Incidence and Characterization of Skin Lesions Among Hospitalized Patients with COVID-19 infection: An Observational Cohort Study from Iraq

Background and objectives: A wide arrays of skin lesions occurred in the setting of COVID-19 infection. We aimed to estimate the incidence and types of skin lesions among hospitalized patients with COVID-19 infection. Methods: We conducted an observational cohort study on 369 hospitalized patients with COVID-19 infection at Basra Teaching Hospital for 3 months. We studied the incidence and types of skin lesions. Results: Among 369 hospitalized patients, skin lesions were developed in 5%. Of those who developed skin lesions, 33% were asymptomatic. Skin lesions were more frequent in young females (83.3%). The most frequent type of skin lesion was maculopapular 44.3% which occurred exclusively in the medical ward and half of them in asymptomatic patients. For pustular lesions, all were female, diabetic, in the ICU, in symptomatic patients with the highest levels of serum C-reactive proteins. Pseudo-chilblain lesions were reported only in younger female and half of patients were asymptomatic. Urticarial lesions were reported in the younger patients, two-thirds were symptomatic and in the ICU. Purpuric lesions were reported in the older patient who developed acute kidney injury. Extremities were the most common distribution of skin lesions in 38.8%. The mean latency period for development of skin lesions was 8.6 ± 9.1, the shortest for urticarial lesions and the longest for purpuric lesions. Conclusions: Different types of skin lesions were reported in patients with COVID-19 and may be the sole manifestations of the disease.


Introduction
A wide range of skin lesions was reported in patients with COVID-19 infection (1)(2)(3). The frequency of skin lesions was reported from 0.2-20% from different studies mainly from Europe and China (4)(5)(6). The association of skin lesions with the severity of the illness was unclear (7). Maculopapular skin lesions involving the trunk were the most common cutaneous manifestations of COVID-19 infection (1,2,5,8,9). Pseudochilblain lesions (COVID toes) was reported in the absence of cold exposure or underlying diseases associated with chilblain lesions (1)(2)(3)(10)(11)(12)(13)(14)(15)(16)(17). The cutaneous manifestations may occurred before, at the time or after the onset of COVID-19 symptoms (5,10,11,(18)(19)(20). The objective of the study was to estimate the incidence of skin lesions among hospitalized patients with COVID-19 infection, to estimate the frequency of skin lesions' types and assess the risk factors for the development of these skin lesions. The purpose of the study was to estimate the incidence of AKI in hospitalized patients with COVID-19 infection in Basra, Iraq. Methods

Study design, setting and participants
This was a single center observational cohort study conducted on patients with PCR confirmed COVID-19 infection who were hospitalized at Basra Teaching Hospital from March 1, 2020, to June 1, 2020. The study was approved by the Institutional Review Board of the University of Basra and Basra Health Directorate. Data were obtained from medical records of patients hospitalized at both medical ward and intensive care unit. Direct examination was performed after wearing full PPE protections and for some cases, photos were sent for characterization of the skin lesions' types. Patients with recent skin lesions were included in this study. Any preexisting skin diseases were excluded from the study.
All patients with COVID-19 infection, whether symptomatic or asymptomatic, were hospitalized in the isolation wards according to the Basra Health Directorate guidelines. Asymptomatic patients were discovered by contact tracing.
Patients' demographics, medical history of chronic diseases, dermatologic diseases and bed location were obtained. Skin lesions' type, distribution, symptoms, latency period and duration were documented. Type of skin lesions were characterized into: Maculopapular, petechial, pseudo-chilblain, pustular, urticarial and purpuric lesions. Distribution of skin lesions were determined according to the affected sites into: Generalized, facial, truncal, inguinal, extremities, palmar and toes. Regarding symptoms of skin lesions, they were either asymptomatic or symptomatic (itching or pain). Acute kidney injury was defined as increased in serum creatinine by 0.3 mg/dL from the baseline within 48 hours or 1.5 times increased in serum creatinine from baseline within 7 days according to KDIGO guideline by the help of nephrologist (21). Inflammatory markers such as serum ferritin and C-reactive proteins were sent. Skin biopsy was not done due to COVID-19 transmission precaution.

Outcomes
The outcome of the study was to study the incidence and types of skin lesions in patients with COVID-19.

Statistics
Baseline demographics, clinical and laboratory characteristics of the overall patients and clinical characteristics of skin lesions were described using mean ± SD for continuous variables and numbers (percentages) for categorical variables. Comparative analyses between patients who developed skin lesions from those who didn't were performed using X 2 test or Fischer's Exact test for categorical variables and independent sample t-test for continuous variables. Demographics, clinical and laboratory characteristics per types of skin lesions were analyzed using X 2 test or Fischer's Exact test for categorical variables and one-way ANOVA for continuous variables. P values less than or equal to 0.05 were considered statistically significant. Statistical analyses were performed using SPSS version 25.

Results
From March 1, 2020, to June 1, 2020, 372 patients were admitted to Basra Teaching Hospital with a diagnosis of COVID-19 infection that was confirmed by polymerase chain reaction. Of these, 369 patients were included in this study after exclusion of 3 patients with preexisting skin diseases. Overall, 18 of 369 patients (5%) developed skin lesion during their hospitalization ( Figure 1). Five patients were in the ICU and 13 patients were in the medical ward.
The baseline clinical characteristics of the patients were provided in Table 1. The mean age was 51 ± 14 SD, 62.1% were female, mean BMI was 28.5 ± 4.2 SD, 40.1% were hypertensive and 27.9% were diabetic, 19.5% were admitted in the ICU, 5% developed skin lesions, 17.3% developed acute kidney injury and 6.5% died. Mean serum CRP and serum ferritin were 38 ± 67 and 325 ± 490 respectively.  Table 2 showed the comparison of patients who developed skin lesions from those who didn't. The characteristics of patients who developed skin lesions were as follow: The mean age was 40 ± 19.6 SD, 83.3% were female, mean BMI was 28.7 ± 3.9 SD, 27.8% were hypertensive, 38.9% were diabetic, 27.8% were in the ICU, 5.6% developed AKI and none was died. The levels of inflammatory markers were non statistically different between the two groups: For CRP, for those with skin lesions vs no skin lesions (49 ± 72.7 vs 37 ± 67; P = 0.470) and for serum ferritin, for those with skin lesions vs no skin lesions ( 343.5 ± 407 vs 324 ± 494; P = 0.870). Values were expressed as mean ± SDs and n (%).  Table 3 showed clinical characteristics of skin lesions. Thirty-three percent of patients who developed skin lesions were asymptomatic regarding COVID-19 presentations. The most common skin lesions were maculopapular in 44.3% with a mean duration was 5 ± 2.9 SD, 50% were itchy, 27.8% was painful and the most common site was the extremities in 38.8%.the mean time from symptoms in symptomatic patients to development of skin lesions was 8.6 days. Demographics, clinical and laboratory characteristics by types of skin lesions were provided in table 4. Pseudo-chilblain and urticarial lesions occurred in younger age group less than 30 years, all types of lesions were more frequent in female, BMI was comparable in all types of lesion, maculopapular lesions were more common in patients with hypertension and diabetes, all pustular lesions were in symptomatic patients, occurred in diabetic and in the ICU, 50% of maculopapular and pseudo-chilblain lesions occurred in asymptomatic patients, 87.5% of maculopapular lesions were itchy and 100% of pustular lesions were painful. Purpuric skin lesions showed longer duration and latency.  Values were expressed as mean ± SDs and n (%).

Discussions
In the present study, the incidence of skin lesions was 5% among hospitalized patients with COVID-19 infection. Italian prospective study by Recalcati reported 20.4% incidence of skin lesions. The lower incidence in our study compared to Recalcati study was because in our country, the guideline of ministry of health at the time of study was to hospitalize all patients and contacts with PCR positive test to reduce the transmission of the virus so inclusion of asymptomatic patients reduces the incidence of skin lesions in our study (22). In binational, multicenter and prospective study recruited 678 patients from China and Italy; incidence of skin lesions was 7.8%. The mean age was higher than our study (55.9 vs 40 years) and less female than our study (40% vs 83%  (10,18,(34)(35)(36)(37). The higher frequency in some of these reports may reflect media attention to pseudochilblain lesions and the health care provider suggested this type of skin lesions even without laboratory confirmation whereas in our study, all pseudo-chilblain lesions (11.1%) were in laboratory-confirmed cases. In these studies, after restriction to laboratory-confirmed cases, the most frequent skin lesions were maculopapular which was similar to our study findings. Furthermore, Spanish case series reported half of laboratory-confirmed lesions as maculopapular, 19% pseudo-chilblain and 19% urticarial (38). Pseudo-chilblain lesions developed in mild cases of COVID-19 in the aforementioned studies which was in agreement with our study where these lesions developed in mild cases and one of them was asymptomatic. Also, pseudo-chilblain lesions in our study were reported exclusively in younger than 40 years old patients and exclusively in female whereas the mean age in this study was 44.7 and 63% were female (38). In our study, all pustular lesions and three-quarter of urticarial lesions were associated with severe disease in the ICU whereas in this study, 7% of pseudo-chilblain, 12% of vesicular, 16% of urticarial, 17% of maculopapular and 41% of livedo/necrosis were reported in the ICU (38). Moreover, pustular lesions were common in older age group and patients with the highest inflammatory markers (CRP).
In the present study, 50% of skin lesion were itchy and 27.8% were painful.  Recalcati et al. (5,15,16,[18][19][20]. No death was reported for those who developed skin lesions even in the 5 patients who were admitted on the ICU which indicate good patients' outcome in those who developed skin lesions whereas 3 deaths were reported in Spanish study (1). The present study has some limitations. First, the study was performed at hospital, so we missed those who developed skin lesions in the community. Second, the pandemic is expanding, so we expect more cases if we extend the study time. Third, skin biopsy was not performed due to infection transmission precaution.
In conclusion, different skin lesions were reported among patients with Covid-19 and up to 33% in asymptomatic patients. Maculopapular was the most frequent type and with pseudochilblain, petechial and purpuric lesions were reported in mild cases while pustular and majority of urticarial lesions occurred in severe cases in the ICU. Nevertheless, the outcome was good.