First, in our study, insulin resistance or insulin sensitivity presented by the homeostasis model assessment for insulin resistance (HOMA-IR) and Matsuda index was not different between the subjects who had 1-hour glucose levels of ≥155 mg/dL with normal glucose tolerance (NGT 1 hour-high) and the subjects with NGT 1 hour-low (<155 mg/dL). In addition, simple indices representing β-cell function such as HOMA-β cell function (HOMA-β) and insulinogenic index were also comparable between the two groups. However, β-cell function should be viewed in the context of insulin sensitivity to reflect the β-cell capacity of compensation for insulin resistance [2]. As such, indices of β-cell function adjusted by insulin sensitivity (e.g., oral disposition index and insulin secretion-sensitivity index-2) were different between the NGT 1 hour-high and NGT 1 hour-low subjects. Therefore, the data presented in our study indicated that insulin resistance (or sensitivity) and simple β-cell function indices were not different, but indices for β-cell function adjusted by insulin sensitivity were different, which consistently substantiated our conclusion.

Second, we did not compare NGT 1 hour-high with impaired fasting glucose (IFG). Because there are different metabolic abnormalities between IFG and IGT [3], it would be worthwhile to examine the differences in insulin sensitivity and β-cell function between NGT 1 hour-high and IFG.

Lastly, Dr. Kim suggested an important implication of our study: NGT 1 hour-high subjects may be associated with non-alcoholic fatty liver disease and dyslipidemia and could be potential candidates for pharmacological or non-pharmacological intervention even though they are classified as normal based on fasting plasma glucose and 2-hour postload plasma glucose levels. Indeed, we showed that fasting plasma glucose and triglyceride levels were significantly higher in NGT 1 hour-high subjects than in NGT 1 hour-low subjects. These parameters are components of metabolic syndrome, which is considered as a risk factor for type 2 diabetes [4]. Further studies are necessary regarding the cost-effectiveness of testing 1-hour postload glucose levels during the standard 75 g oral glucose tolerance test. We all appreciate Dr. Kim's valuable comments and suggestions.