Introduction and Definitions
Pulmonary vasculitis is a general term that encompasses a wide variety of individual disease entities, all of which share a unifying finding of inflammation and destruction of blood vessels within the lung. More specifically, these entities are characterized pathologically by the presence of a variety of types of cellular infiltration within vessel walls, resulting in vessel destruction, and ultimately tissue necrosis. The clinical features of a specific disorder are determined by the site, size, and type of vessel involved as well as the relative amounts of vessel inflammation, destruction, and tissue necrosis. While simple to define, the recognition, diagnosis, and management of the pulmonary vasculitides are among the most demanding challenges in medicine. Protean in their presentation and variable in their clinical evolution, their signs and symptoms fully overlap with infection, adverse medication reaction, connective tissue disease, and malignancy. Even in patients with a known vasculitis, separating active disease from complicating infection, drug toxicity, or some combination thereof, will challenge the most skilled physician. Amplifying the problem, these diseases are often deadly; even when appropriately treated, the long-term survival among patients with antineutrophil cytoplasmic antibody positive (ANCA)- associated vasculitis (AAV) is considerably less than for the general population—88% survival at 1 year and 78% survival at 5 years, or a relative mortality risk of 2.6. Despite these hurdles, the astute clinician can make the diagnosis, initiate and manage therapy, and minimize complications by keeping some broad concepts in mind.
Classification
The major clinical benefit of classifying the vasculitides is to provide a framework that allows for an appreciation of the presenting features of disease. The most current broadly accepted system is the 2012 Revised International Chapel Hill Consensus Conference nomenclature, which is based on clinicopathologic presentations rather than etiology or disease mechanism ( Table 60-1 ). It is important to recall that this classification system cannot be used to inform diagnosis or management. The diagnosis of vasculitis relies upon the bedside clinician recognizing patterns of disease made up of specific clinical, laboratory, imaging, and pathologic features. There are no strict classification criteria or clinical diagnostic guidelines. It is up to the clinician to determine whether or not the preponderance of the data supports the diagnosis of a pulmonary vasculitis.
| PRIMARY IDIOPATHIC |
| Small Vessel |
|
| Medium Vessel |
|
| Large Vessel |
|
| PRIMARY IMMUNE COMPLEX MEDIATED |
|
| SECONDARY VASCULITIS |
|
Nevertheless, there are still useful paradigms by which to organize one’s diagnostic approach. A commonly used scheme uses vessel size (large, medium, and small). The large vessels are made up of the aorta, its largest branches (e.g., carotid, cerebral, iliac, subclavian, and femoral vessels) and the main pulmonary artery. The medium-sized vessels are made up of the main visceral arteries (e.g., renal, hepatic, coronary, and mesenteric vessels), whereas the small vessels are made up of arterioles, capillaries, and venules. There can be overlap, because the small and large vessel vasculitides sometimes involve medium-sized arteries, but large and medium vessel vasculitides generally do not involve vessels smaller than arteries.
A second classification scheme uses ANCA. The identification of these antibodies in the 1980s revolutionized thinking regarding diagnosis and pathogenesis (see later). The vasculitides that are most commonly encountered in the practice of pulmonary medicine, the primary, idiopathic small-vessel vasculitides, are also ANCA-positive and described as AAV. These include granulomatosis with polyangiitis (GPA) (formerly known as Wegener granulomatosis), eosinophilic granulomatosis with polyangiitis (EGPA) (formerly known as Churg-Strauss syndrome), and microscopic polyangiitis (MPA). A third approach to classification is sometimes used, defining the vasculitides by the presence or absence of granulomatous inflammation. Two of the AAVs—GPA and EGPA—as well as the large vessel vasculitides, Takayasu arteritis and giant cell arteritis (GCA), are characterized by the presence of granulomatous inflammation.
Epidemiology
The true overall incidence and prevalence of vasculitis is difficult to gauge, in that all of the available epidemiologic studies contain significant flaws that limit their applicability. The available data are neither geographically nor ethnically diverse, and the case definitions and acquisition methodology vary from study to study. With these caveats in mind, GCA is the most frequently recognized systemic vasculitis with an annual incidence of 150 to 350 per million persons older than age 50 years. Persons of Nordic heritage appear at particularly high risk, especially in older age. The primary systemic vasculitides have an overall prevalence estimated at between 90 and 300 per million in European populations. Among the individual disorders, the annual incidence of GPA ranges from 4.9 to 10.5 per million, of EGPA from 0.5 to 4.2 per million, and of MPA from 2.7 to 11.6 per million. The prevalence of GPA ranges from 24 to 157 per million, of EGPA from 7 to 38 per million, and of MPA from 0 to 66 per million. Among the secondary vasculitides, the best data are for rheumatoid arthritis–associated vasculitis (incidence of 12.5 per million), although with the advent of biologic therapy, this rate appears to have decreased dramatically. Finally, the incidence of systemic lupus erythematosus (SLE)–associated vasculitis is 3.6 per million.
Normal Vascular Anatomy and Histology
In a group of diseases pathologically defined by abnormalities of the vasculature, understanding the normal organization and location of the arterial and venous blood supplies in the lung is useful in appreciating their clinical features. The lung has a dual blood supply: the pulmonary and the bronchial circulations. The bronchial arteries arise from the systemic circulation (aorta and intercostal arteries) and form a plexus in the bronchial wall. The bronchial veins associate closely with the bronchial arteries, although neither is commonly affected in the pulmonary vasculitides (see Chapter 1 ).
During embryogenesis, the pulmonary arterial system forms in tandem with the branching bronchial buds and is recognized on histologic examination by its close proximity to the corresponding bronchi or bronchiole ( Fig. 60-1A ; see also Fig. 1-17 ). In the normal lung, arteries and adjacent airways are similar in size to bronchi and bronchioles, so marked differences can be an indication of pathology. The arterial system consists of four components: elastic arteries, muscular arteries, arterioles, and capillaries. Elastic arteries are larger than 0.5 to 1 mm in diameter and can be recognized macroscopically. They are composed of an endothelial cell lining layer, a smooth muscle medial layer, and have well-developed, multiple elastic laminae. Muscular arteries are between 100 and 500 µm in diameter, and are composed of an endothelial cell lining, a smooth muscle media bound by two elastic laminae (inner and outer), and a collagenous adventitia. The smooth muscle layer progressively decreases in thickness until one reaches the arterioles. Arterioles are defined by a diameter of less than 100 µm and the absence of a muscular media; however, these can become muscularized in a variety of diseased states. The arterioles connect with capillaries. Capillaries are characterized by the presence of a single layer of endothelial cells and an underlying basement membrane. They form part of the alveolar septa, and are the most common vessel affected in pulmonary vasculitis. On hematoxylin and eosin stains of surgical lung biopsies, identifying the capillaries is difficult because a number of cells—endothelial cells, fibroblasts, and mononuclear inflammatory cells—all are normal occupants of the alveolar septa.
The location of the pulmonary veins in the mature lung is typically away from the bronchi and within interlobular septa, because the embryonic pulmonary veins form branches that grow into the mesenchyme surrounding the lung buds. Although pulmonary veins can be distinguished from pulmonary arteries by their single elastic lamina (see Fig. 60-1B and C ), in the diseased lung, there can be reduplication of the elastic lamina, causing “arterialization” of the pulmonary veins, and anatomic location may be the only indicator of vessel type.
Histopathology of Vasculitis
Although the diagnosis of vasculitis never relies on histopathology alone, both the diagnosis and subclassification can be suggested based on the size and location of the affected vessels ( Fig. 60-2 ). The findings common to all the pulmonary vasculitides are inflammatory cell infiltration (inflammation) of the vessel wall with destruction of elastic laminae (in the case of arteries and veins) and often, an accompanying fibrinoid necrosis. The lining endothelial cells may show abnormalities, with subendothelial inflammation (endothelialitis), cellular disruption and even loss of endothelial cells. The type of inflammation can vary widely; neutrophilic, eosinophilic, lymphoplasmacytic, or mixed infiltrates can all be seen. Granulomatous inflammation, consisting of poorly formed granulomas or multinucleated giant cells and/or epithelioid histiocytes, is of particular importance because one level of classification is based on its presence or absence (see earlier).
The term capillaritis is used to describe vasculitis of capillaries; however, it is often difficult to identify the capillary on routine hematoxylin and eosin stains, much less the endothelial disruption necessary to make the diagnosis. Therefore identifying neutrophils and nuclear debris (evidence of apoptosis) within alveolar septal walls is used as an indirect sign of capillaritis.
Not only do the size and type of vessel affected provide important information, but changes to the associated lung parenchyma provide additional clues. For example, a nodular parenchymal infiltrate is commonly seen in GPA, with microabscesses, necrosis, eosinophilic pneumonia, organizing pneumonia and/or hemorrhage. The specific pathologic findings associated with each entity are discussed in more detail later.
Pathogenesis and Etiology
Most of the vasculitic syndromes are hypothesized to be mediated by immunopathogenic mechanisms that arise in response to antigenic stimuli. Why some patients develop vasculitis in response to a particular stimulus, while others do not, is unknown. The etiology is likely multifactorial, including genetic predisposition, environmental exposures, and individual immune responses. A recent large genomewide association study found both major-histocompatibility complex and non–major histocompatibility complex associations with AAV, confirming a genetic contribution to disease development. Three broadly defined mechanisms have been proposed: pathogenic autoantibody formation with neutrophil activation and endothelial damage, immune complex deposition, and pathogenic lymphocyte responses. For each of these mechanisms, some combination of the direct immunologic attack and endothelial and vascular wall response appears responsible for the subsequent clinical and pathologic findings.
A large variety of autoantibodies have been described, including the antiglomerular basement membrane (collagen type IV) antibodies seen in anti-glomerular basement membrane disease, antiendothelial cell antibodies, antilaminin antibodies, antiphospholipid (e.g., anti-beta-2 glycoprotein I and anticardiolipin) antibodies, and ANCA, among others. Of these, the best studied mechanism is that proposed for the AAVs, especially GPA. While the clinical syndrome was described in the 1930s, it was not until 1982 that specific autoantibodies, now known as ANCA, were described. While the role of ANCA in the pathogenesis of disease remains to be fully elucidated, there is compelling in vitro, animal model and clinical evidence supporting a key role for the pathogenicity of ANCA. Some clinical data suggest that, while not sufficient to cause disease alone, the presence of ANCA appears to be required for the development or recurrence of systemic disease. These data include (1) the link between the presence of these autoantibodies, the development of systemic vasculitic complications, and prognosis; (2) the effectiveness of the anti-CD20 monoclonal antibody rituximab at reducing ANCA titers and controlling disease activity; and (3) the finding that patients who become ANCA negative are at low risk for clinical relapse.
Mechanistically, the majority of the clinically significant ANCAs are directed against microbicidal components used by neutrophils in host defense. These antibodies have been shown to have significant proinflammatory effects with activation of neutrophils, monocytes, and endothelial cells. ANCA stimulate the release of chemokines from neutrophils, monocytes, and endothelial cells, enhance the expression of cell adhesion molecules on endothelium, and activate primed neutrophils to release proteolytic enzymes and oxygen radicals. Each of these steps can contribute to the vascular and tissue damage seen. Moreover, an animal model has demonstrated the ability of antimyeloperoxidase antibodies (anti-MPO) to induce necrotizing vasculitis, further strengthening the link between antibody formation and disease development.
The trigger for ANCA production and persistence is poorly understood. A role for active infection/inflammation in the disease pathophysiology has been proposed because there is evidence that chronic or concurrent infections can lead to disease exacerbation or relapse. ANCA positivity has also been directed against a variety of antigens seen with viral, fungal, bacterial, and protozoal infection, as well as subacute bacterial endocarditis and cystic fibrosis. It has been hypothesized that infections may give rise to ANCA through molecular mimicry and contribute to their persistence through T- and B-cell stimulation by microbial superantigens. For example, some patients with AAV have been shown to elaborate antibodies against lysosome-associated membrane protein 2 (LAMP2) found in neutrophils. The LAMP2 protein bears close homology to the bacterial adhesin FimH protein expressed in gram-negative bacteria, and antibodies against LAMP2 and FimH are capable of producing glomerulonephritis in animal models.
Other potential mechanisms for the development of vasculitis include the direct invasion of the vessel wall by pathogenic organisms (e.g., bacterial, mycobacterial, spirochetal, rickettsial, fungal, viral), leading to an acute, direct vasculitic response as well as deposition of circulating immune complexes in the blood vessel wall. This deposition of immune complexes can lead to complement activation, anaphylatoxin production, and mast cell degranulation. Mast cell degranulation leads to release of vasoactive substances and the anaphylatoxins can act as chemotactic agents for neutrophils, eosinophils, and mononuclear inflammatory cells. The prototypical disease for this mechanism is SLE, in which complexes of nuclear antigens such as DNA, immunoglobulins, and complement proteins lead to vascular injury. Other immune complex–mediated vasculitides include rheumatoid vasculitis, Henoch-Schönlein purpura, and cryoglobulinemic vasculitis.
Aberrant lymphocyte responses are hypothesized to contribute to both the granulomatous inflammation seen in GPA and EGPA (T cells) as well as the production of ANCA (B cells). Activated T cells may be found in the peripheral blood of patients with GPA, even when the disease is in remission, and markers of T-cell activation appear to correlate with disease activity. Furthermore, there appears to be an increase in Th17-positive T-cell populations (believed to promote autoimmunity) and reduced regulatory T-cell (Treg) functionality, suggesting a loss of tolerance. Finally, there is increased elaboration of Th1 cytokines in patients with GPA (tumor necrosis factor-α, interleukin (IL)-1 and IL-8), while there is increased interferon-γ, IL-4, IL-5, and IL-13 in patients with EGPA.
Initial Diagnosis
Clinical Scenarios Suggestive of Vasculitis
The importance of the initial history in the evaluation of a patient with suspected vasculitis cannot be overemphasized. Symptoms that initially seem unrelated and of only minor importance may need to be explored more fully because both vasculitis and its mimics (e.g., connective tissue diseases, infection, malignancy, drug toxicity) present with and evolve through a variety of confusing clinical manifestations. Similarly, a careful physical examination may reveal otherwise asymptomatic disease that suggests the presence of a systemic disorder. To put some order to this, the identification of particular clinical scenarios can suggest the presence of a systemic vasculitis.
Destructive Upper Airway Lesions
Chronic refractory sinusitis in which primary infectious, allergic, and anatomic causes have been excluded and/or when destructive soft tissue or bone lesions or chronic ulcerative lesions are present can raise suspicion of an underlying vasculitis.
Chest Imaging Findings of Cavitary or Nodular Disease
Whereas a wide variety of nonspecific findings may be seen on chest imaging, the presence of nodular or cavitary disease should raise one’s suspicion. While infection and malignancy are the most common explanations, in the correct clinical setting a vasculitis, particularly an AAV, should be considered. Illustrating this point, cavities are found in 35% to 50% and nodules in 55% to 70% of patients with GPA.
Diffuse Alveolar Hemorrhage (see Chapter 67 )
Diffuse alveolar hemorrhage (DAH) refers to the presence of diffuse intra-alveolar bleeding generally from the alveolar capillaries and less frequently from the precapillary arterioles and postcapillary venules. While patients “classically” present with hemoptysis, diffuse alveolar opacities, and a drop in hematocrit, DAH must be considered in all patients with unexplained air space disease. Hemoptysis can be difficult to identify because it is often only intermittent and is not seen at all in up to one third of patients. The alveolar opacities do not have to be diffuse, and a drop in hematocrit can be difficult to document. Therefore DAH should be considered in patients with otherwise unexplained diffuse alveolar opacities, particularly when these findings complicate symptoms of a connective tissue disease or new onset renal insufficiency. While an increase in the diffusion capacity of more than 30% over baseline can be suggestive, it is rare to obtain a diffusing capacity in an acutely ill patient.
DAH is diagnosed by bronchoalveolar lavage. With the bronchoscope in wedge position, serial aliquots (30 to 60 mL in volume) of sterile saline are instilled and aspirated (for a total volume of 100 to 300 mL). If the serial aliquots of fluid reveal an increasingly hemorrhagic or, at a minimum, a persistently bloody return, then the diagnosis of DAH is made (see Fig. 67-3 ). The finding of DAH is not diagnostic of vasculitis. DAH can be caused by diseases associated with the histopathologic finding of capillaritis (including the primary idiopathic and secondary vasculitides) ( Fig. 60-3 ), as well as by diseases with diffuse alveolar damage and bland hemorrhage ( Table 60-2 ). When DAH is a complication of an AAV, capillaritis is almost always found; however, bland hemorrhage may be the only finding, particularly if treatment has been initiated. When DAH with pathologic pulmonary capillaritis is the only clinical manifestation of a vasculitis, the term idiopathic pauci-immune pulmonary capillaritis is used, and this syndrome is classified in the family of primary idiopathic small-vessel vasculitides regardless of ANCA status.
| WITH HISTOPATHOLOGIC CAPILLARITIS |
|
| WITHOUT CAPILLARITIS (BLAND HEMORRHAGE) |
|
Acute Glomerulonephritis
Rapidly progressive glomerulonephritis (RPGN) is defined by the identification of an active urinary sediment on urinalysis, including hematuria (especially with dysmorphic red cells), red cell casts, and proteinuria (>500 mg/d) in the setting of a rising blood urea nitrogen and serum creatinine. The microscopic examination of the urine needs to be performed on a fresh urine sample because red cell casts and dysmorphic red cells degenerate within 30 to 60 minutes in a freshly voided sample. Once RPGN is identified, the differential diagnosis includes AAV, idiopathic pauci-immune glomerulonephritis (isolated small-vessel renal vasculitis), SLE, Goodpasture syndrome, post-infectious glomerulonephritis, IgA nephropathy, Henoch-Schönlein purpura, essential cryoglobulinemia, and membranoproliferative glomerulonephritis.
Pulmonary-Renal Syndrome
Pulmonary-renal syndromes are classically defined as the presence of both DAH and RPGN. However, whenever a destructive airway lesion or chest imaging finding of nodules or cavities is seen together with renal insufficiency, vasculitis should also be considered. When this happens, the primary differential includes the AAVs, Goodpasture syndrome, and SLE.
Palpable Purpura
The presence of palpable purpura on physical examination implies a small vessel, cutaneous vasculitis. The most common explanation is a cutaneous (hypersensitivity) vasculitis secondary to a drug reaction; however, the AAV, cryoglobulinemia, connective tissue diseases, infections, and malignancy should all be considered.
Mononeuritis Multiplex
Defined by the development of abnormalities in two or more peripheral nerve distributions, mononeuritis multiplex should raise particular suspicion. A variety of other central or peripheral nervous system symptoms can also be seen, including pain, numbness, paresthesias, weakness, or loss of function (e.g., sudden onset of a foot drop or wrist drop).
Multisystem Disease
Unusual combinations of signs and symptoms that involve multiple organ systems either simultaneously or over time could raise the possibility of a vasculitis. This requires a high index of suspicion from the clinician because items such as constitutional symptoms (e.g., fever of unknown origin), unusual “rashes,” migratory polyarthritis, or “chronic sinus disease” may be relevant when the primary clinical presentation is breathlessness, renal failure, or abnormal findings on chest imaging.
Specific Testing
Antineutrophil Cytoplasmic Antibodies
ANCA were first described by Davies and coworkers in the early 1980s in patients with glomerulonephritis and GPA, and were recognized by a pattern of diffuse immunofluorescent staining of ethanol-fixed neutrophils. At nearly the same time, a pattern of perinuclear immunofluorescent staining of ethanol-fixed neutrophils was described in patients with MPA and pauci-immune glomerulonephirits. Currently, three specific indirect immunofluorescen t (IIF) staining patterns are described: cytoplasmic (c-ANCA) ( Fig. 60-4A ), perinuclear (p-ANCA) (see Fig. 60-4B ), and atypical (a-ANCA). c-ANCA are primarily, but not exclusively, directed against proteinase 3 (PR3, in azurophilic granules), while the p-ANCA are most commonly directed against myeloperoxidase (MPO, also in azurophilic granules), but with a much wider group of potential intracellular targets. Specific enzyme-linked immunosorbent assay (ELISA) testing for PR3 and MPO are commercially available and of considerable clinical utility. These antibodies are closely associated with the small-vessel vasculitides of the lung, GPA, EGPA, and MPA. These “ANCA-associated small-vessel vasculitides” all involve the small vessels and share a number of clinical features including, when present, pauci-immune, crescentic and focal necrotizing glomerulonephritis. However, while ANCA positivity is common in these disorders, it is by no means universal.
The diagnostic utility of ANCA testing is dependent upon the sensitivity, specificity, and positive predictive value of c-ANCA (or anti-PR3) for GPA and p-ANCA (or anti-MPO) for MPA and EGPA. When applied indiscriminately, the positive predictive value of the testing declines dramatically. Mandl and associates demonstrated that by using clinical guidelines to identify at-risk patients, the positive predictive value of the tests increased without reducing sensitivity. While ANCA testing alone or PR3 and MPO ELISA testing alone are used by many centers as their initial screening test, the combination of ANCA IIF testing plus ELISA testing maximizes their sensitivity.
With the caveats noted earlier, c-ANCA is highly sensitive (90% to 95%) in active, systemic GPA but less so (65% to 85% sensitive) in single organ–limited disease, and even less so for GPA in remission. Specificity is approximately 90%. In the proper clinical setting, with a very high pretest probability of disease, a positive c-ANCA/anti-PR3 has sufficient positive predictive value to obviate the necessity of a biopsy. On the other hand, p-ANCA and anti-MPO generally lack sufficient sensitivity and may provide no more than suggestive evidence of EGPA, MPA, or pauci-immune RPGN, because it can also be found in rheumatoid arthritis, Goodpasture syndrome, autoimmune hepatitis, inflammatory bowel disease, and a wide variety of other clinical circumstances.
Considerable attention has been focused on the utility of ANCA to assess disease activity, particularly the role of rising ANCA titers in predicting relapse. Unfortunately there appears to be no clear relationship between antibody titers and disease activity. In a prospective interventional study in which PR3-ANCA levels were routinely obtained, decreasing ANCA titers did not predict time to remission and increasing titers did not predict relapse. Increasing titers were associated with a relapse in only 40% of patients over a 12-month period. Therefore the decision to modify therapy in patients with ANCA-associated vasculitis must be a clinical decision based on clinical evidence of disease activity independent of ANCA titers.
Based on high-quality evidence, recommendations have been offered that ANCA testing by IIF should be performed in the appropriate clinical context to detect the labeling pattern and that all positive samples be tested for anti-PR3 and MPO specificity. A positive test for c-ANCA targeted to PR3 or p-ANCA targeted to MPO has a high sensitivity and specificity for the diagnosis of AAV. It should also be recognized that the absence of a positive test does not rule out a diagnosis of vasculitis. Indeed, while the concept of “ANCA-associated” applies to the patient population as a whole and has implications for pathogenesis, it must be emphasized that an individual patient may well be ANCA (or PR3/MPO) negative and still have what we describe as an AAV. ANCA testing should be performed in accredited laboratories that participate in external quality control programs and undergo regular review of laboratory management and staff performing the assays.
Other Laboratory Studies
Cultures of blood and other potentially affected organs (when tissue is obtained) should be performed to exclude infection. Routine laboratories (complete blood count with differential, chemistries, liver function tests, blood urea nitrogen, and creatinine) should be obtained, although the findings are generally nonspecific. An elevated erythrocyte sedimentation rate and elevated C-reactive protein lack specificity but are common findings. Urinalysis with microscopy should be performed on a fresh sample in all patients, because proteinuria and microscopic hematuria are common early findings in GPA and MPA. Anti-GBM antibodies should be obtained in all patients with pulmonary hemorrhage or a pulmonary-renal syndrome. Antinuclear antibodies and rheumatoid factor can be positive, although high titers; especially with the presence of disease-specific antibodies (e.g., dsDNA, SS-A/SS-B, anti-RNP, anti-Scl-70, anti-centromere antibodies, anti-JO-1) favor a connective tissue disease. IgE and circulating eosinophil counts should be obtained when EGPA is being considered.
Chest Imaging
Chest radiography and computed tomography (CT) findings are often abnormal even in the absence of symptoms, because more than 80% of patients with GPA and EGPA have some radiographic abnormalities. Specific findings are best described in GPA and include nodular opacities ( eFig. 60-1 ) and masses, particularly those that cavitate ( eFig. 60-2 ), diffuse ground-glass opacification ( eFig. 60-3 ) (especially when DAH is a possibility), consolidation ( eFigs. 60-4 and 60-5 ), atelectasis, and airway complications such as stenoses ( eFig. 60-6 ) and ulcerations ( eFig. 60-7 ). Lymphadenopathy ( eFig. 60-8 ) is not common and is more suggestive of infection or malignancy. Patients with EGPA commonly demonstrate patchy, heterogeneous ground-glass opacities or consolidation as well as evidence of airways disease.
Other Imaging Studies
CT of the sinuses (see eFig. 60-7 ) similarly demonstrates abnormalities in a majority (70% to 90%) of patients with GPA and EGPA, and may help identify destructive or ulcerating disease in patients with GPA. Electrocardiograms and echocardiography are useful in identifying cardiac involvement. The heart is involved in only 5% to 15% of patients with GPA but in as many as 30% to 50% of patients with EGPA and potentially carries a high attributable mortality. Routine screening of patients with proven or suspected AAV with electrocardiogram and echocardiography is commonly performed. Additional imaging or functional studies are determined by the clinical scenario and the signs and symptoms present in the individual patient (e.g., abdominal CT, brain CT/MRI, nerve conduction studies).
Bronchoscopy
Bronchoscopy is primarily used to look for malignancy, infection, stenotic or ulcerative upper airway or endobronchial lesions, pulmonary eosinophilia, and alveolar hemorrhage. Bronchoalveolar lavage should be grossly examined for evidence of alveolar hemorrhage, sent for culture (bacterial, fungal, and mycobacterial), cytology, and a differential cell count. Transbronchial biopsies may provide important information that helps exclude infection or malignancy; however, they are rarely useful in making a positive diagnosis of vasculitis. When Hoffman and colleagues performed 59 transbronchial biopsies in 48 patients with GPA, only four provided useful diagnostic features. Schnabel and coworkers found that transbronchial biopsies provided support for a diagnosis of GPA in only 2 of 17 patients, while otolaryngologic examination and biopsy of the clinically involved areas of the upper respiratory tract yielded useful information in 13 of 19 patients with GPA.
Diagnostic Biopsy
While a confident diagnosis may occasionally be made without tissue, diagnostic tissue biopsy remains necessary for a definitive diagnosis. Biopsy of the skin or upper airway is generally safe and these sites are easily accessible, but they less commonly yield diagnostic tissue when compared with a needle biopsy of the kidney or a surgical biopsy of the lung. When Hoffman and colleagues examined 82 open lung biopsies in patients with small-vessel vasculitis, diagnostic features were found in 90%. Renal biopsy is often performed to determine the cause of an acute glomerulonephritis. Specific features indicative of vasculitis such as granulomatous inflammation or vascular necrosis are rarely found; however, the presence of focal, segmental necrotizing glomerulonephritis without immune deposits (pauci-immune) is strongly suggestive of a systemic vasculitis.
It is important to appreciate that the pathologic features of vasculitis often overlap with other inflammatory lesions such as necrotizing infectious granulomas. In addition, not all the histopathologic features of vasculitis may be present because of the timing of the biopsy and/or modification of the histology secondary to prior treatment, particularly corticosteroids. Because of this, it is critical for the surgeon, pulmonologist, and pathologist to have a coordinated plan before the biopsy. The clinical differential will dictate the appropriate specimens to collect, including fresh tissue for culture and immunofluorescence (the presence of characteristic immunofluorescence patterns, such as IgA deposition in Henoch-Schönlein purpura, linear IgG deposition in Goodpasture syndrome, and irregular immunoglobulin and complement deposition in SLE can be diagnostic) as well as for formalin-fixation.
Specific Clinical Disorders
As systemic disorders, essentially all of the vasculitides can involve the lung. This involvement ranges from diffuse alveolar hemorrhage to cavitary nodules, parenchymal inflammation, pleural disease, vascular aneurysms, and thrombotic and thromboembolic phenomena. A full accounting of each described disorder and its pulmonary manifestations is beyond the scope of this chapter. However, the primary idiopathic small-vessel vasculitides deserve special attention.
Granulomatosis with Polyangiitis
GPA is the most common of the ANCA-associated small-vessel vasculitides, arises at any age (generally 40 to 60 years), and equally affects the sexes. It is clinically recognized by its ability to involve the upper airway (e.g., chronic sinusitis and/or otitis, upper airway ulceration and/or structural deformity, subglottic or endobronchial stenosis), lower respiratory tract (e.g., chest symptoms of cough, chest pain, shortness of breath, hemoptysis and/or chest imaging abnormalities), and kidney (e.g., glomerulonephritis) ( Table 60-3 ). However, involvement of all three sites is neither necessary nor common at presentation. For example, even though 80% to 90% of patients ultimately develop renal disease, as few as 40% have renal involvement at time of first presentation. Constitutional symptoms as well as skin, eye, musculoskeletal, peripheral, and central nervous system disease are common. Chest imaging findings are abnormal in most patients, showing alveolar, mixed, or interstitial opacities ( eFigs. 60-3 through 60-5 ), and nodular (see eFig. 60-1 ) or cavitary disease (see eFig. 60-2 ). A positive c-ANCA is seen in 90% to 95% of active systemic disease but in only 60% to 65% in limited disease. The histopathology on surgical lung biopsy is dependent on the stage of the disease and whether there has been prior immunomodulatory treatment. Involvement of a small and medium-sized vessel with necrotizing vasculitis with granulomatous inflammation and parenchymal necrosis, often with a geographic appearance, is characteristic ( Fig. 60-5A ). The pathologic manifestations can be divided into major and minor histologic features. The three major features include (1) lung parenchymal necrosis, either in the form of geographic necrosis or neutrophilic microabscesses; (2) vasculitis (generally involving small to medium-sized arteries, but which can also involve veins and capillaries) (see Fig. 60-5B and C ); and (3) granulomatous inflammation (see Fig. 60-5D ).
| GPA | EGPA | MPA | |
|---|---|---|---|
| Constitutional | Common. Includes fatigue, malaise, fevers, and weight loss. | Common. Weight loss, fatigue, fevers, myalgias and arthralgias. | Very common. Generally precedes renal disease by months. |
| Pulmonary | 70% to 95% of patients with respiratory symptoms or chest imaging abnormalities. Tracheobronchial and endobronchial disease in 10% to 50%. | Asthma essentially universal. Patchy, heterogeneous radiographic opacities in >70%. | 10% to 30% with diffuse alveolar hemorrhage. |
| Renal | 50% to 90% of patients. | 20% to 50% of patients. | RPGN is almost universal. |
| Upper airway | 70% to 95% of patients. Destructive or ulcerating lesions are suggestive. | Sinusitis, polyposis, and/or rhinitis in ≥ 70% of patients. Generally not destructive. | 5% to 30% with sinus disease most common. |
| Musculoskeletal | Arthralgias, synovitis and myalgias in up to 80%. | Arthralgias and myalgias reported in up to 50%. | Arthralgias and myalgias in at least 50% of patients. |
| Eyes | 25% to 60% of patients. Vision-threatening disease including uveitis, ocular ulcers. | <5% | Up to 30% of patients. May be clinically silent. |
| Cardiac | 5% to 25% of patients. Conduction delays or other electrocardiographic abnormality, systolic or diastolic dysfunction, pericarditis, or coronary artery vasculitis. | 30% to 50% of patients and a major cause of mortality. Conduction delays, or other electrocardiographic abnormality, systolic or diastolic dysfunction, pericarditis, or coronary artery vasculitis. | 10% to 20%. Congestive heart failure and pericarditis have been described. |
| Gastrointestinal | <10% | 30% to 50% of patients and a major cause of morbidity and mortality. Hemorrhage, abdominal pain, infarct, or perforated viscus. | 35% to 55% of patients. Findings similar to polyarteritis nodosa. Pain, bleeding, and ischemia. Rare visceral aneurysms. |
| Dermatologic | Up to 60%. Palpable purpura, ulcers, nodules or vesicles. | 50% to 70% purpura, nodules, papules, leukocytoclastic vasculitis with or without eosinophils. | 35% to 60% of patients with purpura common. |
| Neurologic | Both central and peripheral nervous system involvement. | Mononeuritis multiplex in 50% to 75%. Central nervous system in 5% to 40%. | Mononeuritis multiplex in 10% to 50%. |
| Chest imaging | Abnormal in >80%. Alveolar, interstitial, or mixed opacities, often with nodular and/or cavitary disease. | Opacities in >70%. Airways disease common (airway wall thickening, hyperinflation). | Opacities in 10% to 30%. Pleural effusions in 5% to 20%. |
| ANCA | ANCA positive > 90% and c-ANCA/anti-PR3 ELISA positive in > 85% with generalized active disease. | ANCA positive in 30% to 70% with most of these being p-ANCA/anti-MPO positive. | ANCA positive in 50% to 75% with most of these being p-ANCA/anti-MPO positive. |
Although the inflammation in GPA is typically described as granulomatous, it is often mixed and includes granulomas, giant cells, neutrophils, lymphocytes, plasma cells, histiocytes, and eosinophils. The minor histologic criteria include organizing pneumonia (70% of cases), diffuse alveolar hemorrhage (10% of cases), eosinophilia, and bronchocentric granulomatosis (1% of cases). If the biopsy is performed early in the course of disease, some of the classic histologic findings may be absent. With prior treatment, there may be no significant inflammatory infiltrates and the only (nonspecific) clue may be scarring of arteries and/or airways. A distinctive, but uncommon, histologic appearance is isolated capillaritis (see Fig. 60-3 ).
Appropriately treated disease is associated with a 5-year survival rate of 75%. While it is commonly assumed that active vasculitis itself accounts for this excess mortality, the mortality can be attributed to a variety of causes, including infection, malignancy, thromboembolic disease, cardiac disease, renal failure, and drug toxicity. Indeed, the leading cause of death among patients with AAV is infection rather than uncontrolled disease activity. Poor outcomes correlate with advanced age, lack of upper airway involvement, more severe renal impairment, pulmonary involvement (particularly with alveolar hemorrhage), cardiac involvement, and high-level anti-PR3 positivity.
Eosinophilic Granulomatosis with Polyangiitis (see Chapter 68 )
EGPA is a specific ANCA-associated small-vessel vasculitis that is clinically distinct from GPA and MPA (see Table 60-3 ), affects adults of all ages, and affects both genders equally. Its presentation often overlaps with the eosinophilic lung diseases (chronic eosinophilic pneumonia, allergic bronchopulmonary mycosis, drug reactions, hypereosinophilic syndrome, parasitic infection, asthma/atopic disease) and difficult-to-control asthma/atopy. The syndrome has its own triad of (1) asthma, (2) hypereosinophilia, and (3) necrotizing vasculitis that classically has a three-phase presentation with the initial presence of atopy/rhinitis/sinusitis/asthma, followed by an eosinophilia, and finally vasculitis. However, the three phases do not need to present sequentially, and the asthma may even postdate the vasculitis. Asthma is essentially universal and, while EGPA may present with any degree of severity and duration, severe (steroid-requiring) asthma is common with patients having asthma for 7 to 10 years before the vasculitis diagnosis. The upper airway is commonly involved with chronic rhinitis and sinusitis (with or without nasal polyposis), although generally without any of the destructive features associated with GPA.
Chest imaging is abnormal in more than two thirds of patients. Most commonly, waxing and waning parenchymal opacities (ground-glass attenuation and consolidation) ( eFig. 60-9 ) and less commonly nodules are seen on CT of the chest. Effusions can be seen in 10%. In contrast to GPA and MPA, pulmonary hemorrhage and glomerulonephritis are much less common in EGPA. Significant cardiac (conduction abnormalities, systolic or diastolic dysfunction, intracavitary thrombus, pericarditis) or gastrointestinal disease (perforation, ischemia, bleeding) are dreaded and well recognized complications. ANCA is positive in a perinuclear IIF pattern (p-ANCA) in 30% to 70% of cases, with peripheral eosinophilia (absolute eosinophil count > 1500 cells/µL) almost universal at some point in the course.
Recently, it has been identified that patients with EGPA segregate into two distinct clinical phenotypes. One subset is characterized by a greater incidence of neurologic, renal, gastrointestinal, and cutaneous involvement that is more commonly ANCA or MPO-positive (i.e., shares a high degree of overlapping features with GPA and MPA). The other subset of patients shares features with the hypereosinophilic syndromes, namely cardiac manifestations, migratory lung opacities (eosinophilic pneumonia), and an ANCA-negative/MPO-negative serologic profile.
Pathologically, patients with EGPA demonstrate both a necrotizing, small-vessel vasculitis and an eosinophil-rich cellular infiltrate. The diagnostic findings on lung biopsy include eosinophilic pneumonia, necrotizing vasculitis, and granulomatous inflammation ( Fig. 60-6 ). The vasculitis consists of artery, vein, or capillary wall infiltration by lymphocytes and eosinophils. The granulomas often show central areas of necrosis with abundant necrotic eosinophils, surrounded by palisaded histiocytes and multinucleated giant cells. Findings highly suggestive of EGPA include eosinophilic pneumonia and necrotizing vasculitis; findings suggestive of EGPA include eosinophilic pneumonia and parenchymal necrosis.
