Acute Coronary Syndromes and Diabetes High Risk and in need of more attention

1
Acute Coronary Syndromes and DiabetesHigh Risk
and in need of more attention ?

• David Fitchett MD
• St Michaels Hospital
• Toronto

Lake Louise, March 2008
2
Disclosures

• CME, Research or Consulting fees from
• GSK
• Boerhinger
• Merck
• Schering
• Sanofi-Aventis
• BMS
• Pfizer
• Roche
• Biovail
• Servier

3
Cardiologists View of Diabetes

• Diabetes is a state of premature cardiovascular
  death which is associated with chronic
  hyperglycemia and may also be associated with
  blindness and renal failure

Dr Miles Fisher
4
Diabetes and Myocardial Infarction

• The magnitude of the problem
• Why diabetes is associated with a worse prognosis
  
• Does glycemic control matter ?
• Does ACS treatment work in diabetes?
• The care gap

5
Evolution without Natural Selection
Economist, January 2004
6
Changing Epidemiology of Type 2 Diabetes in
Ontario
20 16 12 8 4 0
20 16 12 8 4 0
Prevalence ()
Mortality and Incidence Rate per 1,000 Ontarians
Incidence
94/95
95/96
96/97
97/98
98/99
99/00
00/01
01/02
02/03
03/04
04/05
Fiscal Year
Adapted from Lipscombe LL. Healthc Q 2007
10(3)23-5.
7
(No Transcript)
8
Diabetes Mellitus A Cardiovascular Disease
United Kingdom Prospective Diabetes Study n3867
Incidence
UKPDS Investigators. Lancet 1998 352 837
9
Admission Rate For AMI Ontario 1995-1999

• Admission rates with diabetes
• 7 fold greater
• Age / gender adjusted
• 3 fold greater

9 ?
N104,471 (30 DM)
Hux JE et al, Diabetes in Ontario, an ICES
Practice Atlas, 2003
10
2007
11
Diabetes and Outcome
In-Hospital
of Patients
25
20
15
10
pNS
plt0.001
5
5.7
4.8
3.9
1.8
0
Mortality
(re)MI
In hospital survivors
Yan R et al Am Heart J 2006152676
12
Independent Prognostic Significance of Diabetes
on Clinical Outcome
Predictors of 1 Year Mortality
Clinical Predictors Odds Ratio (95 CI) p
value Age (per 10 year ?) 2.06
(1.83-2.31) lt0.001 Previous MI 1.36
(1.07-1.73) 0.013 Previous CHF 1.56
(1.16-2.10) 0.003 Systolic BP (per 10mmHg
?) 0.91 (0.88-0.95) lt0.001 Heart rate (per 10
bpm ?) 1.15 (1.10-1.20) lt0.001 Creatinine (per
10 umol/L ?) 1.05 (1.04-1.06) lt0.001 Elevated
CK/troponin 1.86 (1.41-2.43) lt0.001 ST
depression 1.24 (0.96-1.61) 0.06 In-hospital
PCI 0.71 (0.48-1.06) 0.09 Diabetes 1.4
7 (1.15-1.87) 0.002
Logistic Multivariable Regression Analysis
predictors from GRACE model
Model Performance c-statistic 0.81
(discrimination) Hosmer-Lemeshow p0.39
(calibration)
Yan R et al Am Heart J 2006152676
13
Diabetes confers a doubling of risk for early MI
mortality despite advances in cardiac care
?
Defibrillation Hemodynamic monitoring
Thrombolysis Beta-blockadeAspirin
PCIIIbIIIa inhibitorsClopidrogelStatins
Early mortality from acute MI
(pre-1962)
(1984-2000)
(2000- )
(1962-1984)
Modified by R Nesto from NEJM 1997 337 1361
14
Worse Outcomes after ACSin the Patient with
Diabetes
Severity of CAD Myocardial dysfunction Endothelial
dysfunction Increased coagulation Autonomic
imbalance Impact of hyperglycemia Clinical
factors

• Recurrent MI
• Heart failure
• Sudden death

15
Why are Patients with Diabetes at Increased Risk
after ACS ?

• Older
• More female
• Atypical symptoms
• Dyspnoea -- Fatigue
• Nausea -- Vomiting
• Disturbance of glycemic control
• Delayed presentation
• Less use of proven treatment
• Response to treatment
• Comorbidity

16
Heart Failure Post ACS in Diabetes

• More CHF even after adjustment for MI size and
  extent of CAD
• Prior silent infarction in 40 first AMI
• Impaired function of non-infarct zone
• Diastolic dysfunction
• Autonomic impairment
• Metabolic disturbance
• ? Fibrosis
• Endothelial dysfunction

17
Diabetes and Cardiogenic Shock
Incidence vs Age
Lindholm et al Eur Hear J 20057834
18
Prior CABG, Diabetes, and AMI? Explains BARI
Result ?
Diabetes, No CABG n23 No Diabetes, No
CABG n38 Diabetes, CABG n27 No
Diabetes CABG n67
1.0
0.8
0.6
RR 0.09 (0.03-0.29)
Mortality
0.4
0.2
0.0
0 1 2 3
4 5
Years after MI
Detre et al NEJM 2000342989
19
Impaired Glucose Metabolism and ACS

• Impact on survival
• Impact on co-morbidity
• Need to identify
• Management
• Acute Hyperglycemia
• Long-term multifactorial control

20
Prevalence of Abnormal Glucose HandlingAMI
Patients and Controls
41
26
65
65
25
9
33
33
34
34
33
33
AMI patients
AMI patients
41
41
34
25
34
25
Relative prevalence ()
Relative prevalence ()
Bartnik et al J Int Med 2004256288
21
Abnormal Glucometabolic State as Predictor of
Outcome after AMI
Admission Glucose lt 11mm/l OGTT at
discharge 113/168 abnormal
Bartnik et al Eur Heart J 2004251990
22
Newly Diagnosed Diabetes Worse Outcome After
AMI

• New diabetics vs Non diabetic
• Similar baseline characteristics
• Similar presentation

30
Previous diabetes
New diabetes
Non diabetic
20
1 year Mortality ()
10
Months
0 3 6
9 12
VALIANT
Aguilar D et al. Circulation 2004
23
Mortality by Admission Glucose Level
lt 5.5 5.5 - 7.0 7.0 8.5
8.5 -12.0 gt 12.0
( mM / l )
Kosiborod et al Circulation 20051113078
24
Association between Mean Blood Glucose and
In-Hospital Mortality.
Reference Mean BG 100-lt110
No Diabetes
No diabetes
All
All
Diabetes
Diabetes
Kosiborod et al Circulation 20081171018
25
Hyperglycemia and Adverse Outcomes After
Myocardial Infarction
Causative
Marker

• Relative insulin deficiency
• ? FFA toxic
• ? glucose for anaerobic metabolism
• Increased thrombosis
• Increased oxidative stress
• Impaired
• Endothelial function
• Microcirculation (no-reflow)
• Ischemic pre-conditioning

• Stress response
• Larger infarct
• Poor correlation
• New dysglycemia
• IFG or IGT or new DM

26
Role of Hyperglycemia on Infarct Size
Infarct size

• Streptozocin induced DM in rats
• Good glucose control with iv insulin (GGC)
• BS 6mmol/L
• Poor glucose control (PGC) BS 22mmol/L

• HIF 1? expression impaired
• in hyperglycemic rat
• Effects reversed with glutathione
• ? Oxidative stress mediated effect
• of hyperglycemia

Marfella et al Diabetologia 2002451172
27
Hyperglycemia Outweighs the Beneficial Effects of
Hyperinsulinemia on Coronary Vasodilation
Myocardial perfusion measured with PET at
baseline and with either Hyperinsulinemia/Euglyce
mia or Hyperinsulinemia/Hyperglycemic clamp
7 6 5 4 3 2 1 0
plt0.001
MPR
p0.1
Baseline Clamp Baseline Clamp
HE
HH
Srinivasan et al. JACC 200546428
28
Benefit of Tight Glycemic Control after Acute
Myocardial Infarction

• STE and Non STE MI
• Blood glucose gt 11 mmol/L
• N1240
• Randomised to
• Intensive insulin for 24 h multi dose insulin
  for 3 months
• Early glucose target 7-10 mm/l
• or
• Usual care
• Mean FU 3.4 years

RR 0.72 (95 CI 0.55-0.92)
NNT9
DIGAMI Malmberg et al JACC 19952657
Malmberg BMJ 19973141512
29
Mortality after Acute Myocardial Infarction
related to Fasting Blood Sugar and Intensive
Glycemic Control
p for trend plt 0.001 p0.1
Mortality
Fasting Blood Sugar (mmol/L)
DIGAMI Circulation 1999992626
30
DIGAMI 2

• Glycemic Control

Note 14 group 3 received iv insulin in hospital,
and 40 received some insulin
31
DIGAMI 2

• Mortality Outcomes
• Total mortality 21.3
• Group 1 23.4, Group 2 22.6 p0.83
• Group 2 22.6, Group 3 19.3 p0.20
• No differences for non-fatal AMI /stroke
• Predictors of mortality Blood glucose, HbA1c,
  Age, renal function and prior CHF

32
Comparison of DIGAMI 1 and 2

• Degree of glucose control
• DIGAMI 1 admission glucose higher (15.7 mmol/l)
• DIGAMI 1 24 hr glucose different
• 11.7 in control and 9.6 in intensive Rx group
• A1c levels differed at 3 mo and 1 year
• DIGAMI 2 no differences between groups
• Mortality rates very different
• DIGAMI 1 control 1 yr 26, 3.4 yrs 44
• DIGAMI 2 2 yrs 19.3 (similar to Rx group in
  DIGAMI 1)

33
Impact of Long-term Glycemic Treatment in
Patients with AMI
Mellbin et al Eur Heart J 200829166
34
Outcomes in Patients Discharged with New Insulin
and No Insulin
All Cause Mortality

• Patient characteristics
• Insulin No Insulin p
• Age 66 68 0.04
• DM dur 5.6 y 4.5 y 0.02
• HTN 49 40 0.02
• A1C 7.8 7.4 0.01
• Glucose 12.9 12.2 0.01
• Killip No difference
• Adm Meds No difference
• D/C Meds No difference

MI and Stroke
Mellbin et al Eur Heart J 200829166
35
DIGAMI 2 Conclusions

• Level of glycemic control more important than
  insulin administration
• Many patients with acute AMI require insulin to
  achieve rapid glycemic control
• Long term management
• Optimal pharmaceutical uncertain
• Probably need to improve insulin sensitivity
• Optimal glycemic target uncertain
• Impact of ACCORD ?

36
Early impact of insulin on mortality for
hyperglycaemic patients without known diabetes
with ACS

• 20035 Registry in 201 UK hospitals
• 38,864 ACS patients with no hx diabetes
• 3835 with glucose gt 11 mmol/l
• 31 received insulin

Mortality 7 days 30 days Received
insulin 11.6 15.8 No insulin 16.5 22.1 H
R (95 CI) 1.56 (1.22 2.0) 1.51 (1.22 -
1.86) p lt O.OO1 lt
O.OO1
Weston et al Heart 2007931542
37
Glucose Insulin PotassiumCREATE-ECLA
Glucose Levels

• Baseline 6hr 24hrs
• Control 9 8.2 7.5
• GIK 9 10.4 8.6

Mortality
Mortality by Tertiles of Blood Glucose

The CREATE-ECLA Trial Group JAMA 2005393437
38
SWEET ACS trial INTENSIFIED MULTIFACTORIAL
INTERVENTION ON HYPERGLYCEMIC PATIENTS WITH ACS
39
Plaque rupture
Non ST-segment elevation ACS
ST-segment elevation ACS
Are These Treatments Effective in the Patient
with Diabetes and ACS ?
40
Diabetes and Fibrinolysis
30-35 Day Mortality
of Patients
20
Placebo/Control
Fibrinolysis
17.3
15
13.6
10
10.7
8.7
5
0
Non-Diabetic
Diabetic
Fibrinolytic Therapy Trialists (FTT)
Collaborative Group Lancet 1994343311-22
41
Non-ST?ACS GP IIb/IIIa Agents
30 Day Death Diabetic Patients
IIb/IIIa
Odds Ratio 95 CI
Placebo
N
Trial
p0.33
PURSUIT
2,163
6.1
5.1
p0.07
PRISM
687
4.2
1.8
p0.17
PRISM-PLUS
362
6.7
3.6
p0.022
GUSTO IV
1,677
7.8
5.0
p0.51
PARAGON A
412
6.2
4.6
p0.93
PARAGON B
1,157
4.8
4.9
0.74 (0.59, 0.92) p0.007
Pooled
6.2
4.6
6,458
1
2
0
Placebo Better
IIb/IIIa Better
Roffi et al Circulation 20001042767-71
42
Early Invasive Strategy vs Conservative Strategy
6 month Death, MI, Rehospitalization
Conserv. ()
Invasive ()
Odds Ratio 95 CI
20.1
27.7
Diabetes (n613 28)
14.2
16.4
Non-diabetes (n1,607 72)
0
1
2
1.5
0.5
Invasive Better
Conservative Better
Cannon et al N Engl J Med 20013441879-87
43
Outcomes in Patients After MIInvasive vs
Noninvasive Strategy
Death or MI
FRISC 2
Norhammer et al J Am Coll Cardiol 200443585
44
Diabetes Remains a Major Risk Factor for
Mortality after AMI for Patients Managed by an
Invasive Strategy
Independent Predictors
FRISC 2
Norhammer et al J Am Coll Cardiol 200443585
45
AMI and Diabetes ACE Inhibition

• Enhanced benefit
• SMILE
• GISSI 3
• TRACE
• Under utilised
• Fear of uremia
• Fear of hypotension
• Fear of ? K
• Preoccupation with glycemic control

6/52 Severe CHF / death
p0.019
p0.14
SMILE Borghi et al Diabetes Care 2003261862
Nesto et al Circulation 19989712
46
Beta-blockers and Diabetes Implications in MI
Enhanced Treatment Effects in Diabetes

• Concerns
• Masking hypoglycemia
• Impaired glycogenolysis
• ? Glucose intolerance
• Hyperosmolar coma
• ? heart failure

Mortality Reduction

• Mortality reduction
• Non DM 33
• DM 48

BHRT Beta-blocker Heart Attack Trial NMTS
Norwegian Multicentre Timolol Study GMT
Goteburg Metoprolol Trial
47
Diabetes and Therapy in the First 24 Hrs
of Patients
100
Non-diabetic (n3,429)
94.5
91.4
90.9
89.3
Diabetic (n1,149)
80
plt0.001
plt0.05
plt0.001
plt0.001
60
60.7
59.1
46.4
44.9
40
Diabetes an independent negative predictor of
receiving fibrinolysis Odds Ratio 0.72
(0.54-0.95), p0.021
20
plt0.005
7.6
5.2
0
Fibrinolysis
Any Reperfusion
Antiplatelet
Heparin
GP IIb/IIIa inhibitor
Yan R et al Am Heart J 2006152676
48
Diabetes and In-hospital Angiography and
Revascularization
of Patients
60
Non-diabetic (n3,429)
50
Diabetic (n1,149)
p0.006
Diabetes an independent negative predictor of
undergoing coronary intervention Odds Ratio 0.80
(0.65-1.00), p0.05
40
30
plt0.001
20
pNS
10
0
Angiography
PCI
CABG
Yan R et al Am Heart J 2006152676
49
Diabetes and Medication Use at Discharge
of Patients
100
p0.008
Non-diabetic (n3,429)
p0.08
Diabetic (n1,149)
80
plt0.0001
pNS
60
40
20
0
Antiplatelet
Beta Blocker
ACE Inhibitor
Lipid Lowering
Or time of transfer or death
Yan R et al Am Heart J 2006152676
50
Associations Between Use of Therapies Less
Frequently Received by Diabetic Patients and 1
Year Mortality
Clinical Predictors Odds Ratio (95 CI) p
value Diabetes 1.48 (1.16-1.90) 0.002 In-hospita
l PCI 0.71 (0.48-1.06) 0.09 Antiplatelet
therapy 0.48 (0.36-0.55) lt0.001 Beta
blocker 0.71 (0.55-0.91) 0.007 Lipid lowering
therapy 0.77 (0.60-0.98) 0.04

• Interaction between treatments and diabetes on
  1-year mortality
• In-hospital PCI (p0.55)
• Antiplatelet therapy (p0.24)
• Beta-blockers (p0.78)

Yan et al Am Heart J 2006152676
51
Underutilization of an Invasive
Management Strategy in Canadian Non-ST? ACS
Diabetic Patients The Problem Still Remains
of Patients
80
Non-diabetic
ACS I (1999-2001) Diabetes 25.9 ACS II
(2002-2004) Diabetes 27.2
p?0.001
ACS I DM
67.1
60
ACS II DM
57.5
p?0.001
43.5
40
p?0.001
36.1
35.2
Multivariable analysis adjusting for other
confounding factors including renal
dysfunction Diabetes independent negative
predictor of use of coronary angiography OR 0.75
(0.60-0.94) plt0.02) and revascularization OR
0.77 (0.62-0.96) p0.02
33.8
23.1
20
p?0.001
plt0.02
14.8
12.5
10.2
10.1
3
2.1
1.3
n757
532
1424
4.5
0
ACS I
ACS II
ACS I
ACS II
ACS I
ACS II
ACS I
ACS II
ACS I
ACS II
Death
Any Revasc.
CABG
PCI
Angiography
Yan R et al Circulation 2005112(Suppl II)694.
52
Conclusions

• Diabetic patients with ACS are high risk
• Yet are
• less likely to receive fibrinolysis, antiplatelet
  agents or anticoagulation in the first 24 hrs
• less likely to undergo invasive management and
  revascularization
• less likely to receive long-term evidence-based
  medical therapies with proven benefit

Yan et al Am Heart J 2006152676
53
Implications

• Under-treatment may have contributed to adverse
  prognosis
• Improved use / adherence to recommended treatment
  likely will have large impact on prognosis

54
Failure to Prescribe Statins

• Post MI
• Ontario patients with AMI (2001-4)
• 35.6 discharged on statin
• Lowest chance of receiving statin
• highest risk patient
• with greatest benefit

Austin et al Am Heart J 2006151969
55
Failure to Prescribe Statin after AMI

• More likely to be
• Older
• Female
• With
• Heart Failure
• Diabetes

• Less likely to receive
• ASA
• B blocker
• ACE inhibitor

Austin et al Am Heart J 2006151969
56
Intensive Intervention in Diabetes

• Behaviour mod
• Diet
• Exercise
• Smoking cessation
• A1C lt 6.5
• SBP lt 140 130
• DBP lt 80-85
• Cholesterol lt 4.5
• Triglycerides lt 1.7
• ACE inhibitor
• ASA

Achievement of Targets
TG lt 1.7
SBP lt130
DBP lt80
A1C lt6.5
Chol lt 4.5
Intensive Rx
Conventional Rx
STENO 2 Gaede et al NEJM 2003348383
57
Steno 2 Effect on CV Outcomes
HR0.47
Gaede P, et al. N Engl J Med. 2003348383-393.
58
Steno 2 Trial 13 year FU
Multifactorial Risk Management Compared to
Conventional Therapy Results in Reduction of
Total Mortality over 13 years Follow up
Conventional Therapy
60
50
End of Trial
40
30
Total Mortality
Intensive Therapy
20
10
HR 0.54 (0.32-0.88) p0.015
0 1 2 3 4 5 6 7 8
9 10 11 12 13
Years of follow-up
Gaede et al N Eng J Med 2008358580-91
59
Steno-2 13 year Follow-up

Intensive Conventional
NNT
Total Mortality
5
24 (30)
40 (50)
CV Mortality
8
9 (11)
19 (24)
CV Events
48 (60)
25 (31)
3-4
HR (95 CI)
0 0.2 0.4 0.6 0.8 1.0
1.2
Favours Intensive Strategy
Gaede et al N Eng J Med 2008358580-91
60
Conclusions

• The patient with diabetes has
• Higher risk of MI
• Worse outcome after MI
• Enhanced benefit from treatments
• Reduced opportunity for optimal Rx

An increasing problem in need of attention