Effect of regional fat loss assessed by DXA on insulin resistance and dyslipidaemia in obese women

Scand J Clin Lab Invest. 2010 Jul;70(4):229-36. doi: 10.3109/00365511003628328.


Objective: To investigate the impact of reduction in total fat mass (FM) and regional FMs on indices of insulin resistance and dyslipidaemia in obese women (BMI > 30 kg/m(2)) after a 1-year weight loss (WL) program; and, secondly, to investigate the potential predictive effect of baseline insulin resistance on reduction in total and regional FMs.

Material and methods: In 35 women with > 4 kg weight loss, body composition by DXA (dual X-ray absorptiometry), fasting insulin, C-Peptide, insulin resistance (HOMA-IR), insulin sensitivity (QUICKI), metabolic clearance rate (MCRestOGTT) and serum lipids were assessed.

Results: Mean WL was 9.6%; trunk and leg FM were reduced proportionally (14.9-14.7%). Improvement in HOMA-IR was 34.7%, insulin 30.7%, QUICKI 8.6% and MCRest OGTT 74%. The reduction in total, trunk and leg FM were similarly correlated with improvement in indices of insulin resistance (p < 0.001-0.05) and also with initial HOMA-IR (p = 0.000-0.02). In linear regressions improvement in HOMA-IR was similarly related with these FMs (p = 0.008), and initial HOMA predicted loss of trunk FM (p = 0.01). In multivariate analysis improvement in HOMA-IR was explained by loss of total FM (R(2) = 0.20, p = 0.004); improvement of QUICKI by loss of leg FM (R(2) = 0.33, p < 0.001).

Conclusion: Loss of leg FM and trunk FM had similar importance for the improvement in insulin resistance. Baseline HOMA-IR predicted the potential for reduction in trunk FM.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Absorptiometry, Photon*
  • Adiposity
  • Adult
  • Aged
  • Dyslipidemias / complications
  • Dyslipidemias / diagnostic imaging*
  • Dyslipidemias / epidemiology
  • Female
  • Humans
  • Insulin Resistance*
  • Middle Aged
  • Obesity / complications
  • Obesity / diagnostic imaging*
  • Obesity / epidemiology
  • Weight Loss*