This 4-month-old female infant was transferred from an outside hospital with a chief complaint of respiratory distress. Four days before her admission to Children's Memorial Hospital, she developed rhinitis and cough and was admitted to an outside hospital with a diagnosis of bronchiolitis. She received albuterol by nebulizer and intravenous methylprednisolone. Because of increasing tachypnea and supplemental oxygen requirements, she underwent emergent endotracheal intubation at the outside hospital and was transferred.
Her medical history was remarkable for a previous episode of bronchiolitis at age 2 months. At that time, results of a chest x-ray were reportedly normal. Her mother stated that the baby "always has cough and congestion." The family had lived in a homeless shelter for the last month, during the winter. The family history was unremarkable.
The infant was the 6-lb, 7-oz product of a full-term pregnancy to a 27-year-old G4P4 female. It was a spontaneous vaginal delivery with no perinatal problems.
On exam, she was intubated and sedated. She was afebrile, her pulse was 150, and her respiratory rate was synchronous with the ventilator. Her weight, length, and head circumference were all in the 25th percentile. HEENT examination was unremarkable. There were coarse inspiratory breath sounds bilaterally with occasional wheezing and mild bilateral intercostal retractions. Sl was normal. S2 was split physiologically. There were no murmurs, rubs, or gallops. Her abdomen was soft and not tender. The liver was palpable 4 cm below the right costal margin. There was no splenomegaly. The remainder of the examination was normal.
Respiratory syncytial virus (RSV) antigen was detected in nasal secretions. The initial chest x-ray revealed bilateral peribronchial thickening and a patchy left lower lobe infiltrate.
Robert Listernick, MD, moderator: Any questions?
Don Zimmerman, MD, pediatric endocrinologist: Did the child's respiratory distress improve following intubation, suggesting the possibility of an upper airway lesion?
Denise Goodman, MD, pediatric critical care physician: Her respiratory distress remained the same. There was nothing in the history to suggest that she had been stridulous prior to the intubation. There was nothing about her history or physical examination that distinguished her from the dozens of children that are admitted to the intensive care unit each winter with RSV infection. Our initial impression of her apparent hepatomegaly was that it was the result of lung hyperinflation.
Dr. Listernick: When this child arrived here, the intensive care physicians "threw the book" at her, treating her with nebulized bronchodilators and intravenous corticosteroids. Is this appropriate, evidence-based therapy?
Dr. Goodman: Most of the studies in patients with uncomplicated bronchiolitis suggest that treatment with either corticosteroids or bronchodilators is ineffective in altering the course of the disease. However, from the vantage point of a critical care physician, intubation of infants with bronchiolitis should be avoided if possible. Intubation itself predisposes to the development of ventilator-associated pneumonia. I believe it's valid to try these therapies in critically ill children who are on the verge of intubation and mechanical ventilation. Obviously, this is a more pragmatic and less "evidence-based" approach, but if it obviates the need for ventilation in some children, I feel that it's justified. Remember that this represents a very small fraction of the total number of children annually who have RSV infections; I would agree that those therapies are not indicated in most of these less ill patients. If we do not observe a response to the bronchodilator in the intensive care unit in an individual child, we discontinue its use.
Unfortunately, there's no easily measurable test by which we can document an objective response other than the traditional parameters that all physicians use - respiratory rate, auscultation, and measurement of oxygen saturation by pulse oximetry.
Oren Lakser, MD, pediatric pulmonologist: I agree with this critical care approach. In addition, it should be pointed out that most of the studies of these therapies in bronchiolitis show no statistically significant benefit. This doesn't mean that either bronchodilators or corticosteroids wouldn't work in an individual patient. Certainly, if the patient has a family history of asthma or atopy, this episode of wheezing triggered by RSV may be her first episode of asthma (ie, wheezing responsive to bronchodilators).
Robert Tanz, MD, general academic pediatrician: Don't we have to be more empiric than that? Essentially you're advocating ignoring the results of double-blind, randomized, controlled studies simply because it might work.
Dr. Lakser: I don't believe that you can rely on empiricism alone when confronted with an individual patient. I'm not advocating ignoring the results of previous, wellperformed studies. However, in looking at the data closely, it appears that approximately 10% of children with bronchiolitis respond to bronchodilators. Perhaps, as has been said, these are the children who are having their first episode of asthma. I'm simply stating that, at times, the therapy needs to be individualized.
Dr. Tanz: I'd counter by pointing out that the studies that used placebos in a blinded fashion demonstrated a positive effect in 30% of the patients receiving the placebo. Perhaps it's the cool mist that helps some patients, although the best data show that even mist has no demonstrable benefit.
Dr. Goodman: Every best practice guideline qualifies its evidence-based approach by saying that therapy should be individualized for a specific patient. The goal is to prevent a shotgun approach to clinical care in which every child with bronchiolitis gets every therapy regardless of the data. The idea is for physicians to use their best clinical judgment guided by the data.
Dr. Listernick: Are there any potential adverse effects associated with the use of these therapies?
Dr. Goodman: There are some adverse effects that generally cause minimal or no problems. Albuterol will induce tachycardia that is generally well-tolerated. Steroids may cause glucose intolerance in some children; however, these patients would only be receiving short courses. The rare patient will develop steroid-induced gastritis and gastrointestinal hemorrhage.
Doug Nordli, MD, pediatric neurologist: How does this child's residence in a homeless shelter affect the differential diagnosis and the therapeutic approach?
Dr. Tanz: During the winter months in crowded conditions, the two most common causes of respiratory illness in infants are RSV and influenza. One should also consider tuberculosis, particularly if the illness was prolonged, had atypical features, or showed other risk factors associated with tuberculosis.
Dr. Listernick: When is bronchiolitis not bronchiolitis? In other words, which features of respiratory infections in infants with wheezing should lead physicians to consider alternate diagnoses?
Dr. Goodman: From the critical care standpoint, normal infants who develop bronchiolitis may wind up ventilated for several weeks. The typical story is an infant who has a mild respiratory illness for several days that becomes progressively worse, eventually requiring mechanical ventilation. After several days of ventilation, the infant slowly improves and is extubated after 1 or 2 weeks. Infants who deviate from this clinical course should be investigated for alternate diagnoses. Examples would be infants who require mechanical ventilation after 7 to 10 days of illness or infants who can't begin to be weaned off the ventilator after several days.
Dr. Lakser: I agree. In addition, I would suspect alternate diagnoses in infants who have recurrent episodes of wheezing, failure to thrive, or chest x-rays that deviate from the typical pattern one sees in RSV bronchiolitis (peribronchial thickening with focal alveolar infiltrates).
Dr. Listernick: What are some of those alternate diagnoses?
Dr. Lakser: It's a long laundry list of conditions. In an infant who has had recurrent or prolonged respiratory difficulties, the differential diagnosis includes gastroesophageal reflux, aspiration, tracheoesophageal fistula, cystic fibrosis, congenital pulmonary lesions such as a pulmonary sequestration, and a host of other conditions. These can be sorted out based on history, physical examination, and radiographic findings.
Dr. Goodman: To give you an idea of the scope of the problem, our hospital usually has 300 to 400 infants with bronchiolitis admitted each winter season. Of those, approximately 10% to 20% end up in the intensive care unit, and 70% of those infants require mechanical ventilation. In this large population of children, we usually establish one or two of these alternate diagnoses each year.
Dr. Listernick: This child was on a ventilator for several days during which time the chest x-ray findings evolved.
Cynthia Rigsby, MD, pediatric radiologist: The repeat chest x-ray performed several days into the admission revealed an abnormal density in the left lower lobe. Initially, it had been called an infiltrate. However, on this film we can appreciate distortion of the lung parenchyma. If you look closely, there are hyperlucent areas within this density. This appears to be either a congenital cystic adenomatoid malformation (CCAM) or a pulmonary sequestration. Computerized tomography (CT) imaging was recommended.
Juda Jona, MD, pediatric surgeon: Had she had any prenatal ultrasounds? Congenital abnormalities of the lung, such as diaphragmatic hernia, CCAM, and bronchogenic cysts, may be detectable on such studies.
Dr. Listernick: There was no prenatal screening performed. If this is a congenital anomaly, why wasn't it identified on the initial x-ray?
Dr. Rigsby: Most probably the hyperlucent areas (areas of lung parenchyma filled with air) were filled with fluid, preventing us from distinguishing this congenital lesion from a simple infiltrate. Sequestrations may be missed on x-rays if they are small and obscured by the diaphragm.
The CT scan identified a left lower lobe mass composed of dense, underaerated sections of lung as well as areas of abnormally large aerated cystic spaces. In addition, there is a large vessel coming off the aorta going into this mass; it's approximately 80% the diameter of the aorta. Finally, there is a large left inferior pulmonary vein that drains its blood supply. These findings identify this lesion as consisting of two elements, both a CCAM and an intralobar pulmonary sequestration.
Dr. Listernick: Does the CT scan need to be ordered in a special way in order to identify a pulmonary sequestration?
Dr. Rigsby: The radiologist would always like to know what the clinicians suspect so that the proper techniques can be used. A CT scan of the chest in young children would almost always involve the administration of intravenous contrast, which is crucial for identification of the abnormal systemic vessel feeding the sequestration. This might not be seen on a routine CT scan. In those circumstances, the ideal study to order would be a CT angiogram.
Dr. Listernick: Let's define our terms.
Dr. Lakser: CCAMs arise as a result of a developmental anomaly of the terminal bronchioles. Usually a single lobe or segment of the lung has multiple cysts that cause compression of surrounding normal lung tissue. If large, they may present in the newborn period with severe respiratory distress. Smaller ones may present as incidental findings on chest x-rays or, if infected, as pneumonia. The diagnosis is often made by prenatal ultrasonography.
A pulmonary sequestration is a mass of nonfunctioning lung tissue, often cystic, whose blood supply comes from the systemic circulation. Its venous drainage goes to the pulmonary circulation, effectively creating a left-to-right shunt. Textbooks generally describe two types of sequestrations, intralobar and extralobar. Extralobar sequestrations almost always involve the left lung; they have their own pleural covering. These may occur with associated anomalies such as diaphragmatic hernia or intestinal duplications. The nature of intralobar sequestrations, which don't have their own pleura, is somewhat controversial. These lesions are often cystic, leading some to believe that they are variant manifestations of CCAM. Sequestrations may be discovered as either an incidental finding on chest x-ray or during the course of treatment of pneumonia.
Marieta Reynolds, MD, pediatric surgeon: I'd like to make two points. First, sequestered lung may become aerated as a result of communication of normal lung tissue with the sequestered segment through the alveolar pores. Thus, sequestrations may be variably translucent or radioopaque, leading to confusion in diagnosis. Second, there is considerable overlap between CCAM and sequestration. The diagnosis of sequestration is made based on the presence of systemic arterial supply to the involved lobe rather than the histology. The more appropriate modem descriptive term is a bronchopulmonary malformation.
Dr. Listernick: What is the significance of these pulmonary malformations?
Dr. Reynolds: Both of these lesions may become infected repeatedly. The arterial blood supply of a sequestration is histologically abnormal. These vessels have an accelerated rate of atherosclerosis that may lead to aneurysms within the lung parenchyma. Finally, there is an increased rate of malignancy in CCAMs, most notably both rhabdomyosarcoma and bronchoalveolar carcinoma. For all these reasons, these lesions should be removed once identified.
Dr Listernick: Do all these lesions need to be removed?
Marybeth Madonna, MD, pediatric surgeon: All postnatally diagnosed lesions should be removed. However, 20% of those discovered prenatally regress. These should be followed conservatively during the first few months of life with repeated CT scans.
Dr. Listernick: Are there any important issues regarding the surgery itself?
Dr. Jona: The blood supply to the lesion should be adequately demonstrated. At a minimum, echocardiography should be used to identify all the pulmonary veins. If they can't be seen, angiography may be necessary. CCAMs tend to be lobar in distribution, necessitating a lobectomy. If only the large cystic components of a lung segment are removed, reinfection and cyst reformation ultimately may require lobectomy. It may be of benefit to perform the surgery during infancy, allowing the lung to regenerate a new segment. If the infant is asymptomatic, I generally wait until age 1 to perform the operation.
Dr. Listernick: We were concerned initially about the hepatomegaly and whether she had a component of congestive heart failure associated with shunting. An echocardiogram demonstrated normal cardiac function. However, not all the pulmonary veins could be visualized. Therefore, she underwent cardiac catheterization, which demonstrated normal anatomy and pulmonary vasculature other than the sequestration. She subsequently underwent successful left lower lobe lobectomy and is doing well.
Thank you everyone.