BackgroundRecent experimental studies have revealed that n-3 fatty acids, such as eicosapentaenoic acid (EPA) regulate postprandial insulin secretion, and correct postprandial glucose and lipid abnormalities. However, the effects of 6-month EPA treatment on postprandial hyperglycemia and hyperlipidemia, insulin secretion, and concomitant endothelial dysfunction remain unknown in patients with impaired glucose metabolism (IGM) and coronary artery disease (CAD).Methods and resultsWe randomized 107 newly diagnosed IGM patients with CAD to receive either 1800 mg/day of EPA (EPA group, n = 53) or no EPA (n = 54). Cookie meal testing (carbohydrates: 75 g, fat: 28.5 g) and endothelial function testing using fasting-state flow-mediated dilatation (FMD) were performed before and after 6 months of treatment. The primary outcome of this study was changes in postprandial glycemic and triglyceridemic control and secondary outcomes were improvement of insulin secretion and endothelial dysfunction. After 6 months, the EPA group exhibited significant improvements in EPA/arachidonic acid, fasting triglyceride (TG), and high-density lipoprotein cholesterol (HDL-C). The EPA group also exhibited significant decreases in the incremental TG peak, area under the curve (AUC) for postprandial TG, incremental glucose peak, AUC for postprandial glucose, and improvements in glycometabolism categorization. No significant changes were observed for hemoglobin A1c and fasting plasma glucose levels. The EPA group exhibited a significant increase in AUC-immune reactive insulin/AUC-plasma glucose ratio (which indicates postprandial insulin secretory ability) and significant improvements in FMD. Multiple regression analysis revealed that decreases in the TG/HDL-C ratio and incremental TG peak were independent predictors of FMD improvement in the EPA group.ConclusionsEPA corrected postprandial hypertriglyceridemia, hyperglycemia and insulin secretion ability. This amelioration of several metabolic abnormalities was accompanied by recovery of concomitant endothelial dysfunction in newly diagnosed IGM patients with CAD.Clinical Trial Registration UMIN Registry number: UMIN000011265 (https://www.upload.umin.ac.jp/cgi-open-bin/ctr/ctr.cgi?function=brows&action=brows&type=summary&recptno=R000013200&language=E)Electronic supplementary materialThe online version of this article (doi:10.1186/s12933-016-0437-y) contains supplementary material, which is available to authorized users.