Chilblains-Like Lesions in Pediatric Patients: A Review of Their Epidemiology, Etiology, Outcomes, and Treatment

Vascular pathologies associated with SARS-CoV-2 infection are poorly understood. Color and sensory changes to the extremities, often referred to as “COVID toes” or chilblains-like lesions, have been widely reported in children and adolescents since the onset of the pandemic, raising the concern that they could be a vasculitis secondary to the infection. However, it is unclear if the lesions are a result of the infection or an epiphenomenon. Most literature focuses on adults, and while there are reports on children and adolescents, many of them are small. This review will help medical care providers better understand the epidemiology, etiology, outcomes, and potential treatments for chilblains-like lesions seen in children and adolescents during the pandemic.


INTRODUCTION
The severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) first appeared in China in late 2019 and was declared a pandemic within months (1). It has infected over 437 million people and has caused over 5.9 million deaths (2). Following the onset of the pandemic, reports of acral skin changes, often referred to as chilblains-like lesions or COVID toes, emerged in children and adolescents.
Chilblains is an uncommon condition in pediatric patients, and its prevalence and incidence are not well reported (3,4). Chilblains occurs most notably in people exposed to the cold and damp, as well as those who participate in cold weather sports such as hockey, figure skating, and speed skating (4). In areas that tend to be drier and/or warmer, the condition is rarely reported; only eight pediatric cases were noted in a 10-year retrospective study done at an academic, ambulatory care center in Colorado (5).
Following a cold exposure, chilblains lesions commonly manifest as cold, cyanotic nodules and papules that may evolve to include swelling, pain, pruritis and ulcerations (3). Extremities are commonly affected, particularly fingers and toes, as well as the ears and nose (3).
Chilblains may be the first symptom of an underlying condition, such as anorexia, antiphospholipid syndrome, Sjögren syndrome, rheumatoid arthritis, Behçet syndrome, STINGassociated vasculopathy of infantile onset, IRAK4 deficiency, systemic lupus erythematosus, Aicardi-Goutierès or Familial chilblain lupus (3,6). The latter two are rare conditions caused by mutations in the TREX1 gene and manifest with chilblain lesions following cold exposure (3,7).
Chilblains also occurs during infection with viruses that have interferon signatures, such as parvovirus B19 and Cytomegalovirus (8,9). As a high interferon signature in SARS-CoV-2 has been observed, there has been concern that chilblains could be a vasculitic manifestation of SARS-CoV-2 infection (10).
However, few patients with chilblains-like lesions during the pandemic have confirmed cases of SARS-CoV-2. This has generated debate about the etiology and significance of the lesions. This article will review the epidemiology of affected patients, the clinical and pathologic findings of the lesions, the current hypotheses about lesion pathophysiology, patient outcomes, and the implications of these findings.

Inclusion Criteria
Inclusion criteria include a pediatric study population, defined as ≤18 years old, the ability to separate the SARS-CoV-2 infection data of pediatric from adult cases if reported together, and potential mechanisms of lesion etiology, regardless of patient age.

Exclusion Criteria
Articles were excluded if the article was not written in English, the study cohort age was >18 years old or not reported, SARS-CoV-2 status of pediatric cases was not separable from adult cases, the primary focus of the paper was not on chilblains-like lesions, or if it was a review.

RESULTS
The literature search generated 359 results. An additional seven articles were found while looking at the articles from the search. Following screening, 53 articles were included in this review. See Figure 1 for compete details. In total, 1,059 reports of chilblainlike lesions were reported (6,.
One study reported that the majority of lesions were on feet (96.3%), followed by hand lesions (11.9%), and lesions on the head or the neck (11.4%) during the spring of 2020 (11). Among patients with feet lesions, toes were the most commonly affected area, but changes were also noted on the periungual areas, heels, and the dorsum of the feet (11).
On dermoscopy, there were 12 reports the background color of the lesions as pink to erythematous to violaceous erythema; a coppery red was also described in six cases (6,34,39). Additional findings included rosettes in 31 cases, red dots in 29 cases, dotted vessels in 26 cases, white streaks in 25 cases, capillary dilation in 15 cases, abnormal capillary morphology in 10 cases, pigmented areas in nine cases, vessel ectasia in nine cases, scales in eight cases, hyperpigmentation in seven cases, microhemorrhages in seven cases, unstructured areas with color changes in seven cases, and hemorrhagic dots in six cases, edema in five cases (6,22,39,51,54). There were winding vessels in three cases, irregular linear vessels in two cases, cuticular vessels in two cases, linear vessels two cases, crusts intwo cases, and desquamation in two cases (22,39,51). There was one case of each of the following: vessels with branches, reduced capillary density, bullae, and multiple short vessels in a perpendicular arrangement (39,51). An unspecified number of ischemic areas and purpuric dots were reported (6) Supplementary Table S1.
Deposition of complement was present in 26 cases, and there were 5 reports of immunoglobulin M, and 1 report of immunoglobulin A (11,26,45,52). Two studies reported the presence of viral structures that could be SARS-CoV-2 particles (18,26). However, biopsies were negative for SARS-CoV-2 in 44 samples (22,33,43,47,51,52). One study reported a positive SARS-CoV-2 testing on a chilblains-like lesion, and there were an additional eight positive tests on other biopsies (18,33,45). Additional details about the histopathology are detailed in Supplementary Table S2.

Characteristics of Patients
Chilblains-like lesions have been reported in children younger than 1 year old, though most cases were in late childhood to early adolescence (11). Table 1. In a Spanish cohort including 22 patients, Andina et al. found that a majority of children affected with chilblains-like lesions were between the ages of 10 and 13 years old; 14 to 17 years old children were the second most common group, and patients 9 years old and younger were the least frequently affected group (6). Caselli et al. reported an Italian cohort with 38 cases (16). Ages ranged from 7 to 18 years old, with a median age of 13.5 years old (16). In a multi-country pediatric dermatology registry, 378 children had a mean age of 13 years old (11).
Most studies have reported a male predominance. Castelo-Soccio et al. reported 229 males and 148 females affected by chilblains-like lesions (11). Among the case reports and case series reviewed, there were 326 males (57.39%) and 242 females (42.60%) (6, 10-57). Most case reports and series did not report race or ethnicity. Most studies did not specify nationality, but as a majority of articles were case reports or series, the population was assumed to be native to the country unless otherwise indicated. Though not common, there are several reports of siblings having chilblains (6,20,23,24,41,49,51,54). See Table 1 for additional demographic details.
The illnesses in potential SARS-CoV-2 contacts varied. In one case, the patient's family was sick: one parent had a cough and the other had a flu-like illness (28). Hubiche et al. reported symptoms of illness in 77 families (48). In several cases, patients were exposed to people with upper respiratory tract infection symptoms (12,20). Three cases had recently returned from international travel (20). One study noted that family members had been ill (headache, gastrointestinal symptoms, cough, or fever) in the 1 to 2 months preceding lesion onset (51).
The time from systemic symptoms of SARS-CoV-2 development to lesion onset was variable. In 58 cases, systemic symptoms preceded the onset of lesions by at least a week, and in 4 cases, the lesions occurred within 7 days of systemic symptom onset (6, 15-17, 19-22, 24, 27, 30-32, 36, 46, 50, 51, 54, 57). There were also concurrent appearances of lesions and systemic symptoms (7). One study noted 3 cases that had simultaneous or overlapping appearance of lesions and systemic symptoms, while another noted 4 cases in which the time from systemic symptom onset to lesion onset ranged from 0 days to 7 days (6, 42). One study reported systemic symptom onset a mean of 22 days before the lesion onset in six cases, with a range of 5 to 46 days. Another study reported a mean of 12.62 days from systemic system onset to lesion onset (14). Lesions preceded systemic symptoms in 4 cases (14,26,51) Supplementary Table S5. Among siblings with a history of chilblains-like lesions, siblings often developed lesions within several days to a week of the index case (20,23).
A variety of co-morbid medical conditions have been reported, including attention-deficit/hyperactivity disorder in eight cases, asthma in four cases, atopic dermatitis in two cases, and celiac disease in two cases (6,14,16,18,30,53). There was one case of each of the following: headaches, allergic rhinoconjunctivitis, seasonal allergies, urticaria, diabetes mellitus, Crohn disease, nephrotic syndrome, alopecia areata universalis, Kawasaki disease, Wolff-Parkinson-White, peripheral neuropathy, drug allergy, and an undisclosed X-linked condition (14-16, 28, 32, 46, 52). Two studies noted a family or personal history of coagulation disorders, though anther study did not find an association (15,16,20) Supplementary Table S4.
Raynaud phenomenon and/or chilblains were reported in 12 cases, but several studies did not elicit this history (6,13,14,18,22,24,25,31,38,41,43,46,49,52,53). Two studies reported ten cases with autoimmune disease, but specific conditions were not reported (15,16). One study reported a patient with systemic lupus erythematosus, and another study reported a case with IgA vasculitis (22,41). One study reported only 18.3% of patients had comorbidities, though the comorbidities were not detailed, with the exception of one patient who was found to have Sjögren disease (11) Supplementary Table S4.
Among studies reporting medication use, 31 patients either did not take medications or had been on stable doses for at least a month before the appearance of chilblains-like lesions (12,13,32). One patient had received acetaminophen intermittently around the time of lesion onset, and one patient had recently started ferric sulfate (41). Seven patients with ADHD reported no changes to their treatment in the 6 months prior to lesion onset (6, 13) Supplementary Table S4.
There is little detail reported about family history. There were 5 cases of autoimmune thyroiditis,4 cases of atopy, 3 cases of chilblains, 2 cases of hemolytic anemia during pregnancy, and one report of each of the following conditions: thromboangiitis obliterans, inflammatory bowel disease, dyshidrosis, and hyperhidrosis (22,34,35). Several articles reported no autoimmune or inflammatory conditions (11,20,24,35,54). There were 8 reports of coagulation disorders in cases or in their families (15,16) Supplementary Table S6.
Twenty-three siblings had lesions: two pairs of brothers (four cases), a pair of sisters (two cases), four pairs of siblings with unspecified sex (eight cases), and three sets of three siblings (nine cases) (6,20,23,24,49,51,54). There were an additional six cases who had unspecified family members with lesions (41) Supplementary Table S6.
While the data was not robust about the lifestyles and activity levels in patients during the SARS-CoV-2 pandemic, 13 cases noted a decrease and 1 case reported a stable amount of activity (52). Fourteen cases were watching more television during the SARS-CoV-2 pandemic than prior to the pandemic (52). One study reported 8 patients going barefoot or only wearing socks during the pandemic (35). Additionally, there were 26 cases of patients walking barefoot and 5 cases of patients walking with socks (41,51). Routine shoe wear was reported in 1 case; 15 cases reported patients not wearing shoes regularly at home  (41,52). Seven cases reported exposure to cold floors, and 19 cases reported no exposure to cold (35,53,54). Accessibility to heating was variable: 10 cases reported living in heated homes, 18 cases reported heating was not available in the home, and 5 cases reported that heating was off (35,41,43) Supplementary Table S5.
In a majority of cases, the chilblains-like lesions were asymptomatic or caused mild symptoms (15,25,42,44,51). For patients with very painful, pruritic, or swollen lesions, treatments included topical steroids in 162 cases, oral analgesics in 49 cases, oral antihistamines in an unspecified number of cases, oral steroids in four cases, and emollients in two cases (1,6,11,13,23,29,32,35,36,38,44,46). There were two cases in which a patient was treated with a calcium channel blocker (one topical and one oral); one of these patients also received a topical anticoagulant (11,24,44). Additional treatments included topical steroids or antibiotics in six cases, prophylactic heparin in one case,  Seventeen patients were excluded from these results as they were reported in a prior study (22). ¶ Study included 4 previously reported patients, of which 1 developed positive serology testing in this study; that patient is reported here and omitted from original study (19). ‡ Two patients in this study were reported as having photosensitive rashes; they are not included in this paper. mometasone furoate cream in one case, hydroxychloroquine in 1 case, azithromycin in 1 case, and instructions to avoid cold and/or damp environments in six cases (29,30,35,42,44). Warming measures were recommended in 74 cases (11). Fifty-seven unspecified treatments were also reported, and 59 cases did not receive treatment (11,19,24,25,51,57). A number of treatments were also reported without specified numbers: Supplementary Table S2.
Four cases that had concomitant erythema multiform also received topical and oral steroids (6). Outpatient therapy was usually sufficient for treatment. In one study, a patient with chilblains-like lesions who was not included in the analysis (reported after initial cohort was selected) required hospitalization for treatment of SARS-CoV-2 infection (11).
Some patients tried medications they had at home before visiting a doctor. Treatments included disinfectants in four cases, topical antibiotics and/or corticosteroids in 4 cases, systemic antibiotics or corticosteroids in three cases, and anti-fungal treatments in three cases (22) Supplementary Table S2.
While lesion improvement followed by resolution was frequently reported, there were reports of outliers. There were 83 cases with improvement, 24 cases with stable or ongoing lesions, three cases with relapse of lesions, and two cases in which the patients worsened (6, 12-14, 17, 32, 37, 44, 48, 51) Table 3. In the study with worsening patients, the rest of the cohort fared well, with stable lesions in eight cases, improvement of lesions in 25 patients, and resolution of lesions in 36 patients (48). Hyperpigmentation was seen in some lesions during follow up (6,22,54). Skin desquamation was rarely reported with lesion healing (22). Nail changes were also noted in some patients. Onychomadesis was observed in two patients 8 to 10 weeks after the onset of skin lesions and was limited to toenails (21) Supplementary Table S1.

DISCUSSION
This review highlights the lack of SARS-CoV-2 positivity among patients with chilblains-like lesions despite temporal association and the positive outcomes among patients with lesions.
While SARS-CoV-2 testing was limited at the start of the pandemic, the majority of studies that utilized testing later on failed to identify the infection. A possible explanation for this could be that the lesions are secondary to lifestyles changes, leading to alterations in extremity blood flow and in color as seen in other conditions, such a frostbite and trench foot (58,59). Blood flow may decrease as much as 90% during cold exposure (58). However, some literature suggests that SARS-CoV-2 may be the inciting agent. The spike (S) protein of the SARS-CoV-2 virus binds with angiotensin-converting enzyme 2 (ACE-2) receptor to enter cells, the envelope protein of the virus was shown to amplify inflammasome activation, and the nucleocapsid protein has an immunomodulatory function that could also increase inflammasome activation, possibly further facilitating this process (1,60). Other studies suggest that SARS-CoV-2 virus may disrupt the communication between endothelial cells and pericytes leading to endothelial cell injury (61).
An increase in interferon is observed in some patients with chilblains-like lesions, though as noted previously, many patients test negative for SARS-CoV-2. Genetic studies done by Frumholtz et al. showed an upregulation in genes associated with natural killer and cytotoxic T cells, both of which occur in the presence of increased interferon (62). The work of Magro et al. also supports the importance of type I interferon signaling in biopsies of chilblains-like lesions in patients, though in this cohort, 2 of the patients had negative testing for SARS-CoV-2 and the third patient was not tested (10). Of note, children have a strong type I interferon response during infection with SARS-CoV-2, and some authors suggest that this response may blunt or entirely suppress the immune system's reaction to the virus (10,20,23). There is a lack of detectable infection or symptoms of illness, these children could still develop chilblains-like lesions in response to the high levels of interferon (24,39).
Despite this possibility, the increased reports of chilblainslike lesions in countries with very low SARS-CoV-2 infection rates raise the possibility of another factor, and possibly multiple other factors, giving rise to chilblains-like lesions (63). In Nordic countries where strict lockdown measures were not initially imposed, there were few cases of chilblains-like lesions (63). When the lesions did occur, patients responded well to topical steroid treatments and lifestyle modifications focused on warming the extremities (63).
Contrary to this, when lockdown measures were in place, an increase in chilblains-like lesions appeared. Four Australian children had chilblains-like lesions during a strict lockdown period (53). At the time, there were 7000 cases of COVID-19 in the country and <1,700 in the region, making an exposure to virus unlikely (53).
An increased number of chilblains-like lesions were also observed in the United States during the first wave of SARS-CoV-2 lockdown (11). Children reported not wearing socks or shoes while doing their schoolwork for prolonged periods of time, which suggests that environmental factors could play a role in the appearance of chilblains-like lesions (11). Additional studies confirmed that many pediatric patients with chilblainslike lesions were not wearing shoes, and only some were wearing socks during the pandemic (35,41,51). In a Spanish study, fifteen of twenty patients reported walking barefoot in their homes, and none of their homes were heated during that time (41).
Furthermore, pediatric patients and their caretakers spent more time together during the pandemic, affording caretakers an increased opportunity to notice the lesions. Media reports on the chilblains-like lesions may have increased awareness of these lesions and may have prompted families to seek medical attention (22).
The SARS-CoV-2 pandemic caused, and continues to cause, a great deal of anxiety and stress, both of which may contribute to the development of amplified musculoskeletal pain syndrome (AMPS) (64,65). This condition may cause changes in skin color and temperature; allodynia may also develop (66). While AMPS does not explain a majority of the reported chilblains-like lesions, it could explain some of the lesions that last for prolonged periods of time and that have an intense sensory component.
It is important to educate families that chilblains-like lesions are not associated with the severe morbidity and mortality that occur with SARS-CoV-2 infection and will resolve without treatment. However, it is equally important that they know the lesions may cause some discomfort and are instructed to seek care for persistent symptoms. The recommendation of supportive care measures, such as warm shoes and socks, are also appropriate as this may prevent the lesions from worsening and may facilitate resolution. If there is concern for AMPS, a trial of physical therapy may be beneficial as increased physical activity is the mainstay of treatment for this condition. Additionally, mental health services may be offered to help ease the psychological burden of the pandemic. If an underlying condition, such as systemic lupus erythematosus, is suspected, a referral to a specialist would be beneficial.
The lack of positive SARS-CoV-2 testing despite the presence of chilblains-like lesions suggests that SARS-CoV-2 infection is not the underlying etiology. The lesions may be incidental findings, possibly secondary to lifestyle and environmental changes. If chilblains-like lesions truly are an epiphenomenon, they may disappear as people resume their activities prior to the SARS-CoV-2 pandemic. Future studies are needed to evaluate the evolution of these lesions. In conclusion, this study supports current literature that the chilblains-like lesions seen during the SARS-CoV-2 pandemic are likely secondary to a multifactorial epiphenomenon as most patients test negative for the virus and that the majority of outcomes are good.