There is little knowledge regarding the relationship between gastrointestinal symptoms and snus. The nicotine intake is usually higher than in smokers, and a great deal of tobacco juice contaminated saliva is swallowed during use. In a Swedish cross-sectional study of symptoms in 130,000 construction workers, smokers reported "ulcer-like" dyspepsia three times as often 5 as the non tobacco users, while snus users reported significantly fewer symptoms than both smokers and non tobacco users [10].It has been shown both in studies on patients and in population-based studies that smoking is a risk factor for peptic ulcer disease (PUD) [8,11], but there are no data concerning snus on this issue. Data from the US show that cigarette smokers have a markedly increased risk for gastric cancer and that use of more than one tobacco product increases the risk in men [12].Male users of smokeless or chewing tobacco have also been shown in US prospective studies to have higher death rates from all causes compared with non-users (CPS-I and II studies) and also higher death rates from cancer of the gastrointestinal tract overall (CPS-I study), but separate data on the upper gastrointestinal tract were not shown [13].Although there is extensive exposure to swallowed tobacco juice contaminated saliva, and high serum levels of nicotine and nitrites, there have been no studies in terms of oesophageal or gastric histology or Helicobacter pylori infection in snus users.The aim of the study was to investigate whether smoking cigarettes and snus use are associated with gastrointestinal symptomatology, macroscopic findings on endoscopy or histological signs of inflammation or cancer risk markers in the esophagus or stomach, including the prevalence of Helicobacter pylori.
MATERIALS AND METHODS
SettingThe Kalixanda study setting consisted of two neighbouring communities in Northern Sweden (Kalix and Haparanda) with 28,988 inhabitants (December 1998). The distribution of age and sex was similar to the national average in Sweden in both communities; although proportion of unemployment, income, and the proportion with a higher education were slightly lower [14].
ParticipantsBy using the computerized national population register, covering all citizens in the two communities by date of birth order, a representative stratified sample was generated. Every seventh adult (n=3,000) from the target population (18-80 years of age, n=21,610 in September 1998) was drawn. The sampled subjects were given an identification number (1-3,000) in a random order [14].
Study design and logisticsThe original study population (n=3,000) was invited by mail to take part. The invitation included information of the study design and of the aims of the study and a validated questionnaire, the Abdominal Symptom Questionnaire (ASQ) to be returned by mail [15]. Up to two remainders were sent when necessary; 140 subjects were unavailable at the time for invitation, thus 2,860 of the original study population were eligible for inclusion [14]. The overall response rate was 7...