Mediterranean Diet Cuts Risk of Diabetes After Heart Attack

In a blog post last year I discussed how the Mediterranean diet reduces the incidence of type 2 diabetes in healthy people.  I found another scientific journal article that examined the effect of various lifestyle factors that might influence the onset of type 2 diabetes in a different population: people who have had a recent heart attack.

Dariush Mozaffarian and colleauges studied 8291 Italians who had suffered a heart attack within the previous three months, but who did not have diabetes at the time of the heart attack.  Each study participant was followed for an average of 3.2 years to see if diabetes developed.  The researchers devised a Mediterranean diet score (range 0-15) incorporating consumption of cooked and raw vegetables, fruit, fish  and olive oil.  They also looked at consumption of butter, oils other than olive oil, cheese, wine, and coffee.  Participants’ dietary habits were assessed and scored three times over 1.5 years.  A number of other demographic, clinical, and lifestyle risk-factors were assessed.

The study did not survey other components of the Mediterranean diet, such as legumes, nuts, and grains.  This is a weakness of the study.  I suspect it relates to the fact they were using information from the GISSI-Prevenzione study, which was designed to evaluate fish oil and vitamin E in people who had had a heart attack, and researchers did not want to burden outpatient cardiology offices with full-scale questionnaires.

Over the three years of the study, 12% of participants developed new-onset diabetes, or 3.7% per year.  If not for the recent heart attack, the expected incidence rate for development of diabetes would be roughly 1.2% per year.  An even larger percentage, over 25%, of participants developed impaired fasting glucose, a kind of prediabetes that often develops into full-blown diabetes over time.

Was there anything about the people who developed diabetes that distinguished them from those who did not?  Yes – they tended to have older age, higher body mass index, high blood pressure, and they smoked.  Current smoking was associated with a 60% higher risk.  Every unit of higher body mass index, e.g, going from BMI 26 to 27, increased the risk by 9%.  High blood pressure increased the risk by 22%.

What about Mediterranean diet score?  The higher Mediterranean diet scores – score of 11-15 compared to 0-5 – were associated with 35% lower risk of diabetes.  A reduction in onset of impaired fasting glucose was similar.

The authors cite another study of 2499 patients with stable angina pectoris or remote heart attack (over 6 months perviously).  Twenty-two percent of them (one in five) developed diabetes or impaired fasting glucose over six years of follow-up, a rate of 4.1% per year.

The researchers write:

The lower risk associated with a Mediterranean-type diet suggests that diet could help reduce incidence of prediabetes and diabetes after a myodcardial infarction.  Many, though not all, trials have indicated that a Mediterranean-type diet lowers risk factors linked to insulin resistance and diabetes, including serum triglycerides, HDL cholesterol, systemic inflammation, endothelial function, and insulin sensitivity.  These physiological effects in short-term randomized trials provide biological plausibility for the inverse association between consumption of a Mediterranean-type diet and incidence of [impaired fasting glucose] and diabetes in this study.

What are the take-home points of this study for people – Italians, at least – who have had a recent heart attack?

1. A recent heart attack is a risk factor for development of diabetes and prediabetes.
2. The risk of developing diabetes and prediabetes may be significantly reduced by smoking cessation, prevention of weight gain, and consumption of typical Mediterranean foods.

Patients with both heart attacks and diabetes  have significantly worse outcomes  than people with only one of these conditions.  Since we can prevent many cases heart attack and diabetes through diet modification, why not?

Steve Parker, M.D.

Reference:  Mozaffarian, Dariush, et al.  Incidence of new-onset diabetes and impaired fasting glucose in patients with recent myocardial infarction and the effect of clinical and lifestyle risk factors.  Lancet, 370 (2007) 667-675.