Reviewer 2's Review

“This paper by Liska et al. evaluated whether there are differences in intramyocellular and hepatic fat content in obese adolescents of three ethnic groups with normal glucose tolerance. Glucose tolerance was determined by OGTT. Insulin sensitivity was estimated using the Matsuda index. Body fat content was measured by DEXA. Intramyocellular fat content was measured using NMR spectroscopy. Liver fat content was measured using fast-MRI. Abdominal MRI was used to measure areas of visceral and subcutaneous adipose tissue. Twenty-one Caucasians, 17 African-Americans and 17 Hispanics were included in this study. Both males and females were analyzed together. African-americans had lower visceral fat area and higher subcutaneous fat when compared to the other ethnic groups. Intramyocellular lipid content was significantly higher in the Hispanics. Hepatic fat content was found to be absent in the African-American group but there were no differences between the Hispanics and the Caucasians. The correlation analysis shows that insulin resistance is related to visceral fat, hepatic fat content and intramyocellular fat content for the three ethnic groups pooled together. The authors conclude that, irrespective of obesity, there are ethnic differences in the amount of lipids accumulated in skeletal muscle, liver and abdominal cavity.

General comments:
This paper addresses an interesting topic. The findings are an extension of previous observations in the adult literature. The main problem with this study is the lack of substantial clinical or mechanistic implications.
The main results are that body composition is different in the three ethnic groups and that these differences are associated with ethnic differences in fat content in liver and muscle. The observation that liver fat content is not detectable in African American adolescents is of interest. This was clearly correlated with decreased visceral fat content. Since insulin resistance is highly associated with visceral fat content, the question arises of whether the reported associations between hepatic fat and insulin resistance are merely reflective of visceral fat content. The correlations reported in figure 3 were performed on the whole group and it does not appear there was any independence from ethnicity or gender. Therefore, my general impression is that these results do not add any information on the role of fat partitioning on insulin resistance and susceptibility to type 2 diabetes in adolescents. This impression is substantiated by the similarity in insulin sensitivity among the three ethnic groups despite the reported differences in fat partitioning (table 2) and the lower 2-hrs insulin concentrations during OGTT of Hispanics.”

N.B. These are the general comments made by the reviewer when reviewing this paper in light of which the manuscript was revised. Specific points addressed during revision of the paper are not shown.