Abstract
Little is known about factors determining the outcome of childhood asthma. The purpose of this longitudinal study was to assess the factors in childhood that determine the level of FEV1 in early adulthood in asthmatic individuals, and to examine factors associated with decline in FEV1 during adulthood. Between 1966 and 1969, 119 allergic asthmatic subjects aged 5 to 14 yr were studied (Visit 1). Of these subjects, 101 (85%) were reinvestigated at ages 22 to 32 yr (Visit 2) and 32 to 42 yr (Visit 3). At the first survey and during follow-up, a standardized questionnaire was used, serum total IgE and peripheral blood eosinophils were measured, and physical examination, skin tests, lung function tests, and histamine challenge (provocative concentration causing a 10% decline in FEV1; PC10) tests were performed according to the same protocol. Multiple linear regression analyses were performed with FEV1 at Visit 2 and with the change of FEV1 from Visit 2 to Visit 3 as outcome variables. A low FEV1% predicted at Visit 1 and PC10 ⩽ 16 mg/ml at Visit 1 were significantly associated with a lower level of FEV1 at Visit 2. Subjects who quit smoking and subjects who continued to use inhaled corticosteroids had a significantly smaller annual decline in FEV1 from Visit 2 to Visit 3, adjusted for attained level of FEV1 at Visit 2. In conclusion, bronchial hyperresponsiveness and a low level of lung function in childhood are independent risk factors for a low level of FEV1 in early adulthood. A smaller decline in FEV1 after ages 22 to 32 yr occurs in asthmatics who quit smoking and who continue to use inhaled corticosteroids. Our data stress the importance of studying intervention strategies for asthma in young childhood and early adulthood in order to prevent or postpone further lung function deficits. Grol MH, Gerritsen J, Vonk JM, Schouten JP, Koëter GH, Rijcken B, Postma DS. Risk factors for growth and decline of lung function in asthmatic individuals up to age 42 years: a 30-year follow-up study.
Many published epidemiologic studies on age-related growth and decline of lung function have as their primary aim the natural history of FEV1 and FVC throughout childhood and adulthood in healthy subjects (1-9). These studies generally show that a growth in FEV1 during childhood is followed by a stable phase from adolescence through early adulthood and a decline in FEV1 after the age of 32 yr (4, 5). Both the obtained maximal level of FEV1 and the rate of decline of FEV1 determine the severity of lung function impairment at older age in symptomatic persons. Therefore, it is important to know which factors are associated with the level of FEV1 in early adulthood and which factors are associated with an accelerated decline of FEV1 throughout adulthood. Most studies of the growth of FEV1 in childhood have shown that lower respiratory tract infections and passive and active smoking by healthy children and adolescents are associated with smaller growth of lung function and a lower maximally attained level of lung function (10-14). Moreover, passive smoking affects lung growth to a greater extent in girls than in boys (10). One study has shown that exposure to indoor allergens is associated with decrements in FEV1 in symptomatic children (13).
Studies of risk factors for accelerated decline in FEV1 have generally focused on the effect of smoking on lung function, and have involved subjects with chronic obstructive pulmonary disease (COPD) or subjects from the general population. Two longitudinal studies, one of middle-aged and older men with no history of asthma, the other of subjects aged 65 yr or older, found that bronchial hyperresponsiveness (BHR) and atopy were both independent predictors of decline of lung function (15, 16). Another longitudinal population study (the Vlagtwedde–Vlaardingen study) showed that increased BHR leads to an accelerated decline in FEV1 (17). All of these studies began their measurements during adulthood. Very few studies of decline in lung function in asthmatic subjects have been reported. Peat and coworkers found that a group of 92 asthmatic subjects aged 22 to 69 yr had lower baseline lung function and a greater rate of decline in lung function than did 186 healthy control subjects (18). Lange and associates found a greater decline of FEV1 in people who identified themselves as having asthma than in a sample of the general population (19). However, to assess the causal relationship between risk factors and growth and decline of lung function in asthmatic subjects, longitudinal studies from early childhood to adulthood are needed.
In this study we investigated a group of 119 allergic asthmatic patients selected from the outpatient department of the University Hospital of Groningen over a period of 30 yr. The first survey started in 1966, when the patients were 5 to 14 yr old, and follow-up studies were performed from 1983 to 1986 (Visit 2) and in 1995 and 1996 (Visit 3). At Visit 2 the subjects were 22 to 32 yr of age, which is the age range in which the plateau phase of lung function has been described (4, 5). We assessed the childhood factors that were associated with the level of FEV1 in early adulthood in the study and the factors associated with decline in lung function from ages 22 to 32 yr to ages 32 to 42 yr (early adulthood), taking into account known risk factors identified in previous studies.
We reexamined a cohort of asthmatic patients who were admitted to the outpatient clinic of the Pediatric Pulmonary Department of the University Hospital of Groningen and who had participated in a study performed between 1966 and 1969 (Visit 1). The design of that study was reported previously (20). The follow-up data were collected between 1983 and 1986 (Visit 2), and most recently in 1995 and 1996 (Visit 3). The study was approved by the medical ethics committee of the University Hospital of Groningen, and all participants gave written informed consent.
At Visit 1, 119 asthmatic children from the outpatient department (113 allergic to house dust), aged 5 to 14 yr, were included in order to determine the relationship between house dust allergy and BHR by means of a house dust test and a histamine provocation test (21). The patients lived in the northern region of the Netherlands. They were included if they had physician-diagnosed asthma, if the parents gave informed consent for a 5-d stay of the patient in the hospital (Pediatric Clinic of the University Hospital Groningen), if the patient was able to perform technically satisfactory lung function tests, and if the patient's asthma was in a stable condition. Exclusion criteria were the presence of specific respiratory diseases such as cystic fibrosis and tuberculosis, identified by a physician, and the presence of other seriously interfering diseases. All therapy was withheld at least 24 h before measurements were made. None of the subjects used oral corticosteroids on a regular basis. Inhaled corticosteroids or cromoglycate were not available at the time of Visit 1.
Between 1983 and 1986, 101 of the 119 subjects (85%) were reevaluated. Sixteen refused to participate and two were lost to follow-up. In 1995 and 1996, 101 of the 119 subjects (85%) were reinvestigated. All subjects were interviewed, and 95 (80%) underwent lung function and skin tests at Visit 3. Eleven subjects refused to participate, one could not participate because of pregnancy, four were lost to follow-up, and two had moved abroad. We analyzed data for 116 of the subjects at Visit 1 instead of 119, because data for two children were missing and one was excluded from the analysis because he appeared to suffer from α1-antitrypsin deficiency. In the initial and the two follow-up studies, measurements conformed to the same protocol as used at Visit 1.
The Dutch version of the British Medical Research Council's standard questionnaire was used, which is comparable to the European Coal and Steel Community's questionnaire (22, 23). During the last visit, the European Community Respiratory Health Study questionnaire was applied as well (24), including additional questions on pets, medication (type and duration), passive smoking, and housing. At all three visits, subjects were interviewed by a trained physician.
In the study data analyses, subjects were regarded as symptomatic if they had wheezing and/or asthma attacks. Nonsmokers were those who had never smoked. Ex-smokers were those who had stopped smoking at least 1 mo before examination. Current cigarette smokers were defined as those who smoked one or more cigarettes a day. Pack-years of cigarettes were calculated as the number of packs of cigarettes a day (20 cigarettes per pack) multiplied by the number of years of smoking.
Intracutaneous skin tests were performed with allergen extracts for house dust, grass pollens, tree pollens, weeds, animal dander, and feather (Diephuis Laboratory, Groningen, The Netherlands). Histamine was used as a positive control and phosphate buffer was used as a negative control. At all three visits, skin tests were considered positive if the diameter of the wheal was at least 5 mm (20). At the last two visits, identical allergen-extract batches were used. Blood eosinophils were counted in a Bürker counting chamber (Scherf; Cecchinato, Venice, Italy). At Visits 2 and 3, serum total and specific IgE (IU/L) were measured with a solid-phase immunoassay (Pharmacia IgE EIA; Pharmacia Diagnostics A, Uppsala, Sweden). This test was not available at Visit 1.
At Visits 2 and 3, inhaled, short-acting β2-agonists, anticholinergic agents, and cromoglycates were stopped at least 8 h before testing; theophylline and oral antihistamines were stopped at least 24 h and long-acting β2-agonists at least 48 h before testing. The use of oral and/or inhaled corticosteroids was continued. Subjects had to be in a stable state without airway infection in the last 2 wk preceding testing. Only one patient used oral corticosteroids; for analytical purposes, this patient was added to the group using inhaled corticosteroids.
Measurements of lung function were made with a water-sealed spirometer (Lode Spirograph Type DL; Lode Instruments, Groningen, The Netherlands). Two valid measurements of FEV1 and slow inspiratory vital capacity (IVC) were obtained, the highest value being recorded. Predicted values used at Visit 1 were those of Zapletal and colleagues (25), the values used at the two follow-up visits were those of the European Coal and Steel Community (26). The measurements were made in the same month as at Visit 1, plus or minus 1 mo.
The method of Tiffeneau as modified by de Vries and coworkers (27) and by Knol (28) was used for the histamine challenge test (29). At the first visit, subjects inhaled nebulized distilled water from a Wiesbaden Doppel inhalator device after baseline measurements of pulmonary function. Phosphate-buffered saline was used at the second and third visits. If there was a decrease in FEV1 of 10% or more, the test was stopped. Sequential aerosols of histamine biphosphate in concentrations of 0.25, 0.50, 1, 2, 4, 8, 16, and 32 mg/ml were inhaled for 30 s. Two FEV1 maneuvers were performed after each challenge, the highest being recorded. The histamine concentration provoking a decrease in FEV1 of 10% or more from baseline (PC10) was taken as the threshold value. The test was terminated when the threshold value was reached or when the highest concentration of histamine had been given. In 1966, at the time of Visit 1, the PC10 challenge protocol of de Vries and coworkers (27) was a standard method. To ensure comparability, this method was used again at the later visits. No test was performed in subjects with an FEV1 < 1.5 L.
Subjects were considered hyperreponsive if they had a PC10 ⩽ 16 mg/ml, a value comparable with PC20 ⩽ 8 mg/ml according to the 2-min inhalation method of Hargreave.
All analyses were done with the statistical package SPSS/PC (version 5.0.2) (SSPC, Inc., Chicago, IL). Variables were checked for normal distributions. To normalize the distributions, values of serum total IgE, blood eosinophils, and histamine PC10 were log-transformed. In the tables these variables are presented with geometric means and percent SDs. Normally distributed variables are presented as their arithmetic means and SDs, age is presented as its median and range.
To investigate which covariables in childhood were associated with the level of FEV1 (ml) at age 22 to 32 yr, we performed a multiple linear regression analysis. Covariables included in this model were: FEV1% predicted at Visit 1, ln eosinophils (106/L) at Visit 1, having a pet at the time of Visit 1, passive smoking at Visit 1, BHR at Visit 1 (PC10 ⩽ 16 mg/ml), number of positive skin tests at Visit 1, gender (male = 1, female = 0), smoking habits (pack-years) at Visit 2, age (in years), and height (in centimeters) at Visit 2.
Another multiple linear regression analysis was done on the mean annual change in FEV1 (ml/yr) between Visit 2 and Visit 3, in order to investigate which of the following covariables were associated with an accelerated decline in FEV1: gender (male = 1, female = 0), use of inhaled corticosteroids at Visits 2 and 3, use of inhaled corticosteroids at only one visit (Visit 2 or 3), ln eosinphils (106 · 1−1) at Visit 2, log serum total IgE (IU/L) at Visit 2, BHR at Visit 2 (PC10 ⩽ 16 mg/ml), age (in years), wheezing and/or asthma attacks at Visit 2, number of positive skin tests at Visit 2, smoking habit at Visits 2 and 3 (continuous smokers, ex-smokers at Visits 2 and 3, having started smoking by the time of Visit 3, and having stopped smoking at the time of Visit 3). Including FEV1 at Visit 2 in the model would have led to regression to the mean, and because of the interdependence of all other covariables with FEV1 at Visit 2 and with each other, this might have led to biased estimates (17, 30). To avoid this bias, we used the residuals derived from a regression of FEV1 at Visit 2 on the covariables of interest as an attained-level variable, instead of the attained level of FEV1 at Visit 2. With the use of these residuals of FEV1, this level variable is determined only by random errors and possibly by other factors not included in the model. This procedure does not prevent bias in the estimated coefficient of the attained-level variable caused by the regression-to-the-mean phenomenon, but it does not lead to bias in the estimates of the effects of other covariables.
The characteristics of subjects studied and those lost to follow-up are summarized in Table 1. More females than males were nonparticipants. At Visit 1, no significant differences existed between the studied and nonstudied groups in mean age, lung function, number of eosinophils, percentage of children with a histamine PC10 ⩽ 16 mg/ml, atopy, or symptoms.
| Visit 1 | Visit 2 | Visit 3 | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| No Follow-up Visits (n = 10) | Only One Follow-up Visit (1983 or 1995) (n = 12) | Two Follow-up Visits (n = 94) | ||||||||
| (n = 100) | (n = 100) | |||||||||
| Age, yr, median (range) | 9 (8-14) | 10 (7-13) | 9 (5-14) | 26 (21-33) | 38 (32-42) | |||||
| Height, cm, mean (SD) | 139 (12) | 138 (9) | 141 (12) | 178 (9) | 178 (10) | |||||
| Sex, % male | 30 | 92* | 68* | 70 | 69 | |||||
| FEV1% predicted, mean (SD) | 82 (17) | 84 (12) | 82 (18) | 84 (15) | 86 (15) | |||||
| FEV1% VC, mean (SD) | 77 (10) | 78 (9) | 75 (12) | 72 (11) | 70 (10) | |||||
| Eosinophils, 106/L, geom · mean (%SD) | 267 (2) | 472 (2) | 315 (2) | 154 (2) | 171 (2) | |||||
| PC10 ⩽ 16 mg/ml, % | 80 | 100 | 81 | 30 | 52 | |||||
| Skintests, % positive | ||||||||||
| Housedust | 86 | 100 | 95 | 97 | 98 | |||||
| Animal dander | 57 | 46 | 73 | 96 | 98 | |||||
| Grasspollen | 14 | 46 | 32 | 85 | 59 | |||||
| Symptoms, % positive | ||||||||||
| Cough | 30 | 73 | 59 | 10 | 21 | |||||
| Phlegm | 20 | 46 | 30 | 10 | 14 | |||||
| Dyspnea | 90 | 82 | 81 | 9 | 8 | |||||
| Wheeze | 100 | 100 | 98 | 52 | 18 | |||||
| Asthma attacks | 80 | 100 | 80 | 39 | 21 | |||||
| Start symptoms, % | ||||||||||
| Before age 3 | 70 | 92 | 66 | 68 | 67 | |||||
| Smoking habits, % | ||||||||||
| Non/ex/current | 100/0/0 | 100/0/0 | 100/0/0 | 47/11/42 | 41/22/37 | |||||
Table 2 shows that a low level of FEV1% predicted at Visit 1 and BHR (PC10 ⩽ 16 mg/ml) at Visit 1 were independently and significantly associated with a lower level of FEV1 at Visit 2. Including absolute FEV1 (height adjusted at Visit 1 or FEV1/ VC at Visit 1, instead of FEV% predicted, did not change the significant association. When we repeated the analysis with log2PC10 at Visit 1, we achieved similar results (data not shown). Figure 1 shows the association between the level of FEV1 at Visit 2 and FEV1% predicted at Visit 1, expressed in tertiles. As compared with the highest tertiles, the lowest tertile of FEV1% predicted at Visit 1 was associated with a markedly lower level of FEV1 at Visit 2. To investigate whether the relationship between FEV1% predicted at Visit 1 and FEV1 at Visit 2 was a linear one, we used in S-plus-two models. In another one of these, FEV1% predicted at Visit 1 was entered as a linear variable, and in FEV1% predicted at Visit 1 was entered as a nonparametric smooth function (Loess). Analysis of variance showed no significant difference between the two models in goodness of fit. It can therefore be concluded that a linear approximation of the relationship between FEV1% predicted at Visit 1 and the FEV1 at Visit 2 is a valid one.
| B | CI | p Value | ||||
|---|---|---|---|---|---|---|
| FEV1% pred Visit 1 | 8.92 | 1.87 to 15.97 | 0.015 | |||
| Ln eos (106 · L−1) Visit 1 | −24.29 | −171.51 to 122.92 | 0.747 | |||
| Having pets Visit 1 | 128.26 | −183.12 to 439.64 | 0.422 | |||
| Passive smoking Visit 1 | −65.46 | −323.43 to 192.51 | 0.620 | |||
| BHR (PC10 ⩽ 16 mg/ml) Visit 1 | −378.96 | −697.76 to −60.16 | 0.022 | |||
| Total positive skin tests Visit 1 | 4.22 | −113.39 to 121.84 | 0.944 | |||
| Sex (male = 1, female = 0) | 269.51 | −99.57 to 638.58 | 0.156 | |||
| Pack years Visit 2 | 11.40 | −7.98 to 30.78 | 0.252 | |||
| Age (yr) Visit 2 | −53.26 | −109.45 to 2.93 | 0.067 | |||
| Height (cm) Visit 2 | 54.49 | 34.55 to 74.43 | 0.000 |
Fig. 1. Association between the level of FEV1 at Visit 2 (ages 22 to 32 yr) and FEV1% predicted, expressed in tertiles, at Visit 1 (ages 5 to 14 yr) (n = 93).
[More] [Minimize]Figure 2 shows that BHR at Visit 1 was associated with a lower level of FEV1 at Visit 2. We found no significant association of ln eosinophils at Visit 1, having pets at the time of Visit 1, passive smoking at Visit 1, number of positive skin tests at Visit 1, male gender, age, or height with the level of FEV1 at Visit 2 (Table 2). Number of pack-years of smoking at Visit 2 was not significantly associated with the level of FEV1 at Visit 2. Using current smoking and ex-smoking instead of pack-years at Visit 2 gave similar results (data not shown).
Fig. 2. Association between the level of FEV1 at Visit 2 (ages 22 to 32 yr) and the presence of BHR at Visit 1 (ages 5 to 14 yr), expressed as PC10 ⩽ 16 mg/ml (n = 93).
[More] [Minimize]To investigate whether FEV1 contains information about unmeasured factors (i.e., factors associated with FEV1 and which therefore may not have an independent effect), we performed an analysis of the level of FEV1 at ages 22 to 32 yr without adjustment for FEV1% predicted at Visit 1. This did not change the estimates for BHR, pack-years of smoking age, or height (data not shown). It seems that these variables had an independent effect in the regression model and did not exert some of their effects through FEV1% predicted at Visit 1. In the analysis with adjustment for FEV1% predicted at Visit 1, the estimates for ln (eosinophils), having pets, passive smoking, total positive skin tests, and gender were smaller than in the analysis without adjustment for FEV1% predicted at Visit 1. This indicates that these variables exerted some of their effects through FEV1% predicted at Visit 1. However, the only variable with a significant association with the level of FEV1 at Visit 2 in both analyses (regardless of FEV1% predicted at Visit 1) was BHR at Visit 1.
Subjects who quit smoking between Visits 2 and 3, and subjects who used inhaled corticosteroids on a regular basis between Visits 2 and 3, had a significantly smaller annual decline in FEV1 from Visit 2 to Visit 3 (both p = 0.026), independent of the attained level of FEV1 at Visit 2. This result was statistically significant, and was driven by two subjects. Removing these subjects from the analysis resulted in lower, nonsignificant estimates for the two covariables (14.2 for ex-smoking at Visit 3, and 18.7 for the use of corticosteroids at Visits 2 and 3). However, both estimates are positive, and the 95% confidence intervals (CIs) of these estimates (−15.5 to 44.0 for ex-smoking at Visit 3 and −27.7 to 45.0 for the use of corticosteroids at Visits 2 and 3) comprised the estimates of the analysis with the subjects included. Because both analyses (with or without these subjects) pointed in the same direction, we included the two outliers in the analysis. Subjects who used inhaled corticosteroids at only one visit (Visit 2 or Visit 3) had a borderline significant association with a steeper annual decline in FEV1 from Visit 2 to Visit 3 (p = 0.05). Having a high number of positive skin tests at Visit 2 had a borderline significant (p = 0.07) association with a smaller decline in FEV1. Male gender, ln eosinophils, log IgE, BHR, age, wheezing and/or asthma attacks at Visit 2, starting smoking by Visit 3, quitting smoking by Visits 2 and 3, and smoking were not significantly associated with a decline in FEV1 from Visit 2 to Visit 3. Including ln slope PC10 or log2PC10 at Visit 2, instead of BHR; including the highest tertile of eosinophils at Visit 2 instead of ln eosinophils; or including pack-years at Visit 3 instead of smoking habits, led to similar results (data not shown).
To investigate whether the residual derived from a regression of FEV1 on the other covariables contained information about unmeasured factors (i.e., factors associated with FEV1 and which therefore might not have had an independent effect), we performed an analysis of the change in FEV1 (in milliliters) between Visits 2 and 3 without adjustment for the residual of FEV1 at Visit 2. Removing the residual of FEV1 at Visit 2 from the analysis did not change the estimates for any of the other covariables (data not shown). This was to be expected. With the method we used for adjustment for FEV1, we removed the correlation between FEV1 at Visit 2 and all the other covariables in the model.
To investigate whether the use of inhaled corticosteroids influenced the findings with regard to smoking in the analysis, and vice versa, we analyzed the data without smoking history and with inhaled corticosteroids, and vice versa. The estimates did not differ from those in the original analysis, which included both smoking habit and the use of inhaled corticosteroids.
Figure 3 shows the association between the annual change in FEV1 from Visit 2 to Visit 3 and smoking habit. Subjects who stopped smoking between Visit 2 and Visit 3 showed an increase in FEV1 (+26.4 ml/yr), whereas nonsmokers at both Visits 2 and 3, subjects who had started smoking by Visit 3, and smokers at Visits 2 and 3 showed decreases in FEV1 of −19.7 ml/yr, −14.3 ml/yr, and −23.4 ml/yr, respectively.
Fig. 3. Association between annual decline in FEV1 from Visit 2 to Visit 3 and smoking habits from Visit 2 (ages 22 to 32 yr) to Visit 3 (ages 32 to 42 yr) (n = 73). * represents outliers.
[More] [Minimize]Figure 4 shows the association between use of inhaled corticosteroids and annual decline in FEV1 from Visit 2 to Visit 3. The group of subjects who used inhaled corticosteroids at Visit 2 and Visit 3 showed an increase in FEV1 from Visit 2 to Visit 3 of 13.7 ml/yr, whereas the group of subjects who did not use inhaled corticosteroids and the group that used inhaled corticosteroids only at one visit (Visit 2 or Visit 3) showed annual declines in FEV1 from Visit 2 to Visit 3 of −18.1 ml/yr and −28.0 ml/yr, respectively. Only 73 individuals were included in this analysis, owing to missing values for eosinophils, lung function, and/or serum IgE.
Fig. 4. Association between annual decline in FEV1 from Visit 2 to Visit 3 and the use of inhaled corticosteroids from Visit 2 (ages 22 to 32 yr) to Visit 3 (ages 32 to 42 yr) (n = 73).
[More] [Minimize]This study has investigated the childhood factors in asthmatic subjects that are associated with the level of FEV1 in early adulthood, as well as the factors in early adulthood that are associated with subsequent decline of FEV1 in adulthood. Both BHR and a low level of FEV1% predicted in childhood were significantly associated with a low level of FEV1 in early adulthood. Individuals who quit smoking and subjects who used inhaled corticosteroids from Visit 2 to Visit 3 had a significantly smaller annual decline of FEV1.
For the outcome of asthma, as assessed through FEV1 in early adulthood, the presence of BHR in childhood was an important predictor in our study. To our knowledge, only a few studies have examined the outcome of childhood asthma through assessment with objective measurements (31-33). Roorda and coworkers performed a follow-up study with a mean interval of 14.8 yr involving 406 asthmatic children aged 8 to 12 yr. They found a significant association between FEV1 in childhood and the level of FEV1 in adulthood, but no association between the ln PC10 for histamine in childhood and level of FEV1 in adulthood. A possible explanation for the difference in their findings and ours is that only 56% of the children in their study had a PC10 ⩽ 16 mg/ml, whereas 83% in our study had this level of BHR.
We found no association between BHR at ages 22 to 32 yr and the rate of decline in FEV1 between this age range and 32 to 42 yr of age. Few studies of asthmatic adults have shown an association between BHR at the start of the study and an accelerated decline of lung function. After a follow-up of 2 yr, Van Schayck and coworkers found that a marked BHR was associated with a more rapid decline in FEV1 in 71 asthmatic adults with a mean age of 51 yr (34). This difference in their results and ours could be explained by the older age of their subjects and the shorter duration of their follow-up. Peat and coworkers, in a population study, found a significant association between the rate of decline of FEV1 and BHR in asthmatic adults aged 22 to 69 yr (35). Having asthma was classified on the basis of a self-administered questionnaire. BHR was assessed only at the end of the study, and the severity of BHR at that time could therefore have been influenced by the previous decline in FEV1, in that individuals with severe BHR had the lowest FEV1 at that time. We also found that BHR in childhood had no association with the rate of decline in FEV1 through adulthood (data not shown). From our results, we deduce that BHR in childhood is an important predictor of the level of FEV1 in early adulthood, but that other factors are important for the decline in FEV1 through adulthood. However, ours was a small group of allergic asthmatic subjects and a nonsignificant result should therefore be interpreted with caution.
Although the group that used inhaled corticosteroids from Visit 2 to Visit 3 of follow-up in our study was rather small (n = 11), there was a significant association between consistent use of inhaled corticosteroids and a smaller decline in lung function in adulthood, emphasizing that this was a relevant observation. This result supports the guidelines for asthma treatment, and is in accord with other, mostly short-term clinical studies of asthmatic subjects with regard to the relationship between inhaled corticosteroids and outcome of asthma (36– 39). The current study suggests that inhaled corticosteroids are also of long-term benefit over a period of 13 yr.
Subjects who quit smoking had a significantly smaller decrease in lung function. This is an important finding and is in accord with results of other studies (40, 41). Special efforts should be made to motivate young adults with asthma to quit smoking. However, subjects who continued smoking did not show a significantly steeper annual decline in FEV1 than non-smokers. This may have been the result of the “healthy smoker effect,” in that persons with more susceptible airways either do not take up smoking or quit at an early age. An analysis of our data suggests that there were no significant differences with respect to gender, lung function, or bronchial responsiveness in childhood between subjects who started smoking during follow-up and those who never smoked. However, in the group of subjects who continued smoking into adulthood, there were significantly more males and subjects with good lung function in childhood (data not shown). Thus, our data support the hypothesis of a healthy smoker effect.
Results of several studies suggest that parental cigarette smoking, and especially maternal smoking, may have adverse effects on children's pulmonary function (10, 11, 42). Most previous studies have been cross-sectional in design, demonstrating an association between current environmental tobacco smoke exposure and lower level of lung function. Two longitudinal studies (12, 41) have each suggested an association of maternal smoking with a reduced rate of increase in FEV1 in school-aged children. In a cross-sectional study, Corbo and coworkers found that a low level of environmental tobacco smoke exposure was associated with a lower level of lung function even in nonsmoking healthy children with nonsmoking parents (43). In our analysis we also expected to find a negative association between the maximally attained level of FEV1 at Visit 2 and environmental tobacco smoke exposure in childhood, but could not find such an association. In the regression analysis, we included both maternal smoking and low level of lung function in childhood. The effect of maternal smoking might already have been reflected in a lower level of FEV1 in childhood. Removing FEV1 in childhood from the analysis resulted in a higher coefficient of maternal smoking in the analysis. This finding supports the idea that the effect of maternal smoking was already reflected in the level of FEV1 in childhood, which is in accord with findings in previous studies.
In the Dutch version of the British Medical Research Council's standard questionnaire, which we used in our study, no information was given about smoking habits during pregnancy, and we were therefore unable to separate the effects of in utero exposure to maternal smoking and early and late environmental tobacco smoke exposure on lung function growth in the asthmatic children in our study. This might be the reason why we found no association between maternal smoking and lung function, because Wang and coworkers showed that a low level of lung function was associated with in utero exposure to maternal smoking and in combination with current exposure in 8,706 nonsmoking children followed annually between the ages of 6 and 18 yr (44). Another possible explanation is that in our study, 40% of the mothers and 92% of the fathers were smokers. Thus, almost the entire group of asthmatic children in the study were exposed to environmental tobacco smoke.
The clinical impact of our study is that such risk factors as a low FEV1 and more severe BHR occurring early in life are predictive of the development of lower lung function in later childhood. This emphasizes the importance of early intervention and prevention to optimize the clinical prognosis in allergic asthma. This is supported by our observation that asthmatic subjects who used inhaled corticosteroids from ages 22 to 32 to 32 to 42 yr had less of a decline in lung function after these ages than did those not using inhaled corticosteroids. It is again important to realize that ours was a small group of asthmatic children, and that our data may not be applicable to all asthmatic individuals in the population. Nevertheless, our sample was unique in that objective data were available from the beginning of the study.
In conclusion, a low FEV1 and more severe BHR in childhood are independent risk factors for a low level of FEV1 at ages 22 to 32 yr. Asthmatic individuals who continue to use inhaled corticosteroids or who quit smoking within this age range have a smaller annual decline in FEV1. Our data stress the importance of studying intervention in both young childhood and early adulthood to prevent or to postpone further lung function deficits in asthmatic individuals.
| B | CI | p Value | ||||
|---|---|---|---|---|---|---|
| Sex (male = 1, female = 0) | −13.78 | −29.59 to 2.03 | 0.093 | |||
| Use of cs at Visits 2 and 3 | 31.33 | 4.44 to 58.22 | 0.026 | |||
| Use of cs at Visits 2 or 3 | −19.21 | −38.17 to −0.24 | 0.052 | |||
| Ln eos (106 · L−1) Visit 2 | −3.80 | −12.82 to 5.22 | 0.412 | |||
| Log IgE (IU/L) visit 2 | −8.05 | −24.16 to 8.06 | 0.331 | |||
| BHR (PC10 ⩽ 16 mg/ml) Visit 2 | 5.91 | −11.71 to 23.52 | 0.514 | |||
| Age (yr) Visit 2 | 0.74 | −2.50 to 3.98 | 0.657 | |||
| Symptoms Visit 2 | −6.61 | −23.46 to 10.23 | 0.445 | |||
| Total positive skin tests Visit 2 | 4.43 | −0.31 to 9.17 | 0.072 | |||
| Start smoking at Visit 3 | 15.63 | −19.27 to 50.54 | 0.384 | |||
| Ex-smoking Visits 2 and 3 | −6.41 | −28.58 to 15.75 | 0.573 | |||
| Ex-smoking Visit 3 | 32.85 | 4.75 to 60.95 | 0.026 | |||
| Smoking Visits 2 and 3 | −7.74 | −24.89 to 9.41 | 0.380 | |||
| Residual Visit 2 | −0.01 | −0.03 to 0.00 | 0.017 |
The authors would like to thank Dr. M. H. F. Wilkinson for correcting the English language in this paper.
Supported by grant 93.64 from The Netherlands Asthma Foundation and by the Stichting Astma Bestrijding.
| 1. | Burrows B., Lebowitz M. D., Camilli A. E., Knudson R. J.Longitudinal changes in forced expiratory volume in one second in adults. Am. Rev. Respir. Dis.1331986974980 |
| 2. | Knudson R. J., Clark D. F., Kennedy T. C., Knudson D. E.Effect of aging alone on mechanical properties of the normal adult lung. J. Appl. Physiol.43197710541062 |
| 3. | Roberts C. M., MacRae K. D., Winning A. J., Adams L., Seed W. A.Reference values and prediction equations for normal lung function in a non-smoking white urban population. Thorax461991643650 |
| 4. | Burrows B., Cline M. C., Knudson R. J., Taussig L. M., Lebowitz M. D.A descriptive analysis of growth and decline of the FVC and FEV1. Chest51983717724 |
| 5. | Tager I. B., Segal M. R., Speizer F. E., Weiss S. T.The natural history of forced expiratory volumes: effect of cigarette smoking and respiratory symptoms. Am. Rev. Respir. Dis.1381988837849 |
| 6. | Sherrill D. L., Morgan W. J., Taussig L. M., Landau L. E., Burrows B.A mathematical procedure for estimating the spatial relationships between lung function, somatic growth, and maturation. Pediatr. Res.251989316321 |
| 7. | Borsboom G. J. J. M., Van Pelt W., Quanjer P. H.Pubertal growth curves of ventilatory function: relationship with childhood respiratory symptoms. Am. Rev. Respir. Dis.1471993372378 |
| 8. | Wang X., Dockery D. W., Wypij D., Fay M. E., Ferris G.Pulmonary function between 6 and 18 years of age. Pediatr. Pulmonol.1519937588 |
| 9. | Sherman C. B., Xu X., Speizer F. E., Ferris B. G., Weiss S. T., Dockery D. W.Longitudinal lung function decline in subjects with respiratory symptoms. Am. Rev. Respir. Dis.1461992855859 |
| 10. | Tager I. B., Hanrahan J. P., Tosteson T. D., Castile R. G., Brown R. W., Weiss S. T., Speizer F. E.Lung function, pre- and postnatal smoke exposure, and wheezing in the first year of life. Am. Rev. Respir. Dis.1471993811817 |
| 11. | Hanrahan J. P., Tager I. B., Segal M. R., Tosteson T. D., Castile R. G., van Vunakis H., Weiss S. T., Speizer F. E.The effect of maternal smoking during pregnancy on early infant lung function. Am. Rev. Respir. Dis.145199211291135 |
| 12. | Sherrill D. L., Martinez F. D., Lebowitz M. D., Holdaway M. D., Flannery E. M., Herbison G. P., Stanton W. R., Silva P. A., Sears M. R.Longitudinal effects of passive smoking on pulmonary function in New Zealand children. Am. Rev. Respir. Dis.145199211361141 |
| 13. | Schwartz J., Weiss S. T.Relationship of skin test reactivity to decrements in pulmonary function in children with asthma or frequent wheezing. Am. J. Respir. Crit. Care Med.152199521762180 |
| 14. | Johnston I. D. A., Strachan D. P., Anderson H. R.Effect of pneumonia and whooping cough in childhood on adult lung function. N. Engl. J. Med.3381998581587 |
| 15. | Gottlieb D. J., Sparrow D., O'Connor G. T., Weiss S. T.Skin test reactivity to common aeroallergens and decline of lung function: The Normative Aging Study. Am. J. Crit. Care Med.1531996561566 |
| 16. | Tracey M., Villar A., Dow L., Coggon D., Lampe F. C., Holgate S. T.The influence of increased bronchial responsiveness, atopy, and serum IgE on decline in FEV1: a longitudinal study in the elderly. Am. J. Respir. Crit. Care Med.1511995656662 |
| 17. | Rijcken B., Schouten J. P., Xu X., Rosner B., Weiss S. C.Airway hyperresponsiveness to histamine associated with accelerated decline in FEV1. Am. J. Respir. Crit. Care Med.51199513771382 |
| 18. | Peat, J. K., A. J. Woolcock, and K. Cullen. 1987. Rate of decline of lung function in subjects with asthma. Eur. J. Respir. Dis. 171–179. |
| 19. | Lange P., Parner J., Vestbo J., Schnohr P., Jensen G.A 15-year follow-up study of ventilatory function in adults with asthma. N. Engl. J. Med.339199811941200 |
| 20. | Gerritsen J., Koëter G. H., Postma D. S., Schouten J. P., Knol K.Prognosis of asthma from childhood to adulthood. Am. Rev. Respir. Dis.140198913251330 |
| 21. | Van Lookersen Campagne, J. G. 1972. The Reaction Patterns After House Dust Provocation in Children with Non-specific Lung Disease (CNSLD). Thesis. University of Groningen, Groningen, The Netherlands. |
| 22. | Van Der Lende R, Orie N. G. M.The MRC-ECCS questionnaire on respiratory symptoms (use in epidemiology). Scand. J. Respir. Dis.531972218226 |
| 23. | Kerrebijn K. F., Hoogeveen-Schroot H. C. A., Van Der Wal M. C.Chronic nonspecific respiratory disease in children: a five year follow-up study. Acta Paediatr. Scand.261(Suppl.)1977172 |
| 24. | Burney P. G. J., Luczynska C., Chinn S., Jarvis D.for the European Community Respiratory Health SurveyThe European Community Respiratory Health Survey. Eur. Respir. J.71994954960 |
| 25. | Zapletal, A., M. Samanek, and T. Paul. 1987. Lung function in children and adolescents. Methods, reference values. In A. Zapletal, editor. Progress in Respiration Research, Vol. 22. Karger, Basel, Switzerland. 114–116, 193. |
| 26. | Quanjer, P. H., G. J. Tammeling, J. E. Cotes, O. F. Pedersen, R. Peslin, and J. C. Yernault. 1993. Lung volumes and forced ventilatory flows: report of the Working Party for Standardization of Lung Function Tests: European Community for Steel and Coal. Official statement of the European Respiratory Society. Eur. Respir. J. 6(Suppl. 16):5–40. |
| 27. | De Vries K., Goei T., Booy-Noord H., Orie N. M. G.Changes during 24 hours in lung function and histamine hyperreactivity of the bronchial tree in asthmatic and bronchiatic patients. Int. Arch. Allergy20196293101 |
| 28. | Know, K. 1965. A clinical and epidemiological study of the significance of bronchial hyperreactivity in children with chronic nonspecific lung disease (CNSLD). Thesis, University of Groningen, Groningen, The Netherlands. |
| 29. | Eiser N. M., Kerrebijn K. F., Quanjer P. H.Guidelines for standardization of bronchial challenges with (nonspecific) bronchoconstricting agents. Bull. Eur. Physiopathol. Respir.191983495514 |
| 30. | Schouten J. P., Tager I. B.Interpretation of longitudinal studies: an overview. Am. J. Respir. Crit. Care Med.1541996S278S284 |
| 31. | Roorda R. J., Gerritsen J., van Aalderen W. M. C., Schouten J. P., Veltman J. C., Weiss S. T., Knol K.Follow-up of asthma from childhood to adulthood: influence of potential childhood risk factors on the outcome of pulmonary function and bronchial responsiveness in adulthood. J. Allergy Clin. Immunol.931994575584 |
| 32. | Oswald H., Phelan P. D., Lanigan A., Hibert M., Bowes G., Olinsky A.Outcome of asthma in mid-adult life. B.M.J.30919949596 |
| 33. | Kokkonen J., Linna O.The state of childhood asthma in young adulthood. Eur. Respir. J.61993657661 |
| 34. | van Schayck C. P., Dompeling E., van Herwaarden C. L. A., Wever A. M., van Weel C.Interacting effects of atopy and bronchial hyperresponsiveness on the annual decline in lung function and the exacerbation rate in asthma. Am. Rev. Respir. Dis.144199112971301 |
| 35. | Peat J. K., Woolcock A. J., Cullen K.Rate of decline of lung function in subjects with asthma. Eur. J. Respir. Dis.701987171179 |
| 36. | Juniper E. F., Kline P. A., Vanzieleghem M. A., Ramsdale E. H., O'Byrne P. M., Hargreave F. E.Effect of long-term treatment with an inhaled corticosteroid (budesonide) on airway hyperresponsiveness and clinical asthma in non-steroid-dependent asthmatics. Am. Rev. Respir. Dis.1421990832836 |
| 37. | Haahtela T., Järvinen M., Kava T., Kiviranda K., Koshinen S., Lehtonen K., Nikander K., Persson T., Reinihainen K., Selroos O., Sovijärvi A., Stenius-Aarniala B., Svahn T., Tammivaara R., Kitinen L.A comparison of a β2-agonist, terbutaline, with an inhaled corticosteroid, budesonide, in newly detected asthma. N. Engl. J. Med.3251991388392 |
| 38. | Kerstjens H. A., Brand P. L. P., Hughes M. D., Robinson N. J., Postma D. S., Sluiter H. J., Bleecker E. R., Dekhuijzen P. N., de Jong P. M., Mengelers H. J., et al.A comparison of bronchodilator therapy with or without inhaled corticosteroid in obstructive airways disease. N. Eng. J. Med.327199214131419 |
| 39. | Hahhtela T., Järvinen M., Kava T., Kiviranta K., Koskinen S., Lehtonen K., Nikander K., Persson T., Selroos O., Sovijarvi A.et alEffects of reducing or discontinuing inhaled budesonide in patients with mild asthma. N. Engl. J. Med.3311994700705 |
| 40. | Bossé R., Sparrow D., Rose C. L., Weiss S. T.Longitudinal effect of age and smoking cessation on pulmonary function. Am. Rev. Respir. Dis.1231981378381 |
| 41. | Sherill D. L., Holberg C. T., Enright P. L., Lebowitz M. D., Burrows B.Longitudinal analysis of the effects of smoking onset and cessation on pulmonary function. Am. J. Respir. Crit. Care Med.1491994591597 |
| 42. | O'Connor S. T., Weiss, Tager I., Speizer F. E.The effect of passive smoking on pulmonary function and nonspecific bronchial responsiveness in a population-based sample of children and young adults. Am. Rev. Respir. Dis.1351987800805 |
| 43. | Corbo G. M., Agabiti N., Forastiere F., Dell'orco V., Pistilli R., Kriebel D., Pacifici R., Zuccaro P., Ciappi G., Perucci C. A.Lung function in children and adolescents with occasional exposure to environmental tobacco smoke. Am. J. Respir. Crit. Care Med.1541996695700 |
| 44. | Wang X., Wypij D., Gold D. R., Speizer F. E., Ware J. H., Ferris B. G., Dockery D. W.A longitudinal study of the effects of parental smoking on pulmonary function in children 6–18 years. Am. J. Respir. Crit. Care Med.149199414201425 |
