Response to 'No evidence of SARS-CoV-2 infection by polymerase chain reaction or serology in children with pseudo-chilblain'.

Dear Editor, We read with interest the article by Caselli et al., which reported a case series of 38 children with chilblain‐like lesions (CLLs). Testing for SARS‐CoV‐2 using polymerase chain reaction (PCR), rapid test serology and enzyme‐linked immunosorbent assay (ELISA) for IgA and IgG antibodies yielded negative results in all cases. The authors concluded that their data do not allow them to support the relationship of CLLs with SARS‐CoV‐2 infection. So far, data in the literature studying CLLs documented a very low percentage of laboratory‐confirmed SARS‐CoV‐2. However, Colmenero et al. were able to detect SARS‐CoV‐2 in endothelial cells of cutaneous chilblain lesions by immunohistochemistry methods in seven paediatric patients with negative nasopharyngeal swabs.

Recently, we updated our case series with additional serological investigations, collecting results from 32 patients with CLLs (15 female patients, 17 male patients; average age 16·3 years). Data are detailed in Table 1. A SARS‐CoV‐2 PCR nasopharyngeal swab test was performed in 11 patients yielding two positive cases. These two cases had been screened 3 weeks before the onset of the cutaneous involvement because of fever and contact with patients positive for COVID‐19. We tested for detection of SARS‐CoV‐2 antibodies in 22 patients using a chemiluminescent immunoassay (LIAISON® SARS‐CoV‐2 IgG kit, DiaSorin®, Saluggia, Italy) at least 14 days after the onset of the cutaneous lesions. Results were consistently negative for specific class IgG antibodies. A new ELISA test was subsequently available (Anti‐SARS‐CoV‐2 ELISA IgM Test, Anti‐SARS‐CoV‐2 ELISA IgG Test ELISA, DIA.PRO Diagnostic Bioprobes, Sesto San Giovanni, Italy) and three previously negative cases were found positive for IgM.

Table 1

Clinical and laboratory data for patients with chilblain‐like lesions (CLLs)

Case	Age (years)	Sex (M/F)	Localization	Cutaneous symptoms	Systemic symptoms	Onset of CLL after systemic symptoms (days)	PCR swab	Serology
								IgGa	IgMb
1	14	F	Feet, hands	None	None	nd	−	−	−
2	15	F	Hands	None	None	nd	np	np	np
3	15	F	Feet	None	None	nd	−	−	−
4	18	F	Feet, hands	None	None	nd	np	−	−
5	13	F	Feet	None	None	nd	np	np	np
6	16	M	Feet	None	Cough	14	np	−	−
7	13	M	Feet	None	None	nd	np	−	−
8	15	F	Hands	None	None	nd	np	np	np
9	14	M	Feet	Itch	Diarrhoea	7	−	−	−
10	11	M	Hands	None	Fever	21	−	np	np
11	14	M	Feet	None	Cold	21	np	−	−
12	17	M	Feet	None	Fever	42	np	np	np
13	12	F	Feet	None	None	nd	np	−	+
14	15	M	Feet, hands	None	None	nd	np	−	+
15	12	F	Feet	Pain	Headache	10	−	−	−
16	8	F	Feet, hands	None	None	nd	−	−	−
17	10	M	Feet	Itch	Fever	56	np	np	np
18	14	M	Feet, hands	None	None	nd	np	−	−
19	16	M	Feet	None	None	nd	np	np	np
20	3	M	Feet, hands	None	Fever	10	np	−	−
21	15	F	Feet, hands	None	None	nd	−	−	−
22	8	M	Hands	None	None	nd	np	−	−
23	14	M	Feet, hands	None	None	nd	−	−	−
24	12	M	Feet	None	Fever	15	−	−	−
25	7	F	Feet, hands	Pain	Fever, headache, ageusia, anosmia	21	+	np	np
26	39	M	Feet	Burning	None	nd	np	−	−
27	23	F	Feet	None	Fever	21	np	−	−
28	25	F	Hands	None	None	nd	np	−	−
29	30	M	Hands	None	None	nd	np	−	−
30	31	M	Hands	None	None	nd	np	np	np
31	23	F	Feet	None	Fever	22	+	np	np
32	31	F	Feet, hands	None	None	nd	np	−	+

M, male; F, female; PCR, polymerase chain reaction; nd, not determined; +, positive; −, negative; np, not performed. aChemiluminescent immunoassay, LIAISON®SARS‐CoV‐2 IgG kit (DiaSorin®, Saluggia, Italy).bAnti‐SARS‐CoV‐2 enzyme‐linked immunosorbent assay (ELISA) IgM Test; Anti‐SARS‐CoV‐2 ELISA IgG Test ELISA (DIA.PRO Diagnostic Bioprobes, Sesto San Giovanni, Italy).

The overall epidemiological and clinical characteristics of CLLs still point to a COVID‐19 related condition. Nasopharyngeal swabs are highly dependent on the sampling technique and timing. As far as serology is concerned, in addition to timing, the method used is also a key factor. Serology negativity could be due to the shortcomings of currently available testing methods. Alternatively, a vigorous innate immune response against the virus could hamper the generation of antibodies through the adaptive response. Taken together, our results suggest that CLLs develop in individuals with mild infection, with low and rapid viral shedding, who are unable in most cases to generate detectable specific antibodies. A better understanding of the underlying immunological mechanism would shed light on the pathogenesis of SARS‐CoV‐2 and could have relevant implications for the development of an effective vaccine.

Author Contribution

Sebastiano Recalcati: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Investigation (equal); Methodology (equal); Writing‐original draft (equal). Silvia Tonolo: Conceptualization (equal); Investigation (equal); Methodology (equal). Francesco Luzzaro: Conceptualization (equal); Methodology (equal). Fabrizio Fantini: Conceptualization (equal); Investigation (equal).