Vital capacity and selected metabolic diseases in middle-aged Japanese men

1Department of Internal Medicine; 2Department of Research and Laboratory, Self-Defense Forces Central Hospital, Tokyo, Japan Correspondence and reprints: Dr Hidenari Sakuta, Department of Education, Ground Self-Defense Force Medical School, 1-2-24 Ikejiri, Setagaya-ku, Tokyo 154-8566, Japan. Telephone and fax 81-03-3411-0151, e-mail hsakuta@msa.biglobe.ne.jp H Sakuta, T Suzuki, H Yasuda, T Ito. Vital capacity and selected metabolic diseases in middle-aged Japanese men. Can Respir J 2006;13(2):79-82.

D ecreased forced vital capacity (FVC) is associated with increased mortality (1).Low levels of FVC or static vital capacity (VC) predict myocardial infarction independently of heavy smoking, productive cough, exertional dyspnea and cardiac enlargement (2,3).A longitudinal study (4) has shown that VC inversely correlates with leg and carotid atherosclerosis.These associations may be explained by the relationship between FVC or VC and metabolic diseases or factors, including diabetes (5-7), hypertension (8,9), triglyceride levels (6,10,11), gamma (γ)-glutamyl transferase levels (12,13) and markers of inflammation such as white blood cell (WBC) count (14), at least in part.However, the association between metabolic disease and FVC or VC has not been well elucidated in an Asian population.It is not known whether total homocysteine, an independent risk factor for cardiovascular disease (15), correlates with FVC or VC.In the present study, we analyzed the associations among FVC, VC as a percentage of that predicted (%VC) and the presence of selected metabolic disorders in middle-aged Japanese men who belong to the Japan Self-Defense Forces.

METHODS
Men (n=954) who underwent retirement medical checkups were studied.Of 1000 men who had undergone a retirement medical checkup from July 1999 to December 2000, 974 men agreed to participate in the study.Twenty of these men were excluded because of missing data.Thus, a total of 954 men were examined.Only men were studied because the proportion of women was less than 1% in the setting.
Serum total cholesterol, triglycerides and γ-glutamyl transferase levels, as well as fasting plasma glucose levels, were measured.
©2006 Pulsus Group Inc.All rights reserved

ORIGINAL ARTICLE
An oral 75 g glucose tolerance test was performed.For the measurement of total homocysteine, fasting blood samples were collected in an EDTA tube in the morning, and the plasma was separated within 1 h after sample collection and stored at -80°C.Plasma total homocysteine was measured by using a fluorescence polarization immunoassay kit (Abbott IMx Homocysteine, Abbott Laboratories, USA) (16).Body mass index was calculated.VC and FVC were measured with an electronic spirometer (FUDAC-70, Fukuda Denshi Co Ltd, Japan).Predicted values of VC were obtained from linear regression analysis of age and height (17): Information about cigarette smoking status and daily number of cigarettes smoked, as well as physical activity, ethanol consumption, vegetable intake and medical history, including episodes of cardiovascular disease, was obtained from a selfadministered questionnaire.The participant's habit of cigarette smoking was categorized as current smoking, ex-smoking or lifelong nonsmoking.Current physical activity was presented as the average weekly frequency of exercise causing sweating.Current ethanol consumption (millilitres of ethanol per day) was calculated from the consumption volume of each type of alcoholic beverage.Current vegetable intake was classified into four categories: very rich intake, rich intake, common intake and poor intake.Each category was given a score of 4, 3, 2 or 1, respectively, which was designated as the vegetable intake score.Plasma folate, which is found in rich concentrations in vegetables and is regarded as a biomarker of vegetable intake (18), was measured with a chemiluminescent kit (Bayer HealthCare LLC, USA).The vegetable intake score was inversely associated with plasma levels of folate (found in vegetables) in the population studied (standard regression coefficient = 0.082; P=0.011).
Hypercholesterolemia was defined as a serum total cholesterol concentration of more than 5.69 mmol/L or receiving medication for hypercholesterolemia. Hypertriglyceridemia was defined as a serum triglyceride concentration of more than 2.24 mmol/L or receiving medication for hypertriglyceridemia.A peripheral WBC count of 6.9×10 9 /L or more (upper 20th percentile of the studied population) was defined as a high WBC count.Type II diabetes was diagnosed based on the criteria of the World Health Organization (19).Persons who showed a 'diabetic pattern' in the tolerance test had diabetes diagnosed.The status of diabetes was determined by the self-reporting of physician-diagnosed diabetes or by an oral 75 g glucose tolerance test.No participants were subject to insulin therapy at the time of admission.Cut-off levels of plasma total homocysteine and serum γ-glutamyl transferase were 14 µmol/L (15) and 70 U/L, respectively.Hypertension was diagnosed based on individual medical history or having received medication for hypertension (blood pressure of at least 140/90 mmHg) based on the criteria of the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (20).Blood pressure was measured with a mercury sphygmomanometer after the participant had been at rest in a sitting position.The participants gave informed consent.The study protocol was approved by the Ethics Committee of the Self-Defense Forces Central Hospital, Tokyo, Japan.
A linear regression analysis was performed with FVC or %VC as the dependent variable, and with lifestyle factors and metabolic parameters as the independent variables.A logistic regression analysis was also performed to obtain odds ratio per 0.54 L (1 SD) decrease in FVC or per 13% (1 SD) decrease in %VC for the presence of the selected metabolic disorders.Statistical analysis was performed using StatView version 5.0 (SAS Institute, USA).

RESULTS
Characteristics of the study population are shown in Table 1.Of lifestyle factors studied, vegetable intake was positively associated with FVC and %VC (Table 2).The daily number of cigarettes smoked and body mass index were inversely associated with FVC and %VC.Neither frequency of sweating exercise nor daily ethanol consumption was associated with FVC or %VC.With regard to metabolic parameters, systolic blood pressure, triglyceride levels and WBC count, but not total homocysteine, γ-glutamyl transferase or total cholesterol levels, correlated with FVC and %VC in a linear regression analysis model adjusted for physical activity, daily ethanol consumption, vegetable consumption, cigarette smoking, body mass index and age.Fasting plasma glucose levels correlated with %VC but not with FVC.
In a logistic regression analysis model adjusted for the abovementioned lifestyle factors, as well as body mass index and age, a decrease of 1 SD of FVC (0.54 L) or a decrease of 1 SD of %VC (13%) was associated with the presence of type II diabetes, hypertension, hypertriglyceridemia, high γ-glutamyl transferase levels and having had an episode of cardiovascular disease, but not with the presence of hyperhomocysteinemia, hypercholesterolemia or a high WBC count (Table 3).
A stratified analysis was performed among lifelong nonsmokers (n=201) to lessen the possible effect of cigarette smoking (Table 4).From the stratification, the odds ratio per 0.54 L decrease in FVC or per 13% decrease in %VC for type II diabetes, hypertension and hypertriglyceridemia, but not for high γ-glutamyl transferase levels, increased.The association became not significant for hypertension or high γ-glutamyl transferase levels after the stratification.

DISCUSSION
It has been shown that FVC and VC predict future development of arterial atherosclerosis (21), diabetes and insulin resistance (5)(6)(7)(22)(23)(24), and hypertension (8,9).In baseline cross-sectional studies (6,10), triglyceride concentrations have been shown to inversely correlate with FVC or VC.Consistent with these previous reports, the present study showed that both FVC and %VC were inversely associated with the presence of type II diabetes, hypertriglyceridemia and hypertension (Table 3).
The mechanism of the association is not clear, although lifestyle factors may contribute to it.Cigarette smoking (10), physical inactivity (25), poor vegetable intake (26), ethanol consumption (11) and obesity are known to inversely correlate with FVC and %VC.In the present study, however, the association between metabolic diseases and VC was independent of these lifestyle factors and body mass index (Table 3).
Another possible mechanism of the association is insulin resistance (6,10).Waist to hip circumference ratio, which is known to correlate with insulin resistance, has been shown to correlate with FVC independently of body mass index (27,28).Obesity, diabetes and hypertension, which are also known to correlate with insulin resistance, inversely correlated with FVC and %VC in the present study, as reported by others (5)(6)(7)(8)(9)29).It has also been shown that a decrease in FVC or %VC correlates with insulin resistance (6,23,24).However, insulin resistance may not be the cause of decreased VC because it has been shown that a decrease in VC precedes insulin resistance (23).In this context, it is speculated that early-life factors can affect lung growth and also program insulin resistance (6,24).Serum γ-glutamyl transferase activity, a marker of oxidative stress (13,30) and also a risk factor for cardiovascular disease (12), predicts cardiovascular disease (12) and diabetes (31).In the present study, high γ-glutamyl transferase levels were inversely associated with FVC and %VC.Although the association between FVC and γ-glutamyl transferase levels has previously been reported, the activity of the enzyme is regarded as a marker of ethanol consumption in that report (11).Thus, as far as we know, the present report is the first to show that γ-glutamyl transferase levels correlate with FVC or %VC independently of ethanol consumption.Because total homocysteine did not correlate with FVC or %VC, total homocysteine does not seem to mediate the association between lung function and cardiovascular disease.

CONCLUSIONS
Decreased FVC and %VC was associated with the presence of type II diabetes, hypertension, hypertriglyceridemia and high γ-glutamyl transferase levels in middle-aged Japanese men.
Vital capacity and metabolic disease Can Respir J Vol 13 No 2 March 2006 81

TABLE 1
Characteristics of the study population (n=954)

TABLE 2
*Analysis of metabolic parameters was adjusted for age, physical activity, daily ethanol consumption, vegetable consumption, daily number of cigarettes smoked and body mass index.β Standard regression coefficient

TABLE 3
Odds ratios for the presence of metabolic disease per 0.54 L (1 SD) decrease in forced vital capacity (FVC) and per 13% (1 SD) decrease in percentage of predicted vital capacity (%VC) for the presence of metabolic disease among all participants irrespective of smoking status (n=954) Analysis was adjusted for age, physical activity, daily ethanol consumption, vegetable consumption, daily number of cigarettes smoked and body mass index