Objective: To investigate the association of simple obesity measures (BMI, WC and WHR) with insulin secretion in obese subjects with normal glucose metabolism (NGM), prediabetes, and T2DM.

Research Design and Methods: 848 obese patients were recruited and divided into three groups (428 NGM, 248 prediabetes, and 172 T2DM). BMI, WC, and WHR were obtained. Insulin secretion was described by three beta cell function indices (HOMA-β, ΔI30/ΔG30 and AUCI60-120). Pearson correlation analysis and multiple linear regression analysis were used. In addition, 284 subjects estimated their body fat by dual x-ray absorptiometry. The joint relationship between obesity measures, body fat and insulin secretion were studied.

Results: As glucose metabolism spectrum changed from NGM to prediabetes and to T2DM, the correlation between WHR and insulin secretion declined (from r=0.286**, r=0.081, r=0.2** to r=-0.012, r=-0.072, r=-0.074 and to r=-0.173*, r=-0.175*, r=-0.171*; *P < 0.05, **P < 0.01). In contrast, BMI and WC showed a consistent association with insulin secretion (r>0). Similar relevant reversal was only found in female subjects. There were significant interactions of WHR and glucose status on insulin secretion in female subjects, in this adjusted model, WHR was associated with higher insulin secretion in NGM group (1.19±0.23**, 0.58±0.32, 0.82±0.26**) and lower in T2DM group (-0.77±0.34*, -1.78±0.47**, -1.36±0.39), not prediabetes (0.15±0.27, 0.08±0.38, 0.28±0.31). Adding HOMA-IR to the core model can increase the inverse associations. Elevated limb fat to trunk fat ratio had significantly lower insulin secretion levels, and the relationship between this fat ratio and WHR was considerably strong in female.

Conclusions: WHR and insulin secretion relationship is dependent on glucose status among obese women. As diabetes developed, higher WHR revealed lower insulin secretion, which may be attributed to more fat distributed in extremities.


J. Wu: None. B. Wang: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.