Adult type 2 diabetes is caused by resistance in many tissues to the action of the hormone insulin and the failure of the pancreas to meet the need for more insulin. In non-diabetics without symptoms, insulin resistance is a strong predictor for the development of diabetes. In July last year Drs Qin Yang and Timothy Graham with colleagues from Harvard Medical School reported that blood levels of a protein called retinol-binding protein-4 (RBP4) produced by fat cells were raised in insulin-resistant mice and also in humans with obesity and adult type 2 diabetes. Injection of RBP4 in mice caused insulin resistance, whilst a genetic deletion to prevent RBP4 production increased insulin sensitivity.

The New England Journal of Medicine has now reported how the Harvard group and colleagues from San Diego in the US, Leipzig in Germany and Göteburg in Sweden have taken this work an important step further. The best way to detect insulin resistance is to give an infusion of glucose and insulin over several hours and change the amount of glucose given until the blood glucose concentration reaches a steady state for a constant insulin infusion rate – this is termed a “glucose clamp”. The teams found close agreement between glucose clamp measurements and the serum concentrations of RBP4 not only in patients with type 2 diabetes but also in subjects with conditions that are known to predispose to the development of frank diabetes. A simple measurement of RBP4 in blood could therefore be used to identify people at high risk of developing diabetes and indicate the need for preventive life-style changes.