COVID-19 related masks increase severity of both acne (maskne) and rosacea (mask rosacea): Multi-center, real-life, telemedical, and observational prospective study.

Masks are essential for COVID-19 prevention, but recently they were suggested to modify cutaneous facial microenvironment and trigger facial dermatoses. To evaluate mask-related rosacea and acne (maskne) in untreated patients during lockdown. In this multi-center, real-life, observational prospective study, we enrolled stable, untreated acne and rosacea patients that wore masks during lockdown at least 6 h/day. They underwent two teledermatological consultations, at the baseline and after 6 weeks. Clinical, pharmacological, and psychological data were recorded. A total 66 patients, 30 (median age: 34.0 [30.25-29.75] yoa) with acne and 36 patients (median age: 48 [43-54] years) with rosacea, were enrolled in this study. After 6 weeks of mask and quarantine, patients with acne displayed an increased Global Acne Grading Scale (GAGS) score in mask-related areas (P < .0001). Likewise, after 6 weeks of mask and quarantine, patients with rosacea displayed a worsen in both physican (P < .0001) and patient (P < .0001) reported outcomes. Remarkably, patients reported also a statistically significant decrease in their quality of life (P < .0001). Masks appear to trigger both acne and rosacea flares. Additional studies are needed to generate evidence and inform clinical decision-making.

INTRODUCTION

During COVID‐19 pandemics, the introduction of personal protective equipments (PPE), such as masks and gloves, drastically changed facial and hand dermatoses dermatoepidemiology. , Focusing on facial dermatoses, Hua et al evaluated the short‐term cutaneous impact of wearing surgical masks (SM) and PPE on the skin finding a relevant modification in the level of hydration, trans‐epidermal water loss, pH, erythema, and sebum production. Masks are usually coupled with dressing and moisturizers to prevent both dehydration and pressure ulcers, but topical agents may be comedogenic and detrimental for preexisting facial inflammatory dermatoses, especially for acne. In fact, maskne or mask‐related acne is a well‐recognized comorbidity due to PPE during COVID‐19 pandemic both in healthcare workers and less in the general population. At the same time, rosacea may be also triggered by the cutaneous modification due to masks, but no conclusive data available.

During the first Italian lockdown several patients refused to attend dermatological visits , and discontinued lifesavings therapies, offering an opportunity for the evaluation of mask‐triggered facial dermatoses with teledermatology.

MATERIALS AND METHODS

This is a prospective observational study focusing on patients with acne vulgaris and rosacea that involved three primary dermatology referral centers with a telemedicine service in Italy (IRCCS Istituto Ortopedico Galeazzi, IRCCS Istituto San Gallicano and Ospedale Maggiore della Carità) during the first Italian lockdown (11 March to 4 May 2020).

Inclusion criteria

We included adult (>18 years old) patients with a diagnosis of acne vulgaris or rosacea that before lockdown underwent topical therapies and during lockdown they were not able to attend dermatological in‐person visits, leading to discontinuation of their topical medications. They were stable (<20% variation in severity scores in the previous two consecutive visits ~2 months without a therapy) and accepted to have teledermatological visits every 4 weeks only for a clinical evaluation. We decided to include only stable patients not undergoing a treatment (topical or systemic ones) to avoid confounders due to therapy, such as drug, posology, administration method, and therapy duration. They had to wear masks for at least 6 h/day, an approved SM or FFP2 for Italian laws (http://www.governo.it/it/articolo/coronavirus‐conte‐firma‐il‐dpcm‐11‐marzo‐2020/14299).

Exclusion criteria

We excluded acne and rosacea adult patients that (a) smoke cigar, cigarettes and e‐cigarettes, (b) did not discontinue topical and/or systemic therapies, (c) have facial dermatoses other than acne and rosacea, (d) undergoing special diets (ie, vegans, vegetarians, Ramadan fasting , , and keto diet ) capable to influence inflammation, (e) infected by SARS‐CoV‐2, (f) with a diagnosis or even a positive medical history for alcoholism, (g) with comorbidities capable to trigger acne or rosacea flares, (h) wear community masks or approved masks for <6 h, and (i) history of patch test positivity for formaldeyde.

Teledermatological evaluation

Patients fulfilling the inclusion criteria were enrolled and underwent a teledermatological evaluation at the baseline (T0) and after 6 weeks of quarantine (T1) with a 15 min videocall on FaceTime or WhatsApp or Skype or Zoom. To increase comparability and ensure consistency we choose for each patient the same videocall program at T0.

To stage acne vulgaris patients we used Global Acne Grading Scale (GAGS) and Dermatology Life Quality Index (DLQI). In particular, we reported both total GAGS score and GAGS for the single location (forehead, left and right cheeks, nose, chin, chest, and upper back) to topographically evaluate mask‐covered vs non mask‐covered sites.

To stage rosacea we used Clinician's Erythema Assessment (CEA), Global Flushing Severity Score (GFSS), Investigator Global Assessment (IGA), Patient Self‐Assessment (PSA), and DLQI. Each patient signed an electronic informed consent form.

Statistical analysis

Normal distribution for each variable was investigated by performing the Kolmogorov‐Smirnov test. Based on the distribution, parametric (Student's t‐test) and nonparametric (Mann‐Whitney test or Wilcoxon test) tests for paired or independent samples were applied to analyze the scores' variation between the baseline (T0) and after 6 weeks of quarantine (T1) in the investigated population. Patients affected by rosacea were further subcategorized based on the type of disease (predominant phenotype or papulopustular [PP] or erythematotelangiectatic phonotype [EE]) and the statistical analysis repeated. Multivariate regression analysis was conducted by using an “Enter” approach to analyze the influence of selected variables on the scores' variation after 6 weeks of quarantine. Data were reported as mean ± SD as well as percentage and all the analyses were conducted by using the MedCalc Statistical Software version v19.0.3 (MedCalc Software bvba, Ostend, Belgium), with a P value considered significant when <.05.

RESULTS

Maskne or acne due to masks

A total of 30 patients (median: 34.0 [30.25‐29.75] yoa) with acne were enrolled in this study, specifically 12 (40%) males and 18 (60%) females. Of those, 6 (33.3%) were affected by poly‐cystic ovary syndrome. Patients reported to wear a mask for an median time of 8 [7‐8.75] h/day.

Acne severity was evaluated by GAGS scores for the six areas (forehead, each cheek, nose, chin and chest/upper back) and by DLQI, at the baseline (T0) and after 6 weeks of quarantine (T1). Results are reported in the Table 1. DLQI and all the GAGS scores mask‐related significantly worsen (P < .0001) after 6 weeks of quarantine, with the exception of GAGS score for forehead and chest/upper back, which did not show any variation. Age, gender, and number of hours/day wearing a mask did not significantly influence the global GAGS scores at T0 (P = .84) and T1 (P = .95), explaining only 3.0% and 1.3% of scores variability in our population, respectively (Table S1).

TABLE 1

GAGS scores and DLQI are reported at the baseline (T0) and after 6 weeks of quarantine (T1)

Variables	T0 (median [IQR])	T1 (median [IQR])
GAGS forehead	2 (2‐4)	2 (2‐4)
GAGS left cheek	2 (0‐3.5)	4 (4‐4)***
GAGS right cheek	2 (0‐3.5)	4 (4‐4)***
GAGS nose	1 (0‐2)	2 (2‐2)***
GAGS chin	1 (0‐1)	2 (1‐2)***
GAGS chest and upper back	3 (0‐3)	3 (0‐3)
GAGS cheeks, nose and chin	5 (4‐6)	12 (11‐13)***
Global GAGS	9.5 (7‐14)	17.5 (15‐21)***
DLQI	8 (6‐9)	11 (10‐13)***

Note: ***P value < .0001.

Abbreviations: DLQI, dermatology life quality index; GAGS, global acne grading scale; IQR, interquartile range.

Mask rosacea or rosacea due to masks

A group of 36 patients (median age: 48 [43‐54] yoa) were included in this study, with a higher prevalence of women (n = 22, 61.1%) than men (n = 14, 38.9%). Patients were mainly affected by papulopustular (n = 23, 63.9%), followed by erythematotelangiectatic (n = 13, 36.1%), and they reported to wear a mask on average for 8 (7‐9) h/day. As reported in Table 2, all the scores (CEA, DLQI, GFSS, IGA, and PSA) showed a significant increase at T1. The same results were obtained when patients were further subcategorized into two rosacea phenotypes, PP and EE. PP patients showed a higher median age (49 [45‐54] years) than EE patients (44 [37‐51] years; P = .039) while no difference was observed in the average time wearing a mask/day (8 [7.5‐9] vs 8 [7‐9] h/day, respectively; P = .94).

TABLE 2

The following scores were calculated to assess the rosacea severity at the baseline (T0) and 6 weeks of quarantine (T1)

	All population (N = 36), median (IQR)		Papulo‐pustular patients (N = 23), median (IQR)		Erythemato‐telangiectatic patients (N = 13), median (IQR)
Variables	T0	T1	T0	T1	T0	T1
CEA	2 (0–3)	3 (0–4)***	1 (0‐2)	2 (0–3)**	2 (2–3)	4 (3–4)**
DLQI	7 (5‐8)	10 (9‐11)***	7 (6‐9)	11 (9‐12)***	6 (4‐6)	9 (8–9)***
GFSS	2.5 (0‐4)	4.5 (0‐6)***	2 (0‐3)	3 (0‐5.5)**	4 (3–5)	6 (5‐7)***
IGA	2 (1‐2.25)	3 (2‐4)***	1 (0‐2)	3 (0–3.5)**	2 (2–3)	4 (3–4)**
PSA	2 (0.75‐3)	3 (2‐4)***	1 (0‐2.5)	2 (0–4)**	2 (2–3)	4 (3–4)**

Note: **P value <.01; ***P value <.0001. P values were calculated in comparison with the relative T0.

Abbreviations: CEA, clinician's erythema assessment; DLQI, dermatology life quality index; GFSS, global flushing severity score; IGA, investigator global assessment, IQR, interquartile range; PSA, patient self‐assessment.

Based on multivariate regression analyses, the variations observed in the scores for the global population analyzed occurred regardless the age, the hours wearing a mask/day and the gender (Table S2).

DISCUSSION

In our real‐world study, acne vulgaris and rosacea appeared to worsen in untreated patients wearing masks. Maskne is a recognized subtype of acne mechanica, and mask rosacea is a potential new entity triggered by mask‐wearing during the pandemic.

Recently, Teo suggested that maskne should be diagnosed in case of (a) de novo acne occurred 6 weeks after a regular mask wear or (b) worsen of pre‐existen acne in mask covered by areas after regular mask wearing, (c) after exclusion of the main differential diagnoses (ie, seborrheic dermatitis).

Remarkably, in this study authors used the same temporal criteria (6 weeks presence) to define also rosacea due to mask. In literature, a few case reports described worsening of rosacea after wearing masks.

Hua et al evaluated cutaneous short‐term changes in patients wearing SM and N95, finding that masks induce microenvironment changes in the skin by dehydration, increased sebum and increased PH. Furthermore, dehydration, transepidermal water loss, and sebum dysregulation are pro‐comedogenic factors, capable to promote Cutibacterium acnes multiplication, and hence innate immune response and leading to inflammatory lesions (papules and pustules). Likewise, also Demodex folliculorum, regarded as the a trigger in rosacea, takes advantage from sebum overproduction producing/amplyfing inflammation (papules, pustules, and erythema).

Laboratory results further support the clinical findings, since in our study severity scores and DLQI worsened in both acne and rosacea patients after 6 weeks of masks, in agreement with previous reports described a detrimental influence on quality of life in patients affected by these dermatoses. ,

Unfortunately, beside the suggestion to not discontinuing isotretinoin (https://www.aad.org/public/diseases/acne/derm‐treat/isotretinoin/coronavirus‐pandemic) in severe acne and to preserve teledermatology consultations every 4 weeks, no other recommendations available to guide dermatologists that treat acne and rosacea patients during pandemics. Thus, real‐world evidence can help inform clinical decision‐making for dermatologists treating acne and rosacea patients.

This study was performed during lockdown where acne and rosacea visits were re‐prioritized in Italian dermatologic referral centers, so the limitation of in‐person visits excluded the possibility to test facial microbiome changes.

Since mask wearing worldwide was maintained due to COVID‐19 pandemics, we expect an increase in the prevalence of maskne and mask rosacea and establishing recommendations/guidelines to support dermatologists and family doctors in the differential diagnosis and management is needed and should be based on big data. Further studies on mask‐related dermatoses are needed to better understand the pathophysiologic mechanism of facemask wearing in rosacea, in order to evaluate the best therapeutic approaches and counteract the pro‐inflammatory effect of masks.

CONFLICTS OF INTEREST

The authors declare no conflict of interest.

AUTHOR CONTRIBUTIONS

CRediT (Contributor Roles Taxonomy): Conceptualization: Giovanni Damiani and Laura C. Gironi; Methodology: Giovanni Damiani, Khalaf Kridin, Renata Finelli, and Alessandra Buja; Software: Renata Finelli and Nicola L. Bragazzi; Validation: Giovanni Damiani, Renata Finelli, Ayman Grada, Nicola L. Bragazzi, and Alessandra Buja; Formal analysis: Renata Finelli; Investigation: Giovanni Damiani, Laura C. Gironi, Paolo D. M. Pigatto, and Paola Savoia; Resources: Giovanni Damiani, Paolo D. M. Pigatto, and Paola Savoia; Data curation: Renata Finelli and Alessandra Buja; Writing—Original Draft: Giovanni Damiani, Ayman Grada, Laura C. Gironi, Khalaf Kridin, Renata Finelli and A.B; Writing—Review and Editing: Giovanni Damiani, Ayman Grada, Laura C. Gironi, Khalaf Kridin, Renata Finelli, Alessandra Buja, Nicola L. Bragazzi, Paolo D. M. Pigatto, and Paola Savoia; Visualization: Khalaf Kridin, Nicola L. Bragazzi, and Paola Savoia; Supervision: Paolo D. M. Pigatto and Paola Savoia; Project Administration: Giovanni Damiani, Laura C. Gironi, Paolo D. M. Pigatto, and Paola Savoia.

TABLE S1: Global GAGS scores at the baseline (T0) and after 6 weeks quarantine (T1).

Click here for additional data file.

TABLE S2: Multivariate regression analysis (Enter approach) to investigate the variables associated with the worsening of rosacea scores after 1 month of quarantine.

Click here for additional data file.