An unusual case of bullous haemorrhagic vasculitis in a COVID‐19 patient

Dear Editor

A novel Coronavirus strain, named ‘Severe Acute Respiratory Syndrome Coronavirus 2’ (SARS‐CoV‐2) was recently identified as the etiological agent of the COronaVIrus Disease 2019 (COVID‐19). Interestingly, a consistent number of COVID‐19‐associated skin manifestations seem to share a certain degree of vascular damage as common pathogenetic mechanism. ¹ Vascular injury may be due to the direct damage of endothelial cells by the virus or may represents an epiphenomenon of a dysregulated host inflammatory responses triggered by the infection. ² Here, we describe an unprecedented case of leukocytoclastic vasculitis presenting with a haemorrhagic bullous eruption in a patient affected by COVID‐19.

A 79‐year‐old man with a history of hypertension, myocardial infarction and chronic obstructive pulmonary disease has been hospitalized for acute heart failure. The patient was tested for COVID‐19 (RT‐PCR on nasopharyngeal swab sample) and resulted negative. Medical treatment for heart failure was started and patient’s conditions progressively improved. On day 15 of the hospitalization, he rapidly developed fever and dyspnea. Chest radiograph and CT scan revealed a radiologic pattern suggestive for COVID‐19 pneumonia and nasopharyngeal swab RT‐PCR confirmed SARS‐CoV‐2 infection. Treatment with hydroxychloroquine (400 mg bid), prophylactic anticoagulation (enoxaparin 4000 IU qd), empiric antibiotics (ceftaroline 600 mg bid) and intravenous corticosteroids (methylprednisolone 80 mg qd) was started. Concomitantly, oxygen therapy was initiated at 8 liters/minute (approximately 40% FiO₂) via a non‐rebreathe mask. After ten days, the patient developed multiple non‐itching vesiculobullous lesions on neck and dorsal areas of hands (Fig. 1a,b). Laboratory tests including whole blood count, biochemical and coagulation parameters were within normal limits. Antinuclear antibody, antineutrophil cytoplasmic antibody and cryoglobulins resulted negative and serum protein electrophoresis as well as complement levels were normal. Moreover, the patient tested negative for enzyme‐linked immunosorbent assay (ELISA) for detecting BP180 and BP230 antibodies. A punch skin biopsy was performed. Histopathologic examination demonstrated irregular hyperplasia of the epidermis and abundant erythrocytes extravasation with formation of intraepithelial haemorrhagic bullae. The epidermis was partly necrotic with keratinocytes focally showing nuclear hyperchromasia and cytoplasmic eosinophilia (Fig. 1c). Within the superficial dermis, there were marked erythrocytes extravasation and severe neutrophilic infiltrate within the wall of small vessels and in their proximity with scant leukocytoclasia (acute vasculitis). Endothelial cells were activated showing nuclear enlargement and hyperchromasia (Fig. 1d). Eosinophils and lymphocytic infiltration were not observed. Fibrinoid vascular changes and thrombi were absent as well as no viral cytopathic changes were observed. The histopathologic findings demonstrated a typical picture of leukocytoclastic vasculitis. Unfortunately, in the following days patient’s respiratory conditions deteriorated and, despite intensive care support, he died of respiratory insufficiency.

Figure 1.

Clinical images of COVID‐19‐associated cutaneous eruption (a, b). Histopathological images of epidermis and dermis. Haematoxylin and eosin staining, original magnification: 20×, 40× (c, d).

The case described is an unusual case of bullous haemorrhagic vasculitis in a COVID‐19 patient. The macroscopic characteristics of the lesions were compatible with a localized bullous pemphigoid (BP) or a heparin‐induced bullous haemorrhagic dermatosis (BHD). ³ , ⁴ In our patient, absence of eosinophilic infiltrates as well as negativity of ELISA for BP180/BP230 autoantibodies reasonably rule out the hypothesis of localized BP. The second diagnostic diagnosis was BHD. Nevertheless, focally necrotic epidermis and vasculitis observed in our case have never been reported in BHD and thus we excluded this diagnosis. Histopathologic features observed in our patient are characteristic of an evolving leukocytoclastic vasculitis (LCV). ⁵ Interestingly, capillary injury and/or neutrophilic infiltrates have been described in lung tissues from COVID‐19 and, in one recent report, also in the skin. ⁶ , ⁷ Nonetheless, we can expect that the number of reports concerning COVID‐19‐related vasculitis is likely to increase since inflammatory vascular damage is emerging as one of the main pathogenic mechanisms of SARS‐CoV‐2 infection, including its cutaneous manifestations. However, only further studies, novel reports including clinical images and detailed histology as well as data from international dermatology registries will be able to confirm this hypothesis.

Conflicts of interest

All authors have nothing to disclose.

Funding sources

None reported.