Prenatal glucocorticoid plus T4 treatment of fetal sheep results in improvements in oxygenation, gas exchange, lung mechanics, and lung volumes after preterm delivery. We have evaluated the morphometric changes in the lungs of lambs exposed to betamethasone and T4 48 h before preterm delivery at 121 and 135 d gestation and related those changes to the physiologic improvements in lung function. The lungs used for the morphometric studies were from lambs with postnatal physiologic responses similar to those of the entire group of lambs reported previously (16). At both 121 and 135 d gestation, lung gas volumes and fixed tissue volumes increased, the percent of collapsed (nonaerated) parenchyma decreased, and the percent of perilobular connective tissue decreased with both gestational age and prenatal hormone exposure. Alveolar size, as estimated by mean linear intercept length, did not change with gestation or hormone exposure, but there was a decrease in alveolar wall thickness with advancing gestation and at each gestation with hormone exposure. The major anatomic effect of prenatal hormone exposure was a decrease in alveolar wall thickness and an increase in aerated parenchyma, effects that were consistent with the physiologic improvements in postnatal lung function.
Subsequent to the observation of Liggins (1) that fetal cortisol infusions caused preterm delivery and increased aeration of the preterm lamb lung, descriptive reports of accelerated morphologic lung development and maturation of type II cells after fetal exposure to glucocorticoids or thyroid hormones appeared (2, 3). Kauffman (4) reported that maternal dexamethasone treatments could increase the volume density of air spaces within 14 h of exposure in preterm mice. Morphometric evaluations after infusions of thyrotropin-releasing hormone, cortisol, and ACTH in various combinations into fetal sheep demonstrated modest increases in lung volumes and air-space volumes, although the effects were often not significant (5-7). These studies were of low resolution because the gestational ages were not uniform within the study groups (6), and fetal catheterization can alter the fetal response to hormone infusions (8). High quality morphometric evaluations 48 h after fetal exposure of rabbits to gluccorticoids demonstrated increased air space and increased maturation and number of type II cells, and more trans-basement membrane epithelial-endothelial contacts (9). Prenatal dexamethasone treatment of pregnant rats resulted in a decreased alveolar surface area at term and a decrease in alveolar number, but not size, in male fetuses (10). However, in these studies with rabbits and rats, prenatal glucocorticoid treatments were associated with global fetal growth retardation and concerns about the specificity of the glucocorticoid effects on the fetal lung (11).
It is remarkable that despite the clinical use of maternal glucocorticosteroids to induce fetal maturation since 1972 (12), and the recent evaluations of thyrotropin hormone to augment the corticosteroid effect (13, 14), there are no reports that directly correlate morphometric effects of maturational agents with physiologic assessments of postnatal lung function. We reported that combination fetal treatment with a single dose of 0.5 mg/kg estimated fetal weight betamethasone and 15 μg/kg T4 followed by delivery at 128 d gestation (term = 150 d) resulted in postnatal lung function that was better than for lambs exposed to betamethasone alone (15). This combination of betamethasone and T4 also enhanced postnatal lung function after preterm delivery at 121 and 135 d gestation (16, 17). To better understand the relationships between postnatal lung function and anatomic effects of maturational agents, we morphometrically evaluated the lungs of the preterm lambs delivered at 121 and 135 d gestation 48 h after fetal exposure to betamethasone and T4.
The fetal treatments and postnatal assessments were reported in detail previously (16, 17). Fetal treatments were given by intramuscular injection using fetal ultrasound to image the fetus (18). Each randomized fetus received 0.5 mg/kg estimated fetal weight betamethasone (Celestone Soluspan®; Schering Corp., Kenilworth, NJ) and 15 μg/kg T4 (Sigma Chemical Co., St. Louis, MO) diluted to a volume of 2.5 ml in saline. Control fetuses received the same volume of saline. The preterm lambs were delivered 48 h after fetal treatment at 121 or 135 d gestation by cesarean section. An endotracheal tube was secured in the fetal trachea, and each lamb was delivered, weighed, and ventilated with a time-cycled, pressure-limited infant ventilator (18). Respiratory rate was 40 breaths/min, positive end-expiratory pressure (PEEP) was 3 cm H2O, inspiratory time was 0.7 s, peak inspiratory pressure (PIP) was initially 35 cm H2O. For the 40-min postnatal evaluation period, oxygen was kept at 100%, and only peak inspiratory pressure was changed to maintain Pco 2 within the physiologic range when possible. PIPs greater than 40 cm H2O were not used to avoid pneumothorax.
Blood gas and pH measurements were made frequently for the 40-min postnatal evaluation period. Compliance was calculated by dividing the tidal volume by the ventilatory pressure (PIP-PEEP) and by weight in kilograms (18). Maximal lung volumes were measured by filling the collapsed lung with air to 30 cm H2O (121 d gestation) or 40 cm H2O (135 d gestation) pressure for 1 min followed by a volume measurement (16). The lower pressure was used for more immature lambs to avoid lung rupture. The deflation limbs of pressure-volume curves were reported previously (16).
After the 40 min postdelivery ventilation period, the right cranial lobe was removed together with its bronchial insertion into the trachea. The bronchus was cannulated and the lobe was inflation-fixed with Karnovsky's solution (0.9% paraformaldehyde, 0.6% glutaraldehyde, cacodylate buffer at pH 7.4) at a constant distending pressure of 30 cm H2O overnight. Similar distending pressures and fixation times were used to ensure that all lungs would be inflated under identical conditions for each gestational age and treatment. The bronchus was tied off and the lung tissue was stored in fixative at 4° C until processed.
The volume of the right cranial lobe was determined by the displacement method of Scherle (19). The entire lobe was cut into serial slices 1 cm thick. One cut surface from each slice was examined using a dissecting microscope with an eyepiece graticule containing 21 uniformly spaced lines. The entire cut surface of each slice was scanned. The number of times the ends of each line fell on lung parenchyma, airways, or blood vessels was recorded to determine the proportion of parenchymal and nonparenchymal tissue. An average of nine to 12 slices was counted for each right cranial lobe.
For further microscopic and morphometric analysis of lung tissues, the method of Cavalieri (20) was used to select three uniformly spaced slices from each cranial lobe. Using a random number generator, the first slice was selected, and the other two slices were taken at equal distances from each other. Each slice was subsequently cut to a thickness of 2 mm, dehydrated in a graded series of ethanol and toluene, and embedded in paraffin. Sections 5 μm thick were cut using a rotary microtome and stained with hematoxylin-eosin (H&E). Each stained section was enlarged and printed on 11-by-14-inch photographic paper at a magnification ×16. An acetate overlay grid consisting of 462 lines and 924 points was overlaid on each of the photographic images and tissues falling under the ends of each line were counted using the following categories: nonparenchyma (consisting of blood vessels greater than 50 μm in diameter and conducting airways), perilobular connective tissue (forming the distinct lobulation of the lung), presumptive aerated parenchyma (consisting of alveoli and ducts in an inflated condition) and nonaerated parenchyma (collapsed alveoli and ducts).
Those proportions of the lung parenchyma that were not collapsed were further evaluated for characteristics of alveolar septal wall thickness and alveolar air-space volume using the morphometric methods described by Bolender and coworkers (20), and Pinkerton and Crapo (21). Ten nonoverlapping fields composed predominantly of parenchymal tissues were randomly selected for analysis. Each field was examined at a final magnification ×380 as a digital image, captured using a video camera interfaced to the microscope and a Macintosh 2CI computer. Each field was overlaid with a test lattice grid consisting of 21 lines, each 2.25 cm long. Structures falling under the ends of each line were tallied as air-space or alveolar tissue. Intercepts of the test lines with the air-tissue interface were also counted. Alveolar air-space and tissue volumes were determined using the formula Vv = Pp = Pn/Pt, where Pp is the point fraction of Pn, the number of test points hitting the structure of interest, divided by Pt, the total points hitting the reference space (i.e., parenchyma). The surface area, represented by the alveoli and alveolar ducts was determined by point and intercept counting using the formula Sv = 2I0/Lr, where I0 is the number of intercepts with the air-tissue interface, and Lr is the length of the test line within the reference volume being the alveolar parenchymal fields. To determine the thickness of the septal wall as a volume per unit area of alveolar surface, the following formula was used: T = Vv/Sv.
Differences between groups at each gestation were evaluated by two-tailed t tests. Comparisons across gestational ages were by analysis of variance followed by the Student-Neuman-Keuls multiple comparison procedure. Values are given as mean ± SEM, and significance was accepted at p < 0.05.
Birth weights, cord blood gas values, and wet to dry lung weights for all animals were reported previously (16). There were nine control lambs and 11 betamethasone plus T4-treated lambs studied at 121 d gestation. At 135 d gestation, eight control and nine betamethasone plus T4-treated lambs were evaluated for postnatal lung function. Five lungs for each group of lambs were used for the morphometry. At 121 d gestation prenatal hormone treatment resulted in significant increases in Po 2 for both the total group and for the lambs that provided the five lungs used for morphometry (Figure 1). Although compliance increased for the five hormone-exposed lungs relative to the five control lungs used for morphometry, the difference was not significant. Differences in these assessments of lung function were significant for the overall group of 121-d gestation lambs, and the same pattern occurred at 135 d gestation. Comparisons of the Po 2 values and compliances for all the lambs and for the subset used for morphometry demonstrate that the lungs used for morphometry were representative of the overall groups.
Maximal gas volume was measured after 40 min of ventilation after preterm birth (Figure 2). Lung volumes of the fixed right cranial lobes were measured by displacement. At both gestations, gas volumes and fixed tissue volumes increased significantly after hormone exposure for the total groups. The differences were similar in magnitude but not always significant for the lungs from each group used for morphometry. Gas volume increased 8-fold for control lambs between 121 and 135 d gestation. However, the increase in fixed tissue volume of the right cranial lobe was only 20% from 9.2 ml/kg at 121 d gestation to 11.1 ml/kg at 135 d gestation (p < 0.03).
The percent of the right cranial lobe composed of parenchyma was about 68% for control and betamethasone plus T4 treatment groups at 121 d gestation (Figure 3). The percent parenchyma increased to 78% for control lambs at 135 d gestation (p < 0.01 versus 121 d control lambs), and increased further to 85% with betamethasone plus T4 treatment at 135 d gestation (p < 0.01). Because all lungs were inflated at a constant pressure and for identical periods of time, we were able to define the proportion of the parenchyma that had become filled with fixative and those parts that remained closed and collapsed. We have assumed that the lung volumes that filled with fixative were likely to have been aerated (Figure 4). The percent of parenchyma that was not aerated decreased from 28% for control lungs to 4% for hormone-exposed lungs at 121 d gestation (p < 0.01). The percent of parenchyma not aerated decreased to only 1% for hormone-exposed lungs at 135 d (p < 0.03 versus control lungs). The percent of perilobular connective tissue interposed between parenchymal tissues of the right cranial lobe was about 16% in 121 d gestation lambs, and fetal hormone exposure did not influence this measurement. In contrast, there was a decrease in the volume of perilobular connective tissue to about 9% of parenchyma at 135 d gestation (p < 0.05 versus 121 d control lambs), and a further decrease to 5% of parenchyma after fetal treatment with betamethasone plus T4 (p = 0.08).
Higher magnification light micrographs of lungs from lambs at 121 and 135 d gestation demonstrated alveoli of similar size and shape (Figures 5 and 6). The mean linear intercepts measured for control and betamethasone plus T4-treated lungs were about 50 μm, independent of gestation or treatment groups (Figure 7). However, measurements of alveolar wall thickness demonstrated a 36% decrease between 121 and 135 d gestation and a decrease of about 25% with hormone exposure at each gestational age.
We evaluated postnatal lung function after preterm treatment with betamethasone and T4 at two gestational ages and quantified the structural changes in lung parenchyma and alveoli in lungs from a subgroup of the same animals. At 121 d gestation hormone exposure resulted in large improvements in oxygenation, lung mechanics, and lung gas volume. The fixed lung tissue volume increased a small amount. The large increase in lung gas volume is consistent with the large decrease in nonaerated parenchyma and the thinning of the alveolar walls. Because alveolar size (mean linear intercept) did not change and fixed tissue volume changed minimally, there was no change in alveolar numbers or potential surface area of the lung. The fetal sheep lung at 120 d gestation is more mature than rodent lungs at term from the perspective of alveolar development (22, 23). By 114 d gestation alveoli are evident, and the increased lung volume with subsequent fetal development occurs by addition of alveoli of equal size (23). The traditional explanation for the decreased nonaerated parenchyma would be an increase in surfactant. However, surfactant lipids do not increase within 48 h of prenatal hormone treatment in this lamb model (15, 16, 18). Therefore, prenatal exposure to hormones improved the micromechanics of the lung to promote aeration of the severely surfactant-deficient 121-d gestation lung (17). Prenatal corticosteroids alter collagen, elastin, and probably other matrix components (24). The striking thinning of the alveolar wall requires rapid remodeling of matrix and interstitial components. The thinning of the barrier together with the decrease in nonaerated parenchyma are consistent with the improved gas exchange.
The results were similar at 135 d gestation but were less striking because the amount of nonaerated parenchyma already was low. Nevertheless, the alveolar wall thinned, lung gas volume increased, and lung mechanics improved. As for the 121-d gestation lungs, there was no evidence to suggest a change in alveolar number or size. A new observation at 135 d gestation is the decrease in perilobular tissue with hormone exposure. The preterm lung has prominent intralobular septae that become less apparent with maturation, and that process was accelerated with prenatal hormone exposure.
Several investigators have reported that midgestation fetal exposure of monkeys to high dose and/or long interval treatments with glucocorticoids can result in lungs that have fewer and larger alveoli later in gestation or at term (25, 26). Prenatal exposure of rats to glucocorticoids results in altered alveolarization postnatally (10), and postnatal exposure can interfere with septation (27). Postnatal exposure of newborn rats also can accelerate the normal maturational fusion of capillaries to become a single capillary layer in alveolar walls (28). Our results with single low dose exposures that are comparable to the clinical use of glucocorticoids did not effect alveolarization within 48 h of fetal exposure. We have reported similar improvements in postnatal lung function for treatment to delivery intervals of 15 h to 7 d (29, 30). This rapid and persistent response is more consistent with a rapid remodeling of the lung interstitium than with an effect on alveolarization, which is thought to be a slow and progressive process. The difference in these responses in the preterm sheep and rodent models results in part from the more structurally mature lung of the fetal sheep with alveoli by 114 d gestation and no apparent double capillary networks (23). The human lung lies somewhere between the rodent and the sheep in terms of the timing of alveolarization (31).
There probably are two roles for glucocorticoids in the developing lung. A complete lack of glucocorticoid results in a maturational arrest at late gestation characterized by no thinning of alveolar walls in mice that are corticotropin-releasing hormone– or glucocorticoid-receptor-deficient (32, 33). The maturation arrest is at the point when glucocorticoids increase spontaneously in the fetus (34). Exposure of the fetus earlier in gestation to glucocorticoid levels comparable to those normally achieved at term results in accelerated or early thinning of alveolar walls. Effects on other aspects of lung development are less consistent (8), and the effects on lung structure are sufficient to explain the physiologic improvements in postnatal lung function.
The lambs were treated with betamethasone plus T4 because we previously noted a modest augmentation in postnatal lung function with the addition of T4 (15). The dose of betamethasone was selected to be the minimal effective dose to consistently give improved postnatal lung function (18). The responses differ from another report that found no effect on alveolar wall thickness in monkeys despite four fetal treatments with betamethasone over 7 d at a higher dose (35). We found very low cortisol levels in cord blood, and no effect of the fetal treatment on fetal catecholamine levels was detected (16, 36). We believe that these experiments with fetal sheep evaluated the “baseline” response to hormones rather than hormonal responses that may be altered by fetal stress (8). The contribution of T4 in these experiments is not known, but the physiologic responses are similar to other results with betamethasone alone (15, 18). The challenge for the future is to learn how prenatal corticosteroids can rapidly alter alveolar wall thickness.
|1.||Liggins G. C.Premature delivery of foetal lambs infused with glucocorticoids. J. Endocrin.451969515523|
|2.||Kotas R. V., Avery M. E.Accelerated appearance of pulmonary surfactant in the fetal rabbit. J. Appl. Physiol.301971358361|
|3.||Kikkawa Y., Kaibara M., Motoyama E. K., Orzalesi M. M., Cook C. D.Morphologic development of fetal rabbit lung and its acceleration with cortisol. Am. J. Pathol.641971423442|
|4.||Kauffman S. L.Acceleration of canalicular development in lungs of fetal mice exposed transplacentally to dexamethasone. Lab. Invest.361977395401|
|5.||Boshier D. P., Holloway H., Liggins G. C., Marshall R. J.Morphometric analyses of the effects of thyrotrophin releasing hormone and cortisol on the lung of fetal sheep. J. Dev. Physiol.1219894954|
|6.||Kendall J. Z., Lakritz J., Plopper C. G., Richards G. E., Randall G. C. B., Nagamanj M., Weir A. J.The effects of hydrocortisone on lung structure in fetal lambs. J. Dev. Physiol.131990165172|
|7.||Kendall J. Z., Rivera-Alcina M. E., Plopper C. G., Weir A. J., Randall G. C.A morphometric study of lung maturation in the adrenocorticotrophin-infused fetal lamb. Biol. Neonate601991258272|
|8.||Tabor B. L., Lewis J. F., Ikegami M., Polk D., Jobe A. H.Corticosteroids and fetal intervention interact alter lung maturation in preterm lambs. Pediatr. Res.351994479483|
|9.||Snyder J. M., Rogers H. F., O'Brien J. A., Mahli N., Magliato S. A., Durham P. L.Glucocorticoid effects on rabbit fetal lung maturation in vivo: an ultrastructural morphometric study. Anat. Rec.2321992133140|
|10.||Massaro G. D., Marraro D.Formation of alveoli in rats: postnatal effect of prenatal dexamethasone. Am. J. Physiol.2631992L37L41|
|11.||Tabor B. L., Rider E. D., Ikegami M., Jobe A. H., Lewis J. F.Dose effects of antenatal corticosteroids for induction of lung maturation in preterm rabbits. Am. J. Obstet. Gynecol.1641991675681|
|12.||Crowley P.Antenatal corticosteroid therapy: a meta-analysis of the randomized trials. Am. J. Obstet. Gynecol.1731995322335|
|13.||Knight D. B., Liggins G. C., Wealthall S. R.A randomized, controlled trial of antepartum thyrotropin-releasing hormone and betamethasone in the prevention of respiratory disease in preterm infants. Am. J. Obstet. Gynecol.17119941116|
|14.||ACTOBAT Study GroupAustralian trial of antenatal thyrotropin-releasing hormone (ACTOBAT) for prevention of neonatal respiratory disease. Lancet3451995877882|
|15.||Polk D. H., Ikegami M., Jobe A. H., Newnham J., Sly P., Kohan R., Kelly R.Postnatal lung function in preterm lambs: effects of a single exposure to betamethasone and thyroid hormones. Am. J. Obstet. Gynecol.1721995872881|
|16.||Ikegami M., Polk D. H., Jobe A. H., Newnham J., Sly P., Kohan R., Kelly R.Postnatal lung function in lambs after fetal hormone treatment. Am. J. Respir. Crit. Care Med152199512561261|
|17.||Willet K. E., Gurrin L., Burton P., Lanteri C. J., Reese A. C., Matsumoto I., Jobe A. H., Ikegami M., Polk D., Newnham J., Kohan R., Kelly R., Sly P. D.Differing patterns of mechanical response to direct fetal hormone treatment. Respir. Physiol.1031996271280|
|18.||Jobe A. H., Polk D., Ikegami M., Newnham J., Sly P., Kohan R., Kelly R.Lung responses to ultrasound guided fetal treatments with corticosteroids in preterm lambs. J. Appl. Physiol.75199320992105|
|19.||Scherle W. F.A simple method of volumetry of organs in quantitative stereology. Mikroscopie2619705760|
|20.||Bolender R. P., Hyde D. M., Dehoff R. T.Lung morphometry: a new generation of tools and experiments for organ, tissue, cell and molecular biology. Am. J. Physiol.2651993L521L548|
|21.||Pinkerton, K. E., and J. D. Crapo. 1985. Morphometry of the alveolar region of the lung. In H. P. Witschi and J. D. Brain, editors. Handbook of Experimental Pharmacology, Vol. 75: Toxicology of Inhaled Materials. Springer-Verlag, New York. 259–285.|
|22.||Alcorn D. G., Adamson T. M., Maloney J. E., Robinson P. M.A morphologic and morphometric analysis of fetal lung development in sheep. Anat. Rec.2011981655667|
|23.||Docimo S. G., Crone R. K., Davies P., Reid L., Retik A. B., Mandell J.Pulmonary development in the fetal lamb: morphometric study of the alveolar phase. Anat. Rec.2291991495498|
|24.||Schellenberg J. C., Liggins G. C., Kitterman J. A., Lee C. C. H.Elastin and collagen in the fetal sheep lung. II. Relationship to mechanical properties of the lung. Pediatr. Res.221987339343|
|25.||Johnson J. W. C., Mitzner W., Beck J. C., London W. T., Sly D. L., Lee P. A., Kouzami V. A., Cavalieri R. L.Long-term effects of betamethasone on fetal development. Am. J. Obstet. Gynecol.141198110531061|
|26.||Bunton T. E., Plopper C. G.Triamcinolone-induced structural alterations in the development of the lung of the fetal rhesus macaque. Am. J. Obstet. Gynecol.1481984203215|
|27.||Massaro D., Teich N., Maxwell S., Massaro G. D., Whitney P.Postnatal development of alveoli. J. Clin. Invest.76199512971305|
|28.||Tschanz S. A., Damke B. M., Burri P. H.Influence of postnatally administered glucocorticoids on rat lung growth. Biol. Neonate681995229245|
|29.||Ikegami M., Polk D., Jobe A. H.Minimum interval from fetal betamethasone treatment to postnatal lung responses in preterm lambs. Am. J. Obstet. Gynecol.174199614081413|
|30.||Ikegami M., Polk D. H., Jobe A. H., Newnham J., Sly P., Kohan R., Kelly R.Effect of interval from fetal corticosteroid treatment to delivery on postnatal lung function of preterm lambs. J. Appl. Physiol.801996591597|
|31.||Langston C., Kida K., Reed M., Thurlbeck W. M.Human lung growth in late gestation and in the neonate. Am. Rev. Respir. Dis.1291984607613|
|32.||Muglia L., Jacobson L., Dikkes P., Majzoub J. A.Corticotropin-releasing hormone deficiency reveals major fetal but not adult glucocorticoid need. Nature3731993139146|
|33.||Cole T. J., Blendy J. A., Monaghan A. P., Krieglstein K., Schmid W., Aguzzi A., Fantuzzi G., Hummler E., Unsicker K., Schutz G.Targeted disruption of the glucocorticoid receptor gene blocks adrenergic chromaffin cell development and severely retards lung maturation. Genes Dev.9199516081621|
|34.||Kitterman J. A., Liggins G. C., Campos G. A., Clements J. A., Forster C. S., Lee C. H., Creasy R. K.Prepartum maturation of the lung in fetal sheep: relation to cortisol. J. Appl. Physiol.511981384390|
|35.||Edwards L. A., Read L. C., Nishio S. J., Weir A. J., Hull W., Barry S., Styne D., Whitsett J. A., Tarantal A. F., George-Nascimento C., Plopper C. G.Comparison of the distinct effects of epidermal growth factor and betamethasone on the morphogenesis of the gas exchange region and differentiation of alveolar type II cells in lungs of fetal rhesus monkeys. J. Pharmacol. Exp. Ther.274199510251032|
|36.||Newnham J. P., Polk D. H., Kelly R. W., Padbury J. F., Evans S. F., Ikegami M., Jobe A. H.Catecholamine response to ultrasonographically guided percutaneous blood sampling in fetal sheep. Am. J. Obstet. Gynecol.1711994460465|