- Email: [email protected]
Influence of daily alcohol consumption on serum adiponectin levels in men
M ET ABOL I SM CL IN I CA L A N D EX PE RI ME N TA L 6 2 ( 2 0 13 ) 41 1–4 1 6
Available online at www.sciencedirect.com
Metabolism www.metabolismjournal.com
Influence of daily alcohol consumption on serum adiponectin levels in men Shinji Makita a,⁎, Akihiko Abiko a , Mizuyoshi Nagai a , Shinetsu Yonezawa b , Makoto Koshiyama b , Mutsuko Ohta b , Motoyuki Nakamura a a b
Department of Internal Medicine, Division of Cardiology, Iwate Medical University, Morioka 020-8505, Japan Iwate Health Service Association, Morioka, Japan
A R T I C LE I N FO Article history:
AB S T R A C T Background. The risk of cardiovascular diseases is lower among moderate alcohol
Received 20 June 2012
drinkers than among both nondrinkers and heavy drinkers. However, factors that can
Accepted 4 September 2012
account for the U-shaped or J-shaped relationship between daily alcohol consumption and incident cardiovascular diseases remain obscure.
Keywords:
Purpose. The present cross-sectional study investigated the relationship between alcohol consumption and serum adiponectin levels.
Alcohol Adiponectin Metabolic syndrome
Method. Total adiponectin was measured in 527 males participating in health check-up programs (age range 40–86 years, mean 60.5 years). Based on questionnaire responses, alcohol intake was categorized into three groups: none or occasional (A1); <50 g/day and ≥3 days/week (A2); and ≥50 g/day and ≥3 days/week (A3). Results. No significant differences in adiponectin levels were observed among the three alcohol consumption groups of subjects without the metabolic syndrome (MetS). In subjects with the MetS, the adiponectin level was significantly higher in the A2 (moderate drinker) group than in both the A1 and A3 groups. MetS subjects in group A2 had higher HDL-C levels than those in A1, but levels in group A3 were not significantly different from those in group A2. Conclusion. An increased adiponectin level in moderate alcohol drinkers who have MetS may contribute to the U-shaped relationship between alcohol consumption and risk of cardiovascular events, in addition to the involvement of HDL-C. © 2013 Elsevier Inc. All rights reserved.
1.
Introduction
The risk of cardiovascular diseases is known to be lower among light to moderate drinkers than among nondrinkers and heavy drinkers. This U-shaped or J-shaped relationship
has been documented between daily alcohol consumption and ischemic stroke [1–3], coronary heart disease [4–9], and all-cause mortality [10]. A recent meta-analysis also demonstrated that moderate alcohol consumption has beneficial effects on cardiovascular disease incidence and mortality [11].
Abbreviations: HDL-C, high-density lipoprotein cholesterol; MetS, metabolic syndrome; EDTA, ethylene diamine tetra acetic acid; IRI, immunoreactive insulin; CVs, coefficients of variation; ELISA, enzyme-linked immunosorbent assay; HOMA-IR, homeostasis model assessment of insulin resistance; eGFR, estimated glomerular filtration rate; HbA1c, glycosylated hemoglobin; HPLC, high performance liquid chromatography; JDS, Japan Diabetes Society; NGSP, National Glycohemoglobin Standardization Program; sBP, systolic blood pressure; dBP, diastolic blood pressure; LDL-C, low-density lipoprotein cholesterol; CRP, C-reactive protein. ⁎ Corresponding author. Tel.: +81 19 651 5111x2324; fax: +81 19 651 0401. E-mail address: [email protected] (S. Makita). 0026-0495/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.metabol.2012.09.003
412
M ET ABOL I SM CL IN I CA L A N D E XP E RI ME N TAL 6 2 ( 2 0 13 ) 41 1–4 16
There is evidence that light to moderate alcohol consumption may have beneficial effects on type 2 diabetes mellitus [12], an effect that appears to be mediated by decreased insulin resistance [13]. Moreover, light to moderate alcohol intake over the short term increases serum adiponectin [13,14] and high-density lipoprotein cholesterol (HDL-C) levels [15]. It has also been reported that the favorable effects of alcohol intake on glucose metabolism are most prominent in obese mice and in glucose-intolerant human subjects. We recently observed a U-shaped relationship between alcohol consumption and incident myocardial infarction in obese subjects from the general population [16]. However, mediating factors that can fully explain the U-shaped relationship between alcohol consumption and cardiovascular events remain unknown. In particular, the mechanism by which heavy alcohol consumption increases the risk of coronary heart disease has been not identified. The present cross-sectional study of the general population sought to investigate the relationship between daily alcohol consumption and glucose and lipid metabolism, as well as obesity-related indices, such as serum adiponectin concentrations and HDL-C levels. Furthermore, these relationships were analyzed in subjects with metabolic syndrome (MetS), who represent a segment of the population that is at high risk of cardiovascular diseases.
2.
Methods
2.1.
Study subjects
A total of 527 men underwent a comprehensive health checkup at the Iwate Health Service Association and agreed to participate in the present study. Participants with missing data and age <40 years were excluded. The mean age of the included participants was 60.5 years (range 40–86 years). All participants underwent a routine clinical examination, including a medical history, lifestyle assessment, and fasting blood sampling. Overall, 133 (25.2%) participants were on antihypertensive therapy, 58 (11.0%) participants used cholesterol-lowering therapies, and 41 (7.8%) participants were taking antidiabetic agents or were on a dietary regimen. The study protocols were approved by our institutional ethics committee, and all participants provided written informed consent.
2.2.
Blood examination
Following an overnight fast, venous blood samples were taken from the antecubital vein of subjects resting in the sitting position. Samples were collected in vacuum tubes containing EDTA or a serum separator gel. After sampling, the tubes were immediately centrifuged for 10 min at 1500×g. Aliquots of serum were stored at −20 °C, and routine hematology and biochemistry tests were performed within a few days following blood sampling. Some of the serum was also stored at −80 °C for measurement of adiponectin and insulin (immunoreactive insulin, IRI) levels. Total adiponectin concentration was measured using an ELISA kit. The intra- and inter-assay coefficients of variation (CVs) for measurement of adiponectin levels were 2.9% and 3.4%, respectively [17]. The
homeostasis model assessment of insulin resistance (HOMAIR; fasting blood glucose [mg/dl]×blood insulin concentration [μU/ml] / 405) was calculated as an index of insulin resistance [18]. The estimated glomerular filtration rate (eGFR) was calculated according to the equation proposed by the Japan Nephrology Society [19]. Glycosylated hemoglobin (HbA1c) was measured by the HPLC method as a Japan Diabetes Society (JDS) value, and the result is presented as a National Glycohemoglobin Standardization Program (NGSP) equivalent value (transformed by uniformly adding 0.4%).
2.3.
Risk factor assessment
Trained nurses measured the systolic and diastolic blood pressures (sBP and dBP, respectively) of patients who were seated and rested using an automatic digital sphygmomanometer placed on the upper arm. The average of two blood pressure measurements was used for the present analysis. A self-reported questionnaire was administered to document subjects’ medical history and lifestyle, including daily alcohol consumption. Alcohol consumption was evaluated by selecting from the 7 items of alcohol consumption status [the combination consisted of amount/day (<2 gou or ≥2 gou, where 2 gou =2 units of Japanese sake =50 g ethanol) and frequency in a week (<3 days, 3–6 days, or every day)] listed in Table 1. An ex-drinker was defined as a person who had stopped alcohol intake for at least 1 year prior. Regarding the time span of alcohol consumption, since the participants were involved in an annual health check-up program, the choice of the items for alcohol consumption status was derived from the estimated average over the previous approximately 1 year. For the analysis, daily alcohol consumption was categorized into three groups based on grams of ethanol consumed: none or occasional (A1); <50 g/day and ≥3 days/week (A2); and ≥50 g/ day and ≥3 days/week (A3). The smoking index was defined as the product of packs per day and years of smoking (packyears). Waist circumference was measured at the level of the umbilicus during normal exhalation, with the subject standing upright. When significant fat accumulation caused the umbilicus to sag downward, waist circumference measurements were taken at a level midway between the lower margin of the ribs and the anterior iliac spine. Hypertension was defined as at least one of sBP ≥140 mmHg, dBP ≥90 mmHg, or current antihypertensive therapy. Dyslipidemia was diagnosed in subjects with serum lowdensity lipoprotein cholesterol (LDL-C) levels ≥140 mg/dl and/ or high-density lipoprotein cholesterol (HDL-C) levels <40 mg/dl
Table 1 – Categories of daily alcohol consumption. Group A1
A2 A3
Alcohol consumption status
n
%
Ex-drinker Non-drinker Occasional (≤2 days /week) Ethanol <50 g/day, 3–6 days/week Ethanol <50 g/day , everyday Ethanol ≥50 g/day, 3–6 days/week Ethanol ≥50 g/day, everyday
12 84 109 80 150 29 63
39%
44% 17%
413
M ET ABOL I SM CL IN I CA L A N D EX PE RI ME N TA L 6 2 ( 2 0 13 ) 41 1–4 1 6
or in subjects with a history of cholesterol-lowering therapy. Diabetes mellitus was diagnosed according to three possible criteria: fasting blood glucose level ≥126 mg/dl, HbA1c level (NGSP) ≥6.5%, or a history of antidiabetic therapy. MetS was defined according to criteria established in Japan in 2005 [20]. Briefly, MetS was defined as an increased waist circumference (≥85 cm for men, ≥90 cm for women) together with at least two of the following three criteria: (1) triglycerides ≥150 mg/dl and/ or high-density lipoprotein cholesterol <40 mg/dl; (2) sBP ≥130 mmHg and/or dBP ≥85 mmHg; and (3) fasting blood glucose ≥110 mg/dl.
2.4.
Statistical analysis
Comparisons of demographic data according to categories of alcohol consumption were performed using n analysis of variance (ANOVA) with Scheffe's post hoc test. The χ2 test was used for comparison of categorical variables. The values in the tables are expressed as means±standard deviation. Ageadjusted comparisons of serum adiponectin, HDL-C, fasting IRI level, and waist circumference according to level of alcohol consumption were performed using analysis of covariance (ANCOVA) with Scheffe's post hoc test. The values and error bars in the graphs indicate estimated means±standard error. A p-value <0.05 was regarded as significant. All analyses were performed using the SPSS statistical package, version 19.0 (Chicago, IL, USA).
3.
Results
The distribution of subjects according to daily alcohol consumption status is presented in Table 1. Although ex-drinkers are potentially at higher risk of cardiovascular diseases, the 12 ex-drinkers in the present study did not have many cardiovascular risk factors. Therefore, these ex-drinkers were included in group A1. The clinical characteristics of subjects according to level of alcohol consumption are shown in Table 2. Group A2 had a lower fasting insulin level, but a higher HDL-C level than group A1. Subjects in group A3 were younger, more obese, and had higher blood pressures, uric acid levels, and smoking index than those in group A2. Age-adjusted relationships between alcohol consumption and adiponectin, HDL-C, fasting insulin, and waist circumference are shown in Fig. 1. No significant association was found between daily alcohol consumption and serum adiponectin level. The average waist circumference was significantly greater in group A3 than in group A2. No factors showing a V-shaped or inverted V-shaped relationship with level of alcohol consumption were identified in the analysis. The clinical characteristics of the subset of subjects with MetS according to the level of alcohol consumption are shown in Table 3. The HDL-C level was significantly higher in groups A2 and A3 than in group A1. The serum adiponectin level was significantly higher in group A2 than in the other groups. Ageadjusted relationships between alcohol consumption and serum adiponectin level or HDL-C level in subjects with or without MetS are shown in Fig. 2. Among subjects without MetS (non-MetS), no significant differences in adiponectin
Table 2 – Clinical characteristics of the three alcohol consumption groups (all subjects). A1 (n=205) Age, years Body mass index, kg/m2 Waist circumference, cm Systolic BP, mmHg Diastolic BP, mmHg Fasting blood glucose, mg/dl Fasting insulin, μU/ml HOMA-IR Hemoglobin A1c, % Serum Cr, mg/dl eGFR, ml/min/ 1.73 m2 Uric acid, mg/dl Triglyceride, mg/dl LDL-cholesterol, mg/dl HDL-cholesterol, mg/dl Smoking index, pack-year Total adiponectin, μg/ml Hypertension, % Diabetes mellitus, % Dyslipidemia, %
A2 (n=230)
A3 (n=92)
p-value
61.6±10.4 61.1±9.8 24.7±3.2 24.3±2.6
56.3±8.9 ⁎,⁎⁎ 25.2±3.0
<0.001 <0.05
88.6±8.6
88.0±7.1
90.8±8.0 ⁎⁎
<0.05
123±15 72±10 105±24
125±16 74±10 106±18
128±14 ⁎ 77±10 ⁎,⁎⁎ 107±17
<0.05 <0.001 ns
7.5±5.2
6.4±4.0 ⁎
6.7±4.5
<0.05
2.0±1.6 5.6±0.8
1.7±1.2 5.5±0.7
1.8±1.4 5.5±0.6
ns ns
0.86±0.16 0.86±0.36 0.80±0.11 ns 69.3±13.4 71.4±14.1 75.8±13.2 ⁎,⁎⁎ <0.01 5.9±1.2 126±71 127±27
5.9±1.4 128±121 120±30
6.4±1.3 ⁎,⁎⁎ 149±97 115±32 ⁎
<0.01 ns <0.01
50±12
57±14 ⁎
60±14 ⁎
<0.001
17±21
17±18
30±25 ⁎,⁎⁎
<0.001
7.2±4.2
7.6±4.0
6.2±2.8 ⁎⁎
<0.05
37.1% 13.7% 39.0%
39.6% 16.1% 38.7%
59.8% 13.0% 44.6%
<0.01 ns ns
BP; blood pressure, HOMA-IR; homeostasis model assessment of insulin resistance, eGFR; estimated glomerular filtration rate, LDL; low-density lipoprotein, HDL; high-density lipoprotein. ⁎ p<0.05 vs. A1. ⁎⁎ p<0.05 vs. A2.
Adiponectin (µg/ml) 8.0
ns
ns
ns 60
7.0 6.0
50
5.0
40
A1
HDL-C (mg/dl)
70
A2
A3
p<0.01
A1
Fasting insulin (µU/ml)
A2 Waist (cm)
92.0
8.0
p<0.05
p<0.05
90.0
ns
ns
7.0
A3
88.0 6.0
86.0
A1
A2
A3
A1
A2
A3
Fig. 1 – Relationships of alcohol consumption to glucose and lipid indices and waist circumference in age-adjusted analyses.
414
M ET ABOL I SM CL IN I CA L A N D E XP E RI ME N TAL 6 2 ( 2 0 13 ) 41 1–4 16
Table 3 – Clinical characteristics of the three alcohol consumption groups (subjects with metabolic syndrome). A1 (n=53)
A2 (n=49)
A3 (n=37)
p-value <0.05
26.0±2.1
57.1± 9.4 ⁎⁎ 26.8±2.2
95.4±7.3
93.3±5.4
95.2±5.8
ns
130±13 76±10 115±19
132±14 79±11 122±23
131±11 79±10 115±18
ns ns ns
10.7±7.5
8.7±5.0
9.0±5.6
ns
3.1±2.4 5.8±0.7 0.86±0.14
2.6±1.6 5.9±0.8 0.85±0.15
2.6±1.7 5.8±0.6 0.79± 0.11 ⁎
ns ns <0.05
69.3±12.2
70.0±13.6
77.6±14.4
<0.01
6.0±1.2 160±77
5.8±1.4 189±224
6.5±1.4 167±109
ns ns
124±27
114±25
123±28
ns
45±11
54±17 ⁎
55±12 ⁎
<0.001
13±18
17±19
30±27 ⁎,⁎⁎
<0.01
5.7±3.4
7.3±3.1 ⁎
5.7±2.5 ⁎⁎
<0.05
66.0% 30.2% 64.2%
75.5% 44.9% 61.2%
83.8% 27.0% 54.1%
ns ns ns
Age, years
61.9±9.8
62.6±10.1
Body mass index, kg/m2 Waist circumference, cm Systolic BP, mmHg Diastolic BP, mmHg Fasting blood glucose, mg/dl Fasting insulin, μU/ml HOMA-IR Hemoglobin A1c, % Serum Cr, mg/dl
27.1±3.0
eGFR, ml/min/1.73 m2 Uric acid, mg/dl Triglyceride, mg/dl LDL-cholesterol, mg/dl HDL-cholesterol, mg/dl Smoking index, pack-year Total adiponectin, μg/ml Hypertension, % Diabetes mellitus, % Dyslipidemia, %
ns
BP; blood pressure, HOMA-IR; homeostasis model assessment of insulin resistance, eGFR; estimated glomerular filtration rate, LDL; low-density lipoprotein, HDL; high-density lipoprotein. ⁎ p<0.05 vs. A1. ⁎⁎ p<0.05 vs. A2.
non-MetS
ns
8.0
MetS
ns
Mean level in non-MetS
Adiponectin (µg/mL) 7.0 Age-adjusted
p<0.05
p<0.05
6.0 5.0
A1
A2
70 HDL-C (mg/dL)
60
Age-adjusted
50
A3
A1
A2
A3
p<0.01 p<0.01
ns
p<0.01
Mean level in non-MetS
40 A1
A2
A3
A1
A2
A3
Fig. 2 – Influence of alcohol consumption on adiponectin and high-density lipoprotein cholesterol (HDL-C) in subjects with and without metabolic syndrome (MetS).
level were observed among the three alcohol consumption groups, despite a linear increase in HDL-C. In MetS subjects, while the adiponectin level was lower than in non-MetS subjects, only group A2 had an adiponectin level comparable to the mean level in non-Mets subjects. These significant differences remained with a multivariate adjusted model including age, eGFR, waist circumference, and smoking index, all of which potentially affect the adiponectin level (p for trend=0.010, p=0.009 for group A1 vs. A2, p=0.010 for group A2 vs. A3, p=0.73 for group A1 vs. A3). On the other hand, MetS subjects in group A2 had higher levels of HDL-C than those in A1, but HDL-C levels in group A3 were not significantly different from those in group A2. The p-values for the interaction between alcohol consumption and MetS were 0.12 for the adiponectin level and 0.50 for the HDL-C level. The post hoc power analysis of ANCOVA for adiponectin levels in the age-adjusted model revealed a statistical power (1−β) of 0.41 for all subjects, 0.13 for the non-MetS subgroup, and 0.68 for the MetS subgroup. The analysis for HDL-C levels showed that 1−β was 1.00, 1.00, and 0.97 for all subjects, the non-MetS subgroup, and the MetS subgroup, respectively.
4.
Discussion
Previous reports suggested that the reduction in ischemic heart disease risk attributed to moderate alcohol consumption may be explained by the beneficial effect of alcohol on the thrombolytic profile [21], insulin sensitivity [22], inflammation, as reflected by the C-reactive protein (CRP) level [23], and fibrinogen concentrations [24]. However, the increased risk of coronary heart disease among heavy drinkers could not be explained by all of these factors. In an effort to explain the U-shaped or J-shaped relationship between cardiovascular disease risk and amount of daily alcohol consumption, the present study examined the associations between various clinical indices, including serum adiponectin and HDL-C levels, and the level of daily alcohol consumption. In the analysis of the overall sample, no factors were identified to explain this relationship. However, among subjects with MetS, who are at high risk for cardiovascular disease, moderate drinkers had a higher serum adiponectin level than non- or occasional drinkers, as well as heavy drinkers. These results suggest that an increased adiponectin level in moderate alcohol drinkers may contribute to the U-shaped relationship between alcohol consumption and risk of cardiovascular events, at least among those at greatest risk for cardiovascular disease. A limited number of controlled trials have investigated the effect of alcohol consumption on the serum adiponectin level [13,14,25,26]. Sierksma et al. reported that plasma adiponectin concentration increased by 11% following 17 days of drinking whisky versus water. These authors concluded that moderate alcohol consumption improved insulin sensitivity in relatively insulin-resistant middle-aged men, and they suggested that the effect may be mediated by an increased adiponectin level [13]. Bullen et al. showed that 4 weeks of moderate alcohol consumption increased plasma adiponectin concentration in men with abdominal obesity, whereas fat distribution and ISI were unaffected [26]. Although these results are in general
M ET ABOL I SM CL IN I CA L A N D EX PE RI ME N TA L 6 2 ( 2 0 13 ) 41 1–4 1 6
agreement with the present findings, differences in the amount of alcohol intake were not examined in their study. Specifically, it was observed that the serum adiponectin level was significantly lower in heavy drinkers than in light to moderate drinkers. This observation may support the theory that serum adiponectin mediates the clinically favorable effects of alcohol consumption. Pischon et al. showed a significant nonlinear association between plasma adiponectin concentrations and alcohol intake in a similar study of men [27]. While the subjects with moderate alcohol intake (5.0–29.9 g/day) had a higher adiponectin level than non-drinkers, the increase in the adiponectin level was lower in heavy drinkers (≥30 g/day, p=0.10 vs. nondrinkers). The present data on the behavior of the adiponectin level are fundamentally consistent with this report. However, significant differences were not confirmed between moderate drinkers and heavy drinkers, and only the overall analysis was shown in their study. Gigleux et al. reported that light to moderate alcohol consumption (>15 g/day) had significant cardioprotective effects that were most apparent in men with MetS [28]. In the present study, an inverted-V-shaped relationship was observed between the serum adiponectin level and the amount of alcohol consumption in MetS subjects. Although the present observations may possibly explain the results reported by Gigleux et al., this latter study did not include information on heavy drinkers. Furthermore, anti-inflammatory effects may also contribute to the U-shaped relationship between alcohol consumption and ischemic heart disease. For instance, Imhof et al. showed that light to moderate daily alcohol intake (up to 20 or 40 g/day) was associated with lower CRP and white blood counts compared to not drinking, as well as heavy drinking, even after adjustment for potential confounders [23]. Although the CRP level was not measured in the present study, a strong correlation has been established between high sensitivity CRP and serum adiponectin levels. An acute increase in HDL-C levels in response to alcohol intake has been shown in a controlled study [13] and a metaanalysis [15]. Additionally, a linear increase in the HDL-C level was documented in association with alcohol consumption ranging from zero to >21 drinks/week (averaged >3 drinks/ day) [29]. Although the linear increase in HDL-C was blunted in heavy drinkers in the present study, their HDL-C level was similar to that of moderate drinkers. Thus, we postulated that the HDL-C level alone did not mediate the increased risk of cardiovascular diseases in heavy drinkers. Alternatively, it is possible that the increased risk of cardiovascular diseases in heavy drinkers may be produced by a combination effect with increased blood pressure. The strength of the present study may be that both the amount and frequency of alcohol consumption were reflected in the categorization. Furthermore, the sample size for evaluation of the adiponectin level was relatively large and comparable to the study of Pischon et al., which had the largest sample size in the previous literature. Nevertheless, the greatest weakness of the present study was its lack of statistical power. Since 1−β for the analysis of the adiponectin level was quite low, especially for the overall subjects and the non-MetS subgroup, findings of no association between alcohol consumption and adiponectin levels in these subjects
415
cannot be statistically guaranteed because of the high probability of type II error. Other limitations of this cross-sectional study warrant consideration. First, medications for hypertension, dyslipidemia, and diabetes that may affect the serum adiponectin level were not taken into consideration. Second, the relationships between alcohol consumption and cardiovascular risk can be altered by the chosen cutoff level for ‘light’ or ‘moderate’ alcohol consumption. It is unclear whether the cutoff used (<50 g/day and >3 days/week) was appropriate for delineating the mild to moderate alcohol consumption category. Although the actual average alcohol consumption cannot be calculated in the present study population, the average alcohol consumption of group A2 of the present study can be estimated as approximately 20–25 g/day based on a similar population involved in our large cohort study [16]. Nevertheless, a metaanalysis revealed that ‘moderate’ consumption of alcohol (up to two drinks or 30 g alcohol a day for men), a level of alcohol consumption that is similar to the cutoff level in the current study, has beneficial effects on a variety of biomarkers linked to the risk of coronary heart disease [15]. Third, drinking patterns were not taken into consideration. An inaccurate estimate of daily alcohol consumption cannot be excluded due to self-reporting of consumption in the present study. In addition, the present study did not investigate the kinds of beverages being consumed. However, beer, wine, and spirits are generally known to have similar effects on HDL-C [21]. Fourth, since female subjects were excluded from the analysis due to their relatively low alcohol consumption level, sex differences remain unclear. Furthermore, many previous investigations showed the U-shaped association of alcohol consumption with risk of cardiovascular disease in many populations, not merely those with MetS. A larger investigation may be needed to fill these gaps. Finally, causality cannot be assumed in the relationship with adiponectin because of the cross-sectional design of the present study. In addition, since both alcohol intake and an increased adiponectin level can potentially improve the features of MetS via increasing the HDL-C level and improving insulin sensitivity, respectively, stratification by MetS can introduce a bias. In conclusion, an increase in the serum adiponectin level in moderate drinkers and a decrease in the serum adiponectin level in heavy drinkers were observed among persons with MetS, a group that is at high risk for cardiovascular disease. Differences in serum adiponectin levels may be associated with the U-shaped relationship between daily alcohol consumption and cardiovascular disease risk, in addition to the involvement of HDL-C. From the translational perspective, evaluation of adiponectin levels may assist in setting the appropriate amount of alcohol consumption to prevent cardiovascular disease in alcohol drinkers.
Author contributions Shinji Makita, MD (statistical analyses, writing the paper); Akihiko Abiko, MD (collecting data); Mizuyoshi Nagai, MD (collecting data); Shinetsu Yonezawa (collecting data); Makoto Koshiyama (collecting data); Mutsuko Ohta (collecting data); Motoyuki Nakamura, MD (critical review of the paper).
416
M ET ABOL I SM CL IN I CA L A N D E XP E RI ME N TAL 6 2 ( 2 0 13 ) 41 1–4 16
Funding The authors did not receive any funding.
Conflict of interest The authors report no conflicts of interest.
REFERENCES
[1] Iso H, Baba S, Mannami T, Sasaki S, Okada K, Konishi M, et al. Alcohol consumption and risk of stroke among middle-age men: the JPHC Study Cohort I. Stroke 2004;35:1124–9. [2] Reynolds K, Lewis B, Nolen JD, Kinney GL, Sathya B, He J. Alcohol consumption and risk of stroke: a meta-analysis. JAMA 2003;289:579–88. [3] Berger K, Ajani UA, Kase CS, Gaziano JM, Buring JE, Glynn RJ, et al. Light-to-moderate alcohol consumption and risk of stroke among U.S. male physicians. N Engl J Med 1999;341:1557–64. [4] Yano K, Rhoads GG, Kagan A. Coffee, alcohol and risk of coronary heart disease among Japanese men living in Hawaii. N Engl J Med 1977;297:405–9. [5] Langer RD, Criqui MH, Reed DM. Lipoproteins and blood pressure as biological pathways for effect of moderate alcohol consumption on coronary heart disease. Circulation 1992;85: 910–5. [6] Shaper AG, Wannamethee G, Walker M. Alcohol and coronary heart disease: a perspective from the British Regional Heart Study. Int J Epidemiol 1994;23:482–94. [7] Gordon T, Kannel WB. Drinking habits and cardiovascular disease: the Framingham Study. Am Heart J 1985;110:331–4. [8] Miller GJ, Beckles GL, Maude GH, Carson DC. Alcohol consumption: protection against coronary heart disease and risks to health. Int J Epidemiol 1990;19:923–30. [9] Kitamura A, Iso H, Sankai T, Naito Y, Sato S, Kiyama M, et al. Alcohol intake and premature coronary heart disease in urban Japanese men. Am J Epidemiol 1998;147:59–65. [10] Emberson JR, Shaper AG, Wannamethee SG, Morris RW, Whincup PH. Alcohol intake in middle age and risk of cardiovascular disease and mortality: accounting for intake variation over time. Am J Epidemiol 2005;161:856–63. [11] Ronksley PE, Brien SE, Turner BJ, Mukamal KJ, Ghali WA. Association of alcohol consumption with selected cardiovascular disease outcomes: a systematic review and metaanalysis. BMJ 2011;342:d671. [12] Howard AA, Arnsten JH, Gourevitch MN. Effect of alcohol consumption on diabetes mellitus: a systematic review. Ann Intern Med 2004;140:211–9. [13] Sierksma A, Patel H, Ouchi N, Kihara S, Funahashi T, Heine RJ, et al. Effect of moderate alcohol consumption on adiponectin, tumor necrosis factor-alpha, and insulin sensitivity. Diabetes Care 2004;27:184–9. [14] Imhof A, Plamper I, Maier S, Trischler G, Koenig W. Effect of drinking on adiponectin in healthy men and women: a randomized intervention study of water, ethanol, red wine, and beer with or without alcohol. Diabetes Care 2009;32: 1101–3.
[15] Brien SE, Ronksley PE, Turner BJ, Mukamal KJ, Ghali WA. Effect of alcohol consumption on biological markers associated with risk of coronary heart disease: systematic review and meta-analysis of interventional studies. BMJ 2011;342: d636. [16] Makita S, Onoda T, Ohsawa M, Tanaka F, Segawa T, Takahash T, et al. Influence of mild-to-moderate alcohol consumption on cardiovascular diseases in men from the general population. Atherosclerosis 2012;224:222–7. [17] Makita S, Abiko A, Naganuma Y, Moriai Y, Nakamura M. Effects of telmisartan on adiponectin levels and body weight in hypertensive patients with glucose intolerance. Metabolism 2008;57:1473–8. [18] Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28: 412–9. [19] Matsuo S, Imai E, Horio M, Yasuda Y, Tomita K, Nitta K, et al. Collaborators developing the Japanese equation for estimated GFR. Revised equations for estimated GFR from serum creatinine in Japan. Am J Kidney Dis 2009;53:982–92. [20] Task Force for Japanese Metabolic Syndrome Criteria. Metabolic syndrome: definition and diagnostic criteria in Japan. J Jpn Soc Intern Med 2005;94:794–809 (in Japanese). [21] Rimm EB, Williams P, Fosher K, Criqui M, Stampfer MJ. Moderate alcohol intake and lower risk of coronary heart disease: meta-analysis of effects on lipids and haemostatic factors. BMJ 1999;319:1523–8. [22] Kiechl S, Willeit J, Poewe W, Egger G, Oberhollenzer F, Muggeo M, et al. Insulin sensitivity and regular alcohol consumption: large, prospective, cross sectional population study (Bruneck study). BMJ 1996;313:1040–4. [23] Imhof A, Woodward M, Doering A, Helbecque N, Loewel H, Amouyel P, et al. Overall alcohol intake, beer, wine, and systemic markers of inflammation in western Europe: results from three MONICA samples (Augsburg, Glasgow, Lille). Eur Heart J 2004;25:2092–100. [24] Sierksma A, van der Gaag MS, Kluft C, Hendriks HF. Moderate alcohol consumption reduces plasma C-reactive protein and fibrinogen levels; a randomized, diet-controlled intervention study. Eur J Clin Nutr 2002;56:1130–6. [25] Beulens JW, de Zoete EC, Kok FJ, Schaafsma G, Hendriks HF. Effect of moderate alcohol consumption on adipokines and insulin sensitivity in lean and overweight men: a diet intervention study. Eur J Clin Nutr 2008;62:1098–105. [26] Beulens JW, van Beers RM, Stolk RP, Schaafsma G, Hendriks HF. The effect of moderate alcohol consumption on fat distribution and adipocytokines. Obesity (Silver Spring) 2006;14:60–6. [27] Pischon T, Girman CJ, Rifai N, Hotamisligil GS, Rimm EB. Association between dietary factors and plasma adiponectin concentrations in men. Am J Clin Nutr 2005;81:780–6. [28] Gigleux I, Gagnon J, St-Pierre A, Cantin B, Dagenais GR, Meyer F, et al. Moderate alcohol consumption is more cardioprotective in men with the metabolic syndrome. J Nutr 2006;136: 3027–32. [29] Suh I, Shaten BJ, Cutler JA, Kuller LH. Alcohol use and mortality from coronary heart disease: the role of highdensity lipoprotein cholesterol. The Multiple Risk Factor Intervention Trial Research Group. Ann Intern Med 1992;116: 881–7.
Available online at www.sciencedirect.com
Metabolism www.metabolismjournal.com
Influence of daily alcohol consumption on serum adiponectin levels in men Shinji Makita a,⁎, Akihiko Abiko a , Mizuyoshi Nagai a , Shinetsu Yonezawa b , Makoto Koshiyama b , Mutsuko Ohta b , Motoyuki Nakamura a a b
Department of Internal Medicine, Division of Cardiology, Iwate Medical University, Morioka 020-8505, Japan Iwate Health Service Association, Morioka, Japan
A R T I C LE I N FO Article history:
AB S T R A C T Background. The risk of cardiovascular diseases is lower among moderate alcohol
Received 20 June 2012
drinkers than among both nondrinkers and heavy drinkers. However, factors that can
Accepted 4 September 2012
account for the U-shaped or J-shaped relationship between daily alcohol consumption and incident cardiovascular diseases remain obscure.
Keywords:
Purpose. The present cross-sectional study investigated the relationship between alcohol consumption and serum adiponectin levels.
Alcohol Adiponectin Metabolic syndrome
Method. Total adiponectin was measured in 527 males participating in health check-up programs (age range 40–86 years, mean 60.5 years). Based on questionnaire responses, alcohol intake was categorized into three groups: none or occasional (A1); <50 g/day and ≥3 days/week (A2); and ≥50 g/day and ≥3 days/week (A3). Results. No significant differences in adiponectin levels were observed among the three alcohol consumption groups of subjects without the metabolic syndrome (MetS). In subjects with the MetS, the adiponectin level was significantly higher in the A2 (moderate drinker) group than in both the A1 and A3 groups. MetS subjects in group A2 had higher HDL-C levels than those in A1, but levels in group A3 were not significantly different from those in group A2. Conclusion. An increased adiponectin level in moderate alcohol drinkers who have MetS may contribute to the U-shaped relationship between alcohol consumption and risk of cardiovascular events, in addition to the involvement of HDL-C. © 2013 Elsevier Inc. All rights reserved.
1.
Introduction
The risk of cardiovascular diseases is known to be lower among light to moderate drinkers than among nondrinkers and heavy drinkers. This U-shaped or J-shaped relationship
has been documented between daily alcohol consumption and ischemic stroke [1–3], coronary heart disease [4–9], and all-cause mortality [10]. A recent meta-analysis also demonstrated that moderate alcohol consumption has beneficial effects on cardiovascular disease incidence and mortality [11].
Abbreviations: HDL-C, high-density lipoprotein cholesterol; MetS, metabolic syndrome; EDTA, ethylene diamine tetra acetic acid; IRI, immunoreactive insulin; CVs, coefficients of variation; ELISA, enzyme-linked immunosorbent assay; HOMA-IR, homeostasis model assessment of insulin resistance; eGFR, estimated glomerular filtration rate; HbA1c, glycosylated hemoglobin; HPLC, high performance liquid chromatography; JDS, Japan Diabetes Society; NGSP, National Glycohemoglobin Standardization Program; sBP, systolic blood pressure; dBP, diastolic blood pressure; LDL-C, low-density lipoprotein cholesterol; CRP, C-reactive protein. ⁎ Corresponding author. Tel.: +81 19 651 5111x2324; fax: +81 19 651 0401. E-mail address: [email protected] (S. Makita). 0026-0495/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.metabol.2012.09.003
412
M ET ABOL I SM CL IN I CA L A N D E XP E RI ME N TAL 6 2 ( 2 0 13 ) 41 1–4 16
There is evidence that light to moderate alcohol consumption may have beneficial effects on type 2 diabetes mellitus [12], an effect that appears to be mediated by decreased insulin resistance [13]. Moreover, light to moderate alcohol intake over the short term increases serum adiponectin [13,14] and high-density lipoprotein cholesterol (HDL-C) levels [15]. It has also been reported that the favorable effects of alcohol intake on glucose metabolism are most prominent in obese mice and in glucose-intolerant human subjects. We recently observed a U-shaped relationship between alcohol consumption and incident myocardial infarction in obese subjects from the general population [16]. However, mediating factors that can fully explain the U-shaped relationship between alcohol consumption and cardiovascular events remain unknown. In particular, the mechanism by which heavy alcohol consumption increases the risk of coronary heart disease has been not identified. The present cross-sectional study of the general population sought to investigate the relationship between daily alcohol consumption and glucose and lipid metabolism, as well as obesity-related indices, such as serum adiponectin concentrations and HDL-C levels. Furthermore, these relationships were analyzed in subjects with metabolic syndrome (MetS), who represent a segment of the population that is at high risk of cardiovascular diseases.
2.
Methods
2.1.
Study subjects
A total of 527 men underwent a comprehensive health checkup at the Iwate Health Service Association and agreed to participate in the present study. Participants with missing data and age <40 years were excluded. The mean age of the included participants was 60.5 years (range 40–86 years). All participants underwent a routine clinical examination, including a medical history, lifestyle assessment, and fasting blood sampling. Overall, 133 (25.2%) participants were on antihypertensive therapy, 58 (11.0%) participants used cholesterol-lowering therapies, and 41 (7.8%) participants were taking antidiabetic agents or were on a dietary regimen. The study protocols were approved by our institutional ethics committee, and all participants provided written informed consent.
2.2.
Blood examination
Following an overnight fast, venous blood samples were taken from the antecubital vein of subjects resting in the sitting position. Samples were collected in vacuum tubes containing EDTA or a serum separator gel. After sampling, the tubes were immediately centrifuged for 10 min at 1500×g. Aliquots of serum were stored at −20 °C, and routine hematology and biochemistry tests were performed within a few days following blood sampling. Some of the serum was also stored at −80 °C for measurement of adiponectin and insulin (immunoreactive insulin, IRI) levels. Total adiponectin concentration was measured using an ELISA kit. The intra- and inter-assay coefficients of variation (CVs) for measurement of adiponectin levels were 2.9% and 3.4%, respectively [17]. The
homeostasis model assessment of insulin resistance (HOMAIR; fasting blood glucose [mg/dl]×blood insulin concentration [μU/ml] / 405) was calculated as an index of insulin resistance [18]. The estimated glomerular filtration rate (eGFR) was calculated according to the equation proposed by the Japan Nephrology Society [19]. Glycosylated hemoglobin (HbA1c) was measured by the HPLC method as a Japan Diabetes Society (JDS) value, and the result is presented as a National Glycohemoglobin Standardization Program (NGSP) equivalent value (transformed by uniformly adding 0.4%).
2.3.
Risk factor assessment
Trained nurses measured the systolic and diastolic blood pressures (sBP and dBP, respectively) of patients who were seated and rested using an automatic digital sphygmomanometer placed on the upper arm. The average of two blood pressure measurements was used for the present analysis. A self-reported questionnaire was administered to document subjects’ medical history and lifestyle, including daily alcohol consumption. Alcohol consumption was evaluated by selecting from the 7 items of alcohol consumption status [the combination consisted of amount/day (<2 gou or ≥2 gou, where 2 gou =2 units of Japanese sake =50 g ethanol) and frequency in a week (<3 days, 3–6 days, or every day)] listed in Table 1. An ex-drinker was defined as a person who had stopped alcohol intake for at least 1 year prior. Regarding the time span of alcohol consumption, since the participants were involved in an annual health check-up program, the choice of the items for alcohol consumption status was derived from the estimated average over the previous approximately 1 year. For the analysis, daily alcohol consumption was categorized into three groups based on grams of ethanol consumed: none or occasional (A1); <50 g/day and ≥3 days/week (A2); and ≥50 g/ day and ≥3 days/week (A3). The smoking index was defined as the product of packs per day and years of smoking (packyears). Waist circumference was measured at the level of the umbilicus during normal exhalation, with the subject standing upright. When significant fat accumulation caused the umbilicus to sag downward, waist circumference measurements were taken at a level midway between the lower margin of the ribs and the anterior iliac spine. Hypertension was defined as at least one of sBP ≥140 mmHg, dBP ≥90 mmHg, or current antihypertensive therapy. Dyslipidemia was diagnosed in subjects with serum lowdensity lipoprotein cholesterol (LDL-C) levels ≥140 mg/dl and/ or high-density lipoprotein cholesterol (HDL-C) levels <40 mg/dl
Table 1 – Categories of daily alcohol consumption. Group A1
A2 A3
Alcohol consumption status
n
%
Ex-drinker Non-drinker Occasional (≤2 days /week) Ethanol <50 g/day, 3–6 days/week Ethanol <50 g/day , everyday Ethanol ≥50 g/day, 3–6 days/week Ethanol ≥50 g/day, everyday
12 84 109 80 150 29 63
39%
44% 17%
413
M ET ABOL I SM CL IN I CA L A N D EX PE RI ME N TA L 6 2 ( 2 0 13 ) 41 1–4 1 6
or in subjects with a history of cholesterol-lowering therapy. Diabetes mellitus was diagnosed according to three possible criteria: fasting blood glucose level ≥126 mg/dl, HbA1c level (NGSP) ≥6.5%, or a history of antidiabetic therapy. MetS was defined according to criteria established in Japan in 2005 [20]. Briefly, MetS was defined as an increased waist circumference (≥85 cm for men, ≥90 cm for women) together with at least two of the following three criteria: (1) triglycerides ≥150 mg/dl and/ or high-density lipoprotein cholesterol <40 mg/dl; (2) sBP ≥130 mmHg and/or dBP ≥85 mmHg; and (3) fasting blood glucose ≥110 mg/dl.
2.4.
Statistical analysis
Comparisons of demographic data according to categories of alcohol consumption were performed using n analysis of variance (ANOVA) with Scheffe's post hoc test. The χ2 test was used for comparison of categorical variables. The values in the tables are expressed as means±standard deviation. Ageadjusted comparisons of serum adiponectin, HDL-C, fasting IRI level, and waist circumference according to level of alcohol consumption were performed using analysis of covariance (ANCOVA) with Scheffe's post hoc test. The values and error bars in the graphs indicate estimated means±standard error. A p-value <0.05 was regarded as significant. All analyses were performed using the SPSS statistical package, version 19.0 (Chicago, IL, USA).
3.
Results
The distribution of subjects according to daily alcohol consumption status is presented in Table 1. Although ex-drinkers are potentially at higher risk of cardiovascular diseases, the 12 ex-drinkers in the present study did not have many cardiovascular risk factors. Therefore, these ex-drinkers were included in group A1. The clinical characteristics of subjects according to level of alcohol consumption are shown in Table 2. Group A2 had a lower fasting insulin level, but a higher HDL-C level than group A1. Subjects in group A3 were younger, more obese, and had higher blood pressures, uric acid levels, and smoking index than those in group A2. Age-adjusted relationships between alcohol consumption and adiponectin, HDL-C, fasting insulin, and waist circumference are shown in Fig. 1. No significant association was found between daily alcohol consumption and serum adiponectin level. The average waist circumference was significantly greater in group A3 than in group A2. No factors showing a V-shaped or inverted V-shaped relationship with level of alcohol consumption were identified in the analysis. The clinical characteristics of the subset of subjects with MetS according to the level of alcohol consumption are shown in Table 3. The HDL-C level was significantly higher in groups A2 and A3 than in group A1. The serum adiponectin level was significantly higher in group A2 than in the other groups. Ageadjusted relationships between alcohol consumption and serum adiponectin level or HDL-C level in subjects with or without MetS are shown in Fig. 2. Among subjects without MetS (non-MetS), no significant differences in adiponectin
Table 2 – Clinical characteristics of the three alcohol consumption groups (all subjects). A1 (n=205) Age, years Body mass index, kg/m2 Waist circumference, cm Systolic BP, mmHg Diastolic BP, mmHg Fasting blood glucose, mg/dl Fasting insulin, μU/ml HOMA-IR Hemoglobin A1c, % Serum Cr, mg/dl eGFR, ml/min/ 1.73 m2 Uric acid, mg/dl Triglyceride, mg/dl LDL-cholesterol, mg/dl HDL-cholesterol, mg/dl Smoking index, pack-year Total adiponectin, μg/ml Hypertension, % Diabetes mellitus, % Dyslipidemia, %
A2 (n=230)
A3 (n=92)
p-value
61.6±10.4 61.1±9.8 24.7±3.2 24.3±2.6
56.3±8.9 ⁎,⁎⁎ 25.2±3.0
<0.001 <0.05
88.6±8.6
88.0±7.1
90.8±8.0 ⁎⁎
<0.05
123±15 72±10 105±24
125±16 74±10 106±18
128±14 ⁎ 77±10 ⁎,⁎⁎ 107±17
<0.05 <0.001 ns
7.5±5.2
6.4±4.0 ⁎
6.7±4.5
<0.05
2.0±1.6 5.6±0.8
1.7±1.2 5.5±0.7
1.8±1.4 5.5±0.6
ns ns
0.86±0.16 0.86±0.36 0.80±0.11 ns 69.3±13.4 71.4±14.1 75.8±13.2 ⁎,⁎⁎ <0.01 5.9±1.2 126±71 127±27
5.9±1.4 128±121 120±30
6.4±1.3 ⁎,⁎⁎ 149±97 115±32 ⁎
<0.01 ns <0.01
50±12
57±14 ⁎
60±14 ⁎
<0.001
17±21
17±18
30±25 ⁎,⁎⁎
<0.001
7.2±4.2
7.6±4.0
6.2±2.8 ⁎⁎
<0.05
37.1% 13.7% 39.0%
39.6% 16.1% 38.7%
59.8% 13.0% 44.6%
<0.01 ns ns
BP; blood pressure, HOMA-IR; homeostasis model assessment of insulin resistance, eGFR; estimated glomerular filtration rate, LDL; low-density lipoprotein, HDL; high-density lipoprotein. ⁎ p<0.05 vs. A1. ⁎⁎ p<0.05 vs. A2.
Adiponectin (µg/ml) 8.0
ns
ns
ns 60
7.0 6.0
50
5.0
40
A1
HDL-C (mg/dl)
70
A2
A3
p<0.01
A1
Fasting insulin (µU/ml)
A2 Waist (cm)
92.0
8.0
p<0.05
p<0.05
90.0
ns
ns
7.0
A3
88.0 6.0
86.0
A1
A2
A3
A1
A2
A3
Fig. 1 – Relationships of alcohol consumption to glucose and lipid indices and waist circumference in age-adjusted analyses.
414
M ET ABOL I SM CL IN I CA L A N D E XP E RI ME N TAL 6 2 ( 2 0 13 ) 41 1–4 16
Table 3 – Clinical characteristics of the three alcohol consumption groups (subjects with metabolic syndrome). A1 (n=53)
A2 (n=49)
A3 (n=37)
p-value <0.05
26.0±2.1
57.1± 9.4 ⁎⁎ 26.8±2.2
95.4±7.3
93.3±5.4
95.2±5.8
ns
130±13 76±10 115±19
132±14 79±11 122±23
131±11 79±10 115±18
ns ns ns
10.7±7.5
8.7±5.0
9.0±5.6
ns
3.1±2.4 5.8±0.7 0.86±0.14
2.6±1.6 5.9±0.8 0.85±0.15
2.6±1.7 5.8±0.6 0.79± 0.11 ⁎
ns ns <0.05
69.3±12.2
70.0±13.6
77.6±14.4
<0.01
6.0±1.2 160±77
5.8±1.4 189±224
6.5±1.4 167±109
ns ns
124±27
114±25
123±28
ns
45±11
54±17 ⁎
55±12 ⁎
<0.001
13±18
17±19
30±27 ⁎,⁎⁎
<0.01
5.7±3.4
7.3±3.1 ⁎
5.7±2.5 ⁎⁎
<0.05
66.0% 30.2% 64.2%
75.5% 44.9% 61.2%
83.8% 27.0% 54.1%
ns ns ns
Age, years
61.9±9.8
62.6±10.1
Body mass index, kg/m2 Waist circumference, cm Systolic BP, mmHg Diastolic BP, mmHg Fasting blood glucose, mg/dl Fasting insulin, μU/ml HOMA-IR Hemoglobin A1c, % Serum Cr, mg/dl
27.1±3.0
eGFR, ml/min/1.73 m2 Uric acid, mg/dl Triglyceride, mg/dl LDL-cholesterol, mg/dl HDL-cholesterol, mg/dl Smoking index, pack-year Total adiponectin, μg/ml Hypertension, % Diabetes mellitus, % Dyslipidemia, %
ns
BP; blood pressure, HOMA-IR; homeostasis model assessment of insulin resistance, eGFR; estimated glomerular filtration rate, LDL; low-density lipoprotein, HDL; high-density lipoprotein. ⁎ p<0.05 vs. A1. ⁎⁎ p<0.05 vs. A2.
non-MetS
ns
8.0
MetS
ns
Mean level in non-MetS
Adiponectin (µg/mL) 7.0 Age-adjusted
p<0.05
p<0.05
6.0 5.0
A1
A2
70 HDL-C (mg/dL)
60
Age-adjusted
50
A3
A1
A2
A3
p<0.01 p<0.01
ns
p<0.01
Mean level in non-MetS
40 A1
A2
A3
A1
A2
A3
Fig. 2 – Influence of alcohol consumption on adiponectin and high-density lipoprotein cholesterol (HDL-C) in subjects with and without metabolic syndrome (MetS).
level were observed among the three alcohol consumption groups, despite a linear increase in HDL-C. In MetS subjects, while the adiponectin level was lower than in non-MetS subjects, only group A2 had an adiponectin level comparable to the mean level in non-Mets subjects. These significant differences remained with a multivariate adjusted model including age, eGFR, waist circumference, and smoking index, all of which potentially affect the adiponectin level (p for trend=0.010, p=0.009 for group A1 vs. A2, p=0.010 for group A2 vs. A3, p=0.73 for group A1 vs. A3). On the other hand, MetS subjects in group A2 had higher levels of HDL-C than those in A1, but HDL-C levels in group A3 were not significantly different from those in group A2. The p-values for the interaction between alcohol consumption and MetS were 0.12 for the adiponectin level and 0.50 for the HDL-C level. The post hoc power analysis of ANCOVA for adiponectin levels in the age-adjusted model revealed a statistical power (1−β) of 0.41 for all subjects, 0.13 for the non-MetS subgroup, and 0.68 for the MetS subgroup. The analysis for HDL-C levels showed that 1−β was 1.00, 1.00, and 0.97 for all subjects, the non-MetS subgroup, and the MetS subgroup, respectively.
4.
Discussion
Previous reports suggested that the reduction in ischemic heart disease risk attributed to moderate alcohol consumption may be explained by the beneficial effect of alcohol on the thrombolytic profile [21], insulin sensitivity [22], inflammation, as reflected by the C-reactive protein (CRP) level [23], and fibrinogen concentrations [24]. However, the increased risk of coronary heart disease among heavy drinkers could not be explained by all of these factors. In an effort to explain the U-shaped or J-shaped relationship between cardiovascular disease risk and amount of daily alcohol consumption, the present study examined the associations between various clinical indices, including serum adiponectin and HDL-C levels, and the level of daily alcohol consumption. In the analysis of the overall sample, no factors were identified to explain this relationship. However, among subjects with MetS, who are at high risk for cardiovascular disease, moderate drinkers had a higher serum adiponectin level than non- or occasional drinkers, as well as heavy drinkers. These results suggest that an increased adiponectin level in moderate alcohol drinkers may contribute to the U-shaped relationship between alcohol consumption and risk of cardiovascular events, at least among those at greatest risk for cardiovascular disease. A limited number of controlled trials have investigated the effect of alcohol consumption on the serum adiponectin level [13,14,25,26]. Sierksma et al. reported that plasma adiponectin concentration increased by 11% following 17 days of drinking whisky versus water. These authors concluded that moderate alcohol consumption improved insulin sensitivity in relatively insulin-resistant middle-aged men, and they suggested that the effect may be mediated by an increased adiponectin level [13]. Bullen et al. showed that 4 weeks of moderate alcohol consumption increased plasma adiponectin concentration in men with abdominal obesity, whereas fat distribution and ISI were unaffected [26]. Although these results are in general
M ET ABOL I SM CL IN I CA L A N D EX PE RI ME N TA L 6 2 ( 2 0 13 ) 41 1–4 1 6
agreement with the present findings, differences in the amount of alcohol intake were not examined in their study. Specifically, it was observed that the serum adiponectin level was significantly lower in heavy drinkers than in light to moderate drinkers. This observation may support the theory that serum adiponectin mediates the clinically favorable effects of alcohol consumption. Pischon et al. showed a significant nonlinear association between plasma adiponectin concentrations and alcohol intake in a similar study of men [27]. While the subjects with moderate alcohol intake (5.0–29.9 g/day) had a higher adiponectin level than non-drinkers, the increase in the adiponectin level was lower in heavy drinkers (≥30 g/day, p=0.10 vs. nondrinkers). The present data on the behavior of the adiponectin level are fundamentally consistent with this report. However, significant differences were not confirmed between moderate drinkers and heavy drinkers, and only the overall analysis was shown in their study. Gigleux et al. reported that light to moderate alcohol consumption (>15 g/day) had significant cardioprotective effects that were most apparent in men with MetS [28]. In the present study, an inverted-V-shaped relationship was observed between the serum adiponectin level and the amount of alcohol consumption in MetS subjects. Although the present observations may possibly explain the results reported by Gigleux et al., this latter study did not include information on heavy drinkers. Furthermore, anti-inflammatory effects may also contribute to the U-shaped relationship between alcohol consumption and ischemic heart disease. For instance, Imhof et al. showed that light to moderate daily alcohol intake (up to 20 or 40 g/day) was associated with lower CRP and white blood counts compared to not drinking, as well as heavy drinking, even after adjustment for potential confounders [23]. Although the CRP level was not measured in the present study, a strong correlation has been established between high sensitivity CRP and serum adiponectin levels. An acute increase in HDL-C levels in response to alcohol intake has been shown in a controlled study [13] and a metaanalysis [15]. Additionally, a linear increase in the HDL-C level was documented in association with alcohol consumption ranging from zero to >21 drinks/week (averaged >3 drinks/ day) [29]. Although the linear increase in HDL-C was blunted in heavy drinkers in the present study, their HDL-C level was similar to that of moderate drinkers. Thus, we postulated that the HDL-C level alone did not mediate the increased risk of cardiovascular diseases in heavy drinkers. Alternatively, it is possible that the increased risk of cardiovascular diseases in heavy drinkers may be produced by a combination effect with increased blood pressure. The strength of the present study may be that both the amount and frequency of alcohol consumption were reflected in the categorization. Furthermore, the sample size for evaluation of the adiponectin level was relatively large and comparable to the study of Pischon et al., which had the largest sample size in the previous literature. Nevertheless, the greatest weakness of the present study was its lack of statistical power. Since 1−β for the analysis of the adiponectin level was quite low, especially for the overall subjects and the non-MetS subgroup, findings of no association between alcohol consumption and adiponectin levels in these subjects
415
cannot be statistically guaranteed because of the high probability of type II error. Other limitations of this cross-sectional study warrant consideration. First, medications for hypertension, dyslipidemia, and diabetes that may affect the serum adiponectin level were not taken into consideration. Second, the relationships between alcohol consumption and cardiovascular risk can be altered by the chosen cutoff level for ‘light’ or ‘moderate’ alcohol consumption. It is unclear whether the cutoff used (<50 g/day and >3 days/week) was appropriate for delineating the mild to moderate alcohol consumption category. Although the actual average alcohol consumption cannot be calculated in the present study population, the average alcohol consumption of group A2 of the present study can be estimated as approximately 20–25 g/day based on a similar population involved in our large cohort study [16]. Nevertheless, a metaanalysis revealed that ‘moderate’ consumption of alcohol (up to two drinks or 30 g alcohol a day for men), a level of alcohol consumption that is similar to the cutoff level in the current study, has beneficial effects on a variety of biomarkers linked to the risk of coronary heart disease [15]. Third, drinking patterns were not taken into consideration. An inaccurate estimate of daily alcohol consumption cannot be excluded due to self-reporting of consumption in the present study. In addition, the present study did not investigate the kinds of beverages being consumed. However, beer, wine, and spirits are generally known to have similar effects on HDL-C [21]. Fourth, since female subjects were excluded from the analysis due to their relatively low alcohol consumption level, sex differences remain unclear. Furthermore, many previous investigations showed the U-shaped association of alcohol consumption with risk of cardiovascular disease in many populations, not merely those with MetS. A larger investigation may be needed to fill these gaps. Finally, causality cannot be assumed in the relationship with adiponectin because of the cross-sectional design of the present study. In addition, since both alcohol intake and an increased adiponectin level can potentially improve the features of MetS via increasing the HDL-C level and improving insulin sensitivity, respectively, stratification by MetS can introduce a bias. In conclusion, an increase in the serum adiponectin level in moderate drinkers and a decrease in the serum adiponectin level in heavy drinkers were observed among persons with MetS, a group that is at high risk for cardiovascular disease. Differences in serum adiponectin levels may be associated with the U-shaped relationship between daily alcohol consumption and cardiovascular disease risk, in addition to the involvement of HDL-C. From the translational perspective, evaluation of adiponectin levels may assist in setting the appropriate amount of alcohol consumption to prevent cardiovascular disease in alcohol drinkers.
Author contributions Shinji Makita, MD (statistical analyses, writing the paper); Akihiko Abiko, MD (collecting data); Mizuyoshi Nagai, MD (collecting data); Shinetsu Yonezawa (collecting data); Makoto Koshiyama (collecting data); Mutsuko Ohta (collecting data); Motoyuki Nakamura, MD (critical review of the paper).
416
M ET ABOL I SM CL IN I CA L A N D E XP E RI ME N TAL 6 2 ( 2 0 13 ) 41 1–4 16
Funding The authors did not receive any funding.
Conflict of interest The authors report no conflicts of interest.
REFERENCES
[1] Iso H, Baba S, Mannami T, Sasaki S, Okada K, Konishi M, et al. Alcohol consumption and risk of stroke among middle-age men: the JPHC Study Cohort I. Stroke 2004;35:1124–9. [2] Reynolds K, Lewis B, Nolen JD, Kinney GL, Sathya B, He J. Alcohol consumption and risk of stroke: a meta-analysis. JAMA 2003;289:579–88. [3] Berger K, Ajani UA, Kase CS, Gaziano JM, Buring JE, Glynn RJ, et al. Light-to-moderate alcohol consumption and risk of stroke among U.S. male physicians. N Engl J Med 1999;341:1557–64. [4] Yano K, Rhoads GG, Kagan A. Coffee, alcohol and risk of coronary heart disease among Japanese men living in Hawaii. N Engl J Med 1977;297:405–9. [5] Langer RD, Criqui MH, Reed DM. Lipoproteins and blood pressure as biological pathways for effect of moderate alcohol consumption on coronary heart disease. Circulation 1992;85: 910–5. [6] Shaper AG, Wannamethee G, Walker M. Alcohol and coronary heart disease: a perspective from the British Regional Heart Study. Int J Epidemiol 1994;23:482–94. [7] Gordon T, Kannel WB. Drinking habits and cardiovascular disease: the Framingham Study. Am Heart J 1985;110:331–4. [8] Miller GJ, Beckles GL, Maude GH, Carson DC. Alcohol consumption: protection against coronary heart disease and risks to health. Int J Epidemiol 1990;19:923–30. [9] Kitamura A, Iso H, Sankai T, Naito Y, Sato S, Kiyama M, et al. Alcohol intake and premature coronary heart disease in urban Japanese men. Am J Epidemiol 1998;147:59–65. [10] Emberson JR, Shaper AG, Wannamethee SG, Morris RW, Whincup PH. Alcohol intake in middle age and risk of cardiovascular disease and mortality: accounting for intake variation over time. Am J Epidemiol 2005;161:856–63. [11] Ronksley PE, Brien SE, Turner BJ, Mukamal KJ, Ghali WA. Association of alcohol consumption with selected cardiovascular disease outcomes: a systematic review and metaanalysis. BMJ 2011;342:d671. [12] Howard AA, Arnsten JH, Gourevitch MN. Effect of alcohol consumption on diabetes mellitus: a systematic review. Ann Intern Med 2004;140:211–9. [13] Sierksma A, Patel H, Ouchi N, Kihara S, Funahashi T, Heine RJ, et al. Effect of moderate alcohol consumption on adiponectin, tumor necrosis factor-alpha, and insulin sensitivity. Diabetes Care 2004;27:184–9. [14] Imhof A, Plamper I, Maier S, Trischler G, Koenig W. Effect of drinking on adiponectin in healthy men and women: a randomized intervention study of water, ethanol, red wine, and beer with or without alcohol. Diabetes Care 2009;32: 1101–3.
[15] Brien SE, Ronksley PE, Turner BJ, Mukamal KJ, Ghali WA. Effect of alcohol consumption on biological markers associated with risk of coronary heart disease: systematic review and meta-analysis of interventional studies. BMJ 2011;342: d636. [16] Makita S, Onoda T, Ohsawa M, Tanaka F, Segawa T, Takahash T, et al. Influence of mild-to-moderate alcohol consumption on cardiovascular diseases in men from the general population. Atherosclerosis 2012;224:222–7. [17] Makita S, Abiko A, Naganuma Y, Moriai Y, Nakamura M. Effects of telmisartan on adiponectin levels and body weight in hypertensive patients with glucose intolerance. Metabolism 2008;57:1473–8. [18] Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28: 412–9. [19] Matsuo S, Imai E, Horio M, Yasuda Y, Tomita K, Nitta K, et al. Collaborators developing the Japanese equation for estimated GFR. Revised equations for estimated GFR from serum creatinine in Japan. Am J Kidney Dis 2009;53:982–92. [20] Task Force for Japanese Metabolic Syndrome Criteria. Metabolic syndrome: definition and diagnostic criteria in Japan. J Jpn Soc Intern Med 2005;94:794–809 (in Japanese). [21] Rimm EB, Williams P, Fosher K, Criqui M, Stampfer MJ. Moderate alcohol intake and lower risk of coronary heart disease: meta-analysis of effects on lipids and haemostatic factors. BMJ 1999;319:1523–8. [22] Kiechl S, Willeit J, Poewe W, Egger G, Oberhollenzer F, Muggeo M, et al. Insulin sensitivity and regular alcohol consumption: large, prospective, cross sectional population study (Bruneck study). BMJ 1996;313:1040–4. [23] Imhof A, Woodward M, Doering A, Helbecque N, Loewel H, Amouyel P, et al. Overall alcohol intake, beer, wine, and systemic markers of inflammation in western Europe: results from three MONICA samples (Augsburg, Glasgow, Lille). Eur Heart J 2004;25:2092–100. [24] Sierksma A, van der Gaag MS, Kluft C, Hendriks HF. Moderate alcohol consumption reduces plasma C-reactive protein and fibrinogen levels; a randomized, diet-controlled intervention study. Eur J Clin Nutr 2002;56:1130–6. [25] Beulens JW, de Zoete EC, Kok FJ, Schaafsma G, Hendriks HF. Effect of moderate alcohol consumption on adipokines and insulin sensitivity in lean and overweight men: a diet intervention study. Eur J Clin Nutr 2008;62:1098–105. [26] Beulens JW, van Beers RM, Stolk RP, Schaafsma G, Hendriks HF. The effect of moderate alcohol consumption on fat distribution and adipocytokines. Obesity (Silver Spring) 2006;14:60–6. [27] Pischon T, Girman CJ, Rifai N, Hotamisligil GS, Rimm EB. Association between dietary factors and plasma adiponectin concentrations in men. Am J Clin Nutr 2005;81:780–6. [28] Gigleux I, Gagnon J, St-Pierre A, Cantin B, Dagenais GR, Meyer F, et al. Moderate alcohol consumption is more cardioprotective in men with the metabolic syndrome. J Nutr 2006;136: 3027–32. [29] Suh I, Shaten BJ, Cutler JA, Kuller LH. Alcohol use and mortality from coronary heart disease: the role of highdensity lipoprotein cholesterol. The Multiple Risk Factor Intervention Trial Research Group. Ann Intern Med 1992;116: 881–7.