Pseudo-Chilblains in Adult Patients with Confirmed COVID-19: A Systematic Review

Background: Pseudo-chilblains have been associated with COVID-19. Many reports, however, lack confirmation of COVID-19 infection. While likely associated, all chilblains/chilblain-like lesions during this time should not be assumed to be COVID-19 related. This study examines the characteristics of adults with pseudo-chilblains and confirmed COVID-19. Methods: A systematic review of PubMed/MEDLINE database was performed using the PRISMA guidelines. Adults (>18 years) with confirmed COVID-19 were included. De-identified registries were excluded to avoid duplication. We extracted study design, age, sex, race, geographic location, relationship of COVID-19 diagnosis to chilblains onset, confirmatory testing, hospitalization status, anatomical location, cold/damp exposure, presence/absence/description of pseudo-chilblains symptoms, presence/absence of biopsies/histopathologic findings, tissue IHC/PCR, presence/absence/details of extracutaneous COVID-19 disease, pre-existing chilblains, treatment and resolution timeline. The search was completed in July 2022. Results: We identified 13 studies (29 patients). In COVID-19-infected adults, pseudo-chilblains were reported primarily from North America and Europe, occurring in both sexes over a wide age-range, affected well and ill patients, favored the hands and feet and could be symptomatic or asymptomatic. Most patients had extracutaneous symptoms. Resolution time ranged from <1 week to >50 days. There was marked variation in treatment strategies and appearance of pseudo-chilblains relative to entire disease course. Biopsies were infrequently performed but findings similar to classical chilblains were described. Conclusions: Many patients reported as pseudo-chilblains of COVID-19 lack confirmed infection. Infection confirmation, photographic documentation and histopathology are critical to establish homogeneity in reported pseudo-chilblains during this global pandemic. Further work clarifying the relationship of acral eruptions and COVID-19 is necessary.


Introduction
Recent reports document cutaneous manifestations of coronavirus disease of 2019 (COVID-19) infection including exanthematous, urticarial, papulovesicular and vascular-related eruptions. 1Acral lesions described early in the pandemic were designated 'pseudo-chilblains', 'COVID-toes' or 'chilblain-like' due to their resemblance to classical chilblains.2][3][4][5] The lesions may be painful, pruritic or asymptomatic and occur in both children and adults, with equal distribution between sexes.While the pathophysiology of pseudo-chilblains is still unclear, viral infection associated increased interferon-α, a strong cytotoxic T-cell and natural killer cell response, along with IgA anti-neutrophil cytoplasmic antibodies have been described. 6is immune response likely contributes to the dense perivascular and periadnexal lymphocytic infiltrate seen on histopathologic sections. 6Cryofibrinogenemia with potential resultant vascular microthrombi has also been reported as a potential pathomechanism. 7In addition to being a marker of COVD-19 positivity, prognostic implications have been suggested, 4 with pseudo-chilblains reportedly associating with mild disease. 4One challenge with the data regarding its association with COVID-19 is the lack of confirmed infection in many studies and whether this eruption is a true manifestation of COVID-19 infection remains controversial. 8In many reports, infection was inferentially deduced using known contact exposure or previous suggestive clinical symptoms rather than confirmed laboratory testing. 5lthough little doubt exists that pseudo-chilblains are a manifestation in some patients with COVID-19 infection, it should not be assumed that it is exclusively seen in COVID-19 infected patients. 9Lack of clinical criteria, variation in appearance and infrequently performed biopsies raise the possibility that pseudochilblains may not be a homogenous condition, potentially

Study Selection
Two authors (SH, MG) independently screened titles/abstracts identifying and including articles describing pseudo-chilblains in patients with confirmed COVID-19 infection (defined as positive reverse transcriptase polymerase chain reaction (RT-PCR), positive serology for IgG/IgM or detection of COVID-19 on biopsies via immunohistochemistry/immunofluorescence (IHC/IF), in situ hybridization (ISH) or tissue PCR.Where there was disagreement on inclusion/exclusion a third author (KW) was consulted for consensus.Eligibility of study based on data available for extraction was determined through full-text review with consensus between two authors (SH, KW, NT, JM) and final review by consultant dermatologist (JH).Studies involving data extracted from de-identified patient registries, such as the American Academy of Dermatology Association COVID-19 Dermatology Registry (https://www.aad.org/member/practice/coronavirus/registry) were excluded to avoid duplicated patient representation.The inclusion/exclusion criteria were decided and vetted using multiple practice runs during planning meetings prior to July 14 th .With the criteria decided, a single run was completed on July 14 th , 2022 for PubMed and July 17 th , 2022 for Web of Science.Microsoft Word was used to organize and manage the yielded citations.Once there was consensus on the included studies, Microsoft Excel was used to extract the required data from the papers.

Data Extraction
Data extracted included study design, number of patients with confirmed COVD-19 and pseudo-chilblains, age, sex, race, geographic

Quality Assessment
The Joanna Briggs Institute critical appraisal checklists (2017) for case reports, case series, cross-sectional and cohort studies 11 were utilized to assess the overall quality of the included studies and estimate the risk for bias.For example, we assigned "Yes" to the question "Was the patient's history clearly described and presented as a timeline?"only if there was well-detailed chronology and timing of events reported.Similarly "Yes" would be assigned to "Were valid methods used for identification of the condition for all participants included in the case series?"only if a standard method of diagnosis was utilized (PCR, antibody testing etc.).All of our case reports and series had at minimum Records after duplicated removed (n=116) Eligibility "Yes" assigned to criteria 1-4 and for cohort and cross-sectional studies, at minimum "Yes" assigned to criteria 1-3 and 7.

General study details
The flow diagram of the search and study selection process is shown in Figure 1.The literature search resulted in 116 articles which were evaluated for relevancy based on their titles and abstracts.Following title and abstract review, 45 studies were excluded for lack of confirmed infection (n=8) or absence of primary data (n=14).Review articles were also excluded (n=23).Seventy-one articles remained for full text reading.Of these, 58 were excluded for lack of confirmed infection in some/all subjects (n=11), inability to extract data due to vague reporting (n=32), lack of confirmed clinical features of chilblains-like lesions (n=13) and global databases (n=2).4][15][16][17][18][19][20][21][22][23][24][25] Extracted data is shown in Table 1 and Table 2

Quality Assessment/Risk of Bias
The majority of included studies fulfilled most of the study-type appropriate Joanna Briggs Institute Critical Assessment checklist parameters (Tables 3-4).For case reports/series missing information was primarily related to the adverse reactions which were generally not relevant based on the subject being studied.
Similarly, for observational studies (cohort and cross-sectional studies), information on confounders was not generally available.
Overall, based on the assessment of the critical appraisal checklists, all but one of our studies had >70% "yes" answers to relevant/applicable criteria (See Table 3-4).Therefore, while not negligible, we assessed the risk of bias as relatively low.

Patient Demographics
The included studies yielded information on 29 patients.Sex and specific ages were evaluable for eleven of the thirteen studies (19 cases).There were 8 males and 11 females.Ages ranged from 19-82 years.The remaining studies provided age ranges for their entire cohorts and minimum (55) and maximum (77) ages could be deduced.Race was generally unreported.24] Four patients were collaboratively reported between the United States of America and Brazil, 14 one study detailing 2 patients from Qatar 17 and a single patient was reported from Southeast Asia (Singapore). 19

Clinical Characteristics
Regarding clinical presentation, twelve studies reported hospitalization status; 13-24 15 outpatient and 16 inpatient cases were reported (unreported in one study of three patients).254] Exposure to cold/damp was excluded in four studies, (10/29 cases) and unreported in the remainder. 15,16,18,20 Antomical locations included toes/feet, hand/fingers, ears, arms and legs.28/29 patients had involvement of hands/feet/digits.There were two reports of ear involvement, one patient with an ear-only lesion. 14,24 Tes/feet were the most commonly reported single location.

Discussion
][28] As in other viral eruptions (e.g., unilateral laterothoracic exanthem), numerous agents may produce similar findings and care must be taken in ascribing causality.Furthermore, the frequent lack histopathologic confirmation, variation in clinical appearance and microscopic features, and absence of clinical photographs for many reports raises the possibility that the designation pseudochilblains/''COVID-toes'' may represent a heterogenous group of conditions with similar anatomic distribution.This study aims to contribute to our evolving understanding of COVID-19associated skin disease by specifically examining the features of pseudo-chilblains in adults from studies where patients were definitively infected.It should be noted a positive serologic test or RT-PCR for COVID-19 is not necessarily an indicator of active infection in otherwise asymptomatic patients, as both may remain positive for some time after infection. 29Perhaps in some patients, pseudo-chilblains represent a delayed reaction to recent but inactive infection. 30r analysis suggests that many reported cases of pseudochilblains do not detail laboratory confirmation of COVID-19 infection.In studies meeting our inclusion criteria, we found pseudo-chilblains in adults occurred in both sexes over a wide age range (2nd-9th decades).Most cases were reported from non-equatorial countries.The apparent geographic distribution and acral localization may implicate environmental factors as concomitant triggers. 3eudo-chilblains have been suggested as a marker for mild disease. 4While the number of cases evaluated in this study is too small to confirm or refute this, it is noteworthy that pseudochilblains occurred in both well outpatients and persons hospitalized with COVID-19 complications. 17,18 hile details of the onset of pseudo-chilblains relative to overall disease-course were not clear in most studies, where evaluable, pseudochilblains could occur from Day 1 of illness to six weeks from initial symptoms, suggesting its potential appearance in acute and more chronic phases of infection, or perhaps in patients with recent but inactive infection.Cold/damp exposure was excluded in 10/29 of the cases.Unfortunately, a history of previous conventional chilblains was generally unreported.Currently pathomechanistic similarities/differences of conventional and pseudo-chilblains are not known.
Pseudo-chilblains could be either asymptomatic or symptomatic.Extracutaneous symptoms were present in greater than two thirds of cases analyzed but no characteristic pattern could be elucidated with respiratory, sensory, gastrointestinal, headache and fever being represented.Resolution time was likewise heterogenous some patients resolving within a week and others longer up to 50 days.Therapeutic approach was not standard and included anti-inflammatory and analgesic agents, anticoagulants, and observation.Regrettably, biopsies were not performed in the majority of cases examined nor in larger global registry reported cases. 5Reported histopathologic features include vacuolar change, spongiosis, necrotic keratinocytes, a superficial and deep perivascular and perieccrine lymphocytic/lymphohistiocytic infiltrate, lymphocytic vasculitis, subepidermal blister formation, papillary dermal edema, extravasation of erythrocytes, increased intradermal mucin and microthrombi. 5,31In our included cases, intraepidermal vesicular (dyshidrotic-like) dermatitis and a superficial and deep perivascular and perieccrine lymphocytic infiltrate were described.While further work outlining histopathologic changes is needed, a perivascular and periadnexal lymphocytic infiltrate similar to conventional chilblains appears to be common, though not universal. 13,25,32 Iterestingly, biopsies may aid in tissuebased confirmation of infection. 25In 4 out of 5 cases, COVID-19 spike protein was visualized via IHC/IF in vascular endothelium and in eccrine epithelium despite negative nasal PCR and/or serology.It is important to note that like nasal/nasopharyngeal RT-PCR and serology, spike protein identification may not equate to active infection.The spike protein is thought to be cleaved, entering endothelium/epithelium via the angiotensin converting enzyme type two receptor 25 but how long it remains within these cells is unclear.
Based on our analysis, features of classical chilblains and pseudochilblains in adults with confirmed COVID-19 infection were compared.Typical chilblains present with painful, acral, erythematous/livid lesions in young, predominantly female patients within the Northern Hemisphere after exposure to cold/damp conditions. 33Microscopic features include superficial and deep perivascular and perieccrine lymphocytic infiltrates, papillary dermal edema and extravasation of erythrocytes. 34imilarities include anatomical and perhaps geographic distribution, morphology and some histopathologic findings.Differences include the often asymptomatic nature, potential for chronicity, lack of exposure to cold/damp, variability in histopathologic findings and the occurrence over a broad age range in both sexes in COVID-19 related lesions compared with classical chilblains.Limitations to this study include the retrospective nature of systematic reviews, occasional methodologic gaps in some of the included studies and the exclusion of cases from large databases where confirmation of COVID-19 status was unavailable and where specific clinical data is often limited at best may have resulted in some true cases of COVID-19 related chilblains being unavailable for analysis.

Conclusion
Many patients reported as pseudo-chilblains of COVID-19 do not have confirmed infection.In adult patients with confirmed COVID-19, chilblain-like lesions have been reported primarily from North America and Europe, occur across the spectrum of age in males and females, favor acral surfaces, may be symptomatic or asymptomatic, lack relationship to cold/damp exposure, display variability in resolution time and association with extracutaneous COVID-19 manifestations, occurs in both well and ill patients and may serve as a trigger for COVID-19 testing.Histopathologic features resemble that of classical chilblains but less common patterns may occur.Further work is needed to clarify the relationship of acral eruptions and COVID-19.Infection confirmation, photographic documentation and histopathology are critical to establish homogeneity in reported pseudo-chilblains during this global pandemic.

Pseudo-Chilblains in Adult Patients with Confirmed COVID-19: A Systematic Review
Many organs can be affected by infection with COVID-19.The skin is no different.One of the earliest skin signs of COVID-19 infection was labeled "COVID-toes", where patients get red-to-purple spots/rashes, primarily on their toes or fingers.In the dermatology world, the preferred name for ''COVID-toes'' is 'pseudo-chilblains' referencing the similarity in appearance of the rash to a condition called chilblains affecting fingers and toes of people who have been exposed to cold and wet conditions for a relatively prolonged time.While little doubt exists that this peculiar rash may be a manifestation of infection with COVID-19, we were struck by the fact that many of the reported cases did not have confirmed infection.In the future, as we look back at the science and data generated during this period, the lack of laboratory confirmation of infection may render some of the conclusions drawn invalid, or at least uncertain.We wished to examine the clinical and laboratory characteristics of adult patients with COVID-toes (pseudo-chilblains) with confirmed infection.
To do this, we performed a systematic review of the published literature on the PubMed/Medline database following the standard guidelines for this type of research (Preferred Reporting Items for Systematic Reviews and Meta-Analyses, PRISMA).We used studies reporting adults (>18 years) with confirmed COVID-19.We recorded the type of study performed, which country the patients came from, age, sex and race of the patients reported, how close the onset of COVID-toes was to the diagnosis of COVID-19 infection, the type of testing used to confirm infection, whether the patient was kept in hospital or not, where on the body the rash occurred, whether the patient had a history of being exposed to cold or wet conditions, whether the rash had any symptoms, whether the patients had any non-skin manifestations of COVID-19 infection, how long the rash took to go away and what treatment if any was prescribed to patients with COVID-toes.We also documented if small pieces of skin were taken (biopsies) to describe what the rash looks like microscopically.Our search identified only 13 studies giving us details on 29 patients.In COVID-19-infected adults, ''COVID toes'' were most commonly reported from North America and Europe, occurred in both males and females over a wide age-range.Both well people and ill patients who were admitted to hospital could be affected.The hands and feet were most commonly affected but lesions on the ear could also be seen.''COVID-toes'' could be symptomatic or not.Many patients had evidence of COVID-19 infection besides rash (e.g.cough or diarrhea).''COVID-toes'' could take <1 week or up to greater than 50 days to resolve.No standard treatment for the rash was found.Biopsies are infrequently performed but when done, findings similar to classical chilblains are described.
In summary, many patients reported as pseudo-chilblains of COVID-19 do not have confirmed infection.Infection confirmation, photographs and biopsies are recommended if we are to be sure that every person reported as "COVID-toes" has the same rash.Further work clarifying the relationship of rashes on the hands and feet with COVID-19 infection is necessary.

Figure 1 .
Figure 1.Study Identification PRISMA Flow Chart; Template Adapted from Page et al. 8

Asymptomatic ( 1
Legend: **An individual case may have more than one anatomic location involved; IF, Direct immunofluorescence; IHC, Immunohistochemistry; IF, immunofluorescence; LMW, low molecular weight; NR, Not reported; NSAID, Non-steroidal anti-inflammatory drug. , Case Report; CS, Case series; Y, Yes; N, No; N/A, Not applicable; U, Unclear; Dash (-), no response necessary based on study type; Q1 (CR), Were patient's demographic characteristics clearly described?;Q1 (CS) Were there clear criteria for inclusion in the case series?; Q2 (CR) Was the patient's history clearly described and presented as a timeline?;Q2 (CS) Was the condition measured in a standard, reliable way for all participants included in the case series?; Q3 (CR) Was the current clinical condition of the patient on presentation clearly described?;Q3 (CS) Were valid methods used for identification of the condition for all participants included in the case series?Q4 (CR) Were diagnostic tests or assessment methods and the results clearly described?;Q4 (CS) Did the case series have consecutive inclusion of participants?; Q5 (CR) Was the intervention(s) or treatment procedure(s) clearly described?;Q5 (CS) Did the case series have complete inclusion of participants?; Q6 (CR) Was the post-intervention clinical condition clearly described?Q6 (CS) Was there clear reporting of the demographics of the participants in the study?Q7 (CR) Were adverse events (harms) or unanticipated events identified and described?Q7 (CS) Was there clear reporting of clinical information of the participants?Q8 CR Does the case report provide takeaway lessons?Q8 (CS) Were the outcomes or follow up results of cases clearly reported?Q9 (CS only) Was there clear reporting of the presenting site(s)/clinic(s) demographic information?Q10 (CS only) Was statistical analysis appropriate?

Table 1 .
Clinical/Laboratory Characteristics of Chilblain-like Lesions in Adults with Confirmed COVID-19 Infection (Part A).
Legend: *Where specific ages not available, age-range of cohort reported; IHC, Immunohistochemistry; NR, Not reported; RT-PCR, reverse transcriptase polymerase change reaction.

Table 2 .
Clinical/Laboratory Characteristics of Chilblain-like Lesions in Adults with Confirmed COVID-19 Infection (Part B).

Table 3 .
Results of Joanna Briggs Institute Critical Appraisal Checklistsfor Case reports and Case Series.

Table 4 .
Results of Joanna Briggs Institute Critical Appraisal Checklistsfor Cross-Sectional and Cohort studies.