adipocyte size after rosiglitazone treatment in high- and low-fat fed rats.

We would like to thank Dr Cohen for his com-ments. We appreciate that rosiglitazone pro-motes adipocyte proliferation, with the cells pro-duced being of smaller size than mature adipocytes and having increased insulin sensitivity.1 Our study suggests that there is an overall reduction in the volume of visceral fat, even though the total numbers of adipocytes have increased. We are aware of, and accept, the potential role of 11-beta hydroxysteroid dehydrogenase-1 (11-beta HSD-1) described by Dr Cohen. Androgens also promote the differentiation of pluripotent cells to the myogenic lin-eage and inhibit their differentiation to adipocytes.2 Testosterone deficiency would thus promote adipocyte dif-ferentiation. Therefore rosiglitazone may have a dual mechanism of reduction of visceral fat volume. As ours was a brief report, there was insufficient space to include these points in the discussion. We thank Dr Cohen for raising these important issues. The purpose of our hypothesis for the hypogonadal-obesity-adipocytokine cycle,3 which is an extension of Dr Cohen’s hypogonadal-obesity cycle,4 is to explain why the hypothalamic-pituitary-testicular axis has an impaired response to low circulating testosterone levels. This hypoth-esis also goes some way to explaining why the majority of men with obesity, metabolic syndrome or type 2 diabetes have luteinising hormone (LH) levels within the normal range in the presence of low serum testosterone concentra-tions. Conflict of interest statement THJ has received honoraria for lectures from GSK