Medium and Small Vessel Vasculitis

Epidemiology

PAN is a rare disease with an estimated annual incidence of 0.9 to 9.7 per million. Prevalence estimates range from 9 to 31 per million, based on studies from Europe. In a study from a hyperendemic area for HBV, an incidence of 7.7 per 100,000 was reported, but a more recent study found that HBV-associated PAN may be declining. In a cohort of 341 cases of PAN, HBV-associated PAN accounted for 17.4% of all cases of PAN in patients diagnosed between 1990 and 2002 compared to earlier decades, when prevalence of HBV among patients with PAN was as high as 52%. There appears to be a slight male predominance in PAN.

Etiology

Based on recent developments, it has been proposed that necrotizing vasculitis represents a broad range of diseases with different etiopathogeneses. While most cases of PAN are idiopathic, the association of HBV with PAN has been well documented. Hepatitis C virus (HCV)-associated PAN has also been reported, but is infrequent. The availability of ANCA testing has allowed us to distinguish PAN from other forms of systemic necrotizing vasculitis like microscopic polyangiitis (MPA), which can cause similar clinical manifestations. Medications like minocycline can rarely cause both c-PAN and systemic PAN. Recently, loss of function mutations in the cat eye syndrome chromosome region, candidate 1 ( CECR1 ) gene, which encodes for adenosine deaminase 2, has been associated with an autoinflammatory condition characterized by early onset vasculopathy with clinical and histopathologic features of PAN, as well as hemorrhagic and ischemic strokes.

Pathogenesis and Histopathology

In the case of HBV-associated PAN, circulating immune complexes or direct injury of the vessel wall from viral replication have been implicated. In contrast, the pathogenesis of the idiopathic or primary form of PAN is not well understood. Immunohistochemistry of the affected biopsy tissue reveals macrophages and T cells in the inflammatory infiltrates, suggesting PAN is a T-cell mediated disease. Elevated circulating soluble adhesion molecules have also been reported and activation of endothelial cells is also thought to play a role in disease pathogenesis.

PAN is segmental and predominates at vessel branching points. Fibrinoid, transmural necrosis of the vessel wall with a mixed inflammatory infiltrate consisting of macrophages, lymphocytes, neutrophils, and eosinophils is observed. Temporal variability of lesions with both active necrotizing lesions and healed or fibrotic lesions representing different stages of the inflammatory process can be seen concurrently in the same specimen. Thrombosis can lead to vascular occlusion, while severe vessel wall injury may result in the formation of microaneurysms.

Clinical Manifestations

Constitutional symptoms including fever, weight loss, myalgias, and arthralgias are frequently observed. In a large series of 348 patients with PAN, the most frequently affected organs were nerves, genitourinary system, and the skin. Mononeuritis multiplex or sensorimotor peripheral neuropathy can occur in up to 74% of patients with PAN. Renal manifestations may include new onset hypertension, proteinuria, and hematuria while testicular tenderness or orchitis has been reported in about 17% patients. PAN does not cause glomerulonephritis; therefore, the presence of an active urine sediment and dysmorphic red blood cells in the urine should prompt evaluation for small vessel vasculitis, mainly ANCA-associated vasculitis. The most frequent cutaneous manifestations are purpura followed by subcutaneous nodules and livedo reticularis ( Fig. 41.1 ). In severe cases, digital ischemia from peripheral arterial occlusions or ischemic ulcerations can also occur. GI manifestations may occur in 38%, with the most frequent manifestation being abdominal pain. However, mesenteric ischemia, hemorrhage, bowel perforation, cholecystitis, appendicitis, and pancreatitis have all been described, and may require surgical intervention. Cardiac involvement with cardiomyopathy and pericarditis have been reported in 5% to 8% of patients with PAN. Pulmonary manifestations including lung infiltrates and pleural effusions are rare, being noted in only about 3% of patients. Indeed, the presence of pulmonary abnormalities in a patient with suspected vasculitis should prompt careful evaluation for ANCA-associated vasculitis. Although rare, PAN may cause ocular manifestations, including retinal vasculitis.

Diagnosis

The 1990 American College of Rheumatology (ACR) classification criteria for PAN are listed in Table 41.1 . However, these criteria predate the discovery of ANCA and are therefore of limited utility in clinical practice. There are ongoing collaborative efforts between the ACR and European League Against Rheumatism (EULAR) for the development of diagnostic and classification criteria for several forms of systemic vasculitis including PAN. In a study evaluating 949 patients with systemic vasculitis (262 with PAN), positive predictors for PAN were presence of hepatitis B viral antigen, arteriographic abnormalities (arteriogram showing aneurysms or occlusions of the visceral arteries not due to arteriosclerosis, fibromuscular dysplasia, or other noninflammatory causes), and mononeuropathy or polyneuropathy while factors that made PAN less likely were presence of ANCA, glomerulonephritis, asthma, ear/nose/throat signs, or cryoglobulinemia.

Laboratory testing in patients with PAN often shows findings of systemic inflammation with elevated sedimentation rate, C-reactive protein, anemia, and leukocytosis. Testing should include evaluation of renal parameters (serum creatinine, urinalysis), liver function testing, chronic hepatitis serologies (HBV and hepatitis C), muscle enzymes, and markers of other forms of systemic vasculitis including ANCA, antinuclear antibody (ANA), cryoglobulins, complement C3, complement C4, and rheumatoid factor (RF). Testing for human immunodeficiency virus (HIV) should be considered in the appropriate clinical context. Other diagnostic testing may include nerve conduction studies and electromyography in cases of suspected neurologic involvement. As there is no diagnostic laboratory marker for PAN, imaging or tissue diagnosis in the form of a biopsy is often essential in the diagnostic evaluation. While computed tomography (CT) of the abdomen can be useful in detecting organ infarcts ( Fig. 41.2 A), hemorrhage, bowel wall thickening or ischemia, these findings are fairly nonspecific and therefore visceral angiography is often needed. Characteristic findings on conventional angiography such as multiple aneurysms (fusiform, sacular, or microaneurysm) of the celiac, mesenteric, and renal artery branches can be diagnostic for PAN (see Fig. 41.2 B and C). Biopsies from affected areas of involvement including skin, muscle, or nerve can allow the histopathologic confirmation of vasculitis. Surgical specimens from patients with visceral involvement (e.g., small intestine, gallbladder) may also provide the diagnosis. Diagnostic confirmation is important, given the medications used for treatment are associated with significant adverse effects.

Treatment and Prognosis

Treatment of PAN should be tailored according to underlying cause (if known), disease severity, and organs that are affected. Patients with idiopathic systemic PAN are generally treated with high-dose glucocorticoids (pulse and/or oral) and cyclophosphamide for induction of remission. Cyclophosphamide is usually given orally at a dose of 1.5 to 2 mg/kg/day or by monthly intravenous infusion (600 to 750 mg/m ). The dose should be carefully adjusted depending on response to therapy, renal function, and hematologic parameters. Treatment with cyclophosphamide is typically continued for at least 6 months, after which it is replaced with a remission maintenance agent such as methotrexate or azathioprine, for total treatment duration of 12 to 18 months. One study showed superiority of treatment with 12 cycles of intravenous pulse cyclophosphamide compared to 6 months of treatment. Although benefit was seen early in the disease course, this was not sustained and at 10 years, there was no difference in survival or disease outcomes between the two groups. In less severe cases of PAN, glucocorticoid monotherapy at a dose of 1 mg/kg/day for 4 weeks followed by a gradual taper over months may be reasonable with the addition of adjunctive immunosuppressive therapy if needed in relapsing cases.

For HBV-related PAN, recommendations for treatment are use of a short course of glucocorticoids (2 weeks) along with antiviral agents to treat the underlying HBV infection, and plasma exchange to remove circulating immune complexes. Treatment should be aimed at achieving seroconversion and stopping viral replication which, generally, results in remission of the disease and prevents long-term hepatic complications. Prolonged glucocorticoid therapy and cyclophosphamide should be avoided in order to allow immunological clearance of HBV-infected hepatocytes.

The treatment of c-PAN also depends on disease severity. Mild cases may respond to nonsteroidal antiinflammatory medications, colchicine, or dapsone, while refractory disease often requires additional immunosuppressive therapy. In the case of isolated SOV that is detected incidentally at surgery, resection is thought to be curative.

The presence of renal, GI, cardiac, or central nervous system involvement in systemic necrotizing vasculitis portends a poorer prognosis with increased risk of mortality. The “Five-Factor Score,” which includes creatinine of > 1.58 mg/dL, proteinuria > 1 g/24 h, cardiac symptoms, central nervous system, or GI involvement, can be used to estimate disease severity and outcomes in patients with systemic necrotizing vasculitis. A higher “Five-Factor Score” has been associated with increased mortality—11% in patients with a score of 0 compared to 46% in those with a score of > 2. Death in the first year was usually associated with poorly controlled vasculitis. In one study, patients with HBV-associated PAN had fewer relapses but more severe disease and increased mortality (34%) compared to non–HBV-associated PAN (20% mortality). In a French cohort study, GI manifestations were the most frequent cause of death in patients with HBV-associated PAN.