However, the evidence for a unique fructose effect is sufficiently weak in both studies that the authors appear left with two solutions in search of a problem. The evidence of

AZD8055 in vivo Abdelmalek et al.1 for a fructose-specific role in NAFLD is not convincing for several reasons. First, in the categorization of subjects by sweetened beverage intake, the designation “no fructose consumers” is misleading; fructose intake from nonbeverage sources was neither calculated nor reported and could have significantly altered clinical and histological conclusions. Second, the raw data in Table 1 of their article (before statistical manipulation) demonstrate inconsistent fructose dose–dependent trends for all but two of the seven metabolic parameters and for all of the histological parameters measured. Third, the caloric difference between the subject groups [the highest fructose consumers (seven or more servings per week) had twice the calorie intake of the no-fructose group (zero servings per week)] cannot be ignored as a far simpler explanation for the few observed differences reported by the authors. This is especially

compelling because the fructose increment in moderate consumers versus zero consumers could not have exceeded 15% of the total energy, whereas that check details for the highest consumers could have been as low as 5% (calculated from Table 1 of their article). Thus, rather than proposing reduced fructose intake as the solution for NAFLD patients at risk for liver fibrosis, Abdelmalek

et al. should recommend that these patients eat less of everything. The experimental design of Li et al.2 provided rats adlibitum access to selleck kinase inhibitor 10% (wt/vol) fructose in water with or without curcumin or pioglitazone and standard rat chow. The data in Fig. 1 of their article show that the fructose rats consumed approximately 1650 mL of fluid per week (236 mL/day), which is equivalent to 94 kcal/day from fructose. For perspective, this would be like feeding a 70-kg human more than 5200 g of fructose per day (21,000 kcal/day). From the combination of such a highly exaggerated exposure and a lack of a suitable nonfructose carbohydrate control, we can conclude only that (1) this study does not prove a meaningful role for fructose in the development of hepatic steatosis at typical intake levels (45 g/day or 180 kcal/day3), and (2) the curcumin therapy solution proposed by Li et al. has academic interest but is likely unnecessary in humans. Because Abdelmalek et al.1 and Li et al.2 have failed to demonstrate a unique effect for fructose in patients with NAFLD or NASH, respectively, they are left with two solutions in search of a problem. John S. White Ph.D. Potential conflict of interest: John S. White is the president of White Technical Research and is a consultant to the food and beverage industry in the area of nutritive sweeteners.