The intricacies of the Mediterranean diet in NAFLD. Response to Duret et al

Response to Duret et al. Daniele Pastori, MD, PhD1,2, Francesco Baratta, MD1,2, Ilaria Ernesti, MD3, Francesco Violi, MD1, Francesco Angelico, MD4 and Maria Del Ben, MD1 Am J Gastroenterol https://doi.org/10.1038/s41395018-0055-2 We read with great interest the letter by Dr. Duret and colleagues [1] regarding our work recently published by the Journal [2], on the relationship between adherence to Mediterranean Diet (Med-Diet) and insulin resistance (IR) in a population of subjects screened for non-alcoholic fatty liver disease (NAFLD). As correctly outlined by Dr. Duret, the pathogenesis of NAFLD is complex and multifactorial, and a predominant role is played by obesity, diabetes, and diet. At this regard, we showed that a high adherence to Med-Diet was associated to a lower prevalence of NAFLD and to a better metabolic profile, as indicated by a lower IR and obesity. We agree that in addition to the cardio-protective foods explored in our study, other nutrients have an important role in the onset of NAFLD. For instance, high fructose intake has been clearly linked to an enhanced prevalence of NAFLD, as it has been shown to promote de novo lipogenesis at liver site [3]. However, it is not firmly clear whether fructose may per se induce intrahepatic lipid storage, or rather this mechanism may be attributable to caloric excess. Thus, a previous metanalysis showed that in studies comparing isocaloric diets with fructose or other carbohydrates, no changes in liver fat content were observed [4]. Conversely, diets supplemented with excess energy from high-dose fructose induced fatty liver and serum liver enzymes increase [4]. In our study, data on whole dietary fructose intake were available in a subgroup of 201 non-diabetic patients with NAFLD. Median fructose intake was 5.8 [2.3–10.0] g/die. We found a significantly higher proportion of patients with high fructose intake (above median) in patients with compared to those without IR (53.7% vs. 69.8%, p = 0.020). This effect seems to be mediated by obesity, as patients with high fructose intake had higher waist circumference than those below median (108.3 ± 11.3 vs. 104.9 ± 10.3 cm, p = 0.027). Conversely, we did not find any association between total fiber intake and IR or ultrasound NAFLD severity. These data suggest that fructose may facilitate NAFLD onset by increasing visceral obesity, that in turn induces IR and NAFLD. Thus, lifestyle intervention, including diet and physical activity, remains the first therapeutic approach in patients with NAFLD. At this regard, recent promising data showed that 3-month intense intervention with Med-Diet significantly induced weight loss in subjects at high cardiometabolic risk [5]. Our belief is that diet modifications may deeply influence body composition and fat distribution, and could reduce the incidence of NAFLD mostly acting on IR at early stages. More over, once NAFLD is established, adherence to Med-Diet may help reducing the degree of hepatic inflammation and oxidative stress, thus lowering the rate of progression to non-alcoholic steatohepatitis. Interventional studies addressing this issue would clarify the role of Med-Diet in this setting. ConfliCt of interest The authors declare that they have no conflict of interest.