| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the University of Western Ontario (Dr. Morgan), London, Ontario, Canada; and Alderson-Broaddus College (Dr. Reger), Phillipi, WV.
Correspondence to: Wm. Keith C. Morgan, MD, Department of Medicine, London Health Sciences Centre, University Campus, 339 Windermere Rd, London, Ontario, Canada N6A 5A5
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Objective: To carry out a prospective study of those factors influencing the rate of decline of the ventilatory capacity in a cohort of automobile workers.
Design: Prospective cohort study.
Setting: Southern Ontario, Canada.
Participants: A cohort of 181 workers employed in assembling and spray painting the chassis of new cars, a minority of whom used paints containing isocyanates.
Measurements: All participants underwent annual anthropometric measurements. Spirometry was carried out at yearly intervals, and a questionnaire relating to respiratory symptoms and smoking habits was completed annually by all participants. Daily monitoring of the isocyanate levels was carried out.
Results: There was no indication of any effect from isocyanate exposure. The annual decline in the FEV1 was similar to that found in other studies, with the respective annual decrements for smokers, ex-smokers, and nonsmokers being 0.055 L, 0.046 L, and 0.035 L, respectively. The decline of the FEV1 in those > 35 years old and < 35 years old differed appreciably. The decrements in the FEV1 in subjects < 35 years old were influenced as much by excessive weight gain as by cigarette smoking. Loss of weight in those significantly overweight was frequently associated with improved lung function.
Conclusions: While age and smoking play an important role in determining the rate of decline in the ventilatory capacity, it is clear that body weight plays a significant role and needs to be taken into account in all epidemiologic studies of the ventilatory capacity.
Key Words: decline of ventilatory capacity • FEV1 • smoking • weight gain
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The ventilatory capacity was measured using a Vitalogram spirometer (Vitalograph Limited; Buckingham, UK). At least three and up to five to six forced expiratory volume maneuvers were performed by a nurse who had been trained to carry out spirometry. The two largest FEV1 measurements and the FVC had to be within 100 mL of each other to be accepted as valid. They also had to conform to the other American Thoracic Society recommendations.1 2 3 All tracings were visually inspected within 2 weeks of being carried out by one of us (WKCM). Approximately 1 in 20 tracings did not meet the criteria noted above, and the tracings were subsequently repeated. The instrument was calibrated on each occasion it was used. Unfortunately, the Vitalogram spirometer only records for 12 s, and it became evident that approximately 50% of those who had airways obstruction had not reached a plateau at 12 s. Thus, it was not possible to measure the FVC accurately in those in whom the measurements were most important (ie, those with obstruction). The FEV1/FVC% was overly high in 13% of subjects and was therefore omitted from our calculations. We relied on the FEV1 and the characteristic wave form of the time vs volume curve to indicate whether obstruction was present or not.3 Many of the men were tall and had large frames, with around 15% > 180 cm in height and several > 198 cm in height. When performing spirometry, we have noticed that taller men take from 12 to 16 s to exhale their FVC completely in the absence of detectable airways obstruction. All subjects but one who had been identified as having obstructive impairment from the FEV1 value and a characteristic wave form had an abnormal FEV1/FVC%.
At the start of the surveillance, all workers had a physical
examination carried out by the company physician. Questions as to
general health and history were asked, with specific attention to
respiratory problems and smoking habits. Subsequently, each subject was
asked to complete a standardized questionnaire when undergoing annual
spirometry. This included questions in regard to chest symptoms, with
specific reference to possible symptoms of any isocyanate-induced
condition. Detailed questions were asked in regard to smoking,
including the age at which the subject started smoking, how many
cigarettes or cigars he smoked, if he was a pipe smoker, and his weekly
consumption of tobacco. If the worker had stopped smoking, he was asked
when he had stopped and for how many years he had smoked. Each subject
was weighed and had his height measured annually. Only those subjects
who had been observed for > 3 years and who had valid spirometry were
included. The majority of subjects had been observed for 
6 years.
The decline in FEV1 was based on the first and
last reliable measurements of the FEV1.
The questionnaires were checked routinely and smoking histories were compared with prior statements. We assumed that the smoking history obtained from the first questionnaire was the most reliable. Some of the workers temporarily stopped smoking but resumed later. Thus, there were transient ex-smokers, who had become smokers again when seen on the next occasion. Five percent of the cohort at some time gave contradictory smoking data that were inaccurate and did not coincide with the data obtained at the preemployment examination and after the original spirometric studies. In such cases, the validity of the smoking data was checked by other means. Most of the subjects had undergone annual spirometry, but in a few instances (about 5%), a participant had been working elsewhere or was on holiday; as a result, there may have been a 2- to 3-year gap between certain observations, but those involved were a minority.
Ninety-five percent confidence intervals are included in the appropriate Tables. Although Canada, in theory, has embraced the metric system, all measurements other than dynamic lung volumes were recorded in imperial units and then converted. Data are included in both sets of units.
While the data in Tables 1 and 2 are mostly descriptive, two-way contrasts in Table 3 were tested using simple t tests. Since multiple testing on the same subgroups existed, Chebychev’s inequality procedure was used as well. Contrasts in Table 4 were tested using an analysis of variance with Duncan’s multiple range test. All statistical testing was performed at the 95% confidence level.
|
|
|
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The mean height and the weight of the cohort were significantly higher than those of the general population (Table 1) . The mean heights of the groups in the cohort were as follows: smokers, 179.5 cm (70.66 inches); ex-smokers, 178.9 cm (70.45 inches); and nonsmokers, 176.7 cm (69.68 inches). Thus, the smokers were the tallest and the nonsmokers were the least tall. The initial mean weights of the three groups were 91.21, 89.12, and 90.72 kg, respectively. However, the disparities in mean height were small, with barely an inch separating the groups. According to the Metropolitan Life Insurance tables, the average weight for smokers should be 71 kg (range, 65 to 80 kg); that for ex-smokers should be 69.4 kg (range, 54 to 79 kg); and that for nonsmokers, 68 kg (range, 62 to 77 kg). Thus, all three groups were overweight. Although the mean height of each group in the cohort was also greater than that of the average white man in North America, the increase in height was appreciably less than the comparable excess weight.
The annual weight gain was least in the cigarette smokers and greatest in the ex-smokers, with the nonsmokers falling between. In this context, the nonsmokers gained appreciably more weight than the smokers, despite being 2.8-cm shorter (Table 1) . The mean years of observation were similar (Table 1) . The distribution of weight according to the approximate age (ie, year of birth) is shown (Table 2) . This surrogate indicates that the ex-smokers are on average a couple of years older than the smokers, while the nonsmokers are 4 to 6 years younger than the other two groups.
|
The rate of decline in the FEV1 (Table 1) is similar to that found in other studies.4 5 6 7 The mean annual decrement for smokers was 0.055 L, for ex-smokers 0.046 L, and for nonsmokers 0.035 L. Among the smokers, eight lost > 0.1 L annually; however, there was a wide variation, with the greatest annual decrement being 0.154 L and the least being 0.006 L. The latter subject was only 26 years old at the start of the study. Hence, it was likely that had he not been a smoker, his FEV1 would have shown a small increase. In several of the nonsmoking subjects < 30 years old, there was a small annual increment in the FEV1, but in other nonsmokers, there was a significant decrement. Further reference will be made to this later.
Table 3 shows the annual decrement in the FEV1 according to smoking status and weight. The greatest decline was in those smokers who weighed between 68.2 kg and 90.5 kg (150 to 199 lb). Within this subgroup, the decrease in FEV1 was significantly greater than for nonsmokers (- 0.058 L vs - 0.029 L). The contrast between ex-smokers and nonsmokers was nearly as severe (- 0.043 vs - 0.029 L) but failed to reach statistical significance. The rate of decline of the FEV1 decreased in the smokers with the heaviest weights, mainly because many lost weight over the period of observation. This also applied to the ex-smokers (Table 3) ; however, among the nonsmokers, the decrease in FEV1 with increasing weight was obvious, marked, and highly significant, with workers > 200 lb showing an 0.066 L decrease, compared to less than half the loss for those < 200 lb. Of the nine workers weighing > 250 lb, seven lost weight between the initial and final examinations, while two gained weight. Four of the nine workers who were > 250 lb showed a large decrease in weight over the period of observation (from 395 to 336 lb, 268 to 196 lb, 267 to 176 lb, and 315 to 268 lb, respectively). The corresponding annual changes in the FEV1 were - 0.080 L (smoker), + 0.015 L (ex-smoker), + 0.73 L (smoker), and + 0.035 L (smoker). The last subject’s FEV1 increased by 0.210 L between the ages of 43 years and 49 years. The only subjects > 30 years old whose FEV1 increased over the period of observation were those who lost > 10 kg in weight. In most smokers, the FEV1 declined despite the subject losing weight, but most of these subjects were heavy smokers.
Table 4 shows the annual decline in the FEV1 according to those who lost or gained weight over the observation period. No matter the smoking habit, the sample of workers who lost weight showed a smaller annual decrement in the FEV1 than those who gained. The implied weight effect on ventilatory capacity necessitates a caveat in view of the fact that only the contrast involving ex-smokers was statistically significant. This phenomenon results from small numbers of observations in certain cells and a large variance, both of which contribute to decreased sensitivity in the statistical tests. Within the weight-classified subgroups who were losing or gaining weight, contrasts involving decreases in the FEV1 across the subgroups according to smoking status were largely unremarkable. However, for both weight classifications, the FEV1 for smokers decreased substantially more than that for nonsmokers, but did not reach statistical significance at the 0.05 level.
There were 20 subjects born on or after January 1, 1960, included in
the study who were < 30 years old at the time of the first set of
lung function tests. Of these, nine were smokers, three were
ex-smokers, and eight were nonsmokers. There is good evidence that the
FEV1 continues to increase up to the age of 30
years. Therefore, it was believed desirable to ascertain whether the
changes in the younger subjects differed from those whose lung function
tests were carried out after attaining the age of 30
years.4
5
6
7
Those smokers < 30 years old at the start of
the study showed a significant annual decrease in the
FEV1, namely 41.5 mL. The mean age of the nine
smokers who were < 30 years old was 27.2 years. At the end of the
period of observation, seven of the nine were 35 years old, with
the remaining two subjects being 34 years old. In contrast and somewhat
surprisingly, the nonsmokers < 30 years old showed a
higher-than-expected mean annual decrease of 34 mL. An explanation,
however, was available, in that one of the nonsmoking subjects had
gained 70 lb (32 kg) over the period of observation (ie, 7.4
lb [3.36 kg]/yr), while another had gained 19 lb (8.6 kg;
ie, 3.2 lb [1.45 kg]/yr), with most of the weight being
added in the last 2 years. In addition, both subjects were 3 cm shorter
than the mean height of the nonsmokers. All of the nonsmokers bar one
were > 30 years old by the end of the observation period, with three
being > 35 years old. Were the two grossly overweight subjects
excluded, the annual decline for the remaining nonsmokers who were
initially < 35 years old was 11 mL.
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Fletcher and coworkers,8 in their seminal survey of chronic airflow limitation, delineated the effects of cigarette smoking on ventilatory capacity. Among their findings was the fact that the presence of chronic bronchitis in their group of workers showed no independent influence on the decline of the FEV1. The cohort selected were men aged from 30 to 59 years "since younger men were thought unlikely to have developed airflow obstruction by this age." Though numerous studies have, for the most part, confirmed their observations, it has become apparent that (1) the exclusion of workers < 30 years old led to some misleading conclusions as to the age of onset of the decline of the FEV1, (2) that chronic bronchitis has an effect on the ventilatory capacity, and (3) it has become apparent that factors other than smoking, such as weight gain, also play a significant role in the decline of the FEV1.
It has been shown that the FEV1 of nonsmokers
usually slowly increases up to the age of 30 years and then remains
relatively unchanged until age 35, at which time it starts to decline
gradually at approximately 20 to 30 mL/yr.4
5
6
7
The
situation in smokers is different, in that between the ages of 25 years
and 30 years, the FEV1 remains relatively
unchanged or increases appreciably less than it does in nonsmokers. At
the age of 30 years, the FEV1 of most smokers
starts to decline and gradually speeds up, when compared to that of
nonsmokers. A second and more progressive decline sets in at
approximately 40 to 45 years of age. These early effects were not
apparent in the study by Fletcher et al8
for reasons given
earlier. When Fletcher and colleagues started their study, it seemed
logical to exclude young persons, since it was known that chronic
airflow limitation evolved slowly and did not exert its main effects
until the age of 45 years.
Airways obstruction and a decrease in the FEV1 have been observed to occur in early adult life, not only in smokers but also those exposed to dust and other irritants.9 10 11 12 The early decline in the FEV1 cannot be accounted for by emphysema or small airways disease. Emphysema gradually evolves, and although slight changes have been observed at postmortem examination in subjects in their 30s, the extent is strictly limited and does not exert any effect on the ventilatory capacity in subjects < 40 years old.13 14 In this connection, there is compelling evidence to indicate that the FEV1 is not affected until at least 25% of the lungs are involved.13 Similarly, small airways disease cannot explain the decline, since it requires obliteration of more than one third of the small airways present in the lungs to affect the FEV1, because of the almost negligible airways resistance located in these airways.15 The resistance to flow in the large airways is much greater, because the total cross-sectional area of their lumina is smaller than that of the small airways. As a result, the flow rates in the large airways are much more affected by relatively minor narrowing and by turbulent flow. Moreover, both emphysema and small airways disease, once established, progress gradually, but seldom cause breathlessness before the age of 40 to 45 years, at which time a gradual reduction in the ventilatory capacity becomes apparent. Nonetheless, it is often difficult differentiating the effects of emphysema and chronic bronchitis. Thus, another explanation must be sought to explain the early airways obstruction noted in young cigarette smokers and in those nonsmokers exposed to dust and other irritants. The onset of chronic bronchitis in cigarette smokers and in those exposed to dust and other irritants occurs early in adult life (in the first 5 to 10 years after starting smoking or exposure to dusts).4 5 6 7 Furthermore, there is compelling evidence that chronic bronchitis by itself will produce airways obstruction, albeit mild and often only evident by comparing large numbers of bronchitic subjects with suitable referents.11 12 16 17 It is not suggested that the small degree of luminal encroachment caused by enlarged mucous glands produces the minor changes in flow; it seems much more probable that the changes in flow are a result of the turbulence that develops as a result of the mucus that is adhering to the walls of the bronchi.
If the effects of weight are taken into account, our observations in this investigation are in accord with those of Tager et al4 and Camilli et al.6 The small number of nonsmokers who were < 30 years old at the start of the study included two subjects who became grossly obese during the period of observation and a third who was moderately overweight throughout the study.
Turning now to the consideration of the effects of weight, it has been recognized for many years that obesity causes ventilation perfusion mismatching and a reduction of the PaO2.18 However, predicted lung volumes have not taken into account excessive weight, mainly because it has been assumed that there is only a small effect on the FEV1 and FVC, unless the subject is moderately or grossly obese. Recently, a number of well-designed studies have shown that even mild-to-moderate obesity influences not only the alveolar-arterial gradient for oxygen, but the PaO2, the diffusing capacity, and also dynamic lung volumes.19 20 The effects of weight are clearly evident in our study, with excessive gains frequently being associated with an increased decrement in the FEV1. Conversely significant increments in the FEV1 occurred in several subjects who had an appreciable loss of weight. It is suggested that in these days when more and more North Americans are becoming increasingly corpulent, it will become essential to take weight into account when predicting lung volumes. Moreover, when comparing the mean FEV1 and FVC of the study and control subjects, disparities in weight and other factors can easily lead to spurious conclusions. Thus, the minor difference in the FVC between those with pleural plaques and control subjects noted by Bourbeau et al21 could easily be accounted for by the difference in weight between the study group and the control subjects. Furthermore, the belief that multiple regression analyses will accurately apportion the contributions of smoking, increasing age, and other colinear factors has been shown to be unjustified.22
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Received for publication October 29, 1999. Accepted for publication June 2, 2000.
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M Skogstad, K Kjaerheim, G Fladseth, M Gjolstad, H L Daae, R Olsen, P Molander, and D G Ellingsen Cross shift changes in lung function among bar and restaurant workers before and after implementation of a smoking ban Occup. Environ. Med., July 1, 2006; 63(7): 482 - 487. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S.-m. Ip, F. W.-s. Ko, A. C.-w. Lau, W.-c. Yu, K.-s. Tang, K. Choo, M. M.-w. Chan-Yeung, and on Behalf of the Hong Kong Thoracic Society and Am Updated spirometric reference values for adult chinese in Hong Kong and implications on clinical utilization. Chest, February 1, 2006; 129(2): 384 - 392. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. D. Goldman Lung Dysfunction in Diabetes Diabetes Care, June 1, 2003; 26(6): 1915 - 1918. [Full Text] [PDF]
|
|
|
|
 |
|
 R. E. Walter, A. Beiser, R. J. Givelber, G. T. O'Connor, and D. J. Gottlieb Association between Glycemic State and Lung Function: The Framingham Heart Study Am. J. Respir. Crit. Care Med., March 15, 2003; 167(6): 911 - 916. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Bottai, F. Pistelli, F. Di Pede, L. Carrozzi, S. Baldacci, G. Matteelli, A. Scognamiglio, and G. Viegi Longitudinal changes of body mass index, spirometry and diffusion in a general population Eur. Respir. J., September 1, 2002; 20(3): 665 - 673. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Reich Improved Survival and Higher Mortality* : The Conundrum of Lung Cancer Screening Chest, July 1, 2002; 122(1): 329 - 337. [Full Text] [PDF]
|
|
|
|
 |
|
 D AHMAD, W K C MORGAN, and L SCHACHTER Obesity and lung function Thorax, September 1, 2001; 56(9): 740c - 741. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
