Frequency and diversity of oral mucosal lesions
To our knowledge this study is the first to assess the frequency and diversity of OML in dermatologic patients, a selected group of the Sudanese adult population. The study group comprised patients with a wide range of dermatological diseases, yielding small numbers in each group, thus limiting the probability for stratified analyses. The most frequently occurring groups of dermatological diseases were spongiotic reaction pattern
, infectious diseases
, and vesiculobullous diseases
. This accords with the results of a recent survey by the International Foundation of Dermatology, reporting that infectious disease, dermatitis, and HIV-related skin disease are the main skin dermatological conditions at the community level worldwide [30
According to the present results, about 58% of the subjects investigated suffered from at least one type of OML, and the occurrence of any OML varied across groups of dermatological diseases from 46.8% in spongiotic to 72.2% in vesiculobullous reaction patterns. Tongue lesions were the most frequently occurring OML group (23.3%) followed by white lesions (19%), red and blue lesions (11%) and vesiculobullous diseases (6%). White lesions and red and blue lesions varied systematically with age, being most frequent in older persons, whereas ulcerative conditions were most common in males. Coated tongue, snuff dippers lesion, erythema and pemphigus vulgaris were the OML most frequently observed in the groups of tongue lesions, white lesions, red and blue lesions and vesiculobullous diseases, respectively.
The present findings should be interpreted with caution due to some limitations. Patients' refusal to volunteer for biopsy might have led to some misclassifications. Thus, some lesions that needed histological confirmation (leukoplakia, lupus erythematosus, pemphigus vulgaris, fibroepithelial polyp, chronic hyperplastic candidiasis, cheilitis glandularis, focal epithelial hyperplasia, Kaposi's sarcoma and some others) were diagnosed clinically and might contain error. Absence of standard methodological approaches and lack of agreed-upon diagnostic criteria, make comparison of epidemiological studies concerning the prevalence of OML difficult. In spite of the limitations associated with diagnostic criteria, all mucosal pathological alterations were identified in the present study.
Being a hospital based study; it is not possible to generalize from the study group to any larger population of skin diseased individuals inside or outside Khartoum. This is due to the rich geographical and socio-cultural diversity within Sudan, as well as the low utilization rate of health facilities generally observed in any developing country [31
]. Although the KTH received patients that have been referred from all over the country, biases in the study group might have been introduced due to differing referral procedures as well as the moderate response rate.
It is unsure how close an approximation the present figures are to the prevalence of OML in the general adult population of Sudan. Probably, the rates of OML presented in this study might be overestimated both with respect to the Sudanese population in general as well as to the population of adults suffering dermatological problems. Self-selection bias was considered to influence the result of the study as patients were more likely to respond when they had OML (the characteristic of interest). Moreover, with respect to the diversity of the types of OML, the present figures might be biased towards those for which people are more inclined to seek treatment, whereas other conditions are less likely to be identified in hospital based prevalence studies. Community based surveys based on random samples from the broader adult population should be recommended for future studies to estimate the actual prevalence and the health burden of OML in this country.
Since the precision of estimates tend to decrease with decreasing prevalence, the prevalence rates of rare conditions (≤ 1%) should be interpreted with particular caution. In addition, populations with different distributions of the risk factors identified for OML are not directly comparable without adjustment. Noteworthy the absence of an official patient's medical journal has created uncertainty regarding participants' self-reported medical condition and lifestyle patterns. A major limitation of self-reported data is recall biases in terms of underreporting of socially undesired events and a tendency to recall events as having occurred more recently than they actually did [33
]. Sensitive events, tobacco and alcohol use and some medical diagnoses would probably be under reported due to social stigma and social desirability.
Comparison of present findings with those of previous studies
In spite of its limitations, the present study provides important information about the frequency and diversity of OML in patients with various dermatological diseases as well as the social and behavioural factors that discriminate between skin diseased patients with and without OML. Moreover, OML in the present study may appear as a part of mucocutaneous diseases, a manifestation of systemic diseases (metabolic or immunological), or an expression of drug reaction. Some OML diagnosed could be attributed to trauma, infection, or denture use, or they could be a manifestation of specific cultural habits, like use of toombak. Due to the cross sectional nature of the present study, any causal relationship could, however, not be concluded upon.
Compared with the frequency of patients with OML observed in this study (57.9%), previous ones have shown point prevalence in the range 25% - 61.6% [6
]. Specifically, the frequency of patients with OML in the present study group was higher than those observed in the Cambodian (4.9%) [37
] Malaysian (9.7%) [7
], Spanish (51%) [12
] and Turkish (42%) populations [36
]. It was lower than that observed in population in Ljubljana (61.6%), but almost similar to the prevalence estimated in Spanish dental patients (58.7%) [8
]. In accordance with the NHANES III [6
] and the Swedish study published by Axell [5
], the present study used the WHO diagnostic criteria and Axell's diagnostic criteria [5
]. Thus, the present results are to some extent comparable with those previous studies, in spite that NHANES III and the study by Axell used large probability samples from the general populations. The frequency observed in this study was higher than that reported in NHANES III, amounting 28% in US adults aged 17 years and above.
Consistent with the results of NHANES III and other studies, the frequency of patients with OML presented in this study varied systematically and positively with being a male and with increasing age. Other epidemiological studies have shown an opposite sex gradient or no systematic variation according to sex [9
]. Sex differences in the occurrence of OML might be attributed to the high consumption of toombak by males, differences in genetic factors, social responsibility and masculinity believes [40
]. Use of toombak was reported by 12.5% of the total study group. In a study emanating from northern Sudan, the frequency of toombak use was estimated to 40% (43, 44). Males adopt a more active outdoor life-style and are exposed to some environmental risk factors to a higher extent than women. In contrast, women are more health conscious and faster to detect abnormality in earlier stages. Older people have higher risk to develop chronic diseases in general because of increased risk with increasing age due to metabolic changes, medications, prosthetic use, and psychological problem. Moreover, economic constraints and physical status of older people may limit their access to health care services [41
Epidemiological studies have revealed that tongue lesions
constitute a considerable proportion of OML, with prevalence rates varying across different parts of the world. Number and type of tongue lesions
involved in different studies have been an important factor in this variability. The present figure amounting to 23%, is lower than that reported in some previous studies [43
], but higher than the rates assessed in NHANES III and in the Hungarian population [6
]. Of interest was that 17 out of 30 patients (56.7%) with psoriasiform reaction pattern
had OML and that tongue lesions
(33.3%) were the most frequently occurring OML in this particular dermatological disease group (Table ). A study of Brazilian psoriatic patients revealed that 59% presented with tongue lesions, which was the most dominant OML [46
]. Similar findings have been reported by Hernandez-Perez et al [19
]. With respect to fissured tongue, the total of 7% of patients with fissured tongue observed in this study corroborates the range reported previously [5
]. Some few studies have reported high frequency of fissured tongue [35
]. Over the past few years an association between geographic tongue, fissured tongue and psoriasis has been postulated. Some authors believe that it is a natural developmental anomaly and a coincidence finding [46
] while others suggest a pathogenic relation between them [50
Snuff dipper's lesion was observed in 5.5% of the study group (Table ). This frequency is higher than that reported in the American and Kenyan population (1.2% and 0.4%, respectively) [6
], but lower than that observed in the Swedish population (15.9%) [52
]. Toombak has been known to play a major role in the aetiology of oral cancer in the Sudan [23
]. It contains at least 100-fold higher concentrations of the carcinogenic factor tobacco specific N-nitrosamines compared with American and Swedish commercial snuff brands [53
]. A recent study showed that toombak induces DNA damage and cell death in normal human oral cells more than the Swedish snuff [54
The frequency of oral leukoplakia (3.1%) disclosed in this study is comparable to findings from Sweden (3.6%), but higher than that reported in NHANES III (0.38%) [6
]. Leukoplakia is a premalignant lesion with transformation rates varying from 15.6% to 39.2% [55
]. It is highly associated with cigarette smoking [8
]. Although we have not done any further analysis of smoking as a possible risk factor of leukoplakia, the low frequency rate of cigarette smoking concomitant with a relatively high frequency of oral leukoplakia as observed in this study deserves further investigation. The high frequency of leukoplakia should be taken seriously as leukoplakia in non-smokers is more likely to undergo malignant transformation than leukoplakia in smokers [55
A total of 4 patients (0.7%) with oral manifestation of discoid lupus erythematosus (DLE) on vermilion border were diagnosed in this study (Table ). This condition has rarely been registered in OML investigation studies. Axel [5
] reported 0.01% in a Swedish population, while Ramirez et al [21
] reported 5% in lupus patients referred to a dermatology clinic because of oral complaints. The difference between the present figure and that reported by Ramirez et al may be attributed to the fact that although both data were collected in dermatology clinic, the selection of patients was different. The precancerous potential of oral DLE is a controversial topic. Lu and Le [57
] reported an incidence of 13.6% epithelial dysplasia in DLE. Another report from Scully et al [58
] postulated that DLE on the lip showed a premalignant potential. Sun exposure plays a crucial role in the induction or exacerbation of the lupus erythematosus and actinic cheilitis [28
]. In connection to that, Wakisa et al [61
] reported oral cell carcinoma on lips of black patients with oral DLE. Noteworthy the tropical climate in Sudan and the summer temperature which often exceed 43°C has to be considered in interpreting such lesions.
Frequency of recurrent aphthous stomatitis (RAS) has been recorded as life time prevalence, point prevalence and as combination of both. The present study revealed a point prevalence of 2.9%, which is higher than 2% and 0.8% reported by Axell [5
] and NHANES III [6
] respectively. Yet, it was lower than 60% and 55% in US female student nurses and professional school students respectively [62
]. This illustrates how RAS varies according to the study group examined. A number of factors have been attributed to the occurrence of this pathology, including immune dysfunction [28