XIV - Congenital, Structural, and Inflammatory Diseases of the Lung

Editors: Shields, Thomas W.; LoCicero, Joseph; Ponn, Ronald B.; Rusch, Valerie W.

Title: General Thoracic Surgery, 6th Edition

Copyright 2005 Lippincott Williams & Wilkins

> Table of Contents > Volume I - The Lung, Pleura, Diaphragm, and Chest Wall > Section XIV - Congenital, Structural, and Inflammatory Diseases of the Lung > Chapter 94 - Diffuse Lung Disease

function show_scrollbar() {}

Chapter 94

Diffuse Lung Disease

Herbert Knight

Ronald B. Ponn

Diffuse parenchymal lung diseases (DPLDs) are characterized by generalized or extensive multifocal involvement of the pulmonary parenchyma. Since Hamman and Rich (1935) described rapidly progressive pulmonary fibrosis, more than 100 disorders have been classified as diffuse diseases. The inclusion of infectious processes raises the number further. Despite variation in etiology, acuity, and outcome, these entities are grouped together because of certain common clinical, histologic, and radiographic features. With the exception of lung transplantation in a tiny fraction of cases and the treatment of complications such as pneumothorax or pleural effusion, the role of the thoracic surgeon in DPLD is mainly diagnostic. Although advances in clinicopathologic correlation have lessened the requirement for surgical biopsy, in many instances, the need for definitive tissue sampling remains, especially as the limitations of pathologic diagnosis from small specimens are increasingly appreciated. When biopsy is considered, the surgeon should not function merely as a technician, but can play an important role in determining the timing, method, and wisdom of invasive diagnostic efforts.

Diffuse lung diseases are often classified clinically as acute or chronic and pathologically as granulomatous or nongranulomatous, but the overlap is considerable. A simpler approach is to divide the disorders into those of known etiology and the larger group in which the cause is unknown (Table 94-1).

PATHOGENESIS: VARIATIONS IN RESPONSE TO INJURY

The microscopic anatomy and cell populations of the lung are discussed in Chapters 3 and 4. DPLD involves the gas-exchanging structures, including bronchioles, alveolar ducts, alveolar sacs, interstitium, and small vessels. Clinical studies and experimental models induced by radiation therapy, inhaled irritants, immune complexes, and toxins such as bleomycin show similarities in the lung's response to injury. Crouch (1990) summarized early concepts of pathogenesis, and Rochester and Elias (1993) emphasized the role of cytokines. More recent models focus attention on varying patterns of repair, as discussed Selman and associates (2001).

Injury to the lung's microstructure produces an accumulation of inflammatory and immune cells in the interstitium and air spaces (Fig. 94-1). Necrosis of type I epithelial cells results in denuded areas of alveolar wall. Damage to capillary endothelial cells may also be the inciting event. Fluid and protein leak into the alveolar space and form fibrin-rich hyaline membranes. After this exudative phase, processes begin that lead either to repair or to further injury and ultimately to fibrosis.

Granulocytes appear early. Activated resident macrophages elaborate cytokines that attract neutrophils, causing them to adhere to capillary endothelium and enter the interstitium and alveoli. Some models suggest that tumor necrosis factor (TNF) is important in the early phase of alveolar injury. Complement activation and platelet products also attract neutrophils. The recruited cells release toxic oxygen products and proteases that break down matrix proteins. Activated macrophages also produce oxidants and proteolytic enzymes.

A mononuclear cell response follows. Macrophages continue to play an important role through chemotaxis for blood monocytes that mature into macrophages. Mediators from activated macrophages also attract and activate T lymphocytes, some of which stimulate eosinophils and mast cells and cause B-cell differentiation into antibody-producing plasma cells. The resultant immune complexes may amplify the inflammatory reaction. In other instances, the balance favors a subpopulation of T lymphocytes that promotes a cell-mediated response. These cells generate cytokines that result in granuloma formation by recruitment of blood monocytes, activation of macrophages, and inhibition of macrophage migration. T cells also enhance injury by stimulation of cytotoxic and killer lymphocytes.

Fibrosis results from an increased number of fibroblasts, their passage into air spaces through damaged alveolar

P.1343


walls, and enhanced extracellular matrix synthesis (Fig. 94-2). Cytokine modulation of collagen elaboration is increasingly appreciated, particularly the role of TNF- and transforming growth factor- 1 (TGF- 1). Macrophages and lymphocytes stimulate fibroblast migration, proliferation, and synthetic activity. Alveolar epithelial cells may also elaborate profibrotic cytokines or fail to function adequately in the process of reepithelialization of the alveolar wall. Peptides produced by matrix molecule degradation are chemotactic for fibroblasts. Fibronectin, present in the alveolar exudate and produced by activated macrophages, attracts fibroblasts and binds them to fibrin. Organization of the exudate produces accretion of fibrous tissue onto the alveolar wall. Type II pneumocytes proliferate, differentiate into type I cells, and epithelialize the surface of the scar. The result is a widening of the normally thin air blood interface.

Table 94-1. Partial List of Noninfectious Diffuse Lung Diseases

Known Cause Unknown Cause
Inhalation
   Inorganic dusts (pneumoconiosis): asbestos, aluminum, barium, beryllium, cadmium, cerium, hafnium, iron, kaolin, mica, niobium, silica, talc, tin, titanium
   Organic dusts (hypersensitivity pneumonitis, extrinsic allergic alveolitis): air-conditioner lung, bagassosis, bird fancier's lung, cheese worker's lung, feather plucker's lung, humidifier lung, malt worker's lung, maple bark stripper's lung, mummy unwrapper's lung, mushroom worker's lung, paprika splitter's lung, sauna lung, sequoiosis, suberosis
   Fumes and gases: chlorine, mercury, metals, paraquat, sulfur dioxide
Neoplasm
   Lymphangitic carcinoma, bronchoalveolar carcinoma, Kaposi's sarcoma, lymphoma
Drugs
   Antibiotics: erythromycin, INH, nitrofurantoin, PAS, penicillin, sulfonamides, tetracycline
   Chemotherapeutic agents: azathioprine, bleomycin, busulfan, chloroambucil, cyclophosphamide, gemcitabine, melphalan, mercaptopurine, methotrexate, mitomycin, nitrosoureas, procarbazine
   Cardiovascular medications: amiodarone, beta-blockers, hydralazine, procainamide
Other drugs: allopurinol, carbamazepine, chlorpromazine, chlorpropamide, cromolyn, dantrolene, gold, mecamylamine, methylphenidate, penicillamine, pentolinium, tolbutamide, tricyclics
Radiation
   Pneumonitis (early), fibrosis (late)
Chronic aspiration
Idiopathic pulmonary fibrosis (cryptogenic fibrosing alveolitis)
Nonspecific interstitial pneumoniaa
   Bronchiolitis obliterans organizing pneumoniaa [cryptogenic organizing pneumonia (COP)]
Collagen vascular disease
   Ankylosing spondylitis
   Dermatomyositis, polymyositis
   Lupus erythematosus
   Mixed connective tissue disease
   Rheumatoid arthritis
   Scleroderma
   Sj gren's syndrome
Sarcoidosis
Eosinophilic lung diseasesa
   Acute eosinophilic pneumonia
   Chronic eosinophilic pneumonia
   Bronchocentric granulomatosis
Churg-Strauss syndrome (allergic angiitis and granulomatosis)b
   Idiopathic hypereosinophilic syndrome
   Simple pulmonary eosinophilia (L effler's syndrome)
Lymphoid infiltrative disease
   Lymphocytic interstitial pneumonia
Pulmonary vasculitis
   Wegener's granulomatosis
   Lymphomatoid granulomatosis (neoplasm)
Histiocytosis X (eosinophilic granuloma)
Lymphangioleiomyomatosis
Pulmonary alveolar proteinosis
Diffuse pulmonary hemorrhage
   Goodpasture's syndrome
   Idiopathic pulmonary hemorrhage
Amyloidosis
Congenital diseases
   Gaucher's disease
   Hermansky-Pudlak syndrome
   Niemann-Pick disease
   Neurofibromatosis
   Tuberous sclerosis
a Some cases are associated with known cause.
b May also be classified as a vasculitis.

The cells and mediators leading to fibrosis are part of the lung's normal repair mechanisms. Many injuries do not cause histologic or physiologic sequelae. The alveolar exudate is cleared, and the denuded alveolar walls undergo reepithelialization without fibrosis. The outcome in any individual case is determined by the severity of the initial damage, the chronicity or repetition of the injury, the specific toxin or disease, dysregulation and autonomy of the inflammatory reaction, and individual susceptibility, including the age of the patient (better reparative capacity in younger people). However, the factors that determine uncomplicated resolution of injury versus progressive fibrosis are incompletely understood. Reconsideration of mechanisms

P.1344


P.1345


of fibrosis has led to the view that disordered healing in response to injury, and not persisting inflammation, may be the pivotal fibrogenic disturbance, as noted by Selman and associates (2001).

Table 94-2. Histologic Patterns and Clinical Correlates in Diffuse Lung Disease

Pattern Distribution Histology Presentation Idiopathic Syndrome Selected Association
Diffuse alveolar damage (DAD) Diffuse Uniform Acute Acute pneumonia (Hamman-Rich) Acute interstitial syndrome (ARDS)
Drugs
Acute radiation therapy
Lupus, acute
Usual interstitial pneumonia (UIP) (mural fibrosing alveolitis) Patchy Nonuniform Chronic Idiopathic pulmonary fibrosis (IPF) (cryptogenic fibrosing alveolitis) Collagen vascular disease (CVD)
Inhalation
Infection
Drugs
Other
Desquamative interstitial pneumonia (DIP) (desquamative fibrosing alveolitis) Diffuse Uniform Subacute/chronic DIP or early IPF Same as UIP
Nonspecific interstitial pneumonia (NSIP) Patchy Uniform Subacute/chronic NSIP Immunocompromised CVD, inhalation
Bronchiolitis obliterans organizing pneumonia (BOOP) [cryptogenic organizing pneumonia (COP)] Patchy Uniform Subacute BOOP (COP) Same as UIP
Transplant
Postinfectious
Radiation
Lymphocytic interstitial pneumonia (LIP) Diffuse Nonuniform Variable LIP Acquired immunodeficiency syndrome (AIDS)
Sj gren's
Other
Giant cell interstitial pneumonia (GIP) Patchy Uniform Variable None Hard metal inhalation
Honeycombing Variable Nonuniform Variable NA Many diffuse parenchymal lung diseases (DPLDs)
Honeycomb lung Diffuse Uniform Chronic NA IPF, lymphangioleio-myomatosis (LAM)
Sarcoidosis
Histiocytosis X
Asbestosis
Scleroderma
NA, not applicable.

Fig. 94-1. Simplified schema of the cellular and humoral mediators of lung inflammation and injury. Only selected mediators are shown, and normal inhibitory pathways are not demonstrated. IL, interleukin; C5a, complement; PDGF, platelet-derived growth factor; TNF, tumor necrosis factor; IFN- , interferon- .

Fig. 94-2. Factors leading to fibroblast stimulation, fibrosis, and lung distortion. IL, interleukin; IGF, insulinlike growth factor; TNF, tumor necrosis factor; TGF- , transforming growth factor- ; AMDGF, alveolar macrophage-derived growth factor; IFN- , interferon- .

PATHOLOGIC FEATURES

Although some diffuse diseases exhibit unique histologic features, and in some cases microorganisms or inhaled irritants are identified, the findings are often nonspecific and reflect stereotyped patterns of response to injury. Pathology reports often describe diffuse alveolar damage, usual, desquamative or nonspecific interstitial pneumonia, bronchiolitis obliterans, or honeycombing. An expanding array of pathologic labels, used by some to connote specific diseases and by others as purely histologic descriptors, may make clinical correlation difficult. Problems in communication are compounded by the use of different terms to describe identical lesions, especially by authors on different sides of the Atlantic Ocean (Table 94-2; Fig. 94-3). A recent international consensus document provides updated standards for the classification of the idiopathic DPLD and is detailed in a communication by the American Thoracic Society and the European Respiratory Society (2002).

Diffuse alveolar damage (DAD) is typical of severe acute injury. Katzenstein and associates (1976) introduced the term to describe the features of adult respiratory distress syndrome (ARDS). Early cases show interstitial and alveolar edema, epithelial necrosis, and sloughing leading to denuded alveolar walls, acute inflammatory cells, and hyaline membranes. Later, there is organization and repair (proliferative phase), with proliferation of fibroblasts in the interstitium and air spaces and of type II pneumocytes along the damaged alveolar walls, along with partial or complete resorption of the hyaline membranes seen earlier. The proliferative phase is thought to occur within 1 to 2 weeks after the initial injury. DAD is generally found diffusely throughout

P.1346


P.1347


the parenchyma and is histologically uniform. These hallmarks suggest a discrete generalized injury. In addition to ARDS, DAD may be seen in other acute syndromes, including acute interstitial pneumonia (Hamman-Rich syndrome), acute radiation pneumonitis, drug reactions, inhalation injury, hypersensitivity pneumonitis, and acute lupus pneumonitis. DAD in immunocompromised patients is often due to infection. The term acute interstitial pneumonitis has been suggested by Katzenstein and associates (1986) to describe idiopathic cases, especially to distinguish this lesion from the more common chronic interstitial pneumonias.

Fig. 94-3. Common histologic patterns in diffuse lung disease. A. Diffuse alveolar damage. Alveolar cells are replaced by hyaline membranes. Inflammatory cell infiltrates are sparse at this early stage of injury. B. Usual interstitial pneumonia. Features of patchy chronic inflammation include fibrosis and interstitial cellular infiltrates, honeycombing, and reactive pneumocytes. C. Desquamative interstitial pneumonia. Alveoli with minimal architectural distortion are filled with macrophages. Pneumocyte hyperplasia is apparent. D. Bronchiolitis obliterans organizing pneumonia. An inflamed bronchiole is filled with a plug of granulation tissue. Chronic active inflammation persists around the airway. E. Honeycomb lung. Normal parenchyma is replaced by fibrosis and epithelium-lined cystic changes. Cysts are filled with mucus and inflammatory cells.

Liebow (1968) described the classic interstitial pneumonias. Usual interstitial pneumonia (UIP) is encountered most frequently. Like DAD, UIP represents a nonspecific response. The clinical syndrome, however, typically evolves over a longer period of time. Hyaline membranes, edema, and alveolar exudates are rarely seen. The distribution is patchy, and the histology is nonuniform, varying from normal alveoli to inflammation to fibrosis. This pattern suggests repetitive or low-grade chronic injury. Areas of UIP may be seen in idiopathic pulmonary fibrosis (IPF), collagen vascular disease (CVD), chronic hypersensitivity pneumonitis, some of the pneumoconioses, sarcoidosis, drug-induced DPLD, granulomatous infections, late radiation pneumonitis, healed infections, organized DAD, chronic aspiration, and histiocytosis X, and even in cases of chronic pulmonary edema due to passive congestion. In desquamative interstitial pneumonia (DIP), the involvement is diffuse and the histology is homogeneous. It was initially believed that the highly cellular histology reflected desquamated alveolar epithelial cells, but it is now known that filling of the alveoli by macrophages is the defining feature. Interstitial monocytic infiltration is also seen, but fibrosis is minimal. A DIP pattern may be found in idiopathic DIP, histiocytosis X, drug reactions (e.g., amiodarone toxicity), alveolar hemorrhage, pneumoconioses, eosinophilic pneumonias, obstructive pneumonia (in which case the lesion is not diffuse), lipid pneumonia and lipid storage diseases, respiratory bronchiolitis, and infections in the immunosuppressed patient. In Europe, UIP and DIP are termed mural and desquamative fibrosing alveolitis, respectively.

Carrington and associates (1978) view UIP and DIP as two diseases because of different pathologic findings and because DIP is more amenable to treatment. Other authorities, however, agree with Scadding and Hinson (1967) and Patchefsky (1973) and Tubbs (1977) and their associates that DIP is the early cellular phase and UIP the later mixed or fibrotic phase of the same reaction. As noted, there is much overlap in clinical processes in which the two patterns may be identified. In addition, areas of DIP can be found in cases of predominant UIP histology, and DIP has been shown to progress to UIP. Recent observations of the frequent association with cigarette smoking and pathologic similarities has linked the condition with the spectrum of respiratory bronchiolitis, as discussed by Yousem and colleagues (1989).

Katzenenstein and Fiorelli (1994) detailed the features of an additional variant of the interstitial pneumonias of unknown etiology [nonspecific interstitial pneumonia (NSIP)]. Although the characteristic lesion contains varying proportions of interstitial inflammation and fibrosis, NSIP differs from UIP in that all areas appear to be temporally uniform. In addition to an idiopathic form, NSIP also occurs in AIDS, as described by Suffredini and associates (1987), in other immunocompromised patients and in cases of CVD, inhalation of organic antigens, and resolving acute lung injury. In a report of 12 cases, Cottin and associates (1998) found that 6 were idiopathic, 3 were associated with CVD, 2 were associated with organic dust exposure, and 1 followed an episode of ARDS. The response to treatment is better than for UIP, but appears to be more variable than in bronchiolitis obliterans organizing pneumonia (BOOP).

In BOOP, organizing connective tissue partially or totally fills alveoli, respiratory bronchioles, and alveolar ducts. Bronchiolar intraluminal polypoid plugs may or may not be seen. The label cryptogenic organizing pneumonia (COP) has been suggested as preferable to BOOP because it conveys the essential features of the syndrome and avoids confusion with other forms of bronchiolitis. Mononuclear cell infiltrates are seen in the airway walls, and the alveoli often contain lipid-filled macrophages. Fibrosis is absent or minimal. The anatomic distribution is patchy but histologically uniform. Background lung architecture is relatively preserved. Similar to other nonspecific reactions, COP can be idiopathic or associated with the response to other forms of lung injury, such as organizing infection, drug reactions, inhalation injury, CVD, hypersensitivity pneumonitis, and chronic eosinophilic pneumonia.

Lymphocytic interstitial pneumonia (LIP) consists of a dense interstitial infiltrate of lymphocytes, plasma cells, and macrophages. The alveolar septa are typically extensively infiltrated. Granulomas, lymphoid follicles, and amyloid deposits are often present. Although an idiopathic form exists, LIP has a strong association with immunologic diseases, acquired immunodeficiency syndrome (AIDS), and lymphoid malignancies. Giant cell interstitial pneumonia (GIP) is a rare pattern in which bizarre multinucleated cells fill the air spaces. Ohori and associates (1989) reported that most cases result from inhalation of hard metal particles.

Progressive severe fibrosis results in cyst formation and loss of normal architecture. When this pattern coexists with normal and inflamed areas, as in UIP, it is called honeycombing. An etiologic diagnosis can often be made if the histology of the less fibrotic areas is characteristic. Honeycomb lung denotes diffuse distortion and represents the potential end-stage of any fibrogenic process, but in practice is seen mainly in IPF, histiocytosis X, scleroderma, rheumatoid lung, asbestosis, lymphangioleiomyomatosis (LAM), and sarcoidosis.

In some DPLDs, necrotizing or nonnecrotizing granulomas are a universal or occasional feature. Processes associated

P.1348


with granuloma formation include sarcoidosis, granulomatous infections, talcosis, berylliosis, Wegener's granulomatosis, necrotizing sarcoid granulomatosis, bronchocentric granulomatosis, certain drug reactions, and hypersensitivity reactions.

RADIOGRAPHIC FEATURES

Despite advances in computed tomography (CT), the chest radiograph remains the basic imaging tool for the initial evaluation and subsequent surveillance of DPLD. Felson (1979) and Fraser and associates (1988) discussed the radiographic features of DPLD. Because the interstitium and air spaces are often affected simultaneously by histology, radiographic division into interstitial and air space or alveolar (acinar) patterns has been criticized as inaccurate. This system, however, is in common use and serves as a reference point. Predominant interstitial disease produces four patterns: reticular, nodular, reticulonodular, and linear. When filling of the air spaces predominates, an acinar pattern is seen.

Reticular infiltrates appear as a network of curvilinear opacities and are described as fine, medium, or coarse, depending on the size of the mesh. These gradations generally correlate with progression from early alveolitis to advanced fibrosis. Transition over time from fine to medium to coarse patterns has been documented by Wholey and associates (1958) for IPF and by Locke (1963) for rheumatoid lung. Coarse reticulation (thick-walled cystic spaces at least 5 mm in diameter) indicates severe fibrosis and, like its pathologic counterpart, is termed honeycombing. A nodular pattern consists of well-circumscribed, small, rounded opacities of varying diameter. More common is a combined reticulonodular pattern that can result from superimposition of nodules and reticulation or from end-on orientation of reticular opacities in the absence of histologic nodules. Thickening of the interlobular septa and perivascular sheaths produces a linear pattern typified by Kerley B lines. Early acinar opacities may simulate nodularity, but the borders are indistinct. Coalescence produces air space consolidation. Patent airways surrounded by opacified acini appear as air bronchograms. Mixed interstitial and air space patterns are common.

Ground-glass opacification results from changes below the resolution of the imaging technique, including fine reticulation, micronodularity, and minimal air space filling. For standard radiographs, this description refers to a generalized, sometimes granular haziness. On CT, ground-glass appearance is often patchy and is defined as an increase in lung density that does not obscure bronchovascular structures.

Table 94-3 lists other features of the chest radiograph, including anatomic distribution, overall lung volume, and extrapulmonary findings, which may aid in diagnosis.

The limitations of conventional radiography in DPLD must be kept in mind and include limited sensitivity, an incomplete view of the lungs and mediastinum, variability in interpretation, and a lack of specificity of the aforementioned plain film patterns. As much as 40% of the lung parenchyma is obscured by superimposed structures on chest radiographs. The frequent inadequacy of standard radiography was highlighted by McLoud and associates (1983). They interpreted films from 365 biopsy-confirmed cases of DPLD and recorded the three most likely diagnoses using a standardized system. Despite a correlation between certain diseases and patterns, the correct diagnosis was included in the first two choices only 50% of the time. Epler and associates (1978) found that the chest film was entirely normal in 10% of cases of DPLD.

High-resolution CT (HRCT) is currently a pivotal study in the assessment of DPLD. The benefits of CT include complete visualization of all lung regions; identification of mediastinal and hilar lymphadenopathy, small effusions,

P.1349


and pleural abnormalities; aid in selecting appropriate sites for biopsy; and, most importantly, the detection of characteristic and often subtle abnormalities. The resolution of HRCT reliably extends to the anatomic level of the secondary pulmonary lobule. Colby and Swenson (1996) provided a comprehensive analysis of the HRCT correlates of the histopathologic patterns in DPLD. The value of HRCT was also confirmed by Mathiesen and co-workers (1989), who were able to make a correct first-choice diagnosis in 76% of cases, compared with 57% by chest radiographs alone. Because of characteristic CT patterns, high accuracy was achieved in silicosis (93%), UIP-IPF (89%), lymphangitic carcinoma (85%), histiocytosis X (83%), and sarcoidosis (77%). Strollo (2003) has provided a valuable recent discussion of HRCT in diffuse lung diseases. As with plain radiography, however, a normal HRCT does not eliminate the possibility of DPLD. Orens and associates (1995) found that 12% of dyspneic patients had normal HRCT findings, despite measured impairment of gas exchange and biopsy-proven IPF, although the false-negative cases had less severe functional and histologic findings than those with abnormal HRCT findings. Although HRCT may be nonspecific and sometimes falsely negative, radiographic-clinical correlation is diagnostic in many instances (e.g., lymphangitic carcinoma, IPF, hypersensitivity, among others). CT imaging may also narrow the list of differential diagnoses at presentation. This improved diagnostic accuracy may obviate the need for surgical biopsy in some cases. The HRCT appearance may also be valuable in determining disease activity and response to treatment, especially in IPF. Ground-glass patterns suggest an improved response to treatment and survival.

Table 94-3. Radiographic Features of Diffuse Lung Diseases

Upper lung zone predominance
   Sarcoidosis
   Histiocytosis X
   Silicosis
   Berylliosis
   Chronic hypersensitivity pneumonitis
   Amiodarone toxicity
Increased lung volumes
   Histiocytosis X
   Chronic hypersensitivity pneumonitis
   Lymphangioleiomyomatosis
   Sarcoidosis, stage III
Pleural effusion
   Collagen vascular disease
   Lymphangitic carcinoma, metastatic disease
   Asbestosis
   Lymphangioleiomyomatosis (chylous)
   Radiation fibrosis
Pleural plaques and thickening
   Asbestosis
   Lymphangitic carcinoma/metastatic disease
Pneumothorax
   Lymphangioleiomyomatosis
   Histiocytosis X
   Advanced sarcoidosis
   Pneumocystis carinii pneumonia
Mediastinal and/or hilar lymphadenopathy
   Sarcoidosis
   Lymphoma
   Lymphangitic carcinoma/metastatic disease
   Berylliosis
Esophageal dilation, air fluid level
   Scleroderma
   Aspiration pneumonitis
Soft tissue calcifications
   Dermatomyositis
   Scleroderma
Skeletal abnormalities
   Lymphangitic carcinoma, metastatic disease (lytic or blastic
      lesions)
Rheumatoid lung (humeral or clavicular erosions)

Radionuclide scanning with gallium 67 has been studied extensively in DPLD. Because the isotope localizes in areas of inflammation, scintigraphy is abnormal in many cases and therefore lacks specificity. Although the intensity of uptake may reflect the degree of alveolitis, the value of scintigraphy for diagnosis, prognosis, and surveillance is limited. Magnetic resonance (MR) imaging is not generally useful in DPLD, and positron emission tomography (PET) is also unlikely to play a significant role in this area.

CLINICAL FEATURES

Most patients come to medical attention because of shortness of breath. Although dyspnea associated with infections or hypersensitivity pneumonitis may arise abruptly and progress rapidly, in most processes, the clinical syndrome is subacute or chronic. In the pneumoconioses, dyspnea not only may develop insidiously, but also may appear many years after exposure. A nonproductive cough is common in both acute and chronic cases. Fever, weight loss, fatigue, and myalgias can occur in infectious and noninfectious disorders. Other nonpulmonary symptoms may be more specific, such as uveitis in sarcoidosis and sinusitis-otitis in Wegener's granulomatosis. Asymptomatic patients with abnormal chest radiographs are most likely to have sarcoidosis or silicosis. Spontaneous pneumothorax may be the presenting event in lymphangioleiomyomatosis and histiocytosis X. Hemoptysis is unusual in most DPLDs, but occurs in cases of diffuse pulmonary hemorrhage or destructive airway disease, particularly when complicated by infection.

A careful medical, occupational, avocational, social, and travel history is essential. Antecedent disease (e.g., prior neoplasm), drug regimens, exposure to dusts or animals, and risk factors for AIDS are examples of information that may suggest a diagnosis.

The physical examination is often normal. Breathlessness and tachypnea may be apparent at rest or provoked by exertion. Jahaveri and Sicilian (1992) demonstrated that ventilation is maintained and hypercapnia is avoided by rapid shallow breathing. Fine crackles reminiscent of the sound of Velcro or cellophane are the classic finding. In fibrosis, these occur in late inspiration and are also called dry rales to distinguish them from the moist rales of pulmonary edema. Wheezing can occur in a subset of DPLDs such as sarcoidosis, histiocytosis X, and lymphangiomyomatosis. Digital clubbing is common in advanced IPF and histiocytosis X, but is uncommon in most other disorders. Physical signs of cor pulmonale (peripheral edema, ascites, hepatomegaly, accentuated pulmonic valve sound) indicate advanced fibrosis or microvascular obstruction. Nonpulmonary physical findings may suggest a diagnosis, as noted later with respect to specific diseases.

Arterial oxygen tension can be normal at rest, but exercise-induced hypoxemia is virtually universal. Hypercarbia is a late finding, reflecting respiratory muscle fatigue and end-stage disease. Pulmonary function testing should include spirometry, lung volumes, and diffusion capacity. Diffusion impairment occurs early and is the most sensitive physiologic sign of DPLD. With advancing fibrosis, pulmonary function tests show a restrictive pattern, with reductions in total lung capacity and its subdivisions. In some cases, however, cystic change, peribronchial fibrosis or granulomas, or bronchiolitis may impair airflow, producing an obstructive or combined pattern. An isolated reduction in diffusing capacity of the lung for carbon monoxide (Dlco) may suggest pulmonary vascular involvement, as is seen in systemic sclerosis.

DIAGNOSIS

A synthesis of the history, physical examination, laboratory data, and radiographs may be diagnostic. A thorough history including a detailed delineation of current and remote occupational and environmental exposures, as well as hobbies, travel, and pets, is an essential starting point. A history suggestive of active exposure to an irritant agent should prompt a trial of avoidance of the suspected agent.

P.1350


Prior chest films should be examined to date the onset of disease and assess change over time. Relevant serologic and microbiologic data should be reviewed. Invasive approaches are warranted when the clinical synthesis does not produce a confident diagnosis, when the course is atypical, when it is necessary to assess disease activity, and occasionally for legal and compensation purposes. Although biopsy of extrathoracic sites, percutaneous aspiration of the lung or pleura, thoracentesis, or mediastinoscopy may be applicable in some cases, the primary diagnostic methods for DPLD are bronchoalveolar lavage (BAL), transbronchial lung biopsy (TBLB), and surgical lung biopsy. The technical aspects of these procedures are described elsewhere in this text.

BAL is often useful in the evaluation of infectious DPLD in the immunocompromised host and may also be definitive in diffuse bronchoalveolar carcinoma, pulmonary alveolar proteinosis (PAP), eosinophilic pneumonias, and histiocytosis X, but has a limited role in most other processes. Although analysis of BAL cell types and subsets of T lymphocytes has been suggested to play a role in diagnosis and in following response to therapy in several lesions, Raghu (1995), among others, cautioned that this analysis may fail to differentiate between common disorders. In fact, cellular analysis may differ in BAL fluid retrieved at the same time from various lobes in the same patient. Thus, the benefit of BAL in the assessment of DPLD in the absence of infection remains limited.

TBLB is a reliable method for detecting processes with characteristic histiologic changes concentrated along bronchovascular sheaths, notably sarcoidosis and lymphangitic carcinoma. TBLB can sporadically diagnose many other lesions with distinctive histology but is of little value in diseases with the less specific features. Reynolds (1998) has observed that, overall, a confident pathologic diagnosis can be made by TBLB in only 25% of cases of diffuse lung disease. Coagulopathy and ventilator dependence with high airway pressures are contraindications to TBLB. The procedure may also be poorly tolerated by the very dyspneic patient. Improvements in surgical technique and the superior diagnostic yield of the larger surgical biopsy specimen have increased the attractiveness of the operative approach to diagnosis.

Surgical biopsy encompasses both open and video-assisted thoracic surgery (VATS) procedures for obtaining large samples of lung tissue necessary for diagnosis and staging of disease activity. We consider the latter technique the approach of choice in patients who can tolerate one-lung ventilation because it allows assessment of most of the lung, as opposed to the limited exposure afforded by standard open biopsy by anterior thoracotomy. VATS biopsy is also cost-effective; one of us (RBP) and associates (1997), and subsequently others, have shown that in many cases, VATS lung biopsy can safely be performed on an outpatient basis. Blewett and colleagues (2001) demonstrated the feasibility of performing open-lung biopsy on an ambulatory basis. In stable chronic DPLD, procedure-related morbidity and mortality are negligible. Gaensler and Carrington (1980) found a mortality rate of 0.3% for open biopsy. Nonetheless, the benefit of proceeding to surgical biopsy when diagnostic assessment including bronchoscopy is inconclusive remains debatable. As with all medical pathways, decisions must fit the individual patient. Some patients are considered too old or frail to tolerate biopsy. Elderly patients with clinical and radiographic features typical of IPF, people with systemic disorders frequently associated with DPLD (e.g., connective tissue disease), and those with known neoplasm likely to have lymphangitic dissemination are among the many patients in whom management without biopsy may be appropriate. If the subsequent clinical or radiographic course is atypical for the presumed diagnosis, however, definitive tissue sampling should be considered. Failure to improve after a period of observation (which may include empiric treatment) may also warrant biopsy. In all instances, a decision to proceed to biopsy should be made jointly by the thoracic surgeon, pulmonologist, and other involved physicians, such as infectious disease specialists and intensivists. Input from a chest radiologist is also crucial.

Specimens should be taken in areas of active disease rather than regions of advanced fibrosis in order to assess the histologic pattern of the active process. A ground-glass appearance on HRCT generally denotes active disease and can be helpful in identifying fruitful biopsy sites. Nishimura and associates (1992b), however, detected fibrosis in many areas of ground-glass opacity. It was axiomatic in the past that lingular and right middle lobe biopsies were unreliable. Wetstein (1986), however, as well as Miller (1987) and Chechani (1992) and their co-workers, have documented the diagnostic reliability of tissue from these lobes in both acute and chronic disorders. In the case of TBLB, several biopsy samples, from radiographically abnormal and adjacent or remote normal-appearing areas, may be helpful. If a surgical biopsy sample is obtained from an appropriate region selected on the basis of HRCT, additional specimens are unlikely to improve the diagnostic yield and may increase the risk for staple-line complications, especially in acute cases and those with advanced parenchymal disease. Monaghan and colleagues (2004), however, suggest that multiple biopsy samples may provide prognostic information in some patients with idiopathic interstitial pneumonias by better delineating those with pure UIP or NSIP from those with mixed UIP and NSIP. Regardless of the biopsy method, proper processing of the specimen is crucial. The pathologist is able to render a meaningful interpretation only when apprised of the clinical situation. In addition to light microscopy, samples should be sent for cultures and a portion saved for ultrastructural study. Inorganic dust analysis may be indicated for suspected pneumoconiosis. In acute processes, frozen sections and touch preparations are examined and stained for organisms, to facilitate early treatment.

P.1351


SELECTED DIFFUSE LUNG DISEASES

Idiopathic Pulmonary Fibrosis (Cryptogenic Fibrosing Alveolitis)

IPF is a chronic fibrosing interstitial pneumonia of unknown etiology that affects men slightly more often than women. Most patients are older than 50 years and present with dyspnea and dry cough that progress over months or years. Many recall an antecedent flulike illness; Crystal and associates (1976) reported constitutional symptoms in 50%. Dry rales and digital clubbing are common physical findings. Chest radiographs usually show a reticular or reticulonodular pattern with basilar predominance, but are normal in a few early cases. Rarely, a primarily acinar pattern is seen and may correlate with DIP histology. The most prominent changes have a characteristic basal and subpleural location by HRCT (Fig. 94-4). Pleural effusion or adenopathy is rare and should suggest an alternative diagnosis.

Although the initiating factor is unknown, immune mechanisms may be operative in the perpetuation of inflammation. Early investigations focused on the presence of immunoglobulins or immune completes in the serum or BAL fluid of affected patients. Immunologic or alternative mechanisms were postulated to result in chronic inflammation, which leads to end-stage fibrosis. Recently, the long-held hypothesis that chronic or repetitive inflammation is the major factor in the pathogenesis of pulmonary fibrosis has been called into question, with attention directed toward dysfunction in the limitation of the repair process after acute injury, as discussed by Selman and colleagues (2001).

The histology of IPF is most commonly the heterogeneous pattern of areas of architectural destruction, fibrosis, and fibroblastic foci, which may coexist with areas of relatively normal lung. Findings from TBLB and BAL may suggest IPF by the presence of inflammation, fibrosis, or both, but a definitive diagnosis requires a surgical biopsy large enough to encompass the patchy, nonuniform histology. Whether all patients with presumed IPF require biopsy is controversial. Confirmation before therapy is considered desirable for the reasons outlined by Raghu (1987), King (1991), and Bjoraker and associates (1998): (a) to eliminate other possibilities; (b) to obtain prognostic information; and (c) to justify withholding treatment for end-stage disease. In practice, however, many pulmonologists would accept a diagnosis based on firm clinical grounds in a patient who refuses or has a contraindication to biopsy. Winterbauer (1991) described a profile that permits a reliable clinical diagnosis without surgical biopsy: patient older than age 65 years with disease duration of 2 years or more, typical chest film, no extrathoracic symptoms or exposure history, negative evaluation for CVD, and consistent TBLB and BAL findings. Raghu (1995) has elaborated similar criteria for omitting surgical biopsy. HRCT should be included in the evaluation of most cases, and characteristic findings may support presumptive diagnosis.

Fig. 94-4. A 59-year-old man with idiopathic pulmonary fibrosis. A. Chest radiograph demonstrates mild reticulonodular densities bilaterally. B. High-resolution computed tomography reveals extensive honeycombing with a characteristic subpleural distribution.

The initial treatment of IPF has traditionally been with high-dose steroids. Response can be followed by a scoring system devised by Watters and associates (1986) that combines symptomatic, radiographic, and pulmonary function data. Favorable prognostic factors include younger age, disease duration of less than 1 year, histologic evidence of active inflammation (cellularity, bronchiolitis, DIP or NSIP histology, pneumonocyte proliferation), BAL lymphocytosis, and the presence of immune complexes. King and co-workers (2001) provided a detailed system for estimating survival in IPF. Unfortunately, most patients do not respond to treatment, and the overall prognosis of IPF is poor. When a response does occur, it is most often within a few weeks

P.1352


of the initiation of treatment. Stack (1972) found a 5-year survival rate of 43% for responders and 20% for nonresponders. The median survival time in IPF is only 2 to 3 years, as reported by Bjoraker (1998) and Douglas (2000) and their colleagues.

Patients who cannot tolerate or who fail steroid therapy are often treated with the immunosuppresants cyclophosphamide or azathioprine. In a randomized trial, Johnson and associates (1989) noted slightly improved survival with combined steroid-cyclophosphamide therapy, compared with steroids alone. The only other randomized study of combined steroid-immunosuppressant regimens, reported by Raghu and co-workers (1991), also failed to document a significant benefit. Nonetheless, the possibility of a minor benefit has led many to support an early combination approach, as suggested in a consensus statement by the American Thoracic Society (2000). It must be kept in mind, however, that combined antiinflammatory and immunosuppressive regimens are associated with more complications than either alone. Douglas and colleagues (1998, 2000) prospectively treated patients with the antifibrogenic agent colchicine, in comparison to steroids, and found a similarly poor response rate but with fewer side effects. Similarly, treatment of IPF with penicillamine and with pirfenidone, alone or in combination with steroids, has also not shown improved results in small numbers of patients, as discussed by Lynch and McCune (1997) and by Raghu and associates (1999). Interest in interferon- -1b as a therapeutic approach was spurred by the report of Ziesche and colleagues (1999). Pulmonary function parameters improved and levels of some inflammatory cytokines decreased in patients treated with both interferon and steroids, as opposed to those who received steroids alone. However, only 18 patients were studied. In a large multinational study reported by Raghu and co-workers (2004), 330 patients whose IPF was not responsive to steroids were randomized to receive interferon- -1b or placebo. Unfortunately, no difference between the groups was observed in progression-free survival, lung function, and quality-of-life assessment. Overall survival was somewhat better for the interferon group, but appeared limited to patients with less impairment of lung function at enrollment. Longer follow-up will help resolve this issue. Care for patients with advanced IPF is supportive, consisting mainly of long-term oxygen therapy. In a minority of younger patients, lung transplantation warrants consideration.

Clinically and pathologically distinct from the syndrome of IPF is the uncommon acute interstitial pneumonia. The patient presents with rapidly progressive symptoms as originally described in the Hamman-Rich syndrome. Katzenstein and colleagues (1986) reported eight cases and applied the term acute interstitial pneumonia. The absence of an obvious etiology separates these patients from those with the usual forms of ARDS. Diffuse alveolar damage is seen histologically. Acute exacerbation can also occur during the course of otherwise typical IPF and correlates with DAD in a background of chronic change. The prognosis for such cases is poor.

Collagen Vascular Disease

The reported frequency of lung involvement in CVD depends on the disorder and the method of detection. In general, lung disease presents after or concurrently with a diagnosis of CVD and correlates with the severity of the primary process, but it may precede or overshadow other manifestations. Although the clinical, radiographic, and histologic features are often similar to those of IPF, a variety of histologic patterns or predominant sites of pulmonary injury can be observed (Table 94-4).

Scleroderma

DPLD is most common in scleroderma (systemic sclerosis). Owens and Follansbee (1987) noted that pulmonary symptoms occur in most patients and that pathologic lung involvement is almost universal. Women are affected more often than men. Pulmonary vasculopathy may occur alone or in combination with interstitial involvement, especially in patients with the CREST variant of this condition (calcinosis, Raynaud's phenomenon, esophageal dysmotility, sclerodactyly, telangiectasia). In patients with diffuse parenchymal involvement, symptoms are similar to those seen in IPF. Chest radiographs typically show fine or coarse reticular markings. Extrapulmonary signs may include cardiomegaly from pulmonary hypertension and esophageal dilation. Aspiration from esophageal dysmotility may produce changing infiltrates. Persistent nodules or infiltrates may indicate adenocarcinoma or bronchoalveolar carcinoma, the incidence of which is increased in scleroderma. Biopsy in cases of diffuse radiographic involvement usually demonstrates UIP with more fibrosis than inflammation. Both the interstitial and vascular involvement are generally resistant to therapy, although penicillamine and systemic corticosteroids with or without an immunosuppressive agent have been reported to be helpful in some cases.

Table 94-4. Lung Involvement in Collagen Vascular Disease

Usual interstitial pneumonia
Diffuse alveolar damage
Bronchiolitis obliterans with or without organizing pneumonia
Small airways disorders
Alveolar hemorrhage
Nodular multifocal involvement
Drug-induced lung disease
Pulmonary hypertension
Infections
Recurrent aspiration pneumonia
Apical fibrobullous disease
Note: List excludes systemic vasculitis.

P.1353


Rheumatoid Arthritis

Many patients with rheumatoid arthritis (RA) develop DPLD, although symptoms are less frequent than in scleroderma. Although RA is more common in women, most patients with lung disease are men who are seropositive and have erosive nodular RA. The reported prevalence of lung involvement is a function of the sensitivity of the testing modality. Signs of fibrosis are present on chest radiographs in only about 5% of cases, usually in the form of bibasilar infiltrates. Using HRCT, however, Fujii and associates (1993) found interstitial disease in 47% overall and in 35% of those patients with normal chest radiographs. Frank and colleagues (1973) detected diminished lung volumes and diffusing capacity in 41% of patients with RA, despite normal chest films in 50%. The diffuse involvement is typically radiographically and histologically similar to IPF, but may be more slowly progressive. Rheumatoid (necrobiotic) nodules are sometimes seen radiographically and must be distinguished from carcinoma. In addition to UIP, pathologic specimens may show rheumatoid nodules, pleural fibrosis, or a characteristic bronchovascular lymphoid hyperplasia described by Yousem and colleagues (1985). Less common causes of diffuse lung involvement in RA include rheumatoid pneumoconiosis (Caplan's syndrome), pulmonary vascular disease, pulmonary nodulosis, and bronchiolitis obliterans. Drug-induced lung disease, in contrast, is a frequent complication of therapy in RA, especially with methotrexate and gold administration. Complicating pulmonary infection must always be considered in the differential diagnosis, particularly in the patient receiving immunomodulating therapy. The broad differential diagnosis of DPLD in RA and the potential for alteration of the therapeutic approach may make surgical biopsy more frequently warranted.

Systemic Lupus Erythematosus

Although pleural abnormalities are found in more than 70% of patients with systemic lupus erythematosus (SLE), DPLD is uncommon. Haupt and associates (1981) reported an autopsy incidence of only 9%. Women are affected far more often than men. The presentation of DPLD in SLE is generally insidious, with progressive cough, dyspnea, and rales. Chest radiographs feature persistent interstitial infiltrates. The UIP of SLE has no pathognomonic histologic features. Lupus can also present as an acute pneumonitis, as described by Matthay and associates (1974), with fever, dyspnea, chest pain, and hemoptysis. In this setting, radiographs show acinar opacities and consolidation. Surgical biopsy may be required, and DAD is the most frequent histologic characteristic. DIP and bronchiolitis are also seen. Mortality is high, and survivors often suffer progression to fibrosis. Diffuse alveolar hemorrhage can also occur in SLE as a fulminant syndrome and often correlates with histologic evidence of capillaritis.

Polymyositis and Dermatomyositis

DPLD is uncommon in patients with polymyositis and dermatomyositis (PM-DM), with abnormal radiographic findings noted in only 5% of cases reviewed by Frazier and Miller (1974) and 9% in the series of Salmeron and colleagues (1981). Other series have reported significantly higher rates. Women are affected twice as often as men. Although the presentation may be indolent, subacute, or rapid, an indolent course predominates, with clinical features similar to those of IPF. As in scleroderma, the radiographic picture may include signs of aspiration pneumonia in PM-DM as a result of myopathy of pharyngeal striated muscle. Tazelaar and associates (1990) reported that the pathologic findings vary with the presentation and include UIP, DAD, BOOP, and, rarely, LIP or capillaritis. Diagnosis may be difficult because DPLD often antedates other signs of PM-DM. However, many patients will have serum antibody directed against the cellular enzyme histidyl-tRNA-synthetase (anti-JO-1 antibody), as noted by Fathi and co-workers (2004).

Other Collagen Vascular Disease

Mixed connective tissue disease (MCTD) combines clinical and serologic features of SLE, RA, PM-DM, and scleroderma. It affects women eight times more often than men. Prakash (1985) and Sullivan (1984) and their colleagues reported a significant incidence of radiographic and functional abnormalities. As an overlap syndrome, the pulmonary manifestations are variable. Lung disease in Sj gren's syndrome, especially the primary type, consists mainly of lymphocytic processes. Although often discussed with CVD-associated DPLD, the apical fibrobullous changes of ankylosing spondylitis, and rarely RA, are not related to a generalized process. Restriction is present only when there is an extensive thoracic wall deformity.

Diagnosis and Treatment

The guidelines for biopsy in CVD are similar to those for IPF. When clinical findings are classic, biopsy is unnecessary and is also generally unwarranted in patients already receiving steroid or cytotoxic therapy for extrapulmonary manifestations. If there is suspicion of carcinoma (e.g., rheumatoid nodule), drug-induced injury, or opportunistic infection, and if bronchoscopic samples are nondiagnostic, surgical biopsy is indicated.

Despite much variability, CVD lung disease has a more favorable natural history and response to treatment than does IPF. Disease stability is common. Steroids and cyclophosphamide are the mainstays of therapy in cases of progressive disease. Agusti and associates (1992) noted that after 2 years of steroid treatment, pulmonary function deteriorated in IPF patients but remained stable in patients with CVD. Severe lung symptoms, however, are ominous.

P.1354


Hakala (1988) found a 3.5-year median survival time and 39% 5-year survival rate in RA patients requiring hospitalization for interstitial disease. As a general rule, response is best in patients with PM-DM or MCTD and intermediate to poor in those with RA, SLE, or scleroderma. The growing consensus is that early cytotoxic therapy is beneficial.

Sarcoidosis

Sarcoidosis is a multisystem disorder characterized by noncaseating granulomas in various tissues in the absence of an identifiable etiology. The lung is affected in almost all cases (94%). Thomas and Hunninghake (1987) stressed the role of activated T cells in the formation of sarcoid granulomas. Numbers of helper T lymphocytes are increased in the lungs of active cases, and these cells proliferate spontaneously in vitro. Paradoxically, evidence of impaired systemic immunity in the form of cutaneous anergy often accompanies augmented pulmonary cell-mediated reactivity. The cause of sarcoidosis remains unknown, although community and workplace clusters suggest a transmissible agent or environmental agent. The role of genetic susceptibility related to variability in the release and modulation of inflammatory cytokines such as TNF is under intense investigation, as reviewed by Baughman and colleagues (2003).

Most patients present in the third or fourth decade of life. Between 30% and 50% of these individuals are asymptomatic, despite abnormal chest radiographs. Blacks have a higher prevalence and severity of the disease When present, symptoms may be systemic or organ specific. The eyes (uveitis, conjunctivitis, retinitis) and skin (nodules, plaques) are the most common symptomatic extrapulmonary sites. Clinical involvement of the liver, synovial joints, central nervous system, larynx, and heart is less frequent. Early pulmonary complaints consist of cough and dyspnea. Wheezing, purulent sputum, and hemoptysis indicate significant lung destruction. The course is usually indolent, but a few patients have an acute multiorgan syndrome (L fgren's syndrome) that typically includes fever, bilateral hilar lymphadenopathy, polyarthralgias, and erythema nodosum. Of note, these particular skin lesions do not contain granulomas.

Physical findings vary from normal breath sounds or dry rales to wheezing and signs of cor pulmonale in advanced cases. Skin and ocular lesions, salivary gland enlargement, hepatosplenomegaly, and peripheral adenopathy may be noted. Laboratory evaluation may show peripheral lymphopenia resulting from lymphocyte accumulation in the lung, polyclonal hypergammaglobulinemia from T-cell activation of B lymphocytes, and hypercalcemia secondary to excess vitamin D produced by macrophages in granulomas. The level of angiotensin-converting enzyme (ACE) is often elevated in serum and BAL fluid, but is nonspecific. Restrictive lung function is typical, but obstructive physiology can result from endobronchial sarcoid, distortion of small airways by granulomas, or airway hyperreactivity.

Table 94-5. Radiographic Stages of Sarcoidosis

Stage Chest Radiographic Findings Frequency at Presentation
0 Normal 8
1 Lymphadenopathy 50
2 Lymphadenopathy and parenchymal infiltrates 30
3 Parenchymal infiltrates only 12
4 End-stage honeycomb lung Rare

The chest radiograph is almost always abnormal and defines the stage of sarcoidosis (Table 94-5). Hilar lymphadenopathy with or without mediastinal node enlargement is noted in about 80% of patients. Bilateral symmetric enlargement is the typical pattern (Fig. 94-5). Unilateral adenopathy warrants consideration of neoplasm or granulomatous infection. Parenchymal findings predominate in the upper lobes and include variable interstitial and acinar lesions. Large nodules are sometimes noted. Honeycombing, increased lung volume, apical bullae, and upward retraction of the hila are late changes. Pleural disease is rare. Pneumothorax can occur in advanced cases. HRCT typically shows small nodules located subpleurally and along bronchovascular sheaths. MR and PET imaging may reveal evidence of particular organ involvement but do not have sufficient specificity to allow secure diagnosis without resorting to tissue biopsy. The previous enthusiasm for gallium scanning in the diagnosis of sarcoidosis has been tempered with experience. Most authorities now believe that this test is insufficiently specific to be of value.

Histologically, in addition to noncaseating granulomas, a mononuclear cell infiltration may be seen but is rarely prominent. An early inflammatory response followed by granuloma formation, which may eventuate in fibrosis, is the postulated pathogenetic sequence. Granulomas are usually identifiable, even in cases of honeycomb lung and, at the opposite extreme, in patients with radiographically normal lungs. Because similar granulomas are found in other

P.1355


diseases, most pathology reports describe noncaseating granulomas consistent with sarcoidosis, emphasizing the necessity of clinicopathologic correlation. Diagnosis depends on exclusion of the spectrum of infectious, neoplastic, environmental, and immunologic conditions associated with granulomatous lung inflammation.

Fig. 94-5. A 21-year-old man with sarcoidosis. Note parenchymal involvement and symmetric bilateral hilar adenopathy (stage 2).

In some cases, biopsy of skin lesions (other than erythema nodosum), lymph nodes, liver, or other overtly affected sites is the simplest approach. The usual source of diagnostic tissue, however, is TBLB, which offers a high yield because of peribronchial clustering of granulomas. Thrasher and Briggs (1982) found granulomas by TBLB in 89% of stage I, 98% of stage II, and 88% of stage III cases. Transtracheal lymph node aspiration may be successful. Mediastinoscopy is a safe outpatient procedure with a high diagnostic yield that may be used as an alternative to TBLB. Mediastinoscopy or VATS is also indicated when bronchoscopic methods are nondiagnostic. There is debate about the universal need for tissue confirmation. Most clinicians advocate biopsy in all cases, whereas others accept a clinical diagnosis in patients with classic features.

Many patients improve or remain stable without treatment. James and associates (1976) reported spontaneous resolution in 65% of stage I, 49% of stage II, and 20% of stage III cases. About 20% suffer progressive pulmonary impairment, most having presented with stage III disease. When indicated, the current standard treatment is with steroids. Other agents, including methotrexate, cyclosporine, chlorambucil, and antimalarial drugs, have been used, but their role remains uncertain. Agents that block TNF, such as pentoxifylline, thalidomide, and infliximab, may be effective in chronic sarcoidosis, as reviewed by Baughman and colleagues (2003). Because of a high remission rate, patients with stage I or II disease at presentation who are asymptomatic and have normal lung function are generally followed clinically and radiographically. Treatment is begun if deterioration occurs and is applied at presentation in cases of stage III sarcoidosis, symptomatic stage II disease, and significant extrapulmonary organ involvement. Baseline lung function, histology, ACE level, and BAL cytology are not accurate predictors of outcome. HRCT may be helpful because response is more likely in patients with signs of active alveolitis (e.g., ground-glass haziness) than in those with radiographic findings of cystic air spaces and advanced fibrosis. The 5-year mortality rate overall is only 4%, but in most cases, death is attributable to pulmonary involvement. As pointed out by Newman and associates (1997), factors associated with a graver prognosis include black race, persistent symptoms for more than 6 months, splenomegaly, involvement of more than three organ systems, absence of erythema nodosum, and radiographic stage III disease. An acute presentation, unlike most other diffuse diseases, has a favorable prognosis with frequent spontaneous resolution and a good response to treatment.

Necrotizing sarcoid granulomatosis (NSG) is a rare disorder characterized by large granulomas with necrosis, cavitation, and vasculitis. Some consider NSG a variant of sarcoidosis, whereas others believe it is a hypersensitivity reaction or a type of angiitis.

Bronchiolitis Obliterans Organizing Pneumonia (Cryptogenic Organizing Pneumonia)

Cryptogenic organizing pneumonia (COP) is a nonspecific histologic pattern that may occur in many settings, including inhalational lung injury, infection, postradiation, transplantation, IPF, hypersensitivity, and CVD. Most cases, however, are idiopathic. Davison and associates (1983) reported an early series of COP and described the distinctive clinical syndrome. Epler and co-workers (1985) described similar cases, employing the term BOOP, which came into popular use. However, recent classification schemes have preferred COP to distinguish this syndrome from other pathologic lesions of the small airways.

Most patients have cough, dyspnea, fever, and malaise of 1 to 3 months' duration. Men and women aged 40 to 60 years are affected equally. Most have dry rales, but clubbing is rare. COP usually causes restriction and impaired diffusion. In the series of King and Mortenson (1992), all patients with obstructive findings were current or former smokers. In contrast, bronchiolitis obliterans associated with other diseases is often manifest by airflow obstruction. Although chest radiographic findings are variable, Chandler and associates (1986) stress that, in contrast to IPF, patchy air space disease is common. The CT features of BOOP, reviewed by M ller and associates (1990) and by Nishimura and Itoh (1992a), include ground-glass density, small nodules, consolidation, and bronchial wall thickening. Areas of air space consolidation constitute the most common pattern. Lee and colleagues (1994) note that consolidation is often subpleural or peribronchovascular in distribution. Unifocal BOOP may simulate bronchogenic carcinoma.

In general, diagnosis requires a surgical biopsy, but large TBLB specimens are sometimes adequate. The histologic hallmark is organizing pneumonia involving alveoli and alveolar ducts, which may be associated with bronchiolar intraluminal polyps. Geographically, the age of the lesions appears uniform. Steroids effect complete recovery in most cases, but prolonged therapy may be needed. Some patients improve without treatment. The mortality rate for BOOP is about 12%.

Hypersensitivity Pneumonitis

Also known as extrinsic allergic alveolitis, hypersensitivity pneumonitis (HP) is caused by inhalation of an allergen, usually an organicantigen. Cell-mediated mechanisms with a strong genetic modulation, as discussed by Salvaggio and deShazo (1986), appear primary. A critical role for neutrophils and the molecules they secrete has more recently

P.1356


been emphasized, as reviewed by Tobin (2001). A long list of antigens has been implicated in specific types of HP, many of which carry unusual occupational or avocational names (see Table 91-1). Exposure to the antigen causes an inflammatory reaction within the lungs, which can eventuate in interstitial fibrosis with honeycombing.

The usual presentation is acute, occasionally with cough and wheezing during exposure, but more often with dyspnea, myalgias, and fever beginning 4 to 8 hours after contact. Resolution is rapid, but symptoms recur, often more intensely, after subsequent exposures. Dry rales may persist after symptoms clear. Decreased lung volumes, impaired diffusion, and hypoxemia are found acutely but resolve between episodes. A mixed obstructive-restrictive pattern is often seen on pulmonary function studies. Continuous exposure causes a chronic illness similar to other fibrogenic diseases. Precipitating antibodies to the offending antigen are found in most cases but lack specificity in diagnosis because they are also noted in exposed asymptomatic people. Specific antigens can be used diagnostically for skin testing, in vitro lymphocyte stimulation, or inhalation challenge. Chest radiographic findings vary from normal in mild cases to hyperexpanded or honeycomb lung in advanced chronic HP. Severe acute exposure produces transient acinar lesions and consolidation. After repeated exacerbations, a fine nodular or reticulonodular pattern may persist between attacks. The HRCT findings include patchy ground-glass opacity and poorly defined nodules. Adenopathy and pleural disease are not seen in HP.

Coleman and Colby (1988) describe characteristic pathologic findings in HP, which include an interstitial infiltrate of lymphocytes, plasma cells and macrophages, a similar mononuclear bronchiolitis, and noncaseating granulomas that are less defined and less constant than in sarcoidosis and often are surrounded by a mantle of lymphocytes. Surgical biopsy is rarely necessary. Early diagnosis is of paramount importance to prevent permanent sequelae. Diagnosis is based on a history of recurrent symptoms temporally related to exposure, supportive laboratory data, and improvement after antigen avoidance. Costabel and colleagues (1984) noted that BAL lymphocytosis is higher in HP than in sarcoidosis and that suppressor T cells outnumber helper cells. When used, TBLB is generally insufficient to encompass the triad of lymphocytic pneumonitis, bronchiolitis, and granulomas. Chronic cases without obvious exposure often elude diagnosis until a surgical biopsy suggests HP and the offending antigen is determined retrospectively.

Elimination of contact is the treatment of choice and may entail job relocation, use of air masks or filtering systems, or cessation of a hobby. Steroids are used for patients with severe symptoms or progressive disease.

Pneumoconioses

The pneumoconioses result from inhalation of fibrogenic inorganic particles that reach and are retained in bronchioles and alveoli. The development of fibrosis depends on the duration and intensity of exposure, individual sensitivity, and the latency between exposure and assessment. Although occupational health measures have decreased the incidence of these diseases in the United States, new cases are still seen. In individuals with fibrotic pneumoconioses, a higher incidence of lung cancer is noted, especially squamous cell type, as discussed in the case of chromate exposure by Katabami and colleagues (2000).

Silicosis

Silicosis is caused by inhalation of silica dust or silicon dioxide in jobs such as sandblasting, mining, foundry casting, glassmaking, and ceramic molding. Patients with simple nodular silicosis may have few or no symptoms. The histologic hallmark is a small nodule with a center of whorled fibrous tissue surrounded by macrophages, plasma cells, lymphocytes, and occasionally giant cells. Some cases evolve into progressive massive fibrosis (PMF) or conglomerate silicosis, resulting from coalescence of nodules to form larger masses. Severe progressive dyspnea is the clinical correlate. In a 9-year, statewide (Michigan) study, Roseman and co-workers (1997) gave a contemporary view of affected individuals: most are older men, frequently black, exposed from the 1920s through 1960s; 30% have PMF, 32% have advanced, symptomatic simple nodular disease, and the remainder have normal lung function. Upper lung field rales may be heard; clubbing is uncommon. Radiographs show a nodular pattern mainly in the upper lobes. Enlarged hilar nodes with peripheral calcium are found in a minority of patients. Although once thought to be pathognomonic of silicosis, Parkes (1994) noted that such eggshell calcification is occasionally found in sarcoidosis and other granulomatous conditions as well as after irradiation. Patients with silicosis are at particular risk for infection withboth Mycobacterium tuberculosis and atypical organisms, possibly because silica-containing macrophages have impaired phagocytic ability. Infection should be considered when unexpected clinical or radiographic deterioration occurs. On occasion, massive exposure causes acute silicosis, characterized clinically by severe symptoms, radiographically by perihilar air space and ground-glass opacities, and histologically by a reaction with features of DIP and alveolar proteinosis.

Asbestosis

Exposure to asbestos in occupations such as welding, pipefitting, boiler making, and insulating is a well-known cause of pulmonary fibrosis. Asbestosis is histologically similar to UIP, but with the pathognomonic finding of beaded or dumbbell-shaped asbestos fibers (ferruginous bodies) made up of a translucent fiber core coated by protein and ferritin. The presence of even a few asbestos bodies indicates significant exposure because they outnumber

P.1357


by a factor of several thousand retained uncoated fibers, which are not seen by light microscopy. Dyspnea and paroxysmal dry cough appear many years after exposure and may progress to respiratory failure. Basilar crackles and clubbing are common. Radiographically, fine reticulation progresses to honeycombing. Staples and colleagues (1989) reported fibrosis demonstrated by HRCT and impaired lung function in exposed workers with normal chest radiographs. Al-Jarad and associates (1992) found HRCT helpful in differentiating IPF and asbestosis. The clinical course of patients with asbestosis may be complicated by benign or malignant pleural disease and by lung cancer.

Coal Worker's Pneumoconiosis

Coal worker's pneumoconiosis is uncommon and usually asymptomatic. Cough and expectoration of black sputum in miners is usually due to coal dust bronchitis and does not indicate fibrosis. A minority of patients, however, progress from asymptomatic small nodules (simple pneumoconiosis) to PMF and respiratory disability.

Beryllium and Other Metals

Inhalation of beryllium can cause an acute illness with DAD histology or a chronic syndrome similar to sarcoidosis, with hilar adenopathy and multiorgan granulomas. Cobalt and tungsten are also heavy metal pulmonary toxins and are associated with a distinctive giant cell reaction.

Diagnosis

Documentation of exposure and typical clinical data generally suffice for the diagnosis of pneumoconiosis for medical and compensation purposes. When surgical biopsy is performed, it is essential to search for retained dusts by light microscopy, electron microscopy, or mineral analysis. There is no primary therapy for the pneumoconioses.

Eosinophilic Lung Disease

A diverse group of disorders are linked by the common finding of DPLD associated with increased tissue or circulating eosinophils. A variety of classification systems have been proposed and have been reviewed by Allen and Davis (1994).

Simple Pulmonary Eosinophilia

Patients with simple pulmonary eosinophilia (L ffler's syndrome) have mild or no symptoms. Transient migratory infiltrates are noted on serial radiographs, and there is blood and sputum eosinophilia. The syndrome is idiopathic in one third of cases and drug induced or caused by parasites (e.g., Ascaris, Toxocara, Strongyloides species) in the remainder. Rapid resolution is the rule and may be spontaneous or follow treatment of infection or drug avoidance.

Acute Eosinophilic Pneumonia

Patients with acute eosinophilic pneumonia present with rapid progression of fever, myalgia, chest pain, often severe respiratory distress, and diffuse alveolar infiltrates. Although peripheral eosinophils are usually not increased, significant BAL eosinophilia is universal. Pleural effusions are sometimes seen and also contain large numbers of eosinophils. The cause of acute eosinophilic pneumonia is unknown. Biopsy shows eosinophils in the alveoli and interstitium, without fibrosis or vasculitis. High-dose steroids effect a rapid resolution in most cases of acute eosinophilic pneumonia, and relapse is rare.

Chronic Eosinophilic Pneumonia

Chronic eosinophilic pneumonia is also an idiopathic syndrome occurring mainly in women in a ratio of 2:1, up to 50% of whom give a history of asthma or rhinosinusitis. Symptoms progress slowly over weeks or months and include fever, cough, dyspnea, wheezing, night sweats, and weight loss. Peripheral eosinophilia may be absent but is seen in most cases. BAL eosinophilia is present during active disease. Serum immunoglobulin E (IgE) is usually elevated. Dense peripheral infiltrates (Fig. 94-6), described by Gaensler and Carrington (1977) as the photographic negative of pulmonary edema, are fairly specific, but were present in less than half the cases reviewed by Jederlinic and colleagues (1988). Peripheral consolidation, however, was noted on CT scans in all cases reported by Mayo and colleagues (1989). Mediastinal lymphadenopathy is often present. Histologically, interstitial fibrosis is noted in many cases, along with an infiltrate of eosinophils and lymphocytes.

P.1358


Although the response to steroids is excellent, relapses are common, especially early after treatment. Naughton and associates (1993) showed that long-term, and in some cases lifetime, treatment may be necessary. Progression to diffuse pulmonary fibrosis can occur sporadically or with delayed diagnosis or inadequate treatment.

Fig. 94-6. A 21-year-old woman with eosinophilic pneumonia. Chest radiograph shows typical peripheral densities that have been called the photographic negative of pulmonary edema.

Idiopathic Hypereosinophilic Syndrome

Idiopathic hypereosinophilic syndrome, once known as eosinophilic leukemia and characterized by extreme BAL and peripheral eosinophilia (30 to 70% of white blood cell count), is a multiorgan disorder associated with DPLD in a minority (40%) of cases. Most patients die from cardiac involvement or thromboembolic complications. Eosinophilic infiltration may also cause symptomatic problems in the gastrointestinal tract, kidneys, skin, and muscles. The hypereosinophilic syndrome occurs with a male to female ratio of 7:1.

Allergic Bronchopulmonary Aspergillosis

Allergic bronchopulmonary aspergillosis (ABPA) is seen in asthmatics and is defined by peripheral and BAL eosinophilia, increased IgE levels and serum, and BAL precipitins to Aspergillus species. CT scans show central bronchiectasis, mucus plugging of airways, and consolidation distal to impacted bronchi. The term toothpaste shadows has been used for the radiographic appearance of dilated, impacted airways and aptly describes the gross findings of thick mucoid material filling the bronchi. Treatment is with steroids and possibly antifungals. Late sequelae of ABPA include pulmonary fibrosis, irreversible airway obstruction, and aspergilloma.

Churg-Strauss Syndrome

Churg-Strauss syndrome (allergic angiitis and granulomatosis) is another eosinophilic lung disorder that occurs in asthmatics. Men and women are affected equally and usually develop vasculitis months or years after a history of allergic disease. Recently, the emergence of Churg-Strauss syndrome has been reported to be associated with the treatment of asthma with the newer oral agents, leukotriene receptor antagonists, as suggested by Knoell and colleagues (1998). Wechsler and co-workers (2000), however, believe that the occurrence of Churg-Strauss syndrome in patients so treated may be due to unmasking of the syndrome that was previously suppressed by steroid administration. The vasculitis is systemic, with frequent involvement of the nasal passages, sinuses, skin, gastrointestinal, central nervous, and musculoskeletal systems. Unlike in Wegener's granulomatosis (see later), significant renal involvement can occur, but is uncommon. Heart failure from coronary vasculitis, pericarditis, and hypertension are common causes of morbidity and death. Radiographs display patchy infiltrates that may wax and wane, as well as pleural effusions in one third of cases, and occasionally thoracic lymphadenopathy. Blood and BAL eosinophil counts are markedly elevated, as is serum IgE. The perinuclear form of antineutrophil cytoplasmic antibody (p-ANCA) may also be elevated. Surgical lung biopsy is generally required for a definitive diagnosis and shows a necrotizing giant cell vasculitis involving arteries and veins along with granulomas and eosinophilic infiltration. Without steroid treatment, about 50% of patients die within 3 months of diagnosis.

Bronchocentric Granulomatosis

Bronchocentric granulomatosis is a rare disorder featuring necrotizing granulomatous inflammation of bronchial epithelium that in some cases (one third) is similar to ABPA in that it is seen in asthmatics and is associated with eosinophilic infiltration, blood eosinophilia, and fungal organisms on biopsy or in sputum. The remaining patients have a neutrophilic reaction. Most cases present as a solitary radiographic density rather than DPLD. Diagnosis usually requires resection.

Diagnosis

The diagnosis of eosinophilic lung disease is most often clinical, based on antecedent illness, symptoms, radiographs, serum studies, BAL, assessment for parasitic or fungal infection or drug reaction, and response to therapy. The finding of eosinophilia on sputum evaluation or BAL may be helpful but is not specific. TBLB sometimes yields a sufficient specimen. Surgical biopsy is indicated before treatment in some instances and in patients with atypical features or a nondiagnostic TBLB.

Neoplasms

Lymphangitic Carcinoma

Primary and metastatic malignancies can involve the lung in a diffuse fashion. In lymphangitic carcinoma, tumor cells are found within the peribronchial and perivascular lymphatic channels. The etiology is believed to be seeding of lymphatics by hematogenous tumor emboli rather than spread from invaded nodes. Adenocarcinomas of the breast, lung, stomach, prostate, and pancreas are the usual sources, but squamous cancer can also occasionally cause lymphangitic carcinoma. Pulmonary disease rarely precedes clinical evidence of the primary tumor. The course may be insidious or rapid. Diffuse microvascular occlusion by tumor emboli can produce a similar syndrome with early cor pulmonale. Chest radiographs show a nonspecific interstitial pattern. Because of the neoplastic etiology, associated adenopathy and pleural effusion are common. The CT features of lymphangitic carcinoma are often characteristic (Fig. 94-7). Stein and associates (1987) described nodularity and irregular thickening

P.1359


of bronchovascular sheaths, interlobular septa, and fissures. When tissue confirmation is indicated, TBLB is usually diagnostic because of the peribronchial location of the malignant cells. BAL has been useful in diagnosis, particularly in coagulopathic patients. With few exceptions, the prognosis is dismal, but response has been seen in tumors sensitive to chemotherapy or hormonal manipulation.

Fig. 94-7. A 70-year-old man with lymphangitic carcinoma. A. Chest radiograph demonstrates diffuse irregular and nodular densities with possible hilar and mediastinal adenopathy. B. High-resolution computed tomography demonstrates peribronchial and perivascular thickening, scattered nodules, septal thickening, and spiculation of the fissure.

Bronchoalveolar Carcinoma

Bronchoalveolar carcinoma manifests in a multinodular or diffuse pneumonic form in about 20% of cases. The mode of dissemination may be aerogenous. Most patients with diffuse bronchoalveolar carcinoma are dyspneic. Bronchorrhea is rare but may be copious and debilitating. Radiographs demonstrate nodularity or air space opacities, air bronchograms, and consolidation. Im and associates (1990) reported that mucin production by bronchoalveolar carcinoma may result in a lower attenuation on CT scan and helps differentiate this tumor from other consolidative processes. The diagnosis of diffuse bronchoalveolar carcinoma is made on the basis of sputum cytology, bronchoscopy, or transthoracic needle aspiration. On occasion, however, the cellular features are so bland that only a surgical specimen allows recognition of the typical architectural pattern.

Lymphoma and Leukemia

Diffuse lymphomatous infiltration of the lung occurs in patients with known, usually non Hodgkin's, lymphoma that originated at extrapulmonary sites. Primary pulmonary lymphoma is uncommon and is distinguished by lung involvement in the absence of extrathoracic lymphoma. Small cell lymphoma can present as a diffuse process, but large cell lymphomas more often are focal. Likewise, primary B-cell pulmonary lymphoma occurring in AIDS is usually nodular and focal. In leukemia, diffuse neoplastic infiltrates are noted at autopsy in up to 50% of cases, but usually do not cause symptoms during life. However, Yoshioka and colleagues (1985) observed that T-cell leukemia is often associated with fibrosis, possibly because of fibroblast activation by interaction of malignant cells and local alveolar macrophages. As in lymphangitic carcinoma, leukemic and lymphomatous infiltrates track along lymphatics, resulting in similar radiographic findings. Despite this location, the yield of TBLB is less than in lymphangitic carcinoma, because, as pointed out by Palosaari and Colby (1986), cytologic atypia may be absent, and peribronchial lymphoid infiltrates are found in association with many benign disorders. Newer techniques for establishing monoclonality, however, may improve the yield of BAL, TBLB, fine-needle aspiration, and pleural fluid cytology, as documented by Keicho and associates (1994), who used gene rearrangement and immunophenotypic analysis to establish both T- and B-cell monoclonality in BAL fluid. Surgical biopsy, however, remains necessary in many cases.

Kaposi's Sarcoma

The most common AIDS-related neoplasm, Kaposi's sarcoma often affects the lungs diffusely. White and Matthay (1989) reported clinical pulmonary Kaposi's sarcoma in 33% of AIDS patients with systemic Kaposi's sarcoma who had lung lesions and at autopsy in 50% of cases of systemic Kaposi's sarcoma. Isolated pulmonary Kaposi's sarcoma is rare. Patients have dry cough, dyspnea, and fever. Endobronchial involvement may cause hemoptysis and wheezing. Radiographic findings include bilateral perihilar

P.1360


interstitial and acinar opacities and poorly marginated nodules. Pleural effusion is present in 30% and lymphadenopathy in 8% of patients. Diffuse alveolar hemorrhage can produce rapid changes in the radiographic appearance. Atelectasis may result from endobronchial lesions. Naidich and colleagues (1989) reported that CT scan findings might be similar to those in lymphangitic carcinoma, but with a perihilar distribution.

Histologically, there are nodular foci of plump spindle cells with frequent mitoses and signs of recent or old hemorrhage. In some areas, spindle cells are sparse, and there is a prominent infiltrate of lymphocytes and plasma cells. Endobronchial Kaposi's sarcoma appears grossly as bright red-violet, flat or raised lesions. Because the appearance is characteristic and the yield of biopsy is inconsistent, presumptive diagnosis is often made by the gross findings. Surgical biopsy was once considered essential for the diagnosis of parenchymal disease, but Purdy and associates (1986), among others, have shown that TBLB is sufficient in many cases. Fears of excessive bleeding from endobronchial biopsy and TBLB have not been substantiated by experience. A clinical diagnosis may be reasonable in some patients with systemic Kaposi's sarcoma, typical radiographic findings, and no evidence of infection by BAL. Life expectancy in AIDS-associated Kaposi's sarcoma is limited, although some patients survive for several years. Prolongation of survival by combination chemotherapy was reported by Gill and associates (1989). Local radiation therapy may palliate hemoptysis and airway obstruction.

Lymphomatoid Granulomatosis

Lymphomatoid granulomatosis is an uncommon disorder that has been variously classified as a lymphoid infiltration, a neoplasm, or a vasculitis. Initially described by Liebow and associates (1972), lymphomatoid granulomatosis is marked by a necrotizing angiocentric granulomatous infiltration of small lymphocytes, macrophages, plasma cells, and atypical lymphoreticular cells. Middle-aged men are affected most often and present with cough, chest pain, dyspnea, fever, and malaise. Involvement of other organs occurs in one third to one half of cases and may manifest as neurologic symptoms, skin rash, nephropathy, peripheral lymphadenopathy, and hepatomegaly. Radiographs show generally lower lung field nodules of variable size, often with cavitation. Small pleural effusions and hilar lymphadenopathy are seen in some cases. Diffuse infiltrates are uncommon. Initial reports indicated that lymphoma developed in 12% of patients. T-cell marker studies and identification of monoclonal atypical cell populations, however, as summarized by Myers (1990), have reclassified some cases as T-cell lymphomas. Pisani and DeRemee (1990), on the other hand, consider lymphomatoid granulomatosis to be a reactive process with a propensity for loss of immunologic control and progression to lymphoma. Cadranel and co-workers (2002) also point out that the clonal nature of the cellular proliferation in lymphomatoid granulomatosis is rarely demonstrated, in contrast to classic primary pulmonary lymphoma. Despite uncertainty, lymphomatoid granulomatosis appears to be a lymphoproliferative disorder with a tendency for the development of diagnosable lymphoma. Although some untreated patients have an indolent course, many die of respiratory failure within 2 years. Fauci and associates (1982) reported long-term remission after treatment with prednisone and cyclophosphamide, but some patients require aggressive chemotherapy. Stem cell support has also been employed. Surgical lung biopsy is usually required for diagnosis, except in patents with accessible cutaneous lesions or peripheral lymphadenopathy.

Lymphocytic Interstitial Pneumonia

Although occasionally an isolated DPLD, LIP usually occurs in association with immune disorders such as Sj gren's syndrome, lupus, myasthenia, and chronic active hepatitis. It is also found in about 50% of HIV-positive children who have pulmonary disease and is considered a criterion for the diagnosis of AIDS in young people. In contrast, only 3% of adults with AIDS and lung disease have LIP. Swigris and colleagues (2002) have provided a discussion of current understanding of this enigmatic entity. Presentation is similar to other interstitial pneumonias. Presenting symptoms may, however, be those of the underlying systemic or autoimmune disease. Pulmonary involvement may be an incidental finding or disclosed in the evaluation of cough or dyspnea. As reported by Oldham and associates (1989), the radiographic appearance is nonspecific and includes combinations of reticular or reticulonodular opacities and air space consolidation. Pleural effusion and lymphadenopathy are unusual.

The etiology of LIP is unknown, although its associated disorders suggest an autoimmune mechanism. Also, some evidence in AIDS patients suggests that LIP is a response to infection with HIV, Epstein-Barr virus, or both. In AIDS-associated LIP, the infiltrating lymphocytes are T cells, whereas in other forms of LIP, B lymphocytes predominate, suggesting a variable pathogenesis. Similar to the situation with lymphomatoid granulomatosis, the relation between LIP and lymphoma is incompletely understood (see Chapter 119). Differentiation may be difficult in the absence of atypia and invasion. Cellular polyclonality favors LIP, whereas lymphoma is associated with a monoclonal infiltrate. Herbert and associates (1985), however, found that polyclonal populations may exist within malignant lymphomas. Other evidence suggests that LIP is a reactive pneumonitis, including cases of spontaneous regression, an often excellent response to steroids, and documented progression to typical fibrosis. However, the application of immunohistochemistry and molecular techniques has enabled the separation of neoplastic disease from other reactive inflammatory patterns. The clinical course of LIP is variable. Some patients improve with no treatment, whereas others

P.1361


progress despite treatment. Corticosteroids are thought to improve symptoms or arrest disease progression in many instances. Nonetheless, one third to one half of patients die within 5 years of diagnosis, and 5% progress to definite lymphoma. Plasma cell interstitial pneumonia, a rare variant of LIP described by Moran and Totten (1970), does not appear to progress to fibrosis or lymphoma.

Pulmonary Vasculitis

The pulmonary vasculitides are variably classified. Fulmer and Kaltreider (1983) divided these lesions into three categories: the lung as the major affected organ; lung involvement as part of a multiorgan syndrome (e.g., cryoglobulinemia, Henoch-Sch nlein purpura, polyarteritis nodosa); and processes in which vasculitis is but one aspect of pulmonary pathology (e.g., CVD, eosinophilic pneumonia). Four processes are often included in the first category: lymphomatoid granulomatosis, necrotizing sarcoid granulomatosis, allergic angiitis and granulomatosis (Churg-Strauss syndrome), and Wegener's granulomatosis. Some recent categorization schemes are based on the size of involved blood vessels (small, medium, large), but substantial overlap exists within this system as well as with respect to other features, as emphasized by Jennette and Falk (1997). The variability in classification of these disorders is emphasized by the fact that the first three syndromes have been discussed under other headings in this chapter: lymphomatoid granulomatosis under Neoplasms; necrotizing sarcoid granulomatosis under Sarcoidosis; and Churg-Strauss syndrome under Eosinophilic Lung Disease.

Wegener's granulomatosis, as originally described, is defined by a triad of necrotizing granulomas of the lung and upper respiratory tract, glomerulonephritis, and systemic vasculitis. Carrington and Liebow (1966) later reported a limited form with little or no extrapulmonary involvement. In addition to pulmonary symptoms, sinusitis, epistaxis, rhinitis, fever, and weight loss are frequent complaints. Lung involvement is noted in about 75% of patients at presentation and ultimately in 95%. Proteinuria or hematuria is detected in most. Large airway obstruction, saddle nose deformity, skin lesions, ocular inflammation, diabetes insipidus, cranial nerve palsy, and coronary vasculitis are seen in some patients. Chest radiographs typically show multiple, often cavitary nodules of variable size, although Leatherman (1988) reported solitary nodules in many patients. Lesions may resolve and reappear spontaneously. Pleural effusion is noted in 20% to 30%. Atelectasis from large airway Wegener's, air space consolidation from pulmonary hemorrhage, and lymphadenopathy are occasional features. Cordier and associates (1990) found CT scan useful for detecting unsuspected cavitation. Radiographs of the sinuses often show Wegener's granulomatosis lesions or the residua of infection. Rosenberg and colleagues (1980) found that pulmonary function tests demonstrate obstructive more often than restrictive disease. Serum antibodies to neutrophil cytoplasmic antigens (ANCA) are a valuable marker, but 10% of patients are ANCA negative. Harrison and associates (1989) reported that a positive ANCA is specific for Wegener's granulomatosis in 86% of cases. The cytoplasmic (c-ANCA) antibody is more specific for Wegener's granulomatosis, as opposed to the perinuclear (p-ANCA) pattern. Histology shows transmural necrotizing granulomatous inflammation of the walls of small and medium arteries and veins. Capillaritis may be found in cases with alveolar hemorrhage and constitutes the most life-threatening manifestation of Wegener's. Because of the sensitivity and specificity of ANCA testing, biopsy is often not required, especially when the history and radiographs are typical. When the clinical synthesis is ambiguous, TBLB is usually inadequate because of the frequent finding of nonspecific necrosis as well as features common to other granulomatous processes, and surgical biopsy may be required for diagnosis. Prolonged therapy with high-dose prednisone and cyclophosphamide achieves 5-year survival rates as high as 90%.

Drug-Induced Lung Disease

The list of drugs that are known to cause DPLD continues to grow. Damage results from direct toxic effects or from induction of the cellular events common to all lung injury. Some agents produce damage in a dose-dependent fashion, whereas in the case of other drugs, the occurrence of DPLD is idiosyncratic. The histology is variable and mostly nonspecific. A comprehensive overview of pathologic patterns and pathogenesis is provided by Pietra (1991). Histologic variations include DAD, hemorrhage, UIP-DIP, eosinophilic pneumonia, lipid pneumonia, COP, and lymphoplasmacytic interstitial pneumonia. The importance of a detailed medical history is obvious. Diagnosis may be complicated by factors such as antecedent or coexisting pulmonary disease, administration of multiple potential toxins, and the possibility of associated diffuse lung infection. The diagnosis is usually clinical. Improvement or resolution following drug withdrawal, of course, supports the diagnosis of drug-induced DPLD. In some cases, bronchoscopic specimens are sufficient for a presumptive diagnosis, but surgical biopsy is sometimes needed. In most cases, drug avoidance is all that is required, but steroids are needed in some instances. Cooper and associates (1986a, 1986b) reviewed the subject, and Rosenow and colleagues (1992) presented an update. A few examples are summarized.

Nitrofurantoin

In addition to an acute eosinophilic pneumonia, nitrofurantoin can cause a syndrome similar to IPF. The acute and chronic reactions are distinct entities and do not overlap. Most patients improve with cessation of the drug, but in a

P.1362


few, the condition progresses, requiring treatment with steroids.

Gold

Gold therapy, used mainly for rheumatoid arthritis and sometimes for ankylosing spondylitis, psoriatic arthritis, pemphigus, and asthma, can cause DPLD. Distinction of drug toxicity from RA-associated UIP may be difficult. Tomioka and King (1997) suggested clinical features supportive of a diagnosis of gold-induced DPLD, including female predominance, acute dyspnea, recent dry cough, fever, skin rash, absence of subcutaneous nodules and digital clubbing, crackles on chest examination, and alveolar opacities along bronchovascular bundles on CT scan. Most patients have peripheral eosinophilia and BAL lymphocytosis. ANCA positivity occurs in about 70% of cases, usually in a perinuclear staining pattern in contrast to the cytoplasmic ANCA often associated with Wegener's. Biopsy is usually nonspecific, including the presence of gold within macrophages because many treated patients have this finding in the absence of DPLD. The key test is drug withdrawal, which usually results in resolution within a few weeks.

Amiodarone

Among cardiovascular drugs, amiodarone is the most common cause of pulmonary toxicity, which occurs in up to 5% to 10% of patients taking a daily dose of 400 mg or more for 2 or more months, as reported by Martin and Rosenow (1988). Pulmonary toxicity is less frequent but still observed with lower dosing, according to Ott and colleagues (2003). In addition to dyspnea and cough, patients may have fever and pleuritic chest pain without effusion. CT may be helpful because areas of amiodarone accumulation are denser than other soft tissues. The histology is variable and includes UIP, DAD, and COP. Foamy macrophages are a nonspecific but characteristic finding and can be retrieved by BAL. In most cases, the diagnosis is based on clinical data and BAL. Resolution usually follows drug withdrawal. A syndrome of fulminant ARDS has been seen following cardiopulmonary bypass in patients currently or previously taking amiodarone who demonstrated preoperative pulmonary toxicity. Prolonged pump times and exposure to high inspired oxygen concentration appear to increase the risk, as reviewed by Nalos and colleagues (1987). Others, including Dimopoulou (1997) and Crystal (2003) and their co-workers, have not found an association between high- or low-dose amiodarone and postoperative pulmonary complications.

Chemotherapeutic Agents

Chemotherapeutic agents are a frequent cause of diagnostic difficulty. Although some can produce an acute hypersensitivity reaction or noncardiogenic pulmonary edema, symptoms develop gradually in most cases. Radiographs may be normal but more often show nonspecific mixed patterns. The histology usually reflects a subacute process with features of DAD, mononuclear cell infiltrates, and mild fibrosis. Atypia of type II epithelial cells is a distinctive feature and may simulate malignancy. Toxicity from bleomycin and mitomycin is dose dependent but is enhanced by prior or concurrent thoracic irradiation and by high levels of inspired oxygen. The danger is particularly acute in patients undergoing operation after induction therapy for thoracic tumors. Our current practice is to administer steroids in the perioperative period and to maintain the lowest possible inspired oxygen level because even brief periods of hyperoxia may be detrimental. Although the mortality rate for advanced toxicity is high, high-dose steroids, as reported by Maher and Daly (1993), may effect reversal. Methotrexate may cause an idiosyncratic reaction in about 5% of patients. Methotrexate pneumonitis has several characteristic features, including peripheral eosinophilia, hilar adenopathy, small pleural effusions, and a characteristic histology (i.e., an intense mononuclear infiltrate of lymphocytes, plasma cells, and eosinophils, along with giant cells and granulomas). The mortality rate is low. Most patients improve following drug withdrawal, but steroids may hasten resolution. Gemcitabine appears to be active against solid tumors such as lung, breast, pancreas, and ovary. Pavlakis and associates (1997) and Vander Els and Miller (1998) reported three cases and one patient, respectively, in whom this drug was associated with DPLD. Two patients in the first report died of respiratory failure and had DAD histology on postmortem examination, whereas the third was treated by drug withdrawal and steroids and survived. The patient in the second report recovered rapidly after institution of prednisone. The increasing use of gemcitabine, especially in lung cancer, warrants recognition of this possible toxicity. Among the many other chemotherapeutic agents that may cause lung damage, busulfan is noteworthy for the often extremely long interval between exposure and the development of symptoms.

Other Drugs

Recent therapeutic advances have been associated with pulmonary toxicity. Combination therapy of hepatitis C with interferon and ribavirin has been associated with interstitial pneumonitis and organizing pneumonia. Fludarabine appears to cause pulmonary toxicity, particularly in patients with chronic lymphocytic leukemia. Illicit drugs are usually associated with noncardiogenic pulmonary edema, but cocaine can also cause diffuse alveolar hemorrhage, DAD, eosinophilic infiltrates, and BOOP-COP. Resolution occurs with supportive care and abstention. In contrast, talc (magnesium silicate) particles from intravenous injection of dissolved pills can lead to progressive granulomatous inflammation and fibrosis resulting in cor pulmonale and death. Radiographic features vary from normal to miliary nodularity to advanced fibrosis. Talc can sometimes be identified by BAL, but TBLB or surgical biopsy may be required.

P.1363


The herbicide paraquat has received much attention because of frequent fatalities from an acute multisystem syndrome, but DPLD and fibrosis may occur on a more chronic basis.

Radiation Toxicity

Movsas and associates (1997) provided a useful review of this subject. Radiation lung damage is divided into early (pneumonitis) and late (fibrosis) and depends on the total dose, fractionation, volume treated, and individual susceptibility (see Chapter 109). The pathogenesis begins with cellular damage and the generation of oxygen radicals. Although only about 10% of patients develop clinical disease, radiographic changes are noted in 40% by plain film, and virtually all cases, by HRCT. Pneumonitis usually occurs 2 to 6 months after treatment but can begin within 2 weeks. Ikezoe and associates (1988) found CT scan changes in most patients within 4 weeks. Symptoms vary from mild cough, fever, and dyspnea to ARDS. Radiographs demonstrate ground-glass and acinar opacities that may progress to consolidation and may extend beyond the radiation port. DAD is the major pathologic pattern, along with pneumocyte atypia, scattered small vessel thromboses, and intimal myoepithelial foam cells. High-dose steroids may reverse the process. Late fibrosis usually follows within 6 to 24 months of acute pneumonitis but can develop insidiously without evidence of early reaction. Symptoms vary from slight to severe. Linear opacities conform to the radiation field and may produce a vertical straight edge sign after mediastinal radiation. Upper-zone volume loss and upward hilar retraction are common. When tangential beams are used, the distribution varies. Histologically, vascular changes are prominent and include microvascular obliteration, medial hyperplasia, fibrosis, and foam cell accumulation in larger arterioles. No treatment is effective at this stage. The risk for radiation damage is increased in patients with preexisting lung disease (e.g., emphysema, asbestosis, other pneumoconiosis) and those treated with agents such as bleomycin, dactinomycin, doxorubicin, cyclophosphamide, and vincristine, especially when administered concomitantly with radiation. The diagnosis is made on the basis of history, review of the radiation details, and absence of evidence of infection or recurrent malignancy. Bronchoscopy with BAL may be used diagnostically to evaluate for infection, but surgical biopsy is rarely necessary. Radiation-induced lung disease and lymphangitic carcinoma may be difficult to distinguish.

Miscellaneous Diffuse Disorders

Histiocytosis X (Langerhans' Cell Histiocytosis)

Pulmonary histiocytosis X is an uncommon idiopathic disorder marked by interstitial infiltration of atypical histiocytes akin to the Langerhans' cell normally seen in skin containing intracytoplasmic tennis racket shaped X bodies on electron microscopy. Histiocytosis X is preferable to the older term eosinophilic granuloma because the process is neither primarily eosinophilic nor granulomatous. Current terminology prefers Langerhans' cell histiocytosis (LCH) for these disorders. There is a spectrum of disease, and pulmonary symptoms are prominent in only a small subset. Most patients with pulmonary LCH are young adults, with men predominating in some series. It is striking that more than 90% are smokers. About 20% are asymptomatic despite an abnormal radiograph. A similar proportion present with or later develop pneumothorax. Constitutional symptoms can accompany cough and dyspnea. Cystic lesions in ribs, skull, and pelvis are found in 20% and may be painful. Diabetes insipidus is noted in 10%. A reticulonodular pattern with cystic honeycombing that spares the bases is the classic radiographic finding. Lung volumes are normal or increased. The additional finding of a solitary lytic rib lesion is highly suggestive of pulmonary LCH. Brauner and associates (1989) described the HRCT features, which include thin-walled cysts and small, sometimes cavitated nodules (Fig. 94-8). Pulmonary hypertension may also be noted from radiographic signs. Pulmonary function tests often indicate a mixed obstructive and restrictive process with reduced diffusion capacity. Histologically, the Langerhans' cells take up stain with S100 protein and are joined by lymphocytes, fibroblasts, macrophages, and eosinophils to form stellate nodules. Despite advanced fibrosis in later stages, Langerhans' cells and a stellate pattern may still be found. Increased numbers of Langerhans' cells have been noted in BAL but are nonspecific because they can occur in association with other diseases. Transbronchial lung biopsy may be diagnostic, but surgical biopsy is often required. The disease may resolve, stabilize, or progress to end-stage fibrosis. Steroids may be helpful early in the course of the disease.

Lymphangioleiomyomatosis

Lymphangioleiomyomatosis (LAM) is a rare disease resulting from hamartomatous proliferation of specialized smooth muscle cells in airways, lymphatics, and blood vessels (see Chapter 118). Sullivan (1998) has recently provided a valuable review of the current knowledge about this lesion. LAM occurs almost exclusively in women, mainly during the childbearing years, and is thought to be hormone related. LAM has been reported, however, in an 11-year-old girl and in postmenopausal women up to age 75 years, most, but not all, of whom were taking estrogens. In addition, onset or exacerbation of LAM has been reported during pregnancy. An interesting and unexplained comorbidity is the occurrence in 44% of LAM patients of an uncommon renal tumor: angiomyolipoma. When present, these tumors are bilateral in 50%. LAM patients present with dyspnea, which may be acute or subacute and result from pneumothorax or chylothorax or slowly progressive from parenchymal infiltration (Fig. 94-9).

P.1364


Ryu and colleagues (2003) found a 10% incidence of chylothorax, 25% of which were bilateral, among 79 patients seen at the Mayo Clinic. As in LCH, pneumothorax may be the presenting finding in LAM. Chu and associates (1999) noted that about two thirds of patients will develop a pneumothorax at some point in the course of the disease. Alveolar hemorrhage and hemoptysis can also occur, as well as cough, chyloptysis, chyluria, chylous ascites, chylous pericardial effusion, and lower extremity lymphedema. In the early stages, radiographs may be normal but usually show nonspecific reticulonodular infiltrates. Lung volumes are normal or increased. Alveolar hemorrhage may produce a changing picture. Effusion or pneumothorax may be found on the initial chest film. Disease progression leads to fine interstitial changes followed by honeycombing. HRCT is abnormal in most cases and shows myriad thin-walled cysts similar to LCH, but without basilar sparing and without nodules. Abdominal and pelvic lymph nodes may be enlarged. Pulmonary function parameters are variable, showing restrictive, obstructive, or mixed patterns. Although progesterone and tamoxifen may be beneficial, Urban and colleagues (1992) showed eventual deterioration in many cases. Oophorectomy has also been employed. The efficacy of treatment is difficult to evaluate because of the rarity of the condition and because the natural history of LAM is variable, as emphasized by Sullivan (1998), with 8-year survival rates varying between 38% and 78%. With respect to diagnosis, surgical biopsy is generally needed, although TBLB may be sufficient in some instances.

Fig. 94-8. A 45-year-old woman with cough and mild exertional dyspnea. Chest radiograph (A) and high-resolution computed tomography (B) show multiple nodules, cavities, and thin-walled cysts. Biopsy confirmed histiocytosis X/Langerhans' cell histiocytosis.

Fig. 94-9. A 35-year-old woman with lymphangioleiomyomatosis who developed severe dyspnea and chest pain. Radiograph shows a right basal pneumothorax under tension (depressed hemidiaphragm). Tube thoracostomy produced immediate relief.

Lung disease that is histologically and symptomatically identical to LAM occurs in 1% to 2% of cases of tuberous sclerosis, an inherited disorder with mental retardation, seizures, and tumors of multiple organs. It is noteworthy that, despite the equal sex incidence of tuberous sclerosis, DPLD is found, with rare exception, in women. Lack and associates (1986), among others, reported overlap cases of LAM associated with tumors in other organs but without the other classic features of tuberous sclerosis.

P.1365


Pulmonary Alveolar Proteinosis

In pulmonary alveolar proteinosis (PAP), the alveoli are filled with a granular proteinaceous surfactant material. Recent reviews of PAP have been provided by Goldstein and colleagues (1998) and by Seymour and Presnelli (2002). The latter authors note that only slightly more than 400 cases have been reported. In contrast to most DPLD, macrophage number and function are decreased, leading to the hypothesis that PAP results from impaired phagocytic clearance of alveolar debris. Recent evidence suggests that deficient stimulation of macrophages by granulocyte-macrophage colony-stimulating factor (GM-CSF) may be the underlying mechanism. Impairment of surfactant clearance by blocking autoantibodies to GM-CSF is likely in many acquired cases, whereas congenital PAP appears as a result of surfactant protein abnormalities.

Interstitial inflammation and fibrosis are not prominent in early PAP, but late fibrosis may rarely occur. PAP is usually idiopathic but can be associated with silicosis, aluminum exposure, fungal or parasitic infection, hematologic malignancies, drug reactions, and immune suppression. Men are affected twice as often as women and present between ages 20 and 50 years. Prakash and colleagues (1987) found that dyspnea, cough, fever, chest pain, and hemoptysis were common symptoms and that 29% of patients had clubbing. Despite the alveolar findings, the cough is usually nonproductive. Bilateral perihilar batwing air space opacities without air bronchograms resemble pulmonary edema but lack Kerley B lines. According to Holbert and associates (2001), CT generally demonstrates a geographic pattern but can show a diffuse process. Some degree of ground-glass opacity was seen all cases, interlobar and air space opacities in most, and significant fibrosis in a few. Restrictive pulmonary function is typical. The natural history varies. Opportunistic infection, especially with Nocardia species, is a common problem and is attributed to impaired macrophage function and the ability of the intraalveolar lipoprotein to serve as a culture medium. The diagnosis of PAP has been made on the basis of TBLB, BAL, and occasionally sputum examination, but surgical biopsy is often required. Stains and cultures for opportunistic organisms are essential. Progressive symptoms are treated by flushing the alveoli by whole-lung lavage, using general anesthesia, selective lung ventilation, and rarely cardiopulmonary bypass. The procedure is repeated as needed. Cheng and co-workers (2002) reported serial lobar lavage under topical anesthesia in three patients and suggested that this approach might be warranted in patients at high risk for general anesthesia and in those with milder disease, whose alveolar protein may be removed with smaller fluid volumes. Steroids are of no benefit in PAP. Treatment with GM-CSF appears promising in patients with idiopathic acquired PAP, as reported by Seymour and colleagues (2001) and Kavuru and co-workers (2000).

Diffuse Pulmonary Hemorrhage

Diffuse pulmonary hemorrhage can complicate many disorders, including CVD, vasculitis, LAM, Kaposi's sarcoma, stem cell transplantation, and drug reactions. Diffuse pulmonary hemorrhage is characterized by dyspnea, cough, hemoptysis, anemia, and changing acinar infiltrates. Goodpasture's syndrome, the most common cause of diffuse pulmonary hemorrhage, includes glomerulonephritis and is marked by immunoglobin deposition along alveolar and glomerular basement membranes. Most patients are men. Hemoptysis is the usual presentation, but renal insufficiency may be the first manifestation. The diagnosis is usually established by a typical history, elevated antibodies, and renal indices, or by renal biopsy. Rarely, TBLB samples are sufficient for demonstrating immunoglobins. Treatment includes steroids, plasmapheresis, and immunosuppression. Idiopathic pulmonary hemorrhage, sometimes termed idiopathic pulmonary hemosiderosis, occurs mainly in children, but young adults are occasionally affected, again with males predominating. The clinical and radiographic picture is similar to that of Goodpasture's, but without antibodies or renal involvement. BAL may yield hemosiderin-filled macrophages. As a diagnosis of exclusion, idiopathic pulmonary hemorrhage may require surgical biopsy to exclude other disorders.

Amyloidosis

Amyloidosis may involve the airways (tracheobronchial amyloidosis) or the lung parenchyma. In the latter case, discrete single or multiple nodules (nodular amyloidosis) are seen more often than diffuse infiltrates (diffuse alveolar amyloidosis). The diffuse form is part of a systemic syndrome and results from infiltration of alveolar septa and vessel walls with acellular eosinophilic material (amyloid) generated by immunoglobin light chains and . Symptoms include dyspnea and cough, and sometimes wheezing and hemoptysis. Signs of pulmonary hypertension may be detected. HRCT may show calcification in micronodules not seen by plain film, as described by Graham and colleagues (1992). The diagnosis is most often made by rectal mucosal or subcutaneous biopsy. TBLB has a high diagnostic yield but may be complicated by bleeding as a result of pulmonary hypertension. Although colchicine, cytotoxic agents, and dimethyl sulfoxide have been used in treatment, the usual course is one of progressive respiratory failure and death.

DIFFUSE LUNG DISEASE IN THE IMMUNOCOMPROMISED HOST

Diffuse lung disease (DPLD) is a common problem in immunocompromised people and may result from a broad spectrum of infectious, neoplastic, and inflammatory processes

P.1366


(Table 94-6). Although certain infections in the normal host can present as generalized pneumonias, such as atypical pneumonias and miliary tuberculosis, the immunocompromised person often suffers from organisms that are usually not normally pathogenic that is, opportunistic infection. Disseminated presentations of normally localized processes, as well as multiorgan and multiorganism infection, are more common. Pulmonary infections are discussed in detail in Chapters 86, 87, 88, 89, 90, 91 and 92.

Table 94-6. Diffuse Lung Disease in the Immunocompromised Host

Noninfectious
   Nonspecific interstitial pneumonitis
   Bronchiolitis obliterans
   Lymphocytic interstitial pneumonia
   Drug induced
   Radiation induced
   Kaposi's sarcoma
   Lymphangitic carcinoma
   Pulmonary lymphoma
   Leukemic cell lysis
   Pulmonary alveolar proteinosis
   Diffuse hemorrhage
   Hypersensitivity pneumonitis
   Acute respiratory distress syndrome (ARDS)
   Vasculitis
   Graft-versus-host reaction
   Recurrence of the primary disease, usually neoplasm
Protozoa
   Pneumocystis
   Toxoplasma
Fungi
   Candida
   Aspergillus
   Cryptococcus
   Histoplasma
   Coccidioides
   Blastomyces
   Mucor
Mycobacteria
   Mycobacterium tuberculosis
   Mycobacterium avium intracellulare and other atypicals
Viruses
   Cytomegalovirus
   Herpes
   Varicella
Bacteria
   Community-acquired (often Streptococcus pneumoniae, Haemophilus influenzae)
Nosocomial (e.g., Pseudomonas, Escherichia coli, Klebsiella, Serratia)

Infectious Disorders

Most noninfectious processes occurring in the immunocompromised host were noted in previous sections, including drug-induced and radiation pneumonitis; BOOP-COP in bone marrow and lung transplant recipients; lymphoma, LIP, and Kaposi's sarcoma in AIDS patients; lymphangitic cancer; pulmonary alveolar proteinosis; pulmonary hemorrhage; and recurrence or exacerbation of the primary disorder. In addition, rejection of lung allografts and graft-versus-host reactions in bone marrow recipients may also cause diffuse infiltrates. Tryka and co-workers (1982) reported acute infiltrates in patients treated for leukemia in whom biopsies showed DAD and degenerating blast cells. The authors termed the process leukemic cell lysis pneumonopathy and postulated that enzyme release from leukemic cell lysis was the initiating injury. The systemic syndrome due to rapid cell lysis in this setting is well recognized.

Nonspecific interstitial pneumonitis is encountered frequently in a variety of subsets of immunocompromised hosts, especially in HIV. This histologic pattern is also observed in apparently immunocompetent hosts who may have a collagen vascular disorder or a hypersensitivity reaction. The clinical syndrome is poorly described, but most patients present with cough, dyspnea, and restrictive physiology on lung function testing. CT images typically show ground-glass attenuation, which may be associated with reticular opacities. Histologic patterns vary but include variable degrees of interstitial inflammation, fibrosis, or a combination of both, which does not fit the pattern of other well-described forms of DPLD. In general, the prognosis is superior to that of UIP. Surgical biopsy is usually required for diagnosis. In patients with human immunodeficiency virus (HIV) infection who have been started on highly active antiretroviral therapy (HAART), a pattern of clinical deterioration has been observed that is believed to reflect restoration of inflammatory immune responses toward infectious and noninfectious antigens. Cough, dyspnea, fever, and recurrent pulmonary infiltrates can be seen that may reflect response to previously acquired (potentially treated) organisms. Recognition of this syndrome may obviate invasive diagnostic intervention. Systemic corticosteroids have been used in treatment.

Protozoal Infection

Pneumocystis carinii pneumonia (PCP) is the modern epitome of opportunistic infection. Long recognized in malnourished infants and renal transplant recipients, PCP was not a common problem before the AIDS epidemic. PCP remains a significant pathogen in AIDS patients, although the incidence is declining as a result of improved prophylactic regimens. Extrapulmonary disease is rare. The clinical presentation may be acute, but more often consists of slowly progressive pulmonary and constitutional symptoms. Pulmonary auscultation may be normal. Pneumothorax with prolonged air leak and failure of lung reexpansion often complicate PCP. Although 10% to 20% of patients have normal chest radiographs, the typical appearance is a diffuse, fine reticular, reticulonodular or ground-glass pattern, sometimes with perihilar accentuation. Sandhu and Goodman (1989) detected pneumatoceles on chest radiographs in 10%

P.1367


of cases. Atypical features include focal consolidation, single or multiple nodules, and upper lobe infiltrates simulating reactivation tuberculosis. A higher incidence of the last finding was noted by Chaffey and associates (1990) in patients treated with aerosolized pentamidine and may reflect poor drug distribution to the lung apices. Adenopathy and pleural effusion are rare. Bergin and associates (1990) found that ground-glass opacification was the most frequent HRCT feature. Cystic areas are also identified more often by CT than by plain film radiography. Tuazon (1985) and Woolfenden (1987) and their colleagues documented 95% and 100% positivity rates of gallium scans, respectively. Although gallium uptake is nonspecific, a negative study makes PCP extremely unlikely. NSIP and PCP have many clinical and radiographic features in common.

The classic finding of a foamy intraalveolar exudate containing Pneumocystis organisms is less common than nonspecific histology including normal structures, DAD, mild or intense inflammation, fibrosis, and sometimes granulomas. Uncommon histologic findings include LIP-like infiltrates or PAP. Although definitive diagnosis requires microscopic demonstration of the organism in sputum, BAL, or biopsy samples, empiric therapy based on clinical criteria has become increasingly accepted, as noted by Tu and colleagues (1993). Treatment consists of trimethoprim-sulfamethoxazole, pentamidine, or both, and steroids. Because the recurrence rate is high, chronic prophylaxis is indicated. The potential for concurrent infectious or noninfectious lung disease in patients with PCP must always be kept in mind.

Toxoplasmosis

Encephalitis is the predominant feature of toxoplasmosis in the immunocompromised host, but pulmonary infection may exist with or, rarely, without apparent CNS involvement. According to Catterall and associates (1986), diffuse infiltrates are seen more frequently than are focal lesions. Advanced HIV disease is the most frequent contemporary context in which the diagnosis is considered. Brain biopsy is often necessary for diagnosis, but identification of trophozoites by BAL, TBLB, or surgical biopsy may be preferable in patients with pulmonary involvement. Recently, polymerase chain reaction (PCR) testing of respiratory specimens has been used for diagnosis.

Mycobacterial Infection

M. tuberculosis (MTB) infection may manifest before the diagnosis of AIDS is made or early in the course, with features similar to standard reactivation cases (upper lobe infiltrates and cavities, positive skin tests). Later presentations, when immune dysfunction has progressed, in contrast, are often atypical. Constitutional, neurologic, and gastrointestinal symptoms from disseminated disease may overshadow pulmonary complaints. The purified protein derivative (PPD) is rarely positive at this stage. Radiographic patterns are often diffuse and may be accompanied by adenopathy, but cavitation is uncommon. The organisms may not respond to staining, and classic caseating granulomas may be absent. Three- or four-drug therapy may be necessary. Small and colleagues (1991) found that radiographic deterioration is usually attributed to a different opportunistic infection.

Infection occurs in other immunocompromised hosts as well. The association of MTB and silicosis, noted earlier, may be due to locally impaired defenses. Bone marrow transplant recipients are another high-risk group, with a variable reported incidence of MTB as high as 5%, as noted by Ip and associates (1998).

According to Horsburgh and Selik (1989), only 78 cases of disseminated atypical mycobacterial infection, of which 37 were Mycobacterium avium intracellulare (MAI), were reported in the preAIDS era. Currently, MAI infection is common and was found in the lung in 34% of AIDS cases in an autopsy series by Wallace and Hannah (1988). MAI typically manifests late in the course of AIDS, often in association with other opportunistic infections. Dissemination is the rule. Fever, malaise, and gastrointestinal symptoms are most common. Lung symptoms may be mild or absent. Radiographs show a diffuse reticulonodular pattern but are normal in 20% of patients or occasionally demonstrate focal lesions. The incidence of adenopathy, rarity of cavitation, and absence of classic granulomas are similar in AIDS-associated MAI and MTB cases. MAI organisms, however, usually can be stained. Blood culture is the primary diagnostic modality, with lung biopsy typically reserved for the exclusion of other diffuse infections. Treatment response is poor.

Fungal Infection

Fungal pneumonia in the immunocompromised host is often manifest by diffuse air space or micronodular opacities, believed to result from aerogenous and hematogenous dissemination, respectively. Infection is usually disseminated and occurs most often in patients with profound T-lymphocyte defects who are often also taking steroids. Fungal disease arises from infection with local endemic organisms or from tissue invasion by infectious pathogens. The ubiquitous Candida and Aspergillus species are the most common pathogens overall, but cryptococcosis is the most frequent AIDS-associated fungal infection, occurring in about 10% of cases. Although meningitis is the predominant feature, concurrent, often asymptomatic lung disease is noted in many patients. Histoplasmosis occurs in endemic and nonendemic geographic regions. Stansell (1991) reported pulmonary symptoms at presentation in 16% to 53% of AIDS patients with histoplasmosis. Coccidioidomycosis is noted less often but predominates in endemic

P.1368


areas. Fish and co-workers (1990) reported pulmonary symptoms in three fourths of AIDS-associated cases, but plain radiographic abnormalities in only one third. Reported but less common fungal pneumonias include blastomycosis and mucormycosis. Bigby and associates (1986) noted that diffuse pulmonary mucormycosis typically develops in patients with chemotherapy-induced neutropenia who are receiving broad-spectrum antibiotics. Unfamiliar fungi such as Fusarium and Penicillium species are emerging pathogens in subsets of immunocompromised hosts.

Fungal infection is diagnosed by microscopic or culture evidence of organisms in body fluids or tissues. The source may be skin, lymph nodes, cerebrospinal fluid, sputum, BAL, TBLB, or surgical lung specimens. Serology is rarely of benefit but may prove helpful in coccidioidomycosis. In addition, serum antigenemia is sensitive and specific in disseminated cryptococcal infection, but may not be as valuable in isolated pulmonary involvement. Encouraging results from antigen determination are emerging in aspergillosis and histoplasmosis.

Viral Infection

Cytomegalovirus (CMV) is frequently isolated from respiratory samples in the immunocompromised host, but its pathogenicity is debated. CMV is detected in about 50% of patients with PCP. Wallace and Hannah (1987) found CMV in the lungs of 80% of autopsied AIDS cases, but CMV was a singular process in only two instances. Waxman and colleagues (1997), in contrast, reported nine AIDS patients with symptomatic DPLD in whom CMV pneumonia was the only active pulmonary process. Three of four untreated patients died within 2 weeks, whereas all five treated with antiviral therapy (ganciclovir) improved and were free of lung disease 3 months later. When tested, CMV antigen was elevated in all instances. In contrast to AIDS cases, the combination of antigenemia and BAL isolates is diagnostic of CMV pneumonitis in bone marrow recipients. In AIDS, cytopathologic confirmation and strict exclusion of other pathogens are required for diagnosis. TBLB or surgical biopsy accomplishes the former. CMV also causes retinitis, colitis, esophagitis, and encephalitis. Less often noted are herpes simplex and varicella-zoster pneumonia, the latter usually associated with a skin rash. Respiratory syncytial virus (RSV) is a recently more frequently recognized treatable pathogen in the compromised and the elderly. Other viral pneumonias, such as influenza, adenovirus, and measles, are uncommon.

Radiographs in viral pneumonitis initially show a diffuse micronodular or reticulonodular pattern that often progresses to diffuse air space opacities and consolidation. Histologic features variably include some degree of DAD, viral inclusion bodies, giant cells, and hemorrhage. Diagnosis depends on a typical clinical syndrome, histology, culture of the virus, and DNA probes or staining with specific antibodies.

Bacterial Infection

Although overshadowed by opportunistic infection, bacterial pneumonia accounts for about 4% of DPLD in people with depressed immunity. According to Murray and Mills (1990), many community-acquired bacterial pneumonias are caused by Streptococcus pneumoniae and Haemophilus influenzae, but the spectrum of potential pathogens is extensive. Nosocomial cases are frequently attributable to gram-negative bacteria or Staphylococcus species. The incidence of associated bacteremia, cavitation, and disseminated tissue infection in all types of bacterial pneumonia is higher in the immunocompromised than in the normal host. Kramer and Uttamchandani (1990) reported that Nocardia species infection in transplant recipients tends to be focal, but diffuse disease is seen in other immunocompromised hosts. Legionella species infection may progress from focal to diffuse lung involvement. It afflicts renal transplant recipients as both a community-acquired and nosocomial process, but is uncommon in AIDS. Chastre and colleagues (1987) showed that pulmonary fibrosis associated with legionellosis could progress even after treatment. Despite its prevalence in the general population, Mycoplasma species infection is not a prominent cause of pneumonitis in the setting of impaired immunity.

Diagnostic Approach

Figure 94-10 presents a general approach to the evaluation of DPLD in the immunocompromised patient. It must be stressed that no algorithm is universally applicable. The patient's condition and prognosis, the local experience with diagnostic techniques and epidemiology, and the tempo of the illness are important factors in decision making. In addition, ongoing experience with these diseases and improving noninvasive and minimally invasive diagnostic modalities play an important role in clinical decision making. The specific immune deficiency and the clinical data often narrow the differential diagnosis. For example, an HIV-positive outpatient with compatible symptoms and radiographic findings and a low CD4 is likely to have PCP. Smith and associates (1992) found that an initial episode of PCP could be diagnosed confidently (95% predictive value) on the basis of history, films, absence of prophylaxis, and desaturation with exercise. Leukocytosis, purulent sputum, and a higher CD4 count favor a bacterial process. DPLD developing before chemotherapy in leukemic patients is most often related to bacterial infection or neoplastic infiltration, whereas opportunistic infection is commonly associated with chemotherapy-induced neutropenia in this population. In bone marrow recipients, noninfectious causes of diffuse lung disease predominate during the first month.

In most cases, empiric therapy based on the individual context is initiated after appropriate microbiologic and cytologic investigations. Discontinuation of toxic drugs is essential,

P.1369


as is correction of coagulopathy, especially if pulmonary hemorrhage is a consideration. When noninvasive studies are diagnostic or the response to treatment is good, therapy is continued or appropriately modified. Invasive diagnostic methods are indicated when the patient fails to improve while receiving empiric or specific therapy. In some instances, confirmation of a suspected diagnosis may be warranted despite improvement, as in the HIV-positive patient without prior documentation of an index disease. Bronchoscopy with BAL and preferably also TBLB is usually the initial procedure. BAL alone is most successful in the AIDS population given the high incidence of PCP. In addition, BAL identification of M. tuberculosis or Histoplasma, Coccidioides, Blastomyces, or species Legionella indicates pathogenicity, although concurrent infections may exist. BAL isolation of viruses, atypical mycobacteria, Cryptococcus, Aspergillus, or Candida species, and bacteria, in contrast, usually requires correlative tissue data for a secure diagnosis. Yu and associates (1986), however, believe that in persistently febrile patients with leukemia or neutropenia, the presence of Aspergillus fumigatus or Aspergillus flavus in respiratory secretions often denotes invasive infection. In a study of a large mixed immunocompromised population evaluated by BAL, Pisani and Wright (1992) documented a diagnostic yield of 39%, with a sensitivity of 82% and specificity of 53%. The addition of TBLB to BAL may improve the yield by demonstrating organisms within lung tissue or by providing specimens that document a noninfectious process. Although the complication rate of TBLB is low (at most, bleeding in 3% and pneumothorax in 5%), many clinicians believe that the yield of BAL is sufficiently high that TBLB need not be routine. Others, including Cazzadori and colleagues (1995), advocate TBLB in all bronchoscopies for DPLD in immunosuppressed patients. These authors found an overall diagnostic rate of 67% for TBLB versus 36% for BAL, with TBLB superior in all categories studied: HIV cases, hematologic malignancies, and renal transplant recipients.

Fig. 94-10. Algorithm for dealing with diffuse lung infiltrates in the immunocompromised host. Dx, diagnostic; Non dx, nondiagnostic; CSF, cerebrospinal fluid; BAL, bronchoalveolar lavage; TBLB, transbronchial lung biopsy.

Surgical lung biopsy offers the greatest sensitivity and specificity but is less often required than in the past, especially in the AIDS population. In the series of Bonfils-Roberts and associates (1990), 18 of 64 patients with AIDS in 1983 underwent biopsy, compared with only 2 of 302 in 1987. Surgical biopsy should be considered as the primary

P.1370


invasive procedure in patients with rapidly progressive disease in whom TBLB is relatively or absolutely contraindicated (ventilator dependence with high airway pressures, coagulopathy) and the clinical suspicion is of a disorder unlikely to be diagnosed by BAL alone and not covered by simple empiric therapy. Occasional patients may be deemed too ill to tolerate bronchoscopy at all. In most cases, however, surgical biopsy is performed after a nondiagnostic bronchoscopy or in patients who fail to improve with therapy dictated by the bronchoscopic findings.

Although surgical biopsy almost always produces a definitive diagnosis, the reported benefit in altering therapy and outcome is variable. Studies by Hiatt (1982), Haverkos (1983), and Potter (1985) and their co-workers, among others, found that information obtained by open biopsy did not translate into improved survival in a variety of non-AIDS populations. In contrast, Cheson (1985), Walker (1989), and Robbins (1992) and their associates found that open biopsy was helpful in mixed non-AIDS groups, as did Catterall and associates (1989) in patients with Hodgkin's disease and White and co-workers (2000) in a group of patients with hematologic malignancies. The data in AIDS cases also are contradictory. Pass and associates (1986) concluded that open biopsy is helpful when TBLB is nondiagnostic or contraindicated and when patients fail to improve despite treatment for diagnoses made bronchoscopically. Fitzgerald and associates (1987) likewise recommended open biopsy when bronchoscopy was nonspecific, but found no benefit for patients who failed treatment after diagnostic bronchoscopy. Bonfils-Roberts and colleagues (1990) reported a beneficial therapeutic change in only 1 of 66 AIDS patients who underwent biopsy.

The reported mortality rate of open biopsy is variable. Cheson and associates (1985) had no deaths in 87 procedures, and Robbins and colleagues (1992) noted only one death (1.4%) in 74 cases, despite coagulopathy in many patients. In contrast, Walker and colleagues (1989) reported a mortality rate of 9%, and Potter and co-workers (1985) implicated surgical problems in the deaths of 3 of 14 biopsy patients (21%). Stover and Kanner (1996) suggest that VATS biopsy in their series of immunosuppressed cancer patients may be less morbid than standard open-lung biopsy. A randomized trial of open-lung versus VATS biopsy in immunocompetent patients conducted by Miller and associates (2000) showed no difference between the two approaches.

We believe that open or thoracoscopic surgical biopsy can be accomplished safely in most cases and should be performed when a diagnosis eludes less invasive techniques. Surgical biopsy should also be considered in some patients who fail to respond to treatment. This bias is based on the frequent coexistence of more than one process, the limitations of BAL TBLB in the diagnosis of many noninfectious lesions, and the occasional gratifying response to therapy of disorders such as Kaposi's sarcoma, diffuse lymphoma, and CMV pneumonia. Surgical biopsy, however, is not likely to alter outcome in patients with respiratory failure despite aggressive specific or empiric treatment. The benefit to individual patients when surgical biopsy yields a specific treatable diagnosis, coupled with current low complication rates, ensures that this approach will retain a place in the management of subgroups of patients.

REFERENCES

Agusti C, et al: Interstitial pulmonary fibrosis with and without associated collagen vascular disease: results of a two-year follow-up. Thorax 47: 1035, 1992.

Al-Jarad N, et al: High resolution computed tomographic assessment of asbestosis and cryptogenic fibrosing alveolitis: a comparative study. Thorax 47:645, 1992.

Allen JN, Davis WB: Eosinophilic lung diseases. Am J Respir Crit Care Med 150:1423, 1994.

American Thoracic Society: Idiopathic pulmonary fibrosis: diagnosis and treatment. International consensus statement. American Thoracic Society (ATS), and the European Respiratory Society (ERS). Am J Respir Crit Care Med 161:646, 2000.

American Thoracic Society; European Respiratory Society: ATS/ERS international multidisciplinary consensus classification of the idiopathic interstitial pneumonias. Am J Respir Crit Care Med 165:277, 2002.

Baughman RP, Lower EE, du Bois RM: Sarcoidosis. Lancet 361:1111, 2003.

Bergin CJ, et al: Pneumocystis carinii pneumonia: CT and HRCT observations. J Comput Assist Tomogr 14:756, 1990.

Bigby TD, et al: Clinical spectrum of pulmonary mucormycosis. Chest 89:435, 1986.

Bjoraker JA, et al: Prognostic significance of histopathologic subsets in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 157:199, 1998.

Blewett CJ, et al: Open lung biopsy as an outpatient procedure. Ann Thorac Surg 71:1113, 2001.

Bonfils-Roberts EA, Nickodem A, Nealon TF Jr: Retrospective analysis of the efficacy of open lung biopsy in acquired immunodeficiency syndrome. Ann Thorac Surg 49:115, 1990.

Brauner MW, et al: Pulmonary histiocytosis X: evaluation with high-resolution CT. Radiology 172:255, 1989.

Cadranel J, Wislez M, Antoine M: Primary pulmonary lymphoma. Eur Respir J 20:750, 2002.

Carrington CB, Liebow AA: Limited forms of angiitis and granulomatosis of Wegener's type. Am J Med 41:497, 1966.

Carrington CB, et al: Natural history and treated course of usual and desquamative interstitial pneumonia. N Engl J Med 298:801, 1978.

Catterall JR, Hofflin JM, Remington JS: Pulmonary toxoplasmosis. Am Rev Respir Dis 133:704, 1986.

Catterall JR, et al: Open lung biopsy in patients with Hodgkin's disease and pulmonary infiltrates. Am Rev Respir Dis 139:1274, 1989.

Cazzadori A, et al: Transbronchial biopsy in the diagnosis of pulmonary infiltrates in immunocompromised patients. Chest 107:101, 1995.

Chaffey MH, et al: Radiographic distribution of Pneumocystis carinii pneumonia in patients with AIDS treated with prophylactic inhaled pentamidine. Radiology 175:715, 1990.

Chandler PW, et al: Radiographic manifestations of bronchiolitis obliterans with organizing pneumonia versus usual interstitial pneumonia. AJR Am J Roentgenol 147:899, 1986.

Chastre J, et al: Pulmonary fibrosis following pneumonia due to acute Legionnaire's disease: clinical, ultrastructural, and immunofluorescent study. Chest 91:57, 1987.

Chechani V, Landrenau RJ, Shaikh SS: Open lung biopsy for diffuse infiltrative lung disease. Ann Thorac Surg 54:296, 1992.

Cheng SL, et al: Pulmonary alveolar proteinosis: treatment by bronchofiberscopic lobar lavage. Chest 122:1480, 2002.

Cheson BD, et al: Value of open-lung biopsy in 87 immunocompromised patients with pulmonary infiltrates. Cancer 55:453, 1985

Chu SC, et al: Comprehensive evaluation of 35 patients with lymphangioleiomyomatosis. Chest 115:1041, 1999.

Colby TV, Swensen SJ: Anatomic distribution and histopathologic patterns in diffuse lung disease: correlation with HRCT. J Thorac Imaging 11:1, 1996.

P.1371


Coleman A, Colby TV: Histologic diagnosis of extrinsic allergic alveolitis. Am J Surg Pathol 12:514, 1988.

Cooper JAD, White DA, Matthay RA: Drug-induced pulmonary disease. I. Cytotoxic drugs. Am Rev Respir Dis 33:321, 1986a.

Cooper JAD, White DA, Matthay RA: Drug-induced pulmonary disease. II. Noncytotoxic drugs. Am Rev Respir Dis 133:488, 1986b.

Cordier JF, et al: Pulmonary Wegener's granulomatosis: a clinical and imaging study of 77 cases. Chest 97:906, 1990.

Costabel U, et al: T-lymphocytosis in bronchoalveolar lavage fluid of hypersensitivity pneumonitis: changes in profile of T-cell subsets during the course of disease. Chest 85:514, 1984.

Cottin V, et al: Nonspecific interstitial pneumonia: individualization of a clinicopathologic entity in a series of twelve patients. Am J Respir Crit Care Med 158:1286, 1998.

Crouch E: Pathobiology of pulmonary fibrosis. Am J Physiol 259:L159, 1990.

Crystal E, et al: Long-term amiodarone therapy and the risk of complications after cardiac surgery. J Thorac Cardiovasc Surg 125:633, 2003.

Crystal RG, et al: Idiopathic pulmonary fibrosis: clinical, histologic, radiographic, physiologic, scintigraphic, cytologic and biochemical aspects. Ann Intern Med 85: 769, 1976.

Davison AG, et al: Cryptogenic organizing pneumonitis. Q J Med 52:382 394, 1983.

Dimopoulou I, et al: Low-dose amiodarone-related complications after cardiac surgery. J Thorac Cardiovasc Surg 114:31, 1997.

Douglas WW, Ryu JH, Schroeder DR: Idiopathic pulmonary fibrosis: impact of oxygen and colchicine, prednisone, or no therapy on survival. Am J Respir Crit Care Med 161:1172, 2000.

Douglas WW, et al: Colchicine versus prednisone in the treatment of idiopathic pulmonary fibrosis. Am J Repir Crit Care Med 158:220, 1998.

Epler GR, et al: Normal chest roentgenograms in chronic diffuse infiltrative lung disease. N Engl J Med 298:934, 1978.

Epler GR, et al: Bronchiolitis obliterans organizing pneumonia. N Engl J Med 312:152, 1985.

Fathi M, et al: Interstitial lung disease, a common manifestation of newly diagnosed polymyositis and dermatomyositis. Ann Rheum Dis 63:297, 2004.

Fauci AS, et al: Lymphomatoid granulomatosis: prospective clinical and therapeutic experience over 10 years. N Engl J Med 306: 68, 1982.

Felson B: A new look at pattern recognition of diffuse pulmonary diseases. AJR Am J Roentgenol 133:183, 1979.

Fish DG, et al: Coccidioidomycosis during human immunodeficiency virus infection: a review of 77 patients. Medicine 69:384, 1990.

Fitzgerald W, et al: The role of open lung biopsy in patients with the acquired immunodeficiency syndrome. Chest 91:659, 1987.

Frank ST, et al: Pulmonary dysfunction in rheumatoid disease. Chest 63: 27, 1973.

Fraser RG, et al: Diagnosis of Diseases of the Chest. 3rd Ed. Philadelphia: WB Saunders, 1988.

Frazier AR, Miller RD: Interstitial pneumonitis in association with polymyositis and dermatomyositis. Chest 65:403, 1974.

Fujii M, et al: Interstitial lung disease in rheumatoid arthritis: assessment with high-resolution computed tomography. J Thorac Imaging 8:54, 1993.

Fulmer JD, Kaltreider HB: The pulmonary vasculitides. Chest 82:615, 1983.

Gaensler EA, Carrington CB: Peripheral opacities in chronic eosinophilic pneumonia: the photographic negative of pulmonary edema. AJR Am J Roentgenol 128:1, 1977.

Gaensler EA, Carrington CB: Open biopsy for chronic diffuse infiltrative lung disease: clinical, roentgenographic and physiologic correlations in 502 patients. Ann Thorac Surg 30:411, 1980.

Gill PS, et al: Pulmonary Kaposi's sarcoma: clinical findings and results of therapy. Am J Med 87:57, 1989.

Goldstein LS, et al: Pulmonary alveolar proteinosis: clinical features and outcomes. Chest 114:1357, 1998.

Graham CM, et al: High-resolution CT appearance of diffuse alveolar septal amyloidosis. AJR Am J Roentgenol 158: 265, 1992.

Hakala M: Poor prognosis in patients with rheumatoid arthritis hospitalized for interstitial lung fibrosis. Chest 93:114, 1988.

Hamman L, Rich AR: Fulminating diffuse interstitial fibrosis of the lungs. Trans Am Clin Climatol Assoc 51:154, 1935.

Harrison DJ, et al: Antibodies to neutrophil-cytoplasmic antigens in Wegener's granulomatosis and other conditions. Thorax 44:373, 1989.

Haupt HM, Moore GW, Hutchins GM: The lung in systemic lupus erythematosus: analysis of the pathologic changes in 120 patients. Am J Med 71:791, 1981.

Haverkos H, et al: Diagnosis of pneumonitis in immunocompromised patients by open lung biopsy. Cancer 52:1093, 1983.

Herbert A, et al: Lymphocytic interstitial pneumonia identified as lymphoma of mucosa-associated lymphoid tissue. J Pathol 146:129, 1985.

Hiatt J, et al: The value of open lung biopsy in the immunocompromised patient. Surgery 92:285, 1982.

Holbert JM, et al: CT features of pulmonary alveolar proteinosis. AJR Am J Roentgenol 176:1287, 2001.

Horsburgh CR, Selik RM: The epidemiology of nontuberculous mycobacterial infection in the acquired immunodeficiency syndrome (AIDS). Am Rev Respir Dis 139:4, 1989.

Ikezoe J, et al: CT appearance of acute radiation-induced injury in the lung. AJR Am J Roentgenol 150:765, 1988.

Im JG, et al: Lobar bronchioloalveolar carcinoma: angiogram sign on CT scans. Radiology 176:749, 1990.

Ip MSM, et al: Risk factors for pulmonary tuberculosis in bone marrow transplant recipients. Am J Respir Crit Care Med 158:1173, 1998.

Jahaveri S, Sicilian L: Lung function, breathing pattern, and gas exchange in interstitial lung disease. Thorax 47:93, 1992.

James DG, Siltzbach LE, Turiaf A: A worldwide review of sarcoidosis. Ann N Y Acad Sci 278:321, 1976.

Jederlinic PJ, Sicilian L, Gaensler EA: Chronic eosinophilic pneumonia: a report of 19 cases and a review of the literature. Medicine 67:154, 1988.

Jennette JC, Falk RJ: Small-vessel vasculitis. N Engl J Med 337:1512, 1997.

Johnson MA, et al: Randomised controlled trial comparing prednisolone alone with cyclophosphamide and low dose prednisolone in combination in cryptogenic fibrosing alveolitis. Thorax 44:280, 1989.

Katabami M, et al: Pneumoconiosis-related lung cancers: preferential occurrence from diffuse interstitial fibrosis-type pneumoconiosis. Am J Respir Crit Care Med 162:295, 2000.

Katzenstein ALA, Bloor CM, Liebow AA: Diffuse alveolar damage: the role of oxygen, shock, and related factors. Am J Pathol 58:210, 1976.

Katzenstein ALA, Fiorelli RF: Nonspecific interstitial pneumonia/fibrosis: histologic features and clinical significance. Am J Surg Pathol 18:136, 1994

Katzenstein ALA, Myers JL, Mazur MT: Acute interstitial pneumonia: a clinicopathological, ultrastructural, and cell kinetic study. Am J Surg Pathol 10:256, 1986.

Kavuru MS, et al: Exogenous granulocyte-macrophage colony-stimulating factor administration for pulmonary alveolar proteinosis. Am J Respir Crit Care Med 161:1143, 2000.

Keicho N, et al: Detection of lymphomatous involvement of the lung by bronchoalveolar lavage: application of immunophenotypic and gene rearrangement analysis. Chest 105:458, 1994.

King TE: Diagnostic advances in idiopathic pulmonary fibrosis. Chest 100:238, 1991.

King TE, Mortenson RL: Cryptogenic organizing pneumonitis: the North American experience. Chest 102[Suppl]:8, 1992.

King TE, et al: Predicting survival in idiopathic pulmonary fibrosis: scoring system and survival model. Am J Respir Crit Care Med 164:1171, 2001.

Knoell D, et al: Churg-Strauss syndrome associated with zafirlukast. Chest 114:332, 1998.

Kramer MR, Uttamchandani RB: The radiographic appearance of pulmonary nocardiosis associated with AIDS. Chest 98:382, 1990.

Lack EE, et al: Pulmonary and extrapulmonary lymphangioleiomyomatosis: report of a case with bilateral angiomyolipomas, multifocal lymphangioleiomyomatosis, and a glial polyp of the endocervix. Am J Surg Pathol 10:650, 1986.

Leatherman JW: The lung in systemic vasculitis. Semin Respir Infect 3: 274, 1988.

Lee JS, et al: Fibrosing alveolitis: prognostic implication of ground-glass attenuation on high-resolution CT. Radiology 184:451, 1992.

Lee KS, et al: Cryptogenic organizing pneumonia: CT findings in 43 patients. AJR Am J Roentgenol 162:543, 1994.

Liebow AA: New concepts and entities in pulmonary diseases. In Liebow AA, Smith DE (eds): The Lung. Baltimore: Williams & Wilkins, 1968.

Liebow AA, Carrington CB, Friedman PJ: Lymphomatoid granulomatosis. Hum Pathol 3:457, 1972.

Locke CB: Rheumatoid lung. Clin Radiol 14:43, 1963.

P.1372


Lynch JP, McCune WJ: Immunosuppressive and cytotoxic pharmacotherapy for pulmonary disorders. Am J Respir Crit Care Med 155:395, 1997.

Maher J, Daly PA: Severe bleomycin lung toxicity: reversal with high dose corticosteroids. Thorax 48:92, 1993.

Martin WJ, Rosenow EC: Amiodarone pulmonary toxicity: recognition and pathogenesis. Chest 93:1067, 1988.

Mathiesen JR, et al: Chronic diffuse infiltrative lung disease: comparison of diagnostic accuracy of CT and chest radiography. Radiology 171:111, 1989.

Matthay RA, et al: Pulmonary manifestations of systemic lupus erythematosus: review of twelve cases of acute lupus pneumonitis. Medicine 54:379, 1974.

Mayo JR, et al: Chronic eosinophilic pneumonia: CT findings in six cases. AJR Am J Roentgenol 153:727, 1989.

McLoud TC, Carrington CB, Gaensler EA: Diffuse infiltrative lung disease: a new scheme for description. Radiology 149:353, 1983.

Miller JD, et al: A randomized, controlled trial comparing thoracoscopy and limited thoracotomy for lung biopsy in interstitial lung disease. Ann Thorac Surg 70:1647, 2000.

Miller RR, et al: Lingular and right middle lobe biopsy in the assessment of diffuse lung disease. Ann Thorac Surg 44:269, 1987.

Monaghan H, et al: Prognostic implications of histologic patterns in multiple surgical lung biopsies from patients with idiopathic interstitial pneumonias. Chest 125:522, 2004.

Moran TJ, Totten RS: Lymphoid interstitial pneumonia with dysproteinemia: report of two cases with plasma cell predominance. Am J Clin Pathol 54:747, 1970.

Movsas B, Raffin TA, Epstein AH: Pulmonary radiation injury. Chest 111: 1061, 1997.

M ller NL, Staples CA, Miller RR: Bronchiolitis obliterans organizing pneumonia: CT features in 14 patients. AJR Am J Roentgenol 154:983, 1990.

Murray JF, Mills J: Pulmonary infectious complications of human immunodeficiency virus infection. Part 1. Am Rev Respir Dis 141:1356, 1990.

Myers JL: Lymphomatoid granulomatosis: past, present, future? Mayo Clin Proc 65:274, 1990.

Naidich DP, et al: Kaposi's sarcoma: CT-radiographic correlation. Chest 96:723, 1989.

Nalos PC, et al: Life-threatening postoperative pulmonary complications in patients with previous amiodarone pulmonary toxicity undergoing cardiothoracic operations. J Thorac Cardiovasc Surg 93:904, 1987.

Naughton M, Fahy J, FitzGerald MX: Chronic eosinophilic pneumonia: a long-term follow-up of 12 patients. Chest 103:162, 1993.

Newman LS, Rose CS, Maier LA: Sarcoidosis. N Engl J Med 336:1224 1234, 1997.

Nishimura K, Itoh H: High-resolution computed tomographic features of bronchiolitis obliterans organizing pneumonia. Chest 102 [Suppl]:27, 1992a.

Nishimura K, et al: Usual interstitial pneumonia: histologic correlation with high-resolution CT. Radiology 182:337, 1992b.

Ohori NP, et al: Giant cell interstitial pneumonia and hard metal pneumoconiosis: a clinicopathologic study of four cases and review of the literature. Am J Surg Pathol 13:581, 1989.

Oldham SAA, et al: HIV-associated lymphocytic interstitial pneumonia: radiologic manifestations and pathologic correlation. Radiology 170:83, 1989.

Orens JB, et al: The sensitivity of high-resolution CT in detecting idiopathic pulmonary fibrosis proved by open lung biopsy: a prospective study. Chest 108:109, 1995.

Ott MC, et al: Pulmonary toxicity in patients receiving low-dose amiodarone. Chest 123:646, 2003.

Owens GR, Follansbee WP: Cardiopulmonary manifestations of systemic sclerosis. Chest 91:118, 1987.

Palosaari DE, Colby TV: Bronchiolocentric chronic lymphocytic leukemia. Cancer 58:1695 1698, 1986.

Parkes WR: Occupational Lung Disorders. 3rd Ed. Oxford: Butterworth-Heinemann, 1994.

Pass H, et al: Indications for and diagnostic efficacy of open-lung biopsy in the patient with acquired immunodeficiency syndrome (AIDS). Ann Thorac Surg 41:307, 1986.

Pavlakis N, et al: Fatal pulmonary toxicity resulting from treatment with gemcitabine. Cancer 80:286, 1997.

Patchefsky AS, et al: Desquamative interstitial pneumonia: relationship to interstitial fibrosis. Thorax 28:680, 1973.

Pietra GG: Pathologic mechanisms of drug-induced lung disorders. J Thorac Imaging 6:1, 1991.

Pisani RJ, DeRemee RA: Clinical implications of the histologic diagnosis of pulmonary lymphomatoid granulomatosis. Mayo Clin Proc 65:151, 1990.

Pisani RJ, Wright AJ: Clinical utility of bronchoalveolar lavage in immunocompromised hosts. Mayo Clin Proc 67:221, 1992.

Ponn RB, Silverman H, Federico JA: Outpatient chest tube management. Ann Thorac Surg 64:1437, 1997.

Potter D, et al: Prospective randomized study of open lung biopsy versus empirical antibiotic therapy for acute pneumonitis in non-neutropenic cancer patients. Ann Thorac Surg 40:422, 1985.

Prakash UBS, Luthra HS, Divertie MB: Intrathoracic manifestations of mixed connective tissue disease. Mayo Clin Proc 60:813, 1985.

Prakash UBS, et al: Pulmonary alveolar phospholipoproteinosis: experience with 34 cases and a review. Mayo Clin Proc 62:499, 1987.

Purdy LJ, et al: Pulmonary Kaposi's sarcoma: premortem histologic diagnosis. Am J Surg Pathol 10:301, 1986.

Raghu G: Idiopathic pulmonary fibrosis: a rational clinical approach. Chest 92:148, 1987.

Raghu G: Interstitial lung disease: a diagnostic approach. Are CT scan and lung biopsy indicated in every patient? Am J Respir Crit Care Med 151:909 914, 1995.

Raghu G, et al: Azathioprine combined with prednisone in the treatment of idiopathic pulmonary fibrosis: a prospective, double-blind, randomized, placebo-controlled trial. Am J Respir Crit Care Med 144:291. 1991.

Raghu G, et al: Treatment of idiopathic pulmonary fibrosis with a new antifibrotic agent, pirfenidone: results of a prospective, open-label Phase II study. Am J Respir Crit Care Med 159:1061, 1999.

Raghu G, et al: A placebo-controlled trial of interferon gamma-1b in patients with idiopathic pulmonary fibrosis. N Engl J Med 350:125, 2004.

Reynolds HY: Diagnostic and management strategies for diffuse interstitial lung disease. Chest 113:192, 1998.

Robbins BE, et al: Diagnosis of acute diffuse pulmonary infiltrates in immunocompromised patients by open biopsy of the lung. Surg Gynecol Obstet 175:8, 1992.

Rochester CL, Elias JA: Cytokines and cytokine networks in the pathogenesis of interstitial and fibrotic lung disorders. Semin Respir Med 14:389, 1993.

Roseman KD, et al: Silicosis in the 1990's. Chest 111:779, 1997.

Rosenberg DM, et al: Functional correlates of lung involvement in Wegener's granulomatosis. Am J Med 69:387, 1980.

Rosenow EC, et al: Drug-induced pulmonary disease: an update. Chest 102:239, 1992.

Ryu JH, et al: Chylothorax in lymphangioleiomyomatosis. Chest 123:623, 2003.

Salmeron G, Greenberg SD, Lidsky MD: Polymyositis and diffuse interstitial lung disease: a review of the pulmonary histopathologic findings. Arch Intern Med 141:1005, 1981.

Salvaggio JE, deShazo RD: Pathogenesis of hypersensitivity pneumonitis. Chest 89[Suppl]:190S, 1986.

Sandhu JS, Goodman PC: Pulmonary cysts associated with Pneumocystis carinii pneumonia in patients with AIDS. Radiology 173:33, 1989.

Scadding JG, Hinson KFW: Diffuse fibrosing alveolitis (diffuse interstitial fibrosis of the lungs). Thorax 22:291, 1967.

Selman M, King TE, Pardo A: Idiopathic pulmonary fibrosis: prevailing and evolving hypotheses about its pathogenesis and implications for therapy. Ann Intern Med 134:136, 2001.

Seymour JF, Presnelli JJ: Pulmonary alveolar proteinosis: progress in the first 44 years. Am J Respir Crit Care Med 166:215, 2002.

Seymour JF, et al: Therapeutic efficacy of granulocyte-macrophage colony-stimulating factor in patients with idiopathic acquired alveolar proteinosis. Am J Respir Crit Care Med 163:524, 2001.

Small PM, et al: Course and outcome of tuberculosis in patients who have or subsequently developed AIDS. N Engl J Med 324:289, 1991.

Smith DE, et al: Diagnosis of Pneumocystis carinii pneumonia in HIV antibody-positive patients by simple outpatient assessments. Thorax 47: 1005, 1992.

Stack BHR, Choo-Kang YF, Heard BE: The prognosis of cryptogenic fibrosing alveolitis. Thorax 27:535, 1972.

Stansell JD: Fungal disease in HIV-infected persons: cryptococcosis, histoplasmosis, and coccidioidomycosis. J Thorac Imaging 6:28, 1991.

P.1373


Staples CA, et al: High resolution computed tomography and lung function in asbestos-exposed workers with normal chest radiographs. Am Rev Respir Dis 139:1502, 1989.

Stein MG, et al: Pulmonary lymphangitic spread of carcinoma: appearance on CT scans. Radiology 162:371, 1987.

Stover DE, Kaner RJ: Pulmonary complications in cancer patients. CA Cancer J Clin 46:303, 1996.

Strollo DC: Imaging of idiopathic interstitial lung diseases: concepts and conundrums. Am J Respir Cell Mol Biol 29[Suppl 3]:10, 2003.

Suffredini AF, et al: Nonspecific interstitial pneumonitis: a common cause of pulmonary disease in the acquired immune deficiency syndrome. Ann Intern Med 107:7, 1987.

Sullivan EJ: Lymphangioleiomyomatosis: a review. Chest 114:1689, 1998.

Sullivan WD, et al: A prospective evaluation emphasizing pulmonary involvement in patients with mixed connective tissue disease. Medicine 63:92, 1984.

Swigris JJ, et al: Lymphoid interstitial pneumonia. Chest 122:2150, 2002.

Tazelaar HD, et al: Interstitial lung disease in polymyositis and dermatomyositis. Am Rev Respir Dis 141:727, 1990.

Thomas D, Hunninghake GW: Current concepts of the pathogenesis of sarcoidosis. Am Rev Respir Dis 135:747, 1987.

Thrasher DR, Briggs DD: Pulmonary sarcoidosis. Clin Chest Med 3:537, 1982.

Tobin MJ: Tuberculosis, lung infections, and interstitial lung disease in AJRCCM 2000. Am J Respir Crit Care Med 164:1774, 2001.

Tomioka H, King TE: Gold-induced pulmonary disease: clinical features, outcome and differentiation from rheumatoid lung disease. Am J Respir Crit Care Med 155:1011, 1997.

Tryka AF, Godleski JJ, Fanta CH: Leukemic cell pneumonopathy: a complication of treated myeloblastic leukemia. Cancer 50:2763, 1982.

Tu JV, Biem J, Detsky AS: Bronchoscopy versus empirical therapy in HIV-infected patients with presumptive Pneumocystis carinii pneumonia: a decision analysis. Am Rev Respir Dis 148:370, 1993.

Tuazon CU, et al: Utility of gallium67 scintigraphy and bronchial washings in the diagnosis and treatment of Pneumocystis carinii pneumonia in patients with the acquired immune deficiency syndrome. Am Rev Respir Dis 132:1087, 1985.

Tubbs RR, et al: Desquamative interstitial pneumonitis. Chest 72:159, 1977.

Urban T, et al: Pulmonary lymphangiomyomatosis: follow-up and long-term outcome with antiestrogen therapy. A report of eight cases. Chest 102:472, 1992.

Vander Els NJ, Miller V: Successful treatment of gemcitabine toxicity with a brief course of oral corticosteroid therapy. Chest 114:1779, 1998.

Walker WA, et al: Does open lung biopsy affect treatment in patients with diffuse pulmonary infiltrates? J Thorac Cardiovasc Surg 97:534, 1989.

Wallace JM, Hannah JB: Cytomegalovirus pneumonitis in patients with AIDS: findings in an autopsy series. Chest 92:198, 1987.

Wallace JM, Hannah JB: Mycobacterium avium complex infection in patients with acquired immunodeficiency syndrome: a clinicopathologic study. Chest 93:926, 1988.

Watters LC, et al: A clinical, radiologic and physiologic scoring system for the longitudinal assessment of patients with idiopathic pulmonary fibrosis. Am Rev Respir Dis 133:97, 1986.

Waxman AB, et al: Cytomegalovirus as a primary pulmonary pathogen in AIDS. Chest 111:128, 1997.

Wechsler ME, et al: Churg-Strauss syndrome in patients receiving montelukast as treatment for asthma. Chest 117:708, 2000.

Wetstein L: Sensitivity and specificity of lingular segmental biopsies of the lung. Chest 90: 383, 1986.

White DA, Matthay RA: Noninfectious pulmonary complications of infection with the human immunodeficiency virus. Am Rev Respir Dis 140: 1763, 1989.

White DA, Wong PW, Downey R: The utility of open lung biopsy in patients with hematologic malignancies. Am J Respir Crit Care Med 161: 723, 2000.

Wholey MH, Good CA, McDonald JR: Disseminated indeterminate pulmonary disease: value of lung biopsy. Radiology 71:651, 1958.

Winterbauer RH: The treatment of idiopathic pulmonary fibrosis. Chest 100:233, 1991.

Woolfenden JM, et al: Acquired immunodeficiency syndrome: Ga-67 citrate imaging. Radiology 162:383, 1987.

Yoshioka R, et al: Pulmonary complications in patients with adult T-cell leukemia. Cancer 55:2491, 1985.

Yousem SA, Colby TV, Carrington CB: Lung biopsy in rheumatoid arthritis. Am Rev Respir Dis 131:770, 1985.

Yousem SA, Colby TV, Gaensler EA: Respiratory bronchiolitis-associated interstitial lung disease and its relationship to desquamative interstitial pneumonia. Mayo Clin Proc 64:1373, 1989.

Yu VL, Muder RR, Poorsattar A: Significance of isolation of aspergillus from the respiratory tract in diagnosis of invasive pulmonary aspergillosis. Am J Med 81:249, 1986.

Ziesche R, et al: A preliminary study of long-term treatment with interferon gamma-1b and low-dose prednisolone in patients with idiopathic pulmonary fibrosis. N Engl J Med 341:1264, 1999.



General Thoracic Surgery. Two Volume Set. 6th Edition
General Thoracic Surgery (General Thoracic Surgery (Shields)) [2 VOLUME SET]
ISBN: 0781779820
EAN: 2147483647
Year: 2004
Pages: 203

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net