This study aimed to determine the effects of a low-carbohydrate diet (LCD) on nonalcoholic fatty liver disease (NAFLD) patients with type 2 diabetes (T2DM) compared with a calorie-restricted diet (CRD). We evaluated 28 patients with NAFLD and T2DM in this three-month-long, prospective, open-label, randomized comparative trial. Patients were randomly allocated to either the LCD group (70-130 g/day of carbohydrates; n =14) or CRD group (25 kcal/kg of ideal body weight per day; n =14). Abdominal computed tomography (CT) was used to evaluate the liver fat deposition by the liver-to-spleen attenuation ratio (L/S ratio) and visceral fat accumulation as the visceral fat area (VFA). Both diets significantly improved the L/S ratio, VFA, AST, ALT, body weight and HbA1c from baseline to 3 months (P < 0.05). There were no significant differences in any findings between the two diets, except for the fact that the VFA significantly decreased in the LCD group. Our findings suggest that LCD may be as effective as CRD in improving fatty liver in patients with NAFLD and T2DM.

The patient was a 76-year-old man. He had been diagnosed with diabetes at 54 years of age and had been treated with an oral hypoglycemic drug. In 2013, insulin glulisine was started, and biphasic insulin aspart was started in February 2015. From April 2016, marked hypoglycemia in the early morning and hyperglycemia during the day repeatedly appeared. His fasting blood glucose level was 39 mg/dL, IRI was 1546.4 μU/mL, and insulin antibody was detected at high concentration, showing a low affinity and high binding capacity on a Scatchard plot. We diagnosed his significant blood glucose fluctuation as being due to this insulin antibody, discontinued the insulin administration, and started the administration of a GLP-1 receptor agonist (duraglutide). As a result, the hypoglycemia from nighttime to early morning disappeared promptly, and thereafter, the IRI value and insulin antibody decreased gradually. Although α-GI and SU drugs administered to improve hyperglycemia during the day were insufficiently effective, combination with an SGLT2 inhibitor (tofogliflozin) was effective. We experienced a case in which the combination of a GLP-1 receptor agonist and an SGLT2 inhibitor was effective for resolving blood glucose fluctuation under conditions of significant blood glucose fluctuation caused by the appearance of insulin antibody.

The patient was a 75-year-old man whose HbA1c level had worsened rapidly from 5.9 % to 8.8 % over 3 months. Fasting hypoglycemia, postprandial hyperglycemia and hyperinsulinemia were observed, and anti-insulin receptor antibody was detected. Because of these findings, we diagnosed him with type B insulin resistance. Flash glucose monitoring (FGM) showed that dapagliflozin, empagliflozin and miglitol reduced the mean blood glucose value±standard deviation from 188±91 mg/dL to 146±82, 140±70 and 142±61 mg/dL, respectively, by improving postprandial hyperglycemia. In contrast, the percentage of time in a hypoglycemic state was increased by dapagliflozin and empagliflozin, but not miglitol. These findings indicate that miglitol is the most effective agent among these three drugs for treating cases of type B insulin resistance with postprandial hyperglycemia because of the lack of any effect on the percentage of time in a hypoglycemic state.

A 26-year-old man was referred to our hospital due to hyperglycemia. He had been diagnosed with autism spectrum disorder in 2002. Since 2010, he had drunk 6 to 12 L of sugar-sweetened beverages every day. The amount of polydipsia had gradually decreased due to the adjustment of psychiatric medications, but in July 2016, he showed weight loss and was found to be in a hyperglycemic state. His plasma glucose level was 377 mg/dL, and his HbA1c was 16.9 %. Furthermore, ketone bodies were positive on a urine test, so he was diagnosed with soft-drink ketosis. Insulin therapy was started, and his blood glucose level gradually decreased. He stopped drinking sugar-sweetened beverages after the diagnosis of diabetes mellitus, and all glucose-lowering medications were able to be stopped soon after. However, his polydipsia was persistent, and a decrease in his antidiuretic hormone release (ADH) was confirmed, although his polydipsia had initially been thought to be merely psychogenic. The administration of desmopressin was started, which improved his polydipsia. Long-term polydipsia may cause the down-regulation of ADH, so it is important to diagnose and treat polydipsia properly in patients consuming sugar-sweetened beverages, as such treatment can prevent the development of diabetes mellitus or improve their glycemic control.