Preventing CAD in Diabetes by Trevor Orchard

1
Preventing CAD in Diabetesby Trevor Orchard

• Definition of Diabetes
• Magnitude of the Risk
• Reasons for the Risk
• Evidence for Preventive Interventions
• Clinical Recommendations

2
M.D. SURVEY DxDIABETES
Frequency of OGTT Use 1 2
2 19 3 16 No mention 68
3
Current Diagnostic Criteria for Diabetes (plasma
glucose mg/dl) WHO/ADA
1979 NDDG/1980 WHO 1997 ADA Fasting ?
140 ? 126 2 hr1 ? 200 ?
2002 Random glucose3 ? 200 ? 200 1Post
75 gm glucose load, Midtest value also has to be
gt 200 mg/dl for NDDG. 2Not recommended for
routine use. 3In the presence of diabetes
symptoms.
4
Prevalence of Diabetes USA40-74 Years Old
1997 ADA Criteria 1985 WHO
Criteria Millions
Millions Undiagnosed 4.4 4.1 6.4 6.0 IFG/IGT 1
0.1 9.6 15.6 14.9 Diagnosed 7.9 7.5 7.9 7.5 T
otal Diabetes 12.3 11.6 14.3 13.5
Harris MI, et al. Diabetes Care 1997 20(1)
1859-1862.
5
CHS Study ADA v WHO
3984 aged 65 yrs followed 5-9 yrs (no known
diabetes/CVD). Adjusted RR compared to common
normal2 for CVD events.
WHO ADA Fasting Criteria n
n Normal 184 1.09 (0.73-1.65)
1142 1.20 (0.99-1.47) IGT or IFG 1264 1.23
(1.01-1.98) 582 1.39 (1.09-1.77) New
Diabetes 563 1.56 (1.23-1.98) 287
1.58 (1.17-2.13) 2FG lt 6.1, 2 hrs lt 7.8
mmol/L. Adjusted for gender, age, ethnicity,
smoking, BMI, LDLc and HT.
Barzilay JL. Lancet 1999 354 622-625.
6
Metaregression Analysis Glucose v CVD Incidence
20 studies, 95,783 people (94 men) followed 12
yrs. (Studies excluded if purely diabetic). RR
(95 CI) FPG ? 110 mg/dl 1.33 (1.06
1.67) 2 hr G ? 140 mg/dl 1.58 (1.19
2.10) Exclude top groupings. FPG p0.056, 2
hr p0.0006
Coutinho, M. Diabetes Care 1999 22 233-240.
7
DeCode Study
22,476 aged 30-89 yrs non-diabetic, 11
cohorts. Followed mean 12 yrs for mortality,
262,811 person years. Adjusted RR of fasting
glucose 2 hr glucose Total 1.10
(1.07-1.13) 1.17 (1.14-1.21) CVD 1.08
(1.03-1.13) 1.15 (1.10-1.20) Non-CVD 1.10
(1.06-1.14) 1.16 (1.12-1.20) Adjusted for age,
gender, center, BP, chol, smoking and BMI. If RR
of fasting glucose adjusted for 2 hr 1.00, 0.99,
1.00, vice versa 1.07, 1.07, 1.07.
Personal Communication. IDF/EDEG, Acapulco, Nov.
2000.
8
Mortality in People with DiabetesCauses of Death
of Deaths
Ischemicheartdisease
Otherheartdisease
Diabetes
Infection
Cancer
Stroke
Other
Geiss LS et al. In Diabetes in America. 2nd ed.
1995 chap 11.
9
(No Transcript)
10
Relative Risks of Cardiovascular EventsDiabetes
v Nondiabetes. Framingham45-74 year old
Any CVD
CVD Death
Cardiac Failure
Brain Infarct
CHD
Int Claud
Unadj.
Unadj.
Adjust.
Adjust.
Kannel, Diabetes Care 1979 2120-126.
11
CHS Study
Diabetes status and presence of
subclinical/clinical CVD at baseline and
incidence of specific events among men and women
in the CHS (outcome death).
Kuller LH. ATVB 2000 20 823-829.
12
CHS Study
Diabetes status and presence of
subclinical/clinical CVD at baseline and
incidence of specific events among men and women
in the CHS.
Kuller LH. ATVB 2000 20 823-829.
13
Effect of Diabetes on 30-Day SurvivalAfter MI
GUSTO-I
14
Cardiovascular Mortality in People with Diabetes
WOMEN
MEN
28 d 1 y Hospitalization 28 d Out of
Hospital
9.1
11.1
4.2
15.4
of Deaths (Crude Rate)
9.6
2.8
22.7
9.0
28.6
22.1
11.9
10.9
No Diabetes
Diabetes
Diabetes
No Diabetes
Adapted from Miettinen H et al. Diabetes Care.
19982169-75.
15
(No Transcript)
16
Trends in Mortality Rates for Ischemic Heart
Disease in NHANES Subjects with and without
Diabetes
Nondiabetes
Diabetes
17.0
Men, cohort 1 Men, cohort 2 Women, cohort 1
Women, cohort 2
14.2
Rate per 1000 person-years
7.6
7.4
6.8
4.2
2.4
1.9
-16.6
10.7
-43.8
-20.4
(P0.46)
(Plt0.001)
(P0.12)
(P0.76)
Defined in 1971-1975, followed up through
1982-1984.Defined in 1982-1984, followed up
through 1992-1993. Gu K et al. JAMA
19992811291-1297.
17
AGE-ADJUSTED RATES OF NONFATAL MI AND FATAL CHD
COMBINED PER 100,000 PERSON-YEARS
452
262
Rate of CHD
133
37
High Cholesterol
Diabetes in Women, Manson et al. Arch Intern
Med, 1991 151 1144.
18
(No Transcript)
19
Glycemia in Diabetes and Heart Disease

• Epidemiological Evidence Type 2
  Type 1
• Clinical Trial Evidence Type 2 Type
  1
• A potential explanation to the paradox Clinical
  evidence Pathology evidence
• Potential explanations for the increased heart
  disease risk in diabetes

20
Hazard Ratio (HR) and 95 Conference Interval
(CI) for Mortality due to Specific Causes for a
1 Increase in Glycosylated Hemoglobin After
Controlling for Other Risk Factors in
Younger-Onset Diabetic Persons
Underlying Cause Any Mention Cause of
Death HR 95 CI HR 95 CI Diabetes
1.25 (1.13-1.38) 1.18 (1.10-1.28) Ischemic
heart disease 1.18 (1.00-1.40) 1.17
(1.03-1.33) Other heart disease . . .
. . . 1.18 (1.06-1.31) Renal disease .
. . . . . 1.07 (0.92-1.25) All causes
1.12 (1.04-1.21) . . . . . .
Moss SE. Arch Intern Med 1994 154 2473-2479.
21
The 14-Year Cumulative Incidence of Amputation
for a Specified Increment in Baseline
Characteristics in Multivariate Logistic
Regression WESDR
Characteristic Increment
P OR (95
CI) Younger-onset Age (years) 10
lt0.0001 1.71 (1.30-2.24) Sex Male
lt0.0001 5.21 (2.50-10.88) Glycosylated
hemoglobin () 1 lt0.0001 1.39
(1.22-1.59) Diastolic blood pressure (mmHg)
10 lt0.005 1.58 (1.20-2.07) History
of ulcers Present lt0.0005 3.19
(1.71-5.95) Retinopathy One step
lt0.0001 1.16 (1.08-1.24)
Moss SE. Diabetes Care 1999 22 951-959.
22
Baseline Risk Factor Levels for CAD in Both
Sexes, by First Event, EDC 10 year Follow-up
No Angina Hard Total Variable
CAD Pectoris CAD CAD N 495 49
42 108 Sex ( Male) 50.1 49.0 61.9 51.9
Age (yrs) 2.597.3 33.4 6.2 32.9
6.6 33.0 6.8 Duration (yrs) 17.6
6.9 25.1 6.5 25.4 6.4 24.9
6.9 HbA1 () 10.4 1.8
9.9 1.9 10.7 1.8 10.3 1.8 Fibrinogen
(mg/dl) 280.1 87.1 305.8 77.9 343.3
97.2 319.6 89.5
23
Baseline Risk Factor Levels for CAD in Both
Sexes, by First Event, EDC 10 year Follow-up
(Cont.)
No Angina Hard
Total Variable CAD
Pectoris CAD
CAD WBC x 103/mm2 6.4 1.8 7.1
2.2 8.1 2.4 7.5 2.3 Triglycerides
(mg/dl) 99.8 82.7 113.4 67.6 156.5
80.1 134.4 90.9 Non-HDLc (mg/dl)
130.7 38.3 151.0 42.0 174.7 48.5 159.2
48.8 LDLc (mg/dl) 111.0 30.8 125.3
32.3 147.0 44.0 132.4 41.8 HDLc
(mg/dl) 54.8 12.2 50.9 13.0 48.3
9.8 50.0 11.8 ApoA1/HDLc 2.6
0.5 2.8 0.6 2.9 0.5 2.9 0.5
Values are given as mean SD or prevalence ().
Mann-Whitney. ?Fishers exact
Log-transformed before t-test Comparisons with
no CAD plt0.05 plt0.01 plt0.001
24
Baseline Risk Factor Levels for CAD in Both
Sexes, by First Event, EDC 10 year Follow-up
(Cont.)
No
Angina Hard Total Variable
CAD Pectoris
CAD CAD Serum Creatinine (mg/dl)
0.96 0.9 1.03 0.5 1.6 1.6
1.3 1.2 Log median AER (µg/min)
3.2 1.8 4.2 2.1 5.9 2.2
4.8 2.3 SBP (mm Hg) 111.1
13.2 118.5 14.1 127.5 21.1 121.3
18.5 QTc 407.1 30.0
414.1 25.9 412.5 29.6 414.1
26.5 Physical Activity 2790.9 2999.8
1779.2 2176.4 1917.4 1766.7 916.9
2053.6 WHR
0.82 0.07 0.84 0.08 0.86
0.07 0.85 0.07 eGDR (mg/kg/min)
8.1 1.8 7.3 2.0
6.4 1.9 7.0 2.0 Beck Depression
Inventory 6.8 6.2 9 .7 7.1
7.7 5.7 8.1 6.5
Values are given as mean SD or prevalence ().
Mann-Whitney. ?Fishers exact Log-transformed
before t-test Comparisons with no CAD plt0.05
plt0.01 plt0.001
25
Baseline Risk Factor Levels for CAD in Both
Sexes, by First Event, EDC 10 year Follow-up
(Cont.)

No
Angina Hard Total Variable CAD
Pectoris CAD CAD Smoke Ever () 32.8
50.0 59.5 54.7 Hypertension () 9.9
34.7 42.9 34.3 DSP () 20.3
61.2 50.0 52.8 E/I lt 1.10 () 12.9
32.6 47.1 37.9 Overt Nephropathy
() 17.2 38.8 69.0 48.1 MA or ON
() 38.8 69.4 85.7 71.3 ABI lt 0.8
or ABD 75 6.4 14.3 26.8? 19.6
eGDRlt6.22 (mg/kg/min)() 14.1
22.4 56.1 34.9
26
EDC 6 Yr Follow-up Multivariate Analysis (Cox
Proportional Hazards)
CHD
LEAD
Men Women Men
Women Duration 0.002 Duration 0.000
Duration 0.004 LDLc 0.02 HDLc 0.009 WH
R 0.001 HbA1 0.000 WHR
0.04 WBC 0.008 BDI 0.040 Smoking
0.03 Fibrinogen 0.092 Hypertension
0.000 Hypertension 0.016 Nephropathy (0.000)
replaces WBC/Fibrinogen/Hypertension and
improves model. Nephropathy doesnt enter
model.
27
Multivariate Models of CVD in EDC and Eurodiab
Prevalence Analyses of Comparable Populations


Standardized
Coefficient Coefficient P
value Males Eurodiab Age 0.071
0.36 0.007 HDL Cholesterol -1.867
-0.38 0.008 EDC Triglycerides 0.40
0.23 0.02 Hypertension 2.163
0.49 0.0001 Females Eurodiab Age 0.043
0.21 0.008 HbA1c -0.288
-0.29 0.008 Hypertension 0.734
0.16 0.032 EDC Age 0.079
0.32 0.01 HbA1 0.266
0.27 0.03 Macroalbuminuria 1.289
0.31 0.006
Int J. of Epidemiology 1998.
28
(No Transcript)
29
Stepwise selection of risk factors, adjusted for
age and sex, in 2693 white patient with Type 2
diabetes mellitus time to first event case model
Non-fatal or fatal MI (n192) Position in
model Variable P value First LDLc 0.0022
Second DBP 0.0074 Third Smoking 0.025 Fo
urth HDLc 0.026 Fifth Haemoglobin
A1c 0.053
UKPDS. BMJ 1998 316 823-828.
30
DOES IMPROVED GLYCEMIC CONTROL REDUCE CVD RISK
IN DIABETES?

• UGDP
• DIS
• KUMAMOTO
• DCCT
• VA FEASIBILITY
• UKPDS

31
GLUCOSE LOWERING AND CARDIOVASCULAR RISK IN
DIABETES
Study Intervention Result UGDP Tolbutamide
Possible increased

cardiovascular risk
Phenformin
Increased lactoacidosis Insulin variable
No benefit Insulin standard
No benefit DCCT/ Intensive(insulin)
Possible decrease in EDIC
glycemic therapy macrovascular
events in type 1 diabetes
(largely lower extremity

arterial disease )
No effect on ankle-brachial index
small
effect on carotid IMT
32
EPIC - Norfolk
4,662 men, 45-79 years (18 of total cohort).
Followed approximately 4 yrs for mortality
(41/131 due to IHD). Adjusted RR of 1
difference in HbA1c for IHD mortality1.31
(1.02-1.67) p0.03 Non CVD mortality1.20
(1.01-1.44) p0.04 Total mortality1.46
(1.00-2.12) p0.05 (excluding diabetes and
h/o CVD) HbA1c replaces diabetes in multivariate
models. Adjusted for age, SBP, TC, BMI, Cigs,
h/o CVD. Khaw KT. BMJ 2001
15-68.
33
The Paradox
Diabetes carries a greatly increased risk of
heart disease that is not explained by
traditional risk factors Type 1 - 5 fold
Type 2 - 2-4 fold.
BUT Hyperglycemia, the hallmark of diabetes,
is only weakly (at best) related to CHD.