Incidence and Outcomes of Valve Hemodynamic

Deterioration in Transcatheter Aortic Valve

Replacement in U.S. Clinical Practice: A Report from
the Society of Thoracic Surgery / American College
of Cardiology Transcatheter Valve Therapy Registry

Sreekanth Vemulapalli MD, David Dai MS, Michael Mack MD, David Holmes MD,
Fred Grover MD, Raj Makkar MD, Vinod H. Thourani MD, Pamela S. Douglas MD
On behalf of the STS/ACC TVT Registry
Disclosures, Funding and Disclaimer

Sreekanth Vemulapalli, MD
Abbott Vascular, American College of Cardiology

This research was supported by the American College of Cardiology Foundation’s National
Cardiovascular Data Registry (NCDR). The views expressed in this presentation represent those
of the author(s), and do not necessarily represent the official views of the NCDR or its
associated professional societies identified at www.ncdr.com.
 Background
• TAVR effectively treats patients with severe aortic stenosis
• There are recent reports of TAVR leaflet abnormalities and valve thrombosis (4DCT / TEE) or
Valve Hemodynamic Deterioration (VHD) (increase in aortic valve mean gradient)
• Planned prospective studies to investigate this using advanced imaging will take years to
complete
• STS / ACC TVT Registry: Collaboration of STS, ACC, CMS, FDA, hospitals, industry, SCAI, AATS,
NIH, and consumer advocates
– Unique opportunity to track current TAVR performance in the community
– All commercial valve implantations in US
– Linked to CMS database for long term follow up
– Prespecified post-procedure, 30-day, 1-year transthoracic echo (TTE)
– TTEs are site read
 Objectives and Methods
Short Term Post-procedure 30-day
Cohort Echo Echo

TAVR

Aim 1: Incidence of VHD

(≥ 10 mm Hg ↑ gradient)
 Objectives and Methods
Short Term Post-procedure 30-day
Cohort Echo Echo

TAVR

Aim 1: Incidence of VHD

(≥ 10 mm Hg ↑ gradient)

Long Term Cohort

TAVR

30-day Echo 1-year Echo

 Objectives and Methods
Aim 2: 18-month
Short Term Post-procedure 30-day cardiovascular
Cohort Echo Echo outcomes

TAVR

Aim 1: Incidence of VHD Mortality

(≥ 10 mm Hg ↑ gradient) Stroke
Heart failure
Long Term Cohort Re-intervention

TAVR

30-day Echo 1-year Echo CMS linkage

 Objectives and Methods
Aim 2: 18-month
Short Term Post-procedure 30-day cardiovascular
Cohort Echo Echo outcomes

TAVR

Aim 1: Incidence of VHD Mortality

(≥ 10 mm Hg ↑ gradient) Aim 3: Predictors of VHD Stroke
Backwards selection regression model Heart failure
Long Term Cohort Re-intervention

TAVR

30-day Echo 1-year Echo CMS linkage

Study 35,268 TAVRs
34,977 patients (365 sites)
Design November 2011 – March 2015
661 cases aborted

12,212 without CMS linkage

124 died in lab

22,271 TAVRs
22,231 patients (360 sites)
 Study 35,268 TAVRs
34,977 patients (365 sites)
Design November 2011 – March 2015
661 cases aborted

12,212 without CMS linkage

124 died in lab

22,271 TAVRs
22,231 patients (360 sites)

269 deaths before post gradient

4340 missing post gradient

769 deaths before next gradient

6794 missing 30-day gradient

Short Term 10,099 TAVRs

Cohort
(↑0–30 days) 10,095 patients (334 sites)
 Study 35,268 TAVRs
34,977 patients (365 sites)
Design November 2011 – March 2015
661 cases aborted

12,212 without CMS linkage

124 died in lab

22,271 TAVRs
22,231 patients (360 sites)

269 deaths before post gradient 1111 deaths before 30-day

gradient
4340 missing post gradient 8716 missing 30-day gradient

769 deaths before next gradient 864 deaths before next gradient

6794 missing 30-day gradient 8405 missing 1-year gradient

Short Term Long Term

10,099 TAVRs 3175 TAVRs Cohort
Cohort
(↑0–30 days) 10,095 patients (334 sites) 3175 patients (254 sites) (↑30 days–1 yr)
 Baseline Patient Characteristics
Short Term Long Term
Cohort
(N=10,099) Cohort
(N=3175)
Age (years), median (IQR) 84.0 (78.0,88.0) 84.0 (78.0,88.0)
Male sex 5182 (51.3) 1487 (46.8)
Hypertension 9003 (89.1) 2801 (88.2)
Diabetes Mellitus 3593 (35.6) 1109 (34.9)
Prior MI 2405 (23.8) 774 (24.4)
Prior stroke or TIA 1891 (18.7) 582 (18.3)
Atrial fibrillation/flutter 4146 (41.1) 1222 (38.5)
Dialysis dependent 379 (3.8) 84 (2.6)
STS PROM Score, median (IQR) 6.7 (4.5,10.0) 6.4 (4.5,9.6)
Aspirin (Discharge) 8798 (87.1) 2816 (88.7)
Warfarin (Discharge) 2510 (24.9) 780 (24.6)
Dabigatran (Discharge) 2602 (25.8) 832 (26.2)
P2Y12 inhibitor (Discharge) 6586 (65.2) 2106 (66.3)
Factor Xa inhibitor (Discharge) 373 (3.7) 59 (1.9)
 Procedure and Echo Variables
Short Term Cohort Long Term Cohort
(N=10,099) (N=3175)
Baseline echo variables
LVEF median (IQR) 58.0 (45.0,64.0) 58.0 (48.0,63.5)
Procedure variables
Balloon expanding valve 8029 (79.5) 2981 (93.9)
Self-expanding valve 2068 (20.5) 194 (6.1)
Valve size = 23 mm 3273 (32.4) 1376 (43.3)
Valve size = 26 mm 4502 (44.6) 1647 (51.9)
Valve size = 29 mm 1612 (16.0) 91 (2.9)
Valve size = 31 mm 710 (7.0) 61 (1.9)
Valve in valve 486 (4.8) 137 (4.3)
Postprocedure echo variables
Valve oversizing 1.2 (1.1,1.4) 1.3 (1.1,1.4)
Mean AV gradient mm Hg, median (IQR) 9.0 (6.0,12.0) 10.0 (7.0,13.0)
EOA index cm2, median (IQR) 1.0 (0.8,1.2) 0.9 (0.7,1.2)
PPM present (moderate/severe) 2957 (29.3%) 847 (26.7%)
 Incidence of VHD
VHD defined as ↑ AS mean gradient ≥ 10 mm Hg
Short Term Cohort (↑ gradient 0–30 days) Long Term Cohort (↑ gradient 30 day–1 yr)

VHD 2.5%
VHD 2.1%

VHD + Death VHD + Death

(0–30 d) (30 d–1 yr)
7.1% 23.5%
 Landmark Cumulative Incidence of
Mortality/Stroke /Aortic Valve Reintervention
Short Term Cohort (↑ gradient 0–30 days)
30
30
Composite % (Mortality/Stroke/AVRI)

<10 >=10
<10 mm Hg
≥ 10 mm Hg
Composite (mortality/stroke/AVRI) (%)

20
20

10
10

00
00 3 6 9 12 15 18

<10 9888 7734

Months
5925
from Index
4423
Procedure
Months from index procedure
3359 2609 2024
No. at >=10 10099
211 173 140 105 87 70 50

Risk 211
 Landmark Cumulative Incidence of
Mortality/Stroke /Aortic Valve Reintervention
Short Term Cohort (↑ gradient 0–30 days)
30
30
Composite % (Mortality/Stroke/AVRI)

<10 >=10
<10 mm Hg
≥ 10 mm Hg
Composite (mortality/stroke/AVRI) (%)

20
20

10
10

00
00 33 66 99 1212 15
15 18
18

<10 9888 7734

Months from Index Procedure
Months
5925
from index
4423
procedure
3359 2609 2024
No. at>=1010099
211 7734
173 5925
140 4423
105 3359
87 2609
70 2024
50

Risk 211 173 140 105 87 70 50

 Landmark Cumulative Incidence of
Mortality/Stroke /Aortic Valve Reintervention
Short Term Cohort (↑ gradient 0–30 days) Long Term Cohort (↑ gradient 30 day–1 yr)
30
30 30
30
<10 >=10
Composite % (Mortality/Stroke/AVRI)

<10 >=10

<10 mm Hg <10 mm Hg
≥ 10 mm Hg ≥ 10 mm Hg

Composite (mortality/stroke/AVRI) (%)

Composite (mortality/stroke/AVRI) (%)

20
20 20
20

10
10 10
10

00 00
00 33 6 6 99 12 12 15
15 18
18 00 3 6 9 12 15 18

<10 9888 7734

MonthsMonths
fromfromIndex
5925 4423
Procedure
index procedure
3359 2609 2024 <10 9888 3031
Months
3004
from Index
2943
Procedure
Months from index procedure
2548 2036 1588
No. at >=10 10099
211 7734 173 5925140 4423
105 3359
87 2609
70 2024 10099
50 211 >=10 77 75 75 63 56 45

Risk 211 173 140 105 87 70 50 211

 Landmark Cumulative Incidence of
Mortality/Stroke /Aortic Valve Reintervention
Short Term Cohort (↑ gradient 0–30 days) Long Term Cohort (↑ gradient 30 day–1 yr)
30
30 30
30
<10 >=10
Composite % (Mortality/Stroke/AVRI)

<10 >=10

<10 mm Hg <10 mm Hg
≥ 10 mm Hg ≥ 10 mm Hg

Composite (mortality/stroke/AVRI) (%)

Composite (mortality/stroke/AVRI) (%)

20
20 20
20

10
10 10
10

00 00
00 33 6 6 9 9 12 12 15
15 18
18 00 33 66 99 1212 15 18

<10 9888 7734

MonthsMonths
fromfromIndex
5925 4423
Procedure
index procedure
3359 2609 2024 <10 9888 3031
Months
3004
from Index
Months from index
2943
Procedure
procedure
2548 2036 1588
No. at>=1010099
211 7734
173 5925
140 4423
105 3359
87 2609
70 2024
50 >=1010099
211 3031
77 3004
75 2943
75 2548
63 56 45

Risk 211 173 140 105 87 70 50 211 77 75 75 63

 Landmark Cumulative Incidence of
Mortality/Stroke /Aortic Valve Reintervention
Short Term Cohort (↑ gradient 0–30 days) Long Term Cohort (↑ gradient 30 day–1 yr)
30
30 30
30
<10 >=10
Composite % (Mortality/Stroke/AVRI)

<10 >=10

<10 mm Hg <10 mm Hg
≥ 10 mm Hg ≥ 10 mm Hg

Composite (mortality/stroke/AVRI) (%)

Composite (mortality/stroke/AVRI) (%)

20
20 20
20

10
10 10
10

00 00
00 33 6 6 9 9 12 12 15
15 18
18 00 33 66 99 1212 15
15 18
18

<10 9888 7734

MonthsMonths
fromfromIndex
5925 4423
Procedure
index procedure
3359 2609 2024 <10 9888 3031
Months
3004
from Index
Months from index
2943
Procedure
procedure
2548 2036 1588
No. at>=1010099
211 7734
173 5925
140 4423
105 3359
87 2609
70 2024
50 >=1010099
211 3031
77 3004
75 2943
75 2548
63 2036
56 1588
45

Risk 211 173 140 105 87 70 50 211 77 75 75 63 56 45

 18-Month Outcomes
Short Term Cohort (↑ gradient 0–30 days) Long Term Cohort (↑ gradient 30 day–1 yr)

35% 35%
<10 mm Hg <10 mm Hg
30%
≥ 10 mm Hg 30% ≥ 10 mm Hg
25%
p=ns 25% p=ns
Rate

Rate
20% 20%

15% 15%

10% 10%

5% 5%

0% 0%
Death Heart Failure MI Stroke Aortic Valve Death Heart Failure MI Stroke Aortic Valve
Reintervention Reintervention

Event
 Factors Associated with VHD
↑Age (per 5 years)
Discharge P2Y12
Male
Discharge Factor Xa inhibitor
↑BMI (per 5 kg / m2)
Severe Chronic Lung Disease
Short Term Cohort

Long Term Cohort

Afib/flutter
↑Baseline AoV Gradient (per 5 mm Hg)
Valve in Valve
26 mm TAVR (vs 23 mm)
26 mm TAVR (vs 23 mm)
29 mm TAVR (vs 23 mm)
29 mm TAVR (vs 23 mm)
31 mm TAVR (vs 23 mm)
31 mm TAVR (vs 23 mm)
↑Pre discharge gradient (per 1 mm Hg)
↑Pre discharge gradient
Severe PPM
0.1 1 5
Odds Ratio for VHD
 Summary
• There is a small but present incidence of Valvular Hemodynamic
Deterioration after TAVR (defined by ↑ AV gradient ≥ 10 mmHg)
– 2.1% in the post-procedure to 30 day timeframe
– 2.5% in the 30 day to 1-year timeframe
• VHD does not appear to be not associated with excess events
– Cumulative incidence of a composite of death, stroke, and aortic valve re-
intervention and of its components are similar between those with and without
VHD
• Predictors of VHD include both patient and procedural factors
– Patient: Male, ↑BMI severe lung disease,
– Procedural: 23 mm TAVR valve, valve-in-valve, ↑Baseline AoV gradient, severe
PPM
 Limitations
• Retrospective analysis using site reported, surveillance echo data
obtained at pre-specified time points
– Uncertain relationship to clinical events, if any
– May also detect asymptomatic or clinically unapparent VHD
• Definition of VHD (↑ 10 mm Hg mean gradient) is not validated
• Incidence of VHD may be underestimated due to death/reoperation
before follow-up gradient measurement
– The incidence of VHD when including death is 2–4x the rate of VHD
• Significant echo data missingness; Clinical follow-up only in CMS pts
• Absence of 4DCT/TEE to determine etiology of VHD or leaflet
abnormalities
 Conclusions
• Incidence of VHD as reported in clinical practice is low; ~2%
• VHD is not clearly associated with adverse CV events
• These findings, especially patient and procedural
predictors, may help to inform TAVR care including patient
selection, surveillance and preventive strategies
• Large, prospective studies using advanced imaging
(4DCT/TEE) are necessary to fully elucidate the incidence,
mechanisms and consequences of VHD