View
1
Download
0
Category
Preview:
Citation preview
Should TAVR be the Standard for Bicuspid
Aortic Stenosis, or Do We Need a Randomized
Clinical Trial?
Raj R. Makkar, MDDirector, Interventional Cardiology & Cardiac Catheterization Laboratories
Associate Director, Cedars-Sinai Heart Institute
Professor of Medicine, University of California, Los Angeles
Stephen Corday Chair in Interventional Cardiology
Disclosures
Consultant and research grant from Edwards LifeSciences,
Medtronic, Abbott, Boston Scientific and Jena Valve
What is the competition for TAVR vs.
SAVR trials?
• Aortic Regurgitation
• Failed surgical bioprosthetic valves
• Mixed valvular heart disease-AS+MS/MR
• Failed TAVR valves
My 5 key arguments for a randomized trial for
Bicuspid AS• Bicuspid AS will be encountered with greater frequency as
TAVR use expands in the younger patients-50% of SAVR
in young are bicuspid
• The available observational data are limited by treatment
bias..perhaps only favorable anatomies were treated
• Surgical outcomes in young Bicuspid AS patients are
excellent; reasonable to expect robust evidence for TAVR to
replace SAVR
• Anatomically heterogeneous group with frequent
aortopathy-unlike what has been treated in previous IDE
studies
• The precedent for label expansion in TAVR in last decade
has been IDE randomized clinical trials
In low risk Bicuspid AS
Is TAVR=SAVR?
• Low risk trials: (Mack et al, NEJM; Popma et al, NEJM)
TAVR equal/better than SAVR
• STS/ACC-TVT registry: (Makkar et al, JAMA 2019)
1 year Death/Stroke
Bicuspid AS =Tricuspid AS
Prevalence of Bicuspid Aortic Valve
20- 30- 40- 50- 60- 70- 80-
Fre
qu
en
cy o
f B
AV
(%
)
Roberts WC et al. Circulation. 2005;111:920-925
33
6064
69
60
42
28
0
10
20
30
40
50
60
70
80
Age (years)
Prevalence of bicuspid valve in patients undergoing
isolated AVR-almost 50%!
Roberts WC. et al. Circulation 2005
Operatively excised, stenotic aortic valves from 932 patients
aged 26 to 91 years
Prevalence of bicuspid aortic valve in TAVR
studies is less than SAVR studies
• Less than 7% of patients
with bicuspid aortic valve
in TAVR registries
• Septugenerians undergoing
SAVR: 41.7%
• Octogenerians undergoing
SAVR: 27.5%
Zhao ZG. et al. Nature Reviews in Cardiology 2015
Pivotal Randomized Trials
Inoperable
High Risk
Intermediate Risk
Low Risk
Key Anatomic Exclusion Criteria
• Aortic annulus diameter < 16mm or 28mm
• Bicuspid valve (CT imaging)
• Severe AR or MR
• Severe LV dysfunction
• Severe calcification of aortic valve complex
• Vascular anatomy not suitable for safe femoral access
• Complex CAD: LM, Syntax score>32
• Low coronary takeoff
Outcomes of TAVR for Bicuspid vs Tricuspid AS
• 546 pairs of patients with bicuspid and tricuspid AS were created with PS-
matching
• Bicuspid had more frequent aortic root injury with Sapien XT and PVL with
CoreValve, but no differences in complications with Sapien 3/Evolut R/Lotus
• No difference in 1-year mortality between bicuspid and tricuspid AS
Yoon et al: J Am Colle Cardiol 2017;69:2579-89
Contemporary TAVR for Bicuspid vs Tricuspid AS
STS/ACC TVT Registry
Makkar et al: JAMA 2019;321:2193-202
22
Characteristic% or mean ± SD
Bicuspid AS(n=2691)
Tricuspid AS(n=2691)
p-value
Device success 96.5 96.6 0.87
Procedure Time, min 100.7 ± 51.80 98.2 ± 52.09 0.08
Fluoroscopy Time, min 18.5 ± 10.96 17.1 ± 10.17 <0.0001
Conversion to open surgery 0.9 0.4 0.03
Annulus Rupture 0.3 0.0 0.02
Cardiopulmonary bypass 1.4 1.0 0.13
Aortic dissection 0.3 0.1 0.34
Coronary Obstruction 0.4 0.3 0.34
Need for a second valve 0.4 0.2 0.16
Procedural Outcomes
23
30-Day Outcomes
KM estimate % Bicuspid Tricuspid AS p-value
All-cause mortality 2.6 2.5 0.82
All stroke 2.4 1.6 0.02
Life-threatening bleeding 0.1 0.1 0.99
Major vascular complication 0.9 1.0 0.68
New pacemaker 9.1 7.5 0.03
Aortic valve reintervention 0.2 0.3 0.79
24
Bicuspid
Tricuspid
1-Year Mortality – Matched
2691 1259 1222 1162 934
2691 1367 1326 1256 977
P= 0.31HR: 0.90 [95% CI: 0.73, 1.10]
10.5%
12.0%
All-C
au
se M
ort
ality
(%
)
0
5
10
15
20
25
30
35
40
Time in Months
0 3 6 9 12
Bicuspid
Tricuspid
Number at risk
25
Bicuspid
Tricuspid
1-Year Stroke – Matched
2691 1234 1196 1135 910
2691 1341 1296 1226 952
P= 0.16HR: 1.28 [95% CI: 0.91, 1.79]
3.1%
3.4%
Str
oke (
%)
Time in Months
0 3 6 9 12
Bicuspid
Tricuspid
Number at risk
0
5
10
15
20
25
30
35
40
26
Paravalvular Leak – Matched
81.5% 83.7%74.4% 77.2% 75.7% 78.8%
17.0% 15.5%23.6% 20.5% 21.1% 18.7%
1.3% 0.7% 2.0% 2.3% 3.0% 2.5%0.1% 0.1% 0.1% 0.1% 0.2% 0.0%
0%
20%
40%
60%
80%
100%
Bicuspid(n=2179)
Tricuspid(n=2233)
Bicuspid(n=1711)
Tricuspid(n=1782)
Biscuspid(n=593)
Tricuspid(n=673)
Severe
Moderate
Mild
None/Trace
Discharge 30-day 1-year
p=0.44p=0.17p=0.08
27
0
15
30
45
60
Mean
Gra
die
nt
(mm
Hg
)
Mean GradientBicuspid Tricuspid
p=0.6
9
Hemodynamics – Matched
p=0.15 p=0.86p=0.51
0
0.5
1
1.5
2
2.5
3
Baseline DischargeA
VA
(cm
2)
Aortic Valve AreaBicuspid Tricuspid
p=0.15 p=0.34
What are the limitations of this
observational studies?
• No central CT core lab to diagnose bicuspid
anatomy
• No central adjudication of events
• No data on aortopathy
• Most importantly.. “Treatment bias-were there
anatomical factors as to why TAVR was
chosen rather than surgey? Only 3% of the
entire cohort were bicuspid.
The “easy case”: Little calcium, No Raphe
60 y/o female undergoing TAVR
Annulus Area 356.2 mm2
Dmin 18.3, Dmax 24.8 mm
Congenital bicuspid aortic valveMinimal calcification
TAVR with 23mm Sapien3 performed
Trace paravalvular AR
87 y/o male referred for TAVRNYHA 3 heart failure, diastolic
Severely calcified aortic
valve
Severe MR
Mean gradient 34mmHg
AVA 0.5cm2
Bicuspid valve with heavily
calcified raphe
29mm Sapien 3 deploymentPredilation with 23mm x 4cm Z-Med II Balloon
Careful predilation, heavily calcified valve
Severely calcified
aortic valve
29mm Sapien 3
Eccentric deployment of Sapien3Well expanded valve, but eccentrically located
Eccentric deployment of Sapien3
Moderate PVL
Persisent PVL despite post-dilation
Post-dilation performed with the
Sapien3 balloon
Persistent PVL despite post-
dilation
PVL closure performed immediately after post-
dilation due to persistent moderate PVLStep 1: Shuttle sheath
across the leak
Leak crossed with Terumo
glidewire with a MPA catheter;
wire exchanged for a Amplatz stiff
wire
Step 2: Amplatz wire
removed from the LV
Step 3: 10mm AVP 2
plug advanced
Step 4: 10mm AVP 2
plug deployed
Final result s/p 10mm AVP 2 plug
deployment
TEE guidance during the procedure
Position of wire through the PVL
confirmed with TEE
Final result s/p TAVR and PVL closure
10mm AVP2
plug
Cardiac CT for aortic valve evaluation
Bicuspid aortic valve
Left main
RCA
Calcified
Raphe
Type 1 bicuspid
aortic valve
Left Right
Valve-in valve with Sapien XT 26 mm
Valve deployment Mild residual PVL
No diastolic flow reversal Improved diastolic gradient
Hemodynamics improvement post TAVR
Pre valve replacement Post valve replacement
Improved LV systolic function immediately post TAVR
Pre-valve replacement Post-valve replacement
Continuous improvement of LV systolic function post TAVR
Pre-TAVR (EF 15%) Day 1 post-TAVR (EF 40%)
Repeat CT 2 years later revealed
pseudoaneurysm of the sinus of ValsalvaNormal leaflet motion
Pseudoaneurysm
Patient underwent surgical repair of the pseudoaneurysm and bioprosthetic aortic valve
replacement
Bicuspid Aortic Valve with Challenging
Anatomy
• Calcified Raphe
• Excess Leaflet Calcification
Multiple Valvuloplasty with 23mm Z-Med
Balloon
Downsizing to 26mm Sapien 3 Ultra Valve
Multiple Post-Dilatation
Post-TAVR TEE
• Calcified Raphe
• Mild Leaflet
Calcification
Bicuspid Aortic Valve with Favorable
Anatomy
Bicuspid Aortic Valve with Favorable
Anatomy
• No Calcified Raphe
• Mild Leaflet
Calcification
Bicuspid Aortic Valve with Favorable
Anatomy
• Very Large Annulus
• No Calcified Raphe
• Mild Leaflet Calcification
Surgical Outcomes for Bicuspid Aortic
Valve
Total No AgeMortality
30 days 1 year 5 years 10 years
Borger at al 1 201 54±11 2.5% NA NA 19%
Girdauskas et al 2153
(aortopathy)54±11
0.7% (hospital death)
NA NA 14%
Itagaki et al 3 2079 55.3±14.9 1.5% NA NA NA
Desai et al 4 1890 50±14 0.4% 1.0% 4.0% NA
Andrei et al 5Total: 628
Men: 478
Women: 150
Men: 56.3±13.6
Women:
60.7±13.8
0.6%Men: 3.2%
Women: 6.0%
Men: 8.3%
Women: 7.8%NA
1 Borger et al: J Thorac Cardiovasc Surg 2004; 128: 677-832 Girdauskas et al: Eur J Cardiothorac Surg 2012; 42:832-83 Itagaki et al: J Am Coll Cardiol 2015; 65: 2363-94 Desai et al: J Thorac Cardiovasc Surg 2016; 151: 1650-95 Andrei et al: Am J Cardiol 2015;116:250-255
TAVR technology is ready and the
procedure is mature..
• In the low risk trials the 1and 2 year death,
stroke,and rehospitalization rates were very
low and lower than surgery
• Imaging has evolved to help better select
suitable anatomical subsets
Non-calcified Raphe Calcified RapheM
ild
Le
afl
et
Ca
lcif
ica
tio
n
Ex
ce
ss
Le
afl
et
Ca
lcif
ica
tio
n
No Raphe
n = 61 (5.5%)
n = 64 (5.7%)
n = 271 (24.3%)
n = 201 (18.0%)
n = 225 (20.2%)
n = 293 (26.3%)
Phenotype Distribution
Yoon S, Makkar R EuroPCR 2019
0 180 360 540 720
0
10
20
30
40A
ll-c
au
se
Mo
rta
lity
(%
)40
0
30
20
10
12.3
10.8
25.5
Overall P < .001 by log-rank test
HR for Calcified raphe plus Excess leaflet calc vs. Calcified raphe or Excess leaflet calc
2.17 (95%CI, 1.46 – 3.25); P < .001
HR for Calcified raphe plus Excess leaflet calcification vs. None
2.97 (95% CI, 1.82 – 4.84); P < .001
0 180 720360 540
Days
Ca-raphe plus excess calc
None
No. at Risk
293
332
156
225
80
108
Ca-raphe or excess calc 490 301 151
Calcified raphe plus
Excess leaflet calc
None
Calcified raphe or
Excess leaflet calc
All-cause Mortality and BAV Phenotype1115 Bicuspid AS patients, Central CT Core Lab, 25 Centers
Yoon S, Makkar R EuroPCR 2019
14.0
4.8
9.2
5.5 6.1
1.2
4.1
1.6 2.8
0.6
4.5
2.1 0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
20.0
PVL ≥ moderate Aortic Root Injury Major VascularCompication
30-day Mortality
Both Calcified raphe plusExcess leaflet calcification
Either Calcified raphe orExcess leaflet Calcification
None
Inc
ide
nc
e (
%)
p < 0.001
p < 0.001 p = 0.006 p = 0.004
Procedural and 30-day Outcomes According to BAV Phenotype
(n = 293) (n = 490) (n = 332)
0 180 360 540 720
0
10
20
30
40A
ll-c
au
se
Mo
rta
lity
(%
)40
0
30
20
10
0 180 720360 540
Days
18.5
12.5
Aortopathy
No Aortopathy
Aortopathy
No Aortopathy
No. at Risk
509
606
302
380
147
192
HR, 1.61 (95% CI, 1.13–2.32)
P = .009 by log-rank test
All-cause Mortality and AortopathyAortopathy was not associated with All-cause Mortality with Multivariate Analysis
Independent Correlates of All-cause Mortality
Univariate Analysis Multivariate Analysis
HR (95% CI) P Value HR (95% CI) P Value
Age 1.03 (1.01 – 1.05) 0.006 – –
STS score 1.06 (1.03 – 1.09) < 0.001 1.04 (1.01 – 1.08) 0.02
MR ≥ moderate at baseline 1.77 (1.10 – 2.87) 0.02 1.65 (1.02 – 2.68) 0.04
Prior MI 1.64 (1.05 – 2.54) 0.028 –
Atrial Fibrillation 1.61 (1.08 – 2.40) 0.021 – –
Type of Bicuspid AV < 0.001 0.001
No raphe (Sievers’ type 0) Reference – Reference –
Non-calcified raphe (Sievers’ type 1) 1.33 (0.59 – 2.99) 0.49 1.55 (0.69 – 3.50) 0.29
Calcified raphe (Sievers’ type 1) 2.70 (1.25 – 5.86) 0.01 2.80 (1.29 – 6.08) 0.009
Excess leaflet calcification 1.72 (1.19 – 2.49) 0.004 1.53 (1.05 – 2.22) 0.03
Aortopathy 1.61 (1.13 – 2.32) 0.009 – –
Non-transfemoral access 2.05 (1.29 – 3.25) 0.002 1.70 (1.05 – 2.75) 0.03
Early-generation devices 1.89 (1.31 – 2.73) 0.001 1.71 (1.17 – 2.50) 0.005
Independent Correlates of All-cause Mortality
Univariate Analysis Multivariate Analysis
HR (95% CI) P Value HR (95% CI) P Value
Age 1.03 (1.01 – 1.05) 0.006 – –
STS score 1.06 (1.03 – 1.09) < 0.001 1.04 (1.01 – 1.08) 0.02
MR ≥ moderate at baseline 1.77 (1.10 – 2.87) 0.02 1.65 (1.02 – 2.68) 0.04
Prior MI 1.64 (1.05 – 2.54) 0.028 –
Atrial Fibrillation 1.61 (1.08 – 2.40) 0.021 – –
Type of Bicuspid AV < 0.001 0.001
No raphe (Sievers’ type 0) Reference – Reference –
Non-calcified raphe (Sievers’ type 1) 1.33 (0.59 – 2.99) 0.49 1.55 (0.69 – 3.50) 0.29
Calcified raphe (Sievers’ type 1) 2.70 (1.25 – 5.86) 0.01 2.80 (1.29 – 6.08) 0.009
Excess leaflet calcification 1.72 (1.19 – 2.49) 0.004 1.53 (1.05 – 2.22) 0.03
Aortopathy 1.61 (1.13 – 2.32) 0.009 – –
Non-transfemoral access 2.05 (1.29 – 3.25) 0.002 1.70 (1.05 – 2.75) 0.03
Early-generation devices 1.89 (1.31 – 2.73) 0.001 1.71 (1.17 – 2.50) 0.005
• CT assessment of morphology in bicuspid aortic stenosis helps assess anatomical risk of TAVR
• In absence of randomized clinical trial data in treating Bicuspid Aortic Stenosis, CT based anatomical assessment may identify patients favorable for TAVR and in conjunction with surgical risk help triage patients to TAVR vs. SAVR
Clinical Implications
Pivotal Randomized Trials have been the standard
for indication expansion for TAVR
Inoperable
High Risk
Intermediate Risk
Low Risk
Meta-Analysis
Systemic Reviews
Randomized Controlled Trial
Cohort Studies
Case Control Studies
Case Report/ Case Series
Background Information & Expert Opinion
Hierarchy of Evidence and Research Designs
In low risk Bicuspid AS
Is TAVR=SAVR?
• Low risk trials: (Mack et al, NEJM; Popma et al, NEJM)
TAVR equal/better than SAVR
• STS/ACC-TVT registry: (Makkar et al, JAMA 2019)
1 year Death/Stroke
Bicuspid AS =Tricuspid AS
Why infer? Let us do the randomized clinical trial!
Practical considerations..
• Be careful of unfavorable anatomical features on CT: excessive
calcium, raphe type especially calcified raphe
• Some degree of undersizing (compared to tricuspid) is appropriate
• Positioning is harder than the tricuspid valve. Cross check with
echo. TEE guidance is is preferable due to higher rates of AI and
risk of aortic root rupture
• Predilation is generally a good idea; avoids difficult crossing and
stresses on aorta which may be diseased; also can help with sizing.
• Post dilation and rarely valve in vavle may be needed to optimize
the expansion and procedural outcomes.
Conclusions
• Though the mortality may be “similar” to the tricuspid TAVR, the acute
outcomes in the published literature are worse with respect to AI, and
pacemaker implantation with the first generation devices
• The data with Sapien 3 valve are excellent, no comparative studies are
available with other next generation valves (Evolut R, Lotus, Portico)
• While Bicuspid TAVR is justifiable in higher surgical risk patients, high
risk anatomical features (extreme calcium, heavy-calcified raphe),
concomitant aortopathy should prompt consideration for surgical AVR in
low risk patients
• Randomized trials/prospective registries especially in patients with lower
surgical risk are needed.
Recommended