Relationships between changes in body composition and changes in cardiovascular risk factors: the SOS Intervention Study. Swedish Obese Subjects.

Relationships between 2-year changes in body composition (estimated from computed tomography-validated anthropometry based on sagittal trunk diameter, weight, and height), adipose tissue (AT) distribution, and cardiovascular risk factors (blood pressure, lipids, glucose, insulin, uric acid) were examined in 842 treated adults with severe obesity with weight changes from -95.5 to +30.6 kg. Although the change (delta) of visceral AT mass (expressed in % total AT) for a given change in body mass index (delta BMI) was 6-fold larger in men than in women, delta waist and delta waist/hip were similar in both sexes. In men, risk factor changes were similarly related to delta waist, delta bodyweight, and delta BMI, whereas in women, delta bodyweight seemed to be the single independent variable with the highest explanatory power. In multivariate regressions adjusted for delta BMI and baseline conditions, delta visceral AT mass was more strongly associated with risk factor changes than were delta waist and delta waist/hip. When using a three-compartment model (lean body mass, subcutaneous and visceral AT masses) plus neck and thigh girths (indicators of subcutaneous AT distribution), risk factor changes were related both to delta subcutaneous and delta visceral AT masses but not to delta lean body mass. In agreement with cross-sectional findings, delta neck was positively and delta thigh was negatively related to some risk factor changes. Thus, the use of waist as a single risk factor indicator seems less effective for epidemiological studies than the simple anthropometric measures presented here, which are able to separate the effects of visceral AT mass, subcutaneous AT mass, and subcutaneous AT distribution on metabolic parameters under both cross-sectional and longitudinal conditions.