Chapter 33. Microscopic Polyangiitis

• Microscopic polyangiitis (MPA) is the most common cause of the pulmonary-renal syndrome of alveolar hemorrhage and glomerulonephritis.
• Usually includes combinations of two or more of the following:
  • Nonspecific constitutional symptoms, including fatigue, myalgias, weight loss, and fevers.
  • Migratory arthralgias or arthritis, either pauciarticular or polyarticular.
  • Palpable purpura, sometimes with skin ulcerations.
  • Sensorimotor mononeuritis multiplex.
  • Alveolar hemorrhage associated with hemoptysis and respiratory compromise.
  • Glomerulonephritis.
• Antineutrophil cytoplasmic antibodies (ANCAs) are often critical in making the diagnosis, but a significant minority of patients are ANCA-negative.
• The majority of patients with MPA who are ANCA positive have antibodies directed against myeloperoxidase (MPO-ANCA).
• ANCA titers are often elevated during disease flares but do not have a consistent temporal relationship with disease activity. Thus, ANCA titers are not reliable predictors of disease flares.

Microscopic polyangiitis (MPA) is a form of systemic vasculitis that may affect many major organs with crippling or fatal effects. Seventy percent of patients with MPA have antineutrophil cytoplasmic antibodies (ANCAs). MPA is recognized to be related to both granulomatosis with polyangiitis (GPA; formerly Wegener granulomatosis) and eosinophilic granulomatosis with polyangiitis (EGPA; the Churg-Strauss syndrome) (see Chapters 32 and 34). These disorders are sometimes considered together as the ANCA-associated vasculitides, but important differences exist among these three conditions and significant percentages of patients with these diagnoses do not have ANCA.

MPA has been recognized increasingly since the first Chapel Hill Consensus Conference on the nomenclature of systemic vasculitides, the results of which were published in 1994. Many cases before then were considered to be forms of polyarteritis nodosa (PAN), a disease with which MPA shares substantial overlap. Table 33–1 compares the features of MPA with those of GPA (Wegener) and PAN.

The term “polyangiitis” is preferred to “polyarteritis” for MPA because of the disease’s tendency to involve veins as well as arteries. The first Chapel Hill Consensus Conference defined MPA as a process that (1) involves necrotizing vasculitis with few or no immune deposits; (2) affects small blood vessels (capillaries, arterioles, or venules) and possibly medium-sized vessels; and (3) demonstrates a tropism for the kidneys and lungs. With an estimated incidence of 4 cases per million per year, MPA is more common than classic PAN but somewhat less common than GPA (Wegener).

MPA occurs in people of all ethnic backgrounds. The male:female ratio is approximately 1:1. The typical patient is middle-aged to elderly, but the disease may affect people of all ages. The mean age at diagnosis for MPA patients (approximately 60 years) is about 10 years older than the mean age of GPA (Wegener) patients at diagnosis. Several epidemiologic studies have tried to elucidate environmental factors associated with the onset of vasculitis. Some authors have found associations with silica and solvent exposure.

The strongest link between an exposure and MPA relates to the use of propylthiouracil (PTU) for the treatment of hyperthyroidism (a handful of other drugs for other indications have also been implicated, but not as strongly). Anti-MPO antibodies are detected frequently in PTU-treated patients, albeit overt vasculitis occurs in only a small minority (<5%). Drug-induced, ANCA-associated vasculitis is discussed further in Chapter 42).

Symptoms and Signs

The interval between the onset of first disease symptoms and diagnosis in MPA is substantially shorter than for patients with GPA (Wegener). This is because of the tendency for GPA (Wegener) to smolder in the upper respiratory tract and cause apparently mundane symptoms for months before leading to medical attention. In contrast, the clinical presentation of MPA is usually more obvious at the time the patient becomes aware of symptoms: cutaneous vasculitis, vasculitis neuropathy, or alveolar hemorrhage. Nevertheless, subtle and subacute presentations of MPA are known to occur, and the range of organ system manifestations is extensive.

Although MPA is classified appropriately as a “pulmonary-renal syndrome,” regarding this disorder exclusively as a disease that affects the kidneys and lungs is a major potential clinical error. The five most common clinical manifestations of MPA are glomerulonephritis (nearly 80% of patients), weight loss (>70%), mononeuritis multiplex (60%), fevers (55%), and cutaneous vasculitis (>60%). Alveolar hemorrhage, in contrast, occurs in only about 12% of patients. The major clinical manifestations of MPA are shown in Table 33–2.

Head, Eyes, Ears, Nose, and Throat

Some vasculitis experts regard the presence of any upper respiratory tract involvement as evidence that the diagnosis is GPA (Wegener), not MPA. Thus, HEENT involvement in MPA is limited generally to rhinitis or mild cases of nondestructive sinusitis. Serous otitis media may occur in MPA but unlike in GPA (Wegener), granulomatous inflammation is absent. Ocular lesions in MPA (eg, episcleritis, conjunctivitis, keratitis, and occasionally scleritis) have been reported but are less common and less severe than in GPA (Wegener).

Lungs

The principal pulmonary manifestation of MPA is capillaritis, which leads to alveolar hemorrhage and often to hemoptysis. Hemoptysis may be only a late indication of bleeding. The typical radiologic features of alveolar hemorrhage are shown in Figure 33–1. Alveolar hemorrhage is associated with a worse prognosis. Interstitial fibrosis and pleuritis occur in some patients with MPA. Pulmonary fibrosis that resembles usual interstitial pneumonitis in clinical presentation is increasingly recognized as a disease manifestation of MPA. Many cases of pulmonary fibrosis are associated with previous alveolar hemorrhage, but the precise relationship between alveolar hemorrhage and fibrosis is not clear.

Kidneys

Renal involvement is seen in at least 80% of patients with MPA. The classic presentation of renal disease in MPA is a rapidly progressive glomerulonephritis reminiscent of GPA (Wegener) (Figure 33–2A). Some patients, however, have renal deterioration that progresses more slowly, over many months. Renal involvement may also present with urinary abnormalities such as proteinuria, microscopic hematuria, and red cell casts. Up to 40% of patients have 24-hour urinary protein excretion of more than 3 g. Proteinuria of this severity is regarded as a poor prognostic factor for renal outcome. The pathologic features of renal disease in MPA are indistinguishable from other forms of pauci-immune glomerulonephritis—namely, a necrotizing, crescentic lesion (Figure 33–2B). Compared with biopsies from patients with ANCA directed against proteinase 3, those with MPO-ANCA have a more chronic pattern of renal injury, with more glomerulosclerosis, tubular atrophy, and interstitial fibrosis.

Nervous System

Vasculitic neuropathy is a potentially devastating complication of MPA. The nerve involvement typically occurs in the pattern of a distal, asymmetric, axonal polyneuropathy (mononeuritis multiplex). The first symptoms of vasculitic neuropathy are usually sensory, with numbness, tingling, and dysesthesias. Muscle weakness and wasting follow the infarction of motor nerves (Figure 33–3). Because the named peripheral nerves are usually mixed nerves, bearing both sensory and motor fibers, patients with vasculitic neuropathy typically have both sensory and motor symptoms. Recovery from vasculitic neuropathy may take months; some patients have residual nerve damage after the disease is controlled. Although peripheral nerve lesions tend to dominate the neurologic features of MPA, central nervous system involvement by vasculitis is also described in this disease.

Small-fiber neuropathy has also been reported in MPA. In patients with small-fiber neuropathy, the predominant symptoms are pain and numbness rather than motor weakness. Electrodiagnostic studies in small-fiber neuropathy patients are normal because the involved fibers are below the resolution of nerve conduction velocity assessments. Diagnosis is made by biopsy of the skin and staining for the density of small nerve fibers.

Skin

The skin manifestations of MPA include all of the cutaneous lesions associated with small-vessel vasculitis (palpable purpura, papules, vesiculobullous lesions, splinter hemorrhages). In the presence of medium-vessel involvement, nodules, ulcers, livedo reticularis (Plate 47), and digital gangrene may occur. As with most forms of cutaneous vasculitis, the lesions favor the lower extremities.

Musculoskeletal System

Nonspecific arthralgias and frank arthritis usually present early in the course of MPA and respond quickly to therapy. Musculoskeletal symptoms may also herald disease flares. The arthritis of MPA is migratory in nature and can assume a variety of joint patterns, from a pauci-articular syndrome of large joints to a polyarthritis of small joints. Destructive joint lesions do not occur in MPA.

Laboratory Findings

The results of routine laboratory tests and specialized assays in MPA are shown in Table 33–3. All of these tests are appropriate at the initial evaluation in patients who demonstrate features consistent with MPA. The exclusion of renal disease through the careful performance of a urinalysis is essential in the evaluation and follow-up of all patients with MPA. The erythrocyte sedimentation rate and serum C-reactive protein level are useful in the longitudinal evaluation of disease activity.

Positive ANCA assays are often instrumental in suggesting the diagnosis, but the titers of these antibodies correlate poorly in time with disease flares. Moreover, approximately 30% of patients with MPA diagnosed on a clinical basis are ANCA negative (see serologic testing, below). Thin-cut computed tomography scans are sensitive in the detection of lung disease in MPA.

Special Tests

Tissue Biopsy

By definition, MPA involves small blood vessels: arterioles, venules, and capillaries. Glomerulonephritis is the renal equivalent of small-vessel vasculitis, akin to palpable purpura in the skin and capillaritis in the lung. Renal biopsy findings, although not specific for MPA, are sufficiently characteristic to establish the diagnosis in appropriate clinical settings. Immunofluorescence studies of renal biopsies in MPA confirm the “pauci-immune” nature of the renal involvement. MPA may also involve medium-sized arteries and veins, but the identification of medium-vessel involvement is not essential to the diagnosis.

MPA is high on the differential diagnosis of leukocytoclastic vasculitis within the small blood vessels of skin lesions. The presence of extracutaneous findings and ANCA increases the likelihood of MPA. If sufficiently deep, skin biopsies may also demonstrate the involvement of medium-sized vessels in the deep dermis subcutaneous tissue layer. The finding of medium-vessel involvement eliminates certain forms of cutaneous vasculitis limited to small-vessel disease, eg, hypersensitivity vasculitis (cutaneous leukocytoclastic angiitis) and Henoch-Schönlein purpura. Direct immunofluorescence of skin biopsy tissue is also important in the exclusion of immune complex-mediated processes such as cryoglobulinemia. The involvement of both veins and arteries distinguishes MPA from classic polyarteritis nodosa, which is confined to arterial lesions.

Nerve Conduction Studies

Nerve conduction studies are an important part of the evaluation for patients with neuropathic symptoms. Nerve conduction studies may reveal the characteristic asymmetric, axonal sensorimotor neuropathy. Nerves such as the sural nerve shown to be involved in this fashion are prime candidates for biopsy, with simultaneous sampling of adjacent muscle (eg, the gastrocnemius). The sural nerve is an excellent candidate for biopsy because, in contrast to most peripheral nerves, it contains only sensory fibers. In some cases, histopathology diagnostic of vasculitis is confined to the muscle as opposed to the nerve, or vice versa. As noted, nerve conduction studies may be negative in patients with small-fiber neuropathies.

Although lung involvement can be a florid manifestation of MPA, demonstration of vasculitis on thoracoscopic or open lung biopsy is often challenging; frank capillaritis may be difficult to detect. Nevertheless, lung biopsies are often essential to exclude other processes (eg, infections or malignancies) if no other tissue options exist for biopsy.

Serologic Testing for ANCA

Three fourths of all patients with clinical diagnoses of MPA are ANCA-positive. A full discussion of ANCA is found in the chapter on GPA (Wegener) (see Chapter 32). In MPA, the classic pattern of serum reactivity upon immunofluorescence testing (with human neutrophils as the substrate) is perinuclear staining (P-ANCA). The P-ANCA pattern in MPA patients is usually caused by antibodies to MPO, a constituent of the primary granules of neutrophils. A variety of nonvasculitic conditions (Table 33–4) can also cause P-ANCA immunofluorescence, but these results are usually caused by antibodies to antigens not associated with vasculitis (eg, lactoferrin).

The combination of both a P-ANCA pattern on immunofluorescence testing and MPO-ANCA demonstrated by enzyme immunoassay has a high positive predictive value for ANCA-associated vasculitis, most commonly MPA. The other type of ANCA found in MPA is PR3-ANCA, directed against proteinase 3. This type of ANCA is usually associated with a cytoplasmic (C-ANCA) pattern of immunofluorescent staining. Despite advances in ANCA testing techniques, histopathology remains the cornerstone of diagnosis in MPA. When the diagnosis is unconfirmed, all reasonable attempts to obtain a tissue diagnosis should be pursued.

The greatest mimickers of MPA are other forms of vasculitis (Table 33–4). Henoch-Schönlein purpura and hypersensitivity vasculitis (also known as cutaneous leukocytoclastic angiitis) can cause identical skin lesions, as can GPA (Wegener), EGPA (Churg-Strauss), mixed cryoglobulinemia, and PAN. The delineation of MPA from these disorders comes from the pattern recognition of extracutaneous involvement (kidneys, lung, nerve), the biopsy of involved organs, and ANCA testing. The difficulties of distinguishing MPA from GPA (Wegener) and PAN are illustrated in Table 33–1.

Since Goodpasture disease (anti-glomerular basement membrane disease) can present in a manner indistinguishable from MPA, such patients may benefit from plasma exchange in addition to glucocorticoids and cytotoxic agents. As discussed below, the potential benefits of plasma exchange in MPA are less clear than in Goodpasture disease.

MPA can lead to lymphoplasmacytic infiltrates within the adventitia and also cause severe headaches, thereby mimicking temporal (giant cell) arteritis both clinically and pathologically. In contrast to “true” temporal arteritis, MPA involving the temporal artery is not associated with giant cells. In addition, some medications, particularly propylthiouracil (used to treat thyroiditis), can cause a drug-induced, ANCA-associated vasculitis associated with high titers of antibodies to MPO.

A variety of pulmonary, renal, and peripheral nerve disorders must be distinguished from MPA by imaging studies, tissue biopsy, nerve conduction studies, and serologic testing. Systemic autoimmune conditions such as systemic lupus erythematosus and rheumatoid arthritis are also prone to imitating MPA because of their abilities to involve multiple organ systems and cause positive P-ANCA results on immunofluorescence testing of serum (see above).

The essentials of management for MPA are shown in Table 33–5. MPA is one of a handful of vasculitic conditions in which the conventional standard of care for many years called for combination therapy with both glucocorticoids and a cytotoxic agent. However, rituximab has been demonstrated to be as effective as cyclophosphamide in the induction of remission in ANCA-associated vasculitis (AAV) (MPA and GPA [Wegener]). The induction of remission with rituximab is preferred to cyclophosphamide because of rituximab’s potentially superior long-term side effect profile. Rituximab has a particular advantage over cyclophosphamide in the treatment of patients with relapsing (as opposed to newly-diagnosed) disease.

If rituximab cannot be given or if the patient has not responded to this regimen, cyclophosphamide is the treatment of choice. The combination of cyclophosphamide (2 mg/kg/d for those with normal renal function) and glucocorticoids (1 mg/kg/d of prednisone, perhaps preceded by a 3-day intravenous “pulse” of methylprednisolone) leads to excellent therapeutic responses if treatment is initiated early enough. Three- to 6-month courses of cyclophosphamide are used followed by longer-term treatment for the maintenance of remission with either azathioprine (up to 2 mg/kg/d) or methotrexate (20–25 mg/wk).

Both intravenous and oral daily administration regimens of cyclophosphamide are effective at inducing disease remission. The preferred route of administration is largely a function of practice style. The important points are that cyclophosphamide be used promptly when indicated and with appropriate cautions. All patients who are receiving treatment for MPA should be given a daily dose of single-strength trimethoprim-sulfamethoxazole, 100 mg/d of dapsone, or 1500 mg/d of atovaquone as prophylaxis against Pneumocystis jiroveci pneumonia.

The need for a remission maintenance regimen following rituximab-induced disease remissions is not clear and must be determined in longitudinal studies. Following the induction of remission by cyclophosphamide, patients may be switched to either azathioprine (up to 2 mg/kg/d) or methotrexate (up to 25 mg/wk, assuming that residual renal dysfunction does not preclude this medication). The optimal duration of these remission maintenance agents is not clear. In general, the continuation of azathioprine or methotrexate for a period of 1 year after the achievement of remission is a reasonable recommendation.

Once the inflammatory process has been controlled with immunosuppressive therapy, primary care clinicians may institute renal preservation therapies for patients with renal damage (blood pressure control, angiotensin-converting enzyme inhibition, and salt restriction).

If MPA is diagnosed early and treated promptly, patients have a high likelihood (>90%) of achieving disease remissions. Approximately one third of patients suffer disease flares after the achievement of remission by conventional therapies (cyclophosphamide and prednisone). In general, MPA is considered less likely than GPA (Wegener) to flare. Unfortunately, significant damage frequently ensues before recognition of the disease. One study indicated that the 5-year renal survival for patients with this disease was only 55%. This prognosis may have improved somewhat since the widespread availability of ANCA testing.

The renal prognosis in MPA may be worse than that of GPA (Wegener), perhaps because of a greater likelihood of delay in diagnosis in MPA, attributable to the involvement of fewer organ systems. Another major disability associated with MPA results from nerve damage and consequent muscle weakness caused by vasculitic neuropathy. Finally, patients with AAV have a high risk of venous thrombotic events. Heightened suspicion for this complication, possibly caused by involvement of the veins by the vasculitic process, should be maintained.