Comment
Our patient presented with severe necrotizing cutaneous vasculitis, likely secondary to levamisole exposure. Some of our patient’s cutaneous findings may be explained exclusively on the basis of cocaine exposure, but the characteristic lesion distribution and histopathologic findings along with the evidence of autoantibody positivity and concurrent arthralgias make the combination of levamisole and cocaine a more likely cause. Similar skin lesions were first described in children treated with levamisole for nephrotic syndrome.2 The most common site of clinical involvement in these children was the ears, as seen in our patient. Our patient tested positive for p-ANCA, which is the most commonly reported autoantibody associated with this patient population. Sixty-one percent (20/33) of patients with levamisole-induced vasculitis from 2 separate reviews showed p-ANCA positivity.7,10
On histopathology, our patient’s skin biopsy findings were consistent with those of prior reports of levamisole-induced vasculitis, which describe patterns of thrombotic vasculitis, leukocytoclasis, and fibrin deposition or occlusive disease.2,6,7,9-14 Mixed histologic findings of vasculitis and thrombosis, usually with varying ages of thrombi, are characteristic of levamisole-induced purpura. In addition, the disease can present nonspecifically with pure microvascular thrombosis without vasculitis, especially later in the course.9
The recommended management of levamisole-induced vasculitis currently involves the withdrawal of the culprit adulterated cocaine along with supportive treatment. Spontaneous and complete clinical resolution of lesions has been reported within 2 to 3 weeks and serology normalization within 2 to 14 months of levamisole cessation.2,6 A 2011 review of patients with levamisole-induced vasculitis reported 66% (19/29) of cases with either full cutaneous resolution after levamisole withdrawal or recurrence with resumed use, supporting a causal relationship.7 Walsh et al9 described 2 patients with recurrent and exacerbated retiform purpura following cocaine binges. Both of these patients had urine samples that tested positive for levamisole.9 In more severe cases, medications shown to be effective include colchicine, polidocanol, antibiotics, methotrexate, anticoagulants, and most commonly systemic corticosteroids.7,10,11,15 Nonsteroidal anti-inflammatory drugs were successful in treating lesions in 2 patients with concurrent arthralgia.7 Rarely, patients have required surgical debridement or skin grafting due to advanced disease at initial presentation.9,12-14 One of the most severe cases of levamisole-induced vasculitis reported in the literature involved 52% of the patient’s total body surface area with skin, soft tissue, and bony necrosis requiring nasal amputation, upper lip excision, skin grafting, and extremity amputation.14 Another severe case with widespread skin involvement was recently reported.16
For unclear reasons, our patient continued to develop cutaneous lesions despite self-reported cocaine cessation. Complete resolution required the combination of vardenafil, prednisone, and aspirin, along with debridement and wound care. Vardenafil, a selective phosphodiesterase 5 inhibitor, enhances the effect of nitrous oxide by increasing levels of cyclic guanosine monophosphate,17 which results in smooth muscle relaxation and vasodilatation. The primary indication for vardenafil is the treatment of erectile dysfunction, but it often is used off label in diseases that may benefit from vasodilatation. Because of its mechanism of action, it is understandable that a vasodilator such as vardenafil could be therapeutic in a condition associated with thrombosis. Moreover, the autoinflammatory nature of levamisole-induced vasculitis makes corticosteroid treatment effective. Given the 10-week delay in improvement, we suspect that it was the combination of treatment or an individual agent that led to our patient’s eventual recovery.
There are few reports in the literature focusing on optimal treatment of levamisole-induced vasculitis and none that mention alternative management for patients who continue to develop new lesions despite cocaine avoidance. Although the discontinuation of levamisole seems to be imperative for resolution of cutaneous lesions, it may not always be enough. It is possible that there is a subpopulation of patients that may not respond to the simple withdrawal of cocaine. It also should be mentioned that there was no urine toxicology screen obtained to support our patient’s reported cocaine cessation. Therefore, it is possible that his worsening condition was secondary to continued cocaine use. However, the patient successfully responded to the combination of vardenafil and prednisone, regardless of whether his condition persisted due to continued use of cocaine or not. This case suggests the possibility of a new treatment option for levamisole-induced vasculitis for patients who continue to use levamisole despite instruction for cessation or for patients with refractory disease.
Conclusion
A trial of prednisone and vardenafil should be considered for patients with refractory levamisole-induced vasculitis. Further studies and discussions of disease course are needed to identify the best treatment of this skin condition, especially for patients with refractory lesions.