CPC01 Eosinophilic cellulitis in the setting of cocaine use

Abstract A 32-year-old man presented to the emergency department with a 2-week history of a vesiculobullous eruption on both arms and legs. He reported use of intranasal cocaine, oral ketamine and benzodiazepine regularly and a background of chronic sinusitis. He was systematically unwell with a temperature of 40 °C, hypotension 90/60 mmHg and tachycardia 113 bpm. On examination, he had multiple vesicles and tense impetiginized bullae with swelling and erythema on the left arm. He had a papulopustular eruption on the right arm and both legs. Laboratory investigations showed white cell count of 18.8 × 109 cells L−1, neutrophilia 8.6 × 109 cells L−1, eosinophilia 5.3 × 109 cells L−1 [normal range (NR) < 0.05 × 109 cells L−1] and C-reactive protein 99 mg L−1. Urine toxicology was positive for cocaine, amphetamines, methamphetamines and benzodiazepines. He was treated empirically with intravenous antibiotics for cellulitis. Skin biopsy showed subepidermal bullae, and dense dermal neutrophilic and eosinophilic infiltration with vascular damage. Gram stain, periodic acid–Schiff stain and perilesional direct immunofluorescence and skin biopsy for bacterial culture were negative. Investigations for persistent eosinophilia showed raised anti-proteinase 3 antineutrophil cytoplasmic antibody (ANCA) titres of 16 IU mL−1 (NR < 1.90 IU mL−1). IgE and tryptase levels were minimally elevated (210 IU mL−1, NR < 100 IU mL−1; and 14.2 μg L−1, NR < 14 μg L−1, respectively). Bone marrow trephine biopsy showed hypercellular marrow with prominent eosinophilia and no evidence of lymphoproliferative disease. Computed tomography of the thorax showed pulmonary inflammatory changes and reactive left axillary lymphadenopathy. Bronchoalveolar lavage for cytology, bacterial and mycobacterial culture was negative. Computed tomography of the sinus showed leftward septal deviation and sinomucosal inflammation likely secondary to cocaine use. Following multidisciplinary team discussion based on laboratory and histological findings, working diagnoses of cocaine-induced hypereosinophilia and eosinophilic cellulitis were made. Oral prednisolone (0.5 mg kg−1 per day) was commenced, with rapid improvement noted within days. Cocaine-induced eosinophilic pneumonia is well described, and the hypothesized pathogenesis is cocaine inducing a hypersensitivity reaction. Herein, we report a potential case of eosinophilic cellulitis in the setting of cocaine use. While the pathogenesis of this rare inflammatory skin disorder remains unclear, it may reflect a reaction pattern to a variety of stimuli, including infections, drugs and insect bites. Eosinophilic cellulitis has also been suggested to result from a type 4 hypersensitivity reaction. In this case it is also possible that the clinical findings were triggered by a contaminated batch of cocaine, as levamisole-adulterated cocaine can cause autoimmunity syndromes such as drug-induced ANCA. This dangerous recreational drug initiates inflammation through formation of neutrophil extracellular traps, which culminates in pronounced autoimmune responses including ANCA production.