Critique 105: For preventing cardiovascular disease, an unrestricted-calorie Mediterranean-type diet provides 30% more benefit than a low-fat diet: Results of a large population-based clinical trial — 6 March 2013

Estruch R, Ros E, Salas-Salvadó J, Covas M-I, Corella D, Arós F, et al, for the PREDIMED Study Investigators.  Primary prevention of cardiovascular disease with a Mediterranean diet.  N Engl J Med 2013. DOI: 10.1056/NEJMoa1200303

Authors’ Abstract

Background  Observational cohort studies and a secondary prevention trial have shown an inverse association between adherence to the Mediterranean diet and cardiovascular risk. We conducted a randomized trial of this diet pattern for the primary prevention of cardiovascular events.

Methods  In a multicenter trial in Spain, we randomly assigned participants who were at high cardiovascular risk, but with no cardiovascular disease at enrollment, to one of three diets: a Mediterranean diet supplemented with extra-virgin olive oil, a Mediterranean diet supplemented with mixed nuts, or a control diet (advice to reduce dietary fat). Participants received quarterly individual and group educational sessions and, depending on group assignment, free provision of extra-virgin olive oil, mixed nuts, or small nonfood gifts. The primary end point was the rate of major cardiovascular events (myocardial infarction, stroke, or death from cardiovascular causes). On the basis of the results of an interim analysis, the trial was stopped after a median follow-up of 4.8 years.

Results  A total of 7,447 persons were enrolled (age range, 55 to 80 years); 57% were women.  The two Mediterranean-diet groups had good adherence to the intervention, according to self-reported intake and biomarker analyses. A primary end-point event occurred in 288 participants. The multivariable-adjusted hazard ratios were 0.70 (95% confidence interval [CI], 0.54 to 0.92) and 0.72 (95% CI, 0.54 to 0.96) for the group assigned to a Mediterranean diet with extra-virgin olive oil (96 events) and the group assigned to a Mediterranean diet with nuts (83 events), respectively, versus the control group (109 events). No diet-related adverse effects were reported.

Conclusions  Among persons at high cardiovascular risk, a Mediterranean diet supplemented with extra-virgin olive oil or nuts reduced the incidence of major cardiovascular events.

Forum Comments

A large number of observational studies have suggested that subjects following a “Mediterranean-type diet” (Med-Diet) tend to have lower levels of cardiovascular risk factors, fewer diseases, and lower mortality.  An excellent meta-analysis of prospective observational studies was published by Sofi et al in 2008  (Sofi F, Cesari F, Abbate R, Gensini GF, Casini A. Adherence to Mediterranean diet and health status: meta-analysis.   BMJ 2008;337:a1344.  doi:10.1136/bmj.a1344).  These authors stated:  “We evaluated studies that analysed prospectively the association between adherence to a Mediterranean diet, mortality, and incidence of diseases; 12 studies, with a total of 1,574,299 subjects followed for a time ranging from three to 18 years were included.  Greater adherence to a Mediterranean diet is associated with a significant improvement in health status, as seen by a significant reduction in overall mortality (9%), mortality from cardiovascular diseases (9%), incidence of or mortality from cancer (6%), and incidence of Parkinson’s disease and Alzheimer’s disease (13%).”

Striking benefits in the secondary prevention of cardiovascular disease from a Med-Diet enriched with the essential omega-3 fatty acid alpha-linolenic acid (ALA), in comparison with the typically recommended low-fat diet, were reported in 1994 from Serge Renaud’s group in Lyon, the Lyon Diet-Heart Study (de Lorgeril M, Renaud S, Mamelle N, et al.  Mediterranean alpha-linolenic acid rich diet in secondary prevention of coronary heart disease. Lancet 1994;343:1454-1459).   That study was also stopped prematurely because of overwhelming results favoring the Med-Diet.  In the Lyon study, after a mean follow up of 27 months, a risk ratio for the two main endpoints combined (cardiac deaths, non-fatal MIs) was 0.27 (95% CI 0.12-0.59, p = 0.001) after adjustment for prognostic variables.  Overall mortality was also much lower in the group given the diet enriched with ALA, an adjusted risk ratio of 0.30 (95% CI 0.11-0.82, p = 0.02).

The present study is the first large randomized clinical trial among subjects with diabetes, hypertension, dyslipidemia, and other factors placing them at high-risk for cardiovascular disease.  The study evaluated the effects on the development of cardiovascular disease of two versions of a Med-Diet, one enriched with extra-virgin olive oil (EVOO) and one enriched with at least three servings per week of mixed nuts.  The “control” group was instructed in the typical low-fat diet.  The a priori primary end point was the occurrence of what was defined as a “major cardiovascular event” (myocardial infarction, stroke, or death from cardiovascular disease).  On the basis of interim results, the study was stopped early by a monitoring board after a median follow up of just under 5 years.

Specific comments on the paper:  The key results of this paper were an approximately 30% lower risk of developing cardiovascular disease for each of the two groups assigned to the Med-Diet.  Forum reviewer Ellison stated: “In this group of subjects in Spain, where a large proportion of subjects were already following somewhat of a Med-Diet, there were few dietary differences during the trial between groups as to the intake of vegetables, whole grains, cereal, fruits, meat, pastries, dairy, and alcohol.  Subjects on either of the two Med-Diets consumed more EVOO, nuts, and slightly more legumes and fish.  The assigned diets resulted in a slight increase in total fat (from about 39% to 41%) in both Med-Diet arms, versus a slight decrease (39-37%) in the control group.  There were slight decreases for saturated fats in all groups (from about 10 to 9.1 – 9.4% of calories), and increases in ALA for the nut group and slight decreases for the other two groups. (It is a shame that the first thing that everyone thinks about for preventing heart disease is cutting down on total fat!)” 

The effects of the Med-Diets were not different according to the presence or absence of most co-existing conditions: smoking, diabetes, or BMI.  However, no effect was seen in the about 20% of subjects who did not have hypertension. The largest effects were in subjects with dyslipidemia.  Ellison added: “It is reassuring that there were larger effects in subjects showing greater adherence to the Med-Diet, suggesting that it was the diet and not chance leading to the results.”

What are the key elements of the Med-Diet leading to better health outcomes?  When reviewing the nutrient data provided, the better results were obtained from a diet that was higher in total fat and with slightly more fish and legumes than the typical “low-fat diet” that was advised for the control group.  However, the main contributors to the effects of the Med-Diet appear to be the supplements provided to subjects: rather large amounts of extra-virgin olive oil or several helpings per week of nuts (supplies of both the EVOO and the nuts were provided free to subjects in these groups).  The authors agreed, as they stated: “Thus, extra virgin olive oil and nuts were probably responsible for most of the observed benefits of the Mediterranean diets.”   The group given nuts was the only one that showed an increase in ALA, which was the key dietary difference associated with the reduction in cardiovascular events and mortality in the Lyon Diet Heart Study almost twenty years ago. 

Effects of changes in fatty acids:  Reviewer Lanzmann-Petithory points out that, in general, Spain has high levels of the ratio of omega-6 linoleic acid (LA) to omga-3 alpha-linolenic acid (ALA), related primarily to a high consumption of sunflower oil.  The work of Renaud repeatedly showed that higher levels of ALA, and lower levels of linoleic acid, are key factors in reducing cardiovascular risk (Renaud S, et al. Cretan Mediterranean diet for prevention of coronary heart disease.  Am J Clin Nutr 1995;61 suppl:1360S-1367S).  This was shown first by the very low levels of heart disease among the men in Crete in the Seven Countries Study ( Keys A. Coronary heart disease in seven countries. Circulation 1970:4l(suppl l):l-211)  A primary feature of the Cretan diet was higher levels of ALA and lower amounts of LA.  

All of the groups of this study had a low intake of ALA at baseline, and only the group given nuts reported an increase, to 1.9 g/day.   This was presumably related to the inclusion of walnuts as a supplement to the Med-Diet; walnuts are the only nuts rich in ALA, but are also high in LA.  Data on plasma fatty acids are not presented in the present study, so it is not possible to judge what the plasma LA/ALA ratios were in the different groups.

Potential use of canola oil to improve fatty acid profile and protect against cardiovascular disease:  Forum members considered it unfortunate that there was not a separate Med-Diet group in this trial given canola oil, which would have increased the ALA, oleic acid, and EPA but also decreased LA and 20:4n-6.  The fact that the LA/ALA ratio apparently remained quite high in both Med-Diet groups in the present study (with the ALA intake increasing only in the group given large amounts of mixed nuts) may explain why the reductions in cardiovascular deaths and in total mortality were not more impressive.  Reviewer Lanzmann-Petithory suggests that had a group been given canola oil, “it would have shown even greater protective effects against cardiovascular disease.”

As described by Reviewer Orgogozo, canola oil (huile de colza), which is related to rapeseed oil and mustard oil, is an ideal oil for cardio-protection.  It is very similar to olive oil, high in non-unsaturated fatty acids but with high levels ALA and very low levels in saturated fat; it has a very favorable ratio (2.8) of n-6 to n-3 fatty acids.   In the past, the wild type of colza contained two different toxins, often combined, which were not lethal but caused gastrointestinal discomfort and hence rejection.  This was solved by natural hybridization/selection (not genetically modified) so that the oil is now palatable and safe.”  Lanzmann-Petithory stated: “In 1981 there was an outbreak of food poisoning in Spain, with 1,000 deaths and 25,000 cases of acute intoxication from organo-phosphates in pesticides (used on tomatoes).  Apparently to protect Spain’s vegetable exports, the poisonings were erroneously attributed to canola oil.  As the compensations of the victims were delayed and there were frequent demonstrations against canola organized in Spain, it would not be feasible to attempt a canola oil study in Spain at this time.”

This is not a study of wine and cardiovascular outcomes:  Reviewer Finkel wondered why there was no more data on the drinking characteristics of the subjects.  The large majority of subjects in all of the intervention groups were consuming alcohol at the onset of and throughout the trial.  From the data shown, it is not possible to judge the role that wine consumption may have played in the more favorable results from the two Med-Diets, as the total alcohol intake of the two Med-Diet groups were only slightly higher than that in the control group.  Over the approximately 5 years of follow up, about one-third of subjects in the Med-Diet groups reported consuming ≥ 7 glasses of wine per week (versus about one-quarter of the control group).  But there are no data presented as to whether the number of subjects in the two Med-Diet groups increased their wine consumption during the trial.   Reviewer Goldfinger stated: “Although mentioned as a component of the Med-Diet, wine was not quantified and was recommended to the subjects as a part of a continued habit, as opposed to an intervention.  I suspect that non-wine drinkers remained non-wine drinkers in each of the study groups.”

Forum Summary

A large well-done clinical trial in Spain was carried out among 7,447 subjects at high-risk of cardiovascular disease (on the basis of diabetes, hypertension, dyslipidemia, and other risk factors).  The trial compared the effects of two versions of an energy-unrestricted Mediterranean-type diet, one with large amounts of extra-virgin olive oil (EVOO) and one with a supplement of mixed nuts, with results among subjects in a “control group,” made up of subjects who were advised to follow a low-fat, low-cholesterol diet (that has typically been recommended for reducing the risk of cardiovascular disease). 

Both of the groups advised to consume a Med-Diet showed an increase in total fat, from an average of 39.3 % of total calories to an average of more than 41%; the group advised to decrease their total fat lowered their intake from 39 to 37 % of calories.  All three groups showed slight decreases in saturated fat and slight decreases in dietary cholesterol intake (but blood cholesterol levels were not reported).  Only the Med-Diet + nuts group showed an increase in alpha-linolenic acid (ALA), the fatty acid related to the lowest risk of cardiovascular disease in the Cretan cohort of the Seven-Countries Study and in the Lyon Diet-Heart Study. 

The large majority of subjects in all groups consumed wine at baseline, and about one-third of the Med-Diet groups and one-quarter of the control group reported consuming ≥ 7 glasses of wine/week during the trial.   However, alcohol consumption was not evaluated specifically for its effects in the present study. 

The study was ended early by its data and safety monitoring board due to marked advantages being shown in terms of the occurrence of the primary end-point (the occurrence of myocardial infarction, stroke, or death from cardiovascular causes) in both Med-Diet groups in comparison with the control group.  The multivariable-adjusted hazard ratios were 0.70 and 0.72 for the Med-Diet with olive oil and the Med-Diet with added nuts, hence an approximately 30% decrease in risk among subjects assigned to a Med-Diet compared with those in the control group.

Forum members (as did the authors) had difficulty in determining which specific components of the Med-Diet may have been the primary reasons for its more favorable results, as significant dietary changes during the trial were mainly related to the supplementary olive oil or nuts given to subjects in the two Med-Diet groups.  Among potential factors were increased mono-unsaturated fat and higher antioxidants from the added EVOO and an increase in ALA from the supplemental nuts.  Forum members noted that the Med-Diet interventions apparently did not provide an increase in the intake of ALA, which could have led to even more protection against cardiovascular disease. 

In any case, the striking reduction in risk shown in the study from the Med-Diet provide further evidence of the effectiveness of such a diet in reducing cardiovascular disease.  It has been shown repeatedly that people advised to consume a Mediterranean-type diet tend to comply much better with such advice than do people advised mainly to decrease their intake of all fats and cholesterol.  A very large amount of  scientific data now support advice to follow a Mediterranean-type diet, one that does not restrict calories but encourages the intake of selected fats (which may include extra-virgin olive oil or canola oil), as well as more fruits, vegetables, grains, nuts, and wine; it may be the most effective  dietary approach for reducing cardiovascular disease.       

Reviewer Keil points out that many good prospective cohort studies have shown that an increase in adherence to components of a Mediterranean-type diet decreases the risk of many diseases and mortality.  He adds: “The great thing about the present study is that when the skeptics raise  their voice and argue that the proponents of the Mediterranean diet have only observational data available, we can now respond that we not only have RCT data on secondary prevention (Lyon Diet-Heart study) but also the present study on primary prevention of cardiovascular disease.”

                                                                  *                   *                   *

Comments on this critique by the International Scientific Forum on Alcohol Research were provided by the following members:

Dominique Lanzmann-Petithory,MD, PhD, Nutrition/Cardiology, Praticien Hospitalier Hôpital Emile Roux, Paris, France

Harvey Finkel, MD, Hematology/Oncology, Boston University Medical Center, Boston, MA, USA

Jean-Marc Orgogozo, MD, Professor of Neurology and Head of the Neurology Divisions, the University Hospital of Bordeaux, Pessac, France

David Van Velden, MD, Dept. of Pathology, Stellenbosch University, Stellenbosch, South Africa

Arne Svilaas, MD, PhD, general practice and lipidology, Oslo University Hospital, Oslo, Norway

Tedd Goldfinger, DO, FACC, Desert Cardiology of Tucson Heart Center, Dept. of Cardiology, University of Arizona School of Medicine, Tucson, Arizona, USA

Ulrich Keil, MD, PhD, Institute of Epidemiology and Social Medicine,  University of Münster, Münster, Germany

Erik Skovenborg, MD, Scandinavian Medical Alcohol Board, Practitioner, Aarhus, Denmark


R. Curtis Ellison, MD, Section of Preventive Medicine & Epidemiology, Boston University School of Medicine, Boston, MA, USA