Comprehensive Meta-Analysis
of DES vs. BMS Randomized
Trials and Registries
Ajay J. Kirtane, M.D., S.M.
Gregg W. Stone, M.D.
Columbia University Medical Center
The Cardiovascular Research Foundation
Conflict of Interest Disclosure
• Ajay J. Kirtane

Past honorarium from Boston Scientific
Corporation (modest)

Consultant/Speaker: Medtronic Vascular,
Abbott Vascular (modest)
• Gregg W. Stone

Research grants from Boston Scientific
and Abbott Vascular
Background: DES vs. BMS RCTs
• In most individual RCTs, DES have reduced the
rates of TLR and TVR compared to BMS, with no
significant differences in death or MI
• However, individual RCTs are underpowered to
assess low frequency endpoints
• RCTs, particularly the pivotal RCTs leading to
regulatory approval, have been criticized for not
reflecting “real-world” DES use
• RCT outcomes may vary based upon differences
in enrollment criteria (e.g. “on-label” vs. “offlabel”), the amount of routine angiographic FU,
and with the duration of clinical FU
Background: DES vs. BMS Registries
• In order to address issues of both sample size as
well as generalizability to the “real-world”,
numerous observational and registry comparisons
of DES vs. BMS have been undertaken
• The outcomes from these studies have varied
• While more generalizable than RCTs, the DES vs.
BMS registries are heterogeneous, with differences
in design and analysis methodology (e.g. adjusted
vs. unadjusted, type of adjustment)
• Registry outcomes may also vary based upon the
types of patients enrolled (e.g. all comers vs. just
ACS or high risk), and the duration of clinical FU
Persistent Questions: DES vs. BMS
• While some of the alarm generated after ESC 2006
has been mitigated by analyses of patient-level
data from the “on-label” RCTs*, there remains
concern regarding DES outcomes in “off-label”
patients and lesions, and with uncontrolled use

Are DES safe in higher risk off-label pts and in
the unregulated environment of real-world use?

Are the benefits of DES in reducing TVR as
robust in the real-world as in the RCTs, given
the impact of routine angio FU and the
oculostenotic reflex in many RCTs?
*Stone et al, Kastrati et al, Spaulding et al, Mauri et al
N Engl J Med 2007; 356(10).
Methods: Goals and Objectives (1)
• We therefore sought to perform a
systematic review and meta-analysis of
DES vs. BMS studies

To derive summary estimates of all-cause
mortality, MI, and TVR in studies with ≥1
year of follow-up

To specifically assess differences
between RCT and registry safety and
effiacy with regard to these endpoints
Methods: Goals and Objectives (2)
• Randomized Trials
To assess differences between RCTs according
to “on-label” vs. “off-label” use, duration of FU,
and baseline risk
• Registries / Observational Analyses
 To assess differences in the estimates derived
from registries using unadjusted and adjusted
analyses (and according to the types of
adjustment)
 To assess differences between registries
according to duration of FU, and baseline risk
• To assess differences in effect size estimates
between the RCTs and registries for each endpoint

Methods: Inclusion Criteria
• English language RCTs or registries which
reported a direct comparison of DES
(commercialized formulations of SES and PES
only) vs. BMS.
• Criteria for each study:

≥100 patients total

Mortality reported (± MI and/or TVR)

≥1 year of cumulative follow-up reported, with
the outcome assessed at the same time point
in both comparator arms
Methods: Exclusion Criteria
• “DES era” vs. pure “BMS era” studies in
which the DES era group did not exclude
BMS pts

Excepting intent-to-treat RCTs
• Study used a control group from another
study already in the meta-analysis
• Study was itself a meta-analysis (although
data abstracted from individual studies in
a published meta-analysis could be used)
Methods: Search Strategy (1)
• 2 PUBMED searches

Focused: (eluting stent OR DES OR drug-eluting
stent) AND (bare OR uncoated OR standard OR
BMS) AND (("2002"[PDat] : "2008"[PDat]) AND
(Humans[Mesh]) AND (English[lang])) AND
(coronary) NOT (cost-effectiveness) NOT review[pt]
NOT case reports[pt] NOT editorial[pt] NOT
comment[pt]

Broad: stent AND bare AND (eluting OR sirolimus
OR paclitaxel)
• Cochrane database
• Eurointervention journal
Methods: Search Strategy (2)
• Abstracts/presentations from 2007 meetings:

ACC SCAI/I2 Summit

ESC

TCT

AHA
• Data requested from study PI’s for large
registries

For most updated data

Where not publicly available or to clarify
methodology
Methods: Analysis
• Pre-specified separate analysis of RCTs and
Registries performed given clinical heterogeneity
• RCTs: Direct randomization to DES vs. BMS
• Registries: Non-rand comparison of DES vs. BMS
(including non-rand comparisons within a RCT)
• All analyses cumulative
• No landmarks
• Single time point estimate for each study assuming
constant hazard of DES vs. BMS through study period
(hence use of HR or RR as the estimate)
• Higher quality estimate picked for primary analyses
(adjusted > unadjusted)
Methods: Statistical Analysis
• All analyses were performed at The Cardiovascular
Research Foundation/Columbia University
• Models (both reported):
• Fixed effects (Inverse-Variance weighted)
• Random effects (DerSimonian and Laird)*
• Fixed effects model was considered the primary
model if significant heterogeneity was not present;
otherwise random effects was considered primary
• Formal heterogeneity testing was performed using
the I2 statistic; heterogeneity was defined as I2 ≥ 25%
*Weights displayed in figures are based upon the primary model used
Selected Excluded Studies
• GRACE

Non-landmark data not available (PI contacted
as well)

Unequal follow-up in comparator arms (data
not presented at fixed timepoint)
• RRISC

Less than 100 patients
• Medicare Data

Comparison of pre-DES era with post-DES
rather than DES vs. BMS
All-Cause Mortality: All RCTs
8,867 patients, 21 trials
%
Weight
ES (95%
CI) CI) Weight
(I-V) (%)
Estimate
(95%
Study ID
SCORPIUS
SESAMI
Typhoon
Passion
BASKET (SES only)
STRATEGY
SES-SMART
Seville
HAAMU-STENT
MISSION!
PRISON II
Pache et al
Ortolani et al
DIABETES
RAVEL
SIRIUS
C-SIRIUS
E-SIRIUS
TAXUS II
TAXUS IV
TAXUS V
I-V Overall Effects
(I-squared = 0.0%, p = 0.918)
Random
D+L Overall
*Fixed
Effects (I2=0.0%)
1.28 (0.35, 4.61)
0.43 (0.11, 1.63)
1.01 (0.38, 2.65)
0.70 (0.36, 1.36)
0.82 (0.37, 1.84)
0.84 (0.36, 1.96)
0.21 (0.02, 1.71)
1.35 (0.23, 7.78)
2.00 (0.63, 6.38)
0.48 (0.09, 2.59)
0.50 (0.09, 2.67)
1.40 (0.45, 4.35)
2.00 (0.19, 21.38)
1.44 (0.48, 4.33)
1.75 (0.73, 4.16)
1.02 (0.67, 1.54)
0.68 (0.11, 4.04)
1.08 (0.25, 2.24)
1.61 (0.57, 4.53)
0.89 (0.63, 1.25)
0.97 (0.57, 1.65)
0.97 (0.81,
1.15)
0.97
(0.81,1.15)
0.97 (0.81, 1.15)
0.97 (0.81,1.15), p=0.72
Favors DES
.1
1.86
1.70
3.27
6.99
4.80
4.30
0.62
1.00
2.30
1.09
1.07
2.40
0.55
2.55
4.08
17.82
0.95
2.57
2.87
26.29
10.92
100.00
Favors BMS
1
10
Ajay J. Kirtane and Gregg W. Stone, 2008
Mean f/u 2.9 yrs
All-Cause Mortality: RCTs (On-Label)
%
4,818 patients, 10 trials
Study ID
Weight
ES (95% CI)
(I-V)
Estimate (95% CI) Weight (%)
SCORPIUS
1.28 (0.35, 4.61)
2.84
Pache et al
1.40 (0.45, 4.35)
3.67
Ortolani et al
2.00 (0.19, 21.38) 0.85
RAVEL
1.75 (0.73, 4.16)
6.23
SIRIUS
1.02 (0.67, 1.54)
27.25
C-SIRIUS
0.68 (0.11, 4.04)
1.45
E-SIRIUS
1.08 (0.25, 2.24)
3.93
TAXUS II
1.61 (0.57, 4.53)
4.39
TAXUS IV
0.89 (0.63, 1.25)
40.20
TAXUS V - Simple
1.09 (0.53, 2.22)
9.20
I-V Overall
(I-squared = 0.0%, p = 0.927)
Random
Effects
1.05 (0.84,
1.30)
(0.84,1.30)
100.00
*Fixed
Effects (I2=0.0%)
D+L Overall
(0.84,1.30),
1.05 (0.84,
1.30) p=0.69
Favors DES
.1
Favors BMS
1
Ajay J. Kirtane and Gregg W. Stone, 2008
10
Mean f/u 4.0 yrs
All-Cause Mortality: RCT’s (Off-Label)
%
4,049 patients, 12 trials
Study ID
Weight
ES (95% CI)
(I-V)
Estimate (95% CI) Weight (%)
SESAMI
0.43 (0.11, 1.63) 4.90
Typhoon
1.01 (0.38, 2.65) 9.44
Passion
0.70 (0.36, 1.36) 20.16
BASKET (SES only)
0.82 (0.37, 1.84) 13.84
STRATEGY
0.84 (0.36, 1.96) 12.40
SES-SMART
0.21 (0.02, 1.71) 1.80
Seville
1.35 (0.23, 7.78) 2.87
HAAMU-STENT
2.00 (0.63, 6.38) 6.64
MISSION!
0.48 (0.09, 2.59) 3.16
PRISON II
0.50 (0.09, 2.67) 3.10
DIABETES
1.44 (0.48, 4.33) 7.36
TAXUS V - complex
0.84 (0.38, 1.84) 14.32
Random
I-V
OverallEffects
(I-squared = 0.0%, p = 0.798)
*Fixed
Effects (I2=0.0%)
D+L
Overall
0.84 (0.62,1.13)
(0.62, 1.13) 100.00
0.84 (0.62,1.13),
(0.62, 1.13) p=0.24
Favors DES
.1
Favors BMS
1
Ajay J. Kirtane and Gregg W. Stone, 2008
10
Mean f/u 1.5 yrs
All-Cause Mortality: All Registries
161,232 patients, 28 registries
Study ID
ES (95% CI)
Estimate
(95% CI)
NHLBI (off label, adjusted)
NHLBI (on label, adjusted)
Germany Metabolic Syndrome
Ontario (matched)
Mayo FFR Substudy
Italian Diabetic Multivessel (adjusted)
McMaster STEMI (adjusted)
Rotterdam Off-Label
Washington Hosp Center (matched)
Asan Korea (adjusted)
SCAAR (adjusted)
Wake Forest (adjusted)
Western Denmark (adjusted)
NY State (adjusted, unmatched)
MIDAS (adjusted)
Massachusetts (matched)
STENT (adjusted)
Liverpool (matched)
GHOST (adjusted)
DEScover (unadjusted)
Cedars Acute MI
REAL (adjusted)
Melbourne
Multicenter SVG (adjusted)
ACUITY (from RCT)
RESTEM
ARTS II (from RCT)
ERACI III (from RCT)
D+L Overall (I-squared =270.1%, p = 0.000)
*Random
Effects (I =70.1%)
I-V Overall
%
Weight
(D+L)
Weight
(%)
0.94 (0.64, 1.38)
3.40
1.47 (0.87, 2.48)
2.31
1.47 (0.65, 3.35)
1.15
0.71 (0.59, 0.84)
5.98
1.00 (0.21, 4.75)
0.36
1.22 (0.36, 4.10)
0.57
0.17 (0.03, 0.97)
0.29
0.98 (0.85, 1.13)
6.44
1.16 (0.78, 1.75)
3.21
0.60 (0.46, 0.79)
4.70
1.03 (0.94, 1.14)
6.98
0.72 (0.55, 0.95)
4.66
1.00 (0.86, 1.17)
6.29
0.84 (0.72, 0.97)
6.35
0.66 (0.59, 0.74)
6.80
0.79 (0.71, 0.89)
6.80
0.69 (0.55, 0.87)
5.25
0.45 (0.24, 0.84)
1.78
0.55 (0.36, 0.83)
3.09
0.53 (0.35, 0.80)
3.13
0.82 (0.37, 1.83)
1.20
0.83 (0.70, 0.98)
6.10
0.67 (0.23, 1.94)
0.73
1.33 (0.47, 3.76)
0.76
0.63 (0.49, 0.82)
4.87
0.73 (0.51, 1.05)
3.63
0.74 (0.41, 1.35)
1.92
1.18 (0.54, 2.58)
1.25
0.80 (0.72, 0.88)
100.00
0.80 (0.72,0.88), p<0.001
0.83 (0.79, 0.86)
Fixed Effects
0.83 (0.79,0.86)
NOTE: Weights are from random effects analysis
.1
Favors DES
1
Favors BMS
Ajay J. Kirtane and Gregg W. Stone, 2008
10
Mean f/u 2.5 yrs
All-Cause Mortality: Unadjusted Registries
122,989 patients, 22 registries
%
Weight
Study ID
ES (95% CI)
(D+L)
(%)
Estimate
(95% CI) Weight
NHLBI (off label, unadjusted)
0.58 (0.42, 0.80)
4.86
NHLBI (on label, unadjusted)
1.04 (0.68, 1.58)
3.70
Germany Metabolic Syndrome
1.47 (0.65, 3.35)
1.43
Mayo FFR Substudy
1.00 (0.21, 4.75)
0.45
McMaster STEMI (unadjusted)
0.27 (0.06, 1.24)
0.47
Rotterdam Off-Label
0.98 (0.85, 1.13)
7.53
Asan Korea (unadjusted)
0.71 (0.58, 0.87)
6.63
Wake Forest (unadjusted)
0.71 (0.54, 0.92)
5.65
Western Denmark (unadjusted)
0.81 (0.70, 0.94)
7.46
NY State (unadjusted, unmatched)
0.79 (0.69, 0.92)
7.51
MIDAS (unadjusted)
0.67 (0.60, 0.75)
7.94
Massachusetts (unadjusted)
0.54 (0.49, 0.59)
8.16
STENT (unadjusted)
0.52 (0.43, 0.62)
6.93
Liverpool (unadjusted)
0.49 (0.30, 0.80)
3.08
DEScover (unadjusted)
0.53 (0.35, 0.80)
3.79
Cedars Acute MI
0.82 (0.37, 1.83)
1.49
REAL (unadjusted)
0.74 (0.66, 0.82)
7.98
Melbourne
0.67 (0.23, 1.94)
0.90
ACUITY (from RCT)
0.63 (0.49, 0.82)
5.78
RESTEM
0.73 (0.51, 1.05)
4.37
ARTS II (from RCT)
0.74 (0.41, 1.35)
2.35
ERACI III (from RCT)
1.18 (0.54, 2.58)
1.54
2=75.3%)
D+L Overall (I-squared
p = 0.000)
*Random
Effects =(I75.3%,
0.70 (0.63,
0.78)
100.00
0.70
(0.63,0.78),
p<0.001
I-V Overall
Fixed
Effects
0.69 (0.66, 0.72)
0.69
(0.66,0.72)
NOTE: Weights are from random effects analysis
.1
Favors DES
1
Favors BMS
Ajay J. Kirtane and Gregg W. Stone, 2008
10
Mean f/u 2.1 yrs
All-Cause Mortality: Adjusted Registries
%
134,534 patients, 18 registries
Study ID
Weight
ES (95% CI)
(D+L)
Estimate (95% CI) Weight (%)
NHLBI (off label, adjusted)
0.94 (0.64, 1.38)
4.59
NHLBI (on label, adjusted)
1.47 (0.87, 2.48)
3.15
Ontario (matched)
0.71 (0.59, 0.84)
7.90
Italian Diabetic Multivessel (adjusted)
1.22 (0.36, 4.10)
0.79
McMaster STEMI (adjusted)
0.17 (0.03, 0.97)
0.40
Washington Hosp Center (matched)
1.16 (0.78, 1.75)
4.34
Asan Korea (adjusted)
0.60 (0.46, 0.79)
6.27
SCAAR (adjusted)
1.03 (0.94, 1.14)
9.14
Wake Forest (adjusted)
0.72 (0.55, 0.95)
6.22
Western Denmark (adjusted)
1.00 (0.86, 1.17)
8.29
NY State (adjusted, unmatched)
0.84 (0.72, 0.97)
8.37
MIDAS (adjusted)
0.66 (0.59, 0.74)
8.92
Massachusetts (matched)
0.79 (0.71, 0.89)
8.92
STENT (adjusted)
0.69 (0.55, 0.87)
6.98
Liverpool (matched)
0.45 (0.24, 0.84)
2.44
GHOST (adjusted)
0.55 (0.36, 0.83)
4.18
REAL (adjusted)
0.83 (0.70, 0.98)
8.05
Multicenter SVG (adjusted)
1.33 (0.47, 3.76)
1.05
D+L OverallEffects
(I-squared
= 76.6%,
*Random
(I2=76.6%)
p = 0.000)
0.80 (0.72,
0.90) p<0.001
100.00
0.80
(0.72,0.90),
0.82
(0.79,0.86)
0.82 (0.79,
0.86)
Fixed
Effects
I-V Overall
NOTE: Weights are from random effects analysis
.1
Favors DES
1
Favors BMS
Ajay J. Kirtane and Gregg W. Stone, 2008
10
Mean f/u 2.7 yrs
All-Cause Mortality: Registries
Begg’s Funnel Plot
p=0.92
log(Hazard Ratio)
2
1
0
-1
-2
0
.5
Standard Error of log(Hazard Ratio)
Ajay J. Kirtane and Gregg W. Stone, 2008
1
MI: All RCTs
%
Weight
ES (95% CI)
(I-V)
Estimate (95% CI) Weight (%)
8,850 patients, 20 trials
StudyIDID
Study
SCORPIUS
SCORPIUS
SESAMI
SESAMI
Typhoon
Typhoon
Passion
Passion
BASKET(All)
(All)
BASKET
STRATEGY
STRATEGY
SES-SMART
SES-SMART
HAAMU-STENT
HAAMU-STENT
MISSION!
MISSION!
PRISONII II
PRISON
Ortolanietetalal
Ortolani
DIABETES
DIABETES
SCANDSTENT
SCANDSTENT
RAVEL
RAVEL
SIRIUS
SIRIUS
C-SIRIUS
C-SIRIUS
E-SIRIUS
E-SIRIUS
TAXUSII II
TAXUS
TAXUSIVIV
TAXUS
TAXUSVV
TAXUS
I-V OverallEffects
D+L
(I-squared = 3.0%, p = 0.420)
Random
D+L
Overall
I-V
Overall
*Fixed
Effects (I2=3.0%)
.1.1
Favors DES
0.82 (0.23, 2.95) 2.02
1.00 (0.20, 4.88) 1.29
0.80 (0.22, 2.97) 1.94
0.83 (0.26, 2.69) 2.40
1.15 (0.64, 2.08) 9.45
0.82 (0.31, 2.40) 3.13
0.16 (0.04, 0.67) 1.65
0.25 (0.03, 2.19) 0.71
0.62 (0.28, 1.39) 5.11
0.83 (0.26, 2.64) 2.44
1.50 (0.26, 8.61) 1.07
0.60 (0.20, 1.50) 3.23
0.33 (0.09, 1.18) 1.98
1.24 (0.49, 3.14) 3.80
0.96 (0.59, 1.55) 14.07
0.59 (0.14, 2.47) 1.59
1.94 (0.93, 4.02) 6.13
0.63 (0.23, 1.72) 3.24
0.99 (0.66, 1.48) 20.13
1.27 (0.79, 2.04) 14.59
0.94 (0.79,
1.13) 100.00
0.94
(0.78,1.13)
0.94 (0.78,
1.13)
0.94
(0.79,1.13),
p=0.54
11
Favors BMS
10
10
Ajay J. Kirtane and Gregg W. Stone, 2008
Mean f/u 2.9 yrs
MI: RCTs (On Label)
%
4,318 patients, 9 trials
Study ID
Weight
ES (95% CI)
(I-V)
Estimate (95% CI) Weight (%)
SCORPIUS
0.82 (0.23, 2.95)
3.33
Ortolani et al
1.50 (0.26, 8.61)
1.77
RAVEL
1.24 (0.49, 3.14)
6.29
SIRIUS
0.96 (0.59, 1.55)
23.26
C-SIRIUS
0.59 (0.14, 2.47)
2.63
E-SIRIUS
1.94 (0.93, 4.02)
10.13
TAXUS II
0.63 (0.23, 1.72)
5.36
TAXUS IV
0.99 (0.66, 1.48)
33.28
TAXUS V - Simple
0.98 (0.52, 1.81)
13.95
Random
I-V OverallEffects
(I-squared = 0.0%, p = 0.761)
*Fixed
Effects (I2=0.0%)
D+L Overall
1.03(0.81,1.30)
(0.81, 1.30)
1.03
100.00
1.03
1.03(0.81,1.30),
(0.81, 1.30) p=0.82
Favors DES
.1
Favors BMS
1
Ajay J. Kirtane and Gregg W. Stone, 2008
10
Mean f/u 4.4 yrs
MI: RCT’s (Off Label)
Study
Study
IDID
%%
4,532 patients, 12 trialsESES(95%
(95%
CI)
CI)
Weight
Weight
(I-V)
(I-V)
Estimate (95% CI) Weight (%)
SESAMI
SESAMI
1.00
1.00
(0.20,
(0.20,
4.88)
4.88)3.30
3.30
Typhoon
Typhoon
0.80
0.80
(0.22,
(0.22,
2.97)
2.97)4.97
4.97
Passion
Passion
0.83
0.83
(0.26,
(0.26,
2.69)
2.69)6.16
6.16
BASKET
BASKET
(All)
(All)
1.15
1.15
(0.64,
(0.64,
2.08)
2.08)24.22
24.22
STRATEGY
STRATEGY
0.82
0.82
(0.31,
(0.31,
2.40)
2.40)8.03
8.03
SES-SMART
SES-SMART
0.16
0.16
(0.04,
(0.04,
0.67)
0.67)4.24
4.24
HAAMU-STENT
HAAMU-STENT
0.25
0.25
(0.03,
(0.03,
2.19)
2.19)1.83
1.83
MISSION!
MISSION!
0.62
0.62
(0.28,
(0.28,
1.39)
1.39)13.11
13.11
PRISON
PRISON
II II
0.83
0.83
(0.26,
(0.26,
2.64)
2.64)6.26
6.26
DIABETES
DIABETES
0.60
0.60
(0.20,
(0.20,
1.50)
1.50)8.29
8.29
SCANDSTENT
SCANDSTENT
0.33
0.33
(0.09,
(0.09,
1.18)
1.18)5.08
5.08
TAXUS
TAXUS
VV
- complex
- complex
1.84
1.84
(0.86,
(0.86,
3.94)
3.94)14.52
14.52
I-VI-V
Overall
Overall
(I-squared
(I-squared= =
25.5%,
25.5%,
p p= =
0.194)
0.194)
Random
Effects
0.83
0.83
(0.62,
(0.62,
1.10)
1.10)100.00
100.00
0.77
(0.54,1.10)
*Fixed
Effects (I2=25.5%)
D+L
D+L
Overall
Overall
0.83
(0.62,1.10),
p=0.19
0.77
0.77
(0.54,
(0.54,
1.10)
1.10)
Favors DES
.1.1
.1
Favors BMS
111
Ajay J. Kirtane and Gregg W. Stone, 2008
10
10
10
Mean f/u 1.5 yrs
MI: All Registries
129,955 patients, 24 registries
Study ID
%
Weight
ES (95% CI)
(D+L)
NHLBI (off label, adjusted)
0.71 (0.50, 1.00)
5.01
NHLBI (on label, adjusted)
0.71 (0.47, 1.05)
4.21
Germany Metabolic Syndrome
0.23 (0.07, 0.78)
0.70
Ontario (matched)
1.10 (0.91, 1.32)
8.26
Mayo FFR Substudy
0.67 (0.12, 3.84)
0.35
Brazil Large Vessels
1.50 (0.25, 8.90)
0.33
Italian Diabetic Multivessel (adjusted)
1.02 (0.46, 2.25)
1.50
McMaster STEMI (adjusted)
0.28 (0.04, 1.71)
0.30
Washington Hosp Center (matched)
0.51 (0.29, 0.88)
2.69
Asan Korea (adjusted)
0.66 (0.42, 1.05)
3.55
SCAAR (adjusted)
1.01 (0.91, 1.11)
10.17
Wake Forest (adjusted)
0.84 (0.60, 1.18)
5.15
Western Denmark (adjusted)
1.29 (1.06, 1.57)
8.02
Massachusetts (matched)
0.92 (0.83, 1.02)
10.10
STENT (adjusted)
0.69 (0.52, 0.92)
6.10
GHOST (adjusted)
1.12 (0.74, 1.70)
4.03
DEScover (unadjusted)
0.69 (0.40, 1.18)
2.80
Cedars Acute MI
0.25 (0.06, 1.16)
0.48
REAL (adjusted)
0.92 (0.76, 1.11)
8.18
Melbourne
1.00 (0.39, 2.58)
1.10
ACUITY (from RCT)
1.07 (0.91, 1.25)
8.90
RESTEM
0.80 (0.52, 1.23)
3.86
ARTS II (from RCT)
0.53 (0.32, 0.88)
3.09
ERACI III (from RCT)
2.30 (0.91, 5.96)
1.11
Estimate (95% CI) Weight (%)
D+L
Overall (I-squared
p = 0.000)
*Random
Effects =(I257.9%,
=57.9%)
0.89 (0.80,0.98),
(0.80, 0.98) 100.00
0.89
p=0.023
I-V
Overall
Fixed
Effects
0.96 (0.91,1.01)
(0.91, 1.01)
0.96
*MI is NOTE:
QWMI in
Washington
Hospital
Center,
RESTEM
Weights
are from
random
effects
analysis
.1
Favors DES
1
Favors BMS
Ajay J. Kirtane and Gregg W. Stone, 2008
10
Mean f/u 2.5 yrs
MI: Unadjusted Registries
%
88,221 patients, 18 registries
Weight
ES (95% CI)
Study ID
(D+L)
Estimate (95% CI) Weight (%)
NHLBI (off label, unadjusted)
0.75 (0.55, 1.01)
8.11
NHLBI (on label, unadjusted)
0.80 (0.56, 1.16)
7.22
Germany Metabolic Syndrome
0.23 (0.07, 0.78)
1.56
Mayo FFR Substudy
0.67 (0.12, 3.84)
0.82
Brazil Large Vessels
1.50 (0.25, 8.90)
0.77
McMaster STEMI (unadjusted)
0.29 (0.04, 1.86)
0.67
Asan Korea (unadjusted)
0.78 (0.55, 1.11)
7.41
Western Denmark (unadjusted)
1.19 (0.99, 1.43)
9.89
Massachusetts (unadjusted)
0.75 (0.69, 0.82)
10.97
STENT (unadjusted)
0.62 (0.49, 0.78)
9.19
DEScover (unadjusted)
0.69 (0.40, 1.18)
5.02
Cedars Acute MI
0.25 (0.06, 1.16)
1.09
REAL (unadjusted)
1.15 (1.02, 1.30)
10.65
Melbourne
1.00 (0.39, 2.58)
2.34
ACUITY (from RCT)
1.07 (0.91, 1.25)
10.22
RESTEM
0.80 (0.52, 1.23)
6.31
ARTS II (from RCT)
0.53 (0.32, 0.88)
5.40
ERACI III (from RCT)
2.30 (0.91, 5.96)
2.36
p = 0.000)
(I-squared(I=277.8%,
D+L Overall Effects
*Random
=77.8%)
I-V Overall
Fixed
Effects
100.00
0.97)
0.83 (0.70,
0.83
(0.70,0.97),
p=0.023
0.93)
0.88 (0.83,
0.88
(0.83,0.93)
*MI
is QWMI
in Washington
RESTEM
effects analysis
random Center,
are from Hospital
Weights
NOTE:
.1
Favors DES
1
Favors BMS
Ajay J. Kirtane and Gregg W. Stone, 2008
10
Mean f/u 2.0 yrs
MI: Adjusted Registries
Study ID
%
107,294 patients, 14 registries
ES (95% CI)
Weight
(D+L)
Estimate (95% CI)
Weight (%)
NHLBI (off label, adjusted)
0.71 (0.50, 1.00)
6.18
NHLBI (on label, adjusted)
0.71 (0.47, 1.05)
5.11
Ontario (matched)
1.10 (0.91, 1.32)
10.94
Italian Diabetic Multivessel (adjusted)
1.02 (0.46, 2.25)
1.72
McMaster STEMI (adjusted)
0.28 (0.04, 1.71)
0.34
Washington Hosp Center (matched)
0.51 (0.29, 0.88)
3.16
Asan Korea (adjusted)
0.66 (0.42, 1.05)
4.25
SCAAR (adjusted)
1.01 (0.91, 1.11)
14.06
Wake Forest (adjusted)
0.84 (0.60, 1.18)
6.37
Western Denmark (adjusted)
1.29 (1.06, 1.57)
10.56
Massachusetts (matched)
0.92 (0.83, 1.02)
13.94
STENT (adjusted)
0.69 (0.52, 0.92)
7.70
GHOST (adjusted)
1.12 (0.74, 1.70)
4.87
REAL (adjusted)
0.92 (0.76, 1.11)
10.81
D+L
OverallEffects
(I-squared
= 60.8%, p = 0.002)
*Random
(I2=60.8%)
0.91 (0.81,
1.01) p=0.083
100.00
0.91
(0.81,1.01),
Fixed
Effects
I-V
Overall
0.96 (0.91,
1.01)
0.96
(0.91,1.01)
*MI
is QWMI
in Washington
Hospital
Centereffects
NOTE:
Weights
are from
random
.1
analysis
Favors DES
1
Favors BMS
Ajay J. Kirtane and Gregg W. Stone, 2008
10
Mean f/u 2.8 yrs
Study ID
TVR: All RCTs
7,291 patients, 16 trials
%
Weight
ES (95% CI)
Estimate (95% CI)
(D+L)
Weight (%)
SESAMI
0.36 (0.17, 0.79)
4.36
Typhoon
0.42 (0.25, 0.69)
7.20
STRATEGY
0.34 (0.16, 0.77)
4.22
HAAMU-STENT
0.33 (0.09, 1.19)
1.91
MISSION!
0.38 (0.17, 0.85)
4.08
PRISON II
0.37 (0.19, 0.69)
5.49
Pache et al
0.38 (0.23, 0.64)
7.14
Ortolani et al
0.58 (0.25, 1.36)
3.78
SCANDSTENT
0.17 (0.09, 0.33)
5.44
RAVEL
0.51 (0.25, 1.04)
4.83
SIRIUS
0.48 (0.37, 0.62)
11.51
C-SIRIUS
0.30 (0.10, 0.93)
2.45
E-SIRIUS
0.35 (0.21, 0.56)
7.45
TAXUS II
0.61 (0.35, 1.08)
6.44
TAXUS IV
0.57 (0.45, 0.72)
11.94
TAXUS V
0.77 (0.60, 0.98)
11.75
D+L OverallEffects
(I-squared
= 53.2%, p = 0.006)
*Random
(I2=53.2%)
Fixed
Effects
I-V Overall
0.45 (0.37,
0.54) p<0.001
100.00
0.45
(0.37,0.54),
0.51 (0.45,
0.57)
0.51
(0.45,0.57)
NOTE: Weights are from random effects analysis
.1
Favors DES
1
Favors BMS
Ajay J. Kirtane and Gregg W. Stone, 2008
10
Mean f/u 3.2 yrs
TVR: RCTs
*Hazard Ratio
.4
.6
.8
Meta-Regression on Percent Angiographic F/U
.2
p=0.73
20
40
60
80
perangfu
Percentage of Patients
with Angiographic F/U
100
tvr_hr
Fitted
*log(HR) regressed on percentage of
pts with angiographic f/u;
figurevalues
displayed on exponentiated scale
Ajay J. Kirtane and Gregg W. Stone, 2008
Study ID
TVR: All Registries
73,819 patients, 17 registries
%
Weight
ES (95% CI)
Estimate (95% CI)
(D+L)
Weight (%)
Ontario (matched)
0.69 (0.60, 0.80)
9.88
Mayo FFR Substudy
0.18 (0.04, 0.78)
0.68
Brazil Large Vessels
0.43 (0.17, 1.10)
1.57
McMaster STEMI (adjusted)
0.32 (0.05, 1.92)
0.46
Washington Hosp Center (matched)
0.65 (0.49, 0.85)
7.35
Asan Korea (adjusted)
0.32 (0.24, 0.43)
7.05
Wake Forest (adjusted)
0.63 (0.48, 0.83)
7.38
NY State (adjusted, unmatched)
0.54 (0.50, 0.60)
10.70
STENT (adjusted)
0.58 (0.47, 0.71)
8.70
GHOST (adjusted)
0.28 (0.20, 0.39)
6.31
Montevergine
0.51 (0.39, 0.68)
7.30
DEScover (adjusted)
0.58 (0.40, 0.83)
5.81
Cedars Acute MI
0.22 (0.08, 0.62)
1.34
REAL (adjusted)
0.67 (0.59, 0.76)
10.17
Multicenter SVG (adjusted)
0.58 (0.28, 1.18)
2.41
RESTEM
0.62 (0.47, 0.80)
7.53
ERACI III (from RCT)
0.58 (0.39, 0.86)
5.35
D+L Overall Effects
(I-squared
71.2%, p = 0.000)
*Random
(I2==71.2%)
I-V Overall
Fixed
Effects
0.53 (0.47,
0.61)
100.00
0.53
(0.47,0.61),
p<0.001
0.57 (0.54,
0.60)
0.57
(0.54,0.60)
NOTE: Weights are from random effects analysis
.1
Favors DES
1
Favors BMS
Ajay J. Kirtane and Gregg W. Stone, 2008
10
Mean f/u 2.2 yrs
Study ID
TVR: Unadjusted Registries
55,531 patients, 12 registries
ES (95% CI)
Estimate (95% CI)
%
Weight
(D+L)
Weight (%)
Mayo FFR Substudy
0.18 (0.04, 0.78)
1.82
Brazil Large Vessels
0.43 (0.17, 1.10)
3.80
McMaster STEMI (unadjusted)
0.29 (0.04, 1.86)
1.16
Wake Forest (unadjusted)
0.62 (0.48, 0.80)
11.08
NY State (unadjusted, unmatched)
0.56 (0.51, 0.62)
12.76
STENT (unadjusted)
0.74 (0.63, 0.87)
12.21
Montevergine
0.51 (0.39, 0.68)
10.77
DEScover (unadjusted)
0.63 (0.46, 0.87)
10.20
Cedars Acute MI
0.22 (0.08, 0.62)
3.32
REAL (unadjusted)
0.99 (0.91, 1.08)
12.84
RESTEM
0.62 (0.47, 0.80)
10.93
ERACI III (from RCT)
0.58 (0.39, 0.86)
9.11
*Random
D+L Overall Effects
(I-squared(I2==88.9%)
88.9%, p = 0.000)
0.60 (0.48,
0.74)
100.00
0.60
(0.48,0.74),
p<0.001
Fixed
Effects
I-V Overall
0.73 (0.69,
0.77)
0.73
(0.69,0.77)
NOTE: Weights are from random effects analysis
.1
Favors DES
1
Favors BMS
Ajay J. Kirtane and Gregg W. Stone, 2008
10 Mean f/u 2.2 yrs
TVR: Adjusted Registries
63,456 patients, 11 registries
Study ID
%
Weight
ES (95% CI)
Estimate (95% CI)
(D+L)
Weight (%)
Ontario (matched)
0.69 (0.60, 0.80)
12.72
McMaster STEMI (adjusted)
0.32 (0.05, 1.92)
0.66
Washington Hosp Center (matched)
0.65 (0.49, 0.85)
9.74
Asan Korea (adjusted)
0.32 (0.24, 0.43)
9.38
Wake Forest (adjusted)
0.63 (0.48, 0.83)
9.78
NY State (adjusted, unmatched)
0.54 (0.50, 0.60)
13.65
STENT (adjusted)
0.58 (0.47, 0.71)
11.34
GHOST (adjusted)
0.28 (0.20, 0.39)
8.46
DEScover (adjusted)
0.58 (0.40, 0.83)
7.83
REAL (adjusted)
0.67 (0.59, 0.76)
13.05
Multicenter SVG (adjusted)
0.58 (0.28, 1.18)
3.39
D+L Overall
(I-squared
= 79.4%, p = 0.000)
*Random
Effects
(I2=79.4%)
0.54 (0.46,0.63),
(0.46, 0.63) p<0.001
100.00
0.54
0.58
0.58 (0.54,0.61)
(0.54, 0.61)
I-V Overall
Fixed
Effects
NOTE: Weights are from random effects analysis
.1
Favors DES
1
Favors BMS
Ajay J. Kirtane and Gregg W. Stone, 2008
10
Mean f/u 2.2 yrs
Summary: DES vs. BMS
Treatment Effect Estimates
RCTs
- Fixed effects
- Random effects
Registries
- Fixed effects
- Random effects
Mortality
MI
TVR
8,867 pts,
21 trials
8,850 pts,
20 trials
7,291 pts,
16 trials
0.97
0.97
0.94
0.94
0.51
0.45*
161,232 pts,
28 studies
129,955 pts,
24 studies
73,819 pts,
17 studies
0.83
0.80*
0.96
0.89*
0.57
0.53*
<1.0  DES better
Study Limitations
• Randomized trial analyses are still underpowered
to assess these clinical endpoints
• Registry analyses are based upon observational,
non-randomized analyses
• Potential for residual confounding
• Significant heterogeneity, despite attempts to
address this through random effects models,
meta-regression and sensitivity analyses
• Analysis was primarily of summary-level data
and included unpublished studies
• Use of hazard ratio / relative risk assumes
constant hazards throughout the FU period
Conclusions (1)
In 22 RCTs in which 9,470 pts were randomized to
DES or BMS and followed for ≥1 yr, DES resulted in:
• Non significant 3% and 6% reductions in mortality
and MI respectively
• A highly significant 55% reduction in TVR
In 30 registries in which 174,302 pts were treated
with either DES or BMS (non-randomized) and
followed for ≥1 yr, DES was associated with:
• A highly significant 20% reduction in mortality
• A significant 11% reduction in MI
• A highly significant 47% reduction in TVR
Conclusions (2)
The favorable results of DES from the RCT and
registry analysis populations were robust and
consistent for both on-label and off-label use,
and for clinical f/u extending to 3-4 years
These findings, derived from more than
180,000 pts treated in 52 studies, strongly
suggest that DES are safe for both on-label
and off-label use, and have comparable
efficacy in both RCTs and in the “real-world”