RESUMO
The effect of fat deficit on bone metabolism is understudied. This study showed that low body fat percent (adipopenia) in non-underweight community adults was associated with elevated odds of osteoporosis independent of low lean mass, highlighting potential clinical importance of adipopenia as an osteoporosis risk factor particularly in older adults. PURPOSE: Although underweight is risk factor for osteoporosis, the association of low body fat percent (adipopenia) with osteoporosis in non-underweight adults remains unclear. METHODS: Among individuals aged ≥ 50 years with body mass index ≥ 18.5 kg/m2 in the Korean National Health and Nutrition Examination Survey 2008-2011, appendicular lean mass (ALM) and body fat percent (BFP) were measured using dual-energy x-ray absorptiometry. Low lean mass was defined as ALM/height2 < 7.0 kg/m2 and < 5.4 kg/m2 in men and women, respectively. Osteoporosis was defined as a bone mineral density (BMD) T-score ≤ -2.5 at the lumbar spine, femoral neck, or total hip. Participants were grouped into adipopenia (BFP < 17% in men; < 30%, in women; 1 standard deviation below the mean), normal, and obesity (BFP > 30% in men; > 40% in women) groups. RESULTS: Of the 5,830 participants (women 50.2%, mean age 63.9 years), 793 had adipopenia. The adipopenia group had a higher prevalence of osteoporosis (31%) than the normal (21%) or obesity groups (27%; p < 0.001). The presence of adipopenia was associated with 61% elevated odds of prevalent osteoporosis (p < 0.001) independent of low lean mass and covariates, which remained robust using different thresholds for adipopenia. Individuals with adipopenia and low lean mass had 3.5-fold elevated odds of osteoporosis compared to those with normal lean mass and fat percent. The association between adipopenia and osteoporosis was stronger in older women compared to middle-aged women (OR 1.93 vs. 0.99, P for interaction = 0.023). CONCLUSION: Adipopenia was associated with osteoporosis in non-underweight adults, independent of low lean mass and covariates.