It is increasingly recognized that obesity as defined by body fat distribution as well as body mass index (BMI) contributes to metabolic disturbances. The relationship between body fat percent (%BF) and insulin resistance in the Asian population remains uncertain. We investigated the association between insulin dynamics and the four body composition phenotypes using BMI and %BF; normal with normal-fat percent (NN), normal with high-fat percent (NH), obese (overweight/obesity) with normal-fat percent (ON), and obese with high-fat percent (OH). We performed a cross-sectional study with 16,578 adult participants aged between 19 and 80 in the fourth and fifth KNHANES with Dual Energy X-ray Absorptiometry (DEXA) data. The subjects with diabetes are excluded. The mean age was 42.72 (standard error (SE) 0.30) years old. The prevalence of impaired fasting glucose, hypertension, and dyslipidemia was higher depending on the body composition phenotypes (p for trend <0.001). Homeostatic model assessment of insulin resistance (HOMA-IR) increased and the quantitative insulin check index (QUICKI) decreased significantly according to the body composition phenotypes (p for trend <0.001). The OH group exhibited significantly the highest HOMA-IR (mean 2.69, SE 0.03) and the lowest QUICKI (mean 0.335, SE 0.001) values. The odds ratio (OR) for high HOMA-IR (men >2.5, women >2.0) and low QUICKI (men <0.347, women <0.343) was increased according to the body composition phenotypes. NH group showed significantly higher OR than ON group, compared to NN group (OR 1.25, 95%CI 1.02-1.53 vs OR 0.91, 95% CI 0.70-1.19, respectively).
In conclusion, using body fat percent with BMI to categorize obesity is useful for estimating HOMA-IR and QUICKI. Although subjects have normal BMI, the amount of their fat mass plays an important role in insulin dynamics. The subjects with high-fat percent need to detect and intervene earlier for future prognosis.
A. Khang: None.
