The numbers 1 through 11 indicate month by number.
eFigure. Scatterplot of Chilblains Incidence (Lagged By 1 Month) In Relation To COVID-19 Incidence by Location-Month
eTable. Morphologic Features and Symptoms of Chilblains Cases, Kaiser Permanente Northern California
Customize your JAMA Network experience by selecting one or more topics from the list below.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
McCleskey PE, Zimmerman B, Lieberman A, et al. Epidemiologic Analysis of Chilblains Cohorts Before and During the COVID-19 Pandemic. JAMA Dermatol. 2021;157(8):947–953. doi:10.1001/jamadermatol.2021.2120
Was the incidence of chilblains correlated with the incidence of COVID-19 in the same location?
In this cohort study of 1319 patients in northern California diagnosed with chilblains before and during the COVID-19 pandemic, the incidence of chilblains increased during the pandemic. However, the incidence of chilblains was only weakly correlated with the incidence of COVID-19 across 207 location-months, representing 23 geographic locations across 9 months; only 2% of chilblains cases were confirmed as potentially secondary to COVID-19.
The findings of this study suggest that the weak correlation between incidence of chilblains and incidence of COVID-19 infection may represent changes in pandemic behavior that led to increased diagnosis of chilblains.
Beginning in March 2020, case reports and case series linked the COVID-19 pandemic with an increased occurrence of chilblains, but this association has not been evaluated in an epidemiologic study.
To assess whether a correlation exists between COVID-19 incidence and chilblains incidence.
Design, Setting, and Participants
A retrospective cohort study was conducted within the Kaiser Permanente Northern California system from January 1, 2016, to December 31, 2020; health plan members of all ages were included.
COVID-19 incidence in 207 location-months, representing 23 geographic locations in northern California across 9 months.
Main Outcome and Measures
Chilblains incidence was the main outcome. The association of chilblains incidence with COVID-19 incidence across the 207 location-months was measured using the Spearman rank correlation coefficient.
Of 780 patients with chilblains reported during the pandemic, 464 were female (59.5%); mean (SD) age was 36.8 (21.8) years. COVID-19 incidence was correlated with chilblains incidence at 207 location-months (Spearman coefficient 0.18; P = .01). However, only 17 of 456 (3.7%) patients with chilblains tested during the pandemic were positive for SARS-CoV-2, and only 9 of 456 (2.0%) were positive for SARS-CoV-2 within 6 weeks of the chilblains diagnosis. Test results of 1 of 97 (1.0%) patients were positive for SARS-CoV-2 IgG antibodies. Latinx patients were disproportionately affected by COVID-19 but not by chilblains.
Conclusions and Relevance
This cohort study found that in northern California, the incidence of chilblains increased during the pandemic but was correlated weakly with the incidence of COVID-19 across 207 location-months. These findings may have resulted from a causal role of COVID-19, increased care-seeking by patients with chilblains during the pandemic, or changes in behavior during shelter in place.
In early 2020, dermatologists in Europe noticed a sharp increase in chilblains, an inflammatory dermatosis that generally affects the dorsal feet or hands during periods of damp and cold, but not freezing, conditions, associated with the COVID-19 pandemic.1-8 Few patients with chilblains tested positive for SARS-CoV-2 or its antibodies, although access to testing was limited.1-4 Subsequent reports of series with complete viral and antibody testing indicated low frequencies of tests with positive results.5-8 The underlying causes of an increase in chilblains related to the pandemic are controversial.5,7-10 Published cases of presumed secondary chilblains in the setting of confirmed COVID-19 infection indicate that COVID-19–associated symptoms preceded chilblains by up to 30 days.2,11
On noticing an increase in chilblains after the shelter-at-home order was initiated in 6 Bay Area counties in California on March 16, 2020 (and statewide on March 18, 2020), we conducted a retrospective cohort study to test the hypothesis that increased cases of chilblains appeared in the same locations as increased cases of COVID-19 over time.
Kaiser Permanente Northern California provides comprehensive, integrated care to 4.4 million people, with all clinical information entered into an electronic medical record. The health care system operates 23 medical centers that are located to serve defined urban, suburban, and exurban population centers. Northern California is known for its microclimates. The geographic features that define these microclimates also define the population centers.
This was a retrospective cohort study of chilblains incidence before and during the COVID-19 pandemic. Age-, sex-, and race/ethnicity-specific incidence rates of pandemic chilblains cases from April 1 to December 31, 2020, and prepandemic chilblains cases limited to April 1 to December 31, 2016-2019, were compared. We also report age-, sex-, and race/ethnicity-specific incidence rates of COVID-19 cases diagnosed from March 1 to November 30, 2020. Human participants research approval was provided by the Kaiser Foundation Research Institute Institutional Review Board; the need for informed consent was waived owing to the retrospective design of the study. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Because reports of chilblains linked to COVID-19 suggest that exposure to COVID-19 infection preceded chilblains diagnosis by up to 30 days,2,11 we compared the number of cases of COVID-19 in a particular month and chilblains the following month for 23 geographic locations. Our analysis used temporal and geographic variation in COVID-19 incidence by location-month as the unit of observation. This method controls for variations in weather because the rates of both diagnoses are being compared in the same geographic location. The analysis used COVID-19 cases diagnosed from March 1 to November 30, 2020, and chilblains cases diagnosed from April 1 to December 31, 2020, for a total of 207 location-months (23 locations across 9 months). We also analyzed the data without the 30-day lag, using COVID-19 and chilblains cases diagnosed the same month from March 1 to November 30, 2020, resulting in 230 location-months (23 locations across 10 months). For a validation substudy, a random sample of 63 patients diagnosed with chilblains before the pandemic who had a photograph of their lesion in the medical record was selected for comparison with 297 chilblains cases diagnosed during March through August of the pandemic. The cohort study included health plan members of all ages seen with chilblains or COVID-19 during a telehealth visit or an ambulatory, emergency department, or inpatient visit by any physician specialty.
COVID-19 was identified from laboratory results for nasopharyngeal testing using polymerase chain reaction (PCR), and chilblains was identified from International Statistical Classification of Diseases, 10th Revision (ICD-10) diagnostic code T69.1XXA recorded during a virtual, clinic, emergency department, or inpatient visit. Race/ethnicity was based on self-reported information when available (>95% of patients). As reported herein, African American includes patients who self-reported as Black, Asian American includes patients who self-reported as Asian or Pacific Islander, and Latinx includes patients who self-reported as Hispanic. Among the chilblains cohort, we identified medical conditions possibly associated with chilblains, using ICD-10 codes, including Raynaud disease12,13 (ICD-10 code I73.0), rheumatologic conditions12-15 (systemic lupus erythematosus, code M32.9), antiphospholipid syndrome (code D68.61), rheumatoid arthritis (code M06.9), blood dyscrasias12,13 (myelodysplastic syndrome, code D46.9; aplastic anemia, code D61.9; and lymphomas and leukemias, codes C81-C96.999), and hyperhidrosis16 (codes L74 and R61).
By early April, dermatologists started sharing chilblains cases via a Health Insurance Portability and Accountability Act–compliant tracking system with the goal of implementing a standard clinical workflow. We used regional-level documentation tools (SmartPhrases; Epic) to provide patient education resources. In addition, a set of clinically relevant questions was formulated for Kaiser Permanente members with chilblains to assess COVID-19 symptoms, the timing of rash, and behavioral factors. For the validation substudy, 297 patients identified early in the pandemic through this tracking system were selected as a convenience sample. Eight board-certified dermatologists (P.E.M., B.Z., A.L., C.C., F.G., C.C.J., D.S.L., and P.M.) performed medical records reviews to confirm the diagnosis of chilblains and discern chilblains symptoms, anatomic sites involved, laterality, and morphologic factors. We compared these patients with chilblains during the pandemic with a random sample of 63 patients diagnosed with chilblains before the pandemic who had a photograph available in the medical record.
Comparison of characteristics between chilblains cohorts before the pandemic and during the pandemic were made using χ2 tests. The age-, sex-, and race/ethnicity-specific incidence rates of chilblains were estimated using as denominators the health system’s enrollment on January 1, 2018, for the prepandemic period and June 1, 2020, for the pandemic period. The 95% CI for the incidence rate was calculated using the Poisson method.
To measure the correlation of COVID-19 with chilblains incidence across location-months, we used the Spearman rank correlation coefficient. Counts were used instead of incidence rates because COVID-19 and chilblains were paired by location-month such that the denominators were the same for the 2 diseases, as were the underlying demographic characteristics. The nonparametric Spearman correlation coefficient was used because some location-months had no cases of COVID-19 or chilblains.
The Figure shows the trends in monthly chilblains cases from January 1, 2016, to December 31, 2020. We identified 780 chilblains cases from April to December 2020, and 539 chilblains cases were identified during the same months from 2016 to 2019 (Table 1). The mean (SD) age of the patients with chilblains was 44.7 (24.3) years before the pandemic and 36.8 (21.8) years during the pandemic (P < .001). The proportion of males with chilblains increased during the pandemic (169 [31.4%] before vs 316 [40.5%] during; P < .001). The chilblains cohort did not vary by ethnicity or body mass index before vs during the pandemic. Most chilblains diagnoses were made by primary care physicians. During the pandemic, a higher percentage of diagnoses was made using telemedicine (before, 112 [20.8%] vs 713 [91.4%]; P < .001). The proportion of patients with chilblains who had a history of Raynaud syndrome and rheumatologic conditions decreased during the pandemic (Raynaud syndrome: from 30 [5.6%] to 9 [1.2%]; P < .001 and rheumatologic conditions: from 17 [3.2%] to 7 [0.9%]; P = .003).
The overall annual incidence rate of chilblains was 11.5 per 100 000 person-years (95% CI, 10.1-13.1) from January 2016 to February 2020 (n = 1986 patients). From April to December 2016-2020, the focus of the study, the annual incidence rate of chilblains per 100 000 person-years was 5.2 (95% CI, 4.8-5.6) before the pandemic in 2016-2019 and 28.6 (95% CI, 26.8-30.4) during the pandemic in 2020 (Table 2). School-aged children (age, 13-19 years) represented the cohort with the highest incidence of chilblains during the pandemic, with more than twice that of other age groups. In contrast, working-aged adults (age, 20-59 years) represented the cohort with the highest incidence of COVID-19 during the pandemic. Females had a higher incidence of chilblains than males before (7.0; 95% CI, 6.3-7.6 vs 3.3; 95% CI, 2.9-3.8) and during (33.1; 95% CI, 30.5-36.0 vs 23.7; 95% CI, 21.5-26.3) the pandemic. The incidence of COVID-19 was only slightly higher among females (29.2; 95% CI, 29.0-29.5) than males (28.0; 95% CI, 27.7-28.2). The highest incidence of chilblains during the pandemic was seen in Asian American (42.5; 95% CI, 37.7-47.8) and White (35.7; 95% CI, 32.6-39.1) patients—approximately 3 times higher that of African American (11.6; 95% CI, 7.8-17.3) and Latinx (12.5; 95% CI, 10.1-15.4) patients. In contrast, the highest incidence of COVID-19 was seen in Latinx patients (62.5; 95% CI, 61.9-63.1)—3 times that of Asian American patients (19.0; 95% CI, 18.6-19.3) and White patients (17.9; 17.7-18.2) and double that of African American patients (29.2; 95% CI, 28.4-29.9).
The Spearman rank correlation coefficient for the association of COVID-19 incidence with chilblains incidence 30 days later among 207 location-months was 0.18 (P = .01) (eFigure in the Supplement). Analysis without the 30-day lag resulted in a correlation of 0.34 (P < .001).
During the pandemic, 456 of the 780 patients (58.5%) with chilblains underwent PCR testing for SARS-CoV-2; of these, 17 (3.7%) were positive for SARS-CoV-2, but only 9 (2.0%) showed positive results within 6 weeks of their chilblains diagnosis. Reliable testing for SARS-CoV-2 IgG antibodies became available in late September 2020 and was performed in 97 chilblains patients, of whom 1 (1.0%) was positive.
The validation substudy included 360 patients with photographs taken at the time of chilblains diagnosis. Three of 63 prepandemic cases were excluded because of a likely alternative diagnosis and 7 because the photograph was unrelated to chilblains. During the pandemic, 22 of 297 chilblains cases (7.4%) were excluded because of a likely alternative diagnosis and 7 (2.4%) because the photograph was unrelated to chilblains. Morphologic features and additional characteristics of cases diagnosed during the pandemic are summarized in the eTable in the Supplement. Compared with chilblains cases diagnosed before the pandemic, those diagnosed during the pandemic were more likely to involve only the feet (31 of 53 [58.5%] before the pandemic vs 219 of 268 [81.7%] during the pandemic; P < .001). Only 6% of the patients reported wearing shoes at home.
In a northern California population, we calculated the incidence of chilblains during the same time frame before and during the COVID-19 pandemic. The incidence of chilblains increased substantially during the pandemic, but increases were not uniformly in the same locations where COVID-19 cases were being identified. We observed a weak correlation of COVID-19 incidence with chilblains incidence across 207 location-months (Spearman coefficient 0.18; P = .01). If chilblains were occurring in the same communities where COVID-19 cases were occurring, the Spearman coefficient would be closer to 1. In addition, although the incidence rate of chilblains increased, few patients with chilblains in our study were confirmed to have COVID-19, similar to the findings of Le Cleach et al.5
Our findings may reflect an etiologic relationship between COVID-19 and chilblains. It is also possible that increased care-seeking following “COVID toes” publicity in April 2020 explains the increase in reports of chilblains. A third possibility is that the increase in chilblains may have resulted from changes in behavior and environment associated with sheltering in place. Our data support this last explanation, as 94% of the patients reported going without shoes at home, and the feet were more involved during pandemic chilblains than in chilblains before the pandemic. The highest incidence of chilblains occurred among patients aged 13 to 19 years who were at home instead of in school. The highest incidence of chilblains with regard to race/ethnicity was found among Asian American and White patients. Cowan and Bellisario17 reported that Asian American and White workers in northern California were more likely to be able to work from home during the pandemic (Asian American: 52%; White: 50%) than African American (33%) or Latinx (30%) individuals. Latinx patients, who compose 21% of our health system members and disproportionately worked outside the home during the pandemic, represented 46% of COVID-19 cases but only 9% of pandemic chilblains cases.
Lester et al18 reported that published images of cutaneous manifestations of COVID-19 infection, including chilblains, have mostly been in lighter-skinned patients, indicating that there may be detection bias and therefore underdiagnosis of chilblains in individuals with darker skin. The highest incidence of pandemic chilblains in our system was among patients who self-identified as of Asian or Asian American race, but we did not separately assess skin color in our study.
Most chilblains is idiopathic. Chilblains has been associated in past studies with Raynaud disease, cryofibrinogenemia, blood dyscrasias, and connective tissue diseases, such as systemic lupus erythematosus, rheumatoid arthritis, and antiphospholipid syndrome.12-15 Chilblains in the setting of an associated autoimmune disorder may be described as secondary chilblains. Until 2020, no infectious disease was credibly reported to potentially cause secondary chilblains. The finding that some patients with COVID-19 developed chilblains at the same time or subsequent to the infection is suggestive of secondary chilblains due to COVID-19. Reports on the histopathologic changes seen in pandemic chilblains support this hypothesis but remain controversial.19,20 Only 9 patients in our study developed chilblains within 6 weeks of COVID-19 infection, and these patients could potentially be regarded as having secondary chilblains due to COVID-19. We found just as many patients who could be regarded as having secondary chilblains due to rheumatologic conditions (7) or blood dyscrasias (2) during the same period.
Most cases of chilblains in 2020 that were suspected to be secondary to COVID-19 were negative for PCR and serologic testing. It is possible that some patients with chilblains were exposed to SARS-CoV-2 but produced such a robust innate immune response that it was later difficult to find any evidence of infection.21 It has been hypothesized that an innate immune response to SARS-CoV-2, including interferon-α and other cytokines, contributes to chilblains symptoms, especially in younger patients who are otherwise asymptomatic.22,23 Perhaps improved testing methods in the future will help better detect past exposure to SARS-CoV-2, and this improvement would increase the number of chilblains cases regarded as secondary to COVID-19. Hubiche et al24 reported that, among 40 patients with chilblains whose PCR test results were negative for SARS-CoV-2, 8 had positive IgA serum antibodies to SARS-CoV-2, compared with only 5 with positive IgG serum antibodies. Similarly, El Hachem et al25 reported that, among 19 patients with chilblains whose PCR test results were negative for SARS-CoV-2, 6 had positive IgA serum antibodies to SARS-CoV-2, compared with only 1 with positive IgG serum antibodies. Despite this finding, in the absence of a confirmed infection with SARS-CoV-2 by any detection method or other associated condition, chilblains remains idiopathic until proven otherwise.
Strengths of this study include its size, community base, historical control group, validation by medical records review, and design using location-month as the unit of observation to control for geographic variation and differences in weather across locations. Limitations include lack of access to reliable antibody testing early in the study, no IgA antibody testing, and potential undercounting of asymptomatic cases of COVID-19. Also, histopathologic testing was not generally performed because chilblains is a self-limited condition. The diagnosis of chilblains was most often made by primary care clinicians, although medical records review by board-certified dermatologists indicated that only 7.4% of reviewed chilblains cases had possible alternative diagnoses.
Although this study found a weak geographic and temporal correlation of COVID-19 cases with chilblains cases, these findings may have resulted from behavioral changes. Very few cases of chilblains were positive for COVID-19.
Accepted for Publication: May 5, 2021.
Published Online: June 23, 2021. doi:10.1001/jamadermatol.2021.2120
Corresponding Author: Patrick E. McCleskey, MD, Department of Dermatology, Kaiser Permanente Oakland Medical Center, 3701 Broadway, 4th Floor, Oakland, CA 94611 (email@example.com).
Author Contributions: Dr Herrinton had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: McCleskey, Zimmerman, Lieberman, Liu, Chen, Gorouhi, Herz, Mirmirani, Herrinton.
Acquisition, analysis, or interpretation of data: McCleskey, Zimmerman, Lieberman, Liu, Chen, Gorouhi, Jacobson, Lee, Sriram, Thornton, Mirmirani, Herrinton.
Drafting of the manuscript: McCleskey, Lieberman, Liu, Lee, Herrinton.
Critical revision of the manuscript for important intellectual content: McCleskey, Zimmerman, Lieberman, Chen, Gorouhi, Jacobson, Sriram, Thornton, Herz, Mirmirani, Herrinton.
Statistical analysis: Liu, Herrinton.
Obtained funding: McCleskey.
Administrative, technical, or material support: McCleskey, Lieberman, Jacobson, Lee, Herrinton.
Supervision: McCleskey, Herz, Mirmirani, Herrinton.
Conflict of Interest Disclosures: None reported.
Funding/Support: Partial salary support was provided for Drs McCleskey, Liu, and Herrinton by The Permanente Medical Group Delivery Science and Applied Research initiative. This project was funded by The Permanente Medical Group Delivery Science and Applied Research initiative.
Role of the Funder/Sponsor: The funding organization had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.