RESULTS

Section:

ChooseTop of pageAbstractMETHODSRESULTS <<DISCUSSIONReferencesCITING ARTICLES

Twenty-six premature infants had a mean (range) gestational age of 32.7 (30–34) weeks and mean (range) birth weight of 1.74 (0.96–2.68) kg. One or more doses of corticosteroids were administered to 18 mothers (69%) before delivery, 9 infants (35%) had intrauterine growth retardation, and 11 (42%) required oxygen for less than 48 hours. The demographics for the premature and full-term control groups are summarized in Table 1. There were no significant differences in sex, race, or exposure to tobacco smoking for the two groups (Table 1). At the time of the first assessment of lung function, the preterm group was significantly younger and smaller; however, they did not differ in their z scores for age-adjusted length and weight. At the time of their second evaluation, there were no significant differences in age or somatic size for the premature and full-term control infants.

Comparison of Lung Function for Premature and Full-term infants

At the first evaluation, there was not a significant difference in FVC for the two groups (Table 2); however, the infants born prematurely had significantly lower forced expiratory flows (FEF₅₀, FEF₇₅, and FEF_25–75), as well as a lower ratio of FEV_0.5/FVC. At the second evaluation, there was still no significant difference in FVC, whereas the preterm group had persistently lower FEF₇₅ and FEF_25–75, as well as a lower ratio of FEV_0.5/FVC. The changes in spirometric values between the two evaluations for the premature and control infants are shown in Figure 1. There were no significant differences for the two groups in the changes in FVC, forced expiratory flows, or the ratio of FEV_0.5/FVC.

Figure 1. Plots of FVC, FEF₅₀, FEF₇₅, FEF_25–75, FEV_0.5, and FEV_0.5/FVC adjusted for length, sex, and smoke exposure at tests 1 and 2 for the control group (solid lines, open circles) and preterm group (dotted lines, solid circles) expressed as least squares mean ± SE. *P < 0.05 between groups at that test (analysis of covariance). No difference detected in the rate of change of all tested variables between the groups.

Download Figure | Download Powerpoint

[More] [Minimize]

TABLE 2. LUNG FUNCTION IN THE PRETERM AND CONTROL GROUPS AT THE FIRST AND SECOND LUNG FUNCTION TEST AND THE CHANGE IN LUNG FUNCTION VARIABLES BETWEEN THE TESTS^*

	Adjusted Least Squares Means							Coefficients of the Variables in Model
	Preterm	Control	95% CI	P	Length	Sex	Smoke
Test 1
FVC	230 ± 10	234 ± 11	−29 to 37	0.814	13 ± 1†	−4 ± 15	−31 ± 17
FEF50	404 ± 26	480 ± 27	−157 to 6	0.068	15 ± 3†	46 ± 36	48 ± 41
FEF75	185 ± 16	242 ± 17	−109 to −6	0.030	10 ± 2†	60 ± 23†	26 ± 26
FEF25–75	346 ± 24	435 ± 25	−165 to −13	0.022	14 ± 3†	51 ± 33	32 ± 38
FEV0.5	179 ± 8	191 ± 9	−39 to 15	0.371	9 ± 1†	7 ± 12	−6 ± 13
FEV0.5/FVC	0.78 ± 0.02	0.86 ± 0.02	−0.14 to −0.02	0.006	−0.01 ± 0.00†	0.06 ± 0.03†	0.06 ± 0.03†
Test 2
FVC	513 ± 14	494 ± 15	−22 to 60	0.364	17 ± 2†	−31 ± 22	−24 ± 24
FEF50	731 ± 32	816 ± 34	−179 to 10	0.078	19 ± 5†	50 ± 52	−107 ± 56
FEF75	360 ± 19	428 ± 20	−124 to −13	0.017	9 ± 3†	35 ± 30	−31 ± 33
FEF25–75	649 ± 29	746 ± 30	−182 to −12	0.027	17 ± 4†	40 ± 46	−85 ± 51
FEV0.5	364 ± 11	379 ± 11	−47 to 17	0.348	11 ± 2†	1 ± 17	−31 ± 19
FEV0.5/FVC	0.72 ± 0.01	0.78 ± 0.01	−0.10 to −0.03	0.001	0.00 ± 0.00†	0.05 ± 0.02†	−0.03 ± 0.02
Change in lung function
FVC	273 ± 14	272 ± 14	−42 to 45	0.949	15 ± 1†	−42 ± 18†	11 ± 22
FEF50	302 ± 33	362 ± 35	−166 to 46	0.263	22 ± 4†	−6 ± 45	−149 ± 53†
FEF75	169 ± 20	193 ± 21	−87 to 39	0.450	12 ± 2†	−20 ± 27	−57 ± 31
FEF25–75	283 ± 30	333 ± 31	−144 to 44	0.289	20 ± 3†	−16 ± 40	−113 ± 47†
FEV0.5	177 ± 11	197 ± 11	−54 to 14	0.238	11 ± 1†	−13 ± 14	−23 ± 17
FEV0.5/FVC	−0.08 ± 0.02	−0.07 ± 0.02	−0.07 to 0.05	0.783	0.00 ± 0.00†	−0.02 ± 0.02	−0.09 ± 0.03†

Definition of abbreviation: CI = confidence interval.

Results are presented as least squares means ± SE. Variables included in the model: Length (cm), sex (male = 0, female = 1), smoke exposure (no = 0, yes = 1).

^*Analysis of covariance adjusted by length, sex, and smoke exposure during pregnancy.

^†P < 0.05 in the model.

DISCUSSION

Section:

ChooseTop of pageAbstractMETHODSRESULTSDISCUSSION <<ReferencesCITING ARTICLES

Our longitudinal evaluation of healthy infants born prematurely between 30 and 34 weeks' gestation demonstrates that these subjects have decreased airway function not only in the first few months of life but also at the 1-year follow-up evaluation in the second year of life. The decreased forced expiratory flows occurred in presence of normal forced vital capacities. These findings suggest that healthy infants born prematurely may have smaller sized airways relative to their lung volume. An alternative explanation for lower forced expiratory flows would be more compliant airways, lower pulmonary elastic recoil, or increased bronchial tone. The increase in spirometric values with growth was the same for infants born prematurely and infants born at full term. Lung function tracked somatic growth in the infants born prematurely; there was no evidence for catch-up growth early in life. These findings suggest relatively healthy infants born prematurely have persistently lower airway function than infants born at full term. The reduced forced expiratory flows in healthy preterm infants may be a factor that contributes to their increased risk of recurrent respiratory illnesses early in life (14).

We found that infants born prematurely had normal forced vital capacities at both evaluations. These findings suggest that, early in life, lung size is increasing appropriately with somatic growth after premature birth. Absolute lung volumes were not measured in our infants, so we cannot be sure that the premature infants do not differ in residual volume and total lung capacity. However, long-term follow-up studies of older children who had premature birth without significant respiratory problems in the neonatal period also have normal values for FVC, as well as normal values for residual volume and total lung capacity (1, 2). Although lung volumes may not differ in subjects born prematurely, it is still not known whether alveolar development (number and size) is affected by premature birth.

Our group of premature infants had persistently lower forced expiratory flows at both evaluations, and there was no evidence for catch-up during this period of time. The lower size-adjusted flows were not secondary to relatively smaller somatic size. The preterm and full-term infants had similar z scores for weight and length at both tests; therefore, the reduced lung function was not a consequence of poor growth. In addition, the reduced flows were not secondary to a smaller lung volume, because FVC was not different for the two groups; however, the ratio of FEV_0.5/FVC was persistently lower in the infants born prematurely compared with those infants born full term. Our finding of lower forced expiratory flows in the second year of life is consistent with the observations by Hoo and colleagues (5). However, these investigators found that, at their initial evaluation in the neonatal period, the infants born prematurely had normal values for max_FRC; therefore, their group of infants born prematurely exhibited a significant worsening of airway function when assessed longitudinally. This contrasts with our findings of persistently lower forced expiratory flows at both evaluations and no change in the rate of increase of lung function when compared with full-term infants. The difference in the findings of these two longitudinal studies may be secondary to Hoo and coworkers who initially assessed infants at a younger age than we did. In addition, those investigators assessed airway function using partial forced expiratory maneuvers, which reference flows to FRC, a variable lung volume during tidal breathing, particularly in neonates.

The mechanism for the persistently lower airway function in infants born prematurely has not been determined. We did not assess airway function using a bronchodilator; therefore, increased bronchial tone could potentially contribute to the lower forced expiratory flows in the infants born prematurely. In addition, lower forced expiratory flows in the infants born prematurely could result from a decrease in pulmonary elastic recoil secondary to abnormal alveolarization of the lung parenchyma, as well as more compliant airways. Alternatively, the premature infants could have airways that are anatomically smaller relative to their lung volume and somatic size. Dysanaptic growth occurs between the airways and the lung parenchyma for full-term infants; lung volume increases faster than airway size (6, 9). Compensatory lung growth after pneumonectomy appears to accentuate this growth pattern. After pneumonectomy, the lung parenchyma exhibits catch-up of compensatory growth; however, the airways do not exhibit comparable increases in growth, and thus the airways are small relative to lung volume (15). Similarly, older children raised at high altitude have larger lung volumes and lower forced expiratory flows than subjects residing at sea level. The infants born prematurely are physically smaller and will have smaller lungs at birth than infants born at full term. Premature extrauterine life may accelerate growth of the lung parenchyma, but not the airways; this could result in a normal lung volume, but airways that are small relative to the size of the lung.

Another interesting observation was the association of oxygen use and lung development. In the preterm group, lung growth appears to be faster in those infants that required supplemental oxygen in the first hours of life. We were unable to find evidence to support the concept that a brief exposure to oxygen could trigger accelerated lung growth. Our observations support the hypothesis of accelerated lung maturation being associated with worse respiratory prognosis (16–18). Among prematurely born infants, supplemental oxygen may be a marker of a less mature lung, which may have a better long-term respiratory prognosis than infants with accelerated maturation due to prenatal events such as infection and inflammation but no oxygen requirement.

Although there were no differences between the infants born prematurely and the full-term infants for the percentage of males or exposure to tobacco smoking during pregnancy, we adjusted for these two factors in our analysis, because these variables have previously be found by us, as well as other investigators, to be important determinants of lung function of infants (4, 9, 19). Among our infants, exposure to tobacco smoking was associated with lower values for several parameters of forced expiratory flows, which agrees with previous studies for both premature and full-term infants (2, 9, 19). We also found that the effect of smoking exposure was present in the rate of change of airway function between the two evaluations (FEF₅₀ and FEV_0.5/FVC), which suggests a long-lasting effect of smoke exposure on lung growth. Presence of smoke exposure during pregnancy relied on information provided by the mothers and may underestimate the actual exposure. We also found among our subjects that male sex was associated with lower forced expiratory, which is consistent with previous reports in premature and full-term infants (9, 20, 21).

Several cross-sectional studies of school-aged children have reported that, in the absence of chronic lung disease of infancy, premature infants have normal spirometric values (1, 2, 22, 23). Our finding of persistently lower forced expiratory flows in the second year of life in infants born prematurely suggests that these infants do not normalize their airway function early in life, during the most rapid period of lung growth. A late recovery of airway function, if it occurs, may explain the increased respiratory symptoms reported for infants born prematurely (14, 22, 24).

One potential limitation of this study is that the subjects were recruited and tested in different centers. However, we believe this had no impact in the observed differences between groups because the testing equipment, software, and technician were the same at both study sites. Furthermore, healthy control subjects previously recruited and tested in both centers have shown similar results (4), validating the comparisons performed in this study (see Figures E1–E4 and Table E1 of the online supplement).

In summary, our study addressed the effects of prematurity on lung function using a very specific population—that is, those who did not require sustained respiratory support in the first days of life. This population of relatively healthy premature infants, although not very premature, enabled us to separate the effect of preterm birth from the significant additional effects of lung injury secondary to mechanical ventilation or prolonged exposure to high concentrations of oxygen. Our findings suggest that the increase in lung function during the first year of life was proportional to somatic growth and similar for premature and full-term infants. In addition, throughout the first year of life, premature infants have persistently reduced airway function in the presence of normal lung volumes. The persistence of reduced expiratory flows in “healthy preterm infants” may contribute to their increased risk of recurrent respiratory illnesses early in life. Future studies are required to determine the mechanisms for the persistently reduced flows in healthy infants born prematurely, as well as to evaluate the effects of even greater prematurity on lung growth.