Table 3.

Relationships Between the Number of Adipokine Abnormalities and Metabolic Characters in Obese Participants

Phenotype	Number of Adipokine Abnormalities				P Values
	0 (n=170)	1 (n=406)	2 (n=415)	3 (n=146)
Unadjusted
Male, %	35.8	40.8	32.4	32.9	0.044a
Age, y, mean±SD	10.1±2.5	11.5±2.9b	12.3±2.7b, c	12.6±2.9b, c	<0.001a
Pubertal stages, %					<0.001a
1	64.6	42.0	34.9	29.8
2	12.2	15.2	19.6	19.1
3	10.4	14.2	15.3	20.6
4	7.9	13.9	19.6	18.4
5	4.9	13.9	9.7	12.1
Residence, rural (%)	37.6	42.4	32.8	30.1	0.011a
Diet score, mean±SD	36.5±4.7	35.8±4.9	35.7±4.8	36.4±4.7	0.161
MVPA, %	61.0	52.6	52.1	52.8	0.251
BMI, kg/m2, mean±SD	24.1±2.9	26.0±3.5b	27.3±3.5b, c	28.3±3.8b, c, d	<0.001a
Adjusted (mean±SEM)e
Waist circumference, cm	76.1±0.70	81.7±0.5	86.3±0.5b, c	88.9±0.9b, c	<0.001a
SBP, mm Hg	108±0.8	113±0.6b	116±0.6b, c	119±1.1b, c	0.034a
DBP, mm Hg	69±0.7	71±0.4	73±0.4b, c	74±0.8b, c	0.054
Triglyceride, mmol/L	0.99±0.04	1.05±0.02	1.34±0.03b, c	1.48±0.06b, c	<0.001a
HDL‐C, mmol/L	1.35±0.02	1.28±0.01b	1.25±0.01b	1.23±0.02b	0.772
FBG, mmol/L	5.10±0.03	5.17±0.02	5.15±0.03	5.15±0.04	0.656
Fasting insulin, mU/Lf	2.07±0.04	2.38±0.03b	2.63±0.03b, c	2.79±0.06b, c, d	<0.001a
HOMA‐IRf	0.58±0.04	0.91±0.03b	1.15±0.03b, c	1.31±0.06b, c, d	<0.001a
Adipokines
RBP‐4, μg/mLf	3.35±0.02	3.46±0.02b	3.61±0.02b, c	3.85±0.02b, c, d	<0.001a
Osteonectin, μg/mLf	−0.24±0.02	−0.01±0.02b	0.23±0.02b, c	0.55±0.03b, c, d	<0.001a
Leptin/adiponectinf	0.12±0.06	0.78±0.05b	1.42±0.04b, c	1.82±0.05b, c, d	<0.001a
MS components (%)
Central obesity	77.6	76.1	84.3	92.5	0.050a
High BP	32.4	46.3b	52.5b	56.2b	0.043a
High triglyceride	21.8	26.6	42.9b	55.5b	<0.001a
Low HDL‐C	11.8	15.8	18.3	19.9	0.948
Hyperglycemia	12.4	16.0	15.4	15.1	0.846
HOMA‐IR ≥3	15.3	37.4b	51.8b	62.3b	<0.001a
MS	14.7	24.9	34.5b	45.9b	0.005a
MHO	38.8	25.1b	12.8b	7.5b	<0.001a

ANOVA (continuous variables) and χ² test (categorical variables) were used in unadjusted analysis, and data were expressed as percentage or mean±SD. GLM (continuous variables) and logistic regression analysis (categorical variables) were used in adjusted analysis, and data were expressed as mean±SEM or percentage. P values were for ANOVA and χ² test of difference in variables including age, sex, pubertal stages, residence, diet score, physical activity, and BMI or for GLM test and logistic regression analysis of other variables adjusted for age, sex, pubertal stages, residence, diet score, physical activity, and BMI. BMI indicates body mass index; BP, blood pressure; DBP, diastolic blood pressure; FBG, fasting blood glucose; GLM, general linear model; HDL‐C, high‐density lipoprotein cholesterol; HOMA‐IR, homeostasis model assessment for insulin resistance; MHO, metabolically healthy obesity; MS, metabolic syndrome; MVPA, moderate‐to‐vigorous physical activity; RBP‐4, retinol binding protein 4; SBP, systolic blood pressure.

^a P<0.05.

^bStatistical significance in a post hoc test after adjusting for age, sex, pubertal stages, residence, diet score, physical activity, and BMI, where differences from the nonadipokine abnormality group are indicated as P<0.05.

^cStatistical significance in a post hoc test after adjusting for age, sex, pubertal stages, residence, diet score, physical activity, and BMI, where differences vs the group with 1 adipokine abnormality are indicated as P<0.05.

^dStatistical significance in a post hoc test after adjusting for age, sex, pubertal stages, residence, diet score, physical activity, and BMI, where differences vs the group with 2 adipokine abnormalities are indicated as P<0.05.

^eAdjusted for age, sex, pubertal stages, residence, diet score, physical activity, and BMI.

^fSkewed distributions were natural logarithmically transformed.