RNAscope in situ hybridization and RT-PCR for detection of SARS-CoV-2 in chilblain-like lesions: A clinical, laboratory and histopathological study

Background: Since the beginning of the coronavirus disease 2019 (COVID-19) pandemic, an increasing number of chilblain-like lesions (ChLL) have been increasingly reported worldwide. To date, the causal link between ChLL and SARS-CoV-2 infection has not been unequivocally established. Methods: In this case series, we present demographic, clinical, laboratory, and histopathological information regarding 27 young patients with a clinical diagnosis of ChLL who referred to the Dermatology Unit of Papa Giovanni XXIII Hospital, Bergamo, Italy, from 1 April 2020 to 1 June 2020. Results: The mean age was 14.2 years, and 21 patients (78%) experienced mild systemic symptoms a median of 28 days before the onset of cutaneous lesions. ChLL mostly involved the feet (20 patients – 74%). Among acral lesions, we identified three different clinical patterns: (i) chilblains in 20 patients (74%); (ii) fixed erythematous macules in 4 children (15%); (iii) erythrocyanosis in 3 female patients (11%). Blood examinations and viral serologies, including parvovirus B19, cytomegalovirus (CMV), Epstein-Barr virus (EBV), and coxsackievirus were normal in all. Three patients (11%) underwent nasopharyngeal swab for RT-PCR for SARS-CoV-2 showing only 1 positive. Histopathological examinations of 7 skin biopsies confirmed the clinical diagnosis of chilblains; vessel thrombi were observed only in 1 case. Our findings failed to demonstrate the direct presence of SARS-CoV-2 RNA in skin biopsies, both with real-time polymerase chain reaction (RT-PCR) and RNAscope in situ hybridization (ISH). Limitations: Limited number of cases, unavailability of laboratory confirmation of COVID-19 in all patients, potential methodological weakness, and latency of skin biopsies in comparison to cutaneous lesions onset. Conclusions: These observations may support the hypothesis of an inflammatory pathogenesis rather than the presence of peripheral viral particles. Although, we could not exclude an early phase of viral endothelial damage followed by an IFN-I or complement-mediated inflammatory phase. Further observations on a large number of patients are needed to confirm this hypothesis.