Adiponectin–leptin ratio for the early detection of lean non-alcoholic fatty liver disease independent of insulin resistance

Figures & data

Table 1. Baseline characteristics among the lean controls, lean NAFLD, simple overweight/obesity and overweight/obesity NAFLD groups.

	BMI < 24		BMI ≥ 24
	Lean controls	Lean NAFLD	Simple overweight/obesity	Overweight/obesity NAFLD
	N = 217	N = 105	N = 53	N = 200
Age (years)	41.20 ± 10.94	42.96 ± 11.59	44.38 ± 11.34	43.33 ± 11.80
Male (%)	45(20.7)^c,d	36(34.3) ^d	25(47.2) ^a	113(56.5) ^a, b
Smoke (%)	14(6.5)^d	11(10.5)	5(9.4)	33(16.5) ^a,
Exercise (%)	95(43.8)	45(42.9)	26(49.1)	86(43.0)
BMI (kg/m^2)	20.66 ± 1.80 ^b, c, d	21.83 ± 1.54 ^a, c, d	25.95 ± 1.75 ^a, b, d	27.98 ± 3.91 ^a, b, c
Fat percentage (%)	27.54 ± 8.08 ^b, c, d	30.76 ± 7.33 ^a, d	33.89 ± 7.12 ^a	36.15 ± 7.89 ^a, b
WC (cm)	73.13 ± 6.13 ^b, c, d	77.56 ± 6.58 ^a, c, d	85.72 ± 5.92 ^a, b, d	90.40 ± 7.88 ^a, b, c
Systolic BP	115.80 ± 15.52 ^b, c, d	122.12 ± 15.15 ^a, d	122.63 ± 17.16 ^a, d	129.99 ± 15.12 ^a, b, c
Diastolic BP	73.00 ± 11.18 ^b, c, d	77.50 ± 9.46 ^a, d	78.06 ± 13.91 ^a	81.75 ± 11.93 ^a, b
TCHO (mg/dL)	189.78 ± 33.99 ^d	197.60 ± 40.05	194.00 ± 29.23	202.64 ± 35.34 ^a
TG (mg/dL)	74.00 ± 35.43 ^b, d	109.36 ± 79.36 ^a, d	96.02 ± 43.21 ^d	157.51 ± 113.57 ^a, b, c
HDL-C (mg/dL)	66.78 ± 14.96 ^b, c, d	57.13 ± 13.35 ^a, d	59.09 ± 13.23 ^a, d	49.93 ± 12.61 ^a, b, c
LDL-C (mg/dL)	114.12 ± 31.13^b, d	126.20 ± 37.28 ^a	122.49 ± 29.21	132.86 ± 32.16 ^a
Glucose (mg/dL)	82.65 ± 8.66 ^d	85.50 ± 8.65 ^d	86.51 ± 9.83 ^d	89.39 ± 9.43 ^a, b, c
Insulin (μIU/mL)	5.13 ± 3.28 ^d	6.71 ± 5.24 ^d	7.15 ± 3.87 ^d	11.30 ± 9.03 ^a, b, c
HOMA-IR	0.66 ± 0.42 ^d	0.86 ± 0.65 ^d	0.92 ± 0.49 ^d	1.44 ± 1.11 ^a, b, c
GOT (U/L)	20.46 ± 6.86 ^d	21.75 ± 7.03 ^d	21.57 ± 6.00 ^d	25.40 ± 10.07 ^a, b, c
GPT (U/L)	17.22 ± 9.48 ^b, d	23.96 ± 16.61 ^a, d	12.48 ± 10.74 ^d	35.69 ± 27.84 ^a, b, c
CRP (mg/dL)	0.114 ± 0.320 ^d	0.091 ± 0.123 ^d	0.173 ± 0.277	0.207 ± 0.250 ^a, b
Metabolic factors (n)	0.39 ± 0.62 ^b, c, d	0.92 ± 0.90 ^a, d -	1.15 ± 0.91 ^a, d	2.01 ± 1.24 ^a, b, c
MetS (%)	1(0.5) ^d	6(6.2) ^d	4(8.5) ^d	56(28.0) ^a, b, c
Adiponectin(μg/mL)	18.13 ± 8.55 ^b, d	13.60 ± 8.00 ^a, d	15.01 ± 8.82 ^d	9.82 ± 5.71 ^a, b, c
Leptin (ng/mL)	8.16 ± 6.30 ^c, d	9.42 ± 7.21 ^d	12.48 ± 10.74 ^a	15.20 ± 11.35 ^a, b
AL ratio (x10^3)	6.43 ± 18.36 ^b. d	2.26 ± 1.93 ^a	2.23 ± 2.32	1.13 ± 1.14 ^a

Abbreviations: NAFLD: non-alcoholic fatty liver disease; BMI: body mass index; WC: waist circumference; BP: blood pressure; TCHO: total cholesterol; TG: triglycerides; HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol; HOMA-IR: homeostasis model assessment of insulin resistance; GOT: glutamic oxaloacetic transaminase; GPT: glutamic pyruvic transaminase; CRP: C-reactive protein; MetS: metabolic syndrome; AL ratio: adiponectin-leptin ratio.

Notes: Data are presented as the mean ± SD for continuous variables and number (percentage) for categorical variables. Differences between the four groups were examined using the chi-squared test for categorical variables and one-way analysis of variance (ANOVA) for continuous variables. Tukey’s post hoc analysis was applied to examine the differences among the healthy control, lean NAFLD, overweight controls and overweight NAFLD groups in terms of basic demographic characteristics, leptin, adiponectin and AL ratio.

The four groups were represented with ^a: lean controls; ^b: lean NAFLD; ^c: simple overweight/obesity; ^d: overweight/obesity NAFLD. If a significant level p < .05 was achieved between any two of the four groups, a superscript was added to the corresponded columns.

Table 2. Relation between the serum adiponectin-leptin ratio and metabolic factors in multivariate linear regression models after adjusting for age and sex.

Variables	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8	Model 9
	ß (SE) p Value	ß (SE) p Value	ß (SE) p Value	ß (SE) p Value	ß (SE) p Value	ß (SE) p Value	ß (SE) p Value	ß (SE) p Value	ß (SE) p Value
BMI (kg/m^2)	−0.252(0.108) <.001
Fat (%)		−0.365(0.060) <.001
WC (cm)			−0.296(0.050) <.001
SBP (mmHg)				−0.141(0.030) .001
DBP (mmHg)					−0.133(0.040) .002
HDL-C (mg/dl)						0.158(0.033) <.001
TG (mg/dl)							−0.165(0.005) <.001
Glucose (mg/dl)								−0.102(0.028) .015
HOMA-IR									−0.203(0.260) <.001

Abbreviations: NAFLD: non-alcoholic fatty liver disease; BMI: body mass index; WC: waist circumference; SBP: systolic blood pressure; DBP: diastolic blood pressure; TG: triglycerides; HDL-C: high-density lipoprotein cholesterol; LDL-C:low-density lipoprotein cholesterol; HOMA-IR: homeostasis model assessment of insulin resistance.

Table 3. Odds ratios of having NAFLD in relation to the serum tertile of adiponectin-leptin (AL × 10³) ratio using multiple logistic regression analyses.

	AL × 10^3 ratio <0.91 N = 190	0.91≤ AL ×10^3 ratio <2.36 N = 193	AL × 10^3 ratio ≥2.36 N = 192	p for trend
Model 1	1.00	0.28(0.17–0.44) **	0.07(0.04–0.12) **	<.001
Model 2	1.00	0.58(0.34–1.00) *	0.22(0.12–0.43) **	<.001
Model 3	1.00	0.66(0.37–1.12)	0.34(0.17–0.71) **	<.001
Model 4	1.00	0.67(0.38–1.21)	0.37(0.18–0.77) *	.008

Notes: Model 1: adjusted for age, sex, current smoking and exercise habits; Model 2: adjusted for variables in Model 1, plus BMI; Model 3: adjusted for variables in Model 2, plus HOMA_IR; Model 4: adjusted for variables in Model 2, plus GOT/GPT ratio.

* p < .05; ** p < .001.

Table 4. Odds ratios of having NAFLD in relation to serum tertile of adiponectin leptin level using multiple logistic regression analyses, stratification by BMI.

Lean NAFLD
	AL <0.91 N = 53	0.91 ≤ AL <2.36 N = 120	AL ≥2.36 N = 149	p for trend
Model 1	1.00	0.58(0.29–1.14)	0.16(0.08–0.36) **	<.001
Model 2	1.00	0.59(0.29–1.21)	0.26(0.11–0.61) *	.002
Model 3	1.00	0.62(0.30–1.29)	0.28(0.12–0.69) *	.005
Overweight NAFLD
	AL <0.91 N = 137	0.91 ≤ AL <2.36 N = 73	AL ≥2.36 N = 43	p for trend
Model 1	1.00	0.34(0.14–0.80) *	0.11(0.04–0.30) **	<.001
Model 2	1.00	0.59(0.23–1.51)	0.28(0.09–0.89) *	.031
Model 3	1.00	0.61(0.23–1.58)	0.30(0.09–0.96) *	.043

Notes: Model 1: adjusted for age, sex, current smoking and exercise habits; Model 2: adjusted for variables in Model 2, plus HOMA_IR; Model 3: adjusted for variables in Model 2, plus GOT/GPT ratio.

*p < .05; **p < .001.

Figure 1. Receiver operating characteristic (ROC) for the diagnosis of NAFLD. Except adiponectin-leptin ratio, BMI, triglyceride and AST/ALT ratio were selected. (A) All subjects, AUROC was 0.85 (95% CI: 0.82–0.88), (B) female subjects, AUROC was 0.83 (0.78–0.87), and (C) male subjects, AUROC was 0.86 (081–0.91). All p < .001.

Data availability statement

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.