Advertisement
Mayo Clinic Proceedings Home

Cocaine-Induced Pseudovasculitis

      Pseudovasculitis is a disease process that mimics the presentation and possibly the laboratory findings of true vasculitis. However, biopsy specimens do not reveal the typical histopathologic findings expected in vasculitis. One often over-looked cause of pseudovasculitis is cocaine use, which has been described in case reports to cause aggressive nasal destruction and various skin lesions and thus has been confused with Wegener granulomatosis or leukocytoclastic vasculitis. Unfortunately, serologic tests such as antinuclear antibody or anti-neutrophil cytoplasmic antibody cannot reliably differentiate between these entities. We describe a patient who presented with what was believed to be Wegener granulomatosis affecting the skin and upper airway. However, findings from repeated biopsies did not support this diagnosis, and the only unifying diagnosis was cocaine-induced pseudovasculitis. The ability to recognize and differentiate between true vasculitis and pseudovasculitis is essential for the clinician because treatment options are radically disparate.
      c-ANCA (cytoplasmic antineutrophil cytoplasmic antibody), ELISA (enzyme-linked immunosorbent assay), IIF (indirect immunofluorescence), p-ANCA (perinuclear antineutrophil cytoplasmic antibody), PR3 (proteinase 3), WG (Wegener granulomatosis)
      A typical constellation of findings on patient history, physical examination, and laboratory testing may suggest the diagnosis of a vasculitic disorder. However, infection, as well as embolic or thrombotic phenomena, can mimic the presentation and systemic nature of vasculitis. Because biopsy specimens do not reveal the diagnostic histopathologic findings expected in vasculitis, these mimicking disorders are referred to as pseudovasculitis.
      • Grau R
      Pseudovasculitis: mechanisms of vascular injury and clinical spectrum.
      • Sack KE
      Mimickers of vasculitis.
      Therefore, a diagnosis of pseudovasculitis should be considered when an initial presumption of vasculitis is not supported on further investigation or when data are inconsistent. The following case shows the need to perform biopsies to differentiate pseudovasculitis from true vasculitis.

      REPORT OF A CASE

      A 27-year-old woman presented with progressive onset of pharyngitis, fatigue, and skin lesions on her face, legs, arms, and back. She had been treated recently with erythromycin for culture-proven streptococcal pharyngitis. Despite antibiotic therapy, the patient's symptoms continued, and her skin developed multiple small ulcers; she then sought further medical attention. The patient had no important medical or surgical history, took no medications, and had no known drug allergies. Her family history was notable for systemic lupus erythematosus in an aunt. The patient admitted to smoking cigarettes daily and snorting cocaine 2 times per week.
      On presentation, the patient looked ill but was afebrile and had normal vital signs. A thick gray membrane coated her posterior oropharynx. There were no oral ulcers. Her skin was remarkable for multiple diffusely spread erythematous pustular and ulcerating lesions with purulent exudates (Figure 1). The rest of her physical examination was unremarkable aside from a grade 2/6 systolic ejection murmur at the left upper sternal border without radiation.
      Figure thumbnail gr1
      Figure 1Yellow-gray exudate on the posterior pharynx and several ulcerated skin lesions.
      Laboratory data revealed a white blood cell count of 12.3 × 109/L, a hematocrit level of 28.5%, and a platelet count of 746 × 109/L. Blood chemistry results including creatinine level were normal. Urinalysis results were normal, and a urinary toxicology screen was positive for cocaine. An anti-nuclear antibody titer was initially negative but soon after converted to greater than 1:640; thereafter, it was elevated consistently. The Westergren erythrocyte sedimentation rate was greater than 150 mm/h. Perinuclear antineutrophil cytoplasmic antibody (p-ANCA) results were positive, whereas antibodies to double-stranded DNA, Smith, smooth muscle, ribonucleoprotein, Ro, and La were negative. Rheumatoid factor, human immunodeficiency virus and hepatitis serologies, and cytoplasmic ANCA (c-ANCA) were also negative. Complement levels were normal. Antibodies to proteinase 3 (PR3) were positive at 75 U/mL.
      Multiple imaging studies and several biopsies were performed to aid in diagnosis. Computed tomography of the lungs was normal. Transthoracic echocardiography revealed no evidence of valvular vegetation. An upper gastrointestinal series with small bowel follow-through was normal. Computed tomography of the sinuses revealed nasal septum destruction and mucosal thickening of the ethmoidal and maxillary sinuses. Nasal endoscopy revealed almost total erosion of the nasal septum and nasopharynx with velopharyngeal insufficiency. Soft palate biopsy specimens showed acute and chronic inflammation with fibrinopurulent exudate. Skin biopsy specimens revealed extensive and destructive neutrophilic infiltrates throughout all levels of the dermis with only a few small blood vessels, surrounded by acute inflammatory infiltrates and edema, displaying fibrinoid necrosis. No granulomas or leukocytoclasia was seen. The final impression was of infection.
      This patient presented with a systemic disease affecting her skin, oropharynx, and nasal cavity, which was clinically suggestive of a vasculitic process. This initial clinical suspicion was bolstered by the serologic data, including the positive p-ANCA results and elevated antinuclear antibody and anti-PR3 levels. Despite initial suspicions, serial biopsy specimens from different sites were consistently normal. The patient died approximately 2 years after her initial presentation. An autopsy revealed a red, raw, and totally perforated nasal septum, scattered indurated and ulcerating lesions on the skin and posterior pharynx, and evidence of gastritis and gastric hemorrhage. Microscopic evaluation of the skin lesions showed only scarring and chronic inflammation. The cause of death was cocaine and narcotic intoxication and gastric hemorrhage. Because multiple biopsy results and ultimately an autopsy failed to support the initial diagnosis of vasculitis, the patient's systemic disorder was likely due to a cocaine-induced pseudovasculitis.

      DISCUSSION

      Pseudovasculitis, as mentioned previously, is a systemic disorder that mimics vasculitis in its clinical presentation. Multiple disease processes have been associated with pseudovasculitis (Table 1). Review articles rarely include cocaine as a mimicker of vasculitic disorders.
      • Grau R
      Pseudovasculitis: mechanisms of vascular injury and clinical spectrum.
      • Sack KE
      Mimickers of vasculitis.
      However, disparate evidence supports the entity of cocaine-induced pseudovasculitis, including case reports describing the systemic effects of cocaine on the lungs, skin, and nasal cavity as well as a case series documenting the similarities and differences between Wegener granulomatosis (WG) and cocaine-induced nasal destruction.
      • Chevalier X
      • Rostoker G
      • Larget-Piet B
      • Gherardi R
      Schoenlein-Henoch purpura with necrotizing vasculitis after cocaine snorting [letter].
      • Heng MC
      • Haberfeld G
      Thrombotic phenomena associated with intravenous cocaine.
      • Hofbauer GF
      • Burg G
      • Nestle FO
      Cocaine-related Stevens-Johnson syndrome.
      • Hofbauer GF
      • Hafner J
      • Trueb RM
      Urticarial vasculitis following cocaine use [letter].
      • Orriols R
      • Munoz X
      • Ferrer J
      • Huget P
      • Morell F
      Cocaine-induced Churg-Strauss vasculitis.
      • Tomecki KJ
      • Wikas SM
      Cocaine-related bullous disease [letter].
      • Trimarchi M
      • Gregorini G
      • Facchetti F
      • et al.
      Cocaine-induced midline destructive lesions: clinical, radiographic, histopathologic, and serologic features and their differentiation from Wegener granulomatosis.
      Although the clinical scenario may suggest vasculitis, the biopsy findings in these case reports show a range of histological patterns, including fibrinoid necrosis, thrombi, perivascular inflammatory infiltrates, and microabscesses.
      Table 1Causes of Pseudovasculitis
      • Grau R
      Pseudovasculitis: mechanisms of vascular injury and clinical spectrum.
      • Sack KE
      Mimickers of vasculitis.
      • Infection
        • Meningitis
        • Infective endocarditis
        • Abscess
      • Thrombotic
        • Thrombotic thrombocytopenic purpura
        • Antiphospholipid syndrome
        • Fibromuscular dysplasia
      • Embolic
        • Myxoma
        • Cholesterol embolus
        • Fat embolus
        • Atherosclerotic embolus
      • Conditions damaging to tissue
        • Calciphylaxis
        • Amyloidosis
      • Neoplasia
      • Drugs
        • Cocaine
        • Ergotamine
        • Tobacco
      • Environmental
        • Pernio (cold exposure)
        • Scurvy
      • Other
        • Sweet syndrome
        • Pyoderma gangrenosum
      The use of serologic tests such as p-ANCA and c-ANCA in the diagnosis of vasculitides has been studied extensively. Authors recommend use of target antigen serologies in conjunction with ANCA patterns. The target antigens, anti-PR3 and myeloperoxidase, are measured by using enzyme-linked immunosorbent assays (ELISAs), whereas the associated ANCAs, c-ANCA and p-ANCA, respectively, are determined with use of indirect immunofluorescence (IIF). With p-ANCA, there are other less commonly associated target antigens in addition to myeloperoxidase. The sensitivity of using IIF and/or ELISA to diagnose vasculitis ranges between 27.3% and 87.9% in different studies. The specificity of these tests alone or in combination is higher, ranging between 75% and 99.6%. Likewise, the positive predictive value of IIF is documented to be 55% to 59% and increases to 79% if ELISA is combined with IIF. The negative predictive value of IIF alone is higher at 84% to 99%.
      • McLaren JS
      • Stimson RH
      • McRorie ER
      • Coia JE
      • Luqmani RA
      The diagnostic value of anti-neutrophil cytoplasmic antibody testing in a routine clinical setting.
      • Russell KA
      • Wiegert E
      • Schroeder DR
      • Homburger HA
      • Specks U
      Detection of anti-neutrophil cytoplasmic antibodies under actual clinical testing conditions.
      • Schmitt WH
      • van der Woude FJ
      Clinical applications of antineutrophil cytoplasmic antibody testing.
      Because of the lower sensitivity and positive predictive value, these tests make poor screening tools. This has been shown in clinical practice, especially in general practices where the prevalence of vasculitis is extremely low.
      • McLaren JS
      • Stimson RH
      • McRorie ER
      • Coia JE
      • Luqmani RA
      The diagnostic value of anti-neutrophil cytoplasmic antibody testing in a routine clinical setting.
      The strength of IIF and ELISA tests is in the high specificity and thus in confirmation of a likely diagnosis. Even so, false-positive IIF patterns are described in disorders such as systemic lupus erythematosus and inflammatory bowel disease.
      • Schmitt WH
      • van der Woude FJ
      Clinical applications of antineutrophil cytoplasmic antibody testing.
      False-positive results also are documented in cocaine users with extensive nasal destruction in whom WG is often suspected. In one case series, patients with cocaine-induced nasal destruction were compared with patients with WG, and inconsistencies between the ANCA pattern and target antigen were described in patients whose disease was due to cocaine use.
      • Trimarchi M
      • Gregorini G
      • Facchetti F
      • et al.
      Cocaine-induced midline destructive lesions: clinical, radiographic, histopathologic, and serologic features and their differentiation from Wegener granulomatosis.
      Accordingly, our patient had a positive p-ANCA pattern with anti-PR3 antibodies. This association may be characteristic of cocaine-induced pseudovasculitis. In the case series, typical biopsy findings such as granulomas or leukocytoclasia were not found in patients whose disease was due to cocaine, whereas these findings were seen in almost all patients with WG. In the patients with cocaine-induced nasal destruction, disease was localized,
      • Trimarchi M
      • Gregorini G
      • Facchetti F
      • et al.
      Cocaine-induced midline destructive lesions: clinical, radiographic, histopathologic, and serologic features and their differentiation from Wegener granulomatosis.
      whereas in patients with cutaneous WG, disease that was initially localized invariably progressed to be systemic, classically affecting organs such as the kidneys and lungs and the nasal cavity.
      • Daoud MS
      • Gibson LE
      • DeRemee RA
      • Specks U
      • el-Azhary RA
      • Su WP
      Cutaneous Wegener's granulomatosis: clinical, histopathologic, and immunopathologic features of thirty patients.
      Thus, inconsistent ANCA pattern and target antibody, lack of typical biopsy findings, and localized rather than systemic disease should be recognized as clues in the diagnosis of pseudovasculitis. Although cocaine use is not included routinely in the differential diagnosis for etiologies of pseudovasculitis, it can cause the systemic clinical findings described in case reports and can induce a nonspecific antibody response that is easily confused with a true vasculitic syndrome.
      This case highlights the fact that mimickers of vasculitis are prevalent and are readily mistaken for true vasculitis. Unfortunately, the available serologic tests cannot be relied on solely for accurate diagnosis. This fact is of particular importance for general internists, who encounter such patients at presentation and must decide on appropriate initial work-up and therapy. Our patient was treated at a dermatology clinic, where she was presumed to have WG; high-dose corticosteroids, immunosuppressants, and intradermal corticosteroid injections to the skin ulcers were initiated. In retrospect, the initial evaluation should have identified that true vasculitis was unlikely so that the patient would have been spared the adverse effects of corticosteroid and immunosuppressant therapy. Given the potential harm in misdiagnosis, recognition of the similarities and differences between pseudovasculitis and true vasculitis is important.

      Acknowledgments

      We gratefully acknowledge Dr Raymond Flores for reviewing the submitted manuscript.

      REFERENCES

        • Grau R
        Pseudovasculitis: mechanisms of vascular injury and clinical spectrum.
        Curr Rheumatol Rep. 2002; 4: 83-89
        • Sack KE
        Mimickers of vasculitis.
        in: Koopman WJ 14th ed. Arthritis and Allied Conditions: A Textbook of Rheumatology. Vol 2. Lippincott Williams & Wilkins, Philadelphia, Pa2001: 1711-1735
        • Chevalier X
        • Rostoker G
        • Larget-Piet B
        • Gherardi R
        Schoenlein-Henoch purpura with necrotizing vasculitis after cocaine snorting [letter].
        Clin Nephrol. 1995; 43: 348-349
        • Heng MC
        • Haberfeld G
        Thrombotic phenomena associated with intravenous cocaine.
        J Am Acad Dermatol. 1987; 16: 462-468
        • Hofbauer GF
        • Burg G
        • Nestle FO
        Cocaine-related Stevens-Johnson syndrome.
        Dermatology. 2000; 201: 258-260
        • Hofbauer GF
        • Hafner J
        • Trueb RM
        Urticarial vasculitis following cocaine use [letter].
        Br J Dermatol. 1999; 141: 600-601
        • Orriols R
        • Munoz X
        • Ferrer J
        • Huget P
        • Morell F
        Cocaine-induced Churg-Strauss vasculitis.
        Eur Respir J. 1996; 9: 175-177
        • Tomecki KJ
        • Wikas SM
        Cocaine-related bullous disease [letter].
        J Am Acad Dermatol. 1985; 12: 585-586
        • Trimarchi M
        • Gregorini G
        • Facchetti F
        • et al.
        Cocaine-induced midline destructive lesions: clinical, radiographic, histopathologic, and serologic features and their differentiation from Wegener granulomatosis.
        Medicine (Baltimore). 2001; 80: 391-404
        • McLaren JS
        • Stimson RH
        • McRorie ER
        • Coia JE
        • Luqmani RA
        The diagnostic value of anti-neutrophil cytoplasmic antibody testing in a routine clinical setting.
        QJM. 2001; 94: 615-621
        • Russell KA
        • Wiegert E
        • Schroeder DR
        • Homburger HA
        • Specks U
        Detection of anti-neutrophil cytoplasmic antibodies under actual clinical testing conditions.
        Clin Immunol. 2002; 103: 196-203
        • Schmitt WH
        • van der Woude FJ
        Clinical applications of antineutrophil cytoplasmic antibody testing.
        Curr Opin Rheumatol. 2004; 16: 9-17
        • Daoud MS
        • Gibson LE
        • DeRemee RA
        • Specks U
        • el-Azhary RA
        • Su WP
        Cutaneous Wegener's granulomatosis: clinical, histopathologic, and immunopathologic features of thirty patients.
        J Am Acad Dermatol. 1994; 31: 605-612