Title: Before It Happens To You Choosing Between the Evidence and the Guidelines Jonathan SacknerBernstein,
1Before It Happens To YouChoosing Betweenthe
Evidence and the Guidelines Jonathan
Sackner-Bernstein, MDJanuary 2004
2Definition of Evidence-Based Medicine
- Application of scientific and clinical
information to the development of preventative,
diagnostic and therapeutic plans. - Sources of information should be hierarchical
- Randomized Clinical Trials
- Controlled mechanistic studies
- Prospective observational studies
- Epidemiology
- Case series
- Clinical experience
3Preventive Cardiology Standard of Care
- Lifestyle modifications
- BP lt 140/90
- lt 135/85 for HF or renal insufficiency
- lt 130/80 for DM
- LDL Cholesterol lt 100 (or lt 130, lt 160)
- HgA1c lt 7
- ASA (75 325 mg daily)
- ACE inhibitors (coronary or vascular disease)
- Beta-Blockers (post-MI and ACS patients)
Smith et al, Circulation 2001
4Lifestyle Management The Ornish Way
- Vegetarian diet, stress management and yoga in
patients with symptoms refractory to medical
therapy, using treatments from 1970s - Studies show
- 48 patients enrolled in randomized phase, less
than ¾ completed study - Completers showed significant regression of
coronary atherosclerosis - JAMA, 1998
5Therapeutic Lifestyle Change DASH Study
- Dietary Approach to Stop Hypertension was a
short-term trial evaluating the effects of a
prepared meal plan on hypertension. The diet can
be complex to implement, but it does help a
little with blood pressure control. - Initial Fruit Fruit Vegetables
- BP Vegetables ?fat, ?K (DASH)
- Hypertension 144 / 89 ? 7/3 ? 11/6
- Pre- 126 / 83 ? 1/0 ? 4/2
- Hypertension
- NEJM 1997
6Therapeutic Lifestyle Change PREMIER Study
- The Premier study showed modest blood pressure
benefit over 6 months, but no lifestyle
modification study has shown reduced risk of
clinical events. - Initial Behavioral Behavioral Modification
- BP Modification DASH
- Hypertension 144 / 88 ? 5/2 ? 6/4
- Pre- 130 / 83 ? 3/2 ? 3/2
- Hypertension
- JAMA 2003
7The Lifestyle Evidence
- Exercise, weight loss, stress reduction have
modest impact. - No clinical outcomes studies prove clinical
benefit of therapeutic lifestyle change (studies
focus on surrogate endpoints) - Quitting smoking has major impact, but is very
difficult.
8The Lifestyle Myth
- People cant change, so recommending lifestyle
will rarely do much good. - Even if someone can change, it has little effect.
9Evidence or Guidelines?
- How big are the risks?
- What are the optimal treatment targets?
- What are the optimal tools?
- Are the benefits class effects?
10Evidence or Guidelines?
11Calculating Risk from Framingham Scores
- 52 year old man without diabetes, no history of
hypertension, who does not smoke. - Lipids
- Cholesterol 224
- HDL 39
- LDL 148
12Calculating Risk from Framingham Scores
- Risk of MI within 1 year is 0.8.
- Since one year mortality post-MI is 25, his risk
of death within 1 year of MI is 0.2, 1 in 500.
http//hin.nhlbi.nih.gov/atpiii/evalData.asp
13Calculating Risk from Framingham Scores
- 56 year old woman, non-smoker, without diabetes,
but with borderline blood pressure of 142/80. - Lipids
- Cholesterol 232
- HDL 49
- LDL 153
14Calculating Risk from Framingham Scores
- Risk of MI within 1 year is 0.4.
- Since one year mortality post-MI is 25, her risk
of death within 1 year of MI is 0.1, 1 in 1,000.
http//hin.nhlbi.nih.gov/atpiii/evalData.asp
15The Risk of Cardiac Death is High
- Risk of MI Risk of Death
- within 1 yr within 1 year from MI
- 52 yo man 1 in 125 1 in 500
- 56 yo woman 1 in 250 1 in 1,000
- How big is this risk?
16The Risk of Cardiac Death is High
- Risk of death from MI within a year
- 52 yo man 1 in 500
- 56 yo woman 1 in 1,000
17The Risk of Cardiac Death is HighWhat Action is
Recommended?
- Risk of death from MI within a year
- 52 yo man 1 in 500
- 56 yo woman 1 in 1,000
- Compare to lifetime risks of death from
- an airplane crash 1 in 659,000
- bad medical care 1 in 83,000
- a car accident 1 in 6,500
- Recommendations?
-
-
- Walsh J, True Odds 1998, NCEP/ATP-3, JNC-7
18The Risk of Cardiac Death is HighWhat Action is
Recommended?
- Risk of death from MI within a year
- 52 yo man 1 in 500
- 56 yo woman 1 in 1,000
- Compare to lifetime risks of death from
- an airplane crash 1 in 659,000
- bad medical care 1 in 83,000
- a car accident 1 in 6,500
- Recommendations?
- 52 yo man aspirin lifestyle change
- 56 yo woman lifestyle change
- Walsh J, True Odds 1998, NCEP/ATP-3, JNC-7
19Classification of Heart Failure (AHA/ACC)
- Stage Description
- A at high risk- HTN, CAD, DM, FH (age gt50)
- B structural heart disease no signs or
symptoms - C current or prior symptoms of HF
- D marked symptoms of HF at rest despite
maximal therapy - Circ 2001
20Cardiac Risk Factors Standard Teaching
- Age
- Sex
- Family History
- Hypertension
- Hypercholesterolemia
- Smoking
- Diabetes
- Obesity (sedentary lifestyle)
21Stage C/D Heart Failure is a Bigger Riskthan
Colon Cancer
10,000
8,000
Incidence Per 100,000
6,000
4,000
2,000
0
Age (years)
Heart Failure (Stage C/D)
Colorectal Cancer
- Kannel, Am Ht J, 1991, NCI, SEER, 2001
22For Women, Coronary Disease (Stage A HF)Is a
Bigger Risk than Breast Cancer
12,000
Coronary Heart Disease
10,000
8,000
Prevalence (per 100,000)
6,000
4,000
Breast Cancer
2,000
0
30
40
50
60
70
Age (years)
National Cancer Institute and American Heart
Association
23Atherosclerosis Starts by Our 20s
- Aorta Right Coronary Artery
- Prevalence
- ()
- n 204
- PDAY Study
24By Now, We Have Coronary Disease
Prevalence of CAD Using IVUS ()
25Cardiac Risk Factors Practical Approach
- Scientifically Patients Easily
- Cant Change Wont Change Modifiable
- Your genes Smoking Blood pressure
- (family history) Sedentary lifestyle Cholesterol
- Gender Obesity/overweight Diabetes
- Age
26Risks of Cardiovascular Disease are Increased
Even If You Are Only a Little Overweight
20 30 Overweight
J Chronic Diseases, 1978
27Evidence or Guidelines?
- How big are the risks?
- What are the optimal treatment targets?
28What is Optimal Blood Pressure?
- JNC-7
- Optimal blood pressure
- Target is 115/75
- Target blood pressure
- Target is 135/90
- Target for diabetics and renal insufficiency
- Target is 13080
29Evidence for Blood Pressure Targets
- 1950s Risk lowest for BP lt 120/80
- 1970s Risk lowest for BP lt 118/77
- 2001 Risk lowest for BP lt 120/80
- 2003 Risk lowest for BP lt 115/75
30Higher Risk of MI, CVA or CV Death in
Prehypertension (above 120/80)
Men
Normal 120-130 Systolic
High Normal 130-139 Systolic
Normal 120-130 Systolic
High Normal 130-139 Systolic
Vasan, NEJM, 2001
31Structural Changes of the Heart Occur in White
Coat Hypertension
? Risk 270
Likelihood of Changes in Heart Structure
? Risk 90
People Whose Blood Pressure is High in the
Doctors Office
People with High Blood Pressure
Muscholl BMJ, 1998
32Goal Blood Pressure Should Be 115/75
- Cause Ages Death Relative Risk
- Stroke
- 40-49 414 0.36 (0.32-0.40)
- 50-59 1372 0.38 (0.35-.040)
- 60-69 2939 0.43 (0.41-0.45)
- 70-79 4327 0.50 (0.48-0.52)
- 80-89 2636 0.67 (0.63-0.71)
- IHD
- 40-49 1022 0.49 (0.49-0.53)
- 50-59 5594 0.50 (0.49-0.52)
- 60-69 10450 0.54 (0.53-0.55)
- 70-79 10852 0.60 (0.58-0.61)
- 80-89 5649 0.67 (0.64-0.70)
- Prospective Studies Collaboration, Lancet 2003
0.25 0.35 0.5 0.7 1.0
33Coronary Death is Less Likely When Blood Pressure
is Minimized
? 256 x ? 128 x ? 64 x ? 32 x ? 16 x ? 8 x ? 4
x ? 2 x 1 x
Age (years) 80-89 70-79 60-69 50-59 40-49
Risk of CHD Death
120 140 160 180 Systolic Blood Pressure
34What is Optimal Blood Pressure?
- JNC-7
- Optimal blood pressure
- Target is 115/75
- Target blood pressure
- Target is 135/90
- Target for diabetics and renal insufficiency
- Target is 13080
- Optimal
- 115/75 or
- at least 20/10 reduction
35What is Optimal LDL-Cholesterol?
- ATP-3/NCEP
- Coronary Heart Disease
- Target is 100 mg/dl
- 2 Risk Factors
- Target is 130 mg/dl
- 0-1 Risk Factors
- Target is 160 mg/dl
36Good LDL Levels Do Not Minimize Risk
3.0
2.5
Risk of Heart Attack, Bypass or Angioplasty
Women
2.0
1.5
Men
1.0
0.5
70
100
130
160
190
LDL Level
ARIC Study, 10 yr follow-up of 45-64 year olds
37Lower LDL Targets Minimize Risk
- Coronary Events at 1 Year
- Based on Level of LDL
- (4S trial)
25 20 15 10 5 0
Per Cent Coronary Events
77 96 116 135 154 174
LDL Level Measured
4S Investigators
38Aggressive Management Reverses Plaques
Effects of Pravastatin and Atorvastatin in the
REVERSAL Study
Pravastatin
Atorvastatin
39What is Optimal LDL-Cholesterol?
- ATP-3/NCEP
- Coronary Heart Disease
- Target is 100 mg/dl
- 2 Risk Factors
- Target is 130 mg/dl
- 0-1 Risk Factors
- Target is 160 mg/dl
- Optimal
- Target lt 100 mg/dl or
- High risk lt 80 mg/dl
40What is Optimal HgA1c Target?
41Diabetic Complications Relate to HgA1c Level
Triple
Risk of Complications Compared to risk if HgA1c lt
6
Double
5
6
7
8
9
10
Glycohemoglobin ()
Before It Happens To You
42What is Optimal HgA1c Target?
43Why Do We Follow the Standard Targets?
- We have been duped about TLC impact.
- Cost-effectiveness is important for society
(statin example gt 500 per year is too much) - Risk-benefit analysis does not accurately reflect
the safety of the available medicines (statins,
ACE inhibitors, beta-blockers and aspirin) - We trust the committees too much.
44Evidence or Guidelines?
- How big are the risks?
- What are the optimal treatment targets?
- What are the optimal tools?
45ACE Inhibition Reduces Risk
Effect of Ramipril in HOPE
25
20
with event
15
10
5
0
N Engl J Med, 2000
46ACE Inhibitors Reduce Risk in Patients with
Normal EF
Effect of Ramipril in HOPE
25
Risk ? 27 p lt 0.0001
Risk ? 25 p lt 0.001
20
with event
15
Risk ? 30 p 0.003
Risk ? 33 p 0.01
10
5
0
MI/Stroke/ CV Death
CV Death
MI
Stroke
G. Dagenais, ESC 1999
47ACE Inhibition Reduces Risk in CAD
Placebo
Perindopril
48Beta-Blockers Reduce Risk of Sudden Death
10
Risk ? 26
Risk ? 40
Risk Sudden Death
Risk ? 28
Risk ? 56
5
Risk ? 30
0
US Carvedilol
Merit-HF
MAPPHY (HTN)
Copernicus
BHAT
Placebo Beta-blocker
- Am J Hypertension 1991, JAMA 1993,NEJM 1996,
Lancet, 1999
49Effect of Candesartan in Heart Failure
Cause-Specific Mortality
All-cause Mortality
30
30
CV Risk ? 13 p 0.006
Risk ? 10 p 0.032
CV
20
Cumulative Incidence ()
20
10
10
Non-CV
0
0
0
1
2
3
4
0
1
2
3
4
Time (years)
Time (years)
50Safety of Simvastatin in HPS
- Parameter Simvastatin Placebo
- n 10,269 10,267
- Liver Function
- 24x ULN 139 (1.35) 131 (1.28)
- gt4x ULN 43 (0.42) 32 (0.31)
- Elevated CK
- 410x ULN 19 (0.19) 13 (0.13)
- gt10x ULN 11 (0.11) 6 (0.06)
- Myopathy
- No rhabdomyolysis 5 (0.05) 1 (0.01)
- Rhabdomyolysis 5 (0.05) 3 (0.03)
- Lancet 2002
51Statins Reduce Risk of MI, CVA and Death
Independent of Baseline LDL Level
LDL Level Before Starting on Statin Therapy
Before It Happens To You
52Statins Reduce the Risk of Stroke
Before It Happens To You
53LDL Reduction Markedly Reduces Risk
Initial LDL 160
160
130
100
130
100
80
80
Target LDL
Women (45-64 years old)
Men (45-64 years old)
Before It Happens To You, ARIC
54Its Too Late for Primary Prevention
- By mid-life, we all have lipid-laden atheromas in
our arterial walls. - 100,000,000 American adults have LDL gt 100
- 150,000,000 American adults have BP gt 120/80
- Preventative cardiology should focus on reversing
functional and structural abnormalities.
55Evidence or Guidelines?
- How big are the risks?
- What are the optimal treatment targets?
- What are the optimal tools?
- Are the benefits class effects?
56Are Drugs Within a Class Interchangeable?
- Benazepril Placebo p
- n 300 283
- Death 8 1 0.04
- Sudden Death 3 1
- MI 3 0
- Pulmonary Embolus 1 0
- Variceal Bleed 1 0
Benazepril Placebo p n 300 283 Death 8 1 0.04 Sud
den Death 3 1 MI 3 0 Pulmonary Embolus 1 0 Varicea
l Bleed 1 0
AIPRI Study, NEJM 1996
57Are Drugs Within a Class Interchangeable?
COMET Trial
- ACE inhibitors? Perhaps not.
- Beta-blockers in HF?
40
Risk ? 17 p0.0017
30
20
prolonged life 1.4 yrs
10
0
0
1
2
3
4
5
Time (years)
Metoprolol
Carvedilol
58Are Drugs Within a Class Interchangeable?
Incidence New Diabetes (COMET Trial)
- ACE inhibitors? Perhaps not.
- Beta-blockers in HF?
Time (years)
59Are Drugs Within a Class Interchangeable?
- Timolol Propranolol Metoprolol
- Norwegian BHAT LIT
- (n1884) (n3837) (n2395)
- ACE inhibitors? Perhaps not.
- Beta-blockers in HF? No.
- Beta-blockers post-MI?
p lt 0 .005
p NS
p lt 0 .0005
Cumulative Mortality Rate
0
24
36
12
0
24
36
12
0
6
18
12
Time (months)
60Are Drugs Within a Class Interchangeable?
- ACE inhibitors? Probably not.
- Beta-blockers in HF? No.
- Beta-blockers post-MI? No.
- Statins for dyslipidemia?
- Lovastatin ? events
- Fluvastatin ? restenosis
- Simvastatin ? events
- Pravastatin ? events
- Atorvastatin ? events
- Cerivastatin ? risks
- Rosuvastatin ? LDL, risks?
61Are Drugs Within a Class Interchangeable?
- ACE inhibitors? Probably not.
- Beta-blockers in HF? No.
- Beta-blockers post-MI? No.
- Statins for dyslipidemia? No.
62The Challenge
- Consider what the patient expects and wants
- Reiteration of our standard approaches?
- Data-based evolution of treatment?
- Optimized care, based on risks and benefits for
the individual.
63Targets and Tools to Deliver Optimal Care
- Targets
- Blood pressure of 115/75
- LDL of 100 (or lt 80 in high risk)
- HDL of 40 (men) or 50 (women)
- HgA1C of 6 (unless hypoglycemia)
- Tools
- ACE inhibitors, beta-blockers and diuretics
- Statins
- Aspirin
64Before It Happens To You
- Scientifically based strategies to minimize the
risk of a heart attack or stroke. This book will
save lives. - Valentin Fuster, MD, PhD
- An elegant and practical game plan for real
health challenges. Education without admonition
for both patients and doctors. - Mehmet Oz, MD
- A unique book that emphasizes maximum risk
reduction, even for those who, like me, have
trouble making a consistent effort at diet and
exercise. - Gervasio Lamas, MD