PEDIATRICS Vol. 111 No. 4 April 2003, pp. 766-776
Collagen Scaffolding During Development and Its Deformation With Chronic Lung Disease
* Department of Pediatrics, Children’s Mercy Hospital, Kansas City, Missouri
University of Missouri-Kansas City School of Medicine, Kansas City, Missouri
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Objective. Infants with chronic lung disease (CLD) have an arrest of primary and secondary septation. We hypothesized that this may be related to damage or abnormal development of lung collagen secondary to positive pressure ventilation. Our aims were to identify the sites and quantity of collagen in control infants 22 to 72 weeks’ postconceptional age and compare these with infants with various degrees of severity of CLD.
Methods. The controls were 22 to 42 weeks’ gestation (n = 30), received minimal ventilator care, and died within 48 hours of birth, plus 5 term infants who died at 43 to 72 weeks’ postconceptional age from nonpulmonary causes. Infants who were 23 to 30 weeks’ gestation, were at risk for CLD, and lived 5 to 94 days (n = 33) were separated into 3 groups on the basis of respiratory score (score group; the integrated area under the curve of the average daily fraction of inspired oxygen x mean airway pressure [cm H2O] over the number of days lived). The score groups, <20, 20 to 69, and 70 to 500, related clinically to mild to moderate and severe lung disease. The lungs were tracheally perfused and formalin fixed. Total lung volume was determined by water displacement. The paraffin-embedded lung blocks were sectioned 5 µm thick, stained with Gomori’s reticulum stain, hematoxylin and eosin, and immunohistochemically for collagen IV. The parenchyma was point-counted, and the volume density of collagen was measured. The chord diameter of the peripheral airway saccules and alveoli was measured. Descriptive collagen data were assessed on en face 40-µm-thick sections through the alveolar or saccular walls on all infants at risk for CLD and in selected controls.
Results. In the controls, the volume density of collagen decreased from a maximum of 9% at 22 weeks to 5% at term and 72 weeks. With Scores 
69, the fraction of collagen was similar to controls, but in infants with scores 70 to 500, it was increased relative to controls. However, when collagen was expressed as the volume density of interstitial tissue, ie, excluding parenchymal air space, it increased from a low of 5% at 22 weeks to 25% at 72 weeks. In infants with scores 70 to 500, 79% of infants had collagens greater than controls. Saccular and alveolar diameter increased from 40 µm at 23 weeks to 100 µm at 72 weeks. Most infants with severe CLD (scores 
70) had diameters more than twice that of controls at the same age. The total lung parenchymal collagen had a similar pattern as the volume density of collagen in interstitial tissue, increasing from 0.4 cm3 at 23 weeks to 9.7 cm3 at 72 weeks in the controls. Eighty-five percent of infants with scores 70 to 500 had total parenchymal collagen greater than the 95% confidence interval of the controls. With en face sections, a fine collagen mesh was seen at 23 weeks, which progressively increased in fiber size and quantity until 72 weeks. With severe CLD, the secondary collagen fibers in the saccular wall were thickened, tortuous, and disorganized relative to same-aged controls. Under 30 weeks, in the controls, the interstitium contained a wide, delicate network of interconnected collagen fibers. After positive pressure ventilation, some saccules markedly increased their diameter, which compressed and obliterated the interstitial network. In contrast with severe CLD, the interstitium was wide, with coarse wavy collagen fibers.
Conclusions. Parenchymal collagen increases throughout development. Before 30 weeks, there is a delicate complex interstitial collagen network, which may be important for primary septation and subsequent normal development. Positive pressure ventilation, if excessive, and depending on lung maturity and disease state, over a short time can severely compress the interstitium and damage this collagen network and prevent normal primary septation and arrest or distort future lung development. With severe CLD, distal air space diameter increases. There is a failure of primary and secondary septation, arrested lung development and remodeling, with thickened collagenous saccular walls, and a wide interstitium with increased quantity and size of collagen fibers that can affect the mechanics of ventilation. We conclude that normal lung development is dependent on a normal interstitium and, perhaps, collagen architecture and that origins of CLD begin early in the course of positive pressure ventilation.
Key Words: lung collagen • chronic lung disease
Abbreviations: GA, gestational age • PCA, postconceptional age • CLD, chronic lung disease • VVCOLL, volume density of collagen tissue • BPD, bronchopulmonary dysplasia
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The elastic theory of lung acinar development hypothesizes that saccular development is guided by the preprogrammed location of an elastic network. This network includes deposits of elastic fibers in a ring at the future saccular mouths.1,2 Secondary septation and alveogenesis, a later process by which saccules transform into alveoli, also involves elastic fibers.3,4 Indeed, without elastic fibers, distal airway branching will not develop normally and the lung becomes emphysematous.5 In the mature lung, the elastic tissue network is a critical mechanical structure supporting lung function during inspiration and expiration. During in utero development, parenchymal elastic tissue progressively increases in quantity, from very low levels at 22 weeks’ gestation to substantial levels at term.6 These reports support the concept that elastic tissue, before 30 weeks’ gestation, is more a map for acinar development than a force-bearing tissue.
Transpulmonary pressures in utero are low, of the order of 3 to 4 cm H2O, and seem to be tightly regulated.7,8 The intra-airway pressure is modulated by the production of fluid by epithelial cells, contraction of airway muscle, and an upper airway pressure control mechanism.7,9 The elastic network theory proposes that acinar elastic fibers anchor growing lung tissues at preprogrammed anatomic points and that airway liquid pressure forms saccules by forcing tissue between the anchors.2 The control of depth, size, and direction of saccular growth assumes that another growth-controlling mechanism and a counteracting tissue force exists. There are very few elastic fibers in the saccular walls and essentially none in the interstitium in fetuses in the pseudoglandular and canalicular stages, so it is not likely that elastic tissue is the controlling counter force. In contrast, lung collagen is thought to be more abundant during the canalicular stages of development.10 Because of its tensile strength and volume-limiting functions, it is a potential interstitial or septal counteracting force during acinar saccular development. The site-specific quantity of collagen and the nature of its network design have not been assessed in human development. The absence of collagen I, a major acinar collagen, during gestation is fatal.11
The ex utero development of the preterm lung is confounded by the use of mechanical assist devices that deliver pressures, at times, far in excess of in utero pressures. The immature lung elastic tissue network is acutely deformed and perturbed by pressure assist devices and chronically responds with an increase in elastic tissue.6 The collagen network can also be anticipated to be stressed by positive pressure devices, a process that may have important implications for preterm ex utero lung development. The aims of this study were to examine the location and quantity of parenchymal collagen during fetal lung development and parenchymal collagen in preterm infants treated with various amounts of positive pressure ventilation and inspired oxygen.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The 30 control infants (gestational age [GA] range: 22–42 weeks) were appropriate weight for GA, received minimal ventilator care, and died within 48 hours of birth (Table 1). Five additional infants, 58 ± 9.4 weeks’ postconceptional age (PCA) (range: 43–72 weeks), were born at term, lived 1 to 32 weeks, then died from nonpulmonary causes, receiving a maximum of 48 hours of minimal ventilator care. The criteria for selection of the controls were all of the infants who had postmortem examinations, were appropriate for GA, did not have congenital anomalies, had relatively normal lungs, and died within 48 hours during the period of the study. Infants were excluded when they had prolonged rupture of fetal membranes (>48 hours) or widespread bronchopneumonia or all lung lobes could not be inflated postmortem, leaked, or had extensive hemorrhage seen on lung histology. The cause of death of these 35 controls were perinatal acidosis and encephalomalacia (9); intracranial hemorrhage (6); sepsis (2); died in delivery room (10); and brain tumors, metabolic disease, disseminated intravascular coagulation, and renal failure accounting for the remainder. Those who died in the delivery room all were
23 weeks’ gestation, except for 1 who was 31 weeks.
|
Thirty-three infants were 25.7 ± 1.7 weeks’ GA (range: 23–30 weeks) and at risk to develop chronic lung disease (CLD). These infants died at PCA 29.3 ± 3.1 weeks, lived a mean of 24 days with a range of 5 to 94 days, and were ventilated for various durations (Table 1). These 33 infants were separated into 3 respiratory score groups—<20, 20 to 69, and 70 to 500—which were generally related clinically to mild, moderate, and severe lung disease, respectively, as previously shown.6 The score for each infant was determined by multiplying the average daily fraction of inspired oxygen by the average daily mean airway pressure, in cm H2O, and integrating the area under the curve, using the Trapezoidal rule, for the total number of days lived. In extubated infants, the mean airway pressure was arbitrarily set at 1 cm H2O.
Lung Preparation
The lungs, collected over an 11-year period (1991–2002), were perfused and fixed uniformly. The time from death to autopsy was 12 ± 11 hours. The lungs were weighed wet. The left lung was deflated by vacuum, warmed to 38°C, and tracheally inflated for 72 hours at 24 cm H2O pressure with 10% formalin. The left lung total lung volume was measured by water displacement. The total lung volume of both lungs was calculated by assuming a constant ratio between the wet weights and lung volumes. The left lower lobe was cut horizontally at 2-mm intervals from the upper to lower portion of lobe. A section was taken at random from each layer, usually 3 to 5 per lobe, depending on the size of the lung. In addition, 5 random sections were taken from other portions of the left lung.
After immersion in 10% buffered formalin, the area of the cut surface of each section was measured by computerized image analysis (Media Cybernetics, Silver Spring, MD). The area of the same section was measured again after processing, paraffin embedding, and staining to determine a shrinkage factor, which was used to correct areal and linear measurements. Sections 5-µm thick from the left lower lobe were stained with hematoxylin and eosin and Gomori reticulum stain.12 In addition, 40-µm-thick sections were obtained on all of the 33 infants with CLD (and selected control lungs) and stained with the reticulum stain to obtain en face views of the collagen in alveolar or saccular walls. Numerous section thicknesses between 10 and 75 µm were made, but 40 µm was the most useful. One section from each left lobe was immunohistochemically stained for collagen IV using a monoclonal antibody (Sigma, St Louis, MO).
With the use of point counting, the volume density of the parenchyma of the whole left lung was determined. The parenchyma included saccular or alveolar and alveolar ductal air space and septal tissue.13 Septal tissue included the interstitial tissue between saccules. The volume density of the parenchymal air space and interstitial tissue and parenchymal collagen tissue (VVCOLL) were then point-counted at 120x on the left lower lobe sections.14 Each field was brought into focus, and without refocusing during the count, the intersections on collagen tissue were recorded. A total of 160 sequential fields were assessed on each section. Fields that contained nonparenchyma or a blood vessel larger than a capillary were skipped.
Because the sections have a finite thickness, collagen fibers within the sample, as well as the surface fibers, can be projected onto the observational plane, thereby overestimating their volume density. Collagen fibers have a tubular shape, with the length many times longer than the diameter. A theoretical correction factor (k) has been derived to be k = 1/(1 + t/d), where t is the thickness of the observational depth and d is the average diameter of the collagen fibers.14 Focal depth, or thickness (t), was considered to be 0.2 µm. The collagen fiber, measured with the light microscope, varies in diameter from 0.5 µm to 1 µm. The calculated correction factor varied from 0.71 to 0.83 and reduced the observed volume density by 17% to 29%. The volume densities were not corrected because of the variation of fiber diameter in any 1 field and the uncertainty of the accuracy of the correction. This means that all of the observed collagen volume densities are overestimated.
The chord length of alveoli or saccules, and alveolar ducts was measured. The diameter was measured at 60x on hematoxylin and eosin-stained sections using an eyepiece ruler on 40 consecutive fields of each of the sections and the average calculated to give the diameter. All quantitative measurements were done with the observers blinded to the infant’s grouping. This project was reviewed and approved by the University of Missouri at Kansas City Health Sciences Pediatric Institutional Review Board.
Statistical Analysis
Values are given as mean ± standard deviation in Table 1. Statistical analysis in Table 1 and the text was performed with t tests or analysis of variance when there were >2 groups. P < .05 was considered significant. Areas under the curve and confidence intervals were calculated by the software program SigmaPlot (SigmaStat: SigmaPlot, SPSS, Inc, Chicago, IL).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The volume density of collagen in the parenchyma (VVCOLL) of the controls decreased from a maximum of 9% at 22 weeks to approximately 5% at term, which then remained constant to 72 weeks (Fig 1, top). The parenchyma includes air space as well as interstitial tissue. Some lungs at the lowest GA had low volume densities of collagen. These infants had prolonged assisted ventilation, beyond delivery room resuscitation. The parenchymal air space was increased in these infants, which decreased the parenchymal collagen volume density compared with infants who received little or no positive pressure.
|
The majority of infants with scores <20 and 20–69 had VVCOLL that were similar to controls. However, 79% of infants with severe lung disease (score: 70–500) had VVCOLL greater than controls at similar ages. The volume density of collagen in the parenchyma is influenced by the ratio of interstitial tissue to air space, as well as the density of collagen in the interstitium. Both the ratio and the quantity of tissue collagen can be anticipated to vary with GA and lung disease. In controls, the parenchyma at 22 to 23 weeks contained 60% to 70% interstitium and 30% to 40% air, which reversed to 20% to 30% interstitium and 70% to 80% air space at 30 weeks, remaining constant to 72 weeks (Fig 1, bottom). The degree of severity of lung disease (score: <20–500) did not seem to change or increase the ratio.
To help resolve the issue of whether tissue collagen was increased in CLD, we plotted the volume density of collagen per unit volume of interstitium, as a function of age (Fig 2, top). Controls increase progressively from a low of 5% at 22 to 23 weeks to approximately 25% at 72 weeks. However, there is a subset of infants, 22 to 25 weeks, with relatively high collagen densities. These are infants with large saccules, secondary to positive pressure ventilation. These large saccules compress and condense the interstitium, thereby raising the collagen density per unit volume of interstitium. Lung collagen density per unit of interstitium in infants with scores 69 seems to be similar to controls. However, in the severe lung disease scores (70–500), 79% had collagen densities per unit interstitium that were greater than same-age controls.
|
Figure 2, bottom, shows saccular chord diameters as a function of age. In controls, the saccular diameters increase progressively, from approximately 40 µm at 22 weeks to 100 µm at 72 weeks. When immature infants received little or no positive pressure in the delivery room and died within the first hour of life, the diameters were small, although the lungs all were fixed with a standard fluid pressure. A subset of larger diameters in some very young lungs is related to positive pressure ventilation, which overdistends and often deforms the saccules. In general, infants with scores
69 were similar to controls but with more severe chronic lung disease (scores: 70–500); 86% of lungs had diameters 2 to 3 times greater than same-age controls. Determining whether tissue collagen is increased with severe CLD by assessing the parenchymal volume density of collagen or the density of collagen per unit volume of interstitium is confounded by the variables of ratio of tissue to airspace and density of interstitial collagen with age and disease. We attempted to eliminate these 2 variables by determining the absolute quantity of collagen in the parenchyma of the left lung by multiplying the volume density of collagen in the parenchyma by the total volume of parenchyma (Fig 3). The total parenchymal collagen of both lungs was then calculated. The controls increased from a low of 0.3 cm3 at 22 weeks to >9 cm3 at 72 weeks. Eighty-five percent of infants with scores 70 to 500 had elevated levels of lung parenchymal collagen. With lesser scores, the collagen was similar to controls.
|
To confirm these quantitative collagen changes, we further determined the site, architecture, and quantity of the acinar parenchymal collagen in controls and in those with lung disease. Figure 4, left, shows an en face plane through a wall of a 23-week control infant saccule. This section and all others showing en face planes were 40 µm thick. The nuclei of the saccular wall cuboidal epithelium are seen, along with a fine mesh of collagen fibers in the wall. The fibers connect to larger diameter collagen fibers surrounding saccules mouths. For comparison, Fig 4, right, shows a septal en face section of a 40-week control lung. Fine collagen fibers are again seen, but they now connect to larger fibers before connecting to still larger fibers at the alveolar mouth. The empty spaces surrounded by relatively large collagen fibers contain blood vessels.
|
The very fine septal wall collagen mesh seen at 23 weeks increases and thickens markedly in controls by 30 weeks (Fig 5, right). This suggests that the 30-week lung has substantial saccular wall support from collagen and is similar to the term infant except that the fibers are thinner. These en face sections support the morphometric data that showed that parenchymal collagen increases with age. Figure 5, left, shows a septal en face section of a 28-week PCA infant who had severe lung disease (score: 190) and lived 31 days. The fibers are coarser than those in 30-week controls, and the pattern is disorganized, with tortuous, spiraling fibers.
|
For further assessing the septal wall collagen, Fig 6, left, shows a thick cut of a 24.4-week GA infant dying at a PCA of 28.8 weeks. The collagen secondary fibers are disorganized, tortuous, and thickened. Figure 6, right, shows a 24.0-week GA infant dying at 27.7 weeks. Again, the fiber arrangement is disorganized, with tortuous, thickened secondary fibers.
|
After 30 weeks’ gestation, in controls, the septal wall is thin and the interstitium forms an intrinsic part of the wall. However, in infants <30 weeks’ gestation and in those with CLD, the septal wall forms a continuum with a thick interstitium, which extends a long distance away from the septal wall. The interstitium of a 23-week control lung is wide but contains an exquisite network of finely, geometrically patterned collagen fibers (Fig 7, top). The saccules are of small diameter, and they undergo primary septation by invaginating into the wide interstitial space with its sparse, fine collagen network. By 30 weeks’ gestation, primary septation decreases because there is very little interstitial space, so that the distal airspaces can only grow by subpleural development or by secondary septation and expansion. This delicate interstitial collagen network is contrasted with the massively wide interstitium with large, thick collagen fibers seen in a 28-week PCA infant with CLD (Fig 7, bottom). The saccules are markedly enlarged with no evidence of beginning secondary septation. All of the severe CLD lungs showed this pattern. Saccules can be seen budding or invaginating into the interstitium in practically every field of view in infants <26 weeks’ gestation. Figure 8, top, is from a control 23-week infant dying within an hour of birth. The budding or invagination into the thick interstitium with its delicate collagen fiber network is shown. Figure 8, bottom, is from a 27-week infant who lived 25 hours on low ventilator settings. The interstitium is still thick with a delicate collagen network and invagination and primary septation is still seen, but not as prominently as at 23 weeks. Figure 9, bottom, shows the distal acinus of a 23.8-week control infant after 18 hours of mechanical ventilation. This portion of the lung looks normal, with normal-sized saccules, and a fine, interstitial collagen network, with adequate space for primary septation. This portion of the lung has a low collagen density per unit volume of interstitium. A large proportion of the same lung shows the effects of mechanical ventilation; the fine, collagen interstitial network has been compressed and damaged, perhaps permanently (Fig 9, top). The collagen density per unit interstitium is increased by the compression. This accounts for the subset of very early gestation patients with high interstitial collagen densities. The saccular size is immense compared with the normal part of the lung. These areas with compressed interstitium can no longer primary-septate because the space and template needed have been obliterated. The distension has distorted and stretched the collagen and elastic fibers in the wall, which may impair or prevent secondary septation later in development.
|
|
|
Collagen IV was immunohistochemically shown to line the subepithelial layer of the distal airways, saccules, and alveoli at all ages from 22 to 72 weeks. The intensity of the stain was similar at all ages. One of the goals in measuring collagen IV was to observe any fractures or frayed ends in the collagen in overdistended saccules. However, with the light microscope, breaks in the collagen IV layer were not seen, despite severe stretching associated with the saccular diameter increase.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This study demonstrates that parenchymal collagen increases throughout development and that before 30 weeks’ gestation there is a wide interstitium with a delicate complex interstitial collagen network, whose integrity may be important for primary septation. Lung overdistension of short duration can compress and damage this interstitium network and is associated with arrest of normal lung development. More chronic, prolonged positive pressure ventilation causes tissue remodeling, increases interstitial collagen, impedes parenchymal development, and adversely affects pulmonary mechanics.
Until recently, the classification of the reticulum fiber and its relationship to lung parenchymal connective tissue was unclear.15 Reticulum does not stain with conventionally used collagen stains such as Masson’s or Gomori’s trichrome or van Giesen’s method. On the basis of this inability of the tissues to stain with conventional stains, it was thought that there were minimal collagen fibers in the parenchyma. Reticulum fibers were, however, found in abundance near alveoli. Eventually it was shown that these reticulum fibers immunostained with collagen I and III antibodies and were identified chemically to be collagen I and III.10,15 Collagen I, III, and IV constitute the bulk of collagens in the parenchymal region. Gomori’s reticulum stains collagens I and III but not IV.
The quantitative and descriptive morphometric approach to assessing parenchymal collagen complements previous biochemical studies on collagen in preterm infants but adds the new dimension of identifying the anatomic site and pattern of collagen parenchymal deposits. Cherukupalli et al16 studied collagen in fresh lung tissue from human control infants with various degrees of bronchopulmonary dysplasia (BPD) from 20 weeks to 1.5 years of age. Collagen was assessed as hydroxyproline concentration and expressed per dry weight. Collagen increased with age in controls and significantly increased beyond control values in BPD infants.
Pierce et al17 assessed lung homogenate collagen as hydroxyproline per dry lung weight in preterm lambs that had received assisted ventilation and inspired oxygen as needed for 3 to 4 weeks. They found no difference in collagen content in the treated groups versus controls. In addition, they found no differences between the groups with a 1-time measurement of 
-1(1) procollagen mRNA. Studies in adults with fibrotic lungs have shown an increased proportion of type III collagen in early fibrosis, but later, in more advanced fibrosis, there is an increased laying down of the more rigid type I collagen.18 Studies that have measured the ratio of collagen I/III on infants with lung disease have been inconclusive or contradictory, and more study is needed.16,19,20
Biochemical studies are inconclusive when trying to decide whether animals or infants with BPD or CLD have a propensity toward fibrosis. Our explanation for these variations is that there are many types of collagen in the lung, with large amounts around vessels and airways, and some types of collagen may increase or decrease with development and disease. These factors confound the interpretation of lung homogenate collagen.
Morphometric analysis gives an added dimension to understanding parenchymal collagen. Hussen et al21 compared alveolar fibrosis in surfactant- and non-surfactant-treated preterm infants who developed BPD. Using the light microscope and a trichrome collagen stain, they morphometrically assessed the quantity of acinar fibrosis. They found mild to moderate alveolar septal fibrosis in surfactant-treated infants with BPD and more fibrosis in non-surfactant-treated infants. We used the reticulum stain to assess the site and quantity of distal acinar collagen and focused on infants with various degrees of severity of CLD, which was quantified by a scoring system. Volume density of parenchymal collagen is difficult to interpret during development and with disease because of variations in the ratio of parenchymal tissue to air, collagen changes with development and disease, and tissue distortion with assisted ventilation. When the VVCOLL is expressed per unit volume of interstitium, a clearer picture emerges, showing that parenchymal collagen is least in the most immature infants and increases with gestation in controls, increasing approximately 4-fold from 22 weeks to term. However, in ventilated infants, especially in those <30 weeks’ gestation, the saccular diameter is increased, which compresses and condenses the fine collagen interstitial network and increases the volume density of collagen per unit volume of interstitium. In CLD infants with scores 69, the volume density of collagen per unit volume of interstitium was not different from controls. However, with severe CLD with higher scores, collagen was significantly increased. To rule out the possibility that interstitial tissue compression alone increased the parenchymal collagen, we calculated the total lung parenchymal collagen, and it paralleled and confirmed the volume density of collagen per unit volume of interstitium findings shown in Fig 3. In general, the most severe lung disease was correlated with the largest chord saccular diameter and the largest amount of collagen, suggesting that the 3 findings are interrelated.
The en face thick sections indicate that the quantity of collagen in the saccular and alveolar walls increases with GA in controls. Rosenquist22 showed 3 types of reticulum fibers in adult human lung septa, which he called primary, secondary, and tertiary fibers. Primary fibers form a continuous mesh and are attached to the endothelial or epithelial basal lamina. Larger diameter, secondary fibers are connected to primary fibers, course around vessels, and connect to the larger, tertiary fibers, which are continuous with the coarse collagen bundles at the alveolar mouth. Our findings indicate qualitatively that the adult fiber pattern is approximated at 30 weeks’ gestation. With severe CLD, the septal walls have more large-caliber secondary fibers that are tortuous compared with age-matched controls and, in addition, the collagen fiber pattern is disorganized.
Of great interest was the pattern and quantity of interstitial collagen that is not directly a part of the septal wall. In controls <30 weeks’ gestation, the interstitium between saccules is spacious and contains a fine interconnecting collagen network. The spaces between the fibers are relatively acellular and provide a space for the distal air spaces to expand and primary-septate (Fig 10). In utero, fluid is produced by the saccule cuboidal epithelial cells, producing an airway pressure a few cm H2O greater than amniotic fluid.7–9 The airway fluid volume and pressure are regulated by the upper airway valve mechanism. Distal acinar fluid pressure is further modulated by distal airway muscle tone.23 This fluid system is thought to provide the force needed to facilitate distal airway and saccular growth. The mouths of future saccules form within the preprogrammed elastic and collagen fiber ring locations. The rings of elastic and collagen fibers direct the fluid force to invaginate the airway walls into the interstitial space containing the fine collagen network. This system would need an interstitial tissue counterforce to direct and determine the size and shape of new saccules. Taking into consideration the low pressure of the system and the positioning and tensile strength of the collagen network, interstitial collagen could possibly function as this counter-force. At approximately 30 weeks’ gestation, the interstitium thins out, signaling the end of primary septation. Secondary septa form as protrusions from the walls of primary saccules into the airway lumen. This process requires normally formed saccules with collagen, elastic fibers, fibroblasts, and vessels properly positioned.3,24
|
In many infants <30 weeks’ gestation, positive pressure ventilation has a profound effect on lung geometry. Some areas of the lung may be normal, with small diameter saccules surrounded by a normal, wide interstitial collagen network, but other areas show very large saccules with stretched walls surrounded by a deformed and compressed interstitial network. The fine interstitial collagen network was not designed to tolerate the large tidal pressure swings associated with assisted ventilation. As described in Fig 9, top, it is difficult to envision any additional primary septation in the overdistended areas of the lung. The wall connective tissue, matrix, and mesenchymal damage may arrest normal lung maturation and preclude later secondary septation.
These findings suggest that any magnitude of positive pressure beyond a few cm H2O to the very immature lung, even with adequate surfactant, may be excessive and adversely affect normal lung development. Lungs of infants who are dying with CLD also have a wide interstitium, but the normal, delicate interstitial collagen network is replaced with a thick, coarse collagen network, and this is associated with large diameter air sacs. The biological window of opportunity for primary and secondary septation in these lungs seems to have passed. The overproduction of thick interstitial collagen is possibly secondary to tissue remodeling that follows positive pressure ventilation. In vitro studies have shown that increased mechanical forces activate epithelial cells to produce signals and inflammatory proteins that stimulate the interstitial fibroblast to produce collagen, metalloproteins, and other inflammatory products.25 This process causes remodeling and, perhaps in human infants, a continuous tissue repair as long as the increased airway pressure persists. By removing the pressure, remodeling may be halted and cause regression of the excessive interstitial collagen. This process could account, in part, for the increases in growth factors, metalloproteins, and cytokines found in tracheal aspirates of ventilated infants.
Because collagen IV lined the subepithelial basement membrane of all of the distal airways and saccules, it was anticipated that lung overdistension would fracture this basement membrane and the frayed ends of collagen IV would be seen by light microscopy. However, fracture in distended air sacs was not seen in this light microscopic study. Ohki et al26 showed elevated levels of collagen IV in the bronchoalveolar lavage fluid in infants who had respiratory distress syndrome in the first 2 weeks of life and went on to develop BPD. This suggests that damage to the basement membrane occurs early in BPD development and, together with our findings of damage to the interstitial matrix and collagen architecture, suggests that the origins of CLD may occur before the hyperoxic and inflammatory damage.
Limitations
It is difficult to obtain nonedematous, noninflamed, nonmechanically damaged, or perfectly inflated lungs. At the time of birth, lung structure may vary from infant to infant, which may also affect the response of the postnatal lung to positive pressure. Routinely, over- and underinflated areas of the lungs are seen radiologically in low birth weight infants at risk for CLD. Nonetheless, inflating lungs at 24 cm H2O fluid pressure may overinflate certain portions of these lungs that could lead to overinterpretation of the geometric changes seen histologically. Hyperoxia and steroid therapies prevent secondary septation without positive pressure, so their contribution to abnormal lung development cannot be clearly separated from the deleterious mechanical effects outlined in this study.27–29 We also studied only infants who were at risk for CLD from 5 to 94 days postnatally. Collagen may change significantly after this period.16 Quantifying lung collagen in 5-µm-thick sections is subject to a large error, as outlined in the Methods section. Nonetheless, the photomicrograph images of the collagen in the septal wall and interstitium, in general, support our quantitative measurements of distal airway collagen during normal fetal development and in CLD infants.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Parenchymal collagen increases throughout lung development. Before 30 weeks’ gestation, the interstitium is spacious, allowing invagination of saccule walls and primary septation. This interstitium also contains a delicate and complex network of collagen. Positive pressure ventilation can compress and damage the interstitial matrix and collagen network and perhaps prevent normal primary septation and lung development. In severe CLD, distal air space diameter increases and primary and secondary septation fails. Lung development is disturbed, with thickened collagen seen around saccule walls, a widened interstitium, and increased quantity and size of interstitial collagen fibers.
|
ACKNOWLEDGMENTS |
|---|
This study was supported, in part, by the Katharine B. Richardson Foundation, by a Physician Scientist Award from Children’s Mercy Hospital (to Dr Truog), by NIH R-01 HL 58125 (to Dr Truog), and by K-23HL04264-01 (to Dr Ekekezie).
|
FOOTNOTES |
|---|
Received for publication May 1, 2002; Accepted Sep 9, 2002.
Reprint requests to (D.W.T.) Children’s Mercy Hospital, 2401 Gillham Rd, Kansas City, MO 64108. E-mail: tos2{at}eastlink.ca
This work was presented at the Annual Meeting of the Society of Pediatric Research, Pediatric Academic Society; May 4–7, 2002; Baltimore, MD.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
- Loosli CG, Potter EL. Pre- and postnatal development of the respiratory portion of the human lung. Am Rev of Respir Dis.1959; 80 :5 –23
- Emery JL. The post natal development of the human lung and its implications for lung pathology. Respiration.1970; 27 :41 –50
- Burri PH. Structural aspects of prenatal and postnatal development and growth of the lungs. Lung Growth Dev.1977; 100 :1 –35
- Lindahl P, Karlsson L, Hellstrom M, et al. Alveogenesis failure in PDGF-A deficient mice is coupled to lack of distal spreading of alveolar smooth muscle cell progenitors during lung development. Development.1997; 124 :3943 –3953[Abstract]
- Wendel DP, Taylor DG, Albertine KH, et al. Impaired distal airway development in mice lacking elastin.
Am J Respir Cell Mol Biol.2000; 23
:320
–326
[Abstract/Free Full Text] - Thibeault DW, Mabry SM, Ekekezie II, Truog WE. Lung elastic tissue maturation and perturbations during the evolution of chronic lung disease (CLD).
Pediatrics.2000; 106
:1452
–1459
[Abstract/Free Full Text] - Liggins GC, Kitterman JA. Development of the fetal lungs. In: The Fetus and Independent Life. Ciba Foundation Symposium 86. London, United Kingdom: Pitman; 1981:308–330
- Vilos GA, Liggins GC. Intrathoracic pressures in fetal sheep. J Dev Physiol.1982; 4 :247 –256[Medline]
- Fewell JE, Johnson P. Upper airway dynamics during breathing and during apnoea in fetal lambs.
J Physiol.1983; 339
:495
–504
[Abstract/Free Full Text] - Bateman E, Turner-Warwick M, Adelmann-Grill BC. Immunohistochemical study of collagen types in human foetal lung and fibrotic lung disease.
Thorax.1981; 36
:645
–653
[Abstract/Free Full Text] - Schnieke A, Harbers K, Jaenisch R. Embryonic lethal mutation in mice induced by retrovirus insertion into the 1(I) collagen gene.
Nature.1983; 304
:315
–320[CrossRef][Medline]
- Gomori G. Silver impregnation of reticulum in paraffin sections. Am J Pathol.1937; 13 :993 –1002[Web of Science]
- Wandel G, Berger LC, Burri PH. Morphometric analysis of adult rat lung after bilobectomy. Am Rev Respir Dis. 193;128 :968 –972
- Weibel ER. Stereological Methods: Vol 1. Practical Methods for Biological Morphometry. London, England: Academic Press; 1980
- Huang TW. Chemical and histochemical studies on human alveolar collagen fibres. Am J Pathol.1977; 86 :81 –93[Abstract]
- Cherukupalli K, Larson JE, Rotschild A, Thurlbeck WM. Biochemical, clinical, and morphologic studies on lungs of infants with bronchopulmonary dysplasia. Pediatr Pulmonol.1996; 22 :215 –229[CrossRef][Web of Science][Medline]
- Pierce RA, Albertine KH, Starcher BC, Bohnsack JF, Carlton DP, Bland RD. Chronic lung injury in preterm lambs: disordered pulmonary elastin deposition.
Am J Physiol.1997; 272
:L452
–L460
[Abstract/Free Full Text] - Low RB, Giancol MS, King TE, Chapitis J, Vacek CP, Davis GS. Serum and bronchoalveolar lavage of N-terminal type III procollagen peptides in idiopathic pulmonary fibrosis. Am Rev Respir Dis.1992; 146 :701 –706[Web of Science][Medline]
- Shoemaker CT, Reiser KM, Goetzman BW, Last JA. Elevated ratios of type I/III collagen in the lungs of chronically ventilated neonates with respiratory distress. Pediatr Res.1984; 18 :1176 –1180[Web of Science][Medline]
- Heikinheimo M, Halila R, Marttinen E, Raivio K. N-terminal propeptide of type III collagen in tracheal fluid and serum in preterm infants at risk for bronchopulmonary dysplasia. Pediatr Res.1992; 31 :340 –344[Web of Science][Medline]
- Husain AN, Siddiqui NH, Stocker JT. Pathology of arrested acinar development in postsurfactant bronchopulmonary dysplasia. Hum Pathol.1998; 29 :710 –717[CrossRef][Web of Science][Medline]
- Rosenquist TH. Organization of collagen in the human pulmonary alveolar wall. Anat Rec.1981; 200 :447 –459[CrossRef][Medline]
- Schittny JC, Miserocchi G, Sparrow M. Spontaneous airway narrowing propels lung liquid through the bronchial tree of intact and fetal lung explants.
Am J Respir Cell Mol Biol.2000; 23
:11
–18
[Abstract/Free Full Text] - Fukuda Y, Ferrans VJ, Crystal RG. The development of alveolar septa in fetal sheep lung. An ultrastructural and immunohistochemical study. Am J Anat.1983; 167 :405 –439[CrossRef][Web of Science][Medline]
- Swartz MA, Tschumperlin DJ, Kamm RD, Drazen JM. Mechanical stress is communicated between different cell types to elicit matrix remodeling.
Proc Natl Acad Sci U S A.2001; 98
:6180
–6185
[Abstract/Free Full Text] - Ohki Y, Kato M, Kimura H, Nako Y, Tokuyama K, Morikawa A. Elevated type IV collagen in bronchoalveolar lavage fluid from infants with bronchopulmonary dysplasia. Biol Neonate.2001; 79 :34 –38[CrossRef][Web of Science][Medline]
- Massaro D, Massaro GD. Dexamethasone accelerates postnatal alveolar wall thinning and alters wall composition. Am J Physiol.1986; 251 :R218 –R224
- Massaro D, Tecih N, Maxwell S, Massaro GD, Whitney P. Postnatal development of alveoli: regulation and evidence for a critical period in rats. J Clin Invest.1985; 76 :1297 –1305
- Thibeault DW, Heimes B, Rezaiekhaligh M, Mabry S. Chronic modifications of lung and heart development in glucocorticoid-treated newborn rats exposed to hyperoxia or room air. Pediatr Pulmonol.1993; 16 :81 –88[Web of Science][Medline]
PEDIATRICS (ISSN 1098-4275). ©2003 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  M. O'Reilly, S. B. Hooper, B. J. Allison, S. J. Flecknoe, K. Snibson, R. Harding, and F. Sozo Persistent bronchiolar remodeling following brief ventilation of the very immature ovine lung Am J Physiol Lung Cell Mol Physiol, November 1, 2009; 297(5): L992 - L1001. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S M Aukland, K Rosendahl, C M Owens, K R Fosse, G E Eide, and T Halvorsen Neonatal bronchopulmonary dysplasia predicts abnormal pulmonary HRCT scans in long-term survivors of extreme preterm birth Thorax, May 1, 2009; 64(5): 405 - 410. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Irwin, K. Helm, N. Campbell, M. Imamura, K. Fagan, J. Harral, M. Carr, K. A. Young, D. Klemm, S. Gebb, et al. Neonatal lung side population cells demonstrate endothelial potential and are altered in response to hyperoxia-induced lung simplification Am J Physiol Lung Cell Mol Physiol, October 1, 2007; 293(4): L941 - L951. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. D. Higgins, E. Bancalari, M. Willinger, and T. N.K. Raju Executive Summary of the Workshop on Oxygen in Neonatal Therapies: Controversies and Opportunities for Research Pediatrics, April 1, 2007; 119(4): 790 - 796. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Alejandre-Alcazar, G. Kwapiszewska, I. Reiss, O. V. Amarie, L. M. Marsh, J. Sevilla-Perez, M. Wygrecka, B. Eul, S. Kobrich, M. Hesse, et al. Hyperoxia modulates TGF-beta/BMP signaling in a mouse model of bronchopulmonary dysplasia Am J Physiol Lung Cell Mol Physiol, February 1, 2007; 292(2): L537 - L549. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. H. Jobe The New BPD NeoReviews, October 1, 2006; 7(10): e531 - e545. [Full Text] [PDF]
|
|
|
|
|
|
M. Sheffield, S. Mabry, D. W. Thibeault, and W. E. Truog Pulmonary Nitric Oxide Synthases and Nitrotyrosine: Findings During Lung Development and in Chronic Lung Disease of Prematurity Pediatrics, September 1, 2006; 118(3): 1056 - 1064. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
