DOI: 10.1148/rg.254055019
RadioGraphics 2005;25:1047-1073
From the Archives of the AFIP
Lung Disease in Premature Neonates: Radiologic-Pathologic Correlation1
Geoffrey A. Agrons, MD,
Sherry E. Courtney, MD,
J. Thomas Stocker, COL, MC, USA and
Richard I. Markowitz, MD
1 From the Department of Radiologic Pathology, Armed Forces Institute of Pathology, 6825 16th St NW, Washington, DC 20306 (G.A.A.); Department of Radiology, Pennsylvania Hospital, Philadelphia, Pa (G.A.A.); Division of Neonatology, Schneider Children’s Hospital, North Shore Long Island Jewish Health System, New Hyde Park, NY (S.E.C.); Department of Pathology, Uniformed Services University of the Health Sciences, Bethesda, Md (J.T.S.); and Department of Radiology, Children’s Hospital of Philadelphia, Pa (R.I.M.). Received February 4, 2005; accepted March 7. S.E.C. participated in a speakers’ bureau for iNO Therapeutics, Clinton, NJ, and was a sponsored speaker for Viasys, Yorba Linda, Calif; all remaining authors have no financial relationships to disclose.
Address correspondence to G.A.A. (e-mail: gagrons{at}mac.com).
 |
Abstract
|
|---|
Pulmonary disease is the most important cause of morbidity in preterm neonates, whose lungs are often physiologically and morphologically immature. Surfactant deficiency in immature lungs triggers a cascade of alveolar instability and collapse, capillary leak edema, and hyaline membrane formation. The term respiratory distress syndrome (RDS) has come to represent the clinical expression of surfactant deficiency and its nonspecific histologic counterpart, hyaline membrane disease. Historically, chest radiographs of infants with RDS predictably demonstrated decreased pulmonary expansion, symmetric generalized reticulogranular lung opacities, and air bronchograms. Refinements in perinatal medicine, including antenatal glucocorticoid administration, surfactant replacement therapy, and increasingly sophisticated ventilatory strategies have decreased the prevalence of RDS and air leak, altered familiar radiographic features, and lowered the threshold of potential viability to a gestational age of approximately 23 weeks. Alveolar paucity and pulmonary interstitial thickness in these profoundly premature neonates impair normal gas exchange and may necessitate prolonged mechanical ventilation, increasing the risk of lung injury. Bronchopulmonary dysplasia (BPD), alternatively termed chronic lung disease of infancy, is a disorder of lung injury and repair originally ascribed to positive-pressure mechanical ventilation and oxygen toxicity. Before the advent of surfactant replacement therapy, chest radiographs of infants with classic BPD demonstrated coarse reticular lung opacities, cystic lucencies, and markedly disordered lung aeration that reflected alternating regions of alveolar septal fibrosis and hyperinflated normal lung parenchyma. In the current era of surfactant replacement, BPD is increasingly a disorder of very low-birth-weight neonates with arrested alveolar and pulmonary vascular development, minimal alveolar septal fibrosis and inflammation, and more subtle radiographic abnormalities.
|
LEARNING OBJECTIVES FOR TEST 6
|
|---|
After reading this article and taking the test, the reader will be able to:
- Describe the anatomic and pathophysiologic features of pulmonary immaturity.
- Recognize the radiologic manifestations of neonatal lung disease related to premature delivery and its treatment complications.
- Discuss the impact of evolving perinatal medical management on radiologic patterns of disease.
|
Introduction
|
|---|
Approximately 487,000 infants were delivered before completing 37 weeks gestation in the United States in 2002, a population that represents 12.1% of all live births and a 14% increase since 1990 (1). The increase in the preterm birth rate is related in part to the steep rise in the number of multiple births over the past 2 decades (2). Many premature newborns require treatment in a neonatal intensive care unit at an annual national cost that exceeds $4 billion. Of the many complications of prematurity, including intracranial hemorrhage, necrotizing enterocolitis, sepsis, and retinitis, lung disease remains the most common cause of neonatal morbidity.
During the past decade, dramatic changes in the medical management of premature infants have lowered the threshold of potential viability to 23–24 weeks of gestation. Below this age, developmental immaturity of multiple organ systems precludes survival. Death often results from pulmonary immaturity, which leads to severe tissue damage and dysfunctional gas exchange in an air-breathing environment. The improved survival of very low-birth-weight neonates, the ubiquity of surfactant replacement therapy, and refinements in mechanical ventilation have transformed the natural history of acute and chronic pulmonary insufficiency in premature newborns and have altered familiar radiologic patterns of disease. This article examines the clinical, pathologic, and radiologic features of pulmonary immaturity, the impact of evolving therapeutic strategies, and the changing face of chronic lung disease of prematurity.
|
Normal Lung Development
|
|---|
A deeper understanding of anatomic and physiologic pulmonary immaturity in the premature neonate is predicated on a review of normal lung development. Intrauterine lung maturation has been divided into five phases: embryonic, pseudoglandular, acinar or canalicular, saccular, and alveolar (3).
During the embryonic phase (26 days to 6 weeks gestation), an endodermlined outpouching or "lung bud" derived from the primitive fore-gut divides and branches dichotomously to form the early tracheobronchial tree (Figs 1, 2). Initially, the primitive airways are surrounded by loose mesenchyme supplied by primitive systemic arteries. The pulmonary arteries arise from the sixth aortic arch near the end of the embryonic period, penetrate the mesenchyme, and ultimately replace the systemic vessels.
View larger version (48K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1. Normal development of the major airways. Schematic shbows induction of dichotomous branching of the lung bud (arrow) by contact with primitive mesenchyme (arrowhead). (Illustration by Aletta Ann Frazier, MD, Department of Radiologic Pathology, Armed Forces Institute of Pathology.)
|
|
View larger version (147K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2. Embryonic phase of normal respiratory tract development. Photomicrograph (original magnification, x100; hematoxylineosin [H-E] stain) of the developing lungs of a 4-mm embryo at approximately 30 days gestation shows the primary bronchial buds (arrows) surrounded by primitive mesenchyme.
|
|
The pseudoglandular phase (6–16 weeks gestation) includes the development of the airways to the level of the terminal bronchioles. At gross examination, the external morphology of the immature lungs is similar to that of neonatal lungs at term. However, at microscopic examination, the bronchioles end blindly within primitive stroma, a histologic feature reminiscent of glandular tissue (Fig 3). Recently, the traditional understanding that only conducting airways are formed during this period of lung development has been challenged (4,5). Nevertheless, a deficient number of true alveolar saccules during this developmental stage prevents meaningful gas exchange, and extrauterine survival is not possible.
View larger version (198K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3. Immature lung in the pseudoglandular phase. Photomicrograph (original magnification, x15; H-E stain) of the lungs at 13 weeks gestation shows blind-ending terminal bronchioles (arrows) surrounded by immature lung parenchyma, which has begun to organize into lobules and clusters of primitive acini (arrowheads).
|
|
During the canalicular or acinar phase (16–28 weeks gestation), multiple alveolar ducts arise from respiratory bronchioles (Fig 4). Alveolar ducts are lined by type II alveolar cells, which are capable of surfactant synthesis. Thin type I alveolar lining cells differentiate from type II cells. Toward the end of this developmental phase (24–28 weeks), primitive distal saccules (primitive alveoli) begin to form through a process known as primary septation. Progressive thinning of the pulmonary interstitium allows gas exchange as the walls of proliferating capillaries and type I alveolar lining cells approximate.
View larger version (168K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4. Immature lung in the canalicular (acinar) phase. Photomicrograph (original magnification, x425; H-E stain) of the lungs shows blood-filled capillaries (arrow) that lie immediately beneath the surface of alveolar duct structures lined by cuboidal epithelium (early type II pneumocytes (arrowhead). (Reprinted, with permission, from reference 6.)
|
|
The saccular phase (28–34 weeks gestation) is characterized by an increase in the number of terminal sacs, further thinning of the interstitium, robust proliferation of the capillary bed, and early development of true alveoli by secondary septation at about 32 weeks gestation (Fig 5) (4).
View larger version (143K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5. Saccular phase of lung development. Photomicrograph (original magnification, x350; H-E stain) of the lungs reveals saccules subdivided by secondary crests (arrows) composed of thinning type I cells immediately adjacent to capillary beds.
|
|
The alveolar phase of lung development extends from approximately 36 weeks gestation until 18 postnatal months, but most alveolarization occurs within 5–6 months following delivery at term (4,5). At term, the mature alveolus is composed exclusively of alveolar lining cells, underlying basement membranes, and capillary endothelia, permitting adequate gas exchange across a thin septum (Fig 6).
View larger version (177K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6. Alveolar phase of lung development. Photomicrograph (original magnification, x25; H-E stain) of the normal lung at 38 weeks gestation shows mature alveolar ducts and alveolar saccules with delicate septa, resulting in a thin air-blood barrier.
|
|
Analogous to a soap bubble, the alveolus represents a dynamic physical structure characterized by high surface tension generated by unequal distribution of molecular forces at an air-liquid interface (7). The physical integrity of normally expanded alveoli is maintained by surfactant, a complex surface-active lipoprotein containing primarily the phospholipids dipalmitoylphosphatidylcholine and phosphatidylglycerol. Synthesis of surfactant phospholipids begins in the endoplasmic reticulum of type II pneumocytes, where they are transported through the Golgi apparatus and ultimately concentrated into intracellular lamellar bodies. Lamellar bodies then migrate to the cell surface, where their contents are expressed onto the alveolar luminal surface by exocytosis (Fig 7). The extruded phospholipids combine with four surface active apoproteins (surfactant proteins A, B, C, and D), which are also produced within type II pneumocytes, to form a highly ordered complex lattice termed tubular myelin (7,8). Tubular myelin is the chief biochemical contribution to the lipid monolayer at the alveolar air-fluid interface, which lowers alveolar surface tension and prevents expiratory acinar collapse.
View larger version (99K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7. Endogenous surfactant delivery. Diagram of a type II pneumocyte demonstrates migration of lamellar bodies (arrow) from the nucleus to the apical cell surface, where surfactant (in pink) is released into the alveolus by exocytosis. (Diagram by Aletta Ann Frazier, MD, Department of Radiologic Pathology, Amred Forces Institute of Pathology.)
|
|
The factors governing structural and physiologic lung maturation are complex, incompletely elucidated, and subject to controversy. Although the precise mechanisms that normally regulate alveolar septation and pulmonary microvascular development in humans remain obscure (9), delayed alveolar and vascular development in experimental animals has been associated with hypoxia, hyperoxia, mechanical ventilation, nutritional deficiencies, elevated levels of proinflammatory cytokines, and the use of exogenous glucocorticoids (9–14). The role of glucocorticoids in lung maturation is particularly enigmatic. Although it is recognized that glucocorticoids may interfere with normal alveolarization, they also promote structural maturation by thinning pulmonary mesenchyme and inducing surfactant axis maturation through increased biosynthesis of phosphatidylcholine (9,10,15–17). Finally, a variety of antepartum stressors, including preeclampsia, prolonged preterm rupture of membranes, and intrauterine growth retardation, have been associated with accelerated lung maturation (9). However, results from several clinical investigations have challenged the axiom that fetal stress triggers maturation of the lungs in neonates delivered before term (18–20).
|
Neonatal Respiratory Distress Syndrome
|
|---|
Clinical Features
Respiratory distress syndrome (RDS) is the clinical expression of surfactant deficiency in neonates. Alveolar instability and collapse results in decreased functional residual capacity and increased dead space. Impaired oxygenation and metabolic acidosis increase pulmonary vascular resistance, which in turn increases the likelihood of right-to-left shunting through the ductus arteriosus, augments ventilation-perfusion mismatch, and exacerbates hypoxemia. The deleterious effects of relative surfactant deficiency may be compounded by qualitative defects in extant surfactant when it is inactivated by acinar fluid from capillary leak, hemorrhage, aspirated meconium, or alveolar epithelial injury (7).
In 2002, the overall rate of RDS was 6.1 cases per 1,000 neonates, or approximately 24,000 newborns, representing a decrease since the highest levels were reported for 1994–1995 (1). Most neonates with RDS are premature, and their deficiency of endogenous surfactant is related to a relative lack of mature type II pneumocytes. Term infants of mothers with poorly controlled diabetes may also present with RDS, because fetal hyperinsulinism interferes with the glucocorticoid axis that governs surfactant biosynthesis (21). Other risk factors for RDS include fetal asphyxia, maternal or fetal hemorrhage, and multiple gestations. RDS is more common and severe in male neonates and occurs more commonly in whites than blacks (22).
Neonates with RDS typically present with non-specific tachypnea, expiratory grunting, nasal flaring, cyanosis, and substernal and intercostal retractions (Fig 8). Grunting is the sound produced by breathing against a partially closed glottis as a neonate with RDS augments alveolar distention by increasing expiratory pressure.
View larger version (54K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8. Clinical features of neonatal respiratory distress. Drawing depicts a representative preterm newborn with RDS exhibiting substernal and intercostal retractions, nasal flaring, and circumoral cyanosis. (Illustration by Aletta Ann Frazier, MD, Department of Radiologic Pathology, Armed Forces Institute of Pathology.)
|
|
Pathologic Features
Hyaline membrane disease, an older term and the synonym of RDS, refers to the microscopic appearance of smooth, homogeneous, eosinophilic membranes that line terminal bronchioles and alveolar ducts (23). Hyaline membranes are composed of necrotic alveolar cells, plasma transudate, aspirated squamae, and fibrin. Polymorphonuclear cells may infiltrate the membranes if infection complicates RDS. Although hyaline membranes may be present as early as 3–4 hours after birth in neonates who later died of RDS, they are usually well established by 12–24 hours (23). Hyaline membranes typically organize and separate from the underlying alveolar or bronchial wall at 36–48 hours if high oxygen tension and ventilator pressures are not required, and they are ultimately cleared by alveolar macrophages. At microscopic examination, the small airways containing the hyaline membranes are surrounded by collapsed acini of the surfactant-deficient lungs (Fig 9). The term hepatization has been applied to the gross appearance of the lungs of infants dying of hyaline membrane disease, because the texture of cut sections of the firm, homogeneous, atelectatic tissue is reminiscent of liver (Fig 10).
View larger version (203K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9. Surfactant deficiency in a premature infant. Histologic features. Photomicrograph (original magnification, x75; H-E stain) shows collapsed acini surrounding dilated alveolar ducts lined by smooth homogeneous hyaline membranes (arrows).
|
|
View larger version (139K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10. Surfactant deficiency in a premature infant. Gross pathologic features. Photograph of an autopsy specimen demonstrates small atelectatic lungs with focal hemorrhage (arrow) visible on the pleural surface.
|
|
Evolving Nomenclature
The term hyaline membrane disease is now less commonly used in clinical practice to describe the constellation of pathologic, clinical, and radiologic findings produced by pulmonary surfactant insufficiency in infants. In general, hyaline membranes are not specific histologic evidence of surfactant deficiency, but they may form in the aftermath of a variety of primary bronchiolar insults. Thus, hyaline membranes are considered a byproduct, not the cause, of respiratory failure in neonates with immature lungs. Respiratory distress syndrome is currently used to denote surfactant deficiency and should not be used for other causes of respiratory distress. However, some authors (24) consider the term nonspecific and imprecise, because it connotes a constellation of signs and symptoms that may accompany other causes of lung disease. In recognition of the underlying pathogenesis of the disease process, the alternative term surfactant deficiency disorder has been proposed (24).
Advances in Clinical Management of RDS
Despite the increased rate of preterm delivery in the United States, recent advances in perinatal medical management have decreased the incidence and severity of RDS. Clinical innovations that have substantially modified the natural history of RDS include antenatal corticosteroid administration, surfactant replacement therapy, and increasingly sophisticated use of assisted ventilation in the delivery room and intensive care nursery (6).
Corticosteroid treatment of pregnant women at risk of premature delivery was introduced in 1972 to enhance structural and biochemical lung maturity in neonates (25). Randomized controlled trials have shown that antenatal corticosteroid administration reduces the incidence of RDS in infants born at 29–34 weeks gestation, but the trial results have not confirmed a definite benefit in infants born at 24–28 weeks gestation (21). However, a large randomized trial of pregnant patients with intact membranes at 24–28 weeks gestation showed a significant reduction in the incidence of high-grade neonatal intracranial hemorrhage (26), and meta-analysis suggests significant benefit of antenatal corticosteroids for reducing the incidence of RDS, intracranial hemorrhage, necrotizing enterocolitis, and neonatal mortality (27). The National Institutes of Health consensus conference statements on antenatal corticosteroid use recommend that all pregnant women should be considered eligible for a single course of corticosteroids during 24–34 weeks gestation (21,28,29). Although the effect of prenatal glucocorticoid administration on RDS in pregnancies less than 24 weeks gestation remains unclear, a course of antenatal steroids is deemed reasonable (30).
After delivery, administration of exogenous surfactant synergizes with antenatal corticosteroid therapy to improve oxygenation, decrease the need for mechanical ventilation, and reduce mortality in neonates with respiratory failure from RDS (31). The efficacy and safety of surfactant replacement therapy have been established in multiple randomized controlled clinical trials (32–47). Surfactant is delivered into the tracheobronchial tree as a liquid bolus through an endotracheal tube. During administration, the infant is turned from side to side to facilitate uniform acinar distribution of surfactant throughout both lungs. Nebulized surfactant delivery, which would eliminate the reliance on tracheal intubation, is under investigation but has not yet been proved effective and is not commercially available (48).
Four surfactant preparations have currently been approved by the U.S. Food and Drug Administration: Exosurf Neonatal (colfosceril palmitate; Glaxo Wellcome, Research Triangle Park, NC), a synthetic formulation; Survanta (beractant; Ross Products Division, Abbott Laboratories, Columbus, Ohio) and Infasurf (calfactant; Forest Laboratories, New York, NY), natural surfactants composed of modified bovine lung extract; and Curosurf (poractant alfa; Chiesi Farmaceutici, Parma, Italy), a porcine-derived minced lung extract. Both modified natural and synthetic surfactant formulations are effective in the treatment and prevention of RDS. However, a meta-analysis of comparative trials indicates that use of natural surfactant results in greater early improvement in the requirement for ventilator support, less frequent pneumothorax, decreased mortality, and a marginal decrease in the risk of chronic lung disease associated with treatment of RDS (49).
Two treatment strategies for surfactant delivery have emerged. In prophylactic treatment, surfactant is administered immediately in the delivery room or very soon thereafter in an effort to prevent RDS in preterm neonates. In rescue (selective) treatment, surfactant is given after the diagnosis of RDS has been established clinically and radiographically. In a comparison of prophylactic surfactant delivery and rescue therapy for confirmed RDS, data analysis of clinical outcomes in eight studies was recently performed in accordance with the standards of the Cochrane Neonatal Review Group (50). The reviewers concluded that, when compared with selective treatment of established RDS, prophylactic surfactant administration to neonates considered at risk for developing RDS (preterm infants less than 30–32 weeks gestation) was associated with a decreased risk of pneumothorax, less frequent interstitial air leak, and decreased mortality. However, the precise criteria by which an infant should be judged "at risk" for developing RDS and thus selected for prophylactic surfactant administration remain unclear. Currently, neonatologists often use their own center’s demographics to make this decision. Because early treatment is more effective, infants not given surfactant prophylactically can be immediately treated if signs of respiratory distress develop.
Before the advent of surfactant replacement therapy, mechanical ventilation was the primary intervention to combat the physiologic consequences of surfactant deficiency in infants with RDS. Positive-pressure ventilation distends collapsed acini while delivering air with high oxygen content (51). However, mechanical ventilation is associated with untoward cardiovascular and pulmonary consequences, including diminished cardiac output, decreased systemic venous return, compromised pulmonary blood flow, air leak complications, and chronic lung disease (52). Current treatment protocols advocate use of ventilatory assistance according to the severity of the lung disease, thus optimizing tissue oxygenation and carbon dioxide removal while minimizing cardiovascular depression, overdistention of small airways, and oxygen toxicity. Neonates with mild RDS often are treated with early nasal continuous positive airway pressure (CPAP). If needed, surfactant can be given and extubation performed very soon thereafter. Thus for some infants, tracheal intubation may be brief or avoided altogether (52,53).
Strategies of mechanical ventilation used today attempt to minimize lung injury by avoiding both atelectasis and overdistention, as well as by optimizing blood gas tensions to avoid hypocarbia and oxygen toxicity. Modern conventional infant ventilators synchronize the mechanical breaths with the infant’s own respiratory effort. Lung injury can be minimized by careful attention to appropriate tidal volumes and adequate end expiratory pressure (54). High-frequency oscillatory ventilation (HFOV) is an alternative to conventional ventilation. The HFOV apparatus delivers small tidal volumes often equal to or less than the infant’s dead space. These volumes are generated by a piston or diaphragm that moves rapidly back and forth, providing both an active inspiratory and expiratory phase and facilitating carbon dioxide removal. Frequencies commonly used in infants are 8–12 Hz. The oscillations are superimposed on a constant mean airway pressure chosen to optimally recruit and oxygenate the lung. Adequate gas exchange thus may occur at substantially lower peak inspiratory pressures than required with conventional mechanical ventilation. Recently, a large, randomized, multicenter clinical trial was conducted in the United States in very low-birth-weight infants with RDS who still required substantial ventilator support after receiving surfactant. Infants were assigned to either early HFOV or conventional synchronized mechanical ventilation. The investigators found a small but significant improvement in pulmonary outcome in the infants managed with HFOV, with no difference between the groups with regard to extrapulmonary complications of prematurity (55). Other investigators, who used somewhat different protocols and HFOV equipment, have found no difference in pulmonary outcome (56).
Surfactant replacement therapy and mechanical ventilation are not universally effective in premature infants with RDS, possibly because of uneven surfactant distribution or concurrent sepsis, acidosis, or patent ductus arteriosus. Extracorporeal membrane oxygenation (ECMO), a form of cardiopulmonary bypass, is reserved for treating reversible respiratory failure refractory to conventional ventilatory measures (57). However, because ECMO requires anticoagulation and because premature infants are at increased risk for intracranial hemorrhage, many treatment centers restrict use of ECMO to newborns of at least 34 weeks gestational age. Because ECMO requires large cannula sizes, the procedure cannot be performed in most infants weighing less than 2000 g.
Radiologic Features
The radiographic findings in untreated RDS predictably reflect the generalized acinar collapse that results from surfactant deficiency. Thus, chest radiography demonstrates decreased lung expansion, symmetric generalized consolidation of variable severity, effacement of normal pulmonary vessels, and air bronchograms. The familiar "reticulogranular" texture of the lung opacities in RDS represents the summation of collapsed alveoli, transudation of fluid into the interstitium from capillary leak, and distention by air of innumerable bronchioles that remain more compliant than surfactant-deficient lung (Fig 11) (58). These radiographic findings are usually present shortly after birth but occasionally do not reach maximum severity until 12–24 hours of life (22). In severe cases of RDS, dense bilateral symmetric lung consolidation (so-called white out) may completely efface the cardiomediastinal and diaphragm contours. Before the widespread use of exogenous surfactant therapy, the chest radiographs of infants with RDS who required only short-term assisted ventilation typically demonstrated granularity that evolved to generalized hazy opacities to clearing over several days to 2–3 weeks (22).
View larger version (117K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 11. Mild RDS. Magnified radiograph of the right lung of a preterm neonate shows effacement of vascular definition by diffuse reticulogranular opacities. Peripheral air bronchograms (arrows) are visible at the medial lung base. The minor fissure (arrowheads) is slightly thickened.
|
|
The radiologic impact of exogenous surfactant therapy in premature infants with RDS has been analyzed in multiple reports. In 1985, Edwards et al (59) described a nonuniform pattern of radiographic improvement in established RDS in 18 newborns who received human lung surfactant derived from amniotic fluid, a phenomenon subsequently reported by other authors (60,61). In a randomized controlled study reported by Wood et al (62) in 1987, preventilatory administration of calf lung surfactant extract produced substantial clinical and radiographic improvement in RDS within 24 hours of life. In 1991, Levine and colleagues (63) published their comparison of radiographic findings following synthetic surfactant administration with those following human surfactant treatment. Radiographic improvement in infants who received synthetic surfactant, compared with age-matched patients given human surfactant, was slightly delayed, but no other significant differences were observed. Delayed radiographic improvement was also found in a multicenter randomized trial, in which natural surfactant extracts initially resulted in greater improvement in oxygenation and pulmonary compliance than did synthetic surfactants (33). A higher incidence of uniform clearing on chest radiographs was associated with newborns treated prophylactically with surfactant, possibly because the study group included neonates who never had RDS (64) or because there was more uniform mixing of surfactant and lung fluid during preventilatory instillation (62). In 1997, Dinger et al (65) reported the radiologic findings in 110 neonates (mean gestational age, 28 weeks) treated with exogenous (bovine) surfactant for established RDS. Uniform improvement was demonstrated on posttreatment chest radiographs in 38% of the infants, asymmetric improvement in 35%, no improvement in 10%, and interstitial gas in 17%. Of the patients who showed uniform improvement in lung opacities, 74% had almost completely normal chest radiographs after a single surfactant dose. Asymmetric or focal improvement tended to be most pronounced in the middle and upper regions of the right lung, and residual areas of lung opacity resolved within several days, often after additional surfactant treatment. Similarly, in a retrospective study of 39 premature infants who received surfactant after RDS was established radiographically, Slama and colleagues (66) noted transient asymmetric clearing of the right lung in 23%.
Several explanations have been proposed for asymmetric radiographic improvement following surfactant therapy (65): (a) maldistribution of surfactant into the right mainstem bronchus, (b) insufficient surfactant requiring additional applications, and (c) regional differences in aeration before surfactant treatment.
"Classic" chest radiographic findings (Fig 12) are less commonly observed in neonates with RDS since surfactant replacement therapy became ubiquitous. Emerging radiographic patterns associated with surfactant administration may complicate image interpretation, particularly when surfactant has been administered before baseline imaging. Asymmetric multifocal areas of opacity may be mistaken for neonatal pneumonia or meconium aspiration syndrome (Fig 13). Localized overaeration of selective acinar populations may produce cystic lucencies that mimic interstitial air leak (22). Asymmetric unilateral improvement in RDS, typically resulting in a hyperlucent right lung with contralateral mediastinal shift, may nominally resemble a tension pneumothorax (Fig 14). Finally, pulmonary hemorrhage, a rare complication following surfactant treatment, is typically heralded by acute respiratory decompensation that occurs after initial clinical improvement and produces sudden dense airspace consolidation on radiographs (Fig 15). Although the precise mechanism of pulmonary hemorrhage following surfactant therapy remains unclear, it has been suggested that improved ventilation and decreased pulmonary vascular resistance following surfactant administration promotes left-to-right shunting through the ductus arteriosus, producing hemorrhagic pulmonary edema (67).
View larger version (147K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 12a. Symmetric surfactant effect in a 36-week-gestationalage infant of a diabetic mother. (a) Pretreatment radiograph shows diminished lung expansion, diffuse bilateral reticulogranular opacities, and air bronchograms, findings consistent with severe RDS. (b) Repeat radiograph, obtained 6 hours after endotracheal administration of one dose of surfactant, reveals marked improvement in lung aeration and vascular definition.
|
|
View larger version (141K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 12b. Symmetric surfactant effect in a 36-week-gestationalage infant of a diabetic mother. (a) Pretreatment radiograph shows diminished lung expansion, diffuse bilateral reticulogranular opacities, and air bronchograms, findings consistent with severe RDS. (b) Repeat radiograph, obtained 6 hours after endotracheal administration of one dose of surfactant, reveals marked improvement in lung aeration and vascular definition.
|
|
View larger version (166K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 13. Asymmetric surfactant effect in a 2-day-old, 32-week-gestational age newborn with RDS. Frontal chest radiograph demonstrates multifocal residual consolidations that mimic pneumonia or meconium aspiration syndrome.
|
|
View larger version (159K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 14. Asymmetric distribution of endotracheal surfactant into the right mainstem bronchus in a 1-day-old preterm neonate with RDS. Frontal radiograph of the chest shows a clear hyperexpanded right lung, shift of mediastinal structures to the left, and persistence of diffuse left lung opacification.
|
|
View larger version (155K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 15a. Pulmonary hemorrhage in a 26-week-gestationalage neonate following prophylactic surfactant therapy. (a) Frontal chest radiograph obtained after one dose of surfactant and during treatment with nasal CPAP shows hyperinflated lungs with faint symmetric residual opacities. (b) Repeat radiograph, obtained after 24 hours for evaluation of sudden respiratory decompensation and bloody endotracheal aspirates, shows dense bilateral airspace consolidation.
|
|
View larger version (156K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 15b. Pulmonary hemorrhage in a 26-week-gestationalage neonate following prophylactic surfactant therapy. (a) Frontal chest radiograph obtained after one dose of surfactant and during treatment with nasal CPAP shows hyperinflated lungs with faint symmetric residual opacities. (b) Repeat radiograph, obtained after 24 hours for evaluation of sudden respiratory decompensation and bloody endotracheal aspirates, shows dense bilateral airspace consolidation.
|
|
The increased survival rate of very low-birth-weight (<1500 g) and extremely low-birth-weight (<1000 g) neonates has introduced radiographic features peculiar to structurally immature lungs. In 1975, Krauss et al (68) recognized that some very low-birth-weight infants with initially normal chest radiographic findings developed delayed respiratory insufficiency at 4–17 days of life. This condition was termed chronic pulmonary insufficiency of prematurity (CPIP). In 1980, Edwards et al (69) identified 36 very low-birth-weight neonates who did not fulfill the clinical criteria for RDS and who had a mature pattern of surfactant phospholipid components in pulmonary effluent. The chest radiographs of these newborns demonstrated radiating perihilar areas of opacity and diffuse fine granularity, which was less pronounced than that typically seen in RDS and was not associated with significant air bronchograms or decreased lung expansion. Nevertheless, more than two-thirds of these infants required ventilatory support for episodes of apnea and bradycardia, left-to-right shunt through a patent ductus arteriosus, necrotizing enterocolitis, and intercurrent sepsis. Because RDS of surfactant deficiency was excluded by definition, the authors speculated that the fine granularity seen in chest radiographs of this group of patients represented thickened alveolar interstitium due to structural immaturity and that the radiating areas of opacity reflected excessive lung fluid. This condition was designated immature lung.
Today, it is not uncommon to encounter profoundly premature newborns weighing substantially less than 1000 g, whose initial chest radiographs may appear normal or display only subtle haziness or interstitial thickening (Figs 16, 17) (22). Despite the absence of granular lung opacities typically associated with RDS, alveolar paucity in extremely low-birth-weight neonates often necessitates continued ventilator support, subjecting the lungs to the deleterious effects of high oxygen concentration and positive pressure. Over several days to weeks, the chest radiographs of these infants may evolve from normal or near-normal appearance to diffuse haziness to a coarse, irregular pattern of chronic lung disease (Fig 17) (22).
View larger version (159K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 16. Immature lungs in a profoundly premature neonate delivered at a gestational age of 22 weeks and 6 days. Anteroposterior chest radiograph obtained on the first postnatal day shows diffuse fine interstitial thickening.
|
|
View larger version (148K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 17a. Immature lungs in a 24-week-gestational-age neonate with a birth weight of 530 g. (a) Anteroposterior radiograph obtained on the second postnatal day shows clear lungs. (b) Repeat radiograph, obtained at 48 days, shows coarse parenchymal opacities with no regions of disordered aeration or cystic lucency, findings consistent with uniform fibrosis in "new" BPD.
|
|
View larger version (139K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 17b. Immature lungs in a 24-week-gestational-age neonate with a birth weight of 530 g. (a) Anteroposterior radiograph obtained on the second postnatal day shows clear lungs. (b) Repeat radiograph, obtained at 48 days, shows coarse parenchymal opacities with no regions of disordered aeration or cystic lucency, findings consistent with uniform fibrosis in "new" BPD.
|
|
|
Bronchopulmonary Dysplasia
|
|---|
Historical Background, Clinical Criteria, and Evolving Terminology
In 1960, Wilson and Mikity (70) described a new form of respiratory disease of unknown cause in five premature infants with a mean weight of 1342 g. None received assisted ventilation or >40% supplemental oxygen, and the onset of tachypnea, dyspnea, and cyanosis occurred between 1 and 5 weeks of age. Chest radiographs demonstrated "coarse, streaky infiltration with small areas of emphysema, occasionally appearing cystic." Histologic findings included alveolar overdistention, septal fibrosis, and interstitial mononuclear infiltration. The original description of this constellation of abnormalities, which became known as the Wilson-Mikity syndrome, predated the widespread use of mechanical assisted ventilation to treat RDS. In 1967, the term bronchopulmonary dysplasia (BPD) was introduced by Northway et al (71) to describe a chronic lung disease occurring in premature infants with RDS who were treated with positive-pressure mechanical ventilation and supplemental oxygen and who developed radiographic abnormalities indistinguishable from those seen in Wilson-Mikity syndrome.
Since its original description almost 4 decades ago, the epidemiology, clinical presentation, and radiologic features of BPD have changed substantially. In the original series published by North-way et al (71), surviving infants with BPD initially presented with severe RDS and had an average gestational age and birth weight of 34 weeks and 2235 g, respectively. Since the introduction of surfactant replacement therapy for RDS, antenatal glucocorticoid treatment, and refinements in assisted ventilation, BPD is now uncommon in larger and more mature infants with a gestational age exceeding 30 weeks or weighing more than 1200 g at birth (72). However, these clinical advances have also improved survival for extremely low-birth-weight infants (<1000 g), who often do not have radiographic evidence of RDS and whose lung injury is not necessarily initiated by oxygen delivery and mechanical ventilation (73).
Traditionally, the diagnosis of BPD was assigned to infants who received mechanical ventilation and oxygen delivery for a minimum of 3 days during the first 2 weeks of life, who exhibited clinical signs of respiratory compromise beyond 28 days, who required supplemental oxygen beyond 28 days of age to maintain a PaO2 above 50 mm Hg, and who developed the characteristic radiographic abnormalities originally described by Northway (74,75). This restrictive definition has been applied in an effort to identify patients who will have significant pulmonary dysfunction during the first year of life (76).
The original criteria for the diagnosis of BPD have grown less reliable as increasingly immature infants have survived. Low-birth-weight infants may fulfill the diagnostic requirements for BPD at 28 days of life but no longer meet the criteria as they approach 40 weeks postmenstrual age (defined as gestational age at birth plus chronologic age) (76,77). In a retrospective study of 119 infants with birth weights less than 1500 g and with follow-up data available for 2 years after birth, Shennan and colleagues (78) found that the requirement for supplemental oxygen at 36 weeks postmenstrual age was a superior predictor of abnormal outcome. On the basis of this revised definition, approximately 30% of infants with birth weight less than 1000 g develop BPD (79). It has also been recognized that radiologic abnormalities in this subset of very immature infants who meet the proposed new clinical requirements for BPD are frequently dissimilar to those originally described by Northway (75). Consequently, the designation chronic lung disease of infancy has gained favor to describe the form of BPD encountered in very low-birth-weight infants (80).
A workshop organized by the National Institute of Heart Disease, the National Heart, Lung, and Blood Institute, and Office of Rare Diseases was recently convened to review the definition of BPD, propose new diagnostic criteria, and establish future research priorities (72). By consensus, the workshop participants elected to retain the term bronchopulmonary dysplasia as the best descriptor of the chronic lung disease peculiar to premature infants. A new definition of BPD (Table) was developed for infants with gestational age less than 32 weeks and greater than 32 weeks based on time of clinical assessment and clinical severity.
Pathogenesis and Pathologic Features
Before the widespread use of surfactant replacement therapy, RDS was considered the precursor of BPD in preterm newborns (81). In this traditional clinical setting, positive-pressure ventilation of surfactant-deficient immature lungs with poorly compliant alveoli was understood to cause repetitive mechanical or ischemic damage to over-distended terminal airways. Airway rupture resulting in interstitial air leak was known to increase the relative risk of BPD by a factor of six (82), and traumatic mechanical ventilation contributed to capillary leak pulmonary edema (81,83). Finally, premature infants were known to be more susceptible to lung damage by free oxygen radicals because of inadequate concentrations of antioxidant enzymes. Lung damage by oxygen radicals, known as oxygen toxicity, may synergize with bronchopulmonary injury from assisted ventilation, originally referred to as barotrauma (76). The type of BPD that resulted from the treatment of immature lungs with oxygen delivered by positive-pressure ventilation was characterized by terms such as necrotizing bronchiolitis, alveolar cell hyperplasia, bronchiolar squamous metaplasia, and focal alveolar septal fibrosis (84,85). Structural changes in the pulmonary arteries were similar to those seen in hypertensive vascular disease and included intimal proliferation, medial hyperplasia, and adventitial thickening.
Alveolar septal fibrosis is the predominant residual feature in the condition known as longstanding healed BPD (84). In 1986, Stocker (84) recognized that the severity of alveolar septal fibrosis in long-standing healed BPD varied from acinus to acinus in the same patient. He postulated that intraluminal plugs of inflammatory debris related to acute necrotizing bronchiolitis in terminal airways protected some acinar populations from the untoward effects of high oxygen tensions and positive-pressure ventilation, while adjacent unprotected acini underwent uniform alveolar septal fibrosis (Fig 18). According to this theory, eventual removal of the occlusive plugs by alveolar macrophages produced alternating populations of hyperinflated but otherwise normal alveoli and scarred alveoli, leading to a "cobblestone" appearance of the lung surface and cleavage of the visceral pleura by multiple deep pseudofissures (Fig 19).
View larger version (56K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 18. Proposed model for the pathogenesis of long-standing healed BPD. Schematic 1 represents three normal acini (a, b, c), each supplied by a terminal bronchiole. As shown in schematic 2, during positive-pressure ventilation and oxygen delivery for treatment of surfactant deficiency, hyaline membranes or occlusive debris from necrotizing bronchiolitis (arrow) "protects" acinus a from alveolar septal injury, whereas varying degrees of partial occlusion of the bronchioles to acini b and c permit alveolar septal necrosis from barotrauma-volutrauma and oxygen toxicity. Schematic 3 depicts the phase following resolution of bronchiolar obstruction, in which acinus a hyper-expands and continues to develop new alveoli, acinus b has undergone septal fibrosis and is inhibited from further alveolar development, and acinus c has atrophied. (Illustration by Aletta Ann Frazier, MD, Department of Radiologic Pathology, Armed Forces Institute of Pathology.) (Adapted, with permission, from reference 85.)
|
|
View larger version (152K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 19a. Long-standing "healed" BPD, pathologic features. (a) Photomicrograph (original magnification, x30; H-E stain) demonstrates diffuse alveolar septal fibrosis within an acinus (bottom) adjacent to another acinus (top) with hyperexpanded but otherwise normal alveoli. (b) Photomicrograph (original magnification, x100; H-E stain) depicts a continuous band of hyperplastic smooth muscle (arrow) surrounding a bronchiole. (c) Photograph of the pleural surface of a lung at 1 month of age shows alternating populations of hyperexpanded and collapsed acini de-fined by irregular pseudofissures (arrows).
|
|
View larger version (189K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 19b. Long-standing "healed" BPD, pathologic features. (a) Photomicrograph (original magnification, x30; H-E stain) demonstrates diffuse alveolar septal fibrosis within an acinus (bottom) adjacent to another acinus (top) with hyperexpanded but otherwise normal alveoli. (b) Photomicrograph (original magnification, x100; H-E stain) depicts a continuous band of hyperplastic smooth muscle (arrow) surrounding a bronchiole. (c) Photograph of the pleural surface of a lung at 1 month of age shows alternating populations of hyperexpanded and collapsed acini de-fined by irregular pseudofissures (arrows).
|
|
View larger version (103K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 19c. Long-standing "healed" BPD, pathologic features. (a) Photomicrograph (original magnification, x30; H-E stain) demonstrates diffuse alveolar septal fibrosis within an acinus (bottom) adjacent to another acinus (top) with hyperexpanded but otherwise normal alveoli. (b) Photomicrograph (original magnification, x100; H-E stain) depicts a continuous band of hyperplastic smooth muscle (arrow) surrounding a bronchiole. (c) Photograph of the pleural surface of a lung at 1 month of age shows alternating populations of hyperexpanded and collapsed acini de-fined by irregular pseudofissures (arrows).
|
|
The widespread adoption of antenatal glucocorticoid administration and postnatal surfactant therapy, the decreased need for high oxygen tensions and ventilatory pressures coupled with close attention to "gentler" ventilation, and the increased survival of very low-birth-weight infants have altered the pathologic characteristics of BPD. The synergy of oxidant injury and mechanical ventilation is no longer considered the major trigger of lung injury in very immature neonates, who now constitute the majority of patients who develop BPD (9). Initially, these very low-birth-weight infants may have little or no radiographic evidence of lung disease, but subsequently they require supplemental oxygen or mechanical ventilation over the first several weeks of life.
Perinatal factors that influence lung maturation likely play an important role in the pathogenesis of BPD in the era of surfactant therapy. Exposure to chronic low-grade chorioamnionitis, a common clinical problem associated with preterm delivery, is known to increase the risk of lung injury (86). In particular, the genital mycoplasma Ureaplasma urealyticum, the most common contaminant of amniotic fluid, has been associated with the development of BPD (87–89). In this clinical setting, antenatal exposure to proinflammatory cytokines in amniotic fluid may prime the lung for postnatal injury, which is subsequently amplified by even gentle mechanical ventilation and oxygen exposure (9). In 1999, Jobe (9) coined the term new BPD to distinguish the emerging type of chronic lung injury in very immature infants from the condition originally described by Northway. In 2003, Hodgman (90) suggested that the "new" BPD and the condition originally described in 1960 by Wilson and Mikity (70) are the same entity, each attributable to an aberrant response of the immature lung to early air breathing rather than damage from barotrauma and oxygen toxicity. Although its pathogenesis is multifactorial and incompletely understood, accumulating evidence strongly suggests that the new BPD is fundamentally an inhibition of acinar and vascular growth during a vulnerable stage of lung development (Fig 20) (9).
To better define the pathologic features of the new BPD, Husain and colleagues (85) reviewed the postmortem histologic lung specimens from 14 preterm infants who were treated with surfactant and eight who received no surfactant, and compared the findings to those from 15 age-matched control infants. They concluded that alveolar septal fibrosis was substantially less extensive and, when present, more diffuse in surfactant-treated lungs than in untreated lungs, and that both populations underwent a similar degree of partial to complete arrest in alveolar saccule development after birth (Fig 21). The authors suggested that the widespread use of surfactant therapy and the concomitant decreased exposure to high oxygen tensions and ventilatory pressures mitigated the necrotizing bronchiolitis and severe alveolar septal fibrosis found in classic BPD. Furthermore, they postulated that the absence of transient occlusion of some bronchioles by inflammatory debris in surfactant-treated lungs resulted in uniform exposure of acini to gentler mechanical ventilation and comparatively low oxygen tensions. Thus, the heterogeneous lung injury and repair characteristic of long-standing healed BPD in the era before surfactant therapy have been replaced by minimal or mild diffuse alveolar septal fibrosis and an oversimplified acinar morphology (Fig 22).
View larger version (55K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 21. Arrested acinar development in post-surfactant BPD. Schematics of a trio of acini show oversimplified anatomy due to alveolar paucity with either thin alveolar septa (center) or uniform mild fibrosis (right), compared with normal alveolar growth and development (left). (Illustration by Aletta Ann Frazier, MD, Department of Radiologic Pathology, Armed Forces Institute of Pathology.) (Adapted, with permission, from reference 85.)
|
|
View larger version (205K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 22. BPD in a surfactant-treated infant. Photomicrograph (original magnification, x50; H-E stain) shows expanded, simplified alveolar ducts, saccules, and alveoli with little or no alveolar septal fibrosis.
|
|
Radiologic Features
Northway and colleagues (71) combined the clinical, pathologic, and radiologic criteria of BPD to define four developmental stages. Chest radiographs in stage I BPD (2–3 days after birth) showed the typical granular opacities of RDS. In stage II BPD (4–10 days), nearly complete opacification of the lungs was identified in the most severe examples. Chest radiographs in stage III BPD (10–20 days) showed small round cystic lucencies alternating with regions of irregular opacity. Stage IV BPD (beyond 1 month) demonstrated further enlargement of lucent regions alternating with thin strands of increased opacity, an appearance that has come to be called "bubbly lungs" (Fig 23 ) (91). Since this original definition, a number of quantitative scoring systems have been proposed that integrate clinical and radiographic criteria in an effort to predict chronic oxygen dependency in premature neonates (92–99). As yet, none of these scoring systems have been universally accepted.
View larger version (148K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 23a. "Classic" severe BPD in a 3-month-old premature infant. (a) Frontal radiograph shows heterogeneous aeration, coarse strandlike areas of opacity, and intervening cystic lucencies. (b) Axial CT scan demonstrates right upper lobe regional air trapping anteriorly, architectural distortion with fibrotic subpleural parenchymal bands (arrows) and subsegmental atelectasis posteriorly, and diffuse coarse reticular opacities in the left lung.
|
|
View larger version (175K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 23b. "Classic" severe BPD in a 3-month-old premature infant. (a) Frontal radiograph shows heterogeneous aeration, coarse strandlike areas of opacity, and intervening cystic lucencies. (b) Axial CT scan demonstrates right upper lobe regional air trapping anteriorly, architectural distortion with fibrotic subpleural parenchymal bands (arrows) and subsegmental atelectasis posteriorly, and diffuse coarse reticular opacities in the left lung.
|
|
The four radiographic stages of BPD originally described by Northway et al (71) are less commonly observed today. Surfactant replacement therapy and frequent survival of very immature neonates have contributed to the changing radiographic picture of chronic lung disease in premature infants. In these infants, sequential chest radiographs often reveal a gradual and subtle progression from clear or minimally abnormal lungs through hazy ground-glass opacity to a relatively uniform pattern of coarse interstitial opacities without cystic lucencies (Fig 17). In those patients who eventually manifest "bubbly" lungs, the radiographic abnormality tends to be symmetric and the cystic lucencies smaller and more uniform than originally described in stage IV BPD (Fig 24).
In a study of 75 premature infants who received rescue surfactant therapy, positive-pressure ventilation, and supplemental oxygen, Swischuk and coworkers (91) reported prompt clearing of granular lung opacities in 45%, all of whom were larger infants. In 31% of the patients in the study group, initial radiographic clearing was followed by "hazy opaque" lungs by day 7 to day 14, of which 35% cleared and 65% progressed to bubbly lungs. In 24% of the study group, granularity never cleared, and most progressed directly to bubbly lungs. The smaller infants with mean gestational ages of 25–27 weeks and mean gestational weights of 832–979 g were more likely to develop chronic lung abnormalities. The authors suggested that hazy opaque lungs on chest radiographs represented capillary leak pulmonary edema due to hypoxia and oxygen toxicity in infants who continued to require ventilatory support, despite initial clearing with surfactant therapy. The term leaky lung syndrome was introduced to emphasize the proposed pathogenesis of this condition, which may or may not progress to bubbly lungs of classic BPD.
In those infants who do develop the bubbly lungs characteristic of severe BPD, chest radiographs may show
Top
Abstract
LEARNING OBJECTIVES FOR TEST...
