Prediabetes Phenotypes and Adiposity Patterns: Findings From a Population-Based Study.
Prediabetes is a heterogeneous condition encompassing three glucose-defined phenotypes (isolated impaired fasting glucose [i-IFG], isolated impaired glucose tolerance [i-IGT], and IFG + IGT), with distinct pathophysiological mechanisms. This study assessed the associations of weight status (body mass index [BMI]), general adiposity (fat mass index [FMI]), total lean mass (lean mass index [LMI]), and body fat distribution (waist circumference [WC] and DEXA-derived appendicular, gynoid, abdominal and visceral adiposity) with prediabetes phenotypes.
This cross-sectional study included 3225 adults without diabetes who had complete data on fasting and 2-h plasma glucose, anthropometric measures (BMI and WC), and DEXA-derived measures (FMI, LMI, and percentages of total fat in appendicular, gynoid, abdominal and visceral regions) from the National Health and Nutrition Examination Survey 2011-2016. BMI and WC were classified according to WHO criteria. DEXA-derived measures were classified using sex-specific tertiles. Glycemic status was classified as normoglycemia, i-IFG, i-IGT, or IFG + IGT based on ADA criteria. Logistic regression and restricted cubic spline analyses were performed.
The weighted mean (SD) age was 37.16 (12.15) years, and 49.5% of participants were male. Overall, 59.6%, 29.1%, 4.2%, and 7.0% had normoglycemia, i-IFG, i-IGT, and IFG + IGT, respectively. Compared with individuals with normal BMI, those with overweight or obesity had higher odds of i-IFG (overweight: OR = 1.56 [1.19, 2.03]; obesity: OR = 2.73 [2.09, 3.56]) and IFG + IGT (overweight: OR = 2.81 [1.61, 4.91]; obesity: OR = 6.72 [4.03, 11.22]), whereas both underweight (OR = 3.22 [1.21, 8.58]) and obesity (OR = 2.57 [1.64, 4.04]) were associated with higher odds of i-IGT, indicating a U-shaped relationship between BMI and i-IGT (pnon-linearity = 0.006). Higher FMI and LMI were associated with higher odds of all three phenotypes (all pT3vs.T1 < 0.05). Compared with normal WC, very-high-risk central obesity was associated with higher odds of all three phenotypes (all p < 0.05). Higher proportions of abdominal or visceral fat and lower proportions of appendicular or gynoid fat were associated with higher odds of i-IFG and IFG + IGT (all pT3vs.T1 < 0.001). For i-IGT, only gynoid fat showed an inverse association (pT3vs.T1 = 0.002).
Adiposity patterns differed across prediabetes phenotypes. These findings provide insights for tailoring intervention strategies by prediabetes phenotype to optimize diabetes prevention.
This cross-sectional study included 3225 adults without diabetes who had complete data on fasting and 2-h plasma glucose, anthropometric measures (BMI and WC), and DEXA-derived measures (FMI, LMI, and percentages of total fat in appendicular, gynoid, abdominal and visceral regions) from the National Health and Nutrition Examination Survey 2011-2016. BMI and WC were classified according to WHO criteria. DEXA-derived measures were classified using sex-specific tertiles. Glycemic status was classified as normoglycemia, i-IFG, i-IGT, or IFG + IGT based on ADA criteria. Logistic regression and restricted cubic spline analyses were performed.
The weighted mean (SD) age was 37.16 (12.15) years, and 49.5% of participants were male. Overall, 59.6%, 29.1%, 4.2%, and 7.0% had normoglycemia, i-IFG, i-IGT, and IFG + IGT, respectively. Compared with individuals with normal BMI, those with overweight or obesity had higher odds of i-IFG (overweight: OR = 1.56 [1.19, 2.03]; obesity: OR = 2.73 [2.09, 3.56]) and IFG + IGT (overweight: OR = 2.81 [1.61, 4.91]; obesity: OR = 6.72 [4.03, 11.22]), whereas both underweight (OR = 3.22 [1.21, 8.58]) and obesity (OR = 2.57 [1.64, 4.04]) were associated with higher odds of i-IGT, indicating a U-shaped relationship between BMI and i-IGT (pnon-linearity = 0.006). Higher FMI and LMI were associated with higher odds of all three phenotypes (all pT3vs.T1 < 0.05). Compared with normal WC, very-high-risk central obesity was associated with higher odds of all three phenotypes (all p < 0.05). Higher proportions of abdominal or visceral fat and lower proportions of appendicular or gynoid fat were associated with higher odds of i-IFG and IFG + IGT (all pT3vs.T1 < 0.001). For i-IGT, only gynoid fat showed an inverse association (pT3vs.T1 = 0.002).
Adiposity patterns differed across prediabetes phenotypes. These findings provide insights for tailoring intervention strategies by prediabetes phenotype to optimize diabetes prevention.