The Characteristics of COVID-19 Symptoms and Skin Manifestations among Nonhospitalized COVID-19 Patients with Psoriasis during the Omicron Pandemic in China: A Single-Center Survey-Based Study

,


Introduction
Te severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has severely afected the whole world.During the coronavirus 2019 (COVID-19) community pandemic, the management of psoriasis, a chronic immune-related disease, particularly those patients who received biologics, has raised concerns among dermatologists [1].At the early stage, the key concern was whether patients who received biologics were more susceptible to COVID-19 [2].On the one hand, previous research indicated that psoriasis might have increased subclinical airway infammation [3]; on the other hand, the biologics-associated immunosuppressive efect and common upper respiratory tract infection were also worrisome [4].However, accumulating evidence has shown that immunosuppressive biologics did not increase the risk of COVID-19 infection and its related hospitalization [1,[5][6][7], and the guidance from the national psoriasis foundation (NPF) recommended that psoriasis patients without current COVID-19 infection should continue biological therapy [8].
In addition, the SARS-CoV-2 mutates in the direction of lower virulence and higher infectious and the COVID-19 infection is spreading wider [9].Tus, the primary concerns are the prevalence of COVID-19-related symptoms and skin manifestations in psoriasis patients.Previous studies have presented that more than one-ffth of psoriasis patients who received biologics sufered psoriasis worsening mainly because of the treatment discontinuation, while a few participants were confrmed to be afected with COVID-19 [10,11].Te shreds of evidence on the direct impact of COVID-19 infection on psoriasis patients who received biologics were limited.Since December 2022, the COVID-19 zero policy has been relaxed in China and then, the cases of Omicron variant (BA.5.2 and BF.7) infection grew dramatically rapid due to the robust transmission in a short period.
In this study, we aim to profle the characteristics of COVID-19-related symptoms and cutaneous signs among psoriasis patients who received diferent therapeutic regimes during the early high epidemic period of the Omicron variant.

Patients and Methods
We conducted this survey-based research from January 9 th to January 22 nd , 2023, during the frst phase of the Omicron pandemic.Participants were psoriasis patients in our cohort, named PSOWCH (psoriasis cohort of West China Hospital).Patients were informed of the purpose of this study and agreed to the patient's consent.Ten, the questionnaires were transferred to them through smartphones and completed with the doctors' guidance to help explain the questions but avoid infuencing their choices.Tis questionnaire contains demographic information, pertinent epidemiologic information about COVID-19 (patients and their cohabitants), comorbidities associated with the risk for relatively severe COVID-19 [12], COVID-19-related symptoms [13], and cutaneous signs (within two weeks after the COVID-19 infection).Te establishment of PSOWCH and the information permission were under the Helsinki Declaration of 1964 (revision in 2013) and approved by the Biomedical Research Ethics Committee of West China Hospital of Sichuan University (approval number: 2021-581).
In total, we collected 269 valid questionnaires from PSOWCH, and the patients were enrolled if they were (1) aged ≥ 18 years old and (2) with a confrmed diagnosis of psoriasis.Patients were excluded if they were asymptomatic without epidemiological history and did not complete the COVID-19 test.According to the COVID-19 case defnition published by the WHO (version 2022) [14], COVID-19 cases are classifed as confrmed, probable, and suspected SARS-CoV-2 cases based on the COVID-19 tests (nucleic acid amplifcation test or the SARS-CoV-2 antigen-rapid diagnostic test), clinical characteristics, and epidemiological criteria, and the specifc number of the three classes are shown in Figure 1.Eventually, we included 240 patients in this study and then, they were subdivided into three groups based on their therapeutic regimes as follows: biologics (BIO), containing adalimumab (ADA), secukinumab (SEC), ixekizumab (IXE), and ustekinumab (UST), nonbiological systematic treatment (NBT), and topical treatment or without any pharmacotherapy (TT).
Te statistical analysis was completed through GraphPad Prism 9.0 and SPSS Statistics version 26.0.Proper descriptive methods were used, as the categorical variables were presented as the count (percentage) and continuous variables were presented as the mean (standard deviation, SD).Te diferences in the variables among multiple subgroups were managed using Pearson's χ 2 test (or Fisher's exact test) or the Kruskal-Wallis test (or one-way ANOVA test).Te multivariable and bivariable logistic regression models were used to adjust for predefned confounders and further analyze whether the diference in therapeutic regimes is associated with the incidence of COVID-19 symptoms and cutaneous signs.In the multivariable logistic regression models, we could only include ten variables at most, as the number of participants in the smallest subgroup should be at least fve times greater than the number of included variables.Tus, we conducted the univariate analysis to screen out comorbidity variables with signifcant diferences and then included them in the models.A P value < 0.05 was considered signifcant.
Te mean age of patients in these three subgroups was 43.06 (SD: 12.51), 38.48 (SD: 11.81), and 37 (SD: 11.01) years old, with a signifcant diference (P � 0.012).Regarding the smoking status, there were 55 (39.86%), 17 (32.69%),and 19 (38.00%) smokers (current smokers and ex-smokers) in BIO, NBT, and TT groups, respectively.Te comorbidities were almost evenly distributed, while the prevalence of hypertension (22, 15.94%) was higher in the BIO group with a statistical signifcance (P � 0.003).In addition, 33 patients (23.91%) in the BIO group did not take the COVID-19 vaccine, which is signifcantly more than in the other two groups (P � 0.005).Other baseline and clinical characteristics showed no statistical diference among these three groups and are detailed in Table 1.
Concerning the COVID-19-related symptoms, 186 patients (77.50%, n = 240) presented with more than one sign, and among which 138 (74.19%, n = 186) patients reported that the COVID-19 symptom duration was less than a week.Fever (141, 75.81%, n = 186), cough (124, 66.67%, n = 186), and fatigue (103, 55.37%, n = 186) were the top three common ones, with the same ranking in the three subgroups.In addition, we focused on dyspnea, a debilitating symptom distinct from the signs and symptoms of mild COVID-19.Out of 24 patients (12.90%, n = 186) who 2 Dermatologic Terapy presented with dyspnea, 13 (11.50%,n = 113) of them were in the BIO group, 8 patients (20.51%, n = 39) were in the NBT group, and 3 patients (8.82%, n = 34) in the TT groups.Tere was no signifcant diference in the prevalence of each symptom in the BIO, NBT, and TT groups (Table 2).In order to attenuate the efect of confounders predefned, logistic regression models were built, focusing on the presence of symptoms of COVID-19.Te minimally adjusted model involved gender, age, and BMI as confounders.Te minimally adjusted OR of the group TT was 0.41 (95% CI: 0.19-0.89).Te fully adjusted model included gender, age, BMI, the comorbidity of hypertension, ever smoked, and vaccination status (received or not), and the adjusted OR of the group TT was 0.40 (95% CI: 0.18-0.90).Both models indicated that the group TT has a lower risk of presenting with COVID-19 symptoms (Table 3).Ten, further exploration in the existence of dyspnea was conducted.Given the sample size limitation, we proceeded with a series of bivariate logistic regression to adjust the confounders, which showed no more risk of presenting dyspnea in the groups BIO and NBT (Table 4).
Intriguingly, we also explored the relationship between skin manifestations of psoriasis and the Omicron variant.83 out of 240 (34.58%) patients had one or one more psoriatic sign in 2 weeks after infection, involving 60 patients (25.00%) who developed new lesions, 33 patients (13.75%) reported aggravating pruritus, and 29 patients (12.08%) had severe desquamation (Table 2).Multivariate logistic regression was performed to adjust the confounders, and we observed that patients with higher BMI (adjusted OR � 1.10, 95% CI � 1.00-1.20)and who received NBT (adjusted OR � 2.15, 95% CI � 1.08-4.27)had a relatively high risk of developing cutaneous signs after COVID-19 infection in the fully adjusted model (Table 5).In the supplementary fle 1, we profled the PASI changes of 33 patients (24 of them in the BIO, 7 patients in the NBT, and 2 patients in the TT group).

Discussion
Tis survey-based study profled the characteristics of COVID-19 symptoms and the skin manifestations among 240 nonhospitalized COVID-19 patients (including confrmed, probable, and suspected ones) with psoriasis in the Omicron outbreak in China.Moreover, our data showed that diferential psoriatic treatment might be a correlative factor in developing symptomatic or asymptomatic Omicron infection and presenting cutaneous signs after infection.
In this study, 74.19% of the COVID-19 symptomatic patients (n � 186) reported that the symptom duration was within a week, consistent with the mean duration of 6.87 days in Omicron dominance presented by a prospective observational study in the UK [15].Regarding the common  Dermatologic Terapy symptoms, we reported that the top three were fever, fatigue, and cough.However, it is diferent from previous knowledge about the distribution of Omicron-induced symptoms that the prevalence of fever was low among COVID-19 patients during the outbreak of Omicron in Hong Kong and the UK [15,16].Te specifc reasons for this diference need to be explored by further studies.
Our data also indicated that the diference in being symptomatic or asymptomatic COVID-19 patients was associated with psoriatic treatment while not with age, BMI, hypertension, or vaccination status.We observed a relatively lower risk of being symptomatic in the group of TT after controlling relevant confounders.In 2020, a monocenter study presented that 33 out of 180 psoriasis patients had at least one COVID-19 symptom and showed no signifcant diference in the distribution of signs and symptoms between psoriasis patients who received biologics (n � 16) and those treated with topical treatment (n � 17), while it did not mention about the asymptomatic rate and related factors [17].Our fndings provide evidence for factors infuencing the onset or absence of symptoms in patients with psoriasis after COVID-19 infection during the Omicron period, but the sample size was limited, and a sizeable real-world study is needed to explore this question further.
Regarding the efect of COVID-19 vaccination, previous research presented that patients who received two or three doses had a lower risk of hospitalization during the Omicron pandemic [15].Whereas, among nonhospitalized Omicron variant-infected patients, evidence suggested that the vaccination status was not associated with the likelihood of being symptomatic or asymptomatic of COVID-19 [16], which is paralleling our results.In this study, we did not present the impact of COVID-19 vaccination on psoriasis fares but we found that previous evidence elucidated this point controversially.In 2021, Sotiriou et al. reported 14 psoriasis exacerbation cases after receiving COVID-19 Continuous variables are performed as the mean (SD) and categorical variables, as count (%).Accurate statistics are chosen for the data, including one-way analysis of variance (ANOVA), Kruskal-Wallis test, and Pearson's χ 2 test (or Fisher's exact test).A P value <0.05 is considered signifcant and marked into bold.† According to the recommendation of Asians' BMI classifcation, overweight is defned as 23 ≤ BMI < 27.5 and general obesity is defned as 27.5 ≤ BMI < 32.5 and the patients were stratifed as BMI < 23 and BMI ≥ 23.‡ Te subgroup of ADA contains the original adalimumab and its biosimilars.Abbreviations: ADA: adalimumab, BMI: body mass index, BIO: biologics, CKD: chronic kidney disease, COPD: chronic obstructive pulmonary disease, CVD: cardiovascular disease, IXE: ixekizumab, M: male, NBT: nonbiological treatment.MTX: methotrexate, SEC: secukinumab, TCM: traditional Chinese medicine, TNF: tumor necrosis factor, TT: topical treatment or without pharmacological therapy, and UST: ustekinumab.
4 Dermatologic Terapy vaccination and three of them were under the treatment of biologics [18].Comparatively, some recent studies found that COVID-19 vaccination does not induce psoriasis or psoriatic arthritis fares among patients treated with biologics [19,20].Following these fndings, in 2023, a multicenter case-control study in Italy also showed no increased fare rate among patients with controlled psoriasis under biological treatment during 24-week observation [21].
Based upon this, we considered that the COVID-19 vaccination should be recommended to most psoriasis patients  according to the NPF COVID-19 task force guidance in this pandemic [8] and more evidence is required for exploring the impact of COVID-19 vaccination on psoriasis.Te COVID-19 infection's impact on the worsening efect of psoriasis remains a hot topic.Previous research suggested that no adherence to the biologics was related to the psoriasis fare-up [22].Although the infection of SARS-CoV-2 might play a crucial direct role in psoriasis aggravation, the pieces of literature about it were mainly presented as cases.We reported the proportion of the incidence of new lesions, pruritus aggravation, and increased desquamation in three treatment groups (BIO, NBT, and TT) within two weeks after the infection, which complemented to the direct impact of the Omicron variant on the exacerbation of psoriasis.Interestingly, high BMI and the NBT were considered as correlative factors to developing cutaneous signs after COVID-19 infection.
Tere are some limitations in this present study.First, the sample size is relatively small, which might lead to bias.Tat might also result in only one comorbidity variable, hypertension, being included in the multivariate logistic model due to its signifcant diference among the three groups.Second, we collected patients' self-report evaluation of the presence of the new lesions after Omicron variant infection, which might difer from the results of lesions evaluation made by dermatologists.In the future, large-scale studies are needed to explore the impact of diferential psoriatic treatment on the Omicron variant symptomatic infection and Omicron's direct impact on the condition of psoriasis.Furthermore, a long-term follow-up is required for the consistent efect of COVID-19 on psoriasis.

Figure 1 :
Figure 1: Te fowchart for enrolment in this study.

Table 1 :
Te baseline characteristics of the participants.
Continuous variables are performed as the mean (SD) and categorical variables, as count (%).Accurate statistics are chosen for the data, including one-way analysis of variance (ANOVA), Kruskal-Wallis test, and Pearson's χ 2 test (or Fisher's exact test).A P value <0.05 is considered signifcant and marked into bold.† Patients might have more than one symptom so that the sum of the percentages is over 1. Abbreviations: BIO: biologics, NBT: nonbiological treatment, and TT: topical treatment or without pharmacological therapy.

Table 3 :
Multivariate logistic regression models for developing symptomatic or asymptomatic COVID-19.P value <0.05 is considered signifcant and marked into bold.† Minimally adjusted OR contains age, gender, BMI, and treatment groups.‡ Fully adjusted OR contains age, gender, BMI, hypertension, smoking status, vaccination, and treatment groups.Abbreviations: BMI: body mass index, BIO: biologics, CI: confdence interval, NBT: nonbiological treatment, OR: odds ratio, and TT: topical treatment or without pharmacological therapy. A