Prevalence of oral mucosal normal variations and lesions in a middle-aged population: a Northern Finland Birth Cohort 1966 study

The study population is part of the longitudinal Northern Finland Birth Cohort 1966 (NFBC 1966) [20]. The entire cohort included 12,055 mothers whose expected delivery was in the year 1966. Altogether 12,231 children were born, 12,058 of whom were born alive. NFBC 1966 data have been collected with health questionnaires and clinical examinations at the ages of around 1, 14, 31 and 46 years. Since the 46-years clinical examination was performed between April 2012 and June 2013, and depending on the month the examinee was born, their age ranged from 45 to 47 years. Health questionnaires and written data from the oral mucosal clinical examination and pictures of 1,961 participants were also collected; see details below [20].

In connection with the 46-year examination, all cohort members were sent the postal questionnaires, which they filled in either on the internet or on paper. The questionnaire covered general health, sex, alcohol intake, smoking and use of snuff. Following the questionnaire, an invitation to a clinical examination was sent to all cohort participants.

Smoking status was assessed in the postal questionnaire as follows: “Have you ever smoked in your life?” (no, yes, I started at the age of XX), “Have you smoked regularly?” (no, yes, I have smoked regularly for a total of XX years), “If you have quit smoking, what age were you when you stopped?”, “Do you smoke currently?” (7 days a week, 5–6 days a week, 2–4 days a week, 1 day a week, occasionally, not at all). Usage of snuff was assessed as follows: “Do you use snuff currently?” (no, occasionally, regularly).

Alcohol consumption was queried in the questionnaire as follows: “How often do you usually drink beer, cider, long drinks or wine, fortified wines, homemade wines or hard liquors?” (never, less than once a year, a couple times a year, 3–4 times a year, once every 2 months, once a month, a couple times a month, once a week, a few times a week, daily), “How much beer, cider or long drinks do you usually drink on one occasion? (1 bottle = 1/3 L)” (less than 1 bottle, 1 bottle, 2 bottles, 3 bottles, 4–5 bottles, 6–9 bottles, 10–14 bottles, 15 bottles or more, I don’t drink any of these beverages), “How much wine, fortified wine or homemade wine do you usually drink on one occasion? (half glass, one glass (= 16 cl), a couple of glasses, about half of the bottle (large bottle = ¾ l), a bit more than a large bottle, about one large bottle, one or two large bottles, more than two large bottles, I don’t drink wine), “How much hard liquor do you usually drink on one occasion? (less than 1 tavern portion (less than 4 cl), 1 tavern portion (about 4 cl), 2 tavern portions, 3–4 tavern portions, 5–6 tavern portions, 7–10 tavern portions, about a half-litre bottle, more than a half-litre bottle, I don’t drink hard liquor).

In the 46-year follow-up survey, altogether 3150 persons currently living in the city of Oulu or within a radius of 100 km were asked to participate in the oral examination, including oral mucosa and dental examination. Of the original subgroup, a total of 1961 persons (62.3%) attended the clinical oral examination and gave their consent to use the data; these persons comprised the present study population. The clinical oral examination was performed between April 2012 and June 2013.

The oral examination was conducted by seven general dentists during the period 2012–2013. All examiners used a standard clinical dental examination protocol. They were trained and calibrated by Oulu University dental clinic teachers, not including oral medicine specialists, before the examination and at the middle of the study period. All examinations were made in the Oulu University dental clinic with modern supplies and optimal light. Clinical examination of the oral cavity included a careful and systematic visual inspection of the whole oral cavity. Dentists also made a palpation for all areas of the oral mucosa. Biopsy was not taken. All oral mucosal changes were written documented and photographed during the clinical examinations. All examiners used the same type of camera (CANON 600D 60 mm Lens Macro 1:2,8, Flash Sigma EM-140 DG).

Afterwards, oral pathologist (TS) and oral mucosal disease specialists (AK, TS) simultaneously re-analysed all the pictures and, based on them and on the documentation by the general dentists, they provided consensus diagnoses for all the findings. Specialists disagreed in 24.3% of the lesions and agreed in 54.8% of the proposed diagnosis by the general dentists. Of the documented lesions, 20.9% had no proposed diagnosis by the general dentists and were diagnosed afterwards based documentation and pictures by two specialists. The total number of examinees who had oral mucosal lesions was 206. There were no false-positive cases.

The study population was described using frequencies and percentages with the categorical variables gender, smoking, snuff use and alcohol intake as shown in Table 1. Smoking was categorised as current smokers, former smokers and non-smokers, and snuff usage as snuff-users and non-users. The consumption of alcohol (g/day) was calculated based on the questions concerning alcohol usage. Alcohol intake was categorised such that the heavy user limit for females was 20 g/day and for males 30 g/day.

The number of mucosal lesions per person was calculated, and lesions and diseases were categorised into 11 main groups based on the Regezi et al. textbook [7] as following: no changes, normal variations, vesiculobullous lesions, ulcerative conditions, white lesions, red-blue lesions, pigmented lesions, verruca-papillary lesions, connective tissue lesions (CTL), divided by region into tongue and buccal mucosal swellings, gingival swellings and floor of mouth swellings.

Cross-tabulation with Chi-square or Fisher’s exact test was used to compare the numbers of mucosal lesions, submucosal swellings, oral mucosal variations and mucosal diseases between genders. Additionally, the association between lesions and smoking within gender as well as association between lesions and gender within smoking have been analysed using cross-tabulation with Chi-square or Fisher’s exact test. Logistic regression analysis was used to present crude odds ratios (OR) and their 95% confidence intervals (CI) when comparing current and former smokers with non-smokers, snuff users with non-users and heavy alcohol users with moderate users or non-users. The adjusted logistic regression model was executed including smoking, snuff use and alcohol in the same model.

The total number of clinically examined participants was 1961; women comprised 53.5% (n = 1050) of the population. The mean age was 46 years. The number of participants who answered the questions concerning smoking was 1876 (95.7%). We divided smokers into three categories: current smokers (18.3%), former smokers (28.5%) and non-smokers (53.3%). Concerning snuff use, 1890 participants (96.4%) answered the questions: 1012 females (96.3%) and 878 males (96.3%). The proportion of snuff users was 3.7%, of which 0.2% were females and 7.7% males of the entire population. Concerning alcohol intake, 1905 participants (97.1%) answered the questions: 882 males (46.3%) and 1023 females (53.7%). In this study, 8.0% of females and 17.0% of males were heavy users (Table 1).

Altogether, 55 different kinds of changes were recorded; 81.8% were diagnosed as OMLs and 18.2% belonged to the normal variations category. When we categorised lesions into the main groups, the most common group was white lesions (6.5%, n = 128), followed by normal variations (3.3%, n = 64) and red-blue lesions (1.3%, n = 25). White lesions were more often diagnosed in males (8.1% vs. 5.1%; p = 0.008), whereas pigmented lesions were more common in females (1.0% vs. 0.3%; p = 0.090). Normal variations were seen more often in males (4.1%, n = 37; p = 0.064) (Table 2).

The most common normal variations were the following: Fordyce granules 1.2% (n = 24), fissured tongue 1.1% (n = 22), geographic tongue 0.9% (n = 18) and hairy tongue 0.2% (n = 3) (Table 3) [3, 6].

Of the oral mucosal lesions, the most common findings were OLR 2.0% (n = 39), OLP 1.5% (n = 29), hyperkeratosis 0.7% (n = 14), fibroma 0.7% (n = 13), snuff-related lesions 0.5% (n = 10), amalgam pigmentation 0.4% (n = 7), median rhomboid glossitis 0.4% (n = 7), hemangioma 0.3% (n = 6), ulcer 0.3% (n = 6) and candidiasis 0.3% (n = 5). Leukoplakia was registered in only 0.5% (n = 9) of all participants (n = 1961) (Table 4).

The two most common lesions, OLR and OLP, grouped together into OLD, were found in 3.5% of the study population: 3.1% in females and 3.9% in males. OLP was slightly more common among females (1.8% vs. 1.2%; p = 0.354), as was hyperpigmentation associated with amalgam (0.6% vs. 0.1%; p = 0.131). Males had OLR significantly more often (2.7% vs. 1.3%; p = 0.026), as well as hyperkeratosis (1.0% vs. 0.5%; p = 0.179). Additionally, Fordyce granules (2.3% vs. 0.3%; p < 0.001) and snuff-related lesions (1.0% vs. 0.0%; p = 0.001) were more common in males than females (Tables 3, 4).

In current smokers, OLP (3.5%, p = 0.001), OLR (4.1%, p = 0.016), hyperkeratosis (1.7%, p = 0.007) and snuff-related lesions (1.5%, p = 0.004) were more common than in non- or former smokers. Of all the current smokers with OLP, 75% (n = 9, p = 0.095) were females. OLR was more common in current smokers (4.1%: males 4.8% and females 3.4%), followed by former smokers (1.7%: males 2.4% and females 0.8%) and non-smokers (1.6%: males 2.4% and females 1.0%). Similarly, hyperkeratosis was more common in current smokers (1.7%, n = 6), compared to former (0.7%, n = 4) and non-smokers (0.2%, n = 2) (p = 0.007). Among males, hyperkeratosis was detected in 2.4% (n = 4) of current smokers, 0.3% (n = 1) of former and 0.5% (n = 2) of non-smokers (p = 0.046). Among females, hyperkeratosis was detected in 1.1% (n = 2) of current smokers, 1.2% (n = 3) of former and in no non-smokers (p = 0.029). Hyperpigmentation (0.6%, n = 2; p = 0.033) was only found in current smokers. Snuff-related lesions were present in 1.5% (n = 5) of current smokers, 0.7% (n = 4) of former smokers and 0.1% (n = 1) of non-smokers (p = 0.004). All snuff-related lesions were found in males (2.4%, n = 4, p = 0.013). Leukoplakia was found in 9 (0.5%) of all participants, and 5 (1.5%) of them were current smokers, 2 (0.4%) were former and 2 (0.2%) were non-smokers (p = 0.024) (Table 5).

Of normal variations, geographic tongue was seen in 1.5% (n = 15) of non-smokers (p = 0.031). In females, 1.1% (n = 2) of current smokers, no former smokers and 1.7% (n = 10) of non-smokers had geographic tongue (p = 0.089). In males, 1.2% (n = 5) of non-smokers had geographic tongue. Fordyce granules were more common in non-smokers (1.5%, n = 15), particularly in males (3.1%, n = 13, p < 0.001), compared to current smokers (0.9%, n = 3) or former smokers (0.7%, n = 4) (p = 0.364) (Table 5).

When we compared non-smokers with current smokers and former smokers, the OR for current smokers was 3.0 (95% CI 2.11–4.28) for any mucosal changes. Use of snuff, occasionally or regularly, raised the risk of having mucosal changes (OR 2.6, 95% CI 1.48–4.70). Alcohol intake affected the risk for mucosal changes (OR 1.3, 95% CI 0.82–1.91). When we included smoking, snuff use and alcohol intake in the same adjusted model, the smoking and snuff remained significant (Table 6).