brand logo

Am Fam Physician. 2009;79(7):571-578

A more recent article on cardiovascular disease prevention is available.

See related editorial on page 541.

Patient information: See related handout on a heart-healthy diet, written by the authors of this article.

Author disclosure: Nothing to disclose.

Patients often initiate commercial dietary plans to reduce obesity and prevent cardiovascular disease. Such plans include very low-carbohydrate, low-carbohydrate, very low-fat, and Mediterranean diets. Published evidence on several popular diets has made it easier for physicians to counsel patients about the health benefits and risks of such plans. Although the Atkins, Zone, Sugar Busters!, and South Beach diets have data proving that they are effective for weight loss and do not increase deleterious disease-oriented outcomes, they have little evidence of patient-oriented benefits. In contrast, the Mediterranean diet has extensive patient-oriented outcome data showing a significant risk reduction in mortality rates and in rates of fatal and nonfatal myocardial infarction. The American Heart Association released guidelines in 2006 that integrate recommendations from a variety of diets into a single plan. Physicians should emphasize diets that are rich in fruits, vegetables, and healthful fatty acids and that limit saturated fat intake. A stepwise individualized patient approach, with incorporation of one or two dietary interventions every three to six months, may be a practical way to help reduce a patient's cardiovascular disease risk.

Increases in obesity rates and in the prevalence of obesity-related illness have made dietary counseling an integral part of medical care.1 Patients often use commercial dietary plans to reach their weight-loss goals. Although some of these diets make unfounded claims, many of the more publicized plans have been studied in randomized, controlled settings. Focused counseling on the risks and benefits of these diets may be beneficial in helping patients reach their goals.

Clinical recommendationEvidence ratingReferences
Very low-carbohydrate diets are more effective than low-fat diets for short-term weight loss (six months or less) and do not cause adverse lipid profile changes.B211
A very low-fat diet, in combination with smoking cessation, exercise, and meditation, may improve cardiac perfusion and decrease angina symptoms.B9, 11, 1622
The Mediterranean diet confers morbidity and mortality benefits in patients with known cardiovascular disease.A2331
The American Heart Association 2006 diet and lifestyle recommendations consist of evidence-based dietary interventions that lower cardiovascular risk.C32

Many commercial diets can be subdivided into four categories based on the combinations of macronutrients recommended: very low-carbohydrate, low-carbohydrate, very low-fat, and Mediterranean diets (Table 1).232 The more restrictive diets include very low-carbohydrate, low-carbohydrate, and very low-fat diets. These plans modify the typical macronutrient ratios by increasing or decreasing the intake of dietary fat, carbohydrate, and protein. Less restrictive diets, such as the Mediterranean diet and American Heart Association (AHA) dietary guidelines,32 aim to increase the intake of beneficial fatty acids, fiber, fruits, and fresh vegetables.

DietCommercial examplesCalorie restriction?Fat intakeFruit and vegetable intakeSaturated and trans -fat intakeChallengesSummary
Very low-carbohydrate211 Atkins dietNoUnlimitedSeverely limitedUnlimited, but discouragedRequires urine ketone monitoringNo patient-oriented outcomes or long-term data; may improve triglyceride and HDL levels; better for short-term weight loss
Low-carbohydrate/low-glycemic index9,1115 South Beach Diet, the Zone, Sugar Busters!NoUnlimitedLimitedLimitedRequires knowledge of glycemic indexOffers sound principles without ample patient-oriented data
Very low-fat9,11,1622 Ornish dietNoLess than 10 percent of total caloriesVegetarian-based dietNoneRequires meditation, vegetarian diet, smoking cessationPerfusion improvements and symptom reductions; very strict diet; lowers HDL levels
Mediterranean diet2331 NANoUnlimitedUnlimitedLimitedCostly, limited in some populations because of lack of produce or fatty acid sourcesExcellent mortality data, but limited study populations
American Heart Association guidelines32 NAYesUnlimitedUnlimitedSaturated fat: less than 7 percent of total calories Trans-fat: Less than 1 percent of total caloriesBroad guidelinesSame as TLC Diet, but with a greater emphasis on overall cardiovascular risk reduction and population health improvements

A useful approach to analyzing these diets includes an evaluation of the evidence for weight loss, disease-oriented outcomes, and patient-oriented outcomes.

Very Low-Carbohydrate Diets

Dr. Robert Atkins published the first version of the Atkins diet in 1972.2 This plan is a very low-carbohydrate diet that does not require calorie counting or portion limitation. It consists of four phases, with strict carbohydrate restriction in the initial phase (less than 20 g per day). Each consecutive phase permits increased carbohydrate intake, using weight loss as the goal for progression. The effectiveness of the diet is attributed to increased fat metabolism and satiety caused by carbohydrate restriction and subsequent ketosis.


Very low-carbohydrate diets have been studied extensively.39,33 Most of the randomized controlled trials (RCTs) had small cohorts and lasted less than one year. Attrition rates were high, with 12 to 40 percent of participants discontinuing the intervention diet. A large meta-analysis of very low-carbohydrate diets evaluated six trials with 447 overweight participants (body mass index [BMI] greater than 25 kg per m2).33 In most cases, the control diet was a low-fat, calorie-restricted plan. Although all participants lost weight, those following the low-carbohydrate plan had improved weight loss at six months, but no difference at 12 months. Of note, there was no significant difference in levels of low-density lipoprotein (LDL) between the diets. High-density lipoprotein (HDL) and triglyceride levels improved at six and 12 months in very low-carbohydrate dieters.

A recent RCT of 311 premenopausal women showed greater mean weight loss at one year in participants following the Atkins diet (4.7 kg [10 lb, 6 oz]) compared with dieters using the Zone (low-carbohydrate; 1.6 kg [3 lb, 8 oz]), Ornish (very low-fat; 2.2 kg [4 lb, 14 oz]), and LEARN (Lifestyle, Exercise, Attitudes, Relationships, and Nutrition; low-carbohydrate; 2.6 kg [5 lb, 12 oz]) diets.9 LDL levels were significantly reduced in the Atkins group, with no reduction in HDL levels or blood pressure. Insulin and fasting glucose levels did not differ among groups. Other studies have shown greater weight loss with very low-carbohydrate diets compared with low-fat plans in the first six months of use.36,33


No studies have measured differences in morbidity, mortality, or cardiovascular outcomes associated with a very low-carbohydrate diet. A recent analysis of a Swedish female cohort showed an increase in overall mortality rates among women with increased protein and decreased carbohydrate intake.10 However, concerns about the risk of coronary heart disease (CHD) may apply only to very low-carbohydrate diets that are high in fat and protein from animal sources. Very low-carbohydrate diets with fat and protein from vegetable sources may have a small beneficial effect on cardiac risk.34 Patient-oriented outcomes are limited by the need for long-term follow-up and by noncompliance. Theoretic concerns about elevated LDL levels, accelerated heart disease, and ketoacidosis have not come to fruition.

Low-Carbohydrate Diets

There are many low-carbohydrate diets, including Sugar Busters!, the Zone diet, and the South Beach Diet.1214 Many of these programs are based on the glycemic index, which is a measure of serum blood glucose response to the ingestion of 50 g of a given food. Foods with a higher glycemic index cause a more rapid increase in postprandial blood glucose and insulin levels over two hours. Many low-carbohydrate diets advocate the use of healthier sources of dietary fat. Of the commercial low-carbohydrate diets, the Zone and South Beach plans are the only ones that have been studied in randomized, controlled settings.


A 2004 trial compared the South Beach Diet with the National Cholesterol Education Plan (NCEP) diet over 12 weeks.35 Participants following the South Beach Diet lost 2.8 kg (6 lb, 2 oz) more than those following the NCEP plan. There was no difference in lipid levels between the groups. In a one-year comparison trial of commercial diets, participants following the Zone diet lost weight and had improved HDL, LDL, and triglyceride levels.11 The improvements were modest and comparable to those in participants following the Ornish, Weight Watchers, and Atkins diets (although Atkins dieters did not have LDL improvements in this study). Large population-based trials show that diets with a higher glycemic index are associated with lower HDL levels, a risk factor for cardiovascular disease.15


No studies of low-carbohydrate diets have measured patient-oriented outcomes.

Very Low-Fat Diets

Very low-fat diets were developed in response to the correlation between saturated fat intake and coronary atherosclerosis.23 These diets typically limit dietary fat intake to levels much lower than those of traditional low-fat diets. The most well-studied plan is the Ornish diet, which integrates a vegetarian diet with a fat intake of 10 percent of total calories.17 The program also integrates exercise, meditation (one to two hours per day), and smoking cessation. Study groups following the diet were also advised against using alcohol or caffeine.


Favorable pilot study findings spurred a long-term evaluation of the Ornish diet.19 All participants were men diagnosed with coronary artery disease (CAD). The initial three-week pilot study showed reductions in total cholesterol levels and a trend toward improved cardiac ejection fraction. However, HDL levels decreased by 20 percent. A five-year RCT of 48 patients with angiographyproven CHD showed that the diet lowered LDL levels by approximately 20 percent.20 Triglyceride and HDL levels were unchanged. Dieters lost an average of 5.8 kg (12 lb, 13 oz) of body weight compared with no change in the control group. Myocardial perfusion improved in the intervention group, with an average improvement in vessel stenosis of 8 percent compared with a 27 percent worsening in the control group.21


Study participants following the Ornish diet had a 91 percent decrease in angina symptoms at one year and a 72 percent decrease at five years.20 The control group had a 186 percent increase in angina symptoms at one year and a 36 percent decrease at five years. The reduction at five years was attributed to greater rates of bypass surgery and angioplasty in the control group.

Mediterranean Diet

The Mediterranean diet first gained interest when the Seven Countries Study showed low cardiovascular mortality rates in a cohort from Crete despite high levels of dietary fat intake (35 to 40 percent of total calories).23,24 Although the diet has many commercial iterations, it consists of several principal components (Table 2).

Plant-based foods (e.g., fruits, vegetables, breads, cereals, potatoes, legumes, nuts)
Locally grown, minimally processed food
Fish and poultry
Infrequent red meat intake
Up to four whole eggs per week
Moderate amount of dairy products
Olive oil as the principal source of fat
Moderate amount of red wine with meals
Desserts primarily of fresh fruits


Disease-oriented outcomes from studies of the Mediterranean diet have been mixed. Several trials have shown improvements in LDL, HDL, C-reactive protein, and insulin levels in persons following the diet.36,37 Other studies, however, have shown minimal improvements in these markers.27,38 This lack of consistency among study results has been supplanted by strong patient-oriented evidence.


The Mediterranean diet has been extensively evaluated for its role in the reduction and prevention of CAD. This focus on patient-oriented outcomes differentiates it from other diets that focus on disease-oriented outcomes. The Lyon Diet Heart Study was a prospective RCT of 300 patients that evaluated the effectiveness of the Mediterranean diet in the secondary prevention of coronary disease.27 Participants in the Mediterranean diet group supplemented with alphalinoleic acid, a common source of omega-3 fatty acids. Those in the control group followed a standard low-fat diet.

After 27 months, there were no significant differences between the groups in serum lipid levels, blood pressure, or BMI.27 However, the Mediterranean diet group had a 73 percent relative risk (RR) reduction for fatal and non-fatal myocardial infarction (MI) and a 70 percent RR reduction for overall mortality. There also was a significant RR reduction for the following end points: angina, stroke, heart failure, pulmonary embolism, and deep venous thrombosis. These findings were independent of total cholesterol levels, systolic blood pressure, male sex, and aspirin use. High levels of serum alphalinoleic acid were associated with improved outcomes. After 46 months, dietary compliance was greater than 70 percent.

Similar results were found in the GISSI (Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto) trial, which showed that supplementation with n-3 polyunsaturated fatty acids (fish oil) after MI reduced the risk of MI recurrence, stroke, and death by 10 percent.28 Differences between the groups were noted within three months of the beginning of the study. Another study showed decreased MI rates in high-risk patients following a Mediterranean diet supplemented with soybean and mustard seed oils.29 These findings suggest that Mediterranean dietary benefits can be replicated without the stringent use of regional Mediterranean foods. The mechanism for this risk reduction is unclear, but some studies suggest decreases in inflammation associated with poly- and monounsaturated fat intake39 and increases in HDL levels.30

The first U.S. investigation of the Mediterranean diet was published as part of the National Institutes of Health–AARP Diet and Health Study.40 It showed an inverse association between adherence to a Mediterranean-style diet and cardiovascular disease, cancer, and overall mortality rates. These effects were greatest in patients who smoked. The results of this trial warrant further investigation using a randomized, controlled approach.

Implementation of the Mediterranean diet in the United States presents several challenges, including the expense of increased fruit and vegetable intake compared with inexpensive processed foods.31

AHA Dietary Guidelines

The 2006 AHA dietary guidelines aim to integrate multiple dietary and lifestyle interventions (Table 3).32 These guidelines have not been studied, but they incorporate many patient-oriented recommendations, including an increased intake of fruits, vegetables, healthful fats, and fiber. Limited salt, sugar, and trans-fat intake is also an integral part of these guidelines. Many diets fulfill these guidelines, including the TLC (Therapeutic Lifestyle Changes) Diet published by the National Heart, Lung, and Blood Institute.41 These guidelines offer physicians a fairly complete list of evidence-based dietary initiatives that patients can undertake in a stepwise fashion.

InterventionImproves lipid profile?Improves morbidity and/or mortality rates?
Count calories to achieve a hypocaloric diet (for weight loss)YesYes
Increase aerobic exercise to 30 minutes or more on most days; increase to 60 minutes for weight lossYesYes
Increase intake of fresh fruits and vegetables (not juice)YesYes
Increase intake of whole grains and other high-fiber foodsYesUnknown
Consider plant stanol supplementationYesUnknown
Consume oily fish twice per weekYesYes
Increase intake of healthful fats (e.g., olive oil, canola oil, nuts)YesYes
Supplement with omega-3 fatty acids in persons who do not eat fishYesUnknown
Limit saturated and trans-fatty acidsYesYes
Limit alcohol intake to two drinks per day in men and one drink per day in womenNoYes*
Limit salt intake to 2 g per dayNoYes (in persons with high blood pressure)
Minimize sugar intakeYesUnknown
Use caution when eating outside of the homeUnknownUnknown

Putting It into Practice

Effective dietary counseling is difficult within the constraints of a typical outpatient visit. Clinical knowledge of popular commercial diets can make it easier for physicians to offer nonbiased, specific patient recommendations. Although all of the diets discussed in this article have some disease-oriented benefits, patient-oriented benefits are more limited. None of the diets show any significant worsening of patient outcomes. Consultation with a nutritionist should be considered for patients with significant comorbidities in whom a restrictive diet may lead to complications.

A survey of patient outcome data shows that diets that are rich in fruits, vegetables, and healthful fatty acids and that limit saturated fat intake seem to result in the best outcomes (Table 4).211,1632 The AHA guidelines incorporate most of these recommendations without severe dietary restrictions. A stepwise, individual patient approach incorporating one or two interventions every three to six months may be a useful way to utilize these guidelines.

Diet typeLDLHDLTriglyceridesCardiovascular eventsBMIStrength of evidence for cardiovascular outcomesReferences
Very low-carbohydrateNo changeIncreaseDecreaseNo dataDecreaseC211
Low-carbohydrate/low-glycemic indexNo changeNo changeNo changeNo dataDecreaseC1720, 22
Very low-fatDecreaseDecreaseDecreaseDecreaseDecreaseB9, 11, 1622
Mediterranean dietDecrease/no changeIncreaseDecreaseDecreaseNo changeA2331
American Heart Association guidelinesDecreaseIncreaseNo changeNo dataDecreaseC*32

Continue Reading

More in AFP

More in PubMed

Copyright © 2009 by the American Academy of Family Physicians.

This content is owned by the AAFP. A person viewing it online may make one printout of the material and may use that printout only for his or her personal, non-commercial reference. This material may not otherwise be downloaded, copied, printed, stored, transmitted or reproduced in any medium, whether now known or later invented, except as authorized in writing by the AAFP.  See permissions for copyright questions and/or permission requests.