Plasma Retinol-Binding Protein-4 Concentrations Are Elevated in Human Subjects With Impaired Glucose Tolerance and Type 2 Diabetes : Response to Cho et al.

In their recent study, Cho et al. (1) demonstrated higher levels of retinol-binding protein-4 (RBP4) in in-dividuals with impaired glucose tolerance and type 2 diabetes. Additionally, higher waist circumferences increased with plasma RBP4 quartiles. This was also demonstrated by Graham et al. (2). Cho et al. suggested that visceral fat may be more important than total body fat in determining circulating RBP4 levels. In our study (3), we investigated whether visceral adipose tissue (VAT) was associated with RBP4 levels in a group of nondiabetic subjects, in whom we also measured adiponectin levels. Serum RBP4 levels were measured by human enzyme-linked immunosorbent assay kits of the same lot number (lot no. 060622, cat no. 30-6110, expiration 24 January 2007; ALPCO Diagnostics, Salem, NH). Serum adiponcetin levels were measured by using a human adiponectin enzyme-linked immunosorbent assay kit (LINCO Research, St. Charles, MO). The coefficient of variation within each assay was <5%. VAT and abdominal subcutaneous adipose tissue (SAT) areas were measured at the fourth lumbar vertebra by computed tomography, as previously described. SAT and total adipose tissue were measured directly, and the VAT was calculated as the difference between the total adipose tissue and SAT. Insulin sensitivity was determined by a hyperinsulinemic-euglycemic clamp. In a study of 16 nondiabetic individuals with a mean ± SD BMI of 24.6 ± 0.8 kg/m², VAT content was inversely correlated to insulin sensitivity (r = −0.69, P = 0.004). Additionally, VAT was inversely correlated to adiponectin levels (r = −0.63, P = 0.009). Interestingly, there was no correlation between VAT and RBP4 levels (r = 0.27, P = 0.29). These data suggest that VAT may not be an important predictor of RBP4 levels. These preliminary findings should be confirmed in larger studies.