Yamauchi et al. indicate a negative correlation between the circulating levels of dehydroepiandrosterone (DHEA) and dehydroepiandrosterone sulfate (DHEAS), and the concomitant blood glucose levels in Japanese patients with either type 2 (non-insulin-dependent) diabetes (NIDDM) or impaired glucose tolerance (IGT) (1). The authors fail to demonstrate a negative correlation between DHEAS and insulinemia even though earlier studies showed that an iv infusion of exogenous insulin marked decreased circulating DHEA and DHEAS concentrations (2, 3).
These interesting results have some limitations. Even though difficult to find, a control group with the same degree of obesity without IGT would have been appreciated. Information on the type of diet should have been given and the measurements of DHEA and DHEAS with the commercial kits have been criticized. For instance, Chasalow et al. (4) have indicated how uncertain results obtained with different kits may be often due to interference by other plasma steroids. Also, Yamauchi