Advertisement

Glucose intolerance and serum aminotransferase activities in Japanese men

      Abstract

      Background/Aims: Elevated activities of serum aminotransferase are commonly observed in patients with diabetes mellitus. Few studies have addressed the relation between glucose intolerance and serum activities of aminotransferase in free-living populations.
      Methods: Using a 75 g oral glucose tolerance test, we examined the association of impaired fasting glycemia (IFG), impaired glucose tolerance (IGT), and type 2 diabetes mellitus with serum aspartate aminotransferase (AST), alanine aminotransferase (ALT), and gamma-glutamyltransferase (GGT) among 4621 men aged 49–59 years of the Japan Self-Defense Forces. Statistical adjustment was made for body mass index, waist-hip ratio, and other possible confounding factors.
      Results: Proportions of an elevated ALT (>50 IU/l) in men with normal glucose tolerance, IFG, IGT, and newly diagnosed diabetes mellitus were 3.5%, 9.5%, 7.7%, and 18.0%, respectively. Adjusted odds ratios of an elevated ALT for IFG, IGT, and newly diagnosed diabetes mellitus were 2.2 (95% confidence interval 1.1–4.3), 1.7 (1.2–2.4), and 4.4 (3.0–6.6), respectively. IGT and diabetes mellitus were also significantly positively associated with elevated AST (>40 IU/l) and GGT (>50 IU/l).
      Conclusions: Glucose intolerance is associated with elevated serum aminotransferase independent of obesity, but even a mildly elevated ALT is relatively uncommon in free-living men with glucose intolerance.

      Keywords

      1. Introduction

      Serum activities of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) are the most frequently used indicators of liver disease [
      • Reichling J.J
      • Kaplan M.M
      Clinical use of serum enzymes in liver disease.
      ,
      • Sherman K.E
      Alanine aminotransferase in clinical practice. A review.
      ], and elevated activities of serum aminotransferase are commonly found in clinical practice and health screening [
      • Sherman K.E
      Alanine aminotransferase in clinical practice. A review.
      ,
      • Byron D
      • Minuk G.Y
      Clinical hepatology: profile of an urban, hospital-based practice.
      ,
      • Salvaggio A
      • Periti M
      • Miano L
      • Tavanelli M
      • Marzorati D
      Body mass index and liver enzyme activity in serum.
      ,
      • Lozano M
      • Cid J
      • Bedini J.L
      • Mazzara R
      • Gimenez N
      • Mas E
      • et al.
      Study of serum alanine-aminotransferase levels in blood donors in Spain.
      ]. Such abnormalities without either hepatitis virus infection or heavy alcohol use have recently drawn much attention as non-alcoholic steatohepatitis [
      • James O.F
      • Day C.P
      Non-alcoholic steatohepatitis (NASH): a disease of emerging identity and importance.
      ,
      • James O
      • Day C
      Non-alcoholic steatohepatitis: another disease of affluence.
      ]. Elevations in serum aminotransferase are frequently observed in patients with diabetes mellitus [
      • Salmela P.I
      • Sotaniemi E.A
      • Niemi M
      • Maentausta O
      Liver function tests in diabetic patients.
      ,
      • Nagore N
      • Scheuer P.J
      The pathology of diabetic hepatitis.
      ,
      • Silverman J.F
      • O'Brien K.F
      • Long S
      • Leggett N
      • Khazanie P.G
      • Pories W.J
      • et al.
      Liver pathology in morbidly obese patients with and without diabetes.
      ], and these changes are mostly attributed to fatty infiltration of the liver [
      • Nagore N
      • Scheuer P.J
      The pathology of diabetic hepatitis.
      ,
      • Silverman J.F
      • O'Brien K.F
      • Long S
      • Leggett N
      • Khazanie P.G
      • Pories W.J
      • et al.
      Liver pathology in morbidly obese patients with and without diabetes.
      ]. Obesity is also known to be associated with elevated activities of serum aminotransferase, especially with elevated ALT activity [
      • Salvaggio A
      • Periti M
      • Miano L
      • Tavanelli M
      • Marzorati D
      Body mass index and liver enzyme activity in serum.
      ,
      • Lozano M
      • Cid J
      • Bedini J.L
      • Mazzara R
      • Gimenez N
      • Mas E
      • et al.
      Study of serum alanine-aminotransferase levels in blood donors in Spain.
      ,
      • James O.F
      • Day C.P
      Non-alcoholic steatohepatitis (NASH): a disease of emerging identity and importance.
      ]. Because obesity is an important risk factor for type 2 diabetes mellitus [
      • DeFronzo R.A
      • Bonadonna R.C
      • Ferrannini E
      Pathogenesis of NIDDM. A balanced overview.
      ,
      • Yki-Järvinen H
      Pathogenesis of non-insulin-dependent diabetes mellitus.
      ], the association between diabetes mellitus and elevated activities of AST and ALT could be attributable to obesity in patients with diabetes mellitus.
      Few studies have addressed the relation between diabetes mellitus and elevated serum aminotransferase in free-living populations. In a cross-sectional study of the Hispanic American population [
      • Meltzer A.A
      • Everhart J.E
      Association between diabetes and elevated serum alanine aminotransferase activity among Mexican Americans.
      ], diabetes mellitus was associated with elevated serum ALT independent of obesity. Another population study in the United States [
      • Erbey J.R
      • Silberman C
      • Lydick E
      Prevalence of abnormal serum alanine aminotransferase levels in obese patients and patients with type 2 diabetes.
      ] also reported that raised serum ALT was independently associated with both obesity and type 2 diabetes mellitus. In these population studies [
      • Meltzer A.A
      • Everhart J.E
      Association between diabetes and elevated serum alanine aminotransferase activity among Mexican Americans.
      ,
      • Erbey J.R
      • Silberman C
      • Lydick E
      Prevalence of abnormal serum alanine aminotransferase levels in obese patients and patients with type 2 diabetes.
      ], only body mass index was used as an index of obesity, and abdominal obesity, which is more important in the development of type 2 diabetes mellitus [
      • DeFronzo R.A
      • Bonadonna R.C
      • Ferrannini E
      Pathogenesis of NIDDM. A balanced overview.
      ,
      • Yki-Järvinen H
      Pathogenesis of non-insulin-dependent diabetes mellitus.
      ], was not taken into consideration. We investigated the relation of glucose tolerance status determined by the 75 g oral glucose tolerance test (OGTT) with serum AST and ALT activities, controlling for not only body mass index but also waist-hip ratio and other factors in Japanese men. We also examined the association with serum gamma-glutamyltransferase (GGT) because serum GGT may be a correlate of hyperinsulinemia or insulin resistance [
      • Ikai E
      • Ishizaki M
      • Suzuki Y
      • Ishida M
      • Noborizaka Y
      • Yamada Y
      Association between hepatic steatosis, insulin resistance and hyperinsulinemia as related to hypertension in alcohol consumers and obese people.
      ,
      • Perry I.J
      • Wannamethee S.G
      • Shaper A.G
      Prospective study of serum gamma-glutamyltransferase and risk of NIDDM.
      ].

      2. Subjects and methods

      2.1 Study subjects

      Subjects were male self-defense officials who received a pre-retirement health examination at the Self-Defense Forces (SDF) Fukuoka Hospital from January 1991 to December 1994, at the SDF Kumamoto Hospital from January 1991 to December 1994, at the SDF Sapporo Hospital from April 1992 to December 1994, or at the SDF Central Hospital from January 1993 to December 1994. In the consecutive series of 4883 men aged 48–59 years, 262 men were excluded for the following reasons: insulin treatment (n=11), undetermined glucose tolerance due to lack of a post-load glucose measurement (n=8), missing information on waist-to-hip ratio (n=3), known chronic hepatitis or liver cirrhosis (n=87), positive test for hepatitis C virus (HCV) antibody (n=43), and positive test for hepatitis B virus surface (HBs) antigen (n=158); some men had two or more conditions for exclusion. While HBs antigen was tested with all the subjects, the test for HCV antibody was introduced in 1993. Thus, the study subjects (n=4621) included 2331 men of unknown status regarding HCV antibody. Details of the health examination have been described previously [
      • Shinchi K
      • Kono S
      • Honjo S
      • Todoroki I
      • Sakurai Y
      • Imanishi K
      • et al.
      Obesity and adenomatous polyps of the sigmoid colon.
      ,
      • Kono S
      • Shinchi K
      • Todoroki I
      • Honjo S
      • Sakurai Y
      • Wakabayashi K
      • et al.
      Gallstone disease among Japanese men in relation to obesity, glucose intolerance, exercise, alcohol use, and smoking.
      ,
      • Sasazuki S
      • Kono S
      • Todoroki I
      • Honjo S
      • Sakurai Y
      • Wakabayashi K
      • et al.
      Impaired glucose tolerance, diabetes mellitus, and gallstone disease: an extended study of male self-defense officials in Japan.
      ,
      • Honjo S
      • Kono S
      • Coleman M.P
      • Shinchi K
      • Sakurai Y
      • Todoroki I
      • et al.
      Coffee consumption and serum aminotransferases in middle-aged Japanese men.
      ]. The study was conducted in accordance with the Helsinki Declaration, and written informed consent was obtained from each subject.

      2.2 Laboratory methods

      A 75 g OGTT was done after subjects fasted overnight. Glucose concentrations were assayed by the glucose oxidase method with a commercial kit (Shino Test Co. Ltd., Tokyo) using plasma at the Fukuoka, Sapporo, and Central Hospitals and whole capillary blood at the Kumamoto Hospital. On the basis of the new criteria of the World Health Organization (WHO) in Ref. [
      • Alberti K.G.M.M
      • Zimmet P.Z
      for the WHO Consultation
      Definition, diagnosis and classification of diabetes mellitus and its complication. Part 1: diagnosis and classification of diabetes mellitus. Provisional report of a WHO consultation.
      ], subjects were classified as having normal glucose tolerance, impaired fasting glycemia (IFG), impaired glucose tolerance (IGF), and type 2 diabetes mellitus. Men under dietary or drug treatment for diabetes mellitus were regarded as having type 2 diabetes mellitus under treatment.
      Venus blood after an overnight fast was used for the determination of serum activities of AST, ALT, and GGT. The assay was done by the enzymatic method with an autoanalyzer at each hospital using commercial reagents [
      • Karmen A
      • Wróblewski F
      • LaDue J.S
      Transaminase activity in human blood.
      ,
      • Szasz G
      A kinetic photometric method for serum-glutamyl transpeptidase.
      ]. Serum AST and ALT values of greater than 40 IU/l were defined as elevated AST and ALT, respectively, and serum GGT greater than 50 IU/l was defined as an elevated GGT, on the basis of reference values provided by manufacturers of the reagents. HBs antigen was tested by reversed passive hemagglutination, and HCV antibody was assayed by passive hemagglutination (second generation) using commercial kits (Dainabot Co. Ltd., Tokyo).

      2.3 Anthropometric and lifestyle data

      Body weight (kg) and height (cm) were recorded, and body mass index (kg/m2) was calculated as an index of overall obesity. Waist and hip circumference were measured to the nearest centimeter, with the participant standing in an erect position. Waist circumference was measured at the umbilical point, and hip circumference around the buttocks. Medical history and current medication were elicited by ward nurses and physicians. Smoking habit, alcohol use, and other lifestyle characteristics were ascertained by using a self-administered questionnaire as described elsewhere in detail [
      • Kono S
      • Shinchi K
      • Todoroki I
      • Honjo S
      • Sakurai Y
      • Wakabayashi K
      • et al.
      Gallstone disease among Japanese men in relation to obesity, glucose intolerance, exercise, alcohol use, and smoking.
      ,
      • Honjo S
      • Kono S
      • Coleman M.P
      • Shinchi K
      • Sakurai Y
      • Todoroki I
      • et al.
      Coffee consumption and serum aminotransferases in middle-aged Japanese men.
      ].

      2.4 Statistical analysis

      The differences in means and proportions of confounding factors according to glucose tolerance status were statistically tested by one-way analysis of variance and chi-square test, respectively. Analysis of covariance was used to calculate the means of AST, ALT, and GGT according to glucose tolerance status with adjustment for possible confounding effects of hospital, rank in the SDF, cigarette smoking, alcohol use, coffee consumption, body mass index, and waist-hip ratio. Because the distributions of values of the liver enzymes were skewed to the right side, the values were transformed to the scale of natural logarithm so that each distribution was approximated to the normal distribution. Presented means of AST, ALT, and GGT are thus always geometric means. Age was in a limited range from 49 to 59 years, and 99.3% were aged 50–54 years. Age was thus omitted in the present analysis. Repeated analyses controlling for age as a continuos variable produced essentially the same results as reported below. Rank in the SDF was categorized into three classes (low, middle, and high). Cigarette smoking was categorized in five groups (never, past, and current smoking with a consumption of <15, 15–24, or 25+ cigarettes per day), alcohol use was in five categories (never, past, and current drinking with a consumption of <30, 30–59, or 60+ ml of alcohol per day), and coffee consumption was in categories of 0, 1–2, 3–4, and 5+ cups per day. Body mass index and waist-hip ratio were categorized at the cutoff points of the 25th, 50th, and 75th percentiles in each distribution. Indicator variables representing categories of the confounding factors were included in the model as independent variables.
      Multiple logistic regression analysis was used to examine the relation of glucose tolerance status to elevated activities of AST and ALT in terms of odds ratio (OR) with adjustment for the above-mentioned confounding variables. The 95% confidence interval (CI) of OR was estimated by using the standard error of a logistic regression coefficient for each indicator variable.
      The trend of the association between the degree of pathologic glucose tolerance and elevation in liver enzymes was assessed by using multiple linear regression for the means and by using multiple logistic regression for the ORs of elevated activities. In these analyses, ordinal scores (0, 1, and 2) were assigned to the three categories of glucose tolerance (normal, IFG and IGT combined, and newly diagnosed diabetes mellitus). P values (two-sided) less than 0.05 were regarded as statistically significant. All statistical analyses were done by using the Statistical Analysis System (SAS) version 6.12 (SAS Institute, Inc., Cary, NC).

      3. Results

      Medians of serum AST, ALT, and GGT were 19 IU/l (range 4–183 IU/l), 16 IU/l (range 1–144 IU/l), and 27 IU/l (range 4–630 IU/l), respectively. Numbers of men with elevated AST, ALT, and GGT were 123 (2.7%), 223 (5.0%), and 1065 (23.0%), respectively. Those having two times the upper limit of normal numbered 16 (0.3%) for AST, 26 (0.6%) for ALT, and 290 (6.3%) for GGT. Numbers of men by glucose tolerance status were as follows: normal, 3410; IFG, 116; IGT, 750; newly diagnosed type 2 diabetes mellitus, 233; and treated diabetes mellitus, 112.
      Table 1 shows the means of body mass index and waist-hip ratio as well as the prevalence of other potential confounding factors according to glucose tolerance status. Body mass index and waist-hip ratio were greater in men with IFG, IGT, and newly diagnosed type 2 diabetes mellitus, but not in those with treated diabetes mellitus, than in those with normal glucose tolerance. While the prevalence of cigarette smoking did not materially vary with glucose tolerance status, heavy smoking was more frequent among men with newly diagnosed diabetes mellitus. Coffee use was less frequent in those with diabetes mellitus whether they were under treatment or not.
      Table 1Potential confounding variables according to glucose tolerance status (n=4621)
      FactorNormalIFGIGTNew type 2 diabetesTreated type 2 diabetesOverall P value
      One-way analysis of variance for mean difference and chi-square test for the difference in proportions.
      Number of subjects3410116750233112
      Body mass index (mean)23.724.724.724.823.60.0001
      Waist–hip ratio (mean)0.8970.9170.9150.9190.9050.0001
      Current smoking (%)49.539.748.953.650.90.18
      Heavy smoking (%)
      Smoking 25+ cigarettes per day.
      17.19.517.625.320.50.003
      Alcohol use (%)
      Drinking at least once per week.
      80.191.480.581.575.90.03
      Heavy alcohol use (%)
      Drinking 60+ ml of alcohol per day.
      23.125.925.730.022.30.10
      Daily use of coffee (%)68.059.561.756.750.00.001
      a One-way analysis of variance for mean difference and chi-square test for the difference in proportions.
      b Smoking 25+ cigarettes per day.
      c Drinking at least once per week.
      d Drinking 60+ ml of alcohol per day.
      Table 2 shows adjusted geometric means of serum AST, ALT, and GGT with and without adjustment for body mass index and waist-hip ratio. ALT levels were progressively higher for IFG, IGT, and newly diagnosed type 2 diabetes mellitus as compared with normal glucose tolerance. The variation of ALT levels according to glucose tolerance status was more profound than that of AST, but less marked than that of GGT. Both AST and ALT activities in the group of treated diabetes mellitus were almost the same as those in the group of normal glucose tolerance. Adjustment for body mass index and waist-hip ratio slightly dwarfed the variation in activities of the liver enzymes according to glucose tolerance status. The increasing trend of means for more pathologic glucose tolerance (normal, IFG and IGT combined, and newly diagnosed diabetes mellitus) was highly significant for each of the three enzymes (trend P=0.0001).
      Table 2Adjusted geometric means and 95% CIs (in parentheses) of serum AST, ALT, and GGT according to glucose tolerance status
      Glucose tolerance statusASTALTGGT
      Adjusted geometric mean
      Adjusted for hospital, rank, smoking, alcohol use, and coffee consumption.
      Normal18.7 (18.5–18.9)15.6 (15.4–15.9)27.6 (27.0–28.2)
      IFG19.5 (18.5–20.7)17.6 (16.0–19.3)29.9 (26.5–33.7)
      IGT20.3 (19.8–20.7)18.9 (18.2–19.6)35.7 (34.0–37.4)
      New type 2 diabetes20.7 (19.9–21.6)21.6 (20.2–23.1)45.7 (42.0–49.8)
      Treated type 2 diabetes17.4 (16.4–18.4)16.4 (14.9–18.1)35.8 (31.7–40.5)
      Fully-adjusted geometric mean
      Adjusted for body mass index and waist-hip ratio in addition to the above factors.
      Normal18.8 (18.6–19.0)15.9 (15.7–16.2)28.1 (27.5–28.7)
      IFG19.1 (18.1–20.3)16.4 (15.0–17.9)28.1 (25.0–31.5)
      IGT20.0 (19.5–20.4)17.9 (17.2–18.5)33.9 (32.4–35.5)
      New type 2 diabetes20.3 (19.5–21.1)20.1 (18.9–21.4)42.9 (39.5–46.5)
      Treated type 2 diabetes17.4 (16.4–18.4)16.4 (15.0–18.0)35.5 (31.6–40.0)
      a Adjusted for hospital, rank, smoking, alcohol use, and coffee consumption.
      b Adjusted for body mass index and waist-hip ratio in addition to the above factors.
      The proportions of elevated AST, ALT, and GGT were higher in men with IFG, IGT, and newly diagnosed type 2 diabetes mellitus than in those with normal glucose tolerance, and the proportions were much higher among men with newly diagnosed diabetes mellitus than among those with IFG or IGT (Table 3) . With adjustment for factors other than obesity indices, substantial increases in the prevalence odds of elevated AST, ALT, and GGT were observed among men with newly diagnosed diabetes mellitus, as compared with those with normal glucose tolerance; the increase in the odds of elevated ALT was slightly more profound. IGT was also associated with a statistically significantly increase in the odds of elevated activities of the liver enzymes. Treated diabetes mellitus was unrelated to either elevated AST or ALT, but was positively associated with an elevated GGT.
      Table 3Numbers of men with elevated serum AST (>40 IU/l), ALT (>40 IU/l), and GGT (>50 IU/l) and adjusted OR with 95% CIs according to glucose tolerance status
      Glucose tolerance statusElevated ASTElevated ALTElevated GGT
      No. (%)OR (95% CI)
      Adjusted for hospital, rank, smoking, alcohol use, and coffee consumption.
      No. (%)OR (95% CI)
      Adjusted for hospital, rank, smoking, alcohol use, and coffee consumption.
      No. (%)OR (95% CI)
      Adjusted for hospital, rank, smoking, alcohol use, and coffee consumption.
      Normal62 (1.8)1.0 (referent)118 (3.5)1.0 (referent)663 (19.4)1.0 (referent)
      IFG5 (4.3)2.1 (0.8–5.4)11 (9.5)3.0 (1.6–5.9)32 (27.6)1.3 (0.9–2.1)
      IGT33 (4.4)2.5 (1.6–3.8)58 (7.7)2.3 (1.6–3.2)220 (29.3)1.6 (1.3–2.0)
      New type 2 diabetes21 (9.0)4.7 (2.8–8.0)42 (18.0)6.0 (4.1–8.9)115 (49.4)4.0 (3.0–5.4)
      Treated type 2 diabetes2 (1.8)0.9 (0.2–3.7)4 (3.6)1.0 (0.4–2.8)35 (31.3)2.0 (1.3–3.1)
      a Adjusted for hospital, rank, smoking, alcohol use, and coffee consumption.
      Table 4 shows adjusted ORs of elevated AST, ALT, and GGT in relation to glucose tolerance status and other covariates in the multivariate analysis including body mass index and waist-hip ratio. Increases in the adjusted ORs of elevated AST, ALT, and GGT associated with newly diagnosed diabetes mellitus remained substantial, and the trend for normal glucose tolerance, IFG and IGT combined, and newly diagnosed diabetes mellitus was again highly significant regarding each of the liver enzymes (P=0.0001). Body mass index and waist-hip ratio were statistically significantly related to increased ORs of elevated ALT and GGT, but not with elevated AST. On the other hand, heavy alcohol use was strongly associated with elevated AST and GGT, and not with elevated ALT. Smoking was not measurably associated with either elevated AST or ALT. Coffee consumption was associated with a decreased OR of elevated AST, and showed a dose-dependent inverse association with GGT.
      Table 4Adjusted ORs and 95% CIs of elevated serum AST (>40 IU/l), ALT (>40 IU/l), and GGT (>50 IU/l) in multivariate analysis
      The multivariate model included indicator variables for hospital, rank, smoking, alcohol use, coffee use, body mass index, waist-hip ratio, and glucose tolerance.
      FactorCategoryElevated ASTElevated ALTElevated GGT
      Glucose toleranceNormal1.0 (referent)1.0 (referent)1.0 (referent)
      IFG1.8 (0.7–4.8)2.2 (1.1–4.3)1.1 (0.7–1.7)
      IGT2.2 (1.4–3.5)1.7 (1.2–2.4)1.4 (1.2–1.7)
      New type 2 diabetes3.8 (2.2–6.5)4.4 (3.0–6.6)3.4 (2.5–4.6)
      Treated type 2 diabetes0.8 (0.2–3.6)0.9 (0.3–2.5)1.9 (1.2–3.0)
      Body mass index
      Categorized at quartiles.
      <22.151.0 (referent)1.0 (referent)1.0 (referent)
      22.15–23.800.4 (0.2–0.7)0.7 (0.4–1.2)1.0 (0.8–1.3)
      23.81–25.460.5 (0.3–0.9)1.3 (0.7–2.2)1.1 (0.9–1.5)
      25.47+1.2 (0.7–2.0)2.6 (1.6–4.4)1.5 (1.1–1.9)
      Waist–hip ratio
      Categorized at quartiles.
      <0.8721.0 (referent)1.0 (referent)1.0 (referent)
      0.872–0.9011.1 (0.5–2.1)2.2 (1.2–4.3)1.6 (1.2–2.0)
      0.902–0.9321.7 (0.9–3.2)3.1 (1.7–6.0)1.8 (1.4–2.3)
      0.933+1.7 (0.9–3.3)4.1 (2.2–7.8)2.6 (2.0–3.4)
      Cigarette smoking (no./day)None1.0 (referent)1.0 (referent)1.0 (referent)
      Former1.2 (0.7–2.0)0.9 (0.6–1.3)1.2 (0.9–1.5)
      <151.1 (0.3–3.8)1.2 (0.6–2.5)1.2 (0.8–1.8)
      15–241.1 (0.6–2.0)0.9 (0.6–1.4)1.4 (1.1–1.7)
      25+1.6 (0.9–2.9)0.8 (0.5–1.2)1.4 (1.1–1.8)
      Alcohol use (ml/day)None1.0 (referent)1.0 (referent)1.0 (referent)
      Former2.3 (0.6–8.9)1.5 (0.7–3.3)1.4 (0.7–2.8)
      <301.6 (0.7–3.8)0.7 (0.5–1.1)2.4 (1.8–3.3)
      30–592.0 (0.9–4.5)0.7 (0.5–1.1)4.9 (3.6–6.7)
      60+4.5 (2.1–9.7)0.8 (0.5–1.2)8.4 (6.1–11.4)
      Coffee use (cups/day)01.0 (referent)1.0 (referent)1.0 (referent)
      1–20.5 (0.3–0.9)0.9 (0.6–1.2)0.8 (0.7–1.0)
      3–40.6 (0.3–1.0)1.0 (0.7–1.4)0.7 (0.6–0.8)
      5+0.8 (0.4–1.5)0.8 (0.5–1.4)0.6 (0.5–0.8)
      a The multivariate model included indicator variables for hospital, rank, smoking, alcohol use, coffee use, body mass index, waist-hip ratio, and glucose tolerance.
      b Categorized at quartiles.

      4. Discussion

      The present study showed a strong, positive association between type 2 diabetes mellitus and elevated activities of serum AST and ALT independent of body mass index, waist-hip ratio, and other confounding factors. The findings consolidate the previous observation that diabetes mellitus was associated with elevated ALT independent of obesity [
      • Meltzer A.A
      • Everhart J.E
      Association between diabetes and elevated serum alanine aminotransferase activity among Mexican Americans.
      ,
      • Erbey J.R
      • Silberman C
      • Lydick E
      Prevalence of abnormal serum alanine aminotransferase levels in obese patients and patients with type 2 diabetes.
      ]. The increase in the odds of elevated ALT associated with type 2 diabetes mellitus was much greater in the present study than in these previous studies; adjusted ORs for diabetes mellitus compared with non-diabetes were reported to be two or three [
      • Meltzer A.A
      • Everhart J.E
      Association between diabetes and elevated serum alanine aminotransferase activity among Mexican Americans.
      ,
      • Erbey J.R
      • Silberman C
      • Lydick E
      Prevalence of abnormal serum alanine aminotransferase levels in obese patients and patients with type 2 diabetes.
      ]. In these studies, however, IFG and IGT were not examined separately, and the physician's diagnosis of diabetes mellitus was combined with unknown (newly diagnosed) diabetes mellitus. The present study demonstrated that both IFG and IGT were also significantly associated with approximately two-fold increases in the odds of elevated ALT. Further, it is noteworthy that treated diabetes mellitus was unrelated to serum aminotransferase activities. Failure to differentiate IFG and IGT from normal glucose tolerance and diabetes mellitus under treatment from newly diagnosed diabetes mellitus necessarily underestimates the effect of diabetes mellitus on serum aminotransferase activities.
      The present study also showed a strong, positive association between type 2 diabetes mellitus and GGT. A greater elevation in the geometric mean was observed for GGT than for ALT among men with type 2 diabetes mellitus, and even men under treatment for diabetes mellitus showed a nearly two-fold, statistically significant increase in the OR of an elevated GGT. Serum GGT has been implicated as a marker of insulin resistance [
      • Ikai E
      • Ishizaki M
      • Suzuki Y
      • Ishida M
      • Noborizaka Y
      • Yamada Y
      Association between hepatic steatosis, insulin resistance and hyperinsulinemia as related to hypertension in alcohol consumers and obese people.
      ,
      • Perry I.J
      • Wannamethee S.G
      • Shaper A.G
      Prospective study of serum gamma-glutamyltransferase and risk of NIDDM.
      ], and elevated levels of GGT were also shown to be predictive of the occurrence of type 2 diabetes mellitus [
      • Perry I.J
      • Wannamethee S.G
      • Shaper A.G
      Prospective study of serum gamma-glutamyltransferase and risk of NIDDM.
      ]. The present findings on GGT are consistent with those of the previous studies [
      • Ikai E
      • Ishizaki M
      • Suzuki Y
      • Ishida M
      • Noborizaka Y
      • Yamada Y
      Association between hepatic steatosis, insulin resistance and hyperinsulinemia as related to hypertension in alcohol consumers and obese people.
      ,
      • Perry I.J
      • Wannamethee S.G
      • Shaper A.G
      Prospective study of serum gamma-glutamyltransferase and risk of NIDDM.
      ].
      Several important confounding factors were taken into account in the present study. Not only body mass index but also waist-hip ratio was controlled for. It has consistently been shown that abdominal obesity is more important as a predictor for type 2 diabetes mellitus than body size [
      • DeFronzo R.A
      • Bonadonna R.C
      • Ferrannini E
      Pathogenesis of NIDDM. A balanced overview.
      ,
      • Yki-Järvinen H
      Pathogenesis of non-insulin-dependent diabetes mellitus.
      ]. In fact, the waist-hip ratio was more strongly associated with elevated ALT than body mass index. The importance of abdominal obesity for elevated ALT was also noted in men, but not in women, in a cross-sectional study in Denmark [
      • Thulstrup A.M
      • Norgard B
      • Steffensen F.H
      • Vilstrup H
      • Sorensen H.T
      • Lauritzen T
      Waist circumference and body mass index as predictors of elevated alanine transaminase in Danes aged 30 to 50 years.
      ]. Several studies have shown an increased risk of type 2 diabetes mellitus associated with high alcohol intake [
      • Holbrook T.L
      • Barrett-Connor E
      • Wingard D.L
      A prospective population-based study of alcohol use and non-insulin-dependent diabetes mellitus.
      ,
      • Kao W.H.L
      • Puddey I.B
      • Boland L.B
      • Watson R.L
      • Brancati F.L
      Alcohol consumption and the risk of type 2 diabetes mellitus. Atherosclerosis Risk in Communities Study.
      ] and smoking [
      • Will J.C
      • Galuska D.A
      • Ford E.S
      • Mokdad A
      • Calle E.E
      Cigarette smoking and diabetes mellitus: evidence of a positive association from a large prospective cohort study.
      ], and thus we included these factors as possible confounding factors. While alcohol use was positively associated with serum ALT [
      • Meltzer A.A
      • Everhart J.E
      Association between diabetes and elevated serum alanine aminotransferase activity among Mexican Americans.
      ,
      • Erbey J.R
      • Silberman C
      • Lydick E
      Prevalence of abnormal serum alanine aminotransferase levels in obese patients and patients with type 2 diabetes.
      ] as well as with serum AST [
      • Steffensen F.H
      • Sorensen H.T
      • Brock A
      • Vilstrup H
      • Lauritzen T
      Alcohol consumption and serum liver-derived enzymes in a Danish population aged 30–50 years.
      ,
      • Whitehead T.P
      • Robinson D
      • Allaway S.L
      The effects of cigarette smoking and alcohol consumption on serum liver enzyme activities: a dose-related study in men.
      ] in other population studies, alcohol use was virtually unrelated to elevated ALT in the present study. We have no clear explanation for the lack of a positive association between alcohol consumption and serum ALT. No material association was noted between smoking and either serum AST or ALT, and the findings are consistent with results in other studies [
      • Meltzer A.A
      • Everhart J.E
      Association between diabetes and elevated serum alanine aminotransferase activity among Mexican Americans.
      ,
      • Steffensen F.H
      • Sorensen H.T
      • Brock A
      • Vilstrup H
      • Lauritzen T
      Alcohol consumption and serum liver-derived enzymes in a Danish population aged 30–50 years.
      ,
      • Whitehead T.P
      • Robinson D
      • Allaway S.L
      The effects of cigarette smoking and alcohol consumption on serum liver enzyme activities: a dose-related study in men.
      ]. Coffee drinking was recently shown to be inversely associated with serum aminotransferase activities [
      • Meltzer A.A
      • Everhart J.E
      Association between diabetes and elevated serum alanine aminotransferase activity among Mexican Americans.
      ,
      • Honjo S
      • Kono S
      • Coleman M.P
      • Shinchi K
      • Sakurai Y
      • Todoroki I
      • et al.
      Coffee consumption and serum aminotransferases in middle-aged Japanese men.
      ]. In the present study population, coffee use was less frequent in men with type 2 diabetes mellitus.
      It is known that liver dysfunction associated with chronic hepatitis or liver cirrhosis results in glucose intolerance [
      • Sherlock S
      • Dooley J
      ]. In this respect, subjects with chronic hepatitis or liver cirrhosis were excluded in the present analysis. Men with a positive test for HBs antigen or HCV antibody were also excluded, but the latter condition was not tested in half of the study subjects. If HCV infection is positively associated with glucose intolerance, positive associations of glucose intolerance with serum ALT and AST may have been overestimated. However, there was no material variation in the positive rate of HCV according to the category of glucose tolerance. In men whose HCV antibody was tested (n=2320), the proportions of HCV antibody positivity among men with normal glucose tolerance, IFG, IGT, and newly diagnosed diabetes mellitus were 1.0%, 3.0%, 1.4%, and 1.6%, respectively. The positive rate was slightly higher in only those with treated diabetes mellitus (7.3%).
      It may be argued that men serving for the SDF until retirement may differ from those of the general population in various characteristics and that the findings may not be generalized. However, the prevalence of IGT and type 2 diabetes mellitus on the basis of the older WHO criteria [

      World Health Organization Expert Committee on Diabetes Mellitus. World Health Organization Tech Rep Ser 1980:646.

      ] did not differ much from the rates reported in other studies in Japan [
      • Sekikawa A
      • Sugiyama K
      • Tominaga M
      • Manaka H
      • Takahashi K
      • Sasaki H
      • et al.
      Prevalence of diabetes and impaired glucose tolerance in Funagata area, Japan.
      ,
      • Ohmura T
      • Ueda K
      • Kiyohara Y
      • Kato I
      • Iwamoto H
      • Nakayama K
      • et al.
      Prevalence of type 2 (non-insulin-dependent) diabetes mellitus and impaired glucose tolerance in the Japanese general population: the Hisayama study.
      ]. The prevalence rates of IGT ranged from 13% to 17%, and those of type 2 diabetes mellitus were from 9% to 14% among middle-aged men in these studies; the corresponding figures in the present population were 16.5% and 7.0%, respectively.
      Many studies have shown that hyperinsulinemia or insulin resistance is associated with non-alcoholic steatohepatitis [
      • Tankurt E
      • Biberoglu S
      • Ellidokuz E
      • Hekimsoy Z
      • Akpinar H
      • Comlekci A
      • et al.
      Hyperinsulinemia and insulin resistance in non-alcoholic steatohepatitis.
      ,
      • Marchesini G
      • Brizi M
      • Morselli-Labate A.M
      • Bianchi G
      • Bugianesi E
      • McCullough A.J
      • et al.
      Association of nonalcoholic fatty liver disease with insulin resistance.
      ,
      • Cömert B
      • Mas M.R
      • Erdem H
      • Dinc A
      • Saglamkaya U
      • Cigerim M
      • et al.
      Insulin resistance in non-alcoholic steatohepatitis.
      ,
      • Marchesini G
      • Brizi M
      • Bianchi G
      • Tomassetti S
      • Bugianesi E
      • Lenzi M
      • et al.
      Nonalcoholic fatty liver disease: a feature of the metabolic syndrome.
      ,
      • Dixon J.B
      • Bhathal P.S
      • O'Brien P.E
      Nonalcoholic fatty liver disease: predictors of nonalcoholic steatohepatitis and liver fibrosis in the severely obese.
      ,
      • Willner I.R
      • Waters B
      • Patil S.R
      • Reuben A
      • Morelli J
      • Riely C.A
      Ninety patients with nonalcoholic steatohepatitis: insulin resistance, familial tendency, and severity of disease.
      ,
      • Pagano G
      • Pacini G
      • Musso G
      • Gambino R
      • Mecca F
      • Depetris N
      • et al.
      Nonalcoholic steatohepatitis, insulin resistance, and metabolic syndrome: further evidence for an etiologic association.
      ,
      • Chitturi S
      • Abeygunasekera S
      • Farrell G.C
      • Holmes-Walker J
      • Hui J.M
      • Fung C
      • et al.
      NASH and insulin resistance: insulin hypersecretion and specific association with the insulin resistance syndrome.
      ]. In some of these studies [
      • Willner I.R
      • Waters B
      • Patil S.R
      • Reuben A
      • Morelli J
      • Riely C.A
      Ninety patients with nonalcoholic steatohepatitis: insulin resistance, familial tendency, and severity of disease.
      ,
      • Chitturi S
      • Abeygunasekera S
      • Farrell G.C
      • Holmes-Walker J
      • Hui J.M
      • Fung C
      • et al.
      NASH and insulin resistance: insulin hypersecretion and specific association with the insulin resistance syndrome.
      ], virtually all patients with non-alcoholic steatohepatitis were found to have insulin resistance. A positive association between insulin resistance and non-alcoholic steatohepatitis was observed even in those without diabetes mellitus [
      • Marchesini G
      • Brizi M
      • Morselli-Labate A.M
      • Bianchi G
      • Bugianesi E
      • McCullough A.J
      • et al.
      Association of nonalcoholic fatty liver disease with insulin resistance.
      ,
      • Pagano G
      • Pacini G
      • Musso G
      • Gambino R
      • Mecca F
      • Depetris N
      • et al.
      Nonalcoholic steatohepatitis, insulin resistance, and metabolic syndrome: further evidence for an etiologic association.
      ]. The present study provides confirmatory evidence that insulin resistance is associated with liver pathology as indicated by raised transaminases. However, it should be noted that even mildly elevated ALT was relatively uncommon in men with glucose intolerance; less than 10% of those with IFG or IGT and no more than 18% of those with newly diagnosed type 2 diabetes mellitus had an elevated ALT. As also suggested in another population study [
      • Erbey J.R
      • Silberman C
      • Lydick E
      Prevalence of abnormal serum alanine aminotransferase levels in obese patients and patients with type 2 diabetes.
      ], liver pathology seems to occur in only a minority of patients with insulin resistance or glucose intolerance. Finally, because the present study was cross-sectional, the observed association of glucose intolerance with elevated aminotransferase activities does not necessarily indicate a causal relationship. Although the finding that treated diabetes mellitus was unrelated to elevated serum aminotransferase activities suggests that treatment of diabetes mellitus may reverse liver pathology, insulin resistance and glucose intolerance could be a result of obesity-related fatty liver as well as contributory causes.

      Acknowledgements

      This study was supported by Grants-in Aid for Scientific Research (B) (12470098) from the Japan Society for the Promotion of Science, and for Scientific Research on Priority Areas (12218226) from the Ministry of Education, Culture, Sports, Science and Technology, Japan.

      References

        • Reichling J.J
        • Kaplan M.M
        Clinical use of serum enzymes in liver disease.
        Dig Dis Sci. 1988; 33: 1601-1614
        • Sherman K.E
        Alanine aminotransferase in clinical practice. A review.
        Arch Intern Med. 1991; 151: 260-265
        • Byron D
        • Minuk G.Y
        Clinical hepatology: profile of an urban, hospital-based practice.
        Hepatology. 1996; 24: 813-815
        • Salvaggio A
        • Periti M
        • Miano L
        • Tavanelli M
        • Marzorati D
        Body mass index and liver enzyme activity in serum.
        Clin Chem. 1991; 37: 720-723
        • Lozano M
        • Cid J
        • Bedini J.L
        • Mazzara R
        • Gimenez N
        • Mas E
        • et al.
        Study of serum alanine-aminotransferase levels in blood donors in Spain.
        Haematologica. 1998; 83: 237-239
        • James O.F
        • Day C.P
        Non-alcoholic steatohepatitis (NASH): a disease of emerging identity and importance.
        J Hepatol. 1998; 29: 495-501
        • James O
        • Day C
        Non-alcoholic steatohepatitis: another disease of affluence.
        Lancet. 1999; 353: 1634-1636
        • Salmela P.I
        • Sotaniemi E.A
        • Niemi M
        • Maentausta O
        Liver function tests in diabetic patients.
        Diabetes Care. 1984; 7: 248-254
        • Nagore N
        • Scheuer P.J
        The pathology of diabetic hepatitis.
        J Pathol. 1988; 156: 155-160
        • Silverman J.F
        • O'Brien K.F
        • Long S
        • Leggett N
        • Khazanie P.G
        • Pories W.J
        • et al.
        Liver pathology in morbidly obese patients with and without diabetes.
        Am J Gastroenterol. 1990; 85: 1349-1355
        • DeFronzo R.A
        • Bonadonna R.C
        • Ferrannini E
        Pathogenesis of NIDDM. A balanced overview.
        Diabetes Care. 1992; 15: 318-368
        • Yki-Järvinen H
        Pathogenesis of non-insulin-dependent diabetes mellitus.
        Lancet. 1994; 343: 91-95
        • Meltzer A.A
        • Everhart J.E
        Association between diabetes and elevated serum alanine aminotransferase activity among Mexican Americans.
        Am J Epidemiol. 1997; 146: 565-571
        • Erbey J.R
        • Silberman C
        • Lydick E
        Prevalence of abnormal serum alanine aminotransferase levels in obese patients and patients with type 2 diabetes.
        Am J Med. 2000; 109: 588-590
        • Ikai E
        • Ishizaki M
        • Suzuki Y
        • Ishida M
        • Noborizaka Y
        • Yamada Y
        Association between hepatic steatosis, insulin resistance and hyperinsulinemia as related to hypertension in alcohol consumers and obese people.
        J Hum Hypertens. 1995; 9: 101-105
        • Perry I.J
        • Wannamethee S.G
        • Shaper A.G
        Prospective study of serum gamma-glutamyltransferase and risk of NIDDM.
        Diabetes Care. 1998; 21: 732-737
        • Shinchi K
        • Kono S
        • Honjo S
        • Todoroki I
        • Sakurai Y
        • Imanishi K
        • et al.
        Obesity and adenomatous polyps of the sigmoid colon.
        Jpn J Cancer Res. 1994; 85: 479-484
        • Kono S
        • Shinchi K
        • Todoroki I
        • Honjo S
        • Sakurai Y
        • Wakabayashi K
        • et al.
        Gallstone disease among Japanese men in relation to obesity, glucose intolerance, exercise, alcohol use, and smoking.
        Scand J Gastroenterol. 1995; 30: 372-376
        • Sasazuki S
        • Kono S
        • Todoroki I
        • Honjo S
        • Sakurai Y
        • Wakabayashi K
        • et al.
        Impaired glucose tolerance, diabetes mellitus, and gallstone disease: an extended study of male self-defense officials in Japan.
        Eur J Epidemiol. 1999; 15: 245-251
        • Honjo S
        • Kono S
        • Coleman M.P
        • Shinchi K
        • Sakurai Y
        • Todoroki I
        • et al.
        Coffee consumption and serum aminotransferases in middle-aged Japanese men.
        J Clin Epidemiol. 2001; 54: 823-829
        • Alberti K.G.M.M
        • Zimmet P.Z
        • for the WHO Consultation
        Definition, diagnosis and classification of diabetes mellitus and its complication. Part 1: diagnosis and classification of diabetes mellitus. Provisional report of a WHO consultation.
        Diabetes Med. 1998; 15: 539-553
        • Karmen A
        • Wróblewski F
        • LaDue J.S
        Transaminase activity in human blood.
        J Clin Invest. 1955; 35: 126-133
        • Szasz G
        A kinetic photometric method for serum-glutamyl transpeptidase.
        Clin Chem. 1969; 15: 124-136
        • Thulstrup A.M
        • Norgard B
        • Steffensen F.H
        • Vilstrup H
        • Sorensen H.T
        • Lauritzen T
        Waist circumference and body mass index as predictors of elevated alanine transaminase in Danes aged 30 to 50 years.
        Dan Med Bull. 1999; 46: 429-431
        • Holbrook T.L
        • Barrett-Connor E
        • Wingard D.L
        A prospective population-based study of alcohol use and non-insulin-dependent diabetes mellitus.
        Am J Epidemiol. 1990; 132: 902-909
        • Kao W.H.L
        • Puddey I.B
        • Boland L.B
        • Watson R.L
        • Brancati F.L
        Alcohol consumption and the risk of type 2 diabetes mellitus. Atherosclerosis Risk in Communities Study.
        Am J Epidemiol. 2001; 154: 748-758
        • Will J.C
        • Galuska D.A
        • Ford E.S
        • Mokdad A
        • Calle E.E
        Cigarette smoking and diabetes mellitus: evidence of a positive association from a large prospective cohort study.
        Int J Epidemiol. 2001; 30: 540-546
        • Steffensen F.H
        • Sorensen H.T
        • Brock A
        • Vilstrup H
        • Lauritzen T
        Alcohol consumption and serum liver-derived enzymes in a Danish population aged 30–50 years.
        Int J Epidemiol. 1997; 26: 92-99
        • Whitehead T.P
        • Robinson D
        • Allaway S.L
        The effects of cigarette smoking and alcohol consumption on serum liver enzyme activities: a dose-related study in men.
        Ann Clin Biochem. 1996; 33: 530-535
        • Sherlock S
        • Dooley J
        Diseases of the liver and biliary system. 9th ed. Blackwell Scientific, Oxford1993
      1. World Health Organization Expert Committee on Diabetes Mellitus. World Health Organization Tech Rep Ser 1980:646.

        • Sekikawa A
        • Sugiyama K
        • Tominaga M
        • Manaka H
        • Takahashi K
        • Sasaki H
        • et al.
        Prevalence of diabetes and impaired glucose tolerance in Funagata area, Japan.
        Diabetes Care. 1993; 16: 570-574
        • Ohmura T
        • Ueda K
        • Kiyohara Y
        • Kato I
        • Iwamoto H
        • Nakayama K
        • et al.
        Prevalence of type 2 (non-insulin-dependent) diabetes mellitus and impaired glucose tolerance in the Japanese general population: the Hisayama study.
        Diabetologia. 1993; 36: 1198-1203
        • Tankurt E
        • Biberoglu S
        • Ellidokuz E
        • Hekimsoy Z
        • Akpinar H
        • Comlekci A
        • et al.
        Hyperinsulinemia and insulin resistance in non-alcoholic steatohepatitis.
        J Hepatol. 1999; 31: 963
        • Marchesini G
        • Brizi M
        • Morselli-Labate A.M
        • Bianchi G
        • Bugianesi E
        • McCullough A.J
        • et al.
        Association of nonalcoholic fatty liver disease with insulin resistance.
        Am J Med. 1999; 107: 450-455
        • Cömert B
        • Mas M.R
        • Erdem H
        • Dinc A
        • Saglamkaya U
        • Cigerim M
        • et al.
        Insulin resistance in non-alcoholic steatohepatitis.
        Dig Liver Dis. 2001; 33: 353-358
        • Marchesini G
        • Brizi M
        • Bianchi G
        • Tomassetti S
        • Bugianesi E
        • Lenzi M
        • et al.
        Nonalcoholic fatty liver disease: a feature of the metabolic syndrome.
        Diabetes. 2001; 50: 1844-1850
        • Dixon J.B
        • Bhathal P.S
        • O'Brien P.E
        Nonalcoholic fatty liver disease: predictors of nonalcoholic steatohepatitis and liver fibrosis in the severely obese.
        Gastroenterology. 2001; 121: 91-100
        • Willner I.R
        • Waters B
        • Patil S.R
        • Reuben A
        • Morelli J
        • Riely C.A
        Ninety patients with nonalcoholic steatohepatitis: insulin resistance, familial tendency, and severity of disease.
        Am J Gastroenterol. 2001; 96: 2957-2961
        • Pagano G
        • Pacini G
        • Musso G
        • Gambino R
        • Mecca F
        • Depetris N
        • et al.
        Nonalcoholic steatohepatitis, insulin resistance, and metabolic syndrome: further evidence for an etiologic association.
        Hepatology. 2002; 35: 367-372
        • Chitturi S
        • Abeygunasekera S
        • Farrell G.C
        • Holmes-Walker J
        • Hui J.M
        • Fung C
        • et al.
        NASH and insulin resistance: insulin hypersecretion and specific association with the insulin resistance syndrome.
        Hepatology. 2002; 35: 373-379