Illicit drug use is common in many societies across the globe and causes significant morbidity and mortality relating to well known harmful physiologic effects of these substances. Many of these drugs, including but not limited to amphetamines (including prescription and over-the-counter medications containing amphetamine-like substances) and cocaine, have also been linked to unusual rheumatic syndromes and vasculitis. Recently, a novel rheumatic syndrome has been reported in cocaine users, and this review will therefore focus primarily on cocaine and its link to newly recognized as well as previously established rheumatic diseases.
Approximately 1.9 million Americans are estimated to have actively used cocaine in 2008, with as many as 1.5% of young adults between the ages of 18 and 25 years having reported using cocaine in the month prior to being surveyed . Cocaine is a powerful, addictive stimulant with well defined pharmacologic effects on the cardiovascular and cerebrovascular systems . Cocaine use has also been implicated in a variety of unusual autoimmune syndromes that can vary in clinical presentation from frank vasculitis (organ-specific or systemic), to necrotic and destructive sinonasal lesions, to autoantibody production [3–5]. The evaluation of a patient with a suspected cocaine-associated rheumatic syndrome requires a high index of suspicion and can be quite complicated to pursue. First, many patients are unwilling to admit to cocaine use. Second, it can induce a clinical syndrome of inflammation, vasculitis and/or autoantibody production, and can closely resemble a primary, idiopathic vasculitis. Finally, cocaine use can promote atherogenesis, arterial stenoses and aneurisms in the absence of vasculitis or systemic autoimmunity.
Many of these syndromes have previously been well described, including in earlier editions of this journal . However, a novel cocaine-associated syndrome has recently emerged as a public health concern and has become the subject of several medical case series and government-issued public health advisories. This recent flurry of cases has been linked not to cocaine itself but to the increasing use of the antihelminthic drug, levamisole, to process and ‘cut’ cocaine. Exposure to levamisole can lead to a distinctive syndrome that includes neutropenia, necrotizing skin lesions with retiform purpura and a characteristic autoantibody profile [6▪▪]. Given the widespread use of cocaine, clinicians should recognize the classic features of this levamisole-associated syndrome while also being aware that its full extent has yet to be completely characterized clinically or mechanistically.
Vascultitis is a rare but well described complication of cocaine use. This vasculitis can vary in the number and type of organs involved and in the overall severity of disease. Several case reports have described what would otherwise be considered a classic picture of primary, idiopathic granulomatosis with polyangiitis (GPA), with the exception that it occurs in the setting of active cocaine use [7,8]. Although the development of GPA in the setting of cocaine use may be coincidental, the authors of these series strongly suggest a causal relationship between the two [7,8]. These patients can present with cutaneous vasculitis, nasal septal destruction, pauci-immune crescentic glomerulonephritis and a characteristic autoantibody profile that includes detection of cytoplasmic-staining antineutrophil cytoplasmic antibodies (c-ANCAs) by immunofluorescence with specificity for the target antigen proteinase-3 (PR-3) confirmed by immunoassay [7–9]. Cerebral angiitis has also been reported in the setting of cocaine use , as have cases of immunoglobulin A (IgA) nephropathy , scrotal vascultitis , urticarial vasculitis  and Churg--Struass vasculitis . Many of these cases have been ANCA negative [10,12,14]. Other patients with cocaine-associated vasculitic syndromes have tested positive for perinuclear ANCA (p-ANCA) by immunofluorescence, with a paradoxically positive subserology against PR-3, a target of c-ANCA . In these cases, the target of p-ANCA may actually be atypical p-ANCA-associated antigens, consistent with what is seen in other forms of drug-induced vascultitis . One of these atypical p-ANCA is human neutrophil elastase (HNE), which shares sequence and structural homology with the classic c-ANCA-associated antigen PR-3 . A more in-depth discussion of ANCA testing will be discussed later in this article. Treatment of cocaine-induced vascultitis usually consists of abstinence from the inciting agent with or without the addition of immunosuppressive therapy with corticosteroids alone or including cytotoxic therapy depending upon the severity of disease [8,12,14].
COCAINE-INDUCED MIDLINE DESTRUCTIVE LESIONS
Chronic use of cocaine can induce sinusitis and necrosis of the sinus mucosa, nasal and palatal perforation, and midline sinonasal destruction that can mimic a vascultitis such as GPA [18–20]. Although many of the lesions demonstrate mixed inflammatory cellular infiltrates, necrotizing inflammation and occasional leukocytoclastic vasculitis and fibrinoid necrosis, other classic features of GPA are usually absent on histopathology [5,18–20].
Serologic markers of inflammation, including erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP), may be elevated in some patients with cocaine-induced midline destructive lesion (CIMDL), making these tests unreliable in distinguishing CIMDL from a systemic vasculitis [18–20]. Unfortunately, ANCA serologies are generally unhelpful as well in distinguishing between these two syndromes, as patients with CIMDL may be either ANCA positive or negative . When positive, ANCA patterns can vary. Some patients test positive for c-ANCA and PR-3 [9,20]. More often, CIMDL patients test positive for p-ANCA with specificity for atypical p-ANCA-associated antigens such as HNE [20,21]. Although some CIMDL patients may also test positive for PR-3 antibodies, the presence of antibodies to HNE provides some diagnostic specificity for CIMDL compared to GPA (which is almost universally anti-HNE negative) . It is unclear whether patients with CIMDL who test dual-positive for antibodies to HNE and PR-3 have one antibody recognizing cross-reacting epitopes on the structurally similar molecules, or as some evidence suggests, two unique antibodies that recognize separate epitopes on the two molecules . In the latter case, data suggest that antibodies directed against PR-3 in CIMDL patients might target different epitopes than do those from patients with GPA .
Treatment of CIMDL is centred upon cessation of cocaine use, and there are limited data regarding use of adjunctive immunotherapy. One report describes some benefit from high-dose prednisilone in the setting of ongoing cocaine use but little benefit from methotrexate or cyclophosphomide . Therefore, distinguishing CIMDL from a primary vasculitis, although difficult, could spare CIMDL patients from potentially unnecessary immunosuppressive therapy.
COCAINE-ASSOCIATED VASCULAR DISEASE
Cocaine can induce vasoconstriction, accelerated atherogenesis and arterial stenoses and aneurisms, which can lead to ischaemic and haemorrhagic stroke, encephalopathy, myocardial ischaemia and infarction, mesenteric ischaemia and even renal infarction [2,23–25]. If not recognized, many of these complications might be confused with those of a primary vasculitis, especially if imaging studies demonstrate vascular abnormalities. Because these complications are not sequelae of a primary autoimmune or inflammatory disorder, patients frequently do not have elevated ESR or CRP levels or positive ANCA serologies.
LEVAMISOLE-ASSOCIATED RHEUMATIC DISEASE
Levamisole is a veterinary antihelminthic medication previously used to treat paediatric nephrotic syndrome, rheumatoid arthritis and colon cancer before being withdrawn from the market because of significant toxicity that included neutropenia. Beginning in 2008–2009, reports surfaced describing severe cases of neutropenia, some of them fatal, occurring in patients who were exposed to cocaine adulterated with levamisole [26–28]. The timing of these cases is most likely linked to the dramatic increase in use of levamisole as a cocaine adulterant. According to the US Drug Enforcement Agency, as much as 70% of the cocaine seized in 2009 contained levamisole compared with fewer than 5% of the cocaine seized in 2007 . Currently, there is no widely available, standardized test for levamisole exposure, but it can be detected in specialized laboratories using liquid chromatography/tandem mass spectroscopy [30▪,31]. Two separate case series from inner-city hospitals detected levamisole in 68--88% of urine samples from confirmed cocaine-positive patients [30▪,31]. The exact reason for this dramatic increase in levamsiole as an adulterant of cocaine is not fully understood.
Exposure to levamisole-adulterated cocaine is also associated with a distinct clinical syndrome characterized primarily by retiform purpura, cutaneous necrosis, thrombotic angiopathy of small dermal vessels and autoantibodies that include high titres of p-ANCA and antiphospholipid antibodies [6▪▪,32,33,34▪,35]. Patients frequently demonstrate high levels of both ESR and CRP [6▪▪]. At first glance, this syndrome of purpura and positive p-ANCA might be confused with microscopic polyangiitis (MPA), but several clinical and laboratory features distinguish these two entities. Levamisole--cocaine associated rheumatic disease is primarily limited to the skin, although some suspected cases of extracutaneous involvement including pulmonary haemorrhage have been identified . The purpura is retiform rather than palpable purpura and has a curious predilection for the ears, cheeks, zygomatic arch and lower extremities (Fig. 1) [6▪▪,33]. Some patients may also have small digital infarctions (Fig. 2). In addition, neutropenia is commonly encountered in this syndrome in contrast to those patients with MPA [6▪▪,34▪].
AUTOANTIBODIES AND HISTOPATHOLOGY
With levamisole-associated disease, p-ANCA is positive by immunofluorescence at unusually high titres compared with those seen with classical MPA [6▪▪,34▪]. The antigen specificities for these p-ANCA are directed against multiple components of neutrophil granules, a feature observed in other drug (medication)-associated vasculitis. In our experience, p-ANCA titres can be greater than 1 : 20 480 in some patients, with the highest titres directed against HNE and other ‘atypical’ p-ANCA-associated antigens such as lactoferrin and cathepsin G [6▪▪]. Antibodies to myeloperoxidase (MPO) are occasionally detected, and when positive, are at relatively low titres compared with titres observed with the p-ANCA immunofluorescence. This discordance between p-ANCA immunofluorescence and MPO immunoassay is a striking hallmark of levamisole-associated autoimmunity that is not seen with classic MPA. Some of these p-ANCA-positive patients seem to paradoxically test positive for antibodies to PR-3. In these cases, it is not known whether or not the target of these PR-3 antibodies is to the same epitopes recognized by sera from GPA patients, different PR-3 epitopes or to homologous (cross-reacting) regions on p-ANCA-associated antigens such as HNE. The majority of patients with levamisole-associated disease also test positive for one or more antiphospholipid antibodies, including lupus anticoagulant and IgM predominant anticardiolipin and anti-beta 2 glycoprotein 1 antibodies [6▪▪]. Other autoantibodies are detected with varying frequency and specificity, including antinuclear and antidouble-stranded DNA antibodies. The predominant histopathology is an obliterative, small vessel thrombosis, with varying amounts of leurkocytoclastic vasculitis [6▪▪,32].
The pathogenesis of this syndrome is unknown. When used therapeutically, levamisole was linked to neutropenia, purpura of the cheeks and ears, ANCA and antiphospholipid antibodies, with many of these adverse events occurring in the setting of long- term use [37–43]. This suggests that prolonged exposure to levamisole alone is sufficient to induce these symptoms. There may be one or more mechanisms by which this might occur. Levamisole may be directly toxic to neutrophils or endothelial cells, may act as a nonspecific immune adjuvant in individuals predisposed to autoimmunity or may induce loss of tolerance to specific autoantigens in a manner that initiates or perpetuates autoimmunity. In an article published before the discovery of ANCA, investigators demonstrated that levamisole could induce IgM-specific granulocytoxic antibodies . Cocaine itself is also associated with ANCA and leukocytoclastic vasculitis. The relevance of concomitant exposure to cocaine and levamisole is not known nor is the importance of underlying patient genetics or other epidemiologic factors that include potential exposure to as-of-yet uncharacterized adulterants or substances.
CLINICAL COURSE AND TREATMENT
The clinical course of patients who suffer from rheumatic complications linked to levamisole-adulterated cocaine is difficult to assess in light of the fact that in our experience, few patients are able to abstain from ongoing use of cocaine. In many instances, necrotic lesions will ulcerate, necessitating debridement and skin grafting. Neutropenia can be quite severe and predispose patients to superinfection of their lesions. Treatment is primarily supportive. In those patients who successfully refrain from cocaine use, skin lesions can begin to resolve over a few weeks . In the setting of worsening skin disease and ongoing cocaine use, some patients have been treated with either anticoagulation, immunosupression or both with varying degrees of success [45▪]. Corticosteroid therapy may be beneficial for those patients with threatening skin lesions that have not fully necrosed or that, on biopsy, have significant components of inflammation and/or leukocytoclstic vasculitis. Although some patients have been treated with anticoagulation, no definitive assessment can be made as to its benefit for this particular cutaneous thrombotic angiopathy.
Cocaine/levamisole-associated rheumatic disease is a newly reported phenomenon, and the extent of its clinical phenotype has likely yet to be fully characterized. Although described primarily as a cutaneous disease, it is conceivable that levamisole-adulterated cocaine is associated with other severe extracutaneous manifestations. Previously, levamisole has been implicated in the development of proliferative glomerulonephritis in a patient with rheumatoid arthritis , and a recent series describes pulmonary haemorrhage as a possible consequence of exposure to levamisole-tainted cocaine . Because both cocaine and levamisole can induce ANCA (perhaps in the absence of clinical disease), it is difficult to definitively know whether or not these drugs were directly involved in these cases or whether these patients developed unrelated idiopathic disease but were coincidentally ANCA positive by virtue of their exposure to cocaine and/or levamisole.
In the time since we reported our initial experience with six patients exposed to levamisole-adulterated cocaine, we have observed other patients develop interesting autoimmune phenomena in the setting of cocaine use and high-titre p-ANCA. One male patient presented with a pauci-immune glomerulonephritis, retiform purpura and digital infarcts. Because he had manifestations of both classic skin involvement and renal disease, it is likely that these phenomena were related to his exposure to levamisole and/or cocaine. Another patient developed alveolar haemorrhage and a pauci-immune glomerulonephritis in the setting of high-titre p-ANCA but without classic retiform purpura. In her case, antibodies to MPO were significantly positive, in contrast to many other patients with levamisole-associated autoimmunity. Both of these patients have been treated with corticosteroids and intravenous cyclophosphamide.
Cocaine use is associated with a wide variety of clinical manifestations that can be difficult to distinguish from idiopathic autoimmune rheumatic diseases. Similarly to patients with idiopathic disease, those with cocaine-associated rheumatic diseases will often, but not always, be ANCA positive; however, the titre and specificities of ANCA may help to distinguish between these different entities. Increased use of levamisole to adulterate cocaine has resulted in a well reported syndrome with features of neutropenia, retiform purpura and cutaneous necrosis (with a predilection for the ear lobes and cheeks), and autoantibodies consisting of high-titre p-ANCA with specificities for ‘atypical’ antigens including HNE. In light of the widespread use of cocaine and its prevalent adulteration with levamisole, clinicians should be familiar with the clinical manifestations of these diseases.
Supported by the Rosalind Russell Medical Research Center for Arthritis.
Conflicts of interest
No conflicts of interest.
REFERENCES AND RECOMMENDED READING
Papers of particular interest, published within the annual period of review, have been highlighted as:
- ▪ of special interest
- ▪▪ of outstanding interest
Additional references related to this topic can also be found in the Current World Literature section in this issue (p. 150).
1. National Institute on Drug Abuse. Cocaine: abuse and addiction. http://http://www.drugabuse.gov
/publications/research-reports/cocaine-abuse-addiction. [Accessed 22 August 2012]
2. Schwartz BG, Rezkalla S, Kloner RA. Cardiovascular effects of cocaine. Circulation 2010; 122:2558–2569.
3. Espinoza LR, Perez Alamino R. Cocaine-induced vasculitis
: clinical and immunological spectrum. Curr Rheumatol Rep 2012. [Epub ahead of print]
4. Bateman H, Rehman A, Valeriano-Marcet J. Vasculitis
-like syndromes. Curr Rheumatol Rep 2009; 11:422–429.
5. Molloy ES, Langford CA. Vasculitis
mimics. Curr Opin Rheumatol 2008; 20:29–34.
6▪▪. Graf J, Lynch K, Yeh C-L, et al. Purpura, cutaneous necrosis, and antineutrophil cytoplasmic antibodies associated with levamisole-adulterated cocaine. Arthritis Rheum 2011; 63:3998–4001.
A concise review of six cases of cocaine--levamisole associated retirform purpura that includes clinical manifestations, autoantibody specificities, cutaneous disease and key features that help distinguish this disease from idiopathic vasculitis.
7. Tervaert JWC, Stegeman CA. A difficult diagnosis. Lancet 2004; 364:1313–1314.
8. Neynaber S, Mistry-Burchardi N, Rust C, et al. PR3-ANCA-positive necrotizing multiorgan vasculitis
following cocaine abuse. Acta Derm Venereol 2008; 88:594–596.
9. Rowshani AT, Schot LJ, ten Berge IJM. c-ANCA as a serological pitfall. Lancet 2004; 363:782.
10. Merkel PA, Koroshetz WJ, Irizarry MC, Cudkowicz ME. Cocaine-associated cerebral vasculitis
. Semin Arthritis Rheum 1995; 25:172–183.
11. Chevalier X, Rostoker G, Larget-Piet B, Gherardi R. Schoenlein-Henoch purpura with necrotizing vasculitis
after cocaine snorting. Clin Nephrol 1995; 43:348–349.
12. Chen S-C, Jang M-Y, Wang C-S, et al. Cocaine-related vasculitis
causing scrotal gangrene. Ann Pharmacother 2009; 43:375–378.
13. Hofbauer GF, Hafner J, Trüeb RM. Urticarial vasculitis
following cocaine use. Br J Dermatol 1999; 141:600–601.
14. Orriols R, Muñoz X, Ferrer J, et al. Cocaine-induced Churg-Strauss vasculitis
. Eur Respir J 1996; 9:175–177.
15. Friedman DR, Wolfsthal SD. Cocaine-induced pseudovasculitis. Mayo Clin Proc 2005; 80:671–673.
16. Wiik A, Stummann L, Kjeldsen L, et al. The diversity of perinuclear antineutrophil cytoplasmic antibodies (pANCA) antigens. Clin Exp Immunol 1995; 101 (Suppl 1):15–17.
17. Fujinaga M, Chernaia MM, Halenbeck R, et al. The crystal structure of PR3, a neutrophil serine proteinase antigen of Wegener's granulomatosis antibodies. J Mol Biol 1996; 261:267–278.
18. Stahelin L, Fialho SC, de MS, et al. Cocaine-induced midline destruction lesions with positive ANCA test mimicking Wegener's granulomatosis. Rev Bras Reumatol 2012; 52:431–437.
19. Rachapalli SM, Kiely PDW. Cocaine-induced midline destructive lesions mimicking ENT-limited Wegener's granulomatosis. Scand J Rheumatol 2008; 37:477–480.
20. 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.
21. Wiesner O, Russell KA, Lee AS, et al. Antineutrophil cytoplasmic antibodies reacting with human neutrophil elastase as a diagnostic marker for cocaine-induced midline destructive lesions but not autoimmune vasculitis
. Arthritis Rheum 2004; 50:2954–2965.
22. Peikert T, Finkielman JD, Hummel AM, et al. Functional characterization of antineutrophil cytoplasmic antibodies in patients with cocaine-induced midline destructive lesions. Arthritis Rheum 2008; 58:1546–1551.
23. Toossi S, Hess CP, Hills NK, Josephson SA. Neurovascular complications of cocaine use at a tertiary stroke center. J Stroke Cerebrovasc Dis 2010; 19:273–278.
24. Sudhakar CB, Al-Hakeem M, MacArthur JD, Sumpio BE. Mesenteric ischemia secondary to cocaine abuse: case reports and literature review. Am J Gastroenterol 1997; 92:1053–1054.
25. Hoefsloot W, de Vries RA, Bruijnen R, Bosch FH. Renal infarction after cocaine abuse: a case report and review. Clin Nephrol 2009; 72:234–236.
26. Zhu NY, Legatt DF, Turner AR. Agranulocytosis after consumption of cocaine adulterated with levamisole. Ann Intern Med 2009; 150:287–289.
27. CDC. Agranulocytosis associated with cocaine use - four States, March 2008-November 2009. MMWR 2009; 58:1381–1385.
28. Nationwide public health alert issued concerning life-threatening risk posed by cocaine laced with veterinary anti-parasite drug. http://http://www.samhsa.gov
/newsroom/advisories/090921vet5101.aspx. [Accessed 22 August 2012]
29. Casale J, Corbeil E, Hays P. Identification of levamisole impurities in illicit cocaine exhibits. DEA Microgram Journal 2008; 6:82–89.
30▪. Lynch KL, Dominy SS, Graf J, Kral AH. Detection of levamisole exposure in cocaine users by liquid chromatography-tandem mass spectrometry. J Anal Toxicol 2011; 35:176–178.
This study describes a straightforward method used to detect levamisole in the urine of suspected patients and reports on the prevalence of levamisole in those who test positive for cocaine.
31. Buchanan JA, Heard K, Burbach C, et al. Prevalence of levamisole in urine toxicology screens positive for cocaine in an inner-city hospital. JAMA 2011; 305:1657–1658.
32. Walsh NMG, Green PJ, Burlingame RW, et al. Cocaine-related retiform purpura: evidence to incriminate the adulterant, levamisole. J Cutan Pathol 2010; 37:1212–1219.
33. Bradford M, Rosenberg B, Moreno J, Dumyati G. Bilateral necrosis of earlobes and cheeks: another complication of cocaine contaminated with levamisole. Ann Intern Med 2010; 152:758–759.
34▪. Poon SH, Baliog CR Jr, Sams RN, et al. Syndrome of cocaine-levamisole-induced cutaneous vasculitis
and immune-mediated leukopenia. Semin Arthritis Rheum 2011; 41:434–444.
This article summarizes four patients exposed to levamisole-contaminated cocaine, including their clinical presentation, laboratory testing and cutaneous disease, and it summarizes several recent case reports and series.
35. Gross RL, Brucker J, Bahce-Altuntas A, et al. A novel cutaneous vasculitis
syndrome induced by levamisole-contaminated cocaine. Clin Rheumatol 2011; 30:1385–1392.
36. McGrath MM, Isakova T, Rennke HG, et al. Contaminated cocaine and antineutrophil cytoplasmic antibody-associated disease. Clin J Am Soc Nephrol 2011; 6:2799–2805.
37. Barbano G, Ginevri F, Ghiggeri GM, Gusmano R. Disseminated autoimmune disease during levamisole treatment of nephrotic syndrome. Pediatr Nephrol 1999; 13:602–603.
38. Laux-End R, Inaebnit D, Gerber HA, Bianchetti MG. Vasculitis
associated with levamisole and circulating autoantibodies. Arch Dis Child 1996; 75:355–356.
39. Macfarlane DG, Bacon PA. Levamisole-induced vasculitis
due to circulating immune complexes. Br Med J 1978; 1:407–408.
40. Menni S, Pistritto G, Gianotti R, et al. Ear lobe bilateral necrosis by levamisole-induced occlusive vasculitis
in a pediatric patient. Pediatr Dermatol 1997; 14:477–479.
41. Powell J, Grech H, Holder J. A boy with cutaneous necrosis occurring during treatment with levamisole. Clin Exp Dermatol 2002; 27:32–33.
42. Rongioletti F, Ghio L, Ginevri F, et al. Purpura of the ears: a distinctive vasculopathy with circulating autoantibodies complicating long-term treatment with levamisole in children. Br J Dermatol 1999; 140:948–951.
43. Scheinberg MA, Bezerra JB, Almeida FA, Silveira LA. Cutaneous necrotising vasculitis
induced by levamisole. Br Med J 1978; 1:408.
44. Thompson JS, Herbick JM, Klassen LW, et al. Studies on levamisole-induced agranulocytosis. Blood 1980; 56:388–396.
45▪. Khan TA, Cuchacovich R, Espinoza LR, et al. Vasculopathy, hematological, and immune abnormalities associated with levamisole-contaminated cocaine use. Semin Arthritis Rheum 2011; 41:445–454.
Another descriptive case series of patients exposed to levamisole-adulterated cocaine and summary of recent literature.
46. Shearn MA, Tu WH. Proliferative glomerulonephritis associated with levamisole therapy of rheumatoid arthritis. J Rheumatol 1981; 8:522–523.