Potential Delay of Diagnosing Infraorbital Skin Tumors Due to Coverage by Face Masks During the COVID-19 Pandemic: An Observational Study

Introduction

Due to safety regulations regarding the COVID-19 (Corona Virus Disease 2019) pandemic, all citizens ≥6 years of age in Bavaria, Germany, were required by law to wear a face mask covering the mouth and nose in all public areas from 27/04/2020¹ to the beginning of 2022 and are still strongly advised to do so as basic measure of protection.² International medical societies recommend a minimum of face masks and face shields for all patients in order to slow the spread of COVID-19.³ Additionally, in Germany, contact with people outside the same household was temporarily restricted to one person and personal contact to residents in nursing homes by relatives was prohibited, therefore oftentimes widely limiting personal contact.

We could find no reports in common literature databases regarding the potential delay of tumor diagnosis due to coverage by face masks which is of importance for the Ophthalmologist regarding infraorbital lesions, but also for ENT (ear, nose and throat) or maxillofacial physicians.

Timely diagnosis of cutaneous tumors will facilitate surgical reconstruction, improve aesthetic results and even impact patient survival in cases of malignancies at risk of developing systemic disease such as eyelid squamous cell carcinoma.^4,5

During the pandemic, examinations were largely symptom-oriented, performed without accompanying people allowed and, for safety reasons, time and potential virus distribution without face masks was reduced to a minimum.

We hypothesized that covering parts of the infraorbital and nasal regions by masks might result in belated diagnosis or potential miss of tumorous lesions in these areas by the treating general Ophthalmologists especially regarding elderly patients and patients living alone. We therefore retrospectively evaluated all patients having presented at our large tertiary university center with infraorbital skin tumors from 04/2020 to 06/2021.

Materials and Methods

For inclusion in this retrospective, mono-centric, observational study, patients were identified by searching the electronic database of the oculoplastic department of the eye clinic of Ludwig-Maximilians-University, Munich/Germany, for the diagnosis “lower eyelid tumor” and “infraorbital tumor”. Patients were clinically diagnosed with infraorbital lesions between 04/2020 and 06/2021. Only patients over 18 years of age at presentation and with histopathology findings were included into this study. Clinical data was extracted from the original patient files.

Approval was obtained from the ethics committee of Ludwig-Maximilians-University, Munich/Germany (vote number 21–0870). The patients photographed in Figures 1 and 2 have seen the photos and have read the article as presented for publication. Individual written informed consent was obtained. The study is in accordance with the Declaration of Helsinki.

Figure 1 Outline demonstrating the average coverage of the mid-face due to masks on frontal portrait photography: Infraorbital line below the lower orbital fat pad over the lower orbital rims following a V-shape over the bridge of the nose (red line). This 83-year-old patient presented with 2 BCCs of which the infraorbital BCC was fully covered by the face mask.

Figure 2 Exemplary image of an infraorbital BCC that was partially covered by the patient’s face mask in an 81-year-old male patient from a single household.

Data collected included histologic diagnosis as well as tumor extent (including largest tumor diameter in mm). Potential confounders included patient gender (female or male), patient age at diagnosis, time from first symptoms to diagnosis and lesion laterality (right or left eye, both eyes). Dependent variables were considered best corrected visual acuity (BCVA) [logMAR], tumor recurrence, presence of multiple lesions and whether the registered contact person was living within the same household.

At presentation in the oculoplastic department of the eye clinic at Ludwig-Maximilians-University, a detailed photodocumentation is performed of the patients’ face in primary position (without mask). Additionally, a tumor close-up in a well-lit room is taken prior to further therapy.

Histopathology findings were taken from the original reports. All tumors included were treated by surgical excision: Depending on the clinical diagnosis, tumor excision was performed by small incision in the eyelid (anterior or posterior incision depending on the location and the size of the lesion) (chalazion) or, excisional biopsy (benign tumors), incisional biopsy (exclusion of malignancy or large malignant tumors requiring definitive treatment decision by the interdisciplinary tumor board with exact histologic diagnosis) or by histologically controlled excision examining the excision margins microscopically for remaining tumor cells until all tumorous cells had been removed (suspected small to medium sized malignant tumors). Histopathologic assessment was performed by a certified ocular histopathologist following formalin-fixation, paraffin embedding, sectioning, staining with haematoxylin and eosin, PAS (periodic acid-Schiff) staining and immunohistology when required.

In order to estimate relevant average coverage of the mid-face by masks, we performed evaluation of portrait photography assuming coverage of the face starting from an infraorbital line below the lower orbital fat pad over the lower orbital rims following a V-shape over the bridge of the nose (Figure 1).

Statistical data collection was performed using Microsoft Excel (Microsoft Corporation, Redmond, WA, USA) for Mac 2019. Data analysis was performed with SPSS 26.0 (IBM Corporation; Armonk, NY, USA). Descriptive statistics, Fisher’s exact test and Mann–Whitney U-test were employed to assess the patients’ clinical and functional data. To avoid bias when evaluating clinical, histopathological data and photography, respective histopathology and clinical data were first recorded separately and combined for analysis only after completion of the data collection process. Statistical significance of differences was accepted for p<0.05 in two-tailed tests. As all statistical procedures are of exploratory nature, no adjustment for multiple testing has been made.

Results

Analysis was completed in 257 patients. Demographic data are shown in Table 1.

Table 1 Demographic Patient Data, SD: Standard Deviation

A total of 112 patients (43.6%) presented with benign and 145 patients (56.4%) with malignant lesions.

Details on final diagnoses are elaborated in Table 2.

Table 2 Final Diagnosis of the Infraorbital Pathology from the Patients’ Medical File

Documented surgical patient management included small incision in the eyelid depending on the location and the size of the lesion (chalazion), or excisional biopsy as a one-step procedure (n = 96, 37.4%), excision with microscopic examination of the margins for remaining tumor cells until all tumorous cells had been removed (two- or multiple step-procedure, respectively) (n = 141, 54.9%), and incisional biopsy (n = 20, 7.8%). Incisional biopsy resulted in malignant histology in 5 (of 20, 25%) cases and was followed by resection in 3 cases (1.2%) including additional adjuvant radiotherapy in one case and oral therapy with a hedgehog pathway inhibitor (vismodegib) alone in one case (0.4%). One patient declined further therapy after incisional biopsy (0.4%).

Fifty-one (19.8%) patients presented with more than one periocular lesion (33 (64.7%) patients with solely benign and 18 (35.3%) with at least one malignant lesion). In these patients, the risk of at least one periocular lesion being covered by the face mask was significantly elevated (p=0.009, Fisher’s exact test; 6 (11.8%) of lesions partially and 3 (5.9%) fully covered).

The lesion in question was partially covered by the face mask in 32 (12.5%, for exemplary image see Figure 2) and fully covered by the face mask in 20 (7.8%) patients. The infraorbital and lower lid lesions were not covered by the face mask in 205 cases (79.8%).

A total of 125 patients (48.6%) reported to be living in a single household. Importantly, significantly more patients with lesions that were partially or fully covered by the face mask came from a single household (p=0.003, Fisher’s exact test). Figure 3 demonstrates that percentages of full coverage of the infraorbital lesions were especially high in patients living in single households.

Figure 3 Diagram showing that 19 (15.2%) and 16 (12.8%) of 125 patients had partial and full coverage of an infraorbital lesion living in single households whereas 13 (9.8%) and 4 (3.0%) of 132 patients had partial and full coverage by the face mask in patients who reportedly lived in a household with family, respectively (p=0.003, Mann–Whitney-U test).

BCVA was significantly lower in patients with larger tumor diameter (p < 0.001, Fisher’s exact test).

Malignant lesions were significantly larger than benign pathologies (largest diameter, malignant median 9.0mm, range 1.3–34.0mm, mean 10.5mm, and benign median 3.0mm, range 1.0–7.0mm, mean 4.3mm, respectively; p<0.001, Mann–Whitney-U test), and patients presenting with malignant lesions were significantly older (median age 78.4, range 33–93, mean 73.7, SD 13.1 years versus median age 57.9, range 18–90, mean 59.8, SD 19.3 years, respectively; p<0.001, Mann–Whitney-U test). In subgroup analysis that compared patients with malignant versus benign lesions, patients presenting with malignant lesions coming from a single household were significantly older (p=0.041, Mann–Whitney-U test) whereas there was no significant correlation between age and family status in patients presenting with benign lesions (p=0.943, Mann–Whitney-U test).

Discussion

The most important results of this study include that approximately 20% of infraorbital pathologies may be overlooked due to face masks if the mask is not transparent or removed during clinical examination. This number is meaningful, as predominantly malignant tumors have been reported to locate in the lower lid which, as in the present study, could be attributed to the majority of basal cell carcinomas (BCCs) with a predilection of the lower eyelid.⁶ Essentially, the present study shows that the presence of one periocular lesion warrants thorough examination for further pathologies. Furthermore, during isolation due to the pandemic, patients living in single households are at risk of underdiagnosis due to permanent coverage of the lesion by the face mask outside of the patients’ private household. This is especially important regarding malignant tumors in the elderly as these patients are at greater risk of presenting with malignant skin tumors^7,8 due to longer duration of exposure to carcinogenic substances.⁷

Of course, large meta-analyses suggest that face masks provide a significant protective effect against the spread of SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2).⁹

Yet, great benefit on the one hand may on the other hand also be accompanied by certain risks that must not be overlooked: As we could prove in this study, caution must be taken concerning potential coverage of lesions by face masks. In the literature, we retrieved a case report entitled “The COVID-19 facemask: Friend or foe?” reporting a non-viable median forehead interpolation flap due to compression caused by said face mask.¹⁰ The present study adds important data drawing attention to potential risks of wearing a face mask by reporting percentages of potential underdiagnosis of infraorbital tumors, more specifically for the elderly and patients living alone. The majority of data collected before the pandemic showed a higher percentage of benign compared to malignant lid tumors,^8,11,12 which was in contrast to the findings in the present study and might be accounted for by the outer circumstances of the pandemic (fear of consultation at the hospital, restrictions in scheduling appointments etc). This is an interesting point for future research.

This study has certain limitations: During the pandemic, patients may have refrained from early presentation at their treating Ophthalmologists’ office due to the COVID-19 pandemic in general. Secondly, patients wear different types of masks that may cover variable parts of the mid-face. For the evaluation of this study, we assumed coverage of the face starting from a slightly V-shaped infraorbital line below the lower orbital fat pad over the lower orbital rims reaching the bridge of the nose centrally.

Furthermore, time from first symptoms to presentation in the clinic varied greatly (0–180 months) with patients with benign diagnosis generally presenting after a longer period in time due to only minor changes of the lesions in question. One could argument that these lesions might have been noticed by the patient and his environment prior to wear of a mask. Yet, for the purpose of this cross-sectional study, we wanted a clear cut-off period of cases presenting to our Ophthalmology department after legal requirement of wearing said face mask.

As expected, compared with the literature, patients presenting with malignant lesions (as cumulative solar damage mostly BCCs) were significantly older^6–8,11,13 and malignant lesions significantly larger. BCC accounted for 83.0% of malignancies in this patient collective, which is in good accordance with previous publications from different parts of the world.^7,8,12

We would assume from these presuppositions, that our meaningful number of patients represents good overall generalizability.

Conclusions

With the initiation of this study, we wanted to create general awareness that a thorough examination of each patient without the face mask should be included in every general ophthalmologic examination. This may specifically help timely diagnosis and treatment of infraorbital lesions in the elderly (who are at greater risk of developing cutaneous malignancies) as well as patients coming from single households and might uncover additional pathologies.