Lung (2014) 192:719–727 DOI 10.1007/s00408-014-9616-3
Relationship of Serum Gamma-Glutamyltransferase Levels with Pulmonary Function and Chronic Obstructive Pulmonary Disease Hyun-Woo Kim • Seock-Hwan Lee Duk-Hee Lee
•
Received: 27 March 2014 / Accepted: 16 June 2014 / Published online: 11 July 2014 Ó Springer Science+Business Media New York 2014
Abstract Background Gamma-glutamyltransferase (GGT) levels within the normal reference range, possibly a biomarker of oxidative stress and/or exposure to various environmental chemicals, are associated with pulmonary function. However, it is unclear whether it is totally independent of cigarette smoking. Also, the potential interaction between serum GGT and cigarette smoking has not ever been evaluated. Therefore, this study investigated (1) whether serum GGT levels are associated with pulmonary function and chronic obstructive pulmonary disease (COPD), independent of cigarette smoking, and (2) whether there is any interaction between serum GGT and cigarette smoking status on pulmonary function. Methods The study subjects were 4,583 participants aged C40 in the 2010–2011 Korean National Health and Nutrition Examination Survey. The outcomes were pulmonary function (forced expiratory volume in 1 second [FEV1] and forced vital capacity [FVC]) and spirometrically defined COPD. Results After adjusting for potential confounders, including cigarette smoking, serum GGT levels were H.-W. Kim S.-H. Lee Department of Family Medicine, Daegu Medical Center, 157 Pyungli-ro, Seo-Gu, Daegu 703-713, Korea D.-H. Lee (&) Department of Preventative Medicine, School of Medicine, Kyungpook National University, 680 Gukchaebosang-ro, Jung-gu, Daegu 700-842, Korea e-mail: [email protected] D.-H. Lee BK21 Plus KNU Biomedical Convergence Program, Department of Biomedical Science, Kyungpook National University, 680 Gukchaebosang-ro, Jung-gu, Daegu 700-842, Korea
inversely associated with FEV1 and FVC in both genders and positively associated with the risk of COPD in men (all P values \0.01). In men, adjusted odds ratios of COPD were 1.0, 1.69, 1.97, and 2.02 across the quartiles of GGT level (Ptrend = 0.002). However, the associations between serum GGT and pulmonary function seemed to differ depending on the smoking status; inverse associations of GGT with FEV1 % and FVC % were clearly observed only among non-current smokers. Conclusions In conclusion, in non-smokers serum GGT levels can be used to detect individuals at high risk of decreased pulmonary function and/or COPD. Keywords Oxidative stress Airway limitation Smoking Air pollution
Introduction Over the last few decades, a number of epidemiological studies have demonstrated that gamma-glutamyltransferase (GGT) levels, within the normal reference range, are predictive of various clinical diseases, including type 2 diabetes, cardiovascular disease, chronic kidney disease, and cancer [1]. Abnormally elevated GGT levels have been used as a clinical marker of alcohol abuse or liver disease affecting the biliary system [2]; however, the association of GGT levels with other clinical diseases was independent of alcohol consumption or liver disease. At present, serum GGT within its reference range has been proposed as a biomarker of oxidative stress associated with glutathione (GSH) metabolism [3] and/or exposure to various environmental chemicals that conjugate to GSH [4]. Cellular GGT is expressed in most human tissues, including the lung, where it is involved in secretory and
123
720
absorptive processes [5, 6]. Although cellular GGT levels in the lung are relatively low compared to the levels detected in other organs, such as the liver or kidney [5, 6], several human studies have suggested an association between serum GGT levels and pulmonary function or chronic obstructive pulmonary disease (COPD) [7–10]. However, these findings are inconsistent. In particular, the effect of cigarette smoking, the most important risk factor in the pathogenesis of lung disease, is unclear. For example, in individuals who underwent routine medical checkups or policemen, the forced expiratory volume in 1 s (FEV1) and the forced vital capacity (FVC) decreased with increasing GGT levels, even after adjustment for smoking [8, 9]. In contrast, another study showed a statistically significant association between serum GGT and airway limitation in middle-aged men almost disappeared after adjusting for smoking [7]. Therefore, it is questionable whether the association between serum GGT and pulmonary function is totally independent of cigarette smoking. In addition, as the burden of oxidative stress is highly increased in smokers, there is a possibility of a synergic interaction between serum GGT and cigarette smoking on the risk of decreased pulmonary function. However, no study has evaluated this possibility. This study was performed to investigate [1] whether serum GGT levels are related to pulmonary function and COPD in a representative sample of the Korean general population, independent of cigarette smoking, and [2] whether serum GGT is more strongly related to pulmonary function and COPD in smokers compared to non-smokers.
Methods Study Participants and Data Collection Our study was based on data collected in 2010–2011 by the Fifth National Health and Nutrition Examination Survey (KNHANES V-1, 2). In KNHNES V, the Division of Chronic Disease Surveillance under the Korea Centers for Disease Control and Prevention (KCDC) collected nationally representative statistics from 2010 to 2012 on the health status, health-related behaviors, diet, and nutritional status of Korean people. Stratified, multistage probability sampling was used for the selection of household units. KNHANES consisted of a health interview survey, a nutrition survey, and a health examination survey. Data on demographic characteristics, diet, and healthrelated variables were collected through personal interviews and self-administered questionnaires. Physical examinations, blood sampling, and pulmonary function tests were carried out at a mobile examination center.
123
Lung (2014) 192:719–727
In total, 9,159 people aged 40 years and older people participated in KNHANES V-1, 2. Of these, spirometry results were available for 7,151 (78.1 %) participants. Out of these, we excluded patients without reproducible or acceptable spirometry data (1,195, 16.7 %), individuals without information on serum GGT or covariates (336 subjects), participants with a history of asthma or pulmonary tuberculosis that could cause airway limitation other than COPD (469 subjects), and those with abnormal levels of serum GGT (serum GGT[73 IU/L in men and[48 IU/L in women, 568 subjects) [11]. Thus, a total of 4,583 subjects were included in the final analyses. All the subjects in the study participated voluntarily under informed consent. The study protocol was approved by the institutional review board of the KCDC. Anthropometric and Biochemical Measurements The subjects’ weights and heights were measured, while they were wearing light clothing and had removed their shoes. Blood samples were collected from the antecubital vein of each participant after overnight fasting. The samples were processed, transported to the Central Testing Institute located in Seoul, Korea, and analyzed within 24 h. The analyses of GGT and alanine aminotransferase (ALT) were performed with a Hitachi Automatic Analyzer 7600 (Hitachi, Tokyo, Japan) by enzymatic or UV methods using commercially available kits (Sekisui, Tokyo, Japan/ Daiichi, Tokyo, Japan). Spirometry Data A pulmonary function test was conducted using dry rolling seal spirometers (Model 2130, Sensor Medics, Yorba Linda, CA, USA), after at least one practice blow, in the seated position with nose clips. The forced expiratory volume in 1 s (FEV1) and the forced vital capacity (FVC) were measured with effort in accordance with the criteria for acceptability, presented in the guidelines of the American Thoracic Society/ European Respiratory Society (ATS/ERS) [12], at least three times and at most eight times. Only subjects who had at least two proper curves of spirometry and a difference of B0.15 (L) between the two highest readings of FEV1 and FVC were included in the study (82.3 % of men and 77.7 % of women). A bronchodilator was not given before the spirometry. The percent predicted values of the following predictive equations, described by Choi et al. [13], were used to determine pulmonary function: In men: Predicted FVC ¼ 4:8434 0:00008633 ðageÞ2 þ 0:05292 ðheightÞ þ 0:01095 ðweightÞ
Lung (2014) 192:719–727
Predicted FEV1 ¼ 3:4132 0:0002484 ðageÞ2 þ 0:04578 ðheightÞ In women: Predicted FVC ¼ 3:0006 0:0001273 ðageÞ2 þ 0:03951 ðheightÞ þ 0:006892 ðweightÞ Predicted FEV1 ¼ 2:4114 0:0001920 ðageÞ2 þ 0:03558 ðheightÞ Subjects with a FEV1/FVC \70 % were identified as having COPD. Statistical Analysis As the serum GGT level, pulmonary function, and prevalence of COPD differed significantly between men and women, sex-specific analyses were performed. The serum GGT levels were divided into sex-specific quartiles. The mean changes in the pulmonary functions (FEV1, FVC %, and FEV1/FVC) and the odd ratios (ORs) of the risk of COPD across the quartiles of serum GGT levels were analyzed using linear regression models and logistic models. Age (continuous), height (continuous), weight (continuous), education (\high school or Chigh school), smoking status (2 groups; 1: current: C5 packs of cigarettes during lifetimes and smoking currently; 2: never or former: C5 packs of cigarettes during lifetimes but not smoking currently, or never), lifetime pack-years (continuous), and alcohol drinking status (regular or none) were selected as covariates associated with pulmonary function or serum GGT levels. Parallel models were repeated after excluding subjects with an elevated ALT level (serum ALT C40 IU/ L) or a history of regular drinking. Statistical tests for linearity were conducted by entering the categorical numbers of quartile of serum GGT into the regression models as a continuous variable. Finally, interactions between smoking status, non-current (never or former) and current, and GGT levels were evaluated. Analyses were conducted using the Stata (version 12.0, College Station, TX, USA) and weighted to the Korean population to provide nationally representative estimates. All tests of hypotheses were based on a type-I error rate of 0.05 using two-sided tests.
721
female current smokers (39.2 and 4.1 %, respectively). The geometric mean of serum GGT was higher in men than women (29.9 and 18.1 IU/L respectively). The means of the FEV1 and FVC % were slightly lower in men (90.9 and 92.4, respectively) than in women (95.7 and 94.9, respectively). Variables associated with an increase in the serum GGT level in both genders were age, weight, and alcohol drinking status. Smoking status was associated with elevated serum GGT only in men, and height and education were associated with elevated serum GGT only in women (Table 1). Next, we assessed whether there were any associations between sex-specific quartiles of serum GGT with FEV1, FVC %, and FEV1/FVC (Table 2). After adjusting for potential confounders, including cigarette smoking status and lifetime pack-years, decreases in the FEV1 and FVC % were observed with increasing serum GGT levels in both genders. The adjusted mean differences between the lowest and highest quartile of FEV1 and FVC % were -5.08 and -3.83, respectively, in men and -3.76 and -3.48, respectively, in women (Model 2). Similar associations were observed, when we restricted analyses to participants with normal liver function (serum ALT \40 IU/L) and non-regular drinking habits in both genders (Model 3). With respect to FEV1/FVC, after excluding subjects with an elevated ALT level or a history of regular drinking, an inverse association with serum GGT was observed only in men. There was a significant association of elevated serum GGT levels and the risk of COPD in men. This association was not observed in women (Table 3). In men, the adjusted ORs were 1.0, 1.69, 1.97, and 2.02 across the quartiles of GGT level (P for trend = 0.002, Model 2). After excluding subjects with abnormal liver function or regular alcohol consumption, the adjusted ORs were 1.0, 1.77, 2.45, and 2.50 (P for trend = 0.001, Model 3). In the stratified analysis (Table 4), the associations between serum GGT levels and pulmonary function parameters (FEV1 and FVC %) differed, depending upon the smoking status of the men and women. Inverse associations between serum GGT and FEV1 and FVC % were clearly observed only in non-current smokers even though p values for interactions between the GGT level and smoking status were marginally significant only in men (P = 0.066 and 0.061 for FEV1 % and FVC %, respectively). In contrast, no significant association of the FEV1/FVC ratio with serum GGT levels and smoking was observed in either gender.
Results Discussion The prevalence of spirometry defined COPD in men was higher than in women (17.3 and 5.4 %, respectively), as was the percentage of male current smokers compared to
This cross-sectional study demonstrated that the serum GGT levels were inversely related to pulmonary function
123
123 93.3 ± 0.7
b
a
50.9 20.8
Ex
Never
16.5
43.2
40.4
32.7
68.7
75.4 ± 0.5
92.2 ± 0.7
90.4 ± 0.8
21.1 ± 1.1
67.5 ± 0.5
168.0 ± 0.4
55.5 ± 0.7
473
22–30
15.4
45.8
38.9
43.4
71.0
76.2 ± 0.5
92.1 ± 0.6
90.5 ± 0.7
19.5 ± 1.0
71.4 ± 0.5
168.8 ± 0.3
53.2 ± 0.6
457
30–43
9.8
40.6
49.6
57.2
65.4
76.7 ± 0.5
91.0 ± 0.7
89.5 ± 0.7
21.3 ± 0.9
71.6 ± 0.5
168.9 ± 0.3
52.5 ± 0.5
451
C43
92.6
3.9
3.5
4.1
\0.001b
53.6
0.111
80.1 ± 0.3
96.8 ± 0.5
96.7 ± 0.5
\0.001
0.020
\0.001
\0.001
0.4 ± 0.1
56.6 ± 0.3
\0.001 0.116
155.9 ± 0.3
54.1 ± 0.6
874
\15
93.7
2.2
4.1
9.2
48.6
79.9 ± 0.4
95.6 ± 0.6
96.0 ± 0.6
0.7 ± 0.2
57.8 ± 0.5
155.6 ± 0.3
55.7 ± 0.7
522
15–18
92.9
3.2
3.9
7.4
37.2
79.7 ± 0.3
94.4 ± 0.6
96.0 ± 0.6
0.6 ± 0.2
58.8 ± 0.4
154.7 ± 0.3
58.0 ± 0.6
647
18–24
Quartiles of serum GGT in women (IU/L)
0.051
\0.001
Ptrend
P values were obtained by chi-square test
FEV1 %, forced expiratory volume in 1 s as the percentage of the predicted value; FVC %, forced vital capacity as the percentage of the predicted value
28.2
20.1
59.3
75.5 ± 0.5
Current
Smoking status
Regular alcohol consumption
Percent (%) CHigh school
FEV1/FVC ratio (%)
94.3 ± 0.6
18.4 ± 1.1
Lifetime pack-years
FEV1 %a
FVC %
65.7 ± 0.5
Weight (kg)
a
57.4 ± 0.7 168.2 ± 0.3
Age (year)
493
\22
Quartiles of serum GGT in men (IU/L)
Height (cm)
Mean ± standard error
No. of participants
Characteristics
Table 1 Subjects’ characteristics by sex-specific quartiles of serum gamma-glutamyltransferase (GGT) level in men and women
91.3
3.5
5.2
10.2
38.9
80.2 ± 0.3
92.4 ± 0.5
94.0 ± 0.6
0.7 ± 0.2
61.1 ± 0.5
155.2 ± 0.3
56.5 ± 0.6
666
C24
0.786b
0.001
\0.001
0.612
\0.001
0.004
0.098
\0.001
0.015
\0.001
Ptrend
722 Lung (2014) 192:719–727
Lung (2014) 192:719–727
723
Table 2 Crude and adjusted mean differences (confidence interval) in pulmonary function by sex-specific quartiles of serum gamma-glutamyltransferase (GGT) level in men and women Quartiles of serum GGT level (IU/L) in men \22
22–30
Ptrend 30–43
C43
FEV1 %a Unadjusted
0.00 (ref.)
-2.93 (-5.21, -0.65)
-2.81 (-4.61, -1.01)
-3.83 (-5.59, -2.08)
\0.001
Model 1b
0.00 (ref.)
-3.31 (-5.58, -1.04)
-3.70 (-5.53, -1.88)
-4.84 (-6.64, -3.05)
\0.001
Model 2c
0.00 (ref.)
-3.14 (-5.36, -0.91)
-3.94 (-5.83,-2.06)
-5.08 (-6.98, -3.19)
\0.001
Model 3d
0.00 (ref.)
-3.72 (-5.98, -1.47)
-5.15 (-7.20, -3.09)
-4.92 (-7.40, -2.44)
\0.001
Unadjusted Model 1b
0.00 (ref.) 0.00 (ref.)
-2.04 (-3.80, -0.27) -1.95 (-3.64, -0.26)
-2.18 (-3.77, -0.58) -1.74 (-3.34, -1.15)
-3.28 (-5.05, -1.52) -3.05 (-4.75, -1.35)
\0.001 0.001
Model 2c
0.00 (ref.)
-1.97 (-3.62, -0.32)
-2.13 (-3.70, -0.55)
-3.83 (-5.50, -2.16)
\0.001
Model 3d
0.00 (ref.)
-2.39 (-4.18, -0.60)
-2.91 (-4.80, -1.02)
-3.39 (-5.90, -0.89)
0.002
FVC %
a
FEV1/FVC % Unadjusted
0.00 (ref.)
-0.14 (-1.46, 1.18)
0.70 (-0.56, 1.96)
Model 1b
0.00 (ref.)
-1.11 (-2.31, 0.09)
-1.46 (-2.49, -0.43)
-1.26 (-2.33, -0.20)
1.17 (0.02, 2.32)
0.016
Model 2c
0.00 (ref.)
-0.96 (-2.17, 0.25)
-1.35 (-2.42, -0.28)
-0.86 (-2.02, -0.30)
0.119
Model 3d
0.00 (ref.)
-1.14 (-2.35, 0.06)
-1.85 (-3.11, -0.59)
-1.09 (-2.49, 0.31)
0.024
Quartiles of serum GGT level (IU/L) in women
0.020
Ptrend
\15
15–18
18–24
C24
FEV1 %a Unadjusted
0.00 (ref.)
-0.67 (-2.21, 0.85)
-0.72 (-2.46, 1.02)
-2.67 (-4.16, -1.18)
0.002
Model 1b
0.00 (ref.)
-1.02 (-2.54, 0.49)
-1.62 (-3.39, -0.14)
-3.52 (-5.05, -2.01)
\0.001
Model 2c
0.00 (ref.)
-1.19 (-2.69, 0.30)
-1.78 (-3.55, -0.01)
-3.76 (-5.32, -2.20)
\0.001
Model 3d
0.00 (ref.)
-1.11 (-2.65, 0.43)
-1.47 (-3.38, 0.44)
-3.05 (-4.63, -1.46)
0.001
FVC %a Unadjusted
0.00 (ref.)
-1.18 (-2.63, 0.27)
-2.40 (-3.87, -0.93)
-4.40 (-5738, -3.08)
\0.001
Model 1b
0.00 (ref.)
-0.79 (-2.27, 0.69)
-1.50 (-2.97, -0.16)
-3.19 (-4.54, -1.84)
\0.001
c
0.00 (ref.)
-0.93 (-2.38, 0.53)
-1.67 (-3.15, -0.20)
-3.48 (-4.84, -2.12)
\0.001
Model 3d FEV1/FVC %
0.00 (ref.)
-0.86 (-2.36, 0.63)
-1.57 (-3.17, 0.03)
-3.00 (-4.40, -1.59)
\0.001
Unadjusted
0.00 (ref.)
-0.21 (-1.04, 0.63)
-0.42 (-1.28, 0.44)
0.10 (-0.76, 0.97)
0.972
Model 1b
0.00 (ref.)
0.04 (-0.66, 0.74)
0.11 (-0.66, 0.88)
0.15 (-0.65, 0.94)
0.695
c
Model 2
0.00 (ref.)
0.02 (-0.67, 0.72)
0.12 (-0.63, 0.87)
0.18 (-0.60, 0.97)
0.628
Model 3d
0.00 (ref.)
0.04 (-0.71, 0.78)
0.24 (-0.56, 1.03)
0.41 (-0.41, 1.23)
0.301
Model 2
a
FEV1 %, forced expiratory volume in 1 s as the percentage of the predicted value; FVC %, forced vital capacity as the percentage of the predicted value b Model 1 Adjustment for age, height, and weight c
Model 2 Model 1 plus adjustment for education, alcohol drinking status, smoking status, and lifetime pack-years
d
Model 3 Analysis excluding participants with abnormal liver function (ALT C 40 IU/L) or a history of regular alcohol consumption from model 2 (n = 1072 for men, n = 2426 for women)
and were positively associated to an increased prevalence of COPD in Korean men aged over 40 years. These associations persisted after adjusting for smoking status and lifetime pack-years, as well as other potential confounders. In women, only pulmonary function was inversely associated with serum GGT. The fact that we did not observe an
association between COPD and serum GGT in women in this study may be due to the low prevalence of COPD in our female study subjects. An abnormally elevated GGT level is generally interpreted as a marker of alcohol abuse and liver damage. However, these factors cannot explain the associations of
123
724
Lung (2014) 192:719–727
Table 3 Crude and adjusted odd ratios (ORs) for the prevalence of COPD by sex-specific quartiles of serum gamma-glutamyltransferase (GGT) level Serum GGT level, men \22
22–30
30–43
C43
All subjects Case/no. at risk
95/493
106/473
73/457
Unadjusted
1.00 (ref.)
1.18 (0.81, 1.74)
0.96 (0.63, 1.47)
0.91 (0.61, 1.34)
79/451
Ptrend 0.421
Model 1a
1.00 (ref.)
1.73 (1.09, 2.76)
1.91 (1.23, 2.97)
2.17 (1.38–3.40)
\0.001
Model 2b
1.00 (ref.)
1.69 (1.05, 2.73)
1.97 (1.25, 3.10)
2.02 (1.23–3.28)
0.002
72/387 1.00 (ref.)
73/317 1.77 (1.02, 3.08)
35/219 2.45 (1.36, 4.41)
27/149 2.50 (1.26, 4.97)
Ptrend 0.001
Subset analysis Case/no. at risk Model 3c
Serum GGT level, women
All subjects Cases/no. at risk
\15
15–18
18–24
C24
42/874
24/522
33/647
28/666
Ptrend
Unadjusted
1.00 (ref.)
1.19 (0.68, 2.08)
1.12 (0.60, 2.10)
0.99 (0.53, 1.86)
0.991
Model 1a
1.00 (ref.)
1.17 (0.65, 2.10)
0.92 (0.47, 1.80)
1.08 (0.54, 2.15)
0.996
b
1.00 (ref.)
1.17 (0.65, 2.14)
0.85 (0.44, 1.66)
1.03 (0.52, 2.03)
0.857
Model 2
Subset analysis Cases/no. at risk Model 3c
40/838
22/474
30/587
24/427
Ptrend
1.00 (ref.)
1.22 (0.66, 2.25)
0.80 (0.41, 1.59)
1.00 (0.52, 1.92)
0.734
a
Model 1: Minimal adjustment for age, height, and weight
b
Model 2: Model 1 plus adjustment for education, alcohol drinking status, smoking status, and lifetime pack-years
c
Model 3: Analysis excluding participants with abnormal liver function (ALT C 40 IU/L) or a history of regular alcohol consumption adjusted for the same confounding factors as in model 2
GGT with pulmonary function and COPD since the subjects with normal ALT levels and non-regular alcohol consumption habits showed a clear decrease in FEV1 and FVC and a clear increase in the prevalence of COPD. Furthermore, these associations were observed within the reference range of serum GGT. At present, the mechanism or mechanisms that link serum GGT, a kind of liver enzyme from a conventional viewpoint, to pulmonary function and COPD are unknown. One possible explanation is that serum GGT is related to the lung function as an early and sensitive biomarker of oxidative stress in humans [3]. Although the relationship between serum GGT and cellular GGT is not known, cellular GGT activity is induced with the burden of oxidative stress to maintain intracellular GSH, which is a major antioxidant within the cell [14]. Several experimental studies have demonstrated that GGT gene expression is increased in response to oxidative stress [15, 16]. In addition, one human study showed that the levels of antioxidant vitamins inversely predicted serum GGT levels in a dose-dependent manner [17], and that serum GGT levels positively predicted F2-isoprostanes, a well-known marker
123
of oxidative stress [18]. An increased oxidant burden has been suggested to play an important role in the development of COPD [19]. Many experimental studies have demonstrated that excessive oxidative stress can lead to inflammatory gene transcription [20], mucus gland hyperplasia [21], and permanent destruction of the antioxidant system [22] in respiratory epithelial cells. Therefore, our study might reflect the association of oxidative stress with pulmonary function and COPD. An alternative and more specific interpretation is also possible. In the analysis stratified by smoking status, we unexpectedly found that GGT levels were associated with pulmonary function only in the group of non-current smokers; very little association was observed between GGT levels and pulmonary function in current smokers. This result contradicts our hypothesis that the GGT level, as a biomarker of oxidative stress, would be more strongly related to pulmonary function in smokers than in nonsmokers. In fact, expanding the concept of GGT as a biomarker of oxidative stress, recently it has been proposed that serum GGT can act as a cumulative biomarker of exposure to various chemicals which conjugated to GSH
Lung (2014) 192:719–727
725
Table 4 Adjusted mean differences in pulmonary function by quartiles of serum GGT level after stratification by smoking status in men and women Quartiles of serum GGT level (IU/L) in men
Ptrend
\22
22–30
30–43
C43
Never or former (n = 1213)
0.00 (ref.)
-4.77 (-7.16, -2.37)
-4.87 (-7.24, -2.49)
-6.60 (-8.88, -4.33) \0.001
Current (n = 661)
0.00 (ref.)
-0.36 (-3.59, 4.31)
-1.49 (-4.61, 1.63)
-1.79 (-5.38, 1.80)
Never or former (n = 1213)
0.00 (ref.)
-3.43 (-5.45, -1.40)
-3.50 (-5.42, -1.59)
-5.65 (-7.77, -3.52) \0.001
Current (n = 661)
0.00 (ref.)
1.17 (-1.47, 3.80)
1.11 (-1.65, 3.87)
-0.26 (-3.12, 2.60)
0.656
Never or former (n = 1213)
0.00 (ref.)
-1.17 (-2.39, 0.05)
-1.22 (-2.34, -0.09)
-0.59 (-2.05, 0.87)
0.335
Current (n = 661)
0.00 (ref.)
-0.59 (-2.79, 1.61)
-1.63 (-3.79, 0.53)
-1.09 (-3.05, 0.88)
0.221
Pinteraction
FEV1 %a Smoking status 0.066
0.160
FVC %a Smoking status 0.061
FEV1/FVC % Smoking status
Quartiles of serum GGT level (IU/L) in women
0.758
Ptrend
Pinteraction
0.526
\15
15–18
18–24
C24
Never or former (n = 2607)
0.00 (ref.)
-1.21 (-2.71, 0.30)
-1.99 (-3.73, -0.26)
-3.86 (-5.46, -2.25)
\0.001
Current (n = 102)
0.00 (ref.)
-1.50 (-9.77, 6.77)
3.45 (-6.52, 13.43)
-2.97 (-11.49, 5.56)
0.807
0.00 (ref.)
-0.82 (-2.29, 0.64)
-1.65 (-3.15, -0.16)
-3.49 (-4.88, -2.11)
\0.001
0.00 (ref.)
-2.78 (-10.53, 4.97)
-3.14 (-10.90, 4.63)
-2.26 (-9.76, 5.23)
Never or former (n = 2607)
0.00 (ref.)
-0.08 (-0.78, 0.63)
-0.02 (-0.77, 0.72)
0.13 (-0.64, 0.91)
0.763
Current (n = 102)
0.00 (ref.)
1.76 (-3.18, 6.70)
4.32 (-0.35, 8.99)
-0.17 (-5.36, 5.02)
0.775
FEV1 %a Smoking status
FVC %a Smoking status Never or former (n = 2607) Current (n = 102)
0.823
0.549
FEV1/FVC % Smoking status 0.404
Adjusted for age, sex, height, weight, education, alcohol drinking status, smoking status, and lifetime pack-years a
FEV1 %, forced expiratory volume in 1 s as the percentage of the predicted value; FVC %, forced vital capacity as the percentage of the predicted value
[4]. Although cigarette smoking exposes individuals to a complex mixture of chemicals, individuals are also exposed to a variety of airborne chemicals in their environment. In this context, the association of GGT levels and pulmonary function in non-smokers could reflect either exposure to indoor and outdoor air pollution or occupational exposure of individuals to various chemicals, which were not considered in this study. On the other hand, although it is difficult to find reasonable explanations for the lack of association among current smokers, as exposure to cigarette smoking itself is strongly related to serum GGT [23], any increase in the serum GGT level due to environmental pollutants may be masked in current smokers.
This study has several strengths and limitations. Seasonal variations and technical errors in the pulmonary function test were minimized using a perennial survey system and criteria for both acceptability and reproducibility. Moreover, the results from this study are applicable to the Korean population due to the assignment of weights. Nevertheless, the cross-sectional nature of the study lessens the ability to infer temporal relations, even though it is difficult to think of any situation in which pulmonary function can affect serum GGT within the normal range. In addition, a fixed FEV1/FVC ratio was used to define COPD, without consideration of the decrease in pulmonary function with age. This may have resulted in
123
726
Lung (2014) 192:719–727
over-diagnosis of COPD especially in elderly subjects [24, 25]. However, these factors are not likely to confound the associations we found. An artifactual positive association between serum GGT level and COPD is possible if serum GGT level is higher in the elderly subjects. However, we observed that serum GGT level was inversely associated with age in men. Finally, recent studies, using a new method based on molecular-size exclusion chromatography, have revealed that circulating GGT consists of four plasma GGT fractions (b-GGT, m-GGT, s-GGT, and f-GGT) [26], and the association of the GGT fractions with specific diseases and organs is beginning to be evaluated [27, 28]. Therefore, a specific GGT fraction, rather than total serum GGT, may be more clearly associated with pulmonary function and COPD.
Conclusions We found that in adults aged 40 years and older, representative of the Korean population, elevated GGT levels are associated with decreased pulmonary function (in both genders) and an increased prevalence of COPD in men. Importantly, however, these associations tended to be restricted to non-current smokers. Although the current findings need to be confirmed in future prospective studies, and the underlying mechanism is unclear, serum GGT levels can be used to detect non-current smokers who are at high risk of decreased pulmonary function and/or COPD. Acknowledgments This work was partly supported by Grants from the Korean Health Technology R&D Project, Ministry of Health & Welfare, Republic of Korea (HI13C0715), and the National Research Foundation of Korea (No. 2013R1A2A2A01068254). Conflict of interest
Nothing to declare.
References 1. Targher G (2010) Elevated serum gamma-glutamyltransferase activity is associated with increased risk of mortality, incident type 2 diabetes, cardiovascular events, chronic kidney disease and cancer—a narrative review. Clin Chem Lab Med 48(2):147–157. doi:10.1515/CCLM.2010.031 2. Conigrave KM, Davies P, Haber P, Whitfield JB (2003) Traditional markers of excessive alcohol use. Addiction 98(Suppl 2):31–43 3. Lee DH, Blomhoff R, Jacobs DR Jr (2004) Is serum gamma glutamyltransferase a marker of oxidative stress? Free Radic Res 38(6):535–539 4. Lee DH, Jacobs DR Jr (2009) Is serum gamma-glutamyltransferase a marker of exposure to various environmental pollutants? Free Radic Res 43(6):533–537. doi:10.1080/1071576 0902893324 5. Whitfield JB (2001) Gamma glutamyl transferase. Crit Rev Clin Lab Sci 38(4):263–355. doi:10.1080/20014091084227
123
6. Mistry D, Stockley RA (2010) Gamma-glutamyl transferase: the silent partner? COPD 7(4):285–290. doi:10.3109/15412555.2010. 496819 7. Sakuta H, Suzuki T, Yasuda H, Ito T (2005) Gamma-glutamyl transferase and airflow obstruction in middle-aged men. Eur J Inter Med 16(5):348–351. doi:10.1016/j.ejim.2005.06.005 8. Kim ES, Moon S, Han JH, Kim SA, Lee EH, Beom SH et al (2012) Association between serum gamma-glutamyltransferase and pulmonary dysfunction. Clin Chem Lab Med 50(11):2053– 2055. doi:10.1515/cclm-2012-0145 9. Zureik M, Liard R, Kauffmann F, Henry C, Neukirch F (1996) Alcohol consumption, gamma-glutamyl transpeptidase (GGT), and pulmonary function: a cross-sectional and longitudinal study in working men. Alcohol Clin Exp Res 20(9):1507–1511 10. Biljak VR, Rumora L, Cepelak I, Pancirov D, Popovic-Grle S, Soric J et al (2013) Gamma-glutamyltransferase and C-reactive protein in stable chronic obstructive pulmonary disease. Collegium antropologicum. 37(1):221–227 11. Persijn JP, van der Slik W (1976) A new method for the determination of gamma-glutamyltransferase in serum. J Clin Chem Clin Biochem 14(9):421–427 12. Miller MR, Hankinson J, Brusasco V, Burgos F, Casaburi R, Coates A et al (2005) Standardisation of spirometry. Eur Respir J 26(2):319–338. doi:10.1183/09031936.05.00034805 13. Choi JK, Paek D, Lee JO (2005) Normal predictive values of spirometry in Korean population. Tuberc Respir Dis 58(3):230–242 14. Zhang H, Forman HJ, Choi J (2005) Gamma-glutamyl transpeptidase in glutathione biosynthesis. Methods Enzymol 401:468–483. doi:10.1016/S0076-6879(05)01028-1 15. Liu RM, Shi MM, Giulivi C, Forman HJ (1998) Quinones increase gamma-glutamyl transpeptidase expression by multiple mechanisms in rat lung epithelial cells. Am J Physiol 274(3 Pt 1):L330–L336 16. Takahashi Y, Oakes SM, Williams MC, Takahashi S, Miura T, Joyce-Brady M (1997) Nitrogen dioxide exposure activates gamma-glutamyl transferase gene expression in rat lung. Toxicol Appl Pharmacol 143(2):388–396. doi:10.1006/taap.1996.8087 17. Lee DH, Gross MD, Jacobs DR Jr (2004) Cardiovascular risk development in young adults S. Association of serum carotenoids and tocopherols with gamma-glutamyltransferase: the cardiovascular risk development in young adults (CARDIA) study. Clin Chem 50(3):582–588. doi:10.1373/clinchem.2003.028852 18. Lee DH, Jacobs DR Jr, Gross M, Kiefe CI, Roseman J, Lewis CE et al (2003) Gamma-glutamyltransferase is a predictor of incident diabetes and hypertension: the coronary artery risk development in young adults (CARDIA) study. Clin Chem 49(8):1358–1366 19. Lin JL, Thomas PS (2010) Current perspectives of oxidative stress and its measurement in chronic obstructive pulmonary disease. COPD 7(4):291–306. doi:10.3109/15412555.2010.496818 20. Tomita K, Barnes PJ, Adcock IM (2003) The effect of oxidative stress on histone acetylation and IL-8 release. Biochem Biophys Res Commun 301(2):572–577 21. Wright DT, Fischer BM, Li C, Rochelle LG, Akley NJ, Adler KB (1996) Oxidant stress stimulates mucin secretion and PLC in airway epithelium via a nitric oxide-dependent mechanism. Am J Physiol 271(5 Pt 1):L854–L861 22. van der Toorn M, Smit-de Vries MP, Slebos DJ, de Bruin HG, Abello N, van Oosterhout AJ et al (2007) Cigarette smoke irreversibly modifies glutathione in airway epithelial cells. Am J Physiol Lung Cell Mol Physiol 293(5):L1156–L1162. doi:10. 1152/ajplung.00081.2007 23. Wannamethee SG, Shaper AG (2010) Cigarette smoking and serum liver enzymes: the role of alcohol and inflammation. Ann Clin Biochem 47(Pt 4):321–326. doi:10.1258/acb.2010.009303
Lung (2014) 192:719–727 24. Hwang YI, Kim CH, Kang HR, Shin T, Park SM, Jang SH et al (2009) Comparison of the prevalence of chronic obstructive pulmonary disease diagnosed by lower limit of normal and fixed ratio criteria. J Korean Med Sci 24(4):621–626. doi:10.3346/ jkms.2009.24.4.621 25. Shirtcliffe P, Weatherall M, Marsh S, Travers J, Hansell A, McNaughton A et al (2007) COPD prevalence in a random population survey: a matter of definition. Eur Respir J 30(2):232–239. doi:10.1183/09031936.00157906 26. Franzini M, Ottaviano V, Fierabracci V, Bramanti E, Zyw L, Barsacchi R et al (2008) Fractions of plasma gamma-glutamyltransferase in healthy individuals: reference values. Clin chim
727 acta 395(1–2):188–189. doi:10.1016/j.cca.2008.06.005 27. Franzini M, Paolicchi A, Fornaciari I, Ottaviano V, Fierabracci V, Maltinti M et al (2010) Cardiovascular risk factors and gamma-glutamyltransferase fractions in healthy individuals. Clin Chem Lab Med. 48(5):713–717. doi:10.1515/CCLM.2010.125 28. Franzini M, Fornaciari I, Fierabracci V, Elawadi HA, Bolognesi V, Maltinti S et al (2012) Accuracy of b-GGT fraction for the diagnosis of non-alcoholic fatty liver disease. Liver Int 32(4):629–634. doi:10.1111/j.1478-3231.2011.02673.x
123
Relationship of Serum Gamma-Glutamyltransferase Levels with Pulmonary Function and Chronic Obstructive Pulmonary Disease Hyun-Woo Kim • Seock-Hwan Lee Duk-Hee Lee
•
Received: 27 March 2014 / Accepted: 16 June 2014 / Published online: 11 July 2014 Ó Springer Science+Business Media New York 2014
Abstract Background Gamma-glutamyltransferase (GGT) levels within the normal reference range, possibly a biomarker of oxidative stress and/or exposure to various environmental chemicals, are associated with pulmonary function. However, it is unclear whether it is totally independent of cigarette smoking. Also, the potential interaction between serum GGT and cigarette smoking has not ever been evaluated. Therefore, this study investigated (1) whether serum GGT levels are associated with pulmonary function and chronic obstructive pulmonary disease (COPD), independent of cigarette smoking, and (2) whether there is any interaction between serum GGT and cigarette smoking status on pulmonary function. Methods The study subjects were 4,583 participants aged C40 in the 2010–2011 Korean National Health and Nutrition Examination Survey. The outcomes were pulmonary function (forced expiratory volume in 1 second [FEV1] and forced vital capacity [FVC]) and spirometrically defined COPD. Results After adjusting for potential confounders, including cigarette smoking, serum GGT levels were H.-W. Kim S.-H. Lee Department of Family Medicine, Daegu Medical Center, 157 Pyungli-ro, Seo-Gu, Daegu 703-713, Korea D.-H. Lee (&) Department of Preventative Medicine, School of Medicine, Kyungpook National University, 680 Gukchaebosang-ro, Jung-gu, Daegu 700-842, Korea e-mail: [email protected] D.-H. Lee BK21 Plus KNU Biomedical Convergence Program, Department of Biomedical Science, Kyungpook National University, 680 Gukchaebosang-ro, Jung-gu, Daegu 700-842, Korea
inversely associated with FEV1 and FVC in both genders and positively associated with the risk of COPD in men (all P values \0.01). In men, adjusted odds ratios of COPD were 1.0, 1.69, 1.97, and 2.02 across the quartiles of GGT level (Ptrend = 0.002). However, the associations between serum GGT and pulmonary function seemed to differ depending on the smoking status; inverse associations of GGT with FEV1 % and FVC % were clearly observed only among non-current smokers. Conclusions In conclusion, in non-smokers serum GGT levels can be used to detect individuals at high risk of decreased pulmonary function and/or COPD. Keywords Oxidative stress Airway limitation Smoking Air pollution
Introduction Over the last few decades, a number of epidemiological studies have demonstrated that gamma-glutamyltransferase (GGT) levels, within the normal reference range, are predictive of various clinical diseases, including type 2 diabetes, cardiovascular disease, chronic kidney disease, and cancer [1]. Abnormally elevated GGT levels have been used as a clinical marker of alcohol abuse or liver disease affecting the biliary system [2]; however, the association of GGT levels with other clinical diseases was independent of alcohol consumption or liver disease. At present, serum GGT within its reference range has been proposed as a biomarker of oxidative stress associated with glutathione (GSH) metabolism [3] and/or exposure to various environmental chemicals that conjugate to GSH [4]. Cellular GGT is expressed in most human tissues, including the lung, where it is involved in secretory and
123
720
absorptive processes [5, 6]. Although cellular GGT levels in the lung are relatively low compared to the levels detected in other organs, such as the liver or kidney [5, 6], several human studies have suggested an association between serum GGT levels and pulmonary function or chronic obstructive pulmonary disease (COPD) [7–10]. However, these findings are inconsistent. In particular, the effect of cigarette smoking, the most important risk factor in the pathogenesis of lung disease, is unclear. For example, in individuals who underwent routine medical checkups or policemen, the forced expiratory volume in 1 s (FEV1) and the forced vital capacity (FVC) decreased with increasing GGT levels, even after adjustment for smoking [8, 9]. In contrast, another study showed a statistically significant association between serum GGT and airway limitation in middle-aged men almost disappeared after adjusting for smoking [7]. Therefore, it is questionable whether the association between serum GGT and pulmonary function is totally independent of cigarette smoking. In addition, as the burden of oxidative stress is highly increased in smokers, there is a possibility of a synergic interaction between serum GGT and cigarette smoking on the risk of decreased pulmonary function. However, no study has evaluated this possibility. This study was performed to investigate [1] whether serum GGT levels are related to pulmonary function and COPD in a representative sample of the Korean general population, independent of cigarette smoking, and [2] whether serum GGT is more strongly related to pulmonary function and COPD in smokers compared to non-smokers.
Methods Study Participants and Data Collection Our study was based on data collected in 2010–2011 by the Fifth National Health and Nutrition Examination Survey (KNHANES V-1, 2). In KNHNES V, the Division of Chronic Disease Surveillance under the Korea Centers for Disease Control and Prevention (KCDC) collected nationally representative statistics from 2010 to 2012 on the health status, health-related behaviors, diet, and nutritional status of Korean people. Stratified, multistage probability sampling was used for the selection of household units. KNHANES consisted of a health interview survey, a nutrition survey, and a health examination survey. Data on demographic characteristics, diet, and healthrelated variables were collected through personal interviews and self-administered questionnaires. Physical examinations, blood sampling, and pulmonary function tests were carried out at a mobile examination center.
123
Lung (2014) 192:719–727
In total, 9,159 people aged 40 years and older people participated in KNHANES V-1, 2. Of these, spirometry results were available for 7,151 (78.1 %) participants. Out of these, we excluded patients without reproducible or acceptable spirometry data (1,195, 16.7 %), individuals without information on serum GGT or covariates (336 subjects), participants with a history of asthma or pulmonary tuberculosis that could cause airway limitation other than COPD (469 subjects), and those with abnormal levels of serum GGT (serum GGT[73 IU/L in men and[48 IU/L in women, 568 subjects) [11]. Thus, a total of 4,583 subjects were included in the final analyses. All the subjects in the study participated voluntarily under informed consent. The study protocol was approved by the institutional review board of the KCDC. Anthropometric and Biochemical Measurements The subjects’ weights and heights were measured, while they were wearing light clothing and had removed their shoes. Blood samples were collected from the antecubital vein of each participant after overnight fasting. The samples were processed, transported to the Central Testing Institute located in Seoul, Korea, and analyzed within 24 h. The analyses of GGT and alanine aminotransferase (ALT) were performed with a Hitachi Automatic Analyzer 7600 (Hitachi, Tokyo, Japan) by enzymatic or UV methods using commercially available kits (Sekisui, Tokyo, Japan/ Daiichi, Tokyo, Japan). Spirometry Data A pulmonary function test was conducted using dry rolling seal spirometers (Model 2130, Sensor Medics, Yorba Linda, CA, USA), after at least one practice blow, in the seated position with nose clips. The forced expiratory volume in 1 s (FEV1) and the forced vital capacity (FVC) were measured with effort in accordance with the criteria for acceptability, presented in the guidelines of the American Thoracic Society/ European Respiratory Society (ATS/ERS) [12], at least three times and at most eight times. Only subjects who had at least two proper curves of spirometry and a difference of B0.15 (L) between the two highest readings of FEV1 and FVC were included in the study (82.3 % of men and 77.7 % of women). A bronchodilator was not given before the spirometry. The percent predicted values of the following predictive equations, described by Choi et al. [13], were used to determine pulmonary function: In men: Predicted FVC ¼ 4:8434 0:00008633 ðageÞ2 þ 0:05292 ðheightÞ þ 0:01095 ðweightÞ
Lung (2014) 192:719–727
Predicted FEV1 ¼ 3:4132 0:0002484 ðageÞ2 þ 0:04578 ðheightÞ In women: Predicted FVC ¼ 3:0006 0:0001273 ðageÞ2 þ 0:03951 ðheightÞ þ 0:006892 ðweightÞ Predicted FEV1 ¼ 2:4114 0:0001920 ðageÞ2 þ 0:03558 ðheightÞ Subjects with a FEV1/FVC \70 % were identified as having COPD. Statistical Analysis As the serum GGT level, pulmonary function, and prevalence of COPD differed significantly between men and women, sex-specific analyses were performed. The serum GGT levels were divided into sex-specific quartiles. The mean changes in the pulmonary functions (FEV1, FVC %, and FEV1/FVC) and the odd ratios (ORs) of the risk of COPD across the quartiles of serum GGT levels were analyzed using linear regression models and logistic models. Age (continuous), height (continuous), weight (continuous), education (\high school or Chigh school), smoking status (2 groups; 1: current: C5 packs of cigarettes during lifetimes and smoking currently; 2: never or former: C5 packs of cigarettes during lifetimes but not smoking currently, or never), lifetime pack-years (continuous), and alcohol drinking status (regular or none) were selected as covariates associated with pulmonary function or serum GGT levels. Parallel models were repeated after excluding subjects with an elevated ALT level (serum ALT C40 IU/ L) or a history of regular drinking. Statistical tests for linearity were conducted by entering the categorical numbers of quartile of serum GGT into the regression models as a continuous variable. Finally, interactions between smoking status, non-current (never or former) and current, and GGT levels were evaluated. Analyses were conducted using the Stata (version 12.0, College Station, TX, USA) and weighted to the Korean population to provide nationally representative estimates. All tests of hypotheses were based on a type-I error rate of 0.05 using two-sided tests.
721
female current smokers (39.2 and 4.1 %, respectively). The geometric mean of serum GGT was higher in men than women (29.9 and 18.1 IU/L respectively). The means of the FEV1 and FVC % were slightly lower in men (90.9 and 92.4, respectively) than in women (95.7 and 94.9, respectively). Variables associated with an increase in the serum GGT level in both genders were age, weight, and alcohol drinking status. Smoking status was associated with elevated serum GGT only in men, and height and education were associated with elevated serum GGT only in women (Table 1). Next, we assessed whether there were any associations between sex-specific quartiles of serum GGT with FEV1, FVC %, and FEV1/FVC (Table 2). After adjusting for potential confounders, including cigarette smoking status and lifetime pack-years, decreases in the FEV1 and FVC % were observed with increasing serum GGT levels in both genders. The adjusted mean differences between the lowest and highest quartile of FEV1 and FVC % were -5.08 and -3.83, respectively, in men and -3.76 and -3.48, respectively, in women (Model 2). Similar associations were observed, when we restricted analyses to participants with normal liver function (serum ALT \40 IU/L) and non-regular drinking habits in both genders (Model 3). With respect to FEV1/FVC, after excluding subjects with an elevated ALT level or a history of regular drinking, an inverse association with serum GGT was observed only in men. There was a significant association of elevated serum GGT levels and the risk of COPD in men. This association was not observed in women (Table 3). In men, the adjusted ORs were 1.0, 1.69, 1.97, and 2.02 across the quartiles of GGT level (P for trend = 0.002, Model 2). After excluding subjects with abnormal liver function or regular alcohol consumption, the adjusted ORs were 1.0, 1.77, 2.45, and 2.50 (P for trend = 0.001, Model 3). In the stratified analysis (Table 4), the associations between serum GGT levels and pulmonary function parameters (FEV1 and FVC %) differed, depending upon the smoking status of the men and women. Inverse associations between serum GGT and FEV1 and FVC % were clearly observed only in non-current smokers even though p values for interactions between the GGT level and smoking status were marginally significant only in men (P = 0.066 and 0.061 for FEV1 % and FVC %, respectively). In contrast, no significant association of the FEV1/FVC ratio with serum GGT levels and smoking was observed in either gender.
Results Discussion The prevalence of spirometry defined COPD in men was higher than in women (17.3 and 5.4 %, respectively), as was the percentage of male current smokers compared to
This cross-sectional study demonstrated that the serum GGT levels were inversely related to pulmonary function
123
123 93.3 ± 0.7
b
a
50.9 20.8
Ex
Never
16.5
43.2
40.4
32.7
68.7
75.4 ± 0.5
92.2 ± 0.7
90.4 ± 0.8
21.1 ± 1.1
67.5 ± 0.5
168.0 ± 0.4
55.5 ± 0.7
473
22–30
15.4
45.8
38.9
43.4
71.0
76.2 ± 0.5
92.1 ± 0.6
90.5 ± 0.7
19.5 ± 1.0
71.4 ± 0.5
168.8 ± 0.3
53.2 ± 0.6
457
30–43
9.8
40.6
49.6
57.2
65.4
76.7 ± 0.5
91.0 ± 0.7
89.5 ± 0.7
21.3 ± 0.9
71.6 ± 0.5
168.9 ± 0.3
52.5 ± 0.5
451
C43
92.6
3.9
3.5
4.1
\0.001b
53.6
0.111
80.1 ± 0.3
96.8 ± 0.5
96.7 ± 0.5
\0.001
0.020
\0.001
\0.001
0.4 ± 0.1
56.6 ± 0.3
\0.001 0.116
155.9 ± 0.3
54.1 ± 0.6
874
\15
93.7
2.2
4.1
9.2
48.6
79.9 ± 0.4
95.6 ± 0.6
96.0 ± 0.6
0.7 ± 0.2
57.8 ± 0.5
155.6 ± 0.3
55.7 ± 0.7
522
15–18
92.9
3.2
3.9
7.4
37.2
79.7 ± 0.3
94.4 ± 0.6
96.0 ± 0.6
0.6 ± 0.2
58.8 ± 0.4
154.7 ± 0.3
58.0 ± 0.6
647
18–24
Quartiles of serum GGT in women (IU/L)
0.051
\0.001
Ptrend
P values were obtained by chi-square test
FEV1 %, forced expiratory volume in 1 s as the percentage of the predicted value; FVC %, forced vital capacity as the percentage of the predicted value
28.2
20.1
59.3
75.5 ± 0.5
Current
Smoking status
Regular alcohol consumption
Percent (%) CHigh school
FEV1/FVC ratio (%)
94.3 ± 0.6
18.4 ± 1.1
Lifetime pack-years
FEV1 %a
FVC %
65.7 ± 0.5
Weight (kg)
a
57.4 ± 0.7 168.2 ± 0.3
Age (year)
493
\22
Quartiles of serum GGT in men (IU/L)
Height (cm)
Mean ± standard error
No. of participants
Characteristics
Table 1 Subjects’ characteristics by sex-specific quartiles of serum gamma-glutamyltransferase (GGT) level in men and women
91.3
3.5
5.2
10.2
38.9
80.2 ± 0.3
92.4 ± 0.5
94.0 ± 0.6
0.7 ± 0.2
61.1 ± 0.5
155.2 ± 0.3
56.5 ± 0.6
666
C24
0.786b
0.001
\0.001
0.612
\0.001
0.004
0.098
\0.001
0.015
\0.001
Ptrend
722 Lung (2014) 192:719–727
Lung (2014) 192:719–727
723
Table 2 Crude and adjusted mean differences (confidence interval) in pulmonary function by sex-specific quartiles of serum gamma-glutamyltransferase (GGT) level in men and women Quartiles of serum GGT level (IU/L) in men \22
22–30
Ptrend 30–43
C43
FEV1 %a Unadjusted
0.00 (ref.)
-2.93 (-5.21, -0.65)
-2.81 (-4.61, -1.01)
-3.83 (-5.59, -2.08)
\0.001
Model 1b
0.00 (ref.)
-3.31 (-5.58, -1.04)
-3.70 (-5.53, -1.88)
-4.84 (-6.64, -3.05)
\0.001
Model 2c
0.00 (ref.)
-3.14 (-5.36, -0.91)
-3.94 (-5.83,-2.06)
-5.08 (-6.98, -3.19)
\0.001
Model 3d
0.00 (ref.)
-3.72 (-5.98, -1.47)
-5.15 (-7.20, -3.09)
-4.92 (-7.40, -2.44)
\0.001
Unadjusted Model 1b
0.00 (ref.) 0.00 (ref.)
-2.04 (-3.80, -0.27) -1.95 (-3.64, -0.26)
-2.18 (-3.77, -0.58) -1.74 (-3.34, -1.15)
-3.28 (-5.05, -1.52) -3.05 (-4.75, -1.35)
\0.001 0.001
Model 2c
0.00 (ref.)
-1.97 (-3.62, -0.32)
-2.13 (-3.70, -0.55)
-3.83 (-5.50, -2.16)
\0.001
Model 3d
0.00 (ref.)
-2.39 (-4.18, -0.60)
-2.91 (-4.80, -1.02)
-3.39 (-5.90, -0.89)
0.002
FVC %
a
FEV1/FVC % Unadjusted
0.00 (ref.)
-0.14 (-1.46, 1.18)
0.70 (-0.56, 1.96)
Model 1b
0.00 (ref.)
-1.11 (-2.31, 0.09)
-1.46 (-2.49, -0.43)
-1.26 (-2.33, -0.20)
1.17 (0.02, 2.32)
0.016
Model 2c
0.00 (ref.)
-0.96 (-2.17, 0.25)
-1.35 (-2.42, -0.28)
-0.86 (-2.02, -0.30)
0.119
Model 3d
0.00 (ref.)
-1.14 (-2.35, 0.06)
-1.85 (-3.11, -0.59)
-1.09 (-2.49, 0.31)
0.024
Quartiles of serum GGT level (IU/L) in women
0.020
Ptrend
\15
15–18
18–24
C24
FEV1 %a Unadjusted
0.00 (ref.)
-0.67 (-2.21, 0.85)
-0.72 (-2.46, 1.02)
-2.67 (-4.16, -1.18)
0.002
Model 1b
0.00 (ref.)
-1.02 (-2.54, 0.49)
-1.62 (-3.39, -0.14)
-3.52 (-5.05, -2.01)
\0.001
Model 2c
0.00 (ref.)
-1.19 (-2.69, 0.30)
-1.78 (-3.55, -0.01)
-3.76 (-5.32, -2.20)
\0.001
Model 3d
0.00 (ref.)
-1.11 (-2.65, 0.43)
-1.47 (-3.38, 0.44)
-3.05 (-4.63, -1.46)
0.001
FVC %a Unadjusted
0.00 (ref.)
-1.18 (-2.63, 0.27)
-2.40 (-3.87, -0.93)
-4.40 (-5738, -3.08)
\0.001
Model 1b
0.00 (ref.)
-0.79 (-2.27, 0.69)
-1.50 (-2.97, -0.16)
-3.19 (-4.54, -1.84)
\0.001
c
0.00 (ref.)
-0.93 (-2.38, 0.53)
-1.67 (-3.15, -0.20)
-3.48 (-4.84, -2.12)
\0.001
Model 3d FEV1/FVC %
0.00 (ref.)
-0.86 (-2.36, 0.63)
-1.57 (-3.17, 0.03)
-3.00 (-4.40, -1.59)
\0.001
Unadjusted
0.00 (ref.)
-0.21 (-1.04, 0.63)
-0.42 (-1.28, 0.44)
0.10 (-0.76, 0.97)
0.972
Model 1b
0.00 (ref.)
0.04 (-0.66, 0.74)
0.11 (-0.66, 0.88)
0.15 (-0.65, 0.94)
0.695
c
Model 2
0.00 (ref.)
0.02 (-0.67, 0.72)
0.12 (-0.63, 0.87)
0.18 (-0.60, 0.97)
0.628
Model 3d
0.00 (ref.)
0.04 (-0.71, 0.78)
0.24 (-0.56, 1.03)
0.41 (-0.41, 1.23)
0.301
Model 2
a
FEV1 %, forced expiratory volume in 1 s as the percentage of the predicted value; FVC %, forced vital capacity as the percentage of the predicted value b Model 1 Adjustment for age, height, and weight c
Model 2 Model 1 plus adjustment for education, alcohol drinking status, smoking status, and lifetime pack-years
d
Model 3 Analysis excluding participants with abnormal liver function (ALT C 40 IU/L) or a history of regular alcohol consumption from model 2 (n = 1072 for men, n = 2426 for women)
and were positively associated to an increased prevalence of COPD in Korean men aged over 40 years. These associations persisted after adjusting for smoking status and lifetime pack-years, as well as other potential confounders. In women, only pulmonary function was inversely associated with serum GGT. The fact that we did not observe an
association between COPD and serum GGT in women in this study may be due to the low prevalence of COPD in our female study subjects. An abnormally elevated GGT level is generally interpreted as a marker of alcohol abuse and liver damage. However, these factors cannot explain the associations of
123
724
Lung (2014) 192:719–727
Table 3 Crude and adjusted odd ratios (ORs) for the prevalence of COPD by sex-specific quartiles of serum gamma-glutamyltransferase (GGT) level Serum GGT level, men \22
22–30
30–43
C43
All subjects Case/no. at risk
95/493
106/473
73/457
Unadjusted
1.00 (ref.)
1.18 (0.81, 1.74)
0.96 (0.63, 1.47)
0.91 (0.61, 1.34)
79/451
Ptrend 0.421
Model 1a
1.00 (ref.)
1.73 (1.09, 2.76)
1.91 (1.23, 2.97)
2.17 (1.38–3.40)
\0.001
Model 2b
1.00 (ref.)
1.69 (1.05, 2.73)
1.97 (1.25, 3.10)
2.02 (1.23–3.28)
0.002
72/387 1.00 (ref.)
73/317 1.77 (1.02, 3.08)
35/219 2.45 (1.36, 4.41)
27/149 2.50 (1.26, 4.97)
Ptrend 0.001
Subset analysis Case/no. at risk Model 3c
Serum GGT level, women
All subjects Cases/no. at risk
\15
15–18
18–24
C24
42/874
24/522
33/647
28/666
Ptrend
Unadjusted
1.00 (ref.)
1.19 (0.68, 2.08)
1.12 (0.60, 2.10)
0.99 (0.53, 1.86)
0.991
Model 1a
1.00 (ref.)
1.17 (0.65, 2.10)
0.92 (0.47, 1.80)
1.08 (0.54, 2.15)
0.996
b
1.00 (ref.)
1.17 (0.65, 2.14)
0.85 (0.44, 1.66)
1.03 (0.52, 2.03)
0.857
Model 2
Subset analysis Cases/no. at risk Model 3c
40/838
22/474
30/587
24/427
Ptrend
1.00 (ref.)
1.22 (0.66, 2.25)
0.80 (0.41, 1.59)
1.00 (0.52, 1.92)
0.734
a
Model 1: Minimal adjustment for age, height, and weight
b
Model 2: Model 1 plus adjustment for education, alcohol drinking status, smoking status, and lifetime pack-years
c
Model 3: Analysis excluding participants with abnormal liver function (ALT C 40 IU/L) or a history of regular alcohol consumption adjusted for the same confounding factors as in model 2
GGT with pulmonary function and COPD since the subjects with normal ALT levels and non-regular alcohol consumption habits showed a clear decrease in FEV1 and FVC and a clear increase in the prevalence of COPD. Furthermore, these associations were observed within the reference range of serum GGT. At present, the mechanism or mechanisms that link serum GGT, a kind of liver enzyme from a conventional viewpoint, to pulmonary function and COPD are unknown. One possible explanation is that serum GGT is related to the lung function as an early and sensitive biomarker of oxidative stress in humans [3]. Although the relationship between serum GGT and cellular GGT is not known, cellular GGT activity is induced with the burden of oxidative stress to maintain intracellular GSH, which is a major antioxidant within the cell [14]. Several experimental studies have demonstrated that GGT gene expression is increased in response to oxidative stress [15, 16]. In addition, one human study showed that the levels of antioxidant vitamins inversely predicted serum GGT levels in a dose-dependent manner [17], and that serum GGT levels positively predicted F2-isoprostanes, a well-known marker
123
of oxidative stress [18]. An increased oxidant burden has been suggested to play an important role in the development of COPD [19]. Many experimental studies have demonstrated that excessive oxidative stress can lead to inflammatory gene transcription [20], mucus gland hyperplasia [21], and permanent destruction of the antioxidant system [22] in respiratory epithelial cells. Therefore, our study might reflect the association of oxidative stress with pulmonary function and COPD. An alternative and more specific interpretation is also possible. In the analysis stratified by smoking status, we unexpectedly found that GGT levels were associated with pulmonary function only in the group of non-current smokers; very little association was observed between GGT levels and pulmonary function in current smokers. This result contradicts our hypothesis that the GGT level, as a biomarker of oxidative stress, would be more strongly related to pulmonary function in smokers than in nonsmokers. In fact, expanding the concept of GGT as a biomarker of oxidative stress, recently it has been proposed that serum GGT can act as a cumulative biomarker of exposure to various chemicals which conjugated to GSH
Lung (2014) 192:719–727
725
Table 4 Adjusted mean differences in pulmonary function by quartiles of serum GGT level after stratification by smoking status in men and women Quartiles of serum GGT level (IU/L) in men
Ptrend
\22
22–30
30–43
C43
Never or former (n = 1213)
0.00 (ref.)
-4.77 (-7.16, -2.37)
-4.87 (-7.24, -2.49)
-6.60 (-8.88, -4.33) \0.001
Current (n = 661)
0.00 (ref.)
-0.36 (-3.59, 4.31)
-1.49 (-4.61, 1.63)
-1.79 (-5.38, 1.80)
Never or former (n = 1213)
0.00 (ref.)
-3.43 (-5.45, -1.40)
-3.50 (-5.42, -1.59)
-5.65 (-7.77, -3.52) \0.001
Current (n = 661)
0.00 (ref.)
1.17 (-1.47, 3.80)
1.11 (-1.65, 3.87)
-0.26 (-3.12, 2.60)
0.656
Never or former (n = 1213)
0.00 (ref.)
-1.17 (-2.39, 0.05)
-1.22 (-2.34, -0.09)
-0.59 (-2.05, 0.87)
0.335
Current (n = 661)
0.00 (ref.)
-0.59 (-2.79, 1.61)
-1.63 (-3.79, 0.53)
-1.09 (-3.05, 0.88)
0.221
Pinteraction
FEV1 %a Smoking status 0.066
0.160
FVC %a Smoking status 0.061
FEV1/FVC % Smoking status
Quartiles of serum GGT level (IU/L) in women
0.758
Ptrend
Pinteraction
0.526
\15
15–18
18–24
C24
Never or former (n = 2607)
0.00 (ref.)
-1.21 (-2.71, 0.30)
-1.99 (-3.73, -0.26)
-3.86 (-5.46, -2.25)
\0.001
Current (n = 102)
0.00 (ref.)
-1.50 (-9.77, 6.77)
3.45 (-6.52, 13.43)
-2.97 (-11.49, 5.56)
0.807
0.00 (ref.)
-0.82 (-2.29, 0.64)
-1.65 (-3.15, -0.16)
-3.49 (-4.88, -2.11)
\0.001
0.00 (ref.)
-2.78 (-10.53, 4.97)
-3.14 (-10.90, 4.63)
-2.26 (-9.76, 5.23)
Never or former (n = 2607)
0.00 (ref.)
-0.08 (-0.78, 0.63)
-0.02 (-0.77, 0.72)
0.13 (-0.64, 0.91)
0.763
Current (n = 102)
0.00 (ref.)
1.76 (-3.18, 6.70)
4.32 (-0.35, 8.99)
-0.17 (-5.36, 5.02)
0.775
FEV1 %a Smoking status
FVC %a Smoking status Never or former (n = 2607) Current (n = 102)
0.823
0.549
FEV1/FVC % Smoking status 0.404
Adjusted for age, sex, height, weight, education, alcohol drinking status, smoking status, and lifetime pack-years a
FEV1 %, forced expiratory volume in 1 s as the percentage of the predicted value; FVC %, forced vital capacity as the percentage of the predicted value
[4]. Although cigarette smoking exposes individuals to a complex mixture of chemicals, individuals are also exposed to a variety of airborne chemicals in their environment. In this context, the association of GGT levels and pulmonary function in non-smokers could reflect either exposure to indoor and outdoor air pollution or occupational exposure of individuals to various chemicals, which were not considered in this study. On the other hand, although it is difficult to find reasonable explanations for the lack of association among current smokers, as exposure to cigarette smoking itself is strongly related to serum GGT [23], any increase in the serum GGT level due to environmental pollutants may be masked in current smokers.
This study has several strengths and limitations. Seasonal variations and technical errors in the pulmonary function test were minimized using a perennial survey system and criteria for both acceptability and reproducibility. Moreover, the results from this study are applicable to the Korean population due to the assignment of weights. Nevertheless, the cross-sectional nature of the study lessens the ability to infer temporal relations, even though it is difficult to think of any situation in which pulmonary function can affect serum GGT within the normal range. In addition, a fixed FEV1/FVC ratio was used to define COPD, without consideration of the decrease in pulmonary function with age. This may have resulted in
123
726
Lung (2014) 192:719–727
over-diagnosis of COPD especially in elderly subjects [24, 25]. However, these factors are not likely to confound the associations we found. An artifactual positive association between serum GGT level and COPD is possible if serum GGT level is higher in the elderly subjects. However, we observed that serum GGT level was inversely associated with age in men. Finally, recent studies, using a new method based on molecular-size exclusion chromatography, have revealed that circulating GGT consists of four plasma GGT fractions (b-GGT, m-GGT, s-GGT, and f-GGT) [26], and the association of the GGT fractions with specific diseases and organs is beginning to be evaluated [27, 28]. Therefore, a specific GGT fraction, rather than total serum GGT, may be more clearly associated with pulmonary function and COPD.
Conclusions We found that in adults aged 40 years and older, representative of the Korean population, elevated GGT levels are associated with decreased pulmonary function (in both genders) and an increased prevalence of COPD in men. Importantly, however, these associations tended to be restricted to non-current smokers. Although the current findings need to be confirmed in future prospective studies, and the underlying mechanism is unclear, serum GGT levels can be used to detect non-current smokers who are at high risk of decreased pulmonary function and/or COPD. Acknowledgments This work was partly supported by Grants from the Korean Health Technology R&D Project, Ministry of Health & Welfare, Republic of Korea (HI13C0715), and the National Research Foundation of Korea (No. 2013R1A2A2A01068254). Conflict of interest
Nothing to declare.
References 1. Targher G (2010) Elevated serum gamma-glutamyltransferase activity is associated with increased risk of mortality, incident type 2 diabetes, cardiovascular events, chronic kidney disease and cancer—a narrative review. Clin Chem Lab Med 48(2):147–157. doi:10.1515/CCLM.2010.031 2. Conigrave KM, Davies P, Haber P, Whitfield JB (2003) Traditional markers of excessive alcohol use. Addiction 98(Suppl 2):31–43 3. Lee DH, Blomhoff R, Jacobs DR Jr (2004) Is serum gamma glutamyltransferase a marker of oxidative stress? Free Radic Res 38(6):535–539 4. Lee DH, Jacobs DR Jr (2009) Is serum gamma-glutamyltransferase a marker of exposure to various environmental pollutants? Free Radic Res 43(6):533–537. doi:10.1080/1071576 0902893324 5. Whitfield JB (2001) Gamma glutamyl transferase. Crit Rev Clin Lab Sci 38(4):263–355. doi:10.1080/20014091084227
123
6. Mistry D, Stockley RA (2010) Gamma-glutamyl transferase: the silent partner? COPD 7(4):285–290. doi:10.3109/15412555.2010. 496819 7. Sakuta H, Suzuki T, Yasuda H, Ito T (2005) Gamma-glutamyl transferase and airflow obstruction in middle-aged men. Eur J Inter Med 16(5):348–351. doi:10.1016/j.ejim.2005.06.005 8. Kim ES, Moon S, Han JH, Kim SA, Lee EH, Beom SH et al (2012) Association between serum gamma-glutamyltransferase and pulmonary dysfunction. Clin Chem Lab Med 50(11):2053– 2055. doi:10.1515/cclm-2012-0145 9. Zureik M, Liard R, Kauffmann F, Henry C, Neukirch F (1996) Alcohol consumption, gamma-glutamyl transpeptidase (GGT), and pulmonary function: a cross-sectional and longitudinal study in working men. Alcohol Clin Exp Res 20(9):1507–1511 10. Biljak VR, Rumora L, Cepelak I, Pancirov D, Popovic-Grle S, Soric J et al (2013) Gamma-glutamyltransferase and C-reactive protein in stable chronic obstructive pulmonary disease. Collegium antropologicum. 37(1):221–227 11. Persijn JP, van der Slik W (1976) A new method for the determination of gamma-glutamyltransferase in serum. J Clin Chem Clin Biochem 14(9):421–427 12. Miller MR, Hankinson J, Brusasco V, Burgos F, Casaburi R, Coates A et al (2005) Standardisation of spirometry. Eur Respir J 26(2):319–338. doi:10.1183/09031936.05.00034805 13. Choi JK, Paek D, Lee JO (2005) Normal predictive values of spirometry in Korean population. Tuberc Respir Dis 58(3):230–242 14. Zhang H, Forman HJ, Choi J (2005) Gamma-glutamyl transpeptidase in glutathione biosynthesis. Methods Enzymol 401:468–483. doi:10.1016/S0076-6879(05)01028-1 15. Liu RM, Shi MM, Giulivi C, Forman HJ (1998) Quinones increase gamma-glutamyl transpeptidase expression by multiple mechanisms in rat lung epithelial cells. Am J Physiol 274(3 Pt 1):L330–L336 16. Takahashi Y, Oakes SM, Williams MC, Takahashi S, Miura T, Joyce-Brady M (1997) Nitrogen dioxide exposure activates gamma-glutamyl transferase gene expression in rat lung. Toxicol Appl Pharmacol 143(2):388–396. doi:10.1006/taap.1996.8087 17. Lee DH, Gross MD, Jacobs DR Jr (2004) Cardiovascular risk development in young adults S. Association of serum carotenoids and tocopherols with gamma-glutamyltransferase: the cardiovascular risk development in young adults (CARDIA) study. Clin Chem 50(3):582–588. doi:10.1373/clinchem.2003.028852 18. Lee DH, Jacobs DR Jr, Gross M, Kiefe CI, Roseman J, Lewis CE et al (2003) Gamma-glutamyltransferase is a predictor of incident diabetes and hypertension: the coronary artery risk development in young adults (CARDIA) study. Clin Chem 49(8):1358–1366 19. Lin JL, Thomas PS (2010) Current perspectives of oxidative stress and its measurement in chronic obstructive pulmonary disease. COPD 7(4):291–306. doi:10.3109/15412555.2010.496818 20. Tomita K, Barnes PJ, Adcock IM (2003) The effect of oxidative stress on histone acetylation and IL-8 release. Biochem Biophys Res Commun 301(2):572–577 21. Wright DT, Fischer BM, Li C, Rochelle LG, Akley NJ, Adler KB (1996) Oxidant stress stimulates mucin secretion and PLC in airway epithelium via a nitric oxide-dependent mechanism. Am J Physiol 271(5 Pt 1):L854–L861 22. van der Toorn M, Smit-de Vries MP, Slebos DJ, de Bruin HG, Abello N, van Oosterhout AJ et al (2007) Cigarette smoke irreversibly modifies glutathione in airway epithelial cells. Am J Physiol Lung Cell Mol Physiol 293(5):L1156–L1162. doi:10. 1152/ajplung.00081.2007 23. Wannamethee SG, Shaper AG (2010) Cigarette smoking and serum liver enzymes: the role of alcohol and inflammation. Ann Clin Biochem 47(Pt 4):321–326. doi:10.1258/acb.2010.009303
Lung (2014) 192:719–727 24. Hwang YI, Kim CH, Kang HR, Shin T, Park SM, Jang SH et al (2009) Comparison of the prevalence of chronic obstructive pulmonary disease diagnosed by lower limit of normal and fixed ratio criteria. J Korean Med Sci 24(4):621–626. doi:10.3346/ jkms.2009.24.4.621 25. Shirtcliffe P, Weatherall M, Marsh S, Travers J, Hansell A, McNaughton A et al (2007) COPD prevalence in a random population survey: a matter of definition. Eur Respir J 30(2):232–239. doi:10.1183/09031936.00157906 26. Franzini M, Ottaviano V, Fierabracci V, Bramanti E, Zyw L, Barsacchi R et al (2008) Fractions of plasma gamma-glutamyltransferase in healthy individuals: reference values. Clin chim
727 acta 395(1–2):188–189. doi:10.1016/j.cca.2008.06.005 27. Franzini M, Paolicchi A, Fornaciari I, Ottaviano V, Fierabracci V, Maltinti M et al (2010) Cardiovascular risk factors and gamma-glutamyltransferase fractions in healthy individuals. Clin Chem Lab Med. 48(5):713–717. doi:10.1515/CCLM.2010.125 28. Franzini M, Fornaciari I, Fierabracci V, Elawadi HA, Bolognesi V, Maltinti S et al (2012) Accuracy of b-GGT fraction for the diagnosis of non-alcoholic fatty liver disease. Liver Int 32(4):629–634. doi:10.1111/j.1478-3231.2011.02673.x
123
