"My five top trials in #interventionalcardiology in 2019". View this extensive slideset by Andreas Baumbach @EAPCIPresident where he covers the potential impact of these trials on clinical practice & their relevance for practice guidelines ow.ly/G64930q7R1K
#ESC
#EuroPCR
6. COMPLETE: Background
● Patients undergoing primary PCI of the culprit lesion for STEMI are often
found to have multivessel CAD, with 1 or more angiographically significant
non-culprit lesions.
● There is uncertainty about how best to manage these non-culprit lesions:
• Routinely revascularise them with PCI?
• Manage them conservatively with guideline-directed medical therapy alone?
● While prior RCTs have shown non-culprit lesion PCI reduces
revascularisation, none were powered to detect moderate reductions in
hard clinical outcomes such as CV death or MI.
Mehta SR, et al. NEJM. 2019.
7. Mehta SR, et al. Am Heart J. 2019.
COMPLETE trial design
STEMI WITH MULTIVESSEL CAD AND SUCCESSFUL PCI TO THE CULPRIT LESION
≥70% stenosis or 50-69% with FFR ≤0.80
RANDOMISATION
Stratified to in-hospital or after discharge
CO-PRIMARY OUTCOMES:1. Composite of CV death or new MI
2. Composite of CV death, new MI or ischaemia-driven revascularisation
KEY SECONDARY OUTCOME: CV death, new MI, IDR, unstable angina, NYHA class IV heart failure
MEDIAN FOLLOW-UP: 3 YEARS
COMPLETE REVASCULARISATION
N=2,000
CULPRIT LESION ONLY REVASCULARISATION
N=2,000
8. Hazard Ratio 0.74
95% CI 0.60-0.91
P=0.004
NNT (median 3 years) = 37
Mehta SR, et al. NEJM. 2019.
Hazard Ratio 0.51
95% CI 0.43-0.61
P < 0.001
NNT (median 3 years) = 13
2nd Co-Primary Outcome: CV Death, MI, or IDRFirst Co-Primary Outcome: CV Death or New MI
COMPLETE: Main Results
9. 158/2016 (2.7) 213/2025 (3.7) 0.74 (0.60-0.91) 179/2016 (3.1
101/1353 (2.7) 130/1349 (3.5) 0.77 (0.59-1.00) 113/1353 (3.0
57/663 (2.7) 83/676 (3.9) 0.69 (0.49-0.97) 66/663 (3.1)
64/820 (2.7) 78/849 (3.1) 0.86 (0.62-1.20) 71/820 (3.0)
87/1097 (2.7) 129/1085 (4.2) 0.65 (0.49-0.85) 97/1097 (3.0)
127/1668 (2.6) 183/1631 (3.9) 0.67 (0.53-0.84) 143/1668 (3.0
31/346 (3.2) 29/392 (2.6) 1.23 (0.74-2.04) 36/346 (3.8)
60/885 (2.3) 90/870 (3.6) 0.65 (0.47-0.90) 69/885 (2.7)
91/1033 (3.0) 117/1070 (3.8) 0.80 (0.61-1.05) 99/1033 (3.3)
118/1623 (2.5) 171/1602 (3.7) 0.67 (0.53-0.85) 136/1623 (2.9
40/393 (3.6) 42/423 (3.5) 1.05 (0.68-1.61) 43/393 (3.9)
82/1233 (2.2) 109/1195 (3.1) 0.72 (0.54-0.96) 97/1233 (2.7)
76/783 (3.5) 104/830 (4.5) 0.77 (0.58-1.04) 82/783 (3.8)
Overall
Main pre-defined:
Intent to perform non-culprit lesion PCI 0.62
During initial hospitalization
After initial hospitalization
Proximal/mid LAD non-culprit stenosis 0.2
Presence
Absence
Non-culprit stenosis severity ≥ 80%visual or ≥ 60%core lab 0.033
Presence
Absence
Residual SYNTAX score 0.32
<6(median)
≥ 6(median)
Other exploratory pre-defined:
Gender 0.08
Male
Female
Age (years) 0.74
<65
≥ 65
CV death/MI
Subgroup Complete Culprit only
no. of events/total no. (%/yr)
HR (95%CI) P value for
Interaction
Complete
no. of eve
CV death or New MI
158/2016 (2.7) 213/2025 (3.7) 0.74 (0.60-0.91) 179/2016 (3.1
101/1353 (2.7) 130/1349 (3.5) 0.77 (0.59-1.00) 113/1353 (3.0
57/663 (2.7) 83/676 (3.9) 0.69 (0.49-0.97) 66/663 (3.1)
64/820 (2.7) 78/849 (3.1) 0.86 (0.62-1.20) 71/820 (3.0)
87/1097 (2.7) 129/1085 (4.2) 0.65 (0.49-0.85) 97/1097 (3.0)
127/1668 (2.6) 183/1631 (3.9) 0.67 (0.53-0.84) 143/1668 (3.0
31/346 (3.2) 29/392 (2.6) 1.23 (0.74-2.04) 36/346 (3.8)
60/885 (2.3) 90/870 (3.6) 0.65 (0.47-0.90) 69/885 (2.7)
Overall
Main pre-defined:
Intent to perform non-culprit lesion PCI 0.62
During initial hospitalization
After initial hospitalization
Proximal/mid LAD non-culprit stenosis 0.2
Presence
Absence
Non-culprit stenosis severity ≥ 80%visual or ≥ 60%core lab 0.033
Presence
Absence
Residual SYNTAX score 0.32
<6(median)
CV death/MI
Subgroup Complete Culprit only
no. of events/total no. (%/yr)
HR (95%CI) P value for
Interaction
Complete
no. of eve
CV death, New MI, or IDR3/2025 (3.7) 0.74 (0.60-0.91) 179/2016 (3.1) 339/2025 (6.2) 0.51 (0.43-0.61)
0/1349 (3.5) 0.77 (0.59-1.00) 113/1353 (3.0) 227/1349 (6.6) 0.47 (0.38-0.59)
676 (3.9) 0.69 (0.49-0.97) 66/663 (3.1) 112/676 (5.4) 0.59 (0.43-0.79)
849 (3.1) 0.86 (0.62-1.20) 71/820 (3.0) 141/849 (6.0) 0.50 (0.38-0.67)
9/1085 (4.2) 0.65 (0.49-0.85) 97/1097 (3.0) 190/1085 (6.5) 0.48 (0.38-0.61)
3/1631 (3.9) 0.67 (0.53-0.84) 143/1668 (3.0) 291/1631 (6.6) 0.46 (0.37-0.56)
392 (2.6) 1.23 (0.74-2.04) 36/346 (3.8) 47/392 (4.4) 0.87 (0.56-1.34)
870 (3.6) 0.65 (0.47-0.90) 69/885 (2.7) 133/870 (5.5) 0.49 (0.37-0.66)
0.62 0.27
0.2 0.82
0.033 0.01
0.32 0.98
CV death/MI CV death/MI/IDR
ulprit only
(%/yr)
HR (95%CI) P value for
Interaction
Complete Culprit only
no. of events/total no. (%/yr)
HR (95%CI) P value for
Interaction
Complete Culprit Only HR (95% CI) Interaction P
no. of events/total no. (%/yr)
Complete Culprit Only HR (95% CI) Interaction P
no. of events/total no. (%/yr)
Median Time to study NCL PCI in Complete Group
During initial hospitalisation: 1 day (IQR 1-3)
After Hospital discharge: 23 days (IQR 12.5-33.5)
/398 (5.6) 0.49 (0.32-0.74) 38/396 (3.2) 84/398 (7.7) 0.42 (0.29-0.62)
/929 (3.0) 0.74 (0.53-1.03) 69/945 (2.5) 151/929 (5.9) 0.43 (0.32-0.57)
/402 (5.0) 0.87 (0.59-1.29) 49/385 (4.8) 84/402 (7.9) 0.61 (0.43-0.87)
6/1623 (3.3) 0.70 (0.55-0.89) 130/1631 (2.7) 255/1623 (5.7) 0.48 (0.39-0.60)
5/1281 (3.3) 0.70 (0.54-0.92) 99/1298 (2.6) 196/1281 (5.5) 0.48 (0.38-0.61)
/169 (2.4) 1.07 (0.48-2.39) 18/192 (3.7) 25/169 (5.9) 0.63 (0.34-1.15)
/568 (4.8) 0.75 (0.53-1.07) 60/513 (4.1) 117/568 (7.9) 0.53 (0.39-0.73)
5/1885 (3.6) 0.77 (0.62-0.95) 168/1853 (3.2) 309/1885 (6.0) 0.53 (0.44-0.64)
/140 (4.7) 0.45 (0.21-0.97) 11/163 (2.3) 30/140 (8.5) 0.28 (0.14-0.56)
6/1861 (3.5) 0.75 (0.60-0.93) 160/1855 (3.0) 302/1861 (6.0) 0.51 (0.42-0.62)
/136 (6.3) 0.65 (0.34-1.23) 17/141 (4.3) 32/136 (9.4) 0.47 (0.26-0.84)
/145 (3.8) 0.62 (0.26-1.49) 10/136 (3.1) 21/145 (6.0) 0.50 (0.24-1.06)
6/1562 (3.2) 0.81 (0.64-1.03) 132/1568 (2.9) 241/1562 (5.6) 0.52 (0.42-0.65)
/272 (5.5) 0.51 (0.31-0.84) 31/276 (3.6) 65/272 (8.4) 0.44 (0.29-0.68)
0.35 0.27
0.62 0.68
0.18 0.07
0.68 0.78
0.23 0.79
0.1 0.2 0.5 1 2 5 10 0.1 0.2 0.5 1 2 5 10
Complete better Culprit only better Complete better Culprit only better
148/1853 (2.7) 195/1885 (3.6) 0.77 (0.62-0.95) 168/1853 (3.2
10/163 (2.1) 18/140 (4.7) 0.45 (0.21-0.97) 11/163 (2.3)
140/1855 (2.6) 186/1861 (3.5) 0.75 (0.60-0.93) 160/1855 (3.0
16/141 (4.0) 23/136 (6.3) 0.65 (0.34-1.23) 17/141 (4.3)
8/136 (2.4) 14/145 (3.8) 0.62 (0.26-1.49) 10/136 (3.1)
120/1568 (2.6) 146/1562 (3.2) 0.81 (0.64-1.03) 132/1568 (2.9
24/276 (2.7) 45/272 (5.5) 0.51 (0.31-0.84) 31/276 (3.6)
Type of PCI 0.18
Primary PCI
Pharmaco-invasive/rescue PCI
Killip class 0.68
I
≥ II
Type of stent 0.23
DES polymer free/DES biodegradable polymer
DES durable polymer
Bare metal stent
0.1 0.2 0.5 1 2 5 10
Complete better Culprit only better
Mehta SR, et al. NEJM. 2019.
Timing of non-culprit lesion PCI:
During or after initial hospitalisation
10. Relevance for my practice
• Confirms our practice to assess and treat bystander
disease
• Reassures practice of bringing patients back for a
staged procedure after 4 weeks
11. Open questions
• Do we need to treat bystander disease during the
index hospitalization ? Not randomized in
COMPLETE
• How do we assess bystander disease
• Angio only / FFR/ non-invasive/ QFR?
12. Guidelines
“As the optimal timing of revascularization
(immediate vs. staged) has not been adequately
investigated, no recommendation in favour of immediate vs.
staged multivessel PCI can be formulated.”
15. BIOSTEMI: Background
● A novel sirolimus eluting stent (SES) with ultrathin struts (60/80 u) and
absorbable polymer was non-inferior to the standard everolimus eluting
stent with durable polymer EES in the Bioscience Study
● A subgroup analysis suggested possible superiority in STEMI
Pilgrim T, et al. Lancet. 2014.
ST-elevation MI
Yes 7/211 17/196 0.38 (0.16-0.91) 0.024 0.014
No 62/852 53/860 1.20(0.83-1.73) 0.33
Favours BP SES Favours DP EES0.25 0.5 1 2 4
16. 651 patients DP-EES649 patients BP-SES
1300 patients with STEMI
We used Bayesian statistical methods with
robust priors incorporating historical data
from 407 patients with acute STEMI included
in the BIOSCIENCE randomized trial
”
“
Methods
BCI, Bayesian credibility interval
Posterior probability of superiority: 0.986
Iglesias JF, et al. The Lancet. 2019.
BIOSTEMI
Superiority
margin
Rate ratio
Probabilitydensity
BIOSTEMI
only
BIOSTEMI
with historical data
95% BCI95% BCI
BIOSCIENCE
STEMI only
3.0
3.5
2.0
2.5
1.0
1.5
0.0
0.5
1.20.8 1.00.4 0.60.0 0.2
Results
17. PRIMARY ENDPOINT
TARGET LESION FAILURE AT 1 YEAR
DP-EES 5.5%
BP-SES 3.9%
0
1
2
3
4
5
6
7
8
9
TARGETLESIONFAILURE(%)
0 30 60 90 120 365
DAYS SINCE INDEX PROCEDURE
330300270150 180 210 240
Difference -1.6 percentage points
Rate ratio (95% BCI)=0.59 (0.37-0.94)
Bayesian posterior probability 0.986
BCI, Bayesian credibility interval
Iglesias JF, et al. The Lancet. 2019.
RESULTS
COMPONENTS OF THE PRIMARY ENDPOINT
AT 1 YEAR
3,9
2,8
0,8
1,4
5,5
2,9
0,9
2,6
0
1
2
3
4
5
6
TLF Cardiac death TV-MI Clinically-indicated
TLR
BP-SES
DP-EES
RR 0.59 0.77 0.55 0.55
95% BCI 0.37-0.94 0.43-1.40 0.19-1.60 0.26-1.13
Bayesian
posterior
probability
0.986 0.806 0.875 0.949
18. Relevance for my practice
• Depends on the availability of this platform
• Generally points towards benefits with thin strut
platforms
• We can still improve outcomes with novel stent
technology
24. Statistical Methodology for the 5-Year Analysis (i)
• Only the 5-year composite primary outcome measure of death, stroke
or MI was powered for superiority testing
• All other individual endpoints were non-powered and not adjusted for
multiplicity, and thus are hypothesis generating
•The only P-value provided is for the primary endpoint at 5 years
• More pts were lost to FU after CABG than after PCI
•Multiple imputation was performed as a sensitivity analysis to
account for missing follow-up data
25. Randomization and Follow-up
1905 pts
with unprotected
left main CAD
were enrolled
between
9/29/2010 and
3/6/2014 at
126 sites in
17 countries
PCI with
CoCr-EES
N=948
CABG
N=957
R
Initial treatment
PCI (n=935)
CABG (n=7)
No revascularization (n=6)
Initial treatment
PCI (n=17)
CABG (n=923)
No revascularization (n=17)
30-day
follow-up
5-year
follow-up
3-year
follow-up
942 (99.4%)
940 (98.2%)
918 (96.8%)
899 (93.9%)
884 (93.2%)
862 (90.1%)
Withdrew; n=16
Lost to follow-up; n=1
Withdrew; n=17
Lost to follow-up; n=24
Withdrew; n=1
Lost to follow-up; n=36
Withdrew; n=6
Lost to follow-up; n=0
Withdrew; n=5
Lost to follow-up; n=19
Withdrew; n=0
Lost to follow-up; n=34
31. Conclusions
“In patients with left main coronary artery disease of low or
intermediate anatomical complexity, there was no significant
difference between PCI and CABG with respect to the rate of the
composite outcome of death, stroke, or myocardial infarction at
5 years.”
Stone et al. NEJM 2019
32. Open questions
• What happens out to ten years?
• How do the results compare to NOBLE 5 years
33. Guideline Recommendations
for LM Revascularization
Levine G, et al. J Am Coll Cardiol. 2011;58:44-122; Neumann et al, EHJ 2018
United States Europe
PCI CABG
Low
SxScore 0-22 IIa B I B
Intermediate
SxScore 23-32 IIb B I B
High
SxScore >32 III B I B
PCI CABG
Low
SxScore 0-22 I B I B
Intermediate
SxScore 23-32 IIa B I B
High
SxScore >32 III B I B
36. Relevance for my practice
• Stenting of LMS appears to give acceptable acute
and longterm results
• With time there is an increase in spontaneous MI in
the PCI group
• Patients should be informed accordingly
• Selection continues via Heart Team
46. Relevance for my practice
• As a strategy Prasugrel appears superior to
Ticagrelor in the setting of rapid access to invasive
diagnosis and treatment
• In my practice, patients admitted with NSTE-ACS
frequently wait for 24-48 hours before their
angiogram
50. Low Risk Patient Group
• Selection according to he Heart Team
• Surgical risk: STS-PROM Score <4% Partner
• Surgical risk <3% at 30d Corevalve LR
• Mean Age 74yrs
57. Procedure and discharge
Leon et al, TVT 2019
Shorter procedure,
Shorter hospital stay
More discharge
home
With TAVI
58. Complications
Leon et al, TVT 2019, Mack et al NEJM 2019 Appendix
New permanent pacemaker 6.5% (32) 4.0% (18)
New onset of AF 5% (21) 39.5% (145)
Low risk of procedure
specific severe
complications for both
procedures
Increased incidence of
AF after surgery
64. Events to 1 year
TAVI SAVR Difference
(95% BCI)
All death 2.9 4.6 −1.8 (−4.0 to 0.4)
All stroke 4.1 4.3 −0.2 (−2.4 to 1.9)
Rehospitalisation for HF 3.2 6.5 −3.4 (−5.9 to −1.0)
Pacemaker 17.4 6.1 12.6 (9.2 to 16.2)
Combined Safety EP 30days 5.3 10.7 −5.4 (−8.3 to −2.6)
Popma et al NEJM 2019
68. Low Risk
• 2 trials showing superior and non-inferior results
• Improved patient journey
• Improved recovery
Should all low risk patients be treated by TAVI ?
69. Metaanalysis
TAVI is associated with
lower mortality across
all surgical risk groups
Should patients be treated
with TAVI
Irrespective of surgical risk ?
70. Summary
• New randomized evidence shows favorable results of TAVI
compared with SAVR for patients at low surgical risk
• TAVI in anatomically suitable patients has now been
shown to be at least as good as SAVR across the risk
spectrum
• TAVI is associated with overall lower periprocedural
complications and faster recovery
71. Relevance for my practice
• Patients suitable for transfemoral TAVI can be given
the option
• The discussion with the surgeons at the Heart Team
meeting has changed…..
72. Open Questions
• Bicuspid valves
• AS and coronary disease
• Embolic protection
• Anticoagulation
• Durability and extended durability with ViV
• Coronary Access for Diagnostic and Intervention
77. MY Top 5 TRIALS 2019
• COMPLETE
• BIOSTEMI
• ISAR REACT 5
• EXCEL 5 Years
• PARTNER 3/ EVOLUT LOW RISK
……there were many more interesting results,
advancing our knowledge and the field in 2019
78. 2019 Comprehensive Review
Eur Heart J 2020; in press
The Year in cardiology 2019 Series
The Year in Cardiology 2019 – Coronary intervention
Baumbach, Bourantas, Serruys, Wijns
The Year in Cardiology 2019 - ACS
Banning, Crea, Lüscher
The Year in Cardiology 2019 – Valvular Heart disease
Prendergast, Binder