�������� ����� ��

A critical review of hemodynamic changes and left ventricular remodelingfollowing surgical aortic valve replacement and percutaneous aortic valvereplacement

Shin-Jae Kim M.D., Zainab Samad M.D., Gerald S. Bloomfield M.D.,M.P.H, Pamela S. Douglas M.D.

PII: S0002-8703(14)00274-9DOI: doi: 10.1016/j.ahj.2014.04.015Reference: YMHJ 4618

To appear in: American Heart Journal

Received date: 29 October 2013Accepted date: 15 April 2014

Please cite this article as: Kim Shin-Jae, Samad Zainab, Bloomfield Gerald S., DouglasPamela S., A critical review of hemodynamic changes and left ventricular remodelingfollowing surgical aortic valve replacement and percutaneous aortic valve replacement,American Heart Journal (2014), doi: 10.1016/j.ahj.2014.04.015

This is a PDF file of an unedited manuscript that has been accepted for publication.As a service to our customers we are providing this early version of the manuscript.The manuscript will undergo copyediting, typesetting, and review of the resulting proofbefore it is published in its final form. Please note that during the production processerrors may be discovered which could affect the content, and all legal disclaimers thatapply to the journal pertain.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

1

A Critical Review of Hemodynamic Changes and Left

Ventricular Remodeling Following Surgical Aortic Valve

Replacement and Percutaneous Aortic Valve Replacement

Shin-Jae Kim, M.D.* ‡, and Zainab Samad, M.D. †, Gerald S. Bloomfield, M.D., M.P.H*†

and Pamela S. Douglas, M.D.* †

*Duke Clinical Research Institute, Durham, NC

† Division of Cardiology, Duke University Medical Center, Durham, NC

‡ Division of Cardiology, Ulsan University Hospital, University of Ulsan College of

Medicine, Ulsan, South Korea

Brief title: Hemodynamics following aortic valve replacement

Funding: None

Correspondence to Shin-Jae Kim, M.D.

Division of Cardiology

Ulsan University Hospital

University of Ulsan College of Medicine

877 Bangeojinsunhwan-doro, Dong-gu

Ulsan, South Korea

Telephone 82-52-250-7029

FAX 82-52-251-8235

E-mail: kimsc226@gmail.com

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

2

Abstract

Background: The introduction of transcatheter aortic valve replacement (TAVR) in clinical

practice has widened options for symptomatic patients at high surgical risk, however, it is not

known whether TAVR has equivalent or prolonged benefits in terms of LV remodeling.

Methods: To explore the relative hemodynamic benefits and postoperative LV remodeling

associated with TAVR and SAVR, we performed a critical review of the available literature. A

total of 60 studies were included in this systematic review.

Results: There is at least equivalent if not slightly superior hemodynamic performance of

TAVR over SAVR and TAVR showed lower PPM compared with SAVR. However, LV mass

appears to regress to a greater degree after SAVR compared to TAVR. Aortic regurgitation,

paravalvular in particular, is more common after TAVR than SAVR although it is rarely more

than moderate in severity. Improvements in diastolic function and mitral regurgitation are

reported in only a handful of studies each and could not compared across prosthesis types.

Conclusions: The published data support the hemodynamic comparability of SAVR and

TAVR, with the higher incidence of PPM in SAVR offset by higher incidence of paravalvular

leak in TAVR. These results highlight the need for further studies focusing on hemodynamic

changes following valve therapy.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

3

Introduction

Aortic stenosis is frequently accompanied by left ventricular (LV) hypertrophy and

remodeling, which is independently associated with cardiac morbidity and mortality. 1 While

conventional surgical aortic valve replacement (SAVR) is effective in reducing afterload, wall

stress, and subsequently, wall thickness in most patients, some degree of left ventricular

hypertrophy (LVH) persists indefinitely after SAVR. 2 It is still controversial whether

incomplete regression of LVH is associated with poorer long-term survival. 3, 4

The introduction of transcatheter aortic valve replacement (TAVR) in clinical practice has

widened options for symptomatic patients at high surgical risk, however, it is not known

whether TAVR has equivalent or prolonged benefits in terms of hemodynamic improvement

and reverse LV remodeling. The Placement of Aortic Transcatheter Valves (PARTNER) I trial

echocardiographic data suggested slightly lower valve gradients and higher valve areas with

TAVR compared to SAVR, 5 but the higher prevalence of paravalvular regurgitation may

modulate the impact of these hemodynamic benefits on overall outcomes. 6 To further explore

the relative hemodynamics and the postoperative LV remodeling associated with TAVR and

SAVR, we performed a critical review of the published literature.

Methods

Data Sources

We performed searches on PubMed, Embase, and Cochrane databases for all studies

published in English between January 1, 1980 and June 30, 2013 related to SAVR and TAVR

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

4

outcomes. Our PubMed subject headings and keywords included “aortic stenosis,” “elderly,”

“operative,” “surgical,” “aortic valve replacement,” “percutaneous,” “aortic valve implantation,”

“left ventricular ejection fraction,” “left ventricular mass.” Publications were also restricted

with a publication date from 1980/01/01 to 2013/06/30. An analogous search was performed in

the other databases. We also reviewed the bibliographies of retrieved articles to obtain

additional citations.

Study selection criteria

Studies were considered eligible for this review if they evaluated the results of SAVR on

>= 30 patients with isolated AS, age more than 18 years, with at least 30 days follow-up in

patients with SAVR or the results of TAVR in >= 10 patients with at least immediate

postprocedural period follow up. Additionally, the included studies had to report sufficient data

to calculate the changes in left ventricular ejection fraction (LVEF) and LV mass (or LV mass

index) pre- and post-intervention. Studies were excluded if they did not report at least one of

three hemodynamic data of interest during both the preoperative (or preprocedural) and

postoperative (or post procedural) periods: effective orifice area (or effective orifice area index),

peak transaortic pressure gradient, or mean transaortic pressure gradient. Randomized

controlled trials, prospective cohort and retrospective cohort studies and cases series were

included. Case reports, review articles, abstracts, editorials and expert opinions were excluded.

We included the PARTNER Cohort A and Cohort B trials in this review. 6, 7

Data extraction and critical appraisal

Two investigators extracted all data from the relevant articles’ texts, tables, and figures.

Discrepancies between the reviewers were resolved by repeated review and discussion. If there

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

5

were duplicate data from a single participant population, these data were included only once in

the analysis. Extracted data included baseline clinical characteristics, number of patients

enrolled, sex, follow-up data, preoperative (or preprocedural) and postoperative (or

postpreprocedural) peak pressure gradient, mean gradient, or (indexed) effective orifice area,

LV mass index (or LV mass), left ventricular LVEF, diastolic parameters (E/A, IVRT, DT, E/e’,

left atrial size), aortic regurgitation, mitral regurgitation, and hemodynamic or

echocardiographic variables associated with mortality. We calculated the fractional change of

peak and mean pressure gradients, (indexed) effective orifice areas, LVEF and LV mass index

(or LV mass). Fractional decrease was calculated by the equation: [(preoperative value –

postoperative value)/preoperative value]  100 and fractional increase calculated by the

equation: [(postoperative value – preoperative value)/preoperative value]  100. Values

provided for each variable are weighted mean values.

No extramural funding was used to support this work. The authors are solely responsible for

the design and conduct of this study, all study analyses, the drafting and editing of the paper and

its final contents.

Results

Characteristics of Included Studies and Patients

A total of 579 studies were identified with the search terms or from bibliographies. We

excluded case reports, studies with multiple prostheses, studies not describing prosthesis type,

studies not satisfying the selection criteria, studies with imaging modalities other than

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

6

echocardiography and studies of other aortic valve procedures such as balloon valvuloplasty

and valve-in-valve implantation (Figure 1).

Sixty seven published studies involving 7101 patients with isolated AS and valve

replacement met our inclusion criteria (Online Table 1). Eighteen publications presented

mechanical SAVR data on 1791 patients with a mean age of 69 years (range 54-75) of whom

45 % were male. The 18 studies examining bioprosthetic SAVR presented data on 2073 patients

with a mean age of 75 years (range 68-85) of whom 50 % were male. The 38 studies of TAVR

presented data on 3237 patients with a mean age of 82 years (range 74-86) of whom 48 % were

male (Online Table 2). The higher age in the TAVR cohort is most likely due to limitation of its

use to those at high surgical risk with conventional SAVR.

Hemodynamic changes: mechanical SAVR, bioprosthetic SAVR and TAVR

The mean postoperative peak aortic pressure gradient in patients with SAVR with

mechanical prosthesis was 26 mmHg (range, 14 to 36 mmHg) representing a mean fractional

decrease of 71 % (range, 57-79%) from preoperative findings.Online 1-5, 7, 8, 10, 12, 14-18

In those who

underwent SAVR with a bioprosthesis, the mean postoperative peak pressure gradient was 23

mmHg (range, 15 to 32 mmHg), representing a mean fractional decrease of 71 % (range 59-82

%).Online 16, 18-26, 28-31, 33

After TAVR, the peak pressure gradient was 17 mmHg (range, 10 to 21

mmHg) with a mean fractional decrease of 77 % (range, 69-84%) (Online Table 2) (Figure 2A

and 2B).Online 19, 23, 25, 26, 35-42, 44-67

The average postoperative mean pressure gradient after SAVR

with mechanical prosthesis was 14 mmHg representing an average fractional decrease of 73 %

(range, 60-83 %). These values were 13 mmHg and 74 % (range, 57-83 %), respectively, after

SAVR with bioprosthesis and 10 mmHg and 79 % (range, 64-86 %), respectively, after TAVR

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

7

(Online Table 2)(Figure 2C and 2D). The mean fractional increase of indexed or non-indexed

effective orifice area (EOA) after SAVR with mechanical prosthesis was 127 % (range, 67-230

%),Online 1, 3, 5-8, 12, 14-16, 18

127 % (range, 25-248 %) after SAVR with bioprosthesis Online 16, 18, 19, 20,

22, 23, 25-27, 29-33 and 180 % (118-250 %) after TAVR

Online 3, 19, 23, 25, 26, 35-39, 41-44, 49-58, 60-67(Online

Table 2) (Figure 2E).

In studies noted above, there were two prospective, randomized, controlled trials that

compared the hemodynamic difference between SAVR and TAVR after intervention. 5, 6, 8

One

study showed that TAVR provided slightly better hemodynamic performance compared with

the SAVR with a stented valve in terms of postoperative mean transvalvular gradient (10 ± 4

mm Hg vs. 13 ± 5 mmHg at follow-up, p <0.05), whereas transcatheter valves hemodynamics

were functionally similar to the stentless surgical valves (10 ± 4 mm Hg vs. 9 ± 4 mmHg at

follow-up). 8 The other study showed that TAVR resulted in slightly lower mean transvalvular

gradients (10.2 ± 4.3 vs. 11.5 ± 5.4 mm Hg, respectively, p = 0.008) and higher aortic valve

areas than porcine SAVR (1.6 ± 0.5 vs. 1.4 ± 0.5 cm2, p = 0.002).

5 Of note, the PARTNER I

study restricted the selection of SAVR types and did not allow concomitant annuloplasty.

LV systolic function: SAVR vs. TAVR

Fifteen studies reported data on LVEF after mechanical prosthesis, Online 1-5, 7-12, 14-17

and 13

studies reported these data after bioprothesis. Online 16, 19-23, 25-27, 29-33

Among the 26 studies with

SAVR performed in patients with a preoperative LVEF above 50%, mean fractional increase of

LVEF was 4 % (range, -5 % to 27 %) determined over a range of final echocardiographic

follow-up of 1 to 102 months (Figure 3A). The 2 studies reporting results in patients with

preoperative LVEF =< 50% showed 14% fractional increase of LVEF after SAVR (Figure 3A).

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

8

Online 3, 14 In studies with short to mid-term follow-up (up to 12 months), the mean fractional

increase of LVEF was 6%, and in studies with long-term follow-up (above 12 months) the mean

fractional increase of LVEF was 4%.

Thirty eight studies reported the follow-up data of LVEF after TAVR. Online 3, 25, 26, 19, 23, 35-67

In the 26 studies with preoperative LVEF above 50% mean fractional increase of LVEF was 6%

(range, -2% to 20%) over a follow up period ranging from immediate to 24 months after TAVR

(Figure 3A). In 13 studies presenting data for patients with preoperative LVEF =< 50%, the

mean fractional increase of LVEF was 20% (range, -6% to 88%) (Figure 3A). Online 3, 38, 39, 46, 47,

50, 54, 65, 67 In studies with follow-up within 12 months, the mean fractional increase of LVEF

was 9%, and the 4 studies with follow-up over 12 months, Online 48-50, 60

the fractional increase of

LVEF was 10%. The fractional increase of LVEF was not closely correlated with the duration

of follow-up in either SAVR and TAVR (Figure 3B). Among included studies, there were only

two prospective, randomized, controlled trials that compared the difference of improvement of

LVEF after SAVR and TAVR. 5, 8

Neither study showed a difference of LVEF between groups

at follow-up after intervention. Thus, regardless of valve type, patients with low preoperative

(or preprocedural) LVEF showed a robust improvement in LVEF compared with patients with

preserved preoperative (or preprocedural) LVEF.

Regression of left ventricular hypertrophy: SAVR vs. TAVR

There was a prospective, randomized, controlled trial that reported the difference in

regression of LV hypertrophy between SAVR and TAVR after intervention.6 The SAVR group

had more LV mass regression early, but there was no difference of LV mass regression rate

over the follow-up period between SAVR and TAVR. In the current review, sixteen studies

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

9

reported the follow-up data of LV mass index (or LV mass) after SAVR with mechanical

prosthesis, Online 1,2, 4-16, 18

16 studies reported these data after SAVR with bioprosthesis, Online 18-

25, 27-34 11 studies reported these data after TAVR

Online 19, 37, 38, 42, 43, 48, 50, 54, 55, 57, 58 (Figure 4). In

studies with short to mid-term follow-up (median, 12 months) after SAVR, the mean fractional

decrease of LV mass index (or LV mass) was 23% (range, 2% to 35%), but in studies after

SAVR with long-term follow-up (median, 34 months), the mean fractional decrease of LV mass

index (or LV mass) was 27% (range, 18% to 52%). All of the studies after TAVR had short to

mid-term follow-up (median, 6 months) and the mean fractional decrease of LV mass index (or

LV mass) was 16% (range, -1% to 31%) (Figure 4A). In 12 studies with immediate to 12

months follow-up after SAVR, Online 4, 9, 14, 15, 18, 19, 28-32, 34

the fractional decreases of LV mass

index (or LV mass) were 10%, 19% and 22 % at immediate to 3 months, 6 months, and 12

months after SAVR, respectively. In 8 studies with immediate to 12 months follow-up after

TAVR, Online 19, 43, 45, 48, 52, 55, 57, 61

the fractional decreases of LV mass index (or LV mass) were

4%, 9% and 11% at immediate to 3months, 6 months, and 12 months after TAVR, respectively

(Figure 4B). In contrast to LV EF, LV mass appears to continue to regress beyond the early

postprocedural period, and possibly beyond one year.

LV diastolic function: SAVR vs. TAVR

There are no prospective, randomized, controlled trials that compare diastolic dysfunction

between SAVR and TAVR after intervention. In one observational study, the E/A ratio and

deceleration time remained unchanged after SAVR. 9 In studies after TAVR, E/A ratio, LA size

and E/e’ ratio decreased in 2 of 2 studies reporting data, 4 of 7 studies and 2 of 2 study,

respectively, but deceleration time remained unchanged in 2 of 2 studies after TAVR (Online

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

10

Table 2). 10, 11

Given the scarcity of the available data the relative effects of TAVR and SAVR

couldn’t be determined.

Prosthesis-patient mismatch: SAVR vs. TAVR

Varying definitions of prosthesis-patient mismatch (PPM) were used in different studies

including effective orifice area indexes of <0.75 cm2/m

2, <0.85 cm

2/m

2, ≤0.85 cm

2/m

2, or <0.9

cm2/m

2 (Online Table 3). There were two prospective, randomized, controlled trials that

evaluated the difference in incidence of PPM. One study showed that incidence of severe PPM

was significantly lower in the TAVR group (6%) compared with the surgical AVR using

stented valve (28%) and stentless valve (20%) groups,8 and the other study showed that the

incidence of PPM was lower in the TAVR group (42%) compared with the SAVR group (56%).

6 In studies after SAVR with mechanical prosthesis sizes of 17, 19, 21, 23, 25 mm, the

prevalence of PPM varied from 23% to 94% Online 1, 2, 5, 7, 8, 10, 11 and the prevalence of severe

PPM varied from 0% to 18%. Online 1, 5, 7, 8, 10

In studies after SAVR with bioprosthesis sizes of

19, 21 and 23mm, the prevalence of severe PPM varied from 7% to 36%. Online 19. 22, 23, 26

In

studies after TAVR with CoreValve (Medtronic, Minneapolis, Minnesota) the prevalence of any

PPM varied from 10% to 39%, Online 19, 35, 36, 42

while severe PPM varied from 2% to 16%. Online

19, 26, 35, 36, 42 In studies after TAVR with Sapien (Edwards Lifesciences, Irvine, California),

Online

36, 23, 52, 54, 58 the prevalence of any PPM varied from 18% to 56%, while severe PPM varied

from 8% to 18%.

Data limited to patients without PPM after SAVR (including patients with mean indexed

EOA > 0.85 cm2/m

2 or studies with no PPM group data) show a mean indexed EOA of 0.99

cm2/m

2 Online 1, 2, 5, 8, 26 which is comparable to results in all patients after TAVR, with or without

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

11

PPM, in whom the mean indexed EOA was 0.98 cm2/m

2 (Figure 5A).

Online 3, 19, 23, 25, 26, 35, 36, 39, 42,

53, 54, 57, 58 In studies of patients without PPM after SAVR,

Online 1, 2, 5, 8, 10, 11 the mean peak

pressure was 26 mmHg, as compared to the mean peak pressures in all patients, with or without

PPM, after TAVR with CoreValve of 16 mmHg Online 23, 25, 36, 39, 40, 45, 52, 54, 58, 59

or Sapien of 18

mmHg (Figure 5B). Online 19, 25, 35, 36, 41, 44, 47, 49, 57, 64

Cumulatively, these data lend support to the

PARTNER IA results where the incidence of PPM was less in TAVR vs SAVR patients. 6

Valvular and paravalvular aortic regurgitation: SAVR vs. TAVR

There were two prospective, randomized, controlled trials that compared post-interventional

total aortic regurgitation (AR). Both studies showed that the incidence of postoperative AR was

higher in patients with TAVR than that with SAVR .5, 21

In studies of patients undergoing

SAVR, the prevalence of any preoperative AR varied from 30% to 86% (median, 71%), Online 3,

19, 25 and in studies undergoing TAVR (Online Table 4), the prevalence of preprocedural AR

varied from 35% to 92% (median, 79%). Online 3, 19, 25, 36, 44, 52, 54-56

The prevalence of preoperative

moderate to severe AR varied from 10% to 19% (median, 14%) in studies evaluating SAVR,

Online 3, 19, 25 and the prevalence of preprocedural moderate to severe AR varied from 5% to 69%

(median, 15%) in studies undergoing TAVR. Online 3, 19, 25, 36, 44, 46, 47, 52, 54-56, 63, 66

After SAVR, the

prevalence of any postoperative total AR varied from 5% to 54% (median, 32%) Online 3, 19, 23, 25,

26 and that of postoperative moderate to severe total AR varied from 0% to 3% (median, 0%).

Online 3, 19, 23, 25, 26 In contrast, after TAVR, the prevalence of any postprocedural total AR varied

from 37% to 96% (median, 69%) Online 3, 19, 23, 25, 26, 36, 37, 39, 44, 46, 47, 52, 54-56, 62

and that of

postprocedural moderate to severe total AR varied from 0% to 41% (median, 11%) (Online

Table 4). Online 3, 19, 23, 25, 26, 36, 37, 39, 44, 46, 47, 52, 54-56, 62, 63, 66

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

12

The presence and severity of paravalvular leaks were reported in one study with SAVR and 6

studies evaluating TAVR. In one study after SAVR with bioprostheses, the prevalence of any

postoperative paravalvular AR was 22% and that of postoperative moderate to severe

paravalvular AR was 2%. 5 After TAVR, the prevalence of any paravalvular AR varied from

39% to 100% (median, 78%), Online 19, 36, 53, 56, 60, 67

but that of moderate to severe paravalvular

AR varied from 7% to 39% (median, 19%). Online 19, 36, 56, 67

The higher severity of total AR and

paravalvular in patients undergoing TAVR is consistent with the recent analysis of the

PARTNER A trial data (Online Table 4).

Mitral regurgitation

The presence of mitral regurgitation was infrequently described especially following SAVR.

In the one study reporting mitral regurgitation after SAVR with mechanical prosthesis, the

prevalence of any preoperative MR and moderate to severe preoperative MR were 52% and

17%, respectively and the prevalence of any postoperative MR and moderate to severe

postoperative MR were 30% and 7%, respectively. 21

In one study after SAVR with

bioprosthesis, the prevalence of any preoperative MR and moderate to severe preoperative MR

were 98% and 21%, respectively and the prevalence of any postoperative MR and moderate to

severe postoperative MR were 90% and 12%, respectively 6 In patients undergoing TAVR, any

MR and moderate to severe MR were observed preoperatively in 74% to 97% and 3% to 49%,

respectively Online 3, 19, 38, 40, 43, 45, 55, 58, 66

and any MR and moderate to severe MR after TAVR

were observed in 71% to 100% and 1% to 29%, respectively. Online 3, 19, 38, 40, 43, 45, 55, 58, 66

Webb

et al 27

reported an improvement in MR in 24 out of the 50 patients (48%) with moderate to

severe MR after TAVR using the Cribier Edwards valve. The improvement of MR was

associated with improvement in LV function and functional class. Since most studies did not

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

13

report the change of MR it is difficult to compare improvement of MR after SAVR and TAVR

interventions (Online Table 5).

Hemodynamic and echocardiographic variables associated with mortality

Among 60 studies that met the selection criteria, 13 studies assessed mortality and its

predictors within study populations following SAVR with a single valve type (ie either

mechanical or bioprosthetic) or TAVR and few assessed variables were associated with

survival. Preoperative LVEF,11, 28

depressed LVEF,29, 30

low gradient AS,29-31

and higher LV

mass11

were not predictors of mortality. Operative success rate,32

paravalvular AR after

TAVR33, 34

and combined bypass-graft operation32

were predictors of mortality, but type of

bioprosthesis,35

and MR grade after TAVR11

were not. 18

Discussion

This systematic review of the large literature on AVR shows that, while both SAVR and

TAVR provide excellent hemodynamic results, the reduction in aortic valve gradients may be

slightly greater with TAVR. There were similar increases in EF after TAVR and SAVR, an

effect which is immediate, if present. In contrast there may be greater regression of LVH

following SAVR, an effect which is at odds with the lower incidence of PPM in these patients.

The time course of the regression of LVH is gradual in both valve types, evolving over months

to years. In contrast the severity of total aortic regurgitation and presence and severity of

paravalvular regurgitation is much higher in TAVR. Finally, limited data preclude a meaningful

comparison of diastolic function or mitral regurgitation between SAVR and TAVR, or of any

possible differential association of hemodynamic variables with survival.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

14

Changes in left ventricle function and hypertrophy after SAVR and TAVR

When LV systolic dysfunction is due to increased afterload with normal myocardial

contractility, systolic function is expected to improve after relief of outflow obstruction in the

absence of significant myocardial dysfunction, regardless of the type of valve used. LVEF is a

powerful predictor of clinical outcome in patients with cardiovascular diseases, including AS. 28

In the current review, when the pre operative LVEF was over 50%, the mean change in LVEF

at follow up after intervention was small, and similar between SAVR and TAVR (4% vs. 6%)

As expected, a more marked improvement of LVEF was seen in patients with a low LVEF prior

to intervention. Although the available data show at least equivalent if not slightly superior

hemodynamic performance of TAVR over SAVR, these data do not support definitive

conclusions about the possible superiority of TAVR over SAVR in terms of improvement of

LVEF.

LVH constitutes a risk factor for cardiac morbidity and mortality. 1 LV afterload reduction

results in regression of LV mass in the majority of patients after surgery for AS. However, in

some patients after SAVR, regression is incomplete because of permanent structural changes in

the myocardium, concomitant disease and the persistent outflow obstruction imposed by

prosthesis-patient mismatch. 29

In the published literature, less PPM was noted in TAVR

compared with SAVR. However, it appears that the regression of LV mass after TAVR was

lower than those after SAVR, 6 an unexpected finding given the lower incidence of PPM. In the

recently published analysis of PARTNER A trial, LV mass was noted to decrease in both

SAVR and TAVR patients but the decrease in LVH over time was similar in the two groups. 6

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

15

Diastolic dysfunction occurs early in the disease course of AS30

and precedes the increase in

LV mass. There are significant age and gender differences in the degree of both hypertrophy

and diastolic dysfunction.41, 42

LV afterload reduction after SAVR for AS increases the

proportion of fibrous tissue in the myocardium because the muscular component of LVH

resolves more rapidly than the fibrous component. 31

As a result, diastolic dysfunction may not

normalize for several years after surgery, if at all. 32

A few studies after TAVR showed that LV

mass regression was not accompanied by an improvement in LV diastolic function.11, 29

In the

current review, some diastolic parameters remained unchanged, but other parameters improved

after TAVR up to one year of follow-up. However, the available data do not distinguish

between effects of TAVR or SAVR in improvement of diastolic function and the long-term

clinical significance is unknown.

Changes of regurgitation at aortic valve and mitral valve after SAVR and TAVR

TAVR is associated with a higher incidence of AR than conventional SAVR. However,

moderate or severe AR occurs in a small minority and the degree of AR following intervention

usually remains stable or even improves slightly over time. 5, 7

Paravalvular AR is presumably

due to suboptimal valve sizing, underexpansion, malpositioning and/or incomplete apposition of

prosthesis in the calcified aortic annulus and/or the anatomy of the left ventricular outflow tract.

33 In the current review, the incidence of postoperative total AR and paravalvular AR is

confirmed to be higher in TAVR than SAVR. Recently published 2-year follow-up of patients

in the PARTNER I trial showed that paravalvular AR was more frequent after TAVR and even

mild paravalvular (or total) AR was associated with increased late mortality. 34

Significant AR

represents a volume load to the LV and may impair systolic or diastolic function as well. The

possibility remains that high incidence of paravalvular AR may adversely affect LV remodeling

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

16

and mortality. Indeed, given the paradoxical presence of both greater PPM and more mass

regression in SAVR, it can be hypothesized that the higher incidence of AR in TAVR may limit

mass regression.

Some degree of MR is commonly found in patients undergoing SAVR and TAVR for AS.

The etiology of MR is often functional in the absence of any significant intrinsic valvular lesion.

35 When there is intrinsic pathology of the mitral apparatus itself, this may result from

calcification of mitral leaflets or annulus, particularly in the elderly, but also from rheumatic

pathology or myxomatous degeneration. Following SAVR, improvement of MR has been

frequently observed by recent prospective data. 36

Functional rather than myxomatous MR may

show more improvement after SAVR. 37

In addition, reduced LV function, increase in LV size

and larger preoperative LV mass have also been associated with a post implant decrease in MR,

supporting the importance of the functional nature of MR in predicting postoperative

improvement. 35

In TAVR, the prosthesis is seated within the aortic annulus with some

component extending into the left ventricular outflow tract. It has been demonstrated that the

anterior mitral valve leaflet or chordae may be damaged during implantation of the transcatheter

heart valve (THV) or due to the left ventricular outflow tract component of the THV. 33

Recently, a multicenter registry of patients undergoing coreValve Revalsing system-TAVR

showed that the baseline MR greater than mild is associated with higher mortality at 1 year after

TAVR. 38

These data ruled out the hypothesis that the nitinol frame of the CoreValve might

interfere with the anterior mitral valve leaflet. MR severity was improved in a significant

proportion of patients, especially with functional MR and without severe pulmonary

hypertension or atrial fibrillation. However, the improvement MR did not independently predict

mortality. 38

In the current review, most of studies did not report the change of MR and the

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

17

available data was inconclusive regarding the improvement of MR after intervention. Further

studies with larger sample size are needed to explain the relationship between the MR and the

LV remodeling after TAVR.

Limitations

While this critical review provides a broad summary of the published literature of the

hemodynamic and structural responses to AVR, it also has important limitations related to the

completeness of the reported data and inability to perform a patient level meta-analysis. First,

there were few prospective, randomized, controlled trials retrieved by our search and most of

the studies were case series or cohort studies. Moreover, as the review embraces the studies

published within the last 33 years, changes in clinical care strategies and technological advances

in SAVR and TAVR over time may be poorly captured, especially in the SAVR group, and

rendered the study groups less comparable.

Second, the mean age of TAVR patients was higher than those undergoing SAVR with

mechanical prosthesis or bioprosthesis (82 yrs vs. 69 yrs vs. 76 yrs). TAVR currently is

restricted to patients with symptomatic, severe AS at high risk or with contraindications to

surgery. Their advanced age means a longer exposure to the harmful effects of severe AS

including LV hypertrophy and probable myocardial fibrosis, 39

as well as a probable greater

burden of comorbidities. The literature is controversial as to the extent to which age limits the

ability to reverse LV remodeling. 21, 26

Third, the mean period of echocardiographic follow-up of LV mass and/or LVEF was

different between SAVR with mechanical prosthesis and that with bioprosthesis and TAVR (32

months vs. 18 months vs. 10 months). In TAVR, there were two long-term echocardiographic

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

18

follow-up (above 12 months) studies of LVEF and there was only one long-term follow-up

study of LV mass. In the systematic review of SAVR, comparisons between preoperative and

postoperative follow-up yielded a consistent regression of LV mass over time. 40

In the current

review, the lack of a long-term available data after TAVR is an important limitation to assess

the difference of LV remodeling between SAVR and TAVR. Further long-term studies after

TAVR are needed to assess this outcome.

Conclusions

We critically evaluated the literature reporting hemodynamic changes and LV remodeling in

SAVR and TAVR. Few trials compared these outcomes directly between the two strategies to

treat severe AS. Despite limitations, the available data show at least equivalent if not slightly

superior hemodynamic performance of TAVR over SAVR with mechanical prosthesis, with

similar performance as SAVR with bioprosthesis. Patients with TAVR showed lower PPM

compared with that with SAVR. However, LV mass appears to regress to a greater degree after

SAVR compared to TAVR. Aortic regurgitation, paravalvular in particular, is more common

after TAVR than SAVR although it is rarely more than moderate in severity. Diastolic function,

when it improves, does so years after SAVR while results are conflicting after TAVR out to 1

year. Mitral regurgitation improves after SAVR but the data are limited after TAVR. While the

data are incomplete in some cases, they support the hemodynamic comparability of SAVR and

TAVR and highlight the need for further studies focusing on hemodynamic changes following

valve therapy to assist clinicians in selecting the appropriate device for patients with high

surgical risk.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

19

References

1. Devereux RB, Wachtell K, Gerdts E, et al. Prognostic Significance of Left Ventricular

Mass Change During Treatment of Hypertension. JAMA: The Journal of the American

Medical Association 2004;292:2350-6.

2. Lund O, Erlandsen M. Changes in left ventricular function and mass during serial

investigations after valve replacement for aortic stenosis. J Heart Valve Dis 2000;9:583-

93.

3. Gaudino M, Alessandrini F, Glieca F, et al. Survival after aortic valve replacement for

aortic stenosis: does left ventricular mass regression have a clinical correlate? European

Heart Journal 2005;26:51-7.

4. Zybach-Benz RE, Aeschbacher BC, Schwerzmann M. Impact of left ventricular

hypertrophy late after aortic valve replacement for aortic stenosis on cardiovascular

morbidity and mortality. International Journal of Cardiology 2006;109:41-7.

5. Smith CR, Leon MB, Mack MJ, et al. Transcatheter versus Surgical Aortic-Valve

Replacement in High-Risk Patients. New England Journal of Medicine 2011;364:2187-

98.

6. Hahn RT, Pibarot P, Stewart WJ, et al. Comparison of Transcatheter and Surgical Aortic

Valve Replacement in Severe Aortic Stenosis: A Longitudinal Study of

Echocardiography Parameters in Cohort A of the PARTNER Trial (Placement of Aortic

Transcatheter Valves). J Am Coll Cardiol 2013;61:2514-21.

7. Leon MB, Smith CR, Mack M, et al. Transcatheter Aortic-Valve Implantation for Aortic

Stenosis in Patients Who Cannot Undergo Surgery. New England Journal of Medicine

2010;363:1597-607.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

20

8. Clavel MA, Webb JG, Pibarot P, et al. Comparison of the Hemodynamic Performance

of Percutaneous and Surgical Bioprostheses for the Treatment of Severe Aortic Stenosis.

Journal of the American College of Cardiology 2009;53:1883-91.

9. Bech-Hanssen O, Caidahl K, Wall B, et al. Influence of aortic valve replacement,

prosthesis type, and size on functional outcome and ventricular mass in patients with

aortic stenosis. Journal of Thoracic and Cardiovascular Surgery 1999;118:57-65.

10. Vizzardi E, D'Aloia A, Fiorina C, et al. Early regression of left ventricular mass

associated with diastolic improvement after transcatheter aortic valve implantation. J

Am Soc Echocardiogr 2012;25:1091-8.

11. Tzikas A, Geleijnse ML, Van Mieghem NM, et al. Left Ventricular Mass Regression

One Year After Transcatheter Aortic Valve Implantation. The Annals of Thoracic

Surgery 2011;91:685-91.

12. Garatti A, Mori F, Innocente F, et al. Aortic Valve Replacement With 17-mm

Mechanical Prostheses: Is Patient–Prosthesis Mismatch a Relevant Phenomenon? The

Annals of Thoracic Surgery 2011;91:71-7.

13. Okamura H, Yamaguchi A, Nagano H, et al. Mid-term outcomes after aortic valve

replacement with the 17-mm St. Jude Medical Regent valve. Circ J 2012;76:365-71.

14. Vicchio M, Della Corte A, De Santo LS, et al. Prosthesis-Patient Mismatch in the

Elderly: Survival, Ventricular Mass Regression, and Quality of Life. The Annals of

Thoracic Surgery 2008;86:1791-7.

15. Okamura H, Yamaguchi A, Noguchi K, et al. Hemodynamics and Outcomes of Aortic

Valve Replacement with a 17- or 19-mm Valve. Asian Cardiovasc Thorac Ann

2010;18:450-5.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

21

16. Clavel M-A, Webb JG, Pibarot P, et al. Comparison of the Hemodynamic Performance

of Percutaneous and Surgical Bioprostheses for the Treatment of Severe Aortic Stenosis.

Journal of the American College of Cardiology 2009;53:1883-91.

17. Pfister R, Wahlers T, Baer FM, et al. Utility of NT-pro-BNP in patients undergoing

transapical aortic valve replacement. Clinical Research in Cardiology 2010;99:301-7.

18. Kukucka M, Pasic M, Dreysse S, et al. Patient-prosthesis mismatch after transapical

aortic valve implantation: incidence and impact on survival. J Thorac Cardiovasc Surg

2013;145:391-7.

19. Ewe SH, Muratori M, Delgado V, et al. Hemodynamic and clinical impact of prosthesis-

patient mismatch after transcatheter aortic valve implantation. J Am Coll Cardiol

2011;58:1910-8.

20. Nombela-Franco L, Ruel M, Radhakrishnan S, et al. Comparison of hemodynamic

performance of self-expandable CoreValve versus balloon-expandable Edwards

SAPIEN aortic valves inserted by catheter for aortic stenosis. Am J Cardiol

2013;111:1026-33.

21. Clavel MA, Webb JG, Rodes-Cabau J, et al. Comparison Between Transcatheter and

Surgical Prosthetic Valve Implantation in Patients With Severe Aortic Stenosis and

Reduced Left Ventricular Ejection Fraction. Circulation 2010;122:1928-36.

22. Tzikas A, Piazza N, Geleijnse ML, et al. Prosthesis-Patient Mismatch After

Transcatheter Aortic Valve Implantation With the Medtronic CoreValve System in

Patients With Aortic Stenosis. The American Journal of Cardiology 2010;106:255-60.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

22

23. Sherif MA, Abdel-Wahab M, Awad O, et al. Early hemodynamic and neurohormonal

response after transcatheter aortic valve implantation. American Heart Journal

2010;160:862-9.

24. Jilaihawi H, Chin D, Spyt T, et al. Prosthesis-patient mismatch after transcatheter aortic

valve implantation with the Medtronic-Corevalve bioprosthesis. European Heart Journal

2010;31:857-64.

25. Guinot P-G, Depoix J-P, Etchegoyen L, et al. Anesthesia and Perioperative Management

of Patients Undergoing Transcatheter Aortic Valve Implantation: Analysis of 90

Consecutive Patients With Focus on Perioperative Complications. Journal of

Cardiothoracic and Vascular Anesthesia 2010;24:752-61.

26. Giannini C, Petronio AS, Nardi C, et al. Left Ventricular Reverse Remodeling in

Percutaneous and Surgical Aortic Bioprostheses: An Echocardiographic Study. Journal

of the American Society of Echocardiography 2011;24:28-36.

27. Webb JG, Pasupati S, Humphries K, et al. Percutaneous Transarterial Aortic Valve

Replacement in Selected High-Risk Patients With Aortic Stenosis. Circulation

2007;116:755-63.

28. Quere J-P, Monin J-L, Levy F, et al. Influence of Preoperative Left Ventricular

Contractile Reserve on Postoperative Ejection Fraction in Low-Gradient Aortic

Stenosis. Circulation 2006;113:1738-44.

29. Monrad ES, Hess OM, Murakami T, et al. Time course of regression of left ventricular

hypertrophy after aortic valve replacement. Circulation 1988;77:1345-55.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

23

30. Douglas PS, Berko B, Lesh M, et al. Alterations in diastolic function in response to

progressive left ventricular hypertrophy. Journal of the American College of Cardiology

1989;13:461-7.

31. Otto CM, Bonow RO. Valvular heart disease: a companion to Braunwald's heart disease.

In. third ed. Philadelphia, PA: Saunders; 2009. p. 131.

32. Villari B, Vassalli G, Monrad ES, et al. Normalization of Diastolic Dysfunction in

Aortic Stenosis Late After Valve Replacement. Circulation 1995;91:2353-8.

33. Bloomfield GS, Gillam LD, Hahn RT, et al. A practical guide to multimodality imaging

of transcatheter aortic valve replacement. JACC Cardiovasc Imaging 2012;5:441-55.

34. Kodali SK, Williams MR, Smith CR, et al. Two-year outcomes after transcatheter or

surgical aortic-valve replacement. N Engl J Med 2012;366:1686-95.

35. Unger P, Dedobbeleer C, Van Camp G, et al. Mitral regurgitation in patients with aortic

stenosis undergoing valve replacement. Heart 2010;96:9-14.

36. Unger P, Plein D, Van Camp G, et al. Effects of Valve Replacement for Aortic Stenosis

on Mitral Regurgitation. The American Journal of Cardiology 2008;102:1378-82.

37. Barreiro CJ, Patel ND, Fitton TP, et al. Aortic Valve Replacement and Concomitant

Mitral Valve Regurgitation in the Elderly: Impact on Survival and Functional Outcome.

Circulation 2005;112:I-443-7.

38. Bedogni F, Latib A, De Marco F, et al. Interplay between mitral regurgitation and

transcatheter aortic valve replacement with the CoreValve Revalving System: a

multicenter registry. Circulation 2013;128:2145-53.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

24

39. Lund O, Emmertsen K, Dørup I, et al. Regression of left ventricular hypertrophy during

10 years after valve replacement for aortic stenosis is related to the preoperative risk

profile. European Heart Journal 2003;24:1437-46.

40. Sharma UC, Barenbrug P, Pokharel S, et al. Systematic review of the outcome of aortic

valve replacement in patients with aortic stenosis. The Annals of Thoracic Surgery

2004;78:90-5.

41. Monrad ES, Hess OM, Murakami T, et al. Time course of regression of left ventricular

hypertrophy after aortic valve replacement. Circulation 1988;77:1345-55.

42. Villa E, Troise G, Cirillo M, et al. Factors affecting left ventricular remodeling after

valve replacement for aortic stenosis. An overview. Cardiovasc Ultrasound 2006;4:25.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

25

Figure Legends.

Figure 1. Flow diagram of literature search and study selection.

Figure 2. Changes in hemodynamic variables after surgical aortic valve replacement or

transcatheter aortic valve replacement. Panels A and B show the mean peak pressure gradient

and the mean fractional decrease of peak pressure gradient after intervention. Panels C and D

show the mean of mean pressure gradient and the mean fractional decrease of mean pressure

gradient after intervention. Panel E shows the fractional increase of (indexed) effective orifice

area after intervention. EOAIFr. increase, fractional increase of (indexed) effective orifice area;

Mean PGFr. decrease, fractional decrease of mean pressure gradient; Mean PGpost, post-intervention

mean pressure gradient; Peak PGFr. decrease, fractional decrease of peak pressure gradient; Peak

PGpost, post-intervention peak pressure gradient.

Figure 3. Changes in left ventricular ejection fraction after surgical aortic valve replacement or

transcatheter aortic valve replacement. (A) Changes in left ventricular ejection fraction (LVEF)

after surgical aortic valve replacement (SAVR) and transcatheter aortic valve implantation

(TAVR) according to the preoperative LVEF. In studies with preoperative (or preprocedural)

LVEF >50%, the mean fractional changes of LVEF after SAVR and TAVR were 4% and 6%,

respectively. In studies with preoperative (or preprocedural) LVEF ≤50%, the mean fractional

changes of LVEF after SAVR and TAVR were 14% and 20%, respectively. (B) The fractional

decrease of left ventricular ejection fraction after SAVR (□) and TAVR (▲) according to the

duration of follow-up. LVEF, left ventricular ejection fraction

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

26

Figure 4. Changes in left ventricular mass after surgical aortic valve replacement or

transcatheter aortic valve replacement. (A) The fractional decrease of left ventricular mass

index (or left ventricular mass) after surgical aortic valve replacement (SAVR) and

transcatheter aortic valve implantation (TAVR) according to the duration of follow-up. The

mean follow-up of left ventricular mass index (or left ventricular mass) was longer in

mechanical SAVR (△) and bioprosthetic SAVR (□) compared to TAVR (▲). (B) The

fractional decreases of LV mass index (or LV mass) during immediately to 3 months, 6 months,

and 12 months after SAVR and TAVR, respectively.

Figure 5. Patient prosthesis mismatch after surgical aortic valve replacement or transcatheter

aortic valve replacement. (A) The mean indexed effective orifice area in studies showing no

PPM or high indexed EOA after SAVR and that in studies regardless of the PPM after TAVR.

(B) The mean peak pressure in studies showing no PPM or high indexed EOA after SAVR and

that after TAVR with CoreValve or Sapien regardless of the PPM.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

27

Figure 1

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

28

Figure 2

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

29

Figure 3

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

30

Figure 4

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

31

Figure 5