Relationship between Alcohol Intake and Risk Factors for Metabolic Syndrome in Men

Transcription

1 ORIGINAL ARTICLE Relationship between Alcohol Intake and Risk Factors for Metabolic Syndrome in Men Miharu Hirakawa, Yasuji Arase, Kazuhisa Amakawa, Yuki Ohmoto-Sekine, Makiko Ishihara, Masato Shiba, Kyoko Ogawa, Chikao Okuda, Toyohisa Jinno, Hisahito Kato, Hiroyuki Tsuji, Mitsuyo Hashimoto, Takashi Yamamoto, Satao Arimoto and Shigeko Hara Abstract Objective The precise relationship between alcohol intake and metabolic syndrome (MetS) is still unclear, and the results from previous studies have been inconclusive. Thus, we examined the effect of alcohol intake on the risk of MetS in men in order to gain more information on a potential relationship. Methods This study included 22,349 men who were divided into four groups according to their average alcohol intake [non-, light (less than 20 g ethanol/day), heavy (equal or more than 20 g and less than 60 g ethanol/day) and very heavy (equal and greater than 60 g ethanol/day) drinkers]. We measured each subject s body mass index (BMI), waist circumference and blood pressure (BP) and conducted a blood test to obtain a complete blood count and biochemical panel. These results were used to obtain the MetS prevalence. Additionally, fatty liver was diagnosed using abdominal ultrasonography. Results Light drinkers had smaller waist circumferences. Heavy and very heavy drinkers had larger waist circumferences, a higher BMI, a higher BP, higher fasting plasma glucose levels, higher triglycerides (TG) levels and higher high-density lipoprotein (HDL) cholesterol levels while they had lower low-density lipoprotein cholesterol levels than nondrinkers. The prevalence of high BP, hyperglycemia and high TG was significantly higher in heavy and very heavy drinkers than in nondrinkers. The prevalence of low HDL cholesterol levels decreased with an increase in alcohol consumption. The prevalence of MetS was significantly lower in light drinkers and higher in very heavy drinkers compared with nondrinkers. Conclusion Alcohol intake significantly influences the risk of MetS in men. A significant association was seen between an alcohol intake of 60 g/day or higher and the prevalence of MetS. Key words: alcohol intake, blood pressure, plasma glucose, triglyceride, HDL cholesterol, metabolic syndrome (Intern Med 54: , 2015) () Introduction The effects of alcohol consumption have been reported in many studies. The adverse effects of alcohol and the merits of moderate alcohol consumption are well established. Furthermore, liver damage caused by the intake of a large quantity of alcohol is well known (1, 2). The National Cancer Center in Japan previously reported that the mortality of people who drink less than 20 g of alcohol daily is lower than people who never consume alcohol (3). More than 100 prospective studies have shown an inverse association between moderate drinking and the risks of heart attack, ischemic (clot-caused) stroke, peripheral vascular disease, sudden cardiac death, and death from all cardiovascular causes (4). The effect is fairly consistent, corresponding to a 25-40% reduction in risk. Many doctors believe that some factors related with metabolic syndrome (MetS), such as obesity, hypertension, dyslipidemia and hyperglycemia, may be caused by excessive alcohol consumption. However, there Health Management Center, Toranomon Hospital, Japan Received for publication February 20, 2014; Accepted for publication December 14, 2014 Correspondence to Dr. Miharu Hirakawa, ZXC00701@nifty.ne.jp 2139

2 are inconsistencies in the results of previous studies on the relationship between alcohol drinking and MetS. The effect of light alcohol consumption has been especially unclear. The relationship between alcohol intake and MetS has been reported to be inversely linear (5), J- or U-shaped (6), positively linear (7) or not significant (8). Furthermore, alcohol consumption has been reported to have a favorable effect on the plasma high-density lipoprotein (HDL) cholesterol levels, a detrimental effect on plasma triglycerides (TG) concentration, and may contribute to an elevation of blood pressure (BP) (8-12). The aim of this study is to examine the effect of the amount of alcohol intake on the factors associated with MetS (13-16) in men. MetS is not clearly defined; however, two major sets of international criteria have been established by the International Diabetes Federation (IDF) (13) and the National Cholesterol Education Program Adult Treatment Panel III (NCEP ATP III) (14). The main difference between the two definitions is that the IDF has an obligatory threshold value for waist circumference (15). The definition put forth by the Japanese Committee for the Diagnostic Criteria of MetS also includes a threshold value for waist circumference (16). For this study, we utilized the IDF criteria (13), NCEP ATP III criteria (14) and Japanese criteria (16). Subjects Materials and Methods In this cross-sectional study, 22,349 men who visited the Health Management Center of Toranomon Hospital for the Ningen Dock (annual comprehensive) health check-up between January 1997 and December 2011 were included. For men who visited our center for the Ningen Dock health check-up more than once during this period, only data from the first visit was included. The mean age of the patients was 48.6±10.2 (range: 18-95). The average alcohol consumption of each subject per day was reported based on answers to standardized questionnaires on which the frequency and amount of habitual alcohol drinking were asked (sample questions include: How frequently do you drink alcohol per week/month? and What amount of alcohol do you drink in a day when you drink? ). The daily alcohol intake was expressed in go (equivalent to 180 ml) for the Japanese unit of the traditional amount of sake, beer was expressed in the number of middle bottles (500 ml volume/bottle), and wine and whisky was expressed in the number of glasses. Twenty grams of ethanol corresponds to approximately one go (180 ml) of sake, 500 ml of beer, 180 ml of wine, 60 ml of whisky, and 110 ml of shochu. The subjects were categorized into four groups based on their daily ethanol consumption (nondrinkers; light drinkers: <20 g ethanol/day; heavy drinkers: 20gand<60gethanol/day;veryheavy drinkers: 60 g ethanol/day). These categories were selected based on the general assumption that alcohol intake should be reduced to g ethanol/day to prevent hypertension. Measurements We measured each subject s height and weight and calculated Body Mass Index (BMI) as weight in kilograms divided by the square height in meters. The waist circumference was measured at the navel level according to the definition in the Japanese Committee for the Diagnostic Criteria of Metabolic Syndrome (18). The systolic and diastolic BP was also measured. A blood test was conducted to obtain the complete blood count, in particular the white blood cell (WBC) count, the hemoglobin (Hb) level, and the platelet (PLT) count, and biochemical factors including aspartate aminotransferase (AST), alanine aminotransferase (ALT), γ- glutamyl transpeptidase (γgtp), fasting plasma glucose (FPG), hemoglobin A1c (HbA1c), TG, total cholesterol (TC), HDL cholesterol, low-density lipoprotein (LDL) cholesterol and ferritin. Additionally, the ratio of fatty liver was determined based on the results from abdominal ultrasonography (US). Criteria of metabolic syndrome The factors considered to be related to MetS in this study were waist circumference, BP, HDL cholesterol, TG and FPG based on criteria from the IDF, NCEP ATP III and the Japanese Committee for the Diagnostic Criteria of Metabolic Syndrome (13, 14, 16). According to the IDF definition, for a person to be defined as having MetS they must have: central obesity (defined as waist circumference with ethnicity-specific values or defined as BMI) plus any two of the following four factors: 1) elevated TG, TG 150 mg/dl; 2) reduced HDL cholesterol, HDL<40 mg/dl in men; 3) elevated BP, systolic BP 130 or diastolic BP 85 mmhg; and 4) elevated FPG, FPG 100 mg/dl. If BMI is >30 kg/m 2, central obesity can be assumed and the waist circumference does not need to be measured. The ethnicity-specific value for waist circumference was 90 cm or higher in men and 80 cm or higher in women for Japanese people. According to the NCEP-ATP III definition, MetS in men is defined as three or more of the the following risk factors: 1) waist circumference >102 cm, 2) elevated TG, TG 150 mg/dl; 3) reduced HDL cholesterol, HDL<40 mg/dl in men; 4) elevated BP, systolic BP 130 or diastolic BP 85 mmhg; and 5) elevated FPG, FPG 100 mg/dl. According to the Japanese criteria, MetS in men is defined as a waist circumference 85 cm and the presence of two or more of the following risk factors: 1) elevated BP, systolic BP 130 or diastolic BP 85 mmhg; 2) dyslipidemia, reduced HDL cholesterol, HDL cholesterol <40 mg/dl and/or raised TG, TG 150 mg/dl; and 3) elevated FPG, FPG 110 mg/dl. Statistical analysis A comparison between the BMI, waist circumference, systolic and diastolic BP, WBC, Hb, PLT, AST, ALT, γgtp, FPG, HbA1c, TG, TC, HDL cholesterol, LDL cholesterol, ferritin, the prevalence of fatty liver, the ratio of men with 2140

3 Table 1. Comparison of Mean Levels of Each Variable among Four Groups. Alcohol group p Nondrinkers Light drinkers (<20 g/day) Heavy drinkers ( 20, <60 g/day) Very heavy drinkers ( 60 g/day) N 2,848 9,788 7,470 2,243 Age 50.3± ±10.6** 48.8±9.5** 48.0±8.3** <0.001 BMI, kg/m ± ± ±2.8** 23.8±2.7** <0.001 Waist circumference, cm 84.4± ±8.7* 85.0±8.1* 86.2±8.8* <0.001 Systolic BP, mmhg 123±14 123±13 126±14** 127±14** <0.001 Diastolic BP, mmhg 78±9 78±9 81±9** 83±10** <0.001 WBC, 10 3 / L 5.8± ±1.5** 5.6±1.5** 5.7±1.5* <0.001 Hb, g/dl 15.0± ± ± ±1.0** <0.001 PLT, 10 4 / L 22.5± ± ± ± AST, IU/L 25±10 25±11 26±11** 29±18** <0.001 ALT, IU/L 28±20 28±21 28±19 31±27* <0.001 GTP, IU/L 44±46 49±56** 68±72** 94±104** <0.001 FPG, mg/dl 100±20 100±18 102±19** 103±18** <0.001 HbA1c(NGSP), % 5.6± ±0.6** 5.5±0.7** 5.4±0.7** <0.001 TG, mg/dl 126±95 124±85 135±98** 147±107** <0.001 TC, mg/dl 201±33 202±32 204±32* 202±34 <0.001 HDL cholesterol, mg/dl 50±12 51±13** 54±14** 55±15** <0.001 LDL cholesterol, mg/dl 126±33 124±30 118±30** 111±33** <0.001 Ferritin, ng/ml 182±83 200±121** 240±160** 299±208** <0.001 Fatty liver, % * With/without smoking history Smoking history, % 805/2, %** 3,216/6,572** 32.9%** 3,431/4,039** 45.9%** 1,296/947** 57.8%** <0.001 Means with standard deviations of each variable are shown. Light drinkers: <20 g ethanol/day; heavy drinkers: 0 and <60 g ethanol/day; very heavy drinkers: 0 g ethanol/day. * p<0.05; ** p<0.001 compared with nondrinkers. smoking history and the prevalence of risk factors for MetS was performed among the 4 groups (nondrinker and 3 drinker groups) utilizing the Kruskal-Wallis test. When a significant difference was found, the results were compared utilizing the Mann-Whitney U test between two groups (i.e., nondrinkers vs. drinkers; light drinkers vs. heavy drinkers; and heavy drinkers vs. very heavy drinkers). Additionally, a comparison of the mean values was performed between men with fatty liver (fatty liver group) and others (non-fatty liver group) in each of the 4 alcohol groups utilizing the Mann- Whitney U test. Statistical analyses were performed using a computer software program (SPSS version 16.0 J for Windows, IBM, Chicago, USA). Results Comparison of the mean values of each variable among the alcohol intake groups The four groups consisted of 2,848 men (12.7%), 9,788 men (43.8%), 7,470 men (33.4%) and 2,243 men (10.0%), respectively. The mean ages were 50.3±11.5, 48.1±10.6, 48.8±9.5 and 48.0±8.3 in each group, respectively. Table 1 shows the comparison of the mean levels of each variable among the four groups. The BMI, waist circumference, systolic and diastolic BP, WBC, Hb, AST, ALT, γgtp, FPG, HbA1c, TG, TC, HDL cholesterol, LDL cholesterol, ferritin, the ratio of fatty liver and the ratio of men with a history of smoking were significantly different among the four groups according to the Kruskal-Wallis test. Therefore, these 18 factors were analyzed using the Mann-Whitney U test. The average BMI was significantly higher in the heavy or very heavy drinkers compared to nondrinkers (p<0.001, p< 0.001, respectively). There was no significant difference in BMI between light drinkers and nondrinkers. The waist circumference was significantly smaller in light drinkers than in nondrinkers (p=0.005) and was significantly larger in heavy and very heavy drinkers than in nondrinkers (p= 0.044, p=0.012, respectively). The smaller waist circumference seen in the light drinkers group remained regardless of age stratification ( 45 years old and 55 years old); thus, no significant difference was seen between the ages of each alcohol group. The systolic and diastolic BP was elevated in heavy and very heavy drinkers compared with nondrinkers (p<0.001, p<0.001, p<0.001, p<0.001, respectively). The BP was also significantly elevated in the heavy drinkers compared with light drinkers (p<0.001, p<0.001, respectively). These results suggest that an alcohol intake 20 g/day may increase the systolic and diastolic BP. The WBC numbers were significantly higher in nondrinkers compared with the three drinker groups (p<0.001, p<0.001, p=0.012, respectively). The Hb levels were significantly higher in very heavy drinkers compared with nondrinkers (p<0.001). No significant difference was seen in the AST and ALT levels between the nondrinker group and light drinker group. However, AST levels were significantly higher in heavy and very heavy drinkers than in nondrinkers (p<0.001, p<0.001, respectively). The AST level appeared to increase with an increase in alcohol intake. The ALT levels were also significantly higher in very heavy drinkers than in nondrinkers (p<0.001). The γgtp levels appeared to increase with an increase in al- 2141

4 cohol intake; the γgtp levels were significantly higher in light, heavy and very heavy drinkers compared with nondrinkers (p<0.001, p<0.001, p<0.001, respectively). The FPG levels were significantly higher in heavy drinkers and in very heavy drinkers than in nondrinkers (p<0.001, p<0.001, respectively). There was also a significant difference in levels between heavy and very heavy drinkers (p< 0.001). The HbA1c levels were significantly lower in light, heavy and very heavy drinkers compared with nondrinkers (p<0.001, p<0.001, p<0.001, respectively). The TG levels were significantly higher in heavy and very heavy drinkers than in nondrinkers (p<0.001, p<0.001, respectively). Additionally, the TG levels in heavy drinkers were higher compared with light drinkers (p<0.001) and the TG levels in very heavy drinkers was higher than in heavy drinkers (p< 0.001). These results suggest that TG levels remain stable when the alcohol intake <20 g/day and increases with an increase in the alcohol intake when the intake 20 g/day. The TC levels were significantly higher in heavy drinkers compared with nondrinkers (p=0.001). However, there was no significant difference in the TC levels in very heavy drinkers compared with nondrinkers (p=0.634). The HDL cholesterol was significantly higher in the three drinker groups than in nondrinkers (p<0.001, p<0.001, p<0.001, respectively). The HDL cholesterol increased with an increase in alcohol intake. The LDL cholesterol was significantly lower in heavy and very heavy drinkers than in nondrinkers (p<0.001, p< 0.001, respectively). There was no significant difference in LDL cholesterol between nondrinkers and light drinkers (p= 0.158). The ferritin levels were significantly higher in the three drinker groups than in nondrinkers (p<0.001, p<0.001, p<0.001, respectively). Thus, ferritin appeared to increase with an increase in alcohol intake. The results of abdominal US indicated that 37.9% of nondrinkers have fatty liver. The ratios were 34.5 and 36.9% in heavy and very heavy drinkers, respectively, and there was a significant difference between heavy and very heavy drinkers (p=0.036).the ratios of men with a history of smoking were significantly higher in the three drinker groups than in nondrinkers (p<0.001, p<0.001, p<0.001, respectively), and the ratio increased as the alcohol intake increased. Comparison of mean values between in men with fatty liver and others Table 2 shows the comparison of the mean levels of some variables between men diagnosed with fatty liver using US (fatty liver group) and men without fatty liver (non-fatty liver group) in the four alcohol groups. The waist circumference, systolic BP, AST, ALT, γgtp, FPG, HbA1c, TG, LDL cholesterol and ferritin levels were significantly higher and the HDL cholesterol was significantly lower in the fatty liver group than in the non-fatty liver group in all alcohol groups (waist circumference: p<0.001, p<0.001, p<0.001, p< 0.001, systolic BP: p<0.001, p<0.001, p<0.001, p=0.002, AST: p<0.001, p<0.001, p<0.001, p<0.001, ALT: p<0.001, p <0.001, p<0.001, p<0.001, γgtp: p<0.001, p<0.001, p< 0.001, p<0.001, FPG: p<0.001, p<0.001, p<0.001, p<0.001, HbA1c: p<0.001, p<0.001, p<0.001, p<0.001, TG: p<0.001, p<0.001, p<0.001, p<0.001, LDL cholesterol: p<0.001, p< 0.001, p<0.001, p=0.034, ferritin: p<0.001, p<0.001, p< 0.001, p<0.001, HDL cholesterol: p<0.001, p<0.001, p< 0.001, p<0.001, respectively). The waist circumference was significantly higher in heavy and very heavy drinkers compared with nondrinkers only in the non-fatty liver group (p< 0.001, p=0.012, respectively). The systolic and diastolic BP, FPG and TG were significantly higher in heavy and very heavy drinkers compared with nondrinkers in both the nonfatty (systolic BP: p<0.001, p=0.005, diastolic BP: p<0.001, p<0.001, FPG: p<0.001, p<0.001, TG: p<0.001, p<0.001, respectively) and in the fatty liver groups (systolic BP: p< 0.001, p=0.004, diastolic BP: p<0.001, p=0.001, FPG: p< 0.001, p<0.001, TG: p=0.01, p<0.001, respectively). The HDL cholesterol and ferritin levels were significantly higher in the three drinker groups than in nondrinkers both in the non-fatty (HDL cholesterol: p<0.001, p<0.001, p<0.001, ferritin: p<0.001, p<0.001, p<0.001, respectively) and in the fatty liver groups (HDL cholesterol: p<0.001, p<0.001, p< 0.001, ferritin: p<0.001, p<0.001, p<0.001, respectively). Comparison of the prevalence of each risk factor for metabolic syndrome among the alcohol intake groups Table 3 shows the prevalence of each risk factor for MetS among the 4 alcohol groups. The prevalence of a large waist circumference was significantly lower in light drinkers than in nondrinkers [Japanese criteria: p=0.003; NCEP-ATP III criteria: p=0.016]. The prevalence of a large waist circumference tended to be greater in heavy and very heavy drinkers than in nondrinkers, but there was no significant difference [Japanese criteria: p=0.585 (heavy drinkers), p=0.172 (very heavy drinkers), respectively; NCEP-ATP III criteria: p=0.050 (heavy drinkers), p=0.596 (very heavy drinkers), respectively]. The prevalence of central obesity was significantly lower in light drinkers than in nondrinkers [IDF criteria: p=0.016] and was significantly higher in very heavy drinkers than in nondrinkers [IDF criteria: p=0.029]. The prevalence of high BP was significantly higher in heavy (p<0.001) and very heavy (p<0.001) drinkers than in nondrinkers. The prevalence of high FPG levels was significantly higher in heavy and very heavy drinkers than in nondrinkers [Japanese criteria: p=0.011 (heavy drinkers), p< (very heavy drinkers), respectively; IDF and NCEP- ATP III criteria: p<0.001 (heavy drinkers), p<0.001 (very heavy drinkers), respectively]. The prevalence of high TG levels was significantly higher in heavy (p<0.001) and very heavy (p<0.001) drinkers than in nondrinkers. The prevalence of low HDL cholesterol was significantly lower in light (p<0.001), heavy (p=0.011) and very heavy (p<0.001) drinkers than in nondrinkers, thus suggesting that the prevalence of low HDL cholesterol decreases with an increase in alcohol intake. The prevalence of MetS was significantly lower in light 2142

5 Table 2. Comparison of Mean Levels of Each Variable between in the Non-fatty Liver Group (NF) and in Fatty Liver Groups (FL) among Four Alcohol Groups. Alcohol group p Nondrinkers Light drinkers (<20 g/day) Heavy drinkers ( 20, <60 g/day) Very heavy drinkers ( 60 g/day) Age NF:50.8±11.5 NF:48.1±10.9** NF:48.7±9.7** NF:47.8±8.3** <0.001 FL:49.4±10.6 FL:47.9±9.9** FL:49.0±9.2 FL:48.2±8.4 <0.001 Waist circumference, cm NF:80.4±7.3 NF:80.2±7.0 NF:81.9±6.8** NF:83.0±7.6* <0.001 FL:90.2±8.8 FL:89.2±8.4* FL:90.2±7.5 FL:90.1±8.6 <0.001 Systolic BP, mmhg NF:120±13 NF:120±13 NF:124±14** NF:124±14* <0.001 FL:126±14 FL:125±13 FL:129±14** FL:130±13* <0.001 Diastolic BP, mmhg NF:76±9 NF:77±9* NF:80±9** NF:81±10** <0.001 FL:80±9 FL:81±9 FL:83±9** FL:84±9* <0.001 AST, IU/L NF:22.6±8.1 NF:22.6±9.3 NF:23.8±9.6** NF:26.0±13.9** <0.001 FL:27.9±11.5 FL:27.8±12.9 FL:29.0±13.0 FL:33.7±22.4** <0.001 ALT, IU/L NF:22.6±14.1 NF:22.5±16.4 NF:23.2±14.1** NF:25.6±25.1** <0.001 FL:39.0±24.3 FL:38.3±25.1 FL:37.3±22.9 FL:40.2±28.4 <0.001 GTP, IU/L NF:37.7±43.0 NF:40.9±46.6** NF:58.2±55.6** NF:77.7±81.8** <0.001 FL:54.3±51.3 FL:63.9±67.3** FL:87.0±91.3** FL:120.9±129.4** <0.001 FPG, mg/dl NF:97±16 NF:97±14 NF:99±15** NF:98±14** <0.001 FL:106±26 FL:104±22 FL:107±23** FL:109±28** <0.001 HbA1c(NGSP), % NF:5.5±0.6 NF:5.4±0.5** NF:5.4±0.5** NF:5.3±0.5** <0.001 FL:5.8±1.0 FL:5.6±0.8** FL:5.6±0.8** FL:5.6±0.9** <0.001 TG, mg/dl NF:102±57 NF:101±53 NF:114±75** NF:125±83** <0.001 FL:167±128 FL:165±110 FL:175±123* FL:185±132** <0.001 HDL cholesterol, mg/dl NF:53±13 NF:54±13** NF:56±14** NF:57±15** <0.001 FL:45±10 FL:46±11** FL:49±12** FL:51±13** <0.001 LDL cholesterol, mg/dl NF:123±31 NF:121±29 NF:116±29** NF:107±30** <0.001 FL:131±35 FL:128±30 FL:123±31* FL:116±35* <0.001 Ferritin, ng/ml NF:166±104 NF:176±103* NF:209±127** NF:262±175** <0.001 FL:207±116 FL:241±139** FL:293±194** FL:347±236** <0.001 Means with standard deviations of each variable are shown. Light drinkers: <20 g ethanol/day; heavy drinkers: 20 and <60 g ethanol/day; very heavy drinkers: 60 g ethanol/day. * p<0.05; ** p<0.001 compared with nondrinkers. p<0.05, p<0.001 compared with non-fatty liver group. drinkers than in nondrinkers [Japanese criteria: p=0.004; IDF criteria: p=0.01; NCEP-ATP III criteria: p=0.016]. The prevalence of MetS was significantly higher in very heavy drinkers than in nondrinkers [Japanese criteria: p= 0.036; NCEP-ATP III criteria: p<0.001]. The prevalence of MetS according to IDF criteria tended to higher in very heavy drinkers than in nondrinkers; however, the difference was not significant (p=0.066). Discussion Our study indicated that the prevalence of risk factors for MetS (e.g., high BP, hyperglycemia, and high TG) was significantly higher in heavy drinkers than in nondrinkers in a general population of Japanese men. Conversely, our study also showed that the prevalence of large waist circumference and low HDL cholesterol was significantly lower in light drinkers than in nondrinkers. In a previous study conducted in Japan, non- and light drinkers were analyzed together and no comparison was made between non- and light drinkers or between nondrinkers and heavy drinkers (17). In another previous study, a significant association between light (<22 g/day) or heavy ( 22 and <44 g/day) alcohol intake and a lower risk for MetS was seen in young men aged 35 and <45 years old (18). Our study showed that an alcohol intake 20 g/day may increase the prevalence of almost all risk factors of MetS for men; the waist circumference, BP, FPG, and TG levels were significantly higher in heavy ( 20 g/day and <60 g/day) and very heavy ( 60 g/day) drinkers than in nondrinkers and the prevalence of high BP, hyperglycemia, and high TG was higher in heavy and very heavy drinkers than in nondrinkers. Alcohol intake was beneficial for the HDL cholesterol levels regardless of how much alcohol was consumed, and the HDL levels decreased in men diagnosed with fatty liver. Thus, the prevalence of MetS was higher in heavy drinkers ( 20 g/day and <60 g/day) and very heavy drinkers ( 60 g/day) than in nondrinkers, but a significant difference was seen only between very heavy and nondrinkers. The prevalence of MetS was significantly lower in light (<20 g/day) drinkers possibly because the prevalence of a large waist circumference was significantly lower in light drinkers. This coincides with the results from a previous study of young men (18). In our study, the variables most sensitive to alcohol intake were γgtp and ferritin. Some papers have reported an association between serum ferritin levels and risk for MetS; high-normal ferritin concentrations are associated with central adiposity (19), hypertension (20), elevated fasting insulin and blood glucose (21), dyslipidemia (22) and MetS (23). The relationships between ferritin and TG and between ferritin and FPG observed our study completely concurred with previous reports. However, the relationship 2143

6 Table 3. Comparison of Prevalence of Each Risk Factor of Metabolic Syndrome among Four Groups. Alcohol group p Nondrinkers Light drinkers (<20 g/day) Heavy drinkers ( 20, <60 g/day) Very heavy drinkers ( 60 g/day) Waist 85 cm, % * <0.001 Waist 90 cm or BMI>30 kg/m 2,% * * <0.001 (IDF) Waist>102 cm, % * (NCEP-ATP III) Systolic BP 130 mmhg, % ** 38.7** <0.001 Diastolic BP 85 mmhg, % ** 32.2** <0.001 FPG 110 mg/dl, % * 18.1** <0.001 (Japanese criteria) FPG 100 mg/dl, % ** 44.6** <0.001 (IDF, NCEP-ATP III) TG 150 mg/dl, % ** 34.6** <0.001 HDL cholesterol<40 mg/dl, % ** 13.3** 12.5** <0.001 MetS by Japanese criteria, % * * <0.001 MetS by IDF, % * <0.001 MetS by NCEP-ATP III, % * ** <0.001 Means with standard deviations of each variable are shown. Light drinkers: <20 g ethanol/day; heavy drinkers: 20 and <60 g ethanol/day; very heavy drinkers: 0 g ethanol/day. * p<0.05; ** p<0.001 compared with nondrinkers. between ferritin and HDL cholesterol appears to be more complex. An elevated ferritin was accompanied by a lower HDL cholesterol and higher FPG in the fatty liver group compared with the non-fatty liver group (Table 2). Conversely, HDL cholesterol levels increased with an increase in alcohol intake, which further led to an increase in FPG and ferritin levels. Thus, we speculate that alcohol intake may increase HDL cholesterol levels, but HDL cholesterol decreases during steatosis. Fatty liver in non- and light drinkers may be diagnosed as non-alcoholic fatty liver disease (NAFLD) and the levels of ferritin during this condition may suggest the possibility of non-alcoholic steatohepatits (NASH) (24). NAFLD/NASH may accompany a decrease in insulin sensitivity and an increase in basal insulin secretion, which causes an increase in HbA1c. Alcohol consumption has been shown to increase the postprandial glucose response and insulin secretion in healthy subjects (25) although FPG increases as the amount of alcohol intake increases, while HbA1c decreases as the amount of alcohol intake decreases (18). The reason for this discrepancy is unclear and future studies are warranted. There are some limitations to this study. One limitation is that the nondrinkers group in this study also included men who had consumed alcohol in the past. There were 2,506 men who had never consumed alcohol; the other 341 men had had at least one drink. Another limitation of this study is the inclusion of only male subjects. The results of the same parameters conducted in women were not mentioned in this paper because the number of very heavy drinkers among women was very small (approximately 1.5% of all women who visited the Health Management Center of Toranomon Hospital for the Ningen Dock health check-up between January 1997 and December 2011). This low rate may have skewed the statistical analyses; thus, we excluded women from the study. In conclusion, we found that alcohol intake was associated with the risk factors of MetS in men. An alcohol intake <20 g/day may decrease the prevalence of a large waist circumference, low HDL cholesterol and MetS, while an intake 20 g/day may increase the risk factors of MetS, such as hypertension, hyperglycemia and high TG. Alcohol consumption and the prevalence of low HDL cholesterol was found to be inversely linear, thus indicating that alcohol consumption may be beneficial for the HDL cholesterol levels regardless of the amount consumed. The authors state that they have no Conflict of Interest (COI). References 1. Lelbach WK. Chirrhosis in the alcoholic and its relation to the volume of alcohol abuse. Ann NY Acad Sci 252: , Lelbach WK. Liver damage in chronic alcoholism. Results of a clinical, clinical-chemical and bioptic-histologic study of 526 alcoholic patients during withdrawal treatment in a public hospital for alcoholics. 3. Bioptic-histologic findings. Acta Hepatosplenol 14: 9-39, Tsugane S. Alcohol, smoking, and obesity epidemiology in Japan. J Gastroenterol Hepatol 27 (Suppl 2): , Goldberg IJ, Mosca L, Piano MR, Fisher EA. AHA Science Advisory: Wine and your heart: a science advisory for healthcare professionals from the Nutrition Committee, Council on Epidemiology and Prevention, and Council on Cardiovascular Nursing of the American Heart Association. Circulation 103: , Freiberg MS, Cabral HJ, Heeren TC, Vasan RS, Curtis Ellison R. Alcohol consumption and the prevalence of the Metabolic Syndrome in the US: a cross-sectional analysis of data from the Third National Health and Nutrition Examination Survey. Diabetes Care 27: , Yoon YS, Oh SW, Baik HW, Park HS, Kim WY. Alcohol consumption and the metabolic syndrome in Korean adults: the 1998 Korean National Health and Nutrition Examination Survey. Am J Clin Nutr 80: , Baik I, Shin C. Prospective study of alcohol consumption and metabolic syndrome. Am J Clin Nutr 87: , Goude D, Fagerberg B, Hulthe J. Alcohol consumption, the meta- 2144

7 bolic syndrome and insulin resistance in 58-year-old clinically healthy men (AIR study). Clin Sci 102: , Lazarus R, Sparrow D, Weiss ST. Alcohol intake and insulin levels. The normative aging study. Am J Epidemiol 145: , Rimm EB, Williams P, Fosher K, Criqui M, Stampfer MJ. Moderate alcohol intake and lower risk of coronary heart disease: metaanalysis of effects on lipids and haemostatic factors. Br Med J 319: , Marmot MG, Elliott P, Shipley MJ, et al. Alcohol and blood pressure: The INTERSALT study. Br Med J 308: , Aneja J, Basu D, Matto SK, Kohli KK. Metabolic syndrome in alcohol-dependent men: a cross-sectional study. Indian J Psychol Med 35: , Alberti KG, Zimmet P, Shaw J.; IDF Epidemiology Task Force Consensus Group. The metabolic syndrome: a new worldwide definition. Lancet 366: , 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 285: , Eckel RH, Alberti KG, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet 375: , Anonymous. Metabolic syndrome-definition and diagnostic criteria in Japan. Nihon Naika Gakkai Zasshi (J Jpn Soc Intern) 94: , 2005 (in Japanese). 17. Urashima M, Wada T, Fukumoto T, et al. Prevalence of metabolic syndrome in a 22,892 Japanese population and its associations with lifestyle. JMAJ 48: , Wakabayashi I. Influence of age on the relationship between alcohol consumption and metabolic syndrome. Genentology 58: 24-31, Gillum RF. Association of serum ferritin and indices of body fat distribution and obesity in Mexican American men: the Third National Health and Nutrition Examination Survey. Int J Obes Relat Metab Disord J Int Assoc Study Obes 25: , Piperno A, Trombini P, Gelosa M, et al. Increased serum ferritin is common in men with essential hypertension. J Hypertens 20: , Kim CH, Kim HK, Bae SJ, Park JY, Lee KU. Association of elevated serum ferritin concentration with insulin resistance and impaired glucose metabolism in Korean men and women. Metabolism 60: , Halle M, Knig D, Berg A, Keul J, Baumstark M. Relationship of serum ferritin concentrations with metabolic cardiovascular risk factors in men without evidence for coronary artery disease. Atherosclerosis 128: , Li J, Wang R, Luo D, Li S, Xiao C. Association between serum ferritin levels and risk of the metabolic syndrome in Chinese adults: a population study. PLoS One 8: 1-7, Sumida Y, Yoneda M, Hyogo H, et al. A simple clinical scoring system using ferritin, fasting insulin, and type IV collagen 7S for predicting steatohepatitis in nonalcoholic fatty liver disease. J Gastroenterol 46: , Hätönen KA, Virtamo J, Eriksson JG, et al. Modifying effects of alcohol on the postprandial glucose and insulin responses in healthy subjects. Am J Clin Nutr 96: 44-49, The Japanese Society of Internal Medicine