Oral and oropharyngeal mucosal lesions: clinical-epidemiological study of patients attended at a reference center for infectious diseases

Highlights • Oral mucosal lesions of infectious diseases and neoplasms were the most frequent.• Clinical-epidemiological characteristics of oral manifestations are often similar.• Systematic oral and oropharyngeal examination is essential for differential diagnosis.• Multidisciplinary teams in medical routine can improve early diagnosis.• Standardized medical records can provide tools for differential diagnosis.


Introduction
The oral cavity plays an important role in the physiology of the human organism, emphasized by the popular saying, 'health comes first, and it enters through the mouth'.The anatomical and functional continuity between the oral cavity and oropharynx highlights the need to understand lesions of these anatomical areas.Oral or Oropharyngeal Mucosal Lesions (OOPML) include any mucosal alteration of the oral cavity/oropharynx, which may result from developmental disturbances, infections, allergic or inflammatory processes, neoplasms, or other histomorphological alterations of the epithelium and soft tissues.OOPML can be caused by primary diseases of the oral cavity/oropharynx or be clinical expressions of other organ or systemic diseases (e.g., autoimmune, infectious, or neoplastic).Therefore, OOPML may be the primary, most significant, or unique signs of diseases, leading to direct or indirect consequences on the individual's health. 1---4ccording to the World Health Organization (WHO), oral diseases affect 3.5 billion people worldwide and the number of cases is increasing globally. 5Thus, a complete, systematic evaluation of the oral cavity/oropharynx is essential for the diagnosis and follow-up of primary diseases of the Upper Aerodigestive (UAD) tract or of other origins.
The data derived from the study of the clinicalepidemiological characteristics of OOPML can assist health professionals in the clinical and laboratory evaluation of patients.Our objective was to determine OOPML prevalence and anatomical location, and to describe the epidemiological profile of the patients, in addition to the first symptom presented, the diagnostic conclusion, and the time of disease evolution.

Methods
A retrospective cross-sectional study of 7551 medical records was performed, and patients with OOPML attended at the Otorhinolaryngology Service of the Evandro Chagas National Institute of Infectious Diseases (INI-FIOCRUZ) from January 2005 to December 2017 were included in the study.Clinical and epidemiological data were collected and stored in a database for statistical analysis.This study was approved by the Research Ethics Committee of INI-FIOCRUZ under protocol number 759873179.0000.5262.
The criteria used for diagnostic confirmation were the presence of OOPML associated with the patient's medical history, clinical characteristics of the lesion, serological tests, direct or histopathological examinations or the culture of specimens obtained from the oral cavity, oropharynx, or other anatomical sites with concomitant manifestations, or clinical/radiological suspicion associated with OOPML remission after specific treatment (Table 1).
OOPML presented by the patients were classified into developmental disturbances, Non-Granulomatous (NGID) and Granulomatous Infectious Diseases (GID), autoimmune diseases, neoplasms (benign and malignant), and epithelial and soft tissue diseases Not Classified in Other Categories (NCOC).The inclusion of OOPML in this last category was based on the classification criteria of Neville et al. (2016). 7yphilis was classified as a NGID, owing to its nonspecific histopathological pattern in the primary and secondary stages.Nonspecific ulcerated lesions were classified as those that resolved spontaneously or without specific treatment.The prevalence of dental caries and periodontal disease was  S1.Two classifications regarding OOPML location were used: general and oral cavity/oropharyngeal subsites.The definition of subsites followed the anatomical division proposed by the TNM classification of malignant tumors, 8 with the following modifications: tonsillar pocket and tonsil were considered as ''tonsil''; uvula was considered as ''soft palate''; and upper/lower lip and labial commissure were considered as ''oral only''.
The Statistical Package for Social Science (SPSS) for Windows, version 16.0 (SPSS Inc., Chicago, IL, USA), was used for data analysis.The simple frequencies of categorical variables were determined, as well as the summary measures (mean ± Standard Deviation [SD], median, Interquartile Range [IQR], and minimum and maximum) of continuous variables.
For GID cases, given the potential link to rural areas, information regarding residence/labor activity in urban/rural areas was collated.Data were available for 123 (79.4%) patients, comprising 86 (69.9%) from urban areas and 37 (30.1%) from rural areas (Supplemental Table S3).
Simultaneous involvement of the oral cavity/oropharynx was uncommon in the patients included in this study (12.6%) (Table 3).
The most affected oral/oropharyngeal subsites, in descending order, were the palatine tonsil, hard palate, tongue, and soft palate (Table 4).
Data on the first mucosal sign/symptom presented by patients with OOPML were available for 286 (46.1%) patients, and local pain and odynophagia were the most common.Information on HIV co-infection was recorded for 203 (32.7%) patients and more than half of these were HIVpositive (Supplemental Tables S4 and S5).

Discussion
OOPML were observed in 10.2% of all patients evaluated during the study period.Most of the identified cases were infectious diseases (Mainly Paracoccidioidomycosis [PCM], candidiasis, and American Tegumentary Leishmaniasis [ATL]), followed by malignant neoplasms.The data collected in this study come from a reference center for infectious diseases, which could explain the high percentage of infectious disease related OOPML.This did not, however, prevent the diagnosis of a variety of non-infectious diseases, which in itself demonstrates the difficulty of OOPML differential diagnosis, since most of these patients were referred to our center as a suspected infectious disease case.In other epidemiological surveys, oral lesions of nonodontogenic/non-periodontal infectious diseases range from 0.8% to 23.2%, mostly restricted to herpes, PCM, and candidiasis, with the latter being the most frequent. 9---13he prevalence of oral lesions is primarily determined through population-based studies, 14---18 or studies carried out in dental centers 19---21 or from oral pathology laboratories. 22---24However, as no standardization in OOPML classification exists, with studies classifying OOPML by a lesion group (e.g., non-neoplastic lesions) 19 or by a specific disease (e.g., oral lesions in syphilis), 25 the reported OOPML frequencies are directly influenced. 26---28Furthermore, the lack of systematic and standardized inclusion of   oral cavity/oropharynx examination in the routine medical examination generates gaps in medical records, 29,30 and epidemiological surveys on oral health only provide information on diseases related to dental elements (e.g., caries, edentulism). 5,31Despite the limitations related to retrospective studies, the present study from an otorhinolaryngology service provides OOPML prevalence data for numerous diseases that were diagnosed.The prevalence of OOPML reported in studies can be influenced by the country or geographic region in which the study is conducted, the socioeconomic level of patients, and the methodologies used, which may explain the large variation observed among publications (4.9%---64.5%). 14---18Our study demonstrated that 10.2% of the total number of patients seen at the Otorhinolaryngology Service of INI-FIOCRUZ had OOPML, similar to the prevalence observed in populationbased studies. 16,18ur sample predominantly comprised white, male individuals, in the fifth and sixth decades of life.Age group and skin color can vary depending on the study, as some studies on OOPML report a similar distribution, 13,16,21,32 while others report a female majority. 6,9,14,16,32The male predominance in the present study may be related to the higher frequency of GID and Squamous Cell Carcinoma (SCC), diseases which are more common in men; 33---35 in contrast, benign lesions, mainly inflammatory fibrous hyperplasia, occur more frequently in women. 6,9,13Most of our patients had low education levels similar to that observed by Souza  et al. (2017).Lower levels of education have been associated with infectious and neoplastic diseases of the UAD tract, which were very frequent in our sample. 36---38moking and alcohol use are generally associated with an increased OOPML incidence in GID and malignant neoplasms. 17,18,38,39However, data on these factors were only available in 34% and 20.6% of patients, respectively.Although studies use different concepts of smoking and alcohol use, our study only considered the reference to smoking or alcohol consumption in the medical records in the data collection and, from this, we were able to observe similar frequencies of smokers and drinkers as in other studies. 13,21ocal pain and odynophagia were the first symptoms most reported by patients, whereas in the study by Santos et al.  (2013), most patients were asymptomatic. 21This difference can be attributed to the most frequent type of lesion found.Santos et al. (2013) reported OOPML of inflammatory fibrous hyperplasia as the most frequent, a disease which is usually asymptomatic, whereas in our study, OOPML of SCC, autoimmune diseases, acute tonsillitis, and pharyngitis, which are usually associated with local pain and/or odynophagia, were more prevalent. 40---42he most frequently affected oral/oropharyngeal subsites were, in descending order, the palatine tonsil, hard palate, tongue, and soft palate, probably influenced by the frequency of acute tonsillitis/pharyngitis, autoimmune diseases, GID, and SCC.Likewise, the OOPML locations in other epidemiological surveys varied according to the diseases observed. 13,14,20,21,32It is worth noting that the anatomical division of the oral cavity/oropharynx between studies is not standardized. 6,9,13As an example, the soft palate, considered as an oral cavity subsite by some authors, 6,13,21 was considered as oropharynx in the present study based on TNM anatomical division criteria. 8 The median time of disease evolution indicated that patients take approximately three months until the first medical consultation, similar to that reported by Santos et al. 21This extended waiting time could affect early diagnosis, which is important in reducing sequelae. 39,43A longer evolution time was observed for patients with GID and neoplasms, which are chronic diseases, often with an insidious and initially oligosymptomatic evolution.As such, patients may delay seeking medical care, in addition, accessing medical resources and laboratory tests for their diagnoses may be hindered.Conversely, NGID generally have more intense and rapidly evolving symptoms, encouraging patients to seek medical care earlier.
Lymphoepithelial cyst, leukoedema, ranula, mucocele, pyogenic granuloma, benign migratory glossitis, fibroma, and leukoplakia were observed at lower frequencies than in other OOPML epidemiological surveys, probably because patients with these OOPML are usually treated in dental, rather than otorhinolaryngology, services. 6,9,19,20,27,32he frequency of autoimmune disease cases observed in the present study corroborates that of Carvalho et al.  (2011).In both studies, the immunologically mediated dermatological diseases with OOPML were diagnosed as Lichen Planus (LP), pemphigus vulgaris, and mucous membrane pemphigoid. 44LP OOPML were the most frequent, as in other studies. 14,16,17,45However, these lesions can still be considered rare, since the global prevalence is ∼1%. 46lmost all the benign neoplasm-related OOPML were diagnosed as squamous papilloma.The presence of this OOPML may be related to HIV infection, since most of these patients were carriers of the virus, a population at greater risk of HPV infection. 47The frequency of squamous papilloma was similar to that observed in other studies, although the rate of immunosuppression was not reported. 13,17,32egarding malignant neoplasms, the occurrence of SCC was equal to or greater than that reported in other OOPML surveys. 9,20,32The high frequency of this disease may be a consequence of the similarity between the clinical and epidemiological characteristics of SCC OOPML with those of GID, which justifies the referral of these patients to our center.This emphasizes the importance of biopsies for diagnosis, to rule out concomitant lesions of other etiologies, and to investigate the association of SCC with HPV.HPV is an important risk factor for SCC, especially for oropharyngeal cases, and an increase in HPV-positive cancers has been observed in Brazilian cohorts. 48,49s in other studies, the total OOPML frequency in the different lymphomas was <1%, confirming its rarity. 50,51We observed a higher frequency of OOPML from sarcomas than that observed in neoplasms of hematological origin, unlike what was reported by Allon et al. 40 HIV co-infection may have influenced this difference, since the literature demonstrates that OOPML can be observed in up to half of patients with Kaposi sarcoma and AIDS. 3,52,53e observed nonspecific ulcerated lesions at frequencies similar to those reported by other studies for traumatic ulcerations and recurrent aphthous stomatitis. 13,14,17,18,21,32,54Ulcerated lesions may be under-diagnosed because their course is short and self-limited, meaning that most patients do not seek medical care.
Candidiasis and PCM were the most frequently observed diagnoses in cases of NGID-and GID-related OOPML, respectively.We found a higher frequency of candidiasis than that reported in other OOPML surveys with no defined age group.This possibly occurred owing to the higher frequency of HIV co-infection in our sample. 9,13,45However, the observed frequency of this fungal infection was lower than that shown in surveys conducted in older adults. 11,45Most patients in our sample were in their fifth and sixth decades of life, and the use of dental prostheses, which has been linked to these infections, was possibly lower.
Acute tonsillitis and pharyngitis were also frequent in the NGID group, and predominantly occurred in patients in their second and third decades of life, in contrast to that observed in other studies, which have reported a higher prevalence in children and adolescents. 10,42,55This difference in age groups can be attributed to the fact that our service mainly meets the demand of adult patients.
OOPML of syphilis were uncommon in our sample, despite the increase in the number of syphilis cases in recent years. 56,57The low prevalence in the current study may be owing to the fact that patients with clinical suspicion of this disease are routinely treated at Sexually Transmitted Disease/AIDS Outpatient Clinics, and, as OOPML improvement typically occurs with the beginning of treatment, patients do not seek evaluation at other services.No cases of oral syphilis lesions have been reported in any other OOPML survey. 16,20,58Despite syphilis being a notifiable disease in Brazil, the OOPML prevalence of this disease is likely underreported, since the clinical form of the Notifiable Diseases Information System does not include the registration of OOPML. 56he low frequency of herpes was similar to that observed in other OOPML surveys. 12,13,17,59As herpes OOPML are usually recurrent and immunocompetent patients are already familiar with the self-limited evolution, 60 they do not typically seek medical or dental care.
The oral cavity/oropharynx are commonly affected in PCM. 61,62The PCM OOPML frequency in our study was proportionally higher than that of other studies when considering the duration of the studies, 51,62 including that observed in a study carried out in a region with high PCM prevalence. 63his higher prevalence is likely owing to the fact that our service is a reference center for infectious diseases and conducts the systematic otorhinolaryngological examination of patients referred by other services.For the same reason, ATL was the second most frequent diagnosis in GID-related OOPML cases.In this disease, the oral cavity/oropharynx are the second most affected anatomical sites in the head and neck. 43,64Underreporting of these OOPML may also occur as a result of the lack of oral/oropharyngeal examination in the medical routine.
The frequency of tuberculosis OOPML was slightly higher than that observed in the literature, 65,66 which is likely related to the systematic oral/oropharyngeal examination performed at our otorhinolaryngology service.Overall, the prevalence of tuberculosis OOPML is difficult to estimate owing to the low frequency, 65,66 in addition to the lack of data in official reports, which generally only report the incidence of extrapulmonary forms of the disease. 67,68tudies that provide the prevalence of histoplasmosis OOPML report the percentage of these lesions in patients with the disease and not in the general population. 69,70espite sporotrichosis being an endemic disease in the state of Rio de Janeiro, 71 the frequency of OOPML was low in our study, confirming the rarity of lesions in this disease. 72,73In addition to OOPML in leprosy being rare, 74 the low frequency in the present study may be related to the fact that most patients are routinely treated at specific leprosy reference centers. 75

Conclusions
Diseases that affect the oral cavity/oropharynx are the subject of study in several areas of health sciences, such as dentistry, otorhinolaryngology, and dermatology.For this reason, lesions in these anatomical areas are often evaluated in a fragmented way.Studies on the general prevalence of OOPML are scarce and surveys are often carried out for specific disease groups or by dental centers.Like dentists, otolaryngologists may be the first professionals to identify OOPML.Therefore, the organization of multidisciplinary teams that include otolaryngologists for routine UAD tract examinations, even in asymptomatic cases, could facilitate the early diagnosis and treatment of many diseases, thus reducing morbidity and improving the prognosis, as in many cases, patients only show symptoms when in a more advanced stage.In addition, the use of standardized medical records for systematic examination of the oral cavity/oropharynx can provide tools for differential diagnosis and relevant information for new clinical-epidemiological studies.

Fig. 1
Fig. 1 Flowchart of the selection of patients with oral or oropharyngeal lesions among the 7551 patients attended at the Otorhinolaryngology Service of the Evandro Chagas National Institute of Infectious Diseases (INI-FIOCRUZ), from 2005 to 2017.

a
NCOC diseases -epithelial and soft tissue diseases not classified in other categories.

Table 1
Diagnoses and the diagnostic methods of patients with oral or oropharyngeal lesions among the 7551 patients attended at the Otorhinolaryngology Service of the Evandro Chagas National Institute of Infectious Diseases (INI-FIOCRUZ), from 2005 to 2017.

Table 2
Epidemiological characteristics of patients with oral or oropharyngeal mucosal lesions attended at the Otorhinolaryngology Service of the Evandro Chagas National Institute of Infectious Diseases (INI-FIOCRUZ), from 2005 to 2017.
a NCOC diseases -epithelial and soft tissue diseases not classified in other categories.b Number of patients with available information.c Range from illiterate up to last year of elementary school.d High school, associate degree, undergraduate degree, and graduate school.e Rio de Janeiro city and metropolitan region.f Other states of Brazil.

Table 3
General location of oral or oropharyngeal mucosal lesions of patients attended at the Otorhinolaryngology Service of the Evandro Chagas National Institute of Infectious

Table 4
Subsites of oral or oropharyngeal mucosal lesions of patients attended at the Otorhinolaryngology Service of the Evandro Chagas National Institute of Infectious Diseases (INI-FIOCRUZ), from 2005 to 2017.
a May be more than 1 subsite per patient.b NCOC diseases -epithelial and soft tissue diseases not classified in other categories.

Table 5
Frequency of oral or oropharyngeal mucosal lesions of patients attended at the Otorhinolaryngology Service of the Evandro Chagas National Institute of Infectious Diseases (INI-FIOCRUZ), from 2005 to 2017.