Dr Akhil Kapur

1
Coronary Artery Revascularisation in Diabetes
Trial
Dr Akhil Kapur London Chest Hospital, Barts and
the London NHS Trust, London, UK On behalf of
the CARDia Investigators Friday 30th January 2009
2
MY CONFLICTS OF INTEREST ARE Grants Cordis,
Eli-Lilly, BMS-Sanofi, Boston Scientific,
Medtronic
3
Revascularisation in Diabetic Patients with
Multivessel Disease

• No specific randomised comparison of CABG and PCI
  until now
• Largest randomised comparison to date

4
CARDia Trial Design
Diabetic patients with multivessel disease or
complex single vessel disease
Suitable for PCI or CABG
Inclusion and exclusion criteria met
CONSENT
Randomisation
Up to date CABG
Optimal PCI stent abciximab
DES 72 BMS 28
5
Main Exclusion Criteria

• Informed consent could not be obtained
• Age gt80 years
• Previous CABG or PCI
• Left main stem disease
• Cardiogenic shock
• Recent ST elevation myocardial infarction
• Contraindications to abciximab, aspirin and
  clopidogrel

6
Endpoints

• Primary endpoint
• Composite event rate at 1 year of death/non-fatal
  MI/non-fatal stroke (time to first event)
• Major secondary
• Further revascularisation at 1 year
• Secondary
• Severe bleeding complications at 30 days
• New requirement for permanent dialysis
• Neurological morbidity
• Quality of life
• Cost difference between treatments
• Change in LV function

7
CARDia Patient flow Chart
510 Pts randomised
CABG
PCI
256 patients 2 withdrew
254 patients 1 died 7 withdrew 2 no procedure
230received CABG 14cross over to PCI
253received PCI 1cross over to CABG
3 lost to follow up 2 withdrawn
4 lost to follow up 2 withdrawn
95 (242) in 1 year follow up
97 (248) in 1 year follow up
8
Baseline Clinical Characteristics

9
Baseline Clinical Characteristics
10
Baseline Clinical Characteristics

11
Procedural details
12
PCI procedural details

• Use prior to procedure of
• aspirin - 100
• clopidogrel - 95
• abciximab - 95
• --------------------------------------------------
  -------------------------------
• 3 vessel disease - 65
• 3 vessels treated in these patients - 88
• --------------------------------------------------
  -------------------------------
• average no. of stents per patient - 3.6
• average stent length - 71mm
• --------------------------------------------------
  -------------------------------
• DES patients (cypher) - 72 (181)
• BMS patients - 28 (72)

13
CABG procedural details

• 3 vessel disease - 60
• 3 vessels treated in these patients - 90
• --------------------------------------------------
  --------------------------
• average no of grafts - 2.9
• LIMAs - 94
• with at least two arterial grafts - 17
• off pump - 31

14
Results - Intention to treat analysis
15
Primary composite outcome at 1 year
16
Individual 1 year outcomes
p0.97 p0.09 p0.07
9.8 (n25)
5.7 (n14)
3.2 (n8)
3.2 (n8)
2.8 (n7)
0.4 (n1)
Non fatal MI
Death
Non fatal stroke
17
CCS Class at 12 months
18
Survival at 1 year CABG vs PCI
19
Death, MI, stroke and repeat revascularisation
20
Subgroup - CABG vs PCI-DES
21
Subgroup - CABG vs PCI-DES Composite
endpoint at 1 year
22
Study Limitations

• Planned recruitment not achieved 510 patients out
  of 600 recruited 85
• Formal non-inferiority parameters not fulfilled
  (insufficient power)
• we did not match the predicted PCI event rate
  originally estimated to be lower than predicted
  CABG rate

23
Test of non inferiority of PCI vs. CABG for
primary endpoint - upper boundary of 1.3 (red
line) shown
24
CARDia Main Conclusions

• First randomised comparison of coronary
  revascularisation in diabetes
• Broadly similar primary endpoint at 1 year
• More repeat revascularisation in the PCI group
• Rate of stroke in respective arms consistent with
  other interventional studies
• We will follow up patients for 5 years to
  increase power

25
CARDia Participating Centres
26
CARDia Participating Centres continued