Commentary February 19, 2014
Peter Libby MD
Observational studies have long associated a food pattern that has come to be called the “Mediterranean diet” with reduced cardiovascular disease. In pioneering work, the late Ancel Keys in the Seven Countries Study documented a gradient in cardiovascular risk from south to north in Europe and spurred interest in a dietary basis for some of this finding. Constituents of the Mediterranean pattern of eating, including fish consumption, olive oil, and fruits and nuts, have received much attention for their health-promoting properties. In a landmark randomized controlled trial (PREDIMED: PREvención con DIeta MEDiterránea), a group of Spanish-based investigators reported striking reductions in cardiovascular events in patients allocated to a Mediterranean diet with extra virgin olive oil or with nuts.1 The cardiovascular benefit of the Mediterranean diet intervention derived primarily from a highly significant reduction in strokes (P = .003.)
In an exploratory, post hoc, non-prespecified analysis, the PREDIMED investigators examined the effect of these interventions on peripheral arterial disease (PAD) endpoints. The endpoints were clinically relevant and well-adjudicated using contemporary techniques. Although the number of individuals who developed incident PAD was small, the diet intervention groups had a significant decline. In the three groups of approximately 2500 individuals, 18 in the extra virgin olive–oil group and 26 in the nut-supplemented group versus 45 in the control group developed incident PAD. The benefit appeared early in Kaplan–Meier analysis, with a visual separation of the curves evident in the second year of the > 5-year observation period.
These results have considerable medical importance as a hypothesis-generating exploratory analysis. Few, if any treatments, have undergone rigorous assessment to reduce the incidence of PAD. A dietary intervention that could achieve this goal safely and easily has obvious attraction.
From a mechanistic perspective, it is not evident why stroke and PAD should decline significantly with the Mediterranean diet pattern while myocardial infarction did not decrease significantly. An anti-atherosclerotic or anti-thrombotic mechanism might be expected to reduce all three types of cardiovascular event. Certainly these exploratory results merit follow-up in a prospective manner. It would be interesting to assess interaction with risk factors for PAD such as smoking. Perhaps an antioxidant effect in smokers might benefit the PAD endpoint particularly. Biomarker analyses in the PREDIMED study looking at variables related to thrombosis, fibrinolysis, inflammation, and oxidative stress might yield mechanistic insights that could be tested prospectively in future trials.