TAVR is Superior to Surgical AVR Paul S.Teirstein, MD SCRIPPS CLINIC Paul S.Teirstein, MD Chief of Cardiology Director of Interventional Cardiology Director Scripps Cardiovascular Institute
TAVR is Superior to Surgical AVR
Paul S.Teirstein, MD
SCRIPPS CLINIC
Paul S.Teirstein, MD
Chief of CardiologyDirector of Interventional CardiologyDirector Scripps Cardiovascular Institute
Paul Teirstein, MD
Disclosures:
Cordis, Boston, Medtronic, Abbott, Edwards:
Research Grants
SCRIPPS CLINIC
Research Grants
Consultant
Speakers Bureau
The elite surgeon:
--Good looking
--Average ht 6’3”
The basic interventionist
--A little “hunched over”
--Average ht 5’7 and. 1/2”
The elite surgeon:
--Good looking
--Average ht 6’3”
The basic interventionist
--A little “hunched over”
--Average ht 5’7 and. 1/2”
How can I prove How can I prove TAVR is superior TAVR is superior to traditional AVR?to traditional AVR?
•• Use Level A Evidence (randomized trial data)Use Level A Evidence (randomized trial data)
•• Focus on the PARTNER 1A randomized trial:Focus on the PARTNER 1A randomized trial:
–– High surgical risk:High surgical risk:
·· Predicted risk of operative mortality ≥ 15% (determined by the heart Predicted risk of operative mortality ≥ 15% (determined by the heart team); guideline = STS score ≥ 10team); guideline = STS score ≥ 10
••
PARTNER Study DesignPARTNER Study Design
N = 358InoperableInoperable
ASSESSMENT: ASSESSMENT:
Symptomatic Severe Aortic Stenosis
ASSESSMENT: High-Risk AVR Candidate3,105 Total Patients Screened
ASSESSMENT: High-Risk AVR Candidate3,105 Total Patients Screened
Total = 1,057 patients
2 Parallel Trials: Individually Powered
N = 699 High RiskHigh Risk
ASSESSMENT: ASSESSMENT: Yes No
N = 179
StandardTherapy
ASSESSMENT:
Transfemoral Access
ASSESSMENT:
Transfemoral Access
Not In Study
TF TAVR
Primary Endpoint: All-Cause Mortality Over Length of Trial (Superiority)
Co-Primary Endpoint: Composite of All-Cause Mortalityand Repeat Hospitalization (Superiority)
1:1 Randomization
VS
Yes No
N = 179
TF TAVR AVR
Primary Endpoint: All-Cause Mortality at 1 yr(Non-inferiority)
TA TAVR AVR
VSVS
N = 248 N = 104 N = 103N = 244
ASSESSMENT:
Transfemoral Access
ASSESSMENT:
Transfemoral Access
Transapical (TA)Transfemoral (TF)
1:1 Randomization1:1 Randomization
Yes No
Procedural Outcomes Procedural Outcomes -- TAVR vs. AVRTAVR vs. AVR
Traditional AVR TAVR
Anesthesia time - min 236
Total procedure time - min 133
Aborted procedure - no. (%) 7 (2.0)
Anesthesia time - min 330
Total procedure time - min 230
Aborted procedure - no. (%) 0
*p<0.001
Reoperation for bleeding - no. (%) 2 (0.6)
Intra-procedural death - no. (%) 3 (0.9)
Aortic perforation - no. (%) 0
Index hospitalization (days) 8*
Median ICU stay (days) 3.0*
Reoperation for bleeding - no. (%) 12 (3.4)
Intra-procedural death - no. (%) 1 (0.3)
Aortic perforation - no. (%) 1 (0.3)
Index hospitalization (days) 12
Median ICU stay (days) 5.0
SixSix--Minute Walk TestMinute Walk TestAll Patients (N=699)All Patients (N=699)
Median Distance, meters
P = 0.73 P = 0.002 P = 0.33 P = 0.76
Median Distance, meters
EchocardiographicEchocardiographic Findings Findings AVA (AT)AVA (AT)
Valve Area (cm
2)
TAVR
AVR
Numbers at RiskNumbers at Risk
TAVRTAVR 301301 269269 223223 210210 139139
AVRAVR 290290 224224 162162 151151 110110
Valve Area (cm
p = 0.002 p = 0.003 p = 0.16p = 0.001
EchocardiographicEchocardiographic Findings Findings Mean and Peak Gradients (AT)Mean and Peak Gradients (AT)
Gradient (m
mHg)
Peak Gradient - TAVR
Mean Gradient - TAVR
Peak Gradient - AVR
Mean Gradient - AVR
Numbers at RiskNumbers at Risk
TAVRTAVR 307307 275275 233233 218218 144144
AVRAVR 295295 228228 168168 155155 112112
Gradient (m
mHg)
Primary Endpoint:Primary Endpoint:AllAll--Cause Mortality at 1 YearCause Mortality at 1 Year
AVR(N = 351)
26.8%
AVR(N = 351)
26.8%
TAVR(N = 348)
24.2%
TAVR(N = 348)
24.2%
Difference -2.6%
Upper 1Upper 1--sided sided 95% CI95% CI 3.0%3.0%
Non-inferiorityP value
= 0.001
Zone of non-inferiority pre-specified
7.0-2.0 -1.0 0.0 6.03.0 4.0 5.0-3.0 1.0 2.0
Non-inferior
Upper one-sided 95% CI
8.0 %
Primary Non-Inferiority Endpoint Met
Zone of non-inferiority pre-specified margin = 7.5%
If two therapies are non inferior and If two therapies are non inferior and one is less invasive,one is less invasive,
then>then>
Even a non-surgeon knows.
then>then>
IT’S SUPERIOR!IT’S SUPERIOR!
Strokes (ITT)Strokes (ITT)Stroke
HR [95% CI] =1.22 [0.67, 2.23]
p (log rank) = 0.517
TAVR
AVR
3.2%
6.0%4.9%
7.7%
Stroke
Months Post Procedure
Numbers at RiskNumbers at Risk
TAVRTAVR 348348 287287 249249 224224 162162 6565 2828
AVRAVR 351351 246246 230230 211211 160160 6262 3131
DiffusionDiffusion--Weighted MRI StudyWeighted MRI Study
Philipp Kahlert, MDWest German Heart Center Essen
Example of an 82-year-old patient two days after successful TAVI
Pre-TAVI Post-TAVI
Embolic Embolic phenomenonphenomenon
NeuroNeuro--imaging with TAVRimaging with TAVR
JACC 2011N=60
JACC 2010N=30
JACC Int 2010N=25
Circulation 2010N=32
EJCTS 2011N=80
Daneault et al., JACC 2011;58: 2143-50
NeuroNeuro--imaging with AVRimaging with AVR
Ann Thor Surg 2006N=15
Stroke 2004N=37
EJCTS 2005N=30
Circulation 2010N=21
Daneault et al., JACC 2011;58: 2143-50
Cerebral Ischemia After TAVICerebral Ischemia After TAVI
8680
60
80
100
%
224
150
200
250
mm3
New Lesions Lesion Volume
48
0
20
40
60
Edwards CoreValve SAVR
81
61
0
50
100
150
Edwards CoreValve SAVR
Kahlert PK et al. Circulation 2010;121:870-878
DWDW--MRI after TAVRMRI after TAVR
•• 32 pts with TAVI; Diffusion32 pts with TAVI; Diffusion--Weighted MRI at baseline, postWeighted MRI at baseline, post--
procedure, and @ 3 procedure, and @ 3 mosmosprocedure, and @ 3 procedure, and @ 3 mosmos
�� 22 balloon22 balloon--expandable and 10 selfexpandable and 10 self--expanding THV devicesexpanding THV devices
•• New foci of restricted perfusion in 27/32 pts (84%)New foci of restricted perfusion in 27/32 pts (84%)
�� Lesions usually multiple and both hemispheres (embolic)Lesions usually multiple and both hemispheres (embolic)
•• No impairment of No impairment of neuroneuro--cognitive function nor clinical cognitive function nor clinical
neurologic events assoc with MRI defectsneurologic events assoc with MRI defects
�� 80% of MRI defects resolved at 3 80% of MRI defects resolved at 3 mosmos imaging studyimaging study
Kahlert P, Knipp SC, Schlamann M, et al.
Circ 2010;121:870-8 Circ 2010;121:870-8
Later Hazard of Neurologic Event
%%
10
15
TAVR
11
6.7
8.0
%%
Mos
5858
130125
243218
278251
344313
TAVRAVR
0
5
6 12 18 24
AVR
6.7
4.5
51% Procedural51% Procedural
(<10 days)(<10 days)
Timing and Types of Neurologic Events (strokes and TIAs)
AVR
AVR
TAVR
TAVR
AVR
AVR
TAVR
TAVR
AVR
AVR
TAVR
TAVR
AVR
AVR
TAVR
TAVR
AVR
AVR
TAVR
TAVR
AVR
AVR
TAVR
TAVR
AVR
AVR
TAVR
TAVR
00--22daysdays
33--55daysdays
3131--364364daysdays
1111--3030daysdays
66--1010daysdays
22--33yearsyears
11--22yearsyears
Embolic Materialafter TAVR
Embolic MaterialEmbolic MaterialEmbolic MaterialEmbolic Material
Embolic MaterialEmbolic MaterialEmbolic MaterialEmbolic Material
Stroke MessagesStroke MessagesPARTNER 1A PARTNER 1A –– 2 year FU2 year FU
•• Stroke is numerically higher in TAVR Vs. AVRStroke is numerically higher in TAVR Vs. AVR
•• 2 2 yryr: 7.7% : 7.7% VsVs 4.9%, p= 0.524.9%, p= 0.52
•• Absolute stroke difference is lowAbsolute stroke difference is low
•• 2.8%; Number Needed to Harm = 362.8%; Number Needed to Harm = 36•• 2.8%; Number Needed to Harm = 362.8%; Number Needed to Harm = 36
•• If you tell patients the chance of stroke is 57% higher If you tell patients the chance of stroke is 57% higher with TAVR, they will likely choose AVRwith TAVR, they will likely choose AVR
•• If you tell patients the difference in stroke is 7.7% Vs. If you tell patients the difference in stroke is 7.7% Vs. 4.9%, they will likely choose TAVR4.9%, they will likely choose TAVR
Stroke MessagesStroke MessagesTargets for ImprovementTargets for Improvement
•• Many strokes (perhaps close to 50%) occur during the Many strokes (perhaps close to 50%) occur during the
procedureprocedure
•• Distal protection may be helpfulDistal protection may be helpful
•• Many patients in this study had significant risk factors for Many patients in this study had significant risk factors for stroke, ie atrial fibrillationstroke, ie atrial fibrillationstroke, ie atrial fibrillationstroke, ie atrial fibrillation
•• Anti platelet and anti thrombotic regimen has not been Anti platelet and anti thrombotic regimen has not been standardized.standardized.
PARTNER Grading Criteria for PARTNER Grading Criteria for ParavalvularParavalvular AR AR
Circumference = 6″AR = 0.1+0.35 = 0.45″Ratio = 8%Severity = Mild (< 10%)
Circumference = 6″AR = 0.5+0.5 = 1.0″AR = 0.5+0.5 = 1.0″Ratio = 17%Severity = Moderate (10 – 20%)(Trans AR also present)
Circumference = 6″AR = 0.6+1.1 = 1.7″Ratio = 28%Severity = Severe (> 20%)
Images courtesy of Pamela Douglas, MD, FASE
Aortic Regurgitation (AT)Aortic Regurgitation (AT)
p < 0.0001 p < 0.0001 p < 0.0001 p < 0.0001
N = 279 N = 228 N = 231 N = 173 N = 217 N = 156 N = 145 N = 113
Paravalvular Aortic Regurgitation (AT)Paravalvular Aortic Regurgitation (AT)
p < 0.0001 p < 0.0001 p < 0.0001 p < 0.0001
N = 277 N = 226 N = 230 N = 172 N = 216 N = 155 N = 145 N = 112
Total AR and MortalityTotal AR and MortalityTAVR Patients (AT)TAVR Patients (AT)
Mortality
None - Trace
Mild
Moderate - Severe 50.7%
33.4%35.3%
26.2%
p (log rank) < 0.001
Months Post Procedure
Mortality
Numbers at RiskNumbers at Risk
NoneNone--TrTr 135135 125125 115115 101101 6868 3131 1111
MildMild 165165 139139 121121 111111 7171 3333 1616
ModMod--SevSev 3434 2525 2222 1919 1515 66 22
26.3%
12.7%
26.2%
Paravalular Regurgitation:A great target for innovation
1 Year 2 Years
OutcomeOutcomeAVRAVR
(N = 351)(N = 351)
TAVRTAVR
(N = 348)(N = 348)pp--valuevalue
AVRAVR
(N = 351)(N = 351)
TAVRTAVR
(N = 348)(N = 348)pp--valuevalue
Major Vascular Major Vascular complicationscomplications
13 (3.8)13 (3.8) 39 (11.3)39 (11.3) <0.001<0.001 13 (3.8)13 (3.8) 40 (11.6)40 (11.6) <0.001<0.001
Major bleeding Major bleeding –– no. (%)no. (%) 88 (26.7)88 (26.7) 52 (15.7)52 (15.7) <0.001<0.001 95 (29.5)95 (29.5) 60 (19.0)60 (19.0) 0.0020.002
Other complications at 1 and 2 Years Other complications at 1 and 2 Years All Patients (N = 699)All Patients (N = 699)
New PM New PM –– no. (%)no. (%) 16 (5.0)16 (5.0) 21 (6.4)21 (6.4) 0.440.44 19 (6.4)19 (6.4) 23 (7.2)23 (7.2) 0.690.69
Endocarditis Endocarditis –– no. (%)no. (%) 3 (1.0)3 (1.0) 2 (0.6)2 (0.6) 0.630.63 3 (1.0)3 (1.0) 4 (1.5)4 (1.5) 0.610.61
SVDSVD§§ requiring AVRrequiring AVR 00 00 00 00
MI MI –– no. (%)no. (%) 2 (0.6)2 (0.6) 00 0.160.16 4 (1.5)4 (1.5) 00 0.050.05
Acute kidney inj* Acute kidney inj* –– no. (%)no. (%) 20 (6.5)20 (6.5) 18 (5.4)18 (5.4) 0.570.57 21 (6.9)21 (6.9) 20 (6.2)20 (6.2) 0.750.75
§SVD = Structural Valve Deterioration*Renal replacement therapy
AllAll--Cause MortalityCause MortalityTransfemoral (N=492)Transfemoral (N=492)
26.4
HR [95% CI] =0.83 [0.60, 1.15]
P (log rank) = 0.25
Months
244 215 188 119 59
248 180 168 109 56
No. at Risk
TAVR
AVR
22.2
AllAll--Cause Mortality at 30 DaysCause Mortality at 30 Days
All Patientsno. of patients ( %)
TF Patientsno. of patients ( %)
TA Patientsno. of patients ( %)
TAVR AVR p-value TAVR AVR p-value TAVR AVR p-value
ITT 12 (3.4) 22 (6.5) 0.07 8 (3.3) 15 (6.2) 0.13 4 (3.8) 7 (7.0) 0.32
AT 18 (5.2) 25 (8.0) 0.15 9 (3.7) 18 (8.2) 0.05 9 (8.7) 7 (7.6) 0.79
AllAll--Cause Mortality at 30 Days Cause Mortality at 30 Days Patient SubgroupsPatient Subgroups
AT 18 (5.2) 25 (8.0) 0.15 9 (3.7) 18 (8.2) 0.05 9 (8.7) 7 (7.6) 0.79
TransfemoralTransfemoralIndex Procedure/Admission Index Procedure/Admission Resource use (perResource use (per--protocol population)protocol population)
Resource CategoryResource CategoryTFTF--TAVRTAVR((N = 234N = 234))
AVRAVR(N= 221)(N= 221)
DifferenceDifference(95% CI)*(95% CI)*
PP--valuevalue
Procedure duration (min) 244±78 330±102 87 (69 – 104) <0.001
Total hospital LOS, days
ICU
10.2 (7)
3.3 (2)
16.4 (12)
5.6 (3)
6.2 (3.8 – 8.2)
2.3 (0.9 – 3.3)
<0.001
<0.001ICU
Non-ICU
Post procedure
3.3 (2)
6.9 (4)
7.4 (5)
5.6 (3)
10.8 (8)
13.5 (10)
2.3 (0.9 – 3.3)
4.0 (2.2 – 5.5)
6.1 (3.7 – 8.0)
<0.001
<0.001
<0.001
Major vasc. complication 13.2% 3.2% 10.1% (5.1 – 15.1) <0.001
Major bleeding 9.4% 22.6% 13.2% (-6.6 to -19.9) <0.001
New pacemaker, n (%) 16 (6.8%) 13 (6.0%) 0.8% (-3.7 – 5.3) 0.73
LOS data are shown as mean (median).*95% CIs from 1,000 bootstrap replications of study data.
DJC6
Slide 33
DJC6 Consider simplifying this table a bit-- new pacemaker doesn't differ and that was reported previously, so perhaps you can get rid of that
Note that some of your confidence intervals have a dash and others have "to"David Cohen, 10/19/2011
$31,192
$4,742$5,773
$60,000
$80,000
Index Admission CostsIndex Admission CostsTransfemoralTransfemoral
$71,955
∆ ∆ ∆ ∆ = ($2,496)P = 0.53
$74,452
$34,863
$14,451
$31,192
$54,228
$0
$20,000
$40,000
TF-TAVR AVR
Procedure Non-Procedure Total MD Fees
1212--Month FollowMonth Follow--up Costsup CostsTransfemoralTransfemoral
∆ = $1,517 Total F/U Costs (12 months)
TF-TAVR $22,251
AVR $21,965
∆ = $287
P = 0.97
∆ = ($2,350)
∆ = 827
∆ = $293
TAVR vs. AVR: TAVR vs. AVR: TransfemoralTransfemoralCost per QALY gainedCost per QALY gained
yr cost (TAVR -AVR)
∆ Cost = - $2210∆ QALYs = + 0.068
ICER = dominant% dominant = 59.7
∆ Cost = - $2210∆ QALYs = + 0.068
ICER = dominant% dominant = 59.7
∆∆∆∆ QALY (TAVR - AVR)
∆∆ ∆∆1-yr cost (TAVR
% <$50,000 per QALY= 74.7
% <$50,000 per QALY= 74.7
Complete Population
Transfemoral (TF) vs Transapical (TA) Transfemoral (TF) vs Transapical (TA) All All Cause MortalityCause Mortality
Transfemoral (TF) subgroupTransfemoral (TF) subgroup Transapical (TA) subgroupTransapical (TA) subgroup
Transapical Aortic Valve Replacement Transapical Aortic Valve Replacement For Critical Aortic Stenosis: Results From For Critical Aortic Stenosis: Results From the Nonthe Non--Randomized Continued Access Randomized Continued Access Cohort of The PARTNER TrialCohort of The PARTNER Trial
Todd M. Dewey, MDon behalf of The PARTNER Trial Investigators
STS 2012 | Fort Lauderdale | Jan 30, 2012
Transapical Enrollment per SiteTransapical Enrollment per Site
Mean # pts enrolled:
PMA-TA (14 sites) = 7.4
Transapical Aortic Valve Replacement For Critical Aortic Stenosis: Results From the NonTransapical Aortic Valve Replacement For Critical Aortic Stenosis: Results From the Non--Randomized Randomized Continued Access Cohort of The PARTNER Trial ... Continued Access Cohort of The PARTNER Trial ... Todd M. Dewey, MDTodd M. Dewey, MD
Transapical Enrollment per SiteTransapical Enrollment per Site
Mean # pts enrolled:
PMA-TA (14 sites) = 7.4
NRCA-TA (22 sites) = 38.3
Transapical Aortic Valve Replacement For Critical Aortic Stenosis: Results From the NonTransapical Aortic Valve Replacement For Critical Aortic Stenosis: Results From the Non--Randomized Randomized Continued Access Cohort of The PARTNER Trial ... Continued Access Cohort of The PARTNER Trial ... Todd M. Dewey, MDTodd M. Dewey, MD
Additional Experience ThroughNon-Randomized Continued Access
500
600
700
800
900
1000
# of Patients Enrolled
Continued Access PeriodRandomized Clinical Trial
PMA-TA = 104
AVR = 103NRCA-TA = 843
0
100
200
300
400
500
May 2007
Sep2009PMA Sep 21, 2011
PMA Data Lock
April 2008
AllAll--Cause Mortality (AT)Cause Mortality (AT)
23.6%25.3%
29.1%
Early Experience
AVRAVR 9292 7676 7171 7070 6767
PMAPMA--TATA 104104 8787 8282 7676 7373
NRCANRCA--TATA 822822 571571 370370 297297 126126
No. at Risk
Later Experience
Stroke (AT)Stroke (AT)
AVRAVR 9292 7272 6767 6666 6363
PMAPMA--TATA 104104 8181 7777 7070 6767
NRCANRCA--TATA 822822 563563 365365 291291 123123
No. at Risk
3.7.%
7.0%
10.8.%
Why is TAVR the Preferred Strategy Compared Why is TAVR the Preferred Strategy Compared to AVR in the High Risk Patient?to AVR in the High Risk Patient?
• Mortality in high risk aortic stenosis patient at 2 years is the same� Mortality in TF subgroup may be lower?
• CVA is only slightly higher at 2 years (numerically, not statistically)
• Anesthesia time cut by 28% (330-236 min)
• Total procedure time cut by 42% (230-133 min)
• ICU stay cut by 40% (5-3 days)
• Total hospitalization days cut by 33% (12-8 days)
SCRIPPS CLINIC
• Total hospitalization days cut by 33% (12-8 days)
• Faster recovery (6 min walk test)
• TF approach is less costly
• Speculation:
• Mortality may be lower than AVR as clinical experience grows
• Mortality may be lower than AVR if paravalvular leak and vascular complications are improved, along with other technology advances
• Stroke may be equivalent to AVR with use of distal protection devices
TAVR Vs AVR: The Future TAVR Vs AVR: The Future
• TAVR is an infant technology� Less invasive than AVR but not yet minimally invasive
• Targets for improvement � Paravalular leak
Better sizing using advanced imaging and larger valvesPost implant dilatation with larger balloonsCircumferential sealing designs to achieve complete annular apposition
SCRIPPS CLINIC
Circumferential sealing designs to achieve complete annular apposition
� StrokeDistal protection deviceSmaller devicesAnti platelet and anti thrombotic therapies
� Vascular complicationsSmaller 14-8Fr delivery devices
� Clinician education/experienceImproved procedural techniqueImproved case selection
esp avoiding cohort C patients-too sick, even for TAVR
Death Incidence (%)
STS 5STS 5--14.914.9STS <5STS <5
Non operative TAVR
STS ≥15STS ≥15
PARTNER 1B PARTNER 1B –– Inoperable patients Inoperable patients
Mortality Stratified by STS Score (ITT)Mortality Stratified by STS Score (ITT)
Death Incidence (%)
Months
Numbers at RiskNumbers at Risk
Months
2828 2626 2525 2424 1616
1212 88 77 66 55
Months
4343 3232 2323 1919 1515
4747 2929 1919 1414 88
108108 8080 7676 6767 5252
119119 8484 5959 4242 292947