tool for qualitative and semi-
`quantitative assessment of AR (39). It is readily available
`during the TAVR procedure and can be quickly and safely
`executed to provide essential information and initiate ad-
`junctive maneuvers if needed in case of significant (para)
`valvular AR. Typically, Sellers criteria are applied to grade
`AR (40): 1) grade 1 or mild AR corresponds to a small
`amount of contrast entering the LV during diastole without
`filling the entire cavity and clearing with each cardiac cycle;
`2) grade 2 or moderate AR corresponds to contrast filling of
`the entire LV in diastole but with less density compared
`with contrast opacification of the ascending aorta; 3) grade
`3 or moderate to severe AR corresponds to contrast filling of
`the entire LV in diastole equal in density to the contrast
`opacification of the ascending aorta; and 4) grade 4 or severe
`AR corresponds to contrast filling of the entire LV in
`diastole on the first beat with greater density compared with
`the contrast opacification of the ascending aorta. During the
`
`contrast injection, no material may cross the aortic valve
`leaflets (e.g., guidewires, catheters) because incomplete valve
`closure may artificially be generated, thus resulting in AR.
`Particularly with self-expanding systems, it is important to
`wait some time (empirically 10 min) after deployment of the
`bioprosthesis to allow the system to expand to its maximum.
`The downside of qualitative aortography AR assessment is
`that it relies on subjective interpretation of unidimensional
`images; therefore, interobserver and intraobserver variability
`can be an issue and additional contrast volume required.
`Moreover, it is difficult to determine the contribution of
`PVL and central AR.
`Classic findings of acute AR (acute drop in the aortic
`diastolic pressure with or without elevated LV end-diastolic
`pressure [LVEDP]) may be seen after TAVR and may be
`suggestive of moderate to severe AR. However,
`these
`findings must be interpreted with caution because the
`concomitant use of sedatives, vasopressors, inotropes, and
`intravenous fluids all impact hemodynamics, and the pres-
`ence of material through the aortic valve (e.g., wire) may
`interfere temporarily with the THV function. Recently, the
`AR index, the ratio of the end-diastolic gradient across the
`aortic valve bioprosthesis to systolic blood pressure ([ADP ⫺
`LVEDP]/ASP; ADP-aortic diastolic pressure, ASP-aortic
`systolic pressure), was described (41). An AR index ⬍25
`was associated with 1-year mortality. Although this associ-
`ation is interesting, more data and validation are needed to
`establish the role of this new index in the therapeutic
`decision process after TAVR.
`Echocardiographic assessment. Although the native valve
`regurgitation quantitative grading scheme has been advo-
`cated for the evaluation of prosthetic valve regurgitation
`(42), there are limited data to support the use of these
`parameters following TAVR. The majority of semiquanti-
`tative parameters for assessing AR apply to central regurgi-
`tant jets, which are more uniform, making semiquantitative
`grading schemes more reliable.
`Unlike central jets, paravalvular regurgitant jets are com-
`monly eccentric with crescentic, irregular orifices. Because
`these jets occur between the annulus and sewing ring, jet
`areas and lengths may not represent the same severity of
`regurgitation compared with central jets and these param-
`eters cannot be used to reliably assess regurgitant severity.
`Although guidelines suggest using the circumferential ex-
`tent of the regurgitant jet as a semiquantitative measure of
`severity (42), this parameter has not been validated against any
`quantitative parameters of regurgitation. Even if we accept the
`limited validation of this scheme for surgical prostheses, the
`anatomy and physiology of THVs are different than that of
`surgical valves. In the balloon-expandable valve, paravalvular
`regurgitation should be assessed just below the skirt; for central
`jets, the regurgitation should be assessed at the coaptation
`point of the leaflets. In addition, there is no scheme that
`specifically addresses the unusual regurgitation that accompa-
`nies the THV. The intact calcified cusps and annulus signifi-
`
`Downloaded From: http://content.onlinejacc.org/ on 08/03/2016
`
`Page 03 of 12
`
`

`

`JACC Vol. 61, No. 11, 2013
`March 19, 2013:1125–36
`
`Continuedonthenextpage
`
`disease
`
`●Peripheralvascular
`●Lowimplantation
`
`diameter
`
`●Largerannulus
`ⱖ2/4AR
`
`1128
`
`Généreux etal.
`Paravalvular Leak After TAVR
`
`deployment
`afterinitialMCV
`gradeofPVL
`correlationwith
`significant
`showed
`
`AgSandDLZ-CS
`
`noncoronarycups
`relationto
`
`●Depthofdevicein
`
`ascendingaorta
`LVOTand
`
`●Increaseangleof
`ⱖ2/4AR
`●Male
`●Renalfailure
`●Cardiogenicshock
`●Annulusbaseline
`●AVAbaseline
`ⱖ2/4AR
`
`experience
`●Operator’s
`●Lowcoverindex
`ⱖ2/4AR
`
`Analysis
`
`Multivariable
`
`PredictorsofARby
`
`4⫽0(0%)
`3⫽3(3.8%)
`2⫽13(16.5%)
`1⫽42(53.2%)
`0⫽21(26.6%)
`Finalresult(angiogram)
`4⫹⫽0(0%)
`3⫹⫽1(0.1%)
`2⫹⫽11(11.1%)
`1⫹⫽49(49.5%)
`0⫽38(38.4%)
`technique(angiogram)
`Earlyafteradjunctive
`4⫹⫽0(0%)
`3⫹⫽8(0.8%)
`2⫹⫽19(19.2%)
`1⫹⫽28(28.3%)
`0⫽35(35.4%)
`Earlypost-TAVR(angiogram)
`4⫽0(0%)
`3⫽4(8%)
`2⫽13(26.0%)
`1⫽24(48.0%)
`0⫽9(18.0%)
`Earlypost-TAVR(TTE)
`4⫽0(0%)
`3⫽7(14.0%)
`2⫽13(26.0%)
`1⫽27(54.0%)
`0⫽3(6.0%)
`Earlypost-TAVR(angiogram)
`4⫽2(0.3%)
`3⫽14(2.0%)
`2⫽103(14.9%)
`1⫽380(55.1%)
`0⫽191(27.7%)
`Earlypost-TAVR(angiogram)
`4⫽0(0%)
`3⫽4(5.0%)
`2⫽12(16.0%)
`1⫽53(72.0%)
`0⫽5(7.0%)
`Earlypost-TAVR(TEE)
`ARPost-TAVR
`
`Valve-in-valve⫽2/79
`Snaretechnique⫽1/79
`Post-dilation⫽21/79
`
`3–4⫽severe
`2⫽moderate
`1⫽mild
`0⫽absent
`
`Sitereported
`Echocardiogram
`Angiogram
`
`MCV
`
`(21.5%)
`SC⫽17
`TF⫽62(78.5%)
`
`79
`
`Takagi,2011(15)
`
`Valve-in-valve⫽3/100
`Snaretechnique⫽4/100
`Post-dilation⫽34/100
`
`4⫹
`3⫹
`2⫹
`
`⫹
`0 1
`
`Echocardiogram
`Angiogram
`
`MCV
`
`SC⫽3(3%)
`TF⫽97(97%)
`
`100
`
`John,2010(78)
`
`Sitereported
`Echocardiogram
`Angiogram
`
`MCV
`
`TF
`
`50
`
`Sherif,2010(14)
`
`Sitereported
`Angiogram
`
`MCV⫽580(84%)
`ES⫽110(16%)
`
`TAo⫽5
`SC⫽22
`TA⫽26
`TF⫽644
`
`690
`
`Abdel-Wahab,2011(3)
`
`—
`
`—
`
`Valve-in-valve⫽2/74
`Post-dilation⫽5/74
`
`4⫽severe
`3⫽moderate/severe
`2⫽moderate
`1⫽trivial/mild
`
`4⫽severe
`3⫽moderate/severe
`2⫽mild/moderate
`1⫽trace/mild
`0⫽absent
`
`4⫽severe
`3⫽moderate/severe
`2⫽mild/moderate
`1⫽trace/mild
`0⫽absent
`
`AdjunctiveTechniques
`
`SeverityGradation
`
`ImagingModality
`
`Prosthesis
`
`echocardiographist)
`Sitereported(blinded
`Echocardiogram(TEE)
`
`ES
`
`TA⫽28(38%)
`TF⫽46(62%)
`Approach
`
`74
`
`n
`
`Detaint,2009(10)
`FirstAuthor,Year(Ref.#)
`
`SelectedPublicationsReportingARAfterTAVR
`
`Table1
`
`Downloaded From: http://content.onlinejacc.org/ on 08/03/2016
`
`Page 04 of 12
`
`

`

`JACC Vol. 61, No. 11, 2013
`March 19, 2013:1125–36
`
`Généreux etal.
`Paravalvular Leak After TAVR
`
`1129
`
`cantly influence the location and shape of paravalvular jets;
`typically, these jets appear smaller and more irregular at the
`level of the intact/calcified cusps and larger just apical to the
`THV stent.
`Quantitative assessment of total AR, or advanced imag-
`ing techniques for assessing paravalvular regurgitant orifices,
`may be a more accurate way of assessing severity and thus a
`more accurate assessment of risk. Quantitative Doppler uses
`comparative flow measurements across a regurgitant valve
`and a nonregurgitant valve to calculate regurgitant volume
`or fraction (17). The effective regurgitant orifice area is then
`calculated by dividing the regurgitant volume by the velocity
`time integral of the regurgitant jet continuous wave spectral
`profile. Alternatively, the LV stroke volume calculated by
`2-dimensional (2D) biplane Simpson method of discs (
`) can
`be used in place of total (regurgitant plus forward) stroke
`volume; however, systematic underestimation of ventricular
`volumes has been reported for this method. Although this
`quantitative assessment has been largely validated in the
`literature (44–51), has shown reproducibility, and is endorsed
`by scientific authorities (17,52), it should be acknowledged that
`this assessment is based on 4 parameters, any one of which may
`be determined with significant inaccuracy.
`Three-dimensional (3D) echocardiography can overcome
`the limitations of 2D and standard Doppler measurements
`for quantifying regurgitation (43). Pitfalls of 2D LV imag-
`ing, including foreshortening, malrotation, and angulation,
`can be overcome by 3D imaging. However, limitations of
`3D imaging (lower line density and low volume rates) may
`reduce the utility of this method for assessing total stroke
`volume. Color Doppler 3D volumes can be useful for the
`identification and localization of regurgitation jets, as well as
`planimetry of the vena contracta area (53,54). This imaging
`modality may be particularly useful for post-TAVR assess-
`ment of PVL (55,56).
`With the increased use of multimodality imaging capable
`of 3D reconstruction of the aortic root (36,57–62), there
`has been intense interest in the shape of the annulus and
`appropriate sizing of the transcatheter heart valve to reduce
`PVL. The oval shape of the annulus has been well documented
`(
`), and a single sagittal plane measurement is
`significantly smaller than the coronary plane measurement.
`Algorithms using 3D imaging tools have been suggested to
`improve annular sizing and reduce PVL (34,35).
`Recently,
`the Valve Academic Research Consortium
`(VARC) published the VARC II definitions and suggested the
`use of TAVR-specific criteria for the assessment of AR and/or
`PVL after TAVR (Table 4) (66). Figures 2
`
`leak;SC⫽subclavian;TA⫽transapical;TAo⫽transaortic;TAVR⫽transcatheteraorticvalvereplacement;TEE⫽transesophagealechocardiography;TF⫽transfemoral;TTE⫽transthoracicechocardiography.
`AgS⫽Agatstonscore;AR⫽aorticregurgitation;AVA⫽aorticvalvereplacement;CEC⫽clinicaleventscommittee;DLZ-CS⫽device-landingzonecalcificationscore;ES⫽EdwardsSapien;LVOT⫽leftventricularoutflowtrack;MCV⫽MedtronicCoreValve;PVL⫽paravalvular
`
`—
`
`—
`
`—
`
`Analysis
`
`Multivariable
`
`PredictorsofARby
`
`AR⬎2⫽115(13.6%)
`ARⱖ1⫽516(61%)
`Severe⫽0(0%)
`Moderate⫽16(13%)
`Mild⫽55(45%)
`30-daysurvivorsonly
`Earlypost-TAVR
`Severe⫽2(1%)
`Moderate⫽23(16%)
`Mild⫽64(44%)
`Earlypost-TAVR
`
`ⱖ2PVL⫽139(21.0%)
`Post-TAVR
`
`surgery⫽6/850
`Conversiontoopen
`
`—
`
`surgery⫽5/663
`Conversiontoopen
`Valve-in-valve⫽139/663
`Post-dilation⫽68/663
`
`—
`
`—
`
`Sitereported
`Angiogram
`
`MCV⫽459(53%)
`ES⫽410(47%)
`
`Other⫽271
`TF⫽599
`
`870
`
`Moat,2011(5)
`
`Severe
`Moderate
`Mild
`
`Sitereported
`
`TTEquality)
`
`Angiogram(Ifpoor
`Echocardiogram
`
`independentCEC
`reviewedby
`
`Sitereported,events
`Echocardiogram
`
`MCV
`
`TF/SC
`
`145
`
`MCV
`
`TF
`
`663
`
`2011(4)
`Gotzmann,
`
`2011(6)
`Tamburino,
`
`ARPost-TAVR
`
`AdjunctiveTechniques
`
`SeverityGradation
`
`ImagingModality
`
`Prosthesis
`
`Approach
`
`n
`
`FirstAuthor,Year(Ref.#)
`
`Table1
`Continued
`
`Downloaded From: http://content.onlinejacc.org/ on 08/03/2016
`
`Page 05 of 12
`
`

`

`1130
`
`Généreux etal.
`Paravalvular Leak After TAVR
`
`JACC Vol. 61, No. 11, 2013
`March 19, 2013:1125–36
`
`
`
`Progression of Aortic and/or Paravalvular Regurgitation Over TimeTable 2
`
`First Author, Year (Ref. #)
`Paravalvular leakage
`Webb, 2009 (24)
`
`n
`
`168
`
`Muñoz-Garcia, 2011 (79)
`
`144
`
`Ussia, 2012 (16)
`
`181
`
`Ye, 2010 (80)
`
`Takagi, 2011 (15)
`
`71
`
`79
`
`Ewe, 2011 (81)
`
`107
`
`Godino, 2010 (82)
`
`137
`
`Significant
`Post-Procedural
`
`Significant at
`6 Months
`
`Significant at
`1 Year
`
`Significant at
`2 Years
`
`Significant at
`3 Years
`
`30 days
`2⫹ ⫽ 37%
`3⫹ ⫽ 5%
`72 h
`Mild ⫽ 40%
`Moderate ⫽ 23%
`Post-procedure
`Mild ⫽ 53%
`Moderate ⫽ 15%
`30 days
`Mild ⫽ 26%
`Moderate ⫽ 5%
`
`30 days
`1⫹ ⫽ 51%
`2⫹ ⫽ 20%
`3⫹ ⫽ 3%
`Post-procedure
`1⫹ ⫽ 58%
`2⫹ ⫽ 16%
`3⫹ ⫽ 5%
`Post-procedure
`1⫹ ⫽ ⬇60%
`2⫹ ⫽ ⬇12%
`3⫹ ⫽ 4%
`4⫹ ⫽ 2%
`
`—
`
`“Stable”
`
`Mild ⫽ 47%
`Moderate ⫽ 19%
`
`—
`
`—
`
`—
`
`—
`
`—
`
`—
`
`—
`
`1⫹ ⫽ 49%
`2⫹ ⫽ 27%
`3⫹ ⫽ 0%
`
`ⱖ6 months
`1⫹ ⫽ 51%
`2⫹ ⫽ 31%
`3⫹ ⫽ 0%
`1⫹ ⫽ ⬇65%
`2⫹ ⫽ ⬇9%
`3⫹ ⫽ ⬇5%
`4⫹ ⫽ 0%
`
`—
`
`Mild ⫽ 48%
`Moderate ⫽ 18%
`
`Mild ⫽ 50%
`Moderate ⫽ 17%
`
`Mild ⫽ 47%
`Moderate ⫽ 10%
`
`—
`
`—
`
`—
`
`—
`
`—
`
`—
`
`—
`
`“Remained unchanged
`and clinically
`insignificant
`during follow-up”
`—
`
`—
`
`—
`
`—
`
`Aortic regurgitation
`Bauer, 2010 (83)
`
`Rajani, 2010 (84)
`
`Clavel, 2009 (85)
`
`Lefevre, 2011 (86)
`
`Buellesfeld, 2011 (20)
`
`Bleiziffer, 2012 (87)
`
`Koos, 2011 (29)
`
`D’Onofrio, 2011 (88)
`
`88
`
`50
`
`2⫹ ⫽ 29%
`3⫹ ⫽ 7%
`46 Within 5 days
`Mild ⫽ 33%
`Moderate ⫽ 19%
`Moderate/severe ⫽ 5%
`Discharge
`Trivial ⫽ 38%
`Mild ⫽ 42%
`Moderate ⫽ 8%
`Severe ⫽ 0%
`Discharge
`2⫹ ⫽ 42%
`3⫹ ⫽ 5%
`30 days
`1⫹ ⫽ 32%
`2⫹ ⫽ 9%
`3⫹ ⫽ 0%
`Discharge
`Mild ⫽ 31%
`Mild/moderate ⫽ 13%
`Moderate ⫽ 8%
`Moderate/severe ⫽ 3%
`After implant
`1⫹ ⫽ 77%
`2⫹ ⫽ 9%
`3⫹ ⫽ 5%
`Discharge
`1⫹ ⫽ 30%
`2⫹ ⫽ 9%
`
`130
`
`126
`
`227
`
`57
`
`504
`
`Gurvitch, 2010 (21)
`
`70
`
`Walther, 2011 (22)
`
`168
`
`Post-procedure
`Trivial ⫽ 40%
`Mild ⫽ 44%
`Moderate ⫽ 6%
`—
`
`2⫹ ⫽ 24%
`3⫹ ⫽ 0%
`Mild ⫽ 31%
`Moderate ⫽ 8%
`Moderate/severe ⫽ 15%
`
`2⫹ ⫽ 25%
`3⫹ ⫽ 0%
`
`1⫹ ⫽ 34%
`2⫹ ⫽ 3%
`3⫹ ⫽ 0%
`
`2⫹ ⫽ 23%
`3⫹ ⫽ 0%
`
`—
`
`—
`
`—
`
`1⫹ ⫽ 37%
`2⫹ ⫽ 0%
`3⫹ ⫽ 0%
`
`Mild ⫽ 40%
`Mild/moderate ⫽ 16%
`Moderate ⫽ 6%
`Moderate/severe ⫽ 0.5%
`Severe ⫽ 0.5%
`—
`
`Mild ⫽ 41%
`Mild/moderate ⫽ 15%
`Moderate ⫽ 5%
`Moderate/severe ⫽ 1%
`Severe ⫽ 1%
`—
`
`—
`
`6–12 months
`Trivial ⫽ 26%
`Mild ⫽ 46%
`Moderate ⫽ 6%
`Severe ⫽ 0%
`—
`
`1⫹ ⫽ 39%
`2⫹ ⫽ 6%
`3⫹ ⫽ 0%
`
`Mild ⫽ 45%
`Mild/moderate ⫽ 11%
`Moderate ⫽ 6%
`Moderate/severe ⫽ 0%
`Severe ⫽ 0%
`Mean 83 ⫾ 80 days
`1⫹ ⫽ 82%
`2⫹ ⫽ 5%
`3⫹ ⫽ 0%
`
`—
`
`—
`
`Mean 9.2 ⫾ 6.5 months
`“No changes in the
`degree of AR were
`found”
`
`—
`
`3–6 months
`1⫹ ⫽ 51%
`2⫹ ⫽ 1%
`3⫹ ⫽ 0%
`
`1⫹ ⫽ 46%
`2⫹ ⫽ 5%
`3⫹ ⫽ 0%
`
`—
`
`—
`
`—
`
`—
`
`—
`
`—
`
`—
`
`—
`
`—
`
`—
`
`Trivial ⫽ 60%
`Mild ⫽ 33%
`Moderate ⫽ 3%
`
`—
`
`Downloaded From: http://content.onlinejacc.org/ on 08/03/2016
`
`Page 06 of 12
`
`

`

`JACC Vol. 61, No. 11, 2013
`March 19, 2013:1125–36
`
`Généreux etal.
`Paravalvular Leak After TAVR
`
`1131
`
`Figure 1
`
`Adapted with permission from Kodali et al. (
`
`).
`
`ization of the valve before implantation. In addition, simul-
`taneous “real-time” imaging, such as echocardiogram (both
`2D and 3D), 3D angiographic reconstruction via rotational
`aortic root angiogram (67), and the use of novel imaging
`systems (68,69), may assist in choosing intraprocedurally the
`optimal projection for THV positioning and deployment,
`leading potentially to less frequent PVL.
`Intraprocedurally, several
`interventional alternatives to
`reduce regurgitation are available (70). Severe calcification
`of the native valve might prevent the implanted valve from
`expanding completely against the annulus, leaving residual
`orifices through which PVL may occur. Post-implantation
`balloon dilation of the valve might be effective in reducing
`PVL and may be considered the initial option for patients
`with PVL (71). A slightly oversized balloon is recom-
`mended to fully expand the valve. Studies have shown that
`post-dilation can be safely performed, with a reduction of
`the regurgitation in a majority of patients (38). Calcification
`of the valve significantly influences the success of this
`intervention. However, in some patients, post-dilation has
`no effect in reducing AR (15); in addition, post-dilation has
`been shown to be associated with a higher incidence of
`cerebrovascular events (38). The effect of post-dilation on
`survival has yet to be determined.
`
`Outcomes Associated With Aortic and/or Paravalvular Regurgitation
`Table 3
`
`Figure 2
`
`Reprinted with permission from Kodali et al. (
`
`). HR ⫽ hazard ratio.
`
`Especially with the CoreValve, implantation of the valve
`that is too low is associated with PVL. Repositioning to a
`higher implantation depth could therefore reduce PVL.
`However, no retrievable valve is currently available on the
`market. Therefore, a snaring maneuver has been de-
`scribed, in which the valve is pulled up by attaching a
`snare to one of
`the frame loops (72,73). Although
`successful cases have been reported (74), the valve may
`also move to the original (too low) position as soon as
`tension is released (70). An extra word of caution is
`warranted when the snaring technique is considered in
`patients with extensively calcified valves because chunks
`of calcium may detach as a result of friction. Further-
`more, there is a risk of damaging the ascending aorta
`during the snaring maneuver.
`A valve-in-valve procedure may be necessary in some
`cases in which post-dilation or other techniques do not
`improve the degree of PVL. This is specifically indicated for
`patients in whom the valve was suboptimally positioned
`(i.e., too shallow or too deep). In the Italian registry, a
`valve-in-valve procedure was used in 3.6% of 663 patients
`
`First Author, Year (Ref. #)
`Abdel-Wahab, 2011 (3)
`Gotzmann, 2011 (4)
`
`n
`690
`122
`
`AR ⱖ2
`AR ⱖ2
`
`Variable
`
`Takagi, 2011 (15)
`Hayashida, 2012 (89)
`Leber, 2011 (90)
`Moat, 2011 (5)
`Sinning, 2012 (91)
`Tamburino, 2011 (6)
`Sinning, 2012 (41)
`Unbehaun, 2012 (26)
`Kodali, 2012 (8)
`
`AR ⱖ2
`41
`AR ⱖ2
`260
`AR ⬎2
`69
`AR ⱖ2
`870
`PVL ⱖ2
`152
`PVL ⱖ2
`663
`146 Moderate/severe PVL
`358
`No vs. trace vs. mild AR
`158 Mild to severe AR
`Mild to severe PVL
`
`Outcome
`In-hospital mortality
`6-month mortality
`No clinical improvement
`6-month mortality
`Median 217 days (IQR: 54–401)
`1-year mortality
`1-year mortality
`1-year mortality
`Late mortality
`1-year survival
`2-year survival
`2-year survival
`2-year survival
`
`Univariate Analysis
`OR ⫽ 2.50 (95% CI 1.37–4.55)
`—
`
`12.2% vs. 25.0% (p ⫽ 0.25)
`HR ⫽ 1.97 (95% CI 1.19–3.28)
`9% vs. 37.5% (95% CI p ⫽ 0.07)
`HR ⫽ 1.49 (95% CI 1.00–2.21)
`HR ⫽ 4.0 (95% CI 2.1–7.5)
`—
`HR ⫽ 3.9 (95% CI 2.0–7.5)
`66% vs. 72% vs. 67% (p ⫽ 0.77)
`HR ⫽ 1.75 (95% CI 1.17–2.61)
`HR ⫽ 2.11 (95% CI 1.43–3.10)
`
`Multivariate Analysis
`OR ⫽ 2.43 (95% CI 1.22–4.85)
`OR ⫽ 4.26 (95% CI 1.59–11.45)
`OR ⫽ 10.1 (95% CI 3.20–31.94)
`—
`—
`—
`HR ⫽ 1.66 (95% CI 1.10–2.51)
`HR ⫽ 4.9 (95% CI 2.5–9.6)
`HR ⫽ 3.79 (95% CI 1.57–9.10)
`HR ⫽ 2.4 (95% CI 1.0–5.4)
`—
`
`Not significant
`Not significant
`
`HR ⫽ hazard ratio; IQR ⫽ interquartile range; OR ⫽ odds ratio; other abbreviations as in
`
`.
`
`Downloaded From: http://content.onlinejacc.org/ on 08/03/2016
`
`Page 07 of 12
`
`

`

`1132
`
`Généreux etal.
`Paravalvular Leak After TAVR
`
`JACC Vol. 61, No. 11, 2013
`March 19, 2013:1125–36
`
`VARC II Recommendations for Evaluation of Aortic and/or Paravalvular Regurgitation After TAVR
`Table 4
`
`Semiquantitative parameters
`Diastolic flow reversal in the descending aorta—pulsed wave
`Circumferential extent of prosthetic valve paravalvular regurgitation (%)*
`Quantitative parameters†
`Regurgitant volume (ml/beat)
`Regurgitant fraction (%)
`Effective regurgitant orifice area (cm2)
`
`Mild
`
`Moderate
`
`Severe
`
`Absent or brief early diastolic
`⬍10
`
`Intermediate
`10–29
`
`Prominent, holodiastolic
`ⱖ30
`
`⬍30
`⬍30
`0.10
`
`30–59
`30–49
`0.10–0.29
`
`ⱖ60
`ⱖ50
`ⱖ0.30
`
`*Not well validated and may overestimate severity compared with quantitative Doppler. †For LVOT ⬎2.5 cm, significant stenosis criteria is ⬍0.20. Adapted with permission from Kappetein et al. (66).
`VARC ⫽ Valve Academic Research Consortium; other abbreviations as in Table 1.
`
`(75). Compared with patients who were implanted with a
`single valve, those who underwent valve-in-valve had similar
`safety and efficacy over a 1-year follow-up. Encouraging
`results have been reported from other series as well (76).
`As a final option for patients with continued severe PVL
`in whom interventional therapy does not suffice, conversion
`to conventional SAVR may be needed (77). SAVR may be
`undesirable because these patients are generally at high or
`extreme risk, but the procedure may be unavoidable in some
`cases.
`
`Emerging TAVR Technologies
`
`Currently, there is no proven or generally accepted treat-
`ment for PVL. However, there are emerging THV systems
`and technologies that are promising in minimizing PVL
`after TAVR (Fig. 5). These devices may reduce PVL by
`better supra-, infra-, or intra-annular sealing (cuff) or by
`allowing controlled deployment, repositioning, or removal
`of the THV. Preimplantation calcification debulking (sur-
`gically or not) also remains one of the most interesting areas
`
`Figure 3
`
`(A) Post–transcatheter heart valve (THV) left ventricular outflow tract (LVOT) diameter (just apical to the THV stent). (B) Right ventricular outflow tract (RVOT) diameter.
`(C) LVOT Doppler with sample volume located just apical to the THV stent aligned in the short-axis view of the