A 44-year-old woman with right ankle pain



Chief symptom

A 44-year-old woman with right ankle pain.

History of the present illness

Eleven months prior to her rheumatology clinic visit, the patient developed right ankle pain unrelated to weight bearing. Within a month, the right ankle pain had spread to the other ankle. The pain was associated with ankle swelling and the sensation of tingling in her feet. Two months after these initial symptoms, pain and stiffness developed in her hands and wrists. She denied swelling of these areas. The pain extended subsequently to her shoulders and elbows.

The patient was evaluated at a local urgent care center. She experienced pain in multiple upper and lower extremity joints, but there was no synovitis, soft tissue swelling, or signs of joint effusion. Despite additional symptoms of tingling and pain in her ankles and feet, the strength in her extremities was normal and the sensory examination was intact to light touch. A complete blood cell count and serum chemistry profile were both normal. The patient was sent home on naproxen and instructed to follow up with her primary care doctor.

The primary care doctor evaluated the patient 2 weeks later and reviewed additional serologies ordered in the urgent care center. The patient's serum was positive for rheumatoid factor (RF; titer 258 IU, normal value <30). Assays for antibodies to cyclic citrullinated peptides (anti-CCPs) and antinuclear antibodies were negative. The patient's erythrocyte sedimentation rate (ESR) was 96 mm/hour (normal value <20) and her serum C-reactive protein (CRP) concentration was 50.6 mg/liter (normal value <8.0). The primary care provider diagnosed rheumatoid arthritis (RA) and prescribed prednisone 30 mg/day, ibuprofen 800 mg 3 times daily, oxycodone 2.5 mg twice daily, and 2 tablets of acetaminophen with codeine at bedtime as needed.

The patient felt somewhat better on this regimen, but over the next 8 months she was unable to taper her prednisone dosage below 30 mg/day because of worsening pain in her extremities and sensations of numbness and tingling in her ankles and feet. She also continued her ibuprofen, oxycodone, and acetaminophen with codeine for pain control. After months of this regimen, the patient developed a purpuric rash over her right lower leg. She was referred to the rheumatology clinic.

The patient's past medical history was remarkable for vitiligo, which she had had for years. The vitiligo had involved large areas over her distal arms, hands, legs, feet, and face. The family history was negative for autoimmune and inflammatory conditions. She did not smoke or drink. She had 3 children, was divorced, and worked as a housekeeper.

On physical examination, the patient had a blood pressure of 143/84 mm Hg. Her temperature was 98.0°C, her respiratory rate was 16/minute, and her pulse was 66 beats/minute. There were no findings on examination of the head, eyes, ears, nose, mouth, neck, heart, lungs, abdomen, or back. However, examination of her ankles revealed severe tenderness of the skin of both ankles. There was no synovitis, joint effusion, or soft tissue swelling. Her skin examination showed a prominent livedo racemosa rash over her lower extremities (Figure 1A), a palpable purpuric lesion on the right lower leg (Figure 1B), and diffuse hypopigmented areas secondary to vitiligo (Figure 1C). Some of the larger purpuric lesions had areas of central necrosis and ulceration.

Figure 1.

A, Prominent livedo racemosa over the patient's lower extremities. B, A palpable purpuric lesion in the right lower leg. C, Diffuse hypopigmented areas due to vitiligo.

Her neurologic examination revealed numbness in the lower extremities distal to the ankles, corresponding to the symptoms of pain and tingling. The strength of all of the muscle groups was 5/5 bilaterally, with no indication of either a foot or wrist drop. The deep tendon reflexes were diminished symmetrically at the ankles.

The patient's full serologic evaluation and other laboratory test results are shown in Table 1.

Table 1. Initial outpatient laboratory evaluation*
ParameterOutpatient rheumatology clinic evaluationNormal range
  • *

    BUN = blood urea nitrogen; ESR = erythrocyte sedimentation rate; CRP = C-reactive protein; RF = rheumatoid factor; anti-CCP = anti–cyclic citrullinated peptide; ANAs = antinuclear antibodies.

Hematocrit, %33.141.0–53.0
Hemoglobin, gm/dl10.813.5–17.5
White cell count, per mm314.54.5–11.0
Differential count, %  
Platelet count, per mm3414150–350
Mean corpuscular volume, μm37980–100
Iron, μg/dl1945–160
Iron-binding capacity, μg/dl437228–428
Sodium, mmoles/liter136135–145
Potassium, mmoles/liter4.03.4–4.8
Chloride, mmoles/liter101100–108
CO2, mmoles/liter27.223.0–31.9
BUN, mg/dl218–25
Creatinine, mg/dl0.720.6–1.5
Glucose, mg/dl13470–110
Calcium, mg/dl9.08.5–10.5
Phosphorous, mg/dl3.52.6–4.5
Albumin, gm/dl3.93.3–5.0
Globulin, gm/dl3.82.6–4.1
Total protein, gm/dl7.76.0–8.3
Creatine kinase, units/liter4660–400
ESR, mm/hour39<20
CRP level, mg/liter16.1<8
RF, IU/ml487<30
Anti-CCP antibodies, units5<20
ANAs1:40, speckled pattern


This 44-year-old woman presented with severe pain, tingling, and numbness in her lower extremities. These symptoms preceded the onset of purpuric lesions, livedo racemosa, and cutaneous ulceration by at least 9 months. She was strongly positive for RF and had elevated acute-phase reactants.


The differential diagnosis is shown in Table 2. The first question is whether or not the patient's given diagnosis of RA is a satisfactory explanation for her presentation. RA complicated by rheumatoid vasculitis could cause the patient's serologic findings and her extremity pain, as well as her livedo racemosa and cutaneous ulcerations. Rheumatoid vasculitis strongly resembles polyarteritis nodosa (PAN) in its tendency to involve medium-sized blood vessels. This patient has several hallmarks of rheumatoid vasculitis: symptoms compatible with a sensory neuropathy, livedo racemosa, and cutaneous ulceration involving the lower extremities (1–3).

Table 2. Differential diagnosis*
  • *

    SLE = systemic lupus erythematosus; ANCA = antineutrophil cytoplasmic antibody.

Rheumatoid arthritis with rheumatoid vasculitis
Other causes of positive rheumatoid factor
 Cryoglobulinemia/hepatitis C
 Sjögren's syndrome
 Other connective tissue disorders (SLE, inflammatory myopathy, systemic sclerosis)
 Waldenström's macroglobulinemia
Livedoid vasculopathy
Antiphospholipid antibody syndrome
Primary systemic vasculitis
 Polyarteritis nodosa
 ANCA-associated disease
  Churg-Strauss syndrome
  Wegener's granulomatosis
Microscopic polyangiitis

Rheumatoid vasculitis typically occurs in patients with years of RF positivity and aggressive joint inflammation. The occurrence of rheumatoid vasculitis within the first year of the onset of RA is unlikely, albeit reported (4). Moreover, the absence of anti-CCP antibodies is evidence against both RA and rheumatoid vasculitis (5). Patients can have RA without anti-CCP antibodies (and, indeed, without RF), but anti-CCP antibodies are a marker for patients at risk for extraarticular disease manifestations such as vasculitis. The early appearance of rheumatoid vasculitis in the absence of anti-CCP antibodies is further unlikely (6, 7).

More importantly, pain in the extremities and seropositivity for RF are insufficient for the diagnosis of RA. This patient lacks the sine qua non of RA; specifically, arthritis. Several physical examinations have failed to document synovitis. The nature of her pain is more suggestive of a sensory neuropathy than an inflammatory arthritis. Therefore, other causes of RF positivity and systemic vasculitis must be considered.

Conditions associated with RF positivity

RF is an autoantibody directed against the Fcγ region of the immunoglobulin molecule. The etiology of RF production is unclear, but RF is associated with a number of rheumatologic diagnoses other than RA, as well an array of nonrheumatic diseases. Some of these common conditions are summarized in Table 3 (8, 9). Some RF-associated conditions present with organ-specific symptoms such as sicca symptoms (Sjögren's syndrome) or muscle weakness (polymyositis/dermatomyositis). Others produce nonspecific systemic manifestations such as low-grade fevers and fatigue (subacute bacterial endocarditis). Our patient had no such features. However, 2 conditions associated with high titers of RF are worthy of further consideration: hepatitis C virus (HCV)–associated mixed cryoglobulinemia and sarcoidosis.

Table 3. Common diseases associated with positive rheumatoid factor
 Frequency of occurrence, %
Autoimmune disease 
 Rheumatoid arthritis60–80
 Primary Sjögren's syndrome70–90
 Mixed cryoglobulinemia (hepatitis C)70
 Systemic lupus erythematosus30
 Mixed connective tissue disease25
 Systemic sclerosis20
 Subacute bacterial endocarditis40
 Hepatitis (A, B, and C)25
 Epstein-Barr virus and cytomegalovirus infections20
 Waldenström's macroglobulinemia25
 Liver cirrhosis25
 Interstitial lung diseases25

HCV–associated mixed cryoglobulinemia

HCV infection is the second most common chronic viral infection in the world, following chronic hepatitis B infection (10). Although HCV primarily affects the liver, it is associated with a wide spectrum of extrahepatic manifestations, including arthralgias, myalgias, sicca symptoms, mixed cryoglobulinemia, autoantibody production, and lymphoproliferative disease (11, 12). The term “mixed” cryoglobulinemia stems from the fact that sera from these patients generally contain excess quantities of both IgG and IgM antibodies (hence the term “mixed”). These antibodies behave as cryoglobulins, i.e., in the laboratory, they precipitate from the serum in conditions of extreme cold, such as those 4°C. The IgM component in mixed cryoglobulinemia typically has the properties of RF: it binds to the Fc portion of IgG. Therefore, essentially all patients with mixed cryoglobulinemia are RF positive, many strikingly so (13).

Cryoglobulinemic vasculitis develops in approximately 5–10% of patients with mixed cryoglobulinemia associated with HCV (14). This complication often manifests itself as arthralgia, palpable purpura, peripheral neuropathy (sensory greater than motor), and glomerulonephritis (11). The palpable purpura typically begins in the lower extremities but later extends to the abdomen and upper extremities. Light microscopy examination of skin biopsy specimens in mixed cryoglobulinemia shows a nonspecific leukocytoclastic vasculitis involving the small and medium-sized vessels, accompanied often by fibrinoid necrosis within the arteriolar walls and endovascular thrombi (11). Direct immunofluorescence studies of skin lesions reveal the deposition of IgG, IgM, and complement components within blood vessel walls, serving to differentiate the condition from other types of immune complex–mediated vasculitis (e.g., Henoch-Schönlein purpura) and pauci-immune disease (e.g., those forms of vasculitis associated with antineutrophil cytoplasmic antibodies [ANCAs]). Our patient had a number of features that suggested cryoglobulinemic vasculitis, including joint pain in the setting of a high titer of RF, ulcerated cutaneous lesions, and peripheral neuropathy.


Sarcoidosis is characterized by the formation of non-caseating epithelioid cell granulomas in an array of organs (15). The clinical presentation of sarcoidosis varies tremendously and symptoms are often nonspecific. Fever, weight loss, sicca symptoms, joint pain, skin lesions, dyspnea, and lymphadenopathy are common (15, 16). A common finding in sarcoidosis is a periarthritis that has a predisposition for the ankles, reminiscent in some respects of our patient's symptoms at presentation. Approximately 28% of patients with sarcoidosis are RF positive (15). Sarcoidosis can be complicated by the concomitant occurrence of autoimmune diseases such as Sjögren's syndrome, RA, lupus, or systemic sclerosis (16, 17). The most characteristic skin lesion of sarcoidosis is erythema nodosum, which does not ulcerate (in contrast to our patient's lesions). In addition, livedo racemosa strongly suggests a primary vascular process and is an unusual finding in sarcoidosis. Although this patient presented with ankle pain and a positive RF, the demonstration of non-caseating granulomas on a tissue biopsy sample would be essential to establish that diagnosis.

Several other categories of disease can lead to cutaneous findings similar to those of our patient. These include livedoid vasculopathy, antiphospholipid syndrome (APS), and primary systemic vasculitides.

Livedoid vasculopathy

Livedoid vasculopathy is a rare skin disorder characterized by recurrent ulceration of the lower extremities. It is often painful and has a strong association with livedo racemosa (18, 19). The skin findings in livedoid vasculopathy include ulceration and atrophic, porcelain-white scars that are known as atrophie blanche. Skin biopsy samples in livedoid vasculopathy show intraluminal thrombosis, endothelial proliferation, and segmental hyalinization of the subintimal layer of dermal vessels (19). The clinical lesions of livedoid vasculopathy often mimic those of systemic vasculitis, particularly PAN, because of their propensity to cause nodules and skin ulcers. Mononeuritis multiplex has been reported in livedoid vasculopathy, albeit only in a minority of cases. Livedoid vasculopathy can be caused by a variety of both inflammatory and noninflammatory disorders, prominent among which is a lengthy list of hypercoagulable states (19). Treatment of livedoid vasculopathy usually targets thrombotic processes rather than primary inflammatory ones.

The diagnosis of livedoid vasculopathy is based on clinical presentation, laboratory findings, and skin biopsy samples. Many of our patient's clinical characteristics were compatible with livedoid vasculopathy, including her age (the mean age for livedoid vasculopathy is 45 years), her sex (the female to male ratio in livedoid vasculopathy is ∼2.5:1), the presence of livedo racemosa, and the finding of ulcerated lesions in the lower extremities. On the other hand, high titers of RF are unusual in livedoid vasculopathy (20).

Antiphospholipid syndrome

APS has a strong association with livedo racemosa. APS is caused (through imprecisely-defined mechanisms) by autoantibodies directed against phospholipids, leading to recurrent arterial or venous thrombosis, pregnancy complications, thrombocytopenia, and a host of other potential manifestations. The clinical manifestations of APS affect almost every organ or system in the body. The most common are cutaneous: leg ulcers, thrombophlebitis, skin necrosis, peripheral ischemia, Raynaud's phenomenon, and livedo racemosa.

APS is a vasculopathy rather than a true vasculitis (21). The diagnosis of APS is based on both typical clinical features and the detection of antiphospholipid antibodies. The presence of antibodies without clinical events is insufficient for the diagnosis of APS (22). Although our patient presented with ulcerated skin lesions and striking livedo racemosa, she did not have any clinical history of thromboembolism to support a diagnosis of APS at the time of the visit. In addition, she had had 3 children without any history of spontaneous abortion. Nevertheless, testing for anticardiolipin antibodies, a lupus anticoagulant, and antibodies to β2-glycoprotein I is appropriate in this patient.

Primary systemic vasculitis

The primary systemic vasculitides are classified by the size of involved vessels into large-, medium-, and small-vessel conditions. In the large-vessel vasculitides, giant cell arteritis and Takayasu arteritis, cutaneous findings are uncommon because the skin does not contain large blood vessels. Cutaneous lesions resulting from medium-vessel disease, however, are reported occasionally. For similar reasons, peripheral neuropathies are highly unusual in large-vessel vasculitides: the vasa nervorum are generally smaller-sized blood vessels. In contrast, the medium- and small-vessel vasculitides are quite likely to involve the skin, and several often cause vasculitic neuropathy. The medium-vessel and small- to medium-vessel vasculitides that can affect both the skin and peripheral nerves are cryoglobulinemic vasculitis (discussed above), PAN, Churg-Strauss syndrome (CSS), Wegener's granulomatosis (WG), and microscopic polyangiitis (MPA).

Polyarteritis nodosa

PAN often poses difficult diagnostic challenges, owing in part to the lack of specific laboratory markers of disease (e.g., autoantibodies). A significant proportion of patients with PAN are RF positive, but high titers of RF as in our patient are unusual (23). According to the 1990 American College of Rheumatology (ACR) classification criteria for PAN, patients with systemic vasculitis can be classified as having PAN if they meet at least 3 of the following 10 criteria: 1) weight loss ≥4 kg, 2) livedo reticularis (i.e., racemosa), 3) testicular pain or tenderness, 4) muscle pain, weakness, or leg tenderness, 5) mono- or polyneuropathy, 6) diastolic blood pressure >99 mm Hg, 7) elevated serum urea and creatinine levels, 8) serologic evidence of hepatitis B virus infection, 9) arteriogram abnormality, and 10) a biopsy sample of a small or medium-sized artery containing polymorphonuclear neutrophils (24).

The dermatologic symptoms of cutaneous PAN are varied, including nodules, ulcers, and livedo racemosa (25). Cutaneous PAN, a disorder marked by skin findings and occasional sensory neuropathies, is sometimes considered to be a condition separate from classic PAN, which has striking involvement of internal organs (25). Patients with cutaneous PAN often present with malaise, fever, or arthralgia, but what distinguishes it from other forms of PAN is the absence of visceral involvement, often indicated by a negative renal and mesenteric angiogram (25).

Classification criteria for PAN are nonspecific and perform better if other forms of vasculitis with more distinct features are excluded first (24). It has been proposed that criteria for PAN should be applied only when all 3 forms of vasculitis associated with ANCAs, i.e., CSS, WG, and MPA, have been excluded.

Churg-Strauss syndrome

CSS is often termed an ANCA-associated vasculitis (AAV), although some studies indicate that only 50% or fewer of patients have ANCAs in their serum (26). The cardinal features of CSS are eosinophil-rich, granulomatous inflammation that involves the respiratory tract; an accumulation of eosinophils in extravascular areas of the medium-sized and small vessels; and a peripheral eosinophilia that exceeds 10% of the total white blood cell count. Vasculitic neuropathy occurs in up to 80% of patients with CSS, but the absence of eosinophilia in our patient excludes CSS from serious consideration.

Wegener's granulomatosis

Diagnostic criteria for WG do not exist (just as they do not exist for any form of primary systemic vasculitis) (26). However, both the 1990 ACR classification criteria for WG and the 1992 Chapel Hill Consensus Conference provided specific guidelines for considering the diagnosis for the purpose of research studies (27, 28). The ACR criteria focused on the common occurrence of WG as a granulomatous vasculitis in the upper respiratory tract, lungs, kidneys, and skin. Two or more positive criteria have a sensitivity of 88.2% and a specificity of 92.0% for WG (27). The Chapel Hill Consensus Conference definition, on the other hand, focused on histopathologic features, emphasizing the presence of granulomatous inflammation on tissue biopsy samples (28). Of the 3 forms of AAV, WG has the strongest tendency to involve the upper respiratory tract and is the only one that has a predilection for ANCA directed against serine proteinase 3 (PR3; i.e., PR3 ANCA). Whereas 75–85% of patients with WG are PR3 ANCA positive, ∼10% have ANCAs directed against myeloperoxidase. In most studies, 10–15% of patients with WG are ANCA negative.

The results of ANCA testing in this patient were not known at this point in the case. However, the absence of physical findings and any other evidence of disease in the upper respiratory tracts, lungs, and kidneys provide compelling evidence against the diagnosis of WG.

Microscopic polyangiitis

MPA was originally described as a microscopic form of PAN that exhibited crescentic glomerulonephritis (29). The Chapel Hill Consensus Conference clearly separated MPA from PAN and emphasized the term “microscopic polyangiitis” rather than “microscopic polyarteritis nodosa” (25, 28). More than 90% of patients with MPA are ANCA positive (30, 31). Among ANCA-positive cases, perinuclear ANCA/myeloperoxidase ANCA is the predominant pattern, varying in frequencies between 65% and 93% (30–33). The remainder of ANCA-positive MPA patients are cytoplasmic ANCA/PR3 ANCA positive.

Up to 90% of patients with MPA develop renal disease (26, 32–34). Presentations of renal dysfunction in MPA range from mild proteinuria and microscopic hematuria with normal serum creatinine measurements to rapidly progressive renal failure as a result of crescentic glomerulonephritis (25). Approximately 58% of patients with MPA present with vasculitic neuropathy (26). The fact that our patient presented without evidence of renal or pulmonary dysfunction despite 11 months of clinical symptoms casts some doubt on MPA as the diagnosis. However, despite the common view of MPA as a “pulmonary-renal syndrome,” only 12% of patients have clinically evident alveolar hemorrhage (26, 32, 33). MPA is distinguished from the other forms of AAV by its absence of granulomatous inflammation.

Secondary vasculitides

The differential diagnosis in cases such as this must include consideration of secondary causes of vasculitis. Some infectious causes of secondary vasculitis, e.g., the associations of hepatitis B and hepatitis C with PAN and mixed cryoglobulinemia, have been considered above. Malignancies such as lymphoma and Waldenström's macroglobulinemia must be excluded with appropriate testing.


The patient underwent a hypercoagulable evaluation that revealed a normal prothrombin time and partial thromboplastin time and normal concentrations of protein C, protein S, anti–thrombin III, and homocysteine. Her lupus anticoagulant assay was negative and the serum anticardiolipin IgG level was normal (8.0 IgG phospholipid units, normal range 0–15). The IgM anticardiolipin antibody was marginally elevated (16.3 IgM phospholipid units, normal range 0–15). Her hepatitis panel, rapid plasma reagin test, cryoglobulins, anti–double-stranded DNA antibodies, and extractable nuclear antibodies were all negative. Her total complement levels, C3, and C4 were normal. Radiographic studies, including a chest radiograph and computed tomography scans of the chest, abdomen, and pelvis, were unremarkable. Electrodiagnostic studies of the peripheral nerves and muscles showed a severe axonal sensorimotor polyneuropathy that primarily affected both lower extremities, the left side slightly more so than the right.

The patient developed more palpable purpura and ulcerated lesions on her lower extremities (Figure 2). She also experienced severe pain in the lower legs, ankles, and feet. Her ESR increased from 39 to 65 mm/hour, and her CRP level increased from 16.1 to 224.4 mg/liter. A diagnostic test was performed.

Figure 2.

Two skin biopsy sites on the right lower leg.


The diagnostic test was a skin biopsy. Biopsies were performed on 2 of the purpuric lesions on the right lower leg (Figure 2). At low magnification, prominent thrombosis was observed in a medium-sized vessel in the dermal layer (Figure 3A). At higher magnifications, focal necrotizing inflammation was seen within the muscularis layer of a medium-sized vessel (Figure 3B). More diffuse necrotizing inflammation was also observed in a small vessel (Figure 3C). Although leukocytoclastic infiltrates were seen in both intravascular and perivascular areas, no granulomatous inflammation was detected. Prominent thrombosis was seen in both small and medium-sized blood vessels within the dermal layer (Figure 3D). Immunofluorescence studies showed positive staining of the epidermal vessels for IgG (whole) and C3 and staining of the dermal vessels for IgM and C3 (data not shown). The combination of a mixed medium- and small-vessel vasculitis, the absence of granulomatous inflammation, and the finding of only moderate immunoreactant deposition were all consistent with MPA. This diagnosis was substantiated by the results of the patient's serum assay for ANCAs, which demonstrated a high titer of myeloperoxidase ANCAs (33.8 units, normal value <2.8).

Figure 3.

Skin biopsy sample of a lower extremity lesion. A, At low magnification, prominent thrombosis is observed in a medium-sized vessel in the dermal layer. B, Focal necrotizing inflammation within the muscularis layer of a medium-sized vessel. No granulomatous inflammation is present. C, Diffuse necrotizing within the walls of a small blood vessel. No granulomatous inflammation is present. D, Prominent thrombosis in both small and medium-sized blood vessels within the dermis. The combination of a mixed medium- and small-vessel vasculitis, the absence of granulomatous inflammation, and the finding of only moderate immunoreactant deposition on immunofluorescence are all consistent with microscopic polyangiitis.


Microscopic polyangiitis.


This case highlights a number of points about the diagnosis of systemic vasculitis and MPA. First, MPA does not always present as a pulmonary-renal syndrome. The fact that our patient presented without evidence of either renal or pulmonary dysfunction despite 11 months of clinical symptoms cast some doubt on the diagnosis of MPA. However, despite the common view of MPA as a pulmonary-renal syndrome, only 12% of patients have clinically evident alveolar hemorrhage (26, 32, 33). Second, the fact that patients with MPA can be strongly positive for RF is often underappreciated. Our patient's RF positivity led first to the misdiagnosis of RA and then to the strong consideration of mixed cryoglobulinemia. One study estimated the prevalence of RF within the serum of WG patients to be 50% (34). The prevalence of RF positivity among patients with MPA has been reported to be 23% (32).

Third, MPA and other forms of AAV often demonstrate a prominent component of thrombosis on tissue biopsy samples. In fact, the abundance of thrombotic lesions led our pathologists to focus initially on the possibility of a thrombotic microangiopathy such as that associated with APS. Only the findings of fibrinoid necrosis within the biopsy sample, the results of serologic testing for ANCA, and the recognition that AAV can be associated with a significant degree of thrombosis led to the correct diagnosis. The potential for AAV to cause thrombosis is likely reflected in its association with certain venous thrombotic events. An increased incidence of deep venous thrombosis and pulmonary emboli has been confirmed in both WG and CSS (32, 35).

MPA generally requires 2 drugs, both high-dose glucocorticoids and cyclophosphamide, to induce remission (36). MPA is treated now with a 3–6-month course of cyclophosphamide, often followed by azathioprine during a remission maintenance period of approximately 1 year (i.e., 15–18 months of total therapy from the time of diagnosis) (37). Glucocorticoids can often be tapered to discontinuation over 6–9 months. A recent randomized controlled trial demonstrated that the combination of rituximab and glucocorticoids was not inferior to cyclophosphamide and glucocorticoids for the induction of remission in AAV, and that the rituximab regimen was superior to the cyclophosphamide regimen for patients who presented with relapsing disease at baseline (38).


The patient was treated with cyclophosphamide 2 mg/kg/day (150 mg), along with a tapering course of prednisone beginning at 50 mg/day. Cyclophosphamide was discontinued in favor of azathioprine (2 mg/kg/day) after 6 months. Prednisone was tapered off after a 9-month course. The ulcerated lesions on her lower legs healed. The pain and numbness in her lower extremities improved slowly over the next year. In addition, the patient's vitiligo also responded dramatically to her vasculitis therapy. The affected portions of the skin on her face and upper body resumed normal skin pigmentation.


All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be submitted for publication. Dr. Stone had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Chou, Mihm, Stone.

Acquisition of data. Chou, Mihm, Stone.

Analysis and interpretation of data. Chou, Mihm, Stone.