Oral mucosal lesions, tobacco use and the long-term outcome in a Swedish population

Abstract

The overall objectives of this thesis were to study the natural history for Oral Lichen (OLL), Oral Leukoplakia (OL) and Snus Induced Lesions (SIL) and to correlate cancer and mortality among users and non-users of Swedish moist snuff (snus). The population sub-sample comprised a historic cohort of 20,333 individuals, established in 1973-74. At the time for follow-up, data on 20,212 wore still available. The cohort was followed for 27-29 years. Using the National Registration Number, a unique identifier for each individual in Sweden, the dataset was linked to the nationwide registers of the Total Population, Migration, Cancer and Deaths - all of which had been in operation for at least 15 years prior to cohort accrual. For the nearly 30 year follow-up, in the first three papers, a sub-section of the cohort was selected, based on district of residence in 1973-74. This restriction was introduced in order to facilitate the subsequent field-work. when the cohort members were to be re-examined. Those who were still alive and residing in the area were offered a reexamination and 289 (68%) of the 422 invited individuals accepted.

The register-based follow-up with data from The Cancer Register revealed oral cancer in one individual in the OLL cohort and three individuals each in the OL and SIL cohorts. There was no statistically significant increase for oral cancer in the three sub-cohorts. The clinical re-examination disclosed that around 40% of the lesions recorded in 1973-74 were no longer clinically detectable. For the OLL cohort this could be explained by the putative relapsing nature of the lesion and that there were no differentiations made between Oral Lichen Planus and Oral Lichenoid Reactions where a trigger factor might be present. For the OL cohort there was a strong association with smoking cessation and the disappearance of the lesion. Snus use was strongly associated with the appearance of SIL. In all cases where the individual reported cessation of using smis the lesion was no longer clinically discernible.

In the fourth study the subjects comprised all males from the original cohort (9,976). Tobacco habits (snus use and/or smoking) were used as exposure and cancer and mortality as outcome. The data set was linked to the health and population-based registers. Cancer had been notified to the cancer register for 1,575 of the subjects (only initial notification of cancer detected after day of entry into the study was included); 3,630 had died during the follow-up time. There was a statistically significant higher cancer incidence for those who were or had been daily smokers than for those who had never smoked (IRR 1.26, 95% CI 1.13-1.40). However, no such difference emerged with respect to snus habits (IRR 1.00).With respect to oropharyngeal cancer, there was a statistically significantly higher incidence among those who were or had been snus users compared to those who had never used snus, based on 11 exposed cases (IRR 3.1, 95% CI 1.5-6.6). For mortality ever daily smoking was strongly associated with death within the categories all cause mortality, cancer death, circulatory death and respiratory death. Among ever daily snus users there was a 10% increase for all cause mortality compared to never snus users.

The following conclusions may be drawn from these results: There is no support for an increased incidence of oral cancer at the lesion sites. Both appearance and location of OLL change over time. There is a strong association between OL and smoking. The presence of SIL may be considered a marker for the use of snus. Use of smis seems to have a negative effect on health. It is associated with some increased risk for oropharyngeal cancer and overall mortality.