pci vs cabg for multivessel and unprotected lm disease€¦ · cabg pci pvalue death 9.3% 10.2%...

PCI Vs CABG for MultivesselPCI Vs CABG for Multivesseland Unprotected LM Diseaseand Unprotected LM Disease

SCRIPPS CLINIC

Paul S. Teirstein

Chief of Cardiology

Director, Interventional Cardiology

Scripps Clinic

Disclosure Statement of Financial InterestDisclosure Statement of Financial Interest

Within the past 12 months, I or my spouse/partner have had a financialinterest/arrangement or affiliation with the organization(s) listed below.

Affiliation/Financial Relationship Company

Grant/Research Support Boston, Abbott, Medtronic, Edwards

Consulting Fees/Honoraria Boston, Abbott, Medtronic, Edwards

President (unpaid) NBPAS

SCRIPPS CLINIC

PCI and CABG for Treating Stable Coronary Artery DiseaseJACC Review Topic of the WeekTorsten Doenst, Axel Haverich, Patrick Serruys, Robert O. Bonow, Pieter Kappetein,Volkmar Falk, Eric Velazquez, Anno Diegeler and Holger Sigusch

Torsten Doenst et al. JACC 2019;73:964-976

2019 American College of Cardiology Foundation

SYNTAX 4-year Outcomes in the 3VD Subgroup • TCT 2011 • November, 2011 • Mohr • Slide 5

Survival = 70.6% PCI Vs 74.4% CABG NNT 26


Survival = 70.8% PCI Vs 78.1% CABG NNT 14


CABG PCI P value

Death 8.8% 17.8% 0.02

CVA 2.6% 5.1% 0.31

MACCE to 5 Years by SYNTAX Score TercileMACCE to 5 Years by SYNTAX Score Tercile3VD Subset3VD Subset High Scores ≥33High Scores ≥33

TAXUS (N=155)

CABG (N=166)

P<0.00141.9%

3-Vessel Disease

Cum

ula

tive

Event

Rate

(%)

50


Site-reported Data; ITT populationCumulative KM Event Rate ± 1.5 SE; log-rank P value

MI 1.9% 8.7% 0.008

Death,CVA or

MI12.5% 26.2% 0.002

Revasc. 12.6% 28.2% <0.001

24.1%

Months Since Allocation

Cum

ula

tive

Event

Rate

(%)

0 12 24

0

25

4836 60

CABG PCI P value

Death 9.6% 16.3% 0.047

CVA 3.6% 2.5% 0.53

MACCE to 5 Years by SYNTAX ScoreMACCE to 5 Years by SYNTAX Score TercileTercile3VD Subset3VD Subset Intermediate Scores 23Intermediate Scores 23--3232

TAXUS (N=207)

CABG (N=208)

37.9%

3-Vessel Disease

Cum

ula

tive

Event

Rate

(%)

50

P<0.001


MI 3.1% 13.8% <0.001

Death,CVA or

MI14.7% 23.2% 0.04

Revasc. 11.0% 25.1% <0.001


22.6%


Cum

ula

tive

Event

Rate

(%)

0 12 24

0

25

4836 60

CABG PCI P value

Death 9.3% 10.2% 0.81

CVA 3.9% 1.8% 0.24

MACCE to 5 Years by SYNTAX ScoreMACCE to 5 Years by SYNTAX Score TercileTercile3VD Subset3VD Subset Low Scores 0Low Scores 0--2222

3-Vessel Disease

TAXUS (N=181)

CABG (N=171)

Cum

ula

tive

Event

Rate

(%)

50

P=0.21


MI 4.9% 8.8% 0.20

Death,CVA or

MI14.8% 17.5% 0.56

Revasc. 14.6% 23.1% 0.04


33.3%

26.8%


Cum

ula

tive

Event

Rate

(%)

0 12 24

0

25

4836 60

CABG Vs PCI: A Different PerspectiveCABG Vs PCI: A Different Perspective

• The BEST trial: In patients with mean SYNTAX score of 25, outcomedifferences were confined to the Repeat Revascularization endpoint:

PCI CABGAny Revascularization: 11% 5.4% (51% relative reduction)

SCRIPPS CLINIC

Any Revascularization: 11% 5.4% (51% relative reduction)

Freedom from Revasc: 89% 94.6% (6.2% absolute increase)

NNT to prevent one revascularization =18

Are we really going to bypass 18 patients to prevent one repeat PCI?

PCI Vs CABG for UnprotectedPCI Vs CABG for UnprotectedLeft MainLeft Main

SCRIPPS CLINIC

Left MainLeft Main

The Synergy between PercutaneousThe Synergy between PercutaneousCoronary Intervention with TAXUS andCoronary Intervention with TAXUS and

Cardiac Surgery: The SYNTAX StudyCardiac Surgery: The SYNTAX Study

The 5The 5--year Outcomes of the SYNTAXyear Outcomes of the SYNTAXTrial in the Subset of Patients WithTrial in the Subset of Patients With

ThreeThree--vessel Diseasevessel Disease

The Synergy between PercutaneousThe Synergy between PercutaneousCoronary Intervention with TAXUS andCoronary Intervention with TAXUS and

Cardiac Surgery: The SYNTAX StudyCardiac Surgery: The SYNTAX Study

The 5The 5--year Outcomes of the SYNTAXyear Outcomes of the SYNTAXTrial in the Subset of Patients WithTrial in the Subset of Patients With

ThreeThree--vessel Diseasevessel Disease


ThreeThree--vessel Diseasevessel DiseaseThreeThree--vessel Diseasevessel Disease

Friedrich W. Mohr, MD PhDFriedrich W. Mohr, MD PhDUniversity of Leipzig Heart CenterUniversity of Leipzig Heart CenterOn behalf of the SYNTAX investigatorsOn behalf of the SYNTAX investigators

Tuesday, November 8th 11:33Tuesday, November 8th 11:33Oral Abstract Sessions: DrugOral Abstract Sessions: Drug--eluting Stents I andeluting Stents I and

BioabsorbableBioabsorbable ScaffoldsScaffoldsConflicts of Interest: NoneConflicts of Interest: None

P=0.12

Cum

ula

tive

Event

Rate

(%)

50

Before 1 year*

13.7% vs 15.8%

P=0.44

1-2 years*

7.5% vs 10.3%P=0.22

2-3 years*

5.2% vs 5.7%P=0.78

3-4 years*

6.4% vs 8.3%P=0.35

36.9%

TAXUS (N=357)CABG (N=348)

MACCE to 5 YearsLeft Main SubsetLeft Main Subset

4-5 years*

5.9% vs 5.5%P=0.82

SYNTAX 4-year Outcomes in the LM Subgroup • TCT 2011 • November 2011 • Serruys • Slide 19

31.0%

0

Cum

ula

tive

Event

Rate

(%)

25

Months Since Allocation0 12 6024 36 48

Serruys PW. TCT2012

SYNTAXES 10 year data


TAXUS (N=135)

CABG (N=149)CABG PCI P value

Death 14.1% 20.9% 0.11

CVA 4.9% 1.6% 0.13

LM Disease

Cum

ula

tive

Event

Rate

(%)

50

MACCE to 5 Years by SYNTAX Score TercileMACCE to 5 Years by SYNTAX Score TercileLM SubsetLM Subset High Scores ≥33High Scores ≥33

46.5%P=0.003


MI 6.1% 11.7% 0.13

Death,CVA or

MI22.1% 26.1% 0.40

Revasc. 11.6% 34.1% <0.001Months Since Allocation

Cum

ula

tive

Event

Rate

(%)

0 12 24

0

25

4836 60

29.7%

Serruys PW. TCT2012

CABG PCI P value

Death 19.3% 8.9% 0.04

CVA 3.6% 1.0% 0.23

TAXUS (N=103)

CABG (N=92)

32.7%

LM Disease

Cum

ula

tive

Event

Rate

(%)

50

MACCE to 5 Years by SYNTAX ScoreMACCE to 5 Years by SYNTAX Score TercileTercileLM SubsetLM Subset Intermediate Scores 23Intermediate Scores 23--3232

P=0.88


MI 4.6% 6.0% 0.71

Death,CVA or

MI24.9% 15.7% 0.11

Revasc. 16.6% 22.2% 0.40

32.7%

32.3%


Cum

ula

tive

Event

Rate

(%)

0 12 24

0

25

4836 60

Serruys PW. TCT2012

CABG PCI P value

Death 11.3% 7.0% 0.28

CVA 4.1% 1.8% 0.28

LM Disease

TAXUS (N=118)

CABG (N=104)

MACCE to 5 Years by SYNTAX ScoreMACCE to 5 Years by SYNTAX Score TercileTercileLM SubsetLM Subset Low Scores 0Low Scores 0--2222

Cum

ula

tive

Event

Rate

(%)

50

P=0.74


MI 3.1% 6.2% 0.32

Death,CVA or

MI15.2% 13.9% 0.71

Revasc. 20.3% 23.0% 0.65

30.4%

31.5%


Cum

ula

tive

Event

Rate

(%)

0 12 24

0

25

4836 60

Serruys PW. TCT2012

EXCELFive-year Outcomes from a Randomized

Trial of PCI vs. CABG in Patients withLeft Main Coronary Artery Disease

Gregg W. Stone MDfor A. Pieter Kappetein, Joseph F. Sabik,

Patrick W. Serruys and the EXCEL investigators

Left Main Coronary Artery Disease

Randomization and Follow-up

1905 ptswith unprotected

left main CADwere enrolled

PCI withCoCr-EES

N=948

Initial treatment

PCI (n=935)

CABG (n=7)

No revascularization (n=6)

942 (99.4%) 918 (96.8%) 884 (93.2%)

Withdrew; n=6Lost to follow-up; n=0



were enrolledbetween

9/29/2010 and3/6/2014 at126 sites in17 countries

CABGN=957

R

Initial treatment

PCI (n=17)

CABG (n=923)

No revascularization (n=17)

30-dayfollow-up

5-yearfollow-up

3-yearfollow-up

940 (98.2%) 899 (93.9%) 862 (90.1%)




Study Design

2900 pts with unprotected left main disease2900 pts with unprotected left main disease

SYNTAX score ≤32SYNTAX score ≤32Consensus agreement of eligibility and equipoise by heart teamConsensus agreement of eligibility and equipoise by heart team

YesYes

NoNo(N=1000)(N=1000)

R

FollowFollow--up: 1 month, 6 months, 1 year, annually through 5 yearsup: 1 month, 6 months, 1 year, annually through 5 yearsPrimary endpoint:Primary endpoint: Measured at a median 3Measured at a median 3--yr FU, minimum 2yr FU, minimum 2--yr FUyr FU

YesYes

(N=1900)(N=1900)EnrollmentEnrollment

registryregistry

PCI (PCI (XienceXience EES)EES)(N=950)(N=950)

CABGCABG(N=950)(N=950)

Stratified by diabetes,Stratified by diabetes, SYNTAX score andSYNTAX score andcentercenter

FU = 93.3% FU = 91.0%

SYNTAX Score

Site Reported Core LabLow (≤22)

Intermediate (23-32)

High (≥33)

PCI59.2%40.8%

42.8%

25.1%

32.2%

CABG61.8%38.2%

Mean 20.6 ± 6.2

Mean 20.5 ± 6.1

P=0.5237.3%

23.4%

39.3%

Mean 26.9 ± 8.8

Mean 26.0 ± 9.8

P=0.005

Core Lab DataPCI (N=942) CABG (N=936)

Qualifying LM lesion*

- LM coronary segment 97.6% 97.0%

- LM equivalent disease** 1.2% 1.5%

- Neither 1.3% 1.5%

Distal LM bifurcation or trifurcation ds. 81.8% 79.2%Distal LM bifurcation or trifurcation ds. 81.8% 79.2%

# Diseased non-LM coronary arteries*

- 0 17.3% 17.8%

- 1 31.0% 31.2%

- 2 34.5% 31.5%

- 3 17.2% 19.4%

*DS ≥50% by QCA**DS of both the ostial left LAD and ostial LCX ≥50% by QCA

15

20

25

CABG (n=957)

PCI (n=948)

str

ok

eo

rM

I(%

)

Primary EndpointAll-cause Death, Stroke or MI at Median 3 Years

15.4%14.7%

0

5

10

De

ath

,s

tro

ke

0 12 24 36 48 60

MonthsPCI 948

Number at risk:

850 784 445

CABG 957 817 763 458

Stone GW et al. N Engl J Med 2016;375:2223-35

HR [95%CI] =1.00 [0.79, 1.26]

P = 0.98

15

20

25

22.0%

19.2%

str

ok

eo

rM

I(%

)

Primary EndpointAll-cause Death, Stroke or MI at 5 Years

CABG (n=957)

PCI (n=948)

OR [95% CI] =1.19 [0.95, 1.50]

P=0.13

0

5

10

De

ath

,s

tro

ke

0 12 24 36 48 60

MonthsPCI 948

Number at risk:

854 809 778 738 486

CABG 957 818 789 763 734 532

Piecewise Hazards

All-cause Death, Stroke or MIThree distinct periods of varying relative risk

15

20

or

MI

(%)

15.1%

0-day to 30-day HR: 0.61 [95% CI: 0.42, 0.88]; P-value = 0.008

30-day to 1-year HR: 1.07 [95% CI: 0.68, 1.70]; P-value = 0.76

1-year to 5-year HR: 1.61 [95% CI: 1.23, 2.12]; P-value <0.001

Treatment-time interaction: P<0.001CABG (n=957)

PCI (n=948)

0

5

10

De

ath

,s

tro

ke

or

MI

(%)

0 12 24 36 48 60

Months

8.0%

3.8%

9.7%

4.9% 4.1%

Number at risk:

PCIPCI

CABGCABG 957

948

889 856 827 794 579

902 854 819 776 511


929

933

1

15

20

25

22.0%

19.2%

str

ok

eo

rM

I(%

)

Primary EndpointAll-cause Death, Stroke or MI at 5 Years

CABG (n=957)

PCI (n=948)

78% vs 80.8% Freedom From Death, Stroke or MIABS diff = 2.8 NNT = 37

Need to do 37 CABG to prevent 1 Death, Stroke or MI

OR [95% CI] =1.19 [0.95, 1.50]

P=0.13

0

5

10

De

ath

,s

tro

ke

0 12 24 36 48 60

MonthsPCI 948

Number at risk:

854 809 778 738 486

CABG 957 818 789 763 734 532

Primary Endpoint at 5 YearsPCI (N=948) CABG (N=957) Difference [95% CI] Odds ratio [95% CI]

Death, stroke or MI 22.0% (203) 19.2% (176) 2.8% [-0.9%, 6.5%] 1.19 [0.95, 1.50]

Death, all-cause 13.0% (119) 9.9% (89) 3.1% [0.2%, 6.1%] 1.38 [1.03, 1.85]

- Cardiovascular 6.8% (61) 5.5% (49) 1.3% [-0.9%, 3.6%] 1.26 [0.85, 1.85]

- Definite cardiovascular 5.0% (45) 4.5% (40) 0.5% [-1.4%, 2.5%] 1.13 [0.73, 1.74]

- Undetermined cause 1.9% (16) 1.1% (9) 0.9% [-0.3%, 2.0%] 1.78 [0.78, 4.06]- Undetermined cause 1.9% (16) 1.1% (9) 0.9% [-0.3%, 2.0%] 1.78 [0.78, 4.06]

- Non-cardiovascular 6.6% (58) 4.6% (40) 2.0% [-0.2%, 4.2%] 1.47 [0.97, 2.23]

Cerebrovascular events 3.3% (29) 5.2% (46) -1.9% [-3.8%, 0.0%] 0.61 [0.38, 0.99]

- Stroke 2.9% (26) 3.7% (33) -0.8% [-2.4%, 0.9%] 0.78 [0.46, 1.31]

- Transient ischemic attack 0.3% (3) 1.6% (14) -1.3% [-2.2%, -0.4%] 0.21 [0.06, 0.74]

Myocardial infarction 10.6% (95) 9.1% (84) 11.4% [-1.3%, 4.2%] 1.14 [0.84, 1.55]

- Peri-procedural 3.9% (37) 6.1% (57) -2.1% [-4.1%, -0.1%] 0.63 [0.41, 0.96]

- Non-peri-procedural 6.8% (59) 3.5% (31) 3.2% [1.2%, 5.3%] 1.96 [1.25, 3.06]

All-cause Death, Stroke or MI

15

20

or

MI

(%)

15.1%

0-day to 30-day HR: 0.61 [95% CI: 0.42, 0.88]; P-value = 0.008

30-day to 1-year HR: 1.07 [95% CI: 0.68, 1.70]; P-value = 0.76


Treatment-time interaction: P<0.001CABG (n=957)

PCI (n=948)

Even if one extrapolates (by doubling theevent rates from 1-5 years), by year 10 allcause death, stroke or MI is 38.9% Vs31.5% favoring CABG. The NNT is 13.5.

Freedom from death, stroke of MI is 61.1%with PCI Vs 68.5% with CABG

0

5

10

De

ath

,s

tro

ke

or

MI

(%)

0 12 24 36 48 60

Months

8.0%

3.8%

9.7%

4.9% 4.1%

Number at risk:

PCIPCI

CABGCABG 957

948

889 856 827 794 579

902 854 819 776 511


929

933

1

Additional Outcomes at 5 Years

PCI (N=948) CABG (N=957) Difference [95% CI] Odds ratio [95% CI]

Death, stroke, MI or IDR 31.3% (290) 24.9% (228) 6.5% [2.4%, 10.6%] 1.39 [1.13, 1.71]

- ID-revascularization 16.9% (150) 10.0% (88) 6.9% [3.7%, 10.0%] 1.84 [1.39, 2.44]

- PCI 14.1% (125) 9.1% (80) 4.9% [1.9%, 7.9%] 1.65 [1.22, 2.22]

- CABG 4.3% (38) 0.9% (8) 3.4% [1.9%, 4.9%] 4.90 [2.27, 10.56]

All revascularization 17.2% (153) 10.5% (92) 6.7% [3.5%, 9.9%] 1.79 [1.36, 2.36]

Stent thrombosis 1.8% (16) 0% (0) - -

- Definite 1.1% (10) 0% (0) - -

- Probable 0.7% (6) 0% (0) - -

Symptomatic graft occlusion 0% (0) 6.5% (58) - -

Therapy failure* 1.1% (10) 6.5% (58) -5.4% [-7.2%, -3.6%] 0.16 [0.08, 0.32]

*Definite stent thrombosis or symptomatic graft occlusion. ID = ischemia-driven.

Conclusions

• In the EXCEL trial, treatment of patients with LMCAD

and visually-assessed low or intermediate SYNTAX

scores with CoCr-EES resulted in similar rates of the

clinically meaningful composite outcome of death, stroke

or MI at 5 yearsor MI at 5 years

• The early benefits of PCI due to reduced peri-procedural

risk were attenuated by the greater number of events

occurring during follow-up with CABG, such that at

5 years the cumulative mean time free from adverse

events was similar with both treatments

Conclusions

• PCI may thus be considered an

acceptable revascularization modality

for selected patients with LMCAD, a

decision which should be made afterdecision which should be made after

heart team discussion, taking into

account each patient’s individual risk

factors and preferences

Subgroup analyses

Distal LMBifurcationsonly

NNT = 17It takes 17open chestsurgeriesto prevent1 repeatPCIPCI

Giustino G, et al; J Am Coll Cardiol. 2018 Aug 14;72(7):754-765

Acute renal failure at 30 days in patients with or without CKDAcute renal failure at 30 days in patients with or without CKDundergoing PCI versus CABGundergoing PCI versus CABG

Chronic Kidney Disease (n = 361) No Chronic Kidney Disease (n = 1508)

pinteractionPCI

(n = 177)

CABG

(n = 184)

Hazard Ratio

(95% CI)

PCI

(n = 757)

CABG

(n = 751)

Hazard Ratio

(95% CI)

Acute renal failure† 4/177 (2.3%) 14/184 (7.6%) 0.28 (0.09-0.87) 2/757 (0.3%) 10/751 (1.3%) 0.20 (0.04-0.90) 0.71

New requirement for

dialysis

2/177 (1.1%) 10/184 (5.4%) 0.20 (0.04-0.92) 1/757 (0.1%) 4/751 (0.5%) 0.25 (0.03-2.22) 0.87

dialysis

Hemodialysis 1/177 (0.6%) 5/184 (2.7%) 0.20 (0.02-1.76) 1/757 (0.1%)* 3/751 (0.4%) 0.33 (0.03-3.18) 0.76

CVVH 1/177 (0.6%) 5/184 (2.7%) 0.20 (0.02-1.76) 1/757 (0.1%)* 1/751 (0.1%) 0.99 (0.06-15.89) 0.38

†Defined as the rise in serum creatinine >5 mg/dL or a new requirement for dialysis. *One patient in the no chronic kidney disease group had bothCVVH and hemodialysis. CVVH: Continuous veno-venous hemofiltration.

Giustino G, et al; J Am Coll Cardiol. 2018 Aug 14;72(7):754-765

Unprotected LM Stenting- Caveats

• EXCEL mostly provisional…results would likely be better with DK crush• Most UPLM bifurcations can be done radially• 6FR can be cumbersome but is certainly doable and may cause slightly less radial

artery occlusion. If you are more comfortable with 7Fr, use Slender sheath radial• Beware of longitudinal compression when treating LM lesions.• Impella can provide very helpful support for these patients. Please think about Impella

before referring UPLM patients for CABG

SCRIPPS CLINIC

before referring UPLM patients for CABG• IVUS should be set up at the start of the case…and used• Data indicates DK crush is preferred, but Coulotte and TAP are also okay…make sure

you are comfortable with the technique you use.• If left main SB FFR >.8 then provisional stenting of SB is not immoral but…• 2 stent data is better, and if you like pretty pictures, you will be happier with 2 stents

for UPLM bifurcations.

When do I send a 3VD +/orUPLM patient for CABG ?

High SYNTAX score with 3VD and/or LM disease

Intermediate SYNTAX score with 3VD

In borderline SYNTAX score patients, IDDM sways metowards CABG

DAPT intolerant patient

Total occlusions with large and important territory at risk,

SCRIPPS CLINICSCRIPPS CLINIC

Total occlusions with large and important territory at risk,not amenable to PCI or failed PCI

Several restenoses, large territory at risk

Poor PCI candidate, ie excessive proximal tortuosity,particularly if calcified with good distal targets

When do I break the above rules:

- Poor CABG candidate, ie frailty, O2 dependent, no targets

- Patient preference? Especially when SYNTAX borderline

The Heart Team and“Shared Decision Making” are overrated.

• Medicine is not a democracy

The heart team and borderline cases

• Few patients can really analyze the data and make an informeddecision

SCRIPPS CLINIC

Patients make decisions based mostly on

the confidence they get from the surgeon or interventionalist

Their fear of major surgery

The experiences of their friends and family

You Can Call Me Now…

• Bypass surgery is very hard to go through more than once

Your saphenous vein grafts will likely close down

Your native vessels will likely shrivel up

Your subsequent PCI will likely be more difficult

…Or You Can Call Me Later

But Remember:

SCRIPPS CLINIC

Your subsequent PCI will likely be more difficult

• But, PCI can be repeated as often as you like

And you can always have a bypass

Sometime in the future

Or, maybe never

Bypass the Bypass!

SCRIPPS CLINIC

pci vs cabg for multivessel and unprotected lm disease€¦ · cabg pci pvalue death 9.3% 10.2%...

Documents