Chilblains and COVID-19: why SARS-CoV-2 endothelial infection is questioned. Reply from the authors.

Dear Editor, We thank Dr Baeck et al. for their interest in our recent article published in the BJD.

The negative reverse‐transcription polymerase chain reaction (RT‐PCR) in nasopharyngeal swabs in patients with coronavirus disease 2019 (COVID‐19) chilblains has been extensively acknowledged in the literature; however, a significant proportion of patients had mild systemic symptoms or contact with confirmed or suspected cases.

Magro et al., recently demonstrated severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) in skin biopsies of three patients with COVID‐19‐related perniosis by immunohistochemistry (SARS‐CoV‐2 envelope protein colocalized with SARS‐CoV‐2 membrane protein) and RNAscope together with evidence of type I interferon signalling activation. The authors propose that a strong type I interferon response may accelerate viral elimination, explaining the reported negativity for RT‐PCR and serological tests. Low sensitivity of the serological tests in asymptomatic patients could also explain the negative results. It is unclear whether serological tests can detect the lower antibody levels likely to be seen in mildly symptomatic or asymptomatic patients.

Although limited to the skin of the distal extremities, the vascular damage seen in COVID‐19 chilblains is severe enough to produce a lymphocytic vasculitis with endothelial disruption, microthrombosis and localized ischaemia. Why the lesions in these patients are limited to the distal feet and hands is still unknown.

We reaffirm our statement that immunohistochemistry for detection of SARS‐CoV/SARS‐CoV‐2 remains restricted and subject to cautious interpretation. The images provided by Baeck et al. show suboptimal nonspecific reactivity. In our research, using an antibody directed against the spike protein of SARS/SARS‐CoV‐2, after optimization of the staining, we obtained a clean background, and our negative controls showed entirely negative endothelial reactivity. We acknowledge that we have no experience with the SARS‐CoV‐2 NP antibody used by Baeck et al.

The observation that our images show positivity limited to relatively healthy vessels is interesting. In fact, in our cases, not all the vessels showed the same degree of positivity, and heavily inflamed vessels appeared to show a lower expression than mildly inflamed ones. Clearance of viruses by the inflammatory process may be a potential reason for this.

The presence of viral particles on electron microscopy (EM) in endothelial cells is supported by several reports describing virus‐like particles in patients with SARS‐CoV‐2 infection. Two of our coauthors have collaborated in a case series of COVID‐19‐related cutaneous lesions, which included biopsies of 11 COVID‐19‐related acroischaemic lesions. EM was performed and demonstrated coronavirus‐like particles in three of five cases of COVID‐19 chilblains.

Definitive characterization of SARS‐CoV‐2 virions requires immuno‐EM. Unfortunately, we do not have remaining tissue adequately processed to perform this study, and we have not seen any other patient presenting with chilblains since the beginning of May. We are prepared to perform immuno‐EM if a second wave of the pandemic causes a new outbreak of similar cases.