Epidemiological studies have established that pear-, compared to apple-, body shape is associated with lower cardiometabolic risk. Up to date, body shape has been considered equivalent to fat distribution and significant research efforts investigated the biology of different fat depots and their metabolic effects. We undertook a systematic analysis of body and tissue compartments in participants in the NHANES 1999-2006 cycles to search for components of body shape, other than fat. As expected, among adults 20-69 years old, those with pear-body shape (defined as waist-to-thigh ratio < 1.65 - bottom quartile) had higher leg fat mass compared to those with apple-body shape (waist-to-thigh ratio > 1.95 - top quartile) by 1,967 g (95% CI 1,801-2,135 g), after adjustment for sex, ethnicity, age and BMI. Surprisingly it was also noted that subjects with pear-body shape had slightly higher leg bone mineral content (by 42 g, 95% CI 30-52 g) and significantly higher leg lean mass (by 1,218 g, 95% CI 1,072-1,365 g). Similar results were observed when analysis was restricted to subjects < 40 years old, women only, or in specific ethnicities. In multiple regression models, leg lean mass (or leg/total lean mass ratio, i.e., lean mass distribution) was an independent predictor of lower systolic blood pressure, HbA1c and CRP after adjustment for sex, age, ethnicity, BMI, and leg fat mass (or leg/total fat mass, i.e., fat distribution), but not a predictor of indices of insulin sensitivity (fasting insulin, triglycerides, HDL-cholesterol).

In conclusion, fat distribution is not the sole determinant of pear-body shape in humans. Rather, coordinated differences in fat and lean mass drive metabolic health and suggest a systemic developmentally-driven pattern of lower body expansion.


K. Karastergiou: None.


National Center for Advancing Translational Sciences (KL2TR001435)

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.