RESUMO
BACKGROUND: Previous studies have reported that high-dose supplementation of n-3 polyunsaturated fatty acids (PUFAs) may reduce the risk of metabolic diseases, but there is limited evidence of an effect on body fat. We examined the associations of erythrocyte n-3 PUFAs with body fat and fat distribution in a general population consuming a normal diet. METHODS: This community-based cross-sectional study included 3075 Chinese (68% women, 40-75 years) recruited between 2008 and 2013. We collected general information and measured anthropometric indices; erythrocyte n-3 PUFAs (including α-C18:3, C20:5, C22:5 and C22:6) by gas-chromatography, and fat mass (FM) and %FM at the total body (TB), android (A) and gynoid (G) regions by dual-energy X-ray absorptiometry (DXA). RESULTS: Both minimally and maximally adjusted models showed dose-dependent inverse associations of total and individual levels of erythrocyte n-3 PUFAs (except C20:5 n-3[EPA]) with adiposity indices. In the full model, the mean differences between quartiles 4 and 1 of total n-3 PUFAs were -1.60% (BMI), -4.06% (TB FM), -5.38% (A FM), -2.05% (G FM), -2.05% (TB %FM), -3.39% (A %FM) and -2.50% (% A/G); the ORs (95% CI) of %FM-derived obesity (≥25% for men, ≥35% for women) for the highest (vs. lowest) quartile were 0.70 (0.57, 0.86) for total n-3 PUFAs and 0.71 (0.58, 0.87), 0.96(0.78, 1.18), 0.82(0.67, 1.00), 0.66 (0.54, 0.81) for α-C18:3/C20:5/C22:5/C22:6 n-3, respectively. The favourable associations were more pronounced for the DXA-derived FM indices, measurements at the android region and for C22:6 n-3. No significant associations between C20:5 n-3 and the adiposity indices were observed. CONCLUSIONS: Higher levels of circulating n-3 PUFAs were dose-dependently associated with better profiles of body fat and fat distribution, particularly in the abdominal regions in this population.