PERIPHERAL NERVE Vas ums / 247 TABLE 5. Differential diagnosis of muttfocial neuropathy ‘ischemic neuropathies 1. Peripheral nerve vascultis 2. Diabetes melitus 8. Diabet anyoopty rari! abet neuropathy) 'b. Mononeuropathy multiplex (cranial nerve, thoracio, limo) 3. Bland microvasculopathy of scleroderma 8. Intammatryimmune-mesated neupates* Sarcoidosis Multifocal demyelinating neuropathy with persistent Conduction Block Mutttocal motor neuropathy Multifocal variants of Guillain-Barré syndrome Idiopathic brachial or lumbosacral plexopathy Eosinophilic syndromes ‘a. Eosinophilia-myalgia syndrome b. Eosinophilic fascis (Shulman syndrome) C. Idiopathic hypereosinophilic syndrome C. Infection-felated neuropathies” 1. Leprosy 2. Lyme disease 3. Retroviral (HIV, HTLV-1) 4. Herpes virus (herpes-zoster, cytomegalovirus) 5. Infective endocarditis 6. Other (leptospirosis, hepatitis A, M. pneumoniae, ascatis, Plasmodium falciparum) . Drug-induced neuropathies 1. Antibiotics (penicilin, sulfonamides) 2. Cromaiyn 3. Thiouracil 4. Allopurinot 5. Interferon-alpha 6. Drugs of abuse (amphetamines, cocaine, heroin) 2 E Herediary neuropaties 1. Hereditary neuropathy with liabilty to pressure palsies (HNPP) Hereditary neuralgic amyotrophy (HNA) HNA with relapsing multfocal sensory neuropathy Porphyria Tangiers disease Traumatic neuropathies 1. Multiple peripheral nerve injuries 2. Burns 3. Multifocal entrapments ‘a. Diabetes melitus b. HNPP Neoplasia-related neuropathies 1, lnfitrative processes a. Non-Hodgkin's lymphoma (B > T cell) b. Leukemia (T, B, NK call) ©. Angiocentric lymphoproiiferative lesions (3) Lymphomatoid granulomatosis (2) Angiotropic Iymphoma (8 >> T cell) «. Hodgkin's lymphoma (rare) . Multiple myeloma (rare) {. Carcinoma (with leptomeningeal metastases) Mass lesions ‘a. Schwannomas (neurofibromatosis-1, b. Extranodal iymphoma ©. Chloroma (granulocytic sarcoma) ther poorly validated or rare associations Wartenbarg's migrant sensory neuritis, Sensory perineurtis Cholesterol emboli syndrome Idiopathic thrombocyfic purpura Gastrointestinal conditions (Crohn's, ulcerative colts, celiac sprue) 2 3 4 2) *Some associated with vasculis. HTLV, human T-cell lymphotropic virus type I; NK, natural Killer. tion (60-62), However, frank necrotizing vasculitis is rare in these patients and multifocal fiber loss can also be observed in other types of diabetic neuropathy (60-63). Moreover, a beneficial response to treatment in patients, with proximal diabetic neuropathy should be interpreted with caution because spontancous recovery frequently cccurs, even in those with inflammatory nerve lesions (64). Further studies are urgently needed to resolve these important issues. EVALUATION AND DIAGNOSIS, ‘Vasculitie neuropathy usually occurs in one of three clinical settings. The first is in the context of a previously diagnosed systemic vasculitis or another collagen vascu- lar disease. Therein, the diagnosis is straightforward, and careful consideration of the clinical, electrophysiologic, and pathologic data will confirm the diagnosis of vas culitic neuropathy. The second scenario is the presenta tion of neuropathy as the predominant manifestation of an undiagnosed systemic condition affecting multiple organ systems. The clinician then attempts to define the extent and distribution of the involved organs and deter- mines the most appropriate tissue to biopsy for diagnosis. In the third presentation, the most difficult to recognize, neuropathy is the only manifestation of vasculitis without other organ involvement. Although each of these clinical scenarios presents ina slightly different fashion, the eval- uation of a patient with a suspected vasculitic neuropathy should always include blood work, electrodiagnostic studies, and, in most cases, cutaneous nerve and muscle biopsy. The suggested routine diagnostic tests for patients ‘with an uncertain diagnosis are listed in Table 6, in addi- tion to other tests that are useful in certain specific clini- cal situations. Approximately 25% of patients with vas- culitic neuropathy will have cerebrospinal fluid protein elevation, but this finding is nonspecific and not usually helpful in making a definite diagnosis (5-7,9,10). Nerve conduction studies and needle electromyogra- phy are mandatory in the evaluation of vasculitic neu- ropathy to document the extent and severity of individual nerve involvement. Patients with distal symmetric248 / Cyarrer 20 TABLE 6. Laboratory stuales in patients with suspected vascultic neuropathy Test “Associated conditions ‘Nonspecific measures of inflammation and ‘organ involvemant Complete blood count ‘Anemia: all systomic vasculitides ~40-50% Leukocytosis: all systemic vascultides ~70% ‘Tarombocytosis: all systemic vasculitides 40-65% Erythrocyte sedimentation rate Chemistry panel Mild 1 in ~60% of isolated PNS vasculitis; marked 7 in ~85% of all systemic vascultides 1 BUN, creatinine: renal involvement 1 LFTs: liver involvement; hepatitis 1 Glucose: diabetes melitus Urinalysis Proteinuri hematuria, RBC casts: renal involvement or underlying ‘connective tissue diseasemalignancy Chest films lesions Antinuclear antibody (ANA) ‘vascultis Serum complements Rheumatoid factor More specific studies frequently useful Cryagiobulins Pulmonary involvement, sarcoidosis, malignancy, or lymphoproliforative (+)>90% connective tissue diseases; ~30% systemic and ~20% nonsystemic 4 in 360% SLE, RY, EMC, Sjégren’s; -25% classic PAN (4) 290% FV, EMC; 40-60% Wegener Sjégren's, PAN, MPA, CSS (+) in EMC/hepatits C, plasma cell dyscrasias, lymphoprolferative disorders, PAN, connective tissue diseases, hepatitis B, other chronic infections Eosinophil count Tin CSS (>>MPA, PAN), eosinophila-myaigia syndrome, eosinophilic fasciitis, hypereosinophilic syndrome, some infections and drug reactions Protein immunofixation, electrophoressis Anti-dsDNA (+) in SLE Anti-Flo and -La Anti-Sm, (4) in SLE Anti-Se 70 and -centromere Hepatitis B serologies Hepatitis C serology or RNA Glycosylated hemoglobin ‘More speciic studies obtained selectively HIV serology HTLV serology Lyme serology CANCA anti-proteinase 3) ‘Serum angiotensin-converting enzyme Monacional protein: plasma cell dyscrasia, lymphoma, leukemia (4) in SLE and Sjogren's syndrome (4) in scleroderma Surface antigen (+) in hepatilis B-associated PAN +) circulating antibodies and/or RNA in hepatitis C-associated EMC in diabetes melitus (4) in HIV infection, CMV multifocal neuropathy (9) in HTLV infection (4) in Lyme disease +) in Wegener's >> MPA, CSS (pANCA Is to0 nonspecific to be useful) In sarcoidosis (leprosy, chronic hepatitis, lymphoma, diabetes, FV, scleroderma) 'BUN, blood urea nitrogen; CANCA, classic antineutrophil oyfoplasmic autoantibody; CSS, Churg-Strauss syndrome; EMC, essential mixed cryoglobulinemia; HNPP, hereditary neuropathy with liability to pressure palsies; LFTs, liver function tests; MPA, microscopic polyangitis; PAN, polyartertis nodosa; pANCA, peri- fhuclear antineutrophil cytoplasmic autoantibody; PNS, periph nervous systam: RV, rheumatoid vas- ‘ulti; SLE, systemic lupus erythematosus; HTLV-1, human T-cell lymphotropic vitus type I; RBC, red blood coll. polyneuropathy on clinical examination have significant asymmetry on electrodiagnostic studies suggesting overlapping multiple mononeuropathy (5,65). Electrical studies can also establish a baseline for the degree and pattern of axon loss to serve as a reference in monitor- ing the patient’s response to therapy (2). This is particu- larly important in patients with severe disease because it may be difficult to assess their progression on clinical grounds alone, These studies also assist in the choice of the best nerve and muscle for biopsy (2). The specific findings on electrodiagnostic studies reflect the primary underlying pathology of vasculitic neuropathy, namely ischemic axonal damage that typically involves motor and sensory nerves (1-7,10,11,23,65). Compound sen- sory and motor action potentials are usually reduced in amplitude or unevokable, whereas conduction velocities are normal or mildly reduced and distal latencies are normal or mildly prolonged. Transient “conduction block” can be seen 6 to 10 days after ischemic injury but disappears as the axon degenerates (23,65). Persistent conduction block is uncommon in vasculitic neuropathy but ean occur as a result of ischemia-induced metabolic failure or secondary paranodal and segmental demyeli- nation (66-68). However, a predominance of demyeli- nating features is distinctly unusual and should prompt reconsideration of the diagnosis. The most commonly affected nerves in the legs are the peroneal and sural 2,410). The ulnar nerve followed by the median and radial nerves are the most frequently affected in the arms (23,65).Blectromyography reveals fibrillation potentials and positive sharp waves in about half of cases and decreased ‘motor unit recruitment in clinically affected muscles 2,455). Proximal muscles can be affected to a similar degree as more distal ones, because vasculitis is not a length-dependent neuropathy (4). Chronic “neurogenic” potentials can be seen in patients with long-standing dis- ease, Although the clinical picture, laboratory testing, and electrophysiologic data may strongly suggest vasculitic neuropathy, definitive diagnosis requires pathologic con- firmation in a sensory nerve. The sural and superficial peroneal nerves are most commonly chosen for biopsy; however, the latter is preferred because a peroneus brevis muscle biopsy can be obtained through the same incision (69). Combining nerve with a muscle biopsy increases the diagnostic yield, even in cases of “nonsystemic™ vas- culitic neuropathy (8). The sensitivity of diagnosing vas- culitic neuropathy by combined biopsy is about 60 to 70% (13,70). The pathologic appearance of vasculitis in a peripheral nerve varies with the age of the lesion. Active lesions are characterized by inflammatory cells that invade the vessel wall associated with necrosis (Fig. 2). Chronic lesions show evidence of vascular repair, includ- ing intimal hyperplasia with variable lumenal narrowing, thrombosis, vascular recanalization, and perivascular macrophages filled with hemosiderin (Fig. 3) (16.28,56,71). Endothelial cell disruption and hemor- rhage can occur but are not sufficient or required for diagnosis. Vasculitis of the nerve has a predilection for epineurial blood vessels with a diameter of 75 to 250 jim. For reasons that are unclear, endoncurial vessels are usu ally spared (4,7,10). ‘The pathologic changes in the nerve fascicles them- selves, although not pathognomonic, are often distinctive vvasculitic neuropathy showing perivascular and transmural ‘mononuclear cel inftration in an epineurial vessel produc Ing fibrinoid necrosis of the vessel wall (hematoxylin and eosin, 350) Bu FIG, 3. Superficial peroneal nerve biopsy from patient with vascultic neuropathy and more chronic vascular lesions. ‘There are several small recanalization channels in tho wall of, the inflamed and damaged vessel (hematoxylin and eosin, 500). enough to strongly support the diagnosis of vasculitis in the appropriate clinical setting. There is nearly always a loss of myelinated nerve fibers that occurs asymmetri- cally between and within fascicles (Fig. 4) (4,7,56,71,72). Multiple fibers undergoing Wallerian degeneration are typically present (7,72). Large myctinated fibers are more susceptible to ischemia, but small myelinated fibers and “unmyelinated fibers can also be affected in severe cases (72). In fulminant cases, Schwann cels, fibroblasts, and all other cellular clements may be lost, and the nerve can appear totally infarcted. Immunofluorescent staining is useful in confirming the diagnosis of vasculitis, Immune deposits of IgG, IgA, IgM, C3, and C5b-9 membrane FIG. 4, Superficial peroneal nerve specimen from patient with rheumatoid vasculitis and vascuitic neuropathy showing marked asymmetry in the degree of myelinated fiber loss. The fascicle on the right demonstrates severe myelinated fiber loss, whereas that on the left is less affected (epoxy resin-embeddad, toluidine blue stained, x 450).250 / Cuarter 20 FIG. 5. Immunotluorescont staining of sural nerve in a case of systemic necrotizing vasculitis showing a single epineurial vessel with transmural deposition of IgM (fluorescein-conju- (ated anti-human IgM, 375). attack complex (MAC) are found in approximately 80% of patient Fig 5) (910, 14,73) and appear tobe specific for vasculitis (74). Perls’ ferrocyanide test for perivas lar iron may also be a sensitive and specific marker for vaseulitie neuropathy (75), PATHOGENESIS Approximately SO% of vasculitie neuropathy cases occur in the setting of a systemic disease, most com- monly polyarteritis nodosa or rheumatoid arthritis (Table 3). Another 10% are associated with infection, malig- nancy, or drug ingestion, An antigen-antibody complex related to the systemic illness presumably triggers the immunologic processes that result in vessel damage. In the remaining 40%, the neuropathy occurs without an associated condition or etiology. Although in all of these settings, the pathogenesis is presumed to involve autoim- munity, the vasculitides must still be considered idio- pathic in that very little is understood concerning their etiology or the antigens that trigger the immunological processes (except for some of the secondary syndromes, such as polyarteritis nodosa related to hepatitis B) (22) The clinical, laboratory, and pathologic features of the various vasculitides are so protean that a single immunopathogenic mechanism cannot underlie all the syndromes and etiologies. It is likely that diverse mecha- nisms specific for the proximate disorder trigger and per= petuate equally diverse sequences of downstream immunologic events (76,77). ‘There has been remarkable progress in the past decade in the elucidation of effector mechanisms in systemic and peripheral nerve vasculitis (77,78). Table 7 summarizes the principal categories of immunologic phenomena that ‘an cause vasculitis. Although these may seem mutually exclusive, in fact there are many similarities, and more than one may operate simultaneously in a given patient to produce vasculitis, (On a fundamental level, the postulated immunopatho- genic mechanisms of vasculitis (Table 7) begin with anti- gen recognition by the host, initiating either a humoral or cellular immune response (or both), with the vascular endothelium assuming a pivotal role as the result of a dis- ruption in the normal homeostasis between circulating leukocytes and endothelial cells (79). Selected leukocytes adhere to the endothelial surface, the cell type and speci- ficity of which are determined by a complex network of, cellular adhesion molecules and soluble mediators that affect their expression (80,81). An equally complex array of eytokines, chemotactic agents, and other inflammatory regulators are locally elaborated that mediate leukocyte and endothelial cell activation, adherence of additional effector cells, and transendothelial migration of leuko- TABLE 7. Proposed pathogenic mechanisms in vascultides Involving peripheral nerves Category ‘Antigen Mechanisms Efiectors Examples Direct infection of Organism ‘Aclivation of inflammatory Neutrophils (bacteriay ___ Oylomegalovirus blood vessels ‘mediators, host's humoral Lymph, macrophages Herpes zoster ‘and cellular response (most viruses: Lyme) HIV, Lyme Cyfopathic etfect by virus Immune complexes Viruses, drugs, 9G, tumor Unknown in most Deposition of complexes in vessel walls, complement activation, effector cell Neutrophils early Monocytes and macrophages late Hypersensitivity Hep Beassoc PAN. Rheumatoid vasculitis recruitment ANCA-mediated Proteinase 3 Activation of circulating Neutrophils early Wegener's Myeloperoxidase neutrophils and monocytes; Eosinophils (CSS) MPA, CSS ‘endothelial cell activation Toell-mediates Unknown, Sensitization and activation of CD8+ cytotoxic T cells? Temporal arteritis (? vessel wall) CD44 T cells, release of Monocytes and Wegener's, CSS, pro-inflammatory cytokines ‘macrophages PAN Isolated PNS vasculitis ‘ANCA, anlinoulrophil oyloplasmic autoantibody, OSS, Churg-Strauss syndrome; MPA, microscopic polyangilis; PAN, polyarteris nodosa,