INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

While the survival of preterm infants may be markedly improved by administration of prenatal glucocorticoids and postnatal surfactant therapy, the high incidence of respiratory problems in this population remains a major source of concern (1). Further studies of the influence of factors such as ethnic group, gender, and postconceptional age on lung and airway function in healthy preterm infants are required in order to interpret measurements made in those with respiratory disease.

Spirometric measurements of forced expiratory maneuvers play a central role in the clinical evaluation of lung function in children and adults. While infants and young children cannot be instructed to perform such maneuvers, they can be encouraged to breathe out rapidly by sudden application of a compressive pressure at end inspiration via an inflatable plastic cuff or jacket wrapped around the thorax and abdomen. This technique, known as the rapid thoracoabdominal compression (RTC) technique, enables maximal flow at functional residual capacity (V'max_FRC) to be measured from the partial expiratory flow volume (PEFV) curves obtained as the infant exhales through a pneumotachograph (2, 3). Measurements of V'max_FRC have been used to characterize the normal growth and development of the lungs during infancy and the pulmonary abnormalities associated with acute and chronic lung disorders during early childhood (4).

Marked ethnic differences in infant mortality and morbidity have been reported (5). Although black infants have a higher risk of neonatal and postneonatal death than white infants, among low birthweight infants, neonatal mortality is lower in black infants (5), who are also less likely to develop the respiratory distress syndrome (RDS) than white infants of similar gestational age (8, 9). These relationships appear to persist even after allowing for socioeconomic status (5, 8), which suggests that the respiratory system may be relatively more mature (10) or that lung and airway function are enhanced (11, 12) in black infants delivered prematurely. In a preliminary study, we found higher values of V'max_FRC among black than white preterm infants (13). However, interpretation of these findings was potentially confounded by gender, which was poorly matched between the groups, thus necessitating a larger study.

The fact that black preterm girls are more likely to survive without ventilatory assistance or chronic lung disease has been attributed, at least partially, to their advanced lung maturity with respect to surfactant production (6, 14). There is also evidence to suggest that there may be ethnic differences in the causes of prematurity (5, 15, 16). The higher incidence of intrauterine infections and pregnancy-induced hypertension reported among black women might temporarily enhance fetal pulmonary maturation, whereas the more frequent association of placental insufficiency with preterm delivery in white women would impair fetal airway growth and development.

In adults and older children, it is well established that ethnic group is an important determinant of lung function (17), with both lung volumes and forced expiratory flows tending to be lower in black than white subjects based on values predicted from standing height (17, 18). These discrepancies have been largely attributed to ethnic differences in trunk:leg ratio. However, comparable information in infants is lacking.

Several studies have suggested that airway function is diminished, and respiratory illness more common, in boys than girls during both infancy and childhood (19). However these studies have largely focused on full-term infants and older children, and gender differences in airway function among preterm infants have not been reported.

The aim of this study was to determine whether airway function varies in preterm infants of different ethnic origin and gender by measuring V'max_FRC, respiratory resistance and compliance, and tidal breathing parameters in black and white infants, of similar gestational and postnatal age and without known respiratory disease, prior to discharge from the neonatal unit.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Study Population

Infants were recruited from the Special Care Baby Unit (SCBU) at the Homerton Hospital, Hackney (January 1994-October 1996). Healthy preterm infants ( 36 weeks of gestation) were eligible if both parents were either African or Caribbean/West Indian (black infants) or white Europeans (white infants), and if the mother did not smoke during pregnancy. Infants with congenital abnormalities, neuromuscular or cardiorespiratory diseases (including RDS of the newborn), and those requiring more than 6 h assisted ventilation or 72 h of added oxygen were ineligible, as were those whose parents were not fluent in English. This study was approved by the East London and City Research Ethics Committee, and informed written consent was obtained from the parents.

Study Protocol

Background details obtained by questionnaire from the mother at recruitment included family history of asthma, hay fever, eczema, and chest problems; maternal age on leaving full-time education; maternal age at delivery; ethnic group; and maternal report of cigarette smoking during pregnancy. Data for other relevant variables (e.g., birth weight, gestational age) were extracted from the medical records. A specimen of urine, obtained within the first 24 h of delivery, was collected from each infant whenever possible. This was subsequently analyzed for cotinine using gas-liquid chromatography (28).

Pulmonary Function Tests

Measurements were performed between 5 and 39 d postnatal age during natural, unsedated sleep in an air-conditioned room in the SCBU when the baby was ready for discharge. Crown-heel and crown-rump length were measured by two adults, using an infant stadiometer, and reported as the mean of two values within 0.5 cm of each other (29). Infants were weighed on digital scales (Seca Ltd, Birmingham, UK). Respiratory measurements were made with infants in the supine position and in quiet sleep (30). Data display, recording, and analysis were performed using RASP software (Respiratory Analysis Program; Physiologic Limited, Berks, UK) on an IBM compatible 486 personal computer.

Partial expiratory flow volume (PEFV) curves were measured by wrapping an inflatable jacket snugly around the infant with the arms outside, taking care not to restrict breathing movements. A transparent face mask, attached to a Hans Rudolph (Kansas City) pneumotachrograph (PNT) (0-35 L · min¹), was placed over the nose and mouth to measure flow, and a leak-free seal created using therapeutic silicone putty. Flow signals were digitally integrated to obtain volume, and the flow/volume plot displayed in real time on the computer monitor. A reproducible end expiratory level was established over at least five breaths prior to performing the rapid thoracoabdominal compression (RTC) maneuver, allowing an interval of at least 30 s between each inflation. Jacket compression pressure commenced at approximately 2 kPa (20 cm H₂O) and was increased in 0.5 kPa (5 cm H₂O) increments to a maximum of 7 kPa (70 cm H₂O), or until evidence of flow limitation was achieved (31). V'max_FRC was computed from PEFV loops meeting the following criteria: a smooth expiratory flow-volume curve, a stable end-expiratory level prior to the maneuver, peak jacket pressure attained within 50-100 msec from onset of expiration, and forced expiration beyond resting end expiratory level. V'max_FRC was reported both as the highest and as the mean (SD) of the three highest technically acceptable flows at FRC (4).

Measurements of tidal volume (VT), respiratory rate (RR), peak tidal expiratory flow (PTEF), time to PTEF as a proportion of total expiratory time (t_PTEF:t_E) (32), and passive respiratory mechanics of the respiratory system (compliance [C_rs], resistance [R_rs] and time constant [_rs], using the multiple occlusion [MOT] and single breath techniques [SBT]) were also attempted in all infants, as described in detail previously (10, 33, 34). Data were only considered to be acceptable if at least five end inspiratory airway occlusions had been obtained wherein: (1) a relaxed pressure plateau was achieved; (2) the flow-volume curve was linear over at least 40% of expiration; (3) there was no evidence of early inspiration or excessive expiratory braking, and (4) the values obtained for C_rs using the SBT agreed within 20% of those obtained in the same infant using the multiple occlusion technique (33, 34). The inflatable jacket was loosened or removed prior to performing these measurements.

Data Analysis and Statistical Approach

V'max_FRC, R_rs, and t_PTEF:t_E were investigated throughout. Differences in V'max_FRC according to ethnic group and gender were of major interest, with differences in R_rs and t_PTEF:t_E data providing supplementary information. One way analysis of variance (ANOVA) was used initially to identify differences between the four gender and ethnic groupings (i.e., black male, black female, white male, white female). Multiple regression was used to quantify the contribution of gender and ethnicity to the observed differences both before and after adjustment for the influence of potential confounding factors.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Technically satisfactory values of V'max_FRC were obtained in 56 of the 61 eligible infants in whom measurements were attempted. Urinary cotinine level within 24 h of birth was less than 7 ng · ml¹ in these infants. Details of infants in whom satisfactory measurements were obtained are summarized in Table 1.

The four gender and ethnic groups (black female, black male, white female, white male) were very similar with respect to: birthweight, gestational and postnatal age, weight and length at time of testing (including the ratio of crown rump to crown heel length), multiple births, maternal age and education, birth centile, and family history of wheeze and asthma (Table 1). There was, however, a lower prevalence of family history of "chest trouble" (predominantly bronchitis) among black than white infants, and among girls than boys (95% confidence intervals [CI] of the difference: 40%, 3% and 39%, 5%, respectively). In addition, when compared with white infants, the percentage of black infants with a positive family history of hay fever was lower (95% CI: 53%; 4%).

Results of respiratory function tests are summarized in Table 2. Technically satisfactory results of passive respiratory mechanisms using the SBT were only obtained in 17 of the black and 15 of the white infants. Application of the MOT was more successful, with technically acceptable results according to previously defined criteria (33) being achieved in all but five of the infants. There were no significant differences between the groups with respect to the mean (SD) jacket pressure at which V'max_FRC was achieved (3.8 [1.0] kPa in the black infants and 4.2 [1.2] kPa in the white infants) nor in the percentage of jacket pressure transmitted to the infant (approximately 40% in both groups) (36). Similar findings were observed if V'max_FRC was reported as the highest or the mean of three technically acceptable maneuvers (Table 2), reflecting the fact that the mean (SD) intrasubject coefficient of variation was 8.8 (6.0%) in this group of infants.

Although this study included only asymptomatic infants without known respiratory disease, the PEFV curves displayed a wide range of patterns, ranging from marked concavity with respect to the volume axis (Figure 1a) to convex curves (Figure 1b). Five white (four male) and two black (both male) infants had evidence of airflow limitation during tidal breathing with concave flow-volume curves even at low jacket pressures.

View larger version (14K): [in this window] [in a new window]   Figure 1.   Examples of tidal and forced flow-volume curves obtained in this study. (a) Results from an infant in whom flow limitation was evident even during tidal breathing and in whom V'maxFRC was 33 ml · s1. Note concave flow-volume loop. (b) Convex flow-volume curve from a baby with a V'maxFRC of 132 ml · s1.

V'max_FRC was higher in girls than boys (115 versus 94 ml · s¹); although this difference just failed to reach statistical significance (95% CI: 5; 47 ml · s¹). No significant difference was found between black and white infants with respect to V'max_FRC (95% CI of the difference: 12; 40 ml · s¹). Tidal volume and respiratory compliance (C_rs) were not influenced by ethnic group or gender. However, when compared with white infants, black infants had significantly lower respiratory resistance (R_rs) (95% CI: 2.9; 0.4 kPa · L¹ · s), shorter respiratory time constants (95% CI: 0.10; 0.00 s), higher respiratory rates (95% CI: 9; 16 breaths per minute), higher peak tidal expiratory flows (95% CI: 2; 16 ml · s¹), shorter expiratory times (95% CI: 0.18; 0.02 s), and a longer t_PTEF:t_E (95% CI: 0.06; 0.20). Girls tended to have a lower R_rs than boys (95% CI: 2.3; 0.2 kPa · L¹ · s), and had a significantly longer t_PTEF and t_PTEF:t_E (95% CI: 0.01; 0.11 s and 0.01; 0.15, respectively). However, they had similar values for C_rs, tidal volume and respiratory rate when compared with the boys. After adjusting for potential confounding factors (family history of asthma, wheezing, chest problems, hay fever, and eczema), gender and ethnic differences diminished by between 30-70% and were no longer significant at the 5% level.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

These results suggest that certain indices of respiratory function may be influenced by gender and ethnic group in infants delivered prematurely. The significance of these findings is strengthened by the fact that we studied a well-matched group of black and white preterm infants, without known respiratory disease, who were born to mothers who did not smoke during pregnancy.

Gender Differences

While just failing to reach statistical significance, the 95% CI suggested that a real gender difference in peripheral airway function may exist with V'max_FRC being 20% higher and R_rs 10% lower in girls than boys. In addition, t_PTEF:t_E was significantly higher in girls than boys, a finding that could not be attributed to differences in respiratory rate which was virtually identical in the two groups. In the present study, gender differences may have been minimized by selective recruitment to ensure equal representation of girls and boys in both ethnic groups, since only the healthiest of preterm infants who required no or minimal ventilatory assistance were recruited. Nevertheless, our findings of increased flows and reduced resistance in the girls is consistent with previous observations that airway function is diminished in boys compared with girls during both infancy and childhood (19, 20, 22, 26, 37). This may contribute to the higher prevalence of asthma and wheezing reported in boys compared with girls at all ages up to puberty (25).

Ethnic Differences

Evidence from older children and adults would argue against any persistent genetically determined enhancement of peripheral airway function among black subjects, since similar values for flows and volumes are found in black and white children and adults when related to sitting height (17). Nevertheless transient changes due to different patterns of lung maturation and breathing pattern shortly after birth, combined with ethnic differences in nasal anatomy and hence resistance, could explain the results from the current study. V'max_FRC is thought to reflect airway caliber upstream (peripheral) to the airway segment subjected to flow limitation, i.e., to give a measure of peripheral airway function that is relatively uninfluenced by the resistance of the upper airways (2). This makes it a particularly useful means of assessing airway function in infants in whom nasal resistance comprises such a large proportion of total resistance (11). However the caliber of the small intrathoracic airways is determined not only by their anatomical dimensions but by the structure and compliance of the airway wall and the distending pressures surrounding them. The latter is itself influenced by lung volume and the elastic recoil of the respiratory system.

Functional residual capacity (FRC) is known to vary between sleep states (29) and was not measured in the present study since unsedated preterm infants spend limited periods of time in quiet sleep, making it extremely difficult to obtain repeat measures of lung volume in addition to satisfactory measurements of V'max_FRC within the same epoch of quiet sleep. Although lung volumes in older children and adults are lower in black than white subjects when related to standing height (17, 18), these differences disappear when volumes are related to sitting rather than standing height. Similarly no ethnic differences in lung volume have been reported among full-term infants in the first year of life (11, 12). In the current study, we did not detect any differences in stature with respect to either trunk (crown-rump) or total (crown-heel) length, suggesting that changes in the trunk-leg ratio according to ethnic group are not present at birth and only become evident with increasing growth during the first years of life. However a recent study has demonstrated higher resting lung volumes both at birth and at one week of age in healthy preterm Afro-Caribbean infants compared with age and gender matched Caucasians (40). This may reflect a transient finding, reflecting more marked expiratory braking and hence dynamic elevation of lung volume among the black babies (see below), or advanced lung maturity with respect to surfactant production (6, 14).

The finding of the lower R_rs and _rs among black infants is in accordance with previous observations suggesting that airway and nasal resistance are lower in black than white infants due to differences in nasal anatomy (11). This may also have contributed to the observed differences in breathing pattern between the two ethnic groups. Throughout life, expiratory flow and timing are closely controlled, interacting with lung volume so that end expiratory lung volume is maintained close to a predetermined desirable level. Normal expiratory flow at rest is longer and slower than if the system were allowed to deflate passively. This is achieved both by some continuing contraction of the inspiratory muscles beyond inspiration and by laryngeal narrowing, both of which brake expiratory airflow during early expiration, resulting in a longer t_PTEF:t_E than would be recording during passive expiration. This braking mechanism is gradually withdrawn as ventilation increases during hypercapnia (41) and in the presence of airway disease (42). In contrast, it is particularly marked in newborn infants in whom it provides an important strategy for defending lung volume during the first weeks of life, when the highly compliant chest wall offers relatively little outward recoil to keep the lungs and airways distended (32, 43). With increase in postnatal age, the need to modulate expiratory flow and timing in order to maintain a stable FRC diminishes, and a reduction in t_PTEF:t_E from approximately 0.5 during the first weeks of life to around 0.3 (the average value reported for healthy children and adults) occurs by about 6 wk of age in healthy, full-term infants (32). Thus the values of t_PTEF:t_E found in the black and the female preterm infants were similar to those observed in healthy full-term infants of similar postnatal age, whereas those in the white and male infants were somewhat lower than anticipated for their postnatal age. This may reflect the higher respiratory resistance and longer passive expiratory time constant found in the white and male infants which would reduce their need for marked expiratory braking beyond the first few days of life.

We have previously reported that C_rs is lower (i.e., elastic recoil higher) among black than white preterm infants during the first week of life, which we postulated could indicate increased maturity in black infants as reflected by a less compliant (i.e., stiffer) chest wall (10). These findings have recently been confirmed by others (40). In the present study, this tendency was still evident, but not significant at the 5% level (95% confidence intervals of the difference [black-white infants]: C_rs,MOT, 0.32; 6.22 ml · kPa¹; C_rs,SBT 1.97; 5.17) which probably reflected the fact that the current population was studied at an older postnatal and postconceptional age, when some stiffening of the chest wall had already occurred. Weight corrected values of C_rs in both ethnic groups approached values that we and others have previously reported from full-term newborn infants (44).

Potential Confounding Factors

Since neonatal respiratory disease and its treatment may affect subsequent airway growth and development (45, 46) strict criteria were adopted for recruitment of preterm infants into this study. All infants were asymptomatic at time of testing, none had received surfactant therapy or a diagnosis of RDS, and only one infant (black male; V'max_FRC, 136 ml · s¹) received any ventilatory assistance, this being given at birth for five hours after which he was extubated to air. Furthermore infants were only recruited if their mothers reported no smoking during pregnancy (47). Because infants were recruited postnatally, direct biochemical validation of maternally reported smoking during pregnancy was not possible. However, in this group of babies, urinary cotinine in the first urine specimen ranged from non-detectable to 7.6 ng · ml¹ compared with values up to 258 ng · ml¹ in samples from infants whose mothers reported smoking during pregnancy (unpublished data). Nevertheless, despite only recruiting infants without known respiratory disease, prematurity itself may influence lung and airway development. In this study, seven infants showed evidence of flow limitation during tidal breathing. Similar observations have been reported in symptom free full-term infants shortly after birth, many of whom subsequently developed asthma (48), with reduced values of V'max_FRC also being reported shortly after birth in apparently healthy full-term infants of atopic parents (23, 48, 49).

The four ethnic and gender groups were extremely well matched with respect to all infant details (age, weight, length, etc.) and to maternal age, education and family history of asthma and wheezing. However, black infants were less likely to have a family history of hay fever or chest trouble (primarily reported as bronchitis). Possible explanations for these differences could include: (1) a reporting bias, i.e., that white mothers were more likely to report hay fever and bronchitis; (2) a true ethnic/gender difference in prevalence of these diseases in the families of preterm infants (although there is no evidence for this in the general population); or (3) sampling variation within this study.

After adjusting for these potential confounding factors, the differences in airway function attributable to either ethnic group or gender diminished markedly. It is possible that if parents were prone to bronchitis and respiratory problems because of congenitally small airways (50), their babies may in turn have inherited this tendency. However since the influence of family history of respiratory disease and atopy was not an a priori hypothesis, and sample size was not designed to address these issues, these findings must be interpreted with caution.

The findings of the present study have important implications when interpreting respiratory function results from infants with chronic lung disease. There have been several reports suggesting that prematurity per se, in the absence of neonatal respiratory disease or assisted ventilation may result in functional airway abnormalities in later childhood (45, 51, 52). There is also increasing awareness of the relevance of family history and the importance of intrauterine development for subsequent lung health, with emphasis on the need to understand both development physiology and the underlying pathophysiology of neonatal lung disorders if adult morbidity and mortality from respiratory disease are to be reduced (50). However, if the results of lung and airway function are to be used as outcome or explanatory variables in such studies, it is important that the potential influence of both gender and ethnic group are taken into account.