We described dermatological reactions after vaccination with three SARS-CoV-2 vaccines (two mRNA and one adenovirus-vectored vaccines) and classified them into six well-defined morphological reactions patterns and new-onset or reactivation of dermatosis.
Initial reports mostly described local injection site reactions and, subsequently, other miscellaneous skin reactions after mRNA vaccination.[4,5,7,13,15,18–27] In 2021, McMahon et al. published a large registry-based study (mostly with the mRNA-1273 vaccine) in healthcare workers and older people, describing not only injection site reactions, but also urticarial and morbilliform rashes.
Unlike the study of McMahon et al., our data entry, description and assignment of clinical patterns were made by dermatologists and were mostly supported by photographs. Case collection throughout Spain and the 3-month recruitment period permitted a more representative sample beyond healthcare workers and older people.
Reactions were more frequent in women (80·2%), which may reflect a real difference or reporting bias, although women are known to have greater reactogenicity to vaccines, and 60% of vaccinated people in Spain were women at the time of the study. Therefore, women's immune systems may be more reactive to SARS-CoV-2 proteins, which would result in lower susceptibility to the disease and greater reactogenicity to vaccines.
Few people had previous atopic dermatitis (6·9%) or urticaria (6·4%). In the general population, the prevalence of atopic dermatitis is around 10%,[30,31] and the lifetime prevalence of acute urticaria is approximately 20%, so it cannot be concluded that previous atopy or acute urticaria predisposes to SARS-CoV-2 vaccine cutaneous reactions. However, 18·6% of patients with acute urticarial reactions to vaccines in our study had a history of urticaria. Case–control studies are needed to clarify this association. Only 7·7% of people were receiving new drugs (mainly acetaminophen) at the time of the reaction, and this factor was therefore unlikely to be related to cutaneous reactions.
There was a previous diagnosis of SARS-CoV-2 in 11·1% of cases, similar to the seroprevalence in Spain at the time of writing (9·9%). The severity of cutaneous reactions in this group did not differ from the rest of the sample. Thus, prior SARS-CoV-2 infection does not seem to predispose to cutaneous reactions or more severe reactions, after vaccination.
The COVID arm, the most reported pattern was described after vaccination with all three vaccines, particularly mRNA-1273 (70·0%), and almost exclusively in women (95·4%). This pattern had the closest association with systemic symptoms (64·6%).
Two-thirds of reported reactions were beyond the injection site. Each morphological pattern seems to correspond to a different spectrum of delayed hypersensitivity reaction, with most of the few skin biopsies that were performed showing nonspecific changes consistent with this reaction. In contrast to previous series, some reactions were scarce (chilblain-like/pernio) or unrepresented (erythromelalgia), while other reactions were more frequently reported (pityriasis rosea-like, VZV reactivations and papulovesicular rashes). The morbilliform and purpuric patterns were reported mostly after BNT162b2 and AZD1222 vaccination, and were associated with more severe reactions. VZV reactivation was more frequent after BNT162b2 vaccination and in men.
UK spontaneous AE reports are the main information source on AZD1222 vaccine cutaneous reactions, which, in our series, were mainly acute urticaria (21·1%), injection site reactions (16·8%) and morbilliform rash (11·6%). Owing to the precautionary suspension of the vaccine in the initial target population, we could not study the second dose.
We found a large number of herpes reactivations (VZV and HSV, 13·8%). For VZV, the number [n = 41 (10·1%)], severity (36·6% took sick leave) and the percentage in healthy people aged < 50 years (29·2%) were particularly striking.[35,36] There were fewer HSV than VZV reactivations, probably because patients with HSV do not usually seek medical care. We also found pityriasis rosea-like eruptions, which might have been due to human herpesvirus 6 and 7 reactivation. These herpetic reactivations were also described after SARS-CoV-2 infection and other vaccinations.[8,9,37,38] Taken together, these data strengthen a causal link between herpesvirus reactivation and the SARS-CoV-2 vaccine. A plausible mechanism is that a strong specific immune response against SARS-CoV-2 or the S protein from vaccines may distract the cell-mediated control of another, latent virus.
New-onset or worsening of inflammatory conditions were also reported, including psoriasis, lichen planus and bullous pemphigoid. These conditions were previously described after SARS-CoV-2 and other vaccinations.[20,28,39–43] As previously stated, vaccines may exacerbate skin manifestations in patients with immune-mediated skin diseases, but further investigation is necessary.
The patterns found in this and previous studies are heterogeneous and similar to those described in association with SARS-CoV-2 infection.[8,9,28] One case repeated the same papulovesicular rash after SARS-CoV-2 infection and vaccination. Therefore, the host immune response to the infection, and not direct viral damage, may cause these skin manifestations. However, a delayed hypersensitivity reaction against vaccine excipients cannot be ruled out.
Although most reactions were classified as mild/moderate, 21% were considered severe/very severe. This degree of severity was not reported in the study by McMahon et al. This percentage is most likely over-represented (reporting bias) but should not be ignored, as some reactions may be life threatening.
The study has some limitations. Firstly, the design did not permit causal associations or the measurement of risks or incidence. We could not compare the incidence or severity of cutaneous reactions by vaccine type, as vaccine distribution depended on availability during the study period. Secondly, the data collection period was short, which might limit study of the comprehensive data and evolution, especially after the second doses and AZD1222 vaccination. Thirdly, only 12·3% of cases were biopsied and histopathology might have prevented misclassification. Fourthly, there was a possible reporting bias towards previously reported or more serious reactions. Fifthly, SARS-CoV-2 infection after vaccination cannot be excluded as a plausible cause of cutaneous reactions. Finally, the lack of ethnic diversity in our sample does not permit generalization of the results.
In conclusion, we have described and classified cutaneous reactions reported after SARS-CoV-2 vaccination in a large Spanish case series. Most reactions were mild/moderate and self-limiting, but some were severe/very severe and required treatment. Better knowledge of these reactions may aid physicians during mass vaccination and reassure patients seeking advice.
Data availability statement
The data that support the findings of this study are available from the corresponding author upon request. The data are not publicly available for privacy/ethical reasons.
We thank the following Spanish dermatologists for their generous contributions: I. García-Doval, J. Sola-Ortigosa, S. Mateo Suárez, M. Elosua González, D. Ruiz Sanchez, B. Roque Quintana, M. Sidro Sarto, C.F. Vásquez Chinchay, J.F. Millán Calletano, A. Rodríguez-Villa Lario, E. Amores Martin, A. Suarez Valle, J. Fernández Vazquez, J. Riera Monroig, L. Armillas-Lliteras, M. Corral Forteza, J. Arandes Marcocci, J. Jimenez Cauhe, G. Selda Enriquez, L. Haya Martinez, A. Frías Sanchez, A. Pérez Pérez, M. Salleras Redonnet, A. Tuneu Valls, I. Fuertes De la Vega, M. Alegre Fernández, E. Baselga Torres, I. Escandel Gonzalez, A. Fernández Orland, A. Lopez Valle, V. García-Patos Briones, M.C. Juarez Dobjanschi, G. Aparicio Español, C. Ferrándiz Pulido, A. Castany Pich, V. Cabezas Calderón and C. Plà Ferrer. We also thank primary care physicians from the Granollers-Valles Oriental area, Spain (J. Alonso, M. Vilageliu, A. Estefanell, M. Manzano, M Martín-Gil), Dr M. Ribell (Internal Medicine Department, Hospital General de Granollers), and the Pharmacovigilance and Occupational Health departments and technical committee of the Hospital Clinic of Barcelona, Maria Angeles Diaz Sotero (Hospital de Móstoles) and Joana Guerrero (Hospital General de Granollers), Spain. This study formed part of the PhD degree of Alba Català, Department of Medicine, Dermatology, Universidad Autónoma de Barcelona, Spain.
The British Journal of Dermatology. 2022;186(1):142-152. © 2022 Blackwell Publishing