Intermuscular lipid: a marker of disordered fat partitioning or the consequence of obesity?

Abstract

In this issue of the Journal, Miljkovic-Gacic et al (1) describe a cross-sectional study of 1249 Afro-Caribbean men from the Tobago Health Study, a study originally designed to examine prostate cancer (and thus limited to men 40 y old). This population has a particularly high prevalence of type 2 diabetes— 20%—and nearly 70% of the population is either overweight or obese. Almost half of the men report a family history of type 2 diabetes. In this large and unique study, the investigators obtained quantitative computed tomography (QCT) measures of the lower calf, with quantitation of total muscle area and adipose tissue distribution of the calf. The authors report that, with aging, intermuscular adipose tissue (IMAT) increased, whereas subcutaneous adipose tissue and total muscle areas decreased. As in many previous studies, measures of obesity and family history were associated with type 2 diabetes, as was IMAT. Perhaps it is counterintuitive, but, as in at least one previous study (2), the amount of subcutaneous adipose tissue in the calf was inversely proportional to diabetes risk. In a logistic regression multivariate analysis, traditional risk factors of age and body mass index were most predictive, subcutaneous fat was protective, family history of diabetes was the most strongly predictive, and IMAT was modestly but significantly predictive of type 2 diabetes. Miljkovic-Gacic et al report that the positive association between fasting plasma glucose and IMAT was stronger in those persons with a family history of diabetes (P 0.02 for the interaction), whereas no such association was observed for subcutaneous fat. This unexpected and provocative finding may suggest a gene environment interaction. Alternatively, IMAT may be a better marker of diabetes risk in the presence of the multiple genetic determinants that characterize a strong family history of diabetes. The strong association of various measures of intermuscular and intramuscular lipid accumulation with insulin resistance, obesity, and type 2 diabetes has been known for at least a decade and has been shown by multiple investigators using methods ranging from muscle biopsy with total lipid extraction to oil red O staining to computerized tomography to magnetic resonance spectroscopy and magnetic resonance imaging (3, 4). Each of these methods, although likely measuring different proportions of intramuscular and intermuscular fat, nonetheless showed a similar association with insulin resistance and diabetes. The study of Miljkovic-Gacic et al specifically measured intermuscular fat, which is the small adipose depot interspersed between muscle fibers. The authors do not provide measures, such as attenuation scores, that may reflect intramyocellular fat. Although it may seem easier to link intramyocellular fat than intermuscular fat to impaired insulin action in muscle, even that measure is but a marker of high triacylglycerols and low fat oxidation in sedentary persons. Paradoxically, endurance exercise training increases intramyocellular fat in persons who are lean, insulin sensitive, and at low risk of type 2 diabetes (5). Whether IMAT is similarly increased with exercise training or whether it, like visceral obesity, represents a redistribution of adipose to less healthy locations is not clear. Plausible hypotheses exist for why visceral fat can alter whole-body insulin action, but why IMAT would do so is unclear. Like intramyocellular fat, IMAT may be a marker of excess caloric intake, particularly fat intake, and of a sedentary lifestyle, rather than being causative of metabolic syndrome. A key finding of the study of Miljkovic-Gacic et al was the interaction of family history and IMAT. That finding may point to a genetic interaction specific to this population that increases the risk of type 2 diabetes in persons with high IMAT, but many other explanations also are plausible. IMAT may be a marker of lipid partition away from subcutaneous depots such as hip and thigh, which appears to improve insulin sensitivity, and genetic factors may determine that distribution of fat. Alternatively, the interaction of family history and IMAT may be similar to the synergistic combination of obesity and family history, and IMAT may simply be a marker of caloric excess and a sedentary lifestyle, which, in the presence of a family history, are known to increase diabetes risk. Unfortunately, this cross-sectional epidemiology study provides few clues about diabetes pathogenesis in this very-high-risk cohort. Finally, whether the findings reported here are unique to this population or whether they, instead, reflect the high prevalence of obesity and the strong family history of diabetes in this older, male population is uncertain. To understand the interaction graphically, it would be helpful to see the IMAT by age curves according to family history. Studies that have compared IMAT among ethnic groups are limited. Many studies suggest that African American persons are more insulin resistant than are whites. Although the current study