Oral Manifestations of COVID-19 in Hospitalized Patients: A Systematic Review

Background: COVID-19 disease first appeared in 2019 and quickly spread worldwide, causing a global pandemic. The oral cavity represents a target of SARS-CoV-2, and oral lesions are observed in both non-hospitalized and hospitalized patients. This systematic review aims to investigate the frequency of oral manifestations in COVID-19 hospitalized patients. Methods: An electronic search was conducted in PubMed, Scopus, and Web of Science databases, including articles published up to September 2021. The review protocol was based on PRISMA-P. The risk of bias of the studies was assessed using the Joana Briggs Institute. The certainty of evidence was assessed using the GRADE instrument. Results: Fifty-nine articles were included: 19 case reports, 17 case series, 2 case-control studies, 13 cross-sectional studies, 4 observational studies, and 4 retrospective studies. Oral ulcers, cheilitis, and tongue lesions were more common in patients before hospitalization, while perioral pressure ulcers, macroglossia, blisters, and oral candidiasis were more recurrent in patients during hospitalization. The first could be related directly to COVID-19, while the latter could be caused by medical devices, treatments, prone position, and immunological impairment. Conclusions: An accurate oral examination during the hospital admission of all confirmed COVID-19 cases is encouraged to recognize oral early manifestations and to apply appropriate treatments.


Introduction
The outbreak of novel coronavirus SARS-CoV-2 has created a global crisis and challenged healthcare systems across the world [1]. It was first identified in Wuhan, China, on 31 December 2019, in association with a severe human respiratory disease . On 30 January 2020, the World Health Organization (WHO) declared the COVID-19 infection as a Public Health Emergency of International Concern.
SARS-CoV-2 is an enveloped, positive-sense, single-stranded RNA virus with multiple spikes on the surface and a genome size of approximately 26-32 kilobases. Fever, dyspnea, body aches, and dry cough are the most common symptoms [2], but severe cases can develop pneumonia, severe acute respiratory syndrome, and kidney failure, representing a life-threatening condition [3,4]. Most patients display moderate symptoms (80%), while 20% of them may develop a severe disease and 5% may become critically ill, developing pneumonia or acute respiratory distress syndrome, which requires mechanical ventilation and intensive care unit hospitalization [5]. SARS-CoV-2 mainly spreads through respiratory droplets, aerosols, contact, and fomites [6][7][8][9].
Studies were selected by title and abstract, and full-text screening. Articles were included if they reported oral manifestations in hospitalized patients with COVID-19 infection.
The full-text articles of all potential studies were evaluated according to the eligibility criteria. The inclusion criteria were: (1) hospitalized patient affected by COVID-19, confirmed through RT-PCR testing; (2) articles reporting oral manifestation associated with and (3) studies published in the English language. The exclusion criteria were: (1) not-confirmed COVID-19 cases; (2) duplicate studies and data; and (3) full-text not accessible. There were no disagreements during the article selection process.
The primary outcome was to highlight the oral signs and symptoms that can occur in hospitalized patients affected by  A flow diagram detailing the process is presented in Figure 1.

Quality Assessment
The risk of bias of each study was assessed by two blinded reviewers using the Joanna Briggs Institute (JBI) critical appraisal checklist for case reports [29,30]. The score system was agreed by all reviewers before the critical appraisal assessments, and studies were classified according to the following categories: (a) low risk of bias, if the studies reached more than 70% scores of "yes"; (b) moderate risk of bias, if "yes" scores ranged from 50% to 69%; and (c) high risk of bias, if "yes" scores were less than 49%. Data are reported in Supplementary Materials (Table S1). Full-text articles assessed for eligibility (n = 187) Full-text articles excluded, with reasons: Studies reporting oral lesions in non-hospitalized patients or studies in which patient status is not specified (n = 58); Studies reporting dental practice or dental management during COVID-19 (n= 11); Data non individualized for oral lesions (n= 24); Reviews, letters to editor without presenting case report or case series (n= 31); Other reasons (n= 4).

Quality Assessment
The risk of bias of each study was assessed by two blinded reviewers using the Joanna Briggs Institute (JBI) critical appraisal checklist for case reports [29,30]. The score system was agreed by all reviewers before the critical appraisal assessments, and studies were classified according to the following categories: (a) low risk of bias, if the studies reached more than 70% scores of "yes"; (b) moderate risk of bias, if "yes" scores ranged from 50% to 69%; and (c) high risk of bias, if "yes" scores were less than 49%. Data are reported in Supplementary Materials (Table S1).

Certainty of Evidence
The evidence levels of the selected studies were assessed using the GRADE instrument (http://gradepro.org (accessed on 28 October 2021)) [31]. The certainty of evidence was rated for oral lesions prevalence in hospitalized patient. This assessment was based on study design, risk of bias, inconsistency, indirectness, imprecision, and other considerations. Evidence quality was characterized as high, moderate, low, or very low [32]. Data are reported in Supplementary Materials (Table S2).

Statistical Analysis
Descriptive statistical analysis was conducted by grouping and comparing data using Microsoft Excel software (2019, Microsoft Corporation, Redmond, WA, USA)

Study Selection and Characteristics
A total of 340 studies were identified from databases, and, after removing the duplicate, 193 studies remained for title and abstract screening. A full-text reading was conducted on 187 studies. A total of 128 studies were excluded according to the eligibility criteria. Therefore, 59 studies were selected: 19 case reports, 17 case series, 2 case-control studies, 13 cross-sectional studies, 4 observational studies, and 4 retrospective studies (Table S1). Due to the lack of available data, case reports and series were also included in this systematic review. Thus, the risk-of-bias of each study was evaluated using the JBI critical appraisal checklist [33]. A flowchart detailing the process is presented in Table S1.
The sample consisted of 1219 patients, of which there were 456 men (37.4%), 374 women (30.7%), and 389 patients with no specified sex (31.9%), with a mean age of 50.8 years. Results from each selected article were subdivided into "oral lesions appeared before hospital admission" and "oral lesions appeared during hospitalization". The studies were set in two tables (Tables 1 and 2), reporting the following data: first author's name and year of publication, study design, sample size and gender, mean age, medical history, oral manifestations, time of onset, affected site, treatment of COVID-19, treatment of oral manifestation, disease duration, reported diagnosis, and the risk of bias.
Before hospitalization, white tongue, necrosis of the dorsal tongue, glossitis, geographic and fissured tongue, strawberry tongue, and depapillation of the tongue were the most common lesions diagnosed, directly related to SARS-CoV-2. One patient with a complex medical history showed a 1.5 × 1.5 cm ulcer in the right border of the tongue, while another patient reported a mucopurulent membrane in the anterior dorsal tongue [37]. Only three studies that reported tongue lesions referred to Kawasaki-like disease, potentially associated with COVID-19 [38,51,65].
During hospital admission, the main SARS-CoV-2 related lesions were tongue ulcers, especially in ICU patients [21]. Moreover, white plaque on tongue dorsum, geographic tongue, tongue redness, fissured tongue, and macroglossia were reported [35,54,66]. Macroglossia was reported in two patients who experienced prolonged pronation cycles for several days. Andrews et al. suggested 10 days of methylprednisolone in addition to a bite block to prevent this complication [35]; on the contrary, according to Mascitti et al., macroglossia could be referred to acute lymphatic and vascular obstruction due to COVID-19-related inflammatory response [54].
Before and during hospitalization, ulcerative and erosive lesions were the most common orofacial manifestations of COVID-19. The ulcers emerged after a latency time of 4 to 7 days after the onset of COVID-19 symptoms, and most of them were diagnosed upon hospital admission. Only in one case did small oral ulcers appear 40 days after a COVID-19 positive test [42]. Some patients displayed painful herpetic and hemorrhagic ulcers with irregular margins, which were variable in size and number. Ulcers that appeared before hospital admission were located on the hard palate and lips, while the lesions that occurred during hospitalization mainly affected the tongue dorsum, lips, and buccal mucosa.
Some authors suggested different factors involved in the development of ulcerative and erosive lesions [12,41,60,61,91]. Ulcers could be related directly to SARS-CoV-2 infection or could be caused by drugs, vasculitis, or thrombotic vasculopathy secondary to COVID-19 [34,41]. Oral ulcerative lesions and erosive plaques appeared a few days after the onset of respiratory symptoms and worsened during hospitalization, due to persistent immunological impairment, and lesions did not heal after SARS-CoV-2 eradication [47]. In four ICU patients, the authors suggested that oral ulcers could be caused by medical devices during the prone positioning phases [21,57]. A wide range of therapies has been used for oral ulcers, including drugs (e.g., dexamethasone, tetracycline) and photobiomodulation therapies [37,57,64].

Aphthous-like Lesions
Aphthous-like lesions were reported in four studies, affecting six minors and three adults (>50 years). Moreover, aphthous-like lesions were showed in 78 cases with various oral signs and symptoms associated with COVID-19 [88]. Lesions were mainly related to COVID-19, probably due to the distribution of the ACE2 receptor on the oral mucosae [11,86], although one patient with minor aphthous ulcers was diagnosed with Sweet syndrome related to COVID-19 [62]. Stress and immunosuppression secondary to COVID-19 infection could be other possible reasons of these lesions [92].
Favia et al. detected oral blisters in 19 hospitalized patients (15.4%) that appeared during the first week after the onset of general symptoms and were mainly located on the tongue and palate [70]. According to the authors, lesions could be related to SARS-CoV-2, medical treatments, and/or poor oral hygiene.
Orolabial recurrent herpes simplex were observed in eight patients [46]. This manifestation was found in at HSV-1-positive patients, suggesting a possible superinfection of herpetic virus with COVID-19.
Lip lesions included fissured lip, angular cheilitis, and perioral pressure ulcers. Fissured lips have been detected mainly in young patients, who were diagnosed with Kawasaki-like disease related to COVID-19 [39,40,51,54,65,71,72]. Mazzotta et al. described a 9-year-old male with Down syndrome and alopecia areata universalis, presenting glossitis and cheilitis, probably due to the excessive and persistent inflammation that occurred during the interstitial pneumonia with acute respiratory failure [74]. Angular cheilitis, due to stress and immunosuppression, was observed in a 53-year-old man a few days after hospital discharge; after the treatment, this lesion completely disappeared [43].
Prolonged prone position and endotracheal intubation were the most common risk factors for perioral and lip pressure ulcers, interfering with mechanical ventilation equipment in the critical care setting [21,49,56,58,59,66,73,77,78,90]. Most cases were treated with antimicrobial dressing, debridement of necrotic tissue, and paraffin gauze dressing. Finally, some articles reported lip lesions directly related to COVID-19, which appeared as a mucopurulent membrane with superficial necrosis or swollen lips and disappeared after 10-12 days of treatment [37,63,71].

Functional Disorders
Functional disorders included xerostomia, viscous saliva, ageusia and dysgeusia, halitosis, burning mouth, masticatory muscle weakness, salivary gland ectasia, temporomandibular article abnormalities, and facial tingling. These disorders were reported by several studies referred to hospitalized patients affected by 34,43,76,82,88]. The authors suggested these alterations could be directly related to SARS-CoV-2 and could be considered as an early manifestation of COVID-19 infection. Indeed, ACE2 receptors have been found in the taste buds and glandular epithelium, leading to salivary gland dysfunction and salivary flow impairment [93].
Masticatory muscle weakness, salivary gland ectasia, temporo-mandibular article abnormalities, and facial tingling were reported by Gherlone et al. [48]. The authors suggested that salivary gland ectasia reflected the hyperinflammatory response to SARS-CoV-2, as demonstrated by the significant relationship with C-reactive protein and lactate dehydrogenase levels at hospital admission, and antibiotics use during acute disease.

Candidiasis
Candidiasis appeared as white plaques on the dorsum of tongue, gingiva, and palate [43,44,66,70,75,80,81]. The authors suggested it could be related to the long-lasting antibiotic therapy, the deterioration of general status, and poor oral hygiene [80]. Salehi et al. isolated 65 species of Candida spp. (70.7% C. albicans) in a cohort of COVID-19 patients with oropharyngeal candidiasis [75]. Interestingly, Dima et al. reported oral candidiasis with diaper erythema in three newborns positive with COVID-19 [44]. The oral candidiasis in young patients was also reported by Bardellini et al. [81], describing two cases of oral pseudomembranous candidiasis diagnosed at the hospitalization.

Nonspecific Lesions (Mucositis)
Petechiae, macules, erythema, stomatitis, brown pigmentation, mucositis, enanthema, and desquamative gingivitis were reported in many hospitalized patients [26]. Vascular disorders could cause mucositis in the affected patients. Indeed, Cruz-Tapia et al. described a 51-year-old female with diffuse vascular-like purple macule on the left palate and a papule-plaque on the right palate [68]. Riad et al. described 13 patients affected by mucositis; enanthema of the buccal mucosa, palate, and gingiva; and depapillation of the tongue, at hospitalization. These lesions disappeared after 7-14 days of "Magic mouthwash" and paracetamol and could be related directly to COVID-19 infection [6,89]. Only one study described a patient affected by oral lichenoid reaction and a case with oral enanthema, directly due to COVID-19 [54]. Marouf et al. suggested that periodontitis was significantly associated with a higher risk of COVID-19 complications, such as ICU admission, assisted ventilation, and the increased markers levels of COVID-19 worse outcome [55].

Discussion
A broad spectrum of signs and symptoms were reported in association with COVID-19; however, only a few studies highlighted oral clinical manifestations observed in hospitalized patients. The pathogenesis remains unclear, but some hypotheses have been formulated. Xu et al. showed a higher expression of ACE2 in the oral mucosa, especially on the tongue and in the salivary glands [11]. Thus, the oral cavity might be an anatomical site susceptible to SARS-CoV-2 infection [94]. Consequently, the interaction between SARS-CoV-2 and ACE2 might dysregulate the oral keratinocytes' function, leading to painful oral ulcers [37]. This mechanism could also be the basis of early manifestations of COVID-19, such as taste alteration and xerostomia [95]. The immune response to infection could activate Langerhans cells and lymphocytes, leading to vasculitis and thrombocytopenia, causing oral lesions related to vascular disorders (e.g., petechiae) [19,41,92,96].
It is still unclear whether oral lesions reflect a direct viral cytopathic effect or represent a consequence of stress, poor oral hygiene, systemic infections, medical treatments, or medical devices used during hospital admission [97]. However, to the best of our knowledge, no attempt has been made to review the available literature regarding oral lesions in hospitalized COVID-19 patients. Therefore, this systematic review is the first to characterizs the patterns of oral lesions that occurred in hospitalized patients affected by COVID-19. As intraoral examination has not yet been considered in the screening of the disease, literature still lacks evidence to better understand the onset of oral manifestations. In this review, the most common oral lesions seen in patients before hospital admission are painful ulcers, cheilitis, and tongue lesions. According to several authors, these lesions are related directly to COVID-19 [36,37,88,89]. Conversely, the most common oral lesions displayed by patients during hospitalization are perioral pressure ulcers, macroglossia, blisters, and oral candidiasis. These lesions may be due to the long-lasting prone position of ICU patients [21,90], increased pressure of the endotracheal tubes [59], prolonged inpatient care [52], persistent immunological impairment [43,47], and medical treatments [80].
The studies reporting that oral manifestations in COVID-19 are very low, leading to a considerable risk of bias. Most of the eligible studies showed a high risk of bias, due to insufficient simple size, poor methods, or inadequate designs. Case reports and case series have profoundly influenced the medical literature and continue to improve the current knowledge [98]. Although the evidence derived from case series and case reports is very low, a strong recommendation for oral lesions of hospitalized patients, affected by COVID-19, can be provide. Moreover, GRADE literature describes five paradigmatic situations in which a strong recommendation can be made based on low quality evidence, such as a condition of life threatening [99].
Considering the results of this systematic review, clinicians should consider the oral lesions affecting the hospitalized patients with COVID-19 to ensure an adequate prevention and therapeutic management. Thus, the importance of the oral examination should be emphasized in hospitalized patients with infectious diseases, to encourage the multidisciplinary management of COVID-19 patient.

Conclusions
The new SARS-CoV-2 has become a worldwide sanitary emergency. Understanding the manifestations and progression of COVID-19 is the first step in developing an adequate prevention and treatment management. Although the typical symptoms include fever, shortness of breath, and dry cough, oral lesions have also been reported. The association between oral diseases and SARS-CoV-2 infection is still unclear and is currently poorly investigated.
Based on this systematic review, it could be suggest that: (1) painful oral ulcers, cheilitis, and tongue lesions are more frequent in patients before hospital admission; (2) perioral pressure ulcers, macroglossia, blisters, and oral candidiasis are more evident in patients during hospitalization; (3) lesions that appeared before hospital admission are mainly related directly to COVID-19; (4) lesions that appeared during hospitalization are mainly associated with medical devices and treatments, prone position, and immunological impairment; (5) all clinicians, during the hospital admission, should be encouraged to perform an accurate oral examination of all confirmed COVID-19 cases to recognize the disease's possible early manifestations; (6) further studies are necessary to establish the pathological significance of oral manifestations during COVID-19.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/ijerph182312511/s1, Table S1: Risk of bias in the reported articles, Table S2: Certainty assessment and grading of the evidence.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.