Abstract
Diffuse alveolar hemorrhage most commonly refers to bleeding from the low-pressure pulmonary vascular bed and is a rare cause of chronic interstitial lung disease in children. Patients can present acutely with respiratory failure or more indolently with a history of worsening exercise intolerance and cough. Diffuse alveolar hemorrhage is divided into two broad categories: immune mediated and nonimmune mediated with immune-mediated hemorrhage being more common. The most common causes of immune-mediated diffuse alveolar hemorrhage are the ANCA-associated vasculitidies and idiopathic pulmonary capillaritis. It is important to distinguish between immune and nonimmune causes of hemorrhage, such as idiopathic pulmonary hemosiderosis, as treatment is significantly different.Regardless of the cause of alveolar hemorrhage patients most often present with diffuse infiltrates on chest x-ray and anemia. Chest CT scan is very useful in establishing a diagnosis and most often reveals diffuse ground glass opacities and septal thickening or nodules. Bronchoscopy and bronchoalveolar lavage are used to establish pulmonary hemorrhage. A lung biopsy is necessary in any patient with diffuse hemorrhage and negative ANCA to help distinguish between immune- and nonimmune-mediated hemorrhage.Treatment of diffuse alveolar hemorrhage is dependent on the type of hemorrhage. Immune-mediated hemorrhage requires aggressive immune suppression to halt progression of the disease. Nonimmune-mediated hemorrhage is managed with much less immune suppression.
Keywords
child, diffuse alveolar hemorrhage, granulomatosis with polyangiitis, microscopic polyangiitis, isolated pulmonary capillaritis, COPA syndrome
The lungs receive blood from two separate systems: the bronchial circulation and the pulmonary circulation. The bronchial circulation is a high-pressure, low-volume circuit supplied by the bronchial arteries, which vary in number and origin, but most often arise directly from the aorta or one of its branches. These vessels provide blood to the conducting airways from the mainstem bronchi to the terminal bronchioles. Because the bronchial circulation is subject to high pressures, bleeding from this system has the potential to be profuse, sometimes resulting in massive hemoptysis and death. In contrast, the pulmonary circulation is a low-pressure, high-capacitance circuit that arises from the right ventricle and provides blood flow to the acinar units involved with gas exchange. Disruption of this system results in alveolar hemorrhage, which is often low-grade, chronic, and diffuse. Although massive hemoptysis is rare, uncontrolled alveolar hemorrhage can be fatal.
Pulmonary hemorrhage arising from either the systemic or pulmonary circulation has multiple etiologies and can be localized or diffuse ( Box 61.1 ). Bleeding from the nasopharynx, oropharynx, or upper digestive tract is common and must be ruled out as a source of “simulated” hemoptysis, or true hemoptysis when the blood is aspirated. Bleeding from the upper airway can occur from intrinsic lesions, such as a subglottic hemangioma or tumor, or from extrinsic causes, such as an inhaled foreign body or intubation. Intubation may induce ulceration and granulation tissue formation in the airway wall, which may eventually lead to hemorrhage. More serious iatrogenic causes capable of inducing massive and fatal hemoptysis include erosion of a tracheostomy tube into the aorta or innominate artery, and perforation of the pulmonary artery by a Swan-Ganz catheter.
Infection
Bronchitis
Bronchiectasis/cystic fibrosis
Primary ciliary dyskinesia
Immunodeficiency
Lung Abscess
Pneumonia
Trauma
Airway laceration
Lung contusion
Artificial airway
Suction catheters
Foreign body
Inhalation injury
Vascular Disorders
Pulmonary embolism/thrombosis
Pulmonary arteriovenous malformation
Pulmonary hemangioma
Coagulopathy
Von Willebrand’s disease
Thrombocytopenia
Anticoagulants
Congenital Lung Malformations
Sequestration
Congenital pulmonary airway malformations
Bronchogenic cyst
Miscellaneous
Catamenial
Factitious
Neoplasm
Diffuse Alveolar Hemorrhage Syndromes
In children with advanced cystic fibrosis, hemoptysis is relatively common because severe, chronic airway inflammation leads to progressive bronchiectasis with increased dilatation and fragility of vessels in the airway walls. Bleeding can also occur with other etiologies of bronchiectasis including primary ciliary dyskinesia and immunodeficiency. Infection of the airways or lung parenchyma from viruses, fungi, and bacteria, particularly Streptococcus pneumoniae and Staphylococcus aureus, is a common cause of hemoptysis in children. In such cases, mechanical trauma from forceful coughing may contribute to bleeding. Finally, factitious hemoptysis should be considered in a child with unusual symptoms and a negative evaluation.
This chapter focuses on diffuse alveolar hemorrhage (DAH) arising from the small vessels of the pulmonary circulation. A cardinal feature of DAH is the presence of hemosiderin-laden macrophages in the acinar units, as macrophages are responsible for clearing free erythrocytes from airspaces. In a simulated alveolar hemorrhage model, hemosiderin-laden macrophages first appeared 3 days following a hemorrhage, peaked at days 7–10 with hemosiderin staining in 60% of macrophages, and continued to be present at 2 months in 10% ( Fig. 61.1 ). The etiology of DAH includes both immune-mediated and nonimmune-mediated disorders ( Box 61.2 ).
Immune Mediated
Idiopathic pulmonary capillaritis
Granulomatosis with polyangiitis
Microscopic polyangiitis
Anti-GBM disease
Systemic lupus erythematosus
Henoch-Schönlein purpura
Behçet’s disease
Cryoglobulinemic vasculitis
Juvenile idiopathic arthritis
COPA Syndrome
Nonimmune Mediated
Idiopathic pulmonary hemosiderosis
Acute idiopathic pulmonary hemorrhage of infancy
Heiner’s syndrome
Asphyxiation/abuse
Cardiovascular causes
Pulmonary vein atresia/stenosis
Total anomalous pulmonary venous return
Pulmonary veno-occlusive disease
Mitral stenosis
Left-sided heart failure
Pulmonary capillary hemangiomatosis
Pulmonary telangiectasia
Etiology of Diffuse Alveolar Hemorrhage
Immune-Mediated Alveolar Hemorrhage
A subgroup of children and adolescents with DAH has the pathologic findings of pulmonary capillaritis (PC). Though a histologic diagnosis, PC usually defines an underlying systemic vasculitis or an immune-mediated disease process. PC can occur as an isolated disorder; as part of an antineutrophil cytoplasmic antibody (ANCA)–associated vasculitis; or a systemic disorder, such as systemic lupus erythematosus (SLE). Of the ANCA-associated vasculitides, DAH from PC has been reported in 12%–55% of patients with microscopic polyangiitis and in 7%–45% of patients with granulomatosis with polyangiitis (formerly known as Wegener’s granulomatosis). In contrast, SLE, a more common disease, has a lower incidence of pulmonary hemorrhage, but the hemorrhage can be life threatening and should be treated aggressively. (See Chapter 58 reviews SLE in detail.) PC also has been reported in Henoch-Schönlein purpura, Behçet’s disease, cryoglobulinemic vasculitis, and juvenile idiopathic arthritis.
Pathophysiology
ANCA is frequently associated with diseases characterized by the presence of vasculitis, affecting small- and medium-sized vessels. These diseases are associated with circulating autoantibodies directed against the neutrophil granule components, myeloperoxidase (MPO) and proteinase 3 (PR3). In the correct clinical context, the specificity of these autoantibodies for ANCA-associated vasculitides is as high as 98%. This group of disorders includes granulomatosis with polyangiitis, microscopic polyangiitis, and eosinophilic granulomatosis with polyangiitis (formerly Churg-Strauss syndrome). Despite some overlap, PR3-ANCA is particularly associated with granulomatosis with polyangiitis, and MPO-ANCA, with microscopic polyangiitis. The primary events in the pathogenesis of these necrotizing vasculitides are not known. Several hypotheses propose that infectious agents trigger and perpetuate such events. The presence of these autoantibodies most likely reflects a pathobiological series of events in which neutrophils and monocytes attack the endothelium of small vessels, causing the release of autoantigens with their eventual presentation to the immune system and consequent autoantibody formation. In ANCA-associated vasculitis, it has been shown that autoantibodies to endothelial cell constituents (antiendothelial cell antibodies [AECA]) are commonly produced following the activation and injury of endothelial cells. During this inflammatory attack on the vessel walls, the basal membrane of some vessels also can be damaged and autoantibodies produced to the basement membrane α3 domain of type IV collagen of the pulmonary vessels and glomeruli.
Other autoantibodies have been described in ANCA-associated vasculitis. These autoantibodies are likely related to other factors in the inflammatory cascade and may even be part of an orchestrated attack on the endothelium and basement membrane. They include AECA, antiglomerular basement (GBM) antibodies, antibasal membrane laminin antibodies, and antiphospholipid antibodies (APLA). With the exception of the antibasal membrane laminin antibodies, these autoantibodies are considered to be clinically significant.
Antibodies to constituents of the endothelium are directed to small-vessel endothelial cells, reflecting the typical distribution of such cells in the lungs, nose, and kidneys. An increase in AECA levels has been described in patients with increasing vasculitic activity in idiopathic ANCA-associated and drug-induced vasculitis. An increase of these autoantibodies has been observed in both ANCA-negative and ANCA-positive patients during disease relapse. Levels of ANCA and AECA fluctuate independently as these autoantibodies do not cross-react with the target antigens.
Among patients with anti-GBM disease (also known as Goodpasture syndrome), antibodies to the alpha domain of type IV collagen have been well described. Several groups have found anti-GBM antibodies cooccurring with ANCAs in patients with idiopathic ANCA-associated vasculitis. Anti-MPO antibodies are the most specific in these patients. Some have proposed that it is essential to test for both types of autoantibodies because the cooccurrence of anti-GBM antibodies and ANCAs may be associated with more severe renal involvement leading to end-stage renal disease and poorer survival.
The presence of APLA, which include anti-beta2 glycoprotein1 antibodies, anticardiolipin antibodies, and the lupus anticoagulant, can be a feature of idiopathic ANCA-associated vasculitis. Several reports have indicated that patients with both ANCA and APLA experience more extensive and more severe disease. In drug-induced vasculitis, APLA of the immunoglobulin M class are more frequently seen. If antihistone antibodies and ANCAs directed to more than one ANCA antigen are found together, a drug-induced condition should be suspected.
Antineutrophil Cytoplasmic Antibody Associated Vasculitis
Granulomatosis With Polyangiitis.
Granulomatosis with polyangiitis is a necrotizing vasculitis of the small- and medium-sized blood vessels associated with granulomatous inflammation, which usually affects the respiratory system first and then the kidneys. Upper and lower respiratory manifestations include chronic rhinitis; sinusitis; serous otitis media; nasal cartilage destruction leading to saddle nose deformity; salivary gland swelling, subglottic stenosis, and tracheobronchial ulceration; parenchymal lung nodules or masses that tend to cavitate; and DAH from PC. Clinically, patients have dyspnea, cough with or without hemoptysis, and hypoxemia. Patients with renal involvement have hematuria and red cell casts. Other affected organs include the eye, heart, gastrointestinal tract, spleen, joints, skin, central nervous system, and pituitary gland.
Laboratory evaluation should include a complete blood count (CBC) to identify anemia, erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) as nonspecific markers of inflammation, and blood chemistries with urinalysis to look for renal disease. Finally, testing for ANCA is essential because the majority of patients will have serum anti-PR3 antibodies (c-ANCA pattern).
Chest radiographs may show impressive nodules or masses with or without cavitation that are disproportionate to respiratory symptoms or pulmonary function studies. Computed tomography (CT) scans of the chest not only better characterize these nodules or cavitary lesions, but also may show diffuse ground glass densities, consolidation related to DAH, or airway abnormalities ( Fig. 61.2 ). Sinus CT is useful in identifying sinus opacification and bony destruction.
Bronchoscopy may identify upper and lower airway lesions including ulceration, granulomata, stenosis, and malacia. In patients with diffuse infiltrates, bronchoalveolar lavage may be overtly bloody and contain red blood cells and hemosiderin-laden macrophages consistent with DAH.
Histopathologic confirmation is still considered the gold standard for the diagnosis of granulomatosis with polyangiitis in children. Characteristic features include vasculitis of the medium and small vessels and capillaritis associated with necrotizing granulomata ( Fig. 61.3 ). Biopsies from the lung or upper respiratory tract are preferred because of their high sensitivity and specificity. Kidney biopsies typically show segmental necrotizing glomerulonephritis with or without crescent formation, which is nonspecific and can be seen in other immune-mediated renal disorders. However, the paucity of immune complexes is an important characteristic feature of this subset of renal disease.
In 1994, the Chapel Hill Consensus Conference declared that histopathologic documentation of granulomatous involvement of the respiratory tract was not explicitly required to diagnose granulomatosis with polyangiitis. Thus, the diagnosis of granulomatosis with polyangiitis now rests on clinical, serologic, radiographic, and pathologic correlations.
Microscopic Polyangiitis.
Microscopic polyangiitis is a small-vessel vasculitis that, in children, appears to be more common than granulomatosis with polyangiitis. The presentation is similar to the other pulmonary-renal syndromes with hemoptysis, anemia, and new chest x-ray infiltrates in adults. However, hemoptysis is not always present in young children. Patients can present with an acute, life-threatening event or with a more indolent course. Hypoxemia, often found at presentation, can be profound and can require intubation and mechanical ventilation. Although renal disease is found in some patients at presentation, its absence does not exclude the diagnosis. Physical exam often reveals diffuse crackles, and imaging studies show characteristic but nonspecific changes of diffuse alveolar infiltrates, sometimes with a tree-in-bud pattern and septal thickening on chest CT ( Fig. 61.4 ). Bronchoscopy and bronchoalveolar lavage reveal evidence of alveolar hemorrhage with blood-tinged lavage and hemosiderin-laden macrophages on cytologic examination. In the absence of renal disease, all of the findings described above are nonspecific and cannot distinguish microscopic polyangiitis from idiopathic pulmonary hemosiderosis (IPH) or any other alveolar hemorrhage syndrome.
A diagnosis of microscopic polyangiitis is suggested by the presence of serum anti-MPO antibodies (p-ANCA pattern). Lung histopathology in microscopic polyangiitis shows PC with neutrophilic infiltration of small arterioles, venules, and capillaries associated with fibrinoid necrosis ( Fig. 61.5 ). Granulomatous vasculitis, characteristic of granulomatosis with polyangiitis, is not seen. Renal involvement, when present, is typically manifest on biopsy as segmental necrotizing glomerulonephritis.
Treatment of ANCA-Associated Vasculitis
Treatment for ANCA-associated vasculitis in children is broadly similar to the approach used for adult patients based on evidence from a number of clinical trials conducted by the European Vasculitis Study Group. With major concern for the high morbidity and mortality, therapy in ANCA-associated vasculitis is aimed at the preservation of organ function, mainly renal and pulmonary, and maintaining clinical remission.
Induction therapy has traditionally included glucocorticoids and cyclophosphamide. In children, we use both pulse IV methylprednisolone (30 mg/kg, maximum 1 g, for 2–3 days, and then once weekly) and oral prednisone (1–2 mg/kg, daily). Regimens involving oral or IV cyclophosphamide therapy, given over a period of 3–6 months, have demonstrated efficacy in inducing remission, lengthening time to relapse, and reducing adverse events. IV pulse cyclophosphamide was shown to be as effective as daily oral cyclophosphamide with a 50% decrease in cumulative cyclophosphamide dose, and therefore fewer adverse effects. Despite the efficacy of glucocorticoids and cyclophosphamide in inducing remission, disease relapses occur in as many as 50% of patients, and 20%–53% may develop a relapse within the first 1–2 years following treatment. Disease relapses occur when the drugs are reduced or withdrawn. Given the adverse effects of cyclophosphamide, newer treatment strategies directed at B cell depletion have been developed. In a recent trial, rituximab was as effective as cyclophosphamide in inducing remission with a similar side-effect profile.
Other important components of induction therapy in children include intravenous immunoglobulin (IVIG) and plasmapheresis. In addition to three open studies demonstrating beneficial effect on ANCA-associated vasculitis, a small randomized, placebo-controlled trial of 34 patients showed a significantly higher rate of remission in the IVIG-treated group compared with placebo (15 vs. 6 remissions, P = .015). Plasmapheresis has been shown to decrease morbidity in patients with worse renal disease, as part of the induction therapy. No controlled trials are available with respect to alveolar hemorrhage and plasmapheresis; however, a retrospective analysis of 20 patients with alveolar hemorrhage who received plasmapheresis reported resolution of hemorrhage in all patients. We believe that plasmapheresis should be considered in the early management of DAH, particularly in an intensive care unit setting due to granulomatosis with polyangiitis, microscopic polyangiitis, macrophage activation syndrome, antiglomerular basement membrane antibody disease, antiphospholipid antibody syndrome, and SLE.
After patients are disease-free for 6 months, maintenance therapy should be initiated with low-dose prednisone and either methotrexate or azathioprine for at least 1–2 years. To date, the optimal duration of maintenance therapy is unknown. A well-defined and structured clinical assessment with urinalysis and basic laboratory testing should be performed regularly to assess for disease activity and relapse, treatment response, and drug adverse effects.
Anti-GBM Disease
Similar to granulomatosis with polyangiitis and microscopic polyangiitis, anti-GBM disease presents with alveolar hemorrhage and renal disease. However, in contrast to the other pulmonary-renal syndromes that may have more systemic involvement, anti-GBM disease is almost exclusively limited to the lung and kidneys. Diagnostic imaging and bronchoalveolar lavage will yield findings consistent with nonspecific DAH. However, the classic distinguishing feature includes the detection of anti-GBM antibodies in serum and by immunofluorescence along the basement membrane in lung and renal tissue ( Fig. 61.6 ). Treatment of anti-GBM disease is similar to that of ANCA-associated vasculitis, with corticosteroids used as primary treatment, and cyclophosphamide added in severe cases. One important difference is that plasmapheresis is used in all cases of anti-GBM disease to remove the circulating anti-GBM antibodies, limiting the amount of damage in the lungs and kidneys.
