Abstract
Aim: Whether the association between serum γ-glutamyltransferase (γ-GTP) levels and total cardiovascular disease (CVD) mortality is independent of alcohol drinking in East Asian populations is not well known. We conducted a pooled analysis of Japanese men and women that enabled an analysis restricted to never-drinkers.
Methods: A total of 15,987 men and 25,053 women aged 40–79 years, pooled from seven cohort studies throughout Japan, were followed-up to examine sex-specific relationship between serum γ-GTP levels and total CVD mortality. Cox regression model was used that was adjusted for age, smoking status, body mass index, and systolic blood pressure and serum triglyceride, total cholesterol, aspartate aminotransferase, and alanine aminotransferase levels.
Results: During an average follow-up of 8.7 years, we documented 361 and 340 deaths from total CVD, 146 and 168 from stroke, and 101 and 53 from coronary heart disease (CHD) for men and women, respectively. Among the never-drinkers, hazard ratios (HRs) for mortality for one standard deviation of log-γ-GTP for men were 1.89 (1.00–3.58) for stroke, 1.04 (0.57–1.90) for CHD, and 1.43 (1.04–1.96) for total CVD. For women, HRs were 1.28 (1.06–1.54), 1.81 (1.34–2.44), and 1.30 (1.14–1.49), respectively.
Conclusion: γ-GTP may be a risk factor for total CVD mortality independent of alcohol drinking status in Japanese men and women.
Keywords: γ-glutamyltransferase, Cardiovascular disease, Stroke, Coronary heart disease, Epidemiology
See editorial vol. 23: 769–770
Introduction
Serum γ-glutamyltransferase (γ-GTP) is commonly used as a diagnostic indicator for liver dysfunction, which is often a consequence of long-term alcohol drinking1). It has also been considered as a potential marker for oxidative stress and inflammation2, 3). A previous meta-analysis showed that serum γ-GTP levels were positively associated with cardiovascular disease (CVD) risks in both men and women even within the normal range4). However, the meta-analysis reported the existence of significant heterogeneity because of Asian studies. So far, only three Asian studies have examined the associations between γ-GTP levels and CVD, including two Japanese studies that found no association in men5, 6) and one Korean study that found an inverse association in the sample combining men and women7). Asians differ from Caucasians in terms of alcohol drinking habits and innate alcohol metabolism, i.e., aldehyde dehydrogenase polymorphism. Furthermore, stroke incidence is higher than coronary heart disease (CHD) in Asians. In addition, the meta-analysis found statistically significant association in nondrinkers only for CHD but not for stroke. Therefore, further studies were warranted in Asian population and for stroke in relation to γ-GTP levels. In the present study, we utilized the merit of large-scale pooled database of major cohort studies in Japan to analyze the association of serum γ-GTP levels with mortalities from CHD and stroke as well as the association in never-drinking men and women.
Methods
Study Participants
The Evidence for Cardiovascular Prevention from Observational Cohorts in Japan is a pooled project incorporating a meta-analysis of individual participant data from 13 well-qualified Japanese cohorts. The project was designed to examine the relationship between health examination measures (laboratory measures and lifestyle factors) and cause-specific mortality in Japanese populations. Each cohort was followed-up for approximately 10 years and included 1,000 or more participants. The details of this project have been described previously8–14). Serum γ-GTP levels at baseline were available in seven cohorts (N = 54,467): the Ohsaki cohort, the Ohasama study, the YKK factory workers cohort, the Radiation Effects Research Foundation cohort, the Hisayama study, the JACC study, and NIPPON DATA 90. We excluded those with histories of CVD at baseline (N = 5,160), those who were < 40 years or > 80 years of age (N = 7,322), and individuals who had high (> 50 IU/L) aspartate aminotransferase (AST) levels (N = 1,349) or alanine aminotransferase (ALT) levels (N = 2,358) in an attempt to exclude possible confounding of apparent liver disease. These exclusions left 41,040 participants comprising 15,987 men and 25,053 women for the present analyses.
Exposure and CVD Outcomes
Serum γ-GTP concentration was determined using a colorimetric assay; ALT and AST levels were measured using the ultraviolet method. In each cohort, mortality ascertainment was systematically conducted by reviewing death certificates. The underlying cause of death was based on either ICD-9 or ICD-10. Classification codes used in the study were as follows: death from stroke (430–438; I60–I69), CHD (410–414; I20–I25), and CVD (390–459; I00–I99). The present study was approved by the Ethical Review Committee of Keio University and Shiga University of Medical Science.
Statistical Analysis
Men and women were separately analyzed in this study. Cox proportional hazards models stratified by cohort15) were performed to estimate hazard ratios (HRs) and their corresponding 95% confidence intervals (95% CI) for CVD outcomes according to baseline serum γ-GTP levels. In the stratified Cox model, individual participant data from all studies are pooled, whereas the model accounts for the clustering of participants within studies. Because cohorts are almost identical to the areas, except two nation-wide cohorts (JACC and NIPPON DATA 90) for which area information had not been incorporated into the present pooled dataset, we did not additionally adjust for area. There were up to 10% of missing values in all the continuous variables of total individuals except for age. Instead of excluding those with missing values from the multivariable analysis, we had included them by using dummy variables that had a value missing. We first adjusted for continuous age (age-adjusted model) and subsequently adjusted for smoking status (never, past, 1–20/day, and ≥ 21/day) and sex-specific quartiles of body mass index (BMI) (kg/m2), systolic blood pressure (mmHg), serum triglyceride levels (mg/dL), serum total cholesterol levels (mg/dL), and AST (IU/L) and ALT (IU/L) levels (multivariable-adjusted model). The analyses were performed among never-drinkers first, then among the whole subjects adjusting for drinking status (never, quit, and regular) in the multivariable model. Statistical analyses were performed using SAS 9.2 for Windows (SAS Inc, Cary, NC, USA), and two sided p < 0.05 was considered to be statistically significant.
Results
Table 1 shows the baseline characteristics according to γ-GTP quartiles. Age, proportion of regular drinker and current smoker, BMI, systolic blood pressure, serum triglycerides levels, serum total cholesterol levels, AST, and ALT were positively associated with γ-GTP quartiles, whereas age was inversely associated with γ-GTP levels in men. Regular drinkers comprised 72% of men and 29% of women.
Table 1. Sex-specific means and proportions of cardiovascular risk factors according to quartiles of γ-GTP at baseline.
| Risk factors | Quartiles of γ-GTP (IU/L) |
||||
|---|---|---|---|---|---|
| Q1 (low) | Q2 | Q3 | Q4 (high) | ||
| Men | Quartile range (IU/L) | 1–16 | 17–24 | 25–40 | 41–837 |
| No. of participants | 4123 | 4074 | 3841 | 3949 | |
| Age (year) | 59.6 (10.8) | 58.6 (10.6) | 57.7 (10.3) | 55.7 (9.7) | |
| Never drinker, n (%) | 1512 (36.7) | 992 (24.4) | 641 (16.7) | 285 (7.2) | |
| Quit drinker, n (%) | 323 (7.8) | 256 (6.3) | 179 (4.7) | 114 (2.9) | |
| Regular drinker, n (%) | 2196 (53.3) | 2729 (67.0) | 2955 (76.9) | 3483 (88.2) | |
| Never smoker, n (%) | 1035 (25.1) | 918 (22.5) | 772 (20.1) | 644 (16.3) | |
| Former smoker, n (%) | 1038 (25.2) | 1029 (25.3) | 985 (25.6) | 892 (22.6) | |
| 1–20 cigarettes a day, n (%) | 1378 (33.4) | 1324 (32.5) | 1183 (30.8) | 1333 (33.8) | |
| ≤ 21 cigarettes a day, n (%) | 438 (10.6) | 488 (12.0) | 552 (14.4) | 736 (18.6) | |
| Body mass index (kg/m2) | 22.0 (2.6) | 22.8 (2.8) | 23.4 (2.9) | 23.9 (2.8) | |
| Systolic blood pressure (mmHg) | 129.0 (18.8) | 130.2 (18.5) | 133.2 (18.7) | 136.1 (18.5) | |
| Serum triglycerides (mg/dL) | 102.2 (55.0) | 120.2 (73.1) | 137.8 (87.3) | 172.6 (122.0) | |
| Serum total cholesterol (mg/dL) | 187.5 (32.6) | 194.8 (32.7) | 199.0 (34.3) | 201.5 (37.7) | |
| Aspartate aminotransferase (IU/L) | 21.5 (5.9) | 22.6 (6.0) | 24.0 (6.3) | 26.9 (7.2) | |
| Alanine aminotransferase (IU/L) | 17.2 (7.2) | 19.6 (7.8) | 22.5 (9.0) | 26.3 (9.5) | |
| Women | Quartile range (IU/L) | 1–9 | 10–13 | 14–18 | 19–435 |
| No. of participants | 5466 | 7785 | 5447 | 6355 | |
| Age (year) | 55.4 (10.4) | 57.3 (10.0) | 58.8 (9.6) | 58.9 (9.1) | |
| Never drinker, n (%) | 4462 (81.6) | 5799 (74.5) | 3904 (71.7) | 4262 (67.1) | |
| Quit drinker, n (%) | 42 (0.8) | 107 (0.4) | 74 (0.3) | 107 (0.4) | |
| Regular drinker, n (%) | 1004 (18.4) | 1986 (25.5) | 1543 (23.3) | 2093 (32.9) | |
| Never smoker, n (%) | 4861 (88.9) | 6196 (79.6) | 4198 (77.1) | 4714 (74.2) | |
| Former smoker, n (%) | 53 (1.0) | 104 (1.3) | 76 (1.4) | 94 (1.5) | |
| 1–20 cigarettes a day, n (%) | 139 (2.5) | 236 (3.0) | 187 (3.4) | 327 (5.2) | |
| ≥ 21 cigarettes a day, n (%) | 9 (0.2) | 15 (0.2) | 17 (0.3) | 42 (0.7) | |
| Body mass index (kg/m2) | 22.4 (2.8) | 23.0 (3.0) | 23.7 (3.3) | 24.4 (3.4) | |
| Systolic blood pressure (mmHg) | 127.6 (19.1) | 128.1 (18.8) | 131.2 (19.7) | 133.0 (19.4) | |
| Serum triglycerides (mg/dL) | 95.6 (51.8) | 106.7 (60.5) | 123.9 (72.6) | 141.2 (87.2) | |
| Serum total cholesterol (mg/dL) | 199.1 (34.9) | 207.2 (35.2) | 214.4 (35.4) | 218.5 (37.3) | |
| Aspartate aminotransferase (IU/L) | 18.9 (5.1) | 20.2 (5.2) | 21.4 (5.5) | 23.7 (6.8) | |
| Alanine aminotransferase (IU/L) | 13.9 (5.7) | 15.1 (6.0) | 17.3 (6.9) | 21.8 (8.9) | |
During an average follow-up of 8.7 years, we documented 361 and 340 deaths from total CVD, 146 and 168 from stroke, and 101 and 53 from CHD for men and women respectively. Among never-drinkers, the respective numbers were 82 and 252 from total CVD, 25 and 126 from stroke, and 31 and 38 from CHD, respectively. The one SD of log-γ-GTP was 0.70 for men and 0.56 for women. Serum γ-GTP level was seemingly positively associated with stroke mortality in never-drinking men. The multivariable HR (95% CI) for one SD increase of log-γ-GTP was 1.89 (1.00–3.58) (Table 2). Although the significant HRs were observed in Q2 and Q3 but not in Q4 compared with Q1 of γ-GTP quartile for total CVD mortality, the HR for one SD increase of log-γ-GTP was statistically significant (multivariable HR: 1.17, 95% CI: 1.03–1.33). However, it was not related to CHD mortality (multivariable HR: 1.04, 95% CI: 0.57–1.90) in never-drinking men. In never-drinking women, one SD increase of log-γ-GTP was significantly positively associated with mortalities from stroke, CHD, and total CVD. Furthermore, never-drinking women in the quartile Q4 had more than four-time higher CHD mortality risk compared with Q1 (multivariable HR: 4.49, 95% CI: 1.41–14.32).
Table 2. Sex-specific, age- and multivariable-adjusted hazard ratios and 95% confidence intervals for mortality from cardiovascular disease according to quartiles of γ-GTP and one SD increment of log γ-GTP in never-drinkers.
| Quartiles of γ-GTP |
HR1† | ||||
|---|---|---|---|---|---|
| Q1 (low) | Q2 | Q3 | Q4 (high) | ||
| Men | |||||
| Quartile range (IU/L) | 1–16 | 17–24 | 25–40 | 41–837 | |
| No. at risk | 1,512 | 992 | 641 | 285 | |
| Person-years | 13,371 | 8,599 | 5,575 | 2,543 | |
| Stroke | |||||
| No. of mortality | 11 | 8 | 4 | 2 | |
| Mortality rate | 0.82 | 0.93 | 0.72 | 0.79 | |
| Age adjusted HR | 1.00 | 1.50 (0.59–3.83) | 1.35 (0.41–4.52) | 2.14 (0.44–10.28) | 1.46 (0.86–2.46) |
| Multivariable HR§ | 1.00 | 1.59 (0.56–4.50) | 1.56 (0.41–5.98) | 4.14 (0.72–23.91) | 1.89 (1.00–3.58) |
| Coronary heart disease | |||||
| No. of mortality | 9 | 12 | 8 | 2 | |
| Mortality rate | 0.67 | 1.40 | 1.44 | 0.79 | |
| Age adjusted HR | 1.00 | 2.19 (0.91–5.29) | 2.49 (0.92–6.72) | 1.67 (0.35–8.00) | 1.69 (1.31–2.19) |
| Multivariable HR§ | 1.00 | 2.02 (0.79–5.13) | 2.10 (0.70–6.27) | 1.69 (0.32–9.02) | 1.04 (0.57–1.90) |
| Total cardiovascular diseases | |||||
| No. of mortality | 30 | 30 | 21 | 5 | |
| Mortality rate | 2.24 | 3.49 | 3.77 | 1.97 | |
| Age adjusted HR | 1.00 | 1.74 (1.04–2.93) | 2.20 (1.22–3.95) | 1.44 (0.55–3.80) | 1.33 (1.00–1.77) |
| Multivariable HR§ | 1.00 | 1.90 (1.09–3.30) | 2.41 (1.27–4.57) | 1.78 (0.64–4.96) | 1.43 (1.04–1.96) |
| Women | |||||
| Quartile range (IU/L) | 1–9 | 10–13 | 14–18 | 19–435 | |
| No. at risk | 4,462 | 5,799 | 3,904 | 4,262 | |
| Person-years | 40,944 | 52,008 | 34,286 | 37,580 | |
| Stroke | |||||
| No. of mortality | 23 | 35 | 25 | 43 | |
| Mortality rate | 0.56 | 0.67 | 0.73 | 1.14 | |
| Age adjusted HR | 1.00 | 1.06 (0.61–1.81) | 0.99 (0.55–1.78) | 1.64 (0.96–2.79) | 1.32 (1.11–1.55) |
| Multivariable HR§ | 1.00 | 1.17 (0.67–2.06) | 1.09 (0.58–2.02) | 1.60 (0.87–2.92) | 1.28 (1.06–1.54) |
| Coronary heart disease | |||||
| No. of mortality | 5 | 8 | 9 | 16 | |
| Mortality rate | 0.12 | 0.15 | 0.26 | 0.43 | |
| Age adjusted HR | 1.00 | 1.52 (0.49–4.71) | 2.46 (0.80–7.56) | 4.26 (1.50–12.07) | 1.43 (0.89–2.29) |
| Multivariable HR§ | 1.00 | 1.53 (0.47–4.98) | 2.51 (0.76–8.25) | 4.49 (1.41–14.32) | 1.81 (1.34–2.44) |
| Total cardiovascular diseases | |||||
| No. of mortality | 47 | 68 | 53 | 84 | |
| Mortality rate | 1.15 | 1.31 | 1.55 | 2.24 | |
| Age adjusted HR | 1.00 | 1.04 (0.71–1.52) | 1.07 (0.71–1.62) | 1.66 (1.13–2.42) | 1.28 (1.14–1.45) |
| Multivariable HR§ | 1.00 | 1.11 (0.75–1.66) | 1.16 (0.75–1.79) | 1.77 (1.15–2.71) | 1.30 (1.14–1.49) |
Mortality rate is expressed as /1000 person-years.
HR1: HR for 1 SD of log γ-GTP.
Multivariable HR: adjusted for age (continuous), smoking status (never, former, 1–20/day and ≥ 21/day), body mass index (sex-specific quartile), systolic blood pressure (sex-specific quartiles), serum triglycerides levels (sex-specific quartiles), serum total cholesterol levels (sex-specific quartiles), aspartate aminotransferase (sex-specific quartiles) and alanine aminotransferase (sex-specific quartiles).
The associations were essentially similar in total participants in both men and women (Table 3).
Table 3. Sex-specific, age- a nd multivariable-adjusted hazard ratios and 95% confidence intervals for mortality from cardiovascular disease according to quartiles of γ-GTP and one SD increment of log γ-GTP in total participants.
| Quartiles of γ-GTP |
HR1† | ||||
|---|---|---|---|---|---|
| Q1 (low) | Q2 | Q3 | Q4 (high) | ||
| Men | |||||
| Quartile range (IU/L) | 1–16 | 17–24 | 25–40 | 41–837 | |
| No. at risk | 4,123 | 4,074 | 3,841 | 3,949 | |
| Person-years | 35,697 | 34,752 | 32,487 | 33,606 | |
| Stroke | |||||
| No. of mortality | 39 | 40 | 31 | 36 | |
| Mortality rate | 1.09 | 1.15 | 0.95 | 1.07 | |
| Age adjusted HR | 1.00 | 1.25 (0.80–1.96) | 1.24 (0.76–2.00) | 1.69 (1.06–2.71) | 1.31 (1.11–1.54) |
| Multivariable HR§ | 1.00 | 1.23 (0.77–1.94) | 1.23 (0.74–2.06) | 1.76 (1.02–3.03) | 1.39 (1.14–1.68) |
| Coronary heart disease | |||||
| No. of mortality | 28 | 29 | 24 | 20 | |
| Mortality rate | 0.78 | 0.83 | 0.74 | 0.60 | |
| Age adjusted HR | 1.00 | 1.19 (0.70–2.02) | 1.15 (0.66–2.00) | 1.08 (0.60–1.96) | 1.01 (0.82–1.25) |
| Multivariable HR§ | 1.00 | 1.25 (0.72–2.15) | 1.15 (0.63–2.11) | 1.03 (0.52–2.06) | 0.98 (0.76–1.26) |
| Total cardiovascular diseases | |||||
| No. of mortality | 97 | 105 | 85 | 74 | |
| Mortality rate | 2.72 | 3.02 | 2.62 | 2.20 | |
| Age adjusted HR | 1.00 | 1.27 (0.96–1.67) | 1.27 (0.95–1.71) | 1.31 (0.96–1.79) | 1.14 (1.02–1.27) |
| Multivariable HR§ | 1.00 | 1.32 (0.99–1.76) | 1.33 (0.96–1.82) | 1.39 (0.97–1.99) | 1.17 (1.03–1.33) |
| Women | |||||
| Quartile range (IU/L) | 1–9 | 10–13 | 14–18 | 19–435 | |
| No. at risk | 5,466 | 7,785 | 5,447 | 6,355 | |
| Person-years | 49,579 | 68,477 | 47,223 | 55,107 | |
| Stroke | |||||
| No. of mortality | 30 | 45 | 35 | 58 | |
| Mortality rate | 0.61 | 0.66 | 0.74 | 1.05 | |
| Age adjusted HR | 1.00 | 0.95 (0.59–1.52) | 0.97 (0.58–1.60) | 1.45 (0.91–2.32) | 1.30 (1.13–1.50) |
| Multivariable HR§ | 1.00 | 1.02 (0.62–1.67) | 1.03 (0.60–1.76) | 1.44 (0.85–2.44) | 1.28 (1.09–1.50) |
| Coronary heart disease | |||||
| No. of mortality | 7 | 9 | 14 | 23 | |
| Mortality rate | 0.14 | 0.13 | 0.30 | 0.42 | |
| Age adjusted HR | 1.00 | 1.04 (0.38–2.83) | 2.28 (0.90–5.82) | 3.50 (1.44–8.49) | 1.72 (1.39–2.12) |
| Multivariable HR§ | 1.00 | 0.97 (0.35–2.75) | 2.11 (0.79–5.68) | 3.33 (1.24–8.93) | 1.82 (1.41–2.34) |
| Total cardiovascular diseases | |||||
| No. of mortality | 59 | 93 | 69 | 119 | |
| Mortality rate | 1.19 | 1.36 | 1.46 | 2.16 | |
| Age adjusted HR | 1.00 | 0.99 (0.71–1.39) | 0.97 (0.68–1.40) | 1.52 (1.09–2.12) | 1.29 (1.17–1.43) |
| Multivariable HR§ | 1.00 | 1.04 (0.73–1.47) | 1.01 (0.69–1.48) | 1.58 (1.08–2.29) | 1.32 (1.17–1.47) |
Mortality rate is expressed as /1000 person-years.
HR1: HR for 1 SD of log γ-GTP.
Multivariable HR: adjusted for age (continuous), drinking status (never, quit, and regular), smoking status (never, former, 1–20/day and ≥ 21/day), body mass index (sex-specific quartile), systolic blood pressure (sex-specific quartiles), serum triglycerides levels (sex-specific quartiles), serum total cholesterol levels (sex-specific quartiles), aspartate aminotransferase (sex-specific quartiles) and alanine aminotransferase (sex-specific quartiles).
Discussion
In this study, we performed meta-analysis of individual participants data consisting of 41,040 Japanese with no history of CVD or overt liver dysfunction. We demonstrated that serum γ-GTP level was positively associated with stroke and total CVD mortality in men and women and CHD mortality in women. These associations did not alter when the analysis was restricted to never-drinker. Our finding extended previously reported positive association mainly in Caucasians4, 16), to East Asian, and to stroke mortality in both men and women. We have confirmed that the association was independent of alcohol drinking habit as well as systolic blood pressure and other potential confounding variables. We found stronger association of γ-GTP with CHD than with stroke in women but not in men where only the association with stroke was observed. The finding for CHD is not consistent with those of several previous studies that found associations in both sexes17) or two other studies including men independent of alcohol consumption18, 19). Although there is no clear explanation for this finding, one possibility could be inaccuracy of participants' self-report of drinking status, particularly in those with high γ-GTP. The fact that alcohol drinking has protective association with CHD might have distorted the current find ing20). However, there would not have been significant stigma of reporting of alcohol consumption as almost four-fifths of the men were regular drinkers at baseline; we did not consider the chance of intentional misreporting being high.
We have examined the association separately for men and women in the present study. The present finding on CVD mortality would be consistent with 24-year follow-up British Regional Heart Study including men that revealed a positive dose–response association for CVD mortality independent of alcohol consumption19). We have extended this finding to stroke mortality in Japanese men and women and also to CHD mortality in Japanese women. Nevertheless, studies exist that reported inconsistent findings for stroke. A Finnish study was consistent with ours in both men and women21). An Austrian study and British Women's Heart and Health Study did not find association in women4, 22). On the contrary, one Japanese study, which was not included in the present individual participant data meta-analysis, observed positive association only in w omen6). The reason for the discrepancies is not clear, but the present finding would be more reliable as it is an individual participant data meta-analysis intended to overcome interstudy variations in the findings.
There could be a few explanations for the present findings. Serum γ-GTP level is a maker of oxidative stress and inflammation, both of which can play a role in the pathophysiology of CVD2, 3). Moreover, γ-GTP may trigger oxidation of low-density lipoproteins and contribute to plaque formation, maturation, and rupture23, 24). Another possible explanation would be the increased γ-GTP as an indicator of nonalcoholic fatty liver disease (NAFLD)25). It was reported that elevation of big fraction of γ-GTP was specific to NAFLD26). Although we have excluded individuals with high ALT (> 50 IU/L) blood level at baseline27) and further adjusted for continuous ALT in the statistical model, it remained to be elucidated whether and how much the present finding was attributed to NAFLD. Further studies are needed to explore pathophysiological mechanisms under the present findings and to evaluate the usefulness of serum γ-GTP levels for risk prediction.
A strength of the present study is that we pooled data from several Japanese cohort studies with long follow-up period (over 10 years). This enabled us to utilize a large number of never-drinking participants to eliminate the confounding of alcohol drinking status on the association between serum γ-GTP levels and CVD mortality, particularly pertinent for East-Asian populations where never-drinking men were scarce. Additionally, the large sample size of current study allows us to exclude individuals with elevated serum AST or ALT in an attempt to eliminate possible confounding of γ-GTP elevation from apparent liver diseases such as long-term medication use and hepatitis C virus infection28), of which the prevalence is high in Japan.
Some limitations of our study merit consideration. First, the participants in most of the cohort studies volunteered to undertake the community health examination, and for that reason, their characteristics might be different to those of unwilling non-participants or the general population. This would influence the absolute effect measure (mortality rate) and might underestimate the risk. However, these differences would have little effect on relative effect measures, such as HR. Second, several potential confounding factors were unavailable in the current study such as treatment of hypertension and diabetes mellitus, high-density lipoprotein cholesterol, and physical activity; however, in the previous studies done by others, adjustments for these factors did not significantly alter the results5, 17, 22). Third, the HRs for the higher quartiles of γ-GTP in the total participant might be overestimated, because we did not collect the information on the alcohol consumption, which are assumed to be correlated with γ-GTP levels. However, the analysis restricted to never-drinkers remained unaltered suggesting that the confounding effect might be weak. Finally, since changes in serum γ-GTP levels could occur over time, a single measurement of γ-GTP levels at baseline may have led to nondifferential misclassification of γ-GTP categories among participants. Thus, real associations may have been stronger.
Conclusion
In conclusion, serum γ-GTP levels were positively associated with CVD and stroke mortality independently of alcohol drinking status in both men and women and with CHD mortality in women in the present individual participant data meta-analysis of Japanese cohorts. Whether measurement of serum γ-GTP levels would help predict future mortality from CVD or understanding pathophysiological mechanisms for the association requires further investigations.
Acknowledgements
We are grateful to all of the participants in each cohort study. We thank Mrs. Toshimi Yoshida (Shiga University of Medical Science) and Mrs. Satoko Narikawa (Keio University) for expert clerical assistance.
Conflicts of Interest
The authors declare no conflicts of interest.
Appendix
The Evidence for Cardiovascular Prevention from Observational Cohorts in Japan (EPOCH– JAPAN) Research Group is composed of the following investigators. Chairperson: Hirotsugu Ueshima (Shiga University of Medical Science); Co–Chairperson: Tomonori Okamura (Keio University); Executive committee: Hirotsugu Ueshima (Shiga University of Medical Science), Yutaka Imai (Tohoku University Graduate School of Pharmaceutical Sciences), Takayoshi Ohkubo (Teikyo University School of Medicine), Fujiko Irie (Ibaraki Prefecture), Hiroyasu Iso, Akihiko Kitamura (Osaka University Graduate School of Medicine), Yutaka Kiyohara (Kyushu University Graduate School of Medicine), Katsuyuki Miura (Shiga University of Medical Science), Yoshitaka Murakami (Toho University), Hideaki Nakagawa (Kanazawa Medical University), Takeo Nakayama (Kyoto University School of Public Health), Akira Okayama (Research Institute of Strategy for Prevention), Toshimi Sairenchi (Dokkyo Medical University), Shigeyuki Saitoh (Sapporo Medical University), Kiyomi Sakata (Iwate Medical University), Akiko Tamakoshi (Hokkaido University Graduate School of Medicine), Ichiro Tsuji (Tohoku University Graduate School of Medicine), Michiko Yamada (Radiation Effects Research Foundation), Masahiko Kiyama (Osaka Center for Cancer and Cardiovascular Disease Prevention), Yoshihiro Miyamoto (National Cerebral and Cardiovascular Center), Shizukiyo Ishikawa (Jichi Medical University), Hiroshi Yatsuya (Fujita Health University) and Tomonori Okamura (Keio University School of Medicine)
Funding Sources
This research was supported by a grant–in–aid from the Ministry of Health, Labour and Welfare, Health and Labor Sciences research grants, Japan (Research on Health Services: H17–Kenkou–007; Comprehensive Research on Cardiovascular Disease and Life–Related Disease: H18–Junkankitou [Seishuu]– Ippan–012; Comprehensive Research on Cardiovascular Disease and Life–Related Disease: H19– Junkankitou [Seishuu]–Ippan–012; Comprehensive Research on Cardiovascular and Life–Style Related Diseases: H20–Junkankitou [Seishuu]–Ippan–013; Comprehensive Research on Cardiovascular and Life– Style Related Diseases: H23–Junkankitou [Seishuu]– Ippan–005), and an Intramural Research Fund (22-4-5) for Cardiovascular Diseases of National Cerebral and Cardiovascular Center; and Comprehensive Research on Cardiovascular and Life-Style Related Diseases: H26-Junkankitou [Seisaku]-Ippan-001.
References
- 1). Whitfield JB: Gamma glutamyl transferase. Crit Rev Clin Lab Sci, 2001; 38: 263-355 [DOI] [PubMed] [Google Scholar]
- 2). Lee DH, Blomhoff R, Jacobs DR, Jr.: Is serum gamma glutamyltransferase a marker of oxidative stress? Free Radic Res, 2004; 38: 535-539 [DOI] [PubMed] [Google Scholar]
- 3). Yamada J, Tomiyama H, Yambe M, Koji Y, Motobe K, Shiina K, Yamamoto Y, Yamashina A: Elevated serum levels of alanine aminotransferase and gamma glutamyltransferase are markers of inflammation and oxidative stress independent of the metabolic syndrome. Atherosclerosis, 2006; 189: 198-205 [DOI] [PubMed] [Google Scholar]
- 4). Fraser A, Harris R, Sattar N, Ebrahim S, Smith GD, Lawlor DA: Gamma-glutamyltransferase is associated with incident vascular events independently of alcohol intake: analysis of the British Women's Heart and Health Study and Meta-Analysis. Arterioscler Thromb Vasc Biol, 2007; 27: 2729-2735 [DOI] [PubMed] [Google Scholar]
- 5). Hozawa A, Okamura T, Kadowaki T, Murakami Y, Nakamura K, Hayakawa T, Kita Y, Nakamura Y, Okayama A, Ueshima H: gamma-Glutamyltransferase predicts cardiovascular death among Japanese women. Atherosclerosis, 2007; 194: 498-504 [DOI] [PubMed] [Google Scholar]
- 6). Shimizu Y, Imano H, Ohira T, Kitamura A, Kiyama M, Okada T, Sato S, Shimamoto T, Yamagishi K, Tanigawa T, Iso H: gamma-Glutamyltranspeptidase and incident stroke among Japanese men and women: the Circulatory Risk in Communities Study (CIRCS). Stroke, 2010; 41: 385-388 [DOI] [PubMed] [Google Scholar]
- 7). Ebrahim S, Sung J, Song YM, Ferrer RL, Lawlor DA, Davey Smith G: Serum cholesterol, haemorrhagic stroke, ischaemic stroke, and myocardial infarction: Korean national health system prospective cohort study. BMJ, 2006; 333: 22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8). Murakami Y, Hozawa A, Okamura T, Ueshima H: Relation of blood pressure and all-cause mortality in 180,000 Japanese participants: pooled analysis of 13 cohort studies. Hypertension, 2008; 51: 1483-1491 [DOI] [PubMed] [Google Scholar]
- 9). Murakami Y, Miura K, Okamura T, Ueshima H: Population attributable numbers and fractions of deaths due to smoking: a pooled analysis of 180,000 Japanese. Prev Med, 2011; 52: 60-65 [DOI] [PubMed] [Google Scholar]
- 10). Nakamura K, Nakagawa H, Sakurai M, Murakami Y, Irie F, Fujiyoshi A, Okamura T, Miura K, Ueshima H, Group E-JR : Influence of smoking combined with another risk factor on the risk of mortality from coronary heart disease and stroke: pooled analysis of 10 Japanese cohort studies. Cerebrovasc Dis, 2012; 33: 480-491 [DOI] [PubMed] [Google Scholar]
- 11). Fujiyoshi A, Ohkubo T, Miura K, Murakami Y, Nagasawa SY, Okamura T, Ueshima H, Observational Cohorts in Japan Research G : Blood pressure categories and longterm risk of cardiovascular disease according to age group in Japanese men and women. Hypertens Res, 2012; 35: 947-953 [DOI] [PubMed] [Google Scholar]
- 12). Nagasawa SY, Okamura T, Iso H, Tamakoshi A, Yamada M, Watanabe M, Murakami Y, Miura K, Ueshima H, Evidence for Cardiovascular Prevention from Observational Cohorts in Japan Research G : Relation between serum total cholesterol level and cardiovascular disease stratified by sex and age group: a pooled analysis of 65 594 individuals from 10 cohort studies in Japan. J Am Heart Assoc, 2012; 1: e001974. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13). Nagata M, Ninomiya T, Kiyohara Y, Murakami Y, Irie F, Sairenchi T, Miura K, Okamura T, Ueshima H, Group E-JR : Prediction of cardiovascular disease mortality by proteinuria and reduced kidney function: pooled analysis of 39,000 individuals from 7 cohort studies in Japan. Am J Epidemiol, 2013; 178: 1-11 [DOI] [PubMed] [Google Scholar]
- 14). Asayama K, Satoh M, Murakami Y, Ohkubo T, Nagasawa SY, Tsuji I, Nakayama T, Okayama A, Miura K, Imai Y, Ueshima H, Okamura T, on behalf of the Evidence for Cardiovascular Prevention From Observational Cohorts in Japan Research G : Cardiovascular Risk With and Without Antihypertensive Drug Treatment in the Japanese General Population: Participant-Level Meta-Analysis. Hypertension, 2014; 63: 1189-1197 [DOI] [PubMed] [Google Scholar]
- 15). Murakami Y: Meta-analyses using individual particiipant data from cardiovascular cohort studies in Japan: current status and future directions. J Epidemiol, 2014; 24: 96-101 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16). Kunutsor SK, Apekey TA, Khan H: Liver enzymes and risk of cardiovascular disease in the general population: a meta-analysis of prospective cohort studies. Atherosclerosis, 2014; 236: 7-17 [DOI] [PubMed] [Google Scholar]
- 17). Lee DH, Silventoinen K, Hu G, Jacobs DR, Jr., Jousilahti P, Sundvall J, Tuomilehto J: Serum gamma-glutamyltransferase predicts non-fatal myocardial infarction and fatal coronary heart disease among 28,838 middle-aged men and women. Eur Heart J, 2006; 27: 2170-2176 [DOI] [PubMed] [Google Scholar]
- 18). Meisinger C, Doring A, Schneider A, Lowel H: Serum gamma-glutamyltransferase is a predictor of incident coronary events in apparently healthy men from the general population. Atherosclerosis, 2006; 189: 297-302 [DOI] [PubMed] [Google Scholar]
- 19). Wannamethee SG, Lennon L, Shaper AG: The value of gamma-glutamyltransferase in cardiovascular risk prediction in men without diagnosed cardiovascular disease or diabetes. Atherosclerosis, 2008; 201: 168-175 [DOI] [PubMed] [Google Scholar]
- 20). Ikehara S, Iso H, Yamagishi K, Yamamoto S, Inoue M, Tsugane S, Group JS: Alcohol consumption, social support, and risk of stroke and coronary heart disease among Japanese men: the JPHC Study. Alcohol Clin Exp Res, 2009; 33: 1025-1032 [DOI] [PubMed] [Google Scholar]
- 21). Jousilahti P, Rastenyte D, Tuomilehto J: Serum gammaglutamyl transferase, self-reported alcohol drinking, and the risk of stroke. Stroke, 2000; 31: 1851-1855 [DOI] [PubMed] [Google Scholar]
- 22). Ruttmann E, Brant LJ, Concin H, Diem G, Rapp K, Ulmer H: Gamma-glutamyltransferase as a risk factor for cardiovascular disease mortality: an epidemiological investigation in a cohort of 163,944 Austrian adults. Circulation, 2005; 112: 2130-2137 [DOI] [PubMed] [Google Scholar]
- 23). Emdin M, Pompella A, Paolicchi A: Gamma-glutamyltransferase, atherosclerosis, and cardiovascular disease: triggering oxidative stress within the plaque. Circulation, 2005; 112: 2078-2080 [DOI] [PubMed] [Google Scholar]
- 24). Paolicchi A, Emdin M, Passino C, Lorenzini E, Titta F, Marchi S, Malvaldi G, Pompella A: Beta-lipoprotein- and LDL-associated serum gamma-glutamyltransferase in patients with coronary atherosclerosis. Atherosclerosis, 2006; 186: 80-85 [DOI] [PubMed] [Google Scholar]
- 25). Ghouri N, Preiss D, Sattar N: Liver enzymes, nonalcoholic fatty liver disease, and incident cardiovascular disease: a narrative review and clinical perspective of prospective data. Hepatology, 2010; 52: 1156-1161 [DOI] [PubMed] [Google Scholar]
- 26). Franzini M, Fornaciari I, Fierabracci V, Elawadi HA, Bolognesi V, Maltinti S, Ricchiuti A, De Bortoli N, Marchi S, Pompella A, Passino C, Emdin M, Paolicchi A: Accuracy of b-GGT fraction for the diagnosis of nonalcoholic fatty liver disease. Liver Int, 2012; 32: 629-634 [DOI] [PubMed] [Google Scholar]
- 27). Chang Y, Ryu S, Sung E, Jang Y: Higher concentrations of alanine aminotransferase within the reference interval predict nonalcoholic fatty liver disease. Clin Chem, 2007; 53: 686-692 [DOI] [PubMed] [Google Scholar]
- 28). Butt AA, Xiaoqiang W, Budoff M, Leaf D, Kuller LH, Justice AC: Hepatitis C virus infection and the risk of coronary disease. Clin Infect Dis, 2009; 49: 225-232 [DOI] [PMC free article] [PubMed] [Google Scholar]