Best Diet for CHD Prevention

Dr. Thomas G. Allison

Mayo Clinic Rochester

Fatty Streaks in Aorta of 19-Year Old Male

Advanced Lesion with Large Lipid Core

Plaque Rupture with Torn Cap

Major Statin Trials

50 210

LDL-C (mg/dL)

70 190170150130110900

5

10

15

20

25

% w

ith

CA

D e

ven

t

WOSCOPS

AFCAPS

CARE

4S

LIPID

HPS

PROSPER

Secondary

Mixed

Primary

ASCOT-LLA

PROVE IT

TNT

JUPITER

SIMVASTATIN: CAUSE-SPECIFIC MORTALITY

Risk ratio and 95% CISTATIN PLACEBOCause ofdeath (10269) (10267) STATIN better STATIN worse

CHD 577 701

Other vascular 214 242

ALL VASCULAR 791 943(7.7%) (9.2%)

17% SE 4.4reduction(2P<0.0002)

Neoplastic 352 337

Respiratory 93 111

Other medical 76 91

Non-medical 16 21

ALL NON-VASCULAR 537 560(5.2%) (5.5%)

5% SE 5.9reduction

ALL CAUSES 1328 1503(12.9%) (14.6%)

12% SE 3.5reduction(2P<0.001)

0.4 0.6 0.8 1.0 1.2 1.4

Nissen, S. E. et al. JAMA 2004;291:1071-1080.

Intravascular Ultrasound Images at Baseline and Follow-up

REVERSAL TrialREVERSAL Trial

Limitations to Pharmacologic Lipid Management

• Cost of treatment– Not an issue if generic drug will control LDL-C– Treatment cost ~ $1000 per year if non-generic

agent needed

• Not all patients tolerant of statins– Myalgia most common complaint (5-15%)– Alternative drugs (intestinal agents, niacin, fibrates)

have limited effect on LDL-C, limited outcome data

• Benefits of add-on drug therapy not established

International Comparisons

44

54

57

87

100

202

420

638

0 100 200 300 400 500 600 700

Korea

China - Rural

Japan

France

China - Urban

USA

Hungary

Russia

CHD Death Rates/100,000(Men ages 35-74)

International rates not due to differences in statin therapy rates!

2002 AHA Heart and Stroke Statistical Update

Diets and CAD: What’s the Evidence?

• Dietary therapy can be an alternative to pharmacologic management of lipids in primary prevention

• Important adjunctive therapy in secondary prevention

• What is the best diet for CHD prevention?

East Finland

Mortality from Coronary Heart DiseaseMen 35-64 Years (1969-1994)

Puska P: Cardiovasc Risk factors 6:203-10, 1996Puska P: Cardiovasc Risk factors 6:203-10, 1996

Cardiac death ratesCardiac death rateshave dropped by 75%!have dropped by 75%!

NorthKarelia

AllFinland

Per100,000

800

700

600

500

400

300

200

100

CP999299-39

Now 80%

Trends in Women’s Lifestyles1980-82 versus 1992-94

• 31% decline in CHD incidence across all ages– 41% decrease in smoking (27% 16%)– Diet changes

• 31% decrease in trans fatty acid intake• 69% increase in P/S ratio• 90% increase in cereal fiber• 180% increase in -3 fatty acids• 12% increase in folate

Nurses’ Health Study -- Hu et al: NEJM 2000;343:530-537

Trends in Women’s Lifestyles1980-82 versus 1992-94

• 38% increase in overweight (BMI>25)– average BMI 24.5 26.1 kg/m2

• 22% increase in glycemic load

Regional Diets with Low CHD Rates

• Seventh Day Adventist

• Japanese

• Rural Chinese

• Eskimo

• Mediterranean

Crete

Adherence to Mediterranean Diet and Survival in a Greek Population

• Prospective, population-based investigation of CHD mortality versus diet

• 22,043 healthy adults in Greece• 44-month follow-up• Diet assessed by 10-point scale (0-9)

– vegetables, legumes, fruits and nuts, cereals, fish, alcohol, monounsaturated/saturated fat ratio (+)

– meat, poultry, dairy products (-)Trichopoulou A et al, NEJM 2003:348:2599-2608

Results• Two single nutrients predicted CHD death

– Fruits and nuts: +200 g/day = 18% reduction– Monounsaturated/saturated fat ratio:

+0.5 = 14% reduction

• 2-point increase in Mediterranean diet score– 25% reduction in total mortality– 33% reduction in CHD mortality– 24% reduction in cancer mortality

• Adjusted for age, sex, WHR, energy expenditure, smoking, BMI, potato and egg consumption, and total caloric intake

Epidemiologic Studies

• Inherently flawed• Problems with ascertainment of both

independent (diets) and dependent (mortality, heart attacks, etc.) variables

• Not all non-dietary variables can be measured (and none controlled)

• Assumes constancy of exposure to dietary factors

Diet-Heart Studies with Outcomes

Location N Year f/u

England (Rose) 80 1965 2y

Middlesex 264 1965 5y

Oslo 412 1966 5y

London 393 1968 5y

Sydney 458 1978 5y

DART 2033 1989 2y

Moradabad 505 1992 1y

LHT (invite) 48 1998 5y

Leon 423 1999 4y

Intervention

control v corn oil v olive oil

control v low fat

control v low fat + PUF

control v soya-bean oil

control v low fat + PUF

low fat v fish v fiber

low fat v fruit/veg+fish+fiber

control v ultra-low fat

control v Mediterranean

-25% -20% -15% -10% -5% 0% 5% 10% 15% 20% 25% 30%

Risk Change

Cholesterol Change

Rose - olive oil (MCE)

Rose -- corn oil (MCE)Middlesex (MCE)

Oslo (MCE) p < .05

London (MCE)

Sydney (death) p < .05

DART - low fat (death)

DART - fish (death) p < .05

Dart - fiber (death)

LHT (MCE)

LHT (MCE + revascularization) p < .001

Moradabad (MCE) p < .01

Moradabad (death) p < .01

Leon 1994 (MCE) p < .001

Leon 1999 (MCE) p < .001

n = 2033

n = 458

n = 2033

n = 2033

n = 393

n = 423

n = 605

n = 505

n = 505

n = 48

n = 48

n = 412

n = 264

n = 52

n = 54

Lifestyle Heart Trial

• Randomized invitational design (recruitment in ~1987)

• 28 experimental patients, 20 usual care

• Intervention:– vegetarian, low fat diet (10% fat, 5 mg

cholesterol/day) – smoking cessation, moderate exercise, stress

managementOrnish et al: Lancet1990;336:129-133

Original Dean Ornish Plan

Fats (<10%)Fats (<10%)

Nonfat dairy products – yogurt, cheese, egg whitesNonfat dairy products – yogurt, cheese, egg whites

Nonfat products – cereal, soups, tofu, crackers, egg beatersNonfat products – cereal, soups, tofu, crackers, egg beaters

Whole grain – corn, rice, oats, wheat, etcWhole grain – corn, rice, oats, wheat, etc

Beans and legumesBeans and legumes

FruitsFruits

VegetablesVegetables

Ban

All oilsAll meatsOlivesAvocadosNuts – seedsHigh or low fat productsSugar – syrup – honeyAlcohol

Ban

All oilsAll meatsOlivesAvocadosNuts – seedsHigh or low fat productsSugar – syrup – honeyAlcohol

CP1095424-1

Moderate exerciseStress reductionSmoking cessation

No calorie restriction

Lifestyle Heart Trial 1-Year Results

Variable Experimental Control p <

LDL mg/dl 95 ± 60 157 ± 45 .0072

HDL mg/dl 37 ± 15 51 ± 15 ns

Progression 18% 53%

Regression 82% 42% stenosis -2.2% +3.4% .001

Not powered (or randomized) for clinical events

Lyon Heart Study

• 423 patients randomized post-MI 1988-92

• Mediterranean diet vs “prudent diet” (Step 1) prescribed by patients’ physicians

• Planned 5-year follow-up

• Study terminated early (4 years) due to favorable interim analysis -- final report on 423 patients

de Lorgeril et al, Circ 1999;99:779-785

CP1059685-22

The Traditional Healthy

Mediterranean Diet Pyramid

The Traditional Healthy

Mediterranean Diet Pyramid

Meat

Sweets

Eggs

Poultry

Fish

Cheese & yogurt

Olive oil

Fruits

Bread, pasta, rice, couscous, polenta,other whole grains & potatoes

Daily physical activity

Daily beverageDaily beveragerecommendationsrecommendations

Daily beverageDaily beveragerecommendationsrecommendations

6 glasses6 glassesof waterof water

6 glasses6 glassesof waterof water

Wine inWine inmoderationmoderation

Wine inWine inmoderationmoderation

VegetablesBeans,legumes& nuts

MonthlyMonthlyMonthlyMonthly

WeeklyWeeklyWeeklyWeekly

DailyDailyDailyDaily

2000 Oldways Preservation & Exchange Trust2000 Oldways Preservation & Exchange Trust2000 Oldways Preservation & Exchange Trust2000 Oldways Preservation & Exchange Trust

Lyon Heart Study - Lipids

Control Experimental

Cholesterol 239 ± 40 239 ± 41

HDL 49 ± 13 50 ± 13

LDL 163 ± 38 161 ± 36

Triglycerides 154 ± 73 171 ± 75

Lyon Heart Study

020406080

100120140160180200

Primary Primary +Secondary

All Endpoints

Control

Experimental

p<.0001

p<.0001

p<.0002

Results consistent with DART and Moradabad trials

Search for the Perfect CHD Prevention Diet

• The Lifestyle Heart Trial achieved marked LDL-C lowering, but adversely affected HDL-C

• The Leon Heart Study lowered CHD risk without affecting lipid levels

• Can we design a diet that lowers LDL-C without lowering HDL-C while providing the heart protective nutrients?

Therapeutic Lifestyle Changes in LDL-Lowering Therapy

Major Features

• TLC Diet (Step 2+)– Reduced intake of cholesterol-raising nutrients (same as

previous Step II Diet)• Saturated fats <7% of total calories• Dietary cholesterol <200 mg per day

– LDL-lowering therapeutic options• Plant stanols/sterols (2 g per day)• Viscous (soluble) fiber (10–25 g per day)

• Weight reduction • Increased physical activity

NCEP

Other Features of TLC Diet

Nutrient Recommended Intake• Polyunsaturated fat Up to 10% of total calories• Monounsaturated fat Up to 20% of total calories• Total fat 25–35% of total calories• Carbohydrate 50–60% of total calories• Fiber 20–30 grams per day• Protein Approximately 15% of total calories• Total calories (energy) Balance energy intake and expenditure

to maintain desirable body weight/prevent weight gain

Dietary Portfolio

• 46 healthy, hyperlipidemic adults randomized – Low saturated fat diet– Low saturated fat diet + Lovastatin 20 mg/day– Diet portfolio (based on Step 2+)

• Phytosterols 1.0 g/1000 kcal

• Soy protein 21.4 g/1000 kcal

• Viscous fiber 9.8 g/1000 kcal

• Almonds 14 g/1000 kcal

• 4-week follow-upJenkins DJA et al, JAMA 2003:290:502-510

Results

Summary:Best CHD Prevention Diet

• Low in saturated fat and cholesterol

• High in monounsaturated fat

• Fish 2+ servings per week– Or omega-3 fatty acids supplement

• Fresh fruits and vegetables 7+ servings/day

• Whole grains in place of refined flour and sugar

Best CHD Prevention Diet

• Nuts 14+ grams/1000 kcal• Added soy protein, soluble fiber, phytosterols• Low glycemic index, especially if overweight• Calorie control should be automatic

– Low caloric density CHO’s– Satiety from monounsaturated fats, proteins

• Highly palatable– Variety of foods and seasonings

BMJ  2004;329:1447-1450 (18 December), doi:10.1136/bmj.329.7480.1447

Oscar H Franco, scientific researcher1, Luc Bonneux, senior researcher2, Chris de Laet, senior researcher1, Anna Peeters, senior researcher3, Ewout W Steyerberg, associate professor1, Johan P Mackenbach, professor1 1 Department of Public Health, Erasmus MC University Medical Centre Rotterdam, PO Box 1738, 3000 DR Rotterdam, Netherlands, 2 Belgian Health Care Knowledge Centre (KCE), Wetstraat 155, B-1040, Brussels, Belgium, 3 Department of Epidemiology and Preventive Medicine, Monash University Central and Eastern Clinical School, Melbourne, Australia

The Polymeal: a more natural, safer, and probably tastier (than the Polypill)strategy to reduce cardiovascular disease by more than 75%

The limits of medicine

IngredientsPercentage reduction (95% CI) in risk of CVD Source

Wine (150 ml/day) 32 (23 to 41) Di Castelnuovo et al (MA)6

Fish (114 g four times/week) 14 (8 to 19) Whelton et al (MA)7

Dark chocolate (100 g/day) 21 (14 to 27) Taubert et al (RCT)8

Fruit and vegetables (400 g/day) 21 (14 to 27) John et al (RCT)10

Garlic (2.7 g/day) 25 (21 to 27) Ackermann et al (MA)11

Almonds (68 g/day) 12.5 (10.5 to 13.5)Jenkins et al (RCT),15 Sabate et al (RCT)16

Combined effect 76 (63 to 84)

Other Aspects of Polymeal• Men at age 50 would live an average of 6.6 years

longer• Women at age 50, 4.8 years longer• Cost of polymeal estimated at $28.10/week• Addition of other components such as oat bran or

olive oil would only enhance effect• No obvious contraindications to combining

polymeal with polypill (or any subset of components)

Weight Loss Controversy

• Americans have substituted refined CHO’s for fats over the past 20 years– Linked to obesity

• Low CHO versus low fat for weight loss– Atkins versus Ornish

• Much speculation, many popular books• Published data only in past 4-5 years• Does losing weight necessarily mean lowering

CHD risk?

Effect of Varying Fat, Protein, and CHO Content on Weight Loss

• 811 overweight adults randomized to 3 weight loss diets for 2 years

• Varying content: fat protein CHO– Diet 1 20% 15% 65%– Diet 2 20% 25% 55%– Diet 3 40% 15% 45%– Diet 4 40% 25% 35%

• 750 kcal per day caloric deficitSacks FM et al. NEJM 2009;360:859-873

Bon Appetit!

• Comments?

• Questions?