INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

There is increasing evidence to suggest that environmental and genetic factors have effects on airway development and function in early life (1). Fetal and early postnatal life are periods of rapid growth and development of the respiratory system (2). Although some remodeling of the lung may occur after prenatal and perinatal insults, there appears to be considerable tracking of respiratory function from the end of the first year of life to late childhood, and this may have an important impact on subsequent respiratory health (2).

In one longitudinal study from Arizona (3), diminished airway function in early life was associated with wheezing in the first year, and impaired airway function at 6 yr. It has been suggested that these findings require confirmation in more temperate and industrialized settings (4). We conducted a prospective community-based study of infants born in inner London to identify whether factors during pregnancy and shortly after birth are associated with subsequent wheezing. Specifically, we investigated whether the odds of wheezing in the first year of life were increased in those with early impairment of airway function, assessed by measurements of specific airway conductance made prior to any respiratory illness. The influence on this association of a family history of asthma and maternal smoking during pregnancy was examined.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Eligible infants were identified from consecutive births to women registered with two primary care health centers in inner London. Healthy Caucasian term (> 35 wk) infants without major congenital abnormalities or neonatal respiratory illness were eligible. Non-Caucasian infants were excluded to minimize heterogeneity due to ethnic origin (5, 6). Families were not approached if social or medical reasons precluded contact, and those planning to leave the area during the next year were ineligible. Parents were contacted by letter, and written informed parental consent was obtained after at least one prearranged home visit by the research nurse. The local Ethics Committee approved this study.

Respiratory function was measured when infants were 1 to 3 mo of age and prior to any upper or lower respiratory symptoms. Measurements were performed, after sedation with triclofos sodium elixir (75 to 100 mg/kg), during behaviorally determined quiet sleep as described previously (7). Body weight was measured to the nearest 10 g and crown-heel length to the nearest 0.1 cm (7).

Airway function was assessed from plethysmographic measurements of inspiratory and expiratory resistance as an index of absolute airway caliber (8). In the absence of any upper airway disease or abnormality, resistance during inspiration tends to reflect the optimal airway caliber during the respiratory cycle, whereas end-expiratory resistance will detect increases in resistance at low lung volumes caused by peripheral airway closure or narrowing. Because the lungs and airways grow rapidly during the first year of life, resistance is higher in smaller or younger infants (11). By dividing the reciprocal of resistance (conductance) by lung volume (FRC_pleth) to obtain specific airway conductance (sGaw), account can be taken of these growth- related changes (8, 11), with sGaw being diminished in the presence of reduced airway caliber in relation to lung size.

Respiratory system compliance was measured using the multiple occlusion technique (12) to allow differences in airway function to be interpreted with respect to underlying differences in elastic recoil. The time to peak tidal expiratory flow as a proportion of total expiratory time (t_PTEF:t_E), a measure of the degree to which expiratory flow and timing are modulated to slow lung-emptying, was also calculated to allow comparison with previously published work (13). Respiratory function parameters were calculated according to previously established criteria (8) by researchers blinded to the child's subsequent respiratory status.

Obstetric, medical, social, and demographic details were obtained from mothers at recruitment. Mothers were asked whether asthma had ever been diagnosed in a first-degree relative of the child. Information was also obtained on current or most recent paternal and maternal occupations (16), as well as maternal smoking during pregnancy. Postnatal exposure to tobacco smoke was assessed at recruitment, and a specimen of urine was collected from each infant at the time of respiratory function testing and was frozen at 20° C within 2 h of voiding. Urinary cotinine, assayed by gas liquid chromatography (17), was expressed as nanograms of cotinine per milligram of creatinine. After the first birthday, physician-diagnosed episodes of wheezing in the first year of life were identified retrospectively from the child's practice medical record by a researcher blinded to the results of the respiratory function tests.

Statistical Methods

Background characteristics and respiratory function were examined in relation to subsequent wheezing, sex, family history of asthma, and maternal smoking. The association between specific airway conductance and the onset of wheezing by the first birthday and the number of episodes experienced was examined using binary and ordinal logistic regression, respectively (Minitab, Release 11; Minitab Inc., State College, PA), with and without prior adjustment for family history of asthma and maternal smoking during pregnancy. The association between age at first episode and early airway function was examined using Cox regression (SPSS for Windows, Release 7; SPSS Inc., Chicago, IL).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Population Characteristics

Over a 5-yr period, the parents of 118 (39%) of 300 eligible infants born in the study practices and successfully contacted consented to their child having respiratory function tests. Although declining testing, parents of a further 148 infants agreed to participate in a parallel morbidity study, so that background data are available for 266 (89%) of all eligible and contacted infants. Infants whose parents agreed to respiratory function tests were similar to those whose parents did not with respect to birthweight, gestational age, sex distribution, maternal age, and educational level and prevalence of a first-degree family history of asthma (data not shown).

Respiratory function was measured in 108 children at a mean (range) age of 7.7 wk (4.9-12.6), prior to the development of any upper or lower respiratory symptoms. The 10 children who developed respiratory symptoms between enrollment and testing were not measured. Seven (6%) of these 108 infants (all boys) were subsequently lost to follow-up (five moved out of the study area and two withdrew from the study), leaving 101 infants, in whom 28 wheezing was diagnosed by a physician before their first birthday. Eleven infants experienced more than one episode of wheezing (range, 2 to 5). The median (interquartile range [IQR]) age at first episode of wheezing was 26 wk (21 to 39) and the interval between respiratory function testing and onset of wheezing was 18 wk (13 to 31). Measurements of specific airway conductance were unsuccessful in 15 infants because of technical problems and successful in 86 (63 without subsequent wheezing, 16 with one episode, and 7 with more than one episode).

At birth, infants with subsequent physician-diagnosed wheezing were significantly lighter and of shorter gestation and, by the time of respiratory function testing, significantly shorter than those without such a diagnosis (Table 1). Infants with subsequent wheezing were more likely to have a father with a manual occupation (Table 1) but had mothers of a similar age to those who did not wheeze: mean (SD) maternal age, 27.3 yr (5.2) and 28.9 yr (5.2), respectively (p = 0.16).

The parents of 27 infants reported a previous or current diagnosis of asthma in one or more first-degree relatives, which in 16 families included the mother. Infants with a positive family history were more likely to wheeze that those without (Table 2): wheezing developed in 14 of 27 (52%) infants with a family history (eight of 16 with an affected mother), compared with 14 of 73 (19%) infants without such a history (p = 0.01). Infants with and without a positive family history were otherwise similar with respect to body size and age at testing, paternal occupational status, and maternal smoking (Table 2).

Boys tended to wheeze more than girls: 35 versus 21%, respectively; 95% confidence interval (CI) of difference (boys-girls), 4% to 31% (p = 0.07). There were no sex differences in the percentage of infants with a first-degree relative with diagnosed asthma or with smoking mothers (data not shown). Among boys, absolute lung volumes were significantly greater, but weight-adjusted values were similar, to those observed in girls, reflecting the fact that boys weighed significantly more and were longer at testing: mean (SD) FRC_pleth, 22.8 (4.8) ml/kg for boys and 23.1 (4.5) ml/kg for girls. There were no significant sex differences in other respiratory parameters, although inspiratory specific airway conductance was lower in boys: mean (SD), 2.45 (1.03) s¹ · kPa¹ and 2.90 (1.07) s¹ · kPa¹ in boys and girls, respectively; 95% CI of difference (boys-girls), 0.90 to 0.004 (p = 0.052). Data for boys and girls were pooled in subsequent analyses.

Of the 44 women reporting smoking in pregnancy, 39 continued to smoke postnatally, whereas only one woman not smoking in pregnancy started postnatally. Geometric mean urinary cotinine:creatinine ratios were significantly higher among infants whose mothers reported smoking during pregnancy, 15 of whom also had a positive family history of asthma (Table 2). Compared with infants of nonsmoking mothers, those whose mothers smoked were more likely to wheeze subsequently and to have fathers in manual occupations, and were lighter at birth as well as lighter and shorter at testing (Table 3).

Association between Premorbid Airway Function and Subsequent Wheeze

When measured at approximately 8 wk of age, specific airway conductance was significantly diminished during expiration in infants with subsequent wheezing, being 2.02 (1.07) s¹ · kPa¹ in infants who subsequently wheezed and 2.60 (0.93) s¹ · kPa¹ in those who did not (p < 0.05). Similar values were observed during inspiration (Table 4). Although t_PTEF:t_E tended to be shorter among infants with subsequent wheezing, this just failed to reach statistical significance (p = 0.07). Other respiratory parameters did not differ significantly between the two groups (Table 4). As the results obtained from statistical analyses were similar for both inspiratory and expiratory specific airway conductance, subsequent findings will be discussed for the latter alone, which will be referred to as specific airway conductance.

Premorbid specific airway conductance and t_PTEF:t_E were significantly diminished among infants whose parents reported a first-degree relative in whom asthma was ever diagnosed (Table 5). These parameters were similar in infants in whose mothers asthma had or had not ever been diagnosed (data not shown). Expiratory resistance was significantly elevated, but specific airway conductance and t_PTEF:t_E were not diminished, in infants of mothers who smoked: mean (SD) expiratory Raw, 5.3 (3.3) kPa · L¹ · s¹ and 4.1 (1.60) kPa · L¹ · s¹ in infants whose mothers did and did not smoke, respectively; 95% CI of difference, 0.0 to 2.46; p < 0.05 (Table 6).

The relative odds of wheezing were increased significantly in infants with lower levels of specific airway conductance at 8 wk, the unadjusted odds ratio (95% CI) for wheezing being 2.1 (1.1 to 3.8) for each unit (s¹ · kPa¹) decline in specific airway conductance (p = 0.02). Including sex in the model did not influence this association. The relative odds were similar when ordinal logistic regression analyses were performed, when a significant association between diminished specific airway conductance and the number of episodes of wheeze experienced in the first year was found: odds ratio (95% CI), 2.2 (1.2 to 4.1) (p = 0.01). Impaired premorbid airway function was also associated with a significantly earlier age at onset of wheezing. At any given age during the first year, infants with diminished premorbid specific airway conductance were more likely to develop wheezing: proportional hazards ratio for each unit (s¹ · kPa¹) decline in sGaw, 1.9; 95% CI, 1.1 to 3.2 (p = 0.02).

The association between specific airway conductance and subsequent wheezing became nonsignificant or less significant after adjustment for a family history of asthma or maternal smoking in pregnancy. The adjusted odds ratio (95% CI) for wheezing was 1.7 (0.9 to 3.3) and 1.9 (1.0 to 3.6), respectively, for each unit decline in specific airway conductance (p = 0.10 and p = 0.05, respectively). After adjustment for specific airway conductance, the odds of subsequent wheezing remained significantly increased among infants with a first-degree relative in whom asthma had been diagnosed and those exposed to maternal smoking in utero: adjusted odds ratio (95% CI), 4.3 (1.3 to 13.8) and 4.9 (1.6 to 15.0), respectively (p = 0.016 and 0.005).

A family history of asthma and exposure to maternal smoking during pregnancy were independently associated with a significantly earlier age at onset of wheezing. After adjustment for maternal smoking and specific airway conductance, infants with a family history of asthma were significantly more likely to develop wheezing at any given age during the first year than were those without such a history (proportional hazards ratio, 3.0; 95% CI, 1.3 to 7.3; p = 0.013) (Figure 1). Similarly, after adjustment for a family history of asthma and specific airway conductance, infants whose mother smoked were significantly more likely to develop wheezing at any given age in the first year than were those whose mothers did not (proportional hazards ratio, 3.2; 95% CI, 1.3 to 7.9; p = 0.013) (Figure 2).

View larger version (11K): [in this window] [in a new window]   Figure 1.   Proportion of infants remaining wheeze-free at any given age among those with a first-degree relative with asthma (dashed line) and those without such a history (solid line).

View larger version (11K): [in this window] [in a new window]   Figure 2.   Proportion of infants remaining wheeze-free at any given age among those whose mother smoked during pregnancy (dashed line) and those without such a history (solid line).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We have found that specific airway conductance was significantly diminished at 8 wk of age and prior to any respiratory illness in previously healthy infants who subsequently developed one or more episodes of wheezing by their first birthday. Infants with lower levels of specific airway conductance were also more likely to have recurrent wheezing in the first year and to start wheezing at an earlier age. Infants with a family history of asthma had significantly reduced specific airway conductance and were also more likely to develop subsequent wheezing. In contrast, we found that specific airway conductance was not significantly diminished in infants whose mothers smoked during pregnancy, although they were more likely to wheeze during the first year of life. Such infants were, however, significantly lighter at birth, lighter and shorter at 8 wk of age, and had higher expiratory resistance than did those not exposed. As mothers who smoke during pregnancy commonly continue to do so after delivery, these findings relate to both prenatal and postnatal exposure to tobacco. There were no significant sex differences in size-corrected parameters of respiratory function, except that boys tended to wheeze more and have lower inspiratory specific airway conductance. These observations are consistent with previously published studies suggesting that airway function is diminished in boys compared with girls during both infancy and childhood (1, 6, 13, 18, 19).

Population Characteristics

The infants included in this study were recruited from consecutive births in two general practices in inner London and were representative of all eligible births in that setting. Only seven (6%) of the 108 infants in whom respiratory function tests were performed at 8 wk of age were subsequently lost to follow-up, suggesting that biases caused by cohort attrition are unlikely. Although having a first-degree relative with a diagnosis of asthma might increase the likelihood of both parental report and physician diagnosis of wheezing in infancy, diagnostic bias is unlikely to explain the finding of diminished conductance in infants with such a history, as those responsible for calculation of respiratory parameters were blinded to the status of the infant. Furthermore, the prevalence of a first- degree relative with asthma was similar in infants whose parents consented to respiratory function testing, when compared with those whose parents declined. Creatinine-standardized urinary cotinine concentrations discriminated well between infants with positive and negative parental reports of postnatal smoking exposure, suggesting that underreporting of smoking by parents was unlikely.

Possible Mechanisms

There have been a number of longitudinal studies examining the association between diminished premorbid respiratory function and subsequent lower respiratory illness and wheezing in infancy and early childhood (1). This study used sensitive and validated measures of infant airway function (7), and it was undertaken in an area of inner London with a high prevalence of smoking, asthma, and environmental pollution (20). The findings confirm and extend observations made in similar longitudinal studies in North America (3, 13, 21, 22) and Australia (14, 15). If airway caliber is diminished in very early life, for whatever reason, then further narrowing during viral infections, or as a consequence of exposure to aeroallergens or continued exposure to tobacco smoke postnatally, may further reduce effective airway caliber to below a critical symptomatic threshold and would increase the likelihood of wheezing (23).

Our findings suggest that different processes may be implicated in the respective contributions of environmental and genetic factors to the development of wheezing in the first year of life (23). Importantly, the influence of a genetic predisposition to asthma appears to operate, at least in part, by reducing effective airway caliber in the first few months of life. In contrast, the airway caliber of infants whose mothers smoke is reduced in absolute terms, as suggested by increased airway resistance, but it appears to be appropriate for lung size, as indicated by normal specific airway conductance. Such infants might be expected to outgrow their wheezing during later infancy as the absolute dimensions of the airways increase. This suggests that the effect of smoking on airway function may be mediated via overall growth retardation rather than by specific effects on airway geometry or tone. Although exposure to maternal smoking was more common in infants whose fathers were in manual occupations, there was a similar trend to lower birth weight and body size at 8 wk among those whose fathers were in nonmanual occupations (data not shown). This provides further support for a specific effect of smoking or a related factor such as maternal diet on fetal growth (24).

Interpretation of Respiratory Function Tests

Specific airway conductance provides a measure of effective airway caliber corrected for lung size and hence somatic growth (11), and it will be reduced if airway caliber is diminished and/or lung volume increased in relation to body size. A potential disadvantage of all measurements of resistance during infancy is that infants are preferential nose breathers and hence the resistance of the nasal passages, which can comprise as much as 50% of total resistance, is included (5). Not only may this reduce the relative sensitivity of these techniques in detecting changes in peripheral airway function, but values may be greatly increased in the presence of any upper respiratory tract infection. For this reason, infants were considered ineligible for this study if they had previously experienced any upper or lower respiratory tract symptoms.

After allowing for differences in body size there were no differences in lung volumes between the groups. As in the current study, all infants were asymptomatic at testing and without clinical signs of hyperinflation, this suggests that the observed reduction in specific airway conductance associated with a family history of asthma and subsequent wheezing was primarily due to a diminished airway conductance (i.e., raised resistance) rather than to an elevated lung volume. Reduced effective airway caliber may reflect a number of factors, which include anatomic changes in lumen size, increased thickness of the airway wall (because of either inflammation or bronchial muscle hypertrophy), airway secretions, bronchoconstriction, or a relative lack of distending pressure (i.e., low elastic recoil). As all infants were asymptomatic at testing and without prior respiratory illness, diminished specific airway conductance is unlikely to be due to airway inflammation or secretions. It is equally unlikely to reflect differences in the elastic recoil of the respiratory system, as respiratory compliance was similar in infants with and without subsequent wheezing, and did not vary according to family history of asthma or maternal smoking. Rather it would appear that diminished specific airway conductance reflects congenitally small airways, as has been proposed previously (13, 21), or some alteration in airway wall compliance or structure. Sex differences may also be important since, relative to girls, boys have diminished airway function and are more susceptible to respiratory illnesses during early childhood (13, 18).

In previous studies of respiratory function and wheezing in infancy (13, 14, 18), airway function has been assessed by measurements of maximal flow at functional residual capacity (max_FRC), which are thought to reflect peripheral airway function (25). These studies have reported increased airway responsiveness, but no significant reduction in baseline max_FRC, during the first months of life in infants with a family history of asthma (14, 18). In the current study, we measured airway resistance rather than airway responsiveness. Although precise partitioning of airway resistance during infancy remains controversial, specific airway conductance is likely to be more sensitive than max_FRC to changes in central airway caliber. Mild increases in airway tone or wall thickness may diminish central airway caliber (and hence specific airway conductance) while protecting against end-expiratory airway closure, thereby maintaining normal or even slightly elevated baseline max_FRC (26). Thus, differences in the parameters used to assess airway function may explain why our observation of diminished baseline airway function in those with a family history of asthma has not been reported previously by others.

In the current study, t_PTEF:t_E was lower in those with subsequent wheezing, although this just failed to reach statistical significance (95% CI, 0.067 to 0.002; p = 0.07). The findings of previously published studies have been inconsistent in this respect, with some reporting a reduction (13, 27), but others no difference (28, 29), in premorbid measures of t_PTEF:t_E among those with subsequent wheezing. The extent to which this parameter reflects airway function remains controversial (1, 30, 31). Although in the presence of severe airflow limitation the tidal breathing curve is defined by intrathoracic airway mechanics, in healthy subjects the shape of this curve and its derived parameters are an integrated response of the entire respiratory system and reflect neural and laryngeal as well as intrathoracic components. Thus, there is a closer association between specific airway conductance and t_PTEF:t_E in the presence of airway narrowing than is found in healthy subjects (32).

Potential Influence of Maternal Smoking

Animal studies suggest that in utero exposure to nicotine will reduce alveolar number and surface area, although absolute lung volume may be largely compensated by increased alveolar size (33) and would therefore appear relatively normal in relation to body size during physiologic studies. A reduction in alveolar number because of intrauterine smoke exposure and/or growth retardation is probably associated with changes in peripheral rather than central airway function. These changes may lead to a reduction in max_FRC in infants exposed to maternal smoking, as reported in some (21), but not all (13, 14, 18), previously published studies. As discussed earlier, we found that airway resistance was significantly elevated during end-expiration, but that specific airway conductance and t_PTEF:t_E were not significantly reduced, in those exposed to maternal smoking during pregnancy. These findings are in keeping with those reported from studies performed in infants at a similar postnatal age (13, 34). By contrast, studies in younger full-term infants have observed that t_PTEF:t_E is reduced in infants whose mothers smoked while pregnant (15, 35). We have recently reported a similar reduction in infants delivered prematurely (mean gestational age, 33 wk) when studied at 3 wk postnatal age (36). Thus, differences reported in the literature may reflect variation in the postnatal age of infants studied. Maturational changes in the modulation of expiratory flow and timing during the first 2 mo of life (10) may explain why some studies have shown a positive association and others none between maternal smoking and t_PTEF:t_E.

In conclusion, These findings suggest that airway function in very early life is an important predictor of wheezing in the first year and that this association is influenced by environmental and genetic factors that may have different physiologic and anatomic effects. Maternal smoking has important and preventable adverse effects on somatic growth and respiratory morbidity in early life. Suboptimal intrauterine growth may have a continuing adverse effect on airway function throughout life (37, 38). Persistence of diminished airway function in early life may be of increasing relevance, given recent observations that impaired airway function in adult life is a major clinical indicator of mortality risk in both men and women (39). It will be important to examine whether diminished specific airway conductance persists to 1 yr of age.