“CABG vs PCI”
What’s the Latest Data?
Should we follow the SYNTAX Score?
Rob Adamson, M.D.
Surgical Director of Cardiac Transplantation
SHARP Memorial
Disclosure
Research funding within the past year:
Thoratec Corporation
Ikaria
No personal conflicts to disclose regarding this topic
Disclosure
Research funding within the past year:
Thoratec Corporation
No personal agreements to disclose, Ethicon, a subsidiary of Johnson and Johnson, the maker of Prolene
suture did not design, sponsor or fund any of the trials that I am going to discuss
Surgeon presenting at Cardiologist meeting
• 600,000 angioplasties/yr. in the US
• 70% indicated for acute MI, (JAMA 2011)
• 30% stable or mild symptoms
• 50% appropriate
• 38% uncertain
• 12% inappropriate
• Courage trial William Boden
• 2,287 patients followed for 5 year showed
• “stents weren’t any better than a cocktail of medicines to treat patients suffering from chronic but stable chest pain”
• Dr. Midei, St Joseph Medical Center, Towson Md. , charged with unprofessional conduct for performing unnecessary angioplasties
Ron Winslow and John Carreyrou WSJ 2011
Doctors Use Euphemism for $2.4 Billion in Needless Stents
• 700,000 stents per year in the U.S.
• ACC changing it’s nomenclature guidelines for stents
• Inappropriate …….. Rarely Appropriate
• Uncertain …………..May be Appropriate
• 14 billion dollars annual stent costs
• Unnecessary stents cost U.S. health care system $2.4 billion a year
Peter Waldman Oct 2013
No one wants any operation
• No patient wants ‘their chest cracked’
if the same effect can be achieved
with a less invasive approach
• Bee sting vs. Buzz Saw euphemism
Bee Sting Bee Sting BUZZ SAW
Bee Sting vs. Bee Sting Caveat
Rates of CABG Surgery, Bare Metal Stents, DES,
and Angioplasty: 2001 to 2008, U.S.
DES
BMS
Angioplasty
CABG
Epstein, A. J. et al. JAMA 2011;305:1769-1776
Management of Coronary Artery Disease
What is the role for CABG ?
• Three large cooperative randomized trials have evaluated the effects of medical and surgical management of ischemic heart disease on survival and other secondary end points. • Both randomized and observational data from these trials show increased survival following coronary artery bypass grafting (CABG) in patients with • left main coronary artery stenosis, triple‐vessel disease, double‐
vessel disease, • left ventricular (LV) functional impairment, or LV aneurysm. 1. Murphy ML, Hultgren HN, Detre K, et al: Treatment of chronic stable angina: a preliminary report of survival data of the randomized Veterans Administration Cooperative Study. N Engl J Med 297621, 1977
2. European Coronary Surgery Study Group: Coronary‐artery bypass surgery in stable angina pectoris: survival at two years. Lancet 12389, 1979
3. CASS Principal Investigators and Their Associates: Coronary Artery Surgery Study (CASS): a randomized trial of coronary artery bypass surgery; survival data. Circulation 68:939, 1983
• Three large cooperative randomized trials have evaluated the effects of medical and surgical management of ischemic heart disease on survival and other secondary end points. • Both randomized and observational data from these trials show increased survival following coronary artery bypass grafting (CABG) in patients with • left main coronary artery stenosis, triple‐vessel disease, double‐
vessel disease, • left ventricular (LV) functional impairment, or LV aneurysm. 1. Murphy ML, Hultgren HN, Detre K, et al: Treatment of chronic stable angina: a preliminary report of survival data of the randomized Veterans Administration Cooperative Study. N Engl J Med 297621, 1977
2. European Coronary Surgery Study Group: Coronary‐artery bypass surgery in stable angina pectoris: survival at two years. Lancet 12389, 1979
3. CASS Principal Investigators and Their Associates: Coronary Artery Surgery Study (CASS): a randomized trial of coronary artery bypass surgery; survival data. Circulation 68:939, 1983
Early and five‐year results for coronary artery bypass grafting. A benchmark for percutaneous transluminal coronary angioplasty.
Daily, PO
• PTCA patients are selected and have lower risk profile than CABG therefore compare outcomes to pre‐PTCA era (1976‐1980)
• 500 consecutive CABG patients 33‐79 yrs., 20% female, 60% 3 vessel
• Complete revasularization 99.8%, 3.2 grafts/pt.
• Hospital mortality 0.2% (1/500)
• Low output +/‐ IABP 1% (5/500)
• Perioperative MI 2.2% (11/500)
• 99.4% follow‐up at 5 years or longer (497/500)
• 92.7 vs. 89.8%% survival (cardiac vs all cause) at 5 years • EF < 50% only univariate predictor of death
• Freedom from reintervention 97.7% at 5 years
J Thor Cardiovasc Surg 1989 Jan;97(1):67‐77
Early and five‐year results for coronary artery bypass grafting. A benchmark for percutaneous transluminal coronary angioplasty.
Daily, PO
Benchmarks
Hospital Mortality <1%
Perioperative MI rate <3%
5 year survival > 90%
5 year freedom from reintervention >95%
J Cardiovasc Thor Surg 1989 Jan;97(1):67‐77
Effect of coronary artery bypass graft surgery on survival: overview of 10‐year results from randomized trials by the Coronary Artery Bypass Graft Surgery Trialists Collaboration.
• 7 randomized trials comparing initial CABG vs. medical management for stable CAD (not severe ↑ Sx or MI)
• 1324 patients CABG vs. 1325 medical management
• Crossover rates from medical management to CABG • 25% at 5 yrs., 33% at 7 yrs., 41% at 10 yrs.
• CABG group lower mortality at 5, 7 and 10 years (p< .003)
• Greatest CABG benefit in left main and 3 vessel disease
• Strategy of initial CABG associated with lower mortality than medical management +/‐ delayed CABG
Yusuf et al Lancet. 1994 Aug 27;344(8922):563‐70
CABG vs Stents
In
“Multi‐Vessel Disease”
15 Trials CABG vs PCI for MVD
Trial
#pt
Stent
Pop
1or 2VD
EF>50%
LM
PLAD
DM
LIMA
RITA
1011
-
4%
88
-
0
-
6
74
ERACI
127
-
9%
55
100
0
-
11
75
LAUSANNE134
-
3%
100
-
0
100
12
100
GABI
359
-
4%
82
-
0
-
10
37
EAST
392
-
4%
60
100
0
70
25
-
CABRI
1054
-
3%
60
100
0
-
12
75
MASS
142
-
69%
-
100
0
100
21
100
BARI
1829
-
12%
59
100
0
36
24
80
TOULOSE 152
-
3%
71
-
0
-
14
58
SIMA
121
-
-
-
100
0
100
11
100
ERACI II
450
+
2%
44
-
0
-
17
88
AWESOME 454
+
-
55
-
0
-
-
70
MASS II
408
+
2%
59
0
ARTS
1205
+
?5%
68
100
0
-
19
93
SOS
988
+
?5%
62
100
0
45
14
81
SUM
8,826
5%
65%
100%
0%
41%
16%
79%
<10%
70%
>20%
>90%
25%
>90%
CABG (current practice)
15 Trials CABG vs PCI for MVD
Trial
#pt
Stent
Pop
1or 2VD
EF>50%
LM
PLAD
DM
LIMA
RITA
1011
-
4%
88
-
0
-
6
74
ERACI
127
-
9%
55
100
0
-
11
75
LAUSANNE134
-
3%
100
-
0
100
12
100
GABI
359
-
4%
82
-
0
-
10
37
EAST
392
-
4%
60
100
0
70
25
-
CABRI
1054
-
3%
60
100
0
-
12
75
MASS
142
-
69%
-
100
0
100
21
100
BARI
1829
-
12%
59
100
0
36
24
80
TOULOSE 152
-
3%
71
-
0
-
14
58
SIMA
121
-
-
-
100
0
100
11
100
ERACI II
450
+
2%
44
-
0
-
17
88
AWESOME 454
+
-
55
-
0
-
-
70
MASS II
408
+
2%
59
0
ARTS
1205
+
?5%
68
100
0
-
19
93
SOS
988
+
?5%
62
100
0
45
14
81
SUM
8,826
5%
65%
100%
0%
41%
16%
79%
<10%
70%
>20%
>90%
25%
>90%
CABG (current practice)
PCI vs. CABG
Meta‐analysis
• University of York
o (National Institute for Health Research)
• ARTS II, CARDia, ERACI III, SYNTAX
• 15 studies (21,228 participants)
o 4 prospective studies (3,895)
o 2 retrospective studies (2,233)
o 11 retrospective, observational (17,333)
• Authors’ Conclusions
“Drug-eluting stents were safe in patients when
compared to CABG, but were associated with a
significantly high risk of target vessel revascularization”
Am et al, EuroIntervention 2010; 6(2): 269‐76
One Size Doesn’t Fit All
Which Fits Best
• CABG
• PCI
• Medical Management
Evidence from a
“Real World”
Randomized Trial
The SYNTAX Trial:
The Synergy between Percutaneous
Coronary Intervention with TAXUS and
Cardiac Surgery
Serruys, et al for SYNTAX: NEJM 360:961‐72, 2009
SYNTAX Eligible Patients
De novo disease
Limited exclusion criteria
• Previous interventions
• Acute MI with CPK >2x
• Concomitant cardiac surgery
Left main disease
(isolated, +1, +2 or +3 vessels)
3 vessel disease
(revasc all 3 vascular territories)
Serruys, et al for SYNTAX: NEJM 360:961‐72, 2009
Dominance
No. &
location
of lesion
Left
main
Calcification
SYNTAX
SCORE
Thrombus
3 vessel
Bifurcation
CTO
Tortuosity
Serruys, et al for SYNTAX: NEJM 360:961‐72, 2009
• 1,800 patients with L Main or 3 vessel CAD
Randomized to TAXUS Express² (PES) vs CABG
Among DM, 3 year adverse event rates
• MACCE 22% CABG vs 37% PES (p=0.002)
• Revascularization rate 12.9% CABG vs 28% PES (p<0.001)
• SYNTAX scores >33 MACCE rate lower for CABG 18.5% vs 45.9% (p<0.001)
• Conclusion:
Diabetic, 3 vessel/L Main MACCE higher with PES
CABG should be revascularization of choice
Mack et al Ann Thorac Surg 2011;92:2140‐6
 Patients not randomized in SYNTAX because deemed inappropriate for one therapy
 3,075 patients in SYNTAX
 198 (6.4%) PCI only
 1,077 (35%) CABG only
 Reasons
 PCI ‐ too high risk for surgery
 CABG ‐ complex coronary anatomy
Head, et al JACC: Cardiovasc Interventions Vol 5 No 6. June 2012:618‐25
SYNTAX Risk Profile
Head, et al JACC: Cardiovasc Interventions Vol 5 No 6. June 2012:618‐25
SYNTAX Non‐Randomized Cohort
Head, et al JACC: Cardiovasc Interventions Vol 5 No 6. June 2012:618‐25
SYNTAX Non‐Randomized Cohort
Head, et al JACC: Cardiovasc Interventions Vol 5 No 6. June 2012:618‐25
SYNTAX Non‐Randomized Cohort
Head, et al JACC: Cardiovasc Interventions Vol 5 No 6. June 2012:618‐25
 During SYNTAX trial agreed upon which vessels to revascularize
 Incomplete revascularization 43.4% PCI vs 36.8% CABG
 Complete revascularization by PCI had  lower MACCE, composite safety endpoints and revascularization rates  Incomplete PCI revascularization
 Higher MACCE
 Incomplete PCI revascularization associated with
 Hyperlipidemia, total occlusion, number of vessels
 Conclusion
 Incomplete revascularization is associated with adverse events in PCI patients but not in CABG patients
Head et al European Journal of Cardio‐Thoracic Surgery 41 (2012) 535‐541







1800 pts randomized
85 centers in USA and Europe
MACCE ‐ 26.9% CABG vs. 37.3% PCI (p<0.0001)
MI – 3.8% CABG vs. 9.7% PCI (p<0.0001
Repeat revascularisation 13.7% CABG vs. 25.9% PCI (p<0.0001) All cause death and CVA not different
Patients with intermediate or high SYNTAX scores MACCE was significantly increased with PCI
Mohr, et al Lancet Vol 381, Feb 23 2013
CABG should remain the standard of care for patients with complex lesions (high or intermediate SYNTAX scores). For patients with less complex disease (low SYNTAX scores) or left main coronary disease (low or intermediate SYNTAX scores), PCI is an acceptable alternative.
Mohr, et al Lancet Vol 381, Feb 23 2013
CABG should remain the standard of care for patients with complex lesions (high or intermediate SYNTAX scores). For patients with less complex disease (low SYNTAX scores) or left main coronary disease (low or intermediate SYNTAX scores), PCI is an acceptable alternative.
All patients with complex multivessel coronary artery disease should be reviewed and discussed by both a cardiac surgeon and interventional cardiologist to reach consensus on optimal treatment
Mohr, et al Lancet Vol 381, Feb 23 2013
What is the advantage
of CABG?
Advantages of an IMA to LAD Graft
vs. SVG:
We Know
• Superior early
and late patency
• Superior late survival
• Superior freedom from late
ischemic events
and
• Superior early results
(30-day mortality)
Advantages of an IMA to LAD Graft
LITA to LAD Superior Late Survival
100
Survival (%)
ITA – 82.6%
90
SV – 71.0%
80
P<0.0001
70
Internal mammary artery n=1,249
60
Saphenous vein graft n=2,116
0
2
4
6
8
10
Years
Loop: NEJM, 1986
CTA of Multiple Arterial Grafts
LIMA
RIMA
Anastomoses
RIMA
LIMA
Anastomoses
Survival Advantage
Bilateral ITA
100
80
Survival
(%)
60
Bilateral ITA
Single ITA
40
P<0.001
20
0
Number at risk 0
1,987
8,059
2
4
6
8
10
1,941
7,680
1,894
7,409
1,799
6,201
1,542
5,249
1,172
4,466
12
14
15
304
67
55
2,714 2,283 1,995
Lytle, JTCVS, 1999
Survival Advantage
Bilateral ITA
100
80
60
Survival
(%)
40
Bilateral ITA
Single ITA
P<0.001
Bilateral ITA reop free
20
Single ITA reop free
P<0.001
0
Number at risk 0
1,987
8,059
2
4
6
8
10
1,941
7,680
1,894
7,409
1,799
6,201
1,542
5,249
1,172
4,466
12
14
15
304
67
55
2,714 2,283 1,995
Lytle, JTCVS, 1999
CABG treats both culprit lesion of any
complexity and future culprit lesions
• PCI deals with “suitable” localized prox disease, no prophylactic benefit new disease
• LIMA-LAD benefit:
- ? Nitric oxide transplant
- ? Endothelial progenitor cells
- ? Limit endothelial dysfunction
Circulation. 2012; 126: 1023‐1030 Methods
• 8,622 consecutive patients at Mayo
Clinic
• primary isolated CABG for multivessel
disease from 1993 to 2009.
• Patients were stratified into 2 groups.
Late Survival
MultArt vs LIMA/SV
94%
Survival (%)
100
84%
80
71%
85%
60
61%
40
36%
20
P<0.001
0
0
5
10
15
Follow‐up time (year)
MultArt
1,177
851
362
83
LIMA/SV
7,281
4,898
2,493
578
Conclusions
• For isolated CABG for MVD at least one more
arterial graft in addition to LIMA to the LAD
conferred a survival advantage throughout a 15year follow up period .
• The survival advantage increased with time,
suggesting that the initial selection of the conduits
was a more important factor in survival than
problems appearing long after surgery, such as the
progression of coronary disease.
Improved Long‐Term Survival With Multiple Arterial Grafting Compared With Percutaneous Coronary Intervention or Conventional Coronary Artery Bypass Grafting
Analysis of 12,615 Patients with Multivessel Disease
Chaim Locker, MD; Hartzell V. Schaff, MD; Richard C. Daly, MD;
Robert L. Frye, MD; Malcolm R. Bell, MD; Joseph A. Dearani, MD;
Zhuo Li; Ryan J. Lennon; Lyle D. Joyce, MD, PhD; Kevin L. Greason,
MD; John M. Stulak, MD; Alberto Pochettino, MD; Amir Lerman, MD
Mayo Clinic, Rochester MN
2013
Overall Survival
MultArt vs LIMA/SV vs PCI
95%
Survival (%)
83%
77%
81%
62%
57%
MultArt
LIMA/SV
PCI
65%
31%
23%
P<0.001
Follow‐up time (year)
MultArt
LIMA/SV
PCI
955
5,712
5,948
676
3,871
3,898
346
2,017
1,861
83
572
381
Overall Survival
MultArt vs LIMA/SV vs PCI Sub Groups
Survival (%)
100
80
60
95%
83%
79%
76%
72%
MultArt
LIMA/SV
BMS PCI
BA PCI
DES PCI
40
20
0
0
1
2
3
4
5
P<0.001
6
7
8
Follow‐up time (year)
MultArt
LIMA/SV
BMS PCI
BA PCI
DES PCI
955
5,712
3,242
1,020
1,686
821
4,872
2,835
879
1,438
727
4,214
2,567
776
982
548
3,358
2,245
665
529
433
2,659
1,877
574
178
Conclusion
• Long history of CABG versus medical management
• CABG proven effective….
o even at the VA hospital
• No evidence of survival advantage of any
revascularization of single vessel disease (proximal
LAD)
• SYNTAX favored CABG for 3 vessel, high SYNTAX
score patients
o The SYNTAX trial sponsored by Boston Scientific
Conclusion of the Conclusion
• PCI and CABG are not mutually exclusive therapies
• Medical therapy may be effective
• Appropriate patient selection and cooperative
consensus is in the patient’s best interest
Rebuttal
We all make choices ……
CABG
PCI (stents)
=
Spinach or Candy
ASPCA complaints
PCI Horse
Conclusion of Conclusions
• CABG proven effective
o major operative procedure
• PCI evolving technology with caveats
o Self referral
o Corporate sponsored trials
o Plavix not always benign especially in the elderly
• These therapies are adjunctive
o PCI and CABG are not mutually exclusive therapies
o Medical therapy may be effective
• Appropriate patient selection and cooperative
consensus is in the patient’s best interest