Chapter 31. Vasculitides

• Vasculitides affecting the kidneys are typically associated with hematuria and proteinuria, frequently presenting as a rapidly progressive glomerulonephritis.
• Glomerular injury occurs in the setting of the small vessel vasculitides, associated with antineutrophil cytoplasmic autoantibodies (ANCA), antiglomerular basement membrane antibodies (anti-GBM), or the presence of immune complex formation such as with Henoch–Schönlein purpura (HSP), cryoglobulinemic vasculitis, and systemic lupus erythematosus (SLE).
• They may affect the kidneys alone, but are more frequently part of a multiorgan disease that may affect the skin, upper and lower respiratory tracts, and the musculoskeletal, gastrointestinal, and nervous systems.

The classification of vasculitides is based on the predilection for injury of the different vascular beds (Table 31–1). The following sections provide an overview of the various groups of vasculitides. The remainder of this chapter will focus on the clinical aspects of small vessel vasculitis (SVV) because of their association with glomerulonephritis.

Large Vessel Vasculitis: Takayasu's Arteritis and Giant Cell Arteritis

Large vessel vasculitis affects the aorta and its major branches. During the acute phase of disease, large vessel vasculitis is characterized pathologically by chronic granulomatous inflammation that often contains giant cells in the inflammatory infiltrates. The chronic phase is characterized by extensive vascular sclerosis with little or no active inflammation. Inflammatory and sclerotic thickening of the aorta and the arteries causes narrowing of lumina, which in turn causes ischemia and the resultant clinical manifestations. Involvement of the renal artery may cause renovascular hypertension.

The two major large vessel vasculitides are Takayasu's arteritis and giant cell arteritis. Takayasu's arteritis most often involves the aorta and its major branches, although the pulmonary arteries may be affected. Takayasu's arteritis is most common in Asia, and rarely occurs in patients older than 40 years. Clinically, it often presents with diminished pulses, vascular bruits, claudication, and renovascular hypertension.

Giant cell arteritis rarely occurs in patients younger than 50 years and is most common in patients of northern European ethnicity. Like Takayasu's arteritis, giant cell arteritis affects the aorta and its major branches; however, it has a much greater predilection for the extracranial branches of the carotid artery. Frequent clinical manifestations include headache, jaw claudication, blindness, deafness, tongue dysfunction, extremity claudication, and reduced peripheral pulses. Pathologic involvement of the renal artery is common in giant cell arteritis, but symptomatic renovascular hypertension is rare. Polymyalgia rheumatica occurs in 40–60% of patients with giant cell arteritis. The presence of polymyalgia rheumatica may be useful for diagnosis, but symptoms may occur before, simultaneously, or after the onset of giant cell arteritis.

Medium-Sized Vessel Vasculitis: Polyarteritis Nodosa and Kawasaki's Disease

The medium-sized vessel vasculitides have a predilection for arteries that lead to major viscera and their initial branches. In the kidneys, the major targets are the interlobar and arcuate arteries, with less frequent involvement of the main renal artery and interlobular arteries. The two major medium-sized vessel vasculitides are polyarteritis nodosa (PAN) and Kawasaki's disease. Pathologically, both are characterized in the acute phase by necrotizing arteritis with transmural inflammation that can lead to the formation of pseudoaneurysms. In only a few days, the lesions evolve from an acute neutrophil-rich inflammation to a “chronic” inflammation with predominantly mononuclear leukocytes. Secondary complications of the arteritis include thrombosis, infarction, and hemorrhage. Sites of thrombosis and necrosis develop progressive scarring. Although medium-sized vessel vasculitides do not cause glomerulonephritis by definition, they can cause hematuria, proteinuria (usually less than 2 g/24 hours), and renal insufficiency as a result of renal infarction. Pseudoaneurysms near the renal surface may rupture and cause severe, even fatal, retroperitoneal, and intraperitoneal hemorrhage.

Although the diagnostic term “polyarteritis nodosa” previously included all patients with any pattern of necrotizing arteritis, the new classification distinguishes it from vasculitides that predominantly affect small arterioles, capillaries, and venules [such as Wegener's granulomatosis, Churg–Strauss syndrome, and microscopic polyangiitis (MPA)]. By this approach, the presence of glomerulonephritis or pulmonary alveolar capillaritis with pulmonary hemorrhage rules out a diagnosis of polyarteritis nodosa and indicates the presence of some type of SVV. Testing for ANCA is typically negative in PAN as opposed to patients with small vessel vasculitides, MPA, Wegener's granulomatosis, or Churg–Strauss syndrome. Table 31–2 compares some of the features of polyarteritis nodosa and MPA.

The necrotizing arteritis of Kawasaki's disease is pathologically indistinguishable from polyarteritis nodosa. Kawasaki's disease is an acute febrile illness of childhood that is characterized by the mucocutaneous lymph node syndrome, which includes nonsuppurative lymphadenopathy, polymorphous erythematosus rash, erythema of the oropharyngeal mucosa, erythema of the palms and soles, conjunctivitis, and indurative edema and desquamation of the extremities. A major cause for morbidity and mortality in patients with Kawasaki's disease is the development of a necrotizing arteritis with a predilection for coronary arteries. Symptomatic renal involvement is rare in Kawasaki's disease.

Because Kawasaki's disease and polyarteritis nodosa are treated differently, differentiation between the two arteritides is very important. The presence or absence of the mucocutaneous lymph node syndrome is an effective diagnostic discriminator. Kawasaki's disease usually is treated with aspirin and intravenous γ-globulin therapy. One of the main goals of treatment of Kawasaki's disease is to reduce the incidence of coronary arteritis. Timely treatment with intravenous immunoglobulin (IVIG) and aspirin has been shown to reduce the incidence of cardiac lesions from 20–40% to <5%. Aspirin alone has not been shown to reduce cardiac sequelae but combined therapy with IVIG appears to have an additive anti-inflammatory effect. If the patient fails initial treatment, repeated doses of IVIG, pulse methylprednisolone, cyclophosphamide, methotrexate, cyclosporin, and plasmapheresis have been utilized. Recently data suggest that infliximab may be an alternative therapy for patients with refractory disease.

Idiopathic polyarteritis nodosa is usually treated with high-dose corticosteroids and cyclophosphamide. This form of therapy has been shown to be beneficial, especially in those who have adverse prognostic factors. The optimal length of treatment is unknown, but a recent study has suggested that 6 months of therapy may be less effective than 12 months. For patients with hepatitis B virus-associated PAN, therapy with antiviral agents improves outcomes.

Small Vessel Vasculitis: Microscopic Polyangiitis, Wegener's Granulomatosis, and Churg–Strauss Syndrome

Small vessel vasculitides are characterized by necrotizing inflammation primarily targeting venules and capillaries, although arteries, arterioles, and veins may be affected. Unlike vasculitides of larger vascular beds, small vessel vasculitides frequently cause glomerulonephritis (Table 31–1).

The two major categories of SVV include the “pauci-immune small vessel vasculitides” and the “immune complex-mediated small vessel vasculitides” (Table 31–3). Immune complex-mediated vasculitides, such as HSP, cryoglobulinemic vasculitis, lupus vasculitis, and anti-GBM vasculitis, have extensive localization of immunoglobulin and complement in vessel walls as a consequence of deposition of circulating immune complexes or in situ immune complex formation between circulating antibodies and planted or constitutive antigens. The pauci-immune necrotizing small vessel vasculitides have little or no vascular wall localization of immunoglobulins, and often present with a necrotizing and crescentic glomerulonephritis, either alone or as a component of a systemic disease. Pauci-immune crescentic glomerulonephritis is the most common type of crescentic glomerulonephritis (Table 31–4).

The three major systemic pauci-immune small vessel vasculitides are MPA, Wegener's granulomatosis, and Churg–Strauss syndrome. MPA is a necrotizing angiitis involving capillaries, venules, and arterioles of one or more organs either simultaneously or at different times, including the kidneys, lungs, skin, spleen, liver, heart, and muscle. Wegener's granulomatosis is characterized by necrotizing granulomatous inflammation found in the upper or lower respiratory tract and may occur without overt vasculitis. The Churg–Strauss syndrome is defined by the presence of eosinophil-rich granulomatous inflammation and vasculitis in patients with a history of eosinophilia and asthma. The necrotizing glomerulonephritis and vasculitis found in Wegener's granulomatosis and Churg–Strauss syndrome can be pathologically identical to that found in microscopic polyangiitis. It is characterized by segmental fibrinoid necrosis, crescent formation, and the absence of immune reactants.

ANCA SVV is more common in whites than in African-Americans, with a ratio of 7–8:1. Females and males are equally affected, and while patients are typically 55 years or older, patients of any age may have this disease.

Conceptually, vasculitides may be organized on the basis of three different mechanisms of injury: (1) Immune complex-mediated vasculitis, (2) direct antibody-mediated attack, or (3) pauci-immune necrotizing vasculitides. Immune complex-mediated vasculitis includes HSP, cryoglobulinemic vasculitis, rheumatoid vasculitis, and lupus vasculitis. Direct antibody attack-mediated diseases include anti-GBM-mediated glomerulonephritis and Goodpasture's syndrome, and Kawasaki's disease mediated by antiendothelial cell antibodies. The pauci-immune small vessel vasculitides associated with ANCA will be the focus of this chapter as HSP, cryoglobulinemic vasculitis, lupus nephritis, and anti-GBM disease are described in detail in other chapters.

Symptoms and Signs

Renal disease is manifested by hematuria with dysmorphic red blood cells and red blood cells casts. Proteinuria is usually moderate (2–3 g/day) but may be as much as 20 g/day. ANCA-associated glomerulonephritis frequently presents as a rapidly progressive glomerulonephritis, although the syndromes of asymptomatic hematuria with minimal amounts of proteinuria or acute nephritis are common as well.

At least 50% of patients with ANCA necrotizing glomerulonephritis have pulmonary disease spanning the spectrum from severe life-threatening pulmonary hemorrhage to fleeting alveolar infiltrates. Massive pulmonary hemorrhage affects about 10% of patients with ANCA glomerulonephritis, and is associated with an elevated risk of death.

In the evaluation of pulmonary disease in the setting of a glomerulonephritis, it is imperative to exclude the possibility of an infectious etiology, especially when pulmonary infiltrates with or without hemoptysis develop in the setting of prior or ongoing immunosuppressive treatment. Recurrent vasculitis must be differentiated from infection, with special attention to Pneumocystis carinii, Mycobacterium tuberculosis, or fungal pathogens. Bronchoscopy and bronchial alveolar lavage help differentiate infection from alveolar hemorrhage. Similarly, infections of the upper respiratory tract mimic vasculitic lesions in the nose, sinus, and ear. Fiberoptic transillumination of the upper airways with biopsy may allow differentiation of vascular inflammation from infection.

Common dermal vasculitic findings are palpable purpura (usually in the lower extremities), petechia, ulcers, nodules, urticaria, ecchymoses, and bullae. The neurologic manifestations of ANCA SVV are typically peripheral neuropathies (mononeuritis multiplex), while central nervous system involvement, specifically granulomatosis meningeal inflammation, occurs uncommonly. Several other organ systems may be involved. Gastrointestinal disease, which affects one-third of patients with ANCA SVV, may present with vasculitic ulcers of the small and large intestine resulting in bleeding or perforation. Iritis, uveitis, and episcleritis result in red, painful eyes. These lesions are frequently present in a “subclinical” fashion and require slitlamp ophthalmologic evaluation. Almost all patients have a prodrome of a “flu-like illness” with malaise, myalgias, and arthralgias. A migratory pattern of arthropathy is a typical presentation.

Laboratory Findings

Testing for ANCA is useful in diagnosing pauci-immune small vessel vasculitis or renal-limited pauci-immune crescentic glomerulonephritis. ANCA reacts with constituents of neutrophils and monocytes. By indirect immunofluorescence microscopy (IIFM) on alcohol-fixed neutrophils, two patterns of staining can be distinguished: A cytoplasmic pattern (C-ANCA) and a perinuclear pattern (P-ANCA). The majority of C-ANCAs react with the lysosomal enzyme proteinase 3 (PR3-ANCA). In necrotizing SVV, P-ANCA reacts with myeloperoxidase (MPO-ANCA). Wegener's granulomatosis is usually associated with circulating PR3-ANCA, but as many as 20% of patients may have MPO-ANCA. Conversely, PR3-ANCA can be found in patients with microscopic polyangiitis and necrotizing glomerulonephritis without evidence of Wegener's granulomatosis. MPO-ANCA is more commonly found in patients with microscopic polyangiitis and those with necrotizing and crescentic glomerulonephritis without evidence of extrarenal SVV. MPO-ANCA is the most common ANCA subtype in Churg–Strauss syndrome. Thus, despite a predominance of antigenic specificity for each of the diseases, neither ANCA subtype differentiates between the three phenotypes of necrotizing vasculitis. While the majority of antibodies associated with a P-ANCA immunofluorescent pattern are antibodies to myeloperoxidase, a number of other antigens have rarely been associated with the P-ANCA pattern. These include antibodies to hyaluronidase, high mobility groups 1 and 2 (associated with gastrointestinal immune disorders), lactoferrin, cathepsin G, enolase, lysozyme, and azurocidin. Occasionally, an antinuclear antibody (ANA) pattern of immunofluorescence may be mistaken for a P-ANCA pattern.

Special Tests

Relationship of Serial ANCA Titers and Disease Activity

Despite numerous retrospective and a few prospective studies, it is still unknown how best to use serial ANCA testing in the management of patients with ANCA vasculitis and glomerulonephritis. In reality, the simple question of whether serial ANCA testing can be used to foresee and prevent the occurrence of relapse corresponds to a more complex problem as ANCA testing represents several tests, ANCA vasculitis represents several diseases, and the risk of relapse is not uniform among all patients with ANCA vasculitis. For serial ANCA testing to be clinically useful in the management of patients, ANCA titers must correlate with disease activity, a relapse must predictably follow a rise in ANCA titer within a reasonable time frame, and an effective, safe, and acceptable therapy must be available to abrogate the relapse. None of these requirements is fully met in ANCA vasculitis.

While ANCA titers have been shown to correlate with disease activity in cohort studies of patients with ANCA vasculitis, this correlation can be very poor for some individual patients. Clinical remission occurs in the setting of a persistently elevated ANCA titer, and conversely, clinical relapses occur in the setting of a persistently negative or low ANCA titer. Furthermore, what constitutes a clinically significant change in ANCA titer has been defined differently in various studies and remains to be determined for each of the various ANCA tests.

Notwithstanding these caveats, most of the available data specifically address the study of C-ANCA or PR3-ANCA in patients with Wegener's granulomatosis, whereas the data for P-ANCA, MPO-ANCA, and patients with microscopic polyangiitis or glomerulonephritis alone are much more limited. The reported rates of relapses following a rise in titer vary from 23% to 82% for C-ANCA determined by IIFM, and from 59% to 100% for PR3- or MPO-ANCA as determined by enzyme-linked immunosorbent assay (ELISA). The time between a rise in ANCA titer and the occurrence of relapse varies significantly. In a study of 85 patients with PR3-ANCA-positive Wegener's granulomatosis, ˜50% of relapses occurred within 6 months after the rise in titer, while some relapses occurred more than 18 months later.

An important concern was recently raised pertaining to patients who maintain a persistently elevated ANCA titer despite a successful induction therapy. Based on relatively small numbers of patients, two retrospective studies suggest that patients with persistently elevated ANCA titers are more likely to suffer a relapse than patients with negative or declining titers. The question of whether to institute immunosuppressive therapy at the time of a rise in ANCA titer has been addressed in only one randomized, prospective study. In this study of patients with Wegener's granulomatosis, of nine patients preemptively treated with cyclophosphamide for 9 months after a ≥4-fold increase in titer none suffered a relapse [versus 6 of 11 (55%) nontreated patients]. Some relapses occurred more than a year after the rise in titer. Although preemptively treated patients received less immunosuppressive drugs by the end of the study, it employed a medication that is associated with a high risk for serious short- and long-term adverse effects.

In summary, although a rise in PR3-ANCA titer appears to presage a subsequent relapse, there is a relative paucity of data regarding microscopic polyangiitis, P-ANCA, and MPO-ANCA and the best way to define a rise in ANCA titer. Considering that the time to relapse can be very long, and that the risk/benefit of preemptive, prolonged immunosuppression is not established, it is fair to state that the use of preemptive immunosuppression is not justified. The decision to initiate a second course of treatment should rather be based on the constellation of findings on clinical course, physical examination, and serologies rather than ANCA titers alone.

Role of the Kidney Biopsy

Whether a renal biopsy is essential for the management of ANCA glomerulonephritis rests on the accuracy of the ANCA methodology, the risks associated with therapy, and the “pretest probability” based on the features of the patient's clinical syndrome. Based on a study of 1000 patients with proliferative and/or necrotizing glomerulonephritis, the sensitivity of ANCA for pauci-immune glomerulonephritis was only 80% with a specificity of 89%, corresponding to a positive predictive value of 86%, a false-positive rate of 14%, and a false-negative rate of 16%. These results preclude the use of ANCA testing alone as the determinant for treatment. Conversely, if the pretest probability of ANCA vasculitis is high based on the presence of characteristic clinical findings, a positive ANCA test may be sufficient to establish a diagnosis of pauci-immune necrotizing glomerulonephritis. In the setting of a rapidly progressive glomerulonephritis, or a pulmonary-renal vasculitic syndrome, initiation of treatment with pulse methylprednisolone should not be delayed until a biopsy result is obtained, as prompt initiation of treatment is an essential determinant of outcome.

The clinical differential diagnosis of ANCA-associated pauci-immune SVV also includes direct antibody attack-mediated and immune complex mediated SVV. Despite a great deal of overlap in organ system involvement among different types of small vessel vasculitides, certain clinical serologic and histologic features—described in Tables 31–4 and 31–5—differentiate among MPA, Wegener's granulomatosis, Churg–Strauss syndrome, cryoglobulinemic vasculitis, and HSP. Testing for ANCA, anti-GBM, cryoglobulins, hepatitis C or B, ANA, and complement component levels helps focus the differential diagnosis. Direct immunofluorescence microscopy of vessels in biopsy specimens, such as glomerular capillaries or dermal venules, demonstrates immunoglobulin (Ig) A-dominant vascular immunoglobulin deposits in HSP, IgG and IgM deposits in cryoglobulinemic vasculitis, and little or no immunoglobulin in pauci-immune SVV.

The differential diagnosis of pauci-immune necrotizing glomerulonephritis includes lupus nephritis, anti-GBM disease, and other aggressive forms of glomerulonephritis. Patients with thrombotic microangiopathies may present a picture that mimics necrotizing glomerulonephritis. However, these patients do not have MPO- or PR3-ANCA and have features typical of a microangiopathic hemolytic disease. About 20% of patients with anti-GBM disease may concomitantly express ANCA.

The very nature of systemic SVV exposes patients to a number of potential complications related to irreversible end-organ damage as a result of the disease itself or its therapy. In fact, as many as 85–90% of patients with Wegener's granulomatosis suffer some permanent end-organ damage that adversely affect their quality of life, such as end-stage renal disease (ESRD), chronic pulmonary dysfunction, hearing loss, destructive sinus disease, proptosis, motor or sensory neurologic deficits, and blindness.

ANCA vasculitis can also present with acute life-threatening manifestations. Diffuse pulmonary hemorrhage presents with gross hemoptysis and respiratory failure requiring mechanical ventilation. Without the use of plasmapheresis in addition to corticosteroids and cyclophosphamide, diffuse pulmonary hemorrhage is associated with a 50% mortality rate.

Patients with Wegener's granulomatosis may present with acute critical subglottic stenosis associated with stridor, and respiratory failure, which may require emergency tracheostomy. Unfortunately, surgical interventions on the trachea in patients with active Wegener's granulomatosis frequently leads to a chronic course characterized by poor healing and scarring, requiring repeat interventions. For this reason, early detection of subglottic involvement and the prompt institution of intralesional as well as systemic corticosteroids and immunosuppressant therapy are essential to avoid surgery or tracheostomy.

End-stage kidney disease (ESKD) may be the result of the acute injury or from progressive chronic scarring. In a cohort of 350 patients with ANCA vasculitis, 22% of patients presented with, or reached ESKD within 2 months of presentation. In addition, 10% of patients who responded to initial therapy later progressed to ESKD in a median of 106 months without clinical evidence of disease relapse.

Perhaps less recognized is the increased risk of venous thrombotic disease among patients with Wegener's granulomatosis. This was illustrated in a recent large clinical trial that detected an incidence of venous thrombosis of 7 per 100 patient-years, with 75% of the events occurring at the time of, or shortly preceding clinically active vasculitis.

The complications met during the course of vasculitis are often iatrogenic. Common to all forms of immunosuppression is the risk of serious infections including bacterial, viral, and fungal opportunistic organisms. The risk of serious infection is compounded by treatment-related neutropenia, which may occur in up to 55% of patients treated with cyclophosphamide or azathioprine. In that respect, the risk of neutropenia and associated infections is substantially lower with the use of monthly pulse cyclophosphamide than the daily oral regime.

Treatment with cyclophosphamide is also associated with long-term risks of gonadal failure and infertility in both men and women, and malignancies, especially cutaneous, hematologic, and urothelial cancers. These risks are commensurate with the cumulative dose of cyclophosphamide and other cytotoxic or antimetabolites received during the course of treatment. Use of glucocorticoids is also associated with multiple short- and long-term complications including diabetes mellitus, osteoporosis, avascular necrosis of bone, myopathy, and cataracts.

It would be useful to describe concomitant assessments and/or prophylactic medications that may be prescribed to minimize complications. For instance, at our institution we are using Bactrim (either SS once/day or DS 3/week) for Pneumocystis prophylaxis, and inconsistently use Mycelex to prevent thrush. We assess the patient for old tuberculosis (we have an immigrant population in which this problem is not uncommon) and we respond accordingly to a positive purified protein derivative (PPD) depending on prior treatment history. I do not know if there are standards for care here, but some attempt to address this issue would be practical and very helpful.

Similarly, on the osteoporosis side, we will use calcium carbonate 1 g/day and vitamin D along with Fosamax for patients with creatinine clearance (Ccr)>50. Again, there may not be evidence-based data, but if any of these things are actually done, it would be useful to begin a non-evidence-based discussion of these issues.

Induction Therapy

The cornerstone of ANCA vasculitis treatment is based on a combination of corticosteroids and cyclophosphamide according to various regimens described in the literature. Since renal prognosis appears to be determined by early treatment, we typically initiate induction therapy with three daily pulses of methylprednisolone (7 mg/kg/day) followed by daily oral prednisone. Prednisone is started at a dose of 1 mg/kg for the first month, and tapered progressively over 3–4 months. Cyclophosphamide is administered either by monthly intravenous pulses (0.5–1 g/m2) or orally (1–2 mg/kg/day). All forms of cyclophosphamide dosage should be titrated to keep the nadir leukocyte count >3000 cells/mm3. When compared to treatment with corticosteroids alone, the use of cyclophosphamide is associated with a 5-fold decrease in the risk of death and a 3-fold decrease in the risk of relapse. Patients treated with either intravenous or oral cyclophosphamide have a long-term remission rate of between 60% and 85%.

Determining the optimum mode (oral versus intravenous) and duration of treatment with cyclophosphamide depends on several factors and considerations. The short- and long-term complications associated with the use of cyclophosphamide are commensurate with the cumulative dose received. In general, the intravenous regimen allows for a two to three times smaller total dose of cyclophosphamide than the oral regimen. In prospective and retrospective analyses, intravenous therapy was associated with a significant decrease in the rate of clinically significant neutropenia and other complications. In a meta-analysis of three randomized controlled trials comparing pulse versus oral continuous cyclophosphamide, intravenous cyclophosphamide resulted in a statistically higher rate of remission (odds ratio for failure to achieve remission 0.29; 95% CI 0.12–0.73) and lower rates of leukopenia (odds ratio 0.36; 95% CI 0.17–0.78) and infections (odds ratio 0.45; 95% CI 0.23–0.89). The final outcomes of patients (death or ESKD) were no different in the two groups despite a (statistically not significant) lower rate of relapse in the oral cyclophosphamide group.

The optimum length of therapy with cyclophosphamide has not been determined and is the subject of ongoing controversy. In patients achieving complete remission within 6 months of therapy, treatment can be stopped with the institution of close patient follow-up. In those individuals with persistently active disease at 6 months, it is reasonable to continue cyclophosphamide therapy for a full 12 months. An alternative regimen consists of switching cyclophosphamide to oral azathioprine at the end of 3 months if the patient is in remission. Azathioprine is then continued for 18 months. This regime offers the advantage of a limited use of cyclophosphamide and results in rates of remission and relapse similar to the cyclophosphamide-only-based therapies.

Patients presenting with pulmonary hemorrhage also benefit from the institution of plasmapheresis in a regimen similar to that used for patients with Goodpasture's disease. Although no controlled data are available, early and aggressive institution of plasmapheresis has substantially diminished the mortality rate associated with massive pulmonary hemorrhage. Plasmapheresis is typically performed daily until the pulmonary hemorrhage ceases and then every other day for a total of seven to ten treatments. The addition of plasma exchange to cyclophosphamide and corticosteroids also improves the chances of recovery of renal function of patients with severe renal dysfunction (serum creatinine >5.6 mg/dL) or needing dialysis at presentation. For patients without profound renal failure or with pulmonary hemorrhage, controlled and observational trials have not demonstrated a beneficial role to the addition of plasmapheresis to cyclophosphamide and corticosteroids.

Adjunctive, Alternative, and Maintenance Treatment Strategies

Several other therapies have been evaluated for a potential role in the management of patients with ANCA vasculitis. Considering the generally excellent response rates to treatment with cyclophosphamide and glucocorticoids, the goals of these therapies can conceptually be delineated in the broad categories of adjunctive therapy for patients at risk for treatment resistance; alternatives to the repeat use of cyclophosphamide; or as maintenance therapy aimed at preventing relapses. With a few notable exceptions, the data available in support of these treatments are limited to open-label, uncontrolled studies that are often limited in size. No agent has as yet been established as a clear alternative to cyclophosphamide for the induction therapy of patients with organ- or life-threatening disease. The following sections review the data regarding alternative, adjunctive, or maintenance therapies for ANCA vasculitis.

Trimethoprim-Sulfamethoxazole

Trimethoprim-sulfamethoxazole is suggested as the initial treatment of selected patients with Wegener's granulomatosis limited to the upper respiratory tract, reserving corticosteroid therapy for patients who fail antibiotic therapy. In assessing the role of trimethoprim-sulfamethoxazole in the prevention of relapse in patients who were already treated with cyclophosphamide and/or prednisone, a prospective placebo-controlled trial with trimethoprim-sulfamethoxazole was performed in 81 patients with Wegener's granulomatosis. The number of relapses was significantly reduced in the groups assigned to trimethoprim-sulfamethoxazole, although the benefit was limited to relapse involving the upper respiratory tract, not the lower respiratory tract or the kidney. In contrast, “maintenance” therapy with trimethoprim-sulfamethoxazole (after cyclophosphamide treatment) has been associated with an increased rate of relapse when compared to no maintenance treatment (42% versus 29%, respectively).

Similarly, in a randomized trial of 65 patients with Wegener's granulomatosis comparing methotrexate and trimethoprim-sulfamethoxazole, methotrexate therapy was more effective in maintaining remission than trimethoprim-sulfamethoxazole used alone or in combination with prednisone. Thus, the use of trimethoprim-sulfamethoxazole in patients with Wegener's granulomatosis remains a matter of controversy, and its beneficial effects seem to be limited to the respiratory tract.

Methotrexate

The use of methotrexate (MTX) has been studied almost exclusively in patients with Wegener's granulomatosis, and primarily in patients with “limited,” mild disease, and with little or no renal involvement. MTX has been evaluated as an alternative to cyclophosphamide for the induction of remission in an open-label, uncontrolled trial that revealed a response rate of 76% and a remission rate of 69%. The value of weekly MTX as an induction agent was compared to daily oral cyclophosphamide in a randomized controlled trial in 100 patients with “early” ANCA vasculitis who did not have organ- or life-threatening manifestations or significant renal involvement (patients with creatinine >150 μmol/L, urinary red cell casts, or proteinuria >1.0 g/day were excluded). Although the remission rate at 6 months was not statistically different between the two groups [MTX (89.8%) versus cyclophosphamide (93.5%) (p = 0.041)], remission was delayed among MTX-treated patients with more extensive disease or pulmonary involvement. MTX treatment was associated with a significantly higher rate of relapse than cyclophosphamide (69.5% versus 46.5%) and 45% of the relapses occurred while the patients were still receiving MTX.

Whether MTX is effective in preventing relapses after induction therapy with cyclophosphamide has also been evaluated in open-label, uncontrolled studies. These studies revealed relapse rates of 37–52% with 61–72% of relapses associated with glomerulonephritis and 69% of relapses involving the upper respiratory tract. In the absence of proper control groups, it is not possible to assess whether maintenance therapy with MTX reduces the rate or severity of relapses.

In summary, MTX appears to be an option in the treatment of patients with mild ANCA vasculitis and normal or near normal renal function. The data regarding its value for maintenance therapy are currently less convincing. The dose of MTX should be decreased by 50% in patients with renal insufficiency and its use is contraindicated if the Ccr is <10 mL/minute. Trimethoprim-sulfamethoxazole for prophylaxis against Pneumocystis carinii pneumonia cannot be used concomitantly with MTX.

Azathioprine

The role of azathioprine in the management of patients with ANCA vasculitis was examined for the maintenance of remission as an alternative to a prolonged course of cyclophosphamide. In a large multicenter European trial, 144 patients with ANCA vasculitis who attained remission with oral cyclophosphamide and prednisolone were randomized to either a continued course of oral cyclophosphamide for a total of 1 year or switched to oral azathioprine. All patients were switched to oral azathioprine after 12 months. At the end of 12 months of follow-up, the risk of relapse was not significantly different between the two groups (13.7% versus 15.5%, p = 0.65) nor was there a significant difference in the risk of adverse events.

In the past few years a number of new immunomodulatory agents, targeting T cells, B cells, or both, and a number of proinflammatory pathways have been introduced. Several of these agents have been evaluated in a limited way as adjunctive or maintenance therapy of ANCA vasculitis. These include the inhibitors of lymphocyte proliferation mycophenolate mofetil and leflunomide, the lymphocyte-depleting humanized anti-CD52 antibody alemtuzumab, the B cell-depleting chimeric antibody rituximab, and blockers of the tumor necrosis factor (TNF)-α pathway infliximab and etanercept. All but etanercept were evaluated only as part of small open-label, uncontrolled studies. The possible role of etanercept for the maintenance of remission was evaluated in a randomized, placebo-controlled trial of 180 patients in the Wegener's Granulomatosis Etanercept Trial (WGET). In addition to etanercept or placebo, patients received standard therapy with glucocorticoids plus cyclophosphamide or MTX. After 3–6 months of cyclophosphamide, patients in remission discontinued cyclophosphamide and were treated with etanercept and either MTX or azathioprine (depending on their degree of renal insufficiency). Over a mean follow-up of 27 months, 72% of patients attained a sustained remission, but only 50% of patients remained in remission for the remainder of the trial. There were no significant differences between the etanercept and control groups in the rates of sustained remission (69.7% versus 75.3%, p = 0.39), in the relative risk of disease flares per 100 person-years of follow-up (0.89, p = 0.54), or in the severity of relapses. During the study, 56.2% of patients in the etanercept group and 57.1% of those in the control group had at least one severe or life-threatening adverse event or died (p = 0.90). It was concluded that etanercept was not effective for the maintenance of remission in patients with Wegener's granulomatosis.

Transplantation

Recurrence of ANCA vasculitis after renal transplantation occurs in about 20% of patients. Time to recurrence varies widely, from a few days to several years posttransplantation. Patients with Wegener's granulomatosis appear more likely to relapse than patients with microscopic polyangiitis or necrotizing crescentic glomerulonephritis alone, whereas the presence of circulating ANCA at the time of transplant does not seem to increase the risk of recurrence after transplantation. If the patient is clinically in remission, it is probably not necessary to delay transplantation until a negative ANCA serology is attained. We believe that transplantation should be delayed in patients with active disease. In the majority of reported cases, recurrent disease after transplantation responded well to treatment with cyclophosphamide and pulse corticosteroids.

Supportive Therapy

As corticosteroids and cyclophosphamide remain the cornerstone of therapy of ANCA SVV, special effort must be exercised to minimize the short- and long-term complications of treatment. Whenever corticosteroids are used, the development of osteoporosis can be minimized with the early institution of calcium and vitamin D supplementation, and in patients with established osteoporosis, calcitonin or bisphosphonates (if not contraindicated by azotemia or esophagitis). Rigorous control of blood pressure with sodium restriction and antihypertensive therapy is essential to minimize the additive effect of hypertension in loss of renal function following active nephritis. Hormonal manipulation during cytotoxic therapy may allow the preservation of gonadal function. The prevention of cyclophosphamide-induced infertility was reported with the use of testosterone in men and leuprolide in women.

Untreated, systemic vasculitis is associated with an 80% 1-year mortality. The introduction of steroids, azathioprine, and cyclophosphamide led to a marked improvement in survival to 84% and 76% at 1 and 5 years, respectively. Predictors of death include increasing age and creatinine at presentation, disease extent and severity at diagnosis, pulmonary hemorrhage, and treatment-related infection.

Response to Treatment

The terms “remission” and “relapse” are defined in Table 31–6. Treatment with cyclophosphamide is beneficial over the use of corticosteroids alone for achieving a remission as well as for patient survival. Patients treated with either intravenous or oral cyclophosphamide have a long-term remission rate of between 70% and 92%. Based on a large observational cohort of 350 patients with ANCA vasculitis, female gender, black race, and potentially older age were associated with a higher likelihood of treatment resistance. Whether socioeconomic factors and access to health care account for these differences in response to treatment is unknown. Patients with PR3-ANCA may have a better chance of remission than patients with MPO-ANCA.

Relapse

The risk of relapse after an initial response to treatment is of the order of 40%, but reports vary between 11% and 57%. The risk of relapse is not uniform among all patients with ANCA vasculitis. An increased risk of relapse has been associated with a diagnosis of Wegener's granulomatosis (as opposed to MPA). Based on a multivariate analysis of 258 patients, the presence of PR3-ANCA and lung and upper respiratory tract involvement were independent risk factors for relapse. Among patients presenting with none of the three risk factors, 26% relapsed in a median of 62 months (median among those who relapsed was 20 months). In contrast, 47% of the patients presenting with a single risk factor experienced a relapse in a median of 39 months (corresponding to a 2-fold increased risk for relapse; 95% CI: 1.1, 3.9, p = 0.038). Among patients presenting with all three risk factors, 73% relapsed in a median of 17 months (median time to relapse among those who relapsed was 15 months), corresponding to a 3.7 times increased risk of relapse (95% CI: 1.4, 9.7, p = 0.007) compared to those with no risk factor.

Relapse typically occurs in the same organ system initially affected by the disease, although new organ system involvement occurs as well. Relapses in the kidney are heralded by the recurrence of microscopic hematuria, red blood cell casts, and worsening renal function. Fluctuations in the amount of proteinuria are not good indicators of active disease, and are related to glomerulosclerosis. Fortunately, a similar rate of response is achieved in the treatment of relapse and initial disease. Full-blown vasculitic relapse should be treated with a repeat course of prednisone and cyclophosphamide. In general, these patients require maintenance on long-term immunosuppression with azathioprine, MTX, mycophenolate mofetil, or cyclophosphamide. How to best treat milder relapses is a matter of substantial investigation. In an effort to limit the exposure of relapsing patients to repetitive cycles of cytotoxic drugs and their associated risks, alternative or adjunctive, less toxic therapies are being evaluated.

Prognostic Factors

Several studies have examined the question of prognostic factors in ANCA SVV. In our experience, the presence of pulmonary hemorrhage was the most important determinant of patient survival, whereas other pulmonary findings (eg, infiltrates, nodules, or cavities) did not increase the risk of death.

The risk of ESRD is largely determined by the degree of renal dysfunction at the time of diagnosis. Serum creatinine is the single most important prognostic marker for long-term renal outcome as exemplified by a 1.24-fold increased risk for ESRD for each 1 mg/dL increase in serum creatinine at baseline. Nevertheless, there is no threshold of renal dysfunction below which treatment is futile, as remission occurs in 57% of individuals with an estimated GFR <10 mL/minute. Histopathologic measures of chronic renal scarring (glomerulosclerosis, interstitial fibrosis, and tubular atrophy) have consistently been associated with poor renal outcomes. The impact of renal damage as a predictor of resistance emphasizes the importance of early diagnosis and prompt institution of therapy.