Published online before print June 27, 2005, 10.1148/radiol.2361041278
(Radiology 2005;236:22-29.)
© RSNA, 2005
Interpretation of Chest Radiographs in Infants with Cough and Fever1
Robert T. Bramson, MD,
N. Thorne Griscom, MD and
Robert H. Cleveland, MD
1 From the Department of Radiology, Children's Hospital, Boston, 300 Longwood Ave, Boston, MA 02115. Received July 22, 2004; revision requested September 22; revision received November 10; accepted December 10.
Address correspondence to R.T.B. (e-mail: robert.bramson{at}childrens.harvard.edu).
 |
ABSTRACT
|
|---|
An understanding of the appearance of the infant chest radiograph requires an understanding of the anatomy and the physiologic, immunologic, and pathologic processes in the infant's chest. The authors describe the features of the infant chest that most influence the appearance of the chest radiograph in infants with cough and fever. They discuss why confusion sometimes occurs when radiology residents and general radiologists familiar with adult chest radiographs first evaluate the infant chest radiograph. The radiographic appearance of acute inflammation does not look the same in infants as it does in older children and adults. The hallmark of inflammatory lung disease in the infant chest is air trapping on the chest radiograph.
© RSNA, 2005
|
INTRODUCTION
|
|---|
EDITOR'S NOTE: Please see the January 2005 From the Editor, where this new feature, Review for Residents, was announced.
Radiology residents have often expressed to us their frustration and confusion when attempting to interpret chest radiographs of infants. Radiologists not trained in pediatric radiology find terms like air trapping, peribronchial thickening, bronchiolitis, lower respiratory tract infection, and peribronchial pneumonia confusing (1). Pediatric radiologists have, in fact, conflicting definitions of these terms and varying opinions about their meaning. The authors hope to clarify the relationship between the pathologic changes within the infant chest and their radiographic appearance, explaining in pathophysiologic terms why the infant chest looks the way it does in inflammatory airways diseases.
An understanding of the infant chest radiograph requires a review of how embryology, anatomy, physiology, pathology, immunology, and the physics of fluid mechanics influence its appearance. This knowledge is critical in the interpretation of the infant chest radiograph.
|
EMBRYOLOGY AND ANATOMY
|
|---|
We will summarize the science that explains the appearance of the infant chest, with the caveat that universal agreement on the details does not exist. All of the generations of the airways have developed by the time the fetus reaches the 16th week of gestational age. There are about 22 generations of airways, depending on how the last generation is counted and where the count is performed. Near the lung hila, there may be as few as 10 generations of airways before the gas-exchange units—the respiratory bronchioles and alveolar sacs—are reached. In the lung periphery, there may be as many as 25 generations of airways before the gas-exchange units are reached. As the child grows and becomes an adult, the airways grow in length and diameter but not in number (2–4) (Fig 1).
View larger version (27K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1. Diagram shows gas exchange that occurs in the respiratory bronchiole and alveolar sac. In the lung periphery there may be as many as 25 generations of airways before the respiratory bronchiole is reached or as few as 10 (near the hila), depending on where the count is performed. Inset shows how a distal bronchiole may become narrowed with edema and mucus during inflammation.
|
|
The alveoli, the gas-exchange units, develop after the airways. They start proliferating about the 29th week of gestational age. By the 40th week of gestation, there are approximately 20 million alveoli in the newborn lung, although the precise number is debated. The mature lung contains approximately 300 million alveoli; that number is reached at about 8 years of age (2). The alveoli then increase in size, and their lining gets progressively thinner as the child becomes an adult. The total alveolar surface area is 70–80 m2 in the adult. In the newborn infant, the total alveolar surface area is only about 3 m2. The ratio of the alveolar surface area to the body surface area is more than three times as great in the adult as in the newborn (5). Thus, the infant is at a distinct disadvantage, compared with the adult and older child, in his or her reserve of alveolar surface area for gas exchange. The alveolar walls contain the pulmonary capillaries, which form a giant blood surface area for the exchange of respiratory gases (4).
The adult lung and that of the older child have communications between the alveoli, the pores of Kohn (intraalveolar pores), and the channels of Lambert (bronchoalveolar channels). These allow collateral air drift between the airspaces and are far fewer in infants. Their relative absence influences the appearance of radiographs in infants with lower respiratory infection, as will be explained later (4).
The structure of the chest wall in the infant differs from that in the adult. The infant's ribs and adjacent soft tissues are more elastic and compliant. As the child grows, the soft tissues and ribs become stiffer (6). Watching a baby breathe makes this obvious. The infant moves his or her chest wall in and out more than the adult does, particularly when in respiratory distress. The more marked compliance of the soft tissues of the infant thorax allows retractions of those tissues between the ribs. The lack of stiffness in the soft tissues requires more work during breathing. When the infant is in respiratory distress, this increased work becomes more obvious. The sight of a sick infant struggling to breathe can be frightening, and the grunting and retractions graphically demonstrate the increased work of breathing. In contrast, adults with pneumonia do not work particularly hard moving air in and out of their lungs unless the pneumonia is extensive.
During inspiration, the intrathoracic airways increase in cross-sectional area. During expiration, the airways narrow and the cross-sectional area decreases. This normal variation in luminal size is accentuated in the infant because the support tissues around the airways are more compliant and allow more narrowing during expiration. Compared with an adult, the number of alveoli is relatively low in the infant, and the proportion of the lung involved in air transport (the airways) is relatively high. In the adult, 80% of the airways are bigger than 2 mm in diameter. The smaller peripheral airways in adults account for less than 20% of the total resistance to the flow of air. In the infant lung, the peripheral airways are considerably smaller, and the resistance to air flow owing to these small airways is 50% of the total resistance (4) (Fig 2).
View larger version (70K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2a. (a) Diagram shows that during normal inspiration and expiration, there is dilation and collapse of the airways. This is most obvious in distal airways. Collapse in infant airways is greater than that in adult airways because cartilaginous soft tissues supporting the airways are more compliant in children. This is illustrated on (b) a lateral chest radiograph of an infant obtained near the end of normal expiration. The trachea (arrows) collapses to a much smaller diameter during normal expiration.
|
|
View larger version (92K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2b. (a) Diagram shows that during normal inspiration and expiration, there is dilation and collapse of the airways. This is most obvious in distal airways. Collapse in infant airways is greater than that in adult airways because cartilaginous soft tissues supporting the airways are more compliant in children. This is illustrated on (b) a lateral chest radiograph of an infant obtained near the end of normal expiration. The trachea (arrows) collapses to a much smaller diameter during normal expiration.
|
|
|
PHYSIOLOGY
|
|---|
Growth is proportionately more rapid in infancy than at any other time of postnatal life. The metabolic requirements for that growth include large amounts of oxygen (6). However, as previously mentioned, the adult lung has a ratio of surface area of the pulmonary capillary bed (gas-exchange area) to body surface area that is more than three times as large as that in the newborn infant (5).
The healthy newborn infant breathes 40 times per minute. The healthy adult breathes 16 times per minute. The high demand for oxygen in the healthy infant plus the relatively small gas-exchange area per unit of body surface area taxes the respiratory system much more in the infant than in the adult. This is one of the causes of the relative tachypnea of healthy infants. Infants compensate for increased oxygen demand primarily by increasing their respiratory rate (4,7).
Resistance to the flow of air through the airways is higher in the infant than in the adult. This is both because the compliance of the tissues surrounding the infant airways makes it easier for the airways to narrow during normal expiration and because of the higher percentage of small airways in the infant lung (4). The resistance of the flow of air through a cylinder, such as an airway, is described by the Poiseuille law. A quick summary of this law is that the resistance to the flow of air through the airways varies inversely with airways radius to the fourth power (4). Thus, a tiny decrease in the diameter of the airways leads to a marked increase in resistance to the flow of air. As previously mentioned, the contribution of the peripheral airways to the resistance of flow is considerably greater in infants than in adults, even in normal circumstances (4).
When breathing at 16 times per minute, the adult has 3.75 seconds to move his or her diaphragm through the full course of inspiration and expiration. When breathing at 40 times per minute, the infant has only 1.5 seconds for this diaphragm movement. The infant increases oxygen exchange primarily by increasing the respiratory rate. When this happens, the diaphragm must change direction more frequently and either move faster or not move as far during each excursion. During periods of respiratory distress, these all occur.
The level of the diaphragm seen on chest radiographs is determined by several things. At all ages, the level of the diaphragm depends on how deep an inspiration the patient has made when the radiograph was obtained. The adult usually takes a deep breath when requested, and the diaphragm level is determined by voluntary action. The infant does not voluntarily take a deep breath and hold it for a chest radiograph. The technologist must guess and acquire the radiograph when the infant appears to have achieved maximum inspiration. If the infant diaphragm moves through the entire respiratory cycle in only 1.5 seconds, or even less time when the rate reaches 60 breaths per minute, the technologist has little time to make a correct guess. The diaphragm sits at a level determined by the resistance to flow of air through the airways. Figure 3 depicts lung volumes and diaphragm movement during respiration, first in health and then in disease, when air trapping has led to an increase in residual volume (8).
View larger version (20K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3. Graph depicts lung volumes at inspiration and expiration. Line on the left shows lung volume at expiration (A) and inspiration (B), as well as maximum expiration (C) and inspiration (D) during normal quiet respiratory cycles. The line on the right shows that when peripheral small-airways resistance is high (a, b), then residual volume (RV) is increased. This is the air trapping depicted on radiographs of infants with small-airways disease. c = Maximum expiration, d = maximum inspiration, ERV = expiratory reserve volume, IRV = inspiratory reserve volume, TLC = total lung capacity, TV = tidal volume, VC = vital capacity. (Reprinted, with permission, from reference 7.)
|
|
Normally, the dome of the right hemidiaphragm at inspiration is projected at about the level of the sixth anterior rib on a chest radiograph. A normal lateral chest radiograph will show a domed, rounded configuration of the diaphragm. Figure 4 shows the typical configuration of the diaphragm on anteroposterior and lateral chest images in a healthy small infant.
View larger version (138K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4a. (a) Anteroposterior radiograph of normal chest in a 4-month-old child referred because of a possible fractured clavicle. (b) Lateral radiograph in the same infant shows rounded configuration of the diaphragm (arrows).
|
|
View larger version (134K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4b. (a) Anteroposterior radiograph of normal chest in a 4-month-old child referred because of a possible fractured clavicle. (b) Lateral radiograph in the same infant shows rounded configuration of the diaphragm (arrows).
|
|
|
IMMUNOLOGY AND PATHOLOGY
|
|---|
Growing children are exposed to many infectious organisms and need to develop immunity to them. The average adult inhales more than 9000 L of air per day; the infant, much less (4). A multitude of organisms enter the airways along with this inspired air. The organisms that infect the respiratory tract in infancy are usually viral. The most severe diseases in the lower respiratory tract are caused by the parainfluenza viruses and the respiratory syncytial virus (9,10). These are also among the most common organisms to infect the infant's respiratory tract (9–11). Adults have some immunity to most of these organisms because they were exposed to them as children and developed an immunity. Although adults may be infected and transmit these viruses to others, they usually become no more than mildly ill. Infants have not yet developed substantial immunity to these viruses and often get sick because of them, especially the parainfluenza and respiratory syncytial viruses.
The terminology for lower respiratory tract infections in infants is confusing. For example the definition of bronchiolitis varies with local pediatric usage. Some physicians limit the term bronchiolitis to respiratory syncytial virus infection in children. Affected children may have retractions, tachypnea, air hunger, and extreme respiratory distress. Other children of a similar age may have the same clinical findings, but respiratory syncytial virus cannot be cultured; some pediatricians also call this bronchiolitis. Other physicians may call all these clinical findings pneumonia, but still others are uncomfortable with that term because, to them, pneumonia means airspace consolidation on a chest radiograph. Sometimes the term peribronchial pneumonia or interstitial pneumonia is used to differentiate this airways infection from airspace pneumonia. Some individuals simply call all of these lower respiratory tract infections in the attempt to avoid the confusion in terminology.
Radiologists should be aware that the nomenclature is confusing. Heated debates may occur between physicians who use different definitions for these terms. We suggest that radiologists use the terms and definitions generally employed by the referring clinicians and be alert to the problem of confusion.
During these respiratory viral infections, the airways react in several ways, most notably with bronchoconstriction and increased secretion of mucus (12). These two factors have the effect of narrowing the cross-sectional area of the airways—particularly the small airways (12th generation airways or smaller) (Fig 5). These airways, which already contribute 50% of the total airways resistance, suddenly have a marked decrease in average radius. Since resistance is inversely proportional to the fourth power of the radius, this has the effect of greatly increasing the total resistance to air flow through the airways.
View larger version (76K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5. Diagram shows that during viral infection, airways secrete increased amounts of mucus and become edematous, particularly in smaller peripheral airways. This narrows the airways, and that narrowing is accentuated during attempts at expiration.
|
|
|
RADIOLOGIC FINDINGS
|
|---|
As was mentioned earlier, the primary way for the infant to increase ventilation is to increase his or her respiratory rate. Infants in respiratory distress due to viral infection often breathe 60–80 times per minute. Their air hunger can be recognized by the use of the accessory respiratory muscles. The increased work of breathing is largely caused by the increased resistance to the flow of air through the small airways. Their air hunger causes retractions and grunting as they work to overcome this resistance.
Infants breathing at 60 or more times per minute have less time to move the diaphragm through its inspiratory-expiratory cycle. They have only 1 second, sometimes less, for each breath. The diaphragm tends to move in a narrow range because of the resistance to the flow of air through their small airways (Fig 3).
The dead space in the upper airways cannot be reduced and remains constant. However, it gets more difficult to propel the dead-space air back up and out of the mouth because of the increased resistance in the peripheral airways. As the rising diaphragm pushes the air out of the lungs, it must almost immediately contract to start inspiration. The inspiration starts before the usual volume of air has been expelled. Because of the difficulty of pushing all the air out during expiration, the lungs are at a more expanded state when inspiration starts again—that is, there is an increase in residual volume, and there is air trapping (Fig 3). The lungs are at a high volume even at the end of expiration. On a radiograph, the diaphragm is projected at a level lower than the sixth anterior rib (Fig 6), and the diaphragm leaves are flattened. The radiologist will notice that there is an increase in the transverse diameter of the chest and a flattened diaphragm on an anteroposterior projection. On a lateral image, there is flattening of the diaphragm, and the sternum may be bowed upward and outward (Fig 7). Some radiologists gauge hyperaeration by counting posterior ribs, particularly if the radiograph is obtained with the patient in an apical lordotic position. Those who use posterior ribs in this situation usually believe the diaphragm is located at the level of the eighth posterior rib during normal inspiration (8).
View larger version (113K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6a. (a) Anteroposterior radiograph shows hyperinflated lungs with suggestion that peribronchial markings are too prominent. (b) Lateral radiograph shows flat slope to the diaphragm, with none of the rounded configuration seen in Figure 4b. The diaphragm now has a straight-line slope rather than a rounded configuration.
|
|
View larger version (109K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6b. (a) Anteroposterior radiograph shows hyperinflated lungs with suggestion that peribronchial markings are too prominent. (b) Lateral radiograph shows flat slope to the diaphragm, with none of the rounded configuration seen in Figure 4b. The diaphragm now has a straight-line slope rather than a rounded configuration.
|
|
View larger version (114K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7a. (a) Anteroposterior radiograph of infant chest shows hyperinflated appearance characteristic of infant inflammatory airways disease. Hemidiaphragm domes are projected at level of the seventh anterior rib or lower. (b) Lateral radiograph of hyperinflated chest shows diaphragm has a straight (not domed) slope.
|
|
View larger version (110K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7b. (a) Anteroposterior radiograph of infant chest shows hyperinflated appearance characteristic of infant inflammatory airways disease. Hemidiaphragm domes are projected at level of the seventh anterior rib or lower. (b) Lateral radiograph of hyperinflated chest shows diaphragm has a straight (not domed) slope.
|
|
To diagnose air trapping, the diaphragm needs to be flattened on both anteroposterior and lateral projections. If the lung volume is high on one radiograph and normal on the other, then the high-volume projection happened to be exposed at a moment of an unusually deep breath. If the diaphragm is well domed on the other view, then the infant is able to move air out of his or her lungs during expiration and there is no air trapping.
To summarize, these sick infants breathe faster and work harder to breathe, their airways narrow during expiration, and the greatly increased airways resistance severely impedes the flow of air. The air trapping revealed by the increase in lung volume on the chest radiograph is the best available indicator of inflammatory lung disease in infancy. Hyperinflation may be the only radiologic clue to illness in these children. The alveoli are usually clear, and there is none of the airspace consolidation associated with classic bacterial pneumonia in adults.
The radiologist should not become wedded to the relationship of the diaphragm to the sixth anterior rib, but this finding does serve as a useful rule of thumb. An experienced pediatric radiologist quickly recognizes hyperaeration at a glance without counting ribs. A less experienced observer will find the sixth anterior rib to be a helpful landmark.
There are often additional signs of inflammatory disease of the small airways. Edema and mucus in these airways can cause peripheral atelectasis. Small plugs in many small airways produce many small patches of atelectasis. If enough of these small patches accumulate in one region, a patch of atelectasis shows up on a chest radiograph (Fig 8). The abnormal appearance is often difficult to define, except that the interstitial lung tissues look prominent. Some observers call this peribronchial cuffing or peribronchial thickening or bronchial wall thickening. Sometimes the radiologist looks at the airways end on. If the radiologist thinks the walls of these airways (usually of the third, fourth, or fifth generation) look thicker, he or she will use the term peribronchial thickening or bronchial wall thickening. In reality, thickening of the smaller airways (12th generation or higher) has a far more deleterious effect on gas exchange than does thickening of the airways that are identifiable on a radiograph. The patches of peripheral atelectasis may shift when the infant coughs and dislodges small mucus plugs. Therefore, the radiographic appearance may change from image to image. Hyperinflation, however, remains the major clue to inflammatory small-airways disease (12).
View larger version (124K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8a. Respiratory syncytial virus infection in a child. (a) Anteroposterior radiograph shows prominent peribronchial markings. (b) Patches of atelectasis (arrow) are best seen on lateral projection of the hyperinflated lungs. Scattered patches of atelectasis tend to follow peribronchial and perivascular structures in a child with respiratory syncytial virus infection.
|
|
View larger version (132K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8b. Respiratory syncytial virus infection in a child. (a) Anteroposterior radiograph shows prominent peribronchial markings. (b) Patches of atelectasis (arrow) are best seen on lateral projection of the hyperinflated lungs. Scattered patches of atelectasis tend to follow peribronchial and perivascular structures in a child with respiratory syncytial virus infection.
|
|
As the infant improves, interesting things happen. The diaphragm returns to a more normal level, but the radiograph may show increasing patches of atelectasis. This can confuse pediatricians and radiologists alike. Perhaps as the hyperexpansion of the lung decreases, some alveoli kept open by the air trapping now succumb and collapse, leading to an appearance of increased atelectasis. In reality, the child is improving even though the radiograph may look mildly worse.
|
CAVEATS
|
|---|
The changes we have described are the most common ones in lower-airways inflammatory disease in infancy. Most respiratory infections in this age group are caused by viruses and lead to the pathophysiologic changes described. Yet, just like adults, infants can also get bacterial infections. In those cases, the radiographic appearance mimics that seen on a chest radiograph in an adult with bacterial pneumonia. Air bronchograms, consolidation, and some volume loss in the consolidated segment usually do not confuse radiology residents and general radiologists, even when seen in infants. Pneumonia looks the same in adults, and, when bacterial, it can look that way in infants too. Infants with bacterial pneumonia can have pleural effusions and adenopathy. Children are not immune to the types of pulmonary infections that older children and adults contract. Nevertheless, viruses are the most common cause of respiratory infections in infants (10).
Bacterial pneumonia in infancy can sometimes produce an unusual and disconcerting appearance because of the anatomic features mentioned earlier. The infant does not have a well-developed system of collateral ventilation—there are fewer pores of Kohn and channels of Lambert. Therefore, exudate that accumulates in the alveoli does not spread to adjacent alveoli as easily as in the adult. The limits of the inflammatory process are difficult to define in adults unless they border on a pleural surface. In infants and younger children, however, the exudate tends to be trapped in the alveoli, unable to spread through the pores of Kohn. Sometimes, because of the lack of collateral air drift openings, the exudate takes the appearance of a spherical consolidation—a "round pneumonia". The inflammatory cells are confined under a mild degree of pressure, and these infants often have a high fever; 104°F or 105°F (40°C or 41°C) is typical. The radiographic appearance can be alarming because a round pneumonia can look like a neoplasm (13). Several children have been referred to us with the suspicion of a primary or metastatic neoplasm, but they really had pneumonia with an unusual spherical appearance (Fig 9).
Other disease processes can produce air trapping and hyperaeration of the lungs. Increased fluid in the interstitial spaces can compress the small airways and cause an increase in small-airways resistance. Enlarged heart chambers and pulmonary vessels can also compress airways. Therefore, severe cardiac disease and increased fluid load can lead to radiographs that show air trapping. Older infants can and do aspirate foreign bodies into their airways; this may cause focal air trapping or atelectasis. Reactive airways disease, which does occur in infants, can cause air trapping at any age. Certain chronic lung diseases such as cystic fibrosis and bronchopulmonary dysplasia demonstrate air trapping on radiographs. Tachypnea from a variety of other causes, such as acidosis, fever, and sometimes even fear, will produce the appearance of air trapping secondary to the mechanism of rapid breathing (4).
Nevertheless, most infants who acutely develop respiratory distress have a viral illness. The appearance of the radiograph reflects the pathologic process occurring in the respiratory system. An understanding of the appearance of the infant chest radiograph requires an understanding of the underlying pathologic process. Hyperaeration of the lungs is often the earliest, and sometimes the only, radiographic sign that the infant has a viral infection involving the lower airways.
|
ESSENTIALS
|
|---|
- Most respiratory infections are caused by viruses in infants and children younger than 24 months and lead to the pathophysiologic changes of air trapping.
- Hyperinflation may be the only radiologic clue to illness.
- Bacterial pneumonia can have the same appearance in adults and infants.
- The radiographic appearance reflects the pathologic process occurring in the respiratory system.
|
|
References
|
|---|
- Coblentz C, Babcook C, Alton D, Riley B, Norman G. Observer variation in detecting the radiologic features associated with bronchiolitis. Invest Radiol 1991; 26:115–118.[Medline]
- Reid L. Lung growth in health and disease. Br J Dis Chest 1984; 78:113–134.[Medline]
- Hislop A, Reid L. Lung development in relation to gas exchange capacity. Bull Physiopathol Respir (Nancy) 1973; 9:1317–1343.[Medline]
- O'Brodovich H, Haddad G. Functional basis of respiratory pathology and disease. In: Chernick V, Boat T, eds. Kendig's disorders of the respiratory tract in children. 6th ed. Philadelphia, Pa: Saunders, 1998; 27–73.
- Dunnill MS. Postnatal growth of the lung. Thorax 1962; 17:329–333.
- Mortola J. Comparative aspects of neonatal respiratory mechanisms. In: Haddad G, Abman S, Chernick V, eds. Chernick-Mellin: basic mechanisms of pediatric respiratory disease. 2nd ed. Hamilton, Ontario, Canada: Decker, 2002; 171–178.
- Lawson E. Respiratory control after birth. In: Chernick V, Mellin R, eds. Basic mechanisms of pediatric respiratory disease: cellular and integrative. Philadelphia, Pa: Decker, 1991; 288–302.
- Griscom NT, Wohl M, Kirkpatrick J. Lower respiratory infections: how infants differ from adults. Radiol Clin North Am 1978; 16:367–387.[Medline]
- Denny F. Acute lower respiratory tract infection: general considerations. In: Taussing L, Landau L, eds. Pediatric respiratory medicine. St Louis, Mo: Mosby, 1999.
- Dubois D, Ray C. Viral infections of the lower respiratory tract. In: Taussing L, Landau L, eds. Pediatric respiratory medicine. St Louis, Mo: Mosby, 1999.
- Gern J. Virus-induced inflammation in airways. In: Haddad G, Abman S, Chernick V, eds. Chernick-Mellins: basic mechanisms of pediatric respiratory disease. 2nd ed. Hamilton, Ontario, Canada: Decker, 2002; 518–527.
- Swischuk LE, Hayden CK Jr. Viral vs. bacterial pneumonia infection in children (is roentgenographic differentiation possible?). Pediatr Radiol 1986; 16:278–284.[CrossRef][Medline]
- Rose R, Ward B. Spherical pneumonias in children simulating pulmonary and mediastinal masses. Radiology 1973; 106:179–182.[Medline]
TOP
ABSTRACT
INTRODUCTION
