Alanine Aminotransferase and Total Bilirubin Are Synergistically Associated with Metabolic Syndrome Among Middle-Aged and Elderly Japanese Women

Abstract

Background:

Metabolic syndrome (MetS) is associated with an increased risk of major cardiovascular events. Alanine aminotransferase (ALT) at high levels and total bilirubin (T-BiL) at low levels were oxidative potentials, but it was uncertain whether ALT and T-BiL had an additive interaction for the risk of MetS.

Methods:

From a single community, we recruited 864 women (70 ± 8 years) during their annual health examination. We cross-sectionally investigated whether ALT and T-BiL are associated with MetS and its components based on the modified criteria of the National Cholesterol Education Program's Adult Treatment Panel (NCEP-ATP) III report.

Results:

Of these subjects, 415 women (48.0%) had MetS. Participants with MetS had a higher ALT and lower T-BiL level than those without MetS. The adjusted-odds ratios (OR) (95% confidence interval [CI]) for MetS across tertiles of ALT and T-BiL were 1.00, 1.19 (0.78–1.81), and 1.86 (1.24–2.80) and 1.00, 0.96 (0.65–1.43), and 0.54 (0.36–0.81), respectively. When ALT and T-BiL were categorized into three binary characteristics by tertiles of ALT and T-BiL, high T-BiL was associated with decreased risk for MetS in a multivariable model (OR: 0.55, 95% CI: 0.37–0.82), especially among those with 1st tertile ALT. Similarly, high ALT was also associated with increased risk for MetS in a multivariate model (OR: 1.81, 95% CI: 1.20–2.71), especially among those with 2nd & 3rd tertiles of T-BiL. In the formal testing of addictive interaction between ALT and T-BiL for MetS, presence of T-BiL <0.72 mg/dL (1st and 2nd tertile) alone was not associated with increased risk of MetS in a multivariate analysis, and presence of ALT ≥16 IU/L (2nd and 3rd tertile) alone was not associated with increased risk of MetS.

Conclusions:

These results suggested that higher ALT and lower T-BiL levels were synergistically associated with MetS, independent of other confounding factors among Japanese women.

Introduction

Metabolic syndrome (MetS) is a cluster of cardiometabolic risk factors, such as visceral obesity, raised blood pressure, hypertriglyceridemia, low high-density lipoprotein cholesterolemia, and impaired fasting glucose.¹ It is known as a predisease state that leads to a fivefold increase in risk of type 2 diabetes, twofold the risk of developing cardiovascular disease (CVD), and all-cause mortality.² The prevalence of MetS has been increasing worldwide with the wake of urbanization, surplus energy intake, increasing obesity, and sedentary life habits.

Many studies have demonstrated that alanine aminotransferase (ALT) levels independently predict type 2 diabetes,^3,4 MetS,^3,5,6 and CVD.⁷ These markers have been shown to be associated with indirect measures of insulin resistance, including fasting insulin levels and homeostasis model assessment-insulin resistance (HOMA-IR).⁸

While serum bilirubin (BiL), a major intravascular product of heme catabolism, is an endogenous compound that can be toxic to infants under certain conditions like excessive production of bilirubin due to hemolysis,⁹ but in adults is a potent physiological antioxidant compound that may provide important protection against CVD^10,11 and inflammation¹² and suppresses oxidation of lipids and lipoprotein.¹³ Some studies focused on the association between serum BiL and MetS; an increase in serum BiL was inversely associated with development of MetS.^14
–16 However, in Japanese community-dwelling persons, there are few studies to demonstrate a relationship between ALT, BiL, and MetS, and it was uncertain whether ALT and BiL had additive interaction for the risk of MetS. It is important for us to be able to evaluate insulin resistance by measuring the liver markers which are inexpensively and routinely measured in clinical setting.

To address this hypothesis, we examined the cross-sectional relationship between ALT, BiL, MetS, and its components based on the modified criteria of the National Cholesterol Education Program's Adult Treatment Panel (NCEP-ATP) III report.¹⁷

Materials and Methods

Subjects

The present study was designed as part of the Nomura study.¹⁸ The study population aged ≥50 years was recruited through a community-based annual checkup process from the Nomura health and welfare center in a rural town located in Ehime prefecture, Japan. For all these individuals, overnight fasting plasma samples were made available. Participants with serum Total BiL (T-BiL) ≥2.0 mg/dL or ALT ≥100 IU/L or gamma glutamyltranspeptidase (GGT) ≥100 IU/L were excluded to avoid confounding factors due to the high possibility of potential Gilbert syndrome and hepatobiliary disease. Participants with an estimated glomerular filtration ratio (eGFR) of <15 mL/min/1.73 m² were also excluded. The study complies with the Declaration of Helsinki and was approved by the Ethics Committee of Ehime University School of Medicine with written informed consent obtained from each subject (institutional review board: 1402009).

Evaluation of risk factors

Information on demographic characteristics and risk factors was collected using clinical files. Body mass index (BMI) was calculated by dividing body weight (in kilograms) by the square of the height (in meters). Waist circumference was measured in centimeters without compression of the soft tissue at midway level between the lower rib margin and iliac crest using nonstretchable measuring tape. Other characteristics, such as smoking, alcohol habit, and medication, were investigated by individual interviews that were conducted using a structured questionnaire. Smoking status was defined as the number of cigarette packs per day multiplied by the number of years smoked (pack year), and the participants were classified into never smokers, past smokers, light smokers (<30 pack year), and heavy smokers (≥30 pack year). Daily alcohol consumption was measured using the Japanese liquor unit in which a unit corresponds to 22.9 g of ethanol, and the participants were classified into never drinkers, occasional drinkers (<1 U/day), and daily drinkers (light, <2 U/day, heavy, 2–3 U/day). We measured blood pressure (BP) with an appropriate-sized cuff on the right upper arm of the subjects in the sedentary position using an automatic oscillometric blood pressure recorder while they were seated after having rested for at least 5 min. Triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), hemoglobin A1c (HbA1c), serum uric acid (SUA), creatinine (Cr), GGT, ALT, and T-BiL were measured during an overnight fast of more than 11 hrs. eGFR was calculated using chronic kidney disease epidemiology collaboration (CKD-EPI) equations modified by a Japanese coefficient: Male, Cr ≤0.9 mg/dL, 141 × (Cr/0.9)^−0.411 × 0.993^age × 0.813; Cr >0.9 mg/dL, 141 × (Cr/0.9)^−1.209 × 0.993^age × 0.813; Female, Cr ≤0.7 mg/dL, 144 × (Cr/0.7)^−0.329 × 0.993^age × 0.813; Cr >0.7 mg/dL, 144 × (Cr/0.7)^−1.209 × 0.993^age × 0.813.¹⁹

Metabolic syndrome

We applied condition-specific cutoff points for MetS based on the modified criteria of the National Cholesterol Education Program's Adult Treatment Panel (NCEP-ATP) III report.¹⁷ MetS was defined as a subject with at least three or more of the following five conditions: (1) visceral obesity with a waist circumference ≥80 cm in women based on the adjusted Japanese waist circumference criterion²⁰; (2) raised BP with systolic blood pressure (SBP) ≥130 mmHg and/or diastolic blood pressure (DBP) ≥85 mmHg, and/or current treatment for hypertension; (3) increased triglyceridemia with a TG level ≥150 mg/dL; (4) low HDL cholesterolemia with a HDL-C <50 mg/dL in women, or current treatment for dyslipidemia; and (5) elevated fasting glucose with a HbA1c ≥5.6% (comparable with fasting plasma glucose [FPG] level ≥100 mg/dL) or current treatment for diabetes mellitus.

Statistical analysis

Statistical analysis was performed using IBM SPSS Statistics Version 21 (Statistical Package for Social Science Japan, Inc., Tokyo, Japan). All values are presented as the mean ± standard deviation (SD) unless otherwise specified, and in the cases of parameters with non-normal distributions (TG, ALT, GGT, HbA1c, and T-BiL), the data are shown as median (interquartile range) values. In all analyses, parameters with non-normal distributions were used after log transformation. Subjects were divided into three groups based on tertiles of ALT and T-BiL after log transformation. The differences among the groups categorized by presence of MetS and tertiles of ALT and T-BiL were analyzed by Student's t-test or ANOVA for continuous variables and χ ²-test for categorical variables. Multiple logistic regression analysis was used to evaluate the contribution of confounding factors for MetS and each component of MetS. The synergistic effect of ALT and T-BiL was evaluated using a general linear model adjusted for the following parameters: age, BMI, smoking status, alcohol consumption, history of CVD, LDL-C, SUA, eGFR, and GGT. A P value <0.05 was considered significant.

Results

Characteristics of subjects by presence of MetS

The characteristics of the subjects categorized according to presence of MetS are illustrated in Table 1. The study included 864 women (mean age, 70 ± 8; range, 50–90 years). In those with MetS, age, BMI, waist circumference, prevalence of exercise habits, history of CVD, SBP, DBP, prevalence of antihypertensive medication, TG, prevalence of antilipidemic medication, HbA1c, prevalence of diabetic medication, SUA, GGT, and ALT were significantly higher than those without MetS, but HDL-C, LDL-C, T-BiL, and eGFR were significantly lower. There were no differences in smoking and alcohol consumption.

Table 1.

Characteristics of Subjects

	MetS	Non-MetS
Characteristics N = 864	N = 415	N = 449	P value ^*
Age (years)	71 ± 7	69 ± 8	<0.001
Body mass index (kg/m²)	24.2 ± 3.0	21.1 ± 2.5	<0.001
Waist circumference (cm)	85.6 ± 7.9	75.6 ± 7.1	<0.001
Smoking status (%)	98.1/1.4/0.5/0	96.4/2.0/0.9/0.7	0.291
Alcohol consumption (%)	73.7/21.0/3.9/1.4	69.7/22.9/4.5/2.9	0.374
Exercise habits (%)	34.5	41.0	0.049
History of cardiovascular disease (%)	5.8	2.9	0.043
Systolic blood pressure (mmHg)	141 ± 15	132 ± 18	<0.001
Diastolic blood pressure (mmHg)	78 ± 9	75 ± 10	<0.001
Antihypertensive medication (%)	62.4	28.1	<0.001
Triglycerides (mM)	104 (75–143)	76 (59–99)	<0.001
HDL cholesterol (mM)	63 ± 15	74 ± 16	<0.001
LDL cholesterol (mM)	122 ± 29	127 ± 28	0.026
Antilipidemic medication (%)	51.6	9.1	<0.001
Hemoglobin A1c (%)	5.8 (5.7–6.1)	5.5 (5.4–5.8)	<0.001
Antidiabetic medication (%)	8.7	2.4	<0.001
Serum uric acid (μM)	5.0 ± 1.2	4.5 ± 1.0	<0.001
eGFR (mL/min/1.73 m²)	70.7 ± 11.4	73.1 ± 10.4	0.001
Gamma glutamyltransferase (IU/L)	18 (14–25)	15 (12–21)	<0.001
Alanine aminotransferase (IU/L)	18 (15–23)	16 (13–19)	<0.001
Total Bilirubin (mg/dL)	0.60 (0.50–0.75)	0.65 (0.50–0.83)	0.006

Data presented are mean ± standard deviation. Data for triglycerides, hemoglobin A1c, gamma glutamyltransferase, alanine aminotransferase, and total bilirubin were skewed and are presented as median (interquartile range) values and were log transformed for analysis. ^* P value from Student's t-test for continuous variables or from χ ²-test for categorical variables.

MetS, metabolic syndrome; HDL, high-density lipoprotein; LDL, low-density lipoprotein; eGFR, estimated glomerular filtration rate.

Significant values (P < 0.05) are presented in bold.

Characteristics of subjects by tertile of ALT and T-BiL

The characteristics of the subjects by tertile of ALT and T-BiL are illustrated in Table 2. BMI, waist circumference, prevalence of antihypertensive medication, TG, prevalence of antilipidemic medication, HbA1c, SUA, eGFR, and GGT were significantly higher in relation to the higher tertiles of ALT, but age was significantly lower in the higher tertiles. In contrast, HDL-C and LDL-C were significantly higher in relation to the higher tertiles of T-BiL, but prevalence of CVD was significantly lower in the higher tertiles.

Table 2.

Characteristics of Subjects by Tertiles of Alanine Aminotransferase and Total Bilirubin

Characteristics N = 864	Tertile of ALT				Tertile of T-BiL
	1^st	2nd	3rd		1st	2nd	3rd
	7–15	16–18	19–98 IU/L		0.20–0.53	0.54–0.71	0.72–1.78 mg/dL
	N = 361	N = 195	N = 308	P value	N = 294	N = 273	N = 297	P value
Age (years)	71 ± 8	71 ± 7	69 ± 8	0.004	70 ± 8	70 ± 8	70 ± 7	0.952
Body mass index (kg/m²)	22.0 ± 2.9	22.4 ± 3.1	23.3 ± 3.4	<0.001	22.6 ± 3.1	22.5 ± 3.1	22.5 ± 3.3	0.890
Waist circumference (cm)	78.9 ± 8.5	79.9 ± 8.4	82.4 ± 9.4	<0.001	80.4 ± 8.8	80.5 ± 8.5	80.3 ± 9.6	0.345
Smoking status (%)	97.5/1.4/0.6/0.6	97.9/1.5/0/0.5	96.4/2.3/1.3/0	0.471	97.3/1.0/1.4/0.3	96.3/2.6/0.7/0.4	98.0/1.7/0/0.3	0.430
Alcohol consumption (%)	72.9/21.3/4.7/1.1	72.3/23.1/3.6/1.0	69.8/22.1/3.9/4.2	0.135	70.4/21.4/6.1/2.0	74.0/22.0/2.6/1.5	70.7/22.6/3.7/3.0	0.363
Exercise habits (%)	40.2	40.5	33.4	0.138	39.8	38.5	35.4	0.521
History of cardiovascular disease (%)	5.3	2.6	4.2	0.324	4.8	6.6	1.7	0.013
Systolic blood pressure (mmHg)	135 ± 17	136 ± 18	138 ± 17	0.096	137 ± 18	137 ± 16	136 ± 18	0.537
Diastolic blood pressure (mmHg)	76 ± 9	76 ± 10	78 ± 10	0.080	77 ± 10	76 ± 9	77 ± 10	0.763
Antihypertensive medication (%)	41.8	39.5	51.0	0.016	44.6	46.2	43.1	0.764
Triglycerides (mg/dL)	84 (65–112)	85 (65–109)	91 (66–130)	0.026	87 (68–121)	84 (66–114)	87 (64–117)	0.249
HDL cholesterol (mg/dL)	68 ± 16	69 ± 17	68 ± 17	0.592	66 ± 16	68 ± 17	71 ± 17	0.006
LDL cholesterol (mg/dL)	127 ± 28	124 ± 27	123 ± 31	0.219	120 ± 28	126 ± 28	128 ± 30	0.006
Antilipidemic medication (%)	21.1	32.8	37.3	<0.001	32.3	27.1	29.0	0.384
Hemoglobin A1c (%)	5.6 (5.4–5.9)	5.7 (5.4–5.9)	5.8 (5.6–6.0)	<0.001	5.7 (5.5–6.0)	5.7 (5.5–5.9)	5.7 (5.4–5.9)	0.075
Diabetic medication (%)	3.9	4.6	7.8	0.071	6.8	5.5	4.0	0.334
Serum uric acid (μM)	4.6 ± 1.2	4.7 ± 1.1	4.9 ± 1.1	0.006	4.8 ± 1.2	4.7 ± 1.1	4.8 ± 1.1	0.705
eGFR (mL/min/1.73 m²)	70.3 ± 13.1	72.0 ± 9.5	73.7 ± 8.7	<0.001	71.1 ± 12.5	72.3 ± 9.6	72.4 ± 10.6	0.290
Gamma glutamyltransferase (IU/L)	14 (12–19)	20 (19–22)	22 (16–32)	<0.001	17 (13–23)	16 (13–23)	17 (13–24)	<0.355
Alanine aminotransferase (IU/L)	13 (11–14)	17 (16–18)	23 (20–28)	<0.001	16 (13–21)	16 (13–21)	17 (14–21)	0.693
Total bilirubin (mg/dL)	0.60 (0.50–0.78)	0.63 (0.50–0.80)	0.64 (0.50–0.81)	0.104	0.46 (0.40–0.50)	0.61 (0.60–0.68)	0.88 (0.80–1.00)	<0.001

Data presented are mean ± standard deviation. Data for triglycerides, fasting plasma glucose, gamma glutamyltransferase (GGT), alanine aminotransferase (ALT), and total bilirubin (T-BiL) were skewed and are presented as median (interquartile range) values, and were log transformed for analysis. ^* P value from ANOVA for continuous variables or from the χ²-test for categorical variables. Significant values (P < 0.05) are presented in bold.

The adjusted odds ratio for MetS and its components by tertiles of ALT and T-BiL

Table 3 shows prevalence and the risk for MetS and abnormalities of its components in relation to tertiles of ALT and T-BiL in 864 women. Among these women, 415 (48.0%) had MetS. Prevalence of MetS increased significantly in relation to tertile of ALT and decreased in relation to tertile of T-BiL. After adjustments for age, BMI, smoking status, alcohol consumption, exercise habits, history of CVD, LDL-C, SUA, eGFR, and GGT, the OR (95% CI) for MetS across tertiles of ALT and T-BiL were 1.00, 1.19 (0.78–1.81), and 1.86 (1.24–2.80) and 1.00, 0.96 (0.65–1.43), and 0.54 (0.36–0.81), respectively. The OR of ALT were significantly high for the MetS components of visceral obesity, hypertriglyceridemia, low HDL cholesterolemia, and impaired fasting glucose.

Table 3.

Adjusted Odds Ratios of Tertiles of ALT and T-BiL for Metabolic Syndrome and Its Components

	Tertile of ALT				Tertile of T-BiL
	1st	2nd	3rd		1st	2nd	3rd
	7–15	16–18	19–98 IU/L		0.20–0.53	0.54–0.71	0.72–1.78 mg/dL
Characteristics N = 864	N = 361	N = 195	N = 308	P value	N = 294	N = 273	N = 297	P value
Metabolic syndrome	137 (38.0)	88 (44.1)	190 (61.7)	<0.001	155 (52.7)	138 (50.5)	122 (41.1)	0.011
Odds ratios (95% CI)	1	1.19 (0.78–1.81)	1.86 (1.24–2.80)	0.011	1	0.96 (0.65–1.43)	0.54 (0.36–0.81)	0.004
Visceral obesity	161 (44.6)	94 (48.2)	183 (59.4)	<0.001	157 (53.4)	140 (51.3)	141 (47.5)	0.344
Odds ratios (95% CI)	1	0.91 (0.55–1.49)	0.83 (0.51–1.35)	0.749	1	0.91 (0.57–1.45)	0.71 (0.44–1.14)	0.340
Raised blood pressure	270 (74.8)	146 (74.9)	248 (80.5)	0.164	227 (77.2)	213 (78.0)	224 (75.4)	0.751
Odds ratios (95% CI)	1	0.93 (0.60–1.44)	1.26 (0.81–1.97)	0.432	1	1.07 (0.70–1.64)	0.94 (0.62–1.43)	0.835
Hypertriglyceridemia	35 (9.7)	18 (9.2)	52 (16.9)	0.007	42 (14.3)	24 (8.8)	39 (13.1)	0.110
Odds ratios (95% CI)	1	0.82 (0.44–1.55)	1.18 (0.68–2.06)	0.520	1	0.58 (0.33–1.02)	0.82 (0.49–1.36)	0.159
Low HDL cholesterolemia	109 (30.2)	73 (37.4)	141 (45.8)	<0.001	118 (40.1)	103 (37.7)	102 (34.3)	0.343
Odds ratios (95% CI)	1	1.33 (0.89–1.98)	1.64 (1.12–2.41)	0.038	1	1.03 (0.71–1.48)	0.85 (0.59–1.23)	0.555
Impaired fasting glucose	224 (62.0)	131 (67.2)	232 (75.3)	0.001	210 (71.4)	186 (68.1)	191 (64.3)	0.179
Odds ratios (95% CI)	1	1.23 (0.84–1.79)	1.75 (1.19–2.56)	0.015	1	0.83 (0.57–1.19)	0.70 (0.49–0.99)	0.142

Adjusted for age, body mass index, smoking status, alcohol consumption, exercise habit, history of cardiovascular disease, LDL cholesterol, serum uric acid, eGFR, and gamma glutamyltransferase. Data for gamma glutamyltransferase were skewed and log transformed for analysis. Significant values (P < 0.05) are presented in bold.

CI, confidence interval.

Associations of tertiles of ALT and T-BiL with MetS

When ALT and T-BiL were categorized into three binary characteristics by tertiles of ALT and T-BiL, high T-BiL was associated with decreased risk for MetS in a multivariable model (OR: 0.55, 95% CI: 0.37–0.82), especially among the 1st tertile of ALT (7–15 IU/L). Similarly, high ALT was also associated with increased risk for MetS in a multivariate model (OR: 1.81, 95% CI: 1.20–2.71), especially among those with 2nd & 3rd tertiles of T-BiL (0.54–1.78 mg/dL) (Table 4).

Table 4.

Adjusted Odds Ratios for Metabolic Syndrome of Subjects by Tertiles of ALT and T-BiL

		Tertile of T-BiL odds ratios (95% CI)
		1st	2nd	3rd
		0.20–0.53	0.54–0.71	0.72–1.78 mg/dL
Characteristics N = 864	N	N = 294	N = 273	N = 297	P value
Tertile of ALT
Total: 7–89 IU/L	864	1	0.96 (0.65–1.43)	0.55 (0.37–0.82)	0.005
1st: 7–15	361	1	0.73 (0.40–1.34)	0.27 (0.14–0.53)	<0.001
2nd: 16–18	195	1	1.22 (0.52–2.91)	0.88 (0.38–2.05)	0.750
3rd: 19–98	308	1	1.06 (0.53–2.13)	0.71 (0.36–1.39)	0.443

		Tertile of ALT odds ratios (95% CI)
		1st	2nd	3rd
		7–15	16–18	19–98 IU/L
		N = 361	N = 195	N = 308
Tertile of T-BiL
Total: 0.20–1.78 mg/dL	864	1	1.16 (0.76–1.77)	1.81 (1.20–2.71)	0.015
1st: 0.20–0.53	294	1	0.68 (0.33–1.41)	0.98 (0.45–2.14)	0.543
2nd: 0.54–0.71	273	1	1.22 (0.57–2.60)	2.10 (1.03–4.25)	0.115
3rd: 0.72–1.78	297	1	2.20 (1.02–4.75)	3.10 (1.49–6.46)	0.008

Synergistic effect of ALT and T-BiL on mean accumulating number of MetS components

In addition to their direct associations, we observed a synergistic effect between ALT and T-BiL (Fig. 1). In Figure 1, subjects were divided into three groups by tertile of ALT and T-BiL levels. Higher T-BiL was significantly associated with decreased number of MetS components only in those with 1st tertile of ALT. We assessed the statistical significance of the synergistic relationship using a general linear model with the following confounding factors: age, BMI, smoking status, alcohol consumption, exercise habits, history of CVD, LDL-C, SUA, eGFR, and GGT. The interaction between ALT and T-BiL was a significant and independent determinant for accumulation of the MetS components (F = 19.3, P < 0.001).

FIG. 1.

Synergistic effect of ALT and T-BiL. Mean accumulated number of metabolic syndrome components: visceral obesity, raised blood pressure, hypertriglyceridemia, low high-density lipoprotein cholesterolemia, and impaired fasting glucose. Study subjects were divided into three groups by tertile of ALT and T-BiL levels. The interaction between increased ALT and decreased T-BiL was a significant and independent determinant for accumulation of the MetS components (F = 19.3, P < 0.001). ALT, alanine aminotransferase; T-BiL, total bilirubin; MetS, metabolic syndrome.

Subgroup effects and interactive effects of ALT and T-BiL on MetS

In the formal testing of addictive interaction between ALT and T-BiL for MetS, presence of T-BiL <0.72 mg/dL (1st and 2nd tertile) alone was not associated with increased risk of MetS in the multivariate analysis. In the same way, presence of ALT ≥16 IU/L (2nd and 3rd tertile) alone was not associated with increased risk of MetS in a multivariate analysis. In contrast, copresence of both factors was associated with a 4.67-fold risk (95% CI, 2.85–7.66) of MetS in a univariate analysis and 3.59-fold risk (95% CI, 1.99–6.45) in a multivariate analysis compared to absence of both risk factors (Table 5).

Table 5.

Interactive Effect of ALT and T-BiL for Metabolic Syndrome

		Metabolic syndrome
			Nonadjusted		Multivariate adjusted
Characteristics N = 864	N	N (%)	Odds ratio (95% CI)	P value	Odds ratio (95% CI)	P value
ALT (1st) and T-BiL (3rd)	116	25 (21.6)	1		1
ALT (2nd–3rd) and T-BiL (3rd)	181	97 (53.6)	4.20 (2.47–7.14)	<0.001	2.68 (1.42–5.04)	0.002
ALT (1st) and T-BiL (1st–2nd)	245	112 (45.7)	3.07 (1.84–5.10)	<0.001	3.03 (1.67–5.49)	<0.001
ALT (2nd–3rd) and T-BiL (1st–2nd)	322	181 (56.2)	4.67 (2.85–7.66)	<0.001	3.59 (1.99–6.45)	<0.001

Multivariate adjusted for age, body mass index, smoking status, alcohol consumption, exercise habits, history of cardiovascular disease, LDL cholesterol, serum uric acid, eGFR, and gamma glutamyltransferase. Data for gamma glutamyltransferase were skewed and log transformed for analysis. Significant values (P < 0.05) are presented in bold.

Discussion

In middle-aged and elderly Japanese women, we determined the prevalence rate of MetS, as defined by the modified NCEP-ATP III criteria,²⁰ and examined the association between ALT, T-BiL, and MetS. MetS was common, occurring in 48.0% of women. The prevalence rate and OR of MetS increased significantly in relation to increased ALT and decreased T-BiL, even after adjusting for potential confounding factors. In addition, we demonstrated that there is an interaction between ALT and T-BiL for MetS, independent of confounding factors, including ALT and T-BiL. To our knowledge, this is the first study to indicate these associations of ALT and T-BiL with MetS in community-dwelling women.

In our study, higher ALT levels were positively associated with MetS, independent of other confounders. Several previous studies have demonstrated that increased ALT was associated with increased odds of MetS after adjusting for potential confounding factors.^21
–23 Of 11,573 Chinese, the prevalence of MetS was 37.3% in males and 45.8% in females, and ALT level, even within the reference range, was independently and positively associated with MetS.²³ Of 4541 participants, 826 MetS cases were reported. Moreover, time-varying changes in ALT and AST levels were independently and positively associated with the dose-dependent incidence of MetS.²⁴ Pei et al.²⁵ demonstrated that among all the various liver function tests, GGT (>16 U/L) in male and ALT (>21 U/L) in female were the best predictors for the development of MetS in healthy elderly. However, in these studies, T-BiL was not considered as a confounding factor for MetS or not associated with the development of MetS.²⁵

Whereas previous studies have consistently reported that serum BiL levels are inversely correlated with MetS.²⁶ We also demonstrated that T-BiL levels were negatively associated with MetS, independent of other confounders. In a Chinese cross-sectional study, including 1728 women aged ≥65 years, T-BiL was an independent predictor of MetS (OR: 0.910, 95% CI: 0.863–0.960; P = 0.001).²⁷ From a 4-year retrospective longitudinal observational study involving 6205 Korean men without MetS, 936 (15.1%) cases of onset of MetS were identified and its incidence decreased across the baseline T-BiL quartile categories (P < 0.001).²⁸ In the meta-analysis of seven cross-sectional studies, the pooled OR (95% CI) for MetS in a comparison of extreme tertiles of T-BiL levels were 0.70 (0.62–0.78), whereas no significant association was found for the pooled estimated relative risk between two prospective studies (0.57, 95% CI: 0.11–2.94).²⁹ Among 11,613 non-MetS participants with a baseline T-BiL level ≤2.0 mg/dL, 2439 (21.0%) cases of onset of MetS developed during 55,407 person-years of follow-up and increased T-BiL levels were positively associated with a higher risk of onset of MetS.³⁰ We believe that the apparent discrepancy regarding an association, if any, between BiL and MetS in the population can be largely explained by differences in the study populations (e.g., ethnicity, gender, different distribution of age, and co-morbidities) and study methodologies. In addition, we found that an interaction between ALT and T-BiL, as well as other confounding factors, including ALT and T-BiL, could also be a significant and independent determinant for MetS. Only in subjects with ALT <16 IU/L (1st tertile), tertile of T-BiL was negatively and significantly associated with decreased risk of MetS in the multivariate analysis.

The mechanisms by which increased ALT and decreased T-BiL reflect the risk for MetS are not completely understood. Fat accumulation in the liver or visceral adipose tissues can induce inflammatory cytokines such as tumor necrosis factor-α, interleukin-6, and interleukin-8.³¹ Systemic inflammation and oxidative stress, which are associated with insulin resistance and obesity, are closely involved in the pathogenesis of MetS.³² Thus, both increased ALT and decreased T-BiL may also reflect inflammation, which impairs insulin signaling in the liver, muscle, and adipose tissues.³³ Furthermore, it has been demonstrated that increased ALT together with decreased T-BiL levels might be a direct marker of oxidative stress or an antioxidant marker (defensive response) to oxidative stress, which is involved directly in the generation of reactive oxygen species. Kumar et al.³⁴ demonstrated that plasma BiL showed significant negative correlation with malondialdehyde but positive correlation with antioxidant enzyme activities (such as superoxide dismutase, catalase, and glutathione peroxidase levels).

There are some limitations to this study. First, our cross-sectional study design does not eliminate potential causal relationships between ALT, T-BiL, and MetS. Second, the prevalence rate of MetS, ALT, and T-BiL categories is based on a single assessment of blood, which may introduce a misclassification bias. Third, ALT levels vary even with the same alcoholic consumption; the possible association of fatty liver with the presence of MetS and with the elevated ALT could not be accurately explored. Fourth, we could not eliminate the possible effects of underlying diseases (e.g., liver disease) and medications used for hypertension and dyslipidemia on the present findings. In our study, participants with serum total bilirubin ≥2.0 mg/dL or ALT ≥100 IU/L or GGT ≥100 IU/L were excluded, but individuals with Gilbert's syndrome may have been included in the 3rd tertile (men, 0.72–1.78 mg/dL) group. Thus, we demonstrated that ALT and T-BiL are independently associated with MetS after adjusting for alcoholic consumption, suggesting that alcohol consumption does not dramatically affect the usefulness of ALT and T-BiL as a biomarker for MetS risk. Fifth, the presence of viral hepatitis must be considered; however, examinations of hepatitis B surface antigen and antibody to hepatitis C were not performed. Therefore the demographics and referral source may limit generalizability.

Conclusion

The present study showed that ALT and T-BiL levels are synergistically associated with MetS among middle-aged and elderly Japanese women. The underlying mechanism behind this relationship is unclear, but seems to be independent of traditional cardiovascular risk factors such as age, BMI, smoking status, alcohol consumption, exercise habits, history of CVD, LDL-C, SUA, eGFR, and GGT. For community-dwelling healthy persons, prospective population-based studies are needed to investigate the mechanisms underlying this association to determine whether intervention, such as effective lifestyle modifications or medication (e.g., antihypertensive, antilipidemic, and diabetic medication) that decrease ALT and increase T-BiL in adults, will decrease the risk of MetS.

Footnotes

Acknowledgment

This work was supported, in part, by a grant-in-aid for Scientific Research from the Foundation for Development of Community (2016).

Author Disclosure Statement

No competing financial interests exist for any of the authors.

References

Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive Summary of The Third Report of The National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, And Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III). JAMA, 2001; 285:2486–2497.

Alberti

KGMM

, Eckel

, Grundy

, et al. Harmonizing the metabolic syndrome: A joint interim statement of the international diabetes federation task force on epidemiology and prevention; National heart, lung, and blood institute; American heart association; World heart federation; International atherosclerosis society; And international association for the study of obesity. Circulation, 2009; 120:1640–1645.

Sattar

, Scherbakova

, Ford

, et al.; west of Scotland coronary prevention study. Elevated alanine aminotransferase predicts new-onset type 2 diabetes independently of classical risk factors, metabolic syndrome, and C-reactive protein in the west of scotland coronary prevention study. Diabetes, 2004; 53:2855–2860.

Ahn

, Shin

, Nam

, et al. The association between liver enzymes and risk of type 2 diabetes: The Namwon study. Diabetol Metab Syndr, 2014; 6:14.

Wannamethee

, Shaper

, Lennon

, et al. Hepatic enzymes, the metabolic syndrome, and the risk of type 2 diabetes in older men. Diabetes Care, 2005; 28:2913–2918.

, Bi

, Xu

, et al. Cross-sectional and longitudinal association of serum alanine aminotransaminase and γ-glutamyltransferase with metabolic syndrome in middle-aged and elderly Chinese people. J Diabetes, 2011; 3:38–47.

Monami

, Bardini

, Lamanna

, et al. Liver enzymes and risk of diabetes and cardiovascular disease: Results of the Firenze Bagno a Ripoli (FIBAR) study. Metabolism, 2008; 57:387–392.

Vozarova

, Stefan

, Lindsay

, et al. High alanine aminotransferase is associated with decreased hepatic insulin sensitivity and predicts the development of type 2 diabetes. Diabetes, 2002; 51:1889–1895.

Tiribelli

, Ostrow

. The molecular basis of bilirubin encephalopathy and toxicity: Report of an EASL Single Topic Conference. J Hepatol, 2005; 43:156–156.

10.

NovotnÝ

, Vítek

. Inverse relationship between serum bilirubin and atherosclerosis in men: A meta-analysis of published studies. Exp Biol Med (Maywood), 2003; 228:568–571.

11.

Erdogan

, Ciçek

, Kocaman

, et al. Increased serum bilirubin level is related to good collateral development in patients with chronic total coronary occlusion. Intern Med, 2012; 51:249–255.

12.

Kang

, Lee

, Kruzliak

. Effects of serum bilirubin on atherosclerotic processes. Ann Med, 2014; 46:138–147.

13.

Macdonagh

. The biliverdin-bilirubin antioxidant cycle of cellular protection: Missing a wheel?. Free Med Biol Med, 2010; 49:814–820.

14.

Lin

, Kuo

, Hwang

, et al. Serum bilirubin is inversely associated with insulin resistance and metabolic syndrome among children and adolescents. Atherosclerosis, 2009; 203:563–568.

15.

, Li

, Xu

, et al. Low serum total bilirubin concentrations are associated with increased prevalence of metabolic syndrome in Chinese. J Diabetes, 2011; 3:217–224.

16.

Choi

, Yun

, Choi

. Relationships between serum total bilirubin levels and metabolic syndrome in Korean adults. Nutr Metab Cardiovasc Dis, 2013; 23:31–37.

17.

Kahn

, Buse

, Ferrannini

, et al. American Diabetes Association; European Association for the Study of Diabetes. The metabolic syndrome: Time for a critical appraisal: Joint statement from the American Diabetes Association and the European Association for the Study of Diabetes. Diabetes Care, 2005; 28:2289–2304.

18.

Kawamoto

, Ninomiya

, Kumagi

. Handgrip strength is positively associated with mildly elevated serum bilirubin levels among community-dwelling adults. Tohoku J Exp Med, 2016; 240:221–226.

19.

Horio

, Imai

, Yasuda

, et al. Modification of the CKD epidemiology collaboration (CKD-EPI) equation for Japanese: Accuracy and use for population estimates. Am J Kidney Dis, 2010; 56:32–38.

20.

Doi

, Ninomiya

, Hata

, et al. Proposed criteria for metabolic syndrome in Japanese based on prospective evidence: The Hisayama study. Stroke, 2009; 40:1187–1194.

21.

Goessling

, Massaro

, Vasan

, et al. Aminotransferase levels and 20-year risk of metabolic syndrome, diabetes, and cardiovascular disease. Gastroenterology, 2008; 135:1935–1944.

22.

Kunutsor

, Seddoh

. Alanine aminotransferase and risk of the metabolic syndrome: A linear dose-response relationship. PLoS One, 2014; 9:e96068.

23.

Chen

, Guo

, Yu

, et al. Metabolic Syndrome and Serum Liver Enzymes in the General Chinese Population. Int J Environ Res Public Health, 2016; 13:223.

24.

Chen

, Xiao

, Zhang

, et al. Longitudinal Changes in Liver Aminotransferases Predict Metabolic Syndrome in Chinese Patients with Nonviral Hepatitis. Biomed Environ Sci, 2016; 29:254–266.

25.

Pei

, Hsia

, Chao

, et al. γ-glutamyl transpeptidase in men and alanine aminotransferase in women are the most suitable parameters among liver function tests for the prediction of metabolic syndrome in nonviral hepatitis and nonfatty liver in the elderly. Saudi J Gastroenterol, 2015; 21:158–164.

26.

Vitek

. The role of bilirubin in diabetes, metabolic syndrome, and cardiovascular diseases. Front Pharmacol, 2012; 3:55.

27.

Zhong

, Sun

, Wu

, et al. Serum total bilirubin levels are negatively correlated with metabolic syndrome in aged Chinese women: A community-based study. Braz J Med Biol Res, 2017; 50:e5252.

28.

Lee

, Jung

, Kang

, et al. Serum bilirubin as a predictor of incident metabolic syndrome: A 4-year retrospective longitudinal study of 6205 initially healthy Korean men. Diabetes Metab, 2014; 40:305–309.

29.

Nano

, Muka

, Cepeda

, et al. Association of circulating total bilirubin with the metabolic syndrome and type 2 diabetes: A systematic review and meta-analysis of observational evidence. Diabetes Metab, 2016; 42:389–397.

30.

Lee

, Lee

, Jun

, et al. Change in Serum Bilirubin Level as a Predictor of Incident Metabolic Syndrome. PLoS One, 2016; 11:e0168253.

31.

Fain

. Release of interleukins and other inflammatory cytokines by human adipose tissue is enhanced in obesity and primarily due to the nonfat cells. Vitam Horm, 2006; 74:443–477.

32.

Lee

, Park

. Pathway-driven approaches of interaction between oxidative balance and genetic polymorphism on metabolic syndrome. Oxid Med Cell Longev, 2017; 2017:6873197.

33.

Hotamisligil

. Inflammatory pathways and insulin action. Int J Obes Relat Metab Disord, 2003; 27 Suppl 3:S53–S55.

34.

Kumar

, Pant

, Basu

, et al. Oxidative stress in neonatal hyperbilirubinemia. J Trop Pediatr, 2007; 53:69–71.