The Detrimental Impact of Chronic Renal Insufficiency

1
CYPHER Stent vs. Taxus StentRandomized Trials
Kastrati A., et al., ACC 2007 Oral Presentation.
2
The ACC/AHA and ESC Guidelines

• American College of Cardiology/American Heart
  Association (ACC/AHA) Style
• Classification of Evidence1
• Level of Evidence A Data derived from multiple
  randomized clinical trials
• Level of Evidence B Data derived from a single
  randomized trial or nonrandomized studies
• Level of Evidence C Consensus opinion of
  experts.

• European Society of Cardiology (ESC)
  Classification of Evidence2
• Level of Evidence A Data derived from multiple
  randomized clinical trials or meta-analyses
• Level of Evidence B Data derived from a single
  randomized trial or large non-randomized studies
• Level of Evidence C Consensus opinion of
  experts and / or small studies, retrospective
  studies, registries

1 Smith, et al., JACC 2001 37 2215-38. 2
Silber S., et al., EHJ 2005 26804-47.
3
Randomized vs. Observational Studies Lessons
from AMI Studies
Meta-analysis of RCTs
Registry Study
100
Thrombolysis
90
survival ()
PTCA
80
70
1
2
3
4
0
time after discharge (years)
Keeley et al, Lancet 2003
Every et al, NEJM 1996
Kastrati A., et al., ACC 2007 Oral Presentation.
4
Uniformity of RCT Results Lessons from Statin
Studies
Uniformity of results across RCTs is rarely
achieved
Simva Prava Prava Lova Prava Simva Fluva Prava Ato
rva
Cheung et al, Br J Clin Pharmacol 2004
Kastrati A., et al., ACC 2007 Oral Presentation.
5
Objective of the Meta-Analysis
To assess and compare long-term outcomes in
randomized trials of sirolimus-eluting stents
(SES) vs. paclitaxel-eluting stents (PES)
Kastrati A., et al., ACC 2007 Oral Presentation.
6
Inclusion Criteria

• Randomized trial, published or presented at
  meetings assessing sirolimus-eluting stent (SES)
  vs. paclitaxel-eluting stent (PES)

Kastrati A., et al., ACC 2007 Oral Presentation.
7
Study Endpoints

• Outcomes of interest
• - Stent thrombosis
• - All-cause mortality
• Composite of death or myocardial infarction (MI)
• Composite of death, MI or reintervention (MACE)
• during the entire available follow-up interval

Kastrati A., et al., ACC 2007 Oral Presentation.
8
Statistical Methods

• Mantel-Cox method for calculating hazard ratios
  for individual trials
• Fixed and random effect models for calculating
  overall harzard ratios and check for
  heterogeneity
• Kaplan-Meier survival curves for all trials
  combined

Kastrati A., et al., ACC 2007 Oral Presentation.
9
Included SES vs. PES Trials
Total No. of patients
Mean Clinical FU in months
Trial
BASKET
545
18.2
CORPAL
652
30.5
ISAR-DESIRE
200
33.9
ISAR-DIABETES
250
32.1
ISAR-SMART 3
360
33.9
LONGDES II
500
13.0
PROSIT
308
12.0
REALITY
1,353
24.1
SIRTAX
1,012
24.2
SORT-OUT II
2,098
9.0
TAXI
202
36.9
Overall
7,480
20.0
Kastrati A., et al., ACC 2007 Oral Presentation.
10
Trials That Provide Individual Patient Data
Total No. of patients
Mean Clinical FU in months
Trial
BASKET
545
18.2
CORPAL
652
30.5
ISAR-DESIRE
200
33.9
ISAR-DIABETES
250
32.1
ISAR-SMART 3
360
33.9
LONGDES II
500
13.0
REALITY
1,353
24.1
SIRTAX
1,012
24.2
TAXI
202
36.9
Overall
5,074
25.1
Kastrati A., et al., ACC 2007 Oral Presentation.
11
Cumulative Incidence of Stent Thrombosis
5
4
3
Probability of Stent Thrombosis,
2
1
Sirolimus-eluting stent
Paclitaxel-eluting stent
0
0
6
12
18
24
30
Patients at Risk
Months After Randomization
SES
2539
2483
2319
2151
1653
573
PES
2535
2457
2311
2121
1603
533
Kastrati A., et al., ACC 2007 Oral Presentation.
12
Risk of Stent Thrombosis
No. of events / Total No. of patients
Favors SES
Favors PES
Trial
SES group
PES group
BASKET
4/264
5/281
2/331
4/321
CORPAL
0/100
2/100
ISAR-DESIRE
ISAR-DIABETES
0/125
2/125
ISAR-SMART 3
1/180
1/180
LONG-DES II
1/250
5/250
PROSIT
0/154
2/154
REALITY
6/684
18/669
SIRTAX
12/503
15/509
SORT-OUT II
NA
TAXI
2/102
2/100
Overall
28/2693
56/2689
0.54 (0.34, 0.85)
.1
10
1
Test for Heterogeneity Cochran Q5.3 (d.f.9)
P.81 Test for Inconsistency I2 0.0
Hazard Ratio
Kastrati A., et al., ACC 2007 Oral Presentation.
13
Influence of Individual Trials
Meta-analysis random-effects estimates
(exponential form)
Study ommited
BASKET
CORPAL
ISAR-DESIRE
ISAR-DIABETES
ISAR-SMART 3
LONG-DES II
PROSIT
REALITY
SIRTAX
TAXI
0.24
0.54
0.34
0.85
1.08

Kastrati A., et al., ACC 2007 Oral Presentation.
14
Kaplan-Meier (K/M) Curves of Survival
100
90
80
Probability of Survival,
70
60
Sirolimus-eluting stent
Paclitaxel-eluting stent
50
0
6
12
18
24
30
Patients at Risk
Months After Randomization
SES
2539
2498
2336
2170
1666
608
PES
2535
2475
2338
2152
1631
552
Kastrati A., et al., ACC 2007 Oral Presentation.
15
Risk of Death
No. of events / Total No. of patients
Favors SES
Favors PES
Trial
SES group
PES group
BASKET
10/264
11/281
18/331
22/321
CORPAL
ISAR-DESIRE
8/100
6/100
ISAR-DIABETES
21/125
22/125
ISAR-SMART 3
10/180
13/180
LONG-DES II
2/250
0/250
PROSIT
5/154
9/154
REALITY
23/684
23/669
SIRTAX
25/503
25/509
SORT-OUT II
19/1065
19/1033
TAXI
7/102
3/100
Overall
148/3758
153/3722
0.93 (0.74, 1.16)
.1
10
1
Test for Heterogeneity Cochran Q5.3 (d.f.10)
P.87 Test for Inconsistency I2 0.0
Hazard Ratio
Kastrati A., et al., ACC 2007 Oral Presentation.
16
K/M Curves of Survival Free of MI
100
90
80
Probability of Survival Free of MI,
70
60
Sirolimus-eluting stent
Paclitaxel-eluting stent
50
0
6
12
18
24
30
Patients at Risk
Months After Randomization
SES
2539
2401
2236
2080
1596
581
PES
2535
2361
2230
2043
1545
529
Kastrati A., et al., ACC 2007 Oral Presentation.
17
Risk of Death or MI
No. of events / Total No. of patients
Favors SES
Favors PES
Trial
SES group
PES group
BASKET
23/264
29/281
33/331
37/321
CORPAL
11/100
10/100
ISAR-DESIRE
ISAR-DIABETES
26/125
26/125
ISAR-SMART 3
19/180
19/180
LONG-DES II
22/250
27/250
REALITY
54/684
69/669
SIRTAX
41/503
46/509
TAXI
11/102
10/100
Overall
240/2539
273/2535
0.86 (0.72, 1.02)
.1
10
1
Test for Heterogeneity Cochran Q1.2 (d.f.8)
P.99 Test for Inconsistency I2 0.0
Hazard Ratio
Kastrati A., et al., ACC 2007 Oral Presentation.
18
K/M Curves of Survival Free of MACE
100
90
80
Probability of Survival Free of MACE,
70
60
Sirolimus-eluting stent
Paclitaxel-eluting stent
50
0
6
12
18
24
30
Patients at Risk
Months After Randomization
SES
2539
2374
2114
1941
1468
527
PES
2535
2313
2061
1855
1371
467
Kastrati A., et al., ACC 2007 Oral Presentation.
19
Risk of MACE
No. of events / Total No. of patients
Hazard Ratio
Favors SES
Favors PES
Trial
SES group
PES group
BASKET
32/264
35/281
51/331
52/321
CORPAL
23/100
37/100
ISAR-DESIRE
ISAR-DIABETES
30/125
30/125
ISAR-SMART 3
34/180
47/180
LONG-DES II
27/250
42/250
PROSIT
9/154
18/154
REALITY
93/684
104/669
SIRTAX
57/503
87/509
SORT-OUT II
83/1065
89/1033
TAXI
14/102
9/100
Overall
453/3758
550/3722
0.79 (0.69, 0.91)
Test for Heterogeneity Cochran Q12.2 (d.f.10)
P.27 Test for Inconsistency I2 18.2
.1
10
1
Kastrati A., et al., ACC 2007 Oral Presentation.
20
Influence of Individual Trials
Meta-analysis random-effects estimates
(exponential form)
Study ommited
BASKET
CORPAL
ISAR-DESIRE
ISAR-DIABETES
ISAR-SMART 3
LONG-DES II
PROSIT
REALITY
SIRTAX
SORT-OUT II
TAXI
0.65
0.79
0.69
0.91
0.95

Kastrati A., et al., ACC 2007 Oral Presentation.
21
Conclusions I

• With 7500 patients randomized between CYPHER
  and Taxus drug-eluting stents, we are provided
  with the most abundant evidence of randomized
  stent vs. stent trials ever accumulated.
• There is no significant heterogeneity across
  trials with respect to the treatment effect.

Kastrati A., et al., ACC 2007 Oral Presentation.
22
Conclusions II

• Compared with Taxus stent, CYPHER stent is
  associated with
• No significant difference in mortality
• A significantly lower risk of stent thrombosis
• A trend toward a lower combined risk of death or
  myocardial infarction
• A significantly lower risk of death, myocardial
  infarction or reintervention

Kastrati A., et al., ACC 2007 Oral Presentation.