Acute acral eruptions in children during the COVID-19 pandemic: Characteristics of 103 children and their family clusters.

BACKGROUND: A marked increase in frequency of acute acral eruptions (AAE) was observed in children during the COVID-19 pandemic in the spring period.
OBJECTIVES: In this observational multicenter study, based on children with AAE, we aimed to assess the proportion of household members possibly infected by SARS-CoV-2.
METHODS: We collected data from all children observed with AAE, prospectively from April 7, 2020 to June 22, 2020, and retrospectively since February 28, 2020. The primary outcome was the household infection rate, defined as the proportion of family clusters having at least one member with COVID-19 infection other than the child with AAE ("index child"). The definition of a case was based on characteristic clinical signs and a positive PCR or serology.
RESULTS: The study included 103 children in 10 French departments and in Quebec. The median age was 13 years and the interquartile range [8-15], with a female-to-male ratio of 1/1.15. In children with AAE, all PCR tests were negative (n=18), and serology was positive in 2/14 (14.3%) cases. We found no significant anomalies in the lab results. A total of 66 of the 103 families (64.1%) of included children had at least one other infected member apart from the index child. The total number of household members was 292, of whom 119 (40.8%) were considered possibly infected by SARS-CoV-2. No index children or households exhibited severe COVID-19. DISCUSSION: Among the 103 households included, 64.1% had at least one infected member. Neither children with AAE nor their households showed severe COVID-19.

Introduction

Several cutaneous signs are associated with COVID-19, but the mechanisms are not fully understood [1], [2]. They have been observed in both adults and children. Severe COVID-19 is less frequent in children than adults. Davies et al. showed that general clinical symptoms manifest in only 21% of young people aged 10 to 19 years with COVID-19 versus 69% of people aged > 70 years old [3]. However, numerous cutaneous signs have been reported in children with COVID-19, such as acute acral eruptions (AAE) [4], [5]. Published cases of AAE display heterogeneous features, including pseudo-chilblains, acral vesicles or papules, and papular-purpuric gloves and socks syndrome [1], [2], [4], [5]. These signs appear late in the course of COVID-19 [5]. Identified risk factors for severe COVID-19 include increased age, male gender, being overweight, and chronic diseases. Genetic risk factors have also been suspected [6]. The high contagiousness of the disease warrants the study of family clusters [7]. Intrafamilial transmission is still incompletely understood and requires documentation, particularly as children have been more exposed since the reopening of schools as of May 2020 in several European countries [3], [8].

In this study, based on children with AAE, we aimed to assess the proportion of household members possibly infected by SARS-CoV-2 with COVID-19 and to describe demographic, clinical, biological and histological signs in these entities.

Methods

Design and setting

This multicenter observational study included centers in France and Quebec, Canada. The study involved all dermatologists belonging to the Group of Research of the French Society of Pediatric Dermatology (GRSFDP) and began on April 7, 2020.

Data collection

We collected data from all children with observed AAE prospectively from April 7, 2020 to June 22, 2020 and retrospectively since February 28, 2020. Data comprised demographic and clinical data (cutaneous and extracutaneous signs) for the child and the household at baseline (at the consultation) and at approximately 1-month of follow-up (all parents were contacted by telephone). Parents were then orally informed about the study and gave their oral consent to participate. Results were also collected for additional examiantions, including lab results, virology (SARS-Cov-2 PCR and serology), imaging, and microscopy of skin biopsy. Investigators entered the data into a standardized electronic case report form (e-CRF). Using the criteria of the European Centre for Disease Prevention and Control, we defined a COVID-19 case as any person presenting fever, cough, shortness of breath, sudden onset of anosmia, ageusia or dysgeusia, or having characteristic CT-scan pulmonary images, or exhibiting positivity for SARS-Cov-2 on PCR (from nasopharyngeal or oropharyngeal swabs) or serology [9]. We used the WHO scale for progression of COVID-19 infection to assess severity (from 0, no infection, to 8, death): severe infection was defined as stage > 3 [10].

Participants

All children (< 18 years old) assessed for AAE were eligible. We excluded children for whom personal or family data was available. The study was approved by the Ethics Committee of Nice, France.

Outcomes

The primary outcome was the proportion of household members possibly infected by SARS-CoV-2 with COVID-19 other than the child with AAE (“index child”). Secondary outcomes were the proportion of severe COVID-19 within the family cluster and the description of demographic, clinical, virology and microscopy data for children with AAE.

Statistical approach

Quantitative data are expressed using mean, standard deviation (SD), median and interquartile range (IQR); categorical data are described with frequency (%). We used the Wilcoxon test to compare cutaneous features of AAE by age of the children. P  < 0.05 was considered statistically significant. We used R v3.6.2 for calculations.

Results

The study included 103 children from 10 French departments and from Quebec (Fig. 1 ). Median age was 13 years [IQR 8; 15] and 55 subjects (53.4%) were male (i.e., female/male ratio 1/1.15) (Table 1 ). Demographic characteristics are given in Table 1, Table 2 . Among the 103 children, 66 (64.1%) were in families having at least one member other than the index child with possible COVID-19. The total number of household members was 292; 119 (40.8%) possibly had COVID-19 according to the previously mentioned diagnostic criteria (Table 3 , Fig. 2 ) [8]. Among these 119 households, 77 had signs of infection before the index child was seen for AAE, and 16 had signs after the 1-month follow-up for the index child (data missing for 26 cases, Table 3). Cutaneous signs were present in 8 children other than the index cases in 7 families. No significant differences in age or sex were observed between these 8 children and those without cutaneous signs in their families. In total, all household members were aged < 60 years. No children or household members exhibited severe infection (no hospitalizations). Three cases among household members were in stage 1 of the WHO scale (2.5%), 113 were in stage 2 (95.0%) and 3 were in stage 3 (2.5%). No children showed multisystem inflammatory disease of childhood [11].

Fig. 1

Geographical distribution of children included.

Table 1

Characteristics of children with acute acral eruptions (AAE) (n = 103).

Characteristics	Number of available data	N (%)
Age, years
Mean ± SD–Median [IQR]	103	11.1 ± 5.2–13 [8–15]
Sex, males	103	55 (53.4)
Phototype	88
I		3 (3.4)
II		21 (23.4)
III		49 (55.7)
IV		9 (10.2)
V		2 (2.3)
VI		4 (4.5)
Overweight/obesity (body mass index > 25 kg/m2)	62	3 (4.8)
Passive smoking (second-hand smoke)	69	11 (15.9)
Personal history of chilblains	4	11 (11.1)
Semiological description of AAEa	103
Chilblain-like		82 (79.6)
Vesicles		19 (18.4)
Palmar/plantar erythema		42 (40.8)
Acrorhigosis		7 (6.8)
Acrocholosis		3 (2.9)
Purpura		9 (8.7)
Eccrine hidradenitis-like		6 (5.8)
Pulpitis		7 (6.8)
Acrocyanosis		13 (12.6)
Hyperhidrosis		10 (9.7)
Telangiectasia		4 (3.9)
Acral edema		14 (13.6)
Erythema multiform-like		3 (2.9)
Papules		2 (1.9)
Pain, numeric scale 0 to 10
Mean ± SD–Median [IQR] Patients with pain (score > 0)	103	1.9 ± 2.7–0 [0–3]50 (48.5)
Pruritus, numeric scale 0 to 10 Mean ± SD–Median [IQR] Patients with acral pruritus (score > 0)	103	2.6 ± 3.0–1 [0–5]58 (56.3)
Location	102
Hands		15 (14.7)
Feet		16 (15.7)
Hands and feet		71 (69.6)
Patients with other skin manifestationsa, b	103	21 (20.4)
Fever	102	9 (8.8)
Extracutaneous infectious signs	102	20 (19.6)

Individual patients may present several signs.

Skin manifestations included urticaria (n = 3), maculopapular eruptions (n = 3), erythema nodosa (n = 2), erythema of the face (n = 10), cheilitis (n = 1), livedo (n = 6); individual patients may present several signs.

Table 2

Age of children with acute acral cutaneous signs. Data are median (IQR); IQR: interquartile range; data were compared by Wilcoxon test.

Chilblain-like + n = 82	Chilblain-like–; n = 21	P-value
14.0 [11.0−15.0]	5.0 [2.0−11.0]	< 0.001
Vesicles + n = 19	Vesicles–; n = 84
14.0 [11.5−15.0]	13.0 [8.0−15.0]	0.37
Palmar/plantar erythema + n = 42	Palmar/plantar erythema −; n = 61
13.5 [3.0−15.0]	13.0 [9.0−15.0]	0.61
Acrorhigosis + n = 7	Acrorhigosis–; n = 96
14.0 [13.0–16.8]	13.0 [8.3–15.0]	0.09
Purpura + n = 9	Purpura–; n = 94
12.0 [2.0–14.0]	13.0 [8.3–15.0]	0.30
Eccrine hidradenitis-like + n = 6	Eccrine hidradenitis-like–; n = 97
6.5 [4.3−8.0]	13.0 [9.0–15.0]	0.02
Pulpitis + n = 7	Pulpitis–; n = 96
3.0 [2.0–7.0]	13.0 [9.0–15.0]	0.02
Acrocyanosis + n = 13	Acrocyanosis–; n = 90
14.0 [13.0–16.0]	13.0 [8.0–15.0]	0.048
Hyperhidrosis + n = 10	Hyperhidrosis–; n = 93
14.0 [13.0–15.8]	13.0 [5.0–15.0]	0.10
Acral edema + n = 14	Acral edema–; n = 89
14.0 [13.3–16.5]	13.0 [5.0–15.0]	0.03

Table 3

Number of cases within family clusters.

	N total	0 cases	1 case	2 cases	3 cases	4 cases	5 cases	6 cases
Families with l membera	16	10	6
Families with 2 membersa	24	9	6	9
Families with 3 membersa	34	10	13	7	4
Families with 4 membersa	21	5	4	3	5	4
Families with 5 membersa	7	3	1	2	1	0	0
Families with 6 membersa	1	0	1	0	0	0	0	0
Total number of families	103	37	31	21	10	4	0	0
Total number of households	291	93	84	61	37	16	0	0
Total number of infected cases among households	119	0	31	42	30	16	0	0
Total number of infected cases among households before the index child	77	0	17	31	18	11	0	0
Total number of infected cases among households after the index child at 1-month follow-upb	16	0	3	7	4	2	0	0

Household infection rate: number of infected cases among households/total number of households = 119/291 = 40.9%. Percentage of families with COVID-19: 66 (103–37)/103 children = 64.1%.

Other than the index child.

26 infected cases had missing data for time to onset of symptoms.

Fig. 2

Number of infected cases by number of members in each family.

The semiological characteristics of cutaneous signs in the 103 children (Fig. 3 ) are given in Table 1, Table 2, Table 4 . AAE cases showed heterogeneous features, with chilblain-like lesions (erythematous and violaceous papules and macules) being the most frequent (79.6%). These features occurred on both hands and feet in 69.6% (n  = 71) of cases and were associated in 21 cases (20.4%) with other cutaneous signs such as urticarial flares. Of these 103 children, 51 (49.5%) had at least 2 different acral skin signs and 20 (19.4%) had at least 3. In most cases, AAE was

Fig. 3

Clinical features of acute acral eruptions in children: 3a and b) chilblain-like lesions of the hands and feet; 3c) chilblain-like lesions; 3d) palmar erythema; 3e) acral vesicles; 3f) purpura; 3 g) eccrine hidradenitis-like lesions; 3 h) pulpitis; 3i) acral papular eruption.

Table 4

Characteristics of acute acral eruptions (AAE) according to whether household members were infected with SARS-CoV-2.

Characteristics	Children from household having at least one case (N = 66)	Children from household having 0 cases (N = 37)	P–value
Semiological description of AAEa, n (%)
Chilblain-like	55 (83.3)	27 (73.0)	0.211
Vesicles	13 (19.7)	6 (16.2)	0.662
Palmar/plantar erythema	28 (42.4)	14 (37.8)	0.650
Acrorhigosis	3 (4.5)	4 (10.8)
Acrocholosis	2 (3.0)	1 (2.7)
Purpura	7 (10.6)	2 (5.4)
Eccrine hidradenitis-like	3 (4.5)	3 (8.1)
Pulpitis	5 (7.6)	2 (5.4)
Acrocyanosis	11 (16.7)	2 (5.4)	0.128
Hyperhidrosis	3 (4.5)	7 (18.9)
Telangiectasia	4 (6.1)	0 (0.0)
Acral edema	7 (10.6)	7 (18.9)
Erythema multiform-like	3 (4.5)	0 (0.0)
Papules	0 (0.0)	2 (5.4)
Evolutionb, n (%)			0.888
Complete regression	23 (47.9)	13 (56.5)
Partial regression	17 (35.4)	8 (34.8)
Stable	6 (12.5)	2 (8.7)
Aggravation	2 (4.2)	0 (0.0)

Individual patients may present several signs.

Data are missing for 18 patients in the “Children from household having at least one case” group and 14 patients in the “Children from household having no cases” group.

not painful and was slightly itchy. Regarding age, chilblain-like AAE was more common in older children [median age 14, (IQR 11–15) with chilblains vs. 5 (IQR 2–11) without, P  < 0.01], as were acrocyanosis and acral edema (Table 2). Pulpitis and eccrine-hidradenitis-like eruptions were more frequent in younger children [median age 3 years (IQR 2–7) and 6.5 years (IQR 4.3–8) respectively].

At 1 month of follow-up, 36 of 71 (50.7%) contactable children with AAE showed complete recovery. AAE partially regressed in 25 (35.2%) children, was stable in 8 (11.3%), and worsened in 2 (2.8%). Median time from the onset of AAE to complete regression was 62.0 days [IQR 47.0–76.0]. The characteristics of AAE in the children did not differ significantly according to whether or not household members were infected with SARS-CoV-2 (Table 4).

Regarding complementary examinations in children with AAE, standard lab results and levels of inflammatory markers and coagulation markers were all either normal or not significantly increased (Table 5 ). Antinuclear antibodies were detected in 13 of 54 (24.1%) children. SARS-Cov-2 serology was positive for IgG in 2 of 14 (14.3%) children, and PCR results were negative in all children (n  = 18). Skin samples of chilblains were analyzed in 5 children (Table 6 ). Microscopy showed dermal perivascular lymphocytic infiltrates of differing intensity in all 5 children, as well as spongiosis (n  = 4) and lymphocytic vasculitis (n  = 2). Two children exhibited mucinosis.

Table 5

Laboratory and virology results for children with acute acral eruptions.

	Number of tests performed	Anomalies N (%)
Standard lab results	60	2a (3.3)
Creatine phosphokinase	48	2b (4.2)
Inflammatory markers	57	1c (1.8)
Coagulation markers	40	1d (2.5)
Antinuclear antibodies	54	13e (24.1)
Cryo-globulinemia/fibrinogenemia	32	1f (3.1)
PCR SARS-Cov-2 (nasopharyngeal swab)	18	0
SARS-Cov-2 serology	14	2g (13.3)

One case of slight thrombocytosis (532 000/mm3, N < 450 000) and one case of slight anemia (Hb 11.2 g/dl, N > 11.5).

Slight increase: 183 and 262 (N < 170).

C-reactive protein level increased in one case (CRP 49.3 mg/l, N < 5).

Very slight elevation of D-dimers (512, N > 500).

When positive, antinuclear antibodies ranged from 1/80 to 1/640.

Moderate polyclonal IgM cryoglobulinemia type 3 (9.80 mg/l).

Positive IgG, IgM (–).

Table 6

Microscopy features of skin samples (5 cases of chilblains).

Vacuoli-zation of basal layer	Dermal perivascular lymphocytic infiltrate	Presence of eosinophils	Lymphocytic vasculitis	Keratinocyte necrosis	Spongiosis	DIFa	Mucin deposition (Alcian blue staining)
+	+++	–	+	+	–	–	+
–	+	+++	-	–	+	-	+
–	++	+	–	–	+	±	NPb
–	+++	–	–	–	+	+	NPb
–	++	–	+	–	+	+	-

DIF: IgM deposition, no IgA, no IgG, no C3.

NP: not performed.

Discussion

In our study of 103 children with AAE, COVID-19 was possibly present in 40.8% (119/291) of household members, and for 66/103 (64.1%) of children, at least one family member than the index child had infectious signs. AAE characteristics differed according to age. Host characteristics related to patient age appeared to be an important factor in skin lesions observed in children during the COVID-19 pandemic.

Since the beginning of the COVID-19 pandemic, many cases of AAE were observed during the same period in different areas in France and Quebec. A causal link between COVID-19 and AAE has not yet been demonstrated and remains debated. Such a link appears likely since the period during which we observed AAE in France and Quebec was associated with COVID-19 progression and was chiefly during the spring months [12], [13].

Evidence of the presence of COVID-19 in skin biopsies of chilblains has recently been reported [14], [15]. However, several studies based on serological results have found no evidence of a causal link between COVID-19 and AAE [16], [17], [18]. Confirmation of COVID-19 in patients presenting AAE may not be based on classical tests such as PCR (since AAE usually occur after the infectious symptoms) or serology. Indeed, serology appears to have diagnostic efficacy, especially in asymptomatic and paucisymptomatic patients [19].

Neither index children nor households showed severe signs of COVID-19. Where available, laboratory data showed no significant abnormalities. The types of clinical manifestations associated with COVID-19 were associated with host factors, with age being the most important [20]. Severe COVID-19 is more frequent in older people. Neither chilblains nor other acral eruptions have been reported in older people but are associated rather with younger subjects [16], [18]. These data underline the importance of host characteristics in COVID-19–associated medical conditions. Efficient innate immune response and/or cross T-cell immunity induced by previous infection with α coronavirus were explored in patients and relatives presenting such dermatological conditions [21], [22], [23]. Patients with chilblains related to COVID-19 may have an efficient immune response not associated with the development of specific COVID-19 antibodies. The estimated percentage of individuals in contact with SARS-CoV-2 is based on serology. If a causal link between COVID-19 and AAE is confirmed, this condition could provide a relevant marker of COVID-19 circulation.

Regarding AAE, chilblain-like lesions were predominant in the children in our study (79.6%); however, we observed a variety of other acral manifestations on the hands or feet, or both, sometimes coexisting in the same patient (49.5%). This finding suggests that AAE could represent a continuous clinical spectrum in which eccrine-hidradenitis–like eruptions or desquamative pulpitis are frequent in infants and chilblain-like lesions are more common in teenagers. The histological findings in our patients were consistent with data from previous larger studies. Findings from chilblain biopsies showed a pattern of interferon (IFN) type I-induced inflammation and vasculitis induced by COVID-19 [4], [17], [24], [25]. Of note is the fact that this skin condition has not been reported in severe COVID-19 patients, who might have impaired IFN type I activity compared to mild COVID-19 cases and to controls [26]. Furthermore, it has been reported that patients with chilblains observed during the COVID-19 pandemic had a significantly higher IFN alpha response compared to severe COVID-19 patients [25]. Some authors have thus hypothesized that patients with chilblains might present an efficient antiviral response triggered by SARS-CoV-2 [27]. Furthermore, in our young population, clinical presentation differed according to age. Host factors play a central role in the clinical presentation of COVID-19. Vessel characteristics and immune status in the host may play a key role. Vessel characteristics and IFN type I response vary according to host characteristics, including age [28]. Such factors could account for the differences between dermatological presentations observed between children and adults as well as those between infants and adolescents. We observed no acral ischemia in our study. This serious complication reported in adults is associated with a pro-thrombotic condition. Values for coagulation markers in the children in our study were normal. Such thrombotic complications observed in COVID-19 may be driven by other factors related to host characteristics.

Only scant data are available on the evolution of acral lesions. Unlike general signs, such as fever or anosmia, which are presenting symptoms of COVID-19, AAE seems to occur later during the disease, from 0 to several weeks after the first symptoms of suspected viral infection or after presumed contact with a carrier of the virus, as reported in previous studies [5], [17], [29]. This finding could account for the negative PCR results in our study and in most previously reported cases. However, 3 adolescents with chilblains recently proved positive on RT-PCR with nasopharyngeal swabs [30]. Most lesions had a long but benign course, with complete spontaneous regression after a median of 47 days.

The main limitation of the study is the absence of a control group to confirm intrafamilial contamination. Further, COVID-19 was based on serology, PCR or clinical symptoms according to criteria of the European Centre for Disease Prevention and Control, which may have either over- or underestimated the proportion of households infected. Family members did not undergo systematic PCR testing with nasopharyngeal swabs, which was only performed in hospitalized patients or in patients with severe respiratory symptoms during the inclusion period. In addition, not all patients or household members underwent serology testing, and the serologic techniques themselves were heterogeneous.

In conclusion, AAE exhibit heterogeneous semiological features in children. Confirmation of their link with COVID-19 requires further investigation. The absence of AAE in severe COVID-19 cases and differences in clinical presentation by age group underline the role of host characteristics in skin conditions associated with COVID-19.

Funding

None

Disclosure of interest

The authors declare that they have no competing interest.

This study was approved by the Ethics Committee of Nice.